Predicting Cognitive Functioning, Activities of Daily Living, and Participation 6 Months after Mild to Moderate Stroke

Predicting Cognitive Functioning, Activities of Daily Living, and Participation 6 Months after... Abstract Objective Predicting neurocognitive and functional outcomes in stroke is an important clinical task, especially in rehabilitation settings. We assessed acute predictors of cognitive and functional outcomes 6 months after mild to moderate stroke. Methods We conducted a retrospective analysis of acute clinical data and 6-month follow-up telephone interviews for 498 mild to moderate stroke patients. Predictors were sociodemographic variables, the National Institute of Health Stroke Scale (NIHSS), basic physical measures, the Mesulam Cancellation Test, the Short Blessed Test (SBT), Trails A/B, and the Boston Naming Test. The outcome variables were the Communication, Memory and Thinking, ADL/IADLs, and Participation subscales from the Stroke Impact Scale. We conducted four hierarchical multiple regression analyses with demographic variables and the NIHSS score entered into the first step, followed by physical variables in the second step, and neuropsychological variables in the final step. Results Physical variables explained more variance in ADL/IADLs and Participation outcomes than in Communication and Memory and Thinking outcomes, while cognitive predictors exhibited the opposite trend. The SBT was the only significant independent predictor of Communication and Memory and Thinking (p’s < .001), while the NIHSS was the only measure that significantly predicted ADL/IADLs (p < .001) and Participation (p = .002). Poorer performance on screening measures predicted worse cognitive and functional outcomes 6 months post-stroke. Conclusions These results support the clinical utility of administering brief screening instruments during acute recovery from mild to moderate stroke. Neuropsychologists should prioritize performance on screening measures assessing acute neurologic status and cognitive dysfunction when making recommendations for post-stroke rehabilitation. Stroke, Rehabilitation, Assessment, Everyday functioning Introduction Mild to moderate stroke is associated with a wide range of distressing and disabling physical, neuropsychological, and functional impairments (Harvey, 2015; Mayo, Wood-Dauphinee, Côté, Durcan, & Carlton, 2002). As a result, a large proportion of patients who suffer such an event have difficulty successfully reintegrating into prior social and occupational roles (Adamit et al., 2015; Van Patten, Merz, Mulhauser, & Fucetola, 2016; Woodman, Riazi, Pereira, & Jones, 2014). Restricted participation in life situations and activities is common after stroke and poses a significant challenge for survivors navigating role changes in multiple life domains, including occupational and family roles (Ch’Ng, French, & Mclean, 2008; Törnbom, Persson, Lundälv, & Sunnerhagen, 2017). As many as 65% of stroke survivors face limitations in work, recreation, or social activity, all of which serve as barriers to return to normal living (Adamit et al., 2015; Mayo et al., 2002; Törnbom et al., 2017). Importantly, the failure to resume meaningful social, vocational, or leisure activities is also associated with worse health outcomes, as well as lower health-related and global quality of life (Bhogal, Teasell, Foley, & Speechley, 2003; Norlander et al., 2016). Therefore, elucidating factors that influence level of participation is essential for understanding recovery trajectory and outcomes for mild to moderate stroke (Adamit et al., 2015). Given the importance of post-stroke engagement to health and quality of life, investigators have sought to determine which factors limit participation. In particular, a number of functional and disability factors limit individuals’ participation in necessary and valued activities after stroke (Jette, Keysor, Coster, Ni, & Haley, 2005; Mayo et al., 2015). Limited mobility, depressive symptoms, apathy, cognitive impairment, fatigue, difficulty performing activities of daily living (ADLs), communication problems, lack of social connection, and lack of self-efficacy all act to restrict participation (Desrosiers et al., 2006; Hoyle, Gustafsson, Meredith, & Ownsworth, 2012; Pang, Eng, & Miller, 2007). Studies of patients recovering from mild to moderate stroke have found that, while they typically achieve independence in self-care and everyday functioning (commonly referred to as ADLs), they may require additional supports when returning to work or leisure activities or have problems with communication or carrying out instrumental activities of daily living (IADLs; Almkvist Muren, Hütler, & Hooper, 2008; Aufman, Bland, Barco, Carr, & Lang, 2013; Edwards, Hahn, Baum, & Dromerick, 2006; Van Patten et al., 2016). In contrast to basic ADLs, IADLs are more complex tasks that are necessary for independent functioning in the community, such as household chores, shopping, taking medications, and managing finances (Hoffmann, McKenna, Cooke, & Tooth, 2003). As many as 54% of stroke survivors have difficulty performing IADLs independently, which is similar to the percentage of those reporting restrictions in participation (Adamit et al., 2015; Mayo et al., 2002; Törnbom et al., 2017). This is particularly significant given the importance of ADLs and IADLs for successful engagement in social, occupational, and recreational activities following stroke (Cioncoloni et al., 2013; Hoffman et al., 2003). Among other factors, cognitive impairment has been shown to affect the ability of individuals to perform ADL/IADLs and to participate in meaningful activities after stroke (Babulal, Huskey, Roe, Goette, & Connor, 2015; Pohjasvaara, Vataja, Leppävuori, Kaste, & Erkinjuntti, 2002; Tatemichi et al., 1994). Post-stroke cognitive deficits are common, even following mild stroke, and can have a substantial affect on recovery and functional outcome (Adamit et al., 2015; Heruti et al., 2002; Jokinen et al., 2015). Although stroke has been associated with deficits in a broad range of cognitive domains including attention, memory, language, motor control, visuospatial skills, and executive functioning, the most consistent effects appear to be impairments in processing speed, attention, and executive functions (Cumming, Marshall, & Lazar, 2013). However, depending on lesion location, domain-specific deficits, such as impaired receptive and/or expressive language abilities, may occur more frequently (Laska, Hellblom, Murray, Kahan, & Von Arbin, 2001). Approximately one-third of stroke patients experience significant language disturbance known as aphasia, though many nonaphasic patients exhibit less severe impairments in complex language and communication skills (Pedersen, Stig Jørgensen, Nakayama, Raaschou, & Olsen, 1995). Srikanth et al. (2003) examined nonaphasic patients with first-ever mild to moderate stroke 3 months post-injury and noted deficits in both complex language and executive functioning. The prevalence of post-acute cognitive impairment has been estimated to be between 20% and 40%, with a similar percentage of stroke survivors reporting persisting difficulties with language and communication (Douiri, Rudd, & Wolfe, 2013; Laska et al., 2001; Makin, Turpin, Dennis, & Wardlaw, 2013; Oksala et al., 2009; Patel, Coshall, Rudd, & Wolfe, 2002). The deleterious effects of impairments in cognition, communication, and ADL/IADLs on health outcomes and quality of life after stroke have been well-established (Hoyle et al., 2012). Consequently, a number of studies have sought to identify acute predictors of these cognitive and functional outcomes, with the goal of tailoring rehabilitation services and enhancing clinical care. With regard to reestablishing independence in ADLs, studies suggest that younger age, decreased stroke severity, and better motor and functional abilities at stroke onset are most associated with recovery in basic ADLs beyond 3 months after stroke (Harvey, 2015; Veerbeek, Kwakkel, van Wegen, Ket, & Heymans, 2011). There is also some evidence that being male, having intact cognitive and language abilities (particularly absence of aphasia), and lack of health comorbidities (e.g., diabetes mellitus) predict patients’ ability to regain independence in basic ADLs (Duarte et al., 2010; Protopsaltis et al., 2009; Loewen & Anderson, 1990; Tilling et al., 2001). In contrast, acute predictors associated with post-stroke functional recovery in IADLs are not well understood (Cioncoloni et al., 2013). Currently, the prognostic variable most associated with independence in IADLs beyond 3 months after stroke is basic ADL status at hospital discharge (Cioncoloni et al., 2013; Hoffman et al., 2003). Older age, being female, greater stroke severity, poor upper limb strength, and acute cognitive impairment have also been implicated in worse recovery of IADLs beyond 3 months after stroke (Cioncoloni et al., 2013; Hoffman et al., 2003; Mok et al., 2004; Nys et al., 2005). While a number of studies have examined acute predictors of functional outcomes after stroke, fewer have examined predictors of post-stroke cognitive impairment (Nys et al., 2005). Thus far, demographic and stroke-related factors that have been associated with cognitive impairment beyond 3 months after stroke include older age, female sex, minority status, lower socioeconomic status, lower education, pre-stroke cognitive decline, previous stroke, left hemisphere stroke, and greater stroke severity (Lazar et al., 2010; Mok et al., 2004; Nys et al., 2005; Patel et al., 2002; Sachdev, Brodaty, Valenzuela, Lorentz, & Koschera, 2004). Additionally, both global cognitive impairment and domain-specific deficits in executive functioning, visuospatial skills (e.g., unilateral neglect), and language (e.g., aphasia) at stroke onset have been implicated in post-stroke cognitive impairment (Hillis, Wityk, Barker, Ulatowski, & Jacobs, 2003; Kalra, Perez, Gupta, & Wittink, 1997; Laska et al., 2001; Leśniak, Bak, Czepiel, Seniów, & Członkowska, 2008; Nys et al., 2005; Pendlebury, Cuthbertson, Welch, Mehta, & Rothwell, 2010). For example, acute language disturbance, including aphasia, is the strongest predictor of impaired communication abilities beyond 3 months after stroke (Harvey, 2015). A probable explanation for the prognostic value of acute cognitive deficits is the relative constancy of post-stroke cognitive impairment, even following mild stroke (Del Sur et al., 2005; Jacquin et al., 2014; Wolf & Rognstad, 2013). Pre-post assessments suggest that some cognitive deficits remain stable from acute measurement (within 24 hr) to 3 months post-stroke, with even fewer changes in cognition occurring between subacute assessment (within 3 weeks of discharge) and follow-up assessment 6 months post-stroke (Riepe, Riss, Bittner, & Huber, 2004; Wolf & Rognstad, 2013). Moreover, the overall prevalence of post-stroke cognitive impairment appears to remain relatively unchanged beyond 3 months after stroke for as long as 14 years (Del Sur et al., 2005; Douiri et al., 2013). Long-term recovery of independence in ADL/IADLs may follow a similar trajectory, owing largely to early recovery of physical abilities and acute ADL status (Cioncoloni et al., 2013; Hoffman et al., 2003). Overall, the literature on cognitive and functional outcomes post-stroke supports the predictive value of cognitive, physical, and functional status at stroke onset, in addition to certain neurologic and demographic factors. Identifying acute predictors of cognitive and functional outcomes following mild to moderate stroke is an important goal for research and clinical care. Recovery from stroke is highly variable and deficits will often resolve naturally over time, with compensation, and/or due to rehabilitation efforts (Cumming et al., 2013). In particular, early initiation and maintenance of targeted rehabilitation services has been associated with better recovery after stroke (Bhogal et al., 2003; Maulden et al., 2005). Therefore, identifying acute factors that meaningfully predict long-term outcomes can improve predictive models for stroke recovery, as well as help inform rehabilitation services for patients recovering from stroke. Although various aspects of acute cognitive, physical, and functional status have been suggested to play a role in determining outcomes after stroke, empirical support for their prognostic value is often limited by methodological challenges (Harvey, 2015). Furthermore, few studies have conducted comprehensive investigations examining the incremental predictive validity of specific cognitive and physical/functional assessments commonly administered in acute stroke. The purpose of the current study was to systematically assess acute predictors of cognitive and functional outcomes 6 months after mild to moderate stroke. Specifically, given previous research suggesting that cognitive tests predict important cognitive and functional outcomes (e.g., Babulal et al., 2015; Harvey, 2015; Hillis et al., 2003; Kalra et al., 1997; Laska et al., 2001; Leśniak et al., 2008; Nys et al., 2005; Pohjasvaara et al., 2002; Tatemichi et al., 1994), we hypothesized that cognitive performance would demonstrate incremental validity above and beyond physical/functional abilities and demographic and stroke-related factors with regard to predicting impairments in cognition, communication, participation, and/or ADL/IADLs 6 months after stroke. Methods Participants The dataset utilized in the current study was acquired from the Brain Recovery Core (BRC; Lang et al., 2011), a collaborative endeavor among a university-affiliated medical center, an acute care hospital, and a rehabilitation institute in a mid-sized Midwestern city in the United States. We collected demographic and clinical data from stroke patients’ acute care hospital records and 6-month follow-up data from recorded responses to telephone interviews. As part of the clinical services provided at these institutions, all patients are afforded the opportunity to provide informed consent for the use of their clinical data for research purposes. The current study sample represents the set of stroke patients who consented to release their data between 2010 and 2014. Data from 602 stroke patients who completed the 6 month follow-up interview were acquired from the BRC. Inclusionary criteria for all patients were simply that (a) the patient received clinical services for an acute stroke through the BRC between 2010 and 2014, (b) the patient voluntarily provided informed consent, and (c) the patient completed the 6 month follow-up telephone interview. Patients were excluded for the following reasons: (a) a National Institute of Health Stroke Scale (NIHSS) score >16 (indicating severe stroke), (b) a NIHSS Aphasia item score of 2 or 3, indicating severe to global aphasia (given that aphasia is likely to compromise the ability of patients to provide valid responses to oral survey questions), and (c) a NIHSS Dysarthria item score of 2 or 3, indicating severe dysarthria or intubation (given that dysarthria will also prevent adequate verbal expression). Following these exclusions, the final sample size was 498 (mean age = 64.50 ± [SD] 14.54 years; mean NIHSS total score = 3.55 ± [SD] 3.76). A full characterization of the sample demographics is provided in Table 1. Table 1. Group characteristics (n = 498) Characteristic Mean ± SD or % N Range Age 64.50 ± 14.54 498 21–98 Education (years) 12.87 ± 2.70 498 0–22 Sex  Male 51.00 254  Female 49.00 244 Race  Caucasian 62.90 313  African American 33.90 169  Asian 0.40 2  Unknown 2.80 14 Marital status  Married 45.20 225  Significant other 4.40 22  Divorced 11.40 57  Never married 0.60 3  Widowed 14.70 73  Separated 1.40 7  Single 18.70 93  Unknown 3.40 17 Working prior to stroke  Yes 27.50 137  No 61.20 305  Unknown 11.20 56  NIHSS total score 3.55 ± 3.76 498 0–16  Short Blessed Test 5.92 ± 5.87 458 0–28 Stroke diagnosis  Ischemic 60.60 302  Hemorrhagic 10.80 54  Unknown 28.50 142  PHQ-9 at 6-month follow-up 6.42 ± 6.35 498 0–27 Minimal-49.40% Mild-24.70% Moderate-13.65% Mod-Sev-6.22% Severe-6.02% Characteristic Mean ± SD or % N Range Age 64.50 ± 14.54 498 21–98 Education (years) 12.87 ± 2.70 498 0–22 Sex  Male 51.00 254  Female 49.00 244 Race  Caucasian 62.90 313  African American 33.90 169  Asian 0.40 2  Unknown 2.80 14 Marital status  Married 45.20 225  Significant other 4.40 22  Divorced 11.40 57  Never married 0.60 3  Widowed 14.70 73  Separated 1.40 7  Single 18.70 93  Unknown 3.40 17 Working prior to stroke  Yes 27.50 137  No 61.20 305  Unknown 11.20 56  NIHSS total score 3.55 ± 3.76 498 0–16  Short Blessed Test 5.92 ± 5.87 458 0–28 Stroke diagnosis  Ischemic 60.60 302  Hemorrhagic 10.80 54  Unknown 28.50 142  PHQ-9 at 6-month follow-up 6.42 ± 6.35 498 0–27 Minimal-49.40% Mild-24.70% Moderate-13.65% Mod-Sev-6.22% Severe-6.02% Note: NIHSS = National Institutes of Health Stroke Scale; PHQ-9 = Patient Health Questionnaire, 9-item. Table 1. Group characteristics (n = 498) Characteristic Mean ± SD or % N Range Age 64.50 ± 14.54 498 21–98 Education (years) 12.87 ± 2.70 498 0–22 Sex  Male 51.00 254  Female 49.00 244 Race  Caucasian 62.90 313  African American 33.90 169  Asian 0.40 2  Unknown 2.80 14 Marital status  Married 45.20 225  Significant other 4.40 22  Divorced 11.40 57  Never married 0.60 3  Widowed 14.70 73  Separated 1.40 7  Single 18.70 93  Unknown 3.40 17 Working prior to stroke  Yes 27.50 137  No 61.20 305  Unknown 11.20 56  NIHSS total score 3.55 ± 3.76 498 0–16  Short Blessed Test 5.92 ± 5.87 458 0–28 Stroke diagnosis  Ischemic 60.60 302  Hemorrhagic 10.80 54  Unknown 28.50 142  PHQ-9 at 6-month follow-up 6.42 ± 6.35 498 0–27 Minimal-49.40% Mild-24.70% Moderate-13.65% Mod-Sev-6.22% Severe-6.02% Characteristic Mean ± SD or % N Range Age 64.50 ± 14.54 498 21–98 Education (years) 12.87 ± 2.70 498 0–22 Sex  Male 51.00 254  Female 49.00 244 Race  Caucasian 62.90 313  African American 33.90 169  Asian 0.40 2  Unknown 2.80 14 Marital status  Married 45.20 225  Significant other 4.40 22  Divorced 11.40 57  Never married 0.60 3  Widowed 14.70 73  Separated 1.40 7  Single 18.70 93  Unknown 3.40 17 Working prior to stroke  Yes 27.50 137  No 61.20 305  Unknown 11.20 56  NIHSS total score 3.55 ± 3.76 498 0–16  Short Blessed Test 5.92 ± 5.87 458 0–28 Stroke diagnosis  Ischemic 60.60 302  Hemorrhagic 10.80 54  Unknown 28.50 142  PHQ-9 at 6-month follow-up 6.42 ± 6.35 498 0–27 Minimal-49.40% Mild-24.70% Moderate-13.65% Mod-Sev-6.22% Severe-6.02% Note: NIHSS = National Institutes of Health Stroke Scale; PHQ-9 = Patient Health Questionnaire, 9-item. Demographic and clinical data included in the current study was obtained from hospital records and typically collected within 7 days of patients’ admission to the hospital (mean number of days = 1.86 ± [SD] 3.04, maximum = 46 days). Follow-up data was obtained from telephone interviews conducted approximately 6-months after stroke onset and hospital admission. Variables Assessed Demographic and stroke-related variables The present study assessed the following demographic and neurologic variables: age, gender, race, marital status, stroke type [ischemic, hemorrhagic, or unknown], and NIHSS total score. The NIHSS is a broad 15-item neurological screening exam of stroke severity that briefly assesses for deficits in the areas of consciousness level, language, neglect, visual-field loss, extraocular movement, motor strength, ataxia, dysarthria, and sensory loss (Goldstein et al., 1989). Physical/functional variables Physical and functional status was assessed with the Berg Balance Scale (BBS) total score, a measure of ambulation (in feet), and functional mobility tasks including Supine to Sit and Sit to Stand. The BBS is a 14-item measure of balance and risk for falls in older adults measured through direct observation on a scale of 0 (inability to complete tw:he task) to 4 (independent task completion; Berg, Wood-Dauphinee, & Williams, 1995). Supine to Sit and Sit to Stand are commonly used measures of fall risk, muscle strength, and weight symmetry (Cheng et al., 1998; Lomaglio & Eng, 2005; Wattanapan, Kovindha, Piravej, & Kuptniratsaikul, 2010). These tasks were measured on a scale of 1 (requiring total assistance) to 7 (complete independence). Cognitive variables Cognitive variables included the Mesulam Cancellation Test total score, Short Blessed Test (SBT), Trails A and B, and the Boston Naming Test (BNT). The Mesulam Cancellation Test is a measure of visual attention and neglect measured by speed of completion (Mesulam, 1985). The SBT is a 6-item screener for deficits in orientation, registration, and attention (Carpenter et al., 2011). Trails A and B are measures of attention and executive functioning, measured by speed of completion, and the BNT is a 60-item language measure of visual confrontation naming (Strauss, Sherman, & Spreen, 2006). Cognitive and functional outcomes We assessed post-stroke cognitive and functional outcomes with selected subscales of the Stroke Impact Scale (SIS; Vellone et al., 2015). The SIS is a self-report instrument that measures quality of life and functional impairments across a range of behaviors and activities following stroke. For the purposes of the current study, the Communication, Memory and Thinking, ADL/IADLs, and Participation subscales from the SIS were included in analyses. The Communication subscale assesses various aspects of both receptive and expressive language. The Memory and Thinking subscale measures retrospective and prospective memory, as well as processing speed, problem solving, and attentional abilities. The ADL/IADLs subscale assesses difficulty in self-care and daily activities (e.g., dressing and bathing), as well as more complex tasks necessary for independent functioning in the community, such as household chores and shopping. Finally, the Participation subscale measures degree of restriction in activities such as work and recreation. Statistical Analyses Data were analyzed using SPSS 24.0. Consistent with other retrospective investigations of the BRC dataset (Aufman et al., 2013; Van Patten et al., 2016), several variables of interest exhibited notable (>10%) missing data. Specifically, data from the following variables were incomplete: stroke type (28.6% missing), the BBS (11.85% missing), FIM Supine to Sit (20.28% missing), FIM Sit to Stand (8.03% missing), Ambulation (22.89% missing), the SBT (8.03% missing), Trails A (31.93% missing) and B (47.19% missing), the Mesulam Cancellation Test Total score (40.16% missing), and the BNT (14.86% missing). Although we were unable to determine the reason for missing data on a case-by-case basis, we did examine the overall mechanism of missingness using the widely accepted Little (1988) criteria. Specifically, we grouped the eight aforementioned variables with >10% missing data into missing and not-missing groups, then assessed group differences across variable with small proportions (≤10%) of missing data (i.e., age, sex, race, years of education, marital status, NIHSS Total Score, SBT, and PHQ-9 Total Score). Such an analysis provides insight into whether data are missing completely at random (MCAR) or missing at random (MAR; see Schafer & Graham, 2002). That is, it determines whether missing data are dependent on other variables in the dataset or not. Importantly, empirical evidence has repeatedly demonstrated that simple case deletion and mean imputation methods lead to biased parameter estimates, most notably when the mechanism of missingness is nonrandom (Newman, 2003; Rubin, 1976; Schafer & Graham, 2002; Scheffer, 2002). Alternatively, the multiple imputation (MI) technique significantly reduces inaccuracy when estimating missing data values, partially by accounting for random error and thereby maintaining the variance structure of a dataset even when data are MAR (Collins, Shafer, & Kam, 2001; Rubin, 1987; Schafer & Graham, 2002). Results of the mechanism of missingness analysis showed that 23/64 (36%) of these nonparametric tests were significant, indicating that our missing data are MAR rather than MCAR. Moreover, 4/8 (50%) of the NIHSS analyses and 5/8 (63%) of the SBT analyses were significant, suggesting that those patients who were more cognitively impaired (as measured by the NIHSS and SBT) were more likely to exhibit missing data. Therefore, given the unambiguous support in simulation studies for the MI approach compared to more simplistic techniques (Newman, 2003; Schafer & Graham, 2002; Scheffer, 2002), we conducted a 5-iteration MI procedure prior to running the primary analysis of interest. To address our primary aim, we conducted four hierarchical multiple regression analyses, with demographic variables (age, gender, race, marital status, stroke type [ischemic or hemorrhagic], and Total NIHSS score) entered into the first step, physical variables (FIM Supine to Sit, FIM Sit to Stand, BBS Total Score, and Ambulation [in feet]) entered into the second step, and neuropsychological variables (Mesulam Cancellation Test Total score, SBT, Trails A and B, and the BNT) entered into the final step. For all physical and neuropsychological variables, raw data were analyzed. Although it would have been possible to use standardized data, this option was considered to be suboptimal given that (a) relevant demographic variables were already controlled in the regression analyses and (b) the process of standardizing raw scores can reduce the variance in a given test, thereby restricting its range and artificially attenuating correlations. The outcome variables were selected subscales from the SIS including Communication, Memory and Thinking, ADL/IADLs, and Participation. Prior to conducting the analyses, the data were examined for outliers, influential cases, and assumptions of linear regression models. Outliers were operationalized as values ±3.29 standard deviations from the mean on the variable of interest. Consistent with established guidelines (Field, 2009), these values were then replaced with values equal to 3 SDs from the mean in the same direction. Variables with identified outliers were FIM Sit to Stand, Ambulation (in feet), the SBT, Trails A and B, and the Mesulam Cancellation Test. Generally accepted criteria (Field, 2009) were utilized to identify influential cases and violations of assumptions. Influential cases were defined as Cook’s distances >1 or DFBetas >1. With respect to statistical assumptions, prior to examining each regression model, multicollinearity was examined through a correlation matrix of the relationships among predictors (and defined as coefficients > ±0.9; see Tables 2 and 3). Next, both linearity of the modeled relationship and homoscedasticity were tested through visual inspection of plotted standardized residuals compared to predicted values. Third, independent errors were defined as residuals with Durbin–Watson values > ±4 and violations of the assumption of normally distributed errors were examined through visual inspection of the histogram of residuals. Unless otherwise specified, no influential cases were present and all assumptions were met for each of the models. Table 2. Correlations among physical/functional variables Measure 1 2 3 4 1. Berg Balance Scale — 2. Supine to Sit 0.763** — 3. Sit to Stand 0.863** 0.823** — 4. Ambulation (feet) 0.533** 0.358** 0.437** — Measure 1 2 3 4 1. Berg Balance Scale — 2. Supine to Sit 0.763** — 3. Sit to Stand 0.863** 0.823** — 4. Ambulation (feet) 0.533** 0.358** 0.437** — *p < .05; **p < .01; ***p < .001. Table 2. Correlations among physical/functional variables Measure 1 2 3 4 1. Berg Balance Scale — 2. Supine to Sit 0.763** — 3. Sit to Stand 0.863** 0.823** — 4. Ambulation (feet) 0.533** 0.358** 0.437** — Measure 1 2 3 4 1. Berg Balance Scale — 2. Supine to Sit 0.763** — 3. Sit to Stand 0.863** 0.823** — 4. Ambulation (feet) 0.533** 0.358** 0.437** — *p < .05; **p < .01; ***p < .001. Table 3. Correlations among cognitive variables Measure 1 2 3 4 5 1. Short Blessed Test — 2. Trails A (seconds) 0.520** — 3. Trails B (seconds) 0.607** 0.789** — 4. Cancellation Test 0.026 0.226** 0.089 — 5. Boston Naming Test −0.618** −0.379** −0.480** 0.007 — Measure 1 2 3 4 5 1. Short Blessed Test — 2. Trails A (seconds) 0.520** — 3. Trails B (seconds) 0.607** 0.789** — 4. Cancellation Test 0.026 0.226** 0.089 — 5. Boston Naming Test −0.618** −0.379** −0.480** 0.007 — *p < .05; **p < .01; ***p < .001. Table 3. Correlations among cognitive variables Measure 1 2 3 4 5 1. Short Blessed Test — 2. Trails A (seconds) 0.520** — 3. Trails B (seconds) 0.607** 0.789** — 4. Cancellation Test 0.026 0.226** 0.089 — 5. Boston Naming Test −0.618** −0.379** −0.480** 0.007 — Measure 1 2 3 4 5 1. Short Blessed Test — 2. Trails A (seconds) 0.520** — 3. Trails B (seconds) 0.607** 0.789** — 4. Cancellation Test 0.026 0.226** 0.089 — 5. Boston Naming Test −0.618** −0.379** −0.480** 0.007 — *p < .05; **p < .01; ***p < .001. Results With respect to the overall hierarchical regressions, adding physical/functional variables to the models while controlling for demographics and stroke-related factors explained more variance in ADL/IADLs (ΔR2 = 4.2%; F(4, 428) = 5.57; p < .001) and Participation (ΔR2 = 2.3%; F(4, 424) = 2.93; p < .001) outcomes than in Communication (ΔR2 = 0.9%; F(4, 454) = 1.09; p = .32) and Memory and Thinking (ΔR2 = 0.6%; F(4, 451) = 0.88; p = .61.) outcomes. Conversely, adding cognitive variables in the final step of the models explained more variance in Communication (ΔR2 = 7.8%; F(5, 449) = 7.93; p < .001) and Memory and Thinking (ΔR2 = 6.9%; F(5, 446) = 6.83; p < .001) than in ADL/IADLs (ΔR2 = 0.9%; F(5, 423) = 1.01; p = .45) and Participation (ΔR2 = 1.7%; F(5, 419) = 1.76; p = .15) subscales. The overall models were all statistically significant (p’s < .001) and accounted for small amounts of variance in SIS cognitive subscales (Communication, 10.7%–11.3%; Memory and Thinking, 8.4%–10.6%) and moderate degrees of variance in functional subscales (ADL/IADLs, 22.0%–24.6%; Participation, 17.7%–18.9%). Significant individual-level predictors of the four outcome variables are presented in Tables 4–7. For the Communication and Memory and Thinking models, only the SBT was significant (p’s < .001). As expected, greater impairment on the SBT predicted poorer self-reported SIS cognition. With regard to the ADL/IADLs model, gender (p = .016), race (p = .021), and NIHSS Total score (p < .001) were significant predictors. Similarly, in the Participation model, gender (p = .069) and race (p = .066) approached significance, while the NIHSS Total Score (p = .002) remained significant. For both functional SIS subscales, the performance of male and Caucasian participants was higher than female and non-Caucasian participants. In terms of NIHSS Total scores, poorer performance predicted worse self-reported functioning in ADL/IADLs and Participation. Pearson’s correlation coefficients for the relationships between the SBT, NIHSS Total score, and SIS cognitive and functional outcomes are presented in Table 8. Overall, all relationships were in the expected direction (i.e., greater symptom severity on one measure correlated with greater symptom severity on the other measure). Descriptive statistics for SIS subscales and physical/functional and cognitive variables are presented in Table 9. Table 4. Hierarchical multiple regression coefficients with memory and thinking as the outcome variable Predictors B SE B β Step 1  Constant 30.36 3.45  Age −0.01 0.02 −.023  Sex −0.54 0.63 −.036  Race −0.87 0.70 −.063  NIHSS −0.06 0.08 −.035  Education 0.20 0.16 .084  Marital status −0.37 0.68 −.027  Stroke diagnosis −0.95 0.89 −.052 Step 2  Constant 28.41 4.63  Age −0.01 0.02 −.016  Sex −0.50 0.65 −.031  Race −0.93 0.70 −.067  NIHSS −0.01 0.11 −.007  Education 0.21 0.17 .086  Marital status −0.23 0.70 −.018  Stroke diagnosis −0.80 0.96 −.043  Berg Balance Scale −0.01 0.04 −.040  Supine to Sit 0.06 0.43 .013  Sit to Stand 0.16 0.53 .038  Ambulation (ft) 0.00 0.00 .076 Step 3  Constant 32.75 4.84  Age −0.00 0.02 −.001  Sex −0.53 0.63 −.040  Race −0.78 0.71 −.057  NIHSS 0.03 0.10 .019  Education 0.13 0.16 .079  Marital status 0.05 0.70 .004  Stroke diagnosis −0.35 0.96 −.019  Berg Balance Scale −0.02 0.04 −.042  Supine to Sit −0.14 0.41 −.031  Sit to Stand 0.15 0.51 .036  Ambulation (ft) 0.00 0.00 .068  Short Blessed Test −0.36*** 0.09 −.325  Trails A −0.00 0.01 −.041  Trails B 0.00 0.01 .132  Cancellation −0.03 0.11 −.075  Boston Naming Test −0.16 0.15 −.075 Predictors B SE B β Step 1  Constant 30.36 3.45  Age −0.01 0.02 −.023  Sex −0.54 0.63 −.036  Race −0.87 0.70 −.063  NIHSS −0.06 0.08 −.035  Education 0.20 0.16 .084  Marital status −0.37 0.68 −.027  Stroke diagnosis −0.95 0.89 −.052 Step 2  Constant 28.41 4.63  Age −0.01 0.02 −.016  Sex −0.50 0.65 −.031  Race −0.93 0.70 −.067  NIHSS −0.01 0.11 −.007  Education 0.21 0.17 .086  Marital status −0.23 0.70 −.018  Stroke diagnosis −0.80 0.96 −.043  Berg Balance Scale −0.01 0.04 −.040  Supine to Sit 0.06 0.43 .013  Sit to Stand 0.16 0.53 .038  Ambulation (ft) 0.00 0.00 .076 Step 3  Constant 32.75 4.84  Age −0.00 0.02 −.001  Sex −0.53 0.63 −.040  Race −0.78 0.71 −.057  NIHSS 0.03 0.10 .019  Education 0.13 0.16 .079  Marital status 0.05 0.70 .004  Stroke diagnosis −0.35 0.96 −.019  Berg Balance Scale −0.02 0.04 −.042  Supine to Sit −0.14 0.41 −.031  Sit to Stand 0.15 0.51 .036  Ambulation (ft) 0.00 0.00 .068  Short Blessed Test −0.36*** 0.09 −.325  Trails A −0.00 0.01 −.041  Trails B 0.00 0.01 .132  Cancellation −0.03 0.11 −.075  Boston Naming Test −0.16 0.15 −.075 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 0.1%–3.2% for Step 1 (p’s = .04–.37), R2 change = 0.4%–0.8% for Step 2 (p’s = .07–.71), R2 change = 5.7%–7.9% for Step 3 (p’s < .01). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 4. Hierarchical multiple regression coefficients with memory and thinking as the outcome variable Predictors B SE B β Step 1  Constant 30.36 3.45  Age −0.01 0.02 −.023  Sex −0.54 0.63 −.036  Race −0.87 0.70 −.063  NIHSS −0.06 0.08 −.035  Education 0.20 0.16 .084  Marital status −0.37 0.68 −.027  Stroke diagnosis −0.95 0.89 −.052 Step 2  Constant 28.41 4.63  Age −0.01 0.02 −.016  Sex −0.50 0.65 −.031  Race −0.93 0.70 −.067  NIHSS −0.01 0.11 −.007  Education 0.21 0.17 .086  Marital status −0.23 0.70 −.018  Stroke diagnosis −0.80 0.96 −.043  Berg Balance Scale −0.01 0.04 −.040  Supine to Sit 0.06 0.43 .013  Sit to Stand 0.16 0.53 .038  Ambulation (ft) 0.00 0.00 .076 Step 3  Constant 32.75 4.84  Age −0.00 0.02 −.001  Sex −0.53 0.63 −.040  Race −0.78 0.71 −.057  NIHSS 0.03 0.10 .019  Education 0.13 0.16 .079  Marital status 0.05 0.70 .004  Stroke diagnosis −0.35 0.96 −.019  Berg Balance Scale −0.02 0.04 −.042  Supine to Sit −0.14 0.41 −.031  Sit to Stand 0.15 0.51 .036  Ambulation (ft) 0.00 0.00 .068  Short Blessed Test −0.36*** 0.09 −.325  Trails A −0.00 0.01 −.041  Trails B 0.00 0.01 .132  Cancellation −0.03 0.11 −.075  Boston Naming Test −0.16 0.15 −.075 Predictors B SE B β Step 1  Constant 30.36 3.45  Age −0.01 0.02 −.023  Sex −0.54 0.63 −.036  Race −0.87 0.70 −.063  NIHSS −0.06 0.08 −.035  Education 0.20 0.16 .084  Marital status −0.37 0.68 −.027  Stroke diagnosis −0.95 0.89 −.052 Step 2  Constant 28.41 4.63  Age −0.01 0.02 −.016  Sex −0.50 0.65 −.031  Race −0.93 0.70 −.067  NIHSS −0.01 0.11 −.007  Education 0.21 0.17 .086  Marital status −0.23 0.70 −.018  Stroke diagnosis −0.80 0.96 −.043  Berg Balance Scale −0.01 0.04 −.040  Supine to Sit 0.06 0.43 .013  Sit to Stand 0.16 0.53 .038  Ambulation (ft) 0.00 0.00 .076 Step 3  Constant 32.75 4.84  Age −0.00 0.02 −.001  Sex −0.53 0.63 −.040  Race −0.78 0.71 −.057  NIHSS 0.03 0.10 .019  Education 0.13 0.16 .079  Marital status 0.05 0.70 .004  Stroke diagnosis −0.35 0.96 −.019  Berg Balance Scale −0.02 0.04 −.042  Supine to Sit −0.14 0.41 −.031  Sit to Stand 0.15 0.51 .036  Ambulation (ft) 0.00 0.00 .068  Short Blessed Test −0.36*** 0.09 −.325  Trails A −0.00 0.01 −.041  Trails B 0.00 0.01 .132  Cancellation −0.03 0.11 −.075  Boston Naming Test −0.16 0.15 −.075 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 0.1%–3.2% for Step 1 (p’s = .04–.37), R2 change = 0.4%–0.8% for Step 2 (p’s = .07–.71), R2 change = 5.7%–7.9% for Step 3 (p’s < .01). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 5. Hierarchical multiple regression coefficients with communication as the outcome variable Predictors B SE B β Step 1  Constant 33.47 2.99  Age −0.012 0.02 −.046  Sex −0.04 0.52 −.004  Race −0.71 0.57 −.063  NIHSS −0.13* 0.07 −.092  Education 0.09 0.16 .043  Marital status −0.40 0.56 −.037  Stroke diagnosis −0.62 0.78 −.041 Step 2  Constant 31.45 3.92  Age −0.02 0.02 −.040  Sex 0.02 0.53 .002  Race −0.77 0.57 −.068  NIHSS −0.11 0.09 −.073  Education 0.09 0.16 .045  Marital status −0.36 0.57 −.033  Stroke diagnosis −0.55 0.84 −.037  Berg Balance Scale −0.04 0.03 −.132  Supine to Sit 0.32 0.34 .124  Sit to Stand 0.11 0.44 .031  Ambulation (ft) 0.00 0.00 .087 Step 3  Constant 34.07 4.23  Age −0.00 0.02 −.012  Sex −0.07 0.51 −.006  Race −0.54 0.57 −.047  NIHSS −0.06 0.09 −.044  Education 0.02 0.16 .009  Marital status −0.10 0.58 −.009  Stroke diagnosis −0.230 0.83 −.015  Berg Balance Scale −0.04 0.03 −.155  Supine to Sit 0.11 0.32 .081  Sit to Stand 0.17 0.41 .049  Ambulation(ft) 0.00 0.00 .079  Short Blessed Test −0.29*** 0.06 −.318  Trails A 0.00 0.01 .014  Trails B −0.00 0.01 −.010  Cancellation −0.03 0.10 −.020  Boston Naming Test −0.06 0.11 −.035 Predictors B SE B β Step 1  Constant 33.47 2.99  Age −0.012 0.02 −.046  Sex −0.04 0.52 −.004  Race −0.71 0.57 −.063  NIHSS −0.13* 0.07 −.092  Education 0.09 0.16 .043  Marital status −0.40 0.56 −.037  Stroke diagnosis −0.62 0.78 −.041 Step 2  Constant 31.45 3.92  Age −0.02 0.02 −.040  Sex 0.02 0.53 .002  Race −0.77 0.57 −.068  NIHSS −0.11 0.09 −.073  Education 0.09 0.16 .045  Marital status −0.36 0.57 −.033  Stroke diagnosis −0.55 0.84 −.037  Berg Balance Scale −0.04 0.03 −.132  Supine to Sit 0.32 0.34 .124  Sit to Stand 0.11 0.44 .031  Ambulation (ft) 0.00 0.00 .087 Step 3  Constant 34.07 4.23  Age −0.00 0.02 −.012  Sex −0.07 0.51 −.006  Race −0.54 0.57 −.047  NIHSS −0.06 0.09 −.044  Education 0.02 0.16 .009  Marital status −0.10 0.58 −.009  Stroke diagnosis −0.230 0.83 −.015  Berg Balance Scale −0.04 0.03 −.155  Supine to Sit 0.11 0.32 .081  Sit to Stand 0.17 0.41 .049  Ambulation(ft) 0.00 0.00 .079  Short Blessed Test −0.29*** 0.06 −.318  Trails A 0.00 0.01 .014  Trails B −0.00 0.01 −.010  Cancellation −0.03 0.10 −.020  Boston Naming Test −0.06 0.11 −.035 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 1.8%–3.2% for Step 1 (p’s = .04–.28), R2 change = 0.7%–1.1% for Step 2 (p’s = .06–.38), R2 change = 6.6%–8.7% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 5. Hierarchical multiple regression coefficients with communication as the outcome variable Predictors B SE B β Step 1  Constant 33.47 2.99  Age −0.012 0.02 −.046  Sex −0.04 0.52 −.004  Race −0.71 0.57 −.063  NIHSS −0.13* 0.07 −.092  Education 0.09 0.16 .043  Marital status −0.40 0.56 −.037  Stroke diagnosis −0.62 0.78 −.041 Step 2  Constant 31.45 3.92  Age −0.02 0.02 −.040  Sex 0.02 0.53 .002  Race −0.77 0.57 −.068  NIHSS −0.11 0.09 −.073  Education 0.09 0.16 .045  Marital status −0.36 0.57 −.033  Stroke diagnosis −0.55 0.84 −.037  Berg Balance Scale −0.04 0.03 −.132  Supine to Sit 0.32 0.34 .124  Sit to Stand 0.11 0.44 .031  Ambulation (ft) 0.00 0.00 .087 Step 3  Constant 34.07 4.23  Age −0.00 0.02 −.012  Sex −0.07 0.51 −.006  Race −0.54 0.57 −.047  NIHSS −0.06 0.09 −.044  Education 0.02 0.16 .009  Marital status −0.10 0.58 −.009  Stroke diagnosis −0.230 0.83 −.015  Berg Balance Scale −0.04 0.03 −.155  Supine to Sit 0.11 0.32 .081  Sit to Stand 0.17 0.41 .049  Ambulation(ft) 0.00 0.00 .079  Short Blessed Test −0.29*** 0.06 −.318  Trails A 0.00 0.01 .014  Trails B −0.00 0.01 −.010  Cancellation −0.03 0.10 −.020  Boston Naming Test −0.06 0.11 −.035 Predictors B SE B β Step 1  Constant 33.47 2.99  Age −0.012 0.02 −.046  Sex −0.04 0.52 −.004  Race −0.71 0.57 −.063  NIHSS −0.13* 0.07 −.092  Education 0.09 0.16 .043  Marital status −0.40 0.56 −.037  Stroke diagnosis −0.62 0.78 −.041 Step 2  Constant 31.45 3.92  Age −0.02 0.02 −.040  Sex 0.02 0.53 .002  Race −0.77 0.57 −.068  NIHSS −0.11 0.09 −.073  Education 0.09 0.16 .045  Marital status −0.36 0.57 −.033  Stroke diagnosis −0.55 0.84 −.037  Berg Balance Scale −0.04 0.03 −.132  Supine to Sit 0.32 0.34 .124  Sit to Stand 0.11 0.44 .031  Ambulation (ft) 0.00 0.00 .087 Step 3  Constant 34.07 4.23  Age −0.00 0.02 −.012  Sex −0.07 0.51 −.006  Race −0.54 0.57 −.047  NIHSS −0.06 0.09 −.044  Education 0.02 0.16 .009  Marital status −0.10 0.58 −.009  Stroke diagnosis −0.230 0.83 −.015  Berg Balance Scale −0.04 0.03 −.155  Supine to Sit 0.11 0.32 .081  Sit to Stand 0.17 0.41 .049  Ambulation(ft) 0.00 0.00 .079  Short Blessed Test −0.29*** 0.06 −.318  Trails A 0.00 0.01 .014  Trails B −0.00 0.01 −.010  Cancellation −0.03 0.10 −.020  Boston Naming Test −0.06 0.11 −.035 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 1.8%–3.2% for Step 1 (p’s = .04–.28), R2 change = 0.7%–1.1% for Step 2 (p’s = .06–.38), R2 change = 6.6%–8.7% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 6. Hierarchical multiple regression coefficients with activities of daily living/instrumental activities of daily living as the outcome variable Predictors B SE B β Step 1  Constant 55.61 4.50  Age −0.08** 0.03 −.119  Sex −2.43** 0.84 −.130  Race −1.97* 0.92 −.101  NIHSS −0.85*** 0.11 −.352  Education 0.18 0.23 .054  Marital status −0.69 0.91 −.037  Stroke diagnosis −1.17 1.32 −.045 Step 2  Constant 45.66 6.10  Age −0.06 0.03 −.089  Sex −2.06* 0.85 −.110  Race −2.24* 0.91 −.115  NIHSS −0.52*** 0.14 −.216  Education 0.21 0.22 .062  Marital status −0.13 0.92 −.006  Stroke diagnosis 0.11 1.29 .004  Berg Balance Scale 0.06 0.05 .134  Supine to Sit 0.98 0.66 .159  Sit to Stand −0.58 0.82 −.095  Ambulation (ft) 0.00 0.00 .094 Step 3  Constant 49.04 7.47  Age −0.05 0.03 −.084  Sex −2.08* 0.86 −.111  Race −2.17* 0.94 −.111  NIHSS −0.52*** 0.14 −.210  Education 0.15 0.26 .044  Marital status −0.10 0.96 −.003  Stroke diagnosis 0.30 1.45 .011  Berg Balance Scale 0.06 0.05 .122  Supine to Sit 0.88 0.68 .142  Sit to Stand −0.58 0.80 −.095  Ambulation (ft) 0.00 0.00 .087  Short Blessed Test −0.11 0.12 −.068  Trails A −0.01 0.02 −.066  Trails B 0.00 0.01 .025  Cancellation −0.05 0.18 −.018  Boston Naming Test −0.10 0.18 −.034 Predictors B SE B β Step 1  Constant 55.61 4.50  Age −0.08** 0.03 −.119  Sex −2.43** 0.84 −.130  Race −1.97* 0.92 −.101  NIHSS −0.85*** 0.11 −.352  Education 0.18 0.23 .054  Marital status −0.69 0.91 −.037  Stroke diagnosis −1.17 1.32 −.045 Step 2  Constant 45.66 6.10  Age −0.06 0.03 −.089  Sex −2.06* 0.85 −.110  Race −2.24* 0.91 −.115  NIHSS −0.52*** 0.14 −.216  Education 0.21 0.22 .062  Marital status −0.13 0.92 −.006  Stroke diagnosis 0.11 1.29 .004  Berg Balance Scale 0.06 0.05 .134  Supine to Sit 0.98 0.66 .159  Sit to Stand −0.58 0.82 −.095  Ambulation (ft) 0.00 0.00 .094 Step 3  Constant 49.04 7.47  Age −0.05 0.03 −.084  Sex −2.08* 0.86 −.111  Race −2.17* 0.94 −.111  NIHSS −0.52*** 0.14 −.210  Education 0.15 0.26 .044  Marital status −0.10 0.96 −.003  Stroke diagnosis 0.30 1.45 .011  Berg Balance Scale 0.06 0.05 .122  Supine to Sit 0.88 0.68 .142  Sit to Stand −0.58 0.80 −.095  Ambulation (ft) 0.00 0.00 .087  Short Blessed Test −0.11 0.12 −.068  Trails A −0.01 0.02 −.066  Trails B 0.00 0.01 .025  Cancellation −0.05 0.18 −.018  Boston Naming Test −0.10 0.18 −.034 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 17.8%–19.1% for Step 1 (p’s < .001), R2 change = 20.7%–5.5% for Step 2 (p’s < .001), R2 change = 0.6%–1.4% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 6. Hierarchical multiple regression coefficients with activities of daily living/instrumental activities of daily living as the outcome variable Predictors B SE B β Step 1  Constant 55.61 4.50  Age −0.08** 0.03 −.119  Sex −2.43** 0.84 −.130  Race −1.97* 0.92 −.101  NIHSS −0.85*** 0.11 −.352  Education 0.18 0.23 .054  Marital status −0.69 0.91 −.037  Stroke diagnosis −1.17 1.32 −.045 Step 2  Constant 45.66 6.10  Age −0.06 0.03 −.089  Sex −2.06* 0.85 −.110  Race −2.24* 0.91 −.115  NIHSS −0.52*** 0.14 −.216  Education 0.21 0.22 .062  Marital status −0.13 0.92 −.006  Stroke diagnosis 0.11 1.29 .004  Berg Balance Scale 0.06 0.05 .134  Supine to Sit 0.98 0.66 .159  Sit to Stand −0.58 0.82 −.095  Ambulation (ft) 0.00 0.00 .094 Step 3  Constant 49.04 7.47  Age −0.05 0.03 −.084  Sex −2.08* 0.86 −.111  Race −2.17* 0.94 −.111  NIHSS −0.52*** 0.14 −.210  Education 0.15 0.26 .044  Marital status −0.10 0.96 −.003  Stroke diagnosis 0.30 1.45 .011  Berg Balance Scale 0.06 0.05 .122  Supine to Sit 0.88 0.68 .142  Sit to Stand −0.58 0.80 −.095  Ambulation (ft) 0.00 0.00 .087  Short Blessed Test −0.11 0.12 −.068  Trails A −0.01 0.02 −.066  Trails B 0.00 0.01 .025  Cancellation −0.05 0.18 −.018  Boston Naming Test −0.10 0.18 −.034 Predictors B SE B β Step 1  Constant 55.61 4.50  Age −0.08** 0.03 −.119  Sex −2.43** 0.84 −.130  Race −1.97* 0.92 −.101  NIHSS −0.85*** 0.11 −.352  Education 0.18 0.23 .054  Marital status −0.69 0.91 −.037  Stroke diagnosis −1.17 1.32 −.045 Step 2  Constant 45.66 6.10  Age −0.06 0.03 −.089  Sex −2.06* 0.85 −.110  Race −2.24* 0.91 −.115  NIHSS −0.52*** 0.14 −.216  Education 0.21 0.22 .062  Marital status −0.13 0.92 −.006  Stroke diagnosis 0.11 1.29 .004  Berg Balance Scale 0.06 0.05 .134  Supine to Sit 0.98 0.66 .159  Sit to Stand −0.58 0.82 −.095  Ambulation (ft) 0.00 0.00 .094 Step 3  Constant 49.04 7.47  Age −0.05 0.03 −.084  Sex −2.08* 0.86 −.111  Race −2.17* 0.94 −.111  NIHSS −0.52*** 0.14 −.210  Education 0.15 0.26 .044  Marital status −0.10 0.96 −.003  Stroke diagnosis 0.30 1.45 .011  Berg Balance Scale 0.06 0.05 .122  Supine to Sit 0.88 0.68 .142  Sit to Stand −0.58 0.80 −.095  Ambulation (ft) 0.00 0.00 .087  Short Blessed Test −0.11 0.12 −.068  Trails A −0.01 0.02 −.066  Trails B 0.00 0.01 .025  Cancellation −0.05 0.18 −.018  Boston Naming Test −0.10 0.18 −.034 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 17.8%–19.1% for Step 1 (p’s < .001), R2 change = 20.7%–5.5% for Step 2 (p’s < .001), R2 change = 0.6%–1.4% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 7. Hierarchical multiple regression coefficients with participation as the outcome variable Predictors B SE B β Step 1  Constant 40.86 4.73  Age 0.019 0.03 .027  Sex −1.923* 0.89 −.100  Race −1.70 0.98 −.085  NIHSS −0.74*** 0.13 −.293  Education 0.04 0.23 .011  Marital status −1.96* 0.96 −.102  Stroke diagnosis −1.99 1.38 −.075 Step 2  Constant 34.30 6.53  Age 0.03 0.03 .049  Sex −1.63* 0.90 −.085  Race −1.93** 0.97 −.096  NIHSS −0.48 0.15 −.190  Education 0.06 0.24 .018  Marital status −1.53 0.99 −.079  Stroke diagnosis −0.93 1.36 −.035  Berg Balance Scale 0.10 0.06 .207  Supine to SIt 0.50 0.58 .078  Sit to Stand −0.59 0.78 −.096  Ambulation (ft) 0.00 0.00 −.003 Step 3  Constant 39.73 7.73  Age 0.03 0.03 .050  Sex −1.66 0.91 −.086  Race −1.86 1.01 −.093  NIHSS −0.47** 0.15 −.185  Education −0.002 0.26 −.002  Marital status −1.48 1.04 −.077  Stroke diagnosis −0.64 1.48 −.024  Berg Balance Scale 0.09 0.06 .198  Supine to Sit 0.36 0.58 .057  Sit to Stand −0.64 0.75 −.104  Ambulation (ft) 0.00 0.00 −.010  Short Blessed Test −0.19 0.13 −.192  Trails A −0.02 0.03 −.018  Trails B 0.01 0.01 .005  Cancellation −0.03 0.22 −.032  Boston Naming Test −0.20 0.19 −.200 Predictors B SE B β Step 1  Constant 40.86 4.73  Age 0.019 0.03 .027  Sex −1.923* 0.89 −.100  Race −1.70 0.98 −.085  NIHSS −0.74*** 0.13 −.293  Education 0.04 0.23 .011  Marital status −1.96* 0.96 −.102  Stroke diagnosis −1.99 1.38 −.075 Step 2  Constant 34.30 6.53  Age 0.03 0.03 .049  Sex −1.63* 0.90 −.085  Race −1.93** 0.97 −.096  NIHSS −0.48 0.15 −.190  Education 0.06 0.24 .018  Marital status −1.53 0.99 −.079  Stroke diagnosis −0.93 1.36 −.035  Berg Balance Scale 0.10 0.06 .207  Supine to SIt 0.50 0.58 .078  Sit to Stand −0.59 0.78 −.096  Ambulation (ft) 0.00 0.00 −.003 Step 3  Constant 39.73 7.73  Age 0.03 0.03 .050  Sex −1.66 0.91 −.086  Race −1.86 1.01 −.093  NIHSS −0.47** 0.15 −.185  Education −0.002 0.26 −.002  Marital status −1.48 1.04 −.077  Stroke diagnosis −0.64 1.48 −.024  Berg Balance Scale 0.09 0.06 .198  Supine to Sit 0.36 0.58 .057  Sit to Stand −0.64 0.75 −.104  Ambulation (ft) 0.00 0.00 −.010  Short Blessed Test −0.19 0.13 −.192  Trails A −0.02 0.03 −.018  Trails B 0.01 0.01 .005  Cancellation −0.03 0.22 −.032  Boston Naming Test −0.20 0.19 −.200 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 13.7%–14.6% for Step 1 (p’s < .001), R2 change = 1.8%–3.3% for Step 2 (p’s < .001), R2 change = 1.2%–2.4% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 7. Hierarchical multiple regression coefficients with participation as the outcome variable Predictors B SE B β Step 1  Constant 40.86 4.73  Age 0.019 0.03 .027  Sex −1.923* 0.89 −.100  Race −1.70 0.98 −.085  NIHSS −0.74*** 0.13 −.293  Education 0.04 0.23 .011  Marital status −1.96* 0.96 −.102  Stroke diagnosis −1.99 1.38 −.075 Step 2  Constant 34.30 6.53  Age 0.03 0.03 .049  Sex −1.63* 0.90 −.085  Race −1.93** 0.97 −.096  NIHSS −0.48 0.15 −.190  Education 0.06 0.24 .018  Marital status −1.53 0.99 −.079  Stroke diagnosis −0.93 1.36 −.035  Berg Balance Scale 0.10 0.06 .207  Supine to SIt 0.50 0.58 .078  Sit to Stand −0.59 0.78 −.096  Ambulation (ft) 0.00 0.00 −.003 Step 3  Constant 39.73 7.73  Age 0.03 0.03 .050  Sex −1.66 0.91 −.086  Race −1.86 1.01 −.093  NIHSS −0.47** 0.15 −.185  Education −0.002 0.26 −.002  Marital status −1.48 1.04 −.077  Stroke diagnosis −0.64 1.48 −.024  Berg Balance Scale 0.09 0.06 .198  Supine to Sit 0.36 0.58 .057  Sit to Stand −0.64 0.75 −.104  Ambulation (ft) 0.00 0.00 −.010  Short Blessed Test −0.19 0.13 −.192  Trails A −0.02 0.03 −.018  Trails B 0.01 0.01 .005  Cancellation −0.03 0.22 −.032  Boston Naming Test −0.20 0.19 −.200 Predictors B SE B β Step 1  Constant 40.86 4.73  Age 0.019 0.03 .027  Sex −1.923* 0.89 −.100  Race −1.70 0.98 −.085  NIHSS −0.74*** 0.13 −.293  Education 0.04 0.23 .011  Marital status −1.96* 0.96 −.102  Stroke diagnosis −1.99 1.38 −.075 Step 2  Constant 34.30 6.53  Age 0.03 0.03 .049  Sex −1.63* 0.90 −.085  Race −1.93** 0.97 −.096  NIHSS −0.48 0.15 −.190  Education 0.06 0.24 .018  Marital status −1.53 0.99 −.079  Stroke diagnosis −0.93 1.36 −.035  Berg Balance Scale 0.10 0.06 .207  Supine to SIt 0.50 0.58 .078  Sit to Stand −0.59 0.78 −.096  Ambulation (ft) 0.00 0.00 −.003 Step 3  Constant 39.73 7.73  Age 0.03 0.03 .050  Sex −1.66 0.91 −.086  Race −1.86 1.01 −.093  NIHSS −0.47** 0.15 −.185  Education −0.002 0.26 −.002  Marital status −1.48 1.04 −.077  Stroke diagnosis −0.64 1.48 −.024  Berg Balance Scale 0.09 0.06 .198  Supine to Sit 0.36 0.58 .057  Sit to Stand −0.64 0.75 −.104  Ambulation (ft) 0.00 0.00 −.010  Short Blessed Test −0.19 0.13 −.192  Trails A −0.02 0.03 −.018  Trails B 0.01 0.01 .005  Cancellation −0.03 0.22 −.032  Boston Naming Test −0.20 0.19 −.200 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 13.7%–14.6% for Step 1 (p’s < .001), R2 change = 1.8%–3.3% for Step 2 (p’s < .001), R2 change = 1.2%–2.4% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 8. Correlations among NIHSS, Short Blessed Test, and Stroke Impact Scale subscales Measure 1 2 3 4 5 6 1. NIHSS — 2. Short Blessed Test 0.23*** — 3. Memory and Thinking −0.05 −0.28*** — 4. Communication −0.11** −0.30*** 0.82*** — 5. ADL/IADLs −0.37*** −0.21*** 0.59*** 0.59*** — 6. Participation −0.32*** −0.20*** 0.57*** 0.57*** 0.79*** — Measure 1 2 3 4 5 6 1. NIHSS — 2. Short Blessed Test 0.23*** — 3. Memory and Thinking −0.05 −0.28*** — 4. Communication −0.11** −0.30*** 0.82*** — 5. ADL/IADLs −0.37*** −0.21*** 0.59*** 0.59*** — 6. Participation −0.32*** −0.20*** 0.57*** 0.57*** 0.79*** — Note: NIHSS = National Institutes of Health Stroke Scale; ADL/IADLs = activities of daily living/instrumental activities of daily living. *p < .05; **p < .01; ***p < .001. Table 8. Correlations among NIHSS, Short Blessed Test, and Stroke Impact Scale subscales Measure 1 2 3 4 5 6 1. NIHSS — 2. Short Blessed Test 0.23*** — 3. Memory and Thinking −0.05 −0.28*** — 4. Communication −0.11** −0.30*** 0.82*** — 5. ADL/IADLs −0.37*** −0.21*** 0.59*** 0.59*** — 6. Participation −0.32*** −0.20*** 0.57*** 0.57*** 0.79*** — Measure 1 2 3 4 5 6 1. NIHSS — 2. Short Blessed Test 0.23*** — 3. Memory and Thinking −0.05 −0.28*** — 4. Communication −0.11** −0.30*** 0.82*** — 5. ADL/IADLs −0.37*** −0.21*** 0.59*** 0.59*** — 6. Participation −0.32*** −0.20*** 0.57*** 0.57*** 0.79*** — Note: NIHSS = National Institutes of Health Stroke Scale; ADL/IADLs = activities of daily living/instrumental activities of daily living. *p < .05; **p < .01; ***p < .001. Table 9. Descriptives for Stroke Impact Scale subscales and physical/functional and cognitive variables Measure M SD Range Stroke Impact Scale subscales  Memory and Thinking 28.42 6.63 7–35  Communication 30.68 5.47 9–35  ADL/IADLs 41.26 9.37 10–50  Participation 29.41 9.63 8–40 Physical/functional variables  Berg Balance Scale 28.21 20.91 0–56  Supine to Sit 5.32 1.51 1–7  Sit to Stand 5.03 1.53 0–7  Ambulation (feet) 190.76 238.42 0–3000 Cognitive variables  Short Blessed Test 5.92 5.87 0–28  Trails A (seconds) 71.27 47.13 16–280  Trails B (seconds) 157.79 83.74 12–416  Cancellation 0.15 3.59 −13–13  Boston Naming Test 11.68 3.20 0–15 Measure M SD Range Stroke Impact Scale subscales  Memory and Thinking 28.42 6.63 7–35  Communication 30.68 5.47 9–35  ADL/IADLs 41.26 9.37 10–50  Participation 29.41 9.63 8–40 Physical/functional variables  Berg Balance Scale 28.21 20.91 0–56  Supine to Sit 5.32 1.51 1–7  Sit to Stand 5.03 1.53 0–7  Ambulation (feet) 190.76 238.42 0–3000 Cognitive variables  Short Blessed Test 5.92 5.87 0–28  Trails A (seconds) 71.27 47.13 16–280  Trails B (seconds) 157.79 83.74 12–416  Cancellation 0.15 3.59 −13–13  Boston Naming Test 11.68 3.20 0–15 Note: ADL/IADLs = activities of daily living/ instrumental activities of daily living. Table 9. Descriptives for Stroke Impact Scale subscales and physical/functional and cognitive variables Measure M SD Range Stroke Impact Scale subscales  Memory and Thinking 28.42 6.63 7–35  Communication 30.68 5.47 9–35  ADL/IADLs 41.26 9.37 10–50  Participation 29.41 9.63 8–40 Physical/functional variables  Berg Balance Scale 28.21 20.91 0–56  Supine to Sit 5.32 1.51 1–7  Sit to Stand 5.03 1.53 0–7  Ambulation (feet) 190.76 238.42 0–3000 Cognitive variables  Short Blessed Test 5.92 5.87 0–28  Trails A (seconds) 71.27 47.13 16–280  Trails B (seconds) 157.79 83.74 12–416  Cancellation 0.15 3.59 −13–13  Boston Naming Test 11.68 3.20 0–15 Measure M SD Range Stroke Impact Scale subscales  Memory and Thinking 28.42 6.63 7–35  Communication 30.68 5.47 9–35  ADL/IADLs 41.26 9.37 10–50  Participation 29.41 9.63 8–40 Physical/functional variables  Berg Balance Scale 28.21 20.91 0–56  Supine to Sit 5.32 1.51 1–7  Sit to Stand 5.03 1.53 0–7  Ambulation (feet) 190.76 238.42 0–3000 Cognitive variables  Short Blessed Test 5.92 5.87 0–28  Trails A (seconds) 71.27 47.13 16–280  Trails B (seconds) 157.79 83.74 12–416  Cancellation 0.15 3.59 −13–13  Boston Naming Test 11.68 3.20 0–15 Note: ADL/IADLs = activities of daily living/ instrumental activities of daily living. Discussion The current study examined acute predictors of cognitive and functional outcomes 6 months after mild to moderate stroke. Overall, the combination of demographic factors and cognitive, physical, and functional status at stroke onset accounted for small amounts of variance in cognitive outcomes and moderate degrees of variance in functional outcomes post-stroke. Furthermore, inclusion of physical/functional variables in the second step of the regression models resulted in better prediction of ADL/IADLs and Participation, but not Communication or Memory and Thinking outcomes. Conversely, adding cognitive variables in the final step of the models improved prediction of Memory and Thinking and Communication outcomes, but not ADL/IADLs or Participation. Finally, in terms of individual predictors, brief screening instruments (i.e., the NIHSS and SBT) exhibited consistent predictive utility, while more domain-specific cognitive tests (e.g., the BNT and Trails A and B) did not. Specifically, the SBT emerged as a significant predictor of the Communication and Memory and Thinking subscales, such that greater impairment on the SBT was associated with worse self-reported communication and cognitive ability 6 months after stroke. Moreover, although none of the cognitive measures included in the present study predicted ADL/IADLs or Participation, the NIHSS Total score was a significant predictor of subsequent functional outcomes. Results of the regression analyses revealed that greater stroke severity was associated with worse self-reported ADL/IADLs and reduced engagement in meaningful activities 6 months after stroke. Consistent with previous studies, being female and/or a racial minority was also associated with decreased independence in ADL/IADLs and participation post-stroke (Cioncoloni et al., 2013; Duarte et al., 2010; Horner, Swanson, Bosworth, & Matchar, 2003). The unfortunate reality is that inequalities in access to care and rehabilitation services exist across gender and racial groups, likely contributing to poorer outcomes for societally disadvantaged individuals (Busch, Coshall, Heuschmann, McKevitt, & Wolfe, 2009). Although not surprising, these findings support the importance of designing and implementing unbiased rehabilitation programs, in order to maximize recovery for individuals across a range of diversity characteristics. Overall, our results support the clinical utility of administering brief, broad screening instruments during acute recovery from mild to moderate stroke. The SBT and NIHSS significantly predicted 6-month cognitive and functional outcomes, and demonstrated superior predictive validity relative to measures assessing specific functional abilities (e.g., Sit to Stand) or cognitive domains (e.g., executive functioning). These findings are consistent with prior studies showing that, relative to more domain-specific cognitive measures, broad-based screening instruments tend to better predict post-stroke outcomes when administered acutely in real-world clinical scenarios (Horstmann, Rizos, Rauch, Arden, & Veltkamp, 2014; Riepe et al., 2004). Researchers have attributed this to the heterogeneity of acute stroke-related cognitive deficits, as well as confounding issues such as delirium, confusion, fatigue, and pre-stroke cognitive decline (Cumming et al., 2013; Lees et al., 2014; Salvadori et al., 2013). Given these complicating factors, brief screening instruments like the SBT and NIHSS that can be readily completed by patients shortly after stroke may be more appropriate for assessing patients in acute care settings than more domain-specific measures. In addition to providing evidence for the clinical utility of brief, broad screening instruments generally, our findings also suggest that measures assessing acute cognitive dysfunction and those examining general neurologic status may differentially predict 6-month cognitive and functional outcomes after stroke. The SBT, which is a screener for global cognitive dysfunction, predicted impairments in cognition and communication abilities, while the NIHSS, which is a broader assessment of neurologic status, predicted recovery of ADL/IADLs and reengagement in meaningful activity (Carpenter et al., 2011; Goldstein et al., 1989). This indicates that brief assessment of both acute neurologic status and global cognitive dysfunction is important for predicting outcomes 6 months after stroke. Specifically, conducting a brief cognitive screen may improve prediction of persistent cognitive deficits, whereas evaluating acute neurologic status may inform expectations for recovery of ADL/IADLs and reengagement in meaningful life activities. Of note, while the SBT has been previously shown to predict dementia diagnosis, neuropathology, and mortality (Bellelli et al., 2015; Marengoni et al., 2013; Katzman et al., 1983; Wilkins et al., 2007), to our knowledge this is the first study to demonstrate that the SBT is a valid predictor of cognitive outcomes following stroke. A probable explanation for the prognostic value of measures assessing acute neurologic status and global cognitive dysfunction is the persistence of certain stroke-related deficits (Cioncoloni et al., 2013; Del Sur et al., 2005; Hoffman et al., 2003; Wolf & Rognstad, 2013). However, even thorough predictive models typically account for no more than moderate amounts of variance in cognitive and functional outcomes due to the heterogeneity of recovery trajectories and outcomes for mild to moderate stroke (Jørgensen et al., 1995; Nichols-Larsen, Clark, Zeringue, Greenspan, & Blanton, 2005; Tilling et al., 2001). Age and other factors contribute to functional adaptation and cognitive plasticity such that in many cases, deficits resolve naturally, with compensation, and/or as a result of direct intervention (Ferrucci et al., 1993; Kleim & Jones, 2008; Miller et al., 2010; Veerbeek et al., 2014). There is strong evidence that rehabilitation services initiated at admission and sustained throughout recovery significantly reduce the likelihood of death and disability after stroke (Maulden, Gassaway, Horn, Smout, & DeJong, 2005; Miller et al., 2010). In particular, interdisciplinary interventions (i.e., physical therapy, occupational therapy, speech-language therapy, etc.) focused on providing intensive, highly repetitive, task-oriented, and task-specific trainings tailored to each phase of recovery have been well-supported (Kalra & Langhorne, 2007; Miller et al., 2010). There is also some evidence to support the effectiveness of cognitive rehabilitation for some cognitive impairments, though more research on the efficacy and effectiveness of cognitive intervention is needed (Gillespie et al., 2015). Acute assessment of cognitive and functional impairments is necessary to determine rehabilitation needs and plan appropriate services for patients recovering from stroke (Lawrence et al., 2001). Because early initiation and maintenance of targeted rehabilitation services is associated with better recovery, identifying acute factors that meaningfully predict long-term outcomes is an important goal for research and clinical care (Bhogal et al., 2003; Maulden et al., 2005). The ability to account for even small amounts of variance in stroke outcomes is of significant clinical utility, given that this information can be used to inform delivery of rehabilitation services. For example, memory problems are a common complaint following stroke. Identification of patients with acute cognitive dysfunction, combined with the knowledge that these early deficits predict long-term cognitive outcomes, can meaningfully inform treatment planning for survivors of mild to moderate stroke. In such cases, early cognitive skills training might aide in the recovery of memory functions or more frequently, enhance the individual’s ability to adapt to or compensate for their deficits. Therefore, improving predictive models for long-term cognitive and functional outcomes can help providers make informed decisions regarding appropriate rehabilitation goals and strategies, leading to better outcomes for survivors of mild to moderate stroke. Limitations and Future Directions Although of significant clinical utility, the findings from the current study should be interpreted in the context of several limitations. First, we did not have access to data pertaining to certain stroke-related characteristics (e.g., region of stroke, history of transient ischemic attacks, etc.) or relevant psychological factors (e.g., psychiatric history, acute depressive symptoms), thus limiting our ability to fully characterize our sample. While this is not ideal, previous investigations of the BRC (e.g., Aufman et al., 2013; Bland, Sturmoski, Whitson, Connor, & Fucetola, 2012; Merz, Van Patten, Mulhauser, & Fucetola, 2017; Van Patten et al., 2016) have faced similar constraints. Broadly, because our sample consisted of 498 consecutive admissions to a large acute care center over a five-year period (minus the exclusions specified in the Methods), we believe that it is a representative sample of mild to moderate stroke that is readily generalizable to patients evaluated in regular neuropsychological practice. Moreover, overall stroke severity, which was assessed in the current study, is likely the most important stroke-related characteristic in terms of predicting functional outcomes. For example, stroke severity was recently demonstrated to reliably predict return to work in stroke patients, whereas other acute biological factors such as location and type of stroke, as well as psychosocial variables such as educational attainment and marital status, did not (Wang, Kapellusch, & Garg, 2014). Additionally, although we were unable to account for affective factors (e.g., depression) that may have affected patients’ performance during acute inpatient testing, the prevalence of depression in our sample at 6-month follow-up is comparable to prior studies (Hackett et al., 2005). Consequently, our inability to report all relevant stroke-related and psychological characteristics of our sample does not render the findings ungeneralizable. Relatedly, although we were unable to incorporate details about rehabilitation participation in our predictive models, this study included a representative sample of stroke patients receiving standard clinical care. Importantly, as part of usual care, performance on acute assessment measures was used to inform recommendations for subsequent rehabilitation services. Therefore, our inability to account for rehabilitation-related factors does not negate our findings pertaining to the clinical utility of acute assessment measures. Second, as part of their clinical assessments, BRC patients complete only a small set of neuropsychological tasks which are administered by occupational therapists rather than neuropsychological assistants and do not capture all relevant cognitive domains. Consequently, we were unable to generate hypotheses with respect to certain aspects of cognitive functioning (e.g., visuospatial skills) and our measurement of other areas (e.g., executive functioning) was only cursory (i.e., Trails B). Involvement of neuropsychologists, particularly during inpatient rehabilitation or other post-acute care, would have allowed for better characterization of patients’ cognitive status post-stroke. Third, our follow-up assessment was restricted to a single time point approximately 6 months post-stroke and we were unable to conduct serial follow-up evaluations. Fourth, our sample was based exclusively out of the BRC database, which represents a geographically limited set of individuals who reside in the Midwestern region of the U.S. Generalizations outside of the greater population of individuals from this area are unwarranted. Finally, our outcome measure (the SIS) reflects an individual’s self-reported level of functioning and although we maximized the accuracy of these reports by excluding individuals with severe strokes, aphasia, and dysarthria, self-report instruments are inherently vulnerable to conscious and unconscious bias (Stone, Bachrach, Jobe, Kurtzman, & Cain, 1999). Future studies should further the current line of inquiry by collecting information from collateral sources during follow-up, as poor insight and/or emotional difficulties associated with stroke may affect the validity of stroke survivor’s self-report data. In addition, investigators should conduct serial assessments that include longer follow-up intervals (e.g., 6, 12, 24, and 36 months post-stroke) in order to better elucidate the predictive capacity of acute physical and cognitive test data across recovery and reintegration into the community. It would also be important to examine and control for relevant rehabilitation-related factors (including specific therapies received, frequency and duration of therapy, treatment adherence, and response to intervention) in order to better understand the relationship between acute performance and outcomes following stroke. Furthermore, investigators should build on this and related studies to determine whether brief screening measures of common stroke-related symptoms including, but not limited to, the NIHSS and SBT predict ecologically relevant post-stroke outcomes (e.g., capacity to drive, ability to successfully engage in relevant therapies). Finally, researchers should capitalize on the demonstrated utility of these broad, brief screening instruments in other acquired brain injury syndromes such as moderate to severe traumatic brain injury and hypoxia. It is likely that, similar to the current findings in stroke, measures such as the SBT and the Montreal Cognitive Assessment (MoCA; Nasreddine et al., 2005) will demonstrate incremental predictive validity with respect to relevant future outcomes in these related populations as well. Conflict of Interest None declared. Acknowledgments We thank the members of the Brain Recovery Core team from Washington University School of Medicine, Barnes-Jewish Hospital, and the Rehabilitation Institute of Saint Louis for the support and sharing of data for the purposes of this project. References Adamit , T. , Maeir , A. , Ben Assayag , E. , Bornstein , N. M. , Korczyn , A. D. , & Katz , N. ( 2015 ). Impact of first-ever mild stroke on participation at 3 and 6 month post-event: the TABASCO study . Disability and Rehabilitation , 37 , 667 – 673 . Google Scholar CrossRef Search ADS PubMed Almkvist Muren , M. , Hütler , M. , & Hooper , J. ( 2008 ). Functional capacity and health-related quality of life in individuals post stroke . Topics in Stroke Rehabilitation , 15 , 51 – 58 . Google Scholar CrossRef Search ADS PubMed Aufman , E. L. , Bland , M. D. , Barco , P. P. , Carr , D. B. , & Lang , C. E. ( 2013 ). Predictors of return to driving after stroke . American Journal of Physical Medicine & Rehabilitation , 92 , 627 – 634 . Google Scholar CrossRef Search ADS PubMed Babulal , G. M. , Huskey , T. N. , Roe , C. M. , Goette , S. A. , & Connor , L. T. ( 2015 ). Cognitive impairments and mood disruptions negatively impact instrumental activities of daily living performance in the first three months after a first stroke . Topics in Stroke Rehabilitation , 22 , 144 – 151 . Google Scholar CrossRef Search ADS PubMed Bellelli , G. , Nobili , A. , Annoni , G. , Morandi , A. , Djade , C. D. , Meagher , D. J. , et al. . ( 2015 ). Under-detection of delirium and impact of neurocognitive deficits on in-hospital mortality among acute geriatric and medical wards . European Journal of Internal Medicine , 26 , 696 – 704 . Google Scholar CrossRef Search ADS PubMed Berg , K. , Wood-Dauphinee , S. , & Williams , J. ( 1995 ). The Balance Scale: Reliability assessment with elderly residents and patients with an acute stroke . Scandinavian Journal of Rehabilitation Medicine , 27 , 27 – 36 . Google Scholar PubMed Bhogal , S. K. , Teasell , R. W. , Foley , N. C. , & Speechley , M. R. ( 2003 ). Community reintegration after stroke . Topics in Stroke Rehabilitation , 10 , 107 – 129 . Google Scholar CrossRef Search ADS PubMed Bland , M. D. , Sturmoski , A. , Whitson , M. , Connor , L. T. , Fucetola , R. , Huskey , T. , et al. . ( 2012 ). Prediction of discharge walking ability from initial assessment in a stroke inpatient rehabilitation facility population . Archives of Physical Medicine and Rehabilitation , 93 , 1441 – 1447 . Google Scholar CrossRef Search ADS PubMed Busch , M. A. , Coshall , C. , Heuschmann , P. U. , McKevitt , C. , & Wolfe , C. D. ( 2009 ). Sociodemographic differences in return to work after stroke: the South London Stroke Register (SLSR) . Journal of Neurology, Neurosurgery & Psychiatry , 80 ( 8 ), 888 – 893 . Google Scholar CrossRef Search ADS Carpenter , C. R. , Bassett , E. R. , Fischer , G. M. , Shirshekan , J. , Galvin , J. E. , & Morris , J. C. ( 2011 ). Four sensitive screening tools to detect cognitive dysfunction in geriatric emergency department patients: Brief Alzheimer’s Screen, Short Blessed Test, Ottawa 3DY, and the caregiver-completed AD8 . Academic Emergency Medicine: Official Journal of the Society for Academic Emergency Medicine , 18 , 374 – 384 . Google Scholar CrossRef Search ADS PubMed Cheng , P. T. , Liaw , M. Y. , Wong , M. K. , Tang , F. T. , Lee , M. Y. , & Lin , P. S. ( 1998 ). The sit-to-stand movement in stroke patients and its correlation with falling . Archives of Physical Medicine and Rehabilitation , 79 , 1043 – 1046 . Google Scholar CrossRef Search ADS PubMed Ch’Ng , A. M. , French , D. , & Mclean , N. ( 2008 ). Coping with the challenges of recovery from stroke long term perspectives of stroke support group members . Journal of Health Psychology , 13 , 1136 – 1146 . Google Scholar CrossRef Search ADS PubMed Cioncoloni , D. , Martini , G. , Piu , P. , Taddei , S. , Acampa , M. , Guideri , F. , et al. . ( 2013 ). Predictors of long-term recovery in complex activities of daily living before discharge from the stroke unit . NeuroRehabilitation , 33 , 217 – 223 . Google Scholar PubMed Collins , L. M. , Schafer , J. L. , & Kam , C. M. ( 2001 ). A comparison of inclusive and restrictive strategies in modern missing data procedures . Psychological Methods , 6 , 330 . Google Scholar CrossRef Search ADS PubMed Cumming , T. B. , Marshall , R. S. , & Lazar , R. M. ( 2013 ). Stroke, cognitive deficits, and rehabilitation: Still an incomplete picture . International Journal of Stroke , 8 , 38 – 45 . Google Scholar CrossRef Search ADS PubMed Del Ser , T. , Barba , R. , Morin , M. M. , Domingo , J. , Cemillan , C. , Pondal , M. , et al. . ( 2005 ). Evolution of cognitive impairment after stroke and risk factors for delayed progression . Stroke; A Journal of Cerebral Circulation , 36 , 2670 – 2675 . Google Scholar CrossRef Search ADS Desrosiers , J. , Noreau , L. , Rochette , A. , Bourbonnais , D. , Bravo , G. , & Bourget , A. ( 2006 ). Predictors of long-term participation after stroke . Disability and Rehabilitation , 28 , 221 – 230 . Google Scholar CrossRef Search ADS PubMed Douiri , A. , Rudd , A. G. , & Wolfe , C. D. ( 2013 ). Prevalence of poststroke cognitive impairment . Stroke; A Journal of Cerebral Circulation , 44 , 138 – 145 . Google Scholar CrossRef Search ADS Duarte , E. , Marco , E. , Muniesa , J. M. , Belmonte , R. , Aguilar , J. J. , & Escalada , F. ( 2010 ). Early detection of non-ambulatory survivors six months after stroke . NeuroRehabilitation , 26 , 317 – 323 . Google Scholar PubMed Edwards , D. F. , Hahn , M. , Baum , C. , & Dromerick , A. W. ( 2006 ). The impact of mild stroke on meaningful activity and life satisfaction . Journal of Stroke and Cerebrovascular Diseases , 15 , 151 – 157 . Google Scholar CrossRef Search ADS PubMed Ferrucci , L. , Bandinelli , S. , Guralnik , J. M. , Lamponi , M. , Bertini , C. , Falchini , M. , et al. . ( 1993 ). Recovery of functional status after stroke. A postrehabilitation follow-up study . Stroke; A Journal of Cerebral Circulation , 24 , 200 – 205 . Google Scholar CrossRef Search ADS Field , A. ( 2009 ). Discovering statistics using SPSS, 3rd edition . Thousand Oaks, CA : Sage Publications . Gillespie , D. C. , Bowen , A. , Chung , C. S. , Cockburn , J. , Knapp , P. , & Pollock , A. ( 2015 ). Rehabilitation for post-stroke cognitive impairment: An overview of recommendations arising from systematic reviews of current evidence . Clinical Rehabilitation , 29 , 120 – 128 . Google Scholar CrossRef Search ADS PubMed Goldstein , L. B. , Bertels , C. , & Davis , J. N. ( 1989 ). Interrater reliability of the NIH Stroke Scale . Archives of Neurology , 46 , 660 – 662 . Google Scholar CrossRef Search ADS PubMed Hackett , M. L. , Yapa , C. , Parag , V. , & Anderson , C. S. ( 2005 ). Frequency of depression after stroke . Stroke; A Journal of Cerebral Circulation , 36 , 1330 – 1340 . Google Scholar CrossRef Search ADS Harvey , R. L. ( 2015 ). Predictors of functional outcome following stroke . Physical Medicine and Rehabilitation Clinics of North America , 26 , 583 – 598 . Google Scholar CrossRef Search ADS PubMed Heruti , R. J. , Lusky , A. , Dankner , R. , Ring , H. , Dolgopiat , M. , Barell , V. , et al. . ( 2002 ). Rehabilitation outcome of elderly patients after a first stroke: Effect of cognitive status at admission on the functional outcome . Archives of Physical Medicine and Rehabilitation , 83 , 742 – 749 . Google Scholar CrossRef Search ADS PubMed Hillis , A. E. , Wityk , R. J. , Barker , P. B. , Ulatowski , J. A. , & Jacobs , M. A. ( 2003 ). Change in perfusion in acute nondominant hemisphere stroke may be better estimated by tests of hemispatial neglect than by the National Institutes of Health Stroke Scale . Stroke; A Journal of Cerebral Circulation , 34 , 2392 – 2396 . Google Scholar CrossRef Search ADS Hoffmann , T. , McKenna , K. , Cooke , D. , & Tooth , L. ( 2003 ). Outcomes after stroke: Basic and instrumental activities of daily living, community reintegration and generic health status . Australian Occupational Therapy Journal , 50 , 225 – 233 . Google Scholar CrossRef Search ADS Horner , R. D. , Swanson , J. W. , Bosworth , H. B. , & Matchar , D. B. ( 2003 ). Effects of race and poverty on the process and outcome of inpatient rehabilitation services among stroke patients . Stroke; A Journal of Cerebral Circulation , 34 , 1027 – 1031 . Google Scholar CrossRef Search ADS Horstmann , S. , Rizos , T. , Rauch , G. , Arden , C. , & Veltkamp , R. ( 2014 ). Feasibility of the Montreal Cognitive Assessment in acute stroke patients . European Journal of Neurology , 21 , 1387 – 1393 . Google Scholar CrossRef Search ADS PubMed Hoyle , M. , Gustafsson , L. , Meredith , P. , & Ownsworth , T. ( 2012 ). Participation after stroke: Do we understand all the components and relationships as categorised in the ICF? Brain Impairment , 13 , 4 – 15 . Google Scholar CrossRef Search ADS Jacquin , A. , Binquet , C. , Rouaud , O. , Graule-Petot , A. , Daubail , B. , Osseby , G. V. , et al. . ( 2014 ). Post-stroke cognitive impairment: high prevalence and determining factors in a cohort of mild stroke . Journal of Alzheimer’s Disease , 40 , 1029 – 1038 . Google Scholar CrossRef Search ADS PubMed Jette , A. M. , Keysor , J. , Coster , W. , Ni , P. , & Haley , S. ( 2005 ). Beyond function: Predicting participation in a rehabilitation cohort . Archives of Physical Medicine and Rehabilitation , 86 , 2087 – 2094 . Google Scholar CrossRef Search ADS PubMed Jokinen , H. , Melkas , S. , Ylikoski , R. , Pohjasvaara , T. , Kaste , M. , Erkinjuntti , T. , et al. . ( 2015 ). Post‐stroke cognitive impairment is common even after successful clinical recovery . European Journal of Neurology , 22 , 1288 – 1294 . Google Scholar CrossRef Search ADS PubMed Jørgensen , H. S. , Nakayama , H. , Raaschou , H. O. , Vive-Larsen , J. , Støier , M. , & Olsen , T. S. ( 1995 ). Outcome and time course of recovery in stroke. Part I: Outcome. The Copenhagen Stroke Study . Archives of Physical Medicine and Rehabilitation , 76 , 399 – 405 . Google Scholar CrossRef Search ADS PubMed Kalra , L. , & Langhorne , P. ( 2007 ). Facilitating recovery: Evidence for organized stroke care . Journal of Rehabilitation Medicine , 39 , 97 – 102 . Google Scholar CrossRef Search ADS PubMed Kalra , L. , Perez , I. , Gupta , S. , & Wittink , M. ( 1997 ). The influence of visual neglect on stroke rehabilitation . Stroke; A Journal of Cerebral Circulation , 28 , 1386 – 1391 . Google Scholar CrossRef Search ADS Katzman , R. , Brown , T. , Fuld , P. , Peck , A. , Schechter , R. , & Schimmel , H. ( 1983 ). Validation of a short Orientation-Memory-Concentration Test of cognitive impairment . The American Journal of Psychiatry , 140 , 734 – 739 . Google Scholar CrossRef Search ADS PubMed Kleim , J. A. , & Jones , T. A. ( 2008 ). Principles of experience-dependent neural plasticity: Implications for rehabilitation after brain damage . Journal of Speech, Language, and Hearing Research , 51 , S225 – S239 . Google Scholar CrossRef Search ADS Lang , C. E. , Bland , M. D. , Connor , L. T. , Fucetola , R. , Whitson , M. , Edmiaston , J. , et al. . ( 2011 ). The brain recovery core: Building a system of organized stroke rehabilitation and outcomes assessment across the continuum of care . Journal of Neurologic Physical Therapy , 35 , 194 – 201 . Google Scholar CrossRef Search ADS PubMed Laska , A. C. , Hellblom , A. , Murray , V. , Kahan , T. , & Von Arbin , M. ( 2001 ). Aphasia in acute stroke and relation to outcome . Journal of Internal Medicine , 249 , 413 – 422 . Google Scholar CrossRef Search ADS PubMed Lawrence , E. S. , Coshall , C. , Dundas , R. , Stewart , J. , Rudd , A. G. , Howard , R. , et al. . ( 2001 ). Estimates of the prevalence of acute stroke impairments and disability in a multiethnic population . Stroke; A Journal of Cerebral Circulation , 32 , 1279 – 1284 . Google Scholar CrossRef Search ADS Lazar , R. M. , Minzer , B. , Antoniello , D. , Festa , J. R. , Krakauer , J. W. , & Marshall , R. S. ( 2010 ). Improvement in aphasia scores after stroke is well predicted by initial severity . Stroke; A Journal of Cerebral Circulation , 41 , 1485 – 1488 . Google Scholar CrossRef Search ADS Lees , R. , Selvarajah , J. , Fenton , C. , Pendlebury , S. T. , Langhorne , P. , Stott , D. J. , et al. . ( 2014 ). Test accuracy of cognitive screening tests for diagnosis of dementia and multidomain cognitive impairment in stroke . Stroke; A Journal of Cerebral Circulation , 45 , 3008 – 3018 . Google Scholar CrossRef Search ADS Leśniak , M. , Bak , T. , Czepiel , W. , Seniów , J. , & Członkowska , A. ( 2008 ). Frequency and prognostic value of cognitive disorders in stroke patients . Dementia and Geriatric Cognitive Disorders , 26 , 356 – 363 . Google Scholar CrossRef Search ADS PubMed Little , R. J. ( 1988 ). A test of missing completely at random for multivariate data with missing values . Journal of the American Statistical Association , 83 , 1198 – 1202 . Google Scholar CrossRef Search ADS Loewen , S. C. , & Anderson , B. A. ( 1990 ). Predictors of stroke outcome using objective measurement scales . Stroke; A Journal of Cerebral Circulation , 21 , 78 – 81 . Google Scholar CrossRef Search ADS Lomaglio , M. J. , & Eng , J. J. ( 2005 ). Muscle strength and weight-bearing symmetry relate to sit-to-stand performance in individuals with stroke . Gait & Posture , 22 , 126 – 131 . Google Scholar CrossRef Search ADS PubMed Makin , S. D. J. , Turpin , S. , Dennis , M. S. , & Wardlaw , J. M. ( 2013 ). Cognitive impairment after lacunar stroke: Systematic review and meta-analysis of incidence, prevalence and comparison with other stroke subtypes . Journal of Neurology, Neurosurgery & Psychiatry , 84 , 893 – 900 . Google Scholar CrossRef Search ADS Marengoni , A. , Nobili , A. , Romano , V. , Tettamanti , M. , Pasina , L. , Djade , S. , et al. . ( 2013 ). Adverse clinical events and mortality during hospitalization and 3 months after discharge in cognitively impaired elderly patients . The Journals of Gerontology Series A: Biological Sciences and Medical Sciences , 68 , 419 – 425 . Google Scholar CrossRef Search ADS Maulden , S. A. , Gassaway , J. , Horn , S. D. , Smout , R. J. , & DeJong , G. ( 2005 ). Timing of initiation of rehabilitation after stroke . Archives of Physical Medicine and Rehabilitation , 86 , 34 – 40 . Google Scholar CrossRef Search ADS Mayo , N. E. , Anderson , S. , Barclay , R. , Cameron , J. I. , Desrosiers , J. , Eng , J. J. , et al. . ( 2015 ). getting on with the rest of your life following stroke: A randomized trial of a complex intervention aimed at enhancing life participation post stroke . Clinical Rehabilitation , 29 , 1198 – 1211 . Google Scholar CrossRef Search ADS PubMed Mayo , N. E. , Wood-Dauphinee , S. , Côté , R. , Durcan , L. , & Carlton , J. ( 2002 ). Activity, participation, and quality of life 6 months poststroke . Archives of Physical Medicine and Rehabilitation , 83 , 1035 – 1042 . Google Scholar CrossRef Search ADS PubMed Merz , Z. C. , Van Patten , R. , Mulhauser , K. , & Fucetola , R. ( 2017 ). Exploratory factor analysis of the reintegration to normal living index in a stroke population . Topics in Stroke Rehabilitation , 24 , 158 – 162 . Google Scholar CrossRef Search ADS PubMed Mesulam , M. ( 1985 ). Principles of behavioral neurology: Verbal and Nonverbal Cancellation Test . Philadelphia : FA Davis Company . Miller , E. L. , Murray , L. , Richards , L. , Zorowitz , R. D. , Bakas , T. , Clark , P. , et al. . ( 2010 ). Comprehensive overview of nursing and interdisciplinary rehabilitation care of the stroke patient . Stroke; A Journal of Cerebral Circulation , 41 , 2402 – 2448 . Google Scholar CrossRef Search ADS Mok , V. C. T. , Wong , A. , Lam , W. W. M. , Fan , Y. H. , Tang , W. K. , Kwok , T. , et al. . ( 2004 ). Cognitive impairment and functional outcome after stroke associated with small vessel disease . Journal of Neurology, Neurosurgery & Psychiatry , 75 , 560 – 566 . Google Scholar CrossRef Search ADS Nasreddine , Z. S. , Phillips , N. A. , Bédirian , V. , Charbonneau , S. , Whitehead , V. , Collin , I. , et al. . ( 2005 ). The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment . Journal of the American Geriatrics Society , 53 ( 4 ), 695 – 699 . Google Scholar CrossRef Search ADS PubMed Newman , D. A. ( 2003 ). Longitudinal modeling with randomly and systematically missing data: A simulation of ad hoc, maximum likelihood, and multiple imputation techniques . Organizational Research Methods , 6 , 328 – 362 . Google Scholar CrossRef Search ADS Nichols-Larsen , D. S. , Clark , P. C. , Zeringue , A. , Greenspan , A. , & Blanton , S. ( 2005 ). Factors influencing stroke survivors’ quality of life during subacute recovery . Stroke; A Journal of Cerebral Circulation , 36 , 1480 – 1484 . Google Scholar CrossRef Search ADS Norlander , A. , Carlstedt , E. , Jönsson , A. C. , Lexell , E. M. , Ståhl , A. , Lindgren , A. , et al. . ( 2016 ). Long-term predictors of social and leisure activity 10 years after stroke . PloSOne , 11 ( 2 ), 1 – 16 . Google Scholar CrossRef Search ADS Nys , G. M. S. , Van Zandvoort , M. J. E. , De Kort , P. L. M. , Van der Worp , H. B. , Jansen , B. P. W. , Algra , A. , et al. . ( 2005 ). The prognostic value of domain-specific cognitive abilities in acute first-ever stroke . Neurology , 64 , 821 – 827 . Google Scholar CrossRef Search ADS PubMed Oksala , N. K. J. , Jokinen , H. , Melkas , S. , Oksala , A. , Pohjasvaara , T. , Hietanen , M. , et al. . ( 2009 ). Cognitive impairment predicts poststroke death in long-term follow-up . Journal of Neurology, Neurosurgery & Psychiatry , 80 , 1230 – 1235 . Google Scholar CrossRef Search ADS Pang , M. Y. , Eng , J. J. , & Miller , W. C. ( 2007 ). Determinants of satisfaction with community reintegration in older adults with chronic stroke: Role of balance self-efficacy . Physical Therapy , 87 , 282 – 291 . Google Scholar CrossRef Search ADS PubMed Patel , M. D. , Coshall , C. , Rudd , A. G. , & Wolfe , C. D. ( 2002 ). Cognitive impairment after stroke: Clinical determinants and its associations with long‐term stroke outcomes . Journal of the American Geriatrics Society , 50 , 700 – 706 . Google Scholar CrossRef Search ADS PubMed Pedersen , P. M. , Stig Jørgensen , H. , Nakayama , H. , Raaschou , H. O. , & Olsen , T. S. ( 1995 ). Aphasia in acute stroke: Incidence, determinants, and recovery . Annals of Neurology , 38 , 659 – 666 . Google Scholar CrossRef Search ADS PubMed Pendlebury , S. T. , Cuthbertson , F. C. , Welch , S. J. , Mehta , Z. , & Rothwell , P. M. ( 2010 ). Underestimation of cognitive impairment by Mini-Mental State Examination versus the Montreal Cognitive Assessment in patients with transient ischemic attack and stroke . Stroke; A Journal of Cerebral Circulation , 41 , 1290 – 1293 . Google Scholar CrossRef Search ADS Pohjasvaara , T. , Vataja , R. , Leppävuori , A. , Kaste , M. , & Erkinjuntti , T. ( 2002 ). Cognitive functions and depression as predictors of poor outcome 15 months after stroke . Cerebrovascular Diseases , 14 , 228 – 233 . Google Scholar CrossRef Search ADS PubMed Protopsaltis , J. , Kokkoris , S. , Korantzopoulos , P. , Milionis , H. J. , Karzi , E. , Anastasopoulou , A. , et al. . ( 2009 ). Prediction of long-term functional outcome in patients with acute ischemic non-embolic stroke . Atherosclerosis , 203 , 228 – 235 . Google Scholar CrossRef Search ADS PubMed Riepe , M. W. , Riss , S. , Bittner , D. , & Huber , R. ( 2004 ). Screening for cognitive impairment in patients with acute stroke . Dementia and Geriatric Cognitive Disorders , 17 , 49 – 53 . Google Scholar CrossRef Search ADS PubMed Rubin , D. B. ( 1976 ). Inference and missing data . Biometrika , 63 , 581 – 592 . Google Scholar CrossRef Search ADS Rubin , D. B. ( 1987 ). Multiple imputation for nonresponse in surveys . New York : Wiley . Sachdev , P. S. , Brodaty , H. , Valenzuela , M. J. , Lorentz , L. , & Koschera , A. ( 2004 ). Progression of cognitive impairment in stroke patients . Neurology , 63 , 1618 – 1623 . Google Scholar CrossRef Search ADS PubMed Salvadori , E. , Pasi , M. , Poggesi , A. , Chiti , G. , Inzitari , D. , & Pantoni , L. ( 2013 ). Predictive value of MoCA in the acute phase of stroke on the diagnosis of mid-term cognitive impairment . Journal of Neurology , 260 , 2220 – 2227 . Google Scholar CrossRef Search ADS PubMed Schafer , J. L. , & Graham , J. W. ( 2002 ). Missing data: Our view of the state of the art . Psychological Methods , 7 , 147 – 177 . Google Scholar CrossRef Search ADS PubMed Scheffer , J. ( 2002 ). Dealing with missing data . Research Letters in the Information and Mathematical Sciences , 3 , 153 – 160 . Srikanth , V. K. , Thrift , A. G. , Saling , M. M. , Anderson , J. F. , Dewey , H. M. , Macdonell , R. A. , et al. . ( 2003 ). Increased risk of cognitive impairment 3 months after mild to moderate first-ever stroke . Stroke; A Journal of Cerebral Circulation , 34 , 1136 – 1143 . Google Scholar CrossRef Search ADS Stone , A. A. , Bachrach , C. A. , Jobe , J. B. , Kurtzman , H. S. , & Cain , V. S. ( 1999 ). The science of self-report: Implications for research and practice . Mahwah, NJ : Erlbaum . Strauss , E. , Sherman , E. M. , & Spreen , O. ( 2006 ). A Compendium of neuropsychological tests: Administration, norms, and commentary (3rd ed.). New York, NY : Oxford University Press . Tatemichi , T. K. , Desmond , D. W. , Stern , Y. , Paik , M. , Sano , M. , & Bagiella , E. ( 1994 ). Cognitive impairment after stroke: Frequency, patterns, and relationship to functional abilities . Journal of Neurology, Neurosurgery & Psychiatry , 57 , 202 – 207 . Google Scholar CrossRef Search ADS Tilling , K. , Sterne , J. A. , Rudd , A. G. , Glass , T. A. , Wityk , R. J. , & Wolfe , C. D. ( 2001 ). A new method for predicting recovery after stroke . Stroke; A Journal of Cerebral Circulation , 32 , 2867 – 2873 . Google Scholar CrossRef Search ADS Törnbom , K. , Persson , H. C. , Lundälv , J. , & Sunnerhagen , K. S. ( 2017 ). Self-assessed physical, cognitive, and emotional impact of stroke at 1 month: The importance of stroke severity and participation . Journal of Stroke and Cerebrovascular Diseases , 26 , 57 – 63 . Google Scholar CrossRef Search ADS PubMed Van Patten , R. , Merz , Z. C. , Mulhauser , K. , & Fucetola , R. ( 2016 ). Multivariable prediction of return to work at 6-month follow-up in patients with mild to moderate acute stroke . Archives of Physical Medicine and Rehabilitation , 97 , 2061 – 2067 . Google Scholar CrossRef Search ADS PubMed Veerbeek , J. M. , Kwakkel , G. , van Wegen , E. E. , Ket , J. C. , & Heymans , M. W. ( 2011 ). Early prediction of outcome of activities of daily living after stroke . Stroke; A Journal of Cerebral Circulation , 42 , 1482 – 1488 . Google Scholar CrossRef Search ADS Veerbeek , J. M. , van Wegen , E. , van Peppen , R. , van der Wees , P. J. , Hendriks , E. , Rietberg , M. , et al. . ( 2014 ). What is the evidence for physical therapy poststroke? A systematic review and meta-analysis . PloSOne , 9 ( 2 ), 1 – 33 . Google Scholar CrossRef Search ADS Vellone , E. , Savini , S. , Fida , R. , Dickson , V. V. , Melkus , G. D. , Carod-Artal , F. J. , et al. . ( 2015 ). Psychometric evaluation of the Stroke Impact Scale 3.0 . The Journal of Cardiovascular Nursing , 30 , 229 – 241 . Google Scholar CrossRef Search ADS PubMed Wang , Y. C. , Kapellusch , J. , & Garg , A. ( 2014 ). Important factors influencing the return to work after stroke . Work (Reading, Mass.) , 47 , 553 – 559 . Google Scholar PubMed Wattanapan , P. , Kovindha , A. , Piravej , K. , & Kuptniratsaikul , V. ( 2010 ). Relationship between the ability to change from a supine to a sitting position at admission and mobility outcomes after stroke rehabilitation . Journal of the Medical Association of Thailand , 93 , S21 – S24 . Google Scholar PubMed Wilkins , C. H. , Wilkins , K. L. , Meisel , M. , Depke , M. , Williams , J. , & Edwards , D. F. ( 2007 ). Dementia undiagnosed in poor older adults with functional impairment . Journal of the American Geriatrics Society , 55 , 1771 – 1776 . Google Scholar CrossRef Search ADS PubMed Wolf , T. J. , & Rognstad , M. C. ( 2013 ). Changes in cognition following mild stroke . Neuropsychological Rehabilitation , 23 ( 2 ), 256 – 266 . Google Scholar CrossRef Search ADS PubMed Woodman , P. , Riazi , A. , Pereira , C. , & Jones , F. ( 2014 ). Social participation post stroke: A meta-ethnographic review of the experiences and views of community-dwelling stroke survivors . Disability and Rehabilitation , 36 , 2031 – 2043 . Google Scholar CrossRef Search ADS PubMed © The Author 2017. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Archives of Clinical Neuropsychology Oxford University Press

Predicting Cognitive Functioning, Activities of Daily Living, and Participation 6 Months after Mild to Moderate Stroke

Loading next page...
 
/lp/ou_press/predicting-cognitive-functioning-activities-of-daily-living-and-th0UlJmyCc
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0887-6177
eISSN
1873-5843
D.O.I.
10.1093/arclin/acx096
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective Predicting neurocognitive and functional outcomes in stroke is an important clinical task, especially in rehabilitation settings. We assessed acute predictors of cognitive and functional outcomes 6 months after mild to moderate stroke. Methods We conducted a retrospective analysis of acute clinical data and 6-month follow-up telephone interviews for 498 mild to moderate stroke patients. Predictors were sociodemographic variables, the National Institute of Health Stroke Scale (NIHSS), basic physical measures, the Mesulam Cancellation Test, the Short Blessed Test (SBT), Trails A/B, and the Boston Naming Test. The outcome variables were the Communication, Memory and Thinking, ADL/IADLs, and Participation subscales from the Stroke Impact Scale. We conducted four hierarchical multiple regression analyses with demographic variables and the NIHSS score entered into the first step, followed by physical variables in the second step, and neuropsychological variables in the final step. Results Physical variables explained more variance in ADL/IADLs and Participation outcomes than in Communication and Memory and Thinking outcomes, while cognitive predictors exhibited the opposite trend. The SBT was the only significant independent predictor of Communication and Memory and Thinking (p’s < .001), while the NIHSS was the only measure that significantly predicted ADL/IADLs (p < .001) and Participation (p = .002). Poorer performance on screening measures predicted worse cognitive and functional outcomes 6 months post-stroke. Conclusions These results support the clinical utility of administering brief screening instruments during acute recovery from mild to moderate stroke. Neuropsychologists should prioritize performance on screening measures assessing acute neurologic status and cognitive dysfunction when making recommendations for post-stroke rehabilitation. Stroke, Rehabilitation, Assessment, Everyday functioning Introduction Mild to moderate stroke is associated with a wide range of distressing and disabling physical, neuropsychological, and functional impairments (Harvey, 2015; Mayo, Wood-Dauphinee, Côté, Durcan, & Carlton, 2002). As a result, a large proportion of patients who suffer such an event have difficulty successfully reintegrating into prior social and occupational roles (Adamit et al., 2015; Van Patten, Merz, Mulhauser, & Fucetola, 2016; Woodman, Riazi, Pereira, & Jones, 2014). Restricted participation in life situations and activities is common after stroke and poses a significant challenge for survivors navigating role changes in multiple life domains, including occupational and family roles (Ch’Ng, French, & Mclean, 2008; Törnbom, Persson, Lundälv, & Sunnerhagen, 2017). As many as 65% of stroke survivors face limitations in work, recreation, or social activity, all of which serve as barriers to return to normal living (Adamit et al., 2015; Mayo et al., 2002; Törnbom et al., 2017). Importantly, the failure to resume meaningful social, vocational, or leisure activities is also associated with worse health outcomes, as well as lower health-related and global quality of life (Bhogal, Teasell, Foley, & Speechley, 2003; Norlander et al., 2016). Therefore, elucidating factors that influence level of participation is essential for understanding recovery trajectory and outcomes for mild to moderate stroke (Adamit et al., 2015). Given the importance of post-stroke engagement to health and quality of life, investigators have sought to determine which factors limit participation. In particular, a number of functional and disability factors limit individuals’ participation in necessary and valued activities after stroke (Jette, Keysor, Coster, Ni, & Haley, 2005; Mayo et al., 2015). Limited mobility, depressive symptoms, apathy, cognitive impairment, fatigue, difficulty performing activities of daily living (ADLs), communication problems, lack of social connection, and lack of self-efficacy all act to restrict participation (Desrosiers et al., 2006; Hoyle, Gustafsson, Meredith, & Ownsworth, 2012; Pang, Eng, & Miller, 2007). Studies of patients recovering from mild to moderate stroke have found that, while they typically achieve independence in self-care and everyday functioning (commonly referred to as ADLs), they may require additional supports when returning to work or leisure activities or have problems with communication or carrying out instrumental activities of daily living (IADLs; Almkvist Muren, Hütler, & Hooper, 2008; Aufman, Bland, Barco, Carr, & Lang, 2013; Edwards, Hahn, Baum, & Dromerick, 2006; Van Patten et al., 2016). In contrast to basic ADLs, IADLs are more complex tasks that are necessary for independent functioning in the community, such as household chores, shopping, taking medications, and managing finances (Hoffmann, McKenna, Cooke, & Tooth, 2003). As many as 54% of stroke survivors have difficulty performing IADLs independently, which is similar to the percentage of those reporting restrictions in participation (Adamit et al., 2015; Mayo et al., 2002; Törnbom et al., 2017). This is particularly significant given the importance of ADLs and IADLs for successful engagement in social, occupational, and recreational activities following stroke (Cioncoloni et al., 2013; Hoffman et al., 2003). Among other factors, cognitive impairment has been shown to affect the ability of individuals to perform ADL/IADLs and to participate in meaningful activities after stroke (Babulal, Huskey, Roe, Goette, & Connor, 2015; Pohjasvaara, Vataja, Leppävuori, Kaste, & Erkinjuntti, 2002; Tatemichi et al., 1994). Post-stroke cognitive deficits are common, even following mild stroke, and can have a substantial affect on recovery and functional outcome (Adamit et al., 2015; Heruti et al., 2002; Jokinen et al., 2015). Although stroke has been associated with deficits in a broad range of cognitive domains including attention, memory, language, motor control, visuospatial skills, and executive functioning, the most consistent effects appear to be impairments in processing speed, attention, and executive functions (Cumming, Marshall, & Lazar, 2013). However, depending on lesion location, domain-specific deficits, such as impaired receptive and/or expressive language abilities, may occur more frequently (Laska, Hellblom, Murray, Kahan, & Von Arbin, 2001). Approximately one-third of stroke patients experience significant language disturbance known as aphasia, though many nonaphasic patients exhibit less severe impairments in complex language and communication skills (Pedersen, Stig Jørgensen, Nakayama, Raaschou, & Olsen, 1995). Srikanth et al. (2003) examined nonaphasic patients with first-ever mild to moderate stroke 3 months post-injury and noted deficits in both complex language and executive functioning. The prevalence of post-acute cognitive impairment has been estimated to be between 20% and 40%, with a similar percentage of stroke survivors reporting persisting difficulties with language and communication (Douiri, Rudd, & Wolfe, 2013; Laska et al., 2001; Makin, Turpin, Dennis, & Wardlaw, 2013; Oksala et al., 2009; Patel, Coshall, Rudd, & Wolfe, 2002). The deleterious effects of impairments in cognition, communication, and ADL/IADLs on health outcomes and quality of life after stroke have been well-established (Hoyle et al., 2012). Consequently, a number of studies have sought to identify acute predictors of these cognitive and functional outcomes, with the goal of tailoring rehabilitation services and enhancing clinical care. With regard to reestablishing independence in ADLs, studies suggest that younger age, decreased stroke severity, and better motor and functional abilities at stroke onset are most associated with recovery in basic ADLs beyond 3 months after stroke (Harvey, 2015; Veerbeek, Kwakkel, van Wegen, Ket, & Heymans, 2011). There is also some evidence that being male, having intact cognitive and language abilities (particularly absence of aphasia), and lack of health comorbidities (e.g., diabetes mellitus) predict patients’ ability to regain independence in basic ADLs (Duarte et al., 2010; Protopsaltis et al., 2009; Loewen & Anderson, 1990; Tilling et al., 2001). In contrast, acute predictors associated with post-stroke functional recovery in IADLs are not well understood (Cioncoloni et al., 2013). Currently, the prognostic variable most associated with independence in IADLs beyond 3 months after stroke is basic ADL status at hospital discharge (Cioncoloni et al., 2013; Hoffman et al., 2003). Older age, being female, greater stroke severity, poor upper limb strength, and acute cognitive impairment have also been implicated in worse recovery of IADLs beyond 3 months after stroke (Cioncoloni et al., 2013; Hoffman et al., 2003; Mok et al., 2004; Nys et al., 2005). While a number of studies have examined acute predictors of functional outcomes after stroke, fewer have examined predictors of post-stroke cognitive impairment (Nys et al., 2005). Thus far, demographic and stroke-related factors that have been associated with cognitive impairment beyond 3 months after stroke include older age, female sex, minority status, lower socioeconomic status, lower education, pre-stroke cognitive decline, previous stroke, left hemisphere stroke, and greater stroke severity (Lazar et al., 2010; Mok et al., 2004; Nys et al., 2005; Patel et al., 2002; Sachdev, Brodaty, Valenzuela, Lorentz, & Koschera, 2004). Additionally, both global cognitive impairment and domain-specific deficits in executive functioning, visuospatial skills (e.g., unilateral neglect), and language (e.g., aphasia) at stroke onset have been implicated in post-stroke cognitive impairment (Hillis, Wityk, Barker, Ulatowski, & Jacobs, 2003; Kalra, Perez, Gupta, & Wittink, 1997; Laska et al., 2001; Leśniak, Bak, Czepiel, Seniów, & Członkowska, 2008; Nys et al., 2005; Pendlebury, Cuthbertson, Welch, Mehta, & Rothwell, 2010). For example, acute language disturbance, including aphasia, is the strongest predictor of impaired communication abilities beyond 3 months after stroke (Harvey, 2015). A probable explanation for the prognostic value of acute cognitive deficits is the relative constancy of post-stroke cognitive impairment, even following mild stroke (Del Sur et al., 2005; Jacquin et al., 2014; Wolf & Rognstad, 2013). Pre-post assessments suggest that some cognitive deficits remain stable from acute measurement (within 24 hr) to 3 months post-stroke, with even fewer changes in cognition occurring between subacute assessment (within 3 weeks of discharge) and follow-up assessment 6 months post-stroke (Riepe, Riss, Bittner, & Huber, 2004; Wolf & Rognstad, 2013). Moreover, the overall prevalence of post-stroke cognitive impairment appears to remain relatively unchanged beyond 3 months after stroke for as long as 14 years (Del Sur et al., 2005; Douiri et al., 2013). Long-term recovery of independence in ADL/IADLs may follow a similar trajectory, owing largely to early recovery of physical abilities and acute ADL status (Cioncoloni et al., 2013; Hoffman et al., 2003). Overall, the literature on cognitive and functional outcomes post-stroke supports the predictive value of cognitive, physical, and functional status at stroke onset, in addition to certain neurologic and demographic factors. Identifying acute predictors of cognitive and functional outcomes following mild to moderate stroke is an important goal for research and clinical care. Recovery from stroke is highly variable and deficits will often resolve naturally over time, with compensation, and/or due to rehabilitation efforts (Cumming et al., 2013). In particular, early initiation and maintenance of targeted rehabilitation services has been associated with better recovery after stroke (Bhogal et al., 2003; Maulden et al., 2005). Therefore, identifying acute factors that meaningfully predict long-term outcomes can improve predictive models for stroke recovery, as well as help inform rehabilitation services for patients recovering from stroke. Although various aspects of acute cognitive, physical, and functional status have been suggested to play a role in determining outcomes after stroke, empirical support for their prognostic value is often limited by methodological challenges (Harvey, 2015). Furthermore, few studies have conducted comprehensive investigations examining the incremental predictive validity of specific cognitive and physical/functional assessments commonly administered in acute stroke. The purpose of the current study was to systematically assess acute predictors of cognitive and functional outcomes 6 months after mild to moderate stroke. Specifically, given previous research suggesting that cognitive tests predict important cognitive and functional outcomes (e.g., Babulal et al., 2015; Harvey, 2015; Hillis et al., 2003; Kalra et al., 1997; Laska et al., 2001; Leśniak et al., 2008; Nys et al., 2005; Pohjasvaara et al., 2002; Tatemichi et al., 1994), we hypothesized that cognitive performance would demonstrate incremental validity above and beyond physical/functional abilities and demographic and stroke-related factors with regard to predicting impairments in cognition, communication, participation, and/or ADL/IADLs 6 months after stroke. Methods Participants The dataset utilized in the current study was acquired from the Brain Recovery Core (BRC; Lang et al., 2011), a collaborative endeavor among a university-affiliated medical center, an acute care hospital, and a rehabilitation institute in a mid-sized Midwestern city in the United States. We collected demographic and clinical data from stroke patients’ acute care hospital records and 6-month follow-up data from recorded responses to telephone interviews. As part of the clinical services provided at these institutions, all patients are afforded the opportunity to provide informed consent for the use of their clinical data for research purposes. The current study sample represents the set of stroke patients who consented to release their data between 2010 and 2014. Data from 602 stroke patients who completed the 6 month follow-up interview were acquired from the BRC. Inclusionary criteria for all patients were simply that (a) the patient received clinical services for an acute stroke through the BRC between 2010 and 2014, (b) the patient voluntarily provided informed consent, and (c) the patient completed the 6 month follow-up telephone interview. Patients were excluded for the following reasons: (a) a National Institute of Health Stroke Scale (NIHSS) score >16 (indicating severe stroke), (b) a NIHSS Aphasia item score of 2 or 3, indicating severe to global aphasia (given that aphasia is likely to compromise the ability of patients to provide valid responses to oral survey questions), and (c) a NIHSS Dysarthria item score of 2 or 3, indicating severe dysarthria or intubation (given that dysarthria will also prevent adequate verbal expression). Following these exclusions, the final sample size was 498 (mean age = 64.50 ± [SD] 14.54 years; mean NIHSS total score = 3.55 ± [SD] 3.76). A full characterization of the sample demographics is provided in Table 1. Table 1. Group characteristics (n = 498) Characteristic Mean ± SD or % N Range Age 64.50 ± 14.54 498 21–98 Education (years) 12.87 ± 2.70 498 0–22 Sex  Male 51.00 254  Female 49.00 244 Race  Caucasian 62.90 313  African American 33.90 169  Asian 0.40 2  Unknown 2.80 14 Marital status  Married 45.20 225  Significant other 4.40 22  Divorced 11.40 57  Never married 0.60 3  Widowed 14.70 73  Separated 1.40 7  Single 18.70 93  Unknown 3.40 17 Working prior to stroke  Yes 27.50 137  No 61.20 305  Unknown 11.20 56  NIHSS total score 3.55 ± 3.76 498 0–16  Short Blessed Test 5.92 ± 5.87 458 0–28 Stroke diagnosis  Ischemic 60.60 302  Hemorrhagic 10.80 54  Unknown 28.50 142  PHQ-9 at 6-month follow-up 6.42 ± 6.35 498 0–27 Minimal-49.40% Mild-24.70% Moderate-13.65% Mod-Sev-6.22% Severe-6.02% Characteristic Mean ± SD or % N Range Age 64.50 ± 14.54 498 21–98 Education (years) 12.87 ± 2.70 498 0–22 Sex  Male 51.00 254  Female 49.00 244 Race  Caucasian 62.90 313  African American 33.90 169  Asian 0.40 2  Unknown 2.80 14 Marital status  Married 45.20 225  Significant other 4.40 22  Divorced 11.40 57  Never married 0.60 3  Widowed 14.70 73  Separated 1.40 7  Single 18.70 93  Unknown 3.40 17 Working prior to stroke  Yes 27.50 137  No 61.20 305  Unknown 11.20 56  NIHSS total score 3.55 ± 3.76 498 0–16  Short Blessed Test 5.92 ± 5.87 458 0–28 Stroke diagnosis  Ischemic 60.60 302  Hemorrhagic 10.80 54  Unknown 28.50 142  PHQ-9 at 6-month follow-up 6.42 ± 6.35 498 0–27 Minimal-49.40% Mild-24.70% Moderate-13.65% Mod-Sev-6.22% Severe-6.02% Note: NIHSS = National Institutes of Health Stroke Scale; PHQ-9 = Patient Health Questionnaire, 9-item. Table 1. Group characteristics (n = 498) Characteristic Mean ± SD or % N Range Age 64.50 ± 14.54 498 21–98 Education (years) 12.87 ± 2.70 498 0–22 Sex  Male 51.00 254  Female 49.00 244 Race  Caucasian 62.90 313  African American 33.90 169  Asian 0.40 2  Unknown 2.80 14 Marital status  Married 45.20 225  Significant other 4.40 22  Divorced 11.40 57  Never married 0.60 3  Widowed 14.70 73  Separated 1.40 7  Single 18.70 93  Unknown 3.40 17 Working prior to stroke  Yes 27.50 137  No 61.20 305  Unknown 11.20 56  NIHSS total score 3.55 ± 3.76 498 0–16  Short Blessed Test 5.92 ± 5.87 458 0–28 Stroke diagnosis  Ischemic 60.60 302  Hemorrhagic 10.80 54  Unknown 28.50 142  PHQ-9 at 6-month follow-up 6.42 ± 6.35 498 0–27 Minimal-49.40% Mild-24.70% Moderate-13.65% Mod-Sev-6.22% Severe-6.02% Characteristic Mean ± SD or % N Range Age 64.50 ± 14.54 498 21–98 Education (years) 12.87 ± 2.70 498 0–22 Sex  Male 51.00 254  Female 49.00 244 Race  Caucasian 62.90 313  African American 33.90 169  Asian 0.40 2  Unknown 2.80 14 Marital status  Married 45.20 225  Significant other 4.40 22  Divorced 11.40 57  Never married 0.60 3  Widowed 14.70 73  Separated 1.40 7  Single 18.70 93  Unknown 3.40 17 Working prior to stroke  Yes 27.50 137  No 61.20 305  Unknown 11.20 56  NIHSS total score 3.55 ± 3.76 498 0–16  Short Blessed Test 5.92 ± 5.87 458 0–28 Stroke diagnosis  Ischemic 60.60 302  Hemorrhagic 10.80 54  Unknown 28.50 142  PHQ-9 at 6-month follow-up 6.42 ± 6.35 498 0–27 Minimal-49.40% Mild-24.70% Moderate-13.65% Mod-Sev-6.22% Severe-6.02% Note: NIHSS = National Institutes of Health Stroke Scale; PHQ-9 = Patient Health Questionnaire, 9-item. Demographic and clinical data included in the current study was obtained from hospital records and typically collected within 7 days of patients’ admission to the hospital (mean number of days = 1.86 ± [SD] 3.04, maximum = 46 days). Follow-up data was obtained from telephone interviews conducted approximately 6-months after stroke onset and hospital admission. Variables Assessed Demographic and stroke-related variables The present study assessed the following demographic and neurologic variables: age, gender, race, marital status, stroke type [ischemic, hemorrhagic, or unknown], and NIHSS total score. The NIHSS is a broad 15-item neurological screening exam of stroke severity that briefly assesses for deficits in the areas of consciousness level, language, neglect, visual-field loss, extraocular movement, motor strength, ataxia, dysarthria, and sensory loss (Goldstein et al., 1989). Physical/functional variables Physical and functional status was assessed with the Berg Balance Scale (BBS) total score, a measure of ambulation (in feet), and functional mobility tasks including Supine to Sit and Sit to Stand. The BBS is a 14-item measure of balance and risk for falls in older adults measured through direct observation on a scale of 0 (inability to complete tw:he task) to 4 (independent task completion; Berg, Wood-Dauphinee, & Williams, 1995). Supine to Sit and Sit to Stand are commonly used measures of fall risk, muscle strength, and weight symmetry (Cheng et al., 1998; Lomaglio & Eng, 2005; Wattanapan, Kovindha, Piravej, & Kuptniratsaikul, 2010). These tasks were measured on a scale of 1 (requiring total assistance) to 7 (complete independence). Cognitive variables Cognitive variables included the Mesulam Cancellation Test total score, Short Blessed Test (SBT), Trails A and B, and the Boston Naming Test (BNT). The Mesulam Cancellation Test is a measure of visual attention and neglect measured by speed of completion (Mesulam, 1985). The SBT is a 6-item screener for deficits in orientation, registration, and attention (Carpenter et al., 2011). Trails A and B are measures of attention and executive functioning, measured by speed of completion, and the BNT is a 60-item language measure of visual confrontation naming (Strauss, Sherman, & Spreen, 2006). Cognitive and functional outcomes We assessed post-stroke cognitive and functional outcomes with selected subscales of the Stroke Impact Scale (SIS; Vellone et al., 2015). The SIS is a self-report instrument that measures quality of life and functional impairments across a range of behaviors and activities following stroke. For the purposes of the current study, the Communication, Memory and Thinking, ADL/IADLs, and Participation subscales from the SIS were included in analyses. The Communication subscale assesses various aspects of both receptive and expressive language. The Memory and Thinking subscale measures retrospective and prospective memory, as well as processing speed, problem solving, and attentional abilities. The ADL/IADLs subscale assesses difficulty in self-care and daily activities (e.g., dressing and bathing), as well as more complex tasks necessary for independent functioning in the community, such as household chores and shopping. Finally, the Participation subscale measures degree of restriction in activities such as work and recreation. Statistical Analyses Data were analyzed using SPSS 24.0. Consistent with other retrospective investigations of the BRC dataset (Aufman et al., 2013; Van Patten et al., 2016), several variables of interest exhibited notable (>10%) missing data. Specifically, data from the following variables were incomplete: stroke type (28.6% missing), the BBS (11.85% missing), FIM Supine to Sit (20.28% missing), FIM Sit to Stand (8.03% missing), Ambulation (22.89% missing), the SBT (8.03% missing), Trails A (31.93% missing) and B (47.19% missing), the Mesulam Cancellation Test Total score (40.16% missing), and the BNT (14.86% missing). Although we were unable to determine the reason for missing data on a case-by-case basis, we did examine the overall mechanism of missingness using the widely accepted Little (1988) criteria. Specifically, we grouped the eight aforementioned variables with >10% missing data into missing and not-missing groups, then assessed group differences across variable with small proportions (≤10%) of missing data (i.e., age, sex, race, years of education, marital status, NIHSS Total Score, SBT, and PHQ-9 Total Score). Such an analysis provides insight into whether data are missing completely at random (MCAR) or missing at random (MAR; see Schafer & Graham, 2002). That is, it determines whether missing data are dependent on other variables in the dataset or not. Importantly, empirical evidence has repeatedly demonstrated that simple case deletion and mean imputation methods lead to biased parameter estimates, most notably when the mechanism of missingness is nonrandom (Newman, 2003; Rubin, 1976; Schafer & Graham, 2002; Scheffer, 2002). Alternatively, the multiple imputation (MI) technique significantly reduces inaccuracy when estimating missing data values, partially by accounting for random error and thereby maintaining the variance structure of a dataset even when data are MAR (Collins, Shafer, & Kam, 2001; Rubin, 1987; Schafer & Graham, 2002). Results of the mechanism of missingness analysis showed that 23/64 (36%) of these nonparametric tests were significant, indicating that our missing data are MAR rather than MCAR. Moreover, 4/8 (50%) of the NIHSS analyses and 5/8 (63%) of the SBT analyses were significant, suggesting that those patients who were more cognitively impaired (as measured by the NIHSS and SBT) were more likely to exhibit missing data. Therefore, given the unambiguous support in simulation studies for the MI approach compared to more simplistic techniques (Newman, 2003; Schafer & Graham, 2002; Scheffer, 2002), we conducted a 5-iteration MI procedure prior to running the primary analysis of interest. To address our primary aim, we conducted four hierarchical multiple regression analyses, with demographic variables (age, gender, race, marital status, stroke type [ischemic or hemorrhagic], and Total NIHSS score) entered into the first step, physical variables (FIM Supine to Sit, FIM Sit to Stand, BBS Total Score, and Ambulation [in feet]) entered into the second step, and neuropsychological variables (Mesulam Cancellation Test Total score, SBT, Trails A and B, and the BNT) entered into the final step. For all physical and neuropsychological variables, raw data were analyzed. Although it would have been possible to use standardized data, this option was considered to be suboptimal given that (a) relevant demographic variables were already controlled in the regression analyses and (b) the process of standardizing raw scores can reduce the variance in a given test, thereby restricting its range and artificially attenuating correlations. The outcome variables were selected subscales from the SIS including Communication, Memory and Thinking, ADL/IADLs, and Participation. Prior to conducting the analyses, the data were examined for outliers, influential cases, and assumptions of linear regression models. Outliers were operationalized as values ±3.29 standard deviations from the mean on the variable of interest. Consistent with established guidelines (Field, 2009), these values were then replaced with values equal to 3 SDs from the mean in the same direction. Variables with identified outliers were FIM Sit to Stand, Ambulation (in feet), the SBT, Trails A and B, and the Mesulam Cancellation Test. Generally accepted criteria (Field, 2009) were utilized to identify influential cases and violations of assumptions. Influential cases were defined as Cook’s distances >1 or DFBetas >1. With respect to statistical assumptions, prior to examining each regression model, multicollinearity was examined through a correlation matrix of the relationships among predictors (and defined as coefficients > ±0.9; see Tables 2 and 3). Next, both linearity of the modeled relationship and homoscedasticity were tested through visual inspection of plotted standardized residuals compared to predicted values. Third, independent errors were defined as residuals with Durbin–Watson values > ±4 and violations of the assumption of normally distributed errors were examined through visual inspection of the histogram of residuals. Unless otherwise specified, no influential cases were present and all assumptions were met for each of the models. Table 2. Correlations among physical/functional variables Measure 1 2 3 4 1. Berg Balance Scale — 2. Supine to Sit 0.763** — 3. Sit to Stand 0.863** 0.823** — 4. Ambulation (feet) 0.533** 0.358** 0.437** — Measure 1 2 3 4 1. Berg Balance Scale — 2. Supine to Sit 0.763** — 3. Sit to Stand 0.863** 0.823** — 4. Ambulation (feet) 0.533** 0.358** 0.437** — *p < .05; **p < .01; ***p < .001. Table 2. Correlations among physical/functional variables Measure 1 2 3 4 1. Berg Balance Scale — 2. Supine to Sit 0.763** — 3. Sit to Stand 0.863** 0.823** — 4. Ambulation (feet) 0.533** 0.358** 0.437** — Measure 1 2 3 4 1. Berg Balance Scale — 2. Supine to Sit 0.763** — 3. Sit to Stand 0.863** 0.823** — 4. Ambulation (feet) 0.533** 0.358** 0.437** — *p < .05; **p < .01; ***p < .001. Table 3. Correlations among cognitive variables Measure 1 2 3 4 5 1. Short Blessed Test — 2. Trails A (seconds) 0.520** — 3. Trails B (seconds) 0.607** 0.789** — 4. Cancellation Test 0.026 0.226** 0.089 — 5. Boston Naming Test −0.618** −0.379** −0.480** 0.007 — Measure 1 2 3 4 5 1. Short Blessed Test — 2. Trails A (seconds) 0.520** — 3. Trails B (seconds) 0.607** 0.789** — 4. Cancellation Test 0.026 0.226** 0.089 — 5. Boston Naming Test −0.618** −0.379** −0.480** 0.007 — *p < .05; **p < .01; ***p < .001. Table 3. Correlations among cognitive variables Measure 1 2 3 4 5 1. Short Blessed Test — 2. Trails A (seconds) 0.520** — 3. Trails B (seconds) 0.607** 0.789** — 4. Cancellation Test 0.026 0.226** 0.089 — 5. Boston Naming Test −0.618** −0.379** −0.480** 0.007 — Measure 1 2 3 4 5 1. Short Blessed Test — 2. Trails A (seconds) 0.520** — 3. Trails B (seconds) 0.607** 0.789** — 4. Cancellation Test 0.026 0.226** 0.089 — 5. Boston Naming Test −0.618** −0.379** −0.480** 0.007 — *p < .05; **p < .01; ***p < .001. Results With respect to the overall hierarchical regressions, adding physical/functional variables to the models while controlling for demographics and stroke-related factors explained more variance in ADL/IADLs (ΔR2 = 4.2%; F(4, 428) = 5.57; p < .001) and Participation (ΔR2 = 2.3%; F(4, 424) = 2.93; p < .001) outcomes than in Communication (ΔR2 = 0.9%; F(4, 454) = 1.09; p = .32) and Memory and Thinking (ΔR2 = 0.6%; F(4, 451) = 0.88; p = .61.) outcomes. Conversely, adding cognitive variables in the final step of the models explained more variance in Communication (ΔR2 = 7.8%; F(5, 449) = 7.93; p < .001) and Memory and Thinking (ΔR2 = 6.9%; F(5, 446) = 6.83; p < .001) than in ADL/IADLs (ΔR2 = 0.9%; F(5, 423) = 1.01; p = .45) and Participation (ΔR2 = 1.7%; F(5, 419) = 1.76; p = .15) subscales. The overall models were all statistically significant (p’s < .001) and accounted for small amounts of variance in SIS cognitive subscales (Communication, 10.7%–11.3%; Memory and Thinking, 8.4%–10.6%) and moderate degrees of variance in functional subscales (ADL/IADLs, 22.0%–24.6%; Participation, 17.7%–18.9%). Significant individual-level predictors of the four outcome variables are presented in Tables 4–7. For the Communication and Memory and Thinking models, only the SBT was significant (p’s < .001). As expected, greater impairment on the SBT predicted poorer self-reported SIS cognition. With regard to the ADL/IADLs model, gender (p = .016), race (p = .021), and NIHSS Total score (p < .001) were significant predictors. Similarly, in the Participation model, gender (p = .069) and race (p = .066) approached significance, while the NIHSS Total Score (p = .002) remained significant. For both functional SIS subscales, the performance of male and Caucasian participants was higher than female and non-Caucasian participants. In terms of NIHSS Total scores, poorer performance predicted worse self-reported functioning in ADL/IADLs and Participation. Pearson’s correlation coefficients for the relationships between the SBT, NIHSS Total score, and SIS cognitive and functional outcomes are presented in Table 8. Overall, all relationships were in the expected direction (i.e., greater symptom severity on one measure correlated with greater symptom severity on the other measure). Descriptive statistics for SIS subscales and physical/functional and cognitive variables are presented in Table 9. Table 4. Hierarchical multiple regression coefficients with memory and thinking as the outcome variable Predictors B SE B β Step 1  Constant 30.36 3.45  Age −0.01 0.02 −.023  Sex −0.54 0.63 −.036  Race −0.87 0.70 −.063  NIHSS −0.06 0.08 −.035  Education 0.20 0.16 .084  Marital status −0.37 0.68 −.027  Stroke diagnosis −0.95 0.89 −.052 Step 2  Constant 28.41 4.63  Age −0.01 0.02 −.016  Sex −0.50 0.65 −.031  Race −0.93 0.70 −.067  NIHSS −0.01 0.11 −.007  Education 0.21 0.17 .086  Marital status −0.23 0.70 −.018  Stroke diagnosis −0.80 0.96 −.043  Berg Balance Scale −0.01 0.04 −.040  Supine to Sit 0.06 0.43 .013  Sit to Stand 0.16 0.53 .038  Ambulation (ft) 0.00 0.00 .076 Step 3  Constant 32.75 4.84  Age −0.00 0.02 −.001  Sex −0.53 0.63 −.040  Race −0.78 0.71 −.057  NIHSS 0.03 0.10 .019  Education 0.13 0.16 .079  Marital status 0.05 0.70 .004  Stroke diagnosis −0.35 0.96 −.019  Berg Balance Scale −0.02 0.04 −.042  Supine to Sit −0.14 0.41 −.031  Sit to Stand 0.15 0.51 .036  Ambulation (ft) 0.00 0.00 .068  Short Blessed Test −0.36*** 0.09 −.325  Trails A −0.00 0.01 −.041  Trails B 0.00 0.01 .132  Cancellation −0.03 0.11 −.075  Boston Naming Test −0.16 0.15 −.075 Predictors B SE B β Step 1  Constant 30.36 3.45  Age −0.01 0.02 −.023  Sex −0.54 0.63 −.036  Race −0.87 0.70 −.063  NIHSS −0.06 0.08 −.035  Education 0.20 0.16 .084  Marital status −0.37 0.68 −.027  Stroke diagnosis −0.95 0.89 −.052 Step 2  Constant 28.41 4.63  Age −0.01 0.02 −.016  Sex −0.50 0.65 −.031  Race −0.93 0.70 −.067  NIHSS −0.01 0.11 −.007  Education 0.21 0.17 .086  Marital status −0.23 0.70 −.018  Stroke diagnosis −0.80 0.96 −.043  Berg Balance Scale −0.01 0.04 −.040  Supine to Sit 0.06 0.43 .013  Sit to Stand 0.16 0.53 .038  Ambulation (ft) 0.00 0.00 .076 Step 3  Constant 32.75 4.84  Age −0.00 0.02 −.001  Sex −0.53 0.63 −.040  Race −0.78 0.71 −.057  NIHSS 0.03 0.10 .019  Education 0.13 0.16 .079  Marital status 0.05 0.70 .004  Stroke diagnosis −0.35 0.96 −.019  Berg Balance Scale −0.02 0.04 −.042  Supine to Sit −0.14 0.41 −.031  Sit to Stand 0.15 0.51 .036  Ambulation (ft) 0.00 0.00 .068  Short Blessed Test −0.36*** 0.09 −.325  Trails A −0.00 0.01 −.041  Trails B 0.00 0.01 .132  Cancellation −0.03 0.11 −.075  Boston Naming Test −0.16 0.15 −.075 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 0.1%–3.2% for Step 1 (p’s = .04–.37), R2 change = 0.4%–0.8% for Step 2 (p’s = .07–.71), R2 change = 5.7%–7.9% for Step 3 (p’s < .01). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 4. Hierarchical multiple regression coefficients with memory and thinking as the outcome variable Predictors B SE B β Step 1  Constant 30.36 3.45  Age −0.01 0.02 −.023  Sex −0.54 0.63 −.036  Race −0.87 0.70 −.063  NIHSS −0.06 0.08 −.035  Education 0.20 0.16 .084  Marital status −0.37 0.68 −.027  Stroke diagnosis −0.95 0.89 −.052 Step 2  Constant 28.41 4.63  Age −0.01 0.02 −.016  Sex −0.50 0.65 −.031  Race −0.93 0.70 −.067  NIHSS −0.01 0.11 −.007  Education 0.21 0.17 .086  Marital status −0.23 0.70 −.018  Stroke diagnosis −0.80 0.96 −.043  Berg Balance Scale −0.01 0.04 −.040  Supine to Sit 0.06 0.43 .013  Sit to Stand 0.16 0.53 .038  Ambulation (ft) 0.00 0.00 .076 Step 3  Constant 32.75 4.84  Age −0.00 0.02 −.001  Sex −0.53 0.63 −.040  Race −0.78 0.71 −.057  NIHSS 0.03 0.10 .019  Education 0.13 0.16 .079  Marital status 0.05 0.70 .004  Stroke diagnosis −0.35 0.96 −.019  Berg Balance Scale −0.02 0.04 −.042  Supine to Sit −0.14 0.41 −.031  Sit to Stand 0.15 0.51 .036  Ambulation (ft) 0.00 0.00 .068  Short Blessed Test −0.36*** 0.09 −.325  Trails A −0.00 0.01 −.041  Trails B 0.00 0.01 .132  Cancellation −0.03 0.11 −.075  Boston Naming Test −0.16 0.15 −.075 Predictors B SE B β Step 1  Constant 30.36 3.45  Age −0.01 0.02 −.023  Sex −0.54 0.63 −.036  Race −0.87 0.70 −.063  NIHSS −0.06 0.08 −.035  Education 0.20 0.16 .084  Marital status −0.37 0.68 −.027  Stroke diagnosis −0.95 0.89 −.052 Step 2  Constant 28.41 4.63  Age −0.01 0.02 −.016  Sex −0.50 0.65 −.031  Race −0.93 0.70 −.067  NIHSS −0.01 0.11 −.007  Education 0.21 0.17 .086  Marital status −0.23 0.70 −.018  Stroke diagnosis −0.80 0.96 −.043  Berg Balance Scale −0.01 0.04 −.040  Supine to Sit 0.06 0.43 .013  Sit to Stand 0.16 0.53 .038  Ambulation (ft) 0.00 0.00 .076 Step 3  Constant 32.75 4.84  Age −0.00 0.02 −.001  Sex −0.53 0.63 −.040  Race −0.78 0.71 −.057  NIHSS 0.03 0.10 .019  Education 0.13 0.16 .079  Marital status 0.05 0.70 .004  Stroke diagnosis −0.35 0.96 −.019  Berg Balance Scale −0.02 0.04 −.042  Supine to Sit −0.14 0.41 −.031  Sit to Stand 0.15 0.51 .036  Ambulation (ft) 0.00 0.00 .068  Short Blessed Test −0.36*** 0.09 −.325  Trails A −0.00 0.01 −.041  Trails B 0.00 0.01 .132  Cancellation −0.03 0.11 −.075  Boston Naming Test −0.16 0.15 −.075 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 0.1%–3.2% for Step 1 (p’s = .04–.37), R2 change = 0.4%–0.8% for Step 2 (p’s = .07–.71), R2 change = 5.7%–7.9% for Step 3 (p’s < .01). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 5. Hierarchical multiple regression coefficients with communication as the outcome variable Predictors B SE B β Step 1  Constant 33.47 2.99  Age −0.012 0.02 −.046  Sex −0.04 0.52 −.004  Race −0.71 0.57 −.063  NIHSS −0.13* 0.07 −.092  Education 0.09 0.16 .043  Marital status −0.40 0.56 −.037  Stroke diagnosis −0.62 0.78 −.041 Step 2  Constant 31.45 3.92  Age −0.02 0.02 −.040  Sex 0.02 0.53 .002  Race −0.77 0.57 −.068  NIHSS −0.11 0.09 −.073  Education 0.09 0.16 .045  Marital status −0.36 0.57 −.033  Stroke diagnosis −0.55 0.84 −.037  Berg Balance Scale −0.04 0.03 −.132  Supine to Sit 0.32 0.34 .124  Sit to Stand 0.11 0.44 .031  Ambulation (ft) 0.00 0.00 .087 Step 3  Constant 34.07 4.23  Age −0.00 0.02 −.012  Sex −0.07 0.51 −.006  Race −0.54 0.57 −.047  NIHSS −0.06 0.09 −.044  Education 0.02 0.16 .009  Marital status −0.10 0.58 −.009  Stroke diagnosis −0.230 0.83 −.015  Berg Balance Scale −0.04 0.03 −.155  Supine to Sit 0.11 0.32 .081  Sit to Stand 0.17 0.41 .049  Ambulation(ft) 0.00 0.00 .079  Short Blessed Test −0.29*** 0.06 −.318  Trails A 0.00 0.01 .014  Trails B −0.00 0.01 −.010  Cancellation −0.03 0.10 −.020  Boston Naming Test −0.06 0.11 −.035 Predictors B SE B β Step 1  Constant 33.47 2.99  Age −0.012 0.02 −.046  Sex −0.04 0.52 −.004  Race −0.71 0.57 −.063  NIHSS −0.13* 0.07 −.092  Education 0.09 0.16 .043  Marital status −0.40 0.56 −.037  Stroke diagnosis −0.62 0.78 −.041 Step 2  Constant 31.45 3.92  Age −0.02 0.02 −.040  Sex 0.02 0.53 .002  Race −0.77 0.57 −.068  NIHSS −0.11 0.09 −.073  Education 0.09 0.16 .045  Marital status −0.36 0.57 −.033  Stroke diagnosis −0.55 0.84 −.037  Berg Balance Scale −0.04 0.03 −.132  Supine to Sit 0.32 0.34 .124  Sit to Stand 0.11 0.44 .031  Ambulation (ft) 0.00 0.00 .087 Step 3  Constant 34.07 4.23  Age −0.00 0.02 −.012  Sex −0.07 0.51 −.006  Race −0.54 0.57 −.047  NIHSS −0.06 0.09 −.044  Education 0.02 0.16 .009  Marital status −0.10 0.58 −.009  Stroke diagnosis −0.230 0.83 −.015  Berg Balance Scale −0.04 0.03 −.155  Supine to Sit 0.11 0.32 .081  Sit to Stand 0.17 0.41 .049  Ambulation(ft) 0.00 0.00 .079  Short Blessed Test −0.29*** 0.06 −.318  Trails A 0.00 0.01 .014  Trails B −0.00 0.01 −.010  Cancellation −0.03 0.10 −.020  Boston Naming Test −0.06 0.11 −.035 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 1.8%–3.2% for Step 1 (p’s = .04–.28), R2 change = 0.7%–1.1% for Step 2 (p’s = .06–.38), R2 change = 6.6%–8.7% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 5. Hierarchical multiple regression coefficients with communication as the outcome variable Predictors B SE B β Step 1  Constant 33.47 2.99  Age −0.012 0.02 −.046  Sex −0.04 0.52 −.004  Race −0.71 0.57 −.063  NIHSS −0.13* 0.07 −.092  Education 0.09 0.16 .043  Marital status −0.40 0.56 −.037  Stroke diagnosis −0.62 0.78 −.041 Step 2  Constant 31.45 3.92  Age −0.02 0.02 −.040  Sex 0.02 0.53 .002  Race −0.77 0.57 −.068  NIHSS −0.11 0.09 −.073  Education 0.09 0.16 .045  Marital status −0.36 0.57 −.033  Stroke diagnosis −0.55 0.84 −.037  Berg Balance Scale −0.04 0.03 −.132  Supine to Sit 0.32 0.34 .124  Sit to Stand 0.11 0.44 .031  Ambulation (ft) 0.00 0.00 .087 Step 3  Constant 34.07 4.23  Age −0.00 0.02 −.012  Sex −0.07 0.51 −.006  Race −0.54 0.57 −.047  NIHSS −0.06 0.09 −.044  Education 0.02 0.16 .009  Marital status −0.10 0.58 −.009  Stroke diagnosis −0.230 0.83 −.015  Berg Balance Scale −0.04 0.03 −.155  Supine to Sit 0.11 0.32 .081  Sit to Stand 0.17 0.41 .049  Ambulation(ft) 0.00 0.00 .079  Short Blessed Test −0.29*** 0.06 −.318  Trails A 0.00 0.01 .014  Trails B −0.00 0.01 −.010  Cancellation −0.03 0.10 −.020  Boston Naming Test −0.06 0.11 −.035 Predictors B SE B β Step 1  Constant 33.47 2.99  Age −0.012 0.02 −.046  Sex −0.04 0.52 −.004  Race −0.71 0.57 −.063  NIHSS −0.13* 0.07 −.092  Education 0.09 0.16 .043  Marital status −0.40 0.56 −.037  Stroke diagnosis −0.62 0.78 −.041 Step 2  Constant 31.45 3.92  Age −0.02 0.02 −.040  Sex 0.02 0.53 .002  Race −0.77 0.57 −.068  NIHSS −0.11 0.09 −.073  Education 0.09 0.16 .045  Marital status −0.36 0.57 −.033  Stroke diagnosis −0.55 0.84 −.037  Berg Balance Scale −0.04 0.03 −.132  Supine to Sit 0.32 0.34 .124  Sit to Stand 0.11 0.44 .031  Ambulation (ft) 0.00 0.00 .087 Step 3  Constant 34.07 4.23  Age −0.00 0.02 −.012  Sex −0.07 0.51 −.006  Race −0.54 0.57 −.047  NIHSS −0.06 0.09 −.044  Education 0.02 0.16 .009  Marital status −0.10 0.58 −.009  Stroke diagnosis −0.230 0.83 −.015  Berg Balance Scale −0.04 0.03 −.155  Supine to Sit 0.11 0.32 .081  Sit to Stand 0.17 0.41 .049  Ambulation(ft) 0.00 0.00 .079  Short Blessed Test −0.29*** 0.06 −.318  Trails A 0.00 0.01 .014  Trails B −0.00 0.01 −.010  Cancellation −0.03 0.10 −.020  Boston Naming Test −0.06 0.11 −.035 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 1.8%–3.2% for Step 1 (p’s = .04–.28), R2 change = 0.7%–1.1% for Step 2 (p’s = .06–.38), R2 change = 6.6%–8.7% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 6. Hierarchical multiple regression coefficients with activities of daily living/instrumental activities of daily living as the outcome variable Predictors B SE B β Step 1  Constant 55.61 4.50  Age −0.08** 0.03 −.119  Sex −2.43** 0.84 −.130  Race −1.97* 0.92 −.101  NIHSS −0.85*** 0.11 −.352  Education 0.18 0.23 .054  Marital status −0.69 0.91 −.037  Stroke diagnosis −1.17 1.32 −.045 Step 2  Constant 45.66 6.10  Age −0.06 0.03 −.089  Sex −2.06* 0.85 −.110  Race −2.24* 0.91 −.115  NIHSS −0.52*** 0.14 −.216  Education 0.21 0.22 .062  Marital status −0.13 0.92 −.006  Stroke diagnosis 0.11 1.29 .004  Berg Balance Scale 0.06 0.05 .134  Supine to Sit 0.98 0.66 .159  Sit to Stand −0.58 0.82 −.095  Ambulation (ft) 0.00 0.00 .094 Step 3  Constant 49.04 7.47  Age −0.05 0.03 −.084  Sex −2.08* 0.86 −.111  Race −2.17* 0.94 −.111  NIHSS −0.52*** 0.14 −.210  Education 0.15 0.26 .044  Marital status −0.10 0.96 −.003  Stroke diagnosis 0.30 1.45 .011  Berg Balance Scale 0.06 0.05 .122  Supine to Sit 0.88 0.68 .142  Sit to Stand −0.58 0.80 −.095  Ambulation (ft) 0.00 0.00 .087  Short Blessed Test −0.11 0.12 −.068  Trails A −0.01 0.02 −.066  Trails B 0.00 0.01 .025  Cancellation −0.05 0.18 −.018  Boston Naming Test −0.10 0.18 −.034 Predictors B SE B β Step 1  Constant 55.61 4.50  Age −0.08** 0.03 −.119  Sex −2.43** 0.84 −.130  Race −1.97* 0.92 −.101  NIHSS −0.85*** 0.11 −.352  Education 0.18 0.23 .054  Marital status −0.69 0.91 −.037  Stroke diagnosis −1.17 1.32 −.045 Step 2  Constant 45.66 6.10  Age −0.06 0.03 −.089  Sex −2.06* 0.85 −.110  Race −2.24* 0.91 −.115  NIHSS −0.52*** 0.14 −.216  Education 0.21 0.22 .062  Marital status −0.13 0.92 −.006  Stroke diagnosis 0.11 1.29 .004  Berg Balance Scale 0.06 0.05 .134  Supine to Sit 0.98 0.66 .159  Sit to Stand −0.58 0.82 −.095  Ambulation (ft) 0.00 0.00 .094 Step 3  Constant 49.04 7.47  Age −0.05 0.03 −.084  Sex −2.08* 0.86 −.111  Race −2.17* 0.94 −.111  NIHSS −0.52*** 0.14 −.210  Education 0.15 0.26 .044  Marital status −0.10 0.96 −.003  Stroke diagnosis 0.30 1.45 .011  Berg Balance Scale 0.06 0.05 .122  Supine to Sit 0.88 0.68 .142  Sit to Stand −0.58 0.80 −.095  Ambulation (ft) 0.00 0.00 .087  Short Blessed Test −0.11 0.12 −.068  Trails A −0.01 0.02 −.066  Trails B 0.00 0.01 .025  Cancellation −0.05 0.18 −.018  Boston Naming Test −0.10 0.18 −.034 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 17.8%–19.1% for Step 1 (p’s < .001), R2 change = 20.7%–5.5% for Step 2 (p’s < .001), R2 change = 0.6%–1.4% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 6. Hierarchical multiple regression coefficients with activities of daily living/instrumental activities of daily living as the outcome variable Predictors B SE B β Step 1  Constant 55.61 4.50  Age −0.08** 0.03 −.119  Sex −2.43** 0.84 −.130  Race −1.97* 0.92 −.101  NIHSS −0.85*** 0.11 −.352  Education 0.18 0.23 .054  Marital status −0.69 0.91 −.037  Stroke diagnosis −1.17 1.32 −.045 Step 2  Constant 45.66 6.10  Age −0.06 0.03 −.089  Sex −2.06* 0.85 −.110  Race −2.24* 0.91 −.115  NIHSS −0.52*** 0.14 −.216  Education 0.21 0.22 .062  Marital status −0.13 0.92 −.006  Stroke diagnosis 0.11 1.29 .004  Berg Balance Scale 0.06 0.05 .134  Supine to Sit 0.98 0.66 .159  Sit to Stand −0.58 0.82 −.095  Ambulation (ft) 0.00 0.00 .094 Step 3  Constant 49.04 7.47  Age −0.05 0.03 −.084  Sex −2.08* 0.86 −.111  Race −2.17* 0.94 −.111  NIHSS −0.52*** 0.14 −.210  Education 0.15 0.26 .044  Marital status −0.10 0.96 −.003  Stroke diagnosis 0.30 1.45 .011  Berg Balance Scale 0.06 0.05 .122  Supine to Sit 0.88 0.68 .142  Sit to Stand −0.58 0.80 −.095  Ambulation (ft) 0.00 0.00 .087  Short Blessed Test −0.11 0.12 −.068  Trails A −0.01 0.02 −.066  Trails B 0.00 0.01 .025  Cancellation −0.05 0.18 −.018  Boston Naming Test −0.10 0.18 −.034 Predictors B SE B β Step 1  Constant 55.61 4.50  Age −0.08** 0.03 −.119  Sex −2.43** 0.84 −.130  Race −1.97* 0.92 −.101  NIHSS −0.85*** 0.11 −.352  Education 0.18 0.23 .054  Marital status −0.69 0.91 −.037  Stroke diagnosis −1.17 1.32 −.045 Step 2  Constant 45.66 6.10  Age −0.06 0.03 −.089  Sex −2.06* 0.85 −.110  Race −2.24* 0.91 −.115  NIHSS −0.52*** 0.14 −.216  Education 0.21 0.22 .062  Marital status −0.13 0.92 −.006  Stroke diagnosis 0.11 1.29 .004  Berg Balance Scale 0.06 0.05 .134  Supine to Sit 0.98 0.66 .159  Sit to Stand −0.58 0.82 −.095  Ambulation (ft) 0.00 0.00 .094 Step 3  Constant 49.04 7.47  Age −0.05 0.03 −.084  Sex −2.08* 0.86 −.111  Race −2.17* 0.94 −.111  NIHSS −0.52*** 0.14 −.210  Education 0.15 0.26 .044  Marital status −0.10 0.96 −.003  Stroke diagnosis 0.30 1.45 .011  Berg Balance Scale 0.06 0.05 .122  Supine to Sit 0.88 0.68 .142  Sit to Stand −0.58 0.80 −.095  Ambulation (ft) 0.00 0.00 .087  Short Blessed Test −0.11 0.12 −.068  Trails A −0.01 0.02 −.066  Trails B 0.00 0.01 .025  Cancellation −0.05 0.18 −.018  Boston Naming Test −0.10 0.18 −.034 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 17.8%–19.1% for Step 1 (p’s < .001), R2 change = 20.7%–5.5% for Step 2 (p’s < .001), R2 change = 0.6%–1.4% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 7. Hierarchical multiple regression coefficients with participation as the outcome variable Predictors B SE B β Step 1  Constant 40.86 4.73  Age 0.019 0.03 .027  Sex −1.923* 0.89 −.100  Race −1.70 0.98 −.085  NIHSS −0.74*** 0.13 −.293  Education 0.04 0.23 .011  Marital status −1.96* 0.96 −.102  Stroke diagnosis −1.99 1.38 −.075 Step 2  Constant 34.30 6.53  Age 0.03 0.03 .049  Sex −1.63* 0.90 −.085  Race −1.93** 0.97 −.096  NIHSS −0.48 0.15 −.190  Education 0.06 0.24 .018  Marital status −1.53 0.99 −.079  Stroke diagnosis −0.93 1.36 −.035  Berg Balance Scale 0.10 0.06 .207  Supine to SIt 0.50 0.58 .078  Sit to Stand −0.59 0.78 −.096  Ambulation (ft) 0.00 0.00 −.003 Step 3  Constant 39.73 7.73  Age 0.03 0.03 .050  Sex −1.66 0.91 −.086  Race −1.86 1.01 −.093  NIHSS −0.47** 0.15 −.185  Education −0.002 0.26 −.002  Marital status −1.48 1.04 −.077  Stroke diagnosis −0.64 1.48 −.024  Berg Balance Scale 0.09 0.06 .198  Supine to Sit 0.36 0.58 .057  Sit to Stand −0.64 0.75 −.104  Ambulation (ft) 0.00 0.00 −.010  Short Blessed Test −0.19 0.13 −.192  Trails A −0.02 0.03 −.018  Trails B 0.01 0.01 .005  Cancellation −0.03 0.22 −.032  Boston Naming Test −0.20 0.19 −.200 Predictors B SE B β Step 1  Constant 40.86 4.73  Age 0.019 0.03 .027  Sex −1.923* 0.89 −.100  Race −1.70 0.98 −.085  NIHSS −0.74*** 0.13 −.293  Education 0.04 0.23 .011  Marital status −1.96* 0.96 −.102  Stroke diagnosis −1.99 1.38 −.075 Step 2  Constant 34.30 6.53  Age 0.03 0.03 .049  Sex −1.63* 0.90 −.085  Race −1.93** 0.97 −.096  NIHSS −0.48 0.15 −.190  Education 0.06 0.24 .018  Marital status −1.53 0.99 −.079  Stroke diagnosis −0.93 1.36 −.035  Berg Balance Scale 0.10 0.06 .207  Supine to SIt 0.50 0.58 .078  Sit to Stand −0.59 0.78 −.096  Ambulation (ft) 0.00 0.00 −.003 Step 3  Constant 39.73 7.73  Age 0.03 0.03 .050  Sex −1.66 0.91 −.086  Race −1.86 1.01 −.093  NIHSS −0.47** 0.15 −.185  Education −0.002 0.26 −.002  Marital status −1.48 1.04 −.077  Stroke diagnosis −0.64 1.48 −.024  Berg Balance Scale 0.09 0.06 .198  Supine to Sit 0.36 0.58 .057  Sit to Stand −0.64 0.75 −.104  Ambulation (ft) 0.00 0.00 −.010  Short Blessed Test −0.19 0.13 −.192  Trails A −0.02 0.03 −.018  Trails B 0.01 0.01 .005  Cancellation −0.03 0.22 −.032  Boston Naming Test −0.20 0.19 −.200 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 13.7%–14.6% for Step 1 (p’s < .001), R2 change = 1.8%–3.3% for Step 2 (p’s < .001), R2 change = 1.2%–2.4% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 7. Hierarchical multiple regression coefficients with participation as the outcome variable Predictors B SE B β Step 1  Constant 40.86 4.73  Age 0.019 0.03 .027  Sex −1.923* 0.89 −.100  Race −1.70 0.98 −.085  NIHSS −0.74*** 0.13 −.293  Education 0.04 0.23 .011  Marital status −1.96* 0.96 −.102  Stroke diagnosis −1.99 1.38 −.075 Step 2  Constant 34.30 6.53  Age 0.03 0.03 .049  Sex −1.63* 0.90 −.085  Race −1.93** 0.97 −.096  NIHSS −0.48 0.15 −.190  Education 0.06 0.24 .018  Marital status −1.53 0.99 −.079  Stroke diagnosis −0.93 1.36 −.035  Berg Balance Scale 0.10 0.06 .207  Supine to SIt 0.50 0.58 .078  Sit to Stand −0.59 0.78 −.096  Ambulation (ft) 0.00 0.00 −.003 Step 3  Constant 39.73 7.73  Age 0.03 0.03 .050  Sex −1.66 0.91 −.086  Race −1.86 1.01 −.093  NIHSS −0.47** 0.15 −.185  Education −0.002 0.26 −.002  Marital status −1.48 1.04 −.077  Stroke diagnosis −0.64 1.48 −.024  Berg Balance Scale 0.09 0.06 .198  Supine to Sit 0.36 0.58 .057  Sit to Stand −0.64 0.75 −.104  Ambulation (ft) 0.00 0.00 −.010  Short Blessed Test −0.19 0.13 −.192  Trails A −0.02 0.03 −.018  Trails B 0.01 0.01 .005  Cancellation −0.03 0.22 −.032  Boston Naming Test −0.20 0.19 −.200 Predictors B SE B β Step 1  Constant 40.86 4.73  Age 0.019 0.03 .027  Sex −1.923* 0.89 −.100  Race −1.70 0.98 −.085  NIHSS −0.74*** 0.13 −.293  Education 0.04 0.23 .011  Marital status −1.96* 0.96 −.102  Stroke diagnosis −1.99 1.38 −.075 Step 2  Constant 34.30 6.53  Age 0.03 0.03 .049  Sex −1.63* 0.90 −.085  Race −1.93** 0.97 −.096  NIHSS −0.48 0.15 −.190  Education 0.06 0.24 .018  Marital status −1.53 0.99 −.079  Stroke diagnosis −0.93 1.36 −.035  Berg Balance Scale 0.10 0.06 .207  Supine to SIt 0.50 0.58 .078  Sit to Stand −0.59 0.78 −.096  Ambulation (ft) 0.00 0.00 −.003 Step 3  Constant 39.73 7.73  Age 0.03 0.03 .050  Sex −1.66 0.91 −.086  Race −1.86 1.01 −.093  NIHSS −0.47** 0.15 −.185  Education −0.002 0.26 −.002  Marital status −1.48 1.04 −.077  Stroke diagnosis −0.64 1.48 −.024  Berg Balance Scale 0.09 0.06 .198  Supine to Sit 0.36 0.58 .057  Sit to Stand −0.64 0.75 −.104  Ambulation (ft) 0.00 0.00 −.010  Short Blessed Test −0.19 0.13 −.192  Trails A −0.02 0.03 −.018  Trails B 0.01 0.01 .005  Cancellation −0.03 0.22 −.032  Boston Naming Test −0.20 0.19 −.200 Note: NIHSS = National Institutes of Health Stroke Scale. R2 = 13.7%–14.6% for Step 1 (p’s < .001), R2 change = 1.8%–3.3% for Step 2 (p’s < .001), R2 change = 1.2%–2.4% for Step 3 (p’s < .001). Bold = statistically significant; *p < .05; **p < .01; ***p < .001. Table 8. Correlations among NIHSS, Short Blessed Test, and Stroke Impact Scale subscales Measure 1 2 3 4 5 6 1. NIHSS — 2. Short Blessed Test 0.23*** — 3. Memory and Thinking −0.05 −0.28*** — 4. Communication −0.11** −0.30*** 0.82*** — 5. ADL/IADLs −0.37*** −0.21*** 0.59*** 0.59*** — 6. Participation −0.32*** −0.20*** 0.57*** 0.57*** 0.79*** — Measure 1 2 3 4 5 6 1. NIHSS — 2. Short Blessed Test 0.23*** — 3. Memory and Thinking −0.05 −0.28*** — 4. Communication −0.11** −0.30*** 0.82*** — 5. ADL/IADLs −0.37*** −0.21*** 0.59*** 0.59*** — 6. Participation −0.32*** −0.20*** 0.57*** 0.57*** 0.79*** — Note: NIHSS = National Institutes of Health Stroke Scale; ADL/IADLs = activities of daily living/instrumental activities of daily living. *p < .05; **p < .01; ***p < .001. Table 8. Correlations among NIHSS, Short Blessed Test, and Stroke Impact Scale subscales Measure 1 2 3 4 5 6 1. NIHSS — 2. Short Blessed Test 0.23*** — 3. Memory and Thinking −0.05 −0.28*** — 4. Communication −0.11** −0.30*** 0.82*** — 5. ADL/IADLs −0.37*** −0.21*** 0.59*** 0.59*** — 6. Participation −0.32*** −0.20*** 0.57*** 0.57*** 0.79*** — Measure 1 2 3 4 5 6 1. NIHSS — 2. Short Blessed Test 0.23*** — 3. Memory and Thinking −0.05 −0.28*** — 4. Communication −0.11** −0.30*** 0.82*** — 5. ADL/IADLs −0.37*** −0.21*** 0.59*** 0.59*** — 6. Participation −0.32*** −0.20*** 0.57*** 0.57*** 0.79*** — Note: NIHSS = National Institutes of Health Stroke Scale; ADL/IADLs = activities of daily living/instrumental activities of daily living. *p < .05; **p < .01; ***p < .001. Table 9. Descriptives for Stroke Impact Scale subscales and physical/functional and cognitive variables Measure M SD Range Stroke Impact Scale subscales  Memory and Thinking 28.42 6.63 7–35  Communication 30.68 5.47 9–35  ADL/IADLs 41.26 9.37 10–50  Participation 29.41 9.63 8–40 Physical/functional variables  Berg Balance Scale 28.21 20.91 0–56  Supine to Sit 5.32 1.51 1–7  Sit to Stand 5.03 1.53 0–7  Ambulation (feet) 190.76 238.42 0–3000 Cognitive variables  Short Blessed Test 5.92 5.87 0–28  Trails A (seconds) 71.27 47.13 16–280  Trails B (seconds) 157.79 83.74 12–416  Cancellation 0.15 3.59 −13–13  Boston Naming Test 11.68 3.20 0–15 Measure M SD Range Stroke Impact Scale subscales  Memory and Thinking 28.42 6.63 7–35  Communication 30.68 5.47 9–35  ADL/IADLs 41.26 9.37 10–50  Participation 29.41 9.63 8–40 Physical/functional variables  Berg Balance Scale 28.21 20.91 0–56  Supine to Sit 5.32 1.51 1–7  Sit to Stand 5.03 1.53 0–7  Ambulation (feet) 190.76 238.42 0–3000 Cognitive variables  Short Blessed Test 5.92 5.87 0–28  Trails A (seconds) 71.27 47.13 16–280  Trails B (seconds) 157.79 83.74 12–416  Cancellation 0.15 3.59 −13–13  Boston Naming Test 11.68 3.20 0–15 Note: ADL/IADLs = activities of daily living/ instrumental activities of daily living. Table 9. Descriptives for Stroke Impact Scale subscales and physical/functional and cognitive variables Measure M SD Range Stroke Impact Scale subscales  Memory and Thinking 28.42 6.63 7–35  Communication 30.68 5.47 9–35  ADL/IADLs 41.26 9.37 10–50  Participation 29.41 9.63 8–40 Physical/functional variables  Berg Balance Scale 28.21 20.91 0–56  Supine to Sit 5.32 1.51 1–7  Sit to Stand 5.03 1.53 0–7  Ambulation (feet) 190.76 238.42 0–3000 Cognitive variables  Short Blessed Test 5.92 5.87 0–28  Trails A (seconds) 71.27 47.13 16–280  Trails B (seconds) 157.79 83.74 12–416  Cancellation 0.15 3.59 −13–13  Boston Naming Test 11.68 3.20 0–15 Measure M SD Range Stroke Impact Scale subscales  Memory and Thinking 28.42 6.63 7–35  Communication 30.68 5.47 9–35  ADL/IADLs 41.26 9.37 10–50  Participation 29.41 9.63 8–40 Physical/functional variables  Berg Balance Scale 28.21 20.91 0–56  Supine to Sit 5.32 1.51 1–7  Sit to Stand 5.03 1.53 0–7  Ambulation (feet) 190.76 238.42 0–3000 Cognitive variables  Short Blessed Test 5.92 5.87 0–28  Trails A (seconds) 71.27 47.13 16–280  Trails B (seconds) 157.79 83.74 12–416  Cancellation 0.15 3.59 −13–13  Boston Naming Test 11.68 3.20 0–15 Note: ADL/IADLs = activities of daily living/ instrumental activities of daily living. Discussion The current study examined acute predictors of cognitive and functional outcomes 6 months after mild to moderate stroke. Overall, the combination of demographic factors and cognitive, physical, and functional status at stroke onset accounted for small amounts of variance in cognitive outcomes and moderate degrees of variance in functional outcomes post-stroke. Furthermore, inclusion of physical/functional variables in the second step of the regression models resulted in better prediction of ADL/IADLs and Participation, but not Communication or Memory and Thinking outcomes. Conversely, adding cognitive variables in the final step of the models improved prediction of Memory and Thinking and Communication outcomes, but not ADL/IADLs or Participation. Finally, in terms of individual predictors, brief screening instruments (i.e., the NIHSS and SBT) exhibited consistent predictive utility, while more domain-specific cognitive tests (e.g., the BNT and Trails A and B) did not. Specifically, the SBT emerged as a significant predictor of the Communication and Memory and Thinking subscales, such that greater impairment on the SBT was associated with worse self-reported communication and cognitive ability 6 months after stroke. Moreover, although none of the cognitive measures included in the present study predicted ADL/IADLs or Participation, the NIHSS Total score was a significant predictor of subsequent functional outcomes. Results of the regression analyses revealed that greater stroke severity was associated with worse self-reported ADL/IADLs and reduced engagement in meaningful activities 6 months after stroke. Consistent with previous studies, being female and/or a racial minority was also associated with decreased independence in ADL/IADLs and participation post-stroke (Cioncoloni et al., 2013; Duarte et al., 2010; Horner, Swanson, Bosworth, & Matchar, 2003). The unfortunate reality is that inequalities in access to care and rehabilitation services exist across gender and racial groups, likely contributing to poorer outcomes for societally disadvantaged individuals (Busch, Coshall, Heuschmann, McKevitt, & Wolfe, 2009). Although not surprising, these findings support the importance of designing and implementing unbiased rehabilitation programs, in order to maximize recovery for individuals across a range of diversity characteristics. Overall, our results support the clinical utility of administering brief, broad screening instruments during acute recovery from mild to moderate stroke. The SBT and NIHSS significantly predicted 6-month cognitive and functional outcomes, and demonstrated superior predictive validity relative to measures assessing specific functional abilities (e.g., Sit to Stand) or cognitive domains (e.g., executive functioning). These findings are consistent with prior studies showing that, relative to more domain-specific cognitive measures, broad-based screening instruments tend to better predict post-stroke outcomes when administered acutely in real-world clinical scenarios (Horstmann, Rizos, Rauch, Arden, & Veltkamp, 2014; Riepe et al., 2004). Researchers have attributed this to the heterogeneity of acute stroke-related cognitive deficits, as well as confounding issues such as delirium, confusion, fatigue, and pre-stroke cognitive decline (Cumming et al., 2013; Lees et al., 2014; Salvadori et al., 2013). Given these complicating factors, brief screening instruments like the SBT and NIHSS that can be readily completed by patients shortly after stroke may be more appropriate for assessing patients in acute care settings than more domain-specific measures. In addition to providing evidence for the clinical utility of brief, broad screening instruments generally, our findings also suggest that measures assessing acute cognitive dysfunction and those examining general neurologic status may differentially predict 6-month cognitive and functional outcomes after stroke. The SBT, which is a screener for global cognitive dysfunction, predicted impairments in cognition and communication abilities, while the NIHSS, which is a broader assessment of neurologic status, predicted recovery of ADL/IADLs and reengagement in meaningful activity (Carpenter et al., 2011; Goldstein et al., 1989). This indicates that brief assessment of both acute neurologic status and global cognitive dysfunction is important for predicting outcomes 6 months after stroke. Specifically, conducting a brief cognitive screen may improve prediction of persistent cognitive deficits, whereas evaluating acute neurologic status may inform expectations for recovery of ADL/IADLs and reengagement in meaningful life activities. Of note, while the SBT has been previously shown to predict dementia diagnosis, neuropathology, and mortality (Bellelli et al., 2015; Marengoni et al., 2013; Katzman et al., 1983; Wilkins et al., 2007), to our knowledge this is the first study to demonstrate that the SBT is a valid predictor of cognitive outcomes following stroke. A probable explanation for the prognostic value of measures assessing acute neurologic status and global cognitive dysfunction is the persistence of certain stroke-related deficits (Cioncoloni et al., 2013; Del Sur et al., 2005; Hoffman et al., 2003; Wolf & Rognstad, 2013). However, even thorough predictive models typically account for no more than moderate amounts of variance in cognitive and functional outcomes due to the heterogeneity of recovery trajectories and outcomes for mild to moderate stroke (Jørgensen et al., 1995; Nichols-Larsen, Clark, Zeringue, Greenspan, & Blanton, 2005; Tilling et al., 2001). Age and other factors contribute to functional adaptation and cognitive plasticity such that in many cases, deficits resolve naturally, with compensation, and/or as a result of direct intervention (Ferrucci et al., 1993; Kleim & Jones, 2008; Miller et al., 2010; Veerbeek et al., 2014). There is strong evidence that rehabilitation services initiated at admission and sustained throughout recovery significantly reduce the likelihood of death and disability after stroke (Maulden, Gassaway, Horn, Smout, & DeJong, 2005; Miller et al., 2010). In particular, interdisciplinary interventions (i.e., physical therapy, occupational therapy, speech-language therapy, etc.) focused on providing intensive, highly repetitive, task-oriented, and task-specific trainings tailored to each phase of recovery have been well-supported (Kalra & Langhorne, 2007; Miller et al., 2010). There is also some evidence to support the effectiveness of cognitive rehabilitation for some cognitive impairments, though more research on the efficacy and effectiveness of cognitive intervention is needed (Gillespie et al., 2015). Acute assessment of cognitive and functional impairments is necessary to determine rehabilitation needs and plan appropriate services for patients recovering from stroke (Lawrence et al., 2001). Because early initiation and maintenance of targeted rehabilitation services is associated with better recovery, identifying acute factors that meaningfully predict long-term outcomes is an important goal for research and clinical care (Bhogal et al., 2003; Maulden et al., 2005). The ability to account for even small amounts of variance in stroke outcomes is of significant clinical utility, given that this information can be used to inform delivery of rehabilitation services. For example, memory problems are a common complaint following stroke. Identification of patients with acute cognitive dysfunction, combined with the knowledge that these early deficits predict long-term cognitive outcomes, can meaningfully inform treatment planning for survivors of mild to moderate stroke. In such cases, early cognitive skills training might aide in the recovery of memory functions or more frequently, enhance the individual’s ability to adapt to or compensate for their deficits. Therefore, improving predictive models for long-term cognitive and functional outcomes can help providers make informed decisions regarding appropriate rehabilitation goals and strategies, leading to better outcomes for survivors of mild to moderate stroke. Limitations and Future Directions Although of significant clinical utility, the findings from the current study should be interpreted in the context of several limitations. First, we did not have access to data pertaining to certain stroke-related characteristics (e.g., region of stroke, history of transient ischemic attacks, etc.) or relevant psychological factors (e.g., psychiatric history, acute depressive symptoms), thus limiting our ability to fully characterize our sample. While this is not ideal, previous investigations of the BRC (e.g., Aufman et al., 2013; Bland, Sturmoski, Whitson, Connor, & Fucetola, 2012; Merz, Van Patten, Mulhauser, & Fucetola, 2017; Van Patten et al., 2016) have faced similar constraints. Broadly, because our sample consisted of 498 consecutive admissions to a large acute care center over a five-year period (minus the exclusions specified in the Methods), we believe that it is a representative sample of mild to moderate stroke that is readily generalizable to patients evaluated in regular neuropsychological practice. Moreover, overall stroke severity, which was assessed in the current study, is likely the most important stroke-related characteristic in terms of predicting functional outcomes. For example, stroke severity was recently demonstrated to reliably predict return to work in stroke patients, whereas other acute biological factors such as location and type of stroke, as well as psychosocial variables such as educational attainment and marital status, did not (Wang, Kapellusch, & Garg, 2014). Additionally, although we were unable to account for affective factors (e.g., depression) that may have affected patients’ performance during acute inpatient testing, the prevalence of depression in our sample at 6-month follow-up is comparable to prior studies (Hackett et al., 2005). Consequently, our inability to report all relevant stroke-related and psychological characteristics of our sample does not render the findings ungeneralizable. Relatedly, although we were unable to incorporate details about rehabilitation participation in our predictive models, this study included a representative sample of stroke patients receiving standard clinical care. Importantly, as part of usual care, performance on acute assessment measures was used to inform recommendations for subsequent rehabilitation services. Therefore, our inability to account for rehabilitation-related factors does not negate our findings pertaining to the clinical utility of acute assessment measures. Second, as part of their clinical assessments, BRC patients complete only a small set of neuropsychological tasks which are administered by occupational therapists rather than neuropsychological assistants and do not capture all relevant cognitive domains. Consequently, we were unable to generate hypotheses with respect to certain aspects of cognitive functioning (e.g., visuospatial skills) and our measurement of other areas (e.g., executive functioning) was only cursory (i.e., Trails B). Involvement of neuropsychologists, particularly during inpatient rehabilitation or other post-acute care, would have allowed for better characterization of patients’ cognitive status post-stroke. Third, our follow-up assessment was restricted to a single time point approximately 6 months post-stroke and we were unable to conduct serial follow-up evaluations. Fourth, our sample was based exclusively out of the BRC database, which represents a geographically limited set of individuals who reside in the Midwestern region of the U.S. Generalizations outside of the greater population of individuals from this area are unwarranted. Finally, our outcome measure (the SIS) reflects an individual’s self-reported level of functioning and although we maximized the accuracy of these reports by excluding individuals with severe strokes, aphasia, and dysarthria, self-report instruments are inherently vulnerable to conscious and unconscious bias (Stone, Bachrach, Jobe, Kurtzman, & Cain, 1999). Future studies should further the current line of inquiry by collecting information from collateral sources during follow-up, as poor insight and/or emotional difficulties associated with stroke may affect the validity of stroke survivor’s self-report data. In addition, investigators should conduct serial assessments that include longer follow-up intervals (e.g., 6, 12, 24, and 36 months post-stroke) in order to better elucidate the predictive capacity of acute physical and cognitive test data across recovery and reintegration into the community. It would also be important to examine and control for relevant rehabilitation-related factors (including specific therapies received, frequency and duration of therapy, treatment adherence, and response to intervention) in order to better understand the relationship between acute performance and outcomes following stroke. Furthermore, investigators should build on this and related studies to determine whether brief screening measures of common stroke-related symptoms including, but not limited to, the NIHSS and SBT predict ecologically relevant post-stroke outcomes (e.g., capacity to drive, ability to successfully engage in relevant therapies). Finally, researchers should capitalize on the demonstrated utility of these broad, brief screening instruments in other acquired brain injury syndromes such as moderate to severe traumatic brain injury and hypoxia. It is likely that, similar to the current findings in stroke, measures such as the SBT and the Montreal Cognitive Assessment (MoCA; Nasreddine et al., 2005) will demonstrate incremental predictive validity with respect to relevant future outcomes in these related populations as well. Conflict of Interest None declared. Acknowledgments We thank the members of the Brain Recovery Core team from Washington University School of Medicine, Barnes-Jewish Hospital, and the Rehabilitation Institute of Saint Louis for the support and sharing of data for the purposes of this project. References Adamit , T. , Maeir , A. , Ben Assayag , E. , Bornstein , N. M. , Korczyn , A. D. , & Katz , N. ( 2015 ). Impact of first-ever mild stroke on participation at 3 and 6 month post-event: the TABASCO study . Disability and Rehabilitation , 37 , 667 – 673 . Google Scholar CrossRef Search ADS PubMed Almkvist Muren , M. , Hütler , M. , & Hooper , J. ( 2008 ). Functional capacity and health-related quality of life in individuals post stroke . Topics in Stroke Rehabilitation , 15 , 51 – 58 . Google Scholar CrossRef Search ADS PubMed Aufman , E. L. , Bland , M. D. , Barco , P. P. , Carr , D. B. , & Lang , C. E. ( 2013 ). Predictors of return to driving after stroke . American Journal of Physical Medicine & Rehabilitation , 92 , 627 – 634 . Google Scholar CrossRef Search ADS PubMed Babulal , G. M. , Huskey , T. N. , Roe , C. M. , Goette , S. A. , & Connor , L. T. ( 2015 ). Cognitive impairments and mood disruptions negatively impact instrumental activities of daily living performance in the first three months after a first stroke . Topics in Stroke Rehabilitation , 22 , 144 – 151 . Google Scholar CrossRef Search ADS PubMed Bellelli , G. , Nobili , A. , Annoni , G. , Morandi , A. , Djade , C. D. , Meagher , D. J. , et al. . ( 2015 ). Under-detection of delirium and impact of neurocognitive deficits on in-hospital mortality among acute geriatric and medical wards . European Journal of Internal Medicine , 26 , 696 – 704 . Google Scholar CrossRef Search ADS PubMed Berg , K. , Wood-Dauphinee , S. , & Williams , J. ( 1995 ). The Balance Scale: Reliability assessment with elderly residents and patients with an acute stroke . Scandinavian Journal of Rehabilitation Medicine , 27 , 27 – 36 . Google Scholar PubMed Bhogal , S. K. , Teasell , R. W. , Foley , N. C. , & Speechley , M. R. ( 2003 ). Community reintegration after stroke . Topics in Stroke Rehabilitation , 10 , 107 – 129 . Google Scholar CrossRef Search ADS PubMed Bland , M. D. , Sturmoski , A. , Whitson , M. , Connor , L. T. , Fucetola , R. , Huskey , T. , et al. . ( 2012 ). Prediction of discharge walking ability from initial assessment in a stroke inpatient rehabilitation facility population . Archives of Physical Medicine and Rehabilitation , 93 , 1441 – 1447 . Google Scholar CrossRef Search ADS PubMed Busch , M. A. , Coshall , C. , Heuschmann , P. U. , McKevitt , C. , & Wolfe , C. D. ( 2009 ). Sociodemographic differences in return to work after stroke: the South London Stroke Register (SLSR) . Journal of Neurology, Neurosurgery & Psychiatry , 80 ( 8 ), 888 – 893 . Google Scholar CrossRef Search ADS Carpenter , C. R. , Bassett , E. R. , Fischer , G. M. , Shirshekan , J. , Galvin , J. E. , & Morris , J. C. ( 2011 ). Four sensitive screening tools to detect cognitive dysfunction in geriatric emergency department patients: Brief Alzheimer’s Screen, Short Blessed Test, Ottawa 3DY, and the caregiver-completed AD8 . Academic Emergency Medicine: Official Journal of the Society for Academic Emergency Medicine , 18 , 374 – 384 . Google Scholar CrossRef Search ADS PubMed Cheng , P. T. , Liaw , M. Y. , Wong , M. K. , Tang , F. T. , Lee , M. Y. , & Lin , P. S. ( 1998 ). The sit-to-stand movement in stroke patients and its correlation with falling . Archives of Physical Medicine and Rehabilitation , 79 , 1043 – 1046 . Google Scholar CrossRef Search ADS PubMed Ch’Ng , A. M. , French , D. , & Mclean , N. ( 2008 ). Coping with the challenges of recovery from stroke long term perspectives of stroke support group members . Journal of Health Psychology , 13 , 1136 – 1146 . Google Scholar CrossRef Search ADS PubMed Cioncoloni , D. , Martini , G. , Piu , P. , Taddei , S. , Acampa , M. , Guideri , F. , et al. . ( 2013 ). Predictors of long-term recovery in complex activities of daily living before discharge from the stroke unit . NeuroRehabilitation , 33 , 217 – 223 . Google Scholar PubMed Collins , L. M. , Schafer , J. L. , & Kam , C. M. ( 2001 ). A comparison of inclusive and restrictive strategies in modern missing data procedures . Psychological Methods , 6 , 330 . Google Scholar CrossRef Search ADS PubMed Cumming , T. B. , Marshall , R. S. , & Lazar , R. M. ( 2013 ). Stroke, cognitive deficits, and rehabilitation: Still an incomplete picture . International Journal of Stroke , 8 , 38 – 45 . Google Scholar CrossRef Search ADS PubMed Del Ser , T. , Barba , R. , Morin , M. M. , Domingo , J. , Cemillan , C. , Pondal , M. , et al. . ( 2005 ). Evolution of cognitive impairment after stroke and risk factors for delayed progression . Stroke; A Journal of Cerebral Circulation , 36 , 2670 – 2675 . Google Scholar CrossRef Search ADS Desrosiers , J. , Noreau , L. , Rochette , A. , Bourbonnais , D. , Bravo , G. , & Bourget , A. ( 2006 ). Predictors of long-term participation after stroke . Disability and Rehabilitation , 28 , 221 – 230 . Google Scholar CrossRef Search ADS PubMed Douiri , A. , Rudd , A. G. , & Wolfe , C. D. ( 2013 ). Prevalence of poststroke cognitive impairment . Stroke; A Journal of Cerebral Circulation , 44 , 138 – 145 . Google Scholar CrossRef Search ADS Duarte , E. , Marco , E. , Muniesa , J. M. , Belmonte , R. , Aguilar , J. J. , & Escalada , F. ( 2010 ). Early detection of non-ambulatory survivors six months after stroke . NeuroRehabilitation , 26 , 317 – 323 . Google Scholar PubMed Edwards , D. F. , Hahn , M. , Baum , C. , & Dromerick , A. W. ( 2006 ). The impact of mild stroke on meaningful activity and life satisfaction . Journal of Stroke and Cerebrovascular Diseases , 15 , 151 – 157 . Google Scholar CrossRef Search ADS PubMed Ferrucci , L. , Bandinelli , S. , Guralnik , J. M. , Lamponi , M. , Bertini , C. , Falchini , M. , et al. . ( 1993 ). Recovery of functional status after stroke. A postrehabilitation follow-up study . Stroke; A Journal of Cerebral Circulation , 24 , 200 – 205 . Google Scholar CrossRef Search ADS Field , A. ( 2009 ). Discovering statistics using SPSS, 3rd edition . Thousand Oaks, CA : Sage Publications . Gillespie , D. C. , Bowen , A. , Chung , C. S. , Cockburn , J. , Knapp , P. , & Pollock , A. ( 2015 ). Rehabilitation for post-stroke cognitive impairment: An overview of recommendations arising from systematic reviews of current evidence . Clinical Rehabilitation , 29 , 120 – 128 . Google Scholar CrossRef Search ADS PubMed Goldstein , L. B. , Bertels , C. , & Davis , J. N. ( 1989 ). Interrater reliability of the NIH Stroke Scale . Archives of Neurology , 46 , 660 – 662 . Google Scholar CrossRef Search ADS PubMed Hackett , M. L. , Yapa , C. , Parag , V. , & Anderson , C. S. ( 2005 ). Frequency of depression after stroke . Stroke; A Journal of Cerebral Circulation , 36 , 1330 – 1340 . Google Scholar CrossRef Search ADS Harvey , R. L. ( 2015 ). Predictors of functional outcome following stroke . Physical Medicine and Rehabilitation Clinics of North America , 26 , 583 – 598 . Google Scholar CrossRef Search ADS PubMed Heruti , R. J. , Lusky , A. , Dankner , R. , Ring , H. , Dolgopiat , M. , Barell , V. , et al. . ( 2002 ). Rehabilitation outcome of elderly patients after a first stroke: Effect of cognitive status at admission on the functional outcome . Archives of Physical Medicine and Rehabilitation , 83 , 742 – 749 . Google Scholar CrossRef Search ADS PubMed Hillis , A. E. , Wityk , R. J. , Barker , P. B. , Ulatowski , J. A. , & Jacobs , M. A. ( 2003 ). Change in perfusion in acute nondominant hemisphere stroke may be better estimated by tests of hemispatial neglect than by the National Institutes of Health Stroke Scale . Stroke; A Journal of Cerebral Circulation , 34 , 2392 – 2396 . Google Scholar CrossRef Search ADS Hoffmann , T. , McKenna , K. , Cooke , D. , & Tooth , L. ( 2003 ). Outcomes after stroke: Basic and instrumental activities of daily living, community reintegration and generic health status . Australian Occupational Therapy Journal , 50 , 225 – 233 . Google Scholar CrossRef Search ADS Horner , R. D. , Swanson , J. W. , Bosworth , H. B. , & Matchar , D. B. ( 2003 ). Effects of race and poverty on the process and outcome of inpatient rehabilitation services among stroke patients . Stroke; A Journal of Cerebral Circulation , 34 , 1027 – 1031 . Google Scholar CrossRef Search ADS Horstmann , S. , Rizos , T. , Rauch , G. , Arden , C. , & Veltkamp , R. ( 2014 ). Feasibility of the Montreal Cognitive Assessment in acute stroke patients . European Journal of Neurology , 21 , 1387 – 1393 . Google Scholar CrossRef Search ADS PubMed Hoyle , M. , Gustafsson , L. , Meredith , P. , & Ownsworth , T. ( 2012 ). Participation after stroke: Do we understand all the components and relationships as categorised in the ICF? Brain Impairment , 13 , 4 – 15 . Google Scholar CrossRef Search ADS Jacquin , A. , Binquet , C. , Rouaud , O. , Graule-Petot , A. , Daubail , B. , Osseby , G. V. , et al. . ( 2014 ). Post-stroke cognitive impairment: high prevalence and determining factors in a cohort of mild stroke . Journal of Alzheimer’s Disease , 40 , 1029 – 1038 . Google Scholar CrossRef Search ADS PubMed Jette , A. M. , Keysor , J. , Coster , W. , Ni , P. , & Haley , S. ( 2005 ). Beyond function: Predicting participation in a rehabilitation cohort . Archives of Physical Medicine and Rehabilitation , 86 , 2087 – 2094 . Google Scholar CrossRef Search ADS PubMed Jokinen , H. , Melkas , S. , Ylikoski , R. , Pohjasvaara , T. , Kaste , M. , Erkinjuntti , T. , et al. . ( 2015 ). Post‐stroke cognitive impairment is common even after successful clinical recovery . European Journal of Neurology , 22 , 1288 – 1294 . Google Scholar CrossRef Search ADS PubMed Jørgensen , H. S. , Nakayama , H. , Raaschou , H. O. , Vive-Larsen , J. , Støier , M. , & Olsen , T. S. ( 1995 ). Outcome and time course of recovery in stroke. Part I: Outcome. The Copenhagen Stroke Study . Archives of Physical Medicine and Rehabilitation , 76 , 399 – 405 . Google Scholar CrossRef Search ADS PubMed Kalra , L. , & Langhorne , P. ( 2007 ). Facilitating recovery: Evidence for organized stroke care . Journal of Rehabilitation Medicine , 39 , 97 – 102 . Google Scholar CrossRef Search ADS PubMed Kalra , L. , Perez , I. , Gupta , S. , & Wittink , M. ( 1997 ). The influence of visual neglect on stroke rehabilitation . Stroke; A Journal of Cerebral Circulation , 28 , 1386 – 1391 . Google Scholar CrossRef Search ADS Katzman , R. , Brown , T. , Fuld , P. , Peck , A. , Schechter , R. , & Schimmel , H. ( 1983 ). Validation of a short Orientation-Memory-Concentration Test of cognitive impairment . The American Journal of Psychiatry , 140 , 734 – 739 . Google Scholar CrossRef Search ADS PubMed Kleim , J. A. , & Jones , T. A. ( 2008 ). Principles of experience-dependent neural plasticity: Implications for rehabilitation after brain damage . Journal of Speech, Language, and Hearing Research , 51 , S225 – S239 . Google Scholar CrossRef Search ADS Lang , C. E. , Bland , M. D. , Connor , L. T. , Fucetola , R. , Whitson , M. , Edmiaston , J. , et al. . ( 2011 ). The brain recovery core: Building a system of organized stroke rehabilitation and outcomes assessment across the continuum of care . Journal of Neurologic Physical Therapy , 35 , 194 – 201 . Google Scholar CrossRef Search ADS PubMed Laska , A. C. , Hellblom , A. , Murray , V. , Kahan , T. , & Von Arbin , M. ( 2001 ). Aphasia in acute stroke and relation to outcome . Journal of Internal Medicine , 249 , 413 – 422 . Google Scholar CrossRef Search ADS PubMed Lawrence , E. S. , Coshall , C. , Dundas , R. , Stewart , J. , Rudd , A. G. , Howard , R. , et al. . ( 2001 ). Estimates of the prevalence of acute stroke impairments and disability in a multiethnic population . Stroke; A Journal of Cerebral Circulation , 32 , 1279 – 1284 . Google Scholar CrossRef Search ADS Lazar , R. M. , Minzer , B. , Antoniello , D. , Festa , J. R. , Krakauer , J. W. , & Marshall , R. S. ( 2010 ). Improvement in aphasia scores after stroke is well predicted by initial severity . Stroke; A Journal of Cerebral Circulation , 41 , 1485 – 1488 . Google Scholar CrossRef Search ADS Lees , R. , Selvarajah , J. , Fenton , C. , Pendlebury , S. T. , Langhorne , P. , Stott , D. J. , et al. . ( 2014 ). Test accuracy of cognitive screening tests for diagnosis of dementia and multidomain cognitive impairment in stroke . Stroke; A Journal of Cerebral Circulation , 45 , 3008 – 3018 . Google Scholar CrossRef Search ADS Leśniak , M. , Bak , T. , Czepiel , W. , Seniów , J. , & Członkowska , A. ( 2008 ). Frequency and prognostic value of cognitive disorders in stroke patients . Dementia and Geriatric Cognitive Disorders , 26 , 356 – 363 . Google Scholar CrossRef Search ADS PubMed Little , R. J. ( 1988 ). A test of missing completely at random for multivariate data with missing values . Journal of the American Statistical Association , 83 , 1198 – 1202 . Google Scholar CrossRef Search ADS Loewen , S. C. , & Anderson , B. A. ( 1990 ). Predictors of stroke outcome using objective measurement scales . Stroke; A Journal of Cerebral Circulation , 21 , 78 – 81 . Google Scholar CrossRef Search ADS Lomaglio , M. J. , & Eng , J. J. ( 2005 ). Muscle strength and weight-bearing symmetry relate to sit-to-stand performance in individuals with stroke . Gait & Posture , 22 , 126 – 131 . Google Scholar CrossRef Search ADS PubMed Makin , S. D. J. , Turpin , S. , Dennis , M. S. , & Wardlaw , J. M. ( 2013 ). Cognitive impairment after lacunar stroke: Systematic review and meta-analysis of incidence, prevalence and comparison with other stroke subtypes . Journal of Neurology, Neurosurgery & Psychiatry , 84 , 893 – 900 . Google Scholar CrossRef Search ADS Marengoni , A. , Nobili , A. , Romano , V. , Tettamanti , M. , Pasina , L. , Djade , S. , et al. . ( 2013 ). Adverse clinical events and mortality during hospitalization and 3 months after discharge in cognitively impaired elderly patients . The Journals of Gerontology Series A: Biological Sciences and Medical Sciences , 68 , 419 – 425 . Google Scholar CrossRef Search ADS Maulden , S. A. , Gassaway , J. , Horn , S. D. , Smout , R. J. , & DeJong , G. ( 2005 ). Timing of initiation of rehabilitation after stroke . Archives of Physical Medicine and Rehabilitation , 86 , 34 – 40 . Google Scholar CrossRef Search ADS Mayo , N. E. , Anderson , S. , Barclay , R. , Cameron , J. I. , Desrosiers , J. , Eng , J. J. , et al. . ( 2015 ). getting on with the rest of your life following stroke: A randomized trial of a complex intervention aimed at enhancing life participation post stroke . Clinical Rehabilitation , 29 , 1198 – 1211 . Google Scholar CrossRef Search ADS PubMed Mayo , N. E. , Wood-Dauphinee , S. , Côté , R. , Durcan , L. , & Carlton , J. ( 2002 ). Activity, participation, and quality of life 6 months poststroke . Archives of Physical Medicine and Rehabilitation , 83 , 1035 – 1042 . Google Scholar CrossRef Search ADS PubMed Merz , Z. C. , Van Patten , R. , Mulhauser , K. , & Fucetola , R. ( 2017 ). Exploratory factor analysis of the reintegration to normal living index in a stroke population . Topics in Stroke Rehabilitation , 24 , 158 – 162 . Google Scholar CrossRef Search ADS PubMed Mesulam , M. ( 1985 ). Principles of behavioral neurology: Verbal and Nonverbal Cancellation Test . Philadelphia : FA Davis Company . Miller , E. L. , Murray , L. , Richards , L. , Zorowitz , R. D. , Bakas , T. , Clark , P. , et al. . ( 2010 ). Comprehensive overview of nursing and interdisciplinary rehabilitation care of the stroke patient . Stroke; A Journal of Cerebral Circulation , 41 , 2402 – 2448 . Google Scholar CrossRef Search ADS Mok , V. C. T. , Wong , A. , Lam , W. W. M. , Fan , Y. H. , Tang , W. K. , Kwok , T. , et al. . ( 2004 ). Cognitive impairment and functional outcome after stroke associated with small vessel disease . Journal of Neurology, Neurosurgery & Psychiatry , 75 , 560 – 566 . Google Scholar CrossRef Search ADS Nasreddine , Z. S. , Phillips , N. A. , Bédirian , V. , Charbonneau , S. , Whitehead , V. , Collin , I. , et al. . ( 2005 ). The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment . Journal of the American Geriatrics Society , 53 ( 4 ), 695 – 699 . Google Scholar CrossRef Search ADS PubMed Newman , D. A. ( 2003 ). Longitudinal modeling with randomly and systematically missing data: A simulation of ad hoc, maximum likelihood, and multiple imputation techniques . Organizational Research Methods , 6 , 328 – 362 . Google Scholar CrossRef Search ADS Nichols-Larsen , D. S. , Clark , P. C. , Zeringue , A. , Greenspan , A. , & Blanton , S. ( 2005 ). Factors influencing stroke survivors’ quality of life during subacute recovery . Stroke; A Journal of Cerebral Circulation , 36 , 1480 – 1484 . Google Scholar CrossRef Search ADS Norlander , A. , Carlstedt , E. , Jönsson , A. C. , Lexell , E. M. , Ståhl , A. , Lindgren , A. , et al. . ( 2016 ). Long-term predictors of social and leisure activity 10 years after stroke . PloSOne , 11 ( 2 ), 1 – 16 . Google Scholar CrossRef Search ADS Nys , G. M. S. , Van Zandvoort , M. J. E. , De Kort , P. L. M. , Van der Worp , H. B. , Jansen , B. P. W. , Algra , A. , et al. . ( 2005 ). The prognostic value of domain-specific cognitive abilities in acute first-ever stroke . Neurology , 64 , 821 – 827 . Google Scholar CrossRef Search ADS PubMed Oksala , N. K. J. , Jokinen , H. , Melkas , S. , Oksala , A. , Pohjasvaara , T. , Hietanen , M. , et al. . ( 2009 ). Cognitive impairment predicts poststroke death in long-term follow-up . Journal of Neurology, Neurosurgery & Psychiatry , 80 , 1230 – 1235 . Google Scholar CrossRef Search ADS Pang , M. Y. , Eng , J. J. , & Miller , W. C. ( 2007 ). Determinants of satisfaction with community reintegration in older adults with chronic stroke: Role of balance self-efficacy . Physical Therapy , 87 , 282 – 291 . Google Scholar CrossRef Search ADS PubMed Patel , M. D. , Coshall , C. , Rudd , A. G. , & Wolfe , C. D. ( 2002 ). Cognitive impairment after stroke: Clinical determinants and its associations with long‐term stroke outcomes . Journal of the American Geriatrics Society , 50 , 700 – 706 . Google Scholar CrossRef Search ADS PubMed Pedersen , P. M. , Stig Jørgensen , H. , Nakayama , H. , Raaschou , H. O. , & Olsen , T. S. ( 1995 ). Aphasia in acute stroke: Incidence, determinants, and recovery . Annals of Neurology , 38 , 659 – 666 . Google Scholar CrossRef Search ADS PubMed Pendlebury , S. T. , Cuthbertson , F. C. , Welch , S. J. , Mehta , Z. , & Rothwell , P. M. ( 2010 ). Underestimation of cognitive impairment by Mini-Mental State Examination versus the Montreal Cognitive Assessment in patients with transient ischemic attack and stroke . Stroke; A Journal of Cerebral Circulation , 41 , 1290 – 1293 . Google Scholar CrossRef Search ADS Pohjasvaara , T. , Vataja , R. , Leppävuori , A. , Kaste , M. , & Erkinjuntti , T. ( 2002 ). Cognitive functions and depression as predictors of poor outcome 15 months after stroke . Cerebrovascular Diseases , 14 , 228 – 233 . Google Scholar CrossRef Search ADS PubMed Protopsaltis , J. , Kokkoris , S. , Korantzopoulos , P. , Milionis , H. J. , Karzi , E. , Anastasopoulou , A. , et al. . ( 2009 ). Prediction of long-term functional outcome in patients with acute ischemic non-embolic stroke . Atherosclerosis , 203 , 228 – 235 . Google Scholar CrossRef Search ADS PubMed Riepe , M. W. , Riss , S. , Bittner , D. , & Huber , R. ( 2004 ). Screening for cognitive impairment in patients with acute stroke . Dementia and Geriatric Cognitive Disorders , 17 , 49 – 53 . Google Scholar CrossRef Search ADS PubMed Rubin , D. B. ( 1976 ). Inference and missing data . Biometrika , 63 , 581 – 592 . Google Scholar CrossRef Search ADS Rubin , D. B. ( 1987 ). Multiple imputation for nonresponse in surveys . New York : Wiley . Sachdev , P. S. , Brodaty , H. , Valenzuela , M. J. , Lorentz , L. , & Koschera , A. ( 2004 ). Progression of cognitive impairment in stroke patients . Neurology , 63 , 1618 – 1623 . Google Scholar CrossRef Search ADS PubMed Salvadori , E. , Pasi , M. , Poggesi , A. , Chiti , G. , Inzitari , D. , & Pantoni , L. ( 2013 ). Predictive value of MoCA in the acute phase of stroke on the diagnosis of mid-term cognitive impairment . Journal of Neurology , 260 , 2220 – 2227 . Google Scholar CrossRef Search ADS PubMed Schafer , J. L. , & Graham , J. W. ( 2002 ). Missing data: Our view of the state of the art . Psychological Methods , 7 , 147 – 177 . Google Scholar CrossRef Search ADS PubMed Scheffer , J. ( 2002 ). Dealing with missing data . Research Letters in the Information and Mathematical Sciences , 3 , 153 – 160 . Srikanth , V. K. , Thrift , A. G. , Saling , M. M. , Anderson , J. F. , Dewey , H. M. , Macdonell , R. A. , et al. . ( 2003 ). Increased risk of cognitive impairment 3 months after mild to moderate first-ever stroke . Stroke; A Journal of Cerebral Circulation , 34 , 1136 – 1143 . Google Scholar CrossRef Search ADS Stone , A. A. , Bachrach , C. A. , Jobe , J. B. , Kurtzman , H. S. , & Cain , V. S. ( 1999 ). The science of self-report: Implications for research and practice . Mahwah, NJ : Erlbaum . Strauss , E. , Sherman , E. M. , & Spreen , O. ( 2006 ). A Compendium of neuropsychological tests: Administration, norms, and commentary (3rd ed.). New York, NY : Oxford University Press . Tatemichi , T. K. , Desmond , D. W. , Stern , Y. , Paik , M. , Sano , M. , & Bagiella , E. ( 1994 ). Cognitive impairment after stroke: Frequency, patterns, and relationship to functional abilities . Journal of Neurology, Neurosurgery & Psychiatry , 57 , 202 – 207 . Google Scholar CrossRef Search ADS Tilling , K. , Sterne , J. A. , Rudd , A. G. , Glass , T. A. , Wityk , R. J. , & Wolfe , C. D. ( 2001 ). A new method for predicting recovery after stroke . Stroke; A Journal of Cerebral Circulation , 32 , 2867 – 2873 . Google Scholar CrossRef Search ADS Törnbom , K. , Persson , H. C. , Lundälv , J. , & Sunnerhagen , K. S. ( 2017 ). Self-assessed physical, cognitive, and emotional impact of stroke at 1 month: The importance of stroke severity and participation . Journal of Stroke and Cerebrovascular Diseases , 26 , 57 – 63 . Google Scholar CrossRef Search ADS PubMed Van Patten , R. , Merz , Z. C. , Mulhauser , K. , & Fucetola , R. ( 2016 ). Multivariable prediction of return to work at 6-month follow-up in patients with mild to moderate acute stroke . Archives of Physical Medicine and Rehabilitation , 97 , 2061 – 2067 . Google Scholar CrossRef Search ADS PubMed Veerbeek , J. M. , Kwakkel , G. , van Wegen , E. E. , Ket , J. C. , & Heymans , M. W. ( 2011 ). Early prediction of outcome of activities of daily living after stroke . Stroke; A Journal of Cerebral Circulation , 42 , 1482 – 1488 . Google Scholar CrossRef Search ADS Veerbeek , J. M. , van Wegen , E. , van Peppen , R. , van der Wees , P. J. , Hendriks , E. , Rietberg , M. , et al. . ( 2014 ). What is the evidence for physical therapy poststroke? A systematic review and meta-analysis . PloSOne , 9 ( 2 ), 1 – 33 . Google Scholar CrossRef Search ADS Vellone , E. , Savini , S. , Fida , R. , Dickson , V. V. , Melkus , G. D. , Carod-Artal , F. J. , et al. . ( 2015 ). Psychometric evaluation of the Stroke Impact Scale 3.0 . The Journal of Cardiovascular Nursing , 30 , 229 – 241 . Google Scholar CrossRef Search ADS PubMed Wang , Y. C. , Kapellusch , J. , & Garg , A. ( 2014 ). Important factors influencing the return to work after stroke . Work (Reading, Mass.) , 47 , 553 – 559 . Google Scholar PubMed Wattanapan , P. , Kovindha , A. , Piravej , K. , & Kuptniratsaikul , V. ( 2010 ). Relationship between the ability to change from a supine to a sitting position at admission and mobility outcomes after stroke rehabilitation . Journal of the Medical Association of Thailand , 93 , S21 – S24 . Google Scholar PubMed Wilkins , C. H. , Wilkins , K. L. , Meisel , M. , Depke , M. , Williams , J. , & Edwards , D. F. ( 2007 ). Dementia undiagnosed in poor older adults with functional impairment . Journal of the American Geriatrics Society , 55 , 1771 – 1776 . Google Scholar CrossRef Search ADS PubMed Wolf , T. J. , & Rognstad , M. C. ( 2013 ). Changes in cognition following mild stroke . Neuropsychological Rehabilitation , 23 ( 2 ), 256 – 266 . Google Scholar CrossRef Search ADS PubMed Woodman , P. , Riazi , A. , Pereira , C. , & Jones , F. ( 2014 ). Social participation post stroke: A meta-ethnographic review of the experiences and views of community-dwelling stroke survivors . Disability and Rehabilitation , 36 , 2031 – 2043 . Google Scholar CrossRef Search ADS PubMed © The Author 2017. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Archives of Clinical NeuropsychologyOxford University Press

Published: Aug 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off