Decision Support for Joint Replacement: Implications for Decisional Conflict and Willingness to Undergo Surgery

Decision Support for Joint Replacement: Implications for Decisional Conflict and Willingness to... Abstract Objectives The present study investigates age differences in the types of decision support that total joint replacement (TJR) candidates desire and receive when making the decision to pursue surgery. We consider the social structural (relationship to the patient) and experiential factors (network members’ experience with TJR) that influence individuals’ support preferences and the interactions of these factors with age. We also examine whether a lack of support is linked with increased decisional conflict and reduced willingness to undergo surgery. Method A telephone survey was conducted with 100 individuals (aged 40+) who were contemplating knee or hip replacement. Results TJR candidates desired and received decision support from health care providers, family members, and individuals who had previously undergone TJR. They reported higher deficits in informational and emotional support than in instrumental support. Overall, a lack of instrumental support was associated with greater decisional conflict; a lack of instrumental support and a lack of informational support were associated with reduced willingness to undergo TJR. Discussion Our findings point to the importance of involving both formal and informal network members in TJR discussions, and the need for informational guidance and practical assistance to reduce decisional conflict and uncertainty among individuals considering TJR. Chronic pain, Decision making, Joint replacement, Social networks, Social support Arthritis is a leading cause of pain and disability among middle-aged and older adults (CDC, 2009). Total joint replacement (TJR) has been shown to be a cost-effective (Daigle, Weinstein, Katz, & Losina, 2012) and highly efficacious treatment for this condition, both in terms of improving health-related quality of life (Ethgen, Bruyere, Richy, Dardennes, & Reginster, 2004) and physical outcomes (Vissers, Bussmann, de Groot, Verhaar, & Reijman, 2013). Although total knee and hip replacements for advanced arthritis are among the most common elective surgeries performed in the United States (Lawson, Gibbons, Ingraham, Shekelle, & Ko, 2011), empirical evidence shows considerable variation in patients’ willingness to undergo these procedures (Hawker et al., 2000). Indeed, many appropriate candidates for TJR are apprehensive to undergo TJR. This reluctance may stem from a variety of reasons, including fear of surgery (Figaro, Williams-Russo, & Allegrante, 2004), concerns about dependency following the procedure (Toye, Barlow, Wright, & Lamb, 2006), and perceptions of current pain severity (Hawker, Wright, Badley, & Coyte, 2004). Although accumulating research has considered the individual-level characteristics associated with the choice to pursue surgery, including patients’ pain symptoms and sociodemographic characteristics (i.e., age, income, gender, living arrangement, rurality, ethnicity; Hawker et al., 2000), less research has examined the broader social context or psychological processes involved in patients’ decisions or willingness to pursue TJR. Instead, studies incorporating psychosocial dimensions of TJR decision making typically focus on the link between social support and health-related quality of life after the surgery is complete (Ethgen et al., 2004). Importantly, a better understanding of the psychosocial factors underlying patients’ decisions to pursue TJR could help promote individualized decision assistance for patients considering this surgical procedure. In this paper, we contribute to the existing literature on people’s pursuit of TJR by examining the types and sources of support patients desire and receive when considering surgery. We approach our investigation from a life-span perspective by considering how patients’ age may be associated with their support preferences. In addition, we explore whether lack of support is associated with greater decisional conflict and reduced willingness to undergo surgery. Definitions Following prior literature, our study defines subjective decisional conflict as the “state of uncertainty about the course of action to take” (O’Connor, 1995, p. 25). We also examine willingness to undergo surgery; or, the self-reported likelihood that the TJR candidate will pursue surgery. With regard to social support, this study differentiates three commonly described dimensions (Cohen & Wills, 1985): informational (knowledge or advice), emotional (reassurance and empathetic listening), and instrumental support (tangible assistance with practical problems). Also following prior definitions of social support systems (Caplan, 1974), we examine both informal (family members, friends who have been through surgery) and formal sources (health care providers) in patients’ decisions to pursue joint replacement. Support for Joint Replacement Decisions Theoretical Frameworks Perspectives from psychology and social gerontology offer guidance regarding the potential mechanisms linking social networks and support provision with the choice to undergo surgery. On the one hand, theoretical frameworks propose that social structural norms and expectations may influence support provision and receipt from one’s formal and informal networks. Sociological perspectives propose that individuals’ support selection is contingent upon the salience of the relationship between the caregiver and care recipient, where kinship is a key dimension (Cantor, 1979). Family members are traditionally expected to offer practical aid (Messeri, Silverstein, & Litwak, 1993) and emotional comfort during times of stress (Cohen & Wills, 1985; Thoits, 2011). Such models have been extended to include formal networks, proposing that the “dual specialization” of informal and formal systems will promote optimal support for the impaired individual (Noelker & Bass, 1989). Health care providers, for example, may be chosen over family during times of illness (Messeri et al., 1993) and are expected to serve as expert advisors regarding health-specific resources and information (Brashers, Goldsmith, & Hsieh, 2002). On the other hand, homophily theory (Pillemer & Suitor, 1996; Thoits, 2011) argues that effective support provision is likely to stem from others who have experienced similar stressors or situations. Because experientially similar others have first-hand knowledge of coping with a particular illness or life event, they may possess the unique ability to validate others’ emotions and concerns and offer detailed advice by providing specific health-related information and knowledge (Gage, 2013; Thoits, 2011). In the present study, experiential similarity explicitly refers to individuals who have previously undergone TJR. Insights from behavioral economics also offer support for the role of experientially similar others in TJR decision making. According to predictions derived from prospect theory, optimal decision making may largely depend on an individual’s ability to accurately predict how s/he will feel in the future (Loewenstein, 2000). Experientially similar others may therefore provide a valuable window into the future by serving as means of predicting one’s future emotional states, also known as affective forecasting (Loewenstein, 2000). Applying this concept to the context of TJR, individuals considering surgery may seek the perspectives of others who have previously undergone TJR to limit projection biases (failure to project one’s present state onto one’s future state; Loewenstein, 2005) and determine their preferences based on others’ prior experience with TJR. Empirical Evidence We review the available evidence from empirical studies of TJR decision making below to illustrate how these predictions map onto prior research and inform our hypotheses regarding patients’ preferences for the types (informational, emotional, instrumental) and sources of support they desire and receive when considering TJR. Informational support With regard to informational support, qualitative studies have shown that the perspectives of individuals who have previously undergone TJR are highly influential in shaping patients’ attitudes toward surgery, both in lieu of and supplementary to physician guidance (McHugh & Luker, 2009; Riffin, Pillemer, Reid, & Löckenhoff, 2015). In addition, quantitative research has suggested that simply knowing someone who has undergone a successful joint replacement may positively influence patients’ willingness to consider the procedure themselves (Hawker et al., 2004), but that patients may be deterred from pursuing surgery if the procedure did not go well for the other (McHugh & Luker, 2009). Because individuals who have been through a successful versus unsuccessful surgery appear to bear differential influences on prospective TJR patients’ decisional outcomes, our study considers these two subgroups separately. Emotional support With regard to emotional support, a recent review and meta-analysis reported that family members positively influence patient well-being through empathic interactions and compassionate care for their loved one (Martire, Lustig, Schulz, Miller, & Helgeson, 2004), which in turn may have implications for patients’ treatment decisions regarding surgery. For example, a study of osteoarthritis patients considering TJR found that emotional encouragement by family members facilitated patients’ choice to undergo the procedure (McHugh & Luker, 2009). Other research has shown that having emotional support provided by close others helped TJR patients maintain continuity throughout the decisional process (Sjöling, Ågren, Olofsson, Hellzén, & Asplund, 2005). Instrumental support TJR almost inevitably requires practical assistance by others (Showalter, Burger, & Salyer, 2000). Yet, with rare exceptions (McHugh & Luker, 2009; Sjöling et al., 2005; Toye et al., 2006), studies have overlooked this important support domain. Critically, concerns about dependency following surgery (e.g., trouble with activities of daily living [ADLs]) may play a vital role in the decision-making process (Toye et al., 2006). For example, TJR candidates tend to solicit instrumental support by family members prior to surgery, which in turn is linked with patients’ sense of control over their decision to undergo the procedure (Sjöling et al., 2005). Based on the theoretical considerations and empirical evidence described above, we hypothesize that TJR candidates will primarily desire and receive informational support from their health care providers and others who have been through TJR, emotional reassurance from informal network ties, including family members and individuals who have been through TJR, and instrumental aid from close relatives. Age-Related Differences in Decision Support Preferences In addition to investigating social structural and experiential factors, the present study also considers how patients’ age may play a role in their decision-making processes. As a guiding framework, socioemotional selectivity theory (SST) offers a life-span perspective of how age-related shifts in time horizons may result in age differences in social support preferences (Carstensen, 2006). Specifically, SST posits that in young adulthood, when time horizons are perceived as expansive, information acquisition is prioritized. With advanced age, however, time horizons constrict and individuals prioritize emotionally meaningful experiences and social relationships over knowledge gathering. This developmental pattern of prioritizing emotional meaning over information has been confirmed in a recent meta-analysis considering age differences in information seeking and decision quality: in general, older adults tend to seek less pre-decisional information compared with younger individuals (Mata & Nunes, 2010). Whereas SST provides a framework for the types of support individuals desire, dyadic exchanges between older adults and their social network members may influence the types of support older adults receive. According to the Social Input Model (Fingerman & Charles, 2010), social network members tend to be more gentle with older adults and offer them preferential treatment, which may stem from a variety of reasons varying from negative stereotypes about aging to respect for one’s elders. Empirical work has substantiated this prediction, revealing that social network members often reinforce satisfying and meaningful relationships for older adults (Fingerman & Pitzer, 2007). In the context of TJR, qualitative reports by physicians also reveal a pronounced hesitation to offer negative feedback to older adults, especially when the patient exhibits signs of fatigue or concern (Gooberman-Hill et al., 2010). Overall, this dyadic process may contribute to older adults receiving greater emotional and less informational support when considering TJR than middle-aged individuals. Implications for Decision Conflict and Willingness to Undergo Surgery In addition to examining the types and sources of support that patients may desire and receive when considering TJR, we further conjecture that a lack of adequate support may have important implications for decisional conflict and the willingness to pursue surgery. Although little research has explicitly investigated the link between social support and preoperative decisional conflict or willingness to undergo TJR, qualitative research with osteoarthritis patients awaiting surgery offers preliminary evidence of this association. Specifically, this research demonstrated patients’ frustration and confusion when they failed to receive the informational guidance they desired from their providers (Sjöling et al., 2005), but that having instrumental assistance by family members helped patients feel more secure in their decisions. In related work, research with primary care patients has shown that when patients’ support expectations are met by their health care providers, they report having significantly higher rates of satisfaction in their clinical interactions (Williams, Weinman, Dale, & Newman, 1995). In turn, patients’ satisfaction regarding communication with their providers has been associated with greater decisional clarity and confidence in patients’ decision making (Edwards et al., 2004). Other research with arthritis patients similarly suggests that a high-quality patient–provider relationship, combined with clear and balanced information, is crucial to promoting decisional clarity when considering whether to undergo surgery (Zaidi, Pfeil, Macgregor, & Goldberg, 2013). For these reasons, we expect that lack of support will be associated with greater decisional conflict and reduced willingness to pursue surgery among patients. The Present Study This study investigates the relative influence of structural (relationship to respondent) and experiential factors (contact with others who have undergone prior successful vs unsuccessful TJR) in determining the types of social support (informational, emotional, instrumental) that patients desire and receive when making the decision to undergo TJR. We also explore potential age differences in patients’ preference for emotional versus informational support and examine whether lack of support is linked with TJR candidates’ subjective decisional conflict or willingness to undergo surgery. All analyses include key covariates previously found to be associated with decision making and social network structure. Five-factor personality traits have been linked with social network formation (Soldz & Vaillant, 1999) and health care decision-making styles (Flynn & Smith, 2007). In addition, poor health has been identified as a strong predictor of a weakened social network; therefore, physical health is included as a covariate (cf. House, Umberson, & Landis, 1988). Pain level and pain location (knee vs hip) are also included given that increased pain has been linked with patients’ willingness to undergo surgery (Brander et al., 2003) and that knee (vs hip) replacement has been associated with less favorable pain outcomes after surgery (Beswick, Wylde, Gooberman-Hill, Blom, & Dieppe, 2012). Consistent with previous methodology, the total number of decision network members (DNMs) is included as a covariate to adjust for differences in respondents’ network size. Finally, because prior TJR experience may influence individuals’ perceptions of and willingness to undergo a second TJR (Ballantyne, Gignac, & Hawker, 2007), this study includes only respondents who have not previously undergone joint replacement. Based on the theoretical considerations outlined above, we propose the following hypotheses: Hypothesis 1: Composition of Decision Support Network. TJR candidates’ decision support networks will include not only include formal ties with health care providers but also informal ties with family members and nonrelatives, including individuals who have previously undergone TJR. Hypothesis 2: Support Desired and Received. TJR candidates will desire and receive informational support from health care providers and from individuals who have previously undergone TJR. Emotional support will be sought from family members and others who have previously undergone TJR, and instrumental support will be desired and received from family members and individuals who have not been through TJR. Hypothesis 3: Age Differences. Older age will be associated with a greater desire for and receipt of emotional rather than informational support from all sources. Hypothesis 4: Decision Conflict and Willingness to Undergo Surgery. A lack of support will be associated with greater decision conflict and reduced willingness to undergo TJR. Given the limited research in this area, we do not make concrete predictions regarding specific types of support. Method Participants A variety of strategies were used to recruit participants at various stages in the decision-making process, from contemplation (respondents who were still considering whether or not to pursue surgery) to active pursuit of joint replacement (respondents who had set a specific date for the procedure). A primary recruitment method was an electronic search of EpicCare health records at two outpatient practices in New York City. This approach was selected for efficiency reasons: Querying the database for patients with a diagnosis of hip or knee osteoarthritis allowed us to more easily identify a subgroup of individuals appropriate for the study (i.e., those who were potential candidates for TJR). These individuals were contacted via postal mail to inform them of the study and invite their participation. This strategy was supplemented by online advertising and posting flyers at local senior centers, senior housing residences, and physician practices. The study protocol was approved by the Cornell University and Weill Cornell Medical College Institutional Review Boards. Individuals were excluded if they were younger than age 40, not fluent in English, had previously undergone joint replacement, or exhibited cognitive impairment, defined by a score of less than 3 on a six-item screener (Callahan, Unverzagt, Hui, Perkins, & Hendrie, 2002). Of the 296 individuals contacted, 72 declined participation and 97 were excluded due to cognitive impairment or prior joint replacement. The final sample consisted of 100 individuals, 84% of whom were women and 72% were non-Hispanic white(Table 1). However, the data for one participant were lost due to a Qualtrics recording failure. Therefore, analyses were conducted on a sample of 99 individuals. Table 1. Respondent Characteristics     n  Demographics   Age (SD)  66.6 (10.6)  99   % women  83.7%  99   % white  72.4%  99   % with college degree  64.6%  99   Physical health  3.3 (1.0)  99   Mental health  3.8 (1.0)  98   # DNMs  2.8 (1.7)  92  Pain location   % knee pain  81.6%  99   % hip pain  18.4%  99  % considered surgery  93.9%  99  % recommended for surgery by physician  71.4%  99  Personality   Extraversion (SD)  5.4 (1.5)  99   Agreeableness (SD)  7.3 (1.4)  98   Conscientiousness (SD)  8.0 (1.2)  99   Neuroticism (SD)  5.3 (1.7)  99   Openness (SD)  7.5 (1.5)  99  Support desired and received (summed across DNM)   Informational support desired (SD)  2.6 (0.8)  92   Informational support received (SD)  2.4 (0.9)  91   Emotional support desired (SD)  2.6 (1.0)  90   Emotional support received (SD)  2.4 (1.0)  90   Instrumental support desired (SD)  2.1 (1.1)  90   Instrumental support received (SD)  1.9 (1.1)  90  Decision conflict (SD)  27.3 (24.3)  90  Willingness to undergo TJR (SD)  2.3 (1.1)  99      n  Demographics   Age (SD)  66.6 (10.6)  99   % women  83.7%  99   % white  72.4%  99   % with college degree  64.6%  99   Physical health  3.3 (1.0)  99   Mental health  3.8 (1.0)  98   # DNMs  2.8 (1.7)  92  Pain location   % knee pain  81.6%  99   % hip pain  18.4%  99  % considered surgery  93.9%  99  % recommended for surgery by physician  71.4%  99  Personality   Extraversion (SD)  5.4 (1.5)  99   Agreeableness (SD)  7.3 (1.4)  98   Conscientiousness (SD)  8.0 (1.2)  99   Neuroticism (SD)  5.3 (1.7)  99   Openness (SD)  7.5 (1.5)  99  Support desired and received (summed across DNM)   Informational support desired (SD)  2.6 (0.8)  92   Informational support received (SD)  2.4 (0.9)  91   Emotional support desired (SD)  2.6 (1.0)  90   Emotional support received (SD)  2.4 (1.0)  90   Instrumental support desired (SD)  2.1 (1.1)  90   Instrumental support received (SD)  1.9 (1.1)  90  Decision conflict (SD)  27.3 (24.3)  90  Willingness to undergo TJR (SD)  2.3 (1.1)  99  Notes. DNM = decision network member; TJR = total joint replacement; SD = standard deviation. Willingness to undergo TJR was assessed on a 5-point Likert scale from 1 = certainly will not pursue surgery to 5 = certainly will pursue surgery. The Decision Conflict Scale ranges from 0 = no decisional conflict to 100 = extremely high decisional conflict. Support desired and received was measured by a 4-point Likert scale ranging from 1 = never to 4 = very often. View Large Measures Primary predictor and outcome variables Social network composition and function. Following previous methodology (Antonucci & Akiyama, 1987; Pillemer & Suitor, 1996), information on the structure and function of participants’ decision support networks was obtained through a series of name-elicitation questions. For each DNM, data were collected on (a) demographic characteristics including age and gender (0 = male; 1 = female), (b) relationship to respondent (open-response coded 1 = family member; 2 = health care provider; 3 = non-family close other), and (c) prior experience with TJR (1 = prior successful TJR; 2 = prior unsuccessful TJR; 3 = no prior TJR). From this point forward, we refer to this variable as underwent TJR. Types of support desired and received. Participants were asked to indicate how often they desired and received support from each DNM. To assess informational support desired and received, respondents were asked “How often does [DNM] give you advice or suggestions about your decision to get a joint replacement (never, sometimes, often, or very often)?” Followed by “Now, think about whether you actually want this type of support from [DNM]. If it were up to you, how often would you want advice or suggestions from [DNM] (1 = never, 2 = sometimes, 3 = often, or 4 = very often)?” Analogous questions were used to examine how often participants received and desired “comfort and reassurance” (emotional support) and “practical aid and assistance” (instrumental support) from each DNM. To capture lack of support, difference scores were calculated for each type of support provided by each DNM (informational, emotional, instrumental) by subtracting the specific types of support received from the type of support desired. Following methodology used previously to examine social ties and support among older adults (Fiori, Antonucci, & Akiyama, 2008), each type of support received was averaged within each DNM category (e.g., health care provider) for each respondent. For example, if two health care providers were named by a single respondent, the amount of emotional support received from each was averaged within the DNM category “health care provider.” The same procedure was used for all other DNM categories and types of support desired. Difference scores were also averaged within each DNM category. Identical analyses were also conducted with support summed (instead of averaged) within each category. Results were comparable using both methods. Decision conflict. Participants responded to the low-literacy 10-item version of the Decision Conflict Scale (DCS) (Linder et al., 2011). The DCS targets uncertainty about choosing among alternatives and modifiable factors contributing to uncertainty (e.g., feeling uninformed, unclear about values). This scale has been shown to reliably discriminate between individuals who make and delay decisions (present study α = .84). Willingness to undergo surgery. All participants were asked the question “How certain are you that you’ll get a knee/hip replacement?” (1 = certainly will not, 2 = probably will not, 3 = could go either way, 4 = probably will, 5 = certainly will). Covariates Personality. Respondents’ personality was assessed using the 10-Item Short Version of the Big Five Inventory (BFI-10; Rammstedt & John, 2007). Individuals rated the extent to which each personality characteristic was true of them, ranging from 1 (strongly disagree) to 5 (strongly agree). The BFI-10 is an acceptable measure of personality when time is limited, as in telephone surveys (Rammstedt & John, 2007). Correlations between the personality characteristics ranged from −.60 to .11. Physical and mental health. Two questions from the PROMIS Global Health Scale (Cella et al., 2010) were used to assess participants’ overall self-reported physical and mental health (1 = poor, 2 = fair, 3 = good, 4 = very good, 5 = excellent). Pain level, function, and location. Pain was assessed using an adapted version of the American Academy of Orthopaedic Surgeons Lower Limb Core Hip/Knee Module (MODEMS, 1996), a region-specific scale measuring pain level and ability to perform daily tasks. Respondents were asked about their pain levels and ability to function when walking on a flat surface, going up/down stairs, sitting/lying down, putting on/off shoes or socks, toileting and bathing, and carrying out light housework. Response options for these items ranged from 0 (no pain) to 5 (extreme pain). The six items were summed to form a pain severity scale (present study α = .66). In addition, participants were asked three questions (also adapted from MODEMS) to assess whether they experienced knee pain (pain or stiffness in one or both of their knee(s); 0 = no, 1 = yes) or hip pain (pain in the top of their thigh or groin, or front section of their thigh; 0 = no, 1 = yes). Sociodemographic characteristics. Respondent sociodemographic characteristics included age, race/ethnicity, gender, education, and marital status. Procedure After providing informed oral consent, each participant completed a structured telephone interview lasting approximately 60min administered by a trained research assistant. Respondents’ sociodemographic information was first collected, followed by an assessment of their current pain level, pain location, and willingness to undergo TJR. Information was next collected on the patient’s social network composition and structure, the decisional support they desired and received, and decisional conflict. The final portion of the survey included measures of respondents’ personality. Several additional self-report questionnaires, not related to the core questions of this study, were also included in the survey (health literacy, aging stereotypes, and risk preferences). Participants’ responses were entered into Qualtrics software, Version 2013 (Qualtrics, Provo, UT). Data Analysis Sample overview Descriptive and univariate analyses were conducted to gain an initial understanding of TJR candidates’ sociodemographic, pain, and personal characteristics. A two-way repeated measures analysis of variance examined differences in TJR candidates’ desire for and receipt of each type of decision support (informational, emotional, instrumental) as well as differences across the types of support. In addition, bivariate correlations examined associations among respondent-level characteristics (age, gender, general health, pain level, pain location, personality, and network size), decision conflict and willingness to undergo surgery, and support desired and received. Hypothesis 1: Descriptive and univariate analyses were conducted to examine TJR candidates’ decision network composition; specifically, the average number of DNM in respondents’ networks, percentage of formal versus informal network ties, and percentage of DNM who had previously undergone TJR. Hypotheses 2 and 3: Hypotheses concerning TJR candidates’ desired and received support were conducted using a multilevel modeling framework. This type of structure was necessary to allow for simultaneous estimation of respondent-level and network-level effects and to accommodate the uneven distributions of social networks across participants. Two separate multilevel models were estimated for the averaged support that TJR candidates desired (Table 2, Model 1) and received (Table 2, Model 3) as dependent variables, with type of support specified as the index variable. Substantive predictor variables and covariates that reached significance in preliminary bivariate analyses were entered into each model. Table 2. Multilevel Models Predicting TJR Candidates’ Desired (Models 1 and 2) and Received (Models 3 and 4) Decision Support Parameter estimates  Model 1  Model 2  Model 3  Model 4  F(df 1, df 2)  F(df 1, df 2)  F(df 1, df 2)  F(df 1, df 2)  Intercept  11.78 (1, 74)***  12.71 (1, 72)***  5.67 (1, 74)*  5.49 (1,75)*   Type of support  15.64 (2, 349)***  13.23 (2,183)***  17.78 (2, 351)***  13.97 (2, 161)***   DNM relationship to respondent  1.84 (2, 377)  2.40 (2, 284)†  3.10 (2, 374)*  3.61 (2, 275)*   DNM underwent TJR  3.66 (2, 366)*  4.76 (2, 276)**  3.21 (2, 363)*  3.67 (1, 266)*   Respondent age  1.84 (1, 66)  1.66 (1, 69)  1.27 (1, 67)  1.14 (1, 68)  Covariates   Respondent marital status (married)  0.66 (1, 59)  0.62 (1, 59)  0.39 (1, 59)  0.34 (1, 58)   Respondent extraversion  0.00 (1, 56)  0.01 (1, 60)  0.31 (1, 57)  0.34 (1, 58)   Respondent agreeableness  2.10 (1, 65)  1.96 (1, 64)  0.22 (1, 66)  0.16 (1, 64)   Respondent openness  0.99 (1, 58)  0.73 (1, 62)  1.29 (1, 58)  1.19 (1, 61)   Network size  0.26 (1, 326)  0.30 (1, 62)  0.00 (1, 322)  0.05 (1, 265)   DNM age  1.01 (1, 273)  1.57 (1, 265)  0.01 (1, 272)  0.04 (1, 245)  Interactions   Type of support × DNM relationship  —  1.56 (4, 333)  —  1.21 (4, 336)   Type of support × DNM underwent TJR  —  3.30 (4, 320)**  —  3.87 (4, 319)**   Respondent age × type of support  —  0.88 (2, 150)  —  0.91 (2, 129)   Respondent age × underwent TJR  —  0.51 (2, 280)  —  1.19 (2, 270)  Parameter estimates  Model 1  Model 2  Model 3  Model 4  F(df 1, df 2)  F(df 1, df 2)  F(df 1, df 2)  F(df 1, df 2)  Intercept  11.78 (1, 74)***  12.71 (1, 72)***  5.67 (1, 74)*  5.49 (1,75)*   Type of support  15.64 (2, 349)***  13.23 (2,183)***  17.78 (2, 351)***  13.97 (2, 161)***   DNM relationship to respondent  1.84 (2, 377)  2.40 (2, 284)†  3.10 (2, 374)*  3.61 (2, 275)*   DNM underwent TJR  3.66 (2, 366)*  4.76 (2, 276)**  3.21 (2, 363)*  3.67 (1, 266)*   Respondent age  1.84 (1, 66)  1.66 (1, 69)  1.27 (1, 67)  1.14 (1, 68)  Covariates   Respondent marital status (married)  0.66 (1, 59)  0.62 (1, 59)  0.39 (1, 59)  0.34 (1, 58)   Respondent extraversion  0.00 (1, 56)  0.01 (1, 60)  0.31 (1, 57)  0.34 (1, 58)   Respondent agreeableness  2.10 (1, 65)  1.96 (1, 64)  0.22 (1, 66)  0.16 (1, 64)   Respondent openness  0.99 (1, 58)  0.73 (1, 62)  1.29 (1, 58)  1.19 (1, 61)   Network size  0.26 (1, 326)  0.30 (1, 62)  0.00 (1, 322)  0.05 (1, 265)   DNM age  1.01 (1, 273)  1.57 (1, 265)  0.01 (1, 272)  0.04 (1, 245)  Interactions   Type of support × DNM relationship  —  1.56 (4, 333)  —  1.21 (4, 336)   Type of support × DNM underwent TJR  —  3.30 (4, 320)**  —  3.87 (4, 319)**   Respondent age × type of support  —  0.88 (2, 150)  —  0.91 (2, 129)   Respondent age × underwent TJR  —  0.51 (2, 280)  —  1.19 (2, 270)  Notes. DNM = decision network member; TJR = total joint replacement. Age dichotomized into 0 = 65 years or younger; 1 = 66 years or older. †p < .10. *p < .05. **p < .01. ***p < .001. Four interaction terms, (a) type of support × DNM relationship to respondent, (b) type of support × underwent TJR, (c) respondent age × type of support, and (d) respondent age × underwent TJR, were then added to allow for comparisons between the types and sources of support that respondents desired and received (Hypothesis 2) and variations by age (Hypothesis 3). In these models (Table 2, Models 2 and 4), age was dichotomized into individuals 65 years and younger and individuals 66 and older. (Analyses using age as a continuous variable yielded comparable results.) If an interaction reached significance, pairwise comparisons using a Bonferroni correction were conducted. Hypothesis 4: The final set of analyses examined TJR candidates’ decisional conflict and willingness to undergo surgery as outcome variables. Difference scores for each type of support were then entered as independent variables into two separate generalized linear models (GLMs) with decision conflict (Table 3) and willingness (Table 4) to undergo surgery as the dependent variables. Each model also included DNM relationship to respondent and underwent TJR, and covariates that were found to be significant in preliminary bivariate analyses. Table 3. General Linear Model Predicting Decision Conflict Variable  β (SE)  t  Difference scores in support desired vs received   Informational support  0.28 (2.92)  0.10   Emotional support  4.34 (3.81)  1.14   Instrumental support  10.41 (3.75)**  2.78  Covariates   Respondent gender (female)  −18.89 (5.88)**  −3.21   Respondent marital status (married)  4.60 (4.05)  1.13   Respondent conscientiousness  −3.95 (1.76)*  1.13   Respondent openness  −3.65 (1.22)**  −2.99   Network size  −0.74 (2.36)  −0.31   DNM relationship to respondent  4.40 (2.38)  1.85   DNM underwent TJR  1.63 (2.44)†  0.67  Variable  β (SE)  t  Difference scores in support desired vs received   Informational support  0.28 (2.92)  0.10   Emotional support  4.34 (3.81)  1.14   Instrumental support  10.41 (3.75)**  2.78  Covariates   Respondent gender (female)  −18.89 (5.88)**  −3.21   Respondent marital status (married)  4.60 (4.05)  1.13   Respondent conscientiousness  −3.95 (1.76)*  1.13   Respondent openness  −3.65 (1.22)**  −2.99   Network size  −0.74 (2.36)  −0.31   DNM relationship to respondent  4.40 (2.38)  1.85   DNM underwent TJR  1.63 (2.44)†  0.67  Notes. DNM = decision network member; SE = standard error; TJR = total joint replacement. The Decision Conflict Scale ranges from 0 = no decisional conflict to 100 = extremely high decisional conflict. †p < .10. *p < .05. **p < .01. Table 4. General Linear Model Predicting Willingness to Undergo Surgery Variable  β (SE)  t  Difference scores in support desired vs received     Informational support  −0.49 (0.12)***  −3.97   Emotional support  −0.12 (0.16)  −0.75   Instrumental support  −0.14 (0.16)*  −1.84  Covariates   Respondent age  −0.04 (0.01)***  −5.01   Respondent marital status (married)  0.17 (0.17)  0.98   Respondent pain level  0.07 (0.02)**  3.07   Network size  0.02 (0.10)  0.20   DNM relationship to respondent  −0.04 (0.10)  −0.41   DNM underwent TJR  −0.11 (0.10)  −1.03  Variable  β (SE)  t  Difference scores in support desired vs received     Informational support  −0.49 (0.12)***  −3.97   Emotional support  −0.12 (0.16)  −0.75   Instrumental support  −0.14 (0.16)*  −1.84  Covariates   Respondent age  −0.04 (0.01)***  −5.01   Respondent marital status (married)  0.17 (0.17)  0.98   Respondent pain level  0.07 (0.02)**  3.07   Network size  0.02 (0.10)  0.20   DNM relationship to respondent  −0.04 (0.10)  −0.41   DNM underwent TJR  −0.11 (0.10)  −1.03  Notes. DNM = decision network member; SE = standard error; TJR = total joint replacement. Willingness to undergo TJR was assessed by a 5-point Likert scale from 1 = certainly will not pursue surgery to 5 = certainly will pursue surgery. *p < .05. **p < .01. ***p < .001. Results Sample Overview Of the total sample, 82% had knee pain and 18% had hip pain. With regard to willingness to undergo surgery, approximately one-fifth (n = 22) were certain they would pursue TJR, 27% (n = 22) were fairly certain they would pursue TJR, 28% (n = 28) were undecided, 15% (n = 15) were fairly certain that they would not pursue TJR, and the remaining 7 (7%) were certain they would not pursue TJR. Participants generally reported good mental health and average physical health, aside from their pain condition. See Table 1 for sample characteristics. In the repeated measures analysis examining differences in TJR candidates’ desire for and receipt of each type of decision support, the overall F statistic was significant, F(5, 153) = 46.0, p < .001. Post hoc comparisons showed that TJR candidates received less informational support than they desired (p < .001), less emotional support than they desired (p < .001), and less instrumental support than they desired (p < .001). Across support types, respondents reported higher deficits in informational and emotional support than in instrumental support. Hypothesis 1: TJR candidates’ decision support networks will include both formal and informal ties Of the 99 respondents, 92 (92%) reported consulting at least one DNM in the decision to pursue surgery. The seven respondents who did not list anyone in their decision network were excluded from subsequent multilevel analyses examining types of support desired and received. These individuals were similar to the rest of the sample with respect to key variables (age, decisional conflict, or willingness to undergo TJR). On average, respondents consulted 2.75 network members regarding the choice to undergo TJR (range: 0–9). Decision networks were evenly composed of family members (34%), providers (33%), and nonrelatives (33%). With regard to prior experience with TJR, respondents’ networks were split between individuals who had previously undergone surgery (32% successful TJR; 23% unsuccessful TJR) and individuals who had not undergone the procedure (45%). Hypothesis 2: Types of support desired and received Multilevel models indicated that the interaction, type of support × underwent TJR, was significant for both support desired (Table 2, Model 2) and received (Table 2, Model 4). The significant pairwise comparisons of this effect are described below. Because the interaction, type of support × DNM relationship, failed to reach significance (Table 2), pairwise comparisons were not conducted. Hypothesis 2a: TJR candidates will desire and receive informational support from health care providers and from individuals who have previously undergone TJR Informational support desired Pairwise comparisons showed that respondents desired more informational support from DNM who had undergone a successful surgery (Predicted Mean = 2.71, SE = 0.17) than from DNM who never had the procedure (Predicted Mean = 2.28, SE = 0.16, p < .05). Informational support received Respondents were also more likely to receive informational support from DNM who had previously undergone a successful surgery (Predicted Mean = 2.54, SE = 0.16) than from DNM who did not have the procedure (Predicted Mean = 2.04, SE = 0.15, p < .05). Hypothesis 2b: Emotional support will be desired and received from family members and others who have previously undergone TJR Emotional support desired and received Pairwise comparisons failed to detect differences across groups, suggesting that respondents were equally likely to desire and receive emotional support from all DNM. Hypothesis 2c: Instrumental support will be desired and received from family members and individuals who have not been through TJR Instrumental support desired Pairwise comparisons showed that respondents were more likely to desire instrumental support from DNM who had not previously undergone surgery (Predicted Mean = 2.02, SE = 0.17, p < .01) than from DNM who had undergone an unsuccessful surgery (Predicted Mean = 1.48, SE = 0.19, p < .01). Instrumental support received A similar pattern of effects was detected for instrumental support received. Respondents were more likely to receive this type of support from DMN who had not previously undergone surgery (Predicted Mean = 2.11, SE = 0.15) than from DNM who had undergone a successful (Predicted Mean = 1.77, SE = 0.16, p < .05) or unsuccessful TJR (Predicted Mean = 1.42, SE = 0.26, p < .05). Hypothesis 2: Additional Findings All multilevel models (Table 2) showed significant main effects for type of support, indicating that respondents desired and received certain types of support more than others. Underwent TJR also reached significance in all four models: respondents desired and received less support from DNM who had previously undergone an unsuccessful surgery compared with the reference group, DNM who never had the procedure. Finally, DNM relationship to respondent was significantly associated only with support received (Models 3 and 4). Respondents received more support from family members than from the reference group, nonrelatives. Hypothesis 3: Increased age will be associated with a greater desire for and receipt of emotional rather than informational support from all sources. Multilevel models were also conducted to examine Hypothesis 3. As shown in Table 2 (Models 2 and 4), neither of the age interactions (respondent age × type of support or age × underwent TJR) were significant. Therefore, post hoc comparisons were not conducted. Hypothesis 4: A lack of support will be associated with greater decision conflict and reduced willingness to undergo TJR. Decisional conflict GLM analyses (Table 3) indicated that a lack of instrumental support was associated with greater decisional conflict (B = 10.41, p < .01). In addition, women were less likely to experience decisional conflict (B = −18.89, p < .01), as were individuals who scored high on the personality traits of Conscientiousness (B = −3.95, p < .05) and Openness (B = −3.65, p < .01). Willingness to undergo surgery As shown in Table 4, results from GLM analyses showed that a lack of informational support (B = −.49, p < .001) and lack of instrumental support (B = −.14, p < .05) were both associated with reduced willingness to pursue TJR. In addition, TJR candidates who were middle-aged (B = −.04. p < .001) and those with greater pain (B = .07, p < .01) were more willing to undergo TJR. Discussion The present study investigated the types of decision support that patients desire and receive when considering TJR, as well as potential variations by age. Analyses also examined whether lack of support was associated with patients’ decisional conflict and willingness to undergo surgery. Overall, results offer partial support for our hypotheses. One key finding from this study was the general lack of support reported by TJR candidates. Across the three support dimensions (informational, emotional, instrumental), respondents desired more support than they received. As shown in our study and elsewhere (Sjöling et al., 2005), patients’ unmet needs leading up to surgery may influence their ability to commit to the procedure. With regard to decision network composition (Hypothesis 1), we found that—as expected—family members and other informal contacts were routinely included along with medical providers. With regard to prior experience with TJR, half of respondents consulted at least one social network member who had previously undergone joint replacement. This finding corroborates qualitative research documenting patients’ reliance on others who have been through TJR (Parks et al., 2014; Zaidi et al., 2013). Our predictions regarding Hypothesis 2 were generally supported. However, contrary to expectation, TJR candidates did not desire or receive informational guidance more from their health care providers than from their informal network ties. Nevertheless, this lack of differentiation appears to be consistent with prior qualitative research suggesting that older adults seek decision support from both formal and informal contacts when considering pain treatment options (Riffin et al., 2015). In the present study, respondents desired and received informational support more from individuals who had previously undergone a successful TJR than from DNM who never had the procedure; a finding that points to a potential “positivity preference” among TJR candidates in their desire for informational guidance from similar others with positive outcomes. Although we had expected that TJR candidates would desire and receive more emotional support from family members and from others who had previously undergone TJR, our analyses failed to detect differences across groups. Finally, as predicted, respondents desired instrumental support more from DNM who had not previously undergone surgery. Contrary to expectation, we found no significant age differences in respondents’ desires for and receipt of emotional versus informational support for TJR (Hypothesis 3). However, it is important to note that most studies comparing age differences in information seeking consider differences between young (ages 18–40) and older respondents (age 65+). In contrast, our study examined a more restricted age range (age 40+). Other studies have also found that older and middle-aged patients actually have similar informational and relational needs (Romito, Corvasce, Montanaro, & Mattioli, 2011), especially with regard to medical information (Mata & Nunes, 2010). It is also plausible that having functional limitations due to pain may have primed limited time horizons even among middle-aged adults and therefore altered respondents’ preference for emotional rather than informational support. Analyses examining patients’ decisional conflict and willingness to undergo TJR (Hypothesis 4) indicated that a lack of informational and instrumental support was associated with greater decisional conflict and reduced willingness to undergo surgery. Thus, obtaining sufficient information about the risks and benefits of the procedure appears to play a critical role in patients’ decisions (Zaidi et al., 2013). Further, having adequate instrumental aid may help to alleviate feelings of uncertainty associated with TJR—and ultimately—determine whether surgery is a tenable option for the patient. Consistent with this idea, studies within the caregiving literature suggest that informal assistance with ADLs is an essential element in planning patients’ medical care (Wolff, Boyd, Gitlin, Bruce, & Roter, 2012), and that insufficient help may have negative consequences for older adults (Depalma et al., 2013). Thus, instrumental support may be a key variable to include in future studies examining social support and medical decision making. Limitations and Future Directions Overall, the results from this study should be viewed in light of several limitations including the cross-sectional design and restricted sociodemographic focus. To address the issue of cross-sectionality, 6-month follow-up assessments of the present sample are currently under way to determine whether (a) reported willingness is associated with actually pursuing surgery and (b) whether congruence in individuals’ received and desired decision support predicts their decisional regret (e.g., regret of the treatment choice) at a 6-month follow up. The present study also sets the stage for other longitudinal investigations. One line of inquiry would be to follow joint replacement candidates across the decisional trajectory, starting at the first treatment discussion of TJR, to target the modifiable factors (e.g., sufficient informational and instrumental support or access others who have already had the procedure) to promote a timely decision. This is an important line of work given prior literature demonstrating that the timing of TJR has strong implications for patients’ long-term recovery (Fortin et al., 2002). A second consideration for future research will be to include measures of specific psychological (e.g., patients’ coping styles, locus of control) and relational (e.g., relationship quality, attachment) factors that could partially explain individuals’ decision network selection and need for emotional reassurance. For example, patients who have high self-efficacy in pain management or motivation may rely on internal resources and therefore may not turn to others for emotional reassurance or practical assistance. Although prior research has examined several psychosocial characteristics in relation to postoperative outcomes, a better understanding of these variables in preoperative contexts would shed light on alternative patient-level characteristics associated with decision support selection. Finally, because our study relied exclusively on TJR candidates’ reports, we were unable to differentiate between network members who were TJR candidates but refused surgery and those who never were candidates themselves. Future studies should make efforts to collect information from DNMs themselves to gain this additional perspective. Despite the clear need for future research to address the caveats noted above, our study extends prior research in several ways. Unlike other studies, we did not restrict our sample to patients who were referred for evaluation for TJR or who were scheduled to have surgery. Instead, we included patients who fell at any point on the decisional spectrum from contemplation (unsure of whether or not they will have the procedure) to active pursuit of TJR (set a date for the surgery) to rejection of the procedure after it had been proposed by a physician. This design allowed us to examine the types of support necessary for individuals at any stage in the decision-making process. Our study is also unique in its focus on decision support networks for TJR patients. To our knowledge, no other studies have quantitatively investigated this topic. Moreover, assessing all three support types offered a more holistic perspective than prior studies that have predominantly focused on advice seeking, generally neglecting the salient component of instrumental aid. A final contribution of this work was assessing whether lack of support was associated with TJR patients’ decisional conflict. Practice Implications Our findings also have implications for patient care. As liaisons between patients and their social networks, health care providers can assist patients in receiving adequate decision support by asking patients about the support they desire from nonmedical sources, and involving informal network members in treatment encounters, as appropriate. Physicians should therefore work with the patient and their informal network to ensure adequate practical assistance leading up to and following the procedure. A second recommendation for physicians is to ask patients whether they have received adequate information about the procedure, and not, what additional information the patient would find useful in helping him/her to make a decision. Finally, health care providers should be aware that patients desire as much emotional support from them as they expect from informal sources, a desire easily overlooked in busy medical settings. Taken together, the findings from this study may help physicians to understand why certain patients may or may not be ready to pursue TJR. Given that both formal and informal DNMs were sought for a spectrum of support functions, future research should aim to incorporate these various perspectives in subsequent studies of TJR decision making. Overall, this research provides an initial understanding of TJR patients’ decisional preferences and points to the need for social network–driven strategies to involve both medical and nonmedical providers in TJR discussions. Funding This work was supported by the Lawrence and Rebecca Stern Family Foundation, the Edward R. Roybal Centers for Translation of the Behavioral and Social Sciences of Aging (P30AG022845), and a President’s Council of Cornell Women Affinito Stewart Grant to C. E. Löckenhoff. C. E. Löckenhoff was also supported in part by 1R21AG043741. C. Riffin is currently supported by a National Institute on Aging Training Grant (T32AG1934). Acknowledgments We thank Elaine Wethington for her advice on analyzing social networks. We also thank Sara Hachey, Rebecca Lampert, and Shayna Ratner for their assistance with participant recruitment and interviewing. This study is based on the doctoral dissertation of C. Riffin, supervised by C. E. Löckenhoff. C. Riffin designed the study, recruited and interviewed participants, performed the analysis, and composed the manuscript. K. Pillemer, M. C. Reid, and C. Lӧckenhoff obtained funding. M. C. Reid and J. Tung supervised participant recruitment. C. Lӧckenhoff supervised data analysis and manuscript preparation. All authors provided input on the study design and approved the final manuscript. References American Academy of Orthopaedic Surgeons/Council of Musculoskeletal Specialty Societies. ( 1996, March). MODEMS Lower Limb Core, Version 1.1 . Antonucci T. C., & Akiyama H. ( 1987). Social networks in adult life and a preliminary examination of the convoy model. Journal of Gerontology , 42, 519– 527. doi: 10.1093/geronj/42.5.519 Google Scholar CrossRef Search ADS PubMed  Ballantyne P. J. Gignac M. A., & Hawker G. A. ( 2007). A patient‐centered perspective on surgery avoidance for hip or knee arthritis: Lessons for the future. Arthritis Care & Research , 57, 27– 34. doi:10.1002/art.22472 Google Scholar CrossRef Search ADS   Beswick, A. D., Wylde, V., Gooberman-Hill, R., Blom, A., & Dieppe, P. (2012). What proportion of patients report long-term pain after total hip or knee replacement for osteoarthritis? A systematic review of prospective studies in unselected patients. BMJ Open, 2, e000435. doi:10.1136/bmjopen-2011-000435 Brander, V. A., Stulberg, S. D., Adams, A. D., Harden, R. N., Bruehl, S., Stanos, S. P., & Houle, T. (2003). Predicting total knee replacement pain: A prospective, observational study. Clinical Orthopaedics and Related Research, 416, 27–36. Brashers D. E. Goldsmith D. J., & Hsieh E. ( 2002). Information seeking and avoiding in health contexts. Human Communication Research , 28, 258– 271. doi: 10.1111/j.1468-2958.2002.tb00807.x Google Scholar CrossRef Search ADS   Centers for Disease Control and Prevention (CDC). (2009). Prevalence and most common causes of disability among adults—United States, 2005. Morbidity and Mortality Weekly Report, 58, 421–426. Callahan C. M. Unverzagt F. W. Hui S. L. Perkins A. J., & Hendrie H. C. ( 2002). Six-item screener to identify cognitive impairment among potential subjects for clinical research. Medical Care , 40, 771– 781. doi: 10.1097/01.MLR.0000024610.33213.C8 Google Scholar CrossRef Search ADS PubMed  Cantor M. H. ( 1979). Neighbors and friends: An overlooked resource in the informal support system. Research on Aging , 1, 434– 463. doi:10.1177/016402757914002 Google Scholar CrossRef Search ADS   Caplan G. ( 1974). Support systems and community mental health: Lectures on concept development . New York, NY: Behavioral Publications. Carstensen L. L. ( 2006). The influence of a sense of time on human development. Science (New York, N.Y.) , 312, 1913– 1915. doi: 10.1126/science.1127488 Google Scholar CrossRef Search ADS PubMed  Cella D. Riley W. Stone A. Rothrock N. Reeve B. Yount S., … Hays R. ( 2010). The patient-reported outcomes measurement information system: 2005–2008. Journal of Clinical Epidemiology , 63, 1179– 1194. doi: 10.1016/j.jclinepi.2010.04.011 Google Scholar CrossRef Search ADS PubMed  Cohen S., & Wills T. A. ( 1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin , 98, 310– 357. doi: 10.1037/0033-2909.98.2.310 Daigle M. E. Weinstein A. M. Katz J. N., & Losina E. ( 2012). The cost-effectiveness of total joint arthroplasty: A systematic review of published literature. Best Practice & Research. Clinical Rheumatology , 26, 649– 658. doi: 10.1016/j.berh.2012.07.013 Google Scholar CrossRef Search ADS PubMed  Depalma G. Xu H. Covinsky K. E. Craig B. A. Stallard E. Thomas J.III, & Sands L. P. ( 2013). Hospital readmission among older adults who return home with unmet need for ADL disability. The Gerontologist , 53, 454– 461. doi: 10.1093/geront/gns103 Google Scholar CrossRef Search ADS PubMed  Edwards A. Elwyn G. Hood K. Atwell C. Robling M. Houston H., … Russell I.; Study Steering Group. ( 2004). Patient-based outcome results from a cluster randomized trial of shared decision making skill development and use of risk communication aids in general practice. Family Practice , 21, 347– 354. doi: 10.1093/fampra/cmh402 Google Scholar CrossRef Search ADS PubMed  Ethgen O. Bruyere O. Richy F. Dardennes C., & Reginster J. Y. ( 2004). Health-related quality of life in total hip and total knee arthroplasty: A qualitative and systematic review of the literature. The Journal of Bone & Joint Surgery , 86, 963– 974. Google Scholar CrossRef Search ADS   Figaro K. M. Williams-Russo P., & Allegrante J. P. ( 2004). Preferences for arthritis care among urban African Americans: “I don’t want to be cut.” Health Psychology , 23, 324– 329. doi: 10.1037/0278-6133.23.3.324 Google Scholar CrossRef Search ADS PubMed  Fingerman K. L., & Charles S. T. ( 2010). It takes two to tango why older people have the best relationships. Current Directions in Psychological Science , 19, 172– 176. doi: 10.1177/0963721410370297 Google Scholar CrossRef Search ADS   Fingerman K. L., & Pitzer L. M. ( 2007). Socialization in old age. In J. E. Grusec P. D. Hastings (Eds.), Handbook of socialization: Theory and research  (pp. 232– 255). New York, NY: Guilford Press. Fiori K. L. Antonucci T. C., & Akiyama H. ( 2008). Profiles of social relations among older adults: A cross-cultural approach. Ageing and Society , 28, 203– 231. doi:10.1017/S0144686X07006472 Google Scholar CrossRef Search ADS   Flynn, K. E., & Smith, M. A. ( 2007). Personality and health care decision-making style. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 62, 261– 267. doi:10.1093/geronb/62.5.P261 Google Scholar CrossRef Search ADS   Fortin P. R. Penrod J. R. Clarke A. E. St-Pierre Y. Joseph L. Bélisle P., … Katz J. N. ( 2002). Timing of total joint replacement affects clinical outcomes among patients with osteoarthritis of the hip or knee. Arthritis and Rheumatism , 46, 3327– 3330. doi: 10.1002/art.10631 Google Scholar CrossRef Search ADS PubMed  Gage E. A. ( 2013). Social networks of experientially similar others: Formation, activation, and consequences of network ties on the health care experience. Social Science & Medicine (1982) , 95, 43– 51. doi: 10.1016/j.socscimed.2012.09.001 Google Scholar CrossRef Search ADS PubMed  Gooberman-Hill, R., Sansom, A., Sanders, C. M., Dieppe, P. A., Horwood, J., Learmonth, I. D., ... Donovan, J. L. (2010). Unstated factors in orthopaedic decision-making: A qualitative study. BMC Musculoskeletal Disorders, 11, 213. doi:10.1186/1471-2474-11-213 Hawker G. A. Wright J. G. Badley E. M., & Coyte P. C. ( 2004). Perceptions of, and willingness to consider, total joint arthroplasty in a population‐based cohort of individuals with disabling hip and knee arthritis. Arthritis Care & Research , 51, 635– 641. doi: 10.1002/art.20524 Google Scholar CrossRef Search ADS   Hawker G. A. Wright J. G. Coyte P. C. Williams J. I. Harvey B. Glazier R., & Badley E. M. ( 2000). Differences between men and women in the rate of use of hip and knee arthroplasty. The New England Journal of Medicine , 342, 1016– 1022. doi: 10.1056/NEJM200004063421405 Google Scholar CrossRef Search ADS PubMed  House J. S. Umberson D., & Landis K. R. ( 1988). Structures and processes of social support. Annual Review of Sociology , 293– 318. doi: 10.1146/annurev.so.14.080188.001453 Lawson E. H. Gibbons M. M. Ingraham A. M. Shekelle P. G., & Ko C. Y. ( 2011). Appropriateness criteria to assess variations in surgical procedure use in the United States. Archives of Surgery (Chicago, Ill.: 1960) , 146, 1433– 1440. doi: 10.1001/archsurg.2011.581 Google Scholar CrossRef Search ADS PubMed  Linder, S. K., Swank, P. R., Vernon, S. W., Mullen, P. D., Morgan, R. O., & Volk, R. J. ( 2011). Validity of a low literacy version of the Decisional Conflict Scale. Patient education and counseling , 85, 521– 524. doi:10.1016/j.pec.2010.12.012 Google Scholar CrossRef Search ADS PubMed  Loewenstein G. ( 2000). Emotions in economic theory and economic behavior. American Economic Review , 426– 432. doi: 10.1257/aer.90.2.426 Loewenstein G. ( 2005). Projection bias in medical decision making. Medical Decision Making , 25, 96– 105. doi: 10.1177/0272989X04273799 Google Scholar CrossRef Search ADS PubMed  Martire L. M. Lustig A. P. Schulz R. Miller G. E., & Helgeson V. S. ( 2004). Is it beneficial to involve a family member? A meta-analysis of psychosocial interventions for chronic illness. Health Psychology , 23, 599– 611. doi: 10.1037/0278-6133.23.6.599 Google Scholar CrossRef Search ADS PubMed  Mata R., & Nunes L. ( 2010). When less is enough: Cognitive aging, information search, and decision quality in consumer choice. Psychology and Aging , 25, 289– 298. doi: 10.1037/a0017927 Google Scholar CrossRef Search ADS PubMed  McHugh G. A., & Luker K. A. ( 2009). Influences on individuals with osteoarthritis in deciding to undergo a hip or knee joint replacement: A qualitative study. Disability and Rehabilitation , 31, 1257– 1266. doi: 10.1080/09638280802535129 Google Scholar CrossRef Search ADS PubMed  Messeri P. Silverstein M., & Litwak E. ( 1993). Choosing optimal support groups: A review and reformulation. Journal of Health and Social Behavior , 34, 122– 137. doi:10.2307/2137239 Google Scholar CrossRef Search ADS PubMed  Noelker L. S., & Bass D. M. ( 1989). Home care for elderly persons: Linkages between formal and informal caregivers. Journal of Gerontology , 44, S63– S70. doi:10.1093/geronj/44.2.S63 Google Scholar CrossRef Search ADS PubMed  O’Connor A. M. ( 1995). Validation of a decisional conflict scale. Medical Decision Making , 15, 25– 30. doi: 10.1177/0272989X9501500105 Google Scholar CrossRef Search ADS PubMed  Parks, M. L., Hebert-Beirne, J., Rojas, M., Tuzzio, L., Nelson, C. L. & Boutin-Foster, C. ( 2014). A qualitative study of factors underlying decision making for joint replacement among African Americans and Latinos with osteoarthritis. Journal of long-term effects of medical implants , 24. doi:10.1615/JLongTermEffMedImplants.2014010428 Pillemer K., & Suitor J. J. ( 1996). “It takes one to help one”: Effects of similar others on the well-being of caregivers. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 51, S250– S257. doi: 10.1093/geronb/51B.5.S250 Google Scholar CrossRef Search ADS   Rammstedt B., & John O. P. ( 2007). Measuring personality in one minute or less: A 10-tem short version of the Big Five Inventory in English and German. Journal of Research in Personality , 41, 203– 212. doi: 10.1016/j.jrp.2006.02.001 Google Scholar CrossRef Search ADS   Riffin C. Pillemer K. Reid M. C., & Löckenhoff C. E. ( 2015). Decision support preferences among Hispanic and non-Hispanic White older adults with chronic pain. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences . Advance online publication. doi: 10.1093/geronb/gbv071. Romito F. Corvasce C. Montanaro R., & Mattioli V. ( 2011). Do elderly cancer patients have different care needs compared with younger ones? Tumori , 97, 374– 379. doi: 10.1700/912.10037 Google Scholar CrossRef Search ADS PubMed  Showalter A. Burger S., & Salyer J. ( 2000). Patients’ and their spouses’ needs after total joint arthroplasty: A pilot study. Orthopedic Nursing , 19, 49– 57. Google Scholar CrossRef Search ADS PubMed  Sjöling M. Ågren Y. Olofsson N. Hellzén O., & Asplund K. ( 2005). Waiting for surgery; living a life on hold—a continuous struggle against a faceless system. International Journal of Nursing Studies , 42, 539– 547. doi: 10.1016/j.ijnurstu.2004.09.009 Google Scholar CrossRef Search ADS PubMed  Soldz S., & Vaillant G. E. ( 1999). The Big Five personality traits and the life course: A 45-year longitudinal study. Journal of Research in Personality , 33, 208– 232. doi:10.1006/jrpe.1999.2243 Google Scholar CrossRef Search ADS   Thoits P. A. ( 2011). Mechanisms linking social ties and support to physical and mental health. Journal of Health and Social Behavior , 52, 145– 161. doi: 10.1177/0022146510395592 Google Scholar CrossRef Search ADS PubMed  Toye F. M. Barlow J. Wright C., & Lamb S. E. ( 2006). Personal meanings in the construction of need for total knee replacement surgery. Social Science & Medicine (1982) , 63, 43– 53. doi: 10.1016/j.socscimed.2005.11.054 Google Scholar CrossRef Search ADS PubMed  Vissers M. M. Bussmann J. B. de Groot I. B. Verhaar J. A., & Reijman M. ( 2013). Physical functioning four years after total hip and knee arthroplasty. Gait & Posture , 38, 310– 315. doi: 10.1016/j.gaitpost.2012.12.007 Google Scholar CrossRef Search ADS PubMed  Williams S. Weinman J. Dale J., & Newman S. ( 1995). Patient expectations: What do primary care patients want from the GP and how far does meeting expectations affect patient satisfaction? Family Practice , 12, 193– 201. doi: 10.1093/fampra/12.2.193 Google Scholar CrossRef Search ADS PubMed  Wolff J. L. Boyd C. M. Gitlin L. N. Bruce M. L., & Roter D. L. ( 2012). Going it together: Persistence of older adults’ accompaniment to physician visits by a family companion. Journal of the American Geriatrics Society , 60, 106– 112. doi: 10.1111/j.1532-5415.2011.03770.x Google Scholar CrossRef Search ADS PubMed  Zaidi R. Pfeil M. Macgregor A. J., & Goldberg A. ( 2013). How do patients with end-stage ankle arthritis decide between two surgical treatments? A qualitative study. BMJ Open , 3, e002782. doi: 10.1136/bmjopen-2013-002782 © The Author(s) 2016. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journals of Gerontology Series B: Psychological Sciences and Social Sciences Oxford University Press

Decision Support for Joint Replacement: Implications for Decisional Conflict and Willingness to Undergo Surgery

Loading next page...
 
/lp/ou_press/decision-support-for-joint-replacement-implications-for-decisional-LSErtC6G08
Publisher
Oxford University Press
Copyright
© The Author(s) 2016. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
1079-5014
eISSN
1758-5368
D.O.I.
10.1093/geronb/gbw023
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives The present study investigates age differences in the types of decision support that total joint replacement (TJR) candidates desire and receive when making the decision to pursue surgery. We consider the social structural (relationship to the patient) and experiential factors (network members’ experience with TJR) that influence individuals’ support preferences and the interactions of these factors with age. We also examine whether a lack of support is linked with increased decisional conflict and reduced willingness to undergo surgery. Method A telephone survey was conducted with 100 individuals (aged 40+) who were contemplating knee or hip replacement. Results TJR candidates desired and received decision support from health care providers, family members, and individuals who had previously undergone TJR. They reported higher deficits in informational and emotional support than in instrumental support. Overall, a lack of instrumental support was associated with greater decisional conflict; a lack of instrumental support and a lack of informational support were associated with reduced willingness to undergo TJR. Discussion Our findings point to the importance of involving both formal and informal network members in TJR discussions, and the need for informational guidance and practical assistance to reduce decisional conflict and uncertainty among individuals considering TJR. Chronic pain, Decision making, Joint replacement, Social networks, Social support Arthritis is a leading cause of pain and disability among middle-aged and older adults (CDC, 2009). Total joint replacement (TJR) has been shown to be a cost-effective (Daigle, Weinstein, Katz, & Losina, 2012) and highly efficacious treatment for this condition, both in terms of improving health-related quality of life (Ethgen, Bruyere, Richy, Dardennes, & Reginster, 2004) and physical outcomes (Vissers, Bussmann, de Groot, Verhaar, & Reijman, 2013). Although total knee and hip replacements for advanced arthritis are among the most common elective surgeries performed in the United States (Lawson, Gibbons, Ingraham, Shekelle, & Ko, 2011), empirical evidence shows considerable variation in patients’ willingness to undergo these procedures (Hawker et al., 2000). Indeed, many appropriate candidates for TJR are apprehensive to undergo TJR. This reluctance may stem from a variety of reasons, including fear of surgery (Figaro, Williams-Russo, & Allegrante, 2004), concerns about dependency following the procedure (Toye, Barlow, Wright, & Lamb, 2006), and perceptions of current pain severity (Hawker, Wright, Badley, & Coyte, 2004). Although accumulating research has considered the individual-level characteristics associated with the choice to pursue surgery, including patients’ pain symptoms and sociodemographic characteristics (i.e., age, income, gender, living arrangement, rurality, ethnicity; Hawker et al., 2000), less research has examined the broader social context or psychological processes involved in patients’ decisions or willingness to pursue TJR. Instead, studies incorporating psychosocial dimensions of TJR decision making typically focus on the link between social support and health-related quality of life after the surgery is complete (Ethgen et al., 2004). Importantly, a better understanding of the psychosocial factors underlying patients’ decisions to pursue TJR could help promote individualized decision assistance for patients considering this surgical procedure. In this paper, we contribute to the existing literature on people’s pursuit of TJR by examining the types and sources of support patients desire and receive when considering surgery. We approach our investigation from a life-span perspective by considering how patients’ age may be associated with their support preferences. In addition, we explore whether lack of support is associated with greater decisional conflict and reduced willingness to undergo surgery. Definitions Following prior literature, our study defines subjective decisional conflict as the “state of uncertainty about the course of action to take” (O’Connor, 1995, p. 25). We also examine willingness to undergo surgery; or, the self-reported likelihood that the TJR candidate will pursue surgery. With regard to social support, this study differentiates three commonly described dimensions (Cohen & Wills, 1985): informational (knowledge or advice), emotional (reassurance and empathetic listening), and instrumental support (tangible assistance with practical problems). Also following prior definitions of social support systems (Caplan, 1974), we examine both informal (family members, friends who have been through surgery) and formal sources (health care providers) in patients’ decisions to pursue joint replacement. Support for Joint Replacement Decisions Theoretical Frameworks Perspectives from psychology and social gerontology offer guidance regarding the potential mechanisms linking social networks and support provision with the choice to undergo surgery. On the one hand, theoretical frameworks propose that social structural norms and expectations may influence support provision and receipt from one’s formal and informal networks. Sociological perspectives propose that individuals’ support selection is contingent upon the salience of the relationship between the caregiver and care recipient, where kinship is a key dimension (Cantor, 1979). Family members are traditionally expected to offer practical aid (Messeri, Silverstein, & Litwak, 1993) and emotional comfort during times of stress (Cohen & Wills, 1985; Thoits, 2011). Such models have been extended to include formal networks, proposing that the “dual specialization” of informal and formal systems will promote optimal support for the impaired individual (Noelker & Bass, 1989). Health care providers, for example, may be chosen over family during times of illness (Messeri et al., 1993) and are expected to serve as expert advisors regarding health-specific resources and information (Brashers, Goldsmith, & Hsieh, 2002). On the other hand, homophily theory (Pillemer & Suitor, 1996; Thoits, 2011) argues that effective support provision is likely to stem from others who have experienced similar stressors or situations. Because experientially similar others have first-hand knowledge of coping with a particular illness or life event, they may possess the unique ability to validate others’ emotions and concerns and offer detailed advice by providing specific health-related information and knowledge (Gage, 2013; Thoits, 2011). In the present study, experiential similarity explicitly refers to individuals who have previously undergone TJR. Insights from behavioral economics also offer support for the role of experientially similar others in TJR decision making. According to predictions derived from prospect theory, optimal decision making may largely depend on an individual’s ability to accurately predict how s/he will feel in the future (Loewenstein, 2000). Experientially similar others may therefore provide a valuable window into the future by serving as means of predicting one’s future emotional states, also known as affective forecasting (Loewenstein, 2000). Applying this concept to the context of TJR, individuals considering surgery may seek the perspectives of others who have previously undergone TJR to limit projection biases (failure to project one’s present state onto one’s future state; Loewenstein, 2005) and determine their preferences based on others’ prior experience with TJR. Empirical Evidence We review the available evidence from empirical studies of TJR decision making below to illustrate how these predictions map onto prior research and inform our hypotheses regarding patients’ preferences for the types (informational, emotional, instrumental) and sources of support they desire and receive when considering TJR. Informational support With regard to informational support, qualitative studies have shown that the perspectives of individuals who have previously undergone TJR are highly influential in shaping patients’ attitudes toward surgery, both in lieu of and supplementary to physician guidance (McHugh & Luker, 2009; Riffin, Pillemer, Reid, & Löckenhoff, 2015). In addition, quantitative research has suggested that simply knowing someone who has undergone a successful joint replacement may positively influence patients’ willingness to consider the procedure themselves (Hawker et al., 2004), but that patients may be deterred from pursuing surgery if the procedure did not go well for the other (McHugh & Luker, 2009). Because individuals who have been through a successful versus unsuccessful surgery appear to bear differential influences on prospective TJR patients’ decisional outcomes, our study considers these two subgroups separately. Emotional support With regard to emotional support, a recent review and meta-analysis reported that family members positively influence patient well-being through empathic interactions and compassionate care for their loved one (Martire, Lustig, Schulz, Miller, & Helgeson, 2004), which in turn may have implications for patients’ treatment decisions regarding surgery. For example, a study of osteoarthritis patients considering TJR found that emotional encouragement by family members facilitated patients’ choice to undergo the procedure (McHugh & Luker, 2009). Other research has shown that having emotional support provided by close others helped TJR patients maintain continuity throughout the decisional process (Sjöling, Ågren, Olofsson, Hellzén, & Asplund, 2005). Instrumental support TJR almost inevitably requires practical assistance by others (Showalter, Burger, & Salyer, 2000). Yet, with rare exceptions (McHugh & Luker, 2009; Sjöling et al., 2005; Toye et al., 2006), studies have overlooked this important support domain. Critically, concerns about dependency following surgery (e.g., trouble with activities of daily living [ADLs]) may play a vital role in the decision-making process (Toye et al., 2006). For example, TJR candidates tend to solicit instrumental support by family members prior to surgery, which in turn is linked with patients’ sense of control over their decision to undergo the procedure (Sjöling et al., 2005). Based on the theoretical considerations and empirical evidence described above, we hypothesize that TJR candidates will primarily desire and receive informational support from their health care providers and others who have been through TJR, emotional reassurance from informal network ties, including family members and individuals who have been through TJR, and instrumental aid from close relatives. Age-Related Differences in Decision Support Preferences In addition to investigating social structural and experiential factors, the present study also considers how patients’ age may play a role in their decision-making processes. As a guiding framework, socioemotional selectivity theory (SST) offers a life-span perspective of how age-related shifts in time horizons may result in age differences in social support preferences (Carstensen, 2006). Specifically, SST posits that in young adulthood, when time horizons are perceived as expansive, information acquisition is prioritized. With advanced age, however, time horizons constrict and individuals prioritize emotionally meaningful experiences and social relationships over knowledge gathering. This developmental pattern of prioritizing emotional meaning over information has been confirmed in a recent meta-analysis considering age differences in information seeking and decision quality: in general, older adults tend to seek less pre-decisional information compared with younger individuals (Mata & Nunes, 2010). Whereas SST provides a framework for the types of support individuals desire, dyadic exchanges between older adults and their social network members may influence the types of support older adults receive. According to the Social Input Model (Fingerman & Charles, 2010), social network members tend to be more gentle with older adults and offer them preferential treatment, which may stem from a variety of reasons varying from negative stereotypes about aging to respect for one’s elders. Empirical work has substantiated this prediction, revealing that social network members often reinforce satisfying and meaningful relationships for older adults (Fingerman & Pitzer, 2007). In the context of TJR, qualitative reports by physicians also reveal a pronounced hesitation to offer negative feedback to older adults, especially when the patient exhibits signs of fatigue or concern (Gooberman-Hill et al., 2010). Overall, this dyadic process may contribute to older adults receiving greater emotional and less informational support when considering TJR than middle-aged individuals. Implications for Decision Conflict and Willingness to Undergo Surgery In addition to examining the types and sources of support that patients may desire and receive when considering TJR, we further conjecture that a lack of adequate support may have important implications for decisional conflict and the willingness to pursue surgery. Although little research has explicitly investigated the link between social support and preoperative decisional conflict or willingness to undergo TJR, qualitative research with osteoarthritis patients awaiting surgery offers preliminary evidence of this association. Specifically, this research demonstrated patients’ frustration and confusion when they failed to receive the informational guidance they desired from their providers (Sjöling et al., 2005), but that having instrumental assistance by family members helped patients feel more secure in their decisions. In related work, research with primary care patients has shown that when patients’ support expectations are met by their health care providers, they report having significantly higher rates of satisfaction in their clinical interactions (Williams, Weinman, Dale, & Newman, 1995). In turn, patients’ satisfaction regarding communication with their providers has been associated with greater decisional clarity and confidence in patients’ decision making (Edwards et al., 2004). Other research with arthritis patients similarly suggests that a high-quality patient–provider relationship, combined with clear and balanced information, is crucial to promoting decisional clarity when considering whether to undergo surgery (Zaidi, Pfeil, Macgregor, & Goldberg, 2013). For these reasons, we expect that lack of support will be associated with greater decisional conflict and reduced willingness to pursue surgery among patients. The Present Study This study investigates the relative influence of structural (relationship to respondent) and experiential factors (contact with others who have undergone prior successful vs unsuccessful TJR) in determining the types of social support (informational, emotional, instrumental) that patients desire and receive when making the decision to undergo TJR. We also explore potential age differences in patients’ preference for emotional versus informational support and examine whether lack of support is linked with TJR candidates’ subjective decisional conflict or willingness to undergo surgery. All analyses include key covariates previously found to be associated with decision making and social network structure. Five-factor personality traits have been linked with social network formation (Soldz & Vaillant, 1999) and health care decision-making styles (Flynn & Smith, 2007). In addition, poor health has been identified as a strong predictor of a weakened social network; therefore, physical health is included as a covariate (cf. House, Umberson, & Landis, 1988). Pain level and pain location (knee vs hip) are also included given that increased pain has been linked with patients’ willingness to undergo surgery (Brander et al., 2003) and that knee (vs hip) replacement has been associated with less favorable pain outcomes after surgery (Beswick, Wylde, Gooberman-Hill, Blom, & Dieppe, 2012). Consistent with previous methodology, the total number of decision network members (DNMs) is included as a covariate to adjust for differences in respondents’ network size. Finally, because prior TJR experience may influence individuals’ perceptions of and willingness to undergo a second TJR (Ballantyne, Gignac, & Hawker, 2007), this study includes only respondents who have not previously undergone joint replacement. Based on the theoretical considerations outlined above, we propose the following hypotheses: Hypothesis 1: Composition of Decision Support Network. TJR candidates’ decision support networks will include not only include formal ties with health care providers but also informal ties with family members and nonrelatives, including individuals who have previously undergone TJR. Hypothesis 2: Support Desired and Received. TJR candidates will desire and receive informational support from health care providers and from individuals who have previously undergone TJR. Emotional support will be sought from family members and others who have previously undergone TJR, and instrumental support will be desired and received from family members and individuals who have not been through TJR. Hypothesis 3: Age Differences. Older age will be associated with a greater desire for and receipt of emotional rather than informational support from all sources. Hypothesis 4: Decision Conflict and Willingness to Undergo Surgery. A lack of support will be associated with greater decision conflict and reduced willingness to undergo TJR. Given the limited research in this area, we do not make concrete predictions regarding specific types of support. Method Participants A variety of strategies were used to recruit participants at various stages in the decision-making process, from contemplation (respondents who were still considering whether or not to pursue surgery) to active pursuit of joint replacement (respondents who had set a specific date for the procedure). A primary recruitment method was an electronic search of EpicCare health records at two outpatient practices in New York City. This approach was selected for efficiency reasons: Querying the database for patients with a diagnosis of hip or knee osteoarthritis allowed us to more easily identify a subgroup of individuals appropriate for the study (i.e., those who were potential candidates for TJR). These individuals were contacted via postal mail to inform them of the study and invite their participation. This strategy was supplemented by online advertising and posting flyers at local senior centers, senior housing residences, and physician practices. The study protocol was approved by the Cornell University and Weill Cornell Medical College Institutional Review Boards. Individuals were excluded if they were younger than age 40, not fluent in English, had previously undergone joint replacement, or exhibited cognitive impairment, defined by a score of less than 3 on a six-item screener (Callahan, Unverzagt, Hui, Perkins, & Hendrie, 2002). Of the 296 individuals contacted, 72 declined participation and 97 were excluded due to cognitive impairment or prior joint replacement. The final sample consisted of 100 individuals, 84% of whom were women and 72% were non-Hispanic white(Table 1). However, the data for one participant were lost due to a Qualtrics recording failure. Therefore, analyses were conducted on a sample of 99 individuals. Table 1. Respondent Characteristics     n  Demographics   Age (SD)  66.6 (10.6)  99   % women  83.7%  99   % white  72.4%  99   % with college degree  64.6%  99   Physical health  3.3 (1.0)  99   Mental health  3.8 (1.0)  98   # DNMs  2.8 (1.7)  92  Pain location   % knee pain  81.6%  99   % hip pain  18.4%  99  % considered surgery  93.9%  99  % recommended for surgery by physician  71.4%  99  Personality   Extraversion (SD)  5.4 (1.5)  99   Agreeableness (SD)  7.3 (1.4)  98   Conscientiousness (SD)  8.0 (1.2)  99   Neuroticism (SD)  5.3 (1.7)  99   Openness (SD)  7.5 (1.5)  99  Support desired and received (summed across DNM)   Informational support desired (SD)  2.6 (0.8)  92   Informational support received (SD)  2.4 (0.9)  91   Emotional support desired (SD)  2.6 (1.0)  90   Emotional support received (SD)  2.4 (1.0)  90   Instrumental support desired (SD)  2.1 (1.1)  90   Instrumental support received (SD)  1.9 (1.1)  90  Decision conflict (SD)  27.3 (24.3)  90  Willingness to undergo TJR (SD)  2.3 (1.1)  99      n  Demographics   Age (SD)  66.6 (10.6)  99   % women  83.7%  99   % white  72.4%  99   % with college degree  64.6%  99   Physical health  3.3 (1.0)  99   Mental health  3.8 (1.0)  98   # DNMs  2.8 (1.7)  92  Pain location   % knee pain  81.6%  99   % hip pain  18.4%  99  % considered surgery  93.9%  99  % recommended for surgery by physician  71.4%  99  Personality   Extraversion (SD)  5.4 (1.5)  99   Agreeableness (SD)  7.3 (1.4)  98   Conscientiousness (SD)  8.0 (1.2)  99   Neuroticism (SD)  5.3 (1.7)  99   Openness (SD)  7.5 (1.5)  99  Support desired and received (summed across DNM)   Informational support desired (SD)  2.6 (0.8)  92   Informational support received (SD)  2.4 (0.9)  91   Emotional support desired (SD)  2.6 (1.0)  90   Emotional support received (SD)  2.4 (1.0)  90   Instrumental support desired (SD)  2.1 (1.1)  90   Instrumental support received (SD)  1.9 (1.1)  90  Decision conflict (SD)  27.3 (24.3)  90  Willingness to undergo TJR (SD)  2.3 (1.1)  99  Notes. DNM = decision network member; TJR = total joint replacement; SD = standard deviation. Willingness to undergo TJR was assessed on a 5-point Likert scale from 1 = certainly will not pursue surgery to 5 = certainly will pursue surgery. The Decision Conflict Scale ranges from 0 = no decisional conflict to 100 = extremely high decisional conflict. Support desired and received was measured by a 4-point Likert scale ranging from 1 = never to 4 = very often. View Large Measures Primary predictor and outcome variables Social network composition and function. Following previous methodology (Antonucci & Akiyama, 1987; Pillemer & Suitor, 1996), information on the structure and function of participants’ decision support networks was obtained through a series of name-elicitation questions. For each DNM, data were collected on (a) demographic characteristics including age and gender (0 = male; 1 = female), (b) relationship to respondent (open-response coded 1 = family member; 2 = health care provider; 3 = non-family close other), and (c) prior experience with TJR (1 = prior successful TJR; 2 = prior unsuccessful TJR; 3 = no prior TJR). From this point forward, we refer to this variable as underwent TJR. Types of support desired and received. Participants were asked to indicate how often they desired and received support from each DNM. To assess informational support desired and received, respondents were asked “How often does [DNM] give you advice or suggestions about your decision to get a joint replacement (never, sometimes, often, or very often)?” Followed by “Now, think about whether you actually want this type of support from [DNM]. If it were up to you, how often would you want advice or suggestions from [DNM] (1 = never, 2 = sometimes, 3 = often, or 4 = very often)?” Analogous questions were used to examine how often participants received and desired “comfort and reassurance” (emotional support) and “practical aid and assistance” (instrumental support) from each DNM. To capture lack of support, difference scores were calculated for each type of support provided by each DNM (informational, emotional, instrumental) by subtracting the specific types of support received from the type of support desired. Following methodology used previously to examine social ties and support among older adults (Fiori, Antonucci, & Akiyama, 2008), each type of support received was averaged within each DNM category (e.g., health care provider) for each respondent. For example, if two health care providers were named by a single respondent, the amount of emotional support received from each was averaged within the DNM category “health care provider.” The same procedure was used for all other DNM categories and types of support desired. Difference scores were also averaged within each DNM category. Identical analyses were also conducted with support summed (instead of averaged) within each category. Results were comparable using both methods. Decision conflict. Participants responded to the low-literacy 10-item version of the Decision Conflict Scale (DCS) (Linder et al., 2011). The DCS targets uncertainty about choosing among alternatives and modifiable factors contributing to uncertainty (e.g., feeling uninformed, unclear about values). This scale has been shown to reliably discriminate between individuals who make and delay decisions (present study α = .84). Willingness to undergo surgery. All participants were asked the question “How certain are you that you’ll get a knee/hip replacement?” (1 = certainly will not, 2 = probably will not, 3 = could go either way, 4 = probably will, 5 = certainly will). Covariates Personality. Respondents’ personality was assessed using the 10-Item Short Version of the Big Five Inventory (BFI-10; Rammstedt & John, 2007). Individuals rated the extent to which each personality characteristic was true of them, ranging from 1 (strongly disagree) to 5 (strongly agree). The BFI-10 is an acceptable measure of personality when time is limited, as in telephone surveys (Rammstedt & John, 2007). Correlations between the personality characteristics ranged from −.60 to .11. Physical and mental health. Two questions from the PROMIS Global Health Scale (Cella et al., 2010) were used to assess participants’ overall self-reported physical and mental health (1 = poor, 2 = fair, 3 = good, 4 = very good, 5 = excellent). Pain level, function, and location. Pain was assessed using an adapted version of the American Academy of Orthopaedic Surgeons Lower Limb Core Hip/Knee Module (MODEMS, 1996), a region-specific scale measuring pain level and ability to perform daily tasks. Respondents were asked about their pain levels and ability to function when walking on a flat surface, going up/down stairs, sitting/lying down, putting on/off shoes or socks, toileting and bathing, and carrying out light housework. Response options for these items ranged from 0 (no pain) to 5 (extreme pain). The six items were summed to form a pain severity scale (present study α = .66). In addition, participants were asked three questions (also adapted from MODEMS) to assess whether they experienced knee pain (pain or stiffness in one or both of their knee(s); 0 = no, 1 = yes) or hip pain (pain in the top of their thigh or groin, or front section of their thigh; 0 = no, 1 = yes). Sociodemographic characteristics. Respondent sociodemographic characteristics included age, race/ethnicity, gender, education, and marital status. Procedure After providing informed oral consent, each participant completed a structured telephone interview lasting approximately 60min administered by a trained research assistant. Respondents’ sociodemographic information was first collected, followed by an assessment of their current pain level, pain location, and willingness to undergo TJR. Information was next collected on the patient’s social network composition and structure, the decisional support they desired and received, and decisional conflict. The final portion of the survey included measures of respondents’ personality. Several additional self-report questionnaires, not related to the core questions of this study, were also included in the survey (health literacy, aging stereotypes, and risk preferences). Participants’ responses were entered into Qualtrics software, Version 2013 (Qualtrics, Provo, UT). Data Analysis Sample overview Descriptive and univariate analyses were conducted to gain an initial understanding of TJR candidates’ sociodemographic, pain, and personal characteristics. A two-way repeated measures analysis of variance examined differences in TJR candidates’ desire for and receipt of each type of decision support (informational, emotional, instrumental) as well as differences across the types of support. In addition, bivariate correlations examined associations among respondent-level characteristics (age, gender, general health, pain level, pain location, personality, and network size), decision conflict and willingness to undergo surgery, and support desired and received. Hypothesis 1: Descriptive and univariate analyses were conducted to examine TJR candidates’ decision network composition; specifically, the average number of DNM in respondents’ networks, percentage of formal versus informal network ties, and percentage of DNM who had previously undergone TJR. Hypotheses 2 and 3: Hypotheses concerning TJR candidates’ desired and received support were conducted using a multilevel modeling framework. This type of structure was necessary to allow for simultaneous estimation of respondent-level and network-level effects and to accommodate the uneven distributions of social networks across participants. Two separate multilevel models were estimated for the averaged support that TJR candidates desired (Table 2, Model 1) and received (Table 2, Model 3) as dependent variables, with type of support specified as the index variable. Substantive predictor variables and covariates that reached significance in preliminary bivariate analyses were entered into each model. Table 2. Multilevel Models Predicting TJR Candidates’ Desired (Models 1 and 2) and Received (Models 3 and 4) Decision Support Parameter estimates  Model 1  Model 2  Model 3  Model 4  F(df 1, df 2)  F(df 1, df 2)  F(df 1, df 2)  F(df 1, df 2)  Intercept  11.78 (1, 74)***  12.71 (1, 72)***  5.67 (1, 74)*  5.49 (1,75)*   Type of support  15.64 (2, 349)***  13.23 (2,183)***  17.78 (2, 351)***  13.97 (2, 161)***   DNM relationship to respondent  1.84 (2, 377)  2.40 (2, 284)†  3.10 (2, 374)*  3.61 (2, 275)*   DNM underwent TJR  3.66 (2, 366)*  4.76 (2, 276)**  3.21 (2, 363)*  3.67 (1, 266)*   Respondent age  1.84 (1, 66)  1.66 (1, 69)  1.27 (1, 67)  1.14 (1, 68)  Covariates   Respondent marital status (married)  0.66 (1, 59)  0.62 (1, 59)  0.39 (1, 59)  0.34 (1, 58)   Respondent extraversion  0.00 (1, 56)  0.01 (1, 60)  0.31 (1, 57)  0.34 (1, 58)   Respondent agreeableness  2.10 (1, 65)  1.96 (1, 64)  0.22 (1, 66)  0.16 (1, 64)   Respondent openness  0.99 (1, 58)  0.73 (1, 62)  1.29 (1, 58)  1.19 (1, 61)   Network size  0.26 (1, 326)  0.30 (1, 62)  0.00 (1, 322)  0.05 (1, 265)   DNM age  1.01 (1, 273)  1.57 (1, 265)  0.01 (1, 272)  0.04 (1, 245)  Interactions   Type of support × DNM relationship  —  1.56 (4, 333)  —  1.21 (4, 336)   Type of support × DNM underwent TJR  —  3.30 (4, 320)**  —  3.87 (4, 319)**   Respondent age × type of support  —  0.88 (2, 150)  —  0.91 (2, 129)   Respondent age × underwent TJR  —  0.51 (2, 280)  —  1.19 (2, 270)  Parameter estimates  Model 1  Model 2  Model 3  Model 4  F(df 1, df 2)  F(df 1, df 2)  F(df 1, df 2)  F(df 1, df 2)  Intercept  11.78 (1, 74)***  12.71 (1, 72)***  5.67 (1, 74)*  5.49 (1,75)*   Type of support  15.64 (2, 349)***  13.23 (2,183)***  17.78 (2, 351)***  13.97 (2, 161)***   DNM relationship to respondent  1.84 (2, 377)  2.40 (2, 284)†  3.10 (2, 374)*  3.61 (2, 275)*   DNM underwent TJR  3.66 (2, 366)*  4.76 (2, 276)**  3.21 (2, 363)*  3.67 (1, 266)*   Respondent age  1.84 (1, 66)  1.66 (1, 69)  1.27 (1, 67)  1.14 (1, 68)  Covariates   Respondent marital status (married)  0.66 (1, 59)  0.62 (1, 59)  0.39 (1, 59)  0.34 (1, 58)   Respondent extraversion  0.00 (1, 56)  0.01 (1, 60)  0.31 (1, 57)  0.34 (1, 58)   Respondent agreeableness  2.10 (1, 65)  1.96 (1, 64)  0.22 (1, 66)  0.16 (1, 64)   Respondent openness  0.99 (1, 58)  0.73 (1, 62)  1.29 (1, 58)  1.19 (1, 61)   Network size  0.26 (1, 326)  0.30 (1, 62)  0.00 (1, 322)  0.05 (1, 265)   DNM age  1.01 (1, 273)  1.57 (1, 265)  0.01 (1, 272)  0.04 (1, 245)  Interactions   Type of support × DNM relationship  —  1.56 (4, 333)  —  1.21 (4, 336)   Type of support × DNM underwent TJR  —  3.30 (4, 320)**  —  3.87 (4, 319)**   Respondent age × type of support  —  0.88 (2, 150)  —  0.91 (2, 129)   Respondent age × underwent TJR  —  0.51 (2, 280)  —  1.19 (2, 270)  Notes. DNM = decision network member; TJR = total joint replacement. Age dichotomized into 0 = 65 years or younger; 1 = 66 years or older. †p < .10. *p < .05. **p < .01. ***p < .001. Four interaction terms, (a) type of support × DNM relationship to respondent, (b) type of support × underwent TJR, (c) respondent age × type of support, and (d) respondent age × underwent TJR, were then added to allow for comparisons between the types and sources of support that respondents desired and received (Hypothesis 2) and variations by age (Hypothesis 3). In these models (Table 2, Models 2 and 4), age was dichotomized into individuals 65 years and younger and individuals 66 and older. (Analyses using age as a continuous variable yielded comparable results.) If an interaction reached significance, pairwise comparisons using a Bonferroni correction were conducted. Hypothesis 4: The final set of analyses examined TJR candidates’ decisional conflict and willingness to undergo surgery as outcome variables. Difference scores for each type of support were then entered as independent variables into two separate generalized linear models (GLMs) with decision conflict (Table 3) and willingness (Table 4) to undergo surgery as the dependent variables. Each model also included DNM relationship to respondent and underwent TJR, and covariates that were found to be significant in preliminary bivariate analyses. Table 3. General Linear Model Predicting Decision Conflict Variable  β (SE)  t  Difference scores in support desired vs received   Informational support  0.28 (2.92)  0.10   Emotional support  4.34 (3.81)  1.14   Instrumental support  10.41 (3.75)**  2.78  Covariates   Respondent gender (female)  −18.89 (5.88)**  −3.21   Respondent marital status (married)  4.60 (4.05)  1.13   Respondent conscientiousness  −3.95 (1.76)*  1.13   Respondent openness  −3.65 (1.22)**  −2.99   Network size  −0.74 (2.36)  −0.31   DNM relationship to respondent  4.40 (2.38)  1.85   DNM underwent TJR  1.63 (2.44)†  0.67  Variable  β (SE)  t  Difference scores in support desired vs received   Informational support  0.28 (2.92)  0.10   Emotional support  4.34 (3.81)  1.14   Instrumental support  10.41 (3.75)**  2.78  Covariates   Respondent gender (female)  −18.89 (5.88)**  −3.21   Respondent marital status (married)  4.60 (4.05)  1.13   Respondent conscientiousness  −3.95 (1.76)*  1.13   Respondent openness  −3.65 (1.22)**  −2.99   Network size  −0.74 (2.36)  −0.31   DNM relationship to respondent  4.40 (2.38)  1.85   DNM underwent TJR  1.63 (2.44)†  0.67  Notes. DNM = decision network member; SE = standard error; TJR = total joint replacement. The Decision Conflict Scale ranges from 0 = no decisional conflict to 100 = extremely high decisional conflict. †p < .10. *p < .05. **p < .01. Table 4. General Linear Model Predicting Willingness to Undergo Surgery Variable  β (SE)  t  Difference scores in support desired vs received     Informational support  −0.49 (0.12)***  −3.97   Emotional support  −0.12 (0.16)  −0.75   Instrumental support  −0.14 (0.16)*  −1.84  Covariates   Respondent age  −0.04 (0.01)***  −5.01   Respondent marital status (married)  0.17 (0.17)  0.98   Respondent pain level  0.07 (0.02)**  3.07   Network size  0.02 (0.10)  0.20   DNM relationship to respondent  −0.04 (0.10)  −0.41   DNM underwent TJR  −0.11 (0.10)  −1.03  Variable  β (SE)  t  Difference scores in support desired vs received     Informational support  −0.49 (0.12)***  −3.97   Emotional support  −0.12 (0.16)  −0.75   Instrumental support  −0.14 (0.16)*  −1.84  Covariates   Respondent age  −0.04 (0.01)***  −5.01   Respondent marital status (married)  0.17 (0.17)  0.98   Respondent pain level  0.07 (0.02)**  3.07   Network size  0.02 (0.10)  0.20   DNM relationship to respondent  −0.04 (0.10)  −0.41   DNM underwent TJR  −0.11 (0.10)  −1.03  Notes. DNM = decision network member; SE = standard error; TJR = total joint replacement. Willingness to undergo TJR was assessed by a 5-point Likert scale from 1 = certainly will not pursue surgery to 5 = certainly will pursue surgery. *p < .05. **p < .01. ***p < .001. Results Sample Overview Of the total sample, 82% had knee pain and 18% had hip pain. With regard to willingness to undergo surgery, approximately one-fifth (n = 22) were certain they would pursue TJR, 27% (n = 22) were fairly certain they would pursue TJR, 28% (n = 28) were undecided, 15% (n = 15) were fairly certain that they would not pursue TJR, and the remaining 7 (7%) were certain they would not pursue TJR. Participants generally reported good mental health and average physical health, aside from their pain condition. See Table 1 for sample characteristics. In the repeated measures analysis examining differences in TJR candidates’ desire for and receipt of each type of decision support, the overall F statistic was significant, F(5, 153) = 46.0, p < .001. Post hoc comparisons showed that TJR candidates received less informational support than they desired (p < .001), less emotional support than they desired (p < .001), and less instrumental support than they desired (p < .001). Across support types, respondents reported higher deficits in informational and emotional support than in instrumental support. Hypothesis 1: TJR candidates’ decision support networks will include both formal and informal ties Of the 99 respondents, 92 (92%) reported consulting at least one DNM in the decision to pursue surgery. The seven respondents who did not list anyone in their decision network were excluded from subsequent multilevel analyses examining types of support desired and received. These individuals were similar to the rest of the sample with respect to key variables (age, decisional conflict, or willingness to undergo TJR). On average, respondents consulted 2.75 network members regarding the choice to undergo TJR (range: 0–9). Decision networks were evenly composed of family members (34%), providers (33%), and nonrelatives (33%). With regard to prior experience with TJR, respondents’ networks were split between individuals who had previously undergone surgery (32% successful TJR; 23% unsuccessful TJR) and individuals who had not undergone the procedure (45%). Hypothesis 2: Types of support desired and received Multilevel models indicated that the interaction, type of support × underwent TJR, was significant for both support desired (Table 2, Model 2) and received (Table 2, Model 4). The significant pairwise comparisons of this effect are described below. Because the interaction, type of support × DNM relationship, failed to reach significance (Table 2), pairwise comparisons were not conducted. Hypothesis 2a: TJR candidates will desire and receive informational support from health care providers and from individuals who have previously undergone TJR Informational support desired Pairwise comparisons showed that respondents desired more informational support from DNM who had undergone a successful surgery (Predicted Mean = 2.71, SE = 0.17) than from DNM who never had the procedure (Predicted Mean = 2.28, SE = 0.16, p < .05). Informational support received Respondents were also more likely to receive informational support from DNM who had previously undergone a successful surgery (Predicted Mean = 2.54, SE = 0.16) than from DNM who did not have the procedure (Predicted Mean = 2.04, SE = 0.15, p < .05). Hypothesis 2b: Emotional support will be desired and received from family members and others who have previously undergone TJR Emotional support desired and received Pairwise comparisons failed to detect differences across groups, suggesting that respondents were equally likely to desire and receive emotional support from all DNM. Hypothesis 2c: Instrumental support will be desired and received from family members and individuals who have not been through TJR Instrumental support desired Pairwise comparisons showed that respondents were more likely to desire instrumental support from DNM who had not previously undergone surgery (Predicted Mean = 2.02, SE = 0.17, p < .01) than from DNM who had undergone an unsuccessful surgery (Predicted Mean = 1.48, SE = 0.19, p < .01). Instrumental support received A similar pattern of effects was detected for instrumental support received. Respondents were more likely to receive this type of support from DMN who had not previously undergone surgery (Predicted Mean = 2.11, SE = 0.15) than from DNM who had undergone a successful (Predicted Mean = 1.77, SE = 0.16, p < .05) or unsuccessful TJR (Predicted Mean = 1.42, SE = 0.26, p < .05). Hypothesis 2: Additional Findings All multilevel models (Table 2) showed significant main effects for type of support, indicating that respondents desired and received certain types of support more than others. Underwent TJR also reached significance in all four models: respondents desired and received less support from DNM who had previously undergone an unsuccessful surgery compared with the reference group, DNM who never had the procedure. Finally, DNM relationship to respondent was significantly associated only with support received (Models 3 and 4). Respondents received more support from family members than from the reference group, nonrelatives. Hypothesis 3: Increased age will be associated with a greater desire for and receipt of emotional rather than informational support from all sources. Multilevel models were also conducted to examine Hypothesis 3. As shown in Table 2 (Models 2 and 4), neither of the age interactions (respondent age × type of support or age × underwent TJR) were significant. Therefore, post hoc comparisons were not conducted. Hypothesis 4: A lack of support will be associated with greater decision conflict and reduced willingness to undergo TJR. Decisional conflict GLM analyses (Table 3) indicated that a lack of instrumental support was associated with greater decisional conflict (B = 10.41, p < .01). In addition, women were less likely to experience decisional conflict (B = −18.89, p < .01), as were individuals who scored high on the personality traits of Conscientiousness (B = −3.95, p < .05) and Openness (B = −3.65, p < .01). Willingness to undergo surgery As shown in Table 4, results from GLM analyses showed that a lack of informational support (B = −.49, p < .001) and lack of instrumental support (B = −.14, p < .05) were both associated with reduced willingness to pursue TJR. In addition, TJR candidates who were middle-aged (B = −.04. p < .001) and those with greater pain (B = .07, p < .01) were more willing to undergo TJR. Discussion The present study investigated the types of decision support that patients desire and receive when considering TJR, as well as potential variations by age. Analyses also examined whether lack of support was associated with patients’ decisional conflict and willingness to undergo surgery. Overall, results offer partial support for our hypotheses. One key finding from this study was the general lack of support reported by TJR candidates. Across the three support dimensions (informational, emotional, instrumental), respondents desired more support than they received. As shown in our study and elsewhere (Sjöling et al., 2005), patients’ unmet needs leading up to surgery may influence their ability to commit to the procedure. With regard to decision network composition (Hypothesis 1), we found that—as expected—family members and other informal contacts were routinely included along with medical providers. With regard to prior experience with TJR, half of respondents consulted at least one social network member who had previously undergone joint replacement. This finding corroborates qualitative research documenting patients’ reliance on others who have been through TJR (Parks et al., 2014; Zaidi et al., 2013). Our predictions regarding Hypothesis 2 were generally supported. However, contrary to expectation, TJR candidates did not desire or receive informational guidance more from their health care providers than from their informal network ties. Nevertheless, this lack of differentiation appears to be consistent with prior qualitative research suggesting that older adults seek decision support from both formal and informal contacts when considering pain treatment options (Riffin et al., 2015). In the present study, respondents desired and received informational support more from individuals who had previously undergone a successful TJR than from DNM who never had the procedure; a finding that points to a potential “positivity preference” among TJR candidates in their desire for informational guidance from similar others with positive outcomes. Although we had expected that TJR candidates would desire and receive more emotional support from family members and from others who had previously undergone TJR, our analyses failed to detect differences across groups. Finally, as predicted, respondents desired instrumental support more from DNM who had not previously undergone surgery. Contrary to expectation, we found no significant age differences in respondents’ desires for and receipt of emotional versus informational support for TJR (Hypothesis 3). However, it is important to note that most studies comparing age differences in information seeking consider differences between young (ages 18–40) and older respondents (age 65+). In contrast, our study examined a more restricted age range (age 40+). Other studies have also found that older and middle-aged patients actually have similar informational and relational needs (Romito, Corvasce, Montanaro, & Mattioli, 2011), especially with regard to medical information (Mata & Nunes, 2010). It is also plausible that having functional limitations due to pain may have primed limited time horizons even among middle-aged adults and therefore altered respondents’ preference for emotional rather than informational support. Analyses examining patients’ decisional conflict and willingness to undergo TJR (Hypothesis 4) indicated that a lack of informational and instrumental support was associated with greater decisional conflict and reduced willingness to undergo surgery. Thus, obtaining sufficient information about the risks and benefits of the procedure appears to play a critical role in patients’ decisions (Zaidi et al., 2013). Further, having adequate instrumental aid may help to alleviate feelings of uncertainty associated with TJR—and ultimately—determine whether surgery is a tenable option for the patient. Consistent with this idea, studies within the caregiving literature suggest that informal assistance with ADLs is an essential element in planning patients’ medical care (Wolff, Boyd, Gitlin, Bruce, & Roter, 2012), and that insufficient help may have negative consequences for older adults (Depalma et al., 2013). Thus, instrumental support may be a key variable to include in future studies examining social support and medical decision making. Limitations and Future Directions Overall, the results from this study should be viewed in light of several limitations including the cross-sectional design and restricted sociodemographic focus. To address the issue of cross-sectionality, 6-month follow-up assessments of the present sample are currently under way to determine whether (a) reported willingness is associated with actually pursuing surgery and (b) whether congruence in individuals’ received and desired decision support predicts their decisional regret (e.g., regret of the treatment choice) at a 6-month follow up. The present study also sets the stage for other longitudinal investigations. One line of inquiry would be to follow joint replacement candidates across the decisional trajectory, starting at the first treatment discussion of TJR, to target the modifiable factors (e.g., sufficient informational and instrumental support or access others who have already had the procedure) to promote a timely decision. This is an important line of work given prior literature demonstrating that the timing of TJR has strong implications for patients’ long-term recovery (Fortin et al., 2002). A second consideration for future research will be to include measures of specific psychological (e.g., patients’ coping styles, locus of control) and relational (e.g., relationship quality, attachment) factors that could partially explain individuals’ decision network selection and need for emotional reassurance. For example, patients who have high self-efficacy in pain management or motivation may rely on internal resources and therefore may not turn to others for emotional reassurance or practical assistance. Although prior research has examined several psychosocial characteristics in relation to postoperative outcomes, a better understanding of these variables in preoperative contexts would shed light on alternative patient-level characteristics associated with decision support selection. Finally, because our study relied exclusively on TJR candidates’ reports, we were unable to differentiate between network members who were TJR candidates but refused surgery and those who never were candidates themselves. Future studies should make efforts to collect information from DNMs themselves to gain this additional perspective. Despite the clear need for future research to address the caveats noted above, our study extends prior research in several ways. Unlike other studies, we did not restrict our sample to patients who were referred for evaluation for TJR or who were scheduled to have surgery. Instead, we included patients who fell at any point on the decisional spectrum from contemplation (unsure of whether or not they will have the procedure) to active pursuit of TJR (set a date for the surgery) to rejection of the procedure after it had been proposed by a physician. This design allowed us to examine the types of support necessary for individuals at any stage in the decision-making process. Our study is also unique in its focus on decision support networks for TJR patients. To our knowledge, no other studies have quantitatively investigated this topic. Moreover, assessing all three support types offered a more holistic perspective than prior studies that have predominantly focused on advice seeking, generally neglecting the salient component of instrumental aid. A final contribution of this work was assessing whether lack of support was associated with TJR patients’ decisional conflict. Practice Implications Our findings also have implications for patient care. As liaisons between patients and their social networks, health care providers can assist patients in receiving adequate decision support by asking patients about the support they desire from nonmedical sources, and involving informal network members in treatment encounters, as appropriate. Physicians should therefore work with the patient and their informal network to ensure adequate practical assistance leading up to and following the procedure. A second recommendation for physicians is to ask patients whether they have received adequate information about the procedure, and not, what additional information the patient would find useful in helping him/her to make a decision. Finally, health care providers should be aware that patients desire as much emotional support from them as they expect from informal sources, a desire easily overlooked in busy medical settings. Taken together, the findings from this study may help physicians to understand why certain patients may or may not be ready to pursue TJR. Given that both formal and informal DNMs were sought for a spectrum of support functions, future research should aim to incorporate these various perspectives in subsequent studies of TJR decision making. Overall, this research provides an initial understanding of TJR patients’ decisional preferences and points to the need for social network–driven strategies to involve both medical and nonmedical providers in TJR discussions. Funding This work was supported by the Lawrence and Rebecca Stern Family Foundation, the Edward R. Roybal Centers for Translation of the Behavioral and Social Sciences of Aging (P30AG022845), and a President’s Council of Cornell Women Affinito Stewart Grant to C. E. Löckenhoff. C. E. Löckenhoff was also supported in part by 1R21AG043741. C. Riffin is currently supported by a National Institute on Aging Training Grant (T32AG1934). Acknowledgments We thank Elaine Wethington for her advice on analyzing social networks. We also thank Sara Hachey, Rebecca Lampert, and Shayna Ratner for their assistance with participant recruitment and interviewing. This study is based on the doctoral dissertation of C. Riffin, supervised by C. E. Löckenhoff. C. Riffin designed the study, recruited and interviewed participants, performed the analysis, and composed the manuscript. K. Pillemer, M. C. Reid, and C. Lӧckenhoff obtained funding. M. C. Reid and J. Tung supervised participant recruitment. C. Lӧckenhoff supervised data analysis and manuscript preparation. All authors provided input on the study design and approved the final manuscript. References American Academy of Orthopaedic Surgeons/Council of Musculoskeletal Specialty Societies. ( 1996, March). MODEMS Lower Limb Core, Version 1.1 . Antonucci T. C., & Akiyama H. ( 1987). Social networks in adult life and a preliminary examination of the convoy model. Journal of Gerontology , 42, 519– 527. doi: 10.1093/geronj/42.5.519 Google Scholar CrossRef Search ADS PubMed  Ballantyne P. J. Gignac M. A., & Hawker G. A. ( 2007). A patient‐centered perspective on surgery avoidance for hip or knee arthritis: Lessons for the future. Arthritis Care & Research , 57, 27– 34. doi:10.1002/art.22472 Google Scholar CrossRef Search ADS   Beswick, A. D., Wylde, V., Gooberman-Hill, R., Blom, A., & Dieppe, P. (2012). What proportion of patients report long-term pain after total hip or knee replacement for osteoarthritis? A systematic review of prospective studies in unselected patients. BMJ Open, 2, e000435. doi:10.1136/bmjopen-2011-000435 Brander, V. A., Stulberg, S. D., Adams, A. D., Harden, R. N., Bruehl, S., Stanos, S. P., & Houle, T. (2003). Predicting total knee replacement pain: A prospective, observational study. Clinical Orthopaedics and Related Research, 416, 27–36. Brashers D. E. Goldsmith D. J., & Hsieh E. ( 2002). Information seeking and avoiding in health contexts. Human Communication Research , 28, 258– 271. doi: 10.1111/j.1468-2958.2002.tb00807.x Google Scholar CrossRef Search ADS   Centers for Disease Control and Prevention (CDC). (2009). Prevalence and most common causes of disability among adults—United States, 2005. Morbidity and Mortality Weekly Report, 58, 421–426. Callahan C. M. Unverzagt F. W. Hui S. L. Perkins A. J., & Hendrie H. C. ( 2002). Six-item screener to identify cognitive impairment among potential subjects for clinical research. Medical Care , 40, 771– 781. doi: 10.1097/01.MLR.0000024610.33213.C8 Google Scholar CrossRef Search ADS PubMed  Cantor M. H. ( 1979). Neighbors and friends: An overlooked resource in the informal support system. Research on Aging , 1, 434– 463. doi:10.1177/016402757914002 Google Scholar CrossRef Search ADS   Caplan G. ( 1974). Support systems and community mental health: Lectures on concept development . New York, NY: Behavioral Publications. Carstensen L. L. ( 2006). The influence of a sense of time on human development. Science (New York, N.Y.) , 312, 1913– 1915. doi: 10.1126/science.1127488 Google Scholar CrossRef Search ADS PubMed  Cella D. Riley W. Stone A. Rothrock N. Reeve B. Yount S., … Hays R. ( 2010). The patient-reported outcomes measurement information system: 2005–2008. Journal of Clinical Epidemiology , 63, 1179– 1194. doi: 10.1016/j.jclinepi.2010.04.011 Google Scholar CrossRef Search ADS PubMed  Cohen S., & Wills T. A. ( 1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin , 98, 310– 357. doi: 10.1037/0033-2909.98.2.310 Daigle M. E. Weinstein A. M. Katz J. N., & Losina E. ( 2012). The cost-effectiveness of total joint arthroplasty: A systematic review of published literature. Best Practice & Research. Clinical Rheumatology , 26, 649– 658. doi: 10.1016/j.berh.2012.07.013 Google Scholar CrossRef Search ADS PubMed  Depalma G. Xu H. Covinsky K. E. Craig B. A. Stallard E. Thomas J.III, & Sands L. P. ( 2013). Hospital readmission among older adults who return home with unmet need for ADL disability. The Gerontologist , 53, 454– 461. doi: 10.1093/geront/gns103 Google Scholar CrossRef Search ADS PubMed  Edwards A. Elwyn G. Hood K. Atwell C. Robling M. Houston H., … Russell I.; Study Steering Group. ( 2004). Patient-based outcome results from a cluster randomized trial of shared decision making skill development and use of risk communication aids in general practice. Family Practice , 21, 347– 354. doi: 10.1093/fampra/cmh402 Google Scholar CrossRef Search ADS PubMed  Ethgen O. Bruyere O. Richy F. Dardennes C., & Reginster J. Y. ( 2004). Health-related quality of life in total hip and total knee arthroplasty: A qualitative and systematic review of the literature. The Journal of Bone & Joint Surgery , 86, 963– 974. Google Scholar CrossRef Search ADS   Figaro K. M. Williams-Russo P., & Allegrante J. P. ( 2004). Preferences for arthritis care among urban African Americans: “I don’t want to be cut.” Health Psychology , 23, 324– 329. doi: 10.1037/0278-6133.23.3.324 Google Scholar CrossRef Search ADS PubMed  Fingerman K. L., & Charles S. T. ( 2010). It takes two to tango why older people have the best relationships. Current Directions in Psychological Science , 19, 172– 176. doi: 10.1177/0963721410370297 Google Scholar CrossRef Search ADS   Fingerman K. L., & Pitzer L. M. ( 2007). Socialization in old age. In J. E. Grusec P. D. Hastings (Eds.), Handbook of socialization: Theory and research  (pp. 232– 255). New York, NY: Guilford Press. Fiori K. L. Antonucci T. C., & Akiyama H. ( 2008). Profiles of social relations among older adults: A cross-cultural approach. Ageing and Society , 28, 203– 231. doi:10.1017/S0144686X07006472 Google Scholar CrossRef Search ADS   Flynn, K. E., & Smith, M. A. ( 2007). Personality and health care decision-making style. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 62, 261– 267. doi:10.1093/geronb/62.5.P261 Google Scholar CrossRef Search ADS   Fortin P. R. Penrod J. R. Clarke A. E. St-Pierre Y. Joseph L. Bélisle P., … Katz J. N. ( 2002). Timing of total joint replacement affects clinical outcomes among patients with osteoarthritis of the hip or knee. Arthritis and Rheumatism , 46, 3327– 3330. doi: 10.1002/art.10631 Google Scholar CrossRef Search ADS PubMed  Gage E. A. ( 2013). Social networks of experientially similar others: Formation, activation, and consequences of network ties on the health care experience. Social Science & Medicine (1982) , 95, 43– 51. doi: 10.1016/j.socscimed.2012.09.001 Google Scholar CrossRef Search ADS PubMed  Gooberman-Hill, R., Sansom, A., Sanders, C. M., Dieppe, P. A., Horwood, J., Learmonth, I. D., ... Donovan, J. L. (2010). Unstated factors in orthopaedic decision-making: A qualitative study. BMC Musculoskeletal Disorders, 11, 213. doi:10.1186/1471-2474-11-213 Hawker G. A. Wright J. G. Badley E. M., & Coyte P. C. ( 2004). Perceptions of, and willingness to consider, total joint arthroplasty in a population‐based cohort of individuals with disabling hip and knee arthritis. Arthritis Care & Research , 51, 635– 641. doi: 10.1002/art.20524 Google Scholar CrossRef Search ADS   Hawker G. A. Wright J. G. Coyte P. C. Williams J. I. Harvey B. Glazier R., & Badley E. M. ( 2000). Differences between men and women in the rate of use of hip and knee arthroplasty. The New England Journal of Medicine , 342, 1016– 1022. doi: 10.1056/NEJM200004063421405 Google Scholar CrossRef Search ADS PubMed  House J. S. Umberson D., & Landis K. R. ( 1988). Structures and processes of social support. Annual Review of Sociology , 293– 318. doi: 10.1146/annurev.so.14.080188.001453 Lawson E. H. Gibbons M. M. Ingraham A. M. Shekelle P. G., & Ko C. Y. ( 2011). Appropriateness criteria to assess variations in surgical procedure use in the United States. Archives of Surgery (Chicago, Ill.: 1960) , 146, 1433– 1440. doi: 10.1001/archsurg.2011.581 Google Scholar CrossRef Search ADS PubMed  Linder, S. K., Swank, P. R., Vernon, S. W., Mullen, P. D., Morgan, R. O., & Volk, R. J. ( 2011). Validity of a low literacy version of the Decisional Conflict Scale. Patient education and counseling , 85, 521– 524. doi:10.1016/j.pec.2010.12.012 Google Scholar CrossRef Search ADS PubMed  Loewenstein G. ( 2000). Emotions in economic theory and economic behavior. American Economic Review , 426– 432. doi: 10.1257/aer.90.2.426 Loewenstein G. ( 2005). Projection bias in medical decision making. Medical Decision Making , 25, 96– 105. doi: 10.1177/0272989X04273799 Google Scholar CrossRef Search ADS PubMed  Martire L. M. Lustig A. P. Schulz R. Miller G. E., & Helgeson V. S. ( 2004). Is it beneficial to involve a family member? A meta-analysis of psychosocial interventions for chronic illness. Health Psychology , 23, 599– 611. doi: 10.1037/0278-6133.23.6.599 Google Scholar CrossRef Search ADS PubMed  Mata R., & Nunes L. ( 2010). When less is enough: Cognitive aging, information search, and decision quality in consumer choice. Psychology and Aging , 25, 289– 298. doi: 10.1037/a0017927 Google Scholar CrossRef Search ADS PubMed  McHugh G. A., & Luker K. A. ( 2009). Influences on individuals with osteoarthritis in deciding to undergo a hip or knee joint replacement: A qualitative study. Disability and Rehabilitation , 31, 1257– 1266. doi: 10.1080/09638280802535129 Google Scholar CrossRef Search ADS PubMed  Messeri P. Silverstein M., & Litwak E. ( 1993). Choosing optimal support groups: A review and reformulation. Journal of Health and Social Behavior , 34, 122– 137. doi:10.2307/2137239 Google Scholar CrossRef Search ADS PubMed  Noelker L. S., & Bass D. M. ( 1989). Home care for elderly persons: Linkages between formal and informal caregivers. Journal of Gerontology , 44, S63– S70. doi:10.1093/geronj/44.2.S63 Google Scholar CrossRef Search ADS PubMed  O’Connor A. M. ( 1995). Validation of a decisional conflict scale. Medical Decision Making , 15, 25– 30. doi: 10.1177/0272989X9501500105 Google Scholar CrossRef Search ADS PubMed  Parks, M. L., Hebert-Beirne, J., Rojas, M., Tuzzio, L., Nelson, C. L. & Boutin-Foster, C. ( 2014). A qualitative study of factors underlying decision making for joint replacement among African Americans and Latinos with osteoarthritis. Journal of long-term effects of medical implants , 24. doi:10.1615/JLongTermEffMedImplants.2014010428 Pillemer K., & Suitor J. J. ( 1996). “It takes one to help one”: Effects of similar others on the well-being of caregivers. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 51, S250– S257. doi: 10.1093/geronb/51B.5.S250 Google Scholar CrossRef Search ADS   Rammstedt B., & John O. P. ( 2007). Measuring personality in one minute or less: A 10-tem short version of the Big Five Inventory in English and German. Journal of Research in Personality , 41, 203– 212. doi: 10.1016/j.jrp.2006.02.001 Google Scholar CrossRef Search ADS   Riffin C. Pillemer K. Reid M. C., & Löckenhoff C. E. ( 2015). Decision support preferences among Hispanic and non-Hispanic White older adults with chronic pain. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences . Advance online publication. doi: 10.1093/geronb/gbv071. Romito F. Corvasce C. Montanaro R., & Mattioli V. ( 2011). Do elderly cancer patients have different care needs compared with younger ones? Tumori , 97, 374– 379. doi: 10.1700/912.10037 Google Scholar CrossRef Search ADS PubMed  Showalter A. Burger S., & Salyer J. ( 2000). Patients’ and their spouses’ needs after total joint arthroplasty: A pilot study. Orthopedic Nursing , 19, 49– 57. Google Scholar CrossRef Search ADS PubMed  Sjöling M. Ågren Y. Olofsson N. Hellzén O., & Asplund K. ( 2005). Waiting for surgery; living a life on hold—a continuous struggle against a faceless system. International Journal of Nursing Studies , 42, 539– 547. doi: 10.1016/j.ijnurstu.2004.09.009 Google Scholar CrossRef Search ADS PubMed  Soldz S., & Vaillant G. E. ( 1999). The Big Five personality traits and the life course: A 45-year longitudinal study. Journal of Research in Personality , 33, 208– 232. doi:10.1006/jrpe.1999.2243 Google Scholar CrossRef Search ADS   Thoits P. A. ( 2011). Mechanisms linking social ties and support to physical and mental health. Journal of Health and Social Behavior , 52, 145– 161. doi: 10.1177/0022146510395592 Google Scholar CrossRef Search ADS PubMed  Toye F. M. Barlow J. Wright C., & Lamb S. E. ( 2006). Personal meanings in the construction of need for total knee replacement surgery. Social Science & Medicine (1982) , 63, 43– 53. doi: 10.1016/j.socscimed.2005.11.054 Google Scholar CrossRef Search ADS PubMed  Vissers M. M. Bussmann J. B. de Groot I. B. Verhaar J. A., & Reijman M. ( 2013). Physical functioning four years after total hip and knee arthroplasty. Gait & Posture , 38, 310– 315. doi: 10.1016/j.gaitpost.2012.12.007 Google Scholar CrossRef Search ADS PubMed  Williams S. Weinman J. Dale J., & Newman S. ( 1995). Patient expectations: What do primary care patients want from the GP and how far does meeting expectations affect patient satisfaction? Family Practice , 12, 193– 201. doi: 10.1093/fampra/12.2.193 Google Scholar CrossRef Search ADS PubMed  Wolff J. L. Boyd C. M. Gitlin L. N. Bruce M. L., & Roter D. L. ( 2012). Going it together: Persistence of older adults’ accompaniment to physician visits by a family companion. Journal of the American Geriatrics Society , 60, 106– 112. doi: 10.1111/j.1532-5415.2011.03770.x Google Scholar CrossRef Search ADS PubMed  Zaidi R. Pfeil M. Macgregor A. J., & Goldberg A. ( 2013). How do patients with end-stage ankle arthritis decide between two surgical treatments? A qualitative study. BMJ Open , 3, e002782. doi: 10.1136/bmjopen-2013-002782 © The Author(s) 2016. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

The Journals of Gerontology Series B: Psychological Sciences and Social SciencesOxford University Press

Published: Mar 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