Associations Between Sedentary Behaviors and Cognitive Function: Cross-Sectional and Prospective Findings From the UK Biobank

Associations Between Sedentary Behaviors and Cognitive Function: Cross-Sectional and Prospective... Abstract We investigated the cross-sectional and prospective associations between different sedentary behaviors and cognitive function in a large sample of adults with data stored in the UK Biobank. Baseline data were available for 502,643 participants (2006–2010, United Kingdom). Cognitive tests included prospective memory (baseline only: n = 171,585), visual-spatial memory (round 1: n = 483,832; round 2: n = 482,762), fluid intelligence (n = 165,492), and short-term numeric memory (n = 50,370). After a mean period of 5.3 years, participants (numbering from 12,091 to 114,373, depending on the test) also provided follow-up cognitive data. Sedentary behaviors (television viewing, driving, and nonoccupational computer-use time) were measured at baseline. At baseline, both television viewing and driving time were inversely associated with cognitive function across all outcomes (e.g., for each additional hour spent watching television, the total number of correct answers in the fluid intelligence test was 0.15 (99% confidence interval: 0.14, 0.16) lower. Computer-use time was positively associated with cognitive function across all outcomes. Both television viewing and driving time at baseline were positively associated with the odds of having cognitive decline at follow-up across most outcomes. Conversely, computer-use time at baseline was inversely associated with the odds of having cognitive decline at follow-up across most outcomes. This study supports health policies designed to reduce television viewing and driving in adults. cognitive decline, cognitive function, computer use, driving, epidemiology, sedentary behavior, television viewing, UK Biobank Currently, there are no effective long-term pharmacological therapies for the treatment or prevention of dementia. Therefore, identifying potentially modifiable risk factors of cognitive decline, a major characteristic of dementia, is a key priority. Engaging in healthy lifestyle practices, including physical activity, has been associated with a reduced risk of dementia and its symptoms, such as cognitive impairment (1, 2), suggesting a potential role for lifestyle therapies. Indeed, physical-activity intervention studies have shown changes to the structure and function of the brain (3–7), supporting the observational associations. Along with physical activity, engaging in sedentary behavior, defined as sitting or reclining with low energy expenditure (8), could also be an important determinant of poor cognitive function. There is cumulative evidence indicating that sedentary time is associated with poor cardiometabolic health, chronic disease, and mortality (9–12). A recent systematic review also suggested that sedentary behavior is negatively associated with cognitive function, although the relationship between the two is complex, and recommended that future studies should focus on determining how different sedentary behaviors are associated with cognitive function (13). Limited observational research has indicated that television viewing is inversely associated with cognition (14–17). However, different sedentary behaviors may have different associations, with some evidence of computer/internet use being linked to cognitive improvement (15–18). Furthermore, most of the existing data have emerged from relatively limited cross-sectional findings (16–18). Therefore, this warrants investigation in large-scale studies with prospective data. The aim of this paper was to use the nationally representative UK Biobank cohort to examine the cross-sectional and prospective associations between domains of sedentary behavior (television viewing, driving, and computer use) and cognitive function (prospective memory, visual-spatial memory, fluid intelligence, and short-term numeric memory). METHODS Study design and population The UK Biobank is a large prospective study of the middle-aged population in the United Kingdom (19–21). Approximately 500,000 adults (aged 37–73 years) were recruited during 2006–2010 via mailed invitations to those registered with the National Health Service (NHS) and living within 25 miles of one of the 22 study assessment centers. Participants provided comprehensive baseline data on a broad range of biological, cognitive, demographic, health, lifestyle, mental, social, and well-being outcomes. Approximately 300,000 participants also provided an e-mail address to allow for remote follow-up for cognitive function measures in the future. During 2014–2015, approximately 125,000 participants provided some follow-up cognitive function data online. For the present study, baseline data were available on 502,643 individuals. Of these, depending on the cognitive test, between 50,370 and 483,832 participants provided baseline cognitive function data (see Web Figure 1, available at https://academic.oup.com/aje). Of these, after a mean period of 5.3 years, participants ranging in number from 12,091 to 114,373, depending on the test, also provided follow-up cognitive function data online (see Web Figure 2). All participants provided written informed consent, and the study was approved by the National Health Service’s National Research Ethics Service (Ref: 11/NW/0382). Further details are available elsewhere (19–21). Cognitive function tests Questionnaires administered through a computerized touchscreen interface assessed cognitive function at baseline. Using the same methodology without the touchscreen ability, follow-up measurements were obtained via online questionnaires that were completed remotely. To ensure effortless application on a large scale and wide response distributions, the cognitive function tests, which were refined in pilot study, were designed comprehensively and specifically for the UK Biobank. Prospective memory (available at baseline only), visual-spatial memory, fluid intelligence, and short-term numeric memory tests were included in this analysis. At baseline, there were variations between the numbers of individuals who completed each cognitive assessment due to tests being abandoned or skipped by participants, incorporated towards the end of recruitment (e.g., fluid intelligence), or phased out during the early stages of recruitment (e.g., short-term numeric memory). For more details on the cognitive function tests, see Web Appendix 1. Sedentary behaviors Data on sedentary behaviors were self-reported and collected at baseline using a computerized questionnaire. Domains of sedentary behavior included television viewing time (in hours/day: <1, 1, 2, 3, ≥4), driving time (in hours/day: <1, 1, 2, ≥3), and nonoccupational computer-use time (in hours/day: <1, 1, 2, ≥3). For more details, see Web Appendix 2. Covariate data Covariate data included anthropometric (body mass index), demographic (age, sex, ethnicity, social deprivation index, employment status, education level), health (number of cancers, number of noncancer illnesses, number of medications/treatments), and lifestyle (smoking status, alcohol drinking status, sleep duration, fruit and vegetable consumption, physical activity) variables. For more details, see Web Appendix 3. Statistical analysis Statistical analyses were performed using Stata/MP, version 14.0 (StataCorp LP, College Station, Texas). Data were analyzed in February 2017. With the intention of maximizing the use of the data, pairwise deletion was used to handle missing data (see Web Figures 1 and 2). Participant characteristics were tabulated. Categorical variables are presented as numbers and proportions, whereas continuous variables were summarized as mean values and standard deviations and are presented with their minimum and maximum values. Cross-sectional analysis Regression analysis was used to examine the cross-sectional associations between the domains of sedentary behavior and cognitive function at baseline. Multiple logistic regression models were fitted for each binary cognitive outcome variable (prospective memory, visual-spatial memory (round 1), and visual-spatial memory (round 2)). Multiple linear regression models were fitted for each continuous cognitive outcome variable (fluid intelligence and short-term numeric memory). For more details on the nature of the cognitive outcome variables used in the cross-sectional analysis, see Web Appendix 1. Model 1 mutually adjusted for the other sedentary behaviors and for age and sex. Model 2 further adjusted for body mass index, ethnicity, social deprivation index, employment status, educational level, smoking status, alcohol drinking status, fruit and vegetable consumption, sleep duration, physical activity (frequency of ≥10 minutes of walking (days/week), frequency of ≥10 minutes of moderate physical activity (days/week), frequency of ≥10 minutes of vigorous physical activity (days/week)), number of cancers, number of noncancer illnesses, and number of medications/treatments. For each sedentary behavior, the “<1 hour/day” category was selected as the reference group. Linear trends (linear terms) across the categories of each sedentary behavior were reported. Interaction terms were separately added to the full-adjustment model (model 2) to observe whether the associations between the sedentary behaviors and cognitive function were modified by age or sex. Significant results for age were stratified at 60 years. Prospective analysis Multiple logistic regression models investigated the prospective associations between the domains of sedentary behavior at baseline and cognitive function at follow-up. These models estimated the odds of having cognitive decline (i.e., a poor outcome) at follow-up. Cognitive outcomes included visual-spatial memory (round 1), visual-spatial memory (round 2), fluid intelligence, and short-term numeric memory. For full details on the definitions and nature of the cognitive outcome variables used in the prospective analysis, see Web Appendix 1. As well as controlling for the baseline result/score of the cognitive test under consideration, models adjusted for all the covariates mentioned previously. Linear trends across the categories of each sedentary behavior were reported. Interactions by age and sex were also investigated. Sensitivity analysis To assess the generalizability of our findings, the cross-sectional and prospective analyses investigating the associations between sedentary behaviors and cognitive function (model 1 and model 2) were repeated across the sample of participants without a medical history of cancer, cardiovascular disease, or cognitive/psychiatric illnesses as sensitivity analyses. For more details on the specific diseases/illnesses, see Web Appendix 4. Statistical reporting For each variable of interest (sedentary behaviors), the β coefficient (linear regression) or odds ratio (logistic regression) with 99% confidence intervals and P values are reported. All analyses employed robust standard errors, and all reported P values are 2-sided. To account for multiple comparisons, P < 0.01 was considered to be statistically significant for the main analyses. For the interaction analyses, P < 0.05 was considered to be statistically significant. RESULTS Cross-sectional findings Table 1 presents the characteristics of the 502,643 participants with baseline data. The mean age of these individuals was 56.5 (standard deviation (SD), 8.1) years, and 273,467 (54.4%) were female. Table 1. Baseline Characteristics of UK Biobank Participants (n = 502,643), United Kingdom, 2006–2010 Characteristic  No.  %  Mean (SD)  Range  Body mass indexa,b      27.4 (4.8)  12.1–74.7   Missing  3,105  0.6      Demographic characteristic           Age, yearsb      56.5 (8.1)  37.0–73.0    Missing  0  0.0       Ethnicityc            White British  442,699  88.1        Other  57,166  11.4        Missing  2,778  0.5       Sexc            Female  273,467  54.4        Male  229,176  45.6        Missing  0  0.0       Social deprivation indexb      −1.3 (3.1)  −6.3–11.0    Missing  627  0.1       Employment statusc            In paid employment or self-employed  287,234  57.1        Not in paid employment or self-employed  212,451  42.3        Missing  2,958  0.6       Educational levelc            No college or university degree  331,291  65.9        College or university degree  161,210  32.1        Missing  10,142  2.0      Lifestyle           Smoking statusc            Never  273,603  54.4        Previous  173,099  34.4        Current  52,989  10.6        Missing  2,952  0.6       Alcohol drinking statusc            Never  22,547  4.5        Previous  18,114  3.6        Current  460,479  91.6        Missing  1,503  0.3       Fruit and vegetable consumption, portions/dayc            <5  300,352  59.8        ≥5  189,979  37.8        Missing  12,132  2.4       Sleep duration, hours/dayb      7.2 (1.1)  1.0–23.0    Missing  4,218  0.8       Frequency of ≥10 minutes of walking, days/weekc            0  12,455  2.5        1  13,459  2.7        2  29,991  6.0        3  39,339  7.8        4  40,036  8.0        5  80,039  15.9        6  50,082  9.9        7  228,697  45.5        Missing  8,545  1.7       Frequency of ≥10 minutes of moderate physical activity, days/weekc            0  61,178  12.2        1  38,290  7.6        2  69,799  13.9        3  71,507  14.2        4  47,201  9.4        5  71,441  14.2        6  26,436  5.3        7  89,506  17.8        Missing  27,285  5.4       Frequency of ≥10 minutes of vigorous physical activity, days/weekc            0  178,275  35.5        1  66,853  13.3        2  75,055  14.9        3  65,276  13.0        4  30,705  6.1        5  32,452  6.5        6  9,430  1.9        7  17,005  3.4        Missing  27,592  5.5      Health           Number of cancersc            0  460,075  91.5        ≥1  41,706  8.3        Missing  862  0.2       Number of noncancer illnessesc            0  126,639  25.2        1  134,113  26.7        2  98,825  19.6        3  62,828  12.5        ≥4  79,376  15.8        Missing  862  0.2       Number of medications/treatmentsc            0  137,704  27.4        1  94,776  18.8        2  77,673  15.4        3  57,819  11.5        4  42,211  8.4        5  29,937  6.0        ≥6  61,661  12.3        Missing  862  0.2       Medical history of cancer, cardiovascular disease, or cognitive/  psychiatric illnessesc            No  402,897  80.2        Yes  99,746  19.8        Missing  0  0.0      Sedentary behaviors           Television viewing time, hours/dayc            <1  39,456  7.8        1  62,503  12.4        2  132,780  26.4        3  116,940  23.3        ≥4  145,546  29.0        Missing  5,418  1.1       Driving time, hours/dayc            <1  259,920  51.7        1  140,144  27.9        2  60,977  12.1        ≥3  31,663  6.3        Missing  9,939  2.0       Computer-use time, hours/dayc            <1  240,648  47.9        1  140,821  28.0        2  62,859  12.5        ≥3  48,939  9.7        Missing  9,376  1.9      Cognitive function at baseline           Prospective memory testc,d            Good result  130,910  26.0        Poor result  40,675  8.1        Missing  331,058  65.9       Visual-spatial memory test (round 1)c,e            Good result  345,685  68.8        Poor result  138,147  27.5        Missing  18,811  3.7       Visual-spatial memory test (round 2)c,f            Good result  82,130  16.3        Poor result  400,632  79.7        Missing  19,881  4.0       Fluid intelligence testb,g            Total number of correct answers      6.0 (2.2)  0.0–13.0    Missing  337,151  67.1       Short-term numeric memory testb,h            Maximum digits remembered correctly      6.7 (1.3)  2.0–12.0    Missing  452,273  90.0      Characteristic  No.  %  Mean (SD)  Range  Body mass indexa,b      27.4 (4.8)  12.1–74.7   Missing  3,105  0.6      Demographic characteristic           Age, yearsb      56.5 (8.1)  37.0–73.0    Missing  0  0.0       Ethnicityc            White British  442,699  88.1        Other  57,166  11.4        Missing  2,778  0.5       Sexc            Female  273,467  54.4        Male  229,176  45.6        Missing  0  0.0       Social deprivation indexb      −1.3 (3.1)  −6.3–11.0    Missing  627  0.1       Employment statusc            In paid employment or self-employed  287,234  57.1        Not in paid employment or self-employed  212,451  42.3        Missing  2,958  0.6       Educational levelc            No college or university degree  331,291  65.9        College or university degree  161,210  32.1        Missing  10,142  2.0      Lifestyle           Smoking statusc            Never  273,603  54.4        Previous  173,099  34.4        Current  52,989  10.6        Missing  2,952  0.6       Alcohol drinking statusc            Never  22,547  4.5        Previous  18,114  3.6        Current  460,479  91.6        Missing  1,503  0.3       Fruit and vegetable consumption, portions/dayc            <5  300,352  59.8        ≥5  189,979  37.8        Missing  12,132  2.4       Sleep duration, hours/dayb      7.2 (1.1)  1.0–23.0    Missing  4,218  0.8       Frequency of ≥10 minutes of walking, days/weekc            0  12,455  2.5        1  13,459  2.7        2  29,991  6.0        3  39,339  7.8        4  40,036  8.0        5  80,039  15.9        6  50,082  9.9        7  228,697  45.5        Missing  8,545  1.7       Frequency of ≥10 minutes of moderate physical activity, days/weekc            0  61,178  12.2        1  38,290  7.6        2  69,799  13.9        3  71,507  14.2        4  47,201  9.4        5  71,441  14.2        6  26,436  5.3        7  89,506  17.8        Missing  27,285  5.4       Frequency of ≥10 minutes of vigorous physical activity, days/weekc            0  178,275  35.5        1  66,853  13.3        2  75,055  14.9        3  65,276  13.0        4  30,705  6.1        5  32,452  6.5        6  9,430  1.9        7  17,005  3.4        Missing  27,592  5.5      Health           Number of cancersc            0  460,075  91.5        ≥1  41,706  8.3        Missing  862  0.2       Number of noncancer illnessesc            0  126,639  25.2        1  134,113  26.7        2  98,825  19.6        3  62,828  12.5        ≥4  79,376  15.8        Missing  862  0.2       Number of medications/treatmentsc            0  137,704  27.4        1  94,776  18.8        2  77,673  15.4        3  57,819  11.5        4  42,211  8.4        5  29,937  6.0        ≥6  61,661  12.3        Missing  862  0.2       Medical history of cancer, cardiovascular disease, or cognitive/  psychiatric illnessesc            No  402,897  80.2        Yes  99,746  19.8        Missing  0  0.0      Sedentary behaviors           Television viewing time, hours/dayc            <1  39,456  7.8        1  62,503  12.4        2  132,780  26.4        3  116,940  23.3        ≥4  145,546  29.0        Missing  5,418  1.1       Driving time, hours/dayc            <1  259,920  51.7        1  140,144  27.9        2  60,977  12.1        ≥3  31,663  6.3        Missing  9,939  2.0       Computer-use time, hours/dayc            <1  240,648  47.9        1  140,821  28.0        2  62,859  12.5        ≥3  48,939  9.7        Missing  9,376  1.9      Cognitive function at baseline           Prospective memory testc,d            Good result  130,910  26.0        Poor result  40,675  8.1        Missing  331,058  65.9       Visual-spatial memory test (round 1)c,e            Good result  345,685  68.8        Poor result  138,147  27.5        Missing  18,811  3.7       Visual-spatial memory test (round 2)c,f            Good result  82,130  16.3        Poor result  400,632  79.7        Missing  19,881  4.0       Fluid intelligence testb,g            Total number of correct answers      6.0 (2.2)  0.0–13.0    Missing  337,151  67.1       Short-term numeric memory testb,h            Maximum digits remembered correctly      6.7 (1.3)  2.0–12.0    Missing  452,273  90.0      Abbreviation: SD, standard deviation. a Weight (kg)/height (m)2. b Continuous variable. c Categorical variable. d Prospective memory test: good result (correct recall on first attempt) or poor result (incorrect recall on first attempt (i.e., correct recall on second attempt, instruction not recalled, skipped or incorrect)). e Pairs matching test (round 1): good result (0 incorrect matches) or poor result (≥1 incorrect matches). f Pairs matching test (round 2): good result (<2 incorrect matches) or poor result (≥2 incorrect matches). g Fluid intelligence score: total number of correct answers. h Numeric memory score: maximum digits remembered correctly. Table 2 presents the associations between the sedentary behaviors and cognitive function. In the full-adjustment models (model 2), the cross-sectional data showed that television viewing time was inversely associated with cognitive function across all outcomes apart from visual-spatial memory (round 2). For example, for each additional hour spent watching television up to ≥4 hours/day, the fluid intelligence and short-term numeric memory scores were 0.15 (99% confidence interval (CI): 0.14, 0.16) and 0.09 (99% CI: 0.07, 0.10) units lower, respectively. Correspondingly, the odds of a poor result in the prospective memory and visual-spatial memory (round 1) tests were 2% (99% CI: 0, 3) and 3% (99% CI: 2, 4) higher, respectively. Driving time was inversely associated with cognitive function across all outcomes. In contrast, computer-use time was positively associated with cognitive function across all outcomes. Table 2. Cross-Sectional Associations at Baseline Between Sedentary Behaviors and Cognitive Function Among UK Biobank Participants, United Kingdom, 2006–2010 Sedentary Behavior  Prospective Memory Testa  Visual-Spatial Memory Test  Fluid Intelligence Testd  Short-Term Numeric Memory Teste  Round 1b  Round 2c  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  β  99% CI  P Valuef  β  99% CI  P Valuef  Model 1g  Television viewing time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.99  0.93, 1.07  0.838  1.05  1.01, 1.09  0.002  1.01  0.96, 1.05  0.712  −0.13  −0.19, −0.07  <0.001  −0.15  −0.22, −0.08  <0.001   2  0.94  0.88, 1.00  0.012  1.07  1.03, 1.11  <0.001  1.02  0.98, 1.06  0.303  −0.30  −0.36, −0.24  <0.001  −0.25  −0.31, −0.19  <0.001   3  0.97  0.91, 1.04  0.303  1.14  1.10, 1.18  <0.001  1.03  0.99, 1.08  0.036  −0.54  −0.60, −0.49  <0.001  −0.35  −0.41, −0.29  <0.001   ≥4  1.23  1.15, 1.30  <0.001  1.26  1.22, 1.31  <0.001  1.08  1.03, 1.12  <0.001  −0.99  −1.04, −0.93  <0.001  −0.55  −0.61, −0.48  <0.001   Linear trend  1.07  1.05, 1.08  <0.001  1.06  1.06, 1.07  <0.001  1.02  1.01, 1.03  <0.001  −0.26  −0.27, −0.25  <0.001  −0.13  −0.15, −0.12  <0.001  Driving time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.05  1.02, 1.09  <0.001  0.99  0.97, 1.01  0.270  1.01  0.98, 1.03  0.569  −0.19  −0.22, −0.16  <0.001  −0.02  −0.06, 0.01  0.139   2  1.12  1.07, 1.18  <0.001  1.05  1.02, 1.08  <0.001  1.03  1.00, 1.06  0.024  −0.37  −0.41, −0.33  <0.001  −0.13  −0.18, −0.08  <0.001   ≥3  1.50  1.41, 1.60  <0.001  1.26  1.22, 1.31  <0.001  1.12  1.07, 1.17  <0.001  −0.88  −0.94, −0.82  <0.001  −0.28  −0.34, −0.21  <0.001   Linear trend  1.11  1.09, 1.12  <0.001  1.05  1.04, 1.06  <0.001  1.03  1.01, 1.04  <0.001  −0.24  −0.26, −0.23  <0.001  −0.08  −0.09, −0.06  <0.001  Computer-use time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.68  0.66, 0.71  <0.001  0.79  0.77, 0.80  <0.001  0.85  0.83, 0.87  <0.001  0.52  0.49, 0.55  <0.001  0.21  0.17, 0.25  <0.001   2  0.69  0.66, 0.72  <0.001  0.77  0.75, 0.79  <0.001  0.80  0.78, 0.83  <0.001  0.58  0.53, 0.62  <0.001  0.21  0.16, 0.26  <0.001   ≥3  0.86  0.82, 0.90  <0.001  0.81  0.79, 0.84  <0.001  0.82  0.79, 0.85  <0.001  0.40  0.35, 0.44  <0.001  0.16  0.11, 0.22  <0.001   Linear trend  0.91  0.89, 0.92  <0.001  0.91  0.90, 0.92  <0.001  0.92  0.91, 0.93  <0.001  0.18  0.17, 0.20  <0.001  0.07  0.06, 0.09  <0.001  Model 2h  Television viewing time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.04  0.96, 1.12  0.232  1.07  1.03, 1.12  <0.001  1.01  0.97, 1.06  0.403  −0.12  −0.18, −0.06  <0.001  −0.13  −0.20, −0.06  <0.001   2  0.96  0.90, 1.03  0.142  1.07  1.03, 1.11  <0.001  1.02  0.98, 1.07  0.128  −0.21  −0.27, −0.16  <0.001  −0.20  −0.26, −0.13  <0.001   3  0.96  0.89, 1.03  0.159  1.09  1.05, 1.14  <0.001  1.03  0.98, 1.07  0.136  −0.33  −0.39, −0.28  <0.001  −0.25  −0.32, −0.19  <0.001   ≥4  1.09  1.01, 1.17  0.003  1.14  1.10, 1.19  <0.001  1.03  0.99, 1.08  0.053  −0.58  −0.64, −0.53  <0.001  −0.38  −0.44, −0.31  <0.001   Linear trend  1.02  1.00, 1.03  0.001  1.03  1.02, 1.04  <0.001  1.01  1.00, 1.02  0.057  −0.15  −0.16, −0.14  <0.001  −0.09  −0.10, −0.07  <0.001  Driving time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.21  1.17, 1.27  <0.001  1.05  1.03, 1.07  <0.001  1.04  1.02, 1.07  <0.001  −0.28  −0.31, −0.25  <0.001  −0.06  −0.10, −0.03  <0.001   2  1.27  1.20, 1.34  <0.001  1.10  1.06, 1.13  <0.001  1.07  1.03, 1.10  <0.001  −0.43  −0.48, −0.39  <0.001  −0.18  −0.23, −0.13  <0.001   ≥3  1.54  1.43, 1.66  <0.001  1.23  1.19, 1.28  <0.001  1.11  1.06, 1.16  <0.001  −0.73  −0.79, −0.68  <0.001  −0.27  −0.34, −0.19  <0.001   Linear trend  1.15  1.13, 1.17  <0.001  1.06  1.05, 1.07  <0.001  1.04  1.02, 1.05  <0.001  −0.24  −0.25, −0.22  <0.001  −0.09  −0.11, −0.07  <0.001  Computer-use time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.77  0.74, 0.81  <0.001  0.85  0.83, 0.87  <0.001  0.88  0.86, 0.90  <0.001  0.32  0.29, 0.35  <0.001  0.14  0.10, 0.17  <0.001   2  0.74  0.70, 0.78  <0.001  0.81  0.79, 0.83  <0.001  0.83  0.80, 0.86  <0.001  0.40  0.36, 0.44  <0.001  0.15  0.10, 0.20  <0.001   ≥3  0.86  0.81, 0.91  <0.001  0.84  0.81, 0.86  <0.001  0.84  0.81, 0.88  <0.001  0.26  0.22, 0.31  <0.001  0.13  0.07, 0.18  <0.001   Linear trend  0.92  0.90, 0.94  <0.001  0.92  0.91, 0.93  <0.001  0.93  0.92, 0.94  <0.001  0.12  0.11, 0.14  <0.001  0.06  0.04, 0.07  <0.001  Sedentary Behavior  Prospective Memory Testa  Visual-Spatial Memory Test  Fluid Intelligence Testd  Short-Term Numeric Memory Teste  Round 1b  Round 2c  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  β  99% CI  P Valuef  β  99% CI  P Valuef  Model 1g  Television viewing time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.99  0.93, 1.07  0.838  1.05  1.01, 1.09  0.002  1.01  0.96, 1.05  0.712  −0.13  −0.19, −0.07  <0.001  −0.15  −0.22, −0.08  <0.001   2  0.94  0.88, 1.00  0.012  1.07  1.03, 1.11  <0.001  1.02  0.98, 1.06  0.303  −0.30  −0.36, −0.24  <0.001  −0.25  −0.31, −0.19  <0.001   3  0.97  0.91, 1.04  0.303  1.14  1.10, 1.18  <0.001  1.03  0.99, 1.08  0.036  −0.54  −0.60, −0.49  <0.001  −0.35  −0.41, −0.29  <0.001   ≥4  1.23  1.15, 1.30  <0.001  1.26  1.22, 1.31  <0.001  1.08  1.03, 1.12  <0.001  −0.99  −1.04, −0.93  <0.001  −0.55  −0.61, −0.48  <0.001   Linear trend  1.07  1.05, 1.08  <0.001  1.06  1.06, 1.07  <0.001  1.02  1.01, 1.03  <0.001  −0.26  −0.27, −0.25  <0.001  −0.13  −0.15, −0.12  <0.001  Driving time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.05  1.02, 1.09  <0.001  0.99  0.97, 1.01  0.270  1.01  0.98, 1.03  0.569  −0.19  −0.22, −0.16  <0.001  −0.02  −0.06, 0.01  0.139   2  1.12  1.07, 1.18  <0.001  1.05  1.02, 1.08  <0.001  1.03  1.00, 1.06  0.024  −0.37  −0.41, −0.33  <0.001  −0.13  −0.18, −0.08  <0.001   ≥3  1.50  1.41, 1.60  <0.001  1.26  1.22, 1.31  <0.001  1.12  1.07, 1.17  <0.001  −0.88  −0.94, −0.82  <0.001  −0.28  −0.34, −0.21  <0.001   Linear trend  1.11  1.09, 1.12  <0.001  1.05  1.04, 1.06  <0.001  1.03  1.01, 1.04  <0.001  −0.24  −0.26, −0.23  <0.001  −0.08  −0.09, −0.06  <0.001  Computer-use time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.68  0.66, 0.71  <0.001  0.79  0.77, 0.80  <0.001  0.85  0.83, 0.87  <0.001  0.52  0.49, 0.55  <0.001  0.21  0.17, 0.25  <0.001   2  0.69  0.66, 0.72  <0.001  0.77  0.75, 0.79  <0.001  0.80  0.78, 0.83  <0.001  0.58  0.53, 0.62  <0.001  0.21  0.16, 0.26  <0.001   ≥3  0.86  0.82, 0.90  <0.001  0.81  0.79, 0.84  <0.001  0.82  0.79, 0.85  <0.001  0.40  0.35, 0.44  <0.001  0.16  0.11, 0.22  <0.001   Linear trend  0.91  0.89, 0.92  <0.001  0.91  0.90, 0.92  <0.001  0.92  0.91, 0.93  <0.001  0.18  0.17, 0.20  <0.001  0.07  0.06, 0.09  <0.001  Model 2h  Television viewing time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.04  0.96, 1.12  0.232  1.07  1.03, 1.12  <0.001  1.01  0.97, 1.06  0.403  −0.12  −0.18, −0.06  <0.001  −0.13  −0.20, −0.06  <0.001   2  0.96  0.90, 1.03  0.142  1.07  1.03, 1.11  <0.001  1.02  0.98, 1.07  0.128  −0.21  −0.27, −0.16  <0.001  −0.20  −0.26, −0.13  <0.001   3  0.96  0.89, 1.03  0.159  1.09  1.05, 1.14  <0.001  1.03  0.98, 1.07  0.136  −0.33  −0.39, −0.28  <0.001  −0.25  −0.32, −0.19  <0.001   ≥4  1.09  1.01, 1.17  0.003  1.14  1.10, 1.19  <0.001  1.03  0.99, 1.08  0.053  −0.58  −0.64, −0.53  <0.001  −0.38  −0.44, −0.31  <0.001   Linear trend  1.02  1.00, 1.03  0.001  1.03  1.02, 1.04  <0.001  1.01  1.00, 1.02  0.057  −0.15  −0.16, −0.14  <0.001  −0.09  −0.10, −0.07  <0.001  Driving time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.21  1.17, 1.27  <0.001  1.05  1.03, 1.07  <0.001  1.04  1.02, 1.07  <0.001  −0.28  −0.31, −0.25  <0.001  −0.06  −0.10, −0.03  <0.001   2  1.27  1.20, 1.34  <0.001  1.10  1.06, 1.13  <0.001  1.07  1.03, 1.10  <0.001  −0.43  −0.48, −0.39  <0.001  −0.18  −0.23, −0.13  <0.001   ≥3  1.54  1.43, 1.66  <0.001  1.23  1.19, 1.28  <0.001  1.11  1.06, 1.16  <0.001  −0.73  −0.79, −0.68  <0.001  −0.27  −0.34, −0.19  <0.001   Linear trend  1.15  1.13, 1.17  <0.001  1.06  1.05, 1.07  <0.001  1.04  1.02, 1.05  <0.001  −0.24  −0.25, −0.22  <0.001  −0.09  −0.11, −0.07  <0.001  Computer-use time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.77  0.74, 0.81  <0.001  0.85  0.83, 0.87  <0.001  0.88  0.86, 0.90  <0.001  0.32  0.29, 0.35  <0.001  0.14  0.10, 0.17  <0.001   2  0.74  0.70, 0.78  <0.001  0.81  0.79, 0.83  <0.001  0.83  0.80, 0.86  <0.001  0.40  0.36, 0.44  <0.001  0.15  0.10, 0.20  <0.001   ≥3  0.86  0.81, 0.91  <0.001  0.84  0.81, 0.86  <0.001  0.84  0.81, 0.88  <0.001  0.26  0.22, 0.31  <0.001  0.13  0.07, 0.18  <0.001   Linear trend  0.92  0.90, 0.94  <0.001  0.92  0.91, 0.93  <0.001  0.93  0.92, 0.94  <0.001  0.12  0.11, 0.14  <0.001  0.06  0.04, 0.07  <0.001  Abbreviations: CI, confidence interval; OR, odds ratio. a Prospective memory result, categorical: good result (referent: correct recall on first attempt) or poor result (incorrect recall on first attempt (i.e., correct recall on second attempt, instruction not recalled, skipped or incorrect)). An odds ratio of less than 1 indicates lower odds of a poor result; an odds ratio of greater than 1 indicates higher odds of a poor result (model 1: n = 166,401; model 2: n = 148,327). b Pairs matching result (round 1), categorical: good result (referent: 0 incorrect matches) or poor result (≥1 incorrect matches). An odds ratio of less than 1 indicates lower odds of a poor result; an odds ratio of greater than 1 indicates higher odds of a poor result (model 1: n = 471,474; model 2: n = 422,731). c Pairs matching result (round 2), categorical: good result (referent: <2 incorrect matches) or poor result (≥2 incorrect matches). An odds ratio of less than 1 indicates lower odds of a poor result; an odds ratio of greater than 1 indicates higher odds of a poor result (model 1: n = 470,433; model 2: n = 421,851). d Fluid intelligence score, continuous: total number of correct answers. A β coefficient of greater than 0 indicates a higher score; a β coefficient of less than 0 indicates a lower score (model 1: n = 161,348; model 2: n = 145,124). e Numeric memory score, continuous: maximum digits remembered correctly. A β coefficient of greater than 0 indicates a higher score; a β coefficient of less than 0 indicates a lower score (model 1: n = 49,035; model 2: n = 44,097). fP < 0.01 indicates statistical significance. g Model 1 mutually adjusted for the other sedentary behaviors and for age and sex. h Model 2 further adjusted for body mass index, ethnicity, social deprivation index, employment status, educational level, smoking status, alcohol drinking status, fruit and vegetable consumption, sleep duration, frequency of ≥10 minutes of walking, frequency of ≥10 minutes of moderate physical activity, frequency of ≥10 minutes of vigorous physical activity, number of cancers, number of noncancer illnesses, and number of medications/treatments. Interaction analyses showed that most findings were modified by age and sex (P < 0.05). Stratification indicated that the associations were generally stronger in older adults (≥60 years of age) and in men (see Web Figure 3 (age) and Web Figure 4 (sex)). Prospective findings Table 3 presents the cognitive function data of the participants with cognitive data at both baseline and follow-up. Cognitive decline over time was apparent—participants performed better in each cognitive test at baseline than at follow-up. For example, the mean fluid intelligence scores (n = 46,704) at baseline and follow-up were 6.7 (SD, 2.1) and 5.5 (SD, 2.0), respectively, with 15,384 (32.9%) individuals reporting a good outcome at follow-up (baseline fluid intelligence score ≤ follow-up fluid intelligence score) and 31,320 (67.1%) individuals reporting a poor outcome at follow-up (baseline fluid intelligence score > follow-up fluid intelligence score). The other tests followed a similar pattern. Table 3. Cognitive Function Data of Participants With Cognitive Data at Both Baseline and Follow-up, UK Biobank, United Kingdom, 2006–2010 Cognitive Functiona  Total No. of Participants  Baseline  Follow-up  No.  %  Mean (SD)  Range  No.  %  Mean (SD)  Range  Visual-spatial memory test (round 1)b,c  114,373                   Good result    89,137  77.9      70,761  61.9       Poor result    25,236  22.1      43,612  38.1       Good outcome at follow-up            70,761  61.9       Poor outcome at follow-up            43,612  38.1      Visual-spatial memory test (round 2)b,d  113,479                   Good result    23,262  20.5      14,886  13.1       Poor result    90,217  79.5      98,593  86.9       Good outcome at follow-up            14,886  13.1       Poor outcome at follow-up            98,593  86.9      Fluid intelligence teste,f  46,704                   Total number of correct answers        6.7 (2.1)  0.0–13.0      5.5 (2.0)  0.0–13.0   Good outcome at follow-up            15,384  32.9       Poor outcome at follow-up            31,320  67.1      Short-term numeric memory teste,g  12,091                   Maximum digits remembered correctly        7.0 (1.2)  2.0–12.0      6.9 (1.5)  2.0–11.0   Good outcome at follow-up            7,791  64.4       Poor outcome at follow-up            4,300  35.6      Cognitive Functiona  Total No. of Participants  Baseline  Follow-up  No.  %  Mean (SD)  Range  No.  %  Mean (SD)  Range  Visual-spatial memory test (round 1)b,c  114,373                   Good result    89,137  77.9      70,761  61.9       Poor result    25,236  22.1      43,612  38.1       Good outcome at follow-up            70,761  61.9       Poor outcome at follow-up            43,612  38.1      Visual-spatial memory test (round 2)b,d  113,479                   Good result    23,262  20.5      14,886  13.1       Poor result    90,217  79.5      98,593  86.9       Good outcome at follow-up            14,886  13.1       Poor outcome at follow-up            98,593  86.9      Fluid intelligence teste,f  46,704                   Total number of correct answers        6.7 (2.1)  0.0–13.0      5.5 (2.0)  0.0–13.0   Good outcome at follow-up            15,384  32.9       Poor outcome at follow-up            31,320  67.1      Short-term numeric memory teste,g  12,091                   Maximum digits remembered correctly        7.0 (1.2)  2.0–12.0      6.9 (1.5)  2.0–11.0   Good outcome at follow-up            7,791  64.4       Poor outcome at follow-up            4,300  35.6      Abbreviation: SD, standard deviation. a Samples sizes were different for different tests, ranging from 12,091 to 114,373. The mean follow-up period was 5.3 years. b Categorical variable. c Pairs matching result (round 1): good result (0 incorrect matches) or poor result (≥1 incorrect matches). Good outcome at follow-up (0 incorrect matches at follow-up) or poor outcome at follow-up (≥1 incorrect matches at follow-up). d Pairs matching result (round 2): good result (<2 incorrect matches) or poor result (≥2 incorrect matches). Good outcome at follow-up (<2 incorrect matches at follow-up) or poor outcome at follow-up (≥2 incorrect matches at follow-up). e Continuous variable. f Fluid intelligence score: total number of correct answers. Good outcome at follow-up (baseline fluid intelligence score ≤ follow-up fluid intelligence score) or poor outcome at follow-up (baseline fluid intelligence score > follow-up fluid intelligence score). g Numeric memory score: Maximum digits remembered correctly. Good outcome at follow-up (baseline numeric memory score ≤ follow-up numeric memory score) or poor outcome at follow-up (baseline numeric memory score > follow-up numeric memory score). Those with follow-up data had similar characteristics to those of the full UK Biobank cohort, although they were better educated and more likely to be employed (see Web Table 1). Table 4 presents the associations between the sedentary behaviors at baseline and cognitive function at follow-up. In the full-adjustment models (model 2), both television viewing and driving time at baseline were positively associated with the odds of having cognitive decline at follow-up across most outcomes. For example, for each additional hour spent watching television up to ≥4 hours/day at baseline, the odds of a lower fluid intelligence score at follow-up were 9% (99% CI: 6, 11) higher. Similarly, for each additional hour spent driving up to ≥3 hours/day at baseline, the odds of a lower fluid intelligence score at follow-up were 11% (99% CI: 7, 15) higher. In contrast, computer-use time at baseline was inversely associated with the odds of having cognitive decline at follow-up across most outcomes. Interaction analyses showed that only the associations between television viewing time and visual-spatial memory (round 2) were modified by age (P < 0.05) (see Web Figure 5). Findings were not modified by sex. Table 4. Prospective Associations Between Sedentary Behaviors at Baseline and Cognitive Function at Follow-upa Among UK Biobank Participants, United Kingdom, 2006–2010 Sedentary Behavior  Visual-Spatial Memory Test  Fluid Intelligence Testd  Short-Term Numeric Memory Teste  Round 1b  Round 2c  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  Model 1g  Television viewing time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.04  0.97, 1.11  0.154  1.02  0.93, 1.11  0.623  1.15  1.02, 1.28  0.002  1.05  0.85, 1.31  0.557   2  1.09  1.03, 1.15  <0.001  1.00  0.92, 1.08  0.961  1.24  1.12, 1.37  <0.001  1.13  0.93, 1.37  0.112   3  1.13  1.07, 1.20  <0.001  1.03  0.94, 1.12  0.439  1.37  1.24, 1.52  <0.001  1.26  1.03, 1.55  0.003   ≥4  1.17  1.10, 1.25  <0.001  1.01  0.93, 1.10  0.672  1.66  1.50, 1.84  <0.001  1.43  1.17, 1.76  <0.001   Linear trend  1.04  1.03, 1.06  <0.001  1.00  0.99, 1.02  0.612  1.13  1.10, 1.15  <0.001  1.10  1.05, 1.15  <0.001  Driving time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.06  1.02, 1.10  <0.001  1.01  0.96, 1.07  0.480  1.15  1.08, 1.22  <0.001  1.05  0.93, 1.18  0.319   2  1.07  1.01, 1.12  0.002  1.00  0.93, 1.08  0.903  1.10  1.00, 1.21  0.008  1.09  0.92, 1.30  0.193   ≥3  1.18  1.09, 1.28  <0.001  1.01  0.90, 1.13  0.831  1.44  1.25, 1.66  <0.001  1.11  0.85, 1.44  0.318   Linear trend  1.05  1.03, 1.07  <0.001  1.00  0.98, 1.03  0.709  1.10  1.06, 1.14  <0.001  1.04  0.98, 1.11  0.108  Computer-use time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  0.96  0.93, 1.00  0.013  0.96  0.91, 1.02  0.068  0.93  0.87, 1.00  0.006  0.90  0.79, 1.02  0.034   2  0.90  0.86, 0.94  <0.001  0.87  0.81, 0.93  <0.001  0.94  0.86, 1.02  0.041  0.77  0.65, 0.90  <0.001   ≥3  0.91  0.86, 0.96  <0.001  0.89  0.83, 0.96  <0.001  0.96  0.88, 1.05  0.293  0.86  0.72, 1.03  0.035   Linear trend  0.96  0.95, 0.98  <0.001  0.95  0.93, 0.97  <0.001  0.98  0.96, 1.01  0.150  0.93  0.88, 0.98  0.001  Model 2h  Television viewing time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.02  0.96, 1.09  0.348  1.03  0.94, 1.12  0.470  1.16  1.03, 1.30  0.001  1.02  0.82, 1.28  0.817   2  1.07  1.00, 1.13  0.006  1.01  0.93, 1.09  0.815  1.21  1.09, 1.35  <0.001  1.08  0.88, 1.33  0.310   3  1.08  1.02, 1.15  0.001  1.03  0.94, 1.12  0.416  1.29  1.15, 1.44  <0.001  1.16  0.94, 1.44  0.066   ≥4  1.09  1.02, 1.17  0.001  1.00  0.91, 1.10  0.993  1.45  1.29, 1.62  <0.001  1.29  1.04, 1.61  0.003   Linear trend  1.02  1.01, 1.04  <0.001  1.00  0.98, 1.02  0.955  1.09  1.06, 1.11  <0.001  1.07  1.02, 1.12  <0.001  Driving time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.07  1.03, 1.12  <0.001  1.02  0.97, 1.08  0.294  1.19  1.11, 1.27  <0.001  1.05  0.92, 1.19  0.363   2  1.08  1.02, 1.14  0.001  1.01  0.94, 1.10  0.624  1.15  1.04, 1.27  <0.001  1.05  0.88, 1.27  0.466   ≥3  1.16  1.06, 1.26  <0.001  1.01  0.90, 1.13  0.895  1.43  1.24, 1.66  <0.001  1.05  0.80, 1.39  0.650   Linear trend  1.05  1.03, 1.07  <0.001  1.01  0.98, 1.04  0.552  1.11  1.07, 1.15  <0.001  1.02  0.96, 1.10  0.363  Computer-use time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  0.97  0.93, 1.01  0.053  0.97  0.92, 1.03  0.250  0.94  0.87, 1.01  0.020  0.92  0.80, 1.05  0.090   2  0.91  0.86, 0.95  <0.001  0.88  0.82, 0.94  <0.001  0.94  0.86, 1.03  0.073  0.76  0.64, 0.90  <0.001   ≥3  0.90  0.85, 0.96  <0.001  0.90  0.83, 0.98  0.001  0.97  0.88, 1.06  0.359  0.84  0.69, 1.01  0.016   Linear trend  0.96  0.95, 0.98  <0.001  0.96  0.93, 0.98  <0.001  0.99  0.96, 1.02  0.207  0.92  0.87, 0.98  <0.001  Sedentary Behavior  Visual-Spatial Memory Test  Fluid Intelligence Testd  Short-Term Numeric Memory Teste  Round 1b  Round 2c  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  Model 1g  Television viewing time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.04  0.97, 1.11  0.154  1.02  0.93, 1.11  0.623  1.15  1.02, 1.28  0.002  1.05  0.85, 1.31  0.557   2  1.09  1.03, 1.15  <0.001  1.00  0.92, 1.08  0.961  1.24  1.12, 1.37  <0.001  1.13  0.93, 1.37  0.112   3  1.13  1.07, 1.20  <0.001  1.03  0.94, 1.12  0.439  1.37  1.24, 1.52  <0.001  1.26  1.03, 1.55  0.003   ≥4  1.17  1.10, 1.25  <0.001  1.01  0.93, 1.10  0.672  1.66  1.50, 1.84  <0.001  1.43  1.17, 1.76  <0.001   Linear trend  1.04  1.03, 1.06  <0.001  1.00  0.99, 1.02  0.612  1.13  1.10, 1.15  <0.001  1.10  1.05, 1.15  <0.001  Driving time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.06  1.02, 1.10  <0.001  1.01  0.96, 1.07  0.480  1.15  1.08, 1.22  <0.001  1.05  0.93, 1.18  0.319   2  1.07  1.01, 1.12  0.002  1.00  0.93, 1.08  0.903  1.10  1.00, 1.21  0.008  1.09  0.92, 1.30  0.193   ≥3  1.18  1.09, 1.28  <0.001  1.01  0.90, 1.13  0.831  1.44  1.25, 1.66  <0.001  1.11  0.85, 1.44  0.318   Linear trend  1.05  1.03, 1.07  <0.001  1.00  0.98, 1.03  0.709  1.10  1.06, 1.14  <0.001  1.04  0.98, 1.11  0.108  Computer-use time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  0.96  0.93, 1.00  0.013  0.96  0.91, 1.02  0.068  0.93  0.87, 1.00  0.006  0.90  0.79, 1.02  0.034   2  0.90  0.86, 0.94  <0.001  0.87  0.81, 0.93  <0.001  0.94  0.86, 1.02  0.041  0.77  0.65, 0.90  <0.001   ≥3  0.91  0.86, 0.96  <0.001  0.89  0.83, 0.96  <0.001  0.96  0.88, 1.05  0.293  0.86  0.72, 1.03  0.035   Linear trend  0.96  0.95, 0.98  <0.001  0.95  0.93, 0.97  <0.001  0.98  0.96, 1.01  0.150  0.93  0.88, 0.98  0.001  Model 2h  Television viewing time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.02  0.96, 1.09  0.348  1.03  0.94, 1.12  0.470  1.16  1.03, 1.30  0.001  1.02  0.82, 1.28  0.817   2  1.07  1.00, 1.13  0.006  1.01  0.93, 1.09  0.815  1.21  1.09, 1.35  <0.001  1.08  0.88, 1.33  0.310   3  1.08  1.02, 1.15  0.001  1.03  0.94, 1.12  0.416  1.29  1.15, 1.44  <0.001  1.16  0.94, 1.44  0.066   ≥4  1.09  1.02, 1.17  0.001  1.00  0.91, 1.10  0.993  1.45  1.29, 1.62  <0.001  1.29  1.04, 1.61  0.003   Linear trend  1.02  1.01, 1.04  <0.001  1.00  0.98, 1.02  0.955  1.09  1.06, 1.11  <0.001  1.07  1.02, 1.12  <0.001  Driving time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.07  1.03, 1.12  <0.001  1.02  0.97, 1.08  0.294  1.19  1.11, 1.27  <0.001  1.05  0.92, 1.19  0.363   2  1.08  1.02, 1.14  0.001  1.01  0.94, 1.10  0.624  1.15  1.04, 1.27  <0.001  1.05  0.88, 1.27  0.466   ≥3  1.16  1.06, 1.26  <0.001  1.01  0.90, 1.13  0.895  1.43  1.24, 1.66  <0.001  1.05  0.80, 1.39  0.650   Linear trend  1.05  1.03, 1.07  <0.001  1.01  0.98, 1.04  0.552  1.11  1.07, 1.15  <0.001  1.02  0.96, 1.10  0.363  Computer-use time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  0.97  0.93, 1.01  0.053  0.97  0.92, 1.03  0.250  0.94  0.87, 1.01  0.020  0.92  0.80, 1.05  0.090   2  0.91  0.86, 0.95  <0.001  0.88  0.82, 0.94  <0.001  0.94  0.86, 1.03  0.073  0.76  0.64, 0.90  <0.001   ≥3  0.90  0.85, 0.96  <0.001  0.90  0.83, 0.98  0.001  0.97  0.88, 1.06  0.359  0.84  0.69, 1.01  0.016   Linear trend  0.96  0.95, 0.98  <0.001  0.96  0.93, 0.98  <0.001  0.99  0.96, 1.02  0.207  0.92  0.87, 0.98  <0.001  Abbreviations: CI, confidence interval; OR, odds ratio. a The mean follow-up period was 5.3 years. b Pairs matching result (round 1), categorical: good outcome at follow-up (0 incorrect matches at follow-up) or poor outcome at follow-up (≥1 incorrect matches at follow-up). An odds ratio of less than 1 indicates lower odds of having cognitive decline at follow-up (i.e., a good outcome at follow-up); an odds ratio of greater than 1 indicates higher odds of having cognitive decline at follow-up (i.e., a poor outcome at follow-up) (model 1: n = 113,129; model 2: n = 106,665). c Pairs matching result (round 2), categorical: good outcome at follow-up (<2 incorrect matches at follow-up) or poor outcome at follow-up (≥2 incorrect matches at follow-up). An odds ratio of less than 1 indicates lower odds of having cognitive decline at follow-up (i.e., a good outcome at follow-up); an odds ratio of greater than 1 indicates higher odds of having cognitive decline at follow-up (i.e., a poor outcome at follow-up) (model 1: n = 112,252; model 2: n = 105,861). d Fluid intelligence score, categorical: good outcome at follow-up (baseline fluid intelligence score ≤ follow-up fluid intelligence score) or poor outcome at follow-up (baseline fluid intelligence score > follow-up fluid intelligence score). An odds ratio of less than 1 indicates lower odds of having cognitive decline at follow-up (i.e., a good outcome at follow-up); an odds ratio of greater than 1 indicates higher odds of having cognitive decline at follow-up (i.e., a poor outcome at follow-up) (model 1: n = 46,158; model 2: n = 43,350). e Numeric memory score, categorical: good outcome at follow-up (baseline numeric memory score ≤ follow-up numeric memory score); or poor outcome at follow-up (baseline numeric memory score > follow-up numeric memory score). An odds ratio of less than 1 indicates lower odds of having cognitive decline at follow-up (i.e., a good outcome at follow-up); an odds ratio of greater than 1 indicates higher odds of having cognitive decline at follow-up (i.e., a poor outcome at follow-up) (model 1: n = 11,957; model 2: n = 11,299). fP < 0.01 indicates statistical significance. g Model 1 mutually adjusted for the other sedentary behaviors and for age, sex, and the baseline result/score of the cognitive test under consideration. h Model 2 further adjusted for body mass index, ethnicity, social deprivation index, employment status, education level, smoking status, alcohol drinking status, fruit and vegetable consumption, sleep duration, frequency of ≥10 minutes of walking, frequency of ≥10 minutes of moderate physical activity, frequency of ≥10 minutes of vigorous physical activity, number of cancers, number of noncancer illnesses, and number of medications/treatments. Sensitivity analyses The cross-sectional and prospective findings were generalizable across the sample of participants without cancer, cardiovascular disease, or cognitive/psychiatric illnesses (see Web Figure 6 (cross-sectional associations) and Web Figure 7 (prospective associations)). DISCUSSION Key findings To our knowledge, this is the first study to quantify the cross-sectional and prospective associations between domains of sedentary behavior and cognitive function in a large cohort of adults in the United Kingdom. At baseline, both television viewing and driving time were inversely associated with cognitive function. In contrast, computer-use time was positively associated with cognitive function. Most findings were modified by age and sex, with stronger relationships generally observed in older adults and in men. These novel results suggest that the influence of sedentary behavior on cognition is enhanced in older age and in men. Both television viewing and driving time at baseline were positively associated with the odds of having cognitive decline at follow-up across most outcomes. In contrast, computer-use time at baseline was inversely associated with the odds of having cognitive decline at follow-up across most outcomes. The cross-sectional and prospective findings were robust and generalizable across the sample of participants without cancer, cardiovascular disease, or cognitive/psychiatric illnesses. Interpretations To our knowledge, a modest number of studies have attempted to examine the prospective associations between the different types of sedentary behaviors and cognitive function (14–17, 22–26). However, these studies have been limited by a small sample size (n values ranging between 469 and 8,462), populations that involved only children or older adults, analyses that considered only one domain or test of cognitive function, or cognitive data that were collected at a single time point. We believe our novel study in a large sample of middle-aged adults representing the general population provides the most comprehensive observational analysis to date. Our findings were consistent with the existing data in this research area. Observational studies have previously demonstrated an inverse association between television viewing and cognition (14–17) and a positive association between computer/internet use and cognition (15–18). However, before this study, the interactions with age or the deleterious influence of driving on cognitive health were less clear. The inverse associations of television viewing and driving time with cognitive function could be due to several factors. Cognition has previously been linked to cardiometabolic health (27, 28), and numerous studies have demonstrated inverse associations of television viewing and driving time with cardiometabolic health (9–12, 29–31). Therefore, it is possible that the observed associations act via pathways linked to the risk of vascular dysfunction and chronic diseases. Because vascular dysfunction and chronic diseases are linked to aging, this mechanism would also help explain the observed interactions with age. Other mediating factors could also explain the results for driving; it is known that driving is related to stress and fatigue (32), and with several studies previously showing the links between these factors and cognitive decline (33–35), it is plausible that the observed relationships are enhanced via this pathway. Furthermore, some types of sedentary behaviors, such as television viewing and driving, could possibly segregate individuals from social networks and restrict external collaborations, factors that are known to affect cognition (36–38); this again could be particularly important in older adults. In contrast, the positive relationship shared between computer use and cognitive function coincides with previous work where improved cognition or a lower risk of dementia was reported in those engaging in cognitively vitalizing sedentary behaviors or leisure activities (15–18). Therefore, as computer use is likely to involve some level of cognitive challenge, stimulate social interactions, and reduce solitariness, it may compensate for the associated sedentary behavior in relation to cognitive health. Some of the mechanisms mentioned above are also linked to and vary according to sex (39, 40), and they could therefore help explain the observed interactions with sex. The differences observed in cognitive function across the categories of sedentary behavior in our analyses are likely to be clinically important beyond the risk of cognitive decline. For example, higher fluid intelligence scores have previously been shown to be strongly associated with a lower risk of all-cause mortality (41, 42). In a sample of 5,572 middle-aged British adults, Sabia et al. (41) observed that a higher fluid intelligence score (1 SD) was associated with a 14% lower risk of all-cause mortality. Similarly, in a sample of 896 older Australian adults, Batterham et al. (42) observed that a higher fluid intelligence score (1 SD) was associated with a 24% lower risk of all-cause mortality. In our analysis at baseline (model 2), the standard deviation of fluid intelligence score was 2.1. Regression analyses investigating the associations of sedentary behaviors with fluid intelligence demonstrated that television viewing and driving time were linearly associated with lower fluid intelligence scores of 0.15 and 0.24 units, respectively. In contrast, computer-use time was linearly associated with a higher fluid intelligence score of 0.12 units. Hence, using the data above, it can be estimated that lower fluid intelligence scores by 0.15 and 0.24 units would equate approximately to a 1.1%–3.2% higher risk of all-cause mortality. In contrast, a higher fluid intelligence score by 0.12 units would equate approximately to a 0.9%–1.6% lower risk of all-cause mortality. For more details on these calculations, see Web Appendix 5. Strengths and limitations This study has several strengths and some limitations. Strengths include access to data on a large sample of adults representing the national population, follow-up cognitive function data enabling prospective investigation of associations, evaluation of dose-response and linear relationships between mutually adjusted and time-quantified sedentary behaviors with a wide range of cognitive outcomes, detailed covariate data enabling control for several important and relevant factors, analysis of interactions with age and sex, and robust sensitivity analyses investigating associations in the healthy population. However, although the UK Biobank is representative of the general population with respect to age, sex, ethnicity, and deprivation within the age range recruited, it may not be representative in other regards (43). While this limits the ability to generalize prevalence rates, estimates of the magnitude of associations in our study are unlikely to have been substantially affected by this due to the large and multifaceted base population (43, 44). Furthermore, the cognitive data from the UK Biobank cohort has recently been shown to be an important and valid resource for investigating predictors and modifiers of cognitive abilities and associated health outcomes in the general population (45). The sedentary behavior data used in this study have strengths and limitations. Only three sedentary domains were included; thus, the findings are restricted and cannot be generalized to other types of sedentary behavior. Self-reported assessments of sedentary behavior are subjective and are influenced by recall and response issues (46, 47); hence, they tend to have low validity and increase the risk of regression dilution. However, although data that are more robust can be obtained using objective measurement tools (e.g., accelerometers) (46, 47), they would not provide information on the specific type of sedentary behavior performed. Furthermore, because the reasons for using the computer outside work were unknown (e.g., computers could be used for such activities as reading, watching videos, internet browsing, or playing games), it is not possible to accurately classify or infer the type of computer use undertaken, and it may have involved crossover into cognitively inert tasks. Additionally, only those who provided an e-mail address at baseline (approximately 300,000) were contacted to participate in the online follow-up of cognitive function. These participants all had computer access and presumably some computer-use experience. This may also have resulted in the small differences in characteristics (including level of education and employment status) in the follow-up sample (see Web Table 1). Consequently, the prospective analysis may be biased and lack generalizability. Moreover, at baseline, the cognitive function tests were implemented using questionnaires that were administered via a touchscreen interface. At follow-up, the measurements were obtained remotely via online questionnaires that were administered on a computer via a mouse interface. This difference in the mode of administration could possibly account for some of the variability in cognitive performance and change over time. Nevertheless, the prospective analysis broadly supports and is consistent with the cross-sectional associations reported for the full cohort at baseline. Although we adjusted for a wide range of covariates, some unmeasured factors (e.g., type of employment/occupation) may have further confounded the reported associations. Our results may be subject to residual confounding or reverse causality. For example, it is possible that the positive association observed between computer use and cognitive function simply reflected greater familiarity with interacting with a computer rather than better cognitive function as such. Correspondingly, individuals with better cognitive function are more likely to engage in healthy behaviors and abstain from unhealthy ones, a concept known as neuroselection (48, 49). Also, although we investigated interactions with age and sex in our study, it must be highlighted that similar differences observed in cognitive function across different groups (i.e., in younger adults vs. older adults and women vs. men) may have different clinical meanings and should be interpreted with caution. For example, a unit difference in the cognitive function test score of a younger adult may not have the same result or significance for cognitive health as a unit difference in the score of an older adult. Last, due to large variations between the numbers of individuals who completed each cognitive assessment at both baseline and follow-up, analyses were based on different sample sizes. Conclusions Our analysis, conducted in a large national sample of adults, demonstrated that some sedentary domains, but not all, are associated with poor cognition. Watching television and driving were inversely associated with cognitive function, whereas computer use was positively associated with cognitive function. Of note, the associations were consistently stronger in older adults. Intervention studies are required to confirm these findings. Nevertheless, these results provide robust observational data supporting public health policies aimed at reducing television viewing and driving time in adults. ACKNOWLEDGMENTS Author affiliations: Department of Health Sciences, University of Leicester, Leicester General Hospital, Leicester, United Kingdom (Kishan Bakrania); Diabetes Research Centre, University of Leicester, Leicester General Hospital, Leicester, United Kingdom (Kishan Bakrania, Charlotte L. Edwardson, Kamlesh Khunti, Melanie J. Davies, Thomas Yates); Leicester Diabetes Centre, University Hospitals of Leicester, Leicester General Hospital, Leicester, United Kingdom (Kishan Bakrania, Charlotte L. Edwardson, Kamlesh Khunti, Melanie J. Davies, Thomas Yates); National Institute for Health Research Leicester Biomedical Research Centre, Leicester General Hospital, Leicester, United Kingdom (Kishan Bakrania, Charlotte L. Edwardson, Melanie J. Davies, Thomas Yates); National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care—East Midlands, Diabetes Research Centre, Leicester General Hospital, Leicester, United Kingdom (Kishan Bakrania, Kamlesh Khunti); and School of Sport, Exercise, and Health Sciences, Loughborough University, Loughborough, United Kingdom (Stephan Bandelow). This research was conducted using the UK Biobank resource (application 10813). The UK Biobank was established by the Wellcome Trust, Medical Research Council, Department of Health, Scottish Government, and the Northwest Regional Development Agency. It has also had funding from the Welsh Assembly Government and the British Heart Foundation, and it is supported by the National Health Service. This research was supported by the National Institute for Health Research Leicester Biomedical Research Center, the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care—East Midlands, and the Leicester Clinical Trials Unit. All the data reported in this study are fully available via application to the United Kingdom Biobank. The views expressed are those of the authors and not necessarily those of the National Health Service, the National Institute for Health Research, or the Department of Health. Conflict of interest: none declared. Abbreviations CI confidence interval SD standard deviation REFERENCES 1 Lautenschlager NT, Cox KL, Flicker L, et al.  . Effect of physical activity on cognitive function in older adults at risk for Alzheimer disease: a randomized trial. JAMA . 2008; 300( 9): 1027– 1037. Google Scholar CrossRef Search ADS PubMed  2 Laurin D, Verreault R, Lindsay J, et al.  . Physical activity and risk of cognitive impairment and dementia in elderly persons. Arch Neurol . 2001; 58( 3): 498– 504. Google Scholar CrossRef Search ADS PubMed  3 Erickson KI, Voss MW, Prakash RS, et al.  . Exercise training increases size of hippocampus and improves memory. Proc Natl Acad Sci USA . 2011; 108( 7): 3017– 3022. Google Scholar CrossRef Search ADS PubMed  4 Colcombe SJ, Kramer AF, Erickson KI, et al.  . Cardiovascular fitness, cortical plasticity, and aging. Proc Natl Acad Sci USA . 2004; 101( 9): 3316– 3321. Google Scholar CrossRef Search ADS PubMed  5 Chapman SB, Aslan S, Spence JS, et al.  . Shorter term aerobic exercise improves brain, cognition, and cardiovascular fitness in aging. Front Aging Neurosci . 2013; 5: 75. Google Scholar CrossRef Search ADS PubMed  6 Colcombe SJ, Erickson KI, Scalf PE, et al.  . Aerobic exercise training increases brain volume in aging humans. J Gerontol A Biol Sci Med Sci . 2006; 61( 11): 1166– 1170. Google Scholar CrossRef Search ADS PubMed  7 Smith JC, Nielson KA, Antuono P, et al.  . Semantic memory functional MRI and cognitive function after exercise intervention in mild cognitive impairment. J Alzheimers Dis . 2013; 37( 1): 197– 215. Google Scholar PubMed  8 Sedentary Behaviour Research Network. Letter to the editor: standardized use of the terms “sedentary” and “sedentary behaviours”. Appl Physiol Nutr Metab . 2012; 37( 3): 540– 542. CrossRef Search ADS PubMed  9 Biswas A, Oh PI, Faulkner GE, et al.  . Sedentary time and its association with risk for disease incidence, mortality, and hospitalization in adults: a systematic review and meta-analysis. Ann Intern Med . 2015; 162( 2): 123– 132. Google Scholar CrossRef Search ADS PubMed  10 Edwardson CL, Gorely T, Davies MJ, et al.  . Association of sedentary behaviour with metabolic syndrome: a meta-analysis. PLoS One . 2012; 7( 4): e34916. Google Scholar CrossRef Search ADS PubMed  11 Grøntved A, Hu FB. Television viewing and risk of type 2 diabetes, cardiovascular disease, and all-cause mortality: a meta-analysis. JAMA . 2011; 305( 23): 2448– 2455. Google Scholar CrossRef Search ADS PubMed  12 Wilmot EG, Edwardson CL, Achana FA, et al.  . Sedentary time in adults and the association with diabetes, cardiovascular disease and death: systematic review and meta-analysis. Diabetologia . 2012; 55( 11): 2895– 2905. Google Scholar CrossRef Search ADS PubMed  13 Falck RS, Davis JC, Liu-Ambrose T. What is the association between sedentary behaviour and cognitive function? A systematic review. Br J Sports Med . 2017; 51( 10): 800– 811. Google Scholar CrossRef Search ADS PubMed  14 Hoang TD, Reis J, Zhu N, et al.  . Effect of early adult patterns of physical activity and television viewing on midlife cognitive function. JAMA Psychiatry . 2016; 73( 1): 73– 79. Google Scholar CrossRef Search ADS PubMed  15 Wang JY, Zhou DH, Li J, et al.  . Leisure activity and risk of cognitive impairment: the Chongqing aging study. Neurology . 2006; 66( 6): 911– 913. Google Scholar CrossRef Search ADS PubMed  16 Hamer M, Stamatakis E. Prospective study of sedentary behavior, risk of depression, and cognitive impairment. Med Sci Sports Exerc . 2014; 46( 4): 718– 723. Google Scholar CrossRef Search ADS PubMed  17 Kesse-Guyot E, Charreire H, Andreeva VA, et al.  . Cross-sectional and longitudinal associations of different sedentary behaviors with cognitive performance in older adults. PLoS One . 2012; 7( 10): e47831. Google Scholar CrossRef Search ADS PubMed  18 Tun PA, Lachman ME. The association between computer use and cognition across adulthood: use it so you won’t lose it? Psychol Aging . 2010; 25( 3): 560– 568. Google Scholar CrossRef Search ADS PubMed  19 Sudlow C, Gallacher J, Allen N, et al.  . UK Biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med . 2015; 12( 3): e1001779. Google Scholar CrossRef Search ADS PubMed  20 Allen N, Sudlow C, Downey P, et al.  . UK Biobank: current status and what it means for epidemiology. Health Policy Technol . 2012; 1( 3): 123– 126. Google Scholar CrossRef Search ADS   21 UK Biobank. UK Biobank: Protocol for a large-scale prospective epidemiological resource. 2016. http://www.ukbiobank.ac.uk/wp-content/uploads/2011/11/UK-Biobank-Protocol.pdf. Accessed March 10, 2017. 22 Iwasa H, Yoshida Y, Kai I, et al.  . Leisure activities and cognitive function in elderly community-dwelling individuals in Japan: a 5-year prospective cohort study. J Psychosom Res . 2012; 72( 2): 159– 164. Google Scholar CrossRef Search ADS PubMed  23 Wilson RS, Mendes De Leon CF, Barnes LL, et al.  . Participation in cognitively stimulating activities and risk of incident Alzheimer disease. JAMA . 2002; 287( 6): 742– 748. Google Scholar CrossRef Search ADS PubMed  24 Verghese J, Lipton RB, Katz MJ, et al.  . Leisure activities and the risk of dementia in the elderly. N Engl J Med . 2003; 348( 25): 2508– 2516. Google Scholar CrossRef Search ADS PubMed  25 Akbaraly TN, Portet F, Fustinoni S, et al.  . Leisure activities and the risk of dementia in the elderly: results from the Three-City Study. Neurology . 2009; 73( 11): 854– 861. Google Scholar CrossRef Search ADS PubMed  26 Aggio D, Smith L, Fisher A, et al.  . Context-specific associations of physical activity and sedentary behavior with cognition in children. Am J Epidemiol . 2016; 183( 12): 1075– 1082. Google Scholar CrossRef Search ADS PubMed  27 Elias MF, Sullivan LM, D’Agostino RB, et al.  . Framingham stroke risk profile and lowered cognitive performance. Stroke . 2004; 35( 2): 404– 409. Google Scholar CrossRef Search ADS PubMed  28 Kaffashian S, Dugravot A, Nabi H, et al.  . Predictive utility of the Framingham general cardiovascular disease risk profile for cognitive function: evidence from the Whitehall II study. Eur Heart J . 2011; 32( 18): 2326– 2332. Google Scholar CrossRef Search ADS PubMed  29 Belkic K, Savic C, Theorell T, et al.  . Mechanisms of cardiac risk among professional drivers. Scand J Work Environ Health . 1994; 20( 2): 73– 86. Google Scholar CrossRef Search ADS PubMed  30 Morris JN, Heady JA, Raffle PA, et al.  . Coronary heart-disease and physical activity of work. Lancet . 1953; 265( 6796): 1111– 1120. Google Scholar CrossRef Search ADS PubMed  31 Robinson CF, Burnett CA. Truck drivers and heart disease in the United States, 1979–1990. Am J Ind Med . 2005; 47( 2): 113– 119. Google Scholar CrossRef Search ADS PubMed  32 Howard A, Joint M. Fatigue and stress in driving. Empl Couns Today . 1994; 6( 6): 3– 7. Google Scholar CrossRef Search ADS   33 van der Linden D, Frese M, Meijman TF. Mental fatigue and the control of cognitive processes: effects on perseveration and planning. Acta Psychol (Amst) . 2003; 113( 1): 45– 65. Google Scholar CrossRef Search ADS PubMed  34 Abd-Elfattah HM, Abdelazeim FH, Elshennawy S. Physical and cognitive consequences of fatigue: a review. J Adv Res . 2015; 6( 3): 351– 358. Google Scholar CrossRef Search ADS PubMed  35 Scott SB, Graham-Engeland JE, Engeland CG, et al.  . The Effects of Stress on Cognitive Aging, Physiology and Emotion (ESCAPE) project. BMC Psychiatry . 2015; 15: 146. Google Scholar CrossRef Search ADS PubMed  36 Shankar A, Hamer M, McMunn A, et al.  . Social isolation and loneliness: relationships with cognitive function during 4 years of follow-up in the English Longitudinal Study of Ageing. Psychosom Med . 2013; 75( 2): 161– 170. Google Scholar CrossRef Search ADS PubMed  37 Bassuk SS, Glass TA, Berkman LF. Social disengagement and incident cognitive decline in community-dwelling elderly persons. Ann Intern Med . 1999; 131( 3): 165– 173. Google Scholar CrossRef Search ADS PubMed  38 Golden J, Conroy RM, Bruce I, et al.  . Loneliness, social support networks, mood and wellbeing in community-dwelling elderly. Int J Geriatr Psychiatry . 2009; 24( 7): 694– 700. Google Scholar CrossRef Search ADS PubMed  39 Mosca L, Barrett-Connor E, Wenger NK. Sex/gender differences in cardiovascular disease prevention: what a difference a decade makes. Circulation . 2011; 124( 19): 2145– 2154. Google Scholar CrossRef Search ADS PubMed  40 McCarrey AC, An Y, Kitner-Triolo MH, et al.  . Sex differences in cognitive trajectories in clinically normal older adults. Psychol Aging . 2016; 31( 2): 166– 175. Google Scholar CrossRef Search ADS PubMed  41 Sabia S, Guéguen A, Marmot MG, et al.  . Does cognition predict mortality in midlife? Results from the Whitehall II cohort study. Neurobiol Aging . 2010; 31( 4): 688– 695. Google Scholar CrossRef Search ADS PubMed  42 Batterham PJ, Christensen H, Mackinnon AJ. Fluid intelligence is independently associated with all-cause mortality over 17 years in an elderly community sample: an investigation of potential mechanisms. Intelligence . 2009; 37( 6): 551– 560. Google Scholar CrossRef Search ADS   43 Collins R. What makes UK Biobank special? Lancet . 2012; 379( 9822): 1173– 1174. Google Scholar CrossRef Search ADS PubMed  44 Manolio TA, Collins R. Enhancing the feasibility of large cohort studies. JAMA . 2010; 304( 20): 2290– 2291. Google Scholar CrossRef Search ADS PubMed  45 Lyall DM, Cullen B, Allerhand M, et al.  . Cognitive test scores in UK Biobank: data reduction in 480,416 participants and longitudinal stability in 20,346 participants. PLoS One . 2016; 11( 4): e0154222. Google Scholar CrossRef Search ADS PubMed  46 Troiano RP, Berrigan D, Dodd KW, et al.  . Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc . 2008; 40( 1): 181– 188. Google Scholar CrossRef Search ADS PubMed  47 Prince SA, Adamo KB, Hamel ME, et al.  . A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act . 2008; 5: 56. Google Scholar CrossRef Search ADS PubMed  48 Belsky DW. Invited commentary: lessons for research on cognitive aging from a study of children. Am J Epidemiol . 2016; 183( 12): 1083– 1085. Google Scholar CrossRef Search ADS PubMed  49 Aggio D, Smith L, Fisher A, et al.  . Aggio et al. Respond to “Lessons for Research on Cognitive Aging”. Am J Epidemiol . 2016; 183( 12): 1086– 1087. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Journal of Epidemiology Oxford University Press

Associations Between Sedentary Behaviors and Cognitive Function: Cross-Sectional and Prospective Findings From the UK Biobank

Loading next page...
 
/lp/ou_press/associations-between-sedentary-behaviors-and-cognitive-function-cross-FgAxgiwxXV
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0002-9262
eISSN
1476-6256
D.O.I.
10.1093/aje/kwx273
Publisher site
See Article on Publisher Site

Abstract

Abstract We investigated the cross-sectional and prospective associations between different sedentary behaviors and cognitive function in a large sample of adults with data stored in the UK Biobank. Baseline data were available for 502,643 participants (2006–2010, United Kingdom). Cognitive tests included prospective memory (baseline only: n = 171,585), visual-spatial memory (round 1: n = 483,832; round 2: n = 482,762), fluid intelligence (n = 165,492), and short-term numeric memory (n = 50,370). After a mean period of 5.3 years, participants (numbering from 12,091 to 114,373, depending on the test) also provided follow-up cognitive data. Sedentary behaviors (television viewing, driving, and nonoccupational computer-use time) were measured at baseline. At baseline, both television viewing and driving time were inversely associated with cognitive function across all outcomes (e.g., for each additional hour spent watching television, the total number of correct answers in the fluid intelligence test was 0.15 (99% confidence interval: 0.14, 0.16) lower. Computer-use time was positively associated with cognitive function across all outcomes. Both television viewing and driving time at baseline were positively associated with the odds of having cognitive decline at follow-up across most outcomes. Conversely, computer-use time at baseline was inversely associated with the odds of having cognitive decline at follow-up across most outcomes. This study supports health policies designed to reduce television viewing and driving in adults. cognitive decline, cognitive function, computer use, driving, epidemiology, sedentary behavior, television viewing, UK Biobank Currently, there are no effective long-term pharmacological therapies for the treatment or prevention of dementia. Therefore, identifying potentially modifiable risk factors of cognitive decline, a major characteristic of dementia, is a key priority. Engaging in healthy lifestyle practices, including physical activity, has been associated with a reduced risk of dementia and its symptoms, such as cognitive impairment (1, 2), suggesting a potential role for lifestyle therapies. Indeed, physical-activity intervention studies have shown changes to the structure and function of the brain (3–7), supporting the observational associations. Along with physical activity, engaging in sedentary behavior, defined as sitting or reclining with low energy expenditure (8), could also be an important determinant of poor cognitive function. There is cumulative evidence indicating that sedentary time is associated with poor cardiometabolic health, chronic disease, and mortality (9–12). A recent systematic review also suggested that sedentary behavior is negatively associated with cognitive function, although the relationship between the two is complex, and recommended that future studies should focus on determining how different sedentary behaviors are associated with cognitive function (13). Limited observational research has indicated that television viewing is inversely associated with cognition (14–17). However, different sedentary behaviors may have different associations, with some evidence of computer/internet use being linked to cognitive improvement (15–18). Furthermore, most of the existing data have emerged from relatively limited cross-sectional findings (16–18). Therefore, this warrants investigation in large-scale studies with prospective data. The aim of this paper was to use the nationally representative UK Biobank cohort to examine the cross-sectional and prospective associations between domains of sedentary behavior (television viewing, driving, and computer use) and cognitive function (prospective memory, visual-spatial memory, fluid intelligence, and short-term numeric memory). METHODS Study design and population The UK Biobank is a large prospective study of the middle-aged population in the United Kingdom (19–21). Approximately 500,000 adults (aged 37–73 years) were recruited during 2006–2010 via mailed invitations to those registered with the National Health Service (NHS) and living within 25 miles of one of the 22 study assessment centers. Participants provided comprehensive baseline data on a broad range of biological, cognitive, demographic, health, lifestyle, mental, social, and well-being outcomes. Approximately 300,000 participants also provided an e-mail address to allow for remote follow-up for cognitive function measures in the future. During 2014–2015, approximately 125,000 participants provided some follow-up cognitive function data online. For the present study, baseline data were available on 502,643 individuals. Of these, depending on the cognitive test, between 50,370 and 483,832 participants provided baseline cognitive function data (see Web Figure 1, available at https://academic.oup.com/aje). Of these, after a mean period of 5.3 years, participants ranging in number from 12,091 to 114,373, depending on the test, also provided follow-up cognitive function data online (see Web Figure 2). All participants provided written informed consent, and the study was approved by the National Health Service’s National Research Ethics Service (Ref: 11/NW/0382). Further details are available elsewhere (19–21). Cognitive function tests Questionnaires administered through a computerized touchscreen interface assessed cognitive function at baseline. Using the same methodology without the touchscreen ability, follow-up measurements were obtained via online questionnaires that were completed remotely. To ensure effortless application on a large scale and wide response distributions, the cognitive function tests, which were refined in pilot study, were designed comprehensively and specifically for the UK Biobank. Prospective memory (available at baseline only), visual-spatial memory, fluid intelligence, and short-term numeric memory tests were included in this analysis. At baseline, there were variations between the numbers of individuals who completed each cognitive assessment due to tests being abandoned or skipped by participants, incorporated towards the end of recruitment (e.g., fluid intelligence), or phased out during the early stages of recruitment (e.g., short-term numeric memory). For more details on the cognitive function tests, see Web Appendix 1. Sedentary behaviors Data on sedentary behaviors were self-reported and collected at baseline using a computerized questionnaire. Domains of sedentary behavior included television viewing time (in hours/day: <1, 1, 2, 3, ≥4), driving time (in hours/day: <1, 1, 2, ≥3), and nonoccupational computer-use time (in hours/day: <1, 1, 2, ≥3). For more details, see Web Appendix 2. Covariate data Covariate data included anthropometric (body mass index), demographic (age, sex, ethnicity, social deprivation index, employment status, education level), health (number of cancers, number of noncancer illnesses, number of medications/treatments), and lifestyle (smoking status, alcohol drinking status, sleep duration, fruit and vegetable consumption, physical activity) variables. For more details, see Web Appendix 3. Statistical analysis Statistical analyses were performed using Stata/MP, version 14.0 (StataCorp LP, College Station, Texas). Data were analyzed in February 2017. With the intention of maximizing the use of the data, pairwise deletion was used to handle missing data (see Web Figures 1 and 2). Participant characteristics were tabulated. Categorical variables are presented as numbers and proportions, whereas continuous variables were summarized as mean values and standard deviations and are presented with their minimum and maximum values. Cross-sectional analysis Regression analysis was used to examine the cross-sectional associations between the domains of sedentary behavior and cognitive function at baseline. Multiple logistic regression models were fitted for each binary cognitive outcome variable (prospective memory, visual-spatial memory (round 1), and visual-spatial memory (round 2)). Multiple linear regression models were fitted for each continuous cognitive outcome variable (fluid intelligence and short-term numeric memory). For more details on the nature of the cognitive outcome variables used in the cross-sectional analysis, see Web Appendix 1. Model 1 mutually adjusted for the other sedentary behaviors and for age and sex. Model 2 further adjusted for body mass index, ethnicity, social deprivation index, employment status, educational level, smoking status, alcohol drinking status, fruit and vegetable consumption, sleep duration, physical activity (frequency of ≥10 minutes of walking (days/week), frequency of ≥10 minutes of moderate physical activity (days/week), frequency of ≥10 minutes of vigorous physical activity (days/week)), number of cancers, number of noncancer illnesses, and number of medications/treatments. For each sedentary behavior, the “<1 hour/day” category was selected as the reference group. Linear trends (linear terms) across the categories of each sedentary behavior were reported. Interaction terms were separately added to the full-adjustment model (model 2) to observe whether the associations between the sedentary behaviors and cognitive function were modified by age or sex. Significant results for age were stratified at 60 years. Prospective analysis Multiple logistic regression models investigated the prospective associations between the domains of sedentary behavior at baseline and cognitive function at follow-up. These models estimated the odds of having cognitive decline (i.e., a poor outcome) at follow-up. Cognitive outcomes included visual-spatial memory (round 1), visual-spatial memory (round 2), fluid intelligence, and short-term numeric memory. For full details on the definitions and nature of the cognitive outcome variables used in the prospective analysis, see Web Appendix 1. As well as controlling for the baseline result/score of the cognitive test under consideration, models adjusted for all the covariates mentioned previously. Linear trends across the categories of each sedentary behavior were reported. Interactions by age and sex were also investigated. Sensitivity analysis To assess the generalizability of our findings, the cross-sectional and prospective analyses investigating the associations between sedentary behaviors and cognitive function (model 1 and model 2) were repeated across the sample of participants without a medical history of cancer, cardiovascular disease, or cognitive/psychiatric illnesses as sensitivity analyses. For more details on the specific diseases/illnesses, see Web Appendix 4. Statistical reporting For each variable of interest (sedentary behaviors), the β coefficient (linear regression) or odds ratio (logistic regression) with 99% confidence intervals and P values are reported. All analyses employed robust standard errors, and all reported P values are 2-sided. To account for multiple comparisons, P < 0.01 was considered to be statistically significant for the main analyses. For the interaction analyses, P < 0.05 was considered to be statistically significant. RESULTS Cross-sectional findings Table 1 presents the characteristics of the 502,643 participants with baseline data. The mean age of these individuals was 56.5 (standard deviation (SD), 8.1) years, and 273,467 (54.4%) were female. Table 1. Baseline Characteristics of UK Biobank Participants (n = 502,643), United Kingdom, 2006–2010 Characteristic  No.  %  Mean (SD)  Range  Body mass indexa,b      27.4 (4.8)  12.1–74.7   Missing  3,105  0.6      Demographic characteristic           Age, yearsb      56.5 (8.1)  37.0–73.0    Missing  0  0.0       Ethnicityc            White British  442,699  88.1        Other  57,166  11.4        Missing  2,778  0.5       Sexc            Female  273,467  54.4        Male  229,176  45.6        Missing  0  0.0       Social deprivation indexb      −1.3 (3.1)  −6.3–11.0    Missing  627  0.1       Employment statusc            In paid employment or self-employed  287,234  57.1        Not in paid employment or self-employed  212,451  42.3        Missing  2,958  0.6       Educational levelc            No college or university degree  331,291  65.9        College or university degree  161,210  32.1        Missing  10,142  2.0      Lifestyle           Smoking statusc            Never  273,603  54.4        Previous  173,099  34.4        Current  52,989  10.6        Missing  2,952  0.6       Alcohol drinking statusc            Never  22,547  4.5        Previous  18,114  3.6        Current  460,479  91.6        Missing  1,503  0.3       Fruit and vegetable consumption, portions/dayc            <5  300,352  59.8        ≥5  189,979  37.8        Missing  12,132  2.4       Sleep duration, hours/dayb      7.2 (1.1)  1.0–23.0    Missing  4,218  0.8       Frequency of ≥10 minutes of walking, days/weekc            0  12,455  2.5        1  13,459  2.7        2  29,991  6.0        3  39,339  7.8        4  40,036  8.0        5  80,039  15.9        6  50,082  9.9        7  228,697  45.5        Missing  8,545  1.7       Frequency of ≥10 minutes of moderate physical activity, days/weekc            0  61,178  12.2        1  38,290  7.6        2  69,799  13.9        3  71,507  14.2        4  47,201  9.4        5  71,441  14.2        6  26,436  5.3        7  89,506  17.8        Missing  27,285  5.4       Frequency of ≥10 minutes of vigorous physical activity, days/weekc            0  178,275  35.5        1  66,853  13.3        2  75,055  14.9        3  65,276  13.0        4  30,705  6.1        5  32,452  6.5        6  9,430  1.9        7  17,005  3.4        Missing  27,592  5.5      Health           Number of cancersc            0  460,075  91.5        ≥1  41,706  8.3        Missing  862  0.2       Number of noncancer illnessesc            0  126,639  25.2        1  134,113  26.7        2  98,825  19.6        3  62,828  12.5        ≥4  79,376  15.8        Missing  862  0.2       Number of medications/treatmentsc            0  137,704  27.4        1  94,776  18.8        2  77,673  15.4        3  57,819  11.5        4  42,211  8.4        5  29,937  6.0        ≥6  61,661  12.3        Missing  862  0.2       Medical history of cancer, cardiovascular disease, or cognitive/  psychiatric illnessesc            No  402,897  80.2        Yes  99,746  19.8        Missing  0  0.0      Sedentary behaviors           Television viewing time, hours/dayc            <1  39,456  7.8        1  62,503  12.4        2  132,780  26.4        3  116,940  23.3        ≥4  145,546  29.0        Missing  5,418  1.1       Driving time, hours/dayc            <1  259,920  51.7        1  140,144  27.9        2  60,977  12.1        ≥3  31,663  6.3        Missing  9,939  2.0       Computer-use time, hours/dayc            <1  240,648  47.9        1  140,821  28.0        2  62,859  12.5        ≥3  48,939  9.7        Missing  9,376  1.9      Cognitive function at baseline           Prospective memory testc,d            Good result  130,910  26.0        Poor result  40,675  8.1        Missing  331,058  65.9       Visual-spatial memory test (round 1)c,e            Good result  345,685  68.8        Poor result  138,147  27.5        Missing  18,811  3.7       Visual-spatial memory test (round 2)c,f            Good result  82,130  16.3        Poor result  400,632  79.7        Missing  19,881  4.0       Fluid intelligence testb,g            Total number of correct answers      6.0 (2.2)  0.0–13.0    Missing  337,151  67.1       Short-term numeric memory testb,h            Maximum digits remembered correctly      6.7 (1.3)  2.0–12.0    Missing  452,273  90.0      Characteristic  No.  %  Mean (SD)  Range  Body mass indexa,b      27.4 (4.8)  12.1–74.7   Missing  3,105  0.6      Demographic characteristic           Age, yearsb      56.5 (8.1)  37.0–73.0    Missing  0  0.0       Ethnicityc            White British  442,699  88.1        Other  57,166  11.4        Missing  2,778  0.5       Sexc            Female  273,467  54.4        Male  229,176  45.6        Missing  0  0.0       Social deprivation indexb      −1.3 (3.1)  −6.3–11.0    Missing  627  0.1       Employment statusc            In paid employment or self-employed  287,234  57.1        Not in paid employment or self-employed  212,451  42.3        Missing  2,958  0.6       Educational levelc            No college or university degree  331,291  65.9        College or university degree  161,210  32.1        Missing  10,142  2.0      Lifestyle           Smoking statusc            Never  273,603  54.4        Previous  173,099  34.4        Current  52,989  10.6        Missing  2,952  0.6       Alcohol drinking statusc            Never  22,547  4.5        Previous  18,114  3.6        Current  460,479  91.6        Missing  1,503  0.3       Fruit and vegetable consumption, portions/dayc            <5  300,352  59.8        ≥5  189,979  37.8        Missing  12,132  2.4       Sleep duration, hours/dayb      7.2 (1.1)  1.0–23.0    Missing  4,218  0.8       Frequency of ≥10 minutes of walking, days/weekc            0  12,455  2.5        1  13,459  2.7        2  29,991  6.0        3  39,339  7.8        4  40,036  8.0        5  80,039  15.9        6  50,082  9.9        7  228,697  45.5        Missing  8,545  1.7       Frequency of ≥10 minutes of moderate physical activity, days/weekc            0  61,178  12.2        1  38,290  7.6        2  69,799  13.9        3  71,507  14.2        4  47,201  9.4        5  71,441  14.2        6  26,436  5.3        7  89,506  17.8        Missing  27,285  5.4       Frequency of ≥10 minutes of vigorous physical activity, days/weekc            0  178,275  35.5        1  66,853  13.3        2  75,055  14.9        3  65,276  13.0        4  30,705  6.1        5  32,452  6.5        6  9,430  1.9        7  17,005  3.4        Missing  27,592  5.5      Health           Number of cancersc            0  460,075  91.5        ≥1  41,706  8.3        Missing  862  0.2       Number of noncancer illnessesc            0  126,639  25.2        1  134,113  26.7        2  98,825  19.6        3  62,828  12.5        ≥4  79,376  15.8        Missing  862  0.2       Number of medications/treatmentsc            0  137,704  27.4        1  94,776  18.8        2  77,673  15.4        3  57,819  11.5        4  42,211  8.4        5  29,937  6.0        ≥6  61,661  12.3        Missing  862  0.2       Medical history of cancer, cardiovascular disease, or cognitive/  psychiatric illnessesc            No  402,897  80.2        Yes  99,746  19.8        Missing  0  0.0      Sedentary behaviors           Television viewing time, hours/dayc            <1  39,456  7.8        1  62,503  12.4        2  132,780  26.4        3  116,940  23.3        ≥4  145,546  29.0        Missing  5,418  1.1       Driving time, hours/dayc            <1  259,920  51.7        1  140,144  27.9        2  60,977  12.1        ≥3  31,663  6.3        Missing  9,939  2.0       Computer-use time, hours/dayc            <1  240,648  47.9        1  140,821  28.0        2  62,859  12.5        ≥3  48,939  9.7        Missing  9,376  1.9      Cognitive function at baseline           Prospective memory testc,d            Good result  130,910  26.0        Poor result  40,675  8.1        Missing  331,058  65.9       Visual-spatial memory test (round 1)c,e            Good result  345,685  68.8        Poor result  138,147  27.5        Missing  18,811  3.7       Visual-spatial memory test (round 2)c,f            Good result  82,130  16.3        Poor result  400,632  79.7        Missing  19,881  4.0       Fluid intelligence testb,g            Total number of correct answers      6.0 (2.2)  0.0–13.0    Missing  337,151  67.1       Short-term numeric memory testb,h            Maximum digits remembered correctly      6.7 (1.3)  2.0–12.0    Missing  452,273  90.0      Abbreviation: SD, standard deviation. a Weight (kg)/height (m)2. b Continuous variable. c Categorical variable. d Prospective memory test: good result (correct recall on first attempt) or poor result (incorrect recall on first attempt (i.e., correct recall on second attempt, instruction not recalled, skipped or incorrect)). e Pairs matching test (round 1): good result (0 incorrect matches) or poor result (≥1 incorrect matches). f Pairs matching test (round 2): good result (<2 incorrect matches) or poor result (≥2 incorrect matches). g Fluid intelligence score: total number of correct answers. h Numeric memory score: maximum digits remembered correctly. Table 2 presents the associations between the sedentary behaviors and cognitive function. In the full-adjustment models (model 2), the cross-sectional data showed that television viewing time was inversely associated with cognitive function across all outcomes apart from visual-spatial memory (round 2). For example, for each additional hour spent watching television up to ≥4 hours/day, the fluid intelligence and short-term numeric memory scores were 0.15 (99% confidence interval (CI): 0.14, 0.16) and 0.09 (99% CI: 0.07, 0.10) units lower, respectively. Correspondingly, the odds of a poor result in the prospective memory and visual-spatial memory (round 1) tests were 2% (99% CI: 0, 3) and 3% (99% CI: 2, 4) higher, respectively. Driving time was inversely associated with cognitive function across all outcomes. In contrast, computer-use time was positively associated with cognitive function across all outcomes. Table 2. Cross-Sectional Associations at Baseline Between Sedentary Behaviors and Cognitive Function Among UK Biobank Participants, United Kingdom, 2006–2010 Sedentary Behavior  Prospective Memory Testa  Visual-Spatial Memory Test  Fluid Intelligence Testd  Short-Term Numeric Memory Teste  Round 1b  Round 2c  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  β  99% CI  P Valuef  β  99% CI  P Valuef  Model 1g  Television viewing time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.99  0.93, 1.07  0.838  1.05  1.01, 1.09  0.002  1.01  0.96, 1.05  0.712  −0.13  −0.19, −0.07  <0.001  −0.15  −0.22, −0.08  <0.001   2  0.94  0.88, 1.00  0.012  1.07  1.03, 1.11  <0.001  1.02  0.98, 1.06  0.303  −0.30  −0.36, −0.24  <0.001  −0.25  −0.31, −0.19  <0.001   3  0.97  0.91, 1.04  0.303  1.14  1.10, 1.18  <0.001  1.03  0.99, 1.08  0.036  −0.54  −0.60, −0.49  <0.001  −0.35  −0.41, −0.29  <0.001   ≥4  1.23  1.15, 1.30  <0.001  1.26  1.22, 1.31  <0.001  1.08  1.03, 1.12  <0.001  −0.99  −1.04, −0.93  <0.001  −0.55  −0.61, −0.48  <0.001   Linear trend  1.07  1.05, 1.08  <0.001  1.06  1.06, 1.07  <0.001  1.02  1.01, 1.03  <0.001  −0.26  −0.27, −0.25  <0.001  −0.13  −0.15, −0.12  <0.001  Driving time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.05  1.02, 1.09  <0.001  0.99  0.97, 1.01  0.270  1.01  0.98, 1.03  0.569  −0.19  −0.22, −0.16  <0.001  −0.02  −0.06, 0.01  0.139   2  1.12  1.07, 1.18  <0.001  1.05  1.02, 1.08  <0.001  1.03  1.00, 1.06  0.024  −0.37  −0.41, −0.33  <0.001  −0.13  −0.18, −0.08  <0.001   ≥3  1.50  1.41, 1.60  <0.001  1.26  1.22, 1.31  <0.001  1.12  1.07, 1.17  <0.001  −0.88  −0.94, −0.82  <0.001  −0.28  −0.34, −0.21  <0.001   Linear trend  1.11  1.09, 1.12  <0.001  1.05  1.04, 1.06  <0.001  1.03  1.01, 1.04  <0.001  −0.24  −0.26, −0.23  <0.001  −0.08  −0.09, −0.06  <0.001  Computer-use time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.68  0.66, 0.71  <0.001  0.79  0.77, 0.80  <0.001  0.85  0.83, 0.87  <0.001  0.52  0.49, 0.55  <0.001  0.21  0.17, 0.25  <0.001   2  0.69  0.66, 0.72  <0.001  0.77  0.75, 0.79  <0.001  0.80  0.78, 0.83  <0.001  0.58  0.53, 0.62  <0.001  0.21  0.16, 0.26  <0.001   ≥3  0.86  0.82, 0.90  <0.001  0.81  0.79, 0.84  <0.001  0.82  0.79, 0.85  <0.001  0.40  0.35, 0.44  <0.001  0.16  0.11, 0.22  <0.001   Linear trend  0.91  0.89, 0.92  <0.001  0.91  0.90, 0.92  <0.001  0.92  0.91, 0.93  <0.001  0.18  0.17, 0.20  <0.001  0.07  0.06, 0.09  <0.001  Model 2h  Television viewing time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.04  0.96, 1.12  0.232  1.07  1.03, 1.12  <0.001  1.01  0.97, 1.06  0.403  −0.12  −0.18, −0.06  <0.001  −0.13  −0.20, −0.06  <0.001   2  0.96  0.90, 1.03  0.142  1.07  1.03, 1.11  <0.001  1.02  0.98, 1.07  0.128  −0.21  −0.27, −0.16  <0.001  −0.20  −0.26, −0.13  <0.001   3  0.96  0.89, 1.03  0.159  1.09  1.05, 1.14  <0.001  1.03  0.98, 1.07  0.136  −0.33  −0.39, −0.28  <0.001  −0.25  −0.32, −0.19  <0.001   ≥4  1.09  1.01, 1.17  0.003  1.14  1.10, 1.19  <0.001  1.03  0.99, 1.08  0.053  −0.58  −0.64, −0.53  <0.001  −0.38  −0.44, −0.31  <0.001   Linear trend  1.02  1.00, 1.03  0.001  1.03  1.02, 1.04  <0.001  1.01  1.00, 1.02  0.057  −0.15  −0.16, −0.14  <0.001  −0.09  −0.10, −0.07  <0.001  Driving time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.21  1.17, 1.27  <0.001  1.05  1.03, 1.07  <0.001  1.04  1.02, 1.07  <0.001  −0.28  −0.31, −0.25  <0.001  −0.06  −0.10, −0.03  <0.001   2  1.27  1.20, 1.34  <0.001  1.10  1.06, 1.13  <0.001  1.07  1.03, 1.10  <0.001  −0.43  −0.48, −0.39  <0.001  −0.18  −0.23, −0.13  <0.001   ≥3  1.54  1.43, 1.66  <0.001  1.23  1.19, 1.28  <0.001  1.11  1.06, 1.16  <0.001  −0.73  −0.79, −0.68  <0.001  −0.27  −0.34, −0.19  <0.001   Linear trend  1.15  1.13, 1.17  <0.001  1.06  1.05, 1.07  <0.001  1.04  1.02, 1.05  <0.001  −0.24  −0.25, −0.22  <0.001  −0.09  −0.11, −0.07  <0.001  Computer-use time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.77  0.74, 0.81  <0.001  0.85  0.83, 0.87  <0.001  0.88  0.86, 0.90  <0.001  0.32  0.29, 0.35  <0.001  0.14  0.10, 0.17  <0.001   2  0.74  0.70, 0.78  <0.001  0.81  0.79, 0.83  <0.001  0.83  0.80, 0.86  <0.001  0.40  0.36, 0.44  <0.001  0.15  0.10, 0.20  <0.001   ≥3  0.86  0.81, 0.91  <0.001  0.84  0.81, 0.86  <0.001  0.84  0.81, 0.88  <0.001  0.26  0.22, 0.31  <0.001  0.13  0.07, 0.18  <0.001   Linear trend  0.92  0.90, 0.94  <0.001  0.92  0.91, 0.93  <0.001  0.93  0.92, 0.94  <0.001  0.12  0.11, 0.14  <0.001  0.06  0.04, 0.07  <0.001  Sedentary Behavior  Prospective Memory Testa  Visual-Spatial Memory Test  Fluid Intelligence Testd  Short-Term Numeric Memory Teste  Round 1b  Round 2c  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  β  99% CI  P Valuef  β  99% CI  P Valuef  Model 1g  Television viewing time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.99  0.93, 1.07  0.838  1.05  1.01, 1.09  0.002  1.01  0.96, 1.05  0.712  −0.13  −0.19, −0.07  <0.001  −0.15  −0.22, −0.08  <0.001   2  0.94  0.88, 1.00  0.012  1.07  1.03, 1.11  <0.001  1.02  0.98, 1.06  0.303  −0.30  −0.36, −0.24  <0.001  −0.25  −0.31, −0.19  <0.001   3  0.97  0.91, 1.04  0.303  1.14  1.10, 1.18  <0.001  1.03  0.99, 1.08  0.036  −0.54  −0.60, −0.49  <0.001  −0.35  −0.41, −0.29  <0.001   ≥4  1.23  1.15, 1.30  <0.001  1.26  1.22, 1.31  <0.001  1.08  1.03, 1.12  <0.001  −0.99  −1.04, −0.93  <0.001  −0.55  −0.61, −0.48  <0.001   Linear trend  1.07  1.05, 1.08  <0.001  1.06  1.06, 1.07  <0.001  1.02  1.01, 1.03  <0.001  −0.26  −0.27, −0.25  <0.001  −0.13  −0.15, −0.12  <0.001  Driving time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.05  1.02, 1.09  <0.001  0.99  0.97, 1.01  0.270  1.01  0.98, 1.03  0.569  −0.19  −0.22, −0.16  <0.001  −0.02  −0.06, 0.01  0.139   2  1.12  1.07, 1.18  <0.001  1.05  1.02, 1.08  <0.001  1.03  1.00, 1.06  0.024  −0.37  −0.41, −0.33  <0.001  −0.13  −0.18, −0.08  <0.001   ≥3  1.50  1.41, 1.60  <0.001  1.26  1.22, 1.31  <0.001  1.12  1.07, 1.17  <0.001  −0.88  −0.94, −0.82  <0.001  −0.28  −0.34, −0.21  <0.001   Linear trend  1.11  1.09, 1.12  <0.001  1.05  1.04, 1.06  <0.001  1.03  1.01, 1.04  <0.001  −0.24  −0.26, −0.23  <0.001  −0.08  −0.09, −0.06  <0.001  Computer-use time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.68  0.66, 0.71  <0.001  0.79  0.77, 0.80  <0.001  0.85  0.83, 0.87  <0.001  0.52  0.49, 0.55  <0.001  0.21  0.17, 0.25  <0.001   2  0.69  0.66, 0.72  <0.001  0.77  0.75, 0.79  <0.001  0.80  0.78, 0.83  <0.001  0.58  0.53, 0.62  <0.001  0.21  0.16, 0.26  <0.001   ≥3  0.86  0.82, 0.90  <0.001  0.81  0.79, 0.84  <0.001  0.82  0.79, 0.85  <0.001  0.40  0.35, 0.44  <0.001  0.16  0.11, 0.22  <0.001   Linear trend  0.91  0.89, 0.92  <0.001  0.91  0.90, 0.92  <0.001  0.92  0.91, 0.93  <0.001  0.18  0.17, 0.20  <0.001  0.07  0.06, 0.09  <0.001  Model 2h  Television viewing time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.04  0.96, 1.12  0.232  1.07  1.03, 1.12  <0.001  1.01  0.97, 1.06  0.403  −0.12  −0.18, −0.06  <0.001  −0.13  −0.20, −0.06  <0.001   2  0.96  0.90, 1.03  0.142  1.07  1.03, 1.11  <0.001  1.02  0.98, 1.07  0.128  −0.21  −0.27, −0.16  <0.001  −0.20  −0.26, −0.13  <0.001   3  0.96  0.89, 1.03  0.159  1.09  1.05, 1.14  <0.001  1.03  0.98, 1.07  0.136  −0.33  −0.39, −0.28  <0.001  −0.25  −0.32, −0.19  <0.001   ≥4  1.09  1.01, 1.17  0.003  1.14  1.10, 1.19  <0.001  1.03  0.99, 1.08  0.053  −0.58  −0.64, −0.53  <0.001  −0.38  −0.44, −0.31  <0.001   Linear trend  1.02  1.00, 1.03  0.001  1.03  1.02, 1.04  <0.001  1.01  1.00, 1.02  0.057  −0.15  −0.16, −0.14  <0.001  −0.09  −0.10, −0.07  <0.001  Driving time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  1.21  1.17, 1.27  <0.001  1.05  1.03, 1.07  <0.001  1.04  1.02, 1.07  <0.001  −0.28  −0.31, −0.25  <0.001  −0.06  −0.10, −0.03  <0.001   2  1.27  1.20, 1.34  <0.001  1.10  1.06, 1.13  <0.001  1.07  1.03, 1.10  <0.001  −0.43  −0.48, −0.39  <0.001  −0.18  −0.23, −0.13  <0.001   ≥3  1.54  1.43, 1.66  <0.001  1.23  1.19, 1.28  <0.001  1.11  1.06, 1.16  <0.001  −0.73  −0.79, −0.68  <0.001  −0.27  −0.34, −0.19  <0.001   Linear trend  1.15  1.13, 1.17  <0.001  1.06  1.05, 1.07  <0.001  1.04  1.02, 1.05  <0.001  −0.24  −0.25, −0.22  <0.001  −0.09  −0.11, −0.07  <0.001  Computer-use time, hours/day                                 <1  1.00  Referent    1.00  Referent    1.00  Referent    0  Referent    0  Referent     1  0.77  0.74, 0.81  <0.001  0.85  0.83, 0.87  <0.001  0.88  0.86, 0.90  <0.001  0.32  0.29, 0.35  <0.001  0.14  0.10, 0.17  <0.001   2  0.74  0.70, 0.78  <0.001  0.81  0.79, 0.83  <0.001  0.83  0.80, 0.86  <0.001  0.40  0.36, 0.44  <0.001  0.15  0.10, 0.20  <0.001   ≥3  0.86  0.81, 0.91  <0.001  0.84  0.81, 0.86  <0.001  0.84  0.81, 0.88  <0.001  0.26  0.22, 0.31  <0.001  0.13  0.07, 0.18  <0.001   Linear trend  0.92  0.90, 0.94  <0.001  0.92  0.91, 0.93  <0.001  0.93  0.92, 0.94  <0.001  0.12  0.11, 0.14  <0.001  0.06  0.04, 0.07  <0.001  Abbreviations: CI, confidence interval; OR, odds ratio. a Prospective memory result, categorical: good result (referent: correct recall on first attempt) or poor result (incorrect recall on first attempt (i.e., correct recall on second attempt, instruction not recalled, skipped or incorrect)). An odds ratio of less than 1 indicates lower odds of a poor result; an odds ratio of greater than 1 indicates higher odds of a poor result (model 1: n = 166,401; model 2: n = 148,327). b Pairs matching result (round 1), categorical: good result (referent: 0 incorrect matches) or poor result (≥1 incorrect matches). An odds ratio of less than 1 indicates lower odds of a poor result; an odds ratio of greater than 1 indicates higher odds of a poor result (model 1: n = 471,474; model 2: n = 422,731). c Pairs matching result (round 2), categorical: good result (referent: <2 incorrect matches) or poor result (≥2 incorrect matches). An odds ratio of less than 1 indicates lower odds of a poor result; an odds ratio of greater than 1 indicates higher odds of a poor result (model 1: n = 470,433; model 2: n = 421,851). d Fluid intelligence score, continuous: total number of correct answers. A β coefficient of greater than 0 indicates a higher score; a β coefficient of less than 0 indicates a lower score (model 1: n = 161,348; model 2: n = 145,124). e Numeric memory score, continuous: maximum digits remembered correctly. A β coefficient of greater than 0 indicates a higher score; a β coefficient of less than 0 indicates a lower score (model 1: n = 49,035; model 2: n = 44,097). fP < 0.01 indicates statistical significance. g Model 1 mutually adjusted for the other sedentary behaviors and for age and sex. h Model 2 further adjusted for body mass index, ethnicity, social deprivation index, employment status, educational level, smoking status, alcohol drinking status, fruit and vegetable consumption, sleep duration, frequency of ≥10 minutes of walking, frequency of ≥10 minutes of moderate physical activity, frequency of ≥10 minutes of vigorous physical activity, number of cancers, number of noncancer illnesses, and number of medications/treatments. Interaction analyses showed that most findings were modified by age and sex (P < 0.05). Stratification indicated that the associations were generally stronger in older adults (≥60 years of age) and in men (see Web Figure 3 (age) and Web Figure 4 (sex)). Prospective findings Table 3 presents the cognitive function data of the participants with cognitive data at both baseline and follow-up. Cognitive decline over time was apparent—participants performed better in each cognitive test at baseline than at follow-up. For example, the mean fluid intelligence scores (n = 46,704) at baseline and follow-up were 6.7 (SD, 2.1) and 5.5 (SD, 2.0), respectively, with 15,384 (32.9%) individuals reporting a good outcome at follow-up (baseline fluid intelligence score ≤ follow-up fluid intelligence score) and 31,320 (67.1%) individuals reporting a poor outcome at follow-up (baseline fluid intelligence score > follow-up fluid intelligence score). The other tests followed a similar pattern. Table 3. Cognitive Function Data of Participants With Cognitive Data at Both Baseline and Follow-up, UK Biobank, United Kingdom, 2006–2010 Cognitive Functiona  Total No. of Participants  Baseline  Follow-up  No.  %  Mean (SD)  Range  No.  %  Mean (SD)  Range  Visual-spatial memory test (round 1)b,c  114,373                   Good result    89,137  77.9      70,761  61.9       Poor result    25,236  22.1      43,612  38.1       Good outcome at follow-up            70,761  61.9       Poor outcome at follow-up            43,612  38.1      Visual-spatial memory test (round 2)b,d  113,479                   Good result    23,262  20.5      14,886  13.1       Poor result    90,217  79.5      98,593  86.9       Good outcome at follow-up            14,886  13.1       Poor outcome at follow-up            98,593  86.9      Fluid intelligence teste,f  46,704                   Total number of correct answers        6.7 (2.1)  0.0–13.0      5.5 (2.0)  0.0–13.0   Good outcome at follow-up            15,384  32.9       Poor outcome at follow-up            31,320  67.1      Short-term numeric memory teste,g  12,091                   Maximum digits remembered correctly        7.0 (1.2)  2.0–12.0      6.9 (1.5)  2.0–11.0   Good outcome at follow-up            7,791  64.4       Poor outcome at follow-up            4,300  35.6      Cognitive Functiona  Total No. of Participants  Baseline  Follow-up  No.  %  Mean (SD)  Range  No.  %  Mean (SD)  Range  Visual-spatial memory test (round 1)b,c  114,373                   Good result    89,137  77.9      70,761  61.9       Poor result    25,236  22.1      43,612  38.1       Good outcome at follow-up            70,761  61.9       Poor outcome at follow-up            43,612  38.1      Visual-spatial memory test (round 2)b,d  113,479                   Good result    23,262  20.5      14,886  13.1       Poor result    90,217  79.5      98,593  86.9       Good outcome at follow-up            14,886  13.1       Poor outcome at follow-up            98,593  86.9      Fluid intelligence teste,f  46,704                   Total number of correct answers        6.7 (2.1)  0.0–13.0      5.5 (2.0)  0.0–13.0   Good outcome at follow-up            15,384  32.9       Poor outcome at follow-up            31,320  67.1      Short-term numeric memory teste,g  12,091                   Maximum digits remembered correctly        7.0 (1.2)  2.0–12.0      6.9 (1.5)  2.0–11.0   Good outcome at follow-up            7,791  64.4       Poor outcome at follow-up            4,300  35.6      Abbreviation: SD, standard deviation. a Samples sizes were different for different tests, ranging from 12,091 to 114,373. The mean follow-up period was 5.3 years. b Categorical variable. c Pairs matching result (round 1): good result (0 incorrect matches) or poor result (≥1 incorrect matches). Good outcome at follow-up (0 incorrect matches at follow-up) or poor outcome at follow-up (≥1 incorrect matches at follow-up). d Pairs matching result (round 2): good result (<2 incorrect matches) or poor result (≥2 incorrect matches). Good outcome at follow-up (<2 incorrect matches at follow-up) or poor outcome at follow-up (≥2 incorrect matches at follow-up). e Continuous variable. f Fluid intelligence score: total number of correct answers. Good outcome at follow-up (baseline fluid intelligence score ≤ follow-up fluid intelligence score) or poor outcome at follow-up (baseline fluid intelligence score > follow-up fluid intelligence score). g Numeric memory score: Maximum digits remembered correctly. Good outcome at follow-up (baseline numeric memory score ≤ follow-up numeric memory score) or poor outcome at follow-up (baseline numeric memory score > follow-up numeric memory score). Those with follow-up data had similar characteristics to those of the full UK Biobank cohort, although they were better educated and more likely to be employed (see Web Table 1). Table 4 presents the associations between the sedentary behaviors at baseline and cognitive function at follow-up. In the full-adjustment models (model 2), both television viewing and driving time at baseline were positively associated with the odds of having cognitive decline at follow-up across most outcomes. For example, for each additional hour spent watching television up to ≥4 hours/day at baseline, the odds of a lower fluid intelligence score at follow-up were 9% (99% CI: 6, 11) higher. Similarly, for each additional hour spent driving up to ≥3 hours/day at baseline, the odds of a lower fluid intelligence score at follow-up were 11% (99% CI: 7, 15) higher. In contrast, computer-use time at baseline was inversely associated with the odds of having cognitive decline at follow-up across most outcomes. Interaction analyses showed that only the associations between television viewing time and visual-spatial memory (round 2) were modified by age (P < 0.05) (see Web Figure 5). Findings were not modified by sex. Table 4. Prospective Associations Between Sedentary Behaviors at Baseline and Cognitive Function at Follow-upa Among UK Biobank Participants, United Kingdom, 2006–2010 Sedentary Behavior  Visual-Spatial Memory Test  Fluid Intelligence Testd  Short-Term Numeric Memory Teste  Round 1b  Round 2c  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  Model 1g  Television viewing time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.04  0.97, 1.11  0.154  1.02  0.93, 1.11  0.623  1.15  1.02, 1.28  0.002  1.05  0.85, 1.31  0.557   2  1.09  1.03, 1.15  <0.001  1.00  0.92, 1.08  0.961  1.24  1.12, 1.37  <0.001  1.13  0.93, 1.37  0.112   3  1.13  1.07, 1.20  <0.001  1.03  0.94, 1.12  0.439  1.37  1.24, 1.52  <0.001  1.26  1.03, 1.55  0.003   ≥4  1.17  1.10, 1.25  <0.001  1.01  0.93, 1.10  0.672  1.66  1.50, 1.84  <0.001  1.43  1.17, 1.76  <0.001   Linear trend  1.04  1.03, 1.06  <0.001  1.00  0.99, 1.02  0.612  1.13  1.10, 1.15  <0.001  1.10  1.05, 1.15  <0.001  Driving time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.06  1.02, 1.10  <0.001  1.01  0.96, 1.07  0.480  1.15  1.08, 1.22  <0.001  1.05  0.93, 1.18  0.319   2  1.07  1.01, 1.12  0.002  1.00  0.93, 1.08  0.903  1.10  1.00, 1.21  0.008  1.09  0.92, 1.30  0.193   ≥3  1.18  1.09, 1.28  <0.001  1.01  0.90, 1.13  0.831  1.44  1.25, 1.66  <0.001  1.11  0.85, 1.44  0.318   Linear trend  1.05  1.03, 1.07  <0.001  1.00  0.98, 1.03  0.709  1.10  1.06, 1.14  <0.001  1.04  0.98, 1.11  0.108  Computer-use time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  0.96  0.93, 1.00  0.013  0.96  0.91, 1.02  0.068  0.93  0.87, 1.00  0.006  0.90  0.79, 1.02  0.034   2  0.90  0.86, 0.94  <0.001  0.87  0.81, 0.93  <0.001  0.94  0.86, 1.02  0.041  0.77  0.65, 0.90  <0.001   ≥3  0.91  0.86, 0.96  <0.001  0.89  0.83, 0.96  <0.001  0.96  0.88, 1.05  0.293  0.86  0.72, 1.03  0.035   Linear trend  0.96  0.95, 0.98  <0.001  0.95  0.93, 0.97  <0.001  0.98  0.96, 1.01  0.150  0.93  0.88, 0.98  0.001  Model 2h  Television viewing time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.02  0.96, 1.09  0.348  1.03  0.94, 1.12  0.470  1.16  1.03, 1.30  0.001  1.02  0.82, 1.28  0.817   2  1.07  1.00, 1.13  0.006  1.01  0.93, 1.09  0.815  1.21  1.09, 1.35  <0.001  1.08  0.88, 1.33  0.310   3  1.08  1.02, 1.15  0.001  1.03  0.94, 1.12  0.416  1.29  1.15, 1.44  <0.001  1.16  0.94, 1.44  0.066   ≥4  1.09  1.02, 1.17  0.001  1.00  0.91, 1.10  0.993  1.45  1.29, 1.62  <0.001  1.29  1.04, 1.61  0.003   Linear trend  1.02  1.01, 1.04  <0.001  1.00  0.98, 1.02  0.955  1.09  1.06, 1.11  <0.001  1.07  1.02, 1.12  <0.001  Driving time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.07  1.03, 1.12  <0.001  1.02  0.97, 1.08  0.294  1.19  1.11, 1.27  <0.001  1.05  0.92, 1.19  0.363   2  1.08  1.02, 1.14  0.001  1.01  0.94, 1.10  0.624  1.15  1.04, 1.27  <0.001  1.05  0.88, 1.27  0.466   ≥3  1.16  1.06, 1.26  <0.001  1.01  0.90, 1.13  0.895  1.43  1.24, 1.66  <0.001  1.05  0.80, 1.39  0.650   Linear trend  1.05  1.03, 1.07  <0.001  1.01  0.98, 1.04  0.552  1.11  1.07, 1.15  <0.001  1.02  0.96, 1.10  0.363  Computer-use time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  0.97  0.93, 1.01  0.053  0.97  0.92, 1.03  0.250  0.94  0.87, 1.01  0.020  0.92  0.80, 1.05  0.090   2  0.91  0.86, 0.95  <0.001  0.88  0.82, 0.94  <0.001  0.94  0.86, 1.03  0.073  0.76  0.64, 0.90  <0.001   ≥3  0.90  0.85, 0.96  <0.001  0.90  0.83, 0.98  0.001  0.97  0.88, 1.06  0.359  0.84  0.69, 1.01  0.016   Linear trend  0.96  0.95, 0.98  <0.001  0.96  0.93, 0.98  <0.001  0.99  0.96, 1.02  0.207  0.92  0.87, 0.98  <0.001  Sedentary Behavior  Visual-Spatial Memory Test  Fluid Intelligence Testd  Short-Term Numeric Memory Teste  Round 1b  Round 2c  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  OR  99% CI  P Valuef  Model 1g  Television viewing time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.04  0.97, 1.11  0.154  1.02  0.93, 1.11  0.623  1.15  1.02, 1.28  0.002  1.05  0.85, 1.31  0.557   2  1.09  1.03, 1.15  <0.001  1.00  0.92, 1.08  0.961  1.24  1.12, 1.37  <0.001  1.13  0.93, 1.37  0.112   3  1.13  1.07, 1.20  <0.001  1.03  0.94, 1.12  0.439  1.37  1.24, 1.52  <0.001  1.26  1.03, 1.55  0.003   ≥4  1.17  1.10, 1.25  <0.001  1.01  0.93, 1.10  0.672  1.66  1.50, 1.84  <0.001  1.43  1.17, 1.76  <0.001   Linear trend  1.04  1.03, 1.06  <0.001  1.00  0.99, 1.02  0.612  1.13  1.10, 1.15  <0.001  1.10  1.05, 1.15  <0.001  Driving time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.06  1.02, 1.10  <0.001  1.01  0.96, 1.07  0.480  1.15  1.08, 1.22  <0.001  1.05  0.93, 1.18  0.319   2  1.07  1.01, 1.12  0.002  1.00  0.93, 1.08  0.903  1.10  1.00, 1.21  0.008  1.09  0.92, 1.30  0.193   ≥3  1.18  1.09, 1.28  <0.001  1.01  0.90, 1.13  0.831  1.44  1.25, 1.66  <0.001  1.11  0.85, 1.44  0.318   Linear trend  1.05  1.03, 1.07  <0.001  1.00  0.98, 1.03  0.709  1.10  1.06, 1.14  <0.001  1.04  0.98, 1.11  0.108  Computer-use time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  0.96  0.93, 1.00  0.013  0.96  0.91, 1.02  0.068  0.93  0.87, 1.00  0.006  0.90  0.79, 1.02  0.034   2  0.90  0.86, 0.94  <0.001  0.87  0.81, 0.93  <0.001  0.94  0.86, 1.02  0.041  0.77  0.65, 0.90  <0.001   ≥3  0.91  0.86, 0.96  <0.001  0.89  0.83, 0.96  <0.001  0.96  0.88, 1.05  0.293  0.86  0.72, 1.03  0.035   Linear trend  0.96  0.95, 0.98  <0.001  0.95  0.93, 0.97  <0.001  0.98  0.96, 1.01  0.150  0.93  0.88, 0.98  0.001  Model 2h  Television viewing time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.02  0.96, 1.09  0.348  1.03  0.94, 1.12  0.470  1.16  1.03, 1.30  0.001  1.02  0.82, 1.28  0.817   2  1.07  1.00, 1.13  0.006  1.01  0.93, 1.09  0.815  1.21  1.09, 1.35  <0.001  1.08  0.88, 1.33  0.310   3  1.08  1.02, 1.15  0.001  1.03  0.94, 1.12  0.416  1.29  1.15, 1.44  <0.001  1.16  0.94, 1.44  0.066   ≥4  1.09  1.02, 1.17  0.001  1.00  0.91, 1.10  0.993  1.45  1.29, 1.62  <0.001  1.29  1.04, 1.61  0.003   Linear trend  1.02  1.01, 1.04  <0.001  1.00  0.98, 1.02  0.955  1.09  1.06, 1.11  <0.001  1.07  1.02, 1.12  <0.001  Driving time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  1.07  1.03, 1.12  <0.001  1.02  0.97, 1.08  0.294  1.19  1.11, 1.27  <0.001  1.05  0.92, 1.19  0.363   2  1.08  1.02, 1.14  0.001  1.01  0.94, 1.10  0.624  1.15  1.04, 1.27  <0.001  1.05  0.88, 1.27  0.466   ≥3  1.16  1.06, 1.26  <0.001  1.01  0.90, 1.13  0.895  1.43  1.24, 1.66  <0.001  1.05  0.80, 1.39  0.650   Linear trend  1.05  1.03, 1.07  <0.001  1.01  0.98, 1.04  0.552  1.11  1.07, 1.15  <0.001  1.02  0.96, 1.10  0.363  Computer-use time, hours/day                           <1  1.00  Referent    1.00  Referent    1.00  Referent    1.00  Referent     1  0.97  0.93, 1.01  0.053  0.97  0.92, 1.03  0.250  0.94  0.87, 1.01  0.020  0.92  0.80, 1.05  0.090   2  0.91  0.86, 0.95  <0.001  0.88  0.82, 0.94  <0.001  0.94  0.86, 1.03  0.073  0.76  0.64, 0.90  <0.001   ≥3  0.90  0.85, 0.96  <0.001  0.90  0.83, 0.98  0.001  0.97  0.88, 1.06  0.359  0.84  0.69, 1.01  0.016   Linear trend  0.96  0.95, 0.98  <0.001  0.96  0.93, 0.98  <0.001  0.99  0.96, 1.02  0.207  0.92  0.87, 0.98  <0.001  Abbreviations: CI, confidence interval; OR, odds ratio. a The mean follow-up period was 5.3 years. b Pairs matching result (round 1), categorical: good outcome at follow-up (0 incorrect matches at follow-up) or poor outcome at follow-up (≥1 incorrect matches at follow-up). An odds ratio of less than 1 indicates lower odds of having cognitive decline at follow-up (i.e., a good outcome at follow-up); an odds ratio of greater than 1 indicates higher odds of having cognitive decline at follow-up (i.e., a poor outcome at follow-up) (model 1: n = 113,129; model 2: n = 106,665). c Pairs matching result (round 2), categorical: good outcome at follow-up (<2 incorrect matches at follow-up) or poor outcome at follow-up (≥2 incorrect matches at follow-up). An odds ratio of less than 1 indicates lower odds of having cognitive decline at follow-up (i.e., a good outcome at follow-up); an odds ratio of greater than 1 indicates higher odds of having cognitive decline at follow-up (i.e., a poor outcome at follow-up) (model 1: n = 112,252; model 2: n = 105,861). d Fluid intelligence score, categorical: good outcome at follow-up (baseline fluid intelligence score ≤ follow-up fluid intelligence score) or poor outcome at follow-up (baseline fluid intelligence score > follow-up fluid intelligence score). An odds ratio of less than 1 indicates lower odds of having cognitive decline at follow-up (i.e., a good outcome at follow-up); an odds ratio of greater than 1 indicates higher odds of having cognitive decline at follow-up (i.e., a poor outcome at follow-up) (model 1: n = 46,158; model 2: n = 43,350). e Numeric memory score, categorical: good outcome at follow-up (baseline numeric memory score ≤ follow-up numeric memory score); or poor outcome at follow-up (baseline numeric memory score > follow-up numeric memory score). An odds ratio of less than 1 indicates lower odds of having cognitive decline at follow-up (i.e., a good outcome at follow-up); an odds ratio of greater than 1 indicates higher odds of having cognitive decline at follow-up (i.e., a poor outcome at follow-up) (model 1: n = 11,957; model 2: n = 11,299). fP < 0.01 indicates statistical significance. g Model 1 mutually adjusted for the other sedentary behaviors and for age, sex, and the baseline result/score of the cognitive test under consideration. h Model 2 further adjusted for body mass index, ethnicity, social deprivation index, employment status, education level, smoking status, alcohol drinking status, fruit and vegetable consumption, sleep duration, frequency of ≥10 minutes of walking, frequency of ≥10 minutes of moderate physical activity, frequency of ≥10 minutes of vigorous physical activity, number of cancers, number of noncancer illnesses, and number of medications/treatments. Sensitivity analyses The cross-sectional and prospective findings were generalizable across the sample of participants without cancer, cardiovascular disease, or cognitive/psychiatric illnesses (see Web Figure 6 (cross-sectional associations) and Web Figure 7 (prospective associations)). DISCUSSION Key findings To our knowledge, this is the first study to quantify the cross-sectional and prospective associations between domains of sedentary behavior and cognitive function in a large cohort of adults in the United Kingdom. At baseline, both television viewing and driving time were inversely associated with cognitive function. In contrast, computer-use time was positively associated with cognitive function. Most findings were modified by age and sex, with stronger relationships generally observed in older adults and in men. These novel results suggest that the influence of sedentary behavior on cognition is enhanced in older age and in men. Both television viewing and driving time at baseline were positively associated with the odds of having cognitive decline at follow-up across most outcomes. In contrast, computer-use time at baseline was inversely associated with the odds of having cognitive decline at follow-up across most outcomes. The cross-sectional and prospective findings were robust and generalizable across the sample of participants without cancer, cardiovascular disease, or cognitive/psychiatric illnesses. Interpretations To our knowledge, a modest number of studies have attempted to examine the prospective associations between the different types of sedentary behaviors and cognitive function (14–17, 22–26). However, these studies have been limited by a small sample size (n values ranging between 469 and 8,462), populations that involved only children or older adults, analyses that considered only one domain or test of cognitive function, or cognitive data that were collected at a single time point. We believe our novel study in a large sample of middle-aged adults representing the general population provides the most comprehensive observational analysis to date. Our findings were consistent with the existing data in this research area. Observational studies have previously demonstrated an inverse association between television viewing and cognition (14–17) and a positive association between computer/internet use and cognition (15–18). However, before this study, the interactions with age or the deleterious influence of driving on cognitive health were less clear. The inverse associations of television viewing and driving time with cognitive function could be due to several factors. Cognition has previously been linked to cardiometabolic health (27, 28), and numerous studies have demonstrated inverse associations of television viewing and driving time with cardiometabolic health (9–12, 29–31). Therefore, it is possible that the observed associations act via pathways linked to the risk of vascular dysfunction and chronic diseases. Because vascular dysfunction and chronic diseases are linked to aging, this mechanism would also help explain the observed interactions with age. Other mediating factors could also explain the results for driving; it is known that driving is related to stress and fatigue (32), and with several studies previously showing the links between these factors and cognitive decline (33–35), it is plausible that the observed relationships are enhanced via this pathway. Furthermore, some types of sedentary behaviors, such as television viewing and driving, could possibly segregate individuals from social networks and restrict external collaborations, factors that are known to affect cognition (36–38); this again could be particularly important in older adults. In contrast, the positive relationship shared between computer use and cognitive function coincides with previous work where improved cognition or a lower risk of dementia was reported in those engaging in cognitively vitalizing sedentary behaviors or leisure activities (15–18). Therefore, as computer use is likely to involve some level of cognitive challenge, stimulate social interactions, and reduce solitariness, it may compensate for the associated sedentary behavior in relation to cognitive health. Some of the mechanisms mentioned above are also linked to and vary according to sex (39, 40), and they could therefore help explain the observed interactions with sex. The differences observed in cognitive function across the categories of sedentary behavior in our analyses are likely to be clinically important beyond the risk of cognitive decline. For example, higher fluid intelligence scores have previously been shown to be strongly associated with a lower risk of all-cause mortality (41, 42). In a sample of 5,572 middle-aged British adults, Sabia et al. (41) observed that a higher fluid intelligence score (1 SD) was associated with a 14% lower risk of all-cause mortality. Similarly, in a sample of 896 older Australian adults, Batterham et al. (42) observed that a higher fluid intelligence score (1 SD) was associated with a 24% lower risk of all-cause mortality. In our analysis at baseline (model 2), the standard deviation of fluid intelligence score was 2.1. Regression analyses investigating the associations of sedentary behaviors with fluid intelligence demonstrated that television viewing and driving time were linearly associated with lower fluid intelligence scores of 0.15 and 0.24 units, respectively. In contrast, computer-use time was linearly associated with a higher fluid intelligence score of 0.12 units. Hence, using the data above, it can be estimated that lower fluid intelligence scores by 0.15 and 0.24 units would equate approximately to a 1.1%–3.2% higher risk of all-cause mortality. In contrast, a higher fluid intelligence score by 0.12 units would equate approximately to a 0.9%–1.6% lower risk of all-cause mortality. For more details on these calculations, see Web Appendix 5. Strengths and limitations This study has several strengths and some limitations. Strengths include access to data on a large sample of adults representing the national population, follow-up cognitive function data enabling prospective investigation of associations, evaluation of dose-response and linear relationships between mutually adjusted and time-quantified sedentary behaviors with a wide range of cognitive outcomes, detailed covariate data enabling control for several important and relevant factors, analysis of interactions with age and sex, and robust sensitivity analyses investigating associations in the healthy population. However, although the UK Biobank is representative of the general population with respect to age, sex, ethnicity, and deprivation within the age range recruited, it may not be representative in other regards (43). While this limits the ability to generalize prevalence rates, estimates of the magnitude of associations in our study are unlikely to have been substantially affected by this due to the large and multifaceted base population (43, 44). Furthermore, the cognitive data from the UK Biobank cohort has recently been shown to be an important and valid resource for investigating predictors and modifiers of cognitive abilities and associated health outcomes in the general population (45). The sedentary behavior data used in this study have strengths and limitations. Only three sedentary domains were included; thus, the findings are restricted and cannot be generalized to other types of sedentary behavior. Self-reported assessments of sedentary behavior are subjective and are influenced by recall and response issues (46, 47); hence, they tend to have low validity and increase the risk of regression dilution. However, although data that are more robust can be obtained using objective measurement tools (e.g., accelerometers) (46, 47), they would not provide information on the specific type of sedentary behavior performed. Furthermore, because the reasons for using the computer outside work were unknown (e.g., computers could be used for such activities as reading, watching videos, internet browsing, or playing games), it is not possible to accurately classify or infer the type of computer use undertaken, and it may have involved crossover into cognitively inert tasks. Additionally, only those who provided an e-mail address at baseline (approximately 300,000) were contacted to participate in the online follow-up of cognitive function. These participants all had computer access and presumably some computer-use experience. This may also have resulted in the small differences in characteristics (including level of education and employment status) in the follow-up sample (see Web Table 1). Consequently, the prospective analysis may be biased and lack generalizability. Moreover, at baseline, the cognitive function tests were implemented using questionnaires that were administered via a touchscreen interface. At follow-up, the measurements were obtained remotely via online questionnaires that were administered on a computer via a mouse interface. This difference in the mode of administration could possibly account for some of the variability in cognitive performance and change over time. Nevertheless, the prospective analysis broadly supports and is consistent with the cross-sectional associations reported for the full cohort at baseline. Although we adjusted for a wide range of covariates, some unmeasured factors (e.g., type of employment/occupation) may have further confounded the reported associations. Our results may be subject to residual confounding or reverse causality. For example, it is possible that the positive association observed between computer use and cognitive function simply reflected greater familiarity with interacting with a computer rather than better cognitive function as such. Correspondingly, individuals with better cognitive function are more likely to engage in healthy behaviors and abstain from unhealthy ones, a concept known as neuroselection (48, 49). Also, although we investigated interactions with age and sex in our study, it must be highlighted that similar differences observed in cognitive function across different groups (i.e., in younger adults vs. older adults and women vs. men) may have different clinical meanings and should be interpreted with caution. For example, a unit difference in the cognitive function test score of a younger adult may not have the same result or significance for cognitive health as a unit difference in the score of an older adult. Last, due to large variations between the numbers of individuals who completed each cognitive assessment at both baseline and follow-up, analyses were based on different sample sizes. Conclusions Our analysis, conducted in a large national sample of adults, demonstrated that some sedentary domains, but not all, are associated with poor cognition. Watching television and driving were inversely associated with cognitive function, whereas computer use was positively associated with cognitive function. Of note, the associations were consistently stronger in older adults. Intervention studies are required to confirm these findings. Nevertheless, these results provide robust observational data supporting public health policies aimed at reducing television viewing and driving time in adults. ACKNOWLEDGMENTS Author affiliations: Department of Health Sciences, University of Leicester, Leicester General Hospital, Leicester, United Kingdom (Kishan Bakrania); Diabetes Research Centre, University of Leicester, Leicester General Hospital, Leicester, United Kingdom (Kishan Bakrania, Charlotte L. Edwardson, Kamlesh Khunti, Melanie J. Davies, Thomas Yates); Leicester Diabetes Centre, University Hospitals of Leicester, Leicester General Hospital, Leicester, United Kingdom (Kishan Bakrania, Charlotte L. Edwardson, Kamlesh Khunti, Melanie J. Davies, Thomas Yates); National Institute for Health Research Leicester Biomedical Research Centre, Leicester General Hospital, Leicester, United Kingdom (Kishan Bakrania, Charlotte L. Edwardson, Melanie J. Davies, Thomas Yates); National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care—East Midlands, Diabetes Research Centre, Leicester General Hospital, Leicester, United Kingdom (Kishan Bakrania, Kamlesh Khunti); and School of Sport, Exercise, and Health Sciences, Loughborough University, Loughborough, United Kingdom (Stephan Bandelow). This research was conducted using the UK Biobank resource (application 10813). The UK Biobank was established by the Wellcome Trust, Medical Research Council, Department of Health, Scottish Government, and the Northwest Regional Development Agency. It has also had funding from the Welsh Assembly Government and the British Heart Foundation, and it is supported by the National Health Service. This research was supported by the National Institute for Health Research Leicester Biomedical Research Center, the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care—East Midlands, and the Leicester Clinical Trials Unit. All the data reported in this study are fully available via application to the United Kingdom Biobank. The views expressed are those of the authors and not necessarily those of the National Health Service, the National Institute for Health Research, or the Department of Health. Conflict of interest: none declared. Abbreviations CI confidence interval SD standard deviation REFERENCES 1 Lautenschlager NT, Cox KL, Flicker L, et al.  . Effect of physical activity on cognitive function in older adults at risk for Alzheimer disease: a randomized trial. JAMA . 2008; 300( 9): 1027– 1037. Google Scholar CrossRef Search ADS PubMed  2 Laurin D, Verreault R, Lindsay J, et al.  . Physical activity and risk of cognitive impairment and dementia in elderly persons. Arch Neurol . 2001; 58( 3): 498– 504. Google Scholar CrossRef Search ADS PubMed  3 Erickson KI, Voss MW, Prakash RS, et al.  . Exercise training increases size of hippocampus and improves memory. Proc Natl Acad Sci USA . 2011; 108( 7): 3017– 3022. Google Scholar CrossRef Search ADS PubMed  4 Colcombe SJ, Kramer AF, Erickson KI, et al.  . Cardiovascular fitness, cortical plasticity, and aging. Proc Natl Acad Sci USA . 2004; 101( 9): 3316– 3321. Google Scholar CrossRef Search ADS PubMed  5 Chapman SB, Aslan S, Spence JS, et al.  . Shorter term aerobic exercise improves brain, cognition, and cardiovascular fitness in aging. Front Aging Neurosci . 2013; 5: 75. Google Scholar CrossRef Search ADS PubMed  6 Colcombe SJ, Erickson KI, Scalf PE, et al.  . Aerobic exercise training increases brain volume in aging humans. J Gerontol A Biol Sci Med Sci . 2006; 61( 11): 1166– 1170. Google Scholar CrossRef Search ADS PubMed  7 Smith JC, Nielson KA, Antuono P, et al.  . Semantic memory functional MRI and cognitive function after exercise intervention in mild cognitive impairment. J Alzheimers Dis . 2013; 37( 1): 197– 215. Google Scholar PubMed  8 Sedentary Behaviour Research Network. Letter to the editor: standardized use of the terms “sedentary” and “sedentary behaviours”. Appl Physiol Nutr Metab . 2012; 37( 3): 540– 542. CrossRef Search ADS PubMed  9 Biswas A, Oh PI, Faulkner GE, et al.  . Sedentary time and its association with risk for disease incidence, mortality, and hospitalization in adults: a systematic review and meta-analysis. Ann Intern Med . 2015; 162( 2): 123– 132. Google Scholar CrossRef Search ADS PubMed  10 Edwardson CL, Gorely T, Davies MJ, et al.  . Association of sedentary behaviour with metabolic syndrome: a meta-analysis. PLoS One . 2012; 7( 4): e34916. Google Scholar CrossRef Search ADS PubMed  11 Grøntved A, Hu FB. Television viewing and risk of type 2 diabetes, cardiovascular disease, and all-cause mortality: a meta-analysis. JAMA . 2011; 305( 23): 2448– 2455. Google Scholar CrossRef Search ADS PubMed  12 Wilmot EG, Edwardson CL, Achana FA, et al.  . Sedentary time in adults and the association with diabetes, cardiovascular disease and death: systematic review and meta-analysis. Diabetologia . 2012; 55( 11): 2895– 2905. Google Scholar CrossRef Search ADS PubMed  13 Falck RS, Davis JC, Liu-Ambrose T. What is the association between sedentary behaviour and cognitive function? A systematic review. Br J Sports Med . 2017; 51( 10): 800– 811. Google Scholar CrossRef Search ADS PubMed  14 Hoang TD, Reis J, Zhu N, et al.  . Effect of early adult patterns of physical activity and television viewing on midlife cognitive function. JAMA Psychiatry . 2016; 73( 1): 73– 79. Google Scholar CrossRef Search ADS PubMed  15 Wang JY, Zhou DH, Li J, et al.  . Leisure activity and risk of cognitive impairment: the Chongqing aging study. Neurology . 2006; 66( 6): 911– 913. Google Scholar CrossRef Search ADS PubMed  16 Hamer M, Stamatakis E. Prospective study of sedentary behavior, risk of depression, and cognitive impairment. Med Sci Sports Exerc . 2014; 46( 4): 718– 723. Google Scholar CrossRef Search ADS PubMed  17 Kesse-Guyot E, Charreire H, Andreeva VA, et al.  . Cross-sectional and longitudinal associations of different sedentary behaviors with cognitive performance in older adults. PLoS One . 2012; 7( 10): e47831. Google Scholar CrossRef Search ADS PubMed  18 Tun PA, Lachman ME. The association between computer use and cognition across adulthood: use it so you won’t lose it? Psychol Aging . 2010; 25( 3): 560– 568. Google Scholar CrossRef Search ADS PubMed  19 Sudlow C, Gallacher J, Allen N, et al.  . UK Biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med . 2015; 12( 3): e1001779. Google Scholar CrossRef Search ADS PubMed  20 Allen N, Sudlow C, Downey P, et al.  . UK Biobank: current status and what it means for epidemiology. Health Policy Technol . 2012; 1( 3): 123– 126. Google Scholar CrossRef Search ADS   21 UK Biobank. UK Biobank: Protocol for a large-scale prospective epidemiological resource. 2016. http://www.ukbiobank.ac.uk/wp-content/uploads/2011/11/UK-Biobank-Protocol.pdf. Accessed March 10, 2017. 22 Iwasa H, Yoshida Y, Kai I, et al.  . Leisure activities and cognitive function in elderly community-dwelling individuals in Japan: a 5-year prospective cohort study. J Psychosom Res . 2012; 72( 2): 159– 164. Google Scholar CrossRef Search ADS PubMed  23 Wilson RS, Mendes De Leon CF, Barnes LL, et al.  . Participation in cognitively stimulating activities and risk of incident Alzheimer disease. JAMA . 2002; 287( 6): 742– 748. Google Scholar CrossRef Search ADS PubMed  24 Verghese J, Lipton RB, Katz MJ, et al.  . Leisure activities and the risk of dementia in the elderly. N Engl J Med . 2003; 348( 25): 2508– 2516. Google Scholar CrossRef Search ADS PubMed  25 Akbaraly TN, Portet F, Fustinoni S, et al.  . Leisure activities and the risk of dementia in the elderly: results from the Three-City Study. Neurology . 2009; 73( 11): 854– 861. Google Scholar CrossRef Search ADS PubMed  26 Aggio D, Smith L, Fisher A, et al.  . Context-specific associations of physical activity and sedentary behavior with cognition in children. Am J Epidemiol . 2016; 183( 12): 1075– 1082. Google Scholar CrossRef Search ADS PubMed  27 Elias MF, Sullivan LM, D’Agostino RB, et al.  . Framingham stroke risk profile and lowered cognitive performance. Stroke . 2004; 35( 2): 404– 409. Google Scholar CrossRef Search ADS PubMed  28 Kaffashian S, Dugravot A, Nabi H, et al.  . Predictive utility of the Framingham general cardiovascular disease risk profile for cognitive function: evidence from the Whitehall II study. Eur Heart J . 2011; 32( 18): 2326– 2332. Google Scholar CrossRef Search ADS PubMed  29 Belkic K, Savic C, Theorell T, et al.  . Mechanisms of cardiac risk among professional drivers. Scand J Work Environ Health . 1994; 20( 2): 73– 86. Google Scholar CrossRef Search ADS PubMed  30 Morris JN, Heady JA, Raffle PA, et al.  . Coronary heart-disease and physical activity of work. Lancet . 1953; 265( 6796): 1111– 1120. Google Scholar CrossRef Search ADS PubMed  31 Robinson CF, Burnett CA. Truck drivers and heart disease in the United States, 1979–1990. Am J Ind Med . 2005; 47( 2): 113– 119. Google Scholar CrossRef Search ADS PubMed  32 Howard A, Joint M. Fatigue and stress in driving. Empl Couns Today . 1994; 6( 6): 3– 7. Google Scholar CrossRef Search ADS   33 van der Linden D, Frese M, Meijman TF. Mental fatigue and the control of cognitive processes: effects on perseveration and planning. Acta Psychol (Amst) . 2003; 113( 1): 45– 65. Google Scholar CrossRef Search ADS PubMed  34 Abd-Elfattah HM, Abdelazeim FH, Elshennawy S. Physical and cognitive consequences of fatigue: a review. J Adv Res . 2015; 6( 3): 351– 358. Google Scholar CrossRef Search ADS PubMed  35 Scott SB, Graham-Engeland JE, Engeland CG, et al.  . The Effects of Stress on Cognitive Aging, Physiology and Emotion (ESCAPE) project. BMC Psychiatry . 2015; 15: 146. Google Scholar CrossRef Search ADS PubMed  36 Shankar A, Hamer M, McMunn A, et al.  . Social isolation and loneliness: relationships with cognitive function during 4 years of follow-up in the English Longitudinal Study of Ageing. Psychosom Med . 2013; 75( 2): 161– 170. Google Scholar CrossRef Search ADS PubMed  37 Bassuk SS, Glass TA, Berkman LF. Social disengagement and incident cognitive decline in community-dwelling elderly persons. Ann Intern Med . 1999; 131( 3): 165– 173. Google Scholar CrossRef Search ADS PubMed  38 Golden J, Conroy RM, Bruce I, et al.  . Loneliness, social support networks, mood and wellbeing in community-dwelling elderly. Int J Geriatr Psychiatry . 2009; 24( 7): 694– 700. Google Scholar CrossRef Search ADS PubMed  39 Mosca L, Barrett-Connor E, Wenger NK. Sex/gender differences in cardiovascular disease prevention: what a difference a decade makes. Circulation . 2011; 124( 19): 2145– 2154. Google Scholar CrossRef Search ADS PubMed  40 McCarrey AC, An Y, Kitner-Triolo MH, et al.  . Sex differences in cognitive trajectories in clinically normal older adults. Psychol Aging . 2016; 31( 2): 166– 175. Google Scholar CrossRef Search ADS PubMed  41 Sabia S, Guéguen A, Marmot MG, et al.  . Does cognition predict mortality in midlife? Results from the Whitehall II cohort study. Neurobiol Aging . 2010; 31( 4): 688– 695. Google Scholar CrossRef Search ADS PubMed  42 Batterham PJ, Christensen H, Mackinnon AJ. Fluid intelligence is independently associated with all-cause mortality over 17 years in an elderly community sample: an investigation of potential mechanisms. Intelligence . 2009; 37( 6): 551– 560. Google Scholar CrossRef Search ADS   43 Collins R. What makes UK Biobank special? Lancet . 2012; 379( 9822): 1173– 1174. Google Scholar CrossRef Search ADS PubMed  44 Manolio TA, Collins R. Enhancing the feasibility of large cohort studies. JAMA . 2010; 304( 20): 2290– 2291. Google Scholar CrossRef Search ADS PubMed  45 Lyall DM, Cullen B, Allerhand M, et al.  . Cognitive test scores in UK Biobank: data reduction in 480,416 participants and longitudinal stability in 20,346 participants. PLoS One . 2016; 11( 4): e0154222. Google Scholar CrossRef Search ADS PubMed  46 Troiano RP, Berrigan D, Dodd KW, et al.  . Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc . 2008; 40( 1): 181– 188. Google Scholar CrossRef Search ADS PubMed  47 Prince SA, Adamo KB, Hamel ME, et al.  . A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act . 2008; 5: 56. Google Scholar CrossRef Search ADS PubMed  48 Belsky DW. Invited commentary: lessons for research on cognitive aging from a study of children. Am J Epidemiol . 2016; 183( 12): 1083– 1085. Google Scholar CrossRef Search ADS PubMed  49 Aggio D, Smith L, Fisher A, et al.  . Aggio et al. Respond to “Lessons for Research on Cognitive Aging”. Am J Epidemiol . 2016; 183( 12): 1086– 1087. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

American Journal of EpidemiologyOxford 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