Association of 12-Year Trajectories of Sitting Time With Frailty in Middle-Aged Women

Association of 12-Year Trajectories of Sitting Time With Frailty in Middle-Aged Women Abstract Prolonged sitting time is associated with several health outcomes; limited evidence indicates associations with frailty. Our aims in this study were to identify patterns of sitting time over 12 years in middle-aged (ages 50–55 years) women and examine associations of these patterns with frailty in older age. We examined 5,462 women born in 1946–1951 from the Australian Longitudinal Study on Women’s Health who provided information on sociodemographic attributes, daily sitting time, and frailty in 2001 and then again every 3 years until 2013. Frailty was assessed using the FRAIL (fatigue, resistance, ambulation, illness, loss of weight) scale (0 = healthy; 1–2 = prefrail; 3–5 = frail), and group-based trajectory analyses identified trajectories of sitting time. We identified 5 sitting-time trajectories: low (26.9%), medium (43.1%; referent), increasing (6.9%), decreasing (18.1%), and high (4.8%). In adjusted models, the likelihoods of being frail were statistically higher for women in the increasing (odds ratio (OR) = 1.29, 95% confidence interval (CI): 1.03, 1.61) and high (OR = 1.42, 95% CI: 1.10, 1.84) trajectories. In contrast, women in the low trajectory group were less likely to be frail (OR = 0.86, 95% CI: 0.75, 0.98), and there was no difference in the likelihood of frailty in the decreasing trajectory group. Our study suggests that patterns of sitting time over 12 years in middle-aged women predict frailty in older age. aging, frailty, longitudinal cohort studies, middle age, sedentary behavior, sitting, women’s health As the populations of most developed countries continue to age, frailty is becoming more prevalent (1). A systematic review found that in community-dwelling elderly adults, frailty prevalence varies from 4.0% to 59.1% (2), with an increasing frailty pattern observed with increased age and with frailty being more prevalent in women than in men (1, 3). Frailty is defined as a syndrome that results from decline in physiological reserve to withstand stressors (1, 4) and that leads to multiple adverse health outcomes, including falls, hospitalization, institutionalization, and premature death (3, 4). Based on recent estimates, one-fifth of Australian women aged 73–78 years are frail, suggesting that it is important to examine the health and behaviors of middle-aged women in order to identify factors associated with the development of frailty (5). Sitting time is a risk factor for poor cardiometabolic health (6, 7), which is closely related to frailty (8). Prolonged sitting time has become a public health concern (9) as the modern environment encourages more sitting, whether at work, at home, or in transport (10). In 2011, Australian adults reported sitting for 5.5 hours per day, with most sitting occurring in leisure domains (61.3%) and the remainder occurring at work (25.6%) or while using transportation (13.1%) (11). In contrast, when it is measured objectively, sitting time is estimated to be 8.9 hours per day (12). While investigators have reported increases in sitting time in cohort studies and the general population (13–15), there is limited evidence on patterns of sitting over time. While prolonged sitting in adults is detrimentally associated with many health outcomes, such as all-cause, cardiovascular, and cancer mortality and incidence of cardiovascular disease, cancer, and type 2 diabetes (16, 17), there have been limited investigations of relationships between sitting time and geriatric conditions such as frailty. In several cross-sectional studies, high levels of sedentary behavior (sitting or lying down while awake with an energy expenditure less than or equal to 1.5 metabolic equivalents (METs)) (18) were associated with frailty in older adults (all aged >50 years) (19–21). In a longitudinal study of 2 cohorts of community-dwelling older adults aged ≥60 years, García-Esquinas et al. (22) reported that baseline television (TV) viewing time was also associated with frailty at follow-up. Using data from a large Australian cohort, we aimed to identify patterns of sitting time over 12 years in middle-aged women, a group at high risk of developing frailty, and to examine associations of these patterns with frailty in older age. Investigation of this relationship is important to help develop programs designed to ameliorate the health outcome of frailty. METHODS Study population and procedures The Australian Longitudinal Study on Women’s Health (http://www.alswh.org.au/), initiated in 1996, is a national longitudinal cohort study examining over 58,000 Australian women (originally) separated into 3 cohorts: those born in 1973–1978, those born in 1946–1951, and those born in 1921–1926 (23, 24). The participants were randomly selected from the Australian national Medicare health insurance database, which includes all Australian citizens and permanent residents (24, 25). It is funded by the Australian Government Department of Health (24, 25). The study was approved by ethics committees at the Universities of Queensland and Newcastle, and informed consent was collected from all participants. When they were first surveyed in 1996 (survey 1), 13,715 women born in 1946–1951 participated, with a 54% response rate (23–25). Women subsequently completed surveys in 1998 (survey 2), 2001 (survey 3), 2004 (survey 4), 2007 (survey 5), 2010 (survey 6), and 2013 (survey 7). Data on sitting time were available only from survey 3 to survey 7. Women were included in the study if they provided data on sitting time and frailty in survey 3, were not frail (i.e., had a FRAIL score (defined below) less than 3) in 2001 (survey 3), and had complete data for all confounders (survey 1 and survey 3) and frailty in 2013 (survey 7). Figure 1 shows the process of participant selection. Figure 1. View largeDownload slide Selection of participants for a study of sitting patterns and frailty in middle-aged (50–55 years) women, Australia, 2001–2013. Figure 1. View largeDownload slide Selection of participants for a study of sitting patterns and frailty in middle-aged (50–55 years) women, Australia, 2001–2013. Sitting time Data on sitting time were collected from survey 3 to survey 7 using the following 2 self-report items: “How many hours EACH DAY do you typically spend sitting down while doing things like visiting friends, driving, reading, watching television, or working at a desk or computer on 1) a usual weekday and 2) a usual weekend day?”. A similar question is part of the International Physical Activity Questionnaire (26), which has good test-retest reliability for self-report measures: a Spearman reliability coefficient greater than 0.75 and low-to-moderate Spearman correlation coefficients greater than 0.25 when validated against accelerometry (26, 27). Daily sitting time was calculated as [(number of hours/day of sitting on a weekday × 5) + (number of hours/day of sitting on a weekend day × 2)/7] (28). Frailty In all surveys from survey 3 to survey 7, frailty was assessed using the FRAIL scale (29, 30), which is based on 5 domains: fatigue, resistance, ambulation, illness, and loss of weight. Frailty was calculated using the Medical Outcomes Study 36-item Short Form Health Survey (31), the presence of chronic conditions, and weight. Scores on the FRAIL scale range from 0 to 5, with 3 possible categories: healthy (score of 0), prefrail (score of 1–2), and frail (score of 3–5) (5). Fatigue was addressed in 3 questions: “Did you feel worn out?,” “Did you feel tired?,” and “Did you have a lot of energy?”. Fatigue was recorded as present if participants answered the first and second questions as “all of the time,” “most of the time,” or “a good bit of the time” or answered the third question as “some of the time,” “a little of the time,” or “none of the time.” Resistance was addressed in the question, “Does your health now limit you in climbing one flight of stairs?”. A positive response was recorded if participants answered “limited a lot” or “limited a little.” Ambulation was defined as the ability to walk 100 m. A positive response was recorded if participants said they were “limited a lot” or “limited a little” in response to the question, “Does your health now limit you in walking 100 meters?”. Participants reported whether they had been diagnosed with or were being treated for diabetes, heart disease, hypertension, stroke, a low iron level (iron deficiency anemia), asthma, chronic obstructive pulmonary disease (i.e., bronchitis/emphysema), osteoporosis, breast cancer, cervical cancer, depression, anxiety, arthritis/rheumatism, or chronic fatigue syndrome. A positive response was recorded if participants answered “yes” to having at least 5 of the above chronic conditions. Weight was self-reported, and a positive score was recorded if participants had lost more than 5% of their weight from the previous survey. The FRAIL scale has been validated for use in longitudinal studies in this age group, where frailty significantly predicted disability, depression, and mortality over 15 years of follow-up (32). Sociodemographic and lifestyle variables Several variables were included as confounders in analyses examining associations of sitting time trajectories with frailty. Education was assessed in survey 1 as the highest educational qualification completed and was categorized into “less than high school,” “high school only,” and “more than high school.” Data for other variables were collected in survey 3. Relationship status was categorized as partnered or single. Body mass index (weight (kg)/height (m)2) was calculated from self-reported weight and height and was categorized as underweight/healthy (<25.0), overweight (25.0–29.9), or obese (≥30.0) (33). Smoking status was classified as never smoker, ex-smoker, or current smoker. Alcohol consumption was classified into 3 categories: nondrinker, rare drinker (<1 drink per week), and low- to high-risk drinker (≥1 drinks per week) (34). Physical activity was assessed using the Active Australia Survey, where weekly frequency and duration of brisk walking and moderate and vigorous leisure-time physical activity were reported in bouts of 10 minutes or more (35). A physical activity score in MET-minutes/week was calculated as the sum of the products of total weekly minutes spent in each of the 3 categories of physical activity [(walking minutes × 3.0 METs) + (moderate-intensity physical activity minutes × 4.0 METs) + (vigorous-intensity physical activity minutes × 7.5 METs)] (36). Employment status was categorized on the basis of hours of paid employment: full-time (≥35 hours/week), part-time (1–34 hours/week), or other (i.e., home duties/not in paid workforce). Geographic location was based on the Accessibility/Remoteness Index of Australia and was categorized as major city, inner regional area, or outer regional/remote area. Country of birth was categorized as Australia, other English-speaking country, or other. Statistical analyses Differences in characteristics between women in the trajectory groups were assessed using 1-way analysis of variance (continuous normally distributed variables), the Kruskal-Wallis test (continuous non–normally distributed variables), and χ2 analyses (categorical variables). Group-based trajectory modeling using a censored normal model across surveys 3–7 was used to identify patterns in sitting over time (37). The final choice of the trajectory models was based on the Bayesian Information Criterion (BIC) and the log Bayes factor (2 × ΔBIC), an average posterior probability of group membership for all participants greater than 70%, and reasonably tight confidence intervals around the trajectory groups (37, 38). Ordinal logistic regression models were used to examine associations of sitting time trajectories with frailty at the time of survey 7. Robust variance estimates were used to account for repeated measures in individuals. To select confounders for these models, ordinal logistic regression models were used to examine associations of each variable with sitting time trajectories and also with frailty at survey 7. Relationship status, education, body mass index, alcohol consumption, smoking status, physical activity, employment status, and the presence of arthritis, depression, and hypertension were associated with both trajectories and frailty (P < 0.1) and were included as confounders in the final model. Logistic regression models, adjusting for the confounders above, were used to examine associations of sitting time trajectories with FRAIL components. Proportions of women who were frail in each trajectory group at the time of each survey and had deficits for each FRAIL component at surveys 3 and 7 were calculated. Differences in the proportions of women reporting deficits in components of the FRAIL scale across trajectory groups at surveys 3 and 7 were examined using χ2 analyses, and McNemar tests were used to examine differences within trajectory groups for deficits between survey 3 and survey 7. All analyses were conducted in STATA (version 14.1; StataCorp LLC, College Station, Texas). RESULTS Of the 6,298 women (aged 62–67 years) who completed survey 7 and were not frail at survey 3 (i.e., FRAIL score <3), 5,462 were included in this study. At survey 7, a total of 396 (7.3%) women were classified as frail (i.e., FRAIL score >2). Characteristics of these women are shown in Table 1. The mean age at survey 3 of women who were included in the study was 52.5 years. The participants were more likely to be partnered, to have less than a high school education at survey 1, to be underweight/healthy as per their body mass index classification, to have never smoked, to be a low- to high-risk alcohol drinker, to have an active level (≥600 MET-minutes/week) of physical activity, to be employed full-time, to live in an inner regional area, and to have been born in Australia. Table 1. Baseline Characteristics of Middle-Aged (50–55 Years) Women Included in Analyses of Sitting Time Trajectories and Frailty (n = 5,462), Australia, 2001a Characteristic Total (n = 5,462) Sitting Time Trajectory Group Low (n = 1,472) Medium (n = 2,359) Increasing (n = 379) Decreasing (n = 989) High (n = 263) P Value Median (IQR) %b Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Age, yearsc 52.5 (1.5) 52.5 (1.5) 52.5 (1.5) 52.4 (1.4) 52.5 (1.5) 52.5 (1.5) 0.553 Sitting time, hours/day  Survey 3 (2001) 5.1 (3.6–7.1) 3.0 (2.3–4.0) 5.0 (4.0–6.3) 5.7 (4.6–7.0) 8.1 (7.0–9.4) 9.7 (8.1–11.3) <0.001  Survey 4 (2004) 5.3 (4.0–7.4) 3.3 (2.6–4.1) 5.3 (4.4–6.4) 6.6 (5.3–8.0) 8.3 (7.0–9.4) 10.3 (9.3–11.6) <0.001  Survey 5 (2007) 5.8 (4.0–7.7) 3.6 (2.7–4.3) 5.6 (4.6–6.6) 7.9 (6.6–9.0) 8.1 (6.9–9.4) 10.6 (9.4–12.0) <0.001  Survey 6 (2010) 5.4 (4.0–7.4) 3.7 (3.0–4.4) 5.4 (4.6–6.4) 8.4 (7.3–10.0) 7.7 (6.0–8.9) 10.6 (9.4–12.0) <0.001  Survey 7 (2013) 5.3 (4.0–7.0) 3.7 (3.0–4.6) 5.3 (4.3–6.3) 9.6 (8.6–10.6) 6.1 (5.0–7.7) 10.0 (8.3–11.7) <0.001 Relationship status <0.001  Partnered 83.9 87.4 85.6 79.7 79.8 70.7  Single 16.1 12.6 14.4 20.3 20.2 29.3 Education at survey 1 (1996) 0.551  Less than high school 43.7 45.2 43.5 44.9 42.3 40.3  High school 16.8 17.2 16.3 17.4 17.6 15.6  More than high school 39.5 37.6 40.3 37.7 40.1 44.1 Body mass index categoryd <0.001  Underweight/healthy (<25.0) 46.7 53.6 46.5 42.0 42.9 31.2  Overweight (25.0–29.9) 32.8 31.9 34.5 29.3 31.9 30.8  Obese (≥30.0) 20.6 14.5 19.1 28.8 25.3 38.0 Smoking status <0.001  Never smoker 63.8 67.6 63.7 61.5 61.7 54.0  Ex-smoker 24.8 21.5 25.4 26.1 26.9 28.5  Current smoker 11.4 10.9 10.9 12.4 11.4 17.5 Alcohol consumption <0.001  Never drinker 10.4 11.8 9.8 10.6 9.9 9.1  Rare drinker (<1 drink per week) 26.1 27.9 25.8 34.0 21.5 25.1  Low- to high-risk drinker (≥1 drinks per week) 63.5 60.4 64.4 55.4 68.6 65.8 Physical activity, MET-minutes/week 540 (180–1,260) 635 (180–1,406) 600 (210–1,260) 480 (135–1,170) 480 (180–1,080) 270 (60–810) <0.001 Physical activity category <0.001  None (<40.0 MET-minutes/week) 13.7 12.4 12.1 19.5 15.2 20.9  Low (40.0–599.9 MET-minutes/week) 37.6 34.4 37.0 36.7 41.1 48.3  Active (≥600.0 MET-minutes/week) 48.8 53.2 51.0 43.8 43.8 30.8 Employment status <0.001  Full-time 36.0 27.6 33.0 37.7 49.8 55.1  Part-time 34.5 38.5 35.4 36.4 28.6 23.6  Other 29.5 33.9 31.5 25.9 21.6 21.3 Area of residence <0.001  Major city 34.6 28.5 33.2 34.9 43.8 47.2  Inner regional area 42.0 45.5 42.6 41.8 36.2 38.0  Outer regional/remote area 23.4 26.0 24.2 23.3 20.0 14.8 Country of birth <0.001  Australia 77.3 80.3 76.7 75.7 76.8 69.6  Other English-speaking country 15.3 11.0 16.4 17.2 16.0 23.2  Other 7.5 8.7 6.9 7.1 7.2 7.2 Frailty status <0.001  Healthy (FRAIL score = 0) 44.3 46.6 46.7 38.3 41.5 29.7  Prefrail (FRAIL score = 1–2) 55.7 53.4 53.3 61.7 58.5 70.3 Characteristic Total (n = 5,462) Sitting Time Trajectory Group Low (n = 1,472) Medium (n = 2,359) Increasing (n = 379) Decreasing (n = 989) High (n = 263) P Value Median (IQR) %b Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Age, yearsc 52.5 (1.5) 52.5 (1.5) 52.5 (1.5) 52.4 (1.4) 52.5 (1.5) 52.5 (1.5) 0.553 Sitting time, hours/day  Survey 3 (2001) 5.1 (3.6–7.1) 3.0 (2.3–4.0) 5.0 (4.0–6.3) 5.7 (4.6–7.0) 8.1 (7.0–9.4) 9.7 (8.1–11.3) <0.001  Survey 4 (2004) 5.3 (4.0–7.4) 3.3 (2.6–4.1) 5.3 (4.4–6.4) 6.6 (5.3–8.0) 8.3 (7.0–9.4) 10.3 (9.3–11.6) <0.001  Survey 5 (2007) 5.8 (4.0–7.7) 3.6 (2.7–4.3) 5.6 (4.6–6.6) 7.9 (6.6–9.0) 8.1 (6.9–9.4) 10.6 (9.4–12.0) <0.001  Survey 6 (2010) 5.4 (4.0–7.4) 3.7 (3.0–4.4) 5.4 (4.6–6.4) 8.4 (7.3–10.0) 7.7 (6.0–8.9) 10.6 (9.4–12.0) <0.001  Survey 7 (2013) 5.3 (4.0–7.0) 3.7 (3.0–4.6) 5.3 (4.3–6.3) 9.6 (8.6–10.6) 6.1 (5.0–7.7) 10.0 (8.3–11.7) <0.001 Relationship status <0.001  Partnered 83.9 87.4 85.6 79.7 79.8 70.7  Single 16.1 12.6 14.4 20.3 20.2 29.3 Education at survey 1 (1996) 0.551  Less than high school 43.7 45.2 43.5 44.9 42.3 40.3  High school 16.8 17.2 16.3 17.4 17.6 15.6  More than high school 39.5 37.6 40.3 37.7 40.1 44.1 Body mass index categoryd <0.001  Underweight/healthy (<25.0) 46.7 53.6 46.5 42.0 42.9 31.2  Overweight (25.0–29.9) 32.8 31.9 34.5 29.3 31.9 30.8  Obese (≥30.0) 20.6 14.5 19.1 28.8 25.3 38.0 Smoking status <0.001  Never smoker 63.8 67.6 63.7 61.5 61.7 54.0  Ex-smoker 24.8 21.5 25.4 26.1 26.9 28.5  Current smoker 11.4 10.9 10.9 12.4 11.4 17.5 Alcohol consumption <0.001  Never drinker 10.4 11.8 9.8 10.6 9.9 9.1  Rare drinker (<1 drink per week) 26.1 27.9 25.8 34.0 21.5 25.1  Low- to high-risk drinker (≥1 drinks per week) 63.5 60.4 64.4 55.4 68.6 65.8 Physical activity, MET-minutes/week 540 (180–1,260) 635 (180–1,406) 600 (210–1,260) 480 (135–1,170) 480 (180–1,080) 270 (60–810) <0.001 Physical activity category <0.001  None (<40.0 MET-minutes/week) 13.7 12.4 12.1 19.5 15.2 20.9  Low (40.0–599.9 MET-minutes/week) 37.6 34.4 37.0 36.7 41.1 48.3  Active (≥600.0 MET-minutes/week) 48.8 53.2 51.0 43.8 43.8 30.8 Employment status <0.001  Full-time 36.0 27.6 33.0 37.7 49.8 55.1  Part-time 34.5 38.5 35.4 36.4 28.6 23.6  Other 29.5 33.9 31.5 25.9 21.6 21.3 Area of residence <0.001  Major city 34.6 28.5 33.2 34.9 43.8 47.2  Inner regional area 42.0 45.5 42.6 41.8 36.2 38.0  Outer regional/remote area 23.4 26.0 24.2 23.3 20.0 14.8 Country of birth <0.001  Australia 77.3 80.3 76.7 75.7 76.8 69.6  Other English-speaking country 15.3 11.0 16.4 17.2 16.0 23.2  Other 7.5 8.7 6.9 7.1 7.2 7.2 Frailty status <0.001  Healthy (FRAIL score = 0) 44.3 46.6 46.7 38.3 41.5 29.7  Prefrail (FRAIL score = 1–2) 55.7 53.4 53.3 61.7 58.5 70.3 Abbreviations: FRAIL, fatigue, resistance, ambulation, illness, loss of weight; IQR, interquartile range; MET, metabolic equivalent. a Data are from survey 3, except where indicated. b Percentages may add up to more than 100% because of rounding. c Values are expressed as mean (standard deviation). d Weight (kg)/height (m)2. Table 1. Baseline Characteristics of Middle-Aged (50–55 Years) Women Included in Analyses of Sitting Time Trajectories and Frailty (n = 5,462), Australia, 2001a Characteristic Total (n = 5,462) Sitting Time Trajectory Group Low (n = 1,472) Medium (n = 2,359) Increasing (n = 379) Decreasing (n = 989) High (n = 263) P Value Median (IQR) %b Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Age, yearsc 52.5 (1.5) 52.5 (1.5) 52.5 (1.5) 52.4 (1.4) 52.5 (1.5) 52.5 (1.5) 0.553 Sitting time, hours/day  Survey 3 (2001) 5.1 (3.6–7.1) 3.0 (2.3–4.0) 5.0 (4.0–6.3) 5.7 (4.6–7.0) 8.1 (7.0–9.4) 9.7 (8.1–11.3) <0.001  Survey 4 (2004) 5.3 (4.0–7.4) 3.3 (2.6–4.1) 5.3 (4.4–6.4) 6.6 (5.3–8.0) 8.3 (7.0–9.4) 10.3 (9.3–11.6) <0.001  Survey 5 (2007) 5.8 (4.0–7.7) 3.6 (2.7–4.3) 5.6 (4.6–6.6) 7.9 (6.6–9.0) 8.1 (6.9–9.4) 10.6 (9.4–12.0) <0.001  Survey 6 (2010) 5.4 (4.0–7.4) 3.7 (3.0–4.4) 5.4 (4.6–6.4) 8.4 (7.3–10.0) 7.7 (6.0–8.9) 10.6 (9.4–12.0) <0.001  Survey 7 (2013) 5.3 (4.0–7.0) 3.7 (3.0–4.6) 5.3 (4.3–6.3) 9.6 (8.6–10.6) 6.1 (5.0–7.7) 10.0 (8.3–11.7) <0.001 Relationship status <0.001  Partnered 83.9 87.4 85.6 79.7 79.8 70.7  Single 16.1 12.6 14.4 20.3 20.2 29.3 Education at survey 1 (1996) 0.551  Less than high school 43.7 45.2 43.5 44.9 42.3 40.3  High school 16.8 17.2 16.3 17.4 17.6 15.6  More than high school 39.5 37.6 40.3 37.7 40.1 44.1 Body mass index categoryd <0.001  Underweight/healthy (<25.0) 46.7 53.6 46.5 42.0 42.9 31.2  Overweight (25.0–29.9) 32.8 31.9 34.5 29.3 31.9 30.8  Obese (≥30.0) 20.6 14.5 19.1 28.8 25.3 38.0 Smoking status <0.001  Never smoker 63.8 67.6 63.7 61.5 61.7 54.0  Ex-smoker 24.8 21.5 25.4 26.1 26.9 28.5  Current smoker 11.4 10.9 10.9 12.4 11.4 17.5 Alcohol consumption <0.001  Never drinker 10.4 11.8 9.8 10.6 9.9 9.1  Rare drinker (<1 drink per week) 26.1 27.9 25.8 34.0 21.5 25.1  Low- to high-risk drinker (≥1 drinks per week) 63.5 60.4 64.4 55.4 68.6 65.8 Physical activity, MET-minutes/week 540 (180–1,260) 635 (180–1,406) 600 (210–1,260) 480 (135–1,170) 480 (180–1,080) 270 (60–810) <0.001 Physical activity category <0.001  None (<40.0 MET-minutes/week) 13.7 12.4 12.1 19.5 15.2 20.9  Low (40.0–599.9 MET-minutes/week) 37.6 34.4 37.0 36.7 41.1 48.3  Active (≥600.0 MET-minutes/week) 48.8 53.2 51.0 43.8 43.8 30.8 Employment status <0.001  Full-time 36.0 27.6 33.0 37.7 49.8 55.1  Part-time 34.5 38.5 35.4 36.4 28.6 23.6  Other 29.5 33.9 31.5 25.9 21.6 21.3 Area of residence <0.001  Major city 34.6 28.5 33.2 34.9 43.8 47.2  Inner regional area 42.0 45.5 42.6 41.8 36.2 38.0  Outer regional/remote area 23.4 26.0 24.2 23.3 20.0 14.8 Country of birth <0.001  Australia 77.3 80.3 76.7 75.7 76.8 69.6  Other English-speaking country 15.3 11.0 16.4 17.2 16.0 23.2  Other 7.5 8.7 6.9 7.1 7.2 7.2 Frailty status <0.001  Healthy (FRAIL score = 0) 44.3 46.6 46.7 38.3 41.5 29.7  Prefrail (FRAIL score = 1–2) 55.7 53.4 53.3 61.7 58.5 70.3 Characteristic Total (n = 5,462) Sitting Time Trajectory Group Low (n = 1,472) Medium (n = 2,359) Increasing (n = 379) Decreasing (n = 989) High (n = 263) P Value Median (IQR) %b Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Age, yearsc 52.5 (1.5) 52.5 (1.5) 52.5 (1.5) 52.4 (1.4) 52.5 (1.5) 52.5 (1.5) 0.553 Sitting time, hours/day  Survey 3 (2001) 5.1 (3.6–7.1) 3.0 (2.3–4.0) 5.0 (4.0–6.3) 5.7 (4.6–7.0) 8.1 (7.0–9.4) 9.7 (8.1–11.3) <0.001  Survey 4 (2004) 5.3 (4.0–7.4) 3.3 (2.6–4.1) 5.3 (4.4–6.4) 6.6 (5.3–8.0) 8.3 (7.0–9.4) 10.3 (9.3–11.6) <0.001  Survey 5 (2007) 5.8 (4.0–7.7) 3.6 (2.7–4.3) 5.6 (4.6–6.6) 7.9 (6.6–9.0) 8.1 (6.9–9.4) 10.6 (9.4–12.0) <0.001  Survey 6 (2010) 5.4 (4.0–7.4) 3.7 (3.0–4.4) 5.4 (4.6–6.4) 8.4 (7.3–10.0) 7.7 (6.0–8.9) 10.6 (9.4–12.0) <0.001  Survey 7 (2013) 5.3 (4.0–7.0) 3.7 (3.0–4.6) 5.3 (4.3–6.3) 9.6 (8.6–10.6) 6.1 (5.0–7.7) 10.0 (8.3–11.7) <0.001 Relationship status <0.001  Partnered 83.9 87.4 85.6 79.7 79.8 70.7  Single 16.1 12.6 14.4 20.3 20.2 29.3 Education at survey 1 (1996) 0.551  Less than high school 43.7 45.2 43.5 44.9 42.3 40.3  High school 16.8 17.2 16.3 17.4 17.6 15.6  More than high school 39.5 37.6 40.3 37.7 40.1 44.1 Body mass index categoryd <0.001  Underweight/healthy (<25.0) 46.7 53.6 46.5 42.0 42.9 31.2  Overweight (25.0–29.9) 32.8 31.9 34.5 29.3 31.9 30.8  Obese (≥30.0) 20.6 14.5 19.1 28.8 25.3 38.0 Smoking status <0.001  Never smoker 63.8 67.6 63.7 61.5 61.7 54.0  Ex-smoker 24.8 21.5 25.4 26.1 26.9 28.5  Current smoker 11.4 10.9 10.9 12.4 11.4 17.5 Alcohol consumption <0.001  Never drinker 10.4 11.8 9.8 10.6 9.9 9.1  Rare drinker (<1 drink per week) 26.1 27.9 25.8 34.0 21.5 25.1  Low- to high-risk drinker (≥1 drinks per week) 63.5 60.4 64.4 55.4 68.6 65.8 Physical activity, MET-minutes/week 540 (180–1,260) 635 (180–1,406) 600 (210–1,260) 480 (135–1,170) 480 (180–1,080) 270 (60–810) <0.001 Physical activity category <0.001  None (<40.0 MET-minutes/week) 13.7 12.4 12.1 19.5 15.2 20.9  Low (40.0–599.9 MET-minutes/week) 37.6 34.4 37.0 36.7 41.1 48.3  Active (≥600.0 MET-minutes/week) 48.8 53.2 51.0 43.8 43.8 30.8 Employment status <0.001  Full-time 36.0 27.6 33.0 37.7 49.8 55.1  Part-time 34.5 38.5 35.4 36.4 28.6 23.6  Other 29.5 33.9 31.5 25.9 21.6 21.3 Area of residence <0.001  Major city 34.6 28.5 33.2 34.9 43.8 47.2  Inner regional area 42.0 45.5 42.6 41.8 36.2 38.0  Outer regional/remote area 23.4 26.0 24.2 23.3 20.0 14.8 Country of birth <0.001  Australia 77.3 80.3 76.7 75.7 76.8 69.6  Other English-speaking country 15.3 11.0 16.4 17.2 16.0 23.2  Other 7.5 8.7 6.9 7.1 7.2 7.2 Frailty status <0.001  Healthy (FRAIL score = 0) 44.3 46.6 46.7 38.3 41.5 29.7  Prefrail (FRAIL score = 1–2) 55.7 53.4 53.3 61.7 58.5 70.3 Abbreviations: FRAIL, fatigue, resistance, ambulation, illness, loss of weight; IQR, interquartile range; MET, metabolic equivalent. a Data are from survey 3, except where indicated. b Percentages may add up to more than 100% because of rounding. c Values are expressed as mean (standard deviation). d Weight (kg)/height (m)2. As shown in Figure 2, 5 participant clusters representing sitting time trajectories were identified: low (n = 1,472 of 5,462; 26.9%); medium (n = 2,359 of 5,462; 43.1% (referent)), increasing (n = 379 of 5,462; 6.9%), decreasing (n = 989 of 5,462; 18.1%), and high (n = 263 of 5,462; 4.8%). As shown in Table 1, there were no differences in age or education across the 5 trajectory groups. Compared with the other 4 trajectory groups, the high trajectory group had more single women, obese women, current smokers, women in full-time employment, women living in major cities, women born in other English-speaking countries, and women with low physical activity. Figure 2. View largeDownload slide Sitting time trajectories observed over the course of 12 years among middle-aged (50–55 years) women, Australia, 2001–2013. Percentage of participants in each trajectory group: low, 26.9%; medium, 43.1%; increasing, 6.9%; decreasing, 18.1%; high, 4.8%. Figure 2. View largeDownload slide Sitting time trajectories observed over the course of 12 years among middle-aged (50–55 years) women, Australia, 2001–2013. Percentage of participants in each trajectory group: low, 26.9%; medium, 43.1%; increasing, 6.9%; decreasing, 18.1%; high, 4.8%. In fully adjusted models, these trajectory groups were associated with frailty at survey 7. The medium trajectory was used as a reference category, since it had the largest number of participants. Compared with women in the medium trajectory group, women in the increasing and high trajectories were more likely to be frail versus prefrail and healthy, with odds ratios of 1.29 (95% confidence interval (CI): 1.03, 1.61) and 1.42 (95% CI: 1.10, 1.84), respectively, with those in the low trajectory group being less likely to be frail (odds ratio = 0.86, 95% CI: 0.75, 0.98) (Table 2). The decreasing trajectory group, however, showed no difference in the risk of frailty at survey 7. Table 2. Distribution of Participants by Frailty Status and Odds of Frailty According to 12-Year Sitting Trajectory (Ordinal Logistic Regression Analysisa) Among Middle-Aged (50–55 Years) Women, Australia, 2001–2013 Trajectory Group Frailty Statusb and No. of Participants Odds of Frailty Healthy (n = 2,500) Prefrail (n = 2,566) Frail (n = 396) OR 95% CI Medium 1,087 1,114 158 1.00 Referent Low 755 639 78 0.86c 0.75, 0.98 Increasing 149 184 46 1.29c 1.03, 1.61 Decreasing 425 479 85 1.14 0.98, 1.32 High 84 150 29 1.42d 1.10, 1.84 Trajectory Group Frailty Statusb and No. of Participants Odds of Frailty Healthy (n = 2,500) Prefrail (n = 2,566) Frail (n = 396) OR 95% CI Medium 1,087 1,114 158 1.00 Referent Low 755 639 78 0.86c 0.75, 0.98 Increasing 149 184 46 1.29c 1.03, 1.61 Decreasing 425 479 85 1.14 0.98, 1.32 High 84 150 29 1.42d 1.10, 1.84 Abbreviations: CI, confidence interval; FRAIL, fatigue, resistance, ambulation, illness, loss of weight; OR, odds ratio. a Models adjusted for relationship status, education, body mass index, smoking status, alcohol consumption, physical activity, employment, and the presence of arthritis, depression, or hypertension. b Healthy: FRAIL score = 0; prefrail: FRAIL score = 1–2; frail: FRAIL score = 3–5. cP < 0.05. dP < 0.01. Table 2. Distribution of Participants by Frailty Status and Odds of Frailty According to 12-Year Sitting Trajectory (Ordinal Logistic Regression Analysisa) Among Middle-Aged (50–55 Years) Women, Australia, 2001–2013 Trajectory Group Frailty Statusb and No. of Participants Odds of Frailty Healthy (n = 2,500) Prefrail (n = 2,566) Frail (n = 396) OR 95% CI Medium 1,087 1,114 158 1.00 Referent Low 755 639 78 0.86c 0.75, 0.98 Increasing 149 184 46 1.29c 1.03, 1.61 Decreasing 425 479 85 1.14 0.98, 1.32 High 84 150 29 1.42d 1.10, 1.84 Trajectory Group Frailty Statusb and No. of Participants Odds of Frailty Healthy (n = 2,500) Prefrail (n = 2,566) Frail (n = 396) OR 95% CI Medium 1,087 1,114 158 1.00 Referent Low 755 639 78 0.86c 0.75, 0.98 Increasing 149 184 46 1.29c 1.03, 1.61 Decreasing 425 479 85 1.14 0.98, 1.32 High 84 150 29 1.42d 1.10, 1.84 Abbreviations: CI, confidence interval; FRAIL, fatigue, resistance, ambulation, illness, loss of weight; OR, odds ratio. a Models adjusted for relationship status, education, body mass index, smoking status, alcohol consumption, physical activity, employment, and the presence of arthritis, depression, or hypertension. b Healthy: FRAIL score = 0; prefrail: FRAIL score = 1–2; frail: FRAIL score = 3–5. cP < 0.05. dP < 0.01. Table 3 shows the relationship of the 5 components of the FRAIL scale with each trajectory group. Following adjustment for the confounding factors, compared with the medium trajectory, women in the increasing and high trajectories had 25% and 56% increased likelihoods of having fatigue, respectively, while women in the low trajectory group had a 16% decreased likelihood of having fatigue. Compared with women in the medium trajectory, women in the increasing trajectory group had a 47% increased likelihood of having a deficit in resistance. However, no association was observed with the other 3 components. Table 3. Association of 12-Year Sitting Trajectory With Components of the FRAIL Scale (Logistic Regression Analysisa) Among Middle-Aged (50–55 Years) Women, Australia, 2001–2013 Trajectory Group Component of FRAIL Scale Fatigue Resistance Ambulation Illness Loss of Weight OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI Medium 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent Low 0.84b 0.73, 0.96 0.97 0.80, 1.16 1.04 0.79, 1.37 0.73 0.44, 1.21 0.89 0.73, 1.08 Increasing 1.25b 1.00, 1.57 1.47c 1.12, 1.92 1.36 0.91, 2.02 1.34 0.71, 2.54 0.98 0.71, 1.33 Decreasing 1.10 0.94, 1.29 1.00 0.81, 1.23 1.26 0.94, 1.69 0.94 0.56, 1.57 1.10 0.89, 1.36 High 1.56c 1.19, 2.04 1.10 0.79, 1.53 1.05 0.65, 1.69 1.16 0.53, 2.50 1.24 0.88, 1.74 Trajectory Group Component of FRAIL Scale Fatigue Resistance Ambulation Illness Loss of Weight OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI Medium 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent Low 0.84b 0.73, 0.96 0.97 0.80, 1.16 1.04 0.79, 1.37 0.73 0.44, 1.21 0.89 0.73, 1.08 Increasing 1.25b 1.00, 1.57 1.47c 1.12, 1.92 1.36 0.91, 2.02 1.34 0.71, 2.54 0.98 0.71, 1.33 Decreasing 1.10 0.94, 1.29 1.00 0.81, 1.23 1.26 0.94, 1.69 0.94 0.56, 1.57 1.10 0.89, 1.36 High 1.56c 1.19, 2.04 1.10 0.79, 1.53 1.05 0.65, 1.69 1.16 0.53, 2.50 1.24 0.88, 1.74 Abbreviations: CI, confidence interval; FRAIL, fatigue, resistance, ambulation, illness, loss of weight; OR, odds ratio. a Models adjusted for relationship status, education, body mass index, smoking status, alcohol consumption, physical activity, employment, and the presence of arthritis, depression, or hypertension. bP < 0.05. cP < 0.01. Table 3. Association of 12-Year Sitting Trajectory With Components of the FRAIL Scale (Logistic Regression Analysisa) Among Middle-Aged (50–55 Years) Women, Australia, 2001–2013 Trajectory Group Component of FRAIL Scale Fatigue Resistance Ambulation Illness Loss of Weight OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI Medium 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent Low 0.84b 0.73, 0.96 0.97 0.80, 1.16 1.04 0.79, 1.37 0.73 0.44, 1.21 0.89 0.73, 1.08 Increasing 1.25b 1.00, 1.57 1.47c 1.12, 1.92 1.36 0.91, 2.02 1.34 0.71, 2.54 0.98 0.71, 1.33 Decreasing 1.10 0.94, 1.29 1.00 0.81, 1.23 1.26 0.94, 1.69 0.94 0.56, 1.57 1.10 0.89, 1.36 High 1.56c 1.19, 2.04 1.10 0.79, 1.53 1.05 0.65, 1.69 1.16 0.53, 2.50 1.24 0.88, 1.74 Trajectory Group Component of FRAIL Scale Fatigue Resistance Ambulation Illness Loss of Weight OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI Medium 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent Low 0.84b 0.73, 0.96 0.97 0.80, 1.16 1.04 0.79, 1.37 0.73 0.44, 1.21 0.89 0.73, 1.08 Increasing 1.25b 1.00, 1.57 1.47c 1.12, 1.92 1.36 0.91, 2.02 1.34 0.71, 2.54 0.98 0.71, 1.33 Decreasing 1.10 0.94, 1.29 1.00 0.81, 1.23 1.26 0.94, 1.69 0.94 0.56, 1.57 1.10 0.89, 1.36 High 1.56c 1.19, 2.04 1.10 0.79, 1.53 1.05 0.65, 1.69 1.16 0.53, 2.50 1.24 0.88, 1.74 Abbreviations: CI, confidence interval; FRAIL, fatigue, resistance, ambulation, illness, loss of weight; OR, odds ratio. a Models adjusted for relationship status, education, body mass index, smoking status, alcohol consumption, physical activity, employment, and the presence of arthritis, depression, or hypertension. bP < 0.05. cP < 0.01. Figure 3 shows the difference in numbers of women reporting deficits in each component of the FRAIL scale in surveys 3 and 7 for each trajectory group. At the time of survey 3, there were differences across trajectory groups in the proportions of women with deficits in fatigue, resistance, and loss of weight and differences across all trajectory groups for all components at survey 7. From survey 3 to survey 7, there was an increase in the proportion of women reporting deficits in all trajectory groups for the resistance, ambulation, and illness components; a decrease in most of the categories of women reporting deficits in fatigue; and an increase in most of the trajectory groups for the loss-of-weight component. As Web Figure 1 (available at https://academic.oup.com/aje) shows, the proportion of women who were frail increased in all trajectory groups, with the highest proportion of frailty being observed in the high (11.0%) and increasing (12.1%) trajectory groups at survey 7. In all surveys, there were lower proportions of healthy women in the high trajectory group than in the other trajectory groups. Figure 3. View largeDownload slide Proportions of middle-aged (50–55 years) women with deficits in components of the FRAIL scale at survey 3 and survey 7, by sitting time trajectory group, Australia, 2001–2013. McNemar’s test was used to assess differences within trajectory group at surveys 3 and 7 for all components of the FRAIL scale. The proportion of women with a deficit in the fatigue component decreased from survey 3 to survey 7 for women in the low, medium, decreasing, and high trajectory groups. The proportions of women with deficits in the resistance, ambulation, and illness components increased from survey 3 to survey 7 in all trajectory groups. The proportion of women with a deficit in the loss-of-weight component increased from survey 3 to survey 7 for women in the medium, decreasing, and high trajectory groups. χ2 analysis was used to assess differences across trajectory groups at surveys 3 and 7. There were significant differences in the proportions of women with deficits in fatigue, resistance, and loss of weight across trajectory groups at survey 3. There was a significant difference in the proportion of women with deficits in all FRAIL scale components across trajectory groups at survey 7. FRAIL, fatigue, resistance, ambulation, illness, loss of weight. Figure 3. View largeDownload slide Proportions of middle-aged (50–55 years) women with deficits in components of the FRAIL scale at survey 3 and survey 7, by sitting time trajectory group, Australia, 2001–2013. McNemar’s test was used to assess differences within trajectory group at surveys 3 and 7 for all components of the FRAIL scale. The proportion of women with a deficit in the fatigue component decreased from survey 3 to survey 7 for women in the low, medium, decreasing, and high trajectory groups. The proportions of women with deficits in the resistance, ambulation, and illness components increased from survey 3 to survey 7 in all trajectory groups. The proportion of women with a deficit in the loss-of-weight component increased from survey 3 to survey 7 for women in the medium, decreasing, and high trajectory groups. χ2 analysis was used to assess differences across trajectory groups at surveys 3 and 7. There were significant differences in the proportions of women with deficits in fatigue, resistance, and loss of weight across trajectory groups at survey 3. There was a significant difference in the proportion of women with deficits in all FRAIL scale components across trajectory groups at survey 7. FRAIL, fatigue, resistance, ambulation, illness, loss of weight. DISCUSSION In this 12-year longitudinal study of middle-aged Australian women, 5 distinct trajectory patterns of sitting time were identified. These patterns of sitting time from ages 50–55 years to ages 62–67 years were associated with frailty, such that women with unfavorable patterns of sitting were more likely to be frail at the end of the study. In contrast, the low sitting time pattern reduced the likelihood of being frail. Women in the decreasing trajectory group had no difference in the likelihood of being frail compared with women with the moderate sitting pattern, despite having higher levels of sitting at baseline. These findings are consistent with previous studies (19–22). Previous cross-sectional studies have found that people with higher levels of sedentary behavior had a higher frailty index score (19) or were frail according to modified Fried criteria (20) or the Frailty Trait Scale (21), with García-Esquinas et al. reporting in a longitudinal study that TV viewing time predicted frailty (Fried’s criteria) at follow-up (22). Furthermore, in one cross-sectional study, Virtuoso Júnior et al. (39) reported that sitting time was an independent indicator of frailty as identified by biomarkers (C-reactive protein and white blood cell count) in hospitalized adults aged ≥60 years. In this study, 4.9% of women aged 50–55 years were frail, which is consistent with previous estimates (40, 41). Investigators have reported an exponential increase in frailty prevalence, from 6.5% in people aged 60–69 years to 65% in people aged >90 years (40), and increases in prevalence of 3% per annum starting from late middle age (42), suggesting that interventions designed to prevent frailty should target persons in middle age. The present study suggests that sitting time is a novel risk factor for the development of frailty. Objectively measured sedentary behavior is associated with mortality in inactive vulnerable or frail people (frailty index score >0.1) (43). Therefore, interventions targeting sitting time may benefit those who are frail. In a meta-analysis, Martin et al. (44) reported that sedentary behavior interventions reduced sitting time by 42 minutes per day, suggesting that reductions in sitting time are feasible and effective. While no researchers implementing a sitting time intervention have reported on frailty outcomes, Rosenberg et al. (45) observed an increase in gait speed (0.5 seconds faster in completing a 3-m walk test from preintervention to postintervention) following a behavioral sitting-time reduction program. It is important to also consider the context in which sedentary time is accumulated. García-Esquinas et al. reported that TV viewing time was associated with frailty but other sedentary behaviors, such as reading, computer/Internet use, listening to music, or time spent in transportation, were not (22). Assessing different domains of sitting (i.e., occupation, transport, leisure) and specific contexts (e.g., TV-watching or using a computer) could potentially assist in developing more targeted interventions. However, these types of data were only available at survey 6, and therefore trajectories of domain- or context-specific sitting could not be investigated. The findings related to the fatigue component of the FRAIL scale were similar to those for frailty. The fatigue component has a stronger correlation with total FRAIL score (Spearman’s ρ = 0.82) than other components (Spearman’s ρ’s ranging from 0.13 to 0.57) (32). This may be due to the method used to operationalize the FRAIL scale in our study: using 3 items from Short Form 36 as compared with 1 item for the other phenotype components of the scale (i.e., resistance and ambulation). It is possible that fatigue may influence sitting time, and this needs further investigation. However, the proportions of women with deficits in resistance, ambulation, illness, and loss-of-weight components increased over the 12 years of the study, while fatigue declined. An accumulating body of evidence suggests that high levels of sedentary time are associated with poorer physical function (46). An interesting finding in the current study was the relationship of resistance with the increasing sitting trajectory group but not the high sitting trajectory group. It is possible that an increase in sitting time is a marker of poor health and may be a simple way to screen people for adverse health outcomes. This highlights the utility of methods such as group-based trajectory modeling to investigate associations of sedentary behavior with health outcomes, as different patterns of behavior may affect health differently. In the Raine Study, 15-year trajectories of TV viewing time from age 5 years onward were associated with body composition at age 20 years, such that people in low and increasing TV trajectories had a lower percentage of body fat and a higher bone mineral content than those in the high TV trajectory group (47, 48). Twelve-year trajectories of TV time are associated with lower body strength but not performance in the Timed Up and Go (TUG) Test (49). A strength of this study was its longitudinal design over the course of 12 years, which included a large and representative sample of middle-aged women in Australia. To the best of our knowledge, this is the first longitudinal study to have examined the relationship between sitting time and frailty (using group-based trajectory modeling) and is the first study to identify trajectories of sitting time. In this study, factors that are related to sitting time and frailty (i.e., sociodemographic and lifestyle factors) were included in the analyses; however, no data on other factors, such as nutrition or cognition, were available. Another strength of this study was the exclusion of women who were frail at survey 3, which allowed us to examine sitting time as a risk factor for the development of frailty. A limitation of this study was the use of self-reported measures of sitting time. Even though the questionnaire we used is similar to the validated International Physical Activity Questionnaire (26, 27), there is a tendency for people to underreport sitting time in comparison with objective measures (50). However, underreporting of sitting time is expected to have been consistent over time and hence is unlikely to have affected the overall findings. While the use of objective measures of sitting time in longitudinal cohort studies is still in early stages, our study had data on sitting collected over 12 years at multiple time points. In future studies, researchers should consider using objective measures of sitting time, such as activPAL (50). activPAL is a physical activity monitor that is based on a uniaxial accelerometer (PAL Technologies Ltd., Glasgow, United Kingdom) (51). Patterns of accumulation of sitting—for example, breaks in sitting time—are known to influence health outcomes such as biomarkers of cardiometabolic risk (6) and premature mortality (52). In a recent cross-sectional study in older adults, Del Pozo-Cruz et al. (21) reported that total sedentary time and sedentary time accumulated in bouts of at least 10 minutes were deleteriously associated with frailty while a higher number of breaks in sedentary time was beneficially associated with frailty. The results are only applicable to women in this age group and not to men. In a 2016 systematic review, Buta et al. (53) identified 67 instruments designed to assess frailty, with Fried’s criteria (3), which combine self-reports and performance-based tests to assess a frailty phenotype, being the most commonly used in the literature, followed by instruments that use a deficit-accumulation model such as the frailty index (42). The FRAIL scale uses self-report data to assess functional deficit accumulation and biological domains of frailty. Thus, the findings of this study may not apply to other models of frailty that utilize performance-based measures or wider sources of data. Further investigation is required to determine whether these findings can be replicated using other models of frailty. In conclusion, in this study, 12-year patterns of sitting time in middle-aged Australian women predicted frailty in older age. Additional studies are required to understand the biological mechanisms of how sitting time may be related to frailty. Programs designed to reduce this novel risk factor may help prevent frailty in this at-risk population. ACKNOWLEDGMENTS Author affiliations: Centre for Health Services Research, Faculty of Medicine, The University of Queensland, Brisbane, Queensland, Australia (Maja Susanto, Ruth E. Hubbard, Paul A. Gardiner); The Princess Alexandra Hospital, Brisbane, Queensland, Australia (Maja Susanto, Ruth E. Hubbard); and Mater Research Institute, The University of Queensland, Brisbane, Queensland, Australia (Paul A. Gardiner). This work was supported by the National Health and Medical Research Council of Australia and the Australian Research Council (Dementia Research Development Fellowship 110331 to P.A.G.). Conflict of interest: none declared. Abbreviations CI confidence interval FRAIL fatigue, resistance, ambulation, illness, loss of weight MET metabolic equivalent TV television REFERENCES 1 Buckinx F , Rolland Y , Reginster JY , et al. . Burden of frailty in the elderly population: perspectives for a public health challenge . Arch Public Health . 2015 ; 73 ( 1 ): 19 . Google Scholar Crossref Search ADS PubMed 2 Collard RM , Boter H , Schoevers RA , et al. . Prevalence of frailty in community-dwelling older persons: a systematic review . J Am Geriatr Soc . 2012 ; 60 ( 8 ): 1487 – 1492 . Google Scholar Crossref Search ADS PubMed 3 Fried LP , Tangen CM , Walston J , et al. . Frailty in older adults: evidence for a phenotype . J Gerontol A Biol Sci Med Sci . 2001 ; 56 ( 3 ): M146 – M156 . Google Scholar Crossref Search ADS PubMed 4 Clegg A , Young J , Iliffe S , et al. . Frailty in elderly people . Lancet . 2013 ; 381 ( 9868 ): 752 – 762 . Google Scholar Crossref Search ADS PubMed 5 Gardiner PA , Mishra GD , Dobson AJ . Validity and responsiveness of the FRAIL scale in a longitudinal cohort study of older Australian women . J Am Med Dir Assoc . 2015 ; 16 ( 9 ): 781 – 783 . Google Scholar Crossref Search ADS PubMed 6 Healy GN , Matthews CE , Dunstan DW , et al. . Sedentary time and cardio-metabolic biomarkers in US adults: NHANES 2003–06 . Eur Heart J . 2011 ; 32 ( 5 ): 590 – 597 . Google Scholar Crossref Search ADS PubMed 7 Saidj M , Jørgensen T , Jacobsen RK , et al. . Work and leisure time sitting and inactivity: effects on cardiorespiratory and metabolic health . Eur J Prev Cardiol . 2016 ; 23 ( 12 ): 1321 – 1329 . Google Scholar Crossref Search ADS PubMed 8 Ramsay SE , Arianayagam DS , Whincup PH , et al. . Cardiovascular risk profile and frailty in a population-based study of older British men . Heart . 2015 ; 101 ( 8 ): 616 – 622 . Google Scholar Crossref Search ADS PubMed 9 Owen N , Sugiyama T , Eakin EE , et al. . Adults’ sedentary behavior determinants and interventions . Am J Prev Med . 2011 ; 41 ( 2 ): 189 – 196 . Google Scholar Crossref Search ADS PubMed 10 Ng SW , Popkin BM . Time use and physical activity: a shift away from movement across the globe . Obes Rev . 2012 ; 13 ( 8 ): 659 – 680 . Google Scholar Crossref Search ADS PubMed 11 Australian Bureau of Statistics . Australian Health Survey: Physical Activity, 2011–12. Canberra, ACT, Australia : Australian Bureau of Statistics ; 2013 . (Catalog no. 4364.0.55.004). http://www.abs.gov.au/AUSSTATS/abs@.nsf/Lookup/4364.0.55.004Main+Features12011-12?OpenDocument. Accessed March 11, 2017. 12 Clark BK , Lynch BM , Winkler EA , et al. . Validity of a multi-context sitting questionnaire across demographically diverse population groups: AusDiab3 . Int J Behav Nutr Phys Act . 2015 ; 12 : 148 . Google Scholar Crossref Search ADS PubMed 13 Chau JY , Merom D , Grunseit A , et al. . Temporal trends in non-occupational sedentary behaviours from Australian Time Use Surveys 1992, 1997 and 2006 . Int J Behav Nutr Phys Act . 2012 ; 9 : 76 . Google Scholar Crossref Search ADS PubMed 14 Church TS , Thomas DM , Tudor-Locke C , et al. . Trends over 5 decades in US occupation-related physical activity and their associations with obesity . PLoS One . 2011 ; 6 ( 5 ): e19657 . Google Scholar Crossref Search ADS PubMed 15 Clark BK , Peeters GM , Gomersall SR , et al. . Nine year changes in sitting time in young and mid-aged Australian women: findings from the Australian Longitudinal Study for Women’s Health . Prev Med . 2014 ; 64 : 1 – 7 . Google Scholar Crossref Search ADS PubMed 16 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 17 de Rezende LF , Rodrigues Lopes M , Rey-López JP , et al. . Sedentary behavior and health outcomes: an overview of systematic reviews . PLoS One . 2014 ; 9 ( 8 ): e105620 . Google Scholar Crossref Search ADS PubMed 18 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 19 Blodgett J , Theou O , Kirkland S , et al. . The association between sedentary behaviour, moderate-vigorous physical activity and frailty in NHANES cohorts . Maturitas . 2015 ; 80 ( 2 ): 187 – 191 . Google Scholar Crossref Search ADS PubMed 20 da Silva Coqueiro R , de Queiroz BM , Oliveira DS , et al. . Cross-sectional relationships between sedentary behavior and frailty in older adults . J Sports Med Phys Fitness . 2017 ; 57 ( 6 ): 825 – 830 . Google Scholar PubMed 21 Del Pozo-Cruz B , Mañas A , Martín-García M , et al. . Frailty is associated with objectively assessed sedentary behaviour patterns in older adults: evidence from the Toledo Study for Healthy Aging (TSHA) . PLoS One . 2017 ; 12 ( 9 ): e0183911 . Google Scholar Crossref Search ADS PubMed 22 García-Esquinas E , Andrade E , Martínez-Gómez D , et al. . Television viewing time as a risk factor for frailty and functional limitations in older adults: results from 2 European prospective cohorts . Int J Behav Nutr Phys Act . 2017 ; 14 : 54 . Google Scholar Crossref Search ADS PubMed 23 Dobson AJ , Hockey R , Brown WJ , et al. . Cohort profile update: Australian Longitudinal Study on Women’s Health . Int J Epidemiol . 2015 ; 44 ( 5 ): 1547 – 1547f . Google Scholar Crossref Search ADS PubMed 24 Lee C , Dobson AJ , Brown WJ , et al. . Cohort profile: the Australian Longitudinal Study on Women’s Health . Int J Epidemiol . 2005 ; 34 ( 5 ): 987 – 991 . Google Scholar Crossref Search ADS PubMed 25 Brown WJ , Bryson L , Byles JE , et al. . Women’s Health Australia: recruitment for a national longitudinal cohort study . Women Health . 1998 ; 28 ( 1 ): 23 – 40 . Google Scholar Crossref Search ADS PubMed 26 Craig CL , Marshall AL , Sjostrom M , et al. . International Physical Activity Questionnaire: 12-country reliability and validity . Med Sci Sports Exerc . 2003 ; 35 ( 8 ): 1381 – 1395 . Google Scholar Crossref Search ADS PubMed 27 Rosenberg DE , Bull FC , Marshall AL , et al. . Assessment of sedentary behavior with the International Physical Activity Questionnaire . J Phys Act Health . 2008 ; 5 ( suppl 1 ): S30 – S44 . Google Scholar Crossref Search ADS PubMed 28 Peeters GM , Mishra GD , Dobson AJ , et al. . Health care costs associated with prolonged sitting and inactivity . Am J Prev Med . 2014 ; 46 ( 3 ): 265 – 272 . Google Scholar Crossref Search ADS PubMed 29 Abellan van Kan G , Rolland Y , Bergman H , et al. . The I.A.N.A. Task Force on frailty assessment of older people in clinical practice . J Nutr Health Aging . 2008 ; 12 ( 1 ): 29 – 37 . Google Scholar Crossref Search ADS PubMed 30 Abellan van Kan G , Rolland YM , Morley JE , et al. . Frailty: toward a clinical definition . J Am Med Dir Assoc . 2008 ; 9 ( 2 ): 71 – 72 . Google Scholar Crossref Search ADS PubMed 31 Ware JE Jr , Snow KK , Kosinski M , et al. . SF-36 Health Survey: Manual and Interpretation Guide . Lincoln, RI : Quality Metric Inc. ; 2000 . 32 Susanto M , Hubbard RE , Gardiner PA . Validity and responsiveness of the FRAIL scale in middle-aged women . J Am Med Dir Assoc . 2018 ; 19 ( 1 ): 65 – 69 . Google Scholar Crossref Search ADS PubMed 33 World Health Organisation . Obesity: Preventing and Managing the Global Epidemic . Geneva, Switzerland : World Health Organization ; 2004 . 34 National Health and Medical Research Council . Australian Alcohol Guidelines. Health Risks and Benefits. Canberra, ACT, Australia : National Health and Medical Research Council ; 2001 . https://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/ds9.pdf. Accessed March 11, 2017. 35 Australian Institute of Health and Welfare . The Active Australia Survey: A Guide and Manual for Implementation, Analysis and Reporting. Canberra, ACT, Australia : Australian Institute of Health and Welfare ; 2003 . https://www.aihw.gov.au/getmedia/ff25c134-5df2-45ba-b4e1-6c214ed157e6/aas.pdf.aspx?inline=true. Accessed March 11, 2017. 36 Brown WJ , Bauman AE . Comparison of estimates of population levels of physical activity using two measures . Aust N Z J Public Health . 2000 ; 24 ( 5 ): 520 – 525 . Google Scholar Crossref Search ADS PubMed 37 Nagin D . Group-Based Modeling of Development . Cambridge, MA : Harvard University Press ; 2005 . 38 Nagin DS , Odgers CL . Group-based trajectory modeling in clinical research . Annu Rev Clin Psychol . 2010 ; 6 : 109 – 138 . Google Scholar Crossref Search ADS PubMed 39 Virtuoso Júnior JS , Roza LB , Tribess S , et al. . Time spent sitting is associated with changes in biomarkers of frailty in hospitalized older adults: a cross sectional study . Front Physiol . 2017 ; 8 : 505 . Google Scholar Crossref Search ADS PubMed 40 Gale CR , Cooper C , Sayer AA . Prevalence of frailty and disability: findings from the English Longitudinal Study of Ageing . Age Ageing . 2015 ; 44 ( 1 ): 162 – 165 . Google Scholar Crossref Search ADS PubMed 41 Santos-Eggimann B , Cuénoud P , Spagnoli J , et al. . Prevalence of frailty in middle-aged and older community-dwelling Europeans living in 10 countries . J Gerontol A Biol Sci Med Sci . 2009 ; 64 ( 6 ): 675 – 681 . Google Scholar Crossref Search ADS PubMed 42 Rockwood K , Mitnitski A . Frailty defined by deficit accumulation and geriatric medicine defined by frailty . Clin Geriatr Med . 2011 ; 27 ( 1 ): 17 – 26 . Google Scholar Crossref Search ADS PubMed 43 Theou O , Blodgett JM , Godin J , et al. . Association between sedentary time and mortality across levels of frailty . CMAJ . 2017 ; 189 ( 33 ): E1056 – E1064 . Google Scholar Crossref Search ADS PubMed 44 Martin A , Fitzsimons C , Jepson R , et al. . Interventions with potential to reduce sedentary time in adults: systematic review and meta-analysis . Br J Sports Med . 2015 ; 49 ( 16 ): 1056 – 1063 . Google Scholar Crossref Search ADS PubMed 45 Rosenberg DE , Gell NM , Jones SM , et al. . The feasibility of reducing sitting time in overweight and obese older adults . Health Educ Behav . 2015 ; 42 ( 5 ): 669 – 676 . Google Scholar Crossref Search ADS PubMed 46 DiPietro L , Jin Y , Talegawkar S , et al. . The joint associations of sedentary time and physical activity with mobility disability in older people: the NIH-AARP Diet and Health Study . J Gerontol A Biol Sci Med Sci . 2018 ; 73 ( 4 ): 523 – 528 . Google Scholar Crossref Search ADS 47 McVeigh J , Smith A , Howie E , et al. . Trajectories of television watching from childhood to early adulthood and their association with body composition and mental health outcomes in young adults . PLoS One . 2016 ; 11 ( 4 ): e0152879 . Google Scholar Crossref Search ADS PubMed 48 McVeigh JA , Zhu K , Mountain J , et al. . Longitudinal trajectories of television watching across childhood and adolescence predict bone mass at age 20 years in the Raine Study . J Bone Miner Res . 2016 ; 31 ( 11 ): 2032 – 2040 . Google Scholar Crossref Search ADS PubMed 49 Reid N , Healy GN , Daly RM , et al. . Twelve-year television viewing time trajectories and physical function in older adults . Med Sci Sports Exerc . 2017 ; 49 ( 7 ): 1359 – 1365 . Google Scholar Crossref Search ADS PubMed 50 Healy GN , Clark BK , Winkler EA , et al. . Measurement of adults’ sedentary time in population-based studies . Am J Prev Med . 2011 ; 41 ( 2 ): 216 – 227 . Google Scholar Crossref Search ADS PubMed 51 Grant PM , Ryan CG , Tigbe WW , et al. . The validation of a novel activity monitor in the measurement of posture and motion during everyday activities . Br J Sports Med . 2006 ; 40 ( 12 ): 992 – 997 . Google Scholar Crossref Search ADS PubMed 52 Diaz KM , Howard VJ , Hutto B , et al. . Patterns of sedentary behavior and mortality in US middle-aged and older adults: a national cohort study . Ann Intern Med . 2017 ; 167 ( 7 ): 465 – 475 . Google Scholar Crossref Search ADS PubMed 53 Buta BJ , Walston JD , Godino JG , et al. . Frailty assessment instruments: systematic characterization of the uses and contexts of highly cited instruments . Ageing Res Rev . 2016 ; 26 : 53 – 61 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. 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. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Journal of Epidemiology Oxford University Press

Association of 12-Year Trajectories of Sitting Time With Frailty in Middle-Aged Women

Loading next page...
 
/lp/ou_press/twelve-year-trajectories-of-sitting-time-are-associated-with-frailty-fUc0J9R2Ay
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. 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/kwy111
Publisher site
See Article on Publisher Site

Abstract

Abstract Prolonged sitting time is associated with several health outcomes; limited evidence indicates associations with frailty. Our aims in this study were to identify patterns of sitting time over 12 years in middle-aged (ages 50–55 years) women and examine associations of these patterns with frailty in older age. We examined 5,462 women born in 1946–1951 from the Australian Longitudinal Study on Women’s Health who provided information on sociodemographic attributes, daily sitting time, and frailty in 2001 and then again every 3 years until 2013. Frailty was assessed using the FRAIL (fatigue, resistance, ambulation, illness, loss of weight) scale (0 = healthy; 1–2 = prefrail; 3–5 = frail), and group-based trajectory analyses identified trajectories of sitting time. We identified 5 sitting-time trajectories: low (26.9%), medium (43.1%; referent), increasing (6.9%), decreasing (18.1%), and high (4.8%). In adjusted models, the likelihoods of being frail were statistically higher for women in the increasing (odds ratio (OR) = 1.29, 95% confidence interval (CI): 1.03, 1.61) and high (OR = 1.42, 95% CI: 1.10, 1.84) trajectories. In contrast, women in the low trajectory group were less likely to be frail (OR = 0.86, 95% CI: 0.75, 0.98), and there was no difference in the likelihood of frailty in the decreasing trajectory group. Our study suggests that patterns of sitting time over 12 years in middle-aged women predict frailty in older age. aging, frailty, longitudinal cohort studies, middle age, sedentary behavior, sitting, women’s health As the populations of most developed countries continue to age, frailty is becoming more prevalent (1). A systematic review found that in community-dwelling elderly adults, frailty prevalence varies from 4.0% to 59.1% (2), with an increasing frailty pattern observed with increased age and with frailty being more prevalent in women than in men (1, 3). Frailty is defined as a syndrome that results from decline in physiological reserve to withstand stressors (1, 4) and that leads to multiple adverse health outcomes, including falls, hospitalization, institutionalization, and premature death (3, 4). Based on recent estimates, one-fifth of Australian women aged 73–78 years are frail, suggesting that it is important to examine the health and behaviors of middle-aged women in order to identify factors associated with the development of frailty (5). Sitting time is a risk factor for poor cardiometabolic health (6, 7), which is closely related to frailty (8). Prolonged sitting time has become a public health concern (9) as the modern environment encourages more sitting, whether at work, at home, or in transport (10). In 2011, Australian adults reported sitting for 5.5 hours per day, with most sitting occurring in leisure domains (61.3%) and the remainder occurring at work (25.6%) or while using transportation (13.1%) (11). In contrast, when it is measured objectively, sitting time is estimated to be 8.9 hours per day (12). While investigators have reported increases in sitting time in cohort studies and the general population (13–15), there is limited evidence on patterns of sitting over time. While prolonged sitting in adults is detrimentally associated with many health outcomes, such as all-cause, cardiovascular, and cancer mortality and incidence of cardiovascular disease, cancer, and type 2 diabetes (16, 17), there have been limited investigations of relationships between sitting time and geriatric conditions such as frailty. In several cross-sectional studies, high levels of sedentary behavior (sitting or lying down while awake with an energy expenditure less than or equal to 1.5 metabolic equivalents (METs)) (18) were associated with frailty in older adults (all aged >50 years) (19–21). In a longitudinal study of 2 cohorts of community-dwelling older adults aged ≥60 years, García-Esquinas et al. (22) reported that baseline television (TV) viewing time was also associated with frailty at follow-up. Using data from a large Australian cohort, we aimed to identify patterns of sitting time over 12 years in middle-aged women, a group at high risk of developing frailty, and to examine associations of these patterns with frailty in older age. Investigation of this relationship is important to help develop programs designed to ameliorate the health outcome of frailty. METHODS Study population and procedures The Australian Longitudinal Study on Women’s Health (http://www.alswh.org.au/), initiated in 1996, is a national longitudinal cohort study examining over 58,000 Australian women (originally) separated into 3 cohorts: those born in 1973–1978, those born in 1946–1951, and those born in 1921–1926 (23, 24). The participants were randomly selected from the Australian national Medicare health insurance database, which includes all Australian citizens and permanent residents (24, 25). It is funded by the Australian Government Department of Health (24, 25). The study was approved by ethics committees at the Universities of Queensland and Newcastle, and informed consent was collected from all participants. When they were first surveyed in 1996 (survey 1), 13,715 women born in 1946–1951 participated, with a 54% response rate (23–25). Women subsequently completed surveys in 1998 (survey 2), 2001 (survey 3), 2004 (survey 4), 2007 (survey 5), 2010 (survey 6), and 2013 (survey 7). Data on sitting time were available only from survey 3 to survey 7. Women were included in the study if they provided data on sitting time and frailty in survey 3, were not frail (i.e., had a FRAIL score (defined below) less than 3) in 2001 (survey 3), and had complete data for all confounders (survey 1 and survey 3) and frailty in 2013 (survey 7). Figure 1 shows the process of participant selection. Figure 1. View largeDownload slide Selection of participants for a study of sitting patterns and frailty in middle-aged (50–55 years) women, Australia, 2001–2013. Figure 1. View largeDownload slide Selection of participants for a study of sitting patterns and frailty in middle-aged (50–55 years) women, Australia, 2001–2013. Sitting time Data on sitting time were collected from survey 3 to survey 7 using the following 2 self-report items: “How many hours EACH DAY do you typically spend sitting down while doing things like visiting friends, driving, reading, watching television, or working at a desk or computer on 1) a usual weekday and 2) a usual weekend day?”. A similar question is part of the International Physical Activity Questionnaire (26), which has good test-retest reliability for self-report measures: a Spearman reliability coefficient greater than 0.75 and low-to-moderate Spearman correlation coefficients greater than 0.25 when validated against accelerometry (26, 27). Daily sitting time was calculated as [(number of hours/day of sitting on a weekday × 5) + (number of hours/day of sitting on a weekend day × 2)/7] (28). Frailty In all surveys from survey 3 to survey 7, frailty was assessed using the FRAIL scale (29, 30), which is based on 5 domains: fatigue, resistance, ambulation, illness, and loss of weight. Frailty was calculated using the Medical Outcomes Study 36-item Short Form Health Survey (31), the presence of chronic conditions, and weight. Scores on the FRAIL scale range from 0 to 5, with 3 possible categories: healthy (score of 0), prefrail (score of 1–2), and frail (score of 3–5) (5). Fatigue was addressed in 3 questions: “Did you feel worn out?,” “Did you feel tired?,” and “Did you have a lot of energy?”. Fatigue was recorded as present if participants answered the first and second questions as “all of the time,” “most of the time,” or “a good bit of the time” or answered the third question as “some of the time,” “a little of the time,” or “none of the time.” Resistance was addressed in the question, “Does your health now limit you in climbing one flight of stairs?”. A positive response was recorded if participants answered “limited a lot” or “limited a little.” Ambulation was defined as the ability to walk 100 m. A positive response was recorded if participants said they were “limited a lot” or “limited a little” in response to the question, “Does your health now limit you in walking 100 meters?”. Participants reported whether they had been diagnosed with or were being treated for diabetes, heart disease, hypertension, stroke, a low iron level (iron deficiency anemia), asthma, chronic obstructive pulmonary disease (i.e., bronchitis/emphysema), osteoporosis, breast cancer, cervical cancer, depression, anxiety, arthritis/rheumatism, or chronic fatigue syndrome. A positive response was recorded if participants answered “yes” to having at least 5 of the above chronic conditions. Weight was self-reported, and a positive score was recorded if participants had lost more than 5% of their weight from the previous survey. The FRAIL scale has been validated for use in longitudinal studies in this age group, where frailty significantly predicted disability, depression, and mortality over 15 years of follow-up (32). Sociodemographic and lifestyle variables Several variables were included as confounders in analyses examining associations of sitting time trajectories with frailty. Education was assessed in survey 1 as the highest educational qualification completed and was categorized into “less than high school,” “high school only,” and “more than high school.” Data for other variables were collected in survey 3. Relationship status was categorized as partnered or single. Body mass index (weight (kg)/height (m)2) was calculated from self-reported weight and height and was categorized as underweight/healthy (<25.0), overweight (25.0–29.9), or obese (≥30.0) (33). Smoking status was classified as never smoker, ex-smoker, or current smoker. Alcohol consumption was classified into 3 categories: nondrinker, rare drinker (<1 drink per week), and low- to high-risk drinker (≥1 drinks per week) (34). Physical activity was assessed using the Active Australia Survey, where weekly frequency and duration of brisk walking and moderate and vigorous leisure-time physical activity were reported in bouts of 10 minutes or more (35). A physical activity score in MET-minutes/week was calculated as the sum of the products of total weekly minutes spent in each of the 3 categories of physical activity [(walking minutes × 3.0 METs) + (moderate-intensity physical activity minutes × 4.0 METs) + (vigorous-intensity physical activity minutes × 7.5 METs)] (36). Employment status was categorized on the basis of hours of paid employment: full-time (≥35 hours/week), part-time (1–34 hours/week), or other (i.e., home duties/not in paid workforce). Geographic location was based on the Accessibility/Remoteness Index of Australia and was categorized as major city, inner regional area, or outer regional/remote area. Country of birth was categorized as Australia, other English-speaking country, or other. Statistical analyses Differences in characteristics between women in the trajectory groups were assessed using 1-way analysis of variance (continuous normally distributed variables), the Kruskal-Wallis test (continuous non–normally distributed variables), and χ2 analyses (categorical variables). Group-based trajectory modeling using a censored normal model across surveys 3–7 was used to identify patterns in sitting over time (37). The final choice of the trajectory models was based on the Bayesian Information Criterion (BIC) and the log Bayes factor (2 × ΔBIC), an average posterior probability of group membership for all participants greater than 70%, and reasonably tight confidence intervals around the trajectory groups (37, 38). Ordinal logistic regression models were used to examine associations of sitting time trajectories with frailty at the time of survey 7. Robust variance estimates were used to account for repeated measures in individuals. To select confounders for these models, ordinal logistic regression models were used to examine associations of each variable with sitting time trajectories and also with frailty at survey 7. Relationship status, education, body mass index, alcohol consumption, smoking status, physical activity, employment status, and the presence of arthritis, depression, and hypertension were associated with both trajectories and frailty (P < 0.1) and were included as confounders in the final model. Logistic regression models, adjusting for the confounders above, were used to examine associations of sitting time trajectories with FRAIL components. Proportions of women who were frail in each trajectory group at the time of each survey and had deficits for each FRAIL component at surveys 3 and 7 were calculated. Differences in the proportions of women reporting deficits in components of the FRAIL scale across trajectory groups at surveys 3 and 7 were examined using χ2 analyses, and McNemar tests were used to examine differences within trajectory groups for deficits between survey 3 and survey 7. All analyses were conducted in STATA (version 14.1; StataCorp LLC, College Station, Texas). RESULTS Of the 6,298 women (aged 62–67 years) who completed survey 7 and were not frail at survey 3 (i.e., FRAIL score <3), 5,462 were included in this study. At survey 7, a total of 396 (7.3%) women were classified as frail (i.e., FRAIL score >2). Characteristics of these women are shown in Table 1. The mean age at survey 3 of women who were included in the study was 52.5 years. The participants were more likely to be partnered, to have less than a high school education at survey 1, to be underweight/healthy as per their body mass index classification, to have never smoked, to be a low- to high-risk alcohol drinker, to have an active level (≥600 MET-minutes/week) of physical activity, to be employed full-time, to live in an inner regional area, and to have been born in Australia. Table 1. Baseline Characteristics of Middle-Aged (50–55 Years) Women Included in Analyses of Sitting Time Trajectories and Frailty (n = 5,462), Australia, 2001a Characteristic Total (n = 5,462) Sitting Time Trajectory Group Low (n = 1,472) Medium (n = 2,359) Increasing (n = 379) Decreasing (n = 989) High (n = 263) P Value Median (IQR) %b Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Age, yearsc 52.5 (1.5) 52.5 (1.5) 52.5 (1.5) 52.4 (1.4) 52.5 (1.5) 52.5 (1.5) 0.553 Sitting time, hours/day  Survey 3 (2001) 5.1 (3.6–7.1) 3.0 (2.3–4.0) 5.0 (4.0–6.3) 5.7 (4.6–7.0) 8.1 (7.0–9.4) 9.7 (8.1–11.3) <0.001  Survey 4 (2004) 5.3 (4.0–7.4) 3.3 (2.6–4.1) 5.3 (4.4–6.4) 6.6 (5.3–8.0) 8.3 (7.0–9.4) 10.3 (9.3–11.6) <0.001  Survey 5 (2007) 5.8 (4.0–7.7) 3.6 (2.7–4.3) 5.6 (4.6–6.6) 7.9 (6.6–9.0) 8.1 (6.9–9.4) 10.6 (9.4–12.0) <0.001  Survey 6 (2010) 5.4 (4.0–7.4) 3.7 (3.0–4.4) 5.4 (4.6–6.4) 8.4 (7.3–10.0) 7.7 (6.0–8.9) 10.6 (9.4–12.0) <0.001  Survey 7 (2013) 5.3 (4.0–7.0) 3.7 (3.0–4.6) 5.3 (4.3–6.3) 9.6 (8.6–10.6) 6.1 (5.0–7.7) 10.0 (8.3–11.7) <0.001 Relationship status <0.001  Partnered 83.9 87.4 85.6 79.7 79.8 70.7  Single 16.1 12.6 14.4 20.3 20.2 29.3 Education at survey 1 (1996) 0.551  Less than high school 43.7 45.2 43.5 44.9 42.3 40.3  High school 16.8 17.2 16.3 17.4 17.6 15.6  More than high school 39.5 37.6 40.3 37.7 40.1 44.1 Body mass index categoryd <0.001  Underweight/healthy (<25.0) 46.7 53.6 46.5 42.0 42.9 31.2  Overweight (25.0–29.9) 32.8 31.9 34.5 29.3 31.9 30.8  Obese (≥30.0) 20.6 14.5 19.1 28.8 25.3 38.0 Smoking status <0.001  Never smoker 63.8 67.6 63.7 61.5 61.7 54.0  Ex-smoker 24.8 21.5 25.4 26.1 26.9 28.5  Current smoker 11.4 10.9 10.9 12.4 11.4 17.5 Alcohol consumption <0.001  Never drinker 10.4 11.8 9.8 10.6 9.9 9.1  Rare drinker (<1 drink per week) 26.1 27.9 25.8 34.0 21.5 25.1  Low- to high-risk drinker (≥1 drinks per week) 63.5 60.4 64.4 55.4 68.6 65.8 Physical activity, MET-minutes/week 540 (180–1,260) 635 (180–1,406) 600 (210–1,260) 480 (135–1,170) 480 (180–1,080) 270 (60–810) <0.001 Physical activity category <0.001  None (<40.0 MET-minutes/week) 13.7 12.4 12.1 19.5 15.2 20.9  Low (40.0–599.9 MET-minutes/week) 37.6 34.4 37.0 36.7 41.1 48.3  Active (≥600.0 MET-minutes/week) 48.8 53.2 51.0 43.8 43.8 30.8 Employment status <0.001  Full-time 36.0 27.6 33.0 37.7 49.8 55.1  Part-time 34.5 38.5 35.4 36.4 28.6 23.6  Other 29.5 33.9 31.5 25.9 21.6 21.3 Area of residence <0.001  Major city 34.6 28.5 33.2 34.9 43.8 47.2  Inner regional area 42.0 45.5 42.6 41.8 36.2 38.0  Outer regional/remote area 23.4 26.0 24.2 23.3 20.0 14.8 Country of birth <0.001  Australia 77.3 80.3 76.7 75.7 76.8 69.6  Other English-speaking country 15.3 11.0 16.4 17.2 16.0 23.2  Other 7.5 8.7 6.9 7.1 7.2 7.2 Frailty status <0.001  Healthy (FRAIL score = 0) 44.3 46.6 46.7 38.3 41.5 29.7  Prefrail (FRAIL score = 1–2) 55.7 53.4 53.3 61.7 58.5 70.3 Characteristic Total (n = 5,462) Sitting Time Trajectory Group Low (n = 1,472) Medium (n = 2,359) Increasing (n = 379) Decreasing (n = 989) High (n = 263) P Value Median (IQR) %b Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Age, yearsc 52.5 (1.5) 52.5 (1.5) 52.5 (1.5) 52.4 (1.4) 52.5 (1.5) 52.5 (1.5) 0.553 Sitting time, hours/day  Survey 3 (2001) 5.1 (3.6–7.1) 3.0 (2.3–4.0) 5.0 (4.0–6.3) 5.7 (4.6–7.0) 8.1 (7.0–9.4) 9.7 (8.1–11.3) <0.001  Survey 4 (2004) 5.3 (4.0–7.4) 3.3 (2.6–4.1) 5.3 (4.4–6.4) 6.6 (5.3–8.0) 8.3 (7.0–9.4) 10.3 (9.3–11.6) <0.001  Survey 5 (2007) 5.8 (4.0–7.7) 3.6 (2.7–4.3) 5.6 (4.6–6.6) 7.9 (6.6–9.0) 8.1 (6.9–9.4) 10.6 (9.4–12.0) <0.001  Survey 6 (2010) 5.4 (4.0–7.4) 3.7 (3.0–4.4) 5.4 (4.6–6.4) 8.4 (7.3–10.0) 7.7 (6.0–8.9) 10.6 (9.4–12.0) <0.001  Survey 7 (2013) 5.3 (4.0–7.0) 3.7 (3.0–4.6) 5.3 (4.3–6.3) 9.6 (8.6–10.6) 6.1 (5.0–7.7) 10.0 (8.3–11.7) <0.001 Relationship status <0.001  Partnered 83.9 87.4 85.6 79.7 79.8 70.7  Single 16.1 12.6 14.4 20.3 20.2 29.3 Education at survey 1 (1996) 0.551  Less than high school 43.7 45.2 43.5 44.9 42.3 40.3  High school 16.8 17.2 16.3 17.4 17.6 15.6  More than high school 39.5 37.6 40.3 37.7 40.1 44.1 Body mass index categoryd <0.001  Underweight/healthy (<25.0) 46.7 53.6 46.5 42.0 42.9 31.2  Overweight (25.0–29.9) 32.8 31.9 34.5 29.3 31.9 30.8  Obese (≥30.0) 20.6 14.5 19.1 28.8 25.3 38.0 Smoking status <0.001  Never smoker 63.8 67.6 63.7 61.5 61.7 54.0  Ex-smoker 24.8 21.5 25.4 26.1 26.9 28.5  Current smoker 11.4 10.9 10.9 12.4 11.4 17.5 Alcohol consumption <0.001  Never drinker 10.4 11.8 9.8 10.6 9.9 9.1  Rare drinker (<1 drink per week) 26.1 27.9 25.8 34.0 21.5 25.1  Low- to high-risk drinker (≥1 drinks per week) 63.5 60.4 64.4 55.4 68.6 65.8 Physical activity, MET-minutes/week 540 (180–1,260) 635 (180–1,406) 600 (210–1,260) 480 (135–1,170) 480 (180–1,080) 270 (60–810) <0.001 Physical activity category <0.001  None (<40.0 MET-minutes/week) 13.7 12.4 12.1 19.5 15.2 20.9  Low (40.0–599.9 MET-minutes/week) 37.6 34.4 37.0 36.7 41.1 48.3  Active (≥600.0 MET-minutes/week) 48.8 53.2 51.0 43.8 43.8 30.8 Employment status <0.001  Full-time 36.0 27.6 33.0 37.7 49.8 55.1  Part-time 34.5 38.5 35.4 36.4 28.6 23.6  Other 29.5 33.9 31.5 25.9 21.6 21.3 Area of residence <0.001  Major city 34.6 28.5 33.2 34.9 43.8 47.2  Inner regional area 42.0 45.5 42.6 41.8 36.2 38.0  Outer regional/remote area 23.4 26.0 24.2 23.3 20.0 14.8 Country of birth <0.001  Australia 77.3 80.3 76.7 75.7 76.8 69.6  Other English-speaking country 15.3 11.0 16.4 17.2 16.0 23.2  Other 7.5 8.7 6.9 7.1 7.2 7.2 Frailty status <0.001  Healthy (FRAIL score = 0) 44.3 46.6 46.7 38.3 41.5 29.7  Prefrail (FRAIL score = 1–2) 55.7 53.4 53.3 61.7 58.5 70.3 Abbreviations: FRAIL, fatigue, resistance, ambulation, illness, loss of weight; IQR, interquartile range; MET, metabolic equivalent. a Data are from survey 3, except where indicated. b Percentages may add up to more than 100% because of rounding. c Values are expressed as mean (standard deviation). d Weight (kg)/height (m)2. Table 1. Baseline Characteristics of Middle-Aged (50–55 Years) Women Included in Analyses of Sitting Time Trajectories and Frailty (n = 5,462), Australia, 2001a Characteristic Total (n = 5,462) Sitting Time Trajectory Group Low (n = 1,472) Medium (n = 2,359) Increasing (n = 379) Decreasing (n = 989) High (n = 263) P Value Median (IQR) %b Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Age, yearsc 52.5 (1.5) 52.5 (1.5) 52.5 (1.5) 52.4 (1.4) 52.5 (1.5) 52.5 (1.5) 0.553 Sitting time, hours/day  Survey 3 (2001) 5.1 (3.6–7.1) 3.0 (2.3–4.0) 5.0 (4.0–6.3) 5.7 (4.6–7.0) 8.1 (7.0–9.4) 9.7 (8.1–11.3) <0.001  Survey 4 (2004) 5.3 (4.0–7.4) 3.3 (2.6–4.1) 5.3 (4.4–6.4) 6.6 (5.3–8.0) 8.3 (7.0–9.4) 10.3 (9.3–11.6) <0.001  Survey 5 (2007) 5.8 (4.0–7.7) 3.6 (2.7–4.3) 5.6 (4.6–6.6) 7.9 (6.6–9.0) 8.1 (6.9–9.4) 10.6 (9.4–12.0) <0.001  Survey 6 (2010) 5.4 (4.0–7.4) 3.7 (3.0–4.4) 5.4 (4.6–6.4) 8.4 (7.3–10.0) 7.7 (6.0–8.9) 10.6 (9.4–12.0) <0.001  Survey 7 (2013) 5.3 (4.0–7.0) 3.7 (3.0–4.6) 5.3 (4.3–6.3) 9.6 (8.6–10.6) 6.1 (5.0–7.7) 10.0 (8.3–11.7) <0.001 Relationship status <0.001  Partnered 83.9 87.4 85.6 79.7 79.8 70.7  Single 16.1 12.6 14.4 20.3 20.2 29.3 Education at survey 1 (1996) 0.551  Less than high school 43.7 45.2 43.5 44.9 42.3 40.3  High school 16.8 17.2 16.3 17.4 17.6 15.6  More than high school 39.5 37.6 40.3 37.7 40.1 44.1 Body mass index categoryd <0.001  Underweight/healthy (<25.0) 46.7 53.6 46.5 42.0 42.9 31.2  Overweight (25.0–29.9) 32.8 31.9 34.5 29.3 31.9 30.8  Obese (≥30.0) 20.6 14.5 19.1 28.8 25.3 38.0 Smoking status <0.001  Never smoker 63.8 67.6 63.7 61.5 61.7 54.0  Ex-smoker 24.8 21.5 25.4 26.1 26.9 28.5  Current smoker 11.4 10.9 10.9 12.4 11.4 17.5 Alcohol consumption <0.001  Never drinker 10.4 11.8 9.8 10.6 9.9 9.1  Rare drinker (<1 drink per week) 26.1 27.9 25.8 34.0 21.5 25.1  Low- to high-risk drinker (≥1 drinks per week) 63.5 60.4 64.4 55.4 68.6 65.8 Physical activity, MET-minutes/week 540 (180–1,260) 635 (180–1,406) 600 (210–1,260) 480 (135–1,170) 480 (180–1,080) 270 (60–810) <0.001 Physical activity category <0.001  None (<40.0 MET-minutes/week) 13.7 12.4 12.1 19.5 15.2 20.9  Low (40.0–599.9 MET-minutes/week) 37.6 34.4 37.0 36.7 41.1 48.3  Active (≥600.0 MET-minutes/week) 48.8 53.2 51.0 43.8 43.8 30.8 Employment status <0.001  Full-time 36.0 27.6 33.0 37.7 49.8 55.1  Part-time 34.5 38.5 35.4 36.4 28.6 23.6  Other 29.5 33.9 31.5 25.9 21.6 21.3 Area of residence <0.001  Major city 34.6 28.5 33.2 34.9 43.8 47.2  Inner regional area 42.0 45.5 42.6 41.8 36.2 38.0  Outer regional/remote area 23.4 26.0 24.2 23.3 20.0 14.8 Country of birth <0.001  Australia 77.3 80.3 76.7 75.7 76.8 69.6  Other English-speaking country 15.3 11.0 16.4 17.2 16.0 23.2  Other 7.5 8.7 6.9 7.1 7.2 7.2 Frailty status <0.001  Healthy (FRAIL score = 0) 44.3 46.6 46.7 38.3 41.5 29.7  Prefrail (FRAIL score = 1–2) 55.7 53.4 53.3 61.7 58.5 70.3 Characteristic Total (n = 5,462) Sitting Time Trajectory Group Low (n = 1,472) Medium (n = 2,359) Increasing (n = 379) Decreasing (n = 989) High (n = 263) P Value Median (IQR) %b Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Median (IQR) % Age, yearsc 52.5 (1.5) 52.5 (1.5) 52.5 (1.5) 52.4 (1.4) 52.5 (1.5) 52.5 (1.5) 0.553 Sitting time, hours/day  Survey 3 (2001) 5.1 (3.6–7.1) 3.0 (2.3–4.0) 5.0 (4.0–6.3) 5.7 (4.6–7.0) 8.1 (7.0–9.4) 9.7 (8.1–11.3) <0.001  Survey 4 (2004) 5.3 (4.0–7.4) 3.3 (2.6–4.1) 5.3 (4.4–6.4) 6.6 (5.3–8.0) 8.3 (7.0–9.4) 10.3 (9.3–11.6) <0.001  Survey 5 (2007) 5.8 (4.0–7.7) 3.6 (2.7–4.3) 5.6 (4.6–6.6) 7.9 (6.6–9.0) 8.1 (6.9–9.4) 10.6 (9.4–12.0) <0.001  Survey 6 (2010) 5.4 (4.0–7.4) 3.7 (3.0–4.4) 5.4 (4.6–6.4) 8.4 (7.3–10.0) 7.7 (6.0–8.9) 10.6 (9.4–12.0) <0.001  Survey 7 (2013) 5.3 (4.0–7.0) 3.7 (3.0–4.6) 5.3 (4.3–6.3) 9.6 (8.6–10.6) 6.1 (5.0–7.7) 10.0 (8.3–11.7) <0.001 Relationship status <0.001  Partnered 83.9 87.4 85.6 79.7 79.8 70.7  Single 16.1 12.6 14.4 20.3 20.2 29.3 Education at survey 1 (1996) 0.551  Less than high school 43.7 45.2 43.5 44.9 42.3 40.3  High school 16.8 17.2 16.3 17.4 17.6 15.6  More than high school 39.5 37.6 40.3 37.7 40.1 44.1 Body mass index categoryd <0.001  Underweight/healthy (<25.0) 46.7 53.6 46.5 42.0 42.9 31.2  Overweight (25.0–29.9) 32.8 31.9 34.5 29.3 31.9 30.8  Obese (≥30.0) 20.6 14.5 19.1 28.8 25.3 38.0 Smoking status <0.001  Never smoker 63.8 67.6 63.7 61.5 61.7 54.0  Ex-smoker 24.8 21.5 25.4 26.1 26.9 28.5  Current smoker 11.4 10.9 10.9 12.4 11.4 17.5 Alcohol consumption <0.001  Never drinker 10.4 11.8 9.8 10.6 9.9 9.1  Rare drinker (<1 drink per week) 26.1 27.9 25.8 34.0 21.5 25.1  Low- to high-risk drinker (≥1 drinks per week) 63.5 60.4 64.4 55.4 68.6 65.8 Physical activity, MET-minutes/week 540 (180–1,260) 635 (180–1,406) 600 (210–1,260) 480 (135–1,170) 480 (180–1,080) 270 (60–810) <0.001 Physical activity category <0.001  None (<40.0 MET-minutes/week) 13.7 12.4 12.1 19.5 15.2 20.9  Low (40.0–599.9 MET-minutes/week) 37.6 34.4 37.0 36.7 41.1 48.3  Active (≥600.0 MET-minutes/week) 48.8 53.2 51.0 43.8 43.8 30.8 Employment status <0.001  Full-time 36.0 27.6 33.0 37.7 49.8 55.1  Part-time 34.5 38.5 35.4 36.4 28.6 23.6  Other 29.5 33.9 31.5 25.9 21.6 21.3 Area of residence <0.001  Major city 34.6 28.5 33.2 34.9 43.8 47.2  Inner regional area 42.0 45.5 42.6 41.8 36.2 38.0  Outer regional/remote area 23.4 26.0 24.2 23.3 20.0 14.8 Country of birth <0.001  Australia 77.3 80.3 76.7 75.7 76.8 69.6  Other English-speaking country 15.3 11.0 16.4 17.2 16.0 23.2  Other 7.5 8.7 6.9 7.1 7.2 7.2 Frailty status <0.001  Healthy (FRAIL score = 0) 44.3 46.6 46.7 38.3 41.5 29.7  Prefrail (FRAIL score = 1–2) 55.7 53.4 53.3 61.7 58.5 70.3 Abbreviations: FRAIL, fatigue, resistance, ambulation, illness, loss of weight; IQR, interquartile range; MET, metabolic equivalent. a Data are from survey 3, except where indicated. b Percentages may add up to more than 100% because of rounding. c Values are expressed as mean (standard deviation). d Weight (kg)/height (m)2. As shown in Figure 2, 5 participant clusters representing sitting time trajectories were identified: low (n = 1,472 of 5,462; 26.9%); medium (n = 2,359 of 5,462; 43.1% (referent)), increasing (n = 379 of 5,462; 6.9%), decreasing (n = 989 of 5,462; 18.1%), and high (n = 263 of 5,462; 4.8%). As shown in Table 1, there were no differences in age or education across the 5 trajectory groups. Compared with the other 4 trajectory groups, the high trajectory group had more single women, obese women, current smokers, women in full-time employment, women living in major cities, women born in other English-speaking countries, and women with low physical activity. Figure 2. View largeDownload slide Sitting time trajectories observed over the course of 12 years among middle-aged (50–55 years) women, Australia, 2001–2013. Percentage of participants in each trajectory group: low, 26.9%; medium, 43.1%; increasing, 6.9%; decreasing, 18.1%; high, 4.8%. Figure 2. View largeDownload slide Sitting time trajectories observed over the course of 12 years among middle-aged (50–55 years) women, Australia, 2001–2013. Percentage of participants in each trajectory group: low, 26.9%; medium, 43.1%; increasing, 6.9%; decreasing, 18.1%; high, 4.8%. In fully adjusted models, these trajectory groups were associated with frailty at survey 7. The medium trajectory was used as a reference category, since it had the largest number of participants. Compared with women in the medium trajectory group, women in the increasing and high trajectories were more likely to be frail versus prefrail and healthy, with odds ratios of 1.29 (95% confidence interval (CI): 1.03, 1.61) and 1.42 (95% CI: 1.10, 1.84), respectively, with those in the low trajectory group being less likely to be frail (odds ratio = 0.86, 95% CI: 0.75, 0.98) (Table 2). The decreasing trajectory group, however, showed no difference in the risk of frailty at survey 7. Table 2. Distribution of Participants by Frailty Status and Odds of Frailty According to 12-Year Sitting Trajectory (Ordinal Logistic Regression Analysisa) Among Middle-Aged (50–55 Years) Women, Australia, 2001–2013 Trajectory Group Frailty Statusb and No. of Participants Odds of Frailty Healthy (n = 2,500) Prefrail (n = 2,566) Frail (n = 396) OR 95% CI Medium 1,087 1,114 158 1.00 Referent Low 755 639 78 0.86c 0.75, 0.98 Increasing 149 184 46 1.29c 1.03, 1.61 Decreasing 425 479 85 1.14 0.98, 1.32 High 84 150 29 1.42d 1.10, 1.84 Trajectory Group Frailty Statusb and No. of Participants Odds of Frailty Healthy (n = 2,500) Prefrail (n = 2,566) Frail (n = 396) OR 95% CI Medium 1,087 1,114 158 1.00 Referent Low 755 639 78 0.86c 0.75, 0.98 Increasing 149 184 46 1.29c 1.03, 1.61 Decreasing 425 479 85 1.14 0.98, 1.32 High 84 150 29 1.42d 1.10, 1.84 Abbreviations: CI, confidence interval; FRAIL, fatigue, resistance, ambulation, illness, loss of weight; OR, odds ratio. a Models adjusted for relationship status, education, body mass index, smoking status, alcohol consumption, physical activity, employment, and the presence of arthritis, depression, or hypertension. b Healthy: FRAIL score = 0; prefrail: FRAIL score = 1–2; frail: FRAIL score = 3–5. cP < 0.05. dP < 0.01. Table 2. Distribution of Participants by Frailty Status and Odds of Frailty According to 12-Year Sitting Trajectory (Ordinal Logistic Regression Analysisa) Among Middle-Aged (50–55 Years) Women, Australia, 2001–2013 Trajectory Group Frailty Statusb and No. of Participants Odds of Frailty Healthy (n = 2,500) Prefrail (n = 2,566) Frail (n = 396) OR 95% CI Medium 1,087 1,114 158 1.00 Referent Low 755 639 78 0.86c 0.75, 0.98 Increasing 149 184 46 1.29c 1.03, 1.61 Decreasing 425 479 85 1.14 0.98, 1.32 High 84 150 29 1.42d 1.10, 1.84 Trajectory Group Frailty Statusb and No. of Participants Odds of Frailty Healthy (n = 2,500) Prefrail (n = 2,566) Frail (n = 396) OR 95% CI Medium 1,087 1,114 158 1.00 Referent Low 755 639 78 0.86c 0.75, 0.98 Increasing 149 184 46 1.29c 1.03, 1.61 Decreasing 425 479 85 1.14 0.98, 1.32 High 84 150 29 1.42d 1.10, 1.84 Abbreviations: CI, confidence interval; FRAIL, fatigue, resistance, ambulation, illness, loss of weight; OR, odds ratio. a Models adjusted for relationship status, education, body mass index, smoking status, alcohol consumption, physical activity, employment, and the presence of arthritis, depression, or hypertension. b Healthy: FRAIL score = 0; prefrail: FRAIL score = 1–2; frail: FRAIL score = 3–5. cP < 0.05. dP < 0.01. Table 3 shows the relationship of the 5 components of the FRAIL scale with each trajectory group. Following adjustment for the confounding factors, compared with the medium trajectory, women in the increasing and high trajectories had 25% and 56% increased likelihoods of having fatigue, respectively, while women in the low trajectory group had a 16% decreased likelihood of having fatigue. Compared with women in the medium trajectory, women in the increasing trajectory group had a 47% increased likelihood of having a deficit in resistance. However, no association was observed with the other 3 components. Table 3. Association of 12-Year Sitting Trajectory With Components of the FRAIL Scale (Logistic Regression Analysisa) Among Middle-Aged (50–55 Years) Women, Australia, 2001–2013 Trajectory Group Component of FRAIL Scale Fatigue Resistance Ambulation Illness Loss of Weight OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI Medium 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent Low 0.84b 0.73, 0.96 0.97 0.80, 1.16 1.04 0.79, 1.37 0.73 0.44, 1.21 0.89 0.73, 1.08 Increasing 1.25b 1.00, 1.57 1.47c 1.12, 1.92 1.36 0.91, 2.02 1.34 0.71, 2.54 0.98 0.71, 1.33 Decreasing 1.10 0.94, 1.29 1.00 0.81, 1.23 1.26 0.94, 1.69 0.94 0.56, 1.57 1.10 0.89, 1.36 High 1.56c 1.19, 2.04 1.10 0.79, 1.53 1.05 0.65, 1.69 1.16 0.53, 2.50 1.24 0.88, 1.74 Trajectory Group Component of FRAIL Scale Fatigue Resistance Ambulation Illness Loss of Weight OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI Medium 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent Low 0.84b 0.73, 0.96 0.97 0.80, 1.16 1.04 0.79, 1.37 0.73 0.44, 1.21 0.89 0.73, 1.08 Increasing 1.25b 1.00, 1.57 1.47c 1.12, 1.92 1.36 0.91, 2.02 1.34 0.71, 2.54 0.98 0.71, 1.33 Decreasing 1.10 0.94, 1.29 1.00 0.81, 1.23 1.26 0.94, 1.69 0.94 0.56, 1.57 1.10 0.89, 1.36 High 1.56c 1.19, 2.04 1.10 0.79, 1.53 1.05 0.65, 1.69 1.16 0.53, 2.50 1.24 0.88, 1.74 Abbreviations: CI, confidence interval; FRAIL, fatigue, resistance, ambulation, illness, loss of weight; OR, odds ratio. a Models adjusted for relationship status, education, body mass index, smoking status, alcohol consumption, physical activity, employment, and the presence of arthritis, depression, or hypertension. bP < 0.05. cP < 0.01. Table 3. Association of 12-Year Sitting Trajectory With Components of the FRAIL Scale (Logistic Regression Analysisa) Among Middle-Aged (50–55 Years) Women, Australia, 2001–2013 Trajectory Group Component of FRAIL Scale Fatigue Resistance Ambulation Illness Loss of Weight OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI Medium 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent Low 0.84b 0.73, 0.96 0.97 0.80, 1.16 1.04 0.79, 1.37 0.73 0.44, 1.21 0.89 0.73, 1.08 Increasing 1.25b 1.00, 1.57 1.47c 1.12, 1.92 1.36 0.91, 2.02 1.34 0.71, 2.54 0.98 0.71, 1.33 Decreasing 1.10 0.94, 1.29 1.00 0.81, 1.23 1.26 0.94, 1.69 0.94 0.56, 1.57 1.10 0.89, 1.36 High 1.56c 1.19, 2.04 1.10 0.79, 1.53 1.05 0.65, 1.69 1.16 0.53, 2.50 1.24 0.88, 1.74 Trajectory Group Component of FRAIL Scale Fatigue Resistance Ambulation Illness Loss of Weight OR 95% CI OR 95% CI OR 95% CI OR 95% CI OR 95% CI Medium 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent 1.00 Referent Low 0.84b 0.73, 0.96 0.97 0.80, 1.16 1.04 0.79, 1.37 0.73 0.44, 1.21 0.89 0.73, 1.08 Increasing 1.25b 1.00, 1.57 1.47c 1.12, 1.92 1.36 0.91, 2.02 1.34 0.71, 2.54 0.98 0.71, 1.33 Decreasing 1.10 0.94, 1.29 1.00 0.81, 1.23 1.26 0.94, 1.69 0.94 0.56, 1.57 1.10 0.89, 1.36 High 1.56c 1.19, 2.04 1.10 0.79, 1.53 1.05 0.65, 1.69 1.16 0.53, 2.50 1.24 0.88, 1.74 Abbreviations: CI, confidence interval; FRAIL, fatigue, resistance, ambulation, illness, loss of weight; OR, odds ratio. a Models adjusted for relationship status, education, body mass index, smoking status, alcohol consumption, physical activity, employment, and the presence of arthritis, depression, or hypertension. bP < 0.05. cP < 0.01. Figure 3 shows the difference in numbers of women reporting deficits in each component of the FRAIL scale in surveys 3 and 7 for each trajectory group. At the time of survey 3, there were differences across trajectory groups in the proportions of women with deficits in fatigue, resistance, and loss of weight and differences across all trajectory groups for all components at survey 7. From survey 3 to survey 7, there was an increase in the proportion of women reporting deficits in all trajectory groups for the resistance, ambulation, and illness components; a decrease in most of the categories of women reporting deficits in fatigue; and an increase in most of the trajectory groups for the loss-of-weight component. As Web Figure 1 (available at https://academic.oup.com/aje) shows, the proportion of women who were frail increased in all trajectory groups, with the highest proportion of frailty being observed in the high (11.0%) and increasing (12.1%) trajectory groups at survey 7. In all surveys, there were lower proportions of healthy women in the high trajectory group than in the other trajectory groups. Figure 3. View largeDownload slide Proportions of middle-aged (50–55 years) women with deficits in components of the FRAIL scale at survey 3 and survey 7, by sitting time trajectory group, Australia, 2001–2013. McNemar’s test was used to assess differences within trajectory group at surveys 3 and 7 for all components of the FRAIL scale. The proportion of women with a deficit in the fatigue component decreased from survey 3 to survey 7 for women in the low, medium, decreasing, and high trajectory groups. The proportions of women with deficits in the resistance, ambulation, and illness components increased from survey 3 to survey 7 in all trajectory groups. The proportion of women with a deficit in the loss-of-weight component increased from survey 3 to survey 7 for women in the medium, decreasing, and high trajectory groups. χ2 analysis was used to assess differences across trajectory groups at surveys 3 and 7. There were significant differences in the proportions of women with deficits in fatigue, resistance, and loss of weight across trajectory groups at survey 3. There was a significant difference in the proportion of women with deficits in all FRAIL scale components across trajectory groups at survey 7. FRAIL, fatigue, resistance, ambulation, illness, loss of weight. Figure 3. View largeDownload slide Proportions of middle-aged (50–55 years) women with deficits in components of the FRAIL scale at survey 3 and survey 7, by sitting time trajectory group, Australia, 2001–2013. McNemar’s test was used to assess differences within trajectory group at surveys 3 and 7 for all components of the FRAIL scale. The proportion of women with a deficit in the fatigue component decreased from survey 3 to survey 7 for women in the low, medium, decreasing, and high trajectory groups. The proportions of women with deficits in the resistance, ambulation, and illness components increased from survey 3 to survey 7 in all trajectory groups. The proportion of women with a deficit in the loss-of-weight component increased from survey 3 to survey 7 for women in the medium, decreasing, and high trajectory groups. χ2 analysis was used to assess differences across trajectory groups at surveys 3 and 7. There were significant differences in the proportions of women with deficits in fatigue, resistance, and loss of weight across trajectory groups at survey 3. There was a significant difference in the proportion of women with deficits in all FRAIL scale components across trajectory groups at survey 7. FRAIL, fatigue, resistance, ambulation, illness, loss of weight. DISCUSSION In this 12-year longitudinal study of middle-aged Australian women, 5 distinct trajectory patterns of sitting time were identified. These patterns of sitting time from ages 50–55 years to ages 62–67 years were associated with frailty, such that women with unfavorable patterns of sitting were more likely to be frail at the end of the study. In contrast, the low sitting time pattern reduced the likelihood of being frail. Women in the decreasing trajectory group had no difference in the likelihood of being frail compared with women with the moderate sitting pattern, despite having higher levels of sitting at baseline. These findings are consistent with previous studies (19–22). Previous cross-sectional studies have found that people with higher levels of sedentary behavior had a higher frailty index score (19) or were frail according to modified Fried criteria (20) or the Frailty Trait Scale (21), with García-Esquinas et al. reporting in a longitudinal study that TV viewing time predicted frailty (Fried’s criteria) at follow-up (22). Furthermore, in one cross-sectional study, Virtuoso Júnior et al. (39) reported that sitting time was an independent indicator of frailty as identified by biomarkers (C-reactive protein and white blood cell count) in hospitalized adults aged ≥60 years. In this study, 4.9% of women aged 50–55 years were frail, which is consistent with previous estimates (40, 41). Investigators have reported an exponential increase in frailty prevalence, from 6.5% in people aged 60–69 years to 65% in people aged >90 years (40), and increases in prevalence of 3% per annum starting from late middle age (42), suggesting that interventions designed to prevent frailty should target persons in middle age. The present study suggests that sitting time is a novel risk factor for the development of frailty. Objectively measured sedentary behavior is associated with mortality in inactive vulnerable or frail people (frailty index score >0.1) (43). Therefore, interventions targeting sitting time may benefit those who are frail. In a meta-analysis, Martin et al. (44) reported that sedentary behavior interventions reduced sitting time by 42 minutes per day, suggesting that reductions in sitting time are feasible and effective. While no researchers implementing a sitting time intervention have reported on frailty outcomes, Rosenberg et al. (45) observed an increase in gait speed (0.5 seconds faster in completing a 3-m walk test from preintervention to postintervention) following a behavioral sitting-time reduction program. It is important to also consider the context in which sedentary time is accumulated. García-Esquinas et al. reported that TV viewing time was associated with frailty but other sedentary behaviors, such as reading, computer/Internet use, listening to music, or time spent in transportation, were not (22). Assessing different domains of sitting (i.e., occupation, transport, leisure) and specific contexts (e.g., TV-watching or using a computer) could potentially assist in developing more targeted interventions. However, these types of data were only available at survey 6, and therefore trajectories of domain- or context-specific sitting could not be investigated. The findings related to the fatigue component of the FRAIL scale were similar to those for frailty. The fatigue component has a stronger correlation with total FRAIL score (Spearman’s ρ = 0.82) than other components (Spearman’s ρ’s ranging from 0.13 to 0.57) (32). This may be due to the method used to operationalize the FRAIL scale in our study: using 3 items from Short Form 36 as compared with 1 item for the other phenotype components of the scale (i.e., resistance and ambulation). It is possible that fatigue may influence sitting time, and this needs further investigation. However, the proportions of women with deficits in resistance, ambulation, illness, and loss-of-weight components increased over the 12 years of the study, while fatigue declined. An accumulating body of evidence suggests that high levels of sedentary time are associated with poorer physical function (46). An interesting finding in the current study was the relationship of resistance with the increasing sitting trajectory group but not the high sitting trajectory group. It is possible that an increase in sitting time is a marker of poor health and may be a simple way to screen people for adverse health outcomes. This highlights the utility of methods such as group-based trajectory modeling to investigate associations of sedentary behavior with health outcomes, as different patterns of behavior may affect health differently. In the Raine Study, 15-year trajectories of TV viewing time from age 5 years onward were associated with body composition at age 20 years, such that people in low and increasing TV trajectories had a lower percentage of body fat and a higher bone mineral content than those in the high TV trajectory group (47, 48). Twelve-year trajectories of TV time are associated with lower body strength but not performance in the Timed Up and Go (TUG) Test (49). A strength of this study was its longitudinal design over the course of 12 years, which included a large and representative sample of middle-aged women in Australia. To the best of our knowledge, this is the first longitudinal study to have examined the relationship between sitting time and frailty (using group-based trajectory modeling) and is the first study to identify trajectories of sitting time. In this study, factors that are related to sitting time and frailty (i.e., sociodemographic and lifestyle factors) were included in the analyses; however, no data on other factors, such as nutrition or cognition, were available. Another strength of this study was the exclusion of women who were frail at survey 3, which allowed us to examine sitting time as a risk factor for the development of frailty. A limitation of this study was the use of self-reported measures of sitting time. Even though the questionnaire we used is similar to the validated International Physical Activity Questionnaire (26, 27), there is a tendency for people to underreport sitting time in comparison with objective measures (50). However, underreporting of sitting time is expected to have been consistent over time and hence is unlikely to have affected the overall findings. While the use of objective measures of sitting time in longitudinal cohort studies is still in early stages, our study had data on sitting collected over 12 years at multiple time points. In future studies, researchers should consider using objective measures of sitting time, such as activPAL (50). activPAL is a physical activity monitor that is based on a uniaxial accelerometer (PAL Technologies Ltd., Glasgow, United Kingdom) (51). Patterns of accumulation of sitting—for example, breaks in sitting time—are known to influence health outcomes such as biomarkers of cardiometabolic risk (6) and premature mortality (52). In a recent cross-sectional study in older adults, Del Pozo-Cruz et al. (21) reported that total sedentary time and sedentary time accumulated in bouts of at least 10 minutes were deleteriously associated with frailty while a higher number of breaks in sedentary time was beneficially associated with frailty. The results are only applicable to women in this age group and not to men. In a 2016 systematic review, Buta et al. (53) identified 67 instruments designed to assess frailty, with Fried’s criteria (3), which combine self-reports and performance-based tests to assess a frailty phenotype, being the most commonly used in the literature, followed by instruments that use a deficit-accumulation model such as the frailty index (42). The FRAIL scale uses self-report data to assess functional deficit accumulation and biological domains of frailty. Thus, the findings of this study may not apply to other models of frailty that utilize performance-based measures or wider sources of data. Further investigation is required to determine whether these findings can be replicated using other models of frailty. In conclusion, in this study, 12-year patterns of sitting time in middle-aged Australian women predicted frailty in older age. Additional studies are required to understand the biological mechanisms of how sitting time may be related to frailty. Programs designed to reduce this novel risk factor may help prevent frailty in this at-risk population. ACKNOWLEDGMENTS Author affiliations: Centre for Health Services Research, Faculty of Medicine, The University of Queensland, Brisbane, Queensland, Australia (Maja Susanto, Ruth E. Hubbard, Paul A. Gardiner); The Princess Alexandra Hospital, Brisbane, Queensland, Australia (Maja Susanto, Ruth E. Hubbard); and Mater Research Institute, The University of Queensland, Brisbane, Queensland, Australia (Paul A. Gardiner). This work was supported by the National Health and Medical Research Council of Australia and the Australian Research Council (Dementia Research Development Fellowship 110331 to P.A.G.). Conflict of interest: none declared. Abbreviations CI confidence interval FRAIL fatigue, resistance, ambulation, illness, loss of weight MET metabolic equivalent TV television REFERENCES 1 Buckinx F , Rolland Y , Reginster JY , et al. . Burden of frailty in the elderly population: perspectives for a public health challenge . Arch Public Health . 2015 ; 73 ( 1 ): 19 . Google Scholar Crossref Search ADS PubMed 2 Collard RM , Boter H , Schoevers RA , et al. . Prevalence of frailty in community-dwelling older persons: a systematic review . J Am Geriatr Soc . 2012 ; 60 ( 8 ): 1487 – 1492 . Google Scholar Crossref Search ADS PubMed 3 Fried LP , Tangen CM , Walston J , et al. . Frailty in older adults: evidence for a phenotype . J Gerontol A Biol Sci Med Sci . 2001 ; 56 ( 3 ): M146 – M156 . Google Scholar Crossref Search ADS PubMed 4 Clegg A , Young J , Iliffe S , et al. . Frailty in elderly people . Lancet . 2013 ; 381 ( 9868 ): 752 – 762 . Google Scholar Crossref Search ADS PubMed 5 Gardiner PA , Mishra GD , Dobson AJ . Validity and responsiveness of the FRAIL scale in a longitudinal cohort study of older Australian women . J Am Med Dir Assoc . 2015 ; 16 ( 9 ): 781 – 783 . Google Scholar Crossref Search ADS PubMed 6 Healy GN , Matthews CE , Dunstan DW , et al. . Sedentary time and cardio-metabolic biomarkers in US adults: NHANES 2003–06 . Eur Heart J . 2011 ; 32 ( 5 ): 590 – 597 . Google Scholar Crossref Search ADS PubMed 7 Saidj M , Jørgensen T , Jacobsen RK , et al. . Work and leisure time sitting and inactivity: effects on cardiorespiratory and metabolic health . Eur J Prev Cardiol . 2016 ; 23 ( 12 ): 1321 – 1329 . Google Scholar Crossref Search ADS PubMed 8 Ramsay SE , Arianayagam DS , Whincup PH , et al. . Cardiovascular risk profile and frailty in a population-based study of older British men . Heart . 2015 ; 101 ( 8 ): 616 – 622 . Google Scholar Crossref Search ADS PubMed 9 Owen N , Sugiyama T , Eakin EE , et al. . Adults’ sedentary behavior determinants and interventions . Am J Prev Med . 2011 ; 41 ( 2 ): 189 – 196 . Google Scholar Crossref Search ADS PubMed 10 Ng SW , Popkin BM . Time use and physical activity: a shift away from movement across the globe . Obes Rev . 2012 ; 13 ( 8 ): 659 – 680 . Google Scholar Crossref Search ADS PubMed 11 Australian Bureau of Statistics . Australian Health Survey: Physical Activity, 2011–12. Canberra, ACT, Australia : Australian Bureau of Statistics ; 2013 . (Catalog no. 4364.0.55.004). http://www.abs.gov.au/AUSSTATS/abs@.nsf/Lookup/4364.0.55.004Main+Features12011-12?OpenDocument. Accessed March 11, 2017. 12 Clark BK , Lynch BM , Winkler EA , et al. . Validity of a multi-context sitting questionnaire across demographically diverse population groups: AusDiab3 . Int J Behav Nutr Phys Act . 2015 ; 12 : 148 . Google Scholar Crossref Search ADS PubMed 13 Chau JY , Merom D , Grunseit A , et al. . Temporal trends in non-occupational sedentary behaviours from Australian Time Use Surveys 1992, 1997 and 2006 . Int J Behav Nutr Phys Act . 2012 ; 9 : 76 . Google Scholar Crossref Search ADS PubMed 14 Church TS , Thomas DM , Tudor-Locke C , et al. . Trends over 5 decades in US occupation-related physical activity and their associations with obesity . PLoS One . 2011 ; 6 ( 5 ): e19657 . Google Scholar Crossref Search ADS PubMed 15 Clark BK , Peeters GM , Gomersall SR , et al. . Nine year changes in sitting time in young and mid-aged Australian women: findings from the Australian Longitudinal Study for Women’s Health . Prev Med . 2014 ; 64 : 1 – 7 . Google Scholar Crossref Search ADS PubMed 16 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 17 de Rezende LF , Rodrigues Lopes M , Rey-López JP , et al. . Sedentary behavior and health outcomes: an overview of systematic reviews . PLoS One . 2014 ; 9 ( 8 ): e105620 . Google Scholar Crossref Search ADS PubMed 18 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 19 Blodgett J , Theou O , Kirkland S , et al. . The association between sedentary behaviour, moderate-vigorous physical activity and frailty in NHANES cohorts . Maturitas . 2015 ; 80 ( 2 ): 187 – 191 . Google Scholar Crossref Search ADS PubMed 20 da Silva Coqueiro R , de Queiroz BM , Oliveira DS , et al. . Cross-sectional relationships between sedentary behavior and frailty in older adults . J Sports Med Phys Fitness . 2017 ; 57 ( 6 ): 825 – 830 . Google Scholar PubMed 21 Del Pozo-Cruz B , Mañas A , Martín-García M , et al. . Frailty is associated with objectively assessed sedentary behaviour patterns in older adults: evidence from the Toledo Study for Healthy Aging (TSHA) . PLoS One . 2017 ; 12 ( 9 ): e0183911 . Google Scholar Crossref Search ADS PubMed 22 García-Esquinas E , Andrade E , Martínez-Gómez D , et al. . Television viewing time as a risk factor for frailty and functional limitations in older adults: results from 2 European prospective cohorts . Int J Behav Nutr Phys Act . 2017 ; 14 : 54 . Google Scholar Crossref Search ADS PubMed 23 Dobson AJ , Hockey R , Brown WJ , et al. . Cohort profile update: Australian Longitudinal Study on Women’s Health . Int J Epidemiol . 2015 ; 44 ( 5 ): 1547 – 1547f . Google Scholar Crossref Search ADS PubMed 24 Lee C , Dobson AJ , Brown WJ , et al. . Cohort profile: the Australian Longitudinal Study on Women’s Health . Int J Epidemiol . 2005 ; 34 ( 5 ): 987 – 991 . Google Scholar Crossref Search ADS PubMed 25 Brown WJ , Bryson L , Byles JE , et al. . Women’s Health Australia: recruitment for a national longitudinal cohort study . Women Health . 1998 ; 28 ( 1 ): 23 – 40 . Google Scholar Crossref Search ADS PubMed 26 Craig CL , Marshall AL , Sjostrom M , et al. . International Physical Activity Questionnaire: 12-country reliability and validity . Med Sci Sports Exerc . 2003 ; 35 ( 8 ): 1381 – 1395 . Google Scholar Crossref Search ADS PubMed 27 Rosenberg DE , Bull FC , Marshall AL , et al. . Assessment of sedentary behavior with the International Physical Activity Questionnaire . J Phys Act Health . 2008 ; 5 ( suppl 1 ): S30 – S44 . Google Scholar Crossref Search ADS PubMed 28 Peeters GM , Mishra GD , Dobson AJ , et al. . Health care costs associated with prolonged sitting and inactivity . Am J Prev Med . 2014 ; 46 ( 3 ): 265 – 272 . Google Scholar Crossref Search ADS PubMed 29 Abellan van Kan G , Rolland Y , Bergman H , et al. . The I.A.N.A. Task Force on frailty assessment of older people in clinical practice . J Nutr Health Aging . 2008 ; 12 ( 1 ): 29 – 37 . Google Scholar Crossref Search ADS PubMed 30 Abellan van Kan G , Rolland YM , Morley JE , et al. . Frailty: toward a clinical definition . J Am Med Dir Assoc . 2008 ; 9 ( 2 ): 71 – 72 . Google Scholar Crossref Search ADS PubMed 31 Ware JE Jr , Snow KK , Kosinski M , et al. . SF-36 Health Survey: Manual and Interpretation Guide . Lincoln, RI : Quality Metric Inc. ; 2000 . 32 Susanto M , Hubbard RE , Gardiner PA . Validity and responsiveness of the FRAIL scale in middle-aged women . J Am Med Dir Assoc . 2018 ; 19 ( 1 ): 65 – 69 . Google Scholar Crossref Search ADS PubMed 33 World Health Organisation . Obesity: Preventing and Managing the Global Epidemic . Geneva, Switzerland : World Health Organization ; 2004 . 34 National Health and Medical Research Council . Australian Alcohol Guidelines. Health Risks and Benefits. Canberra, ACT, Australia : National Health and Medical Research Council ; 2001 . https://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/ds9.pdf. Accessed March 11, 2017. 35 Australian Institute of Health and Welfare . The Active Australia Survey: A Guide and Manual for Implementation, Analysis and Reporting. Canberra, ACT, Australia : Australian Institute of Health and Welfare ; 2003 . https://www.aihw.gov.au/getmedia/ff25c134-5df2-45ba-b4e1-6c214ed157e6/aas.pdf.aspx?inline=true. Accessed March 11, 2017. 36 Brown WJ , Bauman AE . Comparison of estimates of population levels of physical activity using two measures . Aust N Z J Public Health . 2000 ; 24 ( 5 ): 520 – 525 . Google Scholar Crossref Search ADS PubMed 37 Nagin D . Group-Based Modeling of Development . Cambridge, MA : Harvard University Press ; 2005 . 38 Nagin DS , Odgers CL . Group-based trajectory modeling in clinical research . Annu Rev Clin Psychol . 2010 ; 6 : 109 – 138 . Google Scholar Crossref Search ADS PubMed 39 Virtuoso Júnior JS , Roza LB , Tribess S , et al. . Time spent sitting is associated with changes in biomarkers of frailty in hospitalized older adults: a cross sectional study . Front Physiol . 2017 ; 8 : 505 . Google Scholar Crossref Search ADS PubMed 40 Gale CR , Cooper C , Sayer AA . Prevalence of frailty and disability: findings from the English Longitudinal Study of Ageing . Age Ageing . 2015 ; 44 ( 1 ): 162 – 165 . Google Scholar Crossref Search ADS PubMed 41 Santos-Eggimann B , Cuénoud P , Spagnoli J , et al. . Prevalence of frailty in middle-aged and older community-dwelling Europeans living in 10 countries . J Gerontol A Biol Sci Med Sci . 2009 ; 64 ( 6 ): 675 – 681 . Google Scholar Crossref Search ADS PubMed 42 Rockwood K , Mitnitski A . Frailty defined by deficit accumulation and geriatric medicine defined by frailty . Clin Geriatr Med . 2011 ; 27 ( 1 ): 17 – 26 . Google Scholar Crossref Search ADS PubMed 43 Theou O , Blodgett JM , Godin J , et al. . Association between sedentary time and mortality across levels of frailty . CMAJ . 2017 ; 189 ( 33 ): E1056 – E1064 . Google Scholar Crossref Search ADS PubMed 44 Martin A , Fitzsimons C , Jepson R , et al. . Interventions with potential to reduce sedentary time in adults: systematic review and meta-analysis . Br J Sports Med . 2015 ; 49 ( 16 ): 1056 – 1063 . Google Scholar Crossref Search ADS PubMed 45 Rosenberg DE , Gell NM , Jones SM , et al. . The feasibility of reducing sitting time in overweight and obese older adults . Health Educ Behav . 2015 ; 42 ( 5 ): 669 – 676 . Google Scholar Crossref Search ADS PubMed 46 DiPietro L , Jin Y , Talegawkar S , et al. . The joint associations of sedentary time and physical activity with mobility disability in older people: the NIH-AARP Diet and Health Study . J Gerontol A Biol Sci Med Sci . 2018 ; 73 ( 4 ): 523 – 528 . Google Scholar Crossref Search ADS 47 McVeigh J , Smith A , Howie E , et al. . Trajectories of television watching from childhood to early adulthood and their association with body composition and mental health outcomes in young adults . PLoS One . 2016 ; 11 ( 4 ): e0152879 . Google Scholar Crossref Search ADS PubMed 48 McVeigh JA , Zhu K , Mountain J , et al. . Longitudinal trajectories of television watching across childhood and adolescence predict bone mass at age 20 years in the Raine Study . J Bone Miner Res . 2016 ; 31 ( 11 ): 2032 – 2040 . Google Scholar Crossref Search ADS PubMed 49 Reid N , Healy GN , Daly RM , et al. . Twelve-year television viewing time trajectories and physical function in older adults . Med Sci Sports Exerc . 2017 ; 49 ( 7 ): 1359 – 1365 . Google Scholar Crossref Search ADS PubMed 50 Healy GN , Clark BK , Winkler EA , et al. . Measurement of adults’ sedentary time in population-based studies . Am J Prev Med . 2011 ; 41 ( 2 ): 216 – 227 . Google Scholar Crossref Search ADS PubMed 51 Grant PM , Ryan CG , Tigbe WW , et al. . The validation of a novel activity monitor in the measurement of posture and motion during everyday activities . Br J Sports Med . 2006 ; 40 ( 12 ): 992 – 997 . Google Scholar Crossref Search ADS PubMed 52 Diaz KM , Howard VJ , Hutto B , et al. . Patterns of sedentary behavior and mortality in US middle-aged and older adults: a national cohort study . Ann Intern Med . 2017 ; 167 ( 7 ): 465 – 475 . Google Scholar Crossref Search ADS PubMed 53 Buta BJ , Walston JD , Godino JG , et al. . Frailty assessment instruments: systematic characterization of the uses and contexts of highly cited instruments . Ageing Res Rev . 2016 ; 26 : 53 – 61 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. 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. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

American Journal of EpidemiologyOxford University Press

Published: Nov 1, 2018

References

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