2023 年 30 巻 12 号 p. 1817-1827
Aims: We examined the association between television (TV) viewing time and all-cause and cardiovascular disease (CVD) mortality among Japanese adults with and without a history of stroke or myocardial infarction (MI).
Methods: In the Japan Collaborative Cohort Study, 76,572 participants (851 stroke survivors, 1,883 MI survivors, and 73,838 persons without a history of stroke or MI), aged 40–79 years at baseline (1988-1990), completed a lifestyle, diet, and medical history questionnaire, and were followed up regarding mortality until 2009. The Cox proportional hazard model was used to calculate the multivariable-adjusted hazard ratios (HRs) with 95% confidence intervals (CIs) of all-cause and CVD mortality.
Results: During the 19.3-year median follow-up period, 17,387 deaths were documented. TV viewing time was positively associated with all-cause and CVD mortality regardless of stroke or MI history. The multivariable-adjusted HRs of all-cause mortality with 95% CIs for TV viewing time of 3–4.9 h, 5–6.9 h, and ≥ 7 h were 1.18 (0.95–1.48), 1.12 (0.86–1.45), and 1.61 (1.12–2.32) for stroke survivors; 0.97 (0.81–1.17), 1.40 (1.12–1.76), and 1.44 (1.02–2.03) for MI survivors; and 1.00 (0.96–1.03), 1.07 (1.01–1.12), and 1.22 (1.11–1.34) for persons without a history of stroke or MI, respectively, compared with ＜3 h.
Conclusions: Prolonged TV viewing time was associated with higher risks of all-cause and CVD mortality in stroke or MI survivors and in persons without a history of them. It may be recommended to reduce sedentary time for stroke or MI survivors, independent of the level of physical activity.
Non-Standard Abbreviations and Acronyms: CVD: Cardiovascular disease, BMI: Body mass index, HRs: Hazard ratios , CIs: Confidence intervals, JACC Study: Japan Collaborative Cohort Study for Evaluation of Cancer Risk
Television (TV) viewing is a standard leisure-time sedentary behavior in many populations1). Previous studies, including our previous study2), have shown that sedentary behavior, including TV watching time, can be positively associated with risks of all-cause and cardiovascular disease (CVD) mortality in the general population, independent of physical activity levels1, 3, 4). The American Heart Association’s scientific statement on sedentary behavior emphasized reducing sedentary behavior for the primary prevention of CVD5). However, it remains unclear whether reducing sedentary behaviors can improve the prognosis of stroke or MI survivors6), although scientific evidence for better lifestyles among stroke or MI survivors is warranted given their substantially improved case fatality rate7, 8).
Few studies have examined the association between sedentary behavior and the risk of mortality among MI survivors. In the Kailuan study of 989 Chinese MI survivors with a mean follow-up of 7.46 years, prolonged sedentary time (4–8 h/day) was associated with a higher risk of mortality after MI compared with ＜4 h9). In the Women’s Health Initiative Observational Study with 553 female MI survivors, every 1 h/day increase in sitting time after MI was associated with a 9% increased risk of all-cause mortality among those with pre-MI levels of sitting time ＜8 h/day10). These results suggest that prolonged TV viewing time can be a risk factor for mortality among MI survivors.
To our knowledge, no previous study has examined the association between TV viewing time and mortality among stroke survivors. Furthermore, no comparative data are available on the impact of TV viewing time on mortality among stroke or MI survivors and persons without a history of stroke or MI. Therefore, the present study aimed to examine and compare the association between TV viewing time and all-cause and CVD mortality among persons with and without a history of stroke or MI using data from a large long-term cohort study of Japanese people of both sexes.
The Japan Collaborative Cohort Study for Evaluation of Cancer Risk (JACC) is a nationwide community-based prospective study that started between 1988–1990 and enrolled 110,585 individuals (46,395 men and 64,190 women) aged 40–79 years, living in 45 communities across Japan. The methodology of the JACC study has been described elsewhere11). Briefly, a total of 110,585 participants were asked to complete self-administered questionnaires that collected information on their demographic characteristics, medical history, lifestyle, and diet. Before completing the questionnaire, informed individual consent was obtained from each participant in 36 of the 45 study areas (written consent in 35 areas and oral consent in one area). Group consent was obtained from each area leader for the remaining nine areas. The study protocol was approved by the ethics committees of Hokkaido University, Nagoya University, and Osaka University
From 110,585 cohort participants, we excluded 16,442 participants (7,146 men and 9,296 women) living in 8 study areas because questions on TV viewing time were not included in the questionnaire. We also excluded 13,895 participants (5,576 men and 8,319 women) because of missing information on the history of stroke or MI. Furthermore, 958 participants (285 men and 673 women) with a history of cancer and 120 participants (70 men and 50 women) with a history of stroke and MI were excluded. Finally, we excluded 2,598 participants (874 men and 1724 women) due to missing or inappropriate responses (＞12 h) regarding their TV viewing time. A total of 76,572 participants (32,444 men and 44,128 women) were included in the analyses. According to the self-reported history of stroke and MI, we classified the participants into three groups: “history of stroke,” “history of MI,” and “no history of stroke or MI”12).
Assessment of TV Viewing TimeWe asked participants about their average daily time spent viewing TV at baseline using the following question: ‘On average, how many hours do you watch TV?’ The participants reported their time spent viewing TV per day: “approximately h/day.” According to their responses, we classified them into four categories: ＜3 h, 3–4.9 h, 5–6.9 h, and ≥ 7 h/day. We defined ＜3 h as a reference because a recent meta-analysis concluded that 3–4 h/day of TV viewing was a threshold above which all-cause and CVD mortality risk increased13).
Assessment of CovariatesA self-administered questionnaire was used to collect the demographic and lifestyle information. Baseline covariates included age, sex, height, weight, past medical history (such as diabetes and hypertension), smoking and alcohol drinking status, exercise and walking habits, mental status, educational level, occupation, and eating habits. Body mass index (BMI) was calculated as body weight (kg) divided by height in meters squared (m2).
Mortality SurveillanceA systematic review of death certificates was conducted for each area to determine the cause of death. Mortality data were sent to the Ministry of Health and Welfare through the local public health center, and the underlying cause of death was coded for the National Vital Statistics according to the International Classification of Diseases, 10th revision (ICD10). The endpoints of death in this study were all causes and CVD (ICD-10 codes I01–I99). In 37 areas, the end of the follow-up was the end of 1999 in three areas, the end of 2003 in one area, the end of 2008 in two areas, and the end of 2009 in the remaining 31 areas. The date of moving from the community was verified using population registration documents. If participants died after moving from their original community, they were treated as having withdrawn as censored from the study when they moved out.
Statistical AnalysisPerson-years of follow-up were calculated as the duration from the date of the baseline questionnaire to the date of death, emigration from the community, or the end of follow-up, whichever occurred first. Baseline characteristics were reported as mean (standard deviation) for continuous variables and percentages for categorical variables. According to TV viewing time, hazard ratios (HRs) with 95% confidence intervals (CIs) of all-cause and CVD mortality were calculated using Cox proportional hazards regression models. In multivariable analyses, we adjusted for age (continuous), sex (women or men), hours of exercise (seldom, 1–4 h, or ≥ 5 h per week), hours of walking (seldom, 0.5 h, or ＞0.5 h per day) (model 1). We further adjusted for history of hypertension (yes or no), history of diabetes (yes or no), BMI (sex-specific quintile), smoking status (never, ex-smoker, current smoker of 1–19, or a current smoker of ≥ 20 cigarettes per day), alcohol consumption (never drinker, ex-drinker, current drinker of 0.1–45.9, or ≥ 46.0 g ethanol per day), perceived mental stress (low, moderate, or high), educational level (≤ 18 or ≥ 19 years of age upon completion of education), employment status (unemployed or employed), frequency of consuming vegetables, fish, and fruits, soybean intakes (quintile) (model 2). In the sensitivity analyses, we repeated the analyses by excluding all participants who died during the first five years of follow-up to account for potential bias due to reverse causality. We also conducted a stratified analysis by sex. SAS version 9.4 (SAS, Inc., Cary, NC, USA) was used for all statistical analyses.
Table 1 presents the baseline characteristics of the participants according to TV viewing time. Persons with longer TV viewing time were more likely to be older, unemployed, and drink less regardless of a history of stroke or MI. Stroke survivors with longer TV viewing time were less educated and were less likely to walk. In comparison, MI survivors with longer TV viewing time were more likely to be female and diabetic, eat more fruit, and less fish.
| Television viewing time (h/day) | P for trend | ||||
|---|---|---|---|---|---|
| ＜3 | 3-4.9 | 5-6.9 | ≥ 7 | ||
| History of stroke | |||||
| No. of participants | 292 | 334 | 165 | 60 | |
| Age, y | 65.0±8.8 | 64.5±8.3 | 66.3±7.1 | 68.0±8.3 | 0.003 |
| Sex, % of women | 37.0 | 35.6 | 41.2 | 40.0 | 0.53 |
| Body mass index, kg/m2 | 22.7±3.2 | 23.1±3.2 | 22.8±3.5 | 22.9±5.1 | 0.83 |
| History of hypertension, % | 47.6 | 53.0 | 53.1 | 58.3 | 0.18 |
| History of diabetes mellitus, % | 9.3 | 8.1 | 12.8 | 5.0 | 0.42 |
| Current smoker, % | 20.8 | 22.0 | 23.7 | 34.6 | 0.02 |
| Current drinker, % | 36.7 | 37.8 | 28.7 | 22.6 | 0.02 |
| High mental stress, % | 21.0 | 17.9 | 16.1 | 24.4 | 0.51 |
| College or higher education, % | 16.5 | 11.5 | 9.2 | 6.0 | 0.06 |
| Unemployed, % | 48.6 | 47.7 | 73.1 | 82.5 | ＜0.001 |
| Walking ≥ 60min/day, % | 42.8 | 45.3 | 36.2 | 9.1 | ＜0.001 |
| Exercise ≥ 1h/week, % | 30.7 | 34.2 | 40.0 | 22.8 | 0.23 |
| Vegetable intake, times/week | 15.2±8.4 | 13.9±8.1 | 15.9±9.2 | 13.7±8.8 | 0.50 |
| Fish intake, times/week | 6.3±4.0 | 5.6±3.9 | 6.0±3.7 | 6.5±4.0 | 0.47 |
| Fruits intake, times/week | 7.0±4.3 | 7.0±4.6 | 7.4±4.9 | 8.1±5.3 | 0.11 |
| Soybeans intake, times/week | 5.3±3.4 | 4.9±3.3 | 5.6±3.4 | 5.2±2.9 | 0.79 |
| History of MI | |||||
| No. of participants | 690 | 782 | 306 | 105 | |
| Age, y | 62.7±9.2 | 63.7±8.4 | 65.7±7.6 | 65.8±8.0 | ＜0.001 |
| Sex, % of women | 53.0 | 54.7 | 59.2 | 67.6 | 0.004 |
| Body mass index, kg/m2 | 23.1±2.9 | 23.2±3.2 | 23.1±3.1 | 23.4±3.5 | 0.42 |
| History of hypertension, % | 45.5 | 46.1 | 53.0 | 49.0 | 0.39 |
| History of diabetes mellitus, % | 9.7 | 10.6 | 11.8 | 18.5 | 0.007 |
| Current smoker, % | 22.7 | 20.9 | 21.3 | 29.3 | 0.10 |
| Current drinker, % | 41.1 | 37.1 | 34.8 | 18.6 | ＜0.001 |
| High mental stress, % | 27.7 | 21.4 | 19.8 | 23.1 | 0.51 |
| College or higher education, % | 16.7 | 12.0 | 12.9 | 11.1 | 0.31 |
| Unemployed, % | 26.5 | 33.5 | 48.3 | 65.0 | ＜0.001 |
| Walking ≥ 60min/day, % | 46.3 | 43.6 | 34.1 | 30.9 | 0.003 |
| Exercise ≥ 1h/week, % | 29.6 | 27.2 | 29.4 | 21.6 | 0.13 |
| Vegetable intake, times/week | 15.3±8.2 | 14.7±8.1 | 14.9±7.8 | 13.6±7.7 | 0.11 |
| Fish intake, times/week | 6.0±4.1 | 6.0±3.7 | 5.5±3.4 | 4.7±3.1 | 0.003 |
| Fruits intake, times/week | 7.4±4.6 | 7.6±4.6 | 8.1±4.3 | 8.8±4.4 | 0.005 |
| Soybeans intake, times/week | 5.0±3.1 | 5.2±3.1 | 4.9±3.4 | 5.0±3.1 | 0.76 |
| No history of stroke or MI | |||||
| No. of participants | 35873 | 29301 | 7214 | 1450 | |
| Age, y | 55.2±10.0 | 57.2±9.8 | 60.8±9.6 | 63.1±9.7 | ＜0.001 |
| Sex, % of women | 56.2 | 57.2 | 66.7 | 71.4 | ＜0.001 |
| Body mass index, kg/m2 | 22.7±2.9 | 22.9±3.0 | 23.1±3.3 | 23.1±3.5 | ＜0.001 |
| History of hypertension, % | 16.7 | 19.3 | 24.6 | 26.6 | ＜0.001 |
| History of diabetes mellitus, % | 3.5 | 4.3 | 5.3 | 7.4 | ＜0.001 |
| Current smoker, % | 25.2 | 27.3 | 24.7 | 24.0 | 0.09 |
| Current drinker, % | 43.5 | 40.4 | 32.5 | 29.7 | ＜0.001 |
| High mental stress, % | 24.5 | 18.9 | 15.9 | 16.3 | ＜0.001 |
| College or higher education, % | 15.4 | 11.9 | 9.8 | 10.5 | ＜0.001 |
| Unemployed, % | 12.7 | 17.8 | 32.6 | 45.9 | ＜0.001 |
| Walking ≥ 60min/day, % | 52.6 | 51.2 | 45.3 | 37.1 | ＜0.001 |
| Exercise ≥ 1h/week, % | 26.8 | 27.3 | 27.0 | 23.6 | 0.006 |
| egetable intake, times/week | 14.6±7.8 | 14.3±7.8 | 14.4±7.7 | 14.2±8.0 | 0.15 |
| Fish intake, times/week | 6.6±4.0 | 6.5±4.0 | 6.4±4.0 | 6.2±4.0 | 0.001 |
| Fruits intake, times/week | 7.4±4.5 | 7.6±4.5 | 7.9±4.6 | 8.3±4.5 | ＜0.001 |
| Soybeans intake, times/week | 5.1±3.1 | 5.0±3.1 | 5.1±3.2 | 5.0±3.2 | 0.67 |
Data are mean±standard deviation for continuous variables and percentages for categorical variables.
During the 19.3 years of median follow-up, a total of 17,387 deaths were documented. In age- and sex-adjusted analyses, TV viewing time was associated with higher risks of all-cause mortality among persons with and without a history of stroke or MI (Table 2). After adjustment for hours of exercise and walking, the association was slightly attenuated but remained statistically significant. Further adjustment for the other potential confounding factors did not alter the association materially. The multivariable-adjusted HRs of all-cause mortality for TV viewing time of 3–4.9 h, 5–6.9 h, and ≥ 7 h were 1.18 (0.95–1.48), 1.12 (0.86–1.45), and 1.61 (1.12–2.32) for persons with a history of stroke; 0.97 (0.81–1.17), 1.40 (1.12–1.76), and 1.44 (1.02–2.03) for persons with a history of MI; and 1.00 (0.96–1.03), 1.07 (1.01–1.12), and 1.22 (1.11–1.34) for persons without a history of stroke or MI, respectively, compared with a TV viewing time of ＜3 h. A two-hour increment in TV viewing time was associated with an increased risk of all-cause mortality. The multivariable-adjusted HRs were 1.12 (1.03–1.23) for persons with a history of stroke, 1.10 (1.01–1.20) for persons with a history of MI, and 1.04 (1.02–1.06) for persons without a history of stroke or MI.
| Television viewing time (h/day) | P for trend | Increment by 2 h/ day | ||||
|---|---|---|---|---|---|---|
| ＜3 | 3-4.9 | 5-6.9 | ≥ 7 | |||
| History of stroke | ||||||
| Person-years | 3694 | 4131 | 1917 | 537 | ||
| All-cause | ||||||
| No. of cases | 155 | 187 | 108 | 47 | ||
| Mortality rate (per 1000 person-years) | 42.0 | 45.3 | 56.3 | 87.5 | ||
| Age- and sex-adjusted HR (95%CI) | Ref | 1.14 (0.92-1.41) | 1.22 (0.96-1.56) | 2.06 (1.48-2.87) | ＜0.001 | 1.21 (1.11-1.32) |
| Model 1 | Ref | 1.15 (0.92-1.42) | 1.18 (0.92-1.51) | 1.70 (1.21-2.38) | 0.004 | 1.16 (1.06-1.26) |
| Model 2 | Ref | 1.18 (0.95-1.48) | 1.12 (0.86-1.45) | 1.61 (1.12-2.32) | 0.03 | 1.12 (1.03-1.23) |
| CVD | ||||||
| No. of cases | 74 | 93 | 52 | 27 | ||
| Mortality rate (per 1000 person-years) | 20.0 | 22.5 | 27.1 | 50.2 | ||
| Age- and sex-adjusted HR (95%CI) | Ref | 1.19 (0.87-1.61) | 1.23 (0.86-1.75) | 2.32 (1.48-3.62) | 0.001 | 1.23 (1.10-1.39) |
| Model 1 | Ref | 1.20 (0.88-1.63) | 1.17 (0.82-1.67) | 1.81 (1.15-2.87) | 0.03 | 1.16 (1.03-1.31) |
| Model 2 | Ref | 1.28 (0.93-1.77) | 1.12 (0.76-1.63) | 1.77 (1.08-2.91) | 0.08 | 1.12 (0.99-1.27) |
| History of MI | ||||||
| Person-years | 9564 | 10397 | 3517 | 1158 | ||
| All-cause | ||||||
| No. of cases | 234 | 257 | 138 | 46 | ||
| Mortality rate (per 1000 person-years) | 24.5 | 24.7 | 39.2 | 39.7 | ||
| Age- and sex-adjusted HR (95%CI) | Ref | 1.00 (0.84-1.19) | 1.45 (1.17-1.79) | 1.63 (1.18-2.24) | ＜0.001 | 1.16 (1.06-1.25) |
| Model 1 | Ref | 0.99 (0.82-1.18) | 1.41 (1.14-1.75) | 1.61 (1.17-2.23) | ＜0.001 | 1.14 (1.05-1.24) |
| Model 2 | Ref | 0.97 (0.81-1.17) | 1.40 (1.12-1.76) | 1.44 (1.02-2.03) | 0.001 | 1.10 (1.01-1.20) |
| CVD | ||||||
| No. of cases | 103 | 116 | 63 | 26 | ||
| Mortality rate (per 1000 person-years) | 10.8 | 11.2 | 17.9 | 22.5 | ||
| Age- and sex-adjusted HR (95%CI) | Ref | 1.02 (0.78-1.33) | 1.49 (1.09-2.04) | 2.03 (1.32-3.14) | ＜0.001 | 1.20 (1.07-1.36) |
| Model 1 | Ref | 1.00 (0.77-1.31) | 1.44 (1.04-1.98) | 1.98 (1.28-3.08) | ＜0.001 | 1.19 (1.05-1.34) |
| Model 2 | Ref | 1.00 (0.76-1.32) | 1.40 (1.00-1.96) | 1.88 (1.17-3.03) | 0.002 | 1.14 (1.01-1.30) |
| No history of stroke or MI | ||||||
| Person-years | 607856 | 484737 | 110369 | 19959 | ||
| All-cause | ||||||
| No. of cases | 6996 | 6637 | 2076 | 506 | ||
| Mortality rate (per 1000 person-years) | 11.5 | 13.7 | 18.8 | 25.4 | ||
| Age- and sex-adjusted HR (95%CI) | Ref | 1.02 (0.99-1.05) | 1.14 (1.09-1.20) | 1.37 (1.25-1.49) | ＜0.001 | 1.08 (1.05-1.10) |
| Model 1 | Ref | 1.02 (0.98-1.05) | 1.13 (1.08-1.19) | 1.33 (1.22-1.46) | ＜0.001 | 1.07 (1.05-1.09) |
| Model 2 | Ref | 1.00 (0.96-1.03) | 1.07 (1.01-1.12) | 1.22 (1.11-1.34) | ＜0.001 | 1.04 (1.02-1.06) |
| CVD | ||||||
| No. of cases | 2036 | 1901 | 667 | 167 | ||
| Mortality rate (per 1000 person-years) | 3.3 | 3.9 | 6.0 | 8.4 | ||
| Age- and sex-adjusted HR (95%CI) | Ref | 0.98 (0.92-1.04) | 1.16 (1.06-1.27) | 1.37 (1.17-1.60) | ＜0.001 | 1.09 (1.05-1.13) |
| Model 1 | Ref | 0.98 (0.92-1.04) | 1.15 (1.06-1.26) | 1.33 (1.14-1.56) | ＜0.001 | 1.08 (1.05-1.12) |
| Model 2 | Ref | 0.95 (0.89-1.01) | 1.06 (0.97-1.16) | 1.18 (1.00-1.38) | 0.03 | 1.04 (1.00-1.08) |
HR, hazard ratio; CI, confidence interval; CVD, cardiovascular disease; MI, myocardial infarction
Model 1: adjusted for age, sex, hours of exercise, and hours of walking
Model 2: model 1＋adjusted for history of hypertension, history of diabetes, body mass index, smoking status, alcohol consumption, perceived mental stress, educational level, regular employment and dietary intakes of vegetable, fish, fruits and soybeans.
Similar positive associations of TV viewing time with CVD mortality were observed among persons with a history of stroke or MI, while a borderline increased risk of CVD mortality was observed for TV viewing time ≥ 7 h/day among persons without a history of stroke or MI. Similar positive associations were observed between each two-hour increment in TV viewing time and the risk of CVD mortality; the multivariable-adjusted HRs were 1.12 (0.99–1.27) for persons with a history of stroke, 1.14 (1.01–1.30) for persons with a history of MI, and 1.04 (1.00–1.08) for persons without a history of stroke or MI.
In the sensitivity analyses, after excluding all participants who died during the first five years of follow-up (n=2,500), the excess risk of all-cause mortality for TV viewing time ≥ 7 h/day remained statistically significant among stroke survivors and persons without any history of stroke or MI (Supplementary Table 1). The excess risk of all-cause mortality became of borderline statistical significance among MI survivors. Conversely, the excess risk of CVD mortality for TV viewing time ≥ 7 h/day did not change materially among MI survivors, whereas the excess risk was attenuated among persons without any history of stroke or MI. The association was more evident for men than women regardless of a history of stroke and MI (Supplementary Table 2).
| Television viewing time (h/day) | P for trend | ||||
|---|---|---|---|---|---|
| ＜3 | 3–4.9 | 5–6.9 | ≥ 7 | ||
| History of stroke | |||||
| Person-years | 3580 | 4037 | 1849 | 488 | |
| All-cause | |||||
| No. of cases | 115 | 148 | 77 | 26 | |
| Mortality rate (per 1000 person-years) | 32.1 | 36.7 | 41.6 | 53.2 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.22 (0.95–1.55) | 1.17 (0.88–1.56) | 1.77 (1.15–2.72) | 0.03 |
| Model 1 | Ref | 1.23 (0.96–1.57) | 1.13 (0.85–1.51) | 1.54 (0.99–2.40) | 0.11 |
| Model 2 | Ref | 1.25 (0.97–1.61) | 1.09 (0.80–1.49) | 1.60 (1.00–2.57) | 0.14 |
| CVD | |||||
| No. of cases | 51 | 70 | 33 | 14 | |
| Mortality rate (per 1000 person-years) | 14.2 | 17.3 | 17.8 | 28.7 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.30 (0.91–1.87) | 1.12 (0.72–1.73) | 2.01 (1.11–3.67) | 0.09 |
| Model 1 | Ref | 1.31 (0.91–1.89) | 1.08 (0.70–1.68) | 1.65 (0.89–3.05) | 0.28 |
| Model 2 | Ref | 1.41 (0.96–2.07) | 1.06 (0.66–1.70) | 1.87 (0.97–3.62) | 0.24 |
| History of MI | |||||
| Person-years | 9436 | 10266 | 3400 | 1130 | |
| All-cause | |||||
| No. of cases | 190 | 204 | 97 | 36 | |
| Mortality rate (per 1000 person-years) | 20.1 | 19.9 | 28.5 | 31.9 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.00 (0.82–1.21) | 1.33 (1.04–1.70) | 1.70 (1.18–2.43) | ＜0.001 |
| Model 1 | Ref | 0.99 (0.81–1.21) | 1.30 (1.02–1.67) | 1.71 (1.19–2.46) | ＜0.001 |
| Model 2 | Ref | 0.95 (0.77–1.17) | 1.22 (0.93–1.59) | 1.44 (0.97–2.14) | 0.02 |
| CVD | |||||
| No. of cases | 82 | 86 | 45 | 19 | |
| Mortality rate (per 1000 person-years) | 8.7 | 8.4 | 13.2 | 16.8 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 0.97 (0.72–1.32) | 1.43 (0.99–2.06) | 2.03 (1.23–3.36) | 0.001 |
| Model 1 | Ref | 0.95 (0.70–1.29) | 1.39 (0.96–2.01) | 2.01 (1.21–3.35) | 0.002 |
| Model 2 | Ref | 0.94 (0.68–1.29) | 1.35 (0.91–2.00) | 1.88 (1.07–3.31) | 0.01 |
| No history of stroke or MI | |||||
| Person-years | 605263 | 482258 | 109410 | 19663 | |
| All-cause | |||||
| No. of cases | 6094 | 5779 | 1721 | 400 | |
| Mortality rate (per 1000 person-years) | 10.1 | 12.0 | 15.7 | 20.3 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.02 (0.98–1.06) | 1.11 (1.05–1.17) | 1.30 (1.17–1.44) | ＜0.001 |
| Model 1 | Ref | 1.02 (0.98–1.06) | 1.10 (1.04–1.16) | 1.28 (1.15–1.41) | ＜0.001 |
| Model 2 | Ref | 1.00 (0.96–1.03) | 1.04 (0.98–1.10) | 1.17 (1.06–1.30) | 0.005 |
| CVD | |||||
| No. of cases | 1796 | 1650 | 562 | 129 | |
| Mortality rate (per 1000 person-years) | 3.0 | 3.4 | 5.1 | 6.6 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 0.97 (0.90–1.03) | 1.13 (1.03–1.25) | 1.25 (1.05–1.50) | 0.001 |
| Model 1 | Ref | 0.97 (0.90–1.03) | 1.13 (1.02–1.24) | 1.23 (1.03–1.47) | 0.003 |
| Model 2 | Ref | 0.93 (0.87–1.00) | 1.03 (0.94–1.14) | 1.09 (0.91–1.30) | 0.31 |
HR, hazard ratio; CI, confidence interval; CVD, cardiovascular disease; MI, myocardial infarction
Model 1: adjusted for age, sex, hours of exercise, and hours of walking
Model 2: model 1＋adjusted for history of hypertension, history of diabetes, body mass index, smoking status, alcohol consumption, perceived
mental stress, educational level, regular employment and dietary intakes of vegetable, fish, fruits and soybeans.
| Television viewing time (h/day) | P for trend | ||||
|---|---|---|---|---|---|
| ＜3 | 3-4.9 | 5-6.9 | ≥ 7 | ||
| History of stroke | |||||
| Men | |||||
| Person-years | 2282 | 2630 | 1131 | 246 | |
| All-cause | |||||
| No. of cases | 103 | 133 | 69 | 32 | |
| Mortality rate (per 1000 person-years) | 45.1 | 50.6 | 61.0 | 130.2 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.18 (0.91-1.53) | 1.19 (0.88-1.61) | 2.97 (1.99-4.43) | ＜0.001 |
| Model 1 | Ref | 1.21 (0.93-1.57) | 1.15 (0.84-1.56) | 2.46 (1.63-3.72) | ＜0.001 |
| Model 2 | Ref | 1.27 (0.96-1.69) | 1.13 (0.81-1.57) | 2.14 (1.34-3.43) | 0.01 |
| CVD | |||||
| No. of cases | 44 | 60 | 33 | 19 | |
| Mortality rate (per 1000 person-years) | 19.3 | 22.8 | 29.2 | 77.3 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.24 (0.84-1.83) | 1.33 (0.84-2.09) | 3.88 (2.26-6.68) | ＜0.001 |
| Model 1 | Ref | 1.27 (0.86-1.89) | 1.26 (0.80-1.98) | 3.02 (1.71-5.31) | 0.001 |
| Model 2 | Ref | 1.46 (0.95-2.25) | 1.24 (0.75-2.04) | 2.64 (1.40-4.98) | 0.01 |
| Women | |||||
| Person-years | 1412 | 1501 | 786 | 292 | |
| All-cause | |||||
| No. of cases | 52 | 54 | 39 | 15 | |
| Mortality rate (per 1000 person-years) | 36.8 | 36.0 | 49.6 | 51.5 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.07 (0.73-1.57) | 1.31 (0.87-1.99) | 1.17 (0.66-2.08) | 0.33 |
| Model 1 | Ref | 1.09 (0.74-1.60) | 1.30 (0.85-1.99) | 1.02 (0.56-1.84) | 0.59 |
| Model 2 | Ref | 0.88 (0.56-1.38) | 0.99 (0.59-1.66) | 0.66 (0.31-1.39) | 0.43 |
| CVD | |||||
| No. of cases | 30 | 33 | 19 | 8 | |
| Mortality rate (per 1000 person-years) | 21.2 | 22.0 | 24.2 | 27.4 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.14 (0.69-1.87) | 1.11 (0.62-1.97) | 1.07 (0.49-2.35) | 0.85 |
| Model 1 | Ref | 1.13 (0.68-1.87) | 1.11 (0.62-1.99) | 0.89 (0.40-1.98) | 0.83 |
| Model 2 | Ref | 1.00 (0.54-1.82) | 0.87 (0.43-1.78) | 0.53 (0.18-1.54) | 0.23 |
| History of MI | |||||
| Men | |||||
| Person-years | 4517 | 4540 | 1321 | 320 | |
| All-cause | |||||
| No. of cases | 134 | 162 | 82 | 21 | |
| Mortality rate (per 1000 person-years) | 29.7 | 35.7 | 62.1 | 65.5 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.10 (0.87-1.38) | 1.64 (1.24-2.16) | 1.88 (1.18-2.98) | ＜0.001 |
| Model 1 | Ref | 1.09 (0.87-1.38) | 1.59 (1.19-2.11) | 1.96 (1.22-3.14) | ＜0.001 |
| Model 2 | Ref | 1.08 (0.84-1.38) | 1.69 (1.24-2.30) | 1.72 (1.02-2.89) | ＜0.001 |
| CVD | |||||
| No. of cases | 57 | 70 | 36 | 16 | |
| Mortality rate (per 1000 person-years) | 12.6 | 15.4 | 27.3 | 49.9 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.12 (0.79-1.59) | 1.73 (1.13-2.64) | 3.37 (1.93-5.90) | ＜0.001 |
| Model 1 | Ref | 1.13 (0.79-1.61) | 1.67 (1.08-2.57) | 3.56 (2.02-6.28) | ＜0.001 |
| Model 2 | Ref | 1.07 (0.74-1.56) | 1.67 (1.04-2.66) | 3.06 (1.60-5.84) | ＜0.001 |
| Women | |||||
| Person-years | 5047 | 5857 | 2196 | 837 | |
| All-cause | |||||
| No. of cases | 100 | 95 | 56 | 25 | |
| Mortality rate (per 1000 person-years) | 19.8 | 16.2 | 25.5 | 29.9 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 0.88 (0.66-1.16) | 1.28 (0.92-1.77) | 1.43 (0.92-2.21) | 0.03 |
| Model 1 | Ref | 0.88 (0.66-1.17) | 1.25 (0.90-1.74) | 1.34 (0.85-2.09) | 0.06 |
| Model 2 | Ref | 0.83 (0.61-1.12) | 1.20 (0.84-1.70) | 1.37 (0.83-2.25) | 0.07 |
| CVD | |||||
| No. of cases | 46 | 46 | 27 | 10 | |
| Mortality rate (per 1000 person-years) | 9.1 | 7.9 | 12.3 | 11.9 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 0.95 (0.63-1.43) | 1.37 (0.85-2.20) | 1.22 (0.62-2.43) | 0.24 |
| Model 1 | Ref | 0.94 (0.62-1.42) | 1.30 (0.81-2.10) | 1.11 (0.55-2.23) | 0.41 |
| Model 2 | Ref | 1.10 (0.70-1.73) | 1.44 (0.86-2.40) | 1.59 (0.72-3.52) | 0.12 |
| No history of stroke or MI | |||||
| Men | |||||
| Person-years | 260431 | 202958 | 34473 | 5092 | |
| All-cause | |||||
| No. of cases | 4104 | 3797 | 1008 | 220 | |
| Mortality rate (per 1000 person-years) | 15.8 | 18.7 | 29.2 | 43.2 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.03 (0.99-1.08) | 1.20 (1.12-1.29) | 1.48 (1.29-1.70) | ＜0.001 |
| Model 1 | Ref | 1.03 (0.99-1.08) | 1.19 (1.11-1.28) | 1.46 (1.27-1.67) | ＜0.001 |
| Model 2 | Ref | 1.00 (0.96-1.05) | 1.10 (1.03-1.18) | 1.30 (1.13-1.50) | ＜0.001 |
| CVD | |||||
| No. of cases | 1105 | 1011 | 289 | 68 | |
| Mortality rate (per 1000 person-years) | 4.2 | 5.0 | 8.4 | 13.4 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.01 (0.93-1.10) | 1.22 (1.07-1.38) | 1.58 (1.24-2.02) | ＜0.001 |
| Model 1 | Ref | 1.01 (0.93-1.10) | 1.21 (1.06-1.38) | 1.56 (1.22-1.99) | ＜0.001 |
| Model 2 | Ref | 0.97 (0.89-1.06) | 1.09 (0.95-1.24) | 1.32 (1.03-1.69) | 0.03 |
| Women | |||||
| Person-years | 347425 | 281780 | 75896 | 14867 | |
| All-cause | |||||
| No. of cases | 2892 | 2840 | 1068 | 286 | |
| Mortality rate (per 1000 person-years) | 8.3 | 10.1 | 14.1 | 19.2 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 1.00 (0.95-1.05) | 1.08 (1.01-1.16) | 1.26 (1.11-1.42) | ＜0.001 |
| Model 1 | Ref | 1.00 (0.95-1.05) | 1.07 (1.00-1.15) | 1.22 (1.08-1.38) | ＜0.001 |
| Model 2 | Ref | 0.98 (0.93-1.03) | 1.03 (0.96-1.10) | 1.13 (1.00-1.28) | 0.07 |
| CVD | |||||
| No. of cases | 931 | 890 | 378 | 99 | |
| Mortality rate (per 1000 person-years) | 2.7 | 3.2 | 5.0 | 6.7 | |
| Age- and sex-adjusted HR (95%CI) | Ref | 0.95 (0.87-1.04) | 1.12 (0.99-1.26) | 1.21 (0.99-1.49) | 0.01 |
| Model 1 | Ref | 0.95 (0.87-1.04) | 1.11 (0.98-1.25) | 1.17 (0.95-1.44) | 0.03 |
| Model 2 | Ref | 0.92 (0.84-1.01) | 1.04 (0.92-1.18) | 1.07 (0.86-1.32) | 0.34 |
HR, hazard ratio; CI, confidence interval; CVD, cardiovascular disease; MI, myocardial infarction
Model 1: adjusted for age, hours of exercise, and hours of walking
Model 2: model 1＋adjusted for history of hypertension, history of diabetes, body mass index, smoking status, alcohol consumption, perceived mental stress, educational level,
regular employment and dietary intakes of vegetable, fish, fruits and soybeans.
In this large prospective study of Japanese men and women aged 40–79 years with a median follow-up of 19.3 years, we observed that stroke survivors, MI survivors, and persons without stroke or MI who watched TV for ≥ 7 h/day had approximately 60%, 40%, and 20% higher risks of all-cause mortality, respectively, than those who watched TV for ＜3 h/day.
To the best of our knowledge, this is the first study to find associations between TV viewing time and all-cause and CVD mortality among stroke survivors, and TV viewing time with CVD mortality among MI survivors. Our results suggest that prolonged TV viewing time can reduce the prognosis of stroke or MI survivors, independent of other physical activities, such as walking and exercise.
The positive associations between TV viewing time and all-cause and CVD mortality among MI survivors were consistent with the findings of previous studies. A cohort study of 989 Chinese MI survivors with a mean follow-up of 7.46 years showed that prolonged sedentary time (4–8 h per day) was associated with a higher risk of all-cause mortality after MI compared with ＜4 h, whereas no such association was found between sedentary time of ＞8 h per day and all-cause mortality probably due to the limited number of cases (6 cases); the respective multivariable-adjusted HRs were 1.62 (1.14–2.31) and 1.19 (0.48–2.95)9). In another cohort study of 553 female MI survivors from the Women’s Health Initiative Observational Study10), sitting time ＞5 to ＜8 h/day and ≥ 8 h/day tended to be associated with higher risks of all-cause and CVD mortality compared to ≤ 5 h/day. They reported that a 1 h/day increase in sitting time was associated with an increased risk of all-cause mortality. Compared to these previous studies, the larger sample size of stroke or MI survivors in the present study allowed to assess the robust impact of sitting time on mortality.
The underlying mechanisms of the deleterious impacts of prolonged TV viewing time on CVD occurrence and prognosis might be attributable to the increased total cholesterol, triglycerides, waist circumference, decreased glucose uptake, and decreased skeletal muscle lipoprotein lipase activity14, 15). Also, prolonged TV viewing time may lead to systemic inflammation, increased plasma viscosity, and platelet aggregation, resulting in vascular mortality16, 17). A previous experimental study reported that 3 h of sitting resulted in significant impairment of superficial femoral artery flow-mediated dilation, leading to endothelial dysfunction and an elevated risk of vascular mortality18).
The strengths of the present study are that the prospective study design minimizes recall bias of the exposure assessment and the large sample size of stroke and MI survivors enables us to assess their mortality.
However, this study had several limitations. First, since TV viewing time and a history of stroke or MI were self-reported, false reporting could be a potential problem. However, the reliability of self-reported TV viewing time at home was moderate to high19). Second, we did not have information on multiple measurements of TV viewing time. During the long follow-up period, TV viewing time may have changed. Further studies with multiple evaluations of TV viewing time are important to reduce measurement errors and better assess the temporal relationship between TV viewing time and mortality. Third, despite adjusting for potential confounding factors, we cannot rule out the effects of unmeasured factors or residual confounding, such as blood pressure, blood glucose, and cholesterol levels, and the degree of disability attributable to the first stroke or MI. Stroke or MI survivors could spend more time viewing TV because of such disabilities due to the first stroke or MI. Covariates adjusted in the multivariable models, such as hours of exercise and walking and employment status after first stroke or MI, can serve as a proxy for disability. To assess the remaining impact of unmeasured confounders that our present covariates did not capture, we calculated the E-values in the primary analysis20). The E-values for the association between TV viewing time of ≥ 7 h and all-cause mortality were 2.13 for stroke survivors and 1.89 for MI survivors. Therefore, an unmeasured confounder must be associated with a TV viewing time of ≥ 7 h and all-cause mortality by approximately two-fold for stroke and MI survivors, while adjusting for other covariates, to explain the observed HR in the primary analyses. Fourth, reverse causation could have led to the observed associations. However, the association between TV viewing time and mortality remained statistically significant when deaths during the first five years were excluded. Finally, since the present study was an observational study, the causality of prolonged TV viewing time with protective cardiovascular health cannot be determined.
Prolonged TV viewing time was associated with higher risks of all-cause and CVD mortality regardless of a history of stroke or MI after adjustment for physical activities, such as walking, exercise, and other confounding factors. Prolonged TV viewing time can be detrimental among stroke or MI survivors, although confounding due to the severity of the first non-fatal stroke or non-fatal MI and sequelae cannot be ruled out.
We thank all staff members involved in this study for their valuable help in conducting the baseline survey and follow-up.
This study has been supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT) (MonbuKagaku-sho); Grants-in-Aid for Scientific Research on Priority Areas of Cancer; and Grants-in-Aid for Scientific Research on Priority Areas of Cancer Epidemiology from MEXT (Nos. 61010076, 62010074, 63010074, 1010068, 2151065, 3151064, 4151063, 5151069, 6279102, 11181101, 17015022, 18014011, 20014026, 20390156, 26293138), and JSPS KAKENHI No.16H06277. This research was also supported by Grant–in–Aid from the Ministry of Health, Labour and Welfare, Health and Labor Sciences research grants, Japan (Comprehensive Research on Cardiovascular Disease and Life-Style Related Diseases: H20–Junkankitou [Seishuu]–Ippan–013; H23–Junkankitou [Seishuu]–Ippan–005); an Intramural Research Fund (22-4-5) for Cardiovascular Diseases of National Cerebral and Cardiovascular Center; Comprehensive Research on Cardiovascular Diseases and Life-Style Related Diseases (H26-Junkankitou [Seisaku]-Ippan-001) and H29–Junkankitou [Seishuu]–Ippan–003 and 20FA1002.
None.
