Abstract
This study aims to investigated the association between sedentary behavior (SB) and predicted 10-year atherosclerotic cardiovascular disease (ASCVD) risk and determine whether the associations differ by how the behavior is accumulated, in US middle-aged and older adults. Cross-sectional data were derived from national health and nutrition examination survey (NHANES) 2003–2006. Seven-day wearing of accelerometer was used to assess SB pattern, exported as total SB, bouts of 1–9, 10–29, 30–59 and ≥ 60 min SB. Predicted 10-year ASCVD risk was calculated using validated pooled cohort equations. Linear regression was used to estimate adjusted coefficients. A total of 2327 participants were enrolled with mean age of 56.9 and mean predicted 10-year ASCVD risk of 10.7%. We observed significant associations of total SB and its longer accumulated patterns with higher 10-year predicted ASCVD risk, in a linear fashion and independent of a list of covariates. A 30 min increment per day of total SB, bouts in 10–29, bouts in 30–59 and bouts in ≥ 60 min were associated with 0.14, 0.14, 0.23 and 0.12% higher multivariable-adjusted 10-year predicted ASCVD risk. There are significant associations of total SB as well as its longer accumulated patterns with higher 10-year predicted ASCVD risk, independent of a list of covariates and in a linear fashion. The result indicates that reducing total sedentary time and interrupting long duration of prolonged SB, could be meaningful to for public guideline to lessen the personal and public health burden of cardiovascular health.
Keywords: Atherosclerotic cardiovascular disease, Sedentary behavior, Accumulated pattern, Cardiovascular health
Subject terms: Cardiology, Health care, Risk factors
Introduction
Atherosclerotic cardiovascular disease (ASCVD), including coronary heart disease–related death, nonfatal myocardial infarction, and fatal and nonfatal stroke, is found to be the leading cause of mortality worldwide accounting for about 17.6 million deaths annually1. ASCVD has imposed heavy financial burden, with national expenditures for ASCVD projected to increase by over 2.5-fold from 2015 to 20352. In the 2013 guideline on the assessment of cardiovascular risk published by the American college of cardiology (ACC) and the American heart association (AHA), a working group derived a pooled cohort equations (PCE) from several national-sample cohorts for the prediction of 10-year ASCVD risk, which have also been verified to be valid in external cohorts3. Current U.S. prevention guidelines for blood pressure and cholesterol management recommend use of the PCE to assess 10-year ASCVD risk and as an important starting point, for decision-making in primary prevention of ASCVD, which may be particularly useful to those relatively younger adults with unhealthy lifestyle and has not developed symptoms.
Physical activity is a fundamental component of ideal cardiovascular health and is widely recommended for both the primary prevention of CVD. Recently, the AHA published a scientific statement on circulation which emphasized the urgency of PA promotion for optimal cardiovascular health in adults and provided practical suggestions4. Sedentary behavior, often independent of moderate-to-vigorous physical activity (MVPA), is also regarded as a major component of the human movement spectrum, and have found to have adverse impact on human health, especially cardiovascular health5,6. However, most of the existing studies are based on self-reported sedentary behavior which is inaccurate and usually underestimate the magnitude of associations7. The AHA stressed that device-derived measures of sedentary time can provide improved measurement precision over self-report assessments, as well as unique insights into different patterns of behavior, when investigating the association of sedentary behavior with cardiovascular health as well as other health issues8. Sedentary behavior pattern is defined as the manner in which sedentary behavior is accumulated, for example, timing of the day, duration and frequency of bouts and breaks, according to sedentary behavior research network9, which can generally only be measured by devices. Epidemiological evidence suggests that accumulation patterns of sedentary behavior are associated with adverse cardiovascular outcome, such as risk for CVD morbidity5 and mortality10. However, no study has considered the associations between sedentary behavior patterns and PCE-estimated 10-y ACSVD risk, which is necessary when considering primary prevention of CVD. Therefore, this study aims to objectively measure sedentary behavior to (1) confirm the association between total sedentary time and predicted 10-year ASCVD risk, (2) determine whether the associations differ by how the behavior is accumulated, in a national sample of US middle-aged and older adults without prior ASCVD in the national health and nutrition examination survey (NHANES).
Methods
Study design
Data was derived from NHANES, which collects nationally representative data of about 5,000 non-institutionalized US general population in 15 different counties across the country each year. NHANES protocol was approved by the institutional review board of the National center for health statistics, written informed consent was obtained for all participants. Ran by the National center for health statistics, the NHANES collects data from demographic, socioeconomic, dietary, and health-related questions, as well as medical, physiological measurements and laboratory tests administered by highly trained medical personnel. The structure of the manuscript is consistent with the STROBE reporting requirements (Appendix S1) and all methods were performed in accordance with the declaration of Helsinki. The present study obtained data from participants aged 40 to 79 years old in NHANES 2003–2006. Participants were excluded if they already had self-reported congestive heart failure, coronary heart disease, angina/angina pectoris, heart attack/myocardial infarction and stroke (by questioning “Has a doctor or other health professional ever told you that you had a …?”).
Sedentary behavior pattern
Participants from 2003 to 2006 were invited to wear an Actigraph (Actigraph, LLC; Ft. Walton Beach, FL) model 7164 accelerometer on the hip during waking hours for a 7-day period. Non-wear time was defined by an interval of at least 60 consecutive minutes of zero activity intensity counts, with allowance for 1–2 min of counts between 0 and 100. Data for participants with four or more valid days (≥ 10 h each day) were included in the analysis. Details of raw data procession were described elsewhere11. Data of different sedentary behavior patterns were exported from raw data: time spent in 1) total sedentary behavior (total accumulated SB time), 2) different sedentary bouts, including bouts lasting 1–9 min, bouts lasting 10–29, bouts lasting 30–59, and bouts lasting ≥ 60 min (the sum of time spent in accumulated bouts of SB lasting 1 to 9, 10 to 29, 30 to 59 and ≥ 60 min, respectively). MVPA and accelerometer wear time were also exported as covariates.
10-year predicted atherosclerotic cardiovascular disease risk
We used a pooled cohort equations (PCE) endorsed by both the ACC and AHA12, to calculate predicted 10-year ASCVD risk using age, sex, race/ethnicity, systolic blood pressure (SBP), diastolic blood pressure (DBP), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), diabetes, current smoker and hypertension treatment (taking antihypertensive medications)3. Data on age, sex, race/ethnicity, diagnosis of diabetes, current smoker and taking antihypertensive medications were obtained by self-report. NHANES data collection provides the following options for race/ethnicity: Mexican American, other Hispanic, non-hispanic white, non-hispanic black, non-hispanic asian, and other race, including multiracial. In view of PCE, the current analysis recategorized participants as Mexican American, non-hispanic white, non-hispanic black and others (including other hispanic, non-hispanic Asian, and other race). Current smokers were defined as person who smoked ≥ 100 cigarettes in life and still smoke cigarettes at that time. SBP and DBP were measured by trained individuals under standards protocol. Blood samples were collected during mobile examinations for the analysis of TC, HDL-C and hemoglobin A1c levels (HbA1c). Participants who self-reported the diagnosis of diabetes or with a HbA1c ≥ 6.5% were classified as having diabetes mellitus.
Covariates
Other social demographic information (education, poverty index ratio, health insurance), behavioral factors (drinker) and medical informations (family history, diagnose of hypertension, taking aspirins, taking statins) were collected by self-report. Education status was classified to < high school, high school and > high school. Family poverty index ratio (PIR) was calculated by dividing the total family income by the poverty threshold defined by the US Census Bureau with adjustment for family size at the time of the interview, and was grouped into two categories (< 1.0–2.9, ≥ 3.0). Participants were defined as drinkers if they had at least 12 alcohol drinks (such as liquor, beer, wine, wine coolers, and any other type of alcoholic beverage) per year, one drink means a 12 oz. beer, a 4 oz. glass of wine, or an ounce of liquor. Health insurance status was obtained by asking whether they were covered by health insurance or some other kind of health care plan. Family history was collected by asking whether any living or deceased close relative had heart attack before the age of 50. Participants were considered to have hypertension if they reported the diagnosis of hypertension or had a mean SBP ≥ 140 mm Hg or a mean DBP ≥ 90 mm Hg. Height and weight were obtained for the calculation of body mass index (BMI) during mobile examinations by trained technician.
Statistical analysis
Linear regression model adjusted for the NHANES complex survey design was used to investigate the associations of SB patterns with predicted 10-year ASCVD risk. Total time in sedentary behavior was analyzed both in quartile and as a continuous variable, sedentary behavior bouts lasting in different durations were analyzed as continuous variables. Multivariable-adjusted unstandardized coefficients (beta) were estimated for each increase in 30 min of sedentary behavior to aid interpretation. Adjustment for potential covariates was performed in multiple steps. Model 1 was adjusted for age, sex, race/ethnicity and accelerometer wear time. Model 2 was adjusted for covariates in Model 1 and education, poverty index ratio, health insurance. Model 3 was adjusted for covariates in Model 2 and drinker, current smoker, family history, TC, HDL-C, diabetes mellitus, hypertension, taking aspirins, taking statins and BMI. Model 4 was adjusted for covariates in Model 3 and MVPA.
Next, we performed subgroup analyses to investigate the associations of SB patterns with 10-year predicted ASCVD risk according to several important covariates, including age (< 65/ ≥ 65), sex (male/female), race/ethnicity (non-hispanic white, non-hispanic black and Mexican American), smoker (yes/no), taking stains (yes/no), obesity (< 30 kg/m2, ≥ 30 kg/m2) and MVPA (< 150 min/week/ ≥ 150 min/week). Each subgroup was adjusted for all other covariates. Interactions were tested by adding adding multiplicative interaction terms of each subgroup factor and SB patterns into the model. In addition, we used restricted cubic splines to fit nonlinear relationships with the adjustment of all covariates, knots were located at 10, 50 and 90% quantiles of the independent variables13. Analyses were all conducted using Stata V.16.0 (StataCorp) and R V.4.3.1 (R Foundation for Statistical Computing). Two-sided P values < 0.05 were considered statistically significant.
Results
Of 9516 participants in NHANES 2003–2006, 2327 were enrolled in the final analysis (Fig. 1). Table 1 presented the characteristics of participants, with a mean age of 56.9 years old and mean 10-yr predicted ASCVD risk of 10.7%. About a half (49.9%) of participants were women. Participants with more sedentary behavior tend to be older and women, highly educated, PIR ≥ 3, have health insurance, current smoker, have lower HDL-C and higher SBP, have hypertension, taking antihypertensive and aspirin, have longer accelerometer wear time, spent less time in MVPA, and longer time in different accumulated sedentary behavior patterns, and have higher 10-yr predicted ASCVD risk. No significant difference was observed in other variables between sedentary behavior quartile groups.
Figure 1.
Flow diagram of inclusion and exclusion criteria.
Table 1.
Characteristics of participants, in total and by quartile of daily minutes spent in sedentary behavior, measured in 2003–2006 (n = 2327).
| Characteristic | Sedentary behavior min/d, no. (%) | |||||
|---|---|---|---|---|---|---|
| Total (n = 2327) | ≤ 427 (n = 580) | > 427 to ≤ 507 (n = 586) | > 507 to ≤ 593 (n = 578) | > 593 (n = 583) | P | |
| 10-year predicted ASCVD risk (%) | 10.7 (10.2–11.1) | 7.4 (6.7–8.1) | 10.1 (9.2–11.0) | 11.5 (10.6–12.5) | 13.6 (12.6–14.6) | < 0.001 |
| Age, y | 56.9 (56.5–57.4) | 53.2 (52.4–54.0) | 56.7 (55.8–57.5) | 58.4 (57.5–59.3) | 59.4 (58.4–60.3) | < 0.001 |
| Sex, n (%) | < 0.001 | |||||
| Women | 1162 (49.9) | 280 (48.3) | 278 (47.4) | 267 (46.2) | 340 (58.3) | |
| Men | 1165 (50.1) | 300 (51.7) | 308 (52.6) | 311 (53.8) | 243 (41.7) | |
| Race/ethnicity, n (%) | < 0.001 | |||||
| Non-hispanic white | 1285 (55.2) | 265 (45.7) | 333 (56.8) | 342 (59.2) | 345 (59.2) | |
| Non-hispanic black | 437 (18.8) | 98 (16.9) | 104 (17.8) | 112 (19.4) | 123 (21.1) | |
| Mexican American | 489 (21.0) | 187 (32.2) | 118 (20.1) | 96 (16.6) | 88 (15.1) | |
| Other | 116 (5.0) | 30 (5.2) | 31 (5.3) | 28 (4.8) | 27 (4.6) | |
| Education, n (%) | < 0.001 | |||||
| < High school | 598 (25.7) | 214 (36.9) | 136 (23.2) | 137 (23.7) | 111 (19.0) | |
| High school | 573 (24.6) | 157 (27.1) | 144 (24.6) | 144 (24.9) | 128 (22.0) | |
| > High school | 1156 (49.7) | 209 (36.0) | 306 (52.2) | 297 (51.4) | 344 (59.0) | |
| PIR ≥ 3, n (%) | 1146 (49.2) | 230 (39.7) | 288 (49.1) | 312 (54.0) | 316 (54.2) | < 0.001 |
| Health insurance, n (%) | 1978 (85.0) | 442 (76.2) | 497 (84.8) | 517 (89.4) | 522 (89.5) | < 0.001 |
| Drinker, n (%) | 1629 (70.0) | 416 (71.7) | 413 (70.5) | 389 (67.3) | 411 (70.5) | 0.398 |
| Current smoker, n (%) | 450 (19.3) | 129 (22.2) | 91 (15.5) | 104 (18.0) | 126 (21.6) | 0.034 |
| Family history, n (%) | 340 (14.6) | 80 (13.8) | 72 (12.3) | 100 (17.3) | 88 (15.1) | 0.097 |
| TC (mg/dL) | 206.3 (204.7–207.9) | 209.1 (206.0–212.3) | 207.3 (204.0–210.5) | 204.1 (201.0–207.3) | 204.7 (201.5–208.0) | 0.056 |
| HDL-C (mg/dL) | 54.8 (54.2–55.5) | 55.5 (54.2–56.8) | 55.1 (53.7–56.4) | 56.2 (54.8–57.6) | 52.7 (51.5–53.9) | 0.002 |
| HBA1c (%) | 5.7 (5.7– 5.7) | 5.7 (5.6–5.8) | 5.7 (5.7–5.8) | 5.7 (5.6–5.7) | 5.7 (5.6–5.8) | 0.223 |
| SBP (mmHg) | 127.9 (127.1–128.7) | 126.0 (124.5–127.5) | 127.4 (125.9–128.8) | 129.3 (127.6–130.9) | 128.8 (127.2–130.4) | 0.017 |
| DBP (mmHg) | 72.5 (72.0–73.0) | 72.9 (72.0–73.8) | 72.7 (71.8–73.6) | 72.2 (71.1–73.2) | 72.2 (71.2–73.2) | 0.777 |
| Diabetes ‡, n (%) | 326 (14.0) | 62 (10.7) | 84 (14.3) | 88 (15.2) | 92 (15.8) | 0.055 |
| Hypertension †, n (%) | 1137 (48.9) | 247 (42.6) | 277 (47.3) | 315 (54.5) | 298 (51.1) | < 0.001 |
| Taking antihypertensive, n (%) | 782 (33.6) | 155 (26.7) | 195 (33.3) | 227 (39.3) | 205 (35.2) | < 0.001 |
| Taking aspirin, n (%) | 16 (0.7) | 2 (0.3) | 4 (0.7) | 3 (0.5) | 7 (1.2) | < 0.001 |
| Taking statin, n (%) | 372 (16.0) | 60 (10.3) | 106 (18.1) | 108 (18.7) | 99 (17.0) | 0.323 |
| BMI (kg/m2) | 28.9 (28.6–29.1) | 28.4 (28.0–28.9) | 28.7 (28.3–29.1) | 29.1 (28.6–29.6) | 29.3 (28.8–29.8) | 0.121 |
| Accelerometer data (min/day) | ||||||
| Wear time | 885.8 (880.5–891.1) | 812.5 (806.1–819.0) | 849.6 (842.8–856.3) | 872.9 (866.5–879.3) | 1007.9 (994.5–1021.3) | < 0.001 |
| MVPA | 20.4 (19.4–21.4) | 29.3 (26.7–31.8) | 20.8 (18.9–22.8) | 16.8 (25.2–18.5) | 14.8 (13.4–16.1) | < 0.001 |
| Sedentary behavior | ||||||
| Total | 520.5 (514.5–526.5) | 356.9 (352.3–361.5) | 468.2 (466.3–470.1) | 546.4 (544.4–548.3) | 709.9 (699.6–720.3) | < 0.001 |
| Bouts of 1–9 min | 206.6 (204.6–208.5) | 193.1 (190.0–196.2) | 214.8 (211.2–218.3) | 209.0 (205.1–212.9) | 209.2 (204.5–213.9) | < 0.001 |
| Bouts of 10–29 min | 168.2 (165.9–170.6) | 106.1 (103.3–108.8) | 154.6 (152.1–157.0) | 188.1 (185.4–190.9) | 224.0 (220.0–228.1) | < 0.001 |
| Bouts of 30–59 min | 88.5 (86.3–90.7) | 40.7 (38.8–42.6) | 67.9 (65.7–70.3) | 98.6 (95.7–101.6) | 146.7 (142.4–151.1) | < 0.001 |
| Bouts of ≥ 60 min | 57.2 (54.1–60.2) | 17.0 (15.3–18.8) | 30.9 (28.5–33.4) | 50.7 (47.0–54.3) | 130.0 (121.4–138.5) | < 0.001 |
Continuous data are presented as mean (95% CI).
BMI body mass index, HBA1c hemoglobin A1c, HDL-C high density lipoprotein cholesterol, MVPA moderate to vigorous physical activity, PIR poverty index ratio, SBP systolic blood pressure, DBP diastolic blood pressure, TC total cholesterol.
Bold values denote P < 0.05.
†Hypertension was defined as self-reported diagnosis of hypertension or had a mean SBP ≥ 140 mm Hg or a mean DBP ≥ 90 mm Hg.
‡Diabetes was defined as self-reported diagnosis of diabetes or with a HbA1c ≥ 6.5% were classified as having diabetes mellitus.
Table 2 presented the association between total sedentary time and 10-yr predicted ASCVD risk. The results showed that continuous total sedentary time was significantly associated with predicted 10-yr ASCVD in all models. Each 30-min increment of total sedentary time was associated with 0.14% increase of 10-yr predicted ASCVD risk in the fully-adjusted model. For quartiles of total sedentary time, Quartile 2 (> 427 to ≤ 507) and Quartile 3 (> 507 to ≤ 593) were not statistically associated with 10-yr predicted ASCVD in the fully-adjusted model. While same as continuous total sedentary time, the highest quartile (> 593) was significantly associated with 10-yr predicted ASCVD risk in all models, with a 0.18% increase of predict risk compared with the reference (≤ 427) in the fully-adjusted model.
Table 2.
Associations between minutes per day in sedentary behavior with predicted 10-yr atherosclerotic cardiovascular disease risk.
| Sedentary behavior (min/d) | Quartile 1 (n = 580) | Quartile 2 (n = 586) | Quartile 3 (n = 578) | Quartile 4 (n = 583) | Total* (n = 2327) | ||||
|---|---|---|---|---|---|---|---|---|---|
| ≤ 427 | > 427 to ≤ 507 | > 507 to ≤ 593 | > 593 | ||||||
| beta | 95% CI | beta | 95% CI | beta | 95% CI | beta | 95% CI | ||
| Model 1† | Reference | 0.0056 | −0.0010 to 0.0123 | 0.0070 | −0.0003 to 0.0143 | 0.0182 | 0.0091 to 0.0272 | 0.0024 | 0.0016 to 0.0032 |
| Model 2‡ | Reference | 0.0092 | 0.0024 to 0.0161 | 0.0103 | 0.0028 to 0.0178 | 0.0235 | 0.0142 to 0.0328 | 0.0028 | 0.0020 to 0.0036 |
| Model 3§ | Reference | 0.0039 | −0.0019 to 0.0097 | 0.0034 | −0.0029 to 0.0097 | 0.0107 | 0.0032 to 0.0182 | 0.0014 | 0.0008 to 0.0021 |
| Model 4¶ | Reference | 0.0035 | −0.0023 to 0.0094 | 0.0029 | −0.0036 to 0.0093 | 0.0098 | 0.0018 to 0.0177 | 0.0014 | 0.0007 to 0.0021 |
BMI body mass index, HDL-C high density lipoprotein cholesterol, MVPA moderate to vigorous physical activity, TC total cholesterol.
Beta, unstandardized coefficients. Bold values denote P < 0.05.
*Estimated using per 30-min increment of sedentary behavior per day (continuous variable).
†Model 1 = age + sex + race/ethnicity + accelerometer wear time.
‡Model 2 = age + sex + race/ethnicity + accelerometer wear time + education + poverty index ratio + health insurance.
§Model 3 = age + sex + race/ethnicity + accelerometer wear time + education + poverty index ratio + health insurance + drinker + current smoker + family history + TC + HDL-C + diabetes mellitus + hypertension + taking aspirins + taking statins + BMI.
¶Model 4 = age + sex + race/ethnicity + accelerometer wear time + education + poverty index ratio + health insurance + drinker + current smoker + family history + TC + HDL-C + diabetes mellitus + hypertension + taking aspirins + taking statins + BMI + MVPA.
Table 3 presented the association of durations in different sedentary bouts with 10-yr predicted ASCVD risk. The results showed that the association between bouts of 1–9 min and 10-yr predicted ASCVD risk was not significant in any model. While bouts of 10–29, 30–59 and ≥ 60 min were all associated with 10-yr predicted ASCVD risk, with 0.14, 0.23 and 0.12% increase of predicted risk each 30-min increment, respectively, in the fully-adjusted model.
Table 3.
Associations between duration in different bouts of sedentary behavior with 10-yr predicted atherosclerotic cardiovascular disease risk.
| Bouts of 1–9 min | Bouts of 10–29 min | Bouts of 30–59 min | Bouts of ≥ 60 min | |||||
|---|---|---|---|---|---|---|---|---|
| beta* | 95% CI | beta | 95% CI | beta | 95% CI | beta | 95% CI | |
| Model 1† | −0.0003 | −0.0024 to 0.0018 | 0.0023 | 0.0009 to 0.0038 | 0.0037 | 0.0019 to 0.0055 | 0.0025 | 0.0012 to 0.0038 |
| Model 2‡ | −0.0003 | −0.0024 to 0.0018 | 0.0032 | 0.0017 to 0.0047 | 0.0045 | 0.0027 to 0.0064 | 0.0024 | 0.0011 to 0.0037 |
| Model 3§ | −0.0007 | −0.0023 to 0.0010 | 0.0016 | 0.0003 to 0.0028 | 0.0024 | 0.0009 to 0.0039 | 0.0013 | 0.0002 to 0.0025 |
| Model 4¶ | −0.0009 | −0.0025 to 0.0008 | 0.0014 | 0.0001 to 0.0027 | 0.0023 | 0.0007 to 0.0038 | 0.0012 | 0.0001 to 0.0024 |
BMI body mass index, HDL-C high density lipoprotein cholesterol, MVPA moderate to vigorous physical activity, TC total cholesterol.
Beta, unstandardized coefficients. Bold values denote P < 0.05.
*Estimated using per 30-min increment in different bouts of sedentary behavior per day (continuous variable).
†Model 1 = age + sex + race/ethnicity + accelerometer wear time.
‡Model 2 = age + sex + race/ethnicity + accelerometer wear time + education + poverty index ratio + health insurance.
§Model 3 = age + sex + race/ethnicity + accelerometer wear time + education + poverty index ratio + health insurance + drinker + current smoker + family history + TC + HDL-C + diabetes mellitus + hypertension + taking aspirins + taking statins + BMI.
¶Model 4 = age + sex + race/ethnicity + accelerometer wear time + education + poverty index ratio + health insurance + drinker + current smoker + family history + TC + HDL-C + diabetes mellitus + hypertension + taking aspirins + taking statins + BMI + MVPA.
Figure 2 presents the results of subgroup and interaction analysis of total sedentary time. Each 30-min increment of total sedentary time was more strongly associated with the increase of 10-yr predicted ASCVD risk in participants ≥ 65 (beta 0.0053, 95% CI 0.0026–0.0080, P = 0.012 for interaction), didn’t take stains (beta 0.0011, 95% CI 0.0004–0.0019, P = 0.038 for interaction) and with MVPA < 150 min/week (beta 0.0015, 95% CI 0.0004–0.0026, P = 0.019 for interaction) compared to participants in matched groups. There was no statistical evidence of effect modification by sex, race/ethnicity, smoker and BMI with total sedentary time, in relation to predicted 10-year ASCVD risk. The results in different bouts of sedentary behavior were similar to total sedentary behavior, data were shown in Table S1.
Figure 2.
Subgroup- and interaction analyses for the association between every 30-min/day increment in sedentary behavior and 10-yr predicted atherosclerotic cardiovascular disease risk. All models were adjusted for other covariates. Beta, unstandardized coefficients. The size of the squares is proportional to the sample size of the corresponding subgroup. Bold values denote P < 0.05.
Figure 3 presents nonlinear associations between total and duration in different bouts of sedentary behavior with 10-year predicted ASCVD risk fitted using restricted cubic splines. Almost all patterns (except bouts of 1–9 min) including total sedentary time, bouts of 10–29, 30–59 and ≥ 60 min were associated with 10-year predicted ASCVD risk in a linear and dose-dependent fashion (total: P for overall effect < 0.001, P for nonlinear relationship = 0.281 (Fig. 3A); bouts of 1–9 min: P for overall effect = 0.174; P for nonlinear relationship = 0.329 (Fig. 3B); bouts of 10–29 min: P for overall effect = 0.047; P for nonlinear relationship = 0.594 (Fig. 3C); bouts of 30–59 min: P for overall effect < 0.001; P for nonlinear relationship = 0.073 (Fig. 3D); bouts of ≥ 60 min: P for overall effect < 0.043; P for nonlinear relationship = 0.760(Fig. 3E)).
Figure 3.
Nonlinear associations between total and duration in different bouts of sedentary behavior with predicted 10-yr atherosclerotic cardiovascular disease risk. (A) total sedentary time as independent variable; (B) sedentary time in bouts of 1–9 min as independent variable; (C) sedentary time in bouts of 10–29 min as independent variable; (D) sedentary time in bouts of 30–59 min as independent variable; (E) sedentary time in bouts of ≥ 60 min as independent variable). Beta (unstandardized coefficients) were adjusted for all covariates in model. Error bands represent the95% confidence intervals for each effect estimate.
Discussion
In this cross-sectional study of 40- to 79-year-old individuals, we observed significant associations of total sedentary behavior (both in continuous type and quartile) and its longer accumulated patterns with higher 10-year predicted ASCVD risk, in a linear fashion and independent of a list of covariates. A 30-min increment per day of total sedentary behavior, bouts in 10–29, bouts in 30–59 and bouts in ≥ 60 min were associated with 0.14, 0.14, 0.23 and 12% higher multivariable-adjusted 10-year predicted ASCVD risk. While no significant association was found between the shortest accumulated pattern (bouts in 1–9 min) and 10-year predicted ASCVD risk. The total evidence suggests that both total sedentary behavior and the manner in which it is accumulated could be relevant for.
Cardiovascular health in adult
A growing body of studies have investigated the associations between total sedentary time and cardiovascular health along with the development of measurement techniques. The association of total self-reported sedentary time and CVD risk were almost consistently positive pooled by several early meta-analyses14–17. The most recent meta-analysis which enrolled studies all with total self-reported sedentary time based on 448,285 participants, also reported that higher levels of total self-reported sedentary time were still associated with increased CVD risk even after the adjustment of PA18. However, study showed that self-reported measurements of sedentary behavior appear to be less accurate, which may further underestimate the magnitude of the relationship between sedentariness and health risk7. Evidences based on objectively-measured total sedentary time were less. The OPACH Study reported significant association of objectively-measured total sedentary time with increased CVD risk among 5638 older women, in a linear dose-response manner5. The HAI study enrolled 3343 individuals also found that positive association between objectively-measured total sedentary time and CVD mortality19. Our result showed that each 30-min increment of total sedentary time was associated with 0.14% increase of 10-year predicted ASCVD risk, the value was 0.18% when compare the highest quartile with the lowest in the fully-adjusted model. This is in line with the above studies regardless of the methods of measurement. It is remarkable that a work by Loprinzi analyzed the same dataset (NHANES 2003–2006) of this study, also investigated the associations between total sedentary time with 10-yr predicted ASCVD risk, but used total sedentary time as 5 quintiles, and only adjusted age, sex, race/ethnicity, MVPA, wear time and BMI20. Our study may be more statistical powerful where a list of other important covariates were added to models, also, total sedentary time was analyzed both in quartile and continuous variable, as well as in subgroup.
According to a recent scientific statement released by AHA, the association between daily movement and cardiovascular health may differ in populations with different CVD risk factors as well as individual level characteristics4. Our subgroup results of race/ethnicity and obesity were consistent with the OPACH study5 that there was no statistical effect modification by these factors with total sedentary time in relation to 10-yr predicted ASCVD risk. However, the OPACH study didn’t observe statistically interaction of age with total sedentary time in relation to CVD risk5, while our study showed an association between total sedentary time and 10-yr predicted ASCVD risk was modified by age (P for interaction = 0.012) with a stronger effect of exposure on outcome observed in participants ≥ 65. This can be explained by age classification as we analyzed a population aged 40–79 and chose a general classification of older adults (65 years old), while the OPACH study used a cut-off of 80 years old because a relatively older sample (aged 63–97) were enrolled in their study. In addition, the OPACH study found no interactions of MVPA (< 45 min/day, ≥ 45 min/day) and total sedentary time in relation to CVD risk. In contrast, the HAI study observed statistical effect modification by MVPA (< 15 min/day, 16–29 min/day and ≥ 30 min/day) with total sedentary time on CVD mortality. In line with the HAI study, our study showed a stronger association between total sedentary time and 10-year predicted ASCVD risk in participants with less MVPA (P = 0.019 for interaction). The heterogeneity of the results may also be explained by the classification of MVPA. The MVPA cut-off of OPACH study can be transformed to 315 min/week, which is much higher than ours (150 min). While, the result of HAI can be interpreted that there was a stronger association of total sedentary time with CVD mortality in individuals with < 105 min MVPA a week compared to individuals with 105–210 min MVPA or more MVPA a week, which was similar to ours that showed stronger association in group with < 150 min a week. The above results implicates that sedentary behavior may bring stronger cardiovascular hazard to older adults, especially those who don’t follow the World Health Organization released PA guidelines. However, the results in this study should be interpreted cautiously due to smaller sample sizes stratified, the current evidences are exploratory and preliminary which require replication in larger and prospective studies to draw firm conclusions.
The development of accelerometer device enables the exploring of behavior accumulation patterns in recent years. The accumulation patterns of sedentary behavior can range from highly prolonged, which accumulating in long, continuous sedentary bouts, to highly interrupted, which accumulating in uncontinuous and shorter bouts throughout the day. Epidemiological evidences have demonstrated that accumulation patterns of sedentary behavior are associated with some other adverse health outcomes21–24. However, the body of evidence is modest when it comes to cardiovascular health, especially compared with what is known about PA patterns. Yerramalla et al. found that mean sedentary bout duration was associated with incident CVD, while the significance disappeared after the adjustment of MVPA25, this is consistent to Jefferis et al. that find all different duration of time spent in sedentary bouts (lasting 1–15, 16–30, 31–60 and ≥ 61 min) were similarly associated with incident CVD26. Contrast to the above evidences,
the OPACH study found that longer mean sedentary bout duration were associated with higher CVD risk with the independent of MVPA5, which is in line with our results that longer durations of prolonged sedentary behavior were associated with predicted 10-y ASCVD risk in the fully-adjusted model. Yerramalla et al. mentioned that the heterogeneous results may differ depending on the metric used and the level of adjustment25. Due to limited relevant studies, the heterogeneity of the above results was hard to explain, which indicates that more high-quality studies with standardized data processing protocol and cutoffs of sedentary pattern as well as definitions of CVD endpoints are needed, to investigate the associations between objectively-measured sedentary pattern and cardiovascular health outcomes.
While the underly mechanisms, at least partially, may be linked to sedentary behavior induced alterations to traditional cardiovascular risk factors including glucose tolerance, blood press, and lipid profile, as well as impairment in vascular health mediated by reductions in mean and antegrade blood flow and shear rate27. Substantially more research using animal and human models based on recent advances in human genetics and other “omics” technology may be helpful to understand pathophysiological changes that support the epidemiological research findings8.
This study with objective sedentary behavior measurement, controlled analyses for a wide range of CVD factors and conducted multiple analyzing methods on exposures. Also, we excluded participants with existing CVD to reduce bias. Several limitations are necessary to be noted. First, this is a cross-sectional study and the data is a bit old (2003–2006). Although we used a valid and widely accepted PCE to predict 10-year ASCVD risk, a causal relationship cannot be established. Secondly, a 7-day wearing of accelerometer may not be able to assess long-term behavior pattern, and the sedentary behavior measure does not include postural data and could include some standing time.
This is the first study investigated the association of objectively-measured sedentary behavior with predicted 10-year ASCVD risk. We found significant associations of total sedentary behavior as well as its longer accumulated patterns with higher 10-yr predicted ASCVD risk, independent of a list of covariates and in a linear fashion. This result indicates that reducing total sedentary time and interrupting long duration of prolonged sedentary behavior, could be meaningful to for public guideline to lessen the personal and public health burden of cardiovascular health.
Supplementary Information
Acknowledgements
We acknowledge the National Center for Health Statistics of the Centers for Disease Control and Prevention for providing the data of NHANES available on their website.
Author contributions
ZSL analyzed and interpreted the data and wrote the manuscript. PP collected the data and reviewed the manuscript. All authors read and approved the final manuscript.
Funding
This work was supported by grants from the Open Fund Project of Hubei Provincial Key Research Base of Humanities and Social Sciences “East Hubei Education and Culture Research Center” in 2024 (202418504).
Data availability
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: https://wwwn.cdc.gov/nchs/nhanes.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
The online version contains supplementary material available at 10.1038/s41598-024-68627-w.
References
- 1.Virani, S. S. et al. Heart disease and stroke statistics-2020 update: A report from the American heart association. Circulation141, e139 (2020). 10.1161/CIR.0000000000000757 [DOI] [PubMed] [Google Scholar]
- 2.Khera, R., Valero-Elizondo, J. & Nasir, K. Financial toxicity in atherosclerotic cardiovascular disease in the United States: Current state and future directions. J. Am. Heart Assoc.9, e17793 (2020). 10.1161/JAHA.120.017793 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Goff, D. J. et al. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: A report of the American college of cardiology/American heart association task force on practice guidelines. Circulation129, S49 (2014). 10.1161/01.cir.0000437741.48606.98 [DOI] [PubMed] [Google Scholar]
- 4.Jerome, G. J. et al. Increasing equity of physical activity promotion for optimal cardiovascular health in adults: A scientific statement from the American heart association. Circulation147, 1951 (2023). 10.1161/CIR.0000000000001148 [DOI] [PubMed] [Google Scholar]
- 5.Bellettiere, J. et al. Sedentary behavior and cardiovascular disease in older women: The objective physical activity and cardiovascular health (OPACH) study. Circulation139, 1036 (2019). 10.1161/CIRCULATIONAHA.118.035312 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lavie, C. J., Ozemek, C., Carbone, S., Katzmarzyk, P. T. & Blair, S. N. Sedentary behavior, exercise, and cardiovascular health. Circ. Res.124, 799 (2019). 10.1161/CIRCRESAHA.118.312669 [DOI] [PubMed] [Google Scholar]
- 7.Celis-Morales, C. A. et al. Objective vs. self-reported physical activity and sedentary time: Effects of measurement method on relationships with risk biomarkers. PLoS ONE7, e36345 (2012). 10.1371/journal.pone.0036345 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Young, D. R. et al. Sedentary behavior and cardiovascular morbidity and mortality: A science advisory from the American heart association. Circulation134, e262 (2016). 10.1161/CIR.0000000000000440 [DOI] [PubMed] [Google Scholar]
- 9.Tremblay, M. S. et al. Sedentary behavior research network (SBRN)-terminology consensus project process and outcome. Int. J. Behav. Nutr. Phys. Act.14, 75 (2017). 10.1186/s12966-017-0525-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Jefferis, B. J. et al. Objectively measured physical activity, sedentary behaviour and all-cause mortality in older men: Does volume of activity matter more than pattern of accumulation?. Brit. J. Sport Med.53, 1013 (2019). 10.1136/bjsports-2017-098733 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Troiano, R. P. et al. Physical activity in the United States measured by accelerometer. Med. Sci. Sport Exer.40, 181 (2008). 10.1249/mss.0b013e31815a51b3 [DOI] [PubMed] [Google Scholar]
- 12.Grundy, S. M. et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of blood cholesterol: A report of the American college of cardiology/American heart association task force on clinical practice guidelines. Circulation139, e1082 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Harrell, F. E. Regression Modeling Strategies: With Applications to Linear Models, Logistic Regression, and Survival Analysis (Springer, 2001). [Google Scholar]
- 14.Ford, E. S. & Caspersen, C. J. Sedentary behaviour and cardiovascular disease: A review of prospective studies. Int. J. Epidemiol.41, 1338 (2012). 10.1093/ije/dys078 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Grontved, A. & Hu, F. B. Television viewing and risk of type 2 diabetes, cardiovascular disease, and all-cause mortality: A meta-analysis. JAMA J. Am. Med. Assoc.305, 2448 (2011). 10.1001/jama.2011.812 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Biswas, A. 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.162, 123 (2015). 10.7326/M14-1651 [DOI] [PubMed] [Google Scholar]
- 17.Pandey, A. et al. Continuous dose-response association between sedentary time and risk for cardiovascular disease: a meta-analysis. JAMA Cardiol.1, 575 (2016). 10.1001/jamacardio.2016.1567 [DOI] [PubMed] [Google Scholar]
- 18.Bailey, D. P., Hewson, D. J., Champion, R. B. & Sayegh, S. M. Sitting time and risk of cardiovascular disease and diabetes: A systematic review and meta-analysis. Am. J. Prev. Med.57, 408 (2019). 10.1016/j.amepre.2019.04.015 [DOI] [PubMed] [Google Scholar]
- 19.Ballin, M., Nordstrom, P., Niklasson, J. & Nordstrom, A. Associations of objectively measured physical activity and sedentary time with the risk of stroke, myocardial infarction or all-cause mortality in 70-year-old men and women: A prospective cohort study. Sports Med.51, 339 (2021). 10.1007/s40279-020-01356-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Loprinzi, P. D. Sedentary behavior and predicted 10-yr risk for a first atherosclerotic cardiovascular disease (ASCVD) event using the pooled cohort risk equations among US adults. Int. J. Cardiol.203, 443 (2016). 10.1016/j.ijcard.2015.10.213 [DOI] [PubMed] [Google Scholar]
- 21.Manns, P., Ezeugwu, V., Armijo-Olivo, S., Vallance, J. & Healy, G. N. Accelerometer-derived pattern of sedentary and physical activity time in persons with mobility disability: National health and nutrition examination survey 2003 to 2006. J. Am. Geriatr. Soc.63, 1314 (2015). 10.1111/jgs.13490 [DOI] [PubMed] [Google Scholar]
- 22.Wagnild, J. M. & Pollard, T. M. Associations between television time and activPAL-measured duration and pattern of sedentary time among pregnant women at risk of gestational diabetes in the UK. J. Phys. Act. Health17, 471 (2020). 10.1123/jpah.2019-0315 [DOI] [PubMed] [Google Scholar]
- 23.Zhang, X. et al. Associations of accelerometer-measured physical activity, sedentary behavior with self-reported insomnia in older women: Does pattern matter?. Sleep Med.104, 58 (2023). 10.1016/j.sleep.2023.02.015 [DOI] [PubMed] [Google Scholar]
- 24.Du, L. et al. Associations between objectively measured patterns of sedentary behaviour and arterial stiffness in Chinese community-dwelling older women. Eur. J. Cardiovasc. Nur.22, 374 (2023). 10.1093/eurjcn/zvac079 [DOI] [PubMed] [Google Scholar]
- 25.Yerramalla, M. S. et al. Objectively measured total sedentary time and pattern of sedentary accumulation in older adults: Associations with incident cardiovascular disease and all-cause mortality. J. Gerontol. A Biol.77, 842 (2022). 10.1093/gerona/glac023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Jefferis, B. J. et al. Does total volume of physical activity matter more than pattern for onset of CVD? A prospective cohort study of older British men. Int. J. Cardiol.278, 267 (2019). 10.1016/j.ijcard.2018.12.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Carter, S., Hartman, Y., Holder, S., Thijssen, D. H. & Hopkins, N. D. Sedentary behavior and cardiovascular disease risk: Mediating mechanisms. Exerc. Sport Sci. Rev.45, 80 (2017). 10.1249/JES.0000000000000106 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: https://wwwn.cdc.gov/nchs/nhanes.