Light-Intensity Physical Activity and Cardiometabolic Risk Among Older Adults With Multiple Chronic Conditions

Abstract

Purpose:

To assess the relationship between light-intensity physical activity (LIPA) and cardiometabolic risk factors among middle-aged and older adults with multiple chronic conditions.

Design:

Cross-sectional design utilizing data from the Health and Retirement Study (2010, 2012).

Setting:

Laboratory- and survey-based testing of a nationally representative sample of community-dwelling middle aged and older adults.

Participants:

Adults aged 50 years and older (N = 14 996).

Measures:

Weighted metabolic equivalent of tasks was calculated using self-reported frequency of light, moderate, and vigorous physical activity. Cardiometabolic risk factors (systolic and diastolic blood pressure, glycosylated hemoglobin [HbA_1c], high-density lipoprotein cholesterol [HDL-C], total cholesterol, and non-HDL-C) were objectively measured. A multiple chronic condition index was based on 8 self-reported chronic conditions.

Analysis:

Weighted multivariate linear regression models.

Results:

Light-intensity physical activity was independently associated with favorable HDL-C (β = 1.25; 95% confidence interval [CI]: 0.46-2.05) and total cholesterol (β = 2.72; 95% CI: 0.53-4.90) after adjusting for relevant confounders. The HDL-C health benefit was apparent when stratified by number of chronic conditions, for individuals with 2 to 3 conditions (β = 1.73; 95% CI: 0.58-2.89). No significant associations were observed between LIPA and blood pressure, HbA_1c, or non-HDL-C.

Conclusions:

Engaging in LIPA may be an important health promotion activity to manage HDL-C and total cholesterol. Additional longitudinal research is needed to determine the causal association between LIPA and cardiometabolic risk which can potentially inform physical activity guidelines targeting older adults with multiple chronic conditions.

Keywords

light-intensity physical activity cardiometabolic risk factors older adults multiple chronic conditions

Purpose

Cardiovascular disease (CVD) is the leading cause of death in the Unites States and is driven largely by modifiable cardiometabolic risk factors.^1,2 Controlling cardiometabolic risk factors by engaging in moderate to vigorous physical activity is an important primary prevention strategy to manage cardiovascular health and secondary prevention strategy to minimize the severity of CVD.^3
–5 While the benefits of moderate to vigorous physical activity are well established, engagement typically declines with age.^6,7 Current physical activity guidelines recommend older adults engage in muscle-strengthening activities on 2 or more days a week and at least 150 minutes of moderate-intensity aerobic activity or 75 minutes of vigorous-intensity aerobic activity.⁷ However, older adults are least likely to attain the recommended guidelines due to age-related functional limitations, reduced mobility, and the increasing prevalence of multiple chronic conditions.^6,8
–10 Light-intensity physical activity (LIPA), which may consist of casual walking, dancing slowly, light yard work, and housework,^11,12 may be a potential alternative to moderate and vigorous physical activity to maintain cardiovascular health.¹³ However, there is limited empirical data to provide guidance on physical activity recommendations related to LIPA for middle-aged and older adults.¹⁴

While there is an accumulating body of research showing an association between LIPA and cardiometabolic risk factors,^{13,15

–20} the evidence is mixed and may vary by cardiometabolic risk factor assessed.^16,20
–22 For example, Howard and colleagues found significant independent associations between LIPA and favorable outcomes for systolic blood pressure, high-density lipoprotein cholesterol (HDL-C), fasting glucose, and triglyceride levels; however, no significant associations were observed for diastolic blood pressure or low-density lipoprotein cholesterol.¹⁸ Discrepancies in the results of studies demonstrating a positive or null association between LIPA and cardiometabolic risk factors are not wholly attributable to study design (eg, cross-sectional versus prospective) or measurement of physical activity (eg, self-reported versus device-assessed). For example, studies conducted using either a cross-sectional or prospective design and employing accelerometer-measured physical activity have not shown a significant independent benefit from engaging in LIPA on HDL-C, total cholesterol, and glycosylated hemoglobin (HbA_1c).^18,23,24 Moreover, older adults with multiple chronic conditions are a high-risk subpopulation of the older adult population, which can pose a major challenge to manage cardiometabolic risk.²⁵ There has been a dearth of attention to individuals with multiple chronic conditions in research examining physical activity intensity levels and cardiometabolic risk factors. In the literature, most studies have either adjusted for multiple chronic conditions and some studies have excluded individuals with multiple chronic conditions.^5,15,21,22 Understanding the extent to which LIPA may impact cardiometabolic health in this high-risk subpopulation warrants further investigation.

We have the unique opportunity to investigate the role of self-reported LIPA using data from the nationally representative Health and Retirement Study (HRS). While we acknowledge the limitations of self-reported measures of physical activity, including reporting biases (ie, recall and social desirability bias) and measurement error, several nationally representative community-based studies rely on self-reported measures of physical activity because of the feasibility of data collection in large samples.^26
–28 It has been suggested that self-reported physical activity, in comparison to device-assessed physical activity, measures related, but different aspects of physical activity important for improving health.^29
–31 Although older adults may see modest improvements in cardiometabolic risk factor control due to engaging in LIPA,⁵ the nature of the relationship between LIPA and cardiometabolic risk factors is poorly understood. It is unclear whether engaging in LIPA affects cardiometabolic risk differently across chronic condition burden. Such an investigation is warranted given the growing prevalence of multimorbidity among middle-aged and older adults. To address these gaps, we will examine the association between self-reported LIPA and objectively measured cardiometabolic risk factors (ie, HbA_1c, systolic blood pressure, diastolic blood pressure, HDL-C, total cholesterol, and non-HDL-C) among middle-aged and older adults stratified by number of multiple chronic conditions.

Methods

Design

Data for this cross-sectional analysis were obtained from the HRS, a longitudinal biennial survey of a nationally representative sample of US adults aged 50 and older that collects detailed information on physical health and functioning, cognition, disability, socioeconomic factors, and health-care expenditures.^32,33 Data collection began in 1992 and is ongoing. The response rate was 81.4% in 1992 and between 85% and 90% in the following waves.^32,33 Detailed information concerning the sample design, recruitment, response rates, and measurement validation is extensively discussed elsewhere.^32,33 In 2010 and 2012, biomarker data were collected from a randomly selected subsample of the HRS population. Participants with biomarker data from 2010 (N = 7782) and 2012 (N = 7244) were pooled together (N = 15 026). Individuals were excluded if missing all outcome variables (n = 4; 0.02%) and race/ethnicity data (n = 22; 0.15%), yielding 14 996 participants in the final analytic sample.

Measures

Self-reported LIPA

Light-intensity physical activity was collected by self-report from participants via questionnaire. The self-reported physical activity questions used in HRS are similar to validated self-reported physical activity questions that have been used in other surveys and demonstrate good construct, face, and predictive ability.³⁴ To help minimize exposure misclassification, we calculated metabolic equivalent of tasks (METs) to estimate energy expenditure based on the frequency of reported physical activity.^10,35 The following 3 questions were asked to ascertain vigorous, moderate, and light (mildly) physical activity intensity, respectively: “How often do you take part in sports or activities that are vigorous, such as running or jogging, swimming, cycling, aerobics or gym workout, tennis, or digging with a spade or shovel?”; “And how often do you take part in sports or activities that are moderately energetic, such as gardening, cleaning the car, walking at a moderate pace, dancing, or floor or stretching exercises?”; “And how often do you take part in sports or activities that are mildly energetic, such as vacuuming, laundry, and home repairs?” There were 5 response options for each question: “hardly, every, or never”, “1 to 3 times per month”; “once per week”; “greater than once per week”; or “everyday”. Following a method previously described,³⁶ individual responses to each physical activity question was assigned a weight (ie, light = 1.2; moderate = 1.4; vigorous = 1.8) using an average MET calculation.³⁷ The scores were summed across all 3 intensity levels of physical activity (range: 0-17.6) which reflected total physical activity. The thresholds used to determine physical activity intensity levels were based upon established MET thresholds: sedentary ≤2.0; LIPA ≥2.0 to 4.0; moderate ≥4.0 to 8.0; and vigorous ≥8.0.³⁷ We used the cut points to reduce estimation error and increase the comparability of physical activity intensity across other studies.^36,38

Cardiometabolic risk factors

Six cardiometabolic risk factors (HbA_1c, systolic and diastolic blood pressure, HDL-C, total cholesterol, and non-HDL-C) were assessed. Low-density lipoprotein cholesterol was not assessed as this information was not available in the HRS data set. Continuous measures of HbA_1c, HDL-C, and total cholesterol were collected using dried blood spot technique.³⁹ Consistent with prior research,²⁵ non-HDL-C was calculated as total cholesterol minus HDL-C. Standard procedures, previously described in detail, were used to collect and measure blood pressure.⁴⁰ Briefly, both systolic and diastolic blood pressure was measured as the average of 3 measurements. Systolic and diastolic blood pressure was analyzed as continuous variables.

Covariates

Informed by previous studies,^5,13,41 several variables were considered as confounders. Demographic characteristics included age, gender (male, female), race/ethnicity (Hispanic, non-Hispanic black, non-Hispanic white, other), marital status (married or partnered, unmarried, or separated), education level (less than high school, high school, more than high school), health insurance status (insured, uninsured), and household income. Behavioral lifestyle factors included body mass index (BMI) and current smoking status (nonsmoker, current smoker). Clinical factors included in the analysis were self-rated health status (excellent/very good/good, poor/fair), functional limitations (none, at least 1 limitation), and cardiometabolic risk factor medication use (yes, no). Individuals who were categorized as yes self-reported medication use for blood pressure, cholesterol, and/or diabetes. Age, BMI, and household income were all analyzed as continuous variables.

Multiple chronic conditions

Multiple chronic conditions were operationalized based on the following 8 chronic conditions: high blood pressure, diabetes mellitus, cancer (excluding cancer of the skin), lung disease, heart disease, stroke, psychiatric problems, and arthritis. Disease status was collected based on self-report and ascertained by asking respondents, “Has a doctor ever told you that you have…?” The total number of conditions was summed across all participants who indicated yes to a condition, giving a maximum score of 8 chronic conditions. Based on previous research and examination of the distribution of the data in this study, the number of chronic conditions was categorized as 0 to 1, 2 to 3, and 4+ conditions.^42,43

Analysis

Descriptive statistics were reported for all study variables by physical activity intensity, with means and standard deviations computed for continuous variables and frequencies and percentages calculated for categorical variables. Means and standard errors were calculated for continuous variables. Linear correlations between cardiometabolic risk factors and physical activity intensity by multiple chronic condition categories were evaluated. Multivariable linear regression models were used to estimate the association between LIPA and cardiometabolic risk factors in separate models before and after controlling for selected covariates. Two models were constructed for each cardiometabolic risk factor: model 1 (METs, survey year) and model 2 (model 1 + age, gender, race/ethnicity, marital status, education level, health insurance status, household annual income, BMI, self-rated health, current smoking status, functional limitation, and CVD prescription drug use). The linear regression models were stratified by multiple chronic condition categories. To determine whether the association varied by sex, an interaction term between LIPA and sex was tested in the final model. Two-sided P values <.05 were considered statistically significant. Significance for the main effects was .0083 for Bonferroni correction and .10 for interaction terms. Analyses were weighted to account for the complex sampling design. All data management functions and statistical analyses were performed using SAS version 9.3 (Cary, North Carolina).

Results

The overall distribution of selected participant characteristics by physical activity intensity is displayed in Table 1. The mean age among the study population was 67.4, 57.5% were female, 18.3% were black, and 13.0% were Hispanic. Of the 14 996 participants, 20.1% were sedentary, 21.1% reported light-, 32.0% reported moderate-, and 26.8% reported vigorous-intensity physical activity levels. There were significant differences in the means and percentages of participant characteristics where individuals who were sedentary were more likely to be older, have less education, be unmarried, uninsured, current smokers, report higher BMI, have greater multiple chronic conditions, and have more functional limitations.

Table 1.

Distribution of Selected Characteristics of Participants by Physical Activity Intensity^a in the Health and Retirement Study (2010, 2012).

	Overall		Sedentary		Light Intensity		Moderate Intensity		Vigorous Intensity
	N^b	%	n	%	n	%	n	%	n	%
Total	14 996		3000	20.1	3168	21.1	4803	32.0	4025	26.8
Age (mean, SD)	67.4	8.7	77.1	2.2	68.4	1.9	66.7	1.5	64.7	1.5
Gender
Male	6367	42.5	1474	49.1	990	31.3	1903	39.6	2000	49.7
Female	8629	57.5	1526	50.9	2178	68.8	2900	60.4	2025	50.3
Race/ethnicity
Black	2737	18.3	704	23.5	640	20.2	852	17.7	541	13.4
Hispanic	1954	13.0	397	13.2	435	13.7	658	13.7	464	11.5
White	9863	65.7	1807	60.2	2010	63.5	3162	65.8	2884	71.7
Other	442	3.0	92	3.1	83	2.6	131	2.7	136	3.4
Education
Less than high school	3568	23.8	1014	33.8	929	29.3	1034	21.5	591	14.7
High school	4350	29.0	890	29.7	1045	33.0	1472	30.6	943	23.4
More than high school	7076	47.2	1096	36.5	1193	37.7	2296	47.8	2491	61.9
Marital status
Unmarried	5680	37.9	1378	45.9	1379	43.5	1749	36.4	1174	29.2
Married	9316	62.1	1622	54.1	1789	56.5	3054	63.6	2851	70.8
Income (mean, SD)	12 544	541	6465	506	7706	1028	12 187	1142	21 307	1177
Insurance status
Uninsured	1491	10.0	221	7.4	353	11.2	517	10.8	400	10.0
Insured	13462	90.0	2764	92.6	2809	88.8	4274	89.2	3615	90.0
Smoking status
Nonsmoker	12516	83.9	2411	80.9	2565	81.1	3991	83.5	3554	88.8
Current smoker	2400	16.1	568	19.1	595	18.9	788	16.5	449	11.2
Self-rated health
Fair/poor	5352	35.7	1784	59.5	1448	45.7	1490	31.0	630	15.7
Good/very good/excellent	9644	64.3	1216	40.5	1720	54.3	3313	69.0	3395	84.3
BMI (mean, SD)	28.7	5.1	29.5	1.4	29.7	1.2	28.6	0.84	27.3	0.76
CVD drug use^c
Yes	10034	66.9	2314	77.1	2343	74.0	3158	65.7	2219	1806
No	4962	33.1	686	22.9	825	26.0	1645	34.3	55.1	44.9
Chronic conditions
0-1	5191	34.6	601	20.0	806	25.4	1716	35.7	2068	51.4
2-3	6814	45.4	1333	44.4	1543	48.7	2326	48.4	1612	40.1
≥4	2991	20.0	1066	35.5	819	25.9	751	15.8	345	8.6
Functional limitations
0	12413	82.8	1845	61.6	2523	79.7	4239	88.3	3806	94.6
≥1	2578	17.2	1152	38.4	644	20.3	563	11.7	219	5.4

Abbreviations: BMI, body mass index; SD, standard deviation.

^a Physical activity levels measured in metabolic equivalent of tasks (METs).

^b Some categories may not add to total due to a negligible amount of missing.

^c Cardiovascular disease (CVD) drug use includes prescription drugs for hypertension, diabetes, or cholesterol.

Table 2 shows the mean systolic and diastolic blood pressure, HbA_1c, HDL-C, total cholesterol, and non-HDL-C levels by physical activity intensity according to the number of multiple chronic conditions. For the overall population, systolic blood pressure, HbA_1c, HDL-C, total cholesterol, and non-HDL-C were linearly correlated with increasing physical activity intensity levels (P ≤ .0001). Similar patterns were detected among study participants who indicated none or 1 chronic condition. Among study participants who reported 2 to 3 or more than 4 chronic conditions, only HbA_1c, total cholesterol, and HDL-C were significantly correlated with physical activity measures in the expected direction.

Table 2.

Mean Cardiometabolic Risk Factors by Physical Activity Intensity^a According to Number of Multiple Chronic Conditions, Health and Retirement Survey (2010, 2012).

	Total			Sedentary			Light			Moderate			Vigorous			P Value
	N	Mean	SE	N	Mean	SE	N	Mean	SE	N	Mean	SE	N	Mean	SE	P Value
Total
SBP (mm Hg)	14500	130.8	0.17	2822	133.4	0.41	3039	131.2	0.38	4683	130.4	0.30	3956	128.9	0.30	<.001
DBP (mm Hg)	14500	79.6	0.099	2822	79.7	0.24	3039	79.4	0.22	4683	79.7	0.18	3956	79.7	0.18	.3029
HbA_1c	14613	5.88	0.008	2938	6.1	0.22	3078	6.00	0.20	4672	5.88	0.02	3925	5.7	0.01	<.001
HDL-C (mg/dL)	14668	53.8	0.13	2921	51.0	0.28	3102	53.5	0.28	4707	54.1	0.23	3938	55.6	0.26	<.001
Total cholesterol	14724	193.8	0.36	2941	186.1	0.82	3120	193.0	0.78	4713	195.4	0.64	3950	198.4	0.66	<.001
Non-HDL-C (mg/dL)	14592	140.4	0.33	2906	135.4	0.75	3092	139.9	0.71	4678	141.6	0.60	3916	143.0	0.60	<.001
MCC: 0-1
SBP (mm Hg)	5092	128.3	0.16	577	132.4	0.85	781	129.3	0.78	1693	128.7	0.49	2041	126.5	0.41	<.001
DBP (mm Hg)	5092	80.4	0.28	577	81.9	0.51	781	80.6	0.44	1693	80.4	0.29	2041	79.8	0.34	.0036
HbA_1c	5059	5.59	0.01	581	5.68	0.04	785	5.65	0.03	1673	5.62	0.02	2020	5.51	0.01	<.001
HDL-C (mg/dL)	5094	55.7	0.23	586	53.1	0.65	795	55.3	0.57	1688	55.6	0.39	2025	56.8	0.38	<.001
Total cholesterol	5107	202.5	0.59	589	199.6	1.93	798	201.3	1.46	1690	203.2	1.05	2030	203.2	0.90	.0217
Non-HDL-C (mg/dL)	5069	147.1	0.54	582	147.3	1.81	794	146.2	1.36	1679	147.8	0.96	2014	146.8	0.82	.7146
MCC: 2-3
SBP (mm Hg)	6596	132.0	0.25	1280	134.5	0.62	1479	132.1	0.54	2260	131.0	0.43	1577	131.2	0.48	.003
DBP (mm Hg)	6596	79.8	0.15	1280	80.2	0.37	1479	79.7	0.30	2260	79.6	0.25	1577	79.8	0.29	.5794
HbA_1c	6635	5.92	0.01	1311	5.96	0.03	1497	5.99	0.03	2257	5.93	0.02	1570	5.8	0.02	<.001
HDL-C (mg/dL)	6657	53.6	0.19	1300	51.2	0.42	1512	53.9	0.40	2269	53.8	0.32	1576	54.9	0.42	<.001
Total cholesterol	6682	192.5	0.53	1306	186.2	1.19	1520	193.2	1.13	2274	194.1	0.90	1582	195.0	1.07	<.001
Non-HDL-C (mg/dL)	6620	139.2	0.48	1292	135.1	1.08	1506	139.6	1.04	2255	140.6	0.83	1567	140.2	0.96	.002
MCC: ≥4
SBP (mm Hg)	2812	132.3	0.40	965	132.7	0.72	779	131.2	0.74	730	132.6	0.77	338	132.6	1.03	.9140
DBP (mm Hg)	2812	78.0	0.24	965	77.6	0.42	779	77.5	0.43	730	78.6	0.47	338	78.4	0.62	.0770
HbA_1c	2919	6.3	0.02	1046	6.4	0.04	796	6.4	0.04	742	6.2	0.04	335	6.03	0.05	.0011
HDL-C (mg/dL)	2917	50.7	0.27	1035	49.6	0.47	795	50.8	0.52	750	51.4	0.54	337	52.3	0.80	<.001
Total cholesterol	2935	181.7	0.79	1046	178.5	1.34	802	184.3	1.52	749	181.6	1.55	338	185.5	2.14	.0021
Non-HDL-C (mg/dL)	2903	131.3	0.71	1032	129.2	1.2	792	133.9	1.38	744	130.4	1.41	335	133.4	1.94	.0845

Abbreviations: DBP, diastolic blood pressure; HbA_1c, glycosylated hemoglobin; HDL-C, high-density lipoprotein cholesterol; MCC, multiple chronic conditions; SBP, systolic blood pressure; SE, standard error.

^a Physical activity intensity calculated based on metabolic equivalent of tasks (METs) and correlations between cardiometabolic risk factors and physical activity levels are unadjusted.

The unadjusted and adjusted estimates of the association between LIPA and cardiometabolic risk factors by multiple chronic condition categories are presented (Table 3). Light-intensity physical activity was independently associated with higher HDL-C (β = 1.25; 95% confidence interval [CI]: 0.46-2.05) in the fully adjusted models. When stratified by number of chronic conditions, LIPA was positively associated with HDL-C among individuals with 0 to 1 multiple chronic conditions (β = 2.21; 95% CI: 0.47-3.96). However, after adjustment, the association was attenuated and no longer significant (β = 0.89; 95% CI: −0.84 to 2.60). Light-intensity physical activity remained significantly associated with HDL-C for those who indicated 2 to 3 chronic conditions (β = 1.73; 95% CI: 0.58-2.89) in the unadjusted and fully adjusted models. For total cholesterol, we observed an independent association before (β = 6.89; 95% CI: 4.70-9.07) and after (β = 2.72; 95% CI: 0.53-4.90) adjusting for confounders; however, when stratified by multiple chronic conditions, the estimates did not reach statistical significance in the fully adjusted models. Light-intensity physical activity was associated with lower levels of non-HDL-C in the unadjusted models (β = 4.43; 95% CI: 2.44-6.42); however, after adjusting for confounders, the association was no longer significant (β = 1.47; 95% CI: −0.53 to 3.48). Significant associations observed in fully adjusted models remained after applying a Bonferroni correction. There were no significant associations between LIPA and blood pressure (diastolic or systolic) or HbA_1c or significant sex interactions.

Table 3.

Unadjusted and Adjusted^a Estimates and 95% Confidence Intervals of the Associations Between Light-Intensity Physical Activity and Cardiometabolic Risk Factors by Multiple Chronic Conditions, Health and Retirement Study (2010, 2012).

		MCC
	Total	0-1	2-3	>4
Systolic blood pressure (mm Hg)
Unadjusted	−2.13 (−3.18 to −1.09)	−2.93 (−5.07 to −0.80)	−2.19 (−3.73 to −0.67)	−1.35 (−3.34 to 0.63)
Fully adjusted	−0.57 (−1.61 to 0.48)	−0.99 (−3.03 to 1.05)	−0.78 (−2.33 to 0.76)	−0.66 (−2.69 to 1.37)
Diastolic blood pressure (mm Hg)
Unadjusted	−0.18 (−0.79 to 0.43)	−1.21 (−2.47 to 0.04)	−0.37 (−1.27 to 0.52)	0.01 (−1.17 to 1.19)
Fully adjusted	−0.67 (−1.29 to −0.06)	−1.23 (−2.46 to 0.04)	−0.96 (−1.84 to −0.07)	−0.68 (−1.87 to 0.51)
HbA_1c
Unadjusted	−0.01 (−0.10 to 0.14)	−0.03 (−0.11 to 0.06)	0.03 (−0.05 to 0.11)	−0.09 (−0.12 to 0.10)
Fully adjusted	−0.01 (−0.06 to 0.04)	−0.02 (−0.11 to 0.06)	0.02 (−0.06 to 0.10)	0.01 (−0.10 to 0.13)
HDL-C (mg/dL)
Unadjusted	2.47 (1.67 to 3.27)	2.21 (0.47 to 3.96)	2.72 (1.56 to 3.87)	1.28 (−0.09 to 2.65)
Fully adjusted	1.25 (0.46 to 2.05)	0.89 (−0.84 to 2.60)	1.73 (0.58 to 2.89)	0.60 (−0.78 to 1.98)
Total cholesterol (mg/dL)
Unadjusted	6.89 (4.70 to 9.07)	1.76 (−2.77 to 6.29)	7.10 (3.86 to 10.26)	5.84 (1.92 to 9.78)
Fully adjusted	2.72 (0.53 to 4.90)	−0.56 (−5.14 to 4.01)	3.13 (−0.08 to 6.34)	2.93 (−1.00 to 6.86)
Non-HDL-C (mg/dL)
Unadjusted	4.43 (2.44 to 6.42)	−1.04 (−5.19 to 3.11)	4.54 (1.63 to 7.47)	4.72 (1.16 to 8.29)
Fully adjusted	1.47 (−0.53 to 3.48)	−2.06 (−6.25 to 2.13)	1.59 (−1.37 to 4.54)	2.47 (−1.11 to 6.05)

Abbreviations: HbA_1c, glycosylated hemoglobin; HDL-C, high-density lipoprotein cholesterol; MCC, multiple chronic conditions.

^a Adjusted for age, gender, race/ethnicity, education, income, marital status, insurance, smoking status, self-rated health, BMI, functional limitations, and cardiovascular medication use.

Discussion

This study investigated the independent association between LIPA and cardiometabolic risk factors among middle-aged and older adults. The main finding of this study showed that LIPA was associated with favorable HDL-C and total cholesterol after adjusting for several relevant confounders, including moderate and vigorous physical activity levels.

The strongest evidence to support recommending LIPA to improve cardiometabolic risk factor control was observed for HDL-C and total cholesterol. Our cross-sectional results are consistent with studies demonstrating an association between LIPA and favorable HDL-C and total cholesterol levels among middle-aged and older adults.^{5,17,20,41,44} Several of these studies used accelerometer data from the National Health and Nutrition Examination Survey.^24,44 Although these were cross-sectional studies, they approached looking at LIPA differently from the present study. For example, Camhi et al found that a greater amount of time spent performing LIPA was associated with a lower of odds of HDL-C.²⁴ In the present study, the association between LIPA and HDL-C remained robust only among individuals with 2 to 3 chronic conditions, who represent the majority of our study population (45.4%). The unanticipated absence of a gradient relationship between LIPA and the number of multiple chronic conditions may have occurred for several reasons. It is possible that the potential benefit of engaging in LIPA may not extend to relatively healthy middle-aged and older adults (those with none or only 1 chronic condition). Additionally, engaging in LIPA may not be effective for those with a higher chronic disease burden (≥4 chronic conditions) who typically have greater complex health-care needs and functional impairment. These findings contrast with some studies that did not find an association with HDL-C. For example, in a sample of adults with mobility limitations, one study did not find a significant association with LIPA.⁴⁵

Although there are some studies that have reported null associations with blood pressure and HbA_1c,^20,24 the results from our study are somewhat inconsistent with several prior studies that showed engaging in LIPA was associated with lower systolic blood pressure and HbA_1c.^13,45 One explanation for the lack of association observed between LIPA and blood pressure and HbA_1c may be related to diet. The HRS did not allow us to determine dietary patterns, which has a strong influence on HbA_1c and blood pressure,^46,47 and presumably could have an adverse impact by elevating these cardiometabolic factors, thereby diminishing or nullifying the effect of LIPA on these outcomes. Another reason may be related to the overall CVD risk of the population. The findings from some studies suggest that the benefits of LIPA varied by CVD risk of the population. For example, Gay et al did not find an association between LIPA and HbA_1c levels among individuals who were in the moderate- and high-risk diabetes groups but showed a weak association between LIPA and HbA_1c among individuals in the low-risk diabetes groups.²³ Additionally, more recent studies are disaggregating LIPA into low (“low-light”) and high (“high-light”) zones.^16,18,48,49 Among studies that examined high and low LIPA in relation to a health outcome, the pattern of findings suggests the beneficial effect for LIPA is driven by values in the “high” LIPA zones.^17,18,50 In our study, we were not able to differentiate between low and high LIPA, which may in part explain the null association, especially if the benefits of LIPA are largely derived from high LIPA.

There is available evidence from a limited number of studies that found either adverse or null associations,^16,24 which aligns with the results of our analysis. For example, one study that measured physical activity using an ActiGraph accelerometer showed an adverse association between LIPA and systolic blood pressure. Relatedly, there are several studies that measured fasting plasma glucose and did not find an association with LIPA.^13,24,44 Another study showed the LIPA was not independently associated with the Framingham risk score, which is an indicator to predict 10-year risk of developing CVD.⁵¹

Our study has several strengths. The HRS is a diverse nationally representative study cohort. It has a large sample size, and black and Hispanic populations are oversampled. The cardiometabolic risk factors were objectively assessed. Also, our sample included individuals with multiple chronic conditions, where prior studies have typically excluded this population or adjusted for this in the analyses. The results from this study should be interpreted with caution given several limitations. First, our results are cross-sectional and causality between physical activity intensity and cardiometabolic risk factors cannot be determined definitively. Physical activity and demographic characteristics in HRS are based on self-reported questionnaires, which may have reduced the validity and reliability of the exposure and some of the covariates. The use of self-reported physical activity may have resulted in recall and social desirability bias. For example, participants may have over- or underreported their engagement in light, moderate, or vigorous physical activity. Prior studies have shown that social desirability bias may drive participants to overreport their physical activity intensity, duration, and frequency.³¹ Health and Retirement Study does not provide information on the duration (ie, minutes) of physical activity, which could have resulted in nondifferential exposure misclassification and an underestimation of the association between LIPA and cardiometabolic risk factors. Moreover, our measure of physical activity does not capture the totality of activity and does not measure activity related to occupation and transportation. There is some degree of variability as to how self-reported LIPA is defined, making it difficult to compare across studies. For example, in our study, LIPA was defined as vacuuming, laundry, and home repairs; however, other studies defined LIPA to include bicycling and walking, which were categorized as moderate-intensity physical activity in our data.⁵² Device-assessed measures of physical activity are ideal and could enhance the measurement of physical activity intensity and duration.¹⁶ Nonetheless, there are some studies that have shown self-reported physical activity among middle-aged and older adults to have moderate validity against accelerometer-measured physical activity.⁵³ The number of chronic conditions examined in studies investigating multimorbidity varies, ranging from 4 to 147 different conditions.⁵⁴ A minimum of 12 chronic conditions has been suggested to stably estimate the prevalence of multimorbidity.⁵⁵ Although the number of chronic conditions in this study is below the recommended minimum, the chronic conditions captured in our sample include hypertension, arthritis, and diabetes, which represent the most prevalent chronic conditions among middle-aged and older adults.

The present study extends previous knowledge about LIPA and cardiometabolic risk factors. Few published studies have reported on relationships between LIPA and cardiometabolic risk factors stratified by number of multiple chronic conditions. Although there was an absence of a chronic disease burden gradient between LIPA and HDL-C and total cholesterol, our findings still suggest that independent of moderate and vigorous physical activity, engaging in LIPA is associated with favorable HDL-C and total cholesterol outcomes. Light-intensity physical activity may be a practical alternative to moderate and vigorous physical activity for managing CVD health, particularly among population subgroups that suffer from multiple chronic conditions. However, the results of this study need to be replicated in longitudinal studies to further strengthen the quality of the evidence base demonstrating the impact of LIPA on the health outcomes of middle-aged and older adults. As the evidence of quantifiable health benefits of LIPA continues to emerge, a reevaluation of the current physical activity guidelines, particularly for middle-aged and older adults, is warranted.

So What? Implications for Health Promotion Practitioners and Researchers

What is already known on this topic?

Engaging in moderate and vigorous physical activity is an important primary prevention strategy to prevent adverse cardiovascular health outcomes and secondary prevention strategy to minimize the severity of CVD. However, middle-aged and older adults are least likely to meet the recommended physical activity guidelines.

What does this article add?

Light-intensity physical activity is a practical alternative to moderate and vigorous physical activity to manage modifiable cardiometabolic risk factors, such as high-density lipoprotein and total cholesterol, particularly among middle-aged and older adults with multiple chronic conditions.

What are the implications for health promotion practice or research?

Favorable cardiometabolic health outcomes among middle-aged and older adults may be associated with engaging LIPA. If the results of our study are confirmed in longitudinal studies, the findings will underscore the need to reevaluate current physical activity guidelines to include LIPA.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Kellee White, PhD

References

Centers for Disease Control and Prevention, National Center for Health Statistics. Multiple Cause of Death 1999-2015 on CDC WONDER Online Database, released December 2016. Data are from the Multiple Cause of Death Files, 1999-2015, as compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program. http://wonder.cdc.gov/mcd-icd10.html. Accessed May 31, 2017.

Centers for Disease Control and Prevention (CDC). Million hearts: strategies to reduce the prevalence of leading cardiovascular disease risk factors—United States, 2011. MMWR Morb Mortal Wkly Rep. 2011;60(36):1248–1251.

Assah

Brage

Ekelund

Wareham

. The association of intensity and overall level of physical activity energy expenditure with a marker of insulin resistance. Diabetologia. 2008;51(8):1399–1407.

Jung

Bourne

Beauchamp

Robinson

Little

. High-intensity interval training as an efficacious alternative to moderate-intensity continuous training for adults with prediabetes. J Diabetes Res. 2015; 2015:9.

Young

Coleman

Ngor

Reynolds

Sidell

Sallis

. Associations between physical activity and cardiometabolic risk factors assessed in a Southern California health care system, 2010-2012. Prev Chronic Dis. 2014;11:E219.

Jefferis

Sartini

Lee

. Adherence to physical activity guidelines in older adults, using objectively measured physical activity in a population-based study. BMC Public Health. 2014;14:382.

DiPietro

. Physical activity in aging: changes in patterns and their relationship to health and function. J Gerontol A Biol Sci Med Sci. 2001;56(2):13–22.

Cardinal

. Preparticipation physical activity screening within a racially diverse, older adult sample: comparison of the original and revised physical activity readiness questionnaires. Res Q Exerc Sport. 2000;71(3):302–307.

Cardinal

Vuchinich

. Health worry, physical activity participation, and walking difficulty among older adults: a mediation analysis. Res Q Exerc Sport. 2009;80(1):12–21.

10.

Murphy

. Review of physical activity measurement using accelerometers in older adults: considerations for research design and conduct. Prev Med. 2009;48(2):108–114.

11.

Burton

Khan

Brown

Turrell

. The association between sedentary leisure and physical activity in middle-aged adults. Br J Sports Med. 2012;46(10):747–752.

12.

Washburn

Heath

Jackson

. Reliability and validity issues concerning large-scale surveillance of physical activity. Res Q Exerc Sport. 2000;71(2): S104–S113.

13.

Loprinzi

Lee

Cardinal

. Evidence to support including lifestyle light-intensity recommendations in physical activity guidelines for older adults. Am J Health Promot. 2015;29(5):277–284.

14.

Colberg

Sigal

Yardley

. Physical activity/exercise and diabetes: a position statement of the American Diabetes Association. Diabetes Care. 2016;39(11):2065–2079.

15.

Gando

Murakami

Kawakami

. Light-intensity physical activity is associated with insulin resistance in elderly Japanese women independent of moderate- to vigorous-intensity physical activity. J Phys Act Health. 2014;11(2):266–271.

16.

LaMonte

Lewis

Buchner

. Both light intensity and moderate-to-vigorous physical activity measured by accelerometry are favorably associated with cardiometabolic risk factors in older women: the Objective Physical Activity and Cardiovascular Health (OPACH) study. J Am Heart Assoc. 2017;6(10):e007064.

17.

Fuzeki

Engeroff

Banzer

. Health benefits of light-intensity physical activity: a systematic review of accelerometer data of the National Health and Nutrition Examination Survey (NHANES). Sports Med. 2017;47(9):1769–1793.

18.

Howard

Winkler

EAH

Sethi

. Associations of low- and high-intensity light activity with cardiometabolic biomarkers. Med Sci Sports Exerc. 2015;47(10):2093–2101.

19.

Loprinzi

. Effects of light-intensity physical activity on red blood cell distribution width: implications for a novel mechanism through which light-intensity physical activity may influence cardiovascular disease. Int J Cardiol. 2016;203:724–725.

20.

Kim

Tanabe

Yokoyama

Zempo

Kuno

. Objectively measured light-intensity lifestyle activity and sedentary time are independently associated with metabolic syndrome: a cross-sectional study of Japanese adults. Int J Behav Nutr Phys Act. 2013;10:30.

21.

Cornelissen

Arnout

Holvoet

Fagard

. Influence of exercise at lower and higher intensity on blood pressure and cardiovascular risk factors at older age. J Hypertens. 2009;27(4):753–762.

22.

Goto

Higashi

Kimura

. Effect of different intensities of exercise on endothelium-dependent vasodilation in humans: role of endothelium-dependent nitric oxide and oxidative stress. Circulation. 2003;108(5):530–535.

23.

Gay

Buchner

Schmidt

. Dose-response association of physical activity with HbA1c: intensity and bout length. Prev Med. 2016;86:58–63.

24.

Camhi

Sisson

Johnson

Katzmarzyk

Tudor-Locke

. Accelerometer-determined moderate intensity lifestyle activity and cardiometabolic health. Prev Med. 2011;52(5):358–360.

25.

Chen

Shih

Tarng

. Impact of physical activity on the association between lipid profiles and mortality among older people. Sci Rep. 2017;7(1):8399.

26.

Fox

Liao

Sun

Chen

. Prospective associations of objectively assessed physical activity at different intensities with subjective well-being in older adults. Qual Life Res. 2016;25(11):2909–2919.

27.

Sallis

Saelens

. Assessment of physical activity by self-report: status, limitations, and future directions. Res Q Exerc Sport. 2000;71(2):S1–S14.

28.

Prince

Adamo

Hamel

Hardt

Gorber

Tremblay

. A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act. 2008;5:56.

29.

Troiano

Berrigan

Dodd

Masse

Tilert

McDowell

. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008;40(1):181–188.

30.

Pettee Gabriel

Morrow

Jr Woolsey

. Framework for physical activity as a complex and multidimensional behavior. J Phys Act Health. 2012;9(Suppl. 1):S11–S18.

31.

Troiano

Gabriel

KKP

Welk

Owen

Sternfeld

. Reported physical activity and sedentary behavior: why do you ask? J Phys Act Health. 2012;9(suppl 1):S68–S75.

32.

Heeringa

Connor

. Technical Description of the Health and Retirement Study Sample Design. Ann Arbor, MI: Survey Research Center, University of Michigan; 1995.

33.

Juster

Suzman

. An overview of the health and retirement study. J Hum Resour. 1995;30:S7–S56.

34.

XXZ

Baker

. Differences in leisure-time, household, and work-related physical activity by race, ethnicity, and education. J Gen Intern Med. 2005;20(3):259–266.

35.

Dipietro

Caspersen

Ostfeld

Nadel

. A survey for assessing physical activity among older adults. Med Sci Sports Exerc. 1993;25(5):628–642.

36.

Latham

Williams

. Does neighborhood disorder predict recovery from mobility limitation? Findings from the Health and Retirement Study. J Aging Health. 2015;27(8):1415–1442.

37.

Meyer

MRU

Janke

Beaujean

. Predictors of older adults’ personal and community mobility: using a comprehensive theoretical mobility framework. Gerontologist. 2014;54(3):398–408.

38.

Ainsworth

Caspersen

Matthews

Masse

Baranowski

Zhu

. Recommendations to improve the accuracy of estimates of physical activity derived from self report. J Phys Act Health. 2012;9(suppl 1):S76–S84.

39.

Crimmins

Faul

Kim

. Documentation of Biomarkers in the 2006 and 2008 Health and Retirement Study. Ann Arbor, MI: Survey Research Center University of Michigan; 2013.

40.

Eileen

Heidi

Kenneth

Mary Beth

Robert

David

. Documentation of Physical Measures, Anthropometrics and Blood Pressure in the Health and Retirement Study. Ann Arbor, MI: Institute for Social Research, University of Michigan; 2008.

41.

Pescatello

Murphy

Costanzo

. Low-intensity physical activity benefits blood lipids and lipoproteins in older adults living at home. Age Ageing. 2000;29(5):433–439.

42.

Hung

Ross

Boockvar

Siu

. Recent trends in chronic disease, impairment and disability among older adults in the United States. BMC Geriatrics. 2011;11:47.

43.

Stenholm

Westerlund

Head

. Comorbidity and functional trajectories from midlife to old age: the Health and Retirement Study. J Gerontol A Biol Sci Med Sci. 2015;70(3):330–336.

44.

Buman

Winkler

EAH

Kurka

. Reallocating time to sleep, sedentary behaviors, or active behaviors: associations with cardiovascular disease risk biomarkers, NHANES 2005-2006. Am J Epidemiol. 2014;179(3):323–334.

45.

Loprinzi

Sheffield

Tyo

Fittipaldi-Wert

. Accelerometer-determined physical activity, mobility disability, and health. Disabil Health J. 2014;7(4):419–425.

46.

Sami

Ansari

Butt

Hamid

MRA

. Effect of diet on type 2 diabetes mellitus: a review. Int J Health Sci (Qassim). 2017;11(2):65–71.

47.

Ndanuko

Tapsell

Charlton

Neale

Batterham

. Dietary patterns and blood pressure in adults: a systematic review and meta-analysis of randomized controlled trials. Adv Nutr. 2016;7(1):76–89.

48.

Buman

Hekler

Haskell

. Objective light-intensity physical activity associations with rated health in older adults. Am J Epidemiol. 2010;172(10):1155–1165.

49.

Izawa

Shibata

Ishii

Miyawaki

Oka

. Associations of low-intensity light physical activity with physical performance in community-dwelling elderly Japanese: a cross-sectional study. PLoS One. 2017;12(6):e0178654.

50.

Beddhu

Wei

Marcus

Chonchol

Greene

. Light-intensity physical activities and mortality in the United States general population and CKD subpopulation. Clin J Am Soc Nephrol. 2015;10(7):1145–1153.

51.

Hawkins

Gabriel

Conroy

Cooper

Sutton-Tyrrell

. Physical activity intensity and cardiovascular risk by ankle-brachial index. Vasc Med. 2013;18(2):79–84.

52.

Schuna

Johnson

Tudor-Locke

. Adult self-reported and objectively monitored physical activity and sedentary behavior: NHANES 2005-2006. Int J Behav Nutr Phys Act. 2013;10:126.

53.

Falck

McDonald

Beets

Brazendale

Liu-Ambrose

. Measurement of physical activity in older adult interventions: a systematic review. Br J Sports Med. 2016;50(8):464–470.

54.

Willadsen

Bebe

Koster-Rasmussen

. The role of diseases, risk factors and symptoms in the definition of multimorbidity—a systematic review. Scand J Prim Health Care. 2016;34(2):112–121.

55.

Fortin

Stewart

Poitras

M-E

Almirall

Maddocks

. A systematic review of prevalence studies on multimorbidity: toward a more uniform methodology. Ann Fam Med. 2012;10(2):142–151.