Changes in Exercise Patterns in Menopausal Women at Low–Intermediate Risk for Cardiovascular Disease: A Prospective Survey Study

Abstract

Introduction:

We aim to describe changes in exercise habits and barriers to physical activity over 5 years in menopausal women at low–intermediate risk for cardiovascular disease.

Materials and Methods:

Women in the Stress Echocardiography in Menopausal Women at Risk for Coronary Artery Disease trial were prospectively enrolled in a multisite study from 2004 to 2007. Inclusion criteria were as follows: peri- and postmenopausal women with symptoms and/or risk factors for cardiovascular disease resulting in referral for stress echocardiography. A questionnaire, which assessed details of medical history, physical activity, and body mass index (BMI), was administered at baseline and 5 years.

Results:

216 menopausal women (62.5% hypertensive, 15.3% diabetic, 52.3% prior or current smokers) were studied. At baseline, age was 54.9 ± 4.8 years, BMI was 30.7 ± 6.4 kg/m², and Framingham risk score was 4.05% ± 3.76%. One hundred women (46.3%) were obese, 79 (36.6%) overweight, and 37 (17.1%) had a normal BMI. Women changed their self-reported aerobic exercise patterns in similar patterns regardless of their BMI at baseline. There was low participation in strength training among all women. Mean BMI decreased by 0.12 kg/m² in obese women and increased by 1.63 kg/m² in normal BMI women at 5 years (p < 0.0001).

Conclusion:

These data suggest that women can increase their level of physical activity regardless of BMI and that overweight or obese status is not a barrier to initiating an aerobic, nonaerobic, or strength training exercise routine.

Introduction

The American Heart Association and the American College of Sports Medicine recommend that all healthy adults participate in moderate exercise for a minimum of 30 minutes 5 days per week or vigorous exercise for 20 minutes 3 days per week.¹ Similarly, the US Department of Health and Human Services recommends that all healthy adults engage in a total of 150 minutes of moderate intensity activity or 75 minutes of vigorous activity per week, or a combination of the two.² Because there is a dose–response relationship between physical activity and health, persons who wish to improve their personal fitness, reduce their risk for chronic diseases and disability, or prevent unhealthy weight gain may benefit from exceeding the minimum recommended amount of physical activity. Moreover, healthy adults are recommended to perform activities that maintain or increase muscular strength and endurance a minimum of 2 days each week as part of their overall fitness program.¹ There are several interventions that have been demonstrated to be effective in helping patients increase their level of physical activity.³ Lifestyle changes, particularly increasing physical activity, are associated with reduced risk for cardiac events in women.^4
–6 Increased physical activity is also associated with decreased incidence and mortality from all cardiovascular and coronary heart disease.

Aerobic exercise has also been found to decrease blood pressure and raise “good” cholesterol (high density lipoprotein [HDL]).⁷ The American Heart Association recommends using resistance training as an adjunct to aerobic exercise training.⁸ Furthermore, there is evidence to suggest resistance training may reduce blood pressure.^9,10

The Stress Echocardiography in Menopausal Women at Risk for Coronary Artery Disease (SMART) trial was a 5-year prospective study aimed at evaluating the prognostic value of contrast stress echocardiography, coronary artery calcification, and cardiac biomarkers for the prediction of cardiovascular events in peri- or postmenopausal women experiencing chest pain symptoms or risk factors.¹¹ The current study is a 5-year prospective follow-up of a component of the SMART trial. Women in the SMART trial were asked to complete a health questionnaire at baseline, 2 and 5 years. We aim to evaluate how peri- or postmenopausal women may change their self-reported exercise habits and barriers to physical activity over a 5-year period. We hypothesized that women with an elevated body mass index (BMI) at baseline may be more motivated to change their health behaviors at 5-year follow-up.

Materials and Methods

Study design

This prospective multicenter study was approved by the Mayo Clinic Institutional Review Board and conducted at three Mayo Clinic group practice sites: Rochester, MN, Jacksonville, FL, and Scottsdale, AZ. Women were selected for enrollment based on referral to the stress echocardiography laboratory. Eligibility criteria included perimenopausal and postmenopausal women aged 40–65 with one of the following: atypical chest pain or exertional dyspnea with ≥2 cardiovascular risk factors, typical anginal chest pain with at least one cardiovascular risk factor, or no symptoms but ≥3 cardiovascular risk factors. Cardiovascular risk factors included current smoker or quit within the past 5 years, diabetes, family history of premature CAD, hypertension (or on treatment), dyslipidemia (or on treatment), or obesity (BMI ≥30). Women were stratified as low or intermediate risk based on their Framingham risk score (FRS), with a score of <10% corresponding with “low” risk and a score of 10%–20% corresponding with “intermediate” risk. Perimenopausal status was defined as the absence or irregularity of menstrual periods for 6–12 months, while postmenopausal status was defined as absence of periods for greater than or equal to 12 months. Exclusion criteria included previous confirmation of CAD, documented ejection fraction <50%, acute coronary syndrome, contraindication to exercise or pharmacologic stress testing, or known contraindications to the contrast agent. Women who had a cardiac event during the course of the study were also excluded for the purpose of this analysis.

Health behaviors questionnaire

Women were invited to complete the Women's Heart Clinic Risk Assessment Questionnaire (WHCRAQ) at the time of enrollment (baseline clinic visit) as well as 2 and 5 years after enrollment, delivered through US Postal Service. The WHCRAQ addressed employment, educational level, marital status, current health status, and major illness and health-related behaviors, that is, smoking, nutrition, exercise habits, menopausal status (perimenopausal or postmenopausal), hormone replacement therapy (HRT), medication use, over the counter remedies, including vitamin supplements, and BMI.

Outcomes of interest

Physical activity and barriers

The WHCRAQ includes detailed multiple choice questions addressing respondents' physical activity and exercise habits. Physical activity is defined as body movement produced by skeletal muscles and resulting in energy expenditure, while exercise indicates involvement in a regular and structured program of physical activity.

Questions included how often and for how much time women participated in aerobic exercise per week, how many times they participated in nonaerobic activity, and whether or not they participated in strength training activities per week. Examples of each type of activity were provided to help define each category. Aerobic exercise was defined as “physical activity lasting at least 20 minutes, during which you breathe more heavily and your heart beats faster.” Nonaerobic exercise was likened to bowling, softball, golf, walking, or yard work.¹ Women were also provided a list of eight barriers to increasing or maintaining a healthy level of exercise and allowed to select as many as apply (Appendix 1).

Weight and body mass index

Weight was measured in kilograms at baseline and at 5-year follow-up. BMI was defined as weight (kg)/height squared (m²). Subjects were categorized to obese (BMI ≥ 30 kg/m²), overweight (25 ≤ BMI < 30 kg/m²), and normal BMI (18.5 ≤ BMI < 25 kg/m²). Change in weight was defined as weight at 5-year follow-up minus weight at baseline, and measured in kilograms.

Statistical analyses

Clinical characteristics of patients were reported as proportions and percentage for categorical variables and mean (with standard deviation) for continuous variables with normal distribution. Differences between baseline and follow-up values were evaluated using Paired t-test or nonparametric Wilcoxon paired sample testing (for assessment of within-subject changes). Differences between groups were evaluated using ANOVA. All analyses were performed using JMP version 9.1 (SAS Institute, Inc., Cary, NC).

Results

From January 1, 2004, to December 31, 2007, 216 women (4.6% perimenopausal, 93.0% postmenopausal, 2.3% unknown (peri- vs. post-) menopausal status) were studied. Mean age at the beginning of the study was 54.9 ± 4.8 years. Mean BMI at the beginning of the study was 30.7 ± 6.4 kg/m². One hundred (46.3%) women were considered obese at baseline (BMI ≥ 30), 79 (36.3%) were overweight (25 ≤ BMI < 30), and 37 (17.1%) had a normal BMI (18.5 ≤ BMI < 25). At baseline, 62.5% of all women self-reported hypertension or treatment for hypertension, 15.3% had diabetes, and 52.3% were current or past smokers. Baseline average FRS was 4.05% ± 3.76%, with 91% of women considered “low” risk and 9% considered “intermediate” risk. There was no statistically significant difference between BMI groups with respect to age, hypertension, smoking history, FRS, or HRT use (Table 1).

Table 1.

Baseline Clinical Characteristics of Study Population (n = 216) Categorized by Body Mass Index

	Total (n = 216)	Normal BMI (n = 37)	Overweight BMI (n = 79)	Obese BMI (n = 100)	p
Age (years)	54.9 ± 4.8	55.0 ± 4.4	55.3 ± 4.4	54.5 ± 5.3	0.4781
BMI (kg/m²)	30.7 ± 6.4	22.8 ± 1.5	27.3 ± 1.4	36.2 ± 4.9	<0.0001
Hypertensive (%)	62.5	37.8	63.3	62.0	0.1027
Diabetic (%)	15.3	5.4	9.0	20.0	0.0339
Current smokers (%)	9.9	18.9	10.1	6.0	0.0957
Former smokers (%)	43.6	51.3	45.6	37.0	0.2556
Mean framingham risk (%)	4.05 ± 3.76	3.36 ± 3.04	4.45 ± 3.78	4.01 ± 3.98	0.3967
Current HRT (%)	23.3	16.2	27.8	22.3	0.5532
Prior HRT (%)	47.1	48.6	41.8	51.0
Multivitamin use	76	81.1	80.8	70.5	0.5025
High calcium foods	99	100	100	97.9	0.6472
Aerobic exercise per week (%)					0.1039
None	34.7	24.3	34.2	39
Less than 1	14.3	10.8	11.4	18
1–2	22.2	21.6	20.5	24
3+	28.7	43.2	34.2	19
Nonaerobic exercise per week (%)					0.0546
Less than 1 hour	5.6	2.7	16.5	26
1 hour	14.3	18.9	15.2	12
2 hours	22.2	32.4	19.0	21
3 hours	15.3	16.2	17.7	13
4+ hours	24.1	27.0	29.1	19
Strength training (%)	15.7	18.9	19.0	12	0.3744

Continuous variables are presented as mean ± SD; categorical variables are presented as percentage.

BMI, body mass index; HRT, hormone replacement therapy.

All of our participants resided in the continental United States, and 21 different states were represented in our study. Sixty-six percent (66%) of respondents had undergone their initial work-up at our Rochester campus, while 19% had undergone initial work-up at the Jacksonville, FL campus, and 15% at the Scottsdale, AZ campus.

Symptoms

As part of the SMART study, women self-reported their cardiac symptoms at study initiation, including chest pain, shortness of breath, palpitations, lower extremity edema, paroxysmal nocturnal dyspnea, and syncope. Among all women who reported symptoms, 40.6% reported chest pain or pressure with activity, 35.8% reported abnormal shortness of breath with activity, and 55.9% reported palpitations. A greater percentage of normal weight women reported chest pain than women who were overweight (51.3% vs. 35.4%, respectively), and obese women reported more shortness of breath at higher frequency than overweight or normal weight women (47.9% vs. 30.4% vs. 16.2%, respectively). There was no difference in the prevalence of self-reported syncope, paroxysmal nocturnal dyspnea, or orthopnea among the BMI categories at baseline. Women who were obese at baseline reported a greater frequency of lower extremity edema than women who were overweight or normal weight (53.1% vs. 21.5% vs. 18.9%, respectively).

A greater percentage of women who increased their self-reported frequency of aerobic activity over 5 years had chest pain at baseline: 57.4% of women who increased their self-reported frequency of aerobic activity versus 28.3% of women who decreased their self-reported frequency of aerobic activity. Similarly, among women who increased their self-reported frequency of aerobic activity, 46.3% were abnormally short of breath at baseline, whereas only 28.3% of women who decreased their self-reported frequency of aerobic activity were abnormally short of breath at baseline.

Women who decreased their self-reported frequency of light aerobic activity were more likely to have experienced syncope within the prior year than women who increased their self-reported frequency of light activity (5.3% vs. 0.0%, respectively). Women who increased their self-reported frequency of light aerobic activity were more likely to have lower extremity edema than those women who decreased their self-reported frequency of light aerobic activity (36.7% vs. 26.3%, respectively).

Changes in physical activity

At baseline, 141 (65.3%) women self-reported participating in any aerobic exercise. Two hundred four (94.4%) self-reported participating in nonaerobic exercise, and 34 (15.7%) reported participating in strength training each week.

Among the women with a normal BMI at baseline, 75.7% self-reported participating in aerobic exercise, 97.3% reported participating in nonaerobic exercise, and 18.9% reported participating in strength training. Among the women who were overweight at baseline, 65.8% reported participating in aerobic exercise, 97.5% reported participating in nonaerobic exercise, and 19.0% participated in strength training. Among the obese women, 61% participated in aerobic exercise weekly at baseline, 91% participated in nonaerobic exercise at baseline, and 12% participated in strength training at baseline (Table 1).

Comparisons of each woman's self-reported participation in each category of physical activity are listed in Table 2. There was no statistically significant difference among BMI groups in changes in self-reported aerobic exercise or light aerobic activity between baseline and 2 years, or between 2 and 5 years. Notably, women who were obese at baseline were less likely to decrease their self-reported frequency of aerobic exercise after 5 years compared to their normal weight peers (20.0% vs. 31.0%, p = 0.3218). However, more women with a normal BMI at baseline reported no change in their light physical activity, compared to those who were obese (57.6% vs. 40.2%, p = 0.0473). Women increased their self-reported aerobic exercise frequency at 5 years at similar rates (29.0%–31.0%) across all groups. Women with a low FRS at baseline were more likely than their intermediate risk peers to increase their physical activity over 5 years (34.3% vs. 7.1%, p = 0.0601).

Table 2.

Changes in Self-Reported Participation in Physical Activity Categorized by Body Mass Index at Baseline

	Normal (n = 37)	Overweight (n = 79)	Obese (n = 100)	p
Increased aerobic exercise	31.0	29.0	31.0	0.8826
Decreased aerobic exercise	31.0	21.7	20.0	0.3218
Quit aerobic exercise	8.6	8.9	7.0	0.8985
Started aerobic exercise	11.4	15.2	14.0	0.8091
No change in aerobic exercise	38.0	49.3	49.0	0.6719
Increased light exercise	33.3	34.8	29.2	0.6774
Decreased light exercise	9.1	31.8	30.4	0.0382
No change in light exercise	57.6	33.3	40.2	0.0473
Started strength training	8.1	7.7	7.0	0.9409
Quit strength training	8.1	9.2	8.0	0.9933
Never strength trained	59.4	73.8	64.0	0.8510
Always strength trained	10.8	9.2	4.0	0.3190

Categorical variables are presented as percentage.

Changes in BMI

BMI data were available through our electronic medical record at 5 years for 168 of the 216 women. Among all women, the mean change in BMI was 0.65 kg/m² (p = 0.0365). Mean BMI decreased by 0.12 kg/m² in obese women (p = 0.7758) and increased by 1.63 kg/m² in women with a normal BMI at 5 years (p = 0.0185). Mean BMI for all patients at 5 years was 31.2 ± 6.7 kg/m². Thirty-six (21.4%) of participants had a normal BMI at 5 years. Forty-five (26.8%) were overweight at 5 years and 87 (51.8%) were obese at 5 years. Forty-five (26.8% of respondents) had changed category of BMI over the 5-year period, 55% of which had gained weight placing them in a heavier BMI class and 45% of which now fell into a lower BMI class. Overall, 109 women (65% of respondents for whom BMI was available at 5 years) had gained weight over the 5-year period and 57 (35% of respondents for whom BMI was available at 5 years) had lost weight.

Relationship of changes in BMI and changes in physical activity

There was no statistically significant difference in changes in self-reported PA behaviors among the women who gained or lost weight over the 5-year study. Women who gained weight were slightly more likely to increase their physical activity at both 2 and 5 years (40% vs. 27%, p = 0.1957 and 33% vs. 27%, p = 0.6740, respectively). There was a greater difference in changes in self-reported light activity among the women who lost weight. Women who gained weight were less likely to increase their physical activity at 2 years compared to women who lost weight (28% vs. 43%, p = 0.0297). This held true at 5 years, with 23% of women who gained weight increasing their physical activity, compared to 39% of women who lost weight (p = 0.1037).

Changes in barriers to physical activity

Not surprisingly, lack of time and lack of motivation were the most frequently cited barriers to achieving or maintaining an adequate level of physical activity. Both normal weight and obese patients at baseline were more likely to cite no barriers to achieving ideal physical activity at 5 years. Across all patients, the barrier most likely to no longer be present at 5 years was “not enough time.” Complete results of barriers to physical activity are listed in Table 3.

Table 3.

Self-Reported Personal Barriers to Achieving or Maintaining an Ideal Level of Physical Activity

	Normal BMI (n = 37)			Overweight (n = 79)			Obese (n = 100)
	Baseline	5 years	p	Baseline	5 years	p	Baseline	5 years	p
Lack of appropriate facility or equipment	10.8	16.2	0.4217	3.8	5.1	0.7080	10.0	6.0	0.2873
No one to exercise with	13.5	8.1	0.4870	11.4	7.6	0.2594	19.0	22.0	0.4696
Physically unable to exercise	8.1	10.8	0.3240	20.2	21.5	0.8101	25.0	21.0	0.3197
Not enough time	43.2	21.6	0.0094	41.8	21.5	0.0029	45.0	28.0	0.0036
Don't like exercise	24.3	18.9	0.4217	20.2	12.6	0.1586	24.0	27.0	0.5800
Lack of motivation	43.2	48.6	0.4870	43.0	38.0	0.4532	51.0	51.0	1
Cost	0.0	0.0	1	5.1	5.1	1	5.0	11.0	0.0832
Other	5.4	5.4	1	8.9	10.2	0.7835	11.0	11.0	1
None	18.9	21.6	0.6609	19.0	19.0	1	6.0	8.0	0.5955

Categorical variables are presented as percentage.

Discussion

We present a prospective cohort study in menopausal women at low to intermediate risk for CAD addressing the patterns of self-reported physical activity over time. We found no significant relationship between women's BMI and their propensity to change their self-reported physical activity habits over a 5-year period without intervention. This suggests that interventions to promote physical activity should be aimed at all women, without regard to their BMI baseline. We did, however, find that women who are at intermediate cardiovascular risk would benefit from additional counseling to increase their aerobic exercise compared to their low risk peers. We found an overall low participation in strength training among all women, regardless of baseline BMI, identifying another potential area of intervention to increase women's physical activity.

Clinical implications

In their 2010 report, “Women's Health Research: Progress, Pitfalls, and Promise,” the Institute of Medicine examined the results of two decades of federally supported research into women's health. The report questions whether current research is adequately addressing determinants of health and whether findings of research are being communicated effectively to women.¹² In this study, we identify the self-reported barriers to physical activity, which is one social determinant of health. We also identify a large audience of women who would benefit from additional counseling on the importance of physical activity, particularly strength training.

Barriers to initiating and maintaining physical activity are highly individualized in the elderly. However, identifying common barriers and individual factors affecting exercise behavior is essential to help develop a more effective intervention associated with most participation and maintenance of physical activity and exercise behaviors.

Several authors have highlighted the unique challenges facing the elderly with respect to physical activity, including the time commitment and effort.^13,14 In our study, we have shown that lack of time to exercise became a less common barrier to physical activity over time. This may correlate with women's aging during the study. Many of the women in the study were near retirement age, creating flexibility within their schedules and additional time to devote to physical activity.

Our study does not identify or evaluate any mechanism to increase frequency of physical activity. Instead, we describe self-reported changes in women's exercise behaviors over time. Our findings can help guide practitioners in counseling their female patients on physical activity. Providers should be particularly attentive to counseling on resistance training, given the low frequency of weight training among the population studied. In addition, we find that women at low–intermediate risk for cardiovascular disease are at similar likelihood to negatively change their physical activity habits, and thus, the importance of affirming and maintaining optimal behaviors should be emphasized.

Interestingly, there was not a significant change in the percentage of women designating themselves physically incapable of exercise over a 5-year period. Lack of an exercise facility or equipment and cost is potentially confounded by the location and personal resources of the women in the study, but in our population these factors did not appear to be barriers to physical activity. Physical capabilities, personal preferences, and resources will need to be addressed on an individualized patient-by-patient basis when counseling on physical activity.

Our study has a major strength, which is the inclusion of a unique cohort of recently menopausal women identified as at low–intermediate risk for cardiovascular disease (mean age 54 years) and studied prospectively, while the focus of prior exercise literature has been primarily on healthy young men.¹⁵

Limitations

Some limitations in this current study need to be addressed. This study relies on women's responses to a health survey, creating a potential for recall bias, with a possibility of underreporting or overreporting of undesirable and desirable lifestyle behaviors. We worked to overcome this barrier by comparing women to themselves over time and directing the questions to women's present habits, not their prior habits. Moreover, the study has a selection bias. Women who are motivated to return a mailed health questionnaire may be more active in their health and thus may be more likely to exercise and identify fewer barriers to exercise. While our baseline demographics are similar with respect to age, there was a significant difference in our sample size between the various BMI groups, which partially explains the lack of statistical significance. We expect more overweight and obese women to be at elevated risk for cardiovascular disease, and therefore, our group sizes are similar to what may be expected in the population. Moreover, in practice we tend to address physical activity as part of weight loss, so overweight and obese women may be more likely to be directed to an intervention promoting physical activity. We did not explore the effect of physician's advice during a consult on changes of exercise behaviors. However, this was reported in a prior study showing that the elderly are more likely to change their levels of activity as a result of conversations with their physicians than those who do not receive exercise advice.¹⁶ Our study has the strength of a broad geographic catchment. However, we have limited data on these women's socioeconomic status, which may limit its external validity. Ninety-six percent of the women included were Caucasian, which also limits the study's external validity. Finally, we did not test for predictors of long-term exercise behaviors, as was shown in a previous study that illness was found to be the best predictor in a 10-year follow-up study of home-based walking in postmenopausal women.¹⁷

Conclusion

We find that in this cohort of early postmenopausal women at low to intermediate cardiovascular risk, self-reported physical activity habits may change over time, but these changes do not appear to be related to BMI at the beginning of the reporting period. As may be expected, more women with a normal BMI reported participation in aerobic exercise at baseline. However, these women were less likely to continue their aerobic exercise routine and as a result gained weight. Women who were at a higher BMI at baseline were more likely to reduce their BMI over time. This suggests the need for physical activity counseling addressing both intervention and maintenance aimed at all women, regardless of BMI.

Footnotes

Acknowledgments

The SMART trial (clinicalTrial.gov identifier# NCT00162370) was funded solely by Lantheus Medical Imaging (North Billerica, MA). Lantheus Medical Imaging provided the contrast agent (Definity) and additional research funding under grant #DMP 115-407.

Author Disclosure Statement

No competing financial interests exist.

Appendix 1

References

Haskell

, Lee

, Pate

, et al. Physical activity and public health: Updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation, 2007; 116:1081–1093.

Fulton

, Kohl

; US Department of Health and Human Services. 2008. Physical Activity Guildelines for Americans; be active, healthy, and happy! October, 2008.

Artinian

, Fletcher

, Mozaffarian

, et al. Interventions to promote physical activity and dietary lifestyle changes for cardiovascular risk factor reduction in adults: A scientific statement from the American Heart Association. Circulation, 2010; 122:406–441.

Haskell

, Alderman

, Fair

, et al. Effects of intensive multiple risk factor reduction on coronary atherosclerosis and clinical cardiac events in men and women with coronary artery disease. The Stanford Coronary Risk Intervention Project (SCRIP). Circulation, 1994; 89:975–990.

Gregg

, Cauley

, Stone

, et al. Relationship of changes in physical activity and mortality among older women. JAMA, 2003; 289:2379–2386.

Manson

, Greenland

, LaCroix

, et al. Walking compared with vigorous exercise for the prevention of cardiovascular events in women. N Engl J Med, 2002; 347:716–725.

Kesaniemi

, Danforth

Jr. , Jensen

, Kopelman

, Lefebvre

, Reeder

. Dose-response issues concerning physical activity and health: An evidence-based symposium. Med Sci Sports Exerc, 2001; 33:S351–S358.

Williams

, Haskell

, Ades

, et al. Resistance exercise in individuals with and without cardiovascular disease: 2007 update: A scientific statement from the American Heart Association Council on Clinical Cardiology and Council on Nutrition, Physical Activity, and Metabolism. Circulation, 2007; 116:572–584.

Cornelissen

, Fagard

. Effect of resistance training on resting blood pressure: A meta-analysis of randomized controlled trials. J Hypertens, 2005; 23:251–259.

10.

Kelley

, Kelley

. Progressive resistance exercise and resting blood pressure: A meta-analysis of randomized controlled trials. Hypertension, 2000; 35:838–843.

11.

Abdelmoneim

, Bernier

, Hagen

, et al. A multicenter, prospective study to evaluate the use of contrast stress echocardiography in early menopausal women at risk for coronary artery disease: Trial design and baseline findings. J Womens Health (Larchmt), 2013; 22:173–183.

12.

Institute of Medicine (US) Committee on Women's Health Research. Women's health research: Progress, pitfalls, and promise. Washington, DC: National Academies Press, 2010.

13.

Chao

, Foy

, Farmer

. Exercise adherence among older adults: Challenges and strategies. Control Clin Trials, 2000; 21:212S–217S.

14.

Ceari

, Vellas

, Hsu

, et al. A physical activity intervention to treat the frailty syndrome in older persons-results from the LIFE-P study. J Gerontol A Biol Med Sci, 2015; 70:216–222.

15.

King

, Blair

, Bild

, et al. Determinants of physical activity and interventions in adults. Med Sci Sports Exerc, 1992; 24:S221–S236.

16.

Balde

, Figueras

, Hawking

, Miller

. Physician advice to the elderly about physical activity. J Aging Phys Act, 2003; 11:90–97.

17.

Pereira

, Kriska

, Day

, Cauley

, LaPorte

, Kuller

. A randomized walking trial in postmenopausal women: Effects on physical activity and health 10 years later. Arch Intern Med, 1998; 158:1695–1701.