Individuals aged >40 years with cardiovascular disease or its risk factors who were residing in Hongcheon-gun, Gangwon Province, were recruited as study subjects. This study was conducted from April to September 2018. The inclusion criteria for this study were as follows: individuals who had relevant diseases or at least two cardiovascular risk factors and could initiate moderate-intensity exercise without preparticipation medical clearance according to the recommendations of the American College of Sports Medicine (ACSM). (1) Subjects with known cardiovascular (cardiac, peripheral vascular, or cerebrovascular), renal, or metabolic (type 1 or 2 diabetes mellitus) disease who already performed regular exercises [10] or (2) those with no known cardiovascular, renal, or metabolic disease but with two or more risk factors [11] could continue with moderate-intensity exercise without medical clearance according the ACSM’s guidelines. Current exercisers should have a history of performing planned, structured physical activity of at least moderate intensity for at least 30 minutes for 3 or more days per week during the past 3 months. Risk factors included age (≥45 years for men, ≥55 years for women), family history (myocardial infarction, coronary revascularization, or sudden death before 55 years of age in the father or other male first-degree relative or before 65 years of age in the mother or other female first-degree relative), cigarette smoking (current cigarette smoker, those who quit within the previous 6 months, or those exposed to environmental tobacco smoke), sedentary lifestyle (non-participation in at least 30 minutes of moderate-intensity physical activity [40% to <60% VO₂ reserve] for at least 3 days per week for at least 3 months), obesity (body mass index [BMI] ≥25 kg/m² or waist circumference >90 cm for men and >85 cm for women), hypertension, dyslipidemia (total cholesterol ≥200 mg/dL, low-density lipoprotein cholesterol [LDL-C] ≥130 mg/dL, high-density lipoprotein cholesterol [HDL-C] <40 mg/dL, or use of lipid-lowering medications), and prediabetes (hemoglobin A1c [HbA1c] of 5.7–6.4%). HDL-C ≥60 mg/dL was considered a negative risk factor, and one positive risk factor was subtracted from the sum of positive risk factors [11].

The following were the exclusion criteria for this study: pain or discomfort in the chest, neck, jaw, arms, or other areas that might have resulted from ischemia; shortness of breath at rest or with mild exertion; dizziness or syncope; orthopnea or paroxysmal nocturnal dyspnea; ankle edema; palpitations or tachycardia; intermittent claudication; known heart murmur; or unusual fatigue or shortness of breath with usual activities [10]. Subjects with scores <20 on the Korean version of the Mini-Mental State Examination were also excluded.

We assumed that the change in maximal oxygen uptake as a result of the cardiac rehabilitation program would be 2.5±4.2 mL/kg/min. Sample size was estimated as 25 with an alpha level of 0.05 and a beta level of 0.20. We expected the drop-out rate to be 20%, and the final sample size was calculated as 31.

This study was conducted in accordance with the principles embodied in the Declaration of Helsinki. In particular, the study protocol was approved by the Institutional Review Board of Kangwon National University Hospital (No. 2018-02-009), and written informed consent was obtained from all study subjects.

CBCR was carried out at the public health center in Hongcheon-gun, Gangwon Province. A cardiac rehabilitation program was constructed using human resources and equipment available at the health center in order to conduct it on a community basis. The medical staff who participated in this program and the exercise equipment are listed in Figs. 1 and 2, respectively. The program consisted of 10 education sessions and 20 weeks of customized exercises and was run for a total of 22 weeks (5 months), with 1 week for pre-assessment, 20 weeks for education and exercise, and 1 week for post-assessment. Interim assessment was conducted during the 11th week of the program. The exercise program was regularly supervised by physical therapists, and exercises were performed twice a week at moderate intensity, with each exercise session lasting for 1 hour and comprising 30-minute aerobic exercise and 30-minute strengthening training (Table 1). The recommended rating for the perceived exertion score was no more than 11–14.

Blood pressure, chest pain, and respiratory difficulty were evaluated before and during each exercise and assessment session. Chest pain and respiratory difficulty were assessed using the angina scale (0=none, 1=little, 2=medium, 3=very, 4=extreme) and dyspnea scale (0=none, 1=little, 2=medium, 3=very, 4=extreme), respectively. To ensure the safety of the community-based exercise program, we set the criteria for not-participating/stopping exercise or assessment sessions by adapting the blood pressure and chest pain criteria of ACSM’s indications for terminating exercise testing and indications and contraindications for inpatient and outpatient cardiac rehabilitation [10,11]. Subjects with blood pressure of at least 160/100 mmHg, score of ≥1 on the angina scale, and score of ≥1 on the dyspnea scale were not allowed to participate in exercise or assessment sessions. Exercise or assessment was stopped when the blood pressure was ≥250/115 mmHg, the angina scale score was ≥2, and the dyspnea scale score was ≥3, or if the subject wanted the exercise to be terminated. If there were a drop in systolic blood pressure (SBP) of ≥10 mmHg with an increase in exercise intensity, the exercise or assessment session was discontinued.

Three evaluations were conducted prior to the program (pre-assessment, T0), at the 11th week of the program (interim assessment, T1), and after completion (post-assessment, T2) (Table 1). Waist circumference (cm), height (cm), body mass (kg), resting blood pressure (mmHg), and resting heart rate (bpm) were measured. BMI was calculated as weight (kg) divided by height squared (m²). Fasting blood sugar (FBS; mg/dL), HbA1c (%), LDL-C (mg/dL), HDL-C (mg/dL), total cholesterol (mg/dL), and triglyceride (mg/dL) were measured from blood samples.

The Korean Activity Scale/Index (KASI), which consists of 15 items, was used to estimate functional status [12]. For the KASI, the weight is used to calculate the score (range from 0 to 79, with high scores indicating good functional status). Physical activity for 7 days was assessed using the Korean version of the International Physical Activity Questionnaire (IPAQ) short form [13]. We calculated MET-minute per week scores. The European Quality of Life-5 Dimension-3-Level version (EQ-5D-3L), which assesses mobility, self-care, usual activities, pain/discomfort, and anxiety/depression at three levels of functioning, was used to measure health-related quality of life [14]. The EQ-5D-3L index (range from -0.171 to 1, with higher values indicating better health status) was calculated using the valuation set by a previous report [15].

Exercise capacity was evaluated using the 6-minute walk distance (6MWD) test following the recommended guidelines [16]. The 6MWD test was performed by physiotherapists in accordance with the protocol outlined by the American Thoracic Society. The guideline recommends testing in a long corridor; however, owing to the absence of a long corridor at the public health center, we decided to conduct the 6MWD test on a track in a small park adjacent to the health center (Fig. 2C). Recently, a previous study has shown that 6MWD accurately estimates maximal oxygen uptake [17]. Peak VO₂ (mL/kg/min) was estimated as 4.948+0.023×6MWD (m).

The Mann-Whitney U test was used to compare young-old (65–74 years) and old-old (≥75 years) women with respect to baseline characteristics, including height, body mass, BMI, waist circumference, resting SBP/DBP, resting HR, and number of risk factors. The Fisher exact test and chi-square test were employed for the comparison of the history of myocardial infarction and presence of diabetes mellitus, respectively, between young-old and old-old women. The results of evaluation at 3 points were analyzed using repeated-measures analysis of variance (ANOVA). For the post-hoc test, the Tukey method was employed. Comparison between young-old and old-old women was conducted on variables with significant change between T0 and T2 in the post-hoc Tukey test. For these variables, the paired values measured at T0 and T2 were compared using the Wilcoxon rank test. The difference between T0 and T2 values was calculated and compared between young-old and old-old women using the Mann-Whitney U test. Statistical analysis was performed using jamovi version 1.1.9 (https://www.jamovi.org/), a free open-source graphical user interface for the R software, and p<0.05 was considered statistically significant.

Out of 31 subjects, 21 completed the program (completion rate, 67.7%). Altogether, 10 individuals (5 who dropped out, 3 who were admitted to the hospital, and 2 who moved out) were lost to follow-up (Fig. 3). None of the subjects, including the 10 individuals who were lost to follow-up, met the criteria for exclusion from or discontinuation of the exercise program due to an increase in blood pressure, angina scale score, and dyspnea scale score.

The baseline characteristics of subjects are summarized in Table 2. The average age of subjects was 75.0±7.8 years; among the subjects, 1 was under 65 years of age (52 years old), 9 were 65–74 years of age, and 11 were ≥75 years of age. There were 18 female subjects (85.7%). One patient had a history of myocardial infarction, and none had any cardiovascular disease (angina, arrhythmia, heart failure, cerebrovascular disease, or peripheral arteriosclerosis) or renal disease; 10 subjects had diabetes mellitus, and none had a family history of cardiac disorders. All subjects did not have a sedentary lifestyle and were performing moderate-intensity exercise more than 3 times per week. None of the subjects smoked.

Female subjects aged ≥65 years (n=17) were categorized into young-old (65–74 years; n=8) and old-old (≥75 years; n=9) women. Old-old women were lighter in weight and, thus, had a smaller BMI than young-old women (p<0.05). Furthermore, waist circumference and diastolic blood pressure were significantly smaller and lower, respectively, in old-old women (both p<0.05). The old-old women were also shorter, albeit without statistically significant difference. No significant differences in the history of myocardial infarction and presence of diabetes mellitus were noted between young-old and old-old women. Additionally, the number of risk factors did not differ between the two groups (Table 2).

The outcomes of CBCR are summarized in Table 3. Resting SBP did not show a significant difference with time. The BMI and FBS significantly decreased according to T0, T1, and T2. The T0–T1 and T0–T2 differences were significant; however, the T1–T2 difference was not. HbA1c tended to decrease, although this was not statistically significant. The total cholesterol level changed in a different pattern. The total cholesterol level increased during T0–T1 but decreased during T1–T2; these differences were significant, whereas the T0–T2 difference was not. Changes in LDL-C, HDL-C and triglyceride were not statistically significant.

The physical activity derived from the IPAQ showed an increasing trend, and the functional capacity assessed by the KASI gradually increased; nevertheless, these did not reach statistical significance. As for the EQ-5D-3L index results, no significant changes in the quality of life were observed.

The 6MWD was 461.25±39.43 m initially (estimated VO₂, 15.56±0.91 mL/kg/min) and increased to 520.44±53.34 m at post-assessment (estimated VO₂, 16.92±1.23 mL/kg/min). This increase in 6MWD was statistically significant, and the changes in T0–T1, T0–T2, and T1–T2 were significant as well.

As presented in Table 3, BMI, FBS, and 6MWD were the variables with significant changes between T0 and T2 in the post-hoc test. Fig. 4 shows the changes in BMI, FBS, and 6MWD at T0, T1, and T2 in young-old and oldold women. We compared the paired values measured at T0 and T2. The paired differences in FBS and 6MWD were significant in both young-old and old-old groups (p<0.05); however, the paired difference in BMI was not. For all variables, the changes between T0 and T2 were not significantly different between young-old and old-old women (Table 4).