Examining the feasibility and preliminary efficacy of a group-based physical activity intervention integrating strength training among pre- and perimenopausal women: A randomized pilot trial

Abstract

Background:

Despite the well-documented benefits of physical activity, particularly strength training, for managing menopause-related physiological changes such as muscle loss and weight gain, few interventions specifically target pre- and perimenopausal women using scalable, remote delivery methods.

Objectives:

The purpose of this pilot study was to examine the feasibility and preliminary efficacy of a group-based physical activity and strength intervention delivered via Zoom for pre- and perimenopausal women.

Design:

This study was a randomized controlled trial lasting 12 weeks.

Methods:

Low active pre- and perimenopausal women between the ages of 40–50 (n = 39) were randomly assigned to a 12-week physical activity strength intervention or a waitlist control. Participants attended the sessions twice per week, which included both a 40-min cardiovascular and strength training component and a 20-min motivational component informed by Self-Determination Theory and Habit Formation Theory. Participants were provided Apple Watch devices to monitor their physical activity.

Results:

The physical activity strength intervention appears feasible based on recruitment duration, adherence to the Zoom sessions, retention, and the consumer satisfaction survey. The intervention participants increased their physical activity by 21 min/week, and the control group decreased by 2 min, although the differences were not significant. The intervention participants reported greater positive changes on several of the psychosocial variables, including self-efficacy, habit formation, habit automaticity, self-regulation, goal setting, physical activity enjoyment, physical activity feeling, physical activity revitalization, and physical activity tranquility.

Conclusions:

Although there was no evidence that the intervention led to increases in physical activity, this study supports the feasibility of a group-based physical activity strength intervention delivered via Zoom. Additional studies that include larger sample sizes and a longer intervention are needed to better understand the efficacy of the intervention.

Registration/clinical trials: https://clinicaltrials.gov/study/NCT05426070.

Plain language summary

The group-based physical activity and strength intervention delivered via Zoom was found to be feasible for premenopausal women. Additional studies including large sample sizes are needed

The majority of physical activity intervention studies do not adequately integrate strength training. This is particularly problematic for women approaching menopause given menopause is often related to weight gain and loss of muscle.

Objectives:

The purpose of this pilot study was to examine the feasibility and preliminary efficacy of a group-based physical activity and strength intervention delivered via Zoom for premenopausal women.

Design:

This study was a single blind randomized controlled trial lasting 12 weeks.

Methods:

Low active premenopausal women between the ages of 40–50 (n = 39) were randomly assigned to a 12-week physical activity strength intervention or a waitlist control. The sessions included a 40 min cardiovascular and strength training component and a 20 min motivational component informed by Self-Determination Theory and Habit Formation Theory. Participants were provided Apple Watch devices to monitor their physical activity.

Results:

The physical activity strength intervention appears feasible based on recruitment duration, adherence to the Zoom sessions, retention, and the consumer satisfaction survey. There were no differences between the groups on physical activity, however, intervention participants reported greater positive changes on several of the psychosocial variables including self-efficacy, habit formation, habit automaticity, self-regulation, goal setting, physical activity enjoyment, physical activity feeling, physical activity revitalization, and physical activity tranquility.

Conclusions:

This study supports the feasibility of a group-based physical activity strength intervention delivered via Zoom. Additional studies that include larger sample sizes and a longer intervention are needed to better understand the efficacy of the intervention.

Keywords

exercise resistance training women menopause

Introduction

Obesity is related to numerous health problems among women, including cardiovascular disease, type 2 diabetes, hypertension, arthritis, and depression.^1,2 Rapid weight gain is particularly common during menopause (defined as 1 year after cessation of the menstrual cycle).³ At the start of menopause, the rate of fat gain doubles, and lean mass declines. These gains and losses continue for 2 years after the final menstrual period.³ Interventions are needed to prevent this weight gain that often occurs during and after menopause.

Regular physical activity (PA) is related to less weight gain during menopause.⁴ However, many women still gain weight during menopause despite meeting the PA guideline of engaging in PA for 150 min/week.⁵ Most women do not integrate strength training into their PA routine, which is problematic given that strength training results in reduced body mass, body fat, and increased fat-free and muscle mass among women.^6
–8 Therefore, the lack of strength training may contribute to the weight gain associated with menopause.

Studies examining behavioral interventions for weight management during menopause have yielded small changes in weight.⁹ However, many of these studies have not adequately addressed PA adherence, and virtually none have specifically addressed strength training.⁷ Several studies have demonstrated that PA sessions that integrate strength training can be successful among women in highly supervised, lab-based settings with sophisticated equipment.^6,7,10,11 In addition, a recent review indicated that lab-based circuit training alone led to weight loss among middle and older-aged women.¹² Some limitations of these studies are that they are expensive, the potential for real-world dissemination is low given that they tend to be lab-based, and long-term adherence to the PA program is unclear. Another limitation is that the studies tended to be with clinical populations, young women, and/or older women. With the exception of one small lab-based study,¹³ none of the studies examined middle-aged women. This is a disservice to middle-aged women, which is when weight gain occurs most frequently.³

The purpose of this study was to examine the feasibility and preliminary efficacy of a 12-week home-based PA intervention that included strength training among pre- and perimenopausal women ages 40–50. This study addressed several gaps in the literature by emphasizing strength training, delivering the intervention remotely, focusing on middle-aged women, and including behavioral strategies designed to increase adherence. The intervention was informed by Self-Determination Theory (SDT) and Habit Formation Theory.^14,15 As suggested by Bowen et al.,¹⁶ feasibility studies should evaluate both acceptability and implementation, and limited efficacy testing should be conducted. We hypothesized that intervention participants would report more PA minutes per week than the waitlist control. We also hypothesized that participants randomized to the intervention group would report greater increases in psychosocial variables related to PA, including self-efficacy, habit formation, self-regulation, enjoyment, and exercise-induced feelings. Finally, an exploratory aim examined the relationship between group cohesion and attendance at the PA strength sessions.

Methods

Overview of study

This trial was a randomized controlled pilot study conducted from December 2022 to December 2023. Specifically, pre- and perimenopausal women (n = 39) were randomized 1:1 to either a 12-week group-based PA intervention focusing on strength training or a waitlist control group. Acceptability of the intervention and implementation were examined. Regarding preliminary efficacy, the primary dependent variable was PA minutes per week at 12 weeks. Secondary dependent variables included PA self-efficacy, habit formation, self-regulation, PA enjoyment, and PA feelings. The waitlist control group was given the option of receiving an individualized PA intervention program following the 12 weeks. The CONSORT Guidelines were followed.¹⁷ This study was approved by the Institutional Review Board at the University of Minnesota (Code: #00016282) on July 22, 2022. Participants completed an online consent form prior to participating in the study. All participants provided written informed consent prior to enrollment in the study. This research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.

Participants

Pre- and perimenopausal, low-active women between the ages of 40 and 50 were recruited via worksite emails, social media, studyfinder, clinicaltrials.gov, and a previous study in the northern region of the United States. Eligibility was assessed via a telephone screening interview. Specific inclusion criteria included the following: (1) Engaging in <90 min of moderate to vigorous intensity PA per week (to ensure participants could benefit from the intervention); (2) adequate space at home to complete exercise sessions (to ensure consistent participation); (3) capacity to use Zoom via a mobile device or computer; (4) pre- or perimenopausal which was defined as being between the ages of 40–50 and having had at least one menstrual period during the last 12 months (the menstrual period criteria was included to ensure that pre- and perimenopausal women were included); (5) ability to read and write fluently in English; (6) access to an iPhone; (7) not been hospitalized for a psychiatric condition in the past 6 months; and (8) willing to be randomly assigned to either study condition. Exclusion criteria included the following: (1) Two people living together in the same household (to ensure there were not any contamination effects if both received different interventions—one of the two could participate); (2) pregnancy; (3) type 1 diabetes; and (4) history of heart disease, exercise-induced asthma, or another health condition that may be exacerbated by exercise. The remaining exclusion criteria were included because these medical conditions could interfere with PA.

Measures

Feasibility was assessed and is described in more detail below. Questionnaires were administered at baseline and 12 weeks except the demographics survey (baseline only), Consumer Satisfaction Questionnaire (post-test only), and Physical Activity Group Environment Questionnaire (PAGEQ; post-test only).¹⁸ Each measure and the procedures for Apple Watch are described in more detail below.

Feasibility

Feasibility was assessed by evaluating both acceptability and implementation.¹⁶ Acceptability was assessed based on attendance at the PA strength sessions administered twice per week over Zoom and the Consumer Satisfaction Scale (items were on a seven-point Likert scale). Acceptability was considered achieved if participants attended two-thirds or more of the intervention sessions on average and rated the intervention a five or higher on the consumer satisfaction scale (based on a seven-point Likert scale; five was chosen, given that indicates a 70% satisfaction rating). Implementation was based on recruitment (i.e., number of participants recruited each month) and retention rates. Also, as recommended by Eldridge et al.,¹⁹ we examined the number of eligible participants and percent of participants randomized among those who were eligible. Implementation was considered achieved if at least 10 participants/month were recruited during the recruitment phase and 90% of the sample was retained at 12 weeks. Ten participants per month were chosen because this would allow for a future trial to recruit 200 participants in <2 years.

Apple Watch

Apple Watch Series 8 was used to objectively measure PA and move calories. Participants reported exercise minutes and move calories for 7 days at baseline prior to randomization and at 12 weeks based on information they obtained from their Apple Watch. Move calories are based on active calories, which are calories burned via activity throughout the day (i.e., can include light activities). Move calories include any calories burned that are not burned via one’s basal metabolic rate, which refers to energy the body uses to function when at rest.

Research staff connected with participants via a scheduled telephone call. Participants utilized Apple’s Fitness app on their iPhone to report the information indicated above. Participants took a screenshot of each day (7 days at baseline and 7 days at 12 weeks) and emailed the information to the secure university study email address. Apple Watch is valid and reliable for measuring PA.^20
–24 Participants who were not familiar with Apple Watch were provided information regarding how to wear it, check their activity minutes, and how to launch a workout session.

Physical activity self-efficacy

Self-efficacy associated with PA was measured using the 5-item Self-Efficacy for Physical Activity Questionnaire.²⁵ Participants were asked to assess their confidence in participating in PA in the presence of barriers such as feeling tired, being in a bad mood, not having time, vacation, and bad weather. This measure is based on a 5-point Likert scale ranging from not confident to extremely confident. The scale has acceptable reliability and validity.^25,26

Habit Index

The Self-Report Habit Index (SRHI)²⁷ was used to measure habit strength related to PA. The SRHI consists of 12 items and is designed to assess characteristics of habitual behavior including automaticity, frequency of engagement in the behavior, and the link between behavior and self-identity. Participants were asked to rate the degree to which they agree with statements related to habitual exercise behavior on a scale from 1 (strongly disagree) to 7 (strongly agree), with higher scores indicating higher strength of habit. The SRHI has acceptable internal and test–retest reliability, and validity.^27
–30

Self-regulation

Self-regulation was measured using the 31-item Short Form Self-Regulation Questionnaire (SSRQ).³¹ Participants used a five-point Likert scale (1 = strongly disagree to 5 = strongly agree) to rate the degree to which they agreed with statements related to general self-regulation to achieve goals. Participants received a total score based on all of their responses as well as two subscale scores for impulse control and goal setting. The SSRQ has demonstrated good psychometric properties, including a Cronbach’s alpha of 0.92, and high correlations with the original SRQ and other measures of validity and reliability.^31
–33

Physical activity enjoyment

The 18-item Physical Activity Enjoyment Scale (PACES)³⁴ was used to assess enjoyment for PA. Participants answered questions like “how you feel at the moment about the PA you have been doing” using a seven-point Likert scale. An overall enjoyment for PA score was determined by summing the items with higher scores reflecting higher enjoyment. Research indicates that the PACES is reliable and valid in diverse populations.^34
–36

Physical activity feelings

Exercise-induced feeling was assessed using the 12-item Exercise-Induced Feelings Inventory.³⁷ Participants assessed their feeling states during acute bouts of PA on a scale from 0 (do not feel) to 4 (feel very strongly). Four feeling states were assessed, including revitalization, positive enjoyment, physical exhaustion, and tranquility. The subscales have demonstrated excellent internal consistencies and sensitivity to change in PA interventions.³⁷

The PAGEQ

Group cohesion among the intervention participants was examined using the 21-item PAGEQ.¹⁸ This questionnaire uses a nine-point Likert scale ranging from strongly disagree to strongly agree. The four subscales for this measure include the following: (1) Individual Attraction to Group-Task (ATG-T); (2) Group Integration-Task (GI-T); (3) Individual ATG-Social (ATG-S); and (4) GI-Social (GI-S). The “individual” subscales refer to the motives related to the individual staying in the group, and cohesion is the unification of the group. The subscales also differentiate between the task (in this case PA) and social connections with each other. This questionnaire has been shown to have good reliability and concurrent validity.¹⁸

Consumer Satisfaction Questionnaire

The Consumer Satisfaction Questionnaire is a seven-item questionnaire that assessed how satisfied the participants were with the PA strength intervention and the individual components of the interventions including the website, individual exercise session options, motivational messages sent via email, Apple Watch, and the Zoom sessions (the PA strength and motivational components of the Zoom sessions were rated separately). Participants were asked to rate their satisfaction on a seven-point Likert scale ranging from 1 = very dissatisfied to 7 = very satisfied (higher scores indicate higher satisfaction).

Procedure

Potential participants texted, emailed, or called a study line in response to our advertisements and emails. Potential participants completed a telephone screening interview to determine eligibility, which was adapted from the 10-item Physical Activity Readiness Questionnaire.³⁸ PA was self-reported during the telephone screening interview to determine eligibility. The Research Coordinator provided information regarding the study and reviewed the consent form. Interested and eligible participants were emailed the online consent form and baseline questionnaires. Once the online consent form and questionnaires were completed, participants were mailed exercise bands for resistance training and an Apple Watch, which was worn for 7 days prior to the intervention and throughout the intervention. Next, the randomization session was completed over the phone. Specifically, participants were randomized 1:1 to either the 12-week PA strength intervention or waitlist control condition. The random numbers scheme was randomly generated by the PI, and one of the research coordinators conducted the randomization sessions. Participants were randomized in cohorts of ~20 (10 participants in each condition). Participants were given $100 for completing the 12-week assessment session and were allowed to keep their Apple Watch.

PA intervention

The PA intervention consisted of group-based Zoom sessions, motivational email messages, and a website. Participants completed two group-based Zoom PA sessions per week (40-min exercise session and 20 min of motivational counseling) for 12 weeks. The exercise component included various aerobics (e.g., high knees, jogging/walking in place) and muscle strengthening exercises (e.g., lunges, body weight squats, burpees, push-ups). Participants modified the exercise depending on their ability (e.g., wall push-ups versus knee push-ups versus regular push-ups). Exercise bands were also used for resistance training. Participants were instructed to engage in two PA sessions per week outside of the group, which included a variety of options (walking/jogging outside, online exercise videos, prescribed exercise sessions listed on the website, or any other exercise). The intervention was led by a master’s level health coach who is a personal trainer and holds an American College of Sports Medicine Certified Exercise Physiologist^® credential.

SDT and Habit Formation Theory guided the content of the counseling portion of the Zoom sessions^14,15 as outlined in Table 1. According to SDT, there are five types of motivation, which range from external regulation (e.g., engage in PA for appearance reasons) to intrinsic regulation (e.g., engage in PA because it aligns with one’s values or goals). The goal of the intervention was to achieve intrinsic regulation. Therefore, the health coach emphasized positive affective responses to exercise sessions and strategized to make PA, a part of the participants’ self-identity. The ultimate goal was to experience high intrinsic motivation for PA (i.e., individuals engage in exercise due to inherent pleasure) since it is related to higher levels of PA.³⁹ Relatedness, another construct related to SDT, was addressed via the mutual support and social interactions in the group-based sessions. Habit formation was also emphasized in the intervention.¹⁴ The health coach worked with the participants to create exercise habits (e.g., schedule an exercise session like an appointment, exercise at the same time each day). Habit formation theory postulates that long-term adherence to PA is related to PA becoming an automated habit. Group cohesion was enhanced by emphasizing empathy, active listening, and supportive feedback. The participants were encouraged to share personal stories about their journey with PA.

Table 1.

Summary of intervention components targeting psychosocial variables.

Psychosocial variable	Description	Explanation of how intervention targeted the variable	Behavior change technique
Autonomy	Feeling in control of physical activity	Participants did engage in physical activity together but could choose from a variety of workouts to do outside of the Zoom sessions	Prompt review of specific behavioral goals, self-monitoring, and time management
Competence	Learning and mastering new skills can increase physical activity	Participants learned new strength training skills during the Zoom sessions. The counselor helped the participant reflect on everything they were accomplishing (e.g., faster completion rates, new skill, higher weight)	Provided general encouragement and instruction
Enjoyment	More physical activity enjoyment is related to higher levels of physical activity	Participants encouraged to identify associations, positive feelings, and experiences of enjoyment from physical activity. If not enjoying physical activity in the moment, the counselor asked them to focus on feelings after physical activity was completed	Information on consequences
Exercise-induced feeling	Positive feelings associated with physical activity, including revitalization, tranquility, and positive engagement, lead to higher physical activity	Participants identified positive feeling states related to physical activity; the counselor helped participants reframe strategies to address negative feelings related to physical activity	Prompt self-talk
Outcome expectations	Expecting positive outcomes from physical activity is related to higher physical activity	Participants identified benefits of physical activity; the counselor discussed consequences of physical activity (action/inaction)	Provide information about behavior–health link
Self-efficacy	Confidence to engage in physical activity in different situations related to higher physical activity	Participants identified barriers to physical activity, and possible strategies to overcome barriers were discussed	Prompt barrier identification
Social support and relatedness	Social support is related to higher physical activity levels	Participants received support from others in the group and were encouraged to discuss their plans with others outside of the group to gain support; the counselor was also a source of positive social support	Provide feedback on physical activity performance; plan social changes

Motivational website

Participants received access to a mobile phone-friendly study website that included motivational messages, an online log to track completion of home-based sessions, and options for individual-based exercise sessions (e.g., full-length YouTube exercise videos, written out workouts to follow).

Motivational emailed messages

Participants received one motivational email message each week to remind them to complete the exercise sessions. The messages included strategies based on SDT and were tailored to life circumstances (e.g., strategies for adhering to exercise despite a demanding job) as well as the corresponding week’s behavioral session topic.

Waitlist control condition

Participants in the waitlist control condition received an Apple Watch and completed the same assessment sessions as the PA intervention. However, PA information was not provided until after 12 weeks. Participants were given the option of receiving an individualized version of the PA intervention once they completed the 12-week assessment.

Statistical analysis

Descriptive statistics were utilized to summarize the baseline variables. Unadjusted means were reported at baseline and 12 weeks for the primary and secondary dependent variables. Between-group analysis of covariance (ANCOVA) tests were used to examine the effect of the intervention on PA and the psychosocial variables, controlling for baseline. Bivariate correlations were used to examine the relationship between group cohesion and attendance at the PA strength sessions. The alpha level was set at 0.05, and adjustments were not made, given all comparisons were planned. The management and collection of the data was handled via REDCap, and the data analysis was run in SPSS 28.0 (IBM Corporation).

Results

Demographic data are summarized in Table 2. Participants were 44 years of age on average (range was 40–50), most were married (72%), college educated (69%), and all were employed. Thirty-one percent were from racially and/or ethnically diverse backgrounds.

Table 2.

Baseline characteristics by study arms.

Variable	Total sample (n = 39)	Strength (n = 20)	Control (n = 19)
Age (average in years)	43.87 (2.67)	44.00 (3.01)	43.74 (2.33)
Race (%)
Caucasian	71.8	85.0	57.9
Asian	7.7	0.0	15.8
American Indian/Alaskan Native	2.6	0.0	5.3
African-American	12.8	10.0	15.8
Other	5.1	5.0	5.3
Refused	0.0	0.0	0.0
Hispanic (%)	5.1	0.0	10.5
Marital status (married, %)	71.8	80.0	63.2
Education (college grad, %)	92.3	90.0	94.7
Children under 18 in household (%)	59.0	60.0	57.9
Employed (%)	100.0	100.0	100.0
Income (% over $50,000)	92.3	85.0	100.0

Note. Age is reported as means with standard deviations in parentheses.

Feasibility

Recruitment and retention are summarized in Figure 1. Sixty-six participants completed the telephone screening interview, and of these, 42 were eligible. Thirty-nine participants (93% of the eligible participants) were randomized to one of the two study arms. The first cohort was recruited in 1 month (December 14, 2022–January 13, 2023), and the second cohort over 2 months (June 27, 2023–August 30, 2023). All of the intervention participants and 95% of the control participants were retained at 12 weeks (defined as completing the post-test questionnaires). On average, participants completed 17 of the 24 intervention sessions. The overall rating on the consumer satisfaction scale was 5.34 (SD = 1.58) on a scale of 1–7 (see Table 3), which met the acceptability criteria of five or higher. The highest consumer satisfaction rating was for Apple Watch, and the lowest was for the motivational messages via email.

Figure 1.

Flow chart.

Table 3.

Consumer satisfaction survey.

Variable	Mean (SD)
Overall program	5.34 (1.58)
Website	4.76 (1.42)
Zoom strength-training sessions	5.34 (1.68)
Zoom motivational sessions	4.89 (1.59)
Motivational messages via email	4.34 (1.44)
Individual exercise session options	4.79 (1.49)
Apple Watch	6.11 (1.43)

Note. SD are in parentheses. Participants were asked to rate their satisfaction regarding the above components of the intervention on a seven-point Likert scale ranging from very dissatisfied to very satisfied (higher scores indicate higher satisfaction).

SD: standard deviation.

Physical activity

PA minutes per week by study arm are summarized in Table 4. There were no differences between groups for “move” calories or for PA minutes per week at 12 weeks.

Table 4.

The effect of the intervention on physical activity.

Variable	Strength (n = 20)		Control (n = 19)		p value	F-statistic	ηp²
Variable	Baseline	12 weeks	Baseline	12 weeks	p value	F-statistic	ηp²
Move calories per day	458.82 (157.59)	436.97 (171.52)	481.13 (190.74)	498.98 (182.41)	0.564	0.34	0.015
Exercise minutes per week	142.76 (137.20)	163.88 (202.65)	115.25 (88.23)	113.40 (142.15)	0.234	1.50	0.061

Note. Standard deviations are in parentheses. Results are based on ANCOVA analyses.

Psychosocial variables

The effects of the intervention on psychosocial variables are reported in Table 5.

Table 5.

The effect of the intervention on the psychosocial variables.

Variable	Strength (n = 20)		Control (n = 19)		p value	F-statistic	ηp²
Variable	Baseline	12 weeks	Baseline	12 weeks	p value	F-statistic	ηp²
Physical activity self-efficacy	11.90 (3.40)	13.90 (4.60)	10.47 (4.07)	9.84 (4.18)	0.011	7.24	0.167
Habit Index: total	30.25 (13.35)	44.85 (13.40)	28.00 (14.64)	32.11 (16.59)	0.004	9.24	0.204
Habit Index: automaticity	8.50 (4.12)	13.55 (5.20)	8.21 (4.34)	9.84 (5.62)	0.014	6.68	0.157
Self-regulation: total	120.25 (15.42)	126.65 (13.14)	125.16 (14.64)	122.11 (21.28)	0.026	5.39	0.133
Self-regulation: impulse control	46.90 (5.16)	47.30 (5.36)	48.00 (7.36)	46.22 (7.42)	0.193	1.76	0.048
Self-regulation: goal setting	36.75 (6.26)	39.40 (3.39)	40.16 (3.93)	38.50 (6.65)	0.016	6.45	0.156
Exercise enjoyment	88.50 (15.17)	97.15 (14.24)	86.47 (17.57)	83.78 (20.16)	0.018	6.20	0.150
EFI: total	22.15 (5.75)	24.95 (7.16)	19.68 (5.75)	18.72 (7.84)	0.050	4.12	0.105
EFI: positive engagement	6.20 (2.14)	7.15 (2.50)	4.95 (2.93)	4.61 (3.52)	0.102	2.83	0.075
EFI: revitalization	3.95 (2.68)	6.00 (2.75)	3.21 (1.84)	2.72 (3.12)	0.003	10.36	0.228
EFI: tranquility	6.50 (2.91)	7.30 (2.15)	5.57 (2.12)	5.28 (2.32)	0.018	6.21	0.151
EFI: physical exhaustion	5.50 (3.62)	4.50 (2.46)	5.94 (3.47)	6.11 (3.76)	0.158	2.08	0.056

Note. Values are means (standard deviations in parentheses). Results based on ANCOVA.

EFI: Exercise-Induced Feeling Inventory.

Relative to the waitlist control and after adjusting for baseline values, participants in the PA intervention reported higher levels of self-efficacy, habit formation, habit automaticity, self-regulation, self-regulation goal setting, exercise enjoyment, exercise revitalization, and exercise tranquility. There were no differences between the intervention and waitlist participants on impulse control, positive engagement, or physical exhaustion.

Group cohesion

Means for the subscales of the PAGEQ, which were on a 9-point Likert scale, were 6.49 (SD = 2.23) for ATG-T, 5.77 (SD = 1.94) for GI-T, 4.99 (SD = 2.10) for ATG-S, and 4.38 (SD = 1.86) for GI-S. The means are reported as averages per item since the number of items per subscale varied. Results indicated that higher scores on the GI-T subscale were related to better attendance at the PA strength sessions, r = 0.482, p = 0.031.

Discussion

The group-based PA strength intervention delivered via Zoom appears feasible based on the acceptability and implementation parameters. Our retention rate was 97% at 12 weeks, which far exceeded our goal of 90%. In addition, we were within one participant of meeting our goal of recruiting 40 participants. We met our goal of recruiting 10 participants/month on average (20 participants were recruited over 1 month for cohort 1 and 19 were recruited over 2 months for cohort 2). We randomized 93% of the eligible participants, which supports feasibility as well as generalizability of the sample. We also exceeded our intervention adherence goal (participants attended on average at least two-thirds of the intervention sessions). This was a significant accomplishment, since meeting at a specified time twice per week may have been challenging for some participants with work and/or family obligations. The mean score on the consumer satisfaction questionnaire was 5.34, which exceeded our goal of five points on the seven-point Likert scale. Participants rated Apple Watch and Zoom strength training sessions the highest and the emailed motivational messages the lowest, which suggests that technological barriers did not interfere with participation. These findings also suggest that future studies may want to focus on texting and app-based messaging instead of emails.

Contrary to our hypotheses, participants randomized to the PA strength intervention did not have greater move calories per day or exercise minutes per week than participants in the waitlist control condition. Even though it was not statistically significant, possibly due to the small sample size, participants in the PA strength intervention engaged in 28 more min/week of PA than those in the waitlist condition at baseline (143 min/week for the PA strength intervention versus 115 min/week for the control). Therefore, the PA strength intervention participants had less room to improve than the control condition, which could have impacted the results. One of the inclusion criteria was engaging in <90 min/week of PA, and therefore, both groups had higher PA minutes than expected at baseline. Eligibility was assessed via self-report, and the baseline data were objectively measured using Apple Watch, which may have accounted for this discrepancy. It is possible that participants underreported their PA at the telephone screening interview. Also, wearing Apple Watch and emailing their activity screen shots may have motivated them to increase their PA from prior levels. Future trials should consider a lower PA criteria for eligibility and/or assess PA objectively to determine eligibility.

Although not significant, the move calories decreased in the intervention group relative to the control. It is possible to have a low number of move calories throughout the day with high PA minutes if the participant engaged in a low amount of light activities but engaged in moderate to vigorous intensity activity. It is possible that the increased PA minutes were related to less light activities during the rest of the day. Participants may have been physically tired from exercising or felt justified not to engage in light activities throughout the day, given they had exercised. Additional research is needed to better understand the effect of PA minutes on light activities.

This trial was not powered to examine the mediating effect of the psychosocial variables. However, we still examined the effect of the intervention on various psychosocial variables to inform future trials. Relative to the waitlist condition, participants in the PA strength intervention reported increases from baseline to 12 weeks for PA self-efficacy, habit formation, habit automaticity, self-regulation, self-regulation goal setting, PA enjoyment, revitalization related to PA, and tranquility as a result of engaging in PA. This is consistent with previous research suggesting that behavioral interventions lead to improvements in psychosocial variables targeted by the intervention.⁴⁰ The psychosocial findings indicate that the intervention successfully intervened on the intended variables. Since these variables are related to PA increases and maintenance, perhaps future studies that are adequately powered would result in significant increases in PA both during the intervention and in the long term.

To inform future trials regarding the importance of group dynamics, we evaluated group cohesion. Results indicated that participants who rated the intervention as high on group integration related to the shared goal of engaging in PA had higher attendance at the PA strength sessions than those reporting lower ratings. Consistent with previous studies,⁴¹ this finding suggests that the group dynamics can have an impact on individuals consistently attending group-based PA interventions. This result has important implications for future research in that studies should emphasize group bonding and social connections between the group members very early in the intervention (e.g., team building, icebreaker activities, collaborative exercises, encouraging empathy).

Study strengths

There were strengths related to this study. First, the methodology was strong in that the trial included a systematic evaluation of feasibility, randomization occurred between the study arms, an experienced health coach with appropriate credentials conducted the group-based intervention sessions, and validated measures were utilized. Second, evaluating the feasibility and potential efficacy of a group-based Zoom PA intervention focusing on strength is novel and addresses a gap in the literature. The group-based remote format allowed for social support from the other participants while addressing barriers typical for an in-person intervention (e.g., transportation, childcare, high time commitment). Third, the use of Apple Watch to motivate and assess PA has not been extensively studied previously. Finally, targeting pre- and perimenopausal women ages 40–50 is particularly important given the rapid weight gain that can occur as women approach menopause.³

Study limitations

Despite these strengths, there were several limitations. First, data obtained from Apple Watch were not provided directly from Apple Watch devices but rather were provided indirectly through screenshots taken by the participant. This created reliance on participants adhering to the study protocol. We did not have resources to build a study app to passively read data from HealthKit on iPhone, so we relied on data communication via screenshots. Second, the intervention was relatively short in duration (i.e., 12 weeks). Third, there may have been selection bias due to the eligibility criteria, especially related to the technology requirements for the study. Fourth, sample size calculations were not completed for this trial, given it was considered a pilot study and efficacy was not a primary aim of the study. Finally, 28% of the sample was racially or ethnically diverse, which is representative of the recruitment area. However, higher levels of diversity would have enhanced the generalizability of the study.

The focus on strength training has often been ignored in previous PA intervention trials.¹⁰ Typically, the goal of PA interventions is to achieve 150 min of moderate to vigorous intensity activity per week. Meeting the strength training criteria is often not addressed and/or assessed.⁴² This is a disservice to individuals hoping to obtain the health benefits of strength training and is particularly problematic for women entering menopause who often experience a significant loss of muscle and an increase in weight.³ Walking is the most common type of PA,⁴³ but research indicates that walking is not associated with preventing weight gain.⁴⁴ Future intervention trials should target both cardiovascular and strength training (separately or combined). Researchers should keep in mind that cardiovascular and strength training can be combined into one workout (e.g., walking or jogging while taking periodic breaks to do push-ups and air squats). As more research focuses on strength training, it will also be important for researchers and policymakers to determine if the current guideline of strength training at least two times per week is adequate. Finally, given the high consumer satisfaction rating for Apple Watch devices, additional research is needed that examines whether Apple Watch devices can be used for both intervention and assessment purposes or whether other types of accelerometers in which data cannot be seen by the participant, such as the ActiGraph (Ametris), should be added for assessment purposes.

Conclusions

In summary, the group-based PA strength intervention delivered via Zoom appears to be feasible based on recruitment duration, adherence to the Zoom sessions, retention, and the consumer satisfaction survey. However, there was no evidence that the intervention led to increases in PA. Participants were least satisfied with the motivational email messages, suggesting that additional research is needed to determine if these are necessary, or if other delivery modes might be considered. Additional studies could also use artificial intelligence to create more personalized messages, which may improve satisfaction. The group-based sessions were rated highly on the consumer satisfaction survey, which suggests that future studies should continue to examine group-based interventions. Group dynamics are important, as evidenced by group cohesion being related to higher attendance at the PA strength sessions. In addition, even though significant differences for PA were not found between the intervention and control participants, significant increases in key variables addressed in the intervention, such as habit formation and self-efficacy, were observed.

Taken together, even though there were no differences between groups for PA, given the changes in the psychosocial measures and the high retention rates and satisfaction scores, an examination of the intervention in larger, longer randomized controlled trials is warranted. Given the high use of Apple Watch devices among the general population, it will be important for future studies to explore Apple Watch devices as both an intervention tool and assessment method. The use of Apple Watch devices can reduce the burden of adding an additional device to monitor PA in randomized trials. Finally, to address the challenges of menopause,³ PA studies that include a strength component are particularly needed for this population.

Footnotes

Acknowledgements

We would like to thank Laura Polikowsky for her assistance with this project. We are also very grateful to the study participants who made this study possible.

ORCID iDs

Beth A. Lewis

Tyler M. Dregney

Ethical considerations

This study was approved by the Institutional Review Board at the University of Minnesota (Code: #00016282) on July 22, 2022. This research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.

Consent to participate

Participants completed an online consent form prior to participating in the study. All participants provided written informed consent prior to enrollment in the study.

Consent for publication

Participants provided written consent to publish the findings of the study via the consent form.

Author contributions

Beth A. Lewis: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Supervision; Writing – original draft; Writing – review & editing.

Katie J. Schuver: Conceptualization; Methodology; Project administration; Supervision; Writing – original draft; Writing – review & editing.

Kaitlyn B. Swinney: Project administration; Writing – original draft; Writing – review & editing.

Tyler M. Dregney: Project administration; Writing – original draft; Writing – review & editing.

Jennifer A. Linde: Conceptualization; Methodology; Writing – original draft; Writing – review & editing.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported through a grant from the University of Minnesota Grant-in-Aid program. Apple Inc. provided Apple Watch devices for the research.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: There are no conflicts of interest among any of the authors. Apple, Inc. provided Apple Watch devices for the research. Apple was not involved in the design of the research, nor was it involved in the collection, analysis, or interpretation of the research data.

Data availability statement

The dataset is available upon request to the first author of the study.

References

Hagstrom

Marshal

Halaki

, et al The effect of resistance training in women on dynamic strength and muscular hypertrophy: a systematic review with meta-analysis. Sports Med 2020; 50: 1075–1093.

Uzogara

SG.

Obesity epidemic, medical and quality of life consequences: a review. Int J Public Health Res 2017; 5: 1–12.

Greendale

Sternfeld

Huang

, et al Changes in body composition and weight during the menopause transition. JCI Insight 2019; 4: e124865.

Kim

Kang

Regular leisure-time physical activity is effective in boosting neurotrophic factors and alleviating menopause symptoms. Int Journal Environ Res Public Health 2020; 17: 8624.

Davis

Castelo-Branco

Chedraui

, et al Understanding weight gain at menopause. Climacteric 2012; 15: 419–429.

Batrakoulis

Jamurtas

Georgakouli

, et al High intensity, circuit-type integrated neuromuscular training alters energy balance and reduces body mass and fat in obese women: a 10-month training-detraining randomized controlled trial. PLoS One 2018; 13: e0202390.

Cunha

Nunes

Tomeleri

, et al Resistance training performed with single and multiple sets induces similar improvements in muscular strength, muscle mass, muscle quality, and IGF-1 in older women: a randomized controlled trial. J Strength Cond Res 2020; 34: 1008–1016.

U.S. Department of Health and Human Services. Physical Activity Guidelines Advisory Committee scientific report. Department of Health and Human Services, 2018.

Thomson

Kelly

Sainsbury

, et al Weight loss outcomes in premenopausal versus postmenopausal women during behavioral weight loss interventions: a systematic review and meta-analysis. Menopause 2021; 28(3): 337–346.

10.

Holviala

Sallinen

Kraemer

, et al Effects of strength training on muscle strength characteristics, functional capabilities, and balance in middle-aged and older women. J Strength Cond Res 2006; 20: 336–344.

11.

Isenmann

Kaluza

Havers

, et al Resistance training alters body composition in middle-aged women depending on menopause – a 20-week control trial. BMC Womens Health 2023; 23: 526.

12.

Ramos-Campo

Andreu-Caravaca

Carrasco-Poyatos

, et al Effects of circuit resistance training on body composition, strength, and cardiorespiratory fitness in middle-aged and older women: a systematic review and meta-analysis. J Aging Phys Act 2021; 1: 1–14.

13.

Schmitz

Jensen

Kugler

, et al Strength training for obesity prevention in midlife women. Int J Obes 2003; 27: 326–333.

14.

Lally

Wardle

Gardner

Experiences of habit formation: a qualitative study. Psychol Health Med 2011; 16: 484–489.

15.

Ryan

Deci

EL.

Overview of self-determination theory: an organismic dialectical perspective. In: Deci

Ryan

(eds) Handbook self-determination research. University of Rochester Press, 2002, pp.3–33.

16.

Bowen

Kreuter

Spring

, et al How we design feasibility studies. Am J Prev Med 2009; 36: 452–457.

17.

Moher

Hopewell

Schulz

, et al CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trial. Ann Int Med 2010; 152: 726–732.

18.

Estabrooks

Carron

AV.

The Physical Activity Group Environment Questionnaire: an instrument for the assessment of cohesion in exercise classes. Group Dyn Theory Res Pract 2000; 4: 230.

19.

Eldridge

Chan

Campbell

, et al CONSORT 2010 statement: extension to randomised pilot and feasibility trials. Br Med J 2016; 355: i5239.

20.

Khushhal

Nichols

Evans

, et al Validity and reliability of the Apple Watch for measuring heart rate during exercise. Sports Med Int Open 2017; 1: E206–E211.

21.

Khushhal

Mohamed

Elsayed

. Accuracy of Apple Watch to measure cardiovascular indices in patients with chronic diseases: a cross sectional study. J Multidiscip Healthc 2024: 1053–1063.

22.

Nelson

Allen

NB.

Accuracy of consumer wearable heart rate measurement during an ecologically valid 24-hour period: intraindividual validation study. JMIR Mhealth and Uhealth 2019; 7: e10828.

23.

Jeon

Kim

, et al Comparative analysis of energy expenditure assessments from the graded exercise test vs. Galaxy Watch and Apple Watch in Korean college students during a 30-minute workout: a pilot study. Asian J Kinesiol 2024; 26: 60–66.

24.

White

III Finnegan

Tindall

, et al Comparison of raw accelerometry data from ActiGraph, Apple Watch, Garmin, and Fitbit using a mechanical shaker table. PLoS One 2024; 19: e0286898.

25.

Marcus

Selby

Niaura

, et al Self-efficacy and the stages of exercise behavior-change. Res Q Exerc Sport 1992; 63: 60–66.

26.

Mendoza-Vasconez

Marquez

Benitez

, et al Psychometrics of the self-efficacy for physical activity scale among a Latina women sample. BMC Public Health 2018; 18: 1097.

27.

Verplanken

Orbell

Reflections on past behavior: a self-report index of habit strength. J Appl Soc Psychol 2003; 33: 1313–1330.

28.

Morean

DeMartini

Foster

, et al The Self-Report Habit Index: assessing habitual marijuana, alcohol, e-cigarette, and cigarette use. Drug Alcohol Depend 2018; 186: 207–214.

29.

Thurn

Finne

Brandes

, et al Validation of physical activity habit strength with subjective and objective criterion measures. Psychol Sport Exerc 2015; 15: 65–71.

30.

Verplanken

Melkevik

Predicting habit: the case of physical exercise. Psychol Sport Exerc 2008; 9: 15–26.

31.

Carey

Neal

Collins

SE.

A psychometric analysis of the Self-Regulation Questionnaire. Addict Behav 2004; 29: 253–260.

32.

Neal

Carey

KB.

A follow-up psychometric analysis of the Self-Regulation Questionnaire. Psychol Addict Behav 2005; 19: 414–422.

33.

Šebeňa

Orosová

Helmer

, et al Psychometric evaluation of the Short Self-Regulation Questionnaire across three European countries. Stud Psychol 2018; 60: 5–15.

34.

Kendzierski

DeCarlo

KJ.

Physical activity enjoyment scale: two validation studies. J Sport Exerc Psychol 1991; 13: 50–64.

35.

Jung

Nie

, et al Affective and enjoyment responses to sprint interval training in healthy individuals: a systematic review and meta-analysis. Front Psychol 2022; 13: 820228.

36.

Murrock

Bekhet

Zauszniewski

JA.

Psychometric evaluation of the physical activity enjoyment scale in adults with functional limitations. Issues Ment Health Nurs 2016; 37: 164–171.

37.

Gauvin

Rejeski

WJ.

The exercise-induced feeling inventory: development and initial validation. J Sport Exerc Psychol 1993; 15: 403–423.

38.

Thomas

Reading

Shephard

RJ.

Revision of the Physical Activity Readiness Questionnaire (PARQ). Can J Sport Sci. 1992; 17: 338–345.

39.

Thøgersen-Ntoumani

Shepherd

Ntoumanis

, et al Intrinsic motivation in two exercise interventions: associations with fitness and body composition. Health Psychol 2016; 35: 195–198.

40.

Lewis

Williams

Frayeh

, et al Self-efficacy versus perceived enjoyment as predictors of physical activity behaviour. Psychol Health 2016; 31: 456–469.

41.

de Lacy-Vawdon

Klein

Schwarzman

, et al Facilitators of attendance and adherence to group-based physical activity for older adults: a literature synthesis. J Aging Phys Act 2018; 26: 155–167.

42.

Madigan

Fong

Howick

, et al Effectiveness of interventions to maintain physical activity behavior (device-measured): systematic review and meta-analysis of randomized controlled trials. Obes Rev 2021; 22: e13304.

43.

Malone

Patterson

Grunin

, et al Habitual physical activity patterns in a nationally representative sample of U.S. adults. Transl Behav Med 2021; 11: 332–341.

44.

Jakicic

Powell

Campbell

, et al Physical activity and the prevention of weight gain in adults: a systematic review. Med Sci Sports Exerc 2019; 51: 1262.