There has been an increase in the recognition of challenges mothers face in achieving recommended levels of physical activity (PA), underscoring the need for targeted interventions. Mothers are not engaging in enough moderate-to-vigorous-intensity PA to experience health benefits [1]. Instead, mothers experience significant declines in PA over time and report higher levels of physical inactivity, compared to nonparents [2,3]. Barriers to PA that are amplified for mothers include time constraints, childcare responsibilities, and inadequate support from spouse or partner [1,4,5]. A review of PA interventions for mothers suggested group-based programs and behavior change techniques (eg, goal-setting, self-monitoring, and problem-solving) are effective at increasing PA among mothers [6].

Black and Hispanic mothers are less likely to meet national PA guidelines than women from other racial and ethnic groups [7]. Notably, racial and ethnic disparities reflect the cumulative impact of structural inequities, as Black and Hispanic mothers are more likely to have lower socioeconomic status and less access to PA opportunities [8]. Moreover, and consistent with an intersectionality approach [9], the intersection of gender, socioeconomic status, and race creates unique challenges, which can make PA especially difficult for mothers with multiple marginalized identities. Prior interventions have not adequately addressed barriers that are unique to mothers with high-risk, including those from lower-income households and racial and ethnic minority groups [7]. These barriers include time limitations due to child commitments, childcare needs, and lack of access to PA opportunities [10-15].

The Moms on the Move (MOMs) intervention is a PA intervention designed to comprehensively address PA barriers experienced by low-income Black and Hispanic mothers. The MOMs intervention builds on previous research that has focused on developing theory-based interventions that have demonstrated success at increasing PA among Black [16,17] and Hispanic or Latina [18-20] women. The behavioral theories underlying these interventions include Social Cognitive Theory (SCT) [21] and the Transtheoretical Model (TTM) [22]. The SCT suggests behaviors (including PA) can be better understood when considered within the context of personal characteristics, environmental factors, and the reciprocal interactions between the behavior, the person, and the environment [23,24]. The TTM was developed to recognize that people may vary in their readiness or motivation to change behavior, and that identifying this stage of change can be helpful when designing behavior interventions [22,25]. As a result, the MOMs intervention integrated intervention components that focus on theoretical constructs, including social support through the in-person group format of PA sessions, self-efficacy through the instruction of PA by a female fitness coach, and (in 2023) written materials tailored to the participant’s stage of change. A full list of theoretical concepts that guided the MOMs intervention and informed the present study are outlined in Multimedia Appendix 1.

What makes the MOMs intervention unique is that it is tailored specifically to Black and Hispanic low-income mothers and is delivered during children’s sports practices. Delivering the intervention in this context was intentional: It leveraged a setting where mothers were already present, helping to overcome barriers such as childcare needs, transportation, time constraints, and feelings of guilt associated with prioritizing self-care [26]. By integrating PA opportunities into their children’s schedules, the MOMs intervention allows mothers to engage in self-care without guilt while promoting consistent participation through the regularity of practices. Preliminary efficacy, as well as feasibility and acceptability, of the MOMs intervention has been established [27,28], providing support for this intervention. These findings are based on objective and self-report data, limiting an in-depth understanding of the participants’ perceptions of the intervention.

The MOMs intervention is a community-based program designed in partnership with a youth football and cheerleading organization. Community-based research [29] emphasizes the inclusion of community in the development of research. The MOMs intervention facilitated community involvement by soliciting community perspectives both during formative research, to understand barriers to engaging in healthy behaviors among Hispanic and Black mothers [30,31], and again when evaluating the intervention through exit interviews. Exit interviews are a qualitative analysis method that can facilitate process evaluation, specifically in the context of health-related intervention development [32-34].

We aimed to explore participants’ perceptions of the MOMs intervention using data from 2 rounds of exit interviews completed after 2 single-group pilot studies of the MOMs intervention. We were specifically interested in understanding participants’ engagement and satisfaction with the intervention, as well as their perceptions of benefits, barriers, and recommendations for future intervention development. The objective of our study was not only to better understand how to improve the MOMs intervention specifically, but also to guide program development for other mothers and women caregivers with multiple marginalized identities.

This study used a qualitative analysis of exit interview data from participants who completed the MOMs pilot trials. To improve transparency and rigor, the Standards for Reporting Qualitative Research (SRQR) [35] and Template for Intervention Description and Replication (TIDieR) [36] guided information reported about the qualitative methods and intervention.

The MOMs intervention was delivered in partnership with a youth football and cheerleading organization in Providence, Rhode Island, with recruitment conducted during children’s football and cheerleading practices at the start of the season. Inclusion criteria included English-speaking women who had regular access to email and passed a screener to ensure they were not at increased risk for injury or harm by engaging in PA. Eligible participants had to identify as a primary caregiver of a child (age range of 6- to 13-years old) who was a member of one of the organization’s football teams or cheerleading squads (organized by age of the child). The choice to include all women caregivers, instead of exclusively mothers, was intentional to better adapt to the community-based setting and group-based nature of the program.

The first pilot study was completed in 2022; the second, in 2023, with insights from the first year informing intervention refinements in the second year. The primary intervention component during both years was in-person group PA sessions led by a female fitness instructor during children’s sports practices. The same female fitness instructor led the in-person group PA sessions both years (for 6 weeks in 2022 and 8 weeks in 2023). All in-person group PA sessions were an hour long and scheduled 3 times per week. Sessions were scheduled when at least 1, but up to 6, teams from the partner organization were practicing. Participants were invited and welcomed to attend all 3 sessions, regardless of whether their child was on the field and practicing. Child supervision, for younger children not participating on a football team or cheerleading squad, was offered. PA goal-setting calls and individually-tailored written materials (including tip sheets and stage-matched PA manuals) were added in 2023. Table 1 summarizes the MOMs interventions and their components.

Full information about the 2022 pilot trial, and preliminary outcomes, is described elsewhere [28]. Information about the MOMs pilot trial in 2023, including the procedures for exit interviews, is described in the ClinicalTrials.gov registration (NCT05461742); outcomes reporting the quantitative data are reported elsewhere [27].

All study procedures were reviewed and approved by the Institutional Review Board of Brown University (protocol number: 2022003323). Participants provided informed written consent at the time of enrollment in the pilot study, including consent to complete an exit interview at the end of the intervention. Participants were compensated for completing assessments, with a total possible compensation of US $125 (in 2022) and US $130 (in 2023). All data were anonymized and deidentified for analysis.

Convenience sampling was used to recruit participants for the exit interviews, with a desired sample size of 15 participants to capture a range of participant experiences. All participants who consented to participate in the MOMs pilot interventions (2022, n=26; 2023, n=35) were invited to complete an exit interview at the conclusion of the intervention (even if they did not attend any workout session). The 9 participants who participated in both interventions were invited to complete an exit interview at the end of each intervention. Participants who consented to the MOMs pilot interventions were not contacted for exit interviews if they had actively withdrawn from the study and asked not to be contacted (2022, n=4).

Exit interviews were completed by phone with a trained research assistant between October and November 2022 (for participants who completed the intervention in 2022) and between October and November 2023 (for participants who completed the intervention in 2023). A semistructured interview guide began with one broad question, “Overall, what did you think of the Moms on the Move intervention?” Follow-up questions focused on (1) exploring what participants liked and did not like about the intervention (eg, “Tell me what you liked about the intervention”), including barriers to participation (eg, “Can you think of anything else that we could have provided that would have been helpful in increasing your PA?”); (2) understanding participants’ perceptions of specific intervention components (eg, “What about the in-person PA did you find helpful in terms of increasing your PA?”); and (3) soliciting recommended changes for future intervention development (eg, “Do you have any other thoughts or comments about how we could improve the experience of participants in future studies?”). There were 13 follow-up questions for exit interviews conducted in 2022 and 16 follow-up questions for interviews conducted in 2023; the additional 3 questions added to the exit interviews were included to ask participants about additional intervention components (ie, goal-setting calls and written materials), as well as a question asking for participants who completed the intervention both years (2022 and 2023) to compare their experiences.

Interviews ranged between 7 minutes and 27 minutes and were audio recorded. Transcribed deidentified and deidentified interviews were imported to NVivo (Lumivero, version 14.23.3) for analysis.

A directed content analysis [37], informed by the SCT and TTM, was used. First, a codebook was developed using a deductive approach by identifying theory-related (Multimedia Appendix 1) and intervention-specific constructs (eg, social support and group component, respectively). The final codebook can be found in Multimedia Appendix 2. Transcribed interviews were coded by a team of coders (the first 3 authors: SGL, BO, and NR) immediately after codebook development. First, 2 interviews from 2023 were coded by all coders to establish consensus. Remaining interviews were then coded during team meetings by a pair of 2 coders, with a third coder observing and resolving any discrepancies. Pairs of coders were alternated to ensure equal participation across the 3 coders (SGL, BO, and NR). Next, codes were reviewed, and themes were identified by framing patterns from interviews within the SCT and TTM. All coders reviewed the themes to ensure the themes were consistent with the data. This approach was completed for the interviews from 2023 first, and then for the interviews from 2022. Given that coding began immediately after the development of a theory-informed codebook, our codes are considered predetermined [37]. However, we did not want to force the application of all codes; therefore, only those that were applied were reviewed for theme identification. Theory-informed themes based on coded data across both years were finalized during team meetings with the principal investigator (the last author, TvA).

Nineteen participants were interviewed. Table 2 presents a full description of the sample. Participants were mostly Black or Hispanic. About half endorsed 12 or fewer years of education, and about two-fifths reported an annual household income less than or equal to US$ 40,000.

Participants primarily identified their relationship with the children participating in the sports practices as mothers, with 1 participant (Participant 4, 2022 and 2023) identifying as a grandparent.

We sought to examine participants’ perceptions of engagement and satisfaction with the MOMs intervention, as well as benefits, barriers, and recommendations. Overall, participants were generally satisfied with the MOMs intervention. More specifically, participants described the MOMs intervention as “motivating,” “comfortable,” “good,” and “helpful.” Self-reported benefits of the intervention included improved overall health and increased motivation to engage in PA. While barriers were still acknowledged, including inclement weather and guilt associated with prioritizing care for self, the consensus was that the intervention should be continued. Four theory-informed themes were identified. The first centers around takeaways from the interventions. The next 3 identify the essential qualities associated with the perceived success of the intervention: accountability, ability, and accessibility. Exemplar quotes are included with a unique participant identifier, and the years participants completed the intervention.

This study examined the qualitative data from 2 exit interviews completed after each year the MOMs intervention was piloted to explore participants’ perceptions of the program and guide future intervention development. For the most part, the MOMs intervention was described as a helpful way to increase PA for this sample of predominantly Black and Hispanic mothers and women caretakers with low income because of its convenience and components, embedded within a community context. Findings were aligned with the SCT and TTM, given that the most salient characteristics of the intervention were associated with social support (described as accountability), self-efficacy (described as ability), and environmental facilitators (described as accessibility). Coupled with results from a quantitative analysis that found that self-efficacy and self-reported moderate- to vigorous-intensity PA increased among MOMs participants [28] and an exploration of the feasibility and acceptability of the MOMs intervention [27], these findings support the continued delivery of the program, as well as its ability to be expanded to better support high-risk mothers and women caregivers.

Health behavior change theories, including SCT [24] and TTM [22], identify social influences on behavior change, for example, the impact of the social context and of helping relationships. Data from this study add to and extend theory in 3 ways. First, given that participants appreciated a sense of shared identity (or ability) with those in the PA group, it appears that it is not only about having relationships with other people but how relationships transform motivation that is relevant to health behavior change. The support received from a group of similar others can be characterized as a distinct motivating factor that might impact behavior differently. For example, while stages of behavior change are typically thought of as being linear (ie, one stage leads to the next) and circular (eg, people who are in a maintenance stage of change can become precontemplative again) [22], the fluidity of these stages was notable for participants. Additionally, given that this group is embedded within the community, there are opportunities for this intervention to have a sustained impact. For example, participants have the opportunity to engage in PA with group members outside the sessions during other sports practices, or even other sports seasons. Second, given that participants noted how their engagement in the intervention changed how their children viewed their mother’s behavior and the importance of healthy behaviors, it seems that modeling, often understood as identification with a role model [38], might also extend to identification as a role model (ie, recognizing that your behavior has an impact on another).

Interestingly, motivation and self-efficacy (as well as a number of other individual-level factors) were described as outcomes of participating in the MOMs intervention, rather than simply a predictor of engaging in health behaviors. In other words, participants began engaging in behavior change because an opportunity was presented to them, and they appreciated the group nature of it, specifically. Then, after beginning to engage in the intervention, they noticed an increase in motivation. The notion that motivation can come from relationships, rather than oneself, has been identified in relationship science, for example, interdependence theory [39,40], but has infrequently been extended beyond nonromantic contexts. Future studies may benefit from integrating interdependence theory with health behavior change theories, including those presented in this study (the SCT and TTM) as well as others that are relevant to expand our understanding of how social relationships transform motivation to engage in behavior change. In particular, self-determination theory [41] may be important to consider, given how concepts related to the 3 basic psychological needs (competence, autonomy, and relatedness), intrinsic and extrinsic motivation, and motivational climate showed up in the data.

Finally, these participants did not need more internal motivation to get their PA started; they needed opportunity presented by a supportive environment. Opportunity connects to the theme of accessibility. Early health behavior change theory literature originated from understanding preventive behavior [42]; however, increasing healthy behaviors may operate differently (and this could be why some a priori codes were not present, eg, stimulus control). Aligned with the SCT and TTM [22,24] and a growing body of literature on PA specifically [43,44], our findings suggest that creating environments in which healthy lifestyle behaviors, including PA, can occur is an efficacious strategy to facilitate behavior change among underactive populations. In particular, children’s sports practices are a good context for health behavior interventions for mothers and women caretakers with marginalized and intersectional identities.

Future research focused on developing the MOMs intervention can adopt stronger study designs (eg, a fully powered randomized controlled trial) to test the effects of the intervention for high-risk mothers and women caregivers. Recommendations for interventions are drawn from combining themes identified in these exit interviews with participant suggestions (Table 2) and literature on health behavior change. First, given the helpfulness of having accountability and the known influence of helping behaviors [22], it may be beneficial to implement a buddy system (eg, where participants who have completed the intervention at least once can become role models for others who are newer to the intervention). Adding to this social component, opportunities to increase group cohesion and collectiveness may be facilitated in a number of ways, including having ice breakers at the beginning of each session to facilitate bonding and a sense of connection to increase opportunities for mothers to connect about shared identities. Given that bonding with family was also a benefit, it may be important to consider opportunities for introducing shared activities for mothers and their children during practices (eg, completing warmups together). Finally, facilitating more accessibility within the environment could be made possible with opportunities to engage in the intervention either remotely or asynchronously; this would also help reduce disruptions in case of weather disruptions and other schedule changes. This can be made possible through the use of online platforms and technology, as recommended by the participants, and a growing body of research has already begun exploring this in the context of lifestyle interventions [45], especially those developed for women from minoritized backgrounds [46,47]. It is possible that the inclusion of both in-person and technology-based intervention components may allow for increased engagement and improved outcomes.

One of the least liked aspects of the intervention was the assessment. Until this program can be integrated within the community, assessments are necessary for the scientific process of intervention development and evaluation. Therefore, while they cannot be removed at this stage of intervention development, changes can be made to make the assessments less burdensome. The redundancy may be eliminated by gathering data during intervention delivery, for example, during goal-setting calls; this way, the data can be collected in a useful manner for the participant. Incentives for assessment completion can also be adjusted to better reflect what would be appropriate for the moms.

While this study allowed us to gain an in-depth understanding of what worked for participants in the MOMs pilot studies, there are limitations. First, the study combines distinct cohorts with different intervention components and study procedures. Nonetheless, our results suggest that there were overlaps in benefits and recommendations that may help to inform future PA intervention efforts. Second, only 31% (19/61) of participants (19 out of a total of 61 participants enrolled in the 2 pilot studies) completed an exit interview. It is likely that there is a sampling bias in that participants who agreed to complete the exit interview may have had more helpful and positive things to say. For example, while participants overall regarded the fitness instructor in a positive light during the interviews, we know from recruitment efforts that there were participants who declined to participate in the research study upon discovering the fitness instructor’s identity. Future interventions may benefit from offering anonymous surveys at the end for those who did not complete an exit interview. Additionally, because we intentionally developed and tested the MOMs intervention for lower-income Black and Hispanic mothers, there may be limits to generalizability. All our participants were also English-speaking and from a single community organization. It is possible that the sociodemographic factors and history of this organization impacted the context in which the intervention occurred. Finally, while the intervention was designed to focus on mothers whose children were already enrolled in sports to address the barrier of time, this limits the ability to engage mothers whose children are not enrolled. Future interventions may benefit from additional ways to include minoritized mothers, for example, including PA sessions during other children’s activities.

Results from our qualitative analysis of exit interviews from a community-based PA intervention delivered to high-risk mothers and women caregivers during children’s sports practices suggest participants endorsed enjoyable experiences and perceived positive outcomes due to their engagement with the intervention. These findings align with quantitative data on the MOMs intervention [27,28], and support the continued development and testing of PA interventions delivered during children’s sports practices to address the unique needs of this high-risk population. In sum, low-income Black and Hispanic mothers and women caregivers may benefit from programs that include components focused on facilitating accountability, tailoring to ability, and increasing accessibility.