Background
The postpartum period is a critical time-period for interventions designed to prevent subsequent onset of diabetes in vulnerable populations. Worldwide, the number of people with diabetes is projected to rise from 382 million in 2013 to 592 million in 2030 [
1]. At the same time, the age at onset for type 2 diabetes is decreasing, highlighting the importance of identifying high-risk groups early in order to implement prevention efforts [
2]. One such high-risk group is women who develop gestational diabetes mellitus (GDM) or more mild forms of glucose intolerance in pregnancy [
3,
4]. Women who develop GDM in pregnancy have a 7-fold higher risk for future type 2 diabetes [
4]. Indeed, women with recent gestational glucose intolerance are at increased risk of progression to prediabetes or diabetes as early as one year postpartum [
5].
Latinas are the fastest growing minority group in the U.S. [
6] and are disproportionately affected by overweight and obesity, excessive gestational weight gain, and GDM [
7]. In addition, as compared to non-Latinas, Latinas with a history of GDM are less aware of diabetes risk factors and prevention strategies, such as physical activity, dietary behaviors, and weight management [
8]. In spite of these observations, due to socioeconomic circumstances, differences in educational background, health literacy, and language barriers, Latinas have had limited access to interventions that promote healthy lifestyles [
9].
Systematic reviews and meta-analyses have found that postpartum lifestyle interventions have the potential to lead to clinically significant weight reduction, improvements in physical activity, dietary behaviors, and other diabetes risk factors [
10‐
13]. However the number of postpartum interventions has been small, with widely diverse content and varying levels of participant adherence, and therefore their feasibility remains unclear. For example, poor retention rates have been reported in postpartum trials of exercise for weight management, with dropout rates ranging from 17 to 40% [
11]. In addition, the value of lifestyle interventions for women of different ethnic backgrounds remains unclear [
10,
11].
Lifestyle interventions regularly rely on study staff to implement the intervention and collect outcomes data directly from study participants. To date, little has been done to identify the individual components of lifestyle interventions, and the aspects of their implementation, that are critical to success [
14‐
16]. Qualitative research tools can identify staff persons’ experiences and knowledge that might otherwise remain hidden through more traditional quantitative measures [
17,
18]. These staff have a particularly intimate view of the lives of study participants, and may be uniquely qualified to identify factors that influence the ultimate success of an intervention.
To our knowledge, no studies to date have collected information on the experiences of health educators and interviewers involved in implementing a lifestyle intervention to postpartum women. Prior studies implemented during pregnancy have described the perspectives of health educators and interviewers in a randomized trial of prenatal smoking cessation [
14], the perspectives of interviewers in a case-control study of preterm birth [
15], and the perspectives of interviewers in the Danish National Birth Cohort Study [
16].
Therefore, we present results from a focus group held with a team of health educators and interviewers who worked on two postpartum randomized trials of diabetes prevention programs for Latinas: Estudio PARTO (Project Aiming to Reduce Type twO diabetes) and Proyecto Mamá. Our goal was to qualitatively assess the perspectives of these staff to inform future postpartum diabetes prevention programs.
Methods
Proyecto Mamá and Estudio PARTO
The trial protocols for Proyecto Mamá [
19] and Estudio PARTO [
9] have been previously published. Both trials were based at the ambulatory obstetrical practices of Baystate Medical Center in Western Massachusetts and were ongoing at the time of the focus groups.
Proyecto Mamá was a randomized controlled trial conducted from June 2014 to July 2018 to test the efficacy of a culturally and linguistically modified, individually-tailored lifestyle intervention to reduce excess gestational weight gain, increase postpartum weight loss, and improve maternal metabolic status among overweight/obese Latinas. Eligible women were recruited in early pregnancy and randomly assigned to a Lifestyle Intervention (n = 150) or a Comparison Health and Wellness (control) intervention (n = 150). Randomization was stratified based on age (< 30, > 30 years) and prepregnancy BMI (overweight > 25- < 30 kg/m2 vs. obese > 30 kg/m2). Within each stratum, a blocked randomization was used such that both treatment groups were assigned an equal number of times in each set of four sequentially enrolled patients.
Estudio PARTO was a randomized controlled trial conducted from January 2013 to December 2017 to test the efficacy of a culturally and linguistically modified, individually-tailored lifestyle intervention to reduce risk factors for type 2 diabetes and cardiovascular disease among postpartum Latinas with a history of abnormal glucose tolerance during pregnancy. Eligible women were randomly assigned to a Lifestyle Intervention (n = 150) or a Health & Wellness (control) Intervention (n = 150). Randomization occurred after the diagnosis of GDM (24–28 weeks gestation) and after completion of the baseline assessment. Randomization was stratified based on study site and the results of the diagnostic GDM screen (one vs. at least two glucose values during the diagnostic test meeting or exceeding thresholds).
For both trials, women were informed of the aims and procedures of the project, and eligible and interested women were consented in writing for participation in the study. Both studies were approved by the Institutional Review Boards of the University of Massachusetts-Amherst and Baystate Health.
Multimodal contacts (i.e., in-person, telephone, and mailed materials) were used to deliver the intervention from pregnancy (preparatory phase) through 12 months postpartum. This high-reach, low-cost strategy was selected such that findings could readily be translated into clinical practice in underserved and minority populations.
The interventions were delivered by bicultural and bilingual trained health educators. The lifestyle interventions utilized culturally and linguistically modified, motivationally targeted, individually tailored intervention materials. The lifestyle interventions were adapted from evidence-based approaches promulgated by the Institute of Medicine [
7], American College of Obstetrician and Gynecologists [
20]; and the American Diabetes Association [
21]. Specifically, the targets of the intervention were to achieve Institute of Medicine guidelines for gestational weight gain and postpartum weight loss [
7]; American Congress of Obstetrician and Gynecologist guidelines for postpartum physical activity [
20]; and American Diabetes Association guidelines for diet [
21]. The interventions drew from Social Cognitive Theory [
22] and the Transtheoretical Model [
23] and took into account the specific social, cultural, economic, and environmental challenges faced by Latinas [
8,
24,
25].
Assessments were conducted during pregnancy, and at 6-weeks, 6-months, and 12-months postpartum by trained bicultural and bilingual health interviewers blinded to the intervention arm. Measures included weight, physical activity assessed via the Pregnancy Physical Activity Questionnaire [
26], and diet measured via three unannounced 24-h dietary recalls.
Design
An open-ended qualitative focus group was led by an investigator (DL) not known to the health educators/interviewers and who was not involved with the trials. DL is a female doctoral student in Community Health Education with a Master’s degree in Public Health and more than three years of qualitative research experience in one-on-one and focus group interviewing. She moderated the sessions using a focus group discussion guide informed by prior qualitative work among research staff [
14,
17] The focus group began with an introduction where DL described her background, interest in the topic, and reasons for conducting the research. Following this introduction, open-ended questions were used to understand the health educators’ and interviewers’ perspectives on implementing this lifestyle intervention. Topics discussed included: 1) participant recruitment, 2) participant retention, 3) implementation of the lifestyle intervention, 4) assessment of behavior change, 5) overall challenges and rewarding aspects of the trial, and 6) recommended changes for future trials.
The focus group was conducted in a conference room at the University of Massachusetts and was limited to the moderator and the study staff. The focus group lasted ninety minutes, and was audio recorded and transcribed verbatim. Field notes were not made and data saturation was not discussed. Major themes were derived from the data through a grounded theory driven content analysis of the transcript. [
18]. A coding key was developed with definitions, descriptions, exemplary quotes organized underneath the major themes. Quotations were identified through an anonymous numbering system. Transcripts were not returned to participants for comment or correction. Through the thematic content analysis grounded theory [
18], commonalities and distinctions amongst perceptions of the staff were explored, yielding interpretive and illustrative findings.
Focus group participants
The complete roster of eight staff members (five health educators and three health interviewers) were invited in person to participate in the focus group. One additional health interviewer who was currently working on the trial was invited to participate but was not available. Staff provided written informed consent before participating in the focus group. All of the study staff had worked on both studies for a range of 2 to 6 years. Seven of the eight staff members were bicultural; specifically of Puerto Rican, South American, Central American, and Cuban heritage. All of the staff were female.
Staff had been trained for their roles in Estudio Parto during a three-day course, including sessions on the study protocol, and participant tracking and retention. Health educators were further trained on recruitment, obtaining informed consent, randomization, and motivational interviewing. Health interviewers were further trained on objective data collection skills and physical measurements. For all staff, this training was followed by a one-week period spent on site shadowing current staff.
Data management and analysis
We used NVivo 11 to organize and code the transcript data (NVivo qualitative data analysis software; QSR International Pty Ltd. Version 11.4.1). DL conducted multiple listenings and subsequent readings of the text to generate a provisional coding framework for each segment of text, and in consultation with AG, built a coding framework. DL thematically coded the transcript independently and created new codes where these were required. Identified themes are illustrated by selected anonymized quotes, which are characteristic of the data. Focus group participants did not provide feedback on the findings.
Discussion
Our qualitative study of the perceptions of study staff conducting a randomized trial of a diabetes prevention program among postpartum Latinas identified both enabling factors and challenging aspects. Enabling factors included: a) the staff’s belief in the importance of the intervention, b) opportunities associated with the longitudinal nature of the trial, c) belief that they could empower participants to make behavior change, d) benefits of flexible intervention sessions, and e) connection with participants due to shared cultural backgrounds. Barriers included: a) participant stressors: home, food, and financial insecurity, b) low health literacy, c) issues related to recent immigration to the continental U.S., d) handling participant resistance to behavior change, e) involvement of family members in assessment visits, f) limitations of the assessment tools, and g) time limitations.
Overall, the health educators and interviewers valued most the relationships they developed with study participants within the constraints of the research environment. The staff appreciated the insights into participant’s resilience, even in the face of the food, home, and financial insecurity faced by many. The overarching impression from staff was a sense of respect for the study participants. The rewarding aspect of these relationships was viewed, by the staff, as a counterbalance to the more challenging aspects of their roles.
The staff believed that their personal connections with participants was the greatest enabler of positive behavior change and participant retention. Other trials conducted in low-income and minority pregnant and postpartum populations have found that recruitment and retention rates are positively influenced by staff with training in patient-centered techniques grounded in a health equity framework, as well as a flexible protocol tailored to the unique needs of this population [
27,
28]. Our health educators and interviewers shared cultural backgrounds with participants and were similar in age, which they reported as having enhanced the quality of their relationships with participants. Consistent with prior studies in pregnant and postpartum minority women, the staff felt that the flexible nature of the protocol and participant-centered approach of the intervention facilitated both positive behavior change and trial retention [
27,
28].
An important finding of the focus group was the rich information on the stressors experienced by the participants. Staff insights into the extent of food and housing insecurity in participants’ lives would otherwise have been missed by the quantitative assessment tools used in the study. The staff’s perception that these factors were barriers to behavior change are consistent with prior reviews that found that financial, housing, and food insecurity negatively impact participant ability to fully participate in all components of the intervention [
12]. A postpartum lifestyle intervention designed to address the needs of women who are underserved is complicated by challenges that underserved women face. In other words, those women who could benefit the most from a postpartum lifestyle intervention may have the least time, energy, and resources to do so.
Staff on this trial were highly adept at balancing research expectations and their relationships with participants. Staff were directed to follow a fairly demanding study protocol within relatively certain time constraints, while engaging in a personal manner with participants and developing a strong relationship with some of them. Prior findings from qualitative interviews among staff suggest that trials such as ours consider the emotional impact these dual responsibilities place on study staff and include regular efforts to support staff members (i.e., debriefing strategies) [
17]. For example, study staff participated in monthly in-person staff meetings and weekly telephone meetings calls which included dedicated time for health educators and interviewers to discuss their ongoing relationships with study participants, ask each other for help, and openly process their experiences.
While the study staff appreciated the availability of intervention materials tailored to the Latina culture and language, they felt that these materials could be strengthened through further tailoring to specific Latina subgroups or through limiting the study to a single Latina subgroup. They also made specific suggestions for shifting the academic tone of the Spanish translations of the intervention and assessment materials to be more reflective of everyday, colloquial conversation. Overall, staff felt that shortening assessment tools, reducing their number, or converting some of the questions into open-ended responses would increase their time to implement the intervention and build relationships with study participants. Mixed-methods data collection strategies that prioritize process can be effective in informing intervention and health promotion [
29].
Staff noted that participant perceptions of family and cultural expectations were an important factor influencing the participants’ ability to comply with their behavior change goals. Family values were most often raised in the context of decisions about food choices, with staff suggesting that family members be involved in deeper, more meaningful ways in the intervention. These suggestions are consistent with those of a systematic review of lifestyle interventions in overweight/obese pregnant and postpartum women, which call for the development of interventions that target partners and family members [
10]. Partners, in particular, have been identified as important enablers of regular physical activity in childbearing and childrearing women [
10]. Broadening the intervention to engage the wider family network surrounding pregnant and postpartum women would help to address staff concerns that recommended levels of physical activity, dietary guidelines, and weight management behaviors may be in conflict with the participants’ cultural and family expectations.
While this study provides the first insights from study staff conducting a trial of a postpartum lifestyle intervention, it also faces several limitations. For example, it cannot be determined if the staff members’ perspective on the importance of their relationship with participants was similarly valued by the participants themselves. We also cannot establish, from this data, whether the quality of the staff-participant relationship was associated with positive behavior change. Future planned analysis of participant satisfaction surveys administered at the end of participant follow-up in Estudio PARTO and Proyecto Mamá will help to address this question.
While the focus group moderator was unknown to the study staff, and not associated with the study, social desirability bias may have constrained negative feedback. However, the fact that the staff reported a number of challenges to implementing the intervention reduces this concern. In addition, the staff were used to routine debriefing meetings where study challenges were discussed which likely facilitated their comfort level in sharing their perspectives. As compared to individual one-on-one interviews, the focus group method had the advantage of enabling discussion from and between the staff members, rather than being directed by the moderator.
It is important to acknowledge that this trial was focused on pregnant and postpartum Latinas of primarily Puerto Rican background, the largest Latina subgroup in the Northeast [
30]. Findings from this qualitative study may not be generalizable to other Latina subgroups as cultural differences in groups may be important to consider.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.