American Indians (AIs) experience type 2 diabetes mellitus (DM) and cardiovascular disease (CVD) at twice the rate of the US general population, and cardiometabolic disparities in morbidity and mortality rates are increasing in AIs (1–4). High CVD burden is partially attributable to DM prevalence, which increased from 9.3 to 20.2 per 1,000 population among childbearing AI women younger than 35 years from 1994 to 2004 (5). Women with a history of gestational diabetes mellitus (GDM) have a 20% to 50% chance of developing DM in the 5 to 10 years following GDM (6–8), and cumulative incidence approaches 70% in AI women (9). These women are also more likely to experience CVD events, even in the absence of concurrent DM (8). Diabetes is in itself a major risk factor for CVD, and women with DM who experience a CVD event have worse survival and poorer quality of life than men (10). A post-hoc analysis of self-reported women with prior GDM in the Diabetes Prevention Program demonstrated that women, including AI women, randomized to an intensive lifestyle program had a 53% risk reduction for DM (11).

Because GDM diagnosis may heighten a woman’s risk perception for her health and that of her offspring, the childbearing years present a unique window of opportunity for prevention (12–14). Despite the promise of lifestyle change to decrease cardiometabolic risk, implementing interventions to reduce weight and increase physical activity is challenging because of childbearing women’s busy lives (15,16). Furthermore, AI women with prior GDM may face challenges adopting healthy behaviors because of individual, interpersonal, and sociocultural influences (14). Little is known about AI women’s risk perceptions and prevention beliefs or their perspectives about barriers and facilitators to risk-reducing behaviors postpartum. Given the cardiometabolic risk GDM confers and the potential impact of early intervention, we sought to elicit perspectives of AI women with prior GDM to inform the development of a postpartum lifestyle modification intervention to be tested in this tribal community.

We conducted a cross-sectional study of AI women with prior GDM using questionnaires and focus groups and individual interviews from December 2012 to September 2013, in partnership with a large tribal health system in south-central Oklahoma. The institutional review boards of the University of Massachusetts Boston and the Chickasaw Nation of Oklahoma approved this study, and all women provided informed consent.

To recruit participants, providers from the Chickasaw Nation Department of Health (CNDH) and the study team, including staff from the Chickasaw Nation Nutrition Services, which administers the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), displayed study fliers and distributed business cards with the research team’s contact information in the medical center and community clinics. Eligibility criteria were the following: 1) self-identified as AI, 2) aged 19 to 45 years, 3) health care obtained through CNDH, and 4) diagnosed with GDM within past 10 years (confirmed by chart review) but not currently in a first-GDM pregnancy. Of 49 women who expressed interest in participating and were considered eligible, we scheduled 28 for participation. Of those who expressed interest but did not participate, many were either unable to schedule a focus group or interview or did not return follow-up telephone calls or emails. Five of the 28 participants had been diagnosed with DM since their GDM pregnancy; these women did not complete the Risk Perception Survey for Developing Diabetes (RPS-DD) but their data were included in all other analyses.

To our knowledge, this is the first study to elicit perspectives of AI women with prior GDM to inform the development of a postpartum lifestyle program to reduce cardiometabolic risk to be implemented in a tribal community. We found that most Oklahoma AI women believed they would inevitably develop DM, CVD, or both. Although they were optimistic that they could delay onset or decrease severity of disease, women discussed many individual, relational, and social barriers to postpartum lifestyle change. Most women expressed enthusiasm for a lifestyle change program after pregnancy, incorporating facilitators and using the Internet and text messaging, thereby reducing the need for face-to-face contact, a major barrier to participation.

In our study, Oklahoma AI women expressed moderate to high risk perception for DM and CVD following GDM. In contrast to findings from a study that examined DM risk perception among a sample of predominantly white women (17), a larger proportion of women in our study correctly considered themselves at moderate or high risk for DM. This finding is probably due to women’s pervasive family histories but also may have been, in part, because of the GDM education received through the tribal health system. We also found that AI women with a family history of CVD considered themselves at moderate or high risk for CVD, although many women generally perceived a lack of knowledge and familiarity with CVD compared with DM. This finding should be considered when designing programs for AI women with prior GDM, because these women are at increased risk for CVD (8).

Congruent with findings from an earlier study in this tribal health care system (14), most women in this study believed that the development of DM or CVD was inevitable. When asked in greater depth about their beliefs, women were optimistic that they could delay onset or decrease severity of disease with lifestyle changes. This is a crucial finding that should inform future lifestyle modification interventions in this community. It is possible that the public health message of diabetes prevention is not compatible with women’s beliefs and lived experiences, and messages related to delay may be more effective for certain populations than for others. Further research is needed to assess the value of such tailored messaging.

Although their risk perception was high, women in our study reported similar barriers to adopting lifestyle changes as women with prior GDM in other studies (19,20). Building on the lifestyle changes many women adopt during GDM pregnancies, the best strategy to reduce DM may be to implement tailored, relevant, postpartum lifestyle modification interventions (13). Adding to our 2010 study findings (14) and similar to the findings of a recent study in Montreal among women primarily of European origin (20), we found that women had a high perceived need for social support, particularly from partners and family members, and they expressed enthusiasm for a postpartum lifestyle change program incorporating facilitators to promote health behaviors. However, in contrast to the Montreal study’s finding that in-person interactions with peers and professionals were deemed essential, many Oklahoma AI women felt this approach would be too challenging and potentially not feasible for them, unless it was with an existing appointment. Notably, most participants could access the Internet daily from home and reported engaging daily in social networking. All women had a mobile telephone, and most reported having an unlimited texting plan; all but 1 participant reported sending and receiving text messages daily, a finding similar to trends found in a recent survey of this age group (21). This finding is important given the rural location of participants and other time-constraint–related challenges inherent in a face-to-face program; although women thought such a program would be motivating, they stated numerous barriers to attending it. Similar to findings of recent studies (13,20), women in our study were challenged to attend a single focus group, even when offered numerous options at various times of the day and week over a period of months and when several interviews were rescheduled to accommodate them. Additionally, many women who expressed interest in participating did not return telephone calls or emails from research team members or were unable to schedule an in-person meeting. This finding reflects a critical challenge in translating prevention strategies, as interventions must account for women’s competing priorities and time constraints (15).

One important finding, which should prompt further study, was the overall high rate of self-reported depression in this group of AI women with prior GDM. Tailoring postpartum lifestyle interventions to address depression could be critical to promoting healthy lifestyle change and reducing cardiometabolic risk among women who live with depression.

This study builds on our previous work in that it provides AI women’s perspectives about a postpartum lifestyle change program that would be feasible and acceptable in this community. Our findings suggest that a program using text messaging or the Internet that is tailored to address family based changes and that encourages family involvement may be effective for promoting lifestyle change in women with prior GDM in this tribal community. A recent systematic review described 9 lifestyle intervention studies to reduce DM risk in women with prior GDM, and many of these used a combination of in-person and technology-based modes of delivery; although most were pilot or feasibility studies, preliminary findings suggest lifestyle interventions can reduce DM risk (16). These interventions should be tested in larger, well-designed, randomized controlled trials and tailored to be culturally relevant for high-risk populations.

Our study has limitations. The regional, purposive sample of Oklahoma AI women may limit the generalizability of our findings. Furthermore, our final participation rate (26 of 48 eligible women, or 54%) limits findings, because the 22 women who did not participate may have held different views than participants. In addition to potential self-selection bias, recall and social desirability biases may have affected responses.

A family friendly mobile health or technology-based program that provides motivation and promotes accountability for lifestyle behavioral change, while accommodating women’s competing family and work demands, should be tested to reduce rates of DM and CVD in this high-risk group of AI women with prior GDM. Programs that will effectively delay or prevent DM and CVD in this population require an understanding of the greater sociocultural context, with culturally and situationally relevant tailoring of interventions (22–24). For this group of AI women who largely perceived the development of DM and CVD as inevitable, a tailored approach emphasizing the benefits of delaying disease onset, rather than preventing disease, may be the best approach. Because AI women are usually the primary stewards of family health (25), translating DM and CVD delay or prevention in this group is critical to prevention across the life course, the goal being to eliminate cardiometabolic health disparities in AI communities.