Comparing American Indian/Alaska Native Adolescent Daughters’ and Their Mothers’ Awareness, Knowledge, Attitudes, and Behaviors Regarding Risk for Gestational Diabetes: Implications for Mother-Daughter Communication on Reproductive Health

Abstract

Purpose:

The purpose of the study was to describe, compare, and examine associations at baseline of reproductive health awareness, knowledge, health beliefs, communication and behaviors related to gestational diabetes (GDM) and GDM risk reduction in a vulnerable population of both American Indian/Alaska Native (AIAN) adolescent girls and their mothers.

Methods:

Descriptive/comparative/correlational analyses examined multitribal baseline data on 149 mother-daughter (M-D) dyads (N = 298; daughter age = 12–24 years) enrolled in a longitudinal study to adapt and evaluate a culturally relevant diabetes preconception counseling (PC) program (Stopping-GDM). The associations between GDM risk reduction awareness, knowledge, health beliefs, and behaviors (eg, daughters’ eating, physical activity, reproductive-health [RH] choices/planning, M-D communication, daughters’ discussions on PC) were examined. Data collected online from 5 national sites.

Results:

Many M-D lacked awareness/knowledge of GDM and risk reduction. Both M-D were unaware of the girl’s risk for GDM. Mothers’ knowledge and beliefs on GDM prevention/RH were significantly higher than daughters. Younger daughters had greater self-efficacy healthy living. Overall sample reported low to moderate scores for both M-D communication and daughters’ GDM and RH risk-reduction behaviors.

Conclusions:

Knowledge, communication, and behaviors to prevent GDM were low in AIAN M-D, especially daughters. More than daughters, mothers perceive greater risk of GDM for daughters. Early culturally responsive dyadic PC programs could help decrease risk of developing GDM. Implications for M-D communication is compelling.

American Indian/Alaska Native (AIAN) women have a higher risk for gestational diabetes mellitus (GDM) and subsequent type 2 diabetes (T2DM) than non-AIAN, non-Hispanic White females.¹ Obesity increases the risk of GDM, the most common complication of pregnancy.² AIAN adolescents have nearly twice the US prevalence of obesity and GDM.^3–6 In turn, GDM and obesity increase the risk of pregnancy-related complications, including the later development of T2DM, to mothers and offspring.^7–9 GDM exposure in utero also increases the risk of obesity and T2DM in the offspring, creating a vicious cycle.⁹ AIAN adolescents have a much higher rate of teen pregnancy than the US all race population.¹⁰ In 2019, birth rates (births per 1000 females ages 15 to 19 years) for non-Hispanic White teens (11.4 births) were less than half of Hispanic teens (25.3 births), non-Hispanic Black teens (25.8 births), and AIAN teens (29.2 births), with the highest rates among all race/ethnicities.¹⁰ The need for early interventions targeting AIAN female adolescents to prevent obesity and GDM is imperative. Raising awareness about how to plan future pregnancies through preconception counseling (PC) and encouraging adoption of a healthy lifestyle prior to pregnancy could help to prevent obesity, GDM, and T2DM.

This project sought to minimize the risk of GDM among AIAN female adolescents and young adults (AYA) through a multiphase study by first culturally adapting Stopping GDM¹¹ for Native adolescents at risk for GDM from a previously validated diabetes PC program called Reproductive-health Education & Awareness of Diabetes in Youth for Girls [READY-Girls]¹² for non-AIAN adolescents with type 1 diabetes (T1DM) and T2DM. This was followed by phases to evaluate and disseminate the Stopping GDM intervention. READY-Girls is based on the Expanded Health Belief Model (EHBM), which posits that health-related beliefs influence the decision to engage in recommended health behaviors.^13–15 READY-Girls has been adapted for various populations.^16–18 The cultural tailoring of READY-Girls to meet the unique needs of AIAN adolescent girls at risk of developing GDM in the future is key to increasing engagement and effectiveness of the new Stopping GDM program.^19–22

The American Diabetes Association’s (ADA) Practice Recommendations specify that PC should “start at puberty.”⁷ This directive requires support from well-informed adult female caregivers of adolescents. Furthermore, a positive parent-adolescent communication has been associated with positive teen sexual health outcomes.²³ Therefore, the study utilized a mother-daughter (M-D) dyadic approach.

The purpose of this article is to describe, compare, and examine associations at baseline of reproductive health awareness, knowledge, health beliefs, communication, and behaviors related to GDM and GDM risk reduction in a vulnerable population of both the AIAN adolescent girls and their mothers, who were enrolled in a randomized controlled trial (RCT) before engaging in the Stopping GDM dyadic intervention. This is a unique opportunity to present pre-intervention data from the entire cohort regarding perceptions of GDM risk and health beliefs based on EHBM and reproductive-health (RH) characteristics in this representative sample. It is the only study that presents these kinds of data comparing mothers and daughters.

Methods

Research Design

This cross-sectional analysis used data from the baseline time point of the Stopping GDM trial to describe/compare/correlate RH and GDM risk-reduction awareness, knowledge, communication, and health beliefs (eg, benefits, barriers, severity, susceptibility, self-efficacy, intention) of both AIAN AYA daughters and their mothers (or other adult female caregiver) who were enrolled in a longitudinal RCT (during the second phase of this study) and daughters’ perceptions of body types and their behaviors (eg, healthy eating and physical activity, safe RH choices, use of effective family planning).

The Stopping GDM study had 3 phases: (1) revision of focus from T1DM/T2DM to GDM and cultural adaptation of the original READY-Girls program for AIAN AYA and their mothers and pilot testing Stopping GDM, (2) an RCT to implement and evaluate the adapted intervention, and (3) dissemination of the final program. The study protocol was approved by multiple review committees on the protection of human participants, including university Institutional Review Boards (IRBs) as well as Tribal IRBs, an Indian Health Service Area Office IRB, and the Indian Health Service National Institutional Review Board, according to the Declaration of Helsinki.²⁴

Prior to starting the multisite RCT with the Stopping GDM intervention, a feasibility pilot study was conducted among AIAN M-D dyads in an Urban Indian Health Organization to observe the ease of using the Stopping GDM online portal, including responding to the questionnaires.²⁴ Details and results of this pilot study are described elsewhere.²⁴

Participants and Setting

The analyses reported here only used pre-intervention baseline data for all 149 dyads (n = 298 participants). Each dyad included an AIAN AYA who was at risk for developing GDM and her mother. Participants were from 5 sites (Northwest Coast, Southwest, Northeast, 2 in the Southern Plains). Eligibility criteria for daughters included being English-speaking, age 12 to 24 years, and at risk for GDM, defined as being self-reported AIAN, having a first-degree relative with T2DM, having prediabetes or metabolic syndrome, or being overweight. Self-reported AIAN and having a first-degree relative with T2DM were the only required at risk for GDM criteria. Exclusion criteria included having diabetes using ADA criteria⁷ (A1C ≥ 6.5%), being pregnant, and previously having been diagnosed with GDM. Mothers had to be AIAN, English-speaking, and the biological mother of the participating AYA or another adult female caregiver (stepmothers, grandmothers, aunts, and older sisters). Mothers/caregivers had to be ≥18 years old, English speaking, and living in the same household with the younger participant. For the purpose of this article, “mother” is used to refer to biological mothers and other adult female caregivers.

Data Collection and Protocols

Consents were obtained from adults and assents from minors among mothers and daughters at the baseline visit. Online questionnaires were administered pre- and postintervention activities at the baseline visit and again at 3-, 6- and 9-month follow-up visits. Data were collected in a clinic or community setting, done separately for daughter and mother via online parallel questionnaires or via scannable data collection forms. The baseline study visit lasted approximately 2 hours for the intervention group. (These participants completed the baseline questionnaires in addition to viewing a 45-minute video as the first component of the intervention.) This article presents baseline data collected before intervention activities were conducted.

Measurements

In addition to the online questionnaires, physiological parameters were measured.

Body mass index (BMI) of participants’ weight was measured with a balanced scale while they were wearing light clothing and no shoes; height was measured using a stadiometer with participants’ shoes removed. BMI was computed as the measured weight (in kilograms) divided by measured height (in meters) squared.²⁵ BMI percentiles specific to sex and years of age were calculated using CDC growth charts.²⁶

A1C was used to screen for undiagnosed T2DM using a finger stick with blood analyzed immediately using DCA2000, DCAVantage + analyzer (Bayer Corporation, Elkhart, IN) A1C AS100 analyzer (Axis-Shield Inc, Norton, MA). Diagnosis of prediabetes and diabetes was categorized according to ADA guidelines.²⁷ AYA participants identified with prediabetes and diabetes were referred to their health care provider (HCP) for follow-up. Girls with prediabetes were allowed to participate in the RCT.

Participants then completed questionnaires using an online portal on a study laptop. Parallel questionnaires for both daughters and mothers evaluated demographics and outcomes, including knowledge about RH, GDM, and prevention/healthy lifestyle; self-efficacy (daughters only); communication; health beliefs (eg, benefits, barriers, severity, susceptibility, intention); and daughter’s behaviors. These measures were culturally adapted from existing valid and reliable scales that were used in previous READY-Girls studies.^13,28 The constructs and their operational roots remained intact.²⁴

Two knowledge scales were used for this study: Diabetes Prevention Comprehensive Knowledge²⁹ and Reproductive-Health & GDM Knowledge Questionnaire.³⁰ Each measure was computed as the percentage of questions answered correctly. Both scales had “don’t know” as a response, and this response was treated as an incorrect answer. The Diabetes Prevention Comprehensive Knowledge²⁹ scale included 11 multiple choice items addressing knowledge of healthy lifestyle behaviors that can help women prevent GDM. The number of items were computed as being answered correctly (possible item score range = 0–11). The Reproductive-Health & GDM Knowledge³⁰ scale included 13 true/false items. The sum of the items was computed as answered correctly (possible item score range = 0–13). The 2 subscale scores were also computed as (1) general RH knowledge, including birth control items (possible item subscale score range = 0–6), and (2) GDM knowledge (possible item subscale score range = 0–7). Both knowledge questionnaires had adequate reliability (αs = 0.65–0.71).^24,29

Communication was also measured. The degree of openness with general communication between daughters and mothers was evaluated using a subscale of the Parent-Adolescent Communication Scale (PAC),²³ which has 10 Likert-type items using 5-point scales where 1 = strongly disagree and 5 = strongly agree (degree of openness subscale α = 0.87).²³ The PAC assesses perceptions of mothers’ and daughters’ openness to communication and their communication skills.²³ Reliability for this study was strong (α = 0.92).

Both daughters and mothers reported their intention to engage in and actual initiation of discussions regarding GDM (eg, healthy weight) and RH (eg, birth control) with each other and with an HCP. The Initiating Discussion scale has four subscales: (1) intention to talk with HCP (3 items), (2) intention to talk with mother/daughter (3 items), (3) actual discussion with HCP (4 items each for GDM and RH), and (4) actual discussion between mother and daughter (3 items each for GDM and RH). The response scale is dichotomous with yes/no response choices for each item (yes = 1, no = 0). (Intention to initiate discussion and intention behaviors α = 0.90).³⁰

Health beliefs were examined using the Reproductive Health and Diabetes Questionnaire (EHBM scale).³⁰ Based on the EHBM,¹⁴ which assessed perceived susceptibility (4 items), severity (5 items), benefits (8 items), and barriers (3 items) regarding GDM using Likert-type scaling (1 = strongly disagree, 5 = strongly agree). Each construct is computed as a summary score with higher scores suggesting stronger beliefs/attitudes. Internal consistency is adequate based on the EHBM¹⁴ (Cronbach alphas: susceptibility = 0.74; severity = 0.94; benefits = 0.88; barriers = 0.97) and for the Stopping GDM sample (Cronbach alphas: susceptibility = 0.89; benefits = 0.83; barriers = 0.80). Included within the EHBM, self-efficacy was measured as the daughters’ confidence (ie, how sure they were) in their ability to engage in a healthy lifestyle and prevent an unplanned pregnancy to reduce the risk of GDM. Items had Likert-type scaling from 1 = not at all sure to 10 = totally sure. Summative scores were calculated (α = 0.88). Self-efficacy³⁰ was measured as 3 subscale scores: Family Planning (9 items with a possible score range of 9–90), Healthy Living (8 items with a possible score range of 8–80), Pregnancy Planning (12 items with a possible score range of 12–120).²⁴

Intention to seek special care and advice from HCP to attain a healthy weight when planning a pregnancy was measured by a single item. Intention to use family planning was measured by 4 items (α = 0.84).³⁰ Both intention measures used a Likert-type scaling (very unlikely = 1, very likely = 7), with higher scores indicating greater intention.

Healthy lifestyle behavior measures included healthy eating, habitual physical activity, and RH (eg, family planning) for daughters. Eating healthfully and habitual physical activity were measured through self-report using the US Department of Agriculture Household Food Security Module (6 items)³¹ and Youth Risk Behavior Surveillance System Healthy Eating (9 items) and Physical Activity (6 items).³²

Adolescent women can reduce unplanned pregnancies and reduce GDM risk through healthy family planning. Actual RH behaviors in daughters were measured as using family planning/abstinence/delaying sexual debut to reduce the risk of an unplanned pregnancy and seeking PC. The Reproductive-Health and Diabetes Questionnaire measures all of the aforementioned variables (eg, family planning method as weighted summary of the teen’s most frequently used contraception as reported in Trussell’s algorithm).^33,34 For pregnancy planning behaviors, young women’s intentions and actual behaviors for seeking preconception counseling and care were measured by self-report at each time point. Seeking PC is a list of dichotomous items with a “yes or no” response. All participants reported whether they have received any (or additional) PC from their health team and checked from a list based on the ADA PC recommendations^27,35 of actual components of PC received. Another pregnancy planning behavior is the use of effective family planning to reduce risk of a pregnancy until healthy weight is achieved. To enhance participant recall of retrospective data for sexual activity and birth control use, questions were asked according to “ever in your life” and “past 3 months.”³⁴

To evaluate baseline awareness of GDM and risk of pregnancy-related complications, daughters and mothers answered the following open-ended question: What do you know about gestational diabetes and pregnancy? Similar questions and approach were previously used in an earlier study.³⁶

The measure body image, the Body Image Figure³⁷ was used with permission to measure the daughters’ perceived, desired, and healthy body size selections (see Figure 1). Images scored according to body sizes were A = 1, B = 2, C = 3, D = 4, E = 5, F = 6, G = 7, H = 8: an ordinal scale measure with higher numbers equaling larger body sizes.

Figure 1.

Body Image Figure used in assessment of perceived, desired, and healthy body size.³⁷

Image A = 1, B = 2, C = 3, D = 4, E = 5, F = 6, G = 7, H = 8.

Data Analysis

Descriptive/comparative/correlational analyses were conducted to analyze quantitative data collected at the pre-intervention assessment during the baseline visit using IBM SPSS Statistics (Version 28, IBM Corp, Armonk, NY). Based on the level of measurement of the variable and its observed data distribution, appropriate descriptive statistics were computed (ie, means with standard deviations, reported as mean ± SD; median interquartile range [IQR]; ranges, reported as minimum-maximum) for continuous variables with either ratio or interval scaling and frequency counts and percentages for categorical variables with nominal or ordinal scaling) based on the participant’s role (daughter, mother). Comparative statistics were used to examine differences between mother and daughter and within the M-D dyad using either the paired t test or the Wilcoxon signed-rank test with exact estimation of p values. Bivariate correlational analyses were performed to examine the relationship between the variables measuring the key study constructs and daughter’s age and mother’s diabetes history (T1DM, T2DM, and/or GDM) as well as between BMI and body type variables and among body variables. The level of statistical significance was set at .05 for 2-sided hypothesis testing. Descriptive analysis was used for the opened-ended question addressing GDM awareness. Categories were derived by two members of the research team using content analysis and frequency of responses.

Results

Table 1 describes the cohort’s baseline characteristics at the pre-intervention baseline visit. Daughters’ mean age ± SD was 16.7 ± 3.0 years, range = 12.0 to 24.5 years; 86% were currently a student; ≈50% were multiracial. Mothers’ mean age was 44.1 ± 9.3 years, range = 18.3 to 75.2 years; and 36% were college graduates. AIAN adult female-caregiver participants were mostly daughters’ biological mothers (80.5%), followed by grandmothers (8.1%). Forty-nine percent of the biological mothers reported having had some form of diabetes.

Table 1.

Sociodemographic Characteristics of Mother and Daughter Participants at the Initial Assessment Baseline Visit

Characteristic	Mothers (N = 149) Mean ± SD (min-max) or n (%)	Daughters (N = 149) Mean ± SD (min-max) or n (%)
Type of site
Reservation	68 (45.6)	—
Urban	81 (54.4)	—
Age (y)	44.1 ± 9.3 (18.3–75.2)	16.7 ± 3.0 (12.0–24.5)
Race and ethnicity
American Indian	129 (87.2)	118 (79.2)
Alaska Native	1 (0.7)	0 (0.0)
White	6 (4.0)	8 (5.4)
Black or African American	8 (5.4)	14 (9.4)
Hispanic or Latino	2 (1.4)	2 (1.3)
Native Hawaiian or other Pacific Islander	0 (0.0)	2 (1.3)
Other	2 (1.4)	5 (3.4)
More than one racial/ethnic background: yes	75 (50.3)	70 (56.0)
Highest grade completed
8th grade or less	0 (0.0)	39 (26.2)
Some high school	5 (3.4)	66 (44.3)
High school graduate or GED	20 (13.4)	21 (14.1)
Vocational or some college	56 (37.6)	18 (12.1)
Associate’s degree	30 (20.1)	4 (2.7)
Bachelor’s degree	25 (16.8)	1 (0.7)
Graduate degree	10 (8.7)	0 (0.0)
Currently a student	—	133 (89.3)
Household income
<$10 000	12 (8.8)	—
$10 000 – <$20 000	16 (11.8)	—
$20 000 – <$30 000	14 (10.3)	—
$30 000 – <$40 000	14 (10.3)	—
$40 000 – <$50 000	18 (13.2)	—
$50 000 – <$75 000	33 (24.3)	—
≥$75 000	29 (21.3)	—
Unknown	13 (8.7)	—
Employment status: employed	103 (69.6)	41 (27.7)
Marital status
Single	37 (24.8)	146 (98.0)
Married/partnered	87 (58.4)	0 (0.0)
Separated, divorced, or widowed	25 (16.8)	3 (2.0)
Health insurance: yes	124 (83.2)	100 (92.6)
Private health insurance or HMO	68 (54.8)	17 (17.0)
Medicaid or medical assistance	46 (37.1)	29 (29.0)
Other	14 (11.3)	12 (12.0)
Unsure	2 (1.6)	36 (36.0)
Relationship with participating AYA
Mother	120 (80.5)	—
Grandmother	12 (8.1)	—
Aunt	7 (4.7)	—
Other	10 (6.7)	—
Diabetes experience
Gestational diabetes	29 (19.5)	—
Type 1 or 2 diabetes	23 (15.4)	—
Type 2 diabetes	21 (14.1)	—

Abbreviation: AYA, adolescent or young adult.

Table 2 summarizes and reports within-dyad comparisons for diabetes-related constructs for mothers and daughters and behavioral outcomes for daughters at the pre-intervention assessment at the baseline visit. Pre-intervention knowledge scores were low (<60%) on average for both mothers and daughters on GDM and its prevention; daughters also scored low on prevention of T2DM and on RH knowledge. Compared to daughters, all knowledge scores were significantly higher on average (P < .001) for mothers. For health beliefs constructs, all scores, except perceived severity of GDM, were higher on average for mothers than daughters and were significantly higher (P < .001) for perceived susceptibility to developing GDM and benefits of a healthy lifestyle. Compared to their daughters, mothers had a greater perception that their daughters were at risk of developing GDM during a pregnancy, and they had a stronger belief in the benefits of healthy eating and physical activity to mitigate the risk of GDM. Daughters had moderate levels of self-confidence (self-efficacy) in their ability to engage in healthy living, pregnancy planning, and family planning behavior. Regarding communication, mothers perceived significantly greater open communication with their daughters (P < .001) than did their daughters. Mothers initiated discussion on GDM and RH significantly more frequently (P < .001) than did their daughters. Behaviorally, daughters reported low mean scores on healthy eating and physical activity. As expected, based on inclusion criteria, the A1C of all the daughters were in the normal range. The mean BMI was 28.2 ± 8.0 kg/m² (range = 14.5–53.9) for this sample of daughters, falling between the 90th and 95th percentile, indicating overweight status.²⁶

Table 2.

Diabetes-Related Constructs for Mother and Daughter Participants and Behavioral Outcomes for Daughters at the Initial Assessment at the Baseline Visit

Construct	Mothers (N = 149) Mean ± SD (min-max)	Daughters (N = 149) Mean ± SD (min-max)	P value
Knowledge
Diabetes prevention	74.7 ± 15.9 (18.2–100.0)	57.6 ± 16.7 (0.0–100.0)	<.001
RH	86.0 ± 15.9 (0.0–100.0)	56.9 ± 26.3 (0.0–100.0)	<.001
GDM	49.2 ± 20.9 (0.0–71.4)	20.9 ± 20.4 (0.0–71.4)	<.001
Expanded health belief model constructs
Perceived susceptibility (GDM-related)	8.9 ± 4.0 (3.0–15.0)	7.0 ± 3.8 (3.0–15.0)	<.001
Perceived severity (GDM-related)	13.8 ± 2.1 (5.0–15.0)	14.0 ± 1.7 (6.0–15.0)	.273
Perceived benefits (GDM-related)	18.9 ± 1.9 (11.0–20.0)	16.7 ± 3.5 (4.0–20.0)	<.001
Perceived barriers (GDM-related)	4.9 ± 2.2 (2.0–10.0)	4.8 ± 2.1 (2.0–10.0)	.624
Intention (GDM-related), n = 145	—	4.4 ± 2.3 (1.0–7.0)	—
Intention (family planning; n = 142)	—	17.7 ± 5.1 (4.0–28.0)	—
Self-efficacy
Family planning (n = 146)	—	69.8 ± 22.3 (9.0–90.0)	—
Healthy living (n = 141)	—	48.2 ± 13.9 (8.0–76.0)	—
Pregnancy planning (n = 143)	—	85.4 ± 22.3 (24.0–120.0)	—
Communication
Parent-adolescent communication
Degree of openness	41.2 ± 7.8 (11.0–50.0)	37.4 ± 9.5 (10.0–50.0)	<.001
GDM and RH communication
Intention to initiate discussion (total)	28.3 ± 11.0 (6.0–42.0)	20.6 ± 10.7 (6.0–42.0)	<.001
Intention to initiate discussion (HCP)	12.7 ± 6.5 (3.0–21.0)	9.8 ± 5.9 (3.0–21.0)	<.001
Intention to initiate discussion (dyad member)	15.6 ± 5.5 (3.0–21.0)	10.7 ± 5.9 (3.0–21.0)	<.001
Actual initiation of discussion with HCP (GDM)	2.2 ± 1.4 (0.0–4.0)	1.8 ± 1.4 (0.0–4.0)	.008
Actual initiation of discussion with HCP (RH)	1.0 ± 1.2 (0.0–4.0)	0.9 ± 1.2 (0.0–4.0)	.906
Actual initiation of discussion with dyad member (GDM)	2.9 ± 0.9 (0.0–4.0)	2.3 ± 1.3 (0.0–4.0)	<.001
Actual initiation of discussion with HCP (RH)	2.0 ± 1.3 (0.0–4.0)	1.3 ± 1.3 (0.0–4.0)	<.001
Health behaviors
Eating healthfully (n = 145)	—	9.0 ± 5.6 (0.0–31.0)	—
Physical activity (n = 145)	—	3.4 ± 2.2 (0.0–7.0)	—
Clinical outcomes
A1C, % (n = 141)	—	5.3 ± 0.3 (4.5–6.0)	—
Body mass index, kg/m2 (n = 146)	—	28.2 ± 8.0 (14.5–53.9)	—

Abbreviations: GDM, gestational diabetes; HCP, health care provider; RH, reproductive health.

Table 3 presents associations between daughter’s age and mother’s diabetes status with diabetes-related constructs for mothers and daughters and behavioral outcomes for daughters. Adolescent development was dichotomized into younger (age < 16 years) and older (age ≥ 16 years) age groups based on maturity. Older daughters had significantly (P ≤ .01) higher knowledge scores and perceived greater benefit of healthy lifestyle to decrease risk of GDM and initiated more discussions with HCPs and their mothers regarding RH and GDM, and older girls had a significantly higher BMI (P ≤ .001). Younger daughters had significantly (P ≤ .03) greater self-efficacy for healthy living and participated in more physical activity. Mothers with a history of diabetes had significantly (P ≤ 0.03) higher knowledge scores, believed their daughters to be more susceptible to developing GDM, and perceived greater barriers to decreasing their daughters’ risks.

Table 3.

The Associations Between Daughter’s Age and Mother’s Diabetes Status and Diabetes-Related Constructs for Mother and Daughter Participants and Behavioral Outcomes for Daughters

Construct	Daughter’s age (<16 y, ≥16 y) (N = 149) r, P value	Mother’s history of DM (no, yes) (N = 149) r, P value
Knowledge
Diabetes prevention	.323, <.001	.240, .003
RH	.497, <.001	.223, .006
GDM	.212, .010	.208, .011
Expanded health belief model constructs
Perceived susceptibility (GDM-related)	−.008, .924	.235, .004
Perceived severity (GDM-related)	.062, .467	−.001, .990
Perceived benefits (GDM-related)	.213, .010	.014, .865
Perceived barriers (GDM-related)	.103, .217	.188, .024
Intention (GDM-related; n = 145)	−.075, .369	—
Intention (family planning; n = 142)	.077, .361	—
Self-efficacy
Family Planning (n = 146)	.059, .482	—
Healthy living (n = 141)	−.191, .023	—
Pregnancy planning (n = 143)	−.038, .651	—
Health behaviors
Eating healthfully (n = 145)	−.050, .547	—
Physical activity (n = 145)	−.177, .033	—
Communication
Parent-adolescent communication
Degree of openness	.058, .487	−.039, .638
Extent of problems	.031, .713	−.023, .783
GDM and RH communication
Intention to initiate discussion (total)	.145, .083	.058, .483
Intention to initiate discussion (HCP)	.138, .104	.047, .579
Intention to initiate discussion (dyad member)	.133, .116	.071, .398
Actual initiation of discussion with HCP (GDM)	.037, .658	.037, .654
Actual initiation of discussion with HCP (RH)	.401, <.001	.051, .538
Actual initiation of discussion with dyad member (GDM)	.241, .003	.122, .139
Actual initiation of discussion with dyad member (RH)	.325, <.001	.061, .463
Clinical outcomes
A1C, % (n = 141)	.095, .262	—
Body mass index (kg/m2; n = 146)	.343, <.001	—

Table 4 presents RH behavioral outcomes for daughters at the baseline pre-intervention assessment. In this sample, 33% of the daughters had been sexually active, with a mean age of sexual debut of 15 years. Ninety-one percent of them reported ever using birth control. Of those sexually active, 45% always used birth control, and 21% always used a combination method. Most (62%) had at least one episode of sex without birth control. Seven daughters (15%) reported having had a prior pregnancy. The most frequent types of PC they had received had been information about how to keep from getting pregnant, advice about healthy weight, and the importance of not smoking and drinking during a pregnancy. The most frequently reported birth control methods ever used were condoms and withdrawal. As expected, significantly more (P ≤ .02) older adolescent girls had reported having been sexually active, had received more advice about what to use to keep from getting pregnant, and had had a pregnancy test.

Table 4.

Reproductive Health Behavioral Outcomes for Daughters at the Initial Assessment Baseline Visit

Construct	Daughters (N = 149) Mean ± SD (min-max) or n (%) Sexually active = 49	Daughter’s age (<16 y, ≥16 y) (N = 149) r, P value
Major source of RH information	—	.294, .258
Parents	51 (35.9)	—
Health care providers	23 (16.2)	—
Sex education class	17 (12.0)	—
Other	51 (35.9)	—
PC
Ever received PC	3 (2.6)	.026, .999
Types of PC received or given
Information about how to keep from getting pregnant	121 (89.0)	.218, .013
Information about how important it is to plan a pregnancy	82 (65.1)	.027, .851
Advice about what to use to keep from getting pregnant	116 (87.2)	.351, <.001
Tests for health problems	100 (80.6)	.138, .165
Advice about healthy weight	109 (80.7)	.021, .830
Information about pregnancy	106 (78.5)	.165, .060
Advice about not smoking and not drinking	130 (90.3)	.032, .781
Pregnancy test	51 (38.6)	.229, .011
Gynecological exam	27 (22.0)	.102, .276
Advice about pregnancy care	40 (31.7)	.063, .562
Advice about how to lower stress	60 (48.4)	.038, .718
RH behavior
Ever sexually active	49 (33.1)	.351, <.001
Age of sexual debut, y (n = 46)	15.0 ± 2.4	.574, <.001
Ever pregnant	7 (14.9)	.204, .318
Always use BC	21 (44.7)	—
Always use condom with BC	10 (21.3)	—
Risky behavior
Frequency of sex in the last 3 mo	6.6 ± 9.6 (1.0–40)	.062, .809
No. of partners in the last 3 mo	1.2 ± 0.5 (1.0–3.0)	.121, .999
Ever had sex without BC	29 (61.7)	.062, .716
BC use
Ever used BC	42 (91.3)	.586, .001
Types of BC ever used
BC pills	18 (37.5)	.185, .282
Depo shot	11 (22.9)	.280, .089
Patch	1 (2.1)	.075, .999
Vaginal ring	2 (4.2)	.107, .999
IUD	2 (4.2)	.107, .999
Condoms	31 (64.6)	.264, .134
Withdrawal	19 (39.6)	−.004, .999
Implant in the arm	12 (25.0)	.178, .414
Plan B (emergency contraception)	9 (18.8)	.246, .172

Abbreviations: BC, birth control; PC, preconception counseling; RH, reproductive health.

Table 5 summarizes the daughter’s perception of body type at the initial assessment at the baseline visit. The Body Image Figure scale provided a visual assessment of perceived, desired, and healthy body size (Image A = 1, B = 2, C = 3, D = 4, E = 5, F = 6, G = 7, H = 8).³⁷ In this sample, the figure most frequently selected for self-perception “Most like you” was D, with an observed range from A to H. The most frequently selected figure for desired body size, “would you want to look like,” was C, with an observed range of A to F. The “most healthy” figure most frequently selected was C, with an observed range of A to E. The majority of the adolescents in this study indicated that the most healthy size and desired size were smaller than their perceived actual size. Perception of body types was significantly (P ≤ .03) associated with age, whereby older adolescent participants (age ≥ 16 years) selected larger body image figures for all questions compared to younger adolescent participants (age < 16 years). Most daughters (94%; n = 116) reported that they would “like to have a healthier body weight.” BMI was positively associated with “would you like to have a healthier body weight” (P < .0001), whereby those that would like to have a healthier body weight had a higher BMI on average than those who did not want a healthier body weight. Furthermore, BMI was significantly positively associated with daughter’s perceived body type (ie, most like you; r_s = .826, P < .0001) and desired body type (ie, want to look like; r_s = .493, P < .0001), whereby daughters reporting larger perceived and desired body image figures had on average higher BMI values. Although perceived and desired body types were positively associated (r_s = .620, P < .0001), daughters with larger perceived body types (“D” through “H”) generally desired smaller body types than was perceived.

Table 5.

Daughter’s Perception of Body Type at the Initial Assessment at the Baseline Visit

Construct	Daughters (N = 149) Median IQR (min-max)	Daughter’s age (<16 y, ≥16 y) (N = 149) r, P value
Body type
Most like you	4.0 ± 3.0 (1.0–8.0)a	.353, <.001
Would you want to look like	3.0 ± 2.0 (1.0–7.0)a	.300, <.001
Is the most healthy	3.0 ± 1.0 (1.0–5.0)a	.189, .023

^aMedian interquartile range (min-max).

Lastly, the open-ended question on awareness of GDM and pregnancy revealed that most daughters were unaware of GDM or of their own level of risk. Specifically, 78% of daughters reported “no knowledge” of GDM. No daughters linked GDM with diet and/or a woman’s body weight or with complications. Furthermore, 26% of mothers reported “no knowledge” of GDM. Only 1% of mothers listed GDM with diet and/or a woman’s body weight. Only 33% of mothers indicated GDM impacts both baby and mother and “causes problems” and “is bad.” Fifteen percent of mothers (and no daughters) mentioned family experiences with GDM: “Well, I heard about it before. A cousin had [GDM] each time she had a baby, and her babies were born bigger than average. And it can cause complications during pregnancy. But I really do not know very much about gestational diabetes.” Only one mother (and no daughters) described how to prevent GDM: “A healthy lifestyle and eating habits could help prevent gestational diabetes.”

Discussion

This article described, compared, and examined associations between AIAN daughters and their mothers from online questionnaires from 5 sites in the United States regarding GDM awareness, knowledge (about reproductive health, GDM, and prevention/healthy lifestyle), health beliefs, daughters’ GDM risk-reduction behaviors (healthy eating and physical activity, safe RH choices, use of effective family planning), M-D communication regarding these topics, and daughters initiating (or daughter initiated) discussions on PC with HCPs. These quantitative results confirm the previous qualitative findings^19,20 suggesting that participants lack awareness and knowledge of GDM and their risk for developing GDM. These findings highlight the need for new health programming to mitigate risk for unplanned pregnancies and GDM and to promote healthy pregnancy and birth outcomes. The previous qualitative findings from adolescent AIAN girls and AIAN women with T2DM and/or history of GDM^19,20 have been used in conjunction with findings of focus groups of other key stakeholders^21,22 to inform the development of a culturally tailored GDM risk-reduction program for AIAN adolescents called Stopping GDM.^11,38

Participants in this baseline assessment lacked awareness or understanding of GDM and its risk factors, placing daughters at risk for unplanned pregnancies and for developing GDM and possible complications to themselves and their offspring. Achieving a healthy body weight prior to pregnancy is key to minimizing risk of GDM.³⁹ Women should understand the relationship between risk for GDM and their pre-pregnancy weight and receive education, resources, and support to plan and achieve a healthy weight prior to pregnancy.²⁰ GDM risk-reduction education and PC should start at puberty and be reinforced throughout adolescence and childbearing years, well before a woman plans to get pregnant.^7,20 In this study, 7 (15%) daughters reported a previous pregnancy. Because the majority of teenage pregnancies are unplanned,⁴⁰ this poses a challenge for adolescents, especially for AIAN females, for whom the risk for GDM and the prevalence of unplanned adolescent pregnancy is also higher.^10,40–43 Therefore, raising awareness through PC should begin in early adolescence, ideally prior to her first sexual encounter. Both READY-Girls and Stopping GDM are based on the EHBM, and this health behavior change theory supports engaging increased perceptions of self-susceptibility and severity (risk perception) as a step toward healthy behavior change.¹⁴

Stopping GDM is a dyadic program that incorporates maternal support. Maternal communication, especially in providing RH information, has been associated with positive sexual health outcomes among adolescent females.²³ Gurr⁴⁴ notes the importance of traditional forms of RH knowledge sharing among generations of Native women, from mothers and grandmothers, passed on in stories, song, and ceremonies.

Women’s perceptions of having limited knowledge¹⁹ on GDM were confirmed by low knowledge scores by both mothers and daughters regarding GDM. Daughters also lacked an understanding of RH. Mothers had significantly higher scores than daughters in all knowledge categories, especially those mothers who had some type of diabetes. However, mothers with some type of diabetes also perceived greater barriers to decreasing their daughters’ risks. This strongly indicates the need to provide GDM and RH education to mothers and to the adolescents because they will be providing support to their daughters in the future.

Mothers’ knowledge has been associated with daughters’ risk perception.⁴⁵ In this study, mothers overall had a greater perception that their daughters were at risk of developing GDM during a pregnancy and a stronger belief in the benefits of a healthy lifestyle to reduce these risks compared to their daughters. Mothers’ knowledge and beliefs could influence those of their daughters. Although mothers reported initiating discussion on GDM and RH with their daughters, their mean score for actual discussion was only moderate to low, indicating a need for further intervention. Stopping GDM is tailored for female adolescents and young adults, starting at puberty, who are at risk for GDM and targets decision-making regarding effective family planning and seeking PC. This directive requires support from well-informed mothers of teens. Mothers play a vital role in helping and initiating discussions regarding RH with their daughters and reinforcing PC.^23,46 Mother-daughter dyad social support systems have been effective for interventions related to other health-related topics.^40–43,47 Mother-daughter dyadic analyses can be used in the future to explore possible mediating and moderating roles of M-D communication and support about RH on the relationship between the Stopping GDM intervention and sustainable outcomes.³⁶

Older adolescents and young adults (age ≥ 16 years) in this sample tended to be overweight and did not routinely practice healthy lifestyle behaviors such as healthy eating and physical activity. Most daughters reported that they would like to have a healthier body weight, which has implication for PC and education. Yet older adolescent girls (age ≥ 16 years) reported less confidence with healthy living and participated in less physical activity. This may be due to less frequent participation in school or organized team sports compared to younger adolescent girls. These results further indicate the need and opportunity to introduce an early GDM risk-reduction intervention, starting at puberty,⁷ and repeating this message throughout adolescence to maintain and enhance self-efficacy and healthy behavior.

A third of the adolescents in this sample had been sexually active, similar to the previous finding from a READY-Girls study (32%).¹³ Among the sexually active adolescents in this current study, less than half reported having used effective birth control consistently, putting the majority at risk for an unplanned pregnancy. This was also reported in earlier READY-Girls studies.^28,48 Fifteen percent of the current AYA participants reported having had a pregnancy, which is consistent with the reported incidence for pregnancies among AIAN adolescents.¹⁰

This study had many strengths and some limitations. Data were collected from both daughters and mothers, allowing for exploration of dyadic associations. The sample was recruited from 5 sites located at different US geographical locations to enhance generalizability. This article reports on baseline cross-sectional data, therefore temporal relationships and possible causality cannot be concluded. Only perceived available support was examined, and future studies should include actual support provided.

In conclusion, both AIAN mothers and daughters lacked awareness and knowledge of the girls’ risk for developing GDM and mitigating strategies. GDM and diabetes prevention knowledge were the lowest scores. Although mothers’ overall knowledge scores tended to be higher than their daughters, both could benefit from an educational intervention. Dyadic differences and similarities in M-D responses were noted in this study. This is consistent with the previous qualitative findings, whereby AIAN adolescents and adult women did not know about GDM before they were diagnosed, GDM risk-reduction principles, or that they were at higher risk of developing GDM.^19,20 Positive health beliefs that are required for behavior change were also lacking at baseline in this sample, especially among the teens. Most daughters reported no communication with HCPs or their mothers regarding GDM or PC.

Developing culturally relevant, developmentally appropriate PC programs could help decrease these risks. It is imperative to raise awareness of the relationship between body weight and GDM but to do so carefully in a model that is nonblaming/shaming and uplifts, supports, and provides resources to Native youth and their families. Promoting education on the other modifiable risk factors of GDM such as sitting less and being more physically active in AIAN adolescents/young adults and their mothers/caregivers could be a means to increase perceived susceptibility, self-efficacy, and communication⁴⁵ as well as a heightened sense of self-worth and self-esteem. A M-D dyadic team approach for starting PC at puberty in girls at risk for diabetes can be feasible and beneficial.

Implications for education and M-D communication in this vulnerable population is compelling. The goal of Stopping GDM is to provide both AIAN teen girls at risk for GDM and their mothers with PC and knowledge and provide mothers with sex communication training. Mothers can play a vital role discussing RH with their daughters and reinforcing PC.⁴⁹ This research could set new standards of practice for self-management education of AIAN adolescent females and young adults to prevent diabetes.^15,20