Title IX and Its Impact After 40 Years: Understanding Physical Activity Perspectives of Adolescent Girls


Objective: Despite increases in sport participation among girls since the passing of Title IX legislation, girls still tend to have lower physical activity in comparison to boys. The aim of this pilot study was to better understand perspectives of adolescent girls about physical activity. 

Methods: Ten girls aged 13 to 17 years were invited to participate in two, one-hour, structured focus groups using a phenomenological approach. Convenience sampling was used for this pilot study. The girls were queried about the physical activities they do, their enjoyment of that activity, their thoughts about others engaged in physical activity, reasons why girls stop physical activity, and ideas about how girls can be helped to re-engage in physical activity if they ended sport participation in high school. 

Results: Four themes from the focus groups were identified, including Inspiration/Motivation, Comradery, Accomplishment, and Fairness. On a positive note, girls participated in many types of physical activity, both in and out of school, and recognized its benefits from physical, social and psychological perspectives. On a negative note, they spoke at length about school-related discrepancies relative to unequal treatment of boys’ and girls’ sports teams. 

Conclusions: In this group of girls, physical activity was lauded as a healthy and enjoyable behavior, yet displeasure with school preferences for acknowledging and supporting boys’ sports was a stark reminder of the gender gap that still exists in school settings for promoting girls’ exercise activities. 


 It has been 40 years since Title IX legislation was passed in the U.S. One of the issues this Title was designed to end were “barriers in sports for women and girls.”1 Over a decade ago, McCallister, et al. qualitatively queried pre-adolescent children about sports and found that sports participation was viewed as having primarily masculine characteristics and that performing sport “like a girl” was viewed as negative and derogatory.2 In the U.S., boys outnumber girls’ high school sports participation and boys have higher participation in outdoor recreation pursuits than similarly aged girls.3,4 Girls still lag behind boys in school-based physical activity.3,4 

Since recent research points to school teachers as having major influences in girls’ attitudes toward and participation in extracurricular sports, schools are considered an opportune location to promote physical activity for children and adolescents.5,6 While state curricula and school grade level dictate physical education requirements, many school extracurricular opportunities can promote physical activity for girls. As we were working with families impacted by breast cancer and examining the potential impact of physical activity, we wondered whether there has been an increase in positive attitude, perception and practice linked to physical activity in adolescent girls compared to that observed in the past. We conducted a preliminary feasibility study to determine how willing teen girls would be to provide information about their physical activities. 


We used a convenience sample to invite a group of adolescent girls (10 girls aged 13-17 years) to participate in a focus group. At the focus group, we discussed (a) their lived experiences with physical activity and (b) their thoughts, feelings and rationale for participation in physical activities. We used a short interview schedule, of six open-ended questions, that allowed us to guide the participants in the discussion. The conversation lasted 60 minutes and was audio recorded and transcribed. We reviewed the transcript using open coding to identify overarching themes. As we worked through the transcript, we reviewed the information line-by-line to determine what ideas the participants used to describe their experiences with physical activity. The study team discussed the emerging themes in detail to ensure that we had a clear interpretation of the information. The information from these conversations indicated that although they enjoy physical activities in school settings, girls end their participation early due to lack of support and recognition by teachers and school administration. All study procedures were approved by the Institutional Review Board of the University of Utah.


These young women were involved in a number of physical fitness and exercise activities, which included traditional aerobic conditioning exercise such as running, “intense fitness classes,” dance, and physical education class; sports, such as ice hockey, soccer, swimming, track, tennis and water polo; and individual activities such as weight training, hiking and yoga. Based on our conversation with these youth, four overarching themes came to light, i.e., Inspiration/Motivation, Comradery, Accomplishment, and Fairness, which are important for how girls see their participation and value of their efforts in physical activities at school.


The participants were asked what they thought about when they saw other girls participating in physical activities. They noted that “it’s inspiring” when watching peers and/or star athletes. In addition, one participant stated, “You can see how far you have come” when thinking about her own workouts. This inspiration connected the girl’s internal motivation for exercise and developing physical fitness. In terms of positive effect, participants noted different reasons for doing physical activities: 

– “I just do it because it gives me something to do. If I go home right after school, I will just do nothing. Sports kind of give focus because I know I have a limited time to do homework and stuff.” 

– “I do it because I want to stay healthy and fit…just kind of for my own good.” 


Participants noted that having friends and others work out with them enabled them to maintain a level of interest in physical fitness. Having this type of comradery linked to external motivations for completing physical activity and having a companion to work with meant possible accountability and/or fun, and allowed individuals to select an activity for which the impact of friends was key. When asked about what types of things might help them increase their physical activity, the participants noted: 

– “If people encourage them to do it, . . . they will try harder.” 

– “Doing it with their friends.”


The girls were asked how they felt when they moved a lot and/or if they were expected to move a lot in their activities. Their responses related to a sense of accomplishment, both with being involved in physical activity and with putting forth a best effort: 

– “…I know I was there; I was doing my best. I know it is going to pay off and it is worth it to be there. I always feel really good about that.” 

– “I feel accomplished, like . . . I did something besides just sitting there, wasting my time.”


One area that evolved from the conversation was about the lack of community recognition for girls’ activities. Although this was not a topic we specifically intended to query, it was clear that the issue of public recognition for sports participation was of major concern to the girls; they initiated this part of the discussion. The participants were asked why girls stop participating in physical fitness activities, especially in comparison to boys. Their responses were connected to perceived community preference for boys’ activities and realization that boys were more often pushed into sporting activities and recognized for these activities: 

– “I think that sports is not something that is as pressured as much on girls than as boys. I think that sometimes it does not feel as important to girls. So I think it is not as emphasized.” 

– “People talk to guys about sports more. I have seen teachers come into rooms and start asking the guys like how is baseball going, how is basketball coming? But they do not ever ask girls about it.” 

– “At our school, there are so many girls who are so into their sports. It is just no one gives them the recognition. We do not have fan clubs for those [girls’] teams. We have fan clubs for the boys’ teams though.”


The results of this feasibility study provided important insights into the physical activity experiences of teen girls. Our four themes, Inspiration/Motivation, Comradery, Accomplishment, and Fairness, are largely supported by a recent review by Standiford, who categorized themes into somewhat larger categories of Perceptual, Interpersonal and Situational influences for participation.7 The strongest statements from this group of adolescent girls were captured in the construct of Fairness in our study, and articulated as “contending with boys” by Standiford.7 The girls in our study were queried as to why many girls do not persist in sport and exercise participation. The response of the girls was notably strong relative to the role of schools in supporting the boys and affording girls little, if any, recognition. Girls stated that schools allowed students to leave school early or even allowed students to miss school on days when the boys’ football or basketball teams were playing in a championship game: 

– “So if we are in the state championships or anything, we will get out early if it is boys. But if it’s the girls, we do not get out early.” 

– “Our drill team actually went to state and it was not a school excused. Like they encouraged us to go, but it was not school excused. But for the guys’ basketball team, it was a school-excused thing.”

Such actions were taken to ensure that the boys’ teams would have fan support from peers, teachers and other school personnel. Our participants reported that this type of school support was never offered to girls’ teams. Further, school administrators would often announce upcoming boys’ team events during morning announcements, but would rarely provide the same information about girls’ teams. As noted in the statements below, teen girls experienced lack of support due to clear preference for teen boys’ activities: 

– “Like they come over the intercom or the teacher tells you like oh there is a football game tonight. There is also a tennis game tonight, but. . .” 

– “Our school glorifies our football team. They glorify our basketball team. But our girls’ tennis team and our girls’ soccer team like none of the girls’ teams get as much recognition as our basketball team or our football team.” 

– “Swimming will be like region champions and then people will be like, “I didn’t know we had a swim team.” It is like yes we do have a swim team. [laughter] We are there.”

While we suspected that girls would address inequities between girls and boys in the realm of physical activity, exercise and sport, we were surprised at the girls’ level of discontent with schools and their seemingly frank disregard for girls’ roles in sport.

Public Health Implications: 

Existing practices that perpetuate the notion that boys are better at sports and, therefore are favored over girls in sport settings, were described by the girls in the present study and have been reported elsewhere.5 Wetton and colleagues noted that, among a sample of 60 girls aged 15 to 16 years old, perceived lack of ability, negative experiences in physical education classes, and teacher preference for working with skilled students were reasons girls did not participate in team sports.5 The girls in our present study stated that favoritism for boys’ sports was perpetuated by the school environment and, in particular, unsupportive teachers. Participation was considered a highly visible expectation for boys and a low-priority option for girls. Interestingly, one girl in the study described physical activity opportunities as a socially acceptable outlet for her aggression (a sentiment that resonated with the group), especially against boys. 

The role of physical activity in promoting a multitude of health benefits was mentioned by our participants in the present study and by young women from other countries.5,8,9 The participants in this study specifically described the immediate benefits of physical activity such as stress management, sense of accomplishment, and being a healthy person. They also recognized that school credit can be earned through physical education class, which was viewed as positive. In terms of public health, there seems to be an overall lack of physical activity support for young women. This area is one in which public health professionals could conduct more research and develop clear models about why physical activities are important for young women as well as how to create a culture of support for young women’s participation.

Social aspects of physical activity were highly valued by the girls in the current study including friendship, meeting new people and being a team member. The general influence of others is captured by Standiford as Interpersonal Influences and includes parents and teachers in addition to friends; though the present group of girls did not mention teachers as positively influencing their own participation, they were very outspoken about the discrepancy between teachers’ support of boys’ sports versus girls’ sports, suggesting that teachers could have a positive influence on their own participation.7 A recent school-based intervention aimed to improve health behaviors among adolescents, including increased physical activity and decreased sedentary time, was favorable for promoting such changes.10 However, physical activity improvements were most notable among the boys and sedentary time among the girls did not decrease. Thus, a social system of support for girls needs to be developed with a public health lens to promote physical activity options for females in the school system.

This study supports the work of others and brings new information to the effort to increase physical activity among teen girls. Specifically, we found that girls enjoy team sports and recognize the benefits of teamwork; they use physical activity to manage stress and aggression; and they can be discouraged by stereotypical attitudes. Based on a review article about the motivation for participation in sports by Deaner, Balish, and Lombardo, the finding about team sports participation appears to be potentially novel since research has indicated that females’ rationale for sports participation is very different from males’ rationale.11 In fact, one qualitative study, which connects to the work by Deaner, Balish, and Lombardo, noted that physical education teachers’ approaches to increasing girls’ participation in school-based programming were, in fact, largely based upon gender stereotypes, despite recent advancements in physical education curricula.12 Albeit small, this study supports persistent gender stereotypes in school physical education programming and links to the findings from our focus group discussion—that boys’ sports activities are held out as most important illustrating that boys are expected to participate in sports, but girls are not.


In summary, based on the comments made by the girls in the focus group, the school environment is a place that can potentially improve the physical activity participation of adolescent girls. Promoting physical activity can be achieved by having policies that ensure that girls’ sports are promoted and noted at the same level as boys’ sports. For example, if school attendance is waived on days when boys’ sports teams are competing in championship situations, it seems higher level policy makers ought to be part of this support for girls’ participation as well. Schools should ensure that students understand that physical activities through the sports for males and females are on equal footing as “events” that students are expected to support by their attendance. 

Another potential policy change would be to enhance opportunities for girls to participate in the same variety of sports that boys are offered. One example is intramural sports and recreational physical activities that are organized at a variety of skill levels, and another is to not categorize physical education offerings by gender. The proposed variations could help girls increase physical activity while enjoying the social and teamwork benefits of physical activity they like, since these are factors that encourage their participation. Although less is known about the types of variation that might best promote ongoing physical activity for girls, research questions to be considered might include whether same sex physical education classes promote girls participating in physical activity for more years. Answers to these types of questions should be gathered through research to inform revised policies.

Finally, girls in this study identified a variety of benefits associated with participating in physical activity in addition to physical health. These included spending time with friends, feeling a sense of accomplishment, and managing emotions and stress. As schools address the current mental health issues of students, physical activity can be prioritized as a tool to promote mental wellness.

Forty years after the passage of Title IX legislation, it is clear that significant strides still must be made to address the issues related to sex-based discrimination. Schools should play an important role in promoting gender equality by creating a welcoming environment for girls’ participation in organized sports as well as other physical activities. Creating a supportive environment for physical activities and sports for girls in educational settings can promote continued activity into adulthood for women.


1. Title IX (2019). Title IX Info. Available at http://www.titleix.info/.

2. McCallister, S.G., Blinde, E.M., & Phillips, J.M. (2003). Prospects for change in a new millennium: Gender beliefs of young girls in sport and physical activity. Women in Sports and Physical Activity 12(2), 83-96. 

3. Bassett, D.R., John, D., Conger, S.A., Fitzhugh, E.C. & Coe, D.P. (2014). Trends in physical activity and sedentary behaviors of United States youth. Journal of Physical Activity and Health 12(8), 1102-111. 

4. Spencer, R.A., Rehman, L., & Kirk, S.F. (2015). Understanding gender norms, nutrition, and physical activity in adolescent girls: A scoping review. International Journal of Behavior, Nutrition, and Physical Activity 12, 6. 

5. Wetton, A.R., Radely, R., Jones, A.R., & Pearce, M.S. (2013). What are the barriers which discourse 15-16 year-old girls from participating in team sports and how can we overcome them? Biomedical Research International 738705. doi:10.1155/2013/738705. 

6. Carson, R.L., Castelli, D.M., Beighle, A. & Erwin, H. (2014). School-based physical activity promotion: A conceptual framework for research and practice. Childhood Obesity 10(2), 100-106. 

7. Standiford, A. (2013). The secret struggle of the active girl: A qualitative synthesis of interpersonal factors that influence physical activity in adolescent girls. Health Care for Women International 34(10), 860-877. 

8. Peykari, N., Eftekhari, M.B., Tenrani, F.R., et al., (2015). Promoting physical activity participation among adolescents: The barriers and the suggestions. International Journal of Preventive Medicine 6, 12. doi:10.4103/2008-7802.151820

9. Sedibe, H.M., Kahn, K., Edin, K., Gitau, T., Ivarrson, A., & Norris, S.A. (2014). Qualitative study exploring healthy eating practice and physical activity among adolescent girls in South Africa. BMC Pediatrics 14, 211.

10. Sevil, J., García-González, L., Abós, Á., Generelo, E., & Aibar, A. (2018). Can high schools be an effective setting to promote healthy lifestyles? Effects of a multiple behavior change intervention in adolescents. Journal of Adolescent Health 21, S105-139. doi: 10.1016/j.jadohealth.2018.09.027

11. Deaner, R.O., Balish, S.M., & Lombardo, M.P. (2016). Sex differences in sports interest and motivation: An evolutionary perspective. Evolutionary Behavioral Sciences 10(2), 73-97. http://dx.doi.org/10.1037/ebs0000049

12. Murphy, B., Dionigi, R. A. & Litchfield, C. (2014). Physical education and female participation: A case study of teachers’ perspectives and strategies. Issues in Educational Research 24(3), 241-259. 


Frost CJ, Shaw J, O’Toole K, Metos J, Brusseau Jr. T, Moric E, Gren LH. (2020). Title IX and Its Impact After 40 Years: Understanding Physical Activity
Perspectives of Adolescent Girls. Utah Women’s Health Review. doi: 10.26054/0DGHPHTNEG.


View / download

The Baby-Friendly Hospital Initiative (BFHI): An Early Cross-Sectional Analysis of PRAMS Phase 8 Data on Hospital Practices and Breastfeeding Outcomes in Utah and Wyoming


Introduction: Breastfeeding has immediate and long-term benefits for both maternal and child health.  This study examines the association between Baby-Friendly Hospital Initiative (BFHI) experiences and breastfeeding outcomes in the Mountain West region.

Methods: A cross-sectional (retrospective secondary data analysis) was performed using the 2016 Pregnancy Risk Assessment Monitoring System (PRAMS) data. The participants were derived from a stratified random sample of 2,013 women living in Utah and Wyoming who recently had a live birth and who were surveyed on BFHI practices. The association between BFHI experiences and breastfeeding duration were assessed using crude and adjusted Poisson regression models, controlling for other BHFI experiences and maternal age, pre-pregnancy BMI, household income, smoking, alcohol, delivery method, and number of days spent in the hospital post delivery

Results: 82.4% and 82.3% of women from Utah and Wyoming, respectively, reported breastfeeding for 2 months or longer. After controlling for other BFHI experiences and potential confounders, the one shared BFHI experience that was associated with breastfeeding for 2 months or longer vs less than 2 months was starting breastfeeding in the hospital (adjusted prevalence ratio [aPR]=1.49, 95% CI (1.12, 1.98) in Utah and aPR=2.03, 95% CI (1.13, 3.64) in Wyoming. Among women in Utah and Wyoming, only 5 of 7 BFHI steps were significant for breastfeeding duration in at least one state.

Conclusion: There is substantial epidemiological support for health benefits to both mother and infant for exclusive breastfeeding to 6 months and prolonged breastfeeding until at least 1-year. Our findings suggest that women who initiate breastfeeding in the hospital may be more likely to breastfeed for a longer duration.


Breastmilk is the most nutritious food for infant development and studies show that breastfeeding promotes optimal health outcomes for the mother/infant dyad that have lifelong implications. Additionally, the American Academy of Pediatrics promotes sustained breastfeeding practices for at least the first year of life.1–3 Although the U.S. Centers for Disease Control and Prevention (CDC) has reported steady increases in breastfeeding practices in the US, attributed in part to support of the medical community and the Baby-Friendly Hospital Initiative (BFHI),4–7 the prevalence of exclusive breastfeeding is below the Healthy People 2020 targets.2

The Baby-Friendly Hospital Initiative (BFHI) is a joint effort started by UNICEF and the WHO to protect, promote, and support breastfeeding practices.8,9 More specifically, the BFHI seeks to increase positive in-hospital experiences such as initiating breastfeeding within 1 hour of birth, feeding on demand, and breastfeeding education and support, while limiting experiences that hinder early initiation and duration of breastfeeding, such as giving pacifiers or giving gift baskets that include formula.10 Although studies have demonstrated BFHI’s success in improving breastfeeding beyond six weeks, only 28% of U.S. annual births take place in Baby-Friendly certified hospitals. Additionally, implementation strategies vary across hospitals, and there is still widespread use of supplementation and pacifiers.11–14

Furthermore, BFHI implementation in rural hospitals is limited; thus studies that explore BFHI experiences and breastfeeding outcomes in rural regions are needed.4,15 Close to 80% of the population in both Utah and Wyoming live in rural areas,16 and both states have birth rates above the national average, with Utah recording the highest birthrate in 2017.17 In addition, out of the 591 Baby-Friendly facilities in the nation, there is only 1 located in each of these states.18 No previous studies that have included an analysis of BFHI experiences between Utah and Wyoming because prior to 2016, the Wyoming PRAMS did not include any questions about BFHI experiences.

Therefore, the purpose of this study was to describe the current state of Baby-Friendly practices in Utah and Wyoming and to assess whether BFHI experiences impact breastfeeding termination and duration among new mothers. The study was approved by the University Of Utah Institutional Review Board (IRB) and determined exempt.

Materials and Methods

Study Design & Population

This study analyzed cross-sectional population-level data for mothers who had recently given birth to a live infant in Utah or Wyoming in 2016 and who completed the CDC Pregnancy Risk Assessment Monitoring System (PRAMS) Phase 8 questionnaire. The CDC developed PRAMS in 1987 as an ongoing, nationwide surveillance system that is state-specific in its sampling scheme and operated within local health departments to collect data related to behaviors and experiences of mothers pre-pregnancy, during the prenatal period, and in the immediate post-natal period.19,20,21 The PRAMS initiative aims to promote safe motherhood, as well as reduce low birth weight and infant mortality.22 The PRAMS questionnaire collects information on an array of topics such as maternal knowledge, attitudes and behaviors about pregnancy, breastfeeding, infant health, physical abuse, stress and social support, maternal use of alcohol and tobacco products, and contraception, among others.23 The recruitment process involves the random selection of potential participants from a sample of birth certificates indicating a recent live birth between two and six months post-partum.24

Utah oversamples mothers with low education and infants with low birthweight while Wyoming oversamples by maternal race and infant birthweight to ensure that the data is representative of the smaller, higher risk populations.25,26 Using birth certificates, new mothers are randomly selected, within stratified sampling scheme, to participate in completing the PRAMS survey.20 Utah and Wyoming select approximately 200 and 140 women each month, respectively, who delivered live births and are at two to six months post-partum.20,27 Selected mothers receive an introductory letter by mail, followed by a survey that is mailed a week after the introductory letter is sent, followed by third and fourth survey attempts mailed to non-respondents.20 Next, an interviewer contacts those non-respondents who received the mailed survey.20  The surveys and phone interviews are available and may be administered in English and Spanish to accommodate Spanish-speaking mothers when necessary.28 Mothers who recorded “Hispanic” on birth certificate information received surveys in English and Spanish.29 The expected response rate in Utah and Wyoming is 60% – 65% following the CDC protocol.20,28 The actual response rate for UT in 2016 was 65% and 63% for WY. Once the surveys are received by the local health department, responses are grouped to document the self-reported prevalence data.28

Data Sources/Measurement: Breastfeeding Initiation/Duration Measures

For this analysis, breastfeeding (BF) termination and duration measures were informed via the Utah Phase 8 (2016) and Wyoming Phase 8 (2016) surveys, which included the following questions: “Did you ever breastfeed or pump breast milk to feed your new baby, even for a short period of time?” Respondents with a “yes” answer were then asked, “Are you currently breastfeeding or feeding pumped milk to your new baby?” If response was “no”, the respondents were asked, “How many weeks or months did you breastfeed or pump milk to feed your baby?”

Data Sources/Measurement: In-Hospital Newborn Care Enhancement Measures

Mothers who reported “yes” to breastfeeding their newborn or giving them pumped breast milk, regardless of the duration, were asked to respond to the following questions with a “yes” or “no” answer about their BFHI experiences: 1) “Hospital staff gave me information about breastfeeding”; 2)“My baby stayed in the same room with me at the hospital”; 3) “I breastfed my baby in the hospital”; 4) “Hospital staff helped me learn how to breastfeed”; 5) “I breastfed in the first hour after my baby was born?”; 6) “My baby was placed in skin-to-skin contact within the first hour of life?”7) “My baby was fed only breastmilk at the hospital”; 8) Hospital staff told me to breastfeed whenever my baby wanted”; 9) “The hospital gave me a breast pump to use”; 10) “The hospital gave me a gift pack with formula”; 11) “The hospital gave me a telephone number to call for help with breastfeeding”; and 12) “Hospital staff gave my baby a pacifier”. The prevalence of BFHI experiences for the study population are found in Table 2 and in Figure 2.


In this analysis, the Phase 8 Utah-PRAMS (2016) and Phase 8 Wyoming-PRAMS (2016) were examined to understand if there is an association between in-hospital newborn care enhancement measures as well as early initiation and continuation of BF after delivery.  Key demographic, behavioral and experiential factors were identified as potential confounders through a thorough literature review.6,30–33 Final decisions on potential confounding factors to include were informed by confirming that the factor is associated with one of the BFHI experiences and with the outcome variables of interest (BF termination or duration), that the factor is unequally distributed within the study population, and that the factor is not an intermediary step in the causal pathway from BFHI experiences and the outcome variable.34,35 The covariates selected for this analysis included maternal age (<20, 20-24, 25-34, 35+), maternal body mass index (BMI) (WHO categories: underweight, normal weight, overweight, obese), household (HH) income (≤$28,000, $28,001-$57,000, $57,001-$85,000, over $85,000), smoked in previous 2 years (no/yes), drank alcohol in previous 2 years (no/yes), delivery method (vaginal or C-section), and the number of days the baby stayed in hospital post-delivery (<1 day, 1–2 days, 3–5 days, 6–14 days, >14 days, or still in). Table 1 delineates these population characteristics by BF duration status for the Utah & Wyoming PRAMS 2016 analysis.

Study Size, Methods, and Statistical Analysis

The total number of participants between both states included 2,013 women who completed the state-specific PRAMS survey. An overview of the sample selection process is illustrated in Figure 1. In Utah, the total number of recorded live births that occurred in 2016 was 50,486 and of those 1,400 women completed the Utah-PRAMS.36 The Utah analysis excluded 43 women (3%) who did not respond to the ever breastfed question, and 28 women (2%) whose delivery did not occur in the hospital within the state of Utah in 2016. In Wyoming, the total number of recorded live births that occurred in 2016 is 7,384 and of those, 613 women completed the Wyoming PRAMS survey.37 The Wyoming analysis excluded 28 women (4.6%) who did not respond to the ever breastfed question, and 12 women (2%) whose delivery did not occur in the hospital within the state of Wyoming in 2016. The weighted response rate was 63% in Wyoming.26 After exclusions, a total of 1,901 women (n=1328 from Utah and n=573 from Wyoming) were included in the analysis.

Descriptive population characteristics were used to compare mothers from Utah and Wyoming according to BF initiation, termination or duration. Prevalence ratios (PR) were calculated with 95% confidence intervals (CI) to evaluate the relationship between BF termination or duration and mother’s BFHI experiences, using unadjusted and adjusted Poisson regression. Analyses were completed using SAS version 9.4 (SAS University ed.) and STATA 15.1 (Stata Corp, LLC). Survey data were weighted according to PRAMS methodology such that the sample was representative of all mothers who delivered during 2016 in Utah and Wyoming.


Participants from Utah were 15–45 years old, while participants from Wyoming were 15–43. The average age of participants from both states was 28 years old. Participants from both states completed the survey on an average at 16 weeks postpartum. The interquartile range for Utah was 13–20 weeks with a range of 10–32 weeks, while in Wyoming the interquartile range was 13–18 weeks with a range of 10–37 weeks. The percentage of survey respondents who reported BF initiation was 93.4% in Utah and 90.5% in Wyoming. 69.5% of mothers in Utah and 68% in Wyoming reported they were still BF at the time of survey completion. Exclusive BF was reported by 59.9% in Utah and 70.4% in Wyoming. The average number of weeks mothers breastfed was 12.8 weeks in Utah versus. 12.3 weeks in Wyoming.

         Figure 1 outlines the prevalence of each of the BFHI in each state, with “Baby breastfed in the hospital” having the highest prevalence in each state (95.0% in Utah and 95.1% in Wyoming) and staff giving breast pump for breastfeeding having the lowest prevalence (35.5% in Utah and 24.0% in Wyoming).

Figure 1: Flow diagram explaining the final cohort of women in the analysis

17.7% of participants from Wyoming reported BF <2 months and 82.3% breastfed for 2 months or longer compared vs. 17.6% <2 months and 82.4% in Utah. In both states, women who breastfed for 2 months or longer versus <2 months tended to be older, of normal weight, higher income, non-alcohol consumers, non-smokers, having a vaginal delivery, and fewer days in the hospital (Table 1).

Table 1: Population characteristics by breastfeeding duration

Unadjusted analysis of each BFHI experience and the association on breastfeeding duration for each state was assessed. In Utah, findings indicated that 8 of 12 the experiences—feeding baby in hospital, giving gift with formula, giving breast pump, engaging in skin to skin in the first hour, BF in the first hour, feeding only breastmilk, baby staying in the room, and giving pacifier by staff — were associated either positively or negatively with BF duration (PR=1.59, 95% CI (1.23, 2.04), PR=0.88, 95% (CI 0.83, 0.93), PR=0.89, 95% CI (0.83, 0.96), PR=1.15, 95% CI (1.04, 1.27), PR=1.2, 95% (CI 1.09, 1.33), PR=1.20, 95% CI (1.12, 1.29), PR=1.15, 95% CI (1.03, 1.29), and PR=0.87, 95% (CI 0.82, 0.92) respectively) (Table 2). In Wyoming, findings indicated that 4 of 12 the experiences—feeding baby in hospital, staff giving breastfeeding help telephone number, feeding only breastmilk in the hospital, and giving pacifier by staff — were associated either positively or negatively with BF duration (PR=1.69, 95% CI (1.08, 2.65), PR=1.23 95% (CI 1.06, 1.43), PR=1.22, 95% CI (1.07, 1.39), and PR=0.88, 95% (CI 0.80, 0.97 respectively).

Table 2: Breastfeeding duration
Figure 2: Prevalence of Baby-Friendly Hospital Initiative Steps: Utah and Wyoming PRAMS

After adjusting for other BFHI experiences and confounding factors (maternal age, maternal BMI, HH income, alcohol use, smokers, delivery method and hospital length of stay), the only BFHI experience significant for BF duration (≥ 2 months versus less) for both states was starting breastfeeding in the hospital: adjusted prevalence ration [aPR] = 1.49, 95% CI (1.12, 1.98) in Utah and aPR=2.03, 95% CI (1.13, 3.64) in Wyoming.  In Wyoming only, staff giving BF help telephone number or exclusive feeding of breastmilk in the hospital were significant predictors of longer BF duration, aPR=1.18, 95% CI (1.01, 1.39) and aPR=1.16, 95% CI (1.00, 1.34), respectively, while staff giving breastfeeding information or having baby stay in the hospital room with mother were associated with shorter BF duration, aPR=0.78 (95% CI: 0.64, 0.96) and aPR=0.76 (95% CI: 0.65, 0.90), respectively. Conversely, in Utah, staff who gave a gift that included formula were more likely to report early BF termination (<2 months), aPR=0.93, 95% CI (0.87,0.99)], this was not the finding for Wyoming.


This population-based study provides a representative, preliminary description of the current state of breastfeeding for mothers who delivered a live birth in Utah and Wyoming during the first year (2016) Phase 8 PRAMS.  It offers a first ever look at how BFHI experiences impact breastfeeding termination and duration in these rural mountain west states. We found approximately equal prevalence of breastfeeding duration in both states, with approximately 82% of postpartum women reporting breastfeeding for 2 months or more. Additionally, we found that, in relation to BF duration, only women who started breastfeeding in the hospital, had increased likelihood of BF ≥2 months, with Utah having a 49% increase and Wyoming having a 103% increase after adjusting for other BHFI experiences and confounding factors. Wyoming, however, also showed that staff giving breastfeeding help telephone number, and those who fed only breastmilk in the hospital also significantly increased the likelihood of breastfeeding 2 or more months (18% and 16%), respectively. Interestingly those with increased risk of early breastfeeding termination (< 2 months) were those given breastfeeding information by staff and those who had their babies room-in with them 24/7 (22% and 24%), respectively. Those to whom staff gave a pacifier, did not show significant association with breastfeeding duration in either state; however, these were significant risks for early termination for the first multivariate analysis model that controlled only for other BFHI experiences and not the other confounders included in the fully adjusted model. In Utah, those who were given formula were 7% more likely to terminate breastfeeding before 2 months.

Our results regarding starting breastfeeding in the hospital, giving only breast milk, and providing help telephone numbers are consistent with other research.11 Our results showed that giving pacifiers did not significantly impact breastfeeding duration. Although this finding may be counter-intuitive, it is supported by large RCTs that also showed no impact.39 Our findings that showed negative impacts from rooming-in and provision of staff help are contradictory to other findings,12 and may be due to reverse causation (i.e., women who require infant to be in the room with them or who need help from staff breastfeeding may be women who are having greater difficulties breastfeeding and thus it is not the BHFI but rather the difficulty breastfeeding that drives the association).

The limitations present in our study should be considered when interpreting our findings. Reporting biases are likely because data were not available for race/ethnicity or pre-term delivery, both of which are known to impact breastfeeding initiation and duration.31 Similarly, there was no information on the hospitals where the infants were born, and subsequently, no information on the status of the hospital’s Baby-Friendly designation. Possible recall bias may exist in that women who breastfed longer may differ significantly in their recollection of BFHI experiences than those who did not. Additionally, the impact of parity as a potential confounder was not addressed in our analysis. Further, generalizability is limited as the study focuses exclusively on the Mountain West region. However, for states with limited access to BFHI designated hospitals, our findings may be more relevant.

Despite these limitations, there are several strengths of the study.  First, we utilized weighted data to represent all mothers who gave birth from 2016 in Utah and Wyoming. The sample size of the study was also reflective of these populations with weights to ensure the inclusivity of at-risk women. Furthermore, this is the first time Wyoming has included the BFHI experiences question in their PRAMS survey. Thus this study has the unique strength of being the first to compare these two very similar populations.


In conclusion, our results demonstrate the importance of initiating and exclusively breastfeeding in the hospital as well as providing help telephone numbers for women about breastfeeding prior to discharge. More specifically, our findings indicate that small rural hospitals may be able to improve breastfeeding duration by implementing these specific BFHI recommendations. Additionally, our results suggest that giving a gift pack with formula in the hospital is associated with stopping breastfeeding before two months for both states, but providing a pacifier is not associated with breastfeeding duration. There is strong epidemiological support for the health benefits to both mother and infant for exclusive breastfeeding to 6 months and prolonged breastfeeding until at least 1-year, and along with the U.S. Preventive Services Task Force (USPSTF), the WHO and UNICEF we also strongly promote and encourage this practice.1–3,38


Data were provided by the Wyoming and Utah Pregnancy Risk Assessment Monitoring Systems (PRAMS) directed by the Utah and Wyoming Departments of Health in coordination with the US Health and Human Services Department and Centers for Disease Control and Prevention.  This study is not representative of the official views of the entities listed.

Sources of Funding:

Dr. Rogers received supportive funds from the National Cancer Institute of the National Institutes of Health (NIH) [grant number K01CA234319]. This report does not represent the official views of the CDC, Utah Department of Health, or the NIH.


1. Victora CG, Bahl R, Barros AJDD, et al. Breastfeeding in the 21st century: Epidemiology, mechanisms, and lifelong effect. Lancet. 2016;387(10017):475-490. doi:10.1016/S0140-6736(15)01024-7

2. Dieterich CM, Felice JP, O’Sullivan E, Rasmussen KM. Breastfeeding and Health Outcomes for the Mother-Infant Dyad. Pediatr Clin North Am. 2013;60(1):31-48. doi:10.1016/j.pcl.2012.09.010

3. Eideleman AI. Breastfeeding and the Use of Human Milk. Pediatrics. 2012;129(3):e827-e841. doi:10.1542/peds.2011-3552

4. Munn AC, Newman SD, Mueller M, Phillips SM, Taylor SN. The Impact in the United States of the Baby-Friendly Hospital Initiative on Early Infant Health and Breastfeeding Outcomes. Breastfeed Med. 2016;11(5):222-230. doi:10.1089/bfm.2015.0135

5. Myers D, Turner-Maffei C. Improved Breastfeeding Success Through the Baby-Friendly Hospital Initiative. Am Fam Physician. 2008;78(2):180-182.

6. U.S. Centers for Disease Control and Prevention (CDC). Breastfeeding Report Card, United States 2018.; 2018.

7. Sinha B, Chowdhury R, Sankar MJ, et al. Interventions to improve breastfeeding outcomes: A systematic review and meta-analysis. Acta Paediatr Int J Paediatr. 2015;104:114-135. doi:10.1111/apa.13127

8. UNICEF. Baby-Friendly Hospital Initiative Ten Steps to Successful Breastfeeding.

9. World Health Organization (WHO). Implementation guidance: protecting, promoting and supporting breastfeeding in facilities providing maternity and newborn services – the revised Baby-friendly Hospital Initiative. 2018:56.

10. Baby-Friendly USA Inc. Baby-Friendly USA ~ 10 Steps & International Code. https://www.babyfriendlyusa.org/for-facilities/practice-guidelines/10-steps-and-international-code/. Accessed October 19, 2019.

11. Digirolamo AM, Grummer-Strawn LM, Fein SB, University E. Effect of Maternity-Care Practices on Breastfeeding. Pediatrics. 2008;122(Supplement 2):S43-S49. doi:10.1542/peds.2008-1315e

12. Nickel NC, Labbok MH, Hudgens MG, Daniels JL. The extent that noncompliance with the ten steps to successful breastfeeding influences breastfeeding duration. J Hum Lact. 2013;29(1):59-70. doi:10.1177/0890334412464695

13. Baby-Friendly USA Inc. Baby-Friendly USA ~ Upholding the Highest Standards of Infant Feeding Care. https://www.babyfriendlyusa.org/. Accessed October 24, 2019.

14. Declercq E, Labbok MH, Sakala C, O’Hara MA. Hospital practices and women’s likelihood of fulfilling their intention to exclusively breastfeed. Am J Public Health. 2009;99(5):929-935. doi:10.2105/AJPH.2008.135236

15. Lillehoj CJ, Dobson BL. Implementation of the Baby-Friendly Hospital Initiative Steps in Iowa Hospitals. JOGNN – J Obstet Gynecol Neonatal Nurs. 2012;41(6):717-727. doi:10.1111/j.1552-6909.2012.01411.x

16.  United States Census Bureau. Rural America: Geographical Areas and Rural Data. https://gis-portal.data.census.gov/arcgis/apps/MapSeries/index.html?appid=7a41374f6b03456e9d138cb014711e01. Accessed June 16, 2019.

17.  Martin JA, Hamilton BE, Osterman MJK, Driscoll AK, Drake P. Births: Final for 2017. Natl Vital Stat Reports. 2018;67(8):1-49.

18. Baby-Friendly USA Inc. Baby-Friendly Facilities A-Z and by State. https://www.babyfriendlyusa.org/for-parents/baby-friendly-facilities-by-state/. Accessed June 14, 2019.

19. Shulman HB, D’Angelo D V, Harrison L, et al. The Pregnancy Risk Assessment Monitoring System (PRAMS): Overview of design and methodology. Am J Public Health. 2018;108(10):1305-1313. doi:10.2105/AJPH.2018.304563

20. Utah Department of Health. Maternal and Infant Health Program. https://mihp.utah.gov/pregnancy-and-risk-assessment. Accessed June 17, 2019.

21. U.S. Centers for Disease Control and Prevention (CDC). PRAMS Methodology. PRAMS. https://www.cdc.gov/prams/methodology.htm. Accessed June 17, 2019.

22. U.S. Centers for Disease Control and Prevention (CDC). PRAMS. https://www.cdc.gov/prams/index.htm. Published 2018.

23. U.S. Centers for Disease Control and Prevention (CDC). PRAMS Questionnaires. https://www.cdc.gov/prams/questionnaire.htm. Published 2018. Accessed January 8, 2019.

24. Shulman HB, D’Angelo D V., Harrison L, Smith RA, Warner L. The Pregnancy Risk Assessment Monitoring System (PRAMS): Overview of Design and Methodology. Am J Public Health. 2018;108(10):1305-1313. doi:10.2105/AJPH.2018.304563

25. UDOH Maternal & Infant Health Program. Utah PRAMS. Maternal & Infant Health Program.

26. Wyoming Department of Health. Wyoming Pregnancy Risk Assessment Monitoring System (PRAMS) 2017 Surveillance Report.; 2018.

27. Wyoming Department of Health. Hospital-Based Breastfeeding Practices.; 2018.

28. Wyoming Department of Health. Pregnancy Risk Assessment Monitoring System (PRAMS). https://health.wyo.gov/publichealth/chronic-disease-and-maternal-child-health-epidemiology-unit/mch-epi/pregnancy-risk-assessment-monitoring-system-prams/. Accessed May 30, 2019.

29.  Schliep KC, Denhalter D, Gren LH, Panushka KA, Singh TP, Varner MW. Factors in the Hospital Experience Associated with Postpartum Breastfeeding Success. Breastfeed Med. 2019;14(5):334-341. doi:10.1089/bfm.2018.0039

30. Kair LR, Colaizy TT. Association Between In-Hospital Pacifier Use and Breastfeeding Continuation and Exclusivity: Neonatal Intensive Care Unit Admission as a Possible Effect Modifier. Breastfeed Med. 2017;12(1):12-19. doi:10.1089/bfm.2016.0137

31. Pounds L, Shostrom V. Analyzing Factors That Impact Breastfeeding Duration in the Postpartum Period: A Secondary Analysis of PRAMS Data. Breastfeed Med. 2018;13(5):335-340. doi:10.1089/bfm.2018.0020

32. Sipsma HL, Jones K, Nickel NC. Hospital practices to promote breastfeeding: The effect of maternal age. Birth. 2017;44(3):272-280. doi:10.1111/birt.12284

33. Wallwiener S, Müller M, Doster A, et al. Predictors of impaired breastfeeding initiation and maintenance in a diverse sample: what is important? Arch Gynecol Obstet. 2016. doi:10.1007/s00404-015-3994-5

34. LaMorte WW, Sullivan L. Confounding and Effect Measure Modification. http://sphweb.bumc.bu.edu/otlt/MPH-Modules/BS/BS704-EP713_Confounding-EM/BS704-EP713_Confounding-EM_print.html. Accessed September 24, 2019.

35. Aragon TJ. Population Heath Thinking with Bayesian Networks.; 2019. doi:10.1111/ina.12046

36. Office of Vital Records and Statitics. Births and Deaths : Year Ended … Salt Lake City; 2017.

37. Beaudoin G, Storey M. Vital Statistics Services 2016 Annual Report W. S. § 35-1-404 ( a )( V ). Vol 404.; 2017.

38. Hawkins SS, Stern AD, Baum CF, Gillman MW. Evaluating the impact of the Baby-Friendly Hospital Initiative on breast-feeding rates: A multi-state analysis. Public Health Nutr. 2015;18(2):189-197. doi:10.1017/S1368980014000238

39. O’Connor NR, Tanabe KO, Siadaty MS, Hauck FR. Pacifiers and breastfeeding: A systematic review. Arch Pediatr Adolesc Med. 2009;163(4):378-382. doi:10.1111/j.1523-536x.2009.00336_1.x


Bliss J, Mensah NA, Rogers CR, Stanford JB, VanDerslice J, Schliep KC. (2020). The Baby-Friendly Hospital Initiative (BFHI): An early cross-sectional analysis of PRAMS Phase 8 data on hospital practices and breastfeeding outcomes in Utah and Wyoming. Utah Women’s Health Review. doi: 10.26054/0KMTC25CW0


View / download

The Association Between Preconception Body Mass Index and Subfertility Among Hispanic and non-Hispanic Women: A Cross-Sectional Study from Utah’s Pregnancy Risk Assessment Monitoring System Survey (2012–2015)


Objective: To investigate the association between pre-pregnancy body mass index (BMI) and subfertility within a population-based cohort, exploring Hispanic ethnicity as a potential effect modifier.

Methods: We used cross-sectional study data from the Utah Pregnancy Risk Assessment Monitoring System from 2012–2015. Relationships between maternal pre-pregnancy BMI and subfertility were evaluated via Poisson regression models with robust error variance, accounting for the stratified survey sampling. Preconception BMI was analyzed continuously and categorically. Women’s subfertility was defined via self-report in two ways: 1) time trying to achieve pregnancy; and 2) report of using fertility-related drugs/medical procedures.

Results: The median age was 27.0; 18.8% were obese, and 15.9% were Hispanic. Women with preconception obesity (BMI>30kg/m2), compared to normal weight women (18.4kg/ m2<BMI<25kg/m2) had a 1.85 (95% CI 1.43, 2.38) higher adjusted prevalence ratio (aPR) for having subfertility defined by time trying and a 1.73 (95% CI 1.20, 2.32) higher aPR for receiving fertility-enhancing drugs/medical procedures. Continuous models indicated a linear relationship between BMI and subfertility (aPR 1.04, 95% CI 1.03, 1.06 for time trying; and 1.06, 95% CI 1.03, 1.10 for receiving fertility-enhancing drugs/medical procedures).

Conclusions: Obese women, but not underweight or overweight women, reported higher subfertility than normal-weight women. Findings among this cohort of at-risk new mothers, oversampled on low education and birth weight and comprised of higher than the national average of Hispanics, indicated a dose-response relationship between obesity and subfertility.

Implications: Our findings highlight the importance of population-oriented obesity prevention for at-risk women with intentions to conceive.


Body mass index (BMI) in the U.S. has continued to rise over the last two decades, women of reproductive age included. The 2015–2016 National Health and Nutrition Examination Survey estimated that among women ages 20 to 39 years, 36.5% were obese (BMI >30 kg/m2).1 The association between women’s obesity and subfertility has been established,2 however, existing studies focus on the link between maternal pre-pregnancy BMI and pregnancy outcomes among non-Hispanic white women undergoing fertility treatment.2 Prior studies have been conducted among Asians and African Americans,3,4 but few among Hispanic women.5,6 As the proportion of women with obesity in the U.S. continues to rise, evaluating how BMI and obesity directly impact women of various race/ethnicities is critical to address health-related disparities among under-represented women.

The Hispanic population is of interest for several reasons. Although overall fertility rates in the U.S. decreased from 2007–2017, Hispanics had the largest decline compared to non-Hispanic whites and African Americans.7 Furthermore, maternal pre-pregnancy BMI distribution by race indicates that Hispanics have the largest percent of overweight mothers (29.7%) compared to non-Hispanic white (24.1%), African-American (26.9%), American Indian/Alaskan Native (27.2%), or Asian (19.9%) mothers.1 Lastly, although Hispanic women make up 12.5% of the U.S., Hispanics only use 5.4% of the nation’s infertility care/resources—potentially resulting from the disparities in access to care.8 A better understanding of the factors that contribute to this health disparity are needed, warranting research that includes women of Hispanic ethnicity and explores unique attributes that may influence access to infertility care.

Taking into account potential effect modification by Hispanic ethnicity, our study aimed to investigate the association between pre-pregnancy BMI and women’s subfertility within a population-based cohort of Utah women, comprised of roughly 16% Hispanics.


Study Design

This is a cross-sectional study using data from the Utah Pregnancy Risk Assessment Monitoring System (PRAMS) survey, which has the standardized data collection methodology developed by the Centers for Disease Control and Prevention (CDC).

Data Sources

Started by the CDC in 1987, the current study’s population stems from the PRAMS nationwide surveillance project, which has the two-fold purpose of decreasing the morbidity and mortality of mothers and infants, and improving their health by reducing adverse outcomes.9 PRAMS is a population-based and state-specific study of women who delivered a live birth, accompanied by their maternal attitudes, behaviors, and experiences before, during, and shortly after pregnancy. The health topics covered in PRAMS are related to the following: prenatal care, attitudes and feelings about previous pregnancy, health insurance coverage, cigarette smoking, drinking, physical abuse, maternal stress, economic status, and infant health status. One key aspect of PRAMS is the stratified systematic sampling, which oversamples on features related to high risk women (e.g., mothers of low-birth-weight infants, those living in high-risk geographic areas, and racial/ethnic minority groups). The design and methodology of PRAMS have been published elsewhere.10

For the current study, the authors used data from the Utah PRAMS Phase 7 (2012–2015) questionnaire (n = 5,770 reflecting an estimated population of 199,905 women [number of women in the population that each respondent represents]). PRAMS Phase 7 Utah sampling was stratified by maternal education and infant birthweight. The design and sampling frame of PRAMS assure a study sample that is representative of Utah’s population–including a 16% prevalence of Hispanic ethnicity.10

Approximately 200 new mothers are randomly selected from Utah birth certificate data each month to participate in PRAMS. New mothers are contacted via mailed questionnaire (available in English and Spanish) multiple times and telephone follow-up. Response rates in Utah were roughly 72% in 2012, 66% in 2013, 69% in 2014, and 67% in 2015, higher than the 60% response rate that the CDC expects.9 Participating womens’ responses are linked to extracted birth certificate data items for analysis. The availability of birth certificate information for all births is the basis for drawing stratified samples and, ultimately, for generalizing results to the state’s entire population of births.9 The PRAMs weighting process produces an analysis weight taking into account the stratified sampling along with nonresponse and noncoverage components.9 The analysis weight of the PRAMs data can be interpreted as the number of women like herself in the population that each respondent represents.9

The study was evaluated by the University of Utah Institutional Review Board (IRB) and determined exempt. 

Outcome Measures

The primary outcome of interest was women’s subfertility which was assessed in two ways. First, subfertility was defined based on self-reported time trying to achieve pregnancy: >12 months for women ≤35 years of age, and >6 months for women > 35 years.11 In the Utah PRAMS Phase 7 (2012–2015) questionnaire, time trying was assessed by two questions: “When you got pregnant with your new baby, were you trying to get pregnant?” and if women answered “yes” they were then asked “How many months were you trying to get pregnant?” with potential responses of 0–3 months, 4–6 months, 7–12 months, 13–24 months, or >24 months. Second, we defined subfertility based on self-reported fertility treatment. If women answered yes they were trying to get pregnant with their new baby, then they were also asked “Did you take any fertility drugs or receive any medical procedures from a doctor, nurse, or other health care worker to help you get pregnant with your new baby?” with potential responses of fertility-enhancing drugs prescribed by a doctor, artificial insemination, assisted reproductive technology, other medical treatment, or “I wasn’t using fertility treatments during the month that I got pregnant with my new baby.”

Exposure Measures

Pre-pregnancy BMI was calculated using birth certificate reported height and weight, and categorized in standard groups for underweight (<18.5 kg/m2); normal (18.5-24.9 kg/m2); overweight (25.0-29.9 kg/m2); and obese (30 kg/m2 or higher).5 Height and weight data were also available via self-report from the PRAMS questionnaire. Given the high correlation of BMI data from both sources (Pearson correlation coefficient=93.5%), missing birth certificate BMI values (n=71) were replaced with PRAMS BMI data.


Confounding factors thought to impact both women’s preconception BMI and subfertility were determined based on prior literature. Demographic and health factors included race/ethnicity, maternal education, marital status, family income, health insurance, prior pregnancy/live birth, and preconception maternal age, smoking, drinking, diabetes, hypertension, and depression were all considered potential confounding factors.2-6

Statistical Analysis

Participant characteristics were reported across BMI categories while taking into account the complex survey design of PRAMS.9 The continuous variables (e.g. maternal age, BMI) were reported by median and interquartile range (IQR), and the categorical variables were reported by frequencies and percentages. To evaluate associations between preconception BMI and women’s subfertility, modified Poisson regression models were employed with robust error variance, accounting for the stratified sampling, to estimate prevalence ratios (PR) and 95% confidence intervals (CI), with normal weight as the referent group.12 Additionally, adjusted multinomial logistic regression was used to analyze the association between BMI and multiple categories of months trying to get pregnant (0–3 months, 4–6 months, 7–12 months, 13–24 months, or >24 months). Effect modification by Hispanic ethnicity was evaluated via stratified analyses and tested via the interaction term approach. SAS 9.4 and Stata 15 were used for the analysis.


After excluding the 20 missing values for BMI and 107 missing values for whether women were trying to get pregnant with their new baby, 5,644 women (98.2%) were included in the analyses, reflecting an estimated population size of 196,323 women (Figure 1).

Figure 1. Study Participant Flowchart: Utah PRAMS, 2012–2015

Characteristics of the Mothers

Median BMI of the study population was 23.8 (interquartile range [IQR] 21.1, 28.3) (kg/m2), with BMI categories of underweight (4.7%), normal weight (54.3%), overweight (22.3%), and obese (18.8%) (Table 1). Median age was 27.0 (IQR 24.0, 31.0) years old. The majority of women were parous (67.1%), had health insurance (82.1%), were married (83.2%) and had ≥ 12 years of education (90.3%). Most of the women (84.1%) were White non-Hispanic, while 9.2% were Non-white Hispanic and 6.7% were White Hispanic. Prior to pregnancy, nearly a third of women (30.8%) reported consuming alcohol and 11.2% smoked tobacco. Preconception prevalence of depression, diabetes, and hypertension were 10.1%, 1.3%, and 2.2%, respectively.

Table 1: Demographic, lifestyle and clinical characteristics of women by BMI, Utah PRAMS, 2012–2015, n=5644, reflecting an estimated population size of 196,323 women.

Compared to women of normal weight, obese women were more likely to be older, parous, of Hispanic ethnicity, have no health insurance, of lower education and family income, and report smoking or drinking alcohol in the two years prior to pregnancy. Moreover, obese women were more likely to have been previously diagnosed with diabetes, hypertension, or depression.

Association Between BMI and Subfertility Measures

Among women who were trying to get pregnant with their most recent baby, each unit increase in BMI was associated with a higher adjusted prevalence ratio (aPR) of months trying to get pregnant: 1.03 (95% CI 1.01, 1.05) for 4–6 months, 1.03 (95% CI 1.01, 1.05) for 7–12 months, 1.06 (95% CI 1.04, 1.09) for 13–24 months, and 1.08 (95% CI 1.06, 1.10) for >24 months compared to women who attempted to achieve pregnancy for 0–3 months (Table 2). Obese women were 1.58 times (95% CI 1.22, 2.06), 1.39 times (95% CI 1.03, 1.89), 1.92 times (95% CI 1.36, 2.70), and 3.31 times (95% CI 2.43, 4.50) as likely to have tried 4–6, 7–12, 13–24, and >24 months to get pregnant, respectively, compared to women who had tried 0–3 months.

Table 2: Unadjusted association between BMI and self-reported months trying to get pregnant.

After adjusting for maternal age, income, education, marital status, and race/ethnicity, women with preconception obesity, compared to normal weight women, had a 1.85 (95% CI 1.43, 2.38) higher aPR for having subfertility defined by time trying (Table 3). Continuous models indicated a linear relationship between BMI and subfertility (aPR: 1.04, 95% CI 1.03, 1.06); however, no association was found between underweight (aPR: 0.50, 95% CI 0.22, 1.14) or overweight (aPR: 1.06, 95% CI 0.80, 1.41) status and subfertility compared to normal weight. Similar findings were found for receiving any fertility-related drugs, insemination or in vitro fertilization [IVF]) with obese women having a 73% higher prevalence (95% CI 1.29, 2.32) of these procedures compared to normal weight women. Further adjustment for parity and preconception smoking, alcohol consumption, and depression in all models did not appreciably alter the findings (Table 3), nor did further adjustment for a prior diabetes or hypertension diagnosis. As exemplified in the stratified analyses (Table 4), no effect modification by Hispanic ethnicity was identified by the interaction test (Wald test F-value; P=0.73)

Table 3: Relationship between BMI and months trying to get pregnant or fertility treatment.
Table 4: Relationship between BMI and months trying to get pregnant, stratified by Hispanic ethnicity.


Our research among a population-based cohort of women found that obese women, compared to normal weight women, have a 73% and 85% higher probability of experiencing a longer time to pregnancy or using fertility treatment after controlling for a number of sociodemographic and lifestyle factors. No association was found between underweight or overweight women and subfertility; nor was effect modification by Hispanic ethnicity found.

Strengths of the Study

A population-based sample was used and weighted to represent all mothers who gave birth in Utah from 2012–2015. Further, our sample size included a representative proportion of Hispanic ethnicity, and ensured that at-risk women were included. Additionally, the PRAMS questionnaire included detailed information about socio-demographics, reproductive and health history, and lifestyle characteristics, therefore we were able to assess multiple confounding factors that may affect adiposity and subfertility. Finally, subfertility was measured through different ways (time trying and fertility-related drugs/medical procedures).

Limitations of the Data

First, the PRAMS questionnaire collects self-reported data from women who just delivered live births. The reliability of self-reported preconception height, weight, and months trying to get pregnant is dependent on women’s ability to accurately recall, which has been shown to be prone to error.13,14  Second, BMI may not be the best measure to assess women’s obesity because it does not account for ethnicity, age, body composition and shape, or healthy body mass such as muscle.3 The measurements of waist circumstance or waist-hip ratio for central adiposity would be helpful.15 Third, we could not account for BMI of male partners, which might influence the results.16 Fourth, selection bias cannot be ruled out. PRAMS follows a strict protocol for sampling mothers, but Utah’s average response rate for 2012–2015 was 69%. Fifth, certain reproductive disorders such as polycystic ovary syndrome (PCOS) may confound the relationship between BMI and subfertility, but such information was not available in the UT-PRAMS Phase 7 questionnaire. PCOS information has been added to the UT-PRAMS Phase 8 questionnaire (2016 to present) and thus, further research taking into account PCOS diagnosis and/or symptomology is warranted. Finally, perhaps most importantly, this study included only women who had a live birth; the results may differ if women who want to conceive but have not done so successfully yet were included.17


The finding of a relationship between obesity and subfertility agrees with an extensive body of previous literature.2,3,18-34 For instance, Brewer and Balen concluded that obesity impaired both natural and assisted conception, especially in women with a BMI >35 kg/m2.19 Gaskins et al. found that being overweight or obese in female adulthood was associated with modest reductions in fecundity that led to an increase in duration of pregnancy attempt.2 However, the results from our study differ from other studies in that we did not find that preconception overweight (not obese) women and subfertility are associated.2,15,21,24 Conflicting findings may be attributable in part to the fact that prior studies mostly examined women being treated for subfertility.2,21,24 Future studies among non-clinical populations are needed to clarify the relationship between adiposity and subfertility among women not seeking treatment.

Additionally, our findings are consistent with other research conducted in Utah,16 which may be reflective of the relatively good health of the Utah population compared to other states.25 We found no differences in the association of BMI with subfertility among Hispanic women compared to non-Hispanic women. This was most likely due to the relatively small sample size of Hispanic women in our dataset, thus we may have a limited power to detect the disparities between Hispanic and NHW women. However, similarly, Wise and colleagues did not find an association between overweight (BMI of 25.0-29.9 kg/m2) and reduced fecundity among African American women, but did find an association between class 2 and 3 obesity (BMI of ≥ 35.0kg/m2) and reduced fecundity.4 Whether there are clear differences in the effects of adiposity on subfertility among different race and ethnicities has yet to be elucidated. Further population-based research that includes adequate representation of women from various races and ethnicities is warranted before conclusions can be made.

Fertility treatment utilization within our sample was similar to that found in other representative samples. Among our sample of women who reported having sought out fertility treatment, 62.1% reported taking fertility drugs to help them get pregnant while 13.4% reported receiving artificial insemination. A National Survey of Family Growth (NSFG) study reported that nearly half of the women who were trying to get pregnant received drugs to improve ovulation, followed by 13.1% for artificial insemination.26 Because of the small sample sizes for Hispanic women receiving different types of fertility treatment, we were limited in our ability to report the disparities between Hispanic and non-Hispanic women in use of the various fertility treatments. While access to infertility treatment is beyond the scope of this study, given prior research showing that socioeconomic status is significantly associated with the ability to seek out fertility treatment in the US,27 increased equity in access to fertility diagnostics and treatment is needed.28

Health Implications

In brief, this population-based PRAMS study inclusive of at-risk mothers found that preconception obesity, but not overweight or underweight, was associated with women’s subfertility, consistent with prior research. There was no difference by Hispanic ethnicity nor when evaluating subfertility in multiple ways. Given inconsistent findings to date, we are wary to make recommendations for clinicians or policy makers based on our findings. Further population-based research adequately including women and couples of various races and ethnicities is needed to help better understand whether healthy women who are overweight, but not obese, are comparable to normal weight women in regards to ability to achieve a pregnancy. This research is important given that women deserve to have preconception counsel in regards to risk factors for subfertility based not on intuition but rather findings from sound and methodologically rigorous research.


Author Contributions:

Concept and design: Dr. Karen C. Schliepand Qingqing Hu

Data analysis: Qingqing Hu, Jihyun Lee, Jeannette Nelson, Marci Harris, Rebekah H. Ess

Drafting of Manuscript: Qingqing Hu, Jihyun Lee

Critical revision of the manuscript: Drs.Charles R. Rogers, Jessica Sanders, James VanDerslice, and Joseph B. Stanford

Final approval: Dr. Karen C. Schliep

Sources of Funding:

This work was supported by Dr. Rogers’s funds from the National Cancer Institute of the National Institutes of Health (NIH) [grant number K01CA234319].

Disclosure of Potential Conflicts of Interest:

None reported

Additional Contributions:

Data were provided by the Utah Pregnancy Risk Assessment Monitoring System (PRAMS), a project of the Utah Department of Health (UDOH), the Office of Vital Records and Health Statistics of the UDOH, and the Center for Disease Control and Prevention (CDC) of the U.S. Health and Human Services Department. This report does not represent the official views of the CDC, Utah Department of Health, or the NIH.


1.  Branum A, Kirmeyer S, Gregory E. Prepregnancy body mass index by maternal characteristics and state: data from the birth certificate, 2014. National Vital Statistics Reports: From the Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System. 2016;65:1–11.

2.  Gaskins AJ, Rich-Edwards JW, Missmer SA, Rosner B, Chavarro JE. Association of fecundity with changes in adult female weight. Obstetrics and Gynecology 2015;126:850–858.

3. Loy SL, Cheung YB, Soh SE, Ng S, Tint MT, Aris IM, et al. Female adiposity and time-to-pregnancy: a multiethnic prospective cohort. Human Reproduction 2018;33:2141–2149.

4. Wise LA, Palmer JR, Rosenberg L. Body size and time-to-pregnancy in black women. Human Reproduction 2013;28:2856–2864.

5. Luke B. Adverse effects of female obesity and interaction with race on reproductive potential. Fertility and Sterility 2017;107(4):868-877.

6. Quinn M, Fujimoto V. Racial and ethnic disparities in assisted reproductive technology access and outcomes. Fertility and Sterility. 2016;105:1119-1123.

7.  Ely DM, Hamilton BE. Trends in Fertility and Mother’s Age at First Birth Among Rural and Metropolitan Counties: United States, 2007-2017. National Center for Health Statistics Data Brief 2018 Oct;323:1–8.

8.  Feinberg EC, Larsen FW, Wah RM, Alvero RJ, Armstrong AY. Economics may not explain Hispanic underutilization of assisted reproductive technology services. Fertility and Sterility. 2007;88:1439–1441.

9. Pregnancy Risk Assessment Measurement Scale (PRAMS). https://www.cdc.gov/prams/index.htm. Accessed January 26, 2020.

10. Shulman HB, D’Angelo DV, Harrison L, Smith RA, Warner L. The Pregnancy Risk Assessment Monitoring System (PRAMS): Overview of Design and Methodology. American Journal of Public Health. 2018;108:1305–1313.

11.  Practice Committee of American Society for Reproductive Medicine. Definitions of infertility and recurrent pregnancy loss: a committee opinion. Fertility and Sterility 2013;99:63.

12.  Petersen MR, Deddens JA. A comparison of two methods for estimating prevalence ratios. BMC Medical Research Methodology. 2008;8:9.

13.  Han E, Abrahms B, Sridhar S, Xu F, Hedderson M. Validity of self-reported pre-pregnancy weight and body mass index classificaiton in an integrated health care delivery system. Paediatric and Perinatal Epidemiology 2016;30:314–9.

14.  Cooney MA, Buck Louis GM, Sundaram R, McGuiness BM, Lynch CD. Validity of Self-Reported Time to Pregnancy. Epidemiology 2009;20:56–59.

15. Wise LA, Rothman KJ, Mikkelsen EM, Sørensen HT, Riis A, Hatch EE. An internet-based prospective study of body size and time-to-pregnancy. Human Reproduction 2009;25:253–264.

16.  Schliep KC, Mumford SL, Ahrens KA, Hotaling JM, Carrell DT, Link M, et al. Effect of male and female body mass index on pregnancy and live birth success after in vitro fertilization. Fertility and Sterility 2015;103:388–395.

17.  Basso O, Juul S, Olsen J. Time to pregnancy as a correlate of fecundity: differential persistence in trying to become pregnant as a source of bias. International Journal of Epidemiology 2000;29:856–861.

18. Klenov VE, Jungheim ES. Obesity and reproductive function: a review of the evidence. Current Opinion in Obstetrics and Gynecology. 2014;26:455–460.

19.  Brewer CJ, Balen AH. The adverse effects of obesity on conception and implantation. Reproduction. 2010;140:347–364.

20.  Gesink Law D, Maclehose RF, Longnecker MP. Obesity and time to pregnancy. Human Reproduction. 2006;22:414–420.

21.  Koning A, Kuchenbecker W, Groen H, Hoek A, Land JA, Khan KS, et al. Economic consequences of overweight and obesity in infertility: a framework for evaluating the costs and outcomes of fertility care. Human Reproduction Update. 2010;16:246–254.

22.  Lash MM, Armstrong A. Impact of obesity on women’s health. Fertility and Sterility. 2009;91:1712–1716.

23.  Poston L, Caleyachetty R, Cnattingius S, Corvalan c, Uauy R, Herring S, et al. Preconceptional and maternal obesity: epidemiology and health consequences. The Lancet Diabetes & Endocrinology. 2016;4:1025–1036.

24. Kort JD, Winget C, Kim SH, Lathi RB. A retrospective cohort study to evaluate the impact of meaningful weight loss on fertility outcomes in an overweight population with infertility. Fertility and Sterility 2014;101:1400–1403.   

25. United Health Foundation America’s Health Rankings 2018 Annual Report. https://www.americashealthrankings.org/learn/reports/2018-annual-report. Accessed January 26, 2020.

26.  Vahratian A. Utilization of fertility-related services in the United States. Fertility and Sterility 2008;90:1317–1319.

27. Farley Ordovensky Staniec J, Webb NJ. Utilization of infertility services: how much does money matter? Health Services Research. 2007;42:971–989.

28.  Boulet SL, Kawwass J, Session D, Jamieson DJ, Kissin DM, Grosse SD. US State-Level Infertility Insurance Mandates and Health Plan Expenditures on Infertility Treatments. Maternal and Child Health Journal. 2019;23:623–632.

29.  Meldrum DR, Morris MA, Gambone JC. Obesity pandemic: causes, consequences, and solutions—but do we have the will? Fertility and Sterility. 2017;107:833–839.


Hu Q, Lee J, Nelson J, Harris M, Ess R, Rogers CR, Sanders J, VanDerslice J, Stanford JB, & Schliep KC. (2020). The association between preconception body mass index and subfertility among Hispanic and non-Hispanic women: A cross-sectional study from Utah’s Pregnancy Risk Assessment Monitoring System survey (2012–2015). Utah Women’s Health Reviewdoi: 10.26054/0KYE1P6XVT.


View / download