Abstract
Objectives:
To determine associations between primary provider specialty and the contraceptive care that patients receive in a Federally Qualified Health Center setting in Maryland.
Methods:
A study of reproductive-age patients and their providers was performed from January 2018 to December 2021. A pooled crosssectional survey of electronic medical record data for 44 127 encounters of 22 828 patients was performed to calculate the odds of contraceptive care being addressed by patients who had General Practitioner, OB/GYN, pediatrician, or infectious disease (ID) specialists as their primary providers.
Results:
In 19 041 encounters (43%), contraception was addressed through either counseling alone, documentation of a contraceptive prescription, or long-acting reversible contraceptive (LARC) placement procedure. After adjusting for insurance status and race/ethnicity, the odds ratio (OR) of contraceptive care delivery was statistically significantly higher for OB/GYN providers compared to General Practitioners—OR 2.42 (CI 2.29-2.53) and statistically significantly lower for ID providers—OR 0.69 (CI 0.61-0.79). There was a non-statistically significant difference for Pediatricians—OR 0.88 (CI 0.77-1.01).
Conclusion:
The provision of contraceptive care, a critical aspect of comprehensive primary care delivered in an FQHC setting, varies by provider specialty, and may be negatively influenced by Ryan White funding related structures. There is a need to intentionally design robust referral and tracking systems to ensure contraceptive care is equitably accessible to all, regardless of assigned primary care provider specialty or HIV status.
Introduction
Unintended pregnancies impose risks for health and well-being, such as late entry to prenatal care, poorer birth outcomes, and decreased economic and educational lifetime achievements. 1 One proven strategy for avoiding unintended pregnancies is to ensure access to desired, effective contraceptive methods and family planning (FP) services. Access to long-acting reversible contraceptives (LARCs) in particular has been shown to address racial, economic, and age disparities in unintended pregnancies, mitigating the health and social consequences that disproportionately impact groups that have been historically disenfranchised.2,3 Consequently, multiple health and healthcare professional organizations have called for increasing access to contraceptive care in primary care settings, including the United States Health Resources & Services Administration’s funded Women Preventive Services Guidelines, the American College of Obstetricians, and the American Academy of Family Physicians.4,5
Previous research has examined the availability of contraceptive services in federally qualified health centers (FQHCs) nationwide, documenting whether service providers and on-site contraceptive prescription options exist, but not whether contraceptive services are reliably and equitably delivered in those settings.6,7 Persistent differences in preparedness to deliver contraceptive counseling and care by provider specialty have been documented, and efforts to improve pediatrician training around contraceptive care specifically have been associated with increased access to highly effective forms of contraception for adolescents.8,9 Additionally, while contraceptive use among HIV positive individuals has been documented to be lower than among the general population in the U.S., it is not clear the extent to which the background training and specialty of providers caring for HIV positive individuals may influence their contraceptive use outcomes. 10 To the study team’s knowledge, no prior analysis has evaluated the impact of provider specialty on the delivery of contraceptive care in an FQHC that serves a large subpopulation of HIV positive individuals.
We used a retrospective pooled cross-sectional survey design extracting medical record data over 4 years (2018-2021) to examine the impact of primary provider specialty on contraceptive care outcomes in an FQHC with multiple clinic sites in the greater Baltimore area. Our goal was to assess whether contraception was addressed, either through counseling or provision of services, for all eligible patients that visited the FQHC within a year, for every year that was assessed. We then looked for trends among provider specialties, with additional analysis of subgroups with high-risk sexual activity.
Methods
Study Setting and Population
The study location is an FQHC in Maryland that serves approximately 40 000 patients annually. There are sites in Baltimore City and surrounding counties with a payer mix of approximately 40% Medicaid, 7% Medicare, 26% private insurance, and 28% uninsured patients. A total of 88 providers were practicing at the 6 clinic sites between 2018 and 2021: 61 General Practice, with approximately equal numbers of Family Medicine Physician and Family Nurse Practitioner staff members, as well as a small number of Internists and Physician’s Assistants; 15 Reproductive Health, primarily Certified Nurse-Midwives as well as 3 OB/GYN physicians; 7 Pediatricians; and 5 Infectious Disease (ID) specialists.
Study Design, Data Source, and Analysis
A pooled cross-sectional survey analysis was performed using data from 2018, 2019, 2020, and 2021. For each individual year, data was extracted from the health system’s electronic medical record (EMR). Inclusion criteria included women aged 15 to 45 who made at least 1 visit to the FQHC during the corresponding year, starting January 1 to December 31. Those with a previous history of hysterectomy or were biologically male at birth were excluded. For each included patient record, demographic and provider information was extracted from the EMR, including their age, sex, race, ethnicity, gender identity, insurance status, clinic site, and primary provider. These variables were treated as categorical variables. Primary provider was defined as the provider with whom the patient had the most encounters in the corresponding year, and their specialty (General Practitioner, Pediatrician, Infectious Disease, OB/GYN or Certified Nurse-Midwife) was extracted from internal credentialing records. Data analysis was conducted between April and June 2022.
The primary outcome was whether patients had contraceptive care addressed, based on evidence from the EMR, during the year in which they received care (January 1st-December 31st). Contraceptive care was considered addressed based if the patient received contraceptive counseling, prescription for contraceptives, or a relevant procedure. Data on these outcomes were extracted using ICD-10 codes for contraceptive counseling, CPT codes for placement of Long-acting Reversible Contraceptive (LARC) and documented patient medications for prescriptions for other forms of contraception (Supplemental Appendix 1). Individuals having one of these variables documented within the calendar year were classified as having contraception addressed. Encounters for all 4 years were then pooled into 1 dataset to estimate the period prevalence of contraceptive care by medical specialty from 2018 to 2021. Individuals represented across multiple years were treated as separate individuals for each corresponding year.
Next, we performed a subgroup analysis of individuals who had reproductive health history relevant to predicting unplanned pregnancy risk. These were individuals with ICD-10 codes for HIV, syphilis, exposure to sexually transmitted infections, pregnancy, and high-risk sexual behaviors (Supplemental Appendix 2). At this FQHC, infectious disease (ID) providers exclusively provide care to HIV positive patients, and those providers appeared less likely to address contraception during the study period. Therefore, we performed a chi-square test of association between HIV status and contraception addressed.
The demographic characteristics were described, and a univariate analysis was conducted using chi-square test to explore the association between the outcomes and categorical explanatory variables. A univariate logistic regression was then used to describe the odds of having contraception addressed by the levels of each explanatory variables. A multivariate logistic regression was then used to describe the relationship between the primary provider specialty and the primary outcome adjusting for known confounders, including age, race, ethnicity, gender identity, and insurance status. Clinic location and year were also included to account for variations in the populations and changes that may have occurred in service delivery during the COVID-19 pandemic. An observed variation in outcomes of pediatric patients by transgender status led to an investigation of an effect modification of being transgender for those patients seen by pediatric specialists. All statistical analyses were performed using Stata version 17.
Ethical Considerations
The study was reviewed and approved by the Johns Hopkins Bloomberg School of Public Health IRB in September 2021.
Results
Patient and Provider Characteristics
A total of 44 955 patient encounters were initially included over the 4 years, representing 23 123 unique individuals. A total of 670 encounters were dropped as they had either an acute-care provider listed as the patient’s primary provider or a provider with less than 10 encounters over the 4 years. A total of 158 individuals identified as Male to Female transgender and their encounters were dropped. Hence, the final eligible population included 44 127 patient encounters representing 22 828 unique individuals. The average age was 30 years old; roughly 39% identified as black and 84% as cis-female (Table 1).
Patient’s Descriptive Statistics by Provider Specialty.
There were 88 providers: 61 General Practitioners, 15 OB/GYN or Certified Nurse-Midwife (CNM), 7 Pediatricians, and 5 ID providers. A General Practitioner was the primary provider in 31 434 patient encounters. A total of 9672 patient encounters had an OB/GYN or CNM as the primary provider, while 1835 and 1185 had Pediatricians and ID providers, respectively.
Odds of Contraception being Addressed
In a total of 19 041 patient encounters (43.2%), contraception was addressed in the recorded year through either counseling or documentation of a prescription or contraception placement procedure. Primary providers who were OB/GYN physicians or midwives had the highest percentage of patients with contraceptive addressed with 63.2%. General Practitioners and Pediatricians had similar percentages, with 37.8% and 36.2% of their patients having contraception addressed, while 32.8% of patients with an ID primary provider had contraception addressed.
The odds of contraception being addressed were statistically significantly higher for those with private insurance compared to Medicaid, 1.19 (95% CI 1.13-1.26); whites compared to African American, 1.12 (1.06-1.19); and Hispanics compared to non-Hispanics, 1.25 (1.17-1.34).
Odds of Contraception being Addressed by Provider Specialty
Adjusting for known confounders, the odds of contraception being addressed if the primary provider was an OB/GYN or CNM was 2.42 (95% CI 2.29-2.53) (Table 2) compared to General Practitioners, and the odds ratio was statistically significant. The odds if the primary provider was a pediatrician was 0.88 (95% CI 0.77-1.01) compared to General Practitioners, and the odds ratio was not statistically significant. The odds for ID providers was 0.69 (95% CI 0.61-0.79) compared to General Practitioners, and the odds ratio was statistically significant. Evidence for an effect modification of transgender status on pediatricians’ provision of contraceptive services was found. The odds of contraceptive care if seeing a pediatrician and being Transmale was 1.68 (95% CI 1.29-2.19), while the odds for cis-females when seeing a pediatrician was 0.26 (95% CI 0.22-0.31) compared to pediatric patients seeing a generalist primary provider.
Odds of Contraception Addressed.
Sub-analysis of individuals deemed high-risk for sexual activity, including those with a history of sexually transmitted infections, showed slightly modified trends (Table 3). OB/GYN or CNM providers were associated with a 1.75 odds (95% CI 1.63-1.87, P-value: <.005) of contraception being addressed compared to General Practitioners. Pediatricians improved in this population, with an odds of 1.74 (95% CI 1.34-2.26, P-value: <.005) of contraception being addressed for this subgroup. Finally, ID providers were once again statistically significantly less likely to address contraception for this subgroup compared to General Practitioners, with 0.59 odds (95% CI 0.51-0.69, P-value: <.005). In a sub-analysis of individuals with HIV and high-risk sexual activity, adjusted odds fell to 0.41 (95% CI 0.23-0.75, P = .003) for ID providers compared to General Practitioners seeing patients with HIV. The overall association between HIV status and whether contraception was addressed was not significant (X2: 1.23, P-value: .267).
Adjusted Odds of Contraception by Provider Specialty and High-Risk Sexual Activity.
Discussion
Ensuring adequate access and provision of contraceptive services is a critical issue for health systems, payors, and policymakers. Our study shows that provider-level factors such as specialty affect contraceptive access for patients. Providers in the Department of Reproductive Health are more than twice as likely as General Practitioners to address contraception, and ID providers are the least likely to do so for adult patients. Even those providers who are fully committed to meeting contraceptive care needs of their patients report challenges. A lack of time to fully address the topic of contraception amidst competing visit priorities, 11 cost barriers for uninsured patients, 12 lack of confidence to address contraception needs, especially LARCs, by non-OB/GYN providers, 13 and assumptions that contraception care is provided in other settings, are all cited reasons that contraceptive care is inconsistently provided.8,14
Contraception practice behavior comparing General Practitioners to Pediatricians may initially appear similar in the FQHC assessed. However, a closer examination reveals that pediatricians significantly prioritize contraceptive care for trans-male compared to cis-female patients and increase attention to contraception for those with high-risk sexual behavior. General Practitioners on the other hand do not appear to address contraception differently amongst these sub-groups. The positive effect modification of being transmale among the pediatric population prompted our further review of pediatric trans-health provider practices. These providers have a prolonged intake appointment with a standardized clinic note that includes assessment of contraceptive needs, given the risk of teratogenic effects of masculinizing therapy that may be anticipated or prescribed. Clinic note standardization related to contraceptive needs amongst all pediatric and adult patient encounters is recommended to consistently address contraceptive needs and desires for those attending the FQHC, regardless of gender or sexual behavior.
Consistent with prior study, 15 patients with private insurance (which may reflect their socioeconomic status and ability to pay for health services) compared to Medicaid patients, and those who identify as White compared to Black/African American, were more likely to have their contraceptive needs met at the FQHC—which may further worsen the extant health inequities in the represented communities. This finding calls for strategies to address implicit biases that are known to impact the experience of contraceptive care for non-White individuals. 16 The finding that recent immigrants from Central America (the sub-group that self-identify as Hispanic) were more likely to have their contraceptive care needs met compared to non-Hispanic individuals warrants additional analysis to further tailor strategies to sub-groups that have been historically disenfranchised by the health system.
A limitation of our study is that we utilized the EMR to estimate if contraceptive care was received, which may under-estimate actual care delivery due to lack of consistent documentation by providers, or over-estimate provision of prescribed contraceptive methods due to inaccuracies in recorded medication lists. Given that the same individual may have been represented over multiple years, an increased correlation in some of our extracted variables from the patient record is likely which may have affected the precision of our estimated logistic coefficient.
This FQHC is a recipient of Ryan White program funding which supports ID providers who serve as primary care providers, with the expectation that comprehensive primary care needs, including contraceptive services, are addressed, either directly or through referral to colleagues. The statistically significantly lower odds of having contraceptive care addressed for those patients seeing ID providers suggests a need for focused education and awareness among both ID providers and patients cared for under the Ryan White program, along with quality assurance mechanisms for addressing contraceptive needs regardless of HIV diagnosis or provider assignment. Further research is needed to explore any broader impact of Ryan White funding on contraceptive care outcomes in primary care settings.
Access to contraceptive care is of utmost importance to avoid continued disparities in reproductive health outcomes, and provider specialty is a determinant of access to contraceptive care in the FQHC setting and beyond. 17 Hence, without adequate mechanisms to ensure that eligible patients who need contraceptive care can receive the care that they need irrespective of their provider’s specialty, there is a risk of deepening disparities in reproductive health outcomes due to unequal access to contraception. In a clinical setting where comprehensive primary care is expected to be delivered, it may benefit patients to have reproductive health-trained providers assigned to partner with primary providers. Where this is not feasible, robust systems and standardized protocols should be in place to continually assess contraceptive care-related needs and ensure that these needs are being met by a reproductive health-trained secondary provider within the FQHC. Further research is needed to explore clinical interventions, such as internal referral mechanisms, tracking systems, and team-based approaches.
Supplemental Material
sj-docx-1-jpc-10.1177_21501319231173555 – Supplemental material for How do Specialists Address Contraceptive Care Compared to General Practitioners in a Federally Qualified Health Center in Maryland
Supplemental material, sj-docx-1-jpc-10.1177_21501319231173555 for How do Specialists Address Contraceptive Care Compared to General Practitioners in a Federally Qualified Health Center in Maryland by Leah Hart, Jarett Beaudoin, Georgia Parsons and Olakunle Alonge in Journal of Primary Care & Community Health
Footnotes
Acknowledgements
The author team is grateful for Chase Brexton Health Care senior leadership team and research committee members who provided approval for the data use agreement and study proceedings. The authors also extend their thanks to the interview participants.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
