REI clinics and fellowship training—a national snapshot to improve access to reproductive care

Abstract

Purpose

To analyze the geographic distribution of REI fellowships and clinics across the USA and to strategize ways to improve patient access to care.

Methods

Cross-sectional study using population data obtained from publicly available United States Census Bureau, Society for Assisted Reproductive Technology (SART), and National Resident Matching Program websites. Outcomes include the number of REI clinics, REI fellowship-trained physicians, and REI fellowship programs.

Results

In 2020, there were 643 assisted reproductive technology (ART) clinics reporting to SART and 1351 fellowship-trained REI physicians. Most clinics are located in the south (n = 209); however, the northeast has the highest density of REI clinics. Out of 301,316 in vitro fertilization (IVF) cycles in the USA in 2020, northeastern states initiated the most cycles (n = 93,565), and Midwestern states initiated the fewest cycles (n = 50,000). The northeast has the most REI physicians per million women aged 20–44 years (42.4) while the Midwest has the lowest ratio (19.5). There are fewer REI physicians per million women aged 20–44 years in states with a lower proportion of patients with health insurance (r = 0.56, 95% confidence interval ([CI] 0.34–0.73) and in states with a lower average income per resident (r = 0.65, 95% CI 0.46–0.79). Most of the 49 accredited REI fellowship programs in the USA are in the northeast (n = 18), and there are fewest in the south (n = 10) and west (n = 10).

Conclusion

Access to REI care has large geographic disparities from a clinic, physician, and training program perspective. Creative solutions are needed to remedy this problem.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10815-023-02868-y.

Introduction

The demand for reproductive endocrinology and infertility (REI) care is increasing, and more people are seeking assisted reproductive technology (ART). In the USA, in vitro fertilization (IVF) cycle volume rose from 142,241 cycles reported in 2009 to 334,827 cycles in 2019 [1]. This growth likely results from a combination of contributing factors. Maternal age at the time of conception continues to increase in the USA [2]; and older women attempting pregnancy have a higher rate of infertility. Additionally, more women are pursuing oocyte cryopreservation; the number of these cycles increased from 2719 in 2012 to 16,945 in 2020 [1, 3]. Increased access to care in some states may also contribute to this trend as more states mandate private insurance coverage for reproductive services. Studies have shown that financial barriers greatly contribute to underutilization of REI services; of infertile women seeking REI care, only 22% underwent any treatment [4]. Equalization of cost through mandated insurance has improved access to care [5].

Currently, 20 states mandate insurance coverage for diagnosis and treatment of infertility [6]. Of these twenty states, fourteen mandate coverage for IVF—Arkansas, Colorado, Connecticut, Delaware, Hawaii, Illinois, Maine, Maryland, Massachusetts, New Hampshire, New Jersey, New York, Rhode Island, and Utah. Most states limit the maximum number of IVF cycles that will be covered with some covering just one cycle (Hawaii) and others covering up to six cycles (Delaware and Illinois). Others provide a lifetime maximum coverage ranging from $4,000 (Utah) to $100,000 (Maryland and Rhode Island). Lastly, there has been an increase in REI services by lesbian, gay, bisexual, and transgender individuals (LGBT) for both family building and fertility preservation [7, 8]. With this rise in ART utilization, it is imperative to broaden access to reproductive services while simultaneously increasing the workforce to accommodate increased demand.

Formal subspecialty training in REI in the USA began in 1972 and has changed significantly over the past 50 years with the first live birth from IVF in 1978 [9]. For some time, REI fellowships were 2 years in length. Now, REI fellowships are 3-year training programs following a 4-year obstetrics and gynecology (OB/GYN) residency. They include training in ART, complex reproductive endocrine disorders, reproductive surgery, pediatric and adolescent gynecology, male infertility, courses in genetics and biostatistics, and research. More than a decade ago, a study was published analyzing the geographic distribution of REI fellowships in the USA; in 2005, there were 42 fellowship programs, with most located east of the Mississippi River [10]. As of 2022, despite the large increased demand, there are only 49 fellowship programs [11].

We aim to analyze the geographic distribution of REI clinics and fellowships across the USA in relation to state population size, demographics, and fertility insurance coverage to strategize ways to improve access to care.

Methods

Publicly available population data were obtained from the U.S. Census Bureau (USCB). Collected information by state included population density, number of women aged 20–44 years, racial and ethnic distribution, median household income, and percentage of residents with any health insurance (private, public, Medicaid, etc.). We also collected information on the number of major metropolitan areas in each state as defined by the USCB and identified states with fertility insurance mandates as of February 2022. Information regarding states with mandated infertility services coverage, including mandated coverage for IVF cycles, was obtained from Resolve.

Using the Society for Assisted Reproductive Technology (SART) and National Resident Matching Program (NRMP) websites, we obtained information about REI fellowships and clinics. Specifically, we analyzed the number and location of fellowship programs and clinics in each state. In reviewing SART’s website, we included all main and satellite clinics which are also included on the national search. Clinic websites listed on SART were also analyzed to determine the number of physicians who had completed an REI fellowship at each practice, thereby allowing us to determine number of ART clinics and physicians per state as well as number or IVF cycles in 2020. Lastly, we used SART data to determine the number of initiated ART cycles in each state in 2020.

The number of REI clinics per million residents and the number of clinics per million women of reproductive age (20–44 years) were calculated by state. Using the number of REI physicians per state, we calculated the number of REI physicians per million residents and per million reproductive-aged women for each state. We also reported the number of ART clinics and REI physicians per person in each state.

Addresses for each REI fellowship location and each REI clinic and satellite location were geocoded to a USA map using ArcGIS Online (Esri, World Topographical Map). A drive-time analysis was performed using each REI clinic location as a focal point. The ArcGIS Online algorithm models the movement of small automobiles to optimize drive-time with the option to calculate drive-time both with and without traffic. We chose to model 30-, 60-, 90-, and 120-min drive-times towards each clinic facility using each REI clinic as the end point, as this would estimate driving distance for patients to reach a clinic. We specified to estimate driving time in normal traffic patterns for a Monday morning. For areas in which there were overlapping drive-times (e.g., two clinics within a 30-min driving distance of a starting point), we merged these areas as our aim was to show which geographic locations in the country had little or no access to a clinic.

The USA was subdivided into northeastern (Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont), southern (Alabama, Arkansas, Delaware, Florida, Georgia, Kentucky, Louisiana, Maryland, Mississippi, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, West Virginia, Washington D.C.), Midwestern (Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, Wisconsin), and western (Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, Wyoming) states based on the USCB for our analysis. Linear regression was used to investigate relationships between REI clinic and physician geographic distribution as well as population demographics. The number of REI clinics per states was modeled using percent of the population covered by any health insurance (public, private, or alternative group plans) as the predictor variable. Patient access to care was evaluated by using multiple separate univariate linear regressions to investigate both number of REI physicians per state as well as number of REI physicians per million reproductive-aged women in each state using the following predictor variables: % population covered by insurance, median state income, and mandated insurance coverage for fertility services (both diagnostic and treatment). Additionally, given a positive correlation between median state income and mandated insurance coverage, a multivariable linear regression was fit considering both of these variables as predictors with REI physicians per million reproductive-aged women as the outcome to investigate whether this changed any of our results from the univariable regressions. Finally, a separate univariate linear regressions was fit for the number of REI physicians per state as well as number of REI physicians per million reproductive-aged women per state using a variable for % state population that identified as non-White as the predictor. A supplemental analysis was performed that weighted each state observation by total population.

Results

In 2020, there were 643 total ART clinics reporting to SART. This includes both main and satellite clinics. These clinics employ a total of 1351 fellowship-trained REI physicians. Two states, Alaska and Wyoming, have neither REI clinics nor physicians. Most clinics are in Southern USA (n = 209), with Northeastern USA having the highest clinic density (Fig. 1). Drive-times were longest in the Midwest and Western USA, with large swaths of the country more than 120 min away (Fig. 1).

Fig. 1.

Driving distance to REI clinics up to 120 min

In 2020, northeastern states initiated the most cycles (n = 93,565) while Midwestern states initiated the fewest (n = 50,000). The Southern USA has the highest number of physicians (n = 441), while northeast has the most physician density (42.4 REI physicians per million women aged 20–44) and Midwest has the lowest density (10.5 REI physicians per million women aged 20–44) (Fig. 2).

Fig. 2.

Number of REI physicians per million women aged 20–44 and location of REI fellowship programs in 2022

There are currently 49 accredited REI fellowship programs in the USA (Fig. 2). The Midwest (n = 12) and Northeast (n = 18) have the most programs, with concentration in New York state (n = 7). Two states (New Hampshire and Maine) have no fellowship programs. Approximately half of the southern and Midwestern states do not have REI fellowships, and only 5 of the 13 Western states have REI fellowships (California, Colorado, Oregon, Washington, and Utah). Two-thirds of the fellowship programs in the western states are in California.

Patient insurance status and total income were correlated with better access to REI physicians. In states with a higher rates of health insurance, there were more REI physicians per million reproductive-aged women (r = 0.56, 95% confidence interval [CI] 0.34–0.73) (Fig. 3a, Table 1). Similarly, in states with a higher average income per resident, there are more REI physicians per million reproductive-aged women (r = 0.65, 95% CI 0.46–0.79) (Fig. 3b). Additionally, when both median household income and proportion of insured were included in a multivariable model, both variables were independently associated with REI physicians per million reproductive-aged women. However, there was no correlation between the proportion of patients with insurance and the number of clinics (r = −0.04, 95% CI−0.32–0.24). There was a weak relationship between number of REI physicians per million reproductive-aged women and fertility coverage (r = 0.33, 95% CI 0.06–0.55). We found a weak correlation between number of REI physicians and states with a higher percentage of minorities (r = 0.42, 95% CI 0.17–0.62), but this correlation did not remain when analyzed by a number of REI physicians per million reproductive-aged women (r = 0.24, 95% CI −0.03–0.49). Results weighting by total state population are in Supplemental Table 1. Notably, there was a stronger positive correlation between number of clinics and fertility coverage and a positive correlation between median household income and number of clinics.

Fig. 3.

a Insurance status and number of REI physicians per million women aged 20–44 with each data point representing a different state or Washington D.C. b Median household income and number of REI physicians per million women aged 20–44

Table 1.

Correlation coefficients for univariable analyses

Outcome	Predictor	Correlation coefficient (r)	Lower 95% CI	Upper 95% CI	p-value
Number of REI physicians per million women aged 20–44	Median household income	0.56	0.34	0.73	<0.001
	Percent insured	0.56	0.34	0.73	<0.001
	Fertility coveragea	0.33	0.06	0.55	0.018
	Fertility treatment coverageb	0.38	0.12	0.59	0.005
	Percent minority	0.24	−0.03	0.49	0.08
Number of REI physicians	Median household income	0.42	0.17	0.62	0.002
	Percent insured	−0.004	−0.28	0.27	0.98
	Fertility coveragea	0.37	0.10	0.58	0.007
	Fertility treatment coverageb	0.12	−0.16	0.38	0.40
	Percent minority	0.42	0.17	0.62	0.002
Number of REI clinics	Median household income	0.23	−0.05	0.48	0.10
	Percent insured	−0.04	−0.31	0.24	0.77
	Fertility coveragea	0.36	0.10	0.58	0.008
	Fertility treatment coverageb	0.10	−0.18	0.37	0.48
	Percent minority	0.42	0.16	0.62	0.002

^aFertility coverage refers to coverage for either diagnostic testing or treatment

^bFertility treatment coverage refers to mandated coverage for treatment of fertility and encompasses 14 states (Arkansas, Colorado, Connecticut, Delaware, Hawaii, Illinois, Maine, Maryland, Massachusetts, New Hampshire, New Jersey, New York, Rhode Island, Utah)

Discussion

REI clinics in the USA are not equitably positioned. Clinics and physicians are positioned in states with higher population densities; specifically, there is a higher density of clinics and physicians in the northeast and a lower density in the West and Midwest. Furthermore, clinics are clustered close to urban areas, which affects access for rural patients. Although clinics are understandably based in areas with ample patients (and accompanying reimbursement), there is obvious inequitable access to care for many patients, some of whom have zero clinics in their entire state (e.g., Alaska and Wyoming). Additionally, there are fewer REI clinics and physicians in states with a lower median household income and a lower proportion of individuals with any type of health insurance. Distribution of fellowships reasonably mirrors the distribution of clinics and population density as demonstrated by USCB data [12]. Midwestern states have the lowest density of REI physicians and initiated the fewest IVF cycles in 2020. Contrarily, northeastern states initiated the most IVF cycles in 2020 and have the highest density of REI clinics, fellowships, and REI physicians. Most of the clinics are in large, metropolitan areas and far for those outside of cities.

Our findings are similar to previously published studies. A prior study using 2009–2013 data demonstrated that most clinics are in large, metropolitan areas and that 28.8% of the USA population lived in an area with no ART clinics [13]. Additionally, studies have similarly shown that northeastern states continue to have the highest density of REI clinics, fellowships, and physicians [14]. Lastly, our findings are similar to prior studies in showing higher numbers of clinics in states with mandated insurance coverage [15, 16].

Given the above information, there are range of suggestions to help improve access to care. Telemedicine for clinic visits has been touted as a potential antidote for geographic hurdles to care; however, barriers do exist in terms of high-speed internet access, potential “dyssynchronous care,” and the difficulty of having hard conversations across a computer screen [17]. Increasing state-mandated fertility coverage may help improve access to care [18] as there are only fourteen states that require coverage for IVF. There is a large unmet need of underinsured patients and those with lower income. In addition, support for organizations that work to increase access to care may help overcome some barriers [18, 19]. While mandated fertility insurance coverage does improve access to many individuals, these mandates do not apply to those with federally funded insurance through medical assistance programs, leaving a large proportion of people without access to fertility benefits and coverage.

With expansion of fertiliy assistance and coverage, there will be continued need to expand both the physicians workforce and other members of the healthcare team both in sheer numbers and geographically. More physicians are needed overall, but specifically in underserved areas. This can be accomplished with both an increase in the number of and wider distribution of REI fellowships. This results in an increase in the workforce and improve access in REI sparse areas, as up to 77.2% of graduates remain in the state in which they completed their training [19]. In addition, strategies are needed to attract REI physicians to states with low median income and poor infertility coverage. One potential solution is loan payback programs for graduating REI fellows who work in an underserved areas [20]. However, this should not be a sole motivator for geographic location of clinics because infertility is a disease, and there are powerful social and financial incentives for fertility assistance for society as a whole [21].

Improving current operational efficiency can also expand access to care by allowing the physician to see more patients. In the current care model, REI physicians spend hours each day answering patient messages [22]. Web-based applications have been created in an attempt to streamline communication, but they have not yet been successful [23]. To increase the efficiency of time spent in the REI office, the OB/GYN specialist should complete the basic infertility evaluation prior to referral. Where appropriate, ovulation induction could be initiated. Currently, most graduates of OB/GYN residency programs do not spend sufficient time learning the fundamentals of REI, and approximately half of graduating OB/GYN residents felt that their exposure to REI was “too little” [24]. Furthermore, REI physicians can expand geographic access to care by providing educational support to local physicians. It is critical to understand that providing more comprehensive REI education in residency training is not a substitute for REI fellowship training. REI fellowship programs are central to driving progress and innovation. Lastly, increasing ancillary support staff, such as psychologists, and APPs into the workflow for these is needed to allow the physician to practice at the highest level and care for a greater number patients.

Strengths of this study include our use of available data from various resources, including USCB, SART, NRMP, and individual clinic and fellowship websites. With these data, we were able to highlight the relationship between REI clinic and physicians distribution in the USA in relation to state-mandated fertility coverage and population characteristics, including population density, ethnicity, and household income. Another strength of this study is our use of ArcGIS Online to create a heat map demonstrating the drive time to REI clinics in the country, which provides a visual depiction of the current state of access to REI care. Lastly, we provide several potential solutions to improve access to REI care.

This study does have its limitations. Because our analysis contains both main and satellite clinics, we may be over-estimating the current state of access to REI care in the USA. Specifically, our results show 643 total clinics while prior studies have shown 498 clinics [9]. Additionally, data from the approximate 21% of clinics that do not report their ART cycles to SART was not included in the analysis. Additionally, there are 49 clinics that do not report to either SART or the Centers for Disease Control and Prevention whose data were not included [25]. Thus, it is also possible that we are underestimating the current state of access to REI care in the USA. It is not known if clinics that choose not to report to SART are clustered in specific geographic areas. Furthermore, this study captures only access to care from an infertility standpoint, but there are patients that need REI care for other reasons such as reproductive surgery and reproductive medicine evaluation of conditions such as polycystic ovarian syndrome, premature ovarian failure, and recurrent pregnancy loss. Additionally, there are patients seeking care for fertility preservation or for other reasons, such as pre-implantation genetic testing for monogenic or single-gene disorders. Lastly, it is important to note that our study specifically focuses on individuals who identify as female as those accessing REI care, which is not a comprehensive assessment because there is an increasing volume of gay men and transgender individuals who are in need of fertility services. Thus, we are likely underestimating the need for reproductive services, considering the growth and changes in patient demographics over the last couple of decades.

In summary, access to REI care has large geographic disparities in clinic and physician distribution as well as fellowship training programs. Despite the growing need for ART, the physician pipeline has not been adequately expanded. Collaboration from all stakeholders in the field of reproductive medicine is needed for optimal patient care. The number fellowship of spots at high volume-accredited programs should be increased, and de novo ACGME-accredited fellowship training programs should be created. Additionally, expansion of mandated fertility coverage by insurance companies should be implemented. It is important to note that because racial and ethnic minorities are more likely to be uninsured or underinsured and are more likely to live in urban areas, we are likely overestimating access to this group in particular [26]. Research should be expanded to improve the knowledge gaps, including the optimal ratio of REI physicians to patients, anticipated market growth, and ways to improve access to care for racial and ethnic minorities. Additionally, future research is needed to assess age demographics of current REI physicians, anticipated retirement horizon, geographic distribution including clustering of non-SART reporting clinics, and geographic distribution of recently graduated REI fellows. Further analysis of state-specific distribution of clinics and physicians in the context of patient demographics and health insurance status would also elucidate further disparities in access to care. Patients deserve to be cared for by experts who have all the requisite knowledge, skill, and expertise. The goal should be to increase access to care by the expansion of high-quality training programs and to increase the efficiency of practice.

Data availability

Data can be made available upon request.

Declarations

The authors did not receive support from any organization for the submitted work, and they have no relevant financial or non-financial interests to disclose. This is an observational study. The Human Investigation Committee (IRB) of Emory University has confirmed that no ethical approval is required. Daniela Diego, Heather Hipp, and Jennifer Kawwass contributed to the study conception and design. Material preparation and data collection were performed by Daniela Diego. Data analysis was performed by Lisa Shandley. The first draft of the manuscript was written by Daniela Diego, and all authors commented on the previous versions of the manuscript. All authors read and approved the final manuscript.