After adjustment for confounding factors, patients residing in low-income neighborhoods who underwent IVF treatment were less likely to achieve a LB compared to patients from more affluent neighborhoods. There were no neighborhood socioeconomic indicators associated with total number of IVF cycles per patient or number of IVF cycles necessary to achieve LB.
The discrepancy in IVF success rates that we observed may be attributable to several possible factors. Persons living in more affluent neighborhoods generally have better access to healthcare which also includes better access to fertility treatments. This group is less likely to discontinue IVF treatment after an unsuccessful cycle.[13] It has been shown that patients without insurance, of black race, and with poorer prognoses were more likely to prematurely discontinue treatment, and take longer to resume treatment. However this study also demonstrated that estimated mean family income does not directly correlate with premature treatment discontinuation.[17] In addition, differences in LB rates may be attributable to lifestyle habits. Persons from higher socioeconomic classes tend to engage in more physical activity,[18] have more leisure time, and are more likely to consume a healthy diet.[19] Lower fast food intake and higher fruit and vegetable consumption are associated with a modest reduction in time to achieve pregnancy.[20]
Importantly, we showed that neighborhood level socioeconomic factors were not associated with the number of IVF cycles required to achieve first LB, nor were they associated with total number of IVF cycles that couples underwent. These observations precede the New York State IVF and Fertility Preservation mandate that took effect on Jan 1st 2020. [21] State mandates are shown to increase the use of reproductive technologies, decrease the number of embryo transfer per cycle, decrease the number of transfers resulting in pregnancy and the number of multifetal gestations.[22] Our study findings are consistent with prior report that couples with higher income do not necessarily opt to have more children; in fact, woman’s income is negatively associated with the number of children.[23] The objective of our study is to evaluate socioeconomic factors association with the outcome of LB, since the significance of patient race and ethnicity has been thoroughly studied, nevertheless our findings regarding patient race and ethnicity are mixed, we show that race and ethnicity in our patient population are not associated with LB after IVF, however Hispanic race is associated with the total number of IVF cycles. Numerous prior reports note that race and ethnicity are strongly associated with LB rates after ART. [24]–[28]
Our finding that patients without a documented language preference were more likely to have a LB compared to those who selected English is difficult to interpret. Our centers are equipped with robust resources that provide language translation services. Although statistically significant, this finding is likely attributable to other unknown and unmeasured factors and does not by itself have clinical importance.
Our study has limitations, including a retrospective design. Individual patient socioeconomic characteristics were not evaluated due to inconsistent documentation. Some patients may elect to reside in more socially vulnerable neighborhoods for family or cultural reasons even if their own income level significantly exceeds local averages. Nevertheless, most families in the United States are segregated by income.[29] Our study is limited to New York City and Long Island and our findings may not be generalizable to other regions. We did not account for spatial mismatch between ZIP codes and ZCTAs; in such cases, local socioeconomic conditions may be inaccurate. In addition, some patients may have moved to a new residence between IVF cycles, and we only evaluated the first associated ZIP code. Finally, some patients may have migrated out of the area or pursued further treatment with fertility specialists outside of our health system.
Our study has several strengths, including a large sample size representing a diverse and densely populated area of New York state. Data was obtained from a health system-wide electronic medical record system that reduces inconsistency in documentation and ensures more reliable outcome data. Lastly, our data collection precedes the New York State Insurance Law §§ 3221(k)(6)(C) and 4303(s)(3), which mandate inclusion of IVF coverage in major employer-based insurance policies and may be more generalizable to the majority of states that do not mandate IVF coverage. [21]
In conclusion: Patients residing in lower income neighborhoods have lower LB rates after IVF despite undergoing a similar number of IVF stimulation cycles. Few prior studies have evaluated whether neighborhood-level socioeconomic factors affect access to fertility treatments and subsequent outcomes. Larger population-based studies may better identify opportunities to promote health equity and reduce disparities in access to effective fertility treatments.