IMPORTANCE Hospital-based obstetric services have decreased in rural US counties, but whether this has been associated with changes in birth location and outcomes is unknown. OBJECTIVE To examine the relationship between loss of hospital-based obstetric services and location of childbirth and birth outcomes in rural counties. DESIGN, SETTING, AND PARTICIPANTS A retrospective cohort study, using county-level regression models in an annual interrupted time series approach. Births occurring from 2004 to 2014 in rural US counties were identified using birth certificates linked to American Hospital Association Annual Surveys. Participants included 4 941 387 births in all 1086 rural counties with hospital-based obstetric services in 2004. EXPOSURES Loss of hospital-based obstetric services in the county of maternal residence, stratified by adjacency to urban areas. MAIN OUTCOMES AND MEASURES Primary outcomeswere county rates of (1) out-of-hospital births; (2) births in hospitals without obstetric units; and (3) preterm births (<37 weeks' gestation). RESULTS Between 2004 and 2014, 179 rural counties lost hospital-based obstetric services. Of the 4 941 387 births studied, the mean (SD) maternal age was 26.2 (5.8) years. A mean (SD) of 75.9%(23.2%) of women who gave birth were non-Hispanic white, and 49.7% (15.6%) were college graduates. Rural counties not adjacent to urban areas that lost hospital-based obstetric services had significant increases in out-of-hospital births (0.70 percentage points [95%CI, 0.30 to 1.10]); births in a hospital without an obstetric unit (3.06 percentage points [95%CI, 2.66 to 3.46]); and preterm births (0.67 percentage points [95% CI, 0.02 to 1.33]), in the year after loss of services, compared with those with continual obstetric services. Rural counties adjacent to urban areas that lost hospital-based obstetric services also had significant increases in births in a hospital without obstetric services (1.80 percentage points [95%CI, 1.55 to 2.05]) in the year after loss of services, compared with those with continual obstetric services, and this was followed by a decreasing trend (-0.19 percentage points per year [95%CI,-0.25 to-0.14]). CONCLUSIONS AND RELEVANCE In rural US counties not adjacent to urban areas, loss of hospital-based obstetric services, compared with counties with continual services, was associated with increases in out-of-hospital and preterm births and births in hospitals without obstetric units in the following year; the latter also occurred in urban-adjacent counties. These findings may inform planning and policy regarding rural obstetric services.
|Original language||English (US)|
|Number of pages||9|
|Journal||JAMA - Journal of the American Medical Association|
|State||Published - Mar 27 2018|
Bibliographical noteFunding Information:
completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Kozhimannil reported receiving grants from the Federal Office of Rural Health Policy, Health Resources and Services Administration, Department of Health and Human Services. No other disclosures were reported.
article was supported by the Federal Office of Rural Health Policy, Health Resources and Services Administration, Department of Health and Human Services (cooperative agreement No. U1CRH03717-13-00).
Additional Contributions: We gratefully acknowledge the assistance of Alex Evenson, MA, in the preparation of the figures and editing of this manuscript; Alexandra Ecklund, MPH, in reviewing and editing the manuscript; and Ira Moscovice, PhD, for the guidance, comments, and input provided. Dr Moscovice, Ms Ecklund, and Mr Evenson are affiliated with the Rural Health Research Center, University of Minnesota School of Public Health and received salary support from the above-named grant (cooperative agreement No. U1CRH03717-13-00). We also thank the mothers, clinicians, clinic administrators, and staff in a rural community that lost hospital-based obstetric services for the insights provided during focus group discussions, which helped contextualize study findings.