Associations between unstable housing, obstetric outcomes, and perinatal health care utilization

Abstract

BACKGROUND: While there is a growing interest in addressing social determinants of health in clinical settings, there are limited data on the relationship between unstable housing and both obstetric outcomes and health care utilization. OBJECTIVE: The objective of the study was to investigate the relationship between unstable housing, obstetric outcomes, and health care utilization after birth. STUDY DESIGN: This was a retrospective cohort study. Data were drawn from a database of liveborn neonates linked to their mothers' hospital discharge records (2007e2012) maintained by the California Office of Statewide Health Planning and Development. The analytic sample included singleton pregnancies with both maternal and infant data available, restricted to births between the gestational age of 20 and 44 weeks, who presented at a hospital that documented at least 1 woman as having unstable housing using the International Classification of Diseases, ninth edition, codes (n ¼ 2,898,035). Infants with chromosomal abnormalities and major birth defects were excluded. Women with unstable housing (lack of housing or inadequate housing) were identified using International Classification of Diseases, ninth edition, codes from clinical records. Outcomes of interest included preterm birth (<37 weeks' gestational age), early term birth (37e38 weeks gestational age), preterm labor, preeclampsia, chorioamnionitis, small for gestational age, long birth hospitalization length of stay after delivery (vaginal birth, >2 days; cesarean delivery, >4 days), emergency department visit within 3 months and 1 year after delivery, and readmission within 3 months and 1 year after delivery. We used exact propensity score matching without replacement to select a reference population to compare with the sample of women with unstable housing using a one-to-one ratio, matching for maternal age, race/ethnicity, parity, prior preterm birth, body mass index, tobacco use during pregnancy, drug/alcohol abuse during pregnancy, hypertension, diabetes, mental health condition during pregnancy, adequacy of prenatal care, education, and type of hospital. Odds of an adverse obstetric outcome were estimated using logistic regression. RESULTS: Of 2794 women with unstable housing identified, 83.0% (n ¼ 2318) had an exact propensity scoreematched control. Women with an unstable housing code had higher odds of preterm birth (odds ratio, 1.2, 95% confidence interval, 1.0e1.4, P < .05), preterm labor (odds ratio, 1.4, 95% confidence interval, 1.2e1.6, P < .001), long length of stay (odds ratio, 1.6, 95% confidence interval, 1.4e1.8, P < .001), emergency department visits within 3 months (odds ratio, 2.4, 95% confidence interval, 2.1e2.8, P < .001) and 1 year after birth (odds ratio, 2.7, 95% confidence interval, 2.4e3.0, P < .001), and readmission within 3 months (odds ratio, 2.7, 95% confidence interval, 2.2e3.4, P < .0014) and 1 year after birth (odds ratio, 2.6, 95% confidence interval, 2.2e3.0, P < .001). CONCLUSION: Unstable housing documentation is associated wi...

Housing instability and adverse perinatal outcomes: a systematic review

“…22,28 Housing instability or homelessness was consistently associated with an increased odds of preterm birth in multiple studies, with increased adjusted odds ranging from 20% to 43%. 1,22,29,32 One study found no association, but this study had a limited sample size with less than 20 individuals in each group experiencing preterm birth and was rated very low GRADE and poor internal validity. 31 One study found that severe housing instability was associated with a 73% greater odds of low birthweight and/or preterm birth (aOR 1.73, 95% CI 1.28-2.32).…”

“…The screening process yielded 14 studies meeting inclusion criteria (Figure 1). 1,2,[21][22][23][24][25][26][27][28][29][30][31][32] All studies were observational; studies were either retrospective cohort studies (n=10, 71.4%), cross-sectional observational studies (n=3, 21.4%), or prospective cohort studies (n=1, 7.1%). (Table 1)…”

“…The majority of studies included individual race, age, mental health history, or substance use as covariates in adjusted logistic regression models. 1,2,18,21,23,25,26,29,30,32…”

“…12,13 Social determinants of health are known to influence perinatal morbidity. 14,15 With regard to housing, pregnant individuals who report housing instability are more likely to self-identify as non-Hispanic Black, be younger, less educated, and low income compared to those with stable housing; [16][17][18] all of these sociodemographic factors have been associated with adverse pregnancy outcomes. 14…”

Housing Intervention For Medically Complex Families Associated With Improved Family Health: Pilot Randomized Trial

“…10 Studies have also documented increased health care use and costs associated with housing instability, homelessness, or both. [11][12][13] Strategies to prevent homelessness and reduce housing instability may improve health and reduce avoidable health care use. 12 Innovative solutions that provide health and housing resources, commonly known as sup-portive housing, may be essential for improving population health.…”