Service Penetration and Community Health: A Look at Home Visitation and Prenatal Smoking Cessation

Introduction: Since 2010, there has been substantial federal investment to states for scale-up of maternal and child home visiting services. These services often self-identify as community-based programs and federal law now requires home visiting services be targeted for implementation in high-risk communities as a part of a community-based coordinated system of care.  However, while there is evidence to support the effectiveness of these programs among program recipients, there is a lack of information on the effect of home visiting services on community-level health.  The current context of unprecedented scale-up of home visiting programs presents an opportunity to better understand the relation between penetration of services and community health.  

Methods: Observational study design using a retrospective population-based cohort of women in Pennsylvania who gave birth between 2006 and 2012 and smoked during the first trimester of pregnancy. Community-level penetration of home visiting services was calculated as the proportion of prenatally enrolled  clients served within a given year in a given county to welfare-enrolled women within that county with a birth in the same year, and was categorized into quartiles for analysis. The main outcome measure was smoking cessation in the third trimester as reported on the birth certificate. Multivariable logistic regression analysis examined the association of county-level penetration of home visiting with prenatal smoking cessation, adjusted for temporal trends, socio-demographic factors, and urbanicity, and accommodating the correlation due to clustering of women within counties.

Results: An annual average of 25,000 women in 67 counties were included for analysis. County-level annual penetration ranged from 0.0% to 27.3%, with quartiles of <0.1%, 0.1-1.4%, 1.4-5.5%, and 5.5-27.3%. The adjusted odds of prenatal smoking cessation were significantly increased for women in counties with the highest penetration of home visiting (>5.5%) compared to those with the lowest levels (<0.1%) (OR=1.24; 95% CI: 1.04, 1.47; p=0.015); the predicted probability of prenatal smoking cessation was 21.1% and 17.9%, respectively. In a sensitivity analysis adjusting for individual-level program enrollment, the effect of county-level penetration remained stable and statistically significant.

Conclusions:  County-level penetration of home visiting services is a significant predictor of prenatal smoking cessation, independent of program participation. Service penetration within a community may be an important consideration of scale-up efforts, as increased reach may afford benefits beyond program participants. The mechanisms driving the effect of penetration within communities require further study but may include diffusion of norms or a robust service infrastructure.