Abstract
Background: Integrating family planning into child immunization services may address unmet need for contraception by offering family planning information and services to postpartum women during routine child immunization visits. However, policies and programs promoting integration are often based on insubstantial or conflicting evidence about its effects on service delivery and health outcomes. Since most studies measure integration as binary (a facility is integrated or not) rather than a multidimensional and varying continuum, it is difficult to understand the determinants and effects of integration. This study creates Facility and Provider Integration Indexes, which measure capacity to support integrated family planning and child immunization services, and applies them to analyze the extent of integration across 400 health facilities. Methods: This study utilizes cross-sectional health facility (N= 400; 49% hospitals, 51% primary healthcare centers) and healthcare provider (N= 1,479) survey data that were collected in six urban areas of Nigeria for the impact evaluation of the Nigerian Urban Reproductive Health Initiative. Principal Components Analysis was used to develop Provider and Facility Integration Indexes that estimate the extent of integration in these health facilities. The Provider Integration Index measures provider skills and practices that support integrated service delivery while the Facility Integration Index measures facility norms that support integrated service delivery. Index scores range from zero (low) to ten (high). Results: Mean Provider Integration Index score is 5.42 (SD 3.10), and mean Facility Integration Index score is 6.22 (SD 2.72). Twenty-three percent of facilities were classified as having low Provider Integration scores, 32% as medium, and 45% as high. Fourteen percent of facilities were classified as having low Facility Integration scores, 38% as medium, and 48% as high. Conclusion: Many facilities in our sample have achieved high levels of integration, while many others have not. Results suggest that measuring integration as a binary variable does not (a) reflect the true variation in integration within and across health facilities, (b) enable nuanced measurement of the determinants or effects of integration, or (c) provide tailored, actionable information about how best to improve integration. Overall, results reinforce the importance of utilizing more nuanced measures of facility-level integration.