Mapathons versus automated feature extraction: a comparative analysis for strengthening immunization microplanning.

BACKGROUND: Social instability and logistical factors like the displacement of vulnerable populations, the difficulty of accessing these populations, and the lack of geographic information for hard-to-reach areas continue to serve as barriers to global essential immunizations (EI). Microplanning, a population-based, healthcare intervention planning method has begun to leverage geographic information system (GIS) technology and geospatial methods to improve the remote identification and mapping of vulnerable populations to ensure inclusion in outreach and immunization services, when feasible. We compare two methods of accomplishing a remote inventory of building locations to assess their accuracy and similarity to currently employed microplan line-lists in the study area. METHODS: The outputs of a crowd-sourced digitization effort, or mapathon, were compared to those of a machine-learning algorithm for digitization, referred to as automatic feature extraction (AFE). The following accuracy assessments were employed to determine the performance of each feature generation method: (1) an agreement analysis of the two methods assessed the occurrence of matches across the two outputs, where agreements were labeled as "befriended" and disagreements as "lonely"; (2) true and false positive percentages of each method were calculated in comparison to satellite imagery; (3) counts of features generated from both the mapathon and AFE were statistically compared to the number of features listed in the microplan line-list for the study area; and (4) population estimates for both feature generation method were determined for every structure identified assuming a total of three households per compound, with each household averaging two adults and 5 children. RESULTS: The mapathon and AFE outputs detected 92,713 and 53,150 features, respectively. A higher proportion (30%) of AFE features were befriended compared with befriended mapathon points (28%). The AFE had a higher true positive rate (90.5%) of identifying structures than the mapathon (84.5%). The difference in the average number of features identified per area between the microplan and mapathon points was larger (t = 3.56) than the microplan and AFE (t = - 2.09) (alpha = 0.05). CONCLUSIONS: Our findings indicate AFE outputs had higher agreement (i.e., befriended), slightly higher likelihood of correctly identifying a structure, and were more similar to the local microplan line-lists than the mapathon outputs. These findings suggest AFE may be more accurate for identifying structures in high-resolution satellite imagery than mapathons. However, they both had their advantages and the ideal method would utilize both methods in tandem.

Postpartum long-acting contraception uptake and service delivery outcomes after a multilevel intervention in Kigali, Rwanda.

INTRODUCTION: Postpartum family planning (PPFP) is critical to reduce maternal-child mortality, abortion and unintended pregnancy. As in most countries, the majority of PP women in Rwanda have an unmet need for PPFP. In particular, increasing use of the highly effective PP long-acting reversible contraceptive (LARC) methods (the intrauterine device (IUD) and implant) is a national priority. We developed a multilevel intervention to increase supply and demand for PPFP services in Kigali, Rwanda. METHODS: We implemented our intervention (which included PPFP promotional counselling for clients, training for providers, and Ministry of Health stakeholder involvement) in six government health facilities from August 2017 to October 2018. While increasing knowledge and uptake of the IUD was a primary objective, all contraceptive method options were discussed and made available. Here, we report a secondary analysis of PP implant uptake and present already published data on PPIUD uptake for reference. RESULTS: Over a 15-month implementation period, 12 068 women received PPFP educational counselling and delivered at a study facility. Of these women, 1252 chose a PP implant (10.4% uptake) and 3372 chose a PPIUD (27.9% uptake). On average providers at our intervention facilities inserted 83.5 PP implants/month and 224.8 PPIUDs/month. Prior to our intervention, 30 PP implants/month and 8 PPIUDs/month were inserted at our selected facilities. Providers reported high ease of LARC insertion, and clients reported minimal insertion anxiety and pain. CONCLUSIONS: PP implant and PPIUD uptake significantly increased after implementation of our multilevel intervention. PPFP methods were well received by clients and providers.

Social vulnerability and county stay-at-home behavior during COVID-19 stay-at-home orders, United States, April 7-April 20, 2020.

PURPOSE: Early COVID-19 mitigation relied on people staying home except for essential trips. The ability to stay home may differ by sociodemographic factors. We analyzed how factors related to social vulnerability impact a community's ability to stay home during a stay-at-home order. METHODS: Using generalized, linear mixed models stratified by stay-at-home order (mandatory or not mandatory), we analyzed county-level stay-at-home behavior (inferred from mobile devices) during a period when a majority of United States counties had stay-at-home orders (April 7-April 20, 2020) with the Centers for Disease Control and Prevention Social Vulnerability Index (CDC SVI). RESULTS: Counties with higher percentages of single-parent households, mobile homes, and persons with lower educational attainment were associated with lower stay-at-home behavior compared with counties with lower respective percentages. Counties with higher unemployment, higher percentages of limited-English-language speakers, and more multi-unit housing were associated with increases in stay-at-home behavior compared with counties with lower respective percentages. Stronger effects were found in counties with mandatory orders. CONCLUSIONS: Sociodemographic factors impact a community's ability to stay home during COVID-19 stay-at-home orders. Communities with higher social vulnerability may have more essential workers without work-from-home options or fewer resources to stay home for extended periods, which may increase risk for COVID-19. Results are useful for tailoring messaging, COVID-19 vaccine delivery, and public health responses to future outbreaks.