School of Public Health
Evaluating facility accessibility and quality as predictors of contraceptive use in Nigeria
Contraceptive use is a key proximate determinant of fertility, enabling women to more effectively control the timing and number of births (Bongaarts, 1982). Although the causal mechanism by which family planning programs influence contraceptive use and fertility intentions continues to be discussed, existing evidence demonstrates that myriad aspects of family planning service delivery are associated with increased likelihood of contraceptive use. Using Performance Monitoring and Accountability 2020 (PMA2020) data, I will work with PMA2020 Baltimore and Nigeria staff to build a multi-level, random effects logistic regression prediction model that can be used to better understand how the family planning service environment – independent of individual and community level factors – can explain differences in contraceptive use in Kaduna, Nigeria. I will link female and household level data to service delivery points via global positioning system (GPS) data, enabling me to develop a prediction model at the individual level. The resulting deliverable will be a draft manuscript that will be submitted for peer review. In addition, I will be working with Nigeria PMA2020 Principal Investigators to generate data analysis training materials in anticipation of leading a 5-day training in country. Using my epidemiologic and biostatistical knowledge to build the analytic capacity of our PMA2020 partners will increase the impact of the PMA2020 initiative, which is only funded to conduct data collection and monitoring; trainings and work related to in-depth data analyses are not within the mission of PMA2020, which is why data are publicly released.
Global Health Project Grant Advisor/Mentor: Dr. Scott Radloff
Many people’s knowledge of Nigeria is limited to what is presented in the media, which highlights the threat of unpredictable violence from the terrorist group Boko Haram. Little to nothing positive is portrayed of Africa’s most populous country, a nation of more than 175 million people comprised of diverse ethnic groups, religions, and languages. I wanted to learn more about this country and its residents, and I wanted to use the knowledge acquired through my schooling at Johns Hopkins and elsewhere to enrich the skills of young researchers in Nigeria.
As a first year PhD student, I worked part-time with the Gates Institute’s Performance, Monitoring and Accountability 2020 project (PMA2020), introducing me to the project’s Nigerian partners. The Nigerian team, including the co-PIs, is an enthusiastic group of individuals who were eager to improve their data management and analysis skills in the statistical software, Stata. As the only two female co-PIs, I was particularly excited at the opportunity to be working with strong African women who have thrived in a predominantly patriarchal society. Receiving a Global Health Field Research Award from the Center for Global Health allowed me to work with these Nigerian colleagues to prepare and implement a 5-day Data Analysis Workshop in Ife, Osun State. Early in the summer, my Nigerian colleagues and I began coordinating the logistics and contacting potential participants to gauge their previous exposure to epidemiology and biostatistics concepts, and their experience with Stata. Understanding the range of proficiency with this material, I was able to prepare an extensive amount of materials, including several PowerPoints, a training dataset, training code, and several training exercises.
Arriving in Lagos, Nigeria at night, I was surrounded by a controlled chaos; throngs of people eagerly awaiting friends’ and family members’ arrival, crowding en masse around the airport exit. After spending the night in Lagos, I left with my driver early in the morning, the city already awake, roads filled with traffic. Once in Ife, I was anxious to finalize training materials and begin the training the following week. I was warmly greeted by our Nigerian co-PI’s colleague, who brought me into his home, introduced me to his family, toured me around Ife and the local university campus, and integrated me into his family’s day-to-day activities. It was an incredibly welcoming experience and a stark contrast to the safety concerns I had been warned of.
After weeks of preparation, my Data Analysis Workshop began on Monday August 3rd at 9:00AM. Thirteen participants, including the two co-PIs, attended the week’s training. The trainees were ready, their attention rapt. They were excited to be learning hands-on skills that would improve their performance at their current job, as well as their future job prospects. Following some initial trepidation, the trainees began asking questions, clarifying concepts, and providing answers to the many questions I posed to the group. I was thrilled with their performance and felt gratified throughout the experience. I found many of them reviewing materials and asking questions in the evenings and early in the morning before the training began. They all expressed appreciation for the opportunity and the materials I shared with them, conveying their discontent with the limited amount of exposure to these concepts they receive at their universities, and the limited formal training in Stata they have access to.
Following the Data Analysis Workshop, I traveled to Kaduna with my Nigerian colleagues to aid in a PMA2020 training. During this training, I continued to field data analysis and data management questions from the Data Analysis Workshop participants while providing broader support to the training of data collectors for the upcoming round of PMA2020 data collection in Kaduna State. During this time, I consulted with our co-PI, a trained demographer with an interest in fertility, family planning, and abortion, to get her input on the specific context of Kaduna and Nigeria at large. These conversations are helping to inform the initial phases of my dissertation, which will use PMA2020 data to estimate abortion rates in Nigeria using model-based approaches. As soon as round 2 data collection is complete in the coming months, I will be pooling round 1 and round 2 data together to generate more precise estimates of the inputs for the models. The Nigerian co-PI is very interested in asking questions about abortion and we have discussed a new method for obtaining abortion estimates from survey respondents that we hope to pilot in a future round of PMA2020 data collection.
Both trainings were highly successful and I was delighted to have had the opportunity to work with such wonderful, bright Nigerians, providing a wholly different and more complete image of Nigeria and its residents than I could have acquired stateside. I am grateful for this experience, which has cemented my desire to continue teaching and training in Africa following graduation and has ignited my dissertation process.