Field Note: It Takes Two: The People Behind The Data

They awake before dawn, rolling out of sleeping bags thrown onto the floor of the elementary school, stretching and sipping tea, organizing the questionnaires and pens for the day’s work.  They arrived yesterday evening, after an almost 20 mile hike along the unpaved roads and wading across a river, carrying satchels containing several 60 page questionnaires, 14 lbs of specialized scales, and other equipment. They had to hike because the roads were impassable by vehicle—at this time at the close of the rainy season, the roads are muddy swamps, so rutted that even the Land Rovers can’t get across safely—or because the villages they are trying to reach are so remote, there ARE no roads. Today, their day starts at 5:30, as they have to conduct their first interview at 6 am so that the household can participate before the mom and dad head out to the rice paddy where they farm. Twenty fit, strong Malagasy men and women, of varying ages, make up 5 teams organized and directed by INSTAT (the Madagascar National Institute of Statistics). They will be collecting health and economic data from 1600 households in Ifanadiana district, Southeastern Madagascar. They are impressive. They are tired. But they are continuing on, because they are professionals. This is what they do.

The data they are collecting will be taken to the capital, entered into a database and used to provide a realistic picture of what the current health and economic conditions are in the district.  What proportion of pregnant women receive antenatal care? What proportion of the children have had vaccines? When a child is sick, where is he taken for care? What proportion of people in the household are currently unable to work because of illness or injury? And we hope to be able to show improved access to care after PIVOT’s work has gotten underway.

As an epidemiologist, sitting at my desk in Boston, I spend a lot of time immersed in grouped data. Proportions, means, regression analyses. From time to time, I remind myself to stop and think about the people they represent. I am not often privileged to see names or faces, but it helps to think that if our vaccination rates for children under age 2 increase from 60% to 80%, that may represent 500 more children that got vaccinated against polio, measles, rubella. I can see that this child on the screen in front of me, who got vaccinated this year, has 2 siblings. His mother lost another, older child. His family has 2 bednets, and he and his little sister slept under them last night, to protect them from the mosquitoes carrying malaria. They have a radio, but no TV, no refrigerator, and it takes his mother 45 minutes to get water for the family each day.

So, when I look at these data, I remind myself to stop and consider the individuals it represents, the people sitting patiently in their homes, answering question after question. But it is not often that I remember that each the data represent two sets of individuals—those answering the questions at 6 am in their homes, and those who have hiked 20 miles and arisen at 5 am to ask them. We thank both of them.

______________________________

En français:

Ils se réveillent avant l’aube, le déploiement des sacs de couchage jetés sur le sol de l’école élémentaire, d’étirement et en sirotant un thé, l’organisation des questionnaires et des stylos pour le travail de la journée. Ils sont arrivés hier soir, après une hausse de près de 20 mile le long des routes non revêtues et patauger dans une rivière, en portant des cartables contenant plusieurs questionnaires 60 pages, 6.5 kilos de balances spécialisées, et d’autres équipements. Ils avaient de la randonnée parce que les routes étaient impraticables par véhicule à ce moment, à la fin de la saison des pluies, les routes sont marais boueux, donc défoncée que même les Land Rover peut pas traverser en toute sécurité ou parce que les villages qu’ils essaient atteindre sont si éloignées, il n’y a pas de routes. Aujourd’hui, la journée commence à 5h30, car ils doivent effectuer leur première entrevue à 6h du matin afin que le ménage peut participer avant la mère et le père partent vers la rizière où ils agricole. Vingt hommes et femmes malgaches forts, de différents âges, représentent 5 équipes organisé et supervisé par l‘INSTAT. Ils recueilleront des données sanitaires et économiques de 1600 ménages dans le District de Ifanadiana, sud-est de Madagascar. Ils sont impressionnants. Ils sont fatigués. Mais ils continuent, parce que ce sont des professionnels. C’est leur travail.

Les données qu’ils recueillent seront prises pour la capitale, a conclu une base de données et utilisée pour fournir une image réaliste de ce que l’état de santé actuel et les conditions économiques sont dans le district. Quelle est la proportion de femmes enceintes reçoivent des soins prénatals? Quelle est la proportion des enfants ont eu les vaccins? Quand un enfant est malade, où est-il pris pour les soins? Quelle est la proportion de personnes dans le ménage sont actuellement incapables de travailler en raison de maladie ou de blessure? Et nous espérons être en mesure de démontrer l’amélioration de l’accès aux soins après le travail de PIVOT a obtenu en cours.

Parce que je suis un épidémiologiste, assis à mon bureau à Boston, je passe beaucoup de temps plongé dans des données groupées. Proportions, des moyens, des analyses de régression. De temps en temps, je me souviens d’arrêter et de penser aux gens qu’ils représentent. Je ne suis pas souvent le privilège de voir des noms ou des visages, mais cela m’aide à penser que si nos taux de vaccination pour les enfants de moins de 2 ans augmentation de 60% à 80%, qui peuvent représenter 500 autres enfants qui s’étaient fait vacciner contre la polio, la rougeole, la rubéole.  Je peux voir que cet enfant sur l’écran en face de moi, qui s’étaient fait vacciner cette année, a 2 frères et sœurs. Sa mère a perdu un autre enfant, plus âgé. Sa famille dispose de 2 moustiquaires, et lui et sa petite sœur dormait sous eux la nuit dernière, pour les protéger contre les moustiques qui peuvent transmettre du paludisme. Ils ont une radio, mais pas de télévision, pas de réfrigérateur, et il faut que sa mère de 45 minutes pour obtenir de l’eau pour la famille chaque jour.

Donc, quand je regarde ces données, je me rappelle d’arrêter et de considérer les personnes qu’il représente, les gens assis patiemment dans leurs maisons, répondant à la question après question. Mais ce n’est pas souvent que je me souviens que chaque les données représentent deux ensembles d’individus- les répondre aux questions à 06 heures dans leurs maisons, et ceux qui ont parcouru 20 miles et posé à 05 heures pour leur demander. Nous remercions tous les deux.

 



    • Sign up for our newsletter
      Stay up to date on Pivot’s latest news

    • Molecular Diagnostics

      Pivot has partnered with the Ministry of Public Health and Centre ValBio to develop the first molecular diagnostics laboratory for COVID-19 outside of the capital city.  We additionally have partnered with the Pasteur Institute of Madagascar on novel analysis of dried blood spots from I-HOPE survey for serological analysis of measles, malaria, schistosomiasis, Hep B, and COVID-19.

       

      Learn more: 

      Integrating Health Systems and Science to Respond to COVID-19 in a Model District of Madagascar, Rakotonanahary, R.J.L., et al., 2021, Frontiers in Public Health

      Reconciling model predictions with low reported cases of COVID-19 in Sub-Saharan Africa: Insights from Madagascar, Evans, M.V., et al., 2020, Global Health Action

    • Eco-Epidemiological Surveillance

      We collate environmental surveillance data (e.g. vector surveillance) with biomarkers from household surveys (e.g. rapid tests and antibody tests from dried blood spots of participants in the I-HOPE household survey) to inform the spatio-temporal dynamics of infectious diseases.

       

      Learn more:

      Estimating the local spatio‐temporal distribution of malaria from routine health information systems in areas of low health care access and reporting, Hyde, E, et al., 2021, International Journal of Health Geographics

      Reconciling model predictions with low reported cases of COVID-19 in Sub-Saharan Africa: Insights from Madagascar, Evans, M.V., et al., 2020, Global Health Action

    • GIS (geographical)

      We use a participatory approach to map over 20,000 kilometers of footpaths and 100,000 structures. These data have been combined with high resolution data on land cover, a digital elevation model, rainfall data, and geolocated data from the IHOPE cohort (see above). These data are used to determine travel times to health care, measure geographic equity, and study geospatial determinants of disease. Accessibility modeling results are available on an e-health platform developed with R Shiny.

    • IHOPE (household-level longitudinal cohort)

      The Ifanadiana Health Outcomes and Prosperity longitudinal Evaluation (IHOPE) cohort was established in 2014 at the start of Pivot’s work in Ifanadiana District. Modeled after Demographic and Health Surveys, it tracks standard international health, demographic, and socioeconomic indicators. IHOPE has the following combination of characteristics that are novel for localized interventions:

      • True baseline
      • True representative sample
      • Sample from inside and outside of the initial catchment population
      • Collection by third party professionals at the National Institute of Statistics who collect identical data nationally
      • Tracks same individuals over time
      • Includes biomarkers such as dried blood spots used for molecular analyses

       

      Learn more: 

      Baseline Population Health Conditions Ahead of a Health System Strengthening Program in Rural Madagascar, Miller, A., et al. 2017, Global Health Action

    • HMIS and Routine M&E

      Health management information systems (HMIS) collect information on health system utilization and care provision at public sector facilities. These data are combined with additional, routine monitoring and evaluation (M&E) data to track over 1000 indicators of health system performance in Ifanadiana District, including rates of treatment and service utilization, quality of care, supply stock-outs, and human resource capacity at every level. These data are accessible via a real-time dashboard.

    • Capacity-Building

      The purpose of Pivot Science is to improve health outcomes over the long-term. Central to this goal is a new capacity building program, which focuses on both increasing research skills among clinicians/implementers, as well as helping researchers better understand and inform clinical priorities. The training includes a series of workshops on understanding clinical priorities, developing research questions, methodology, and dissemination. Trainees include a range of personnel who work within Madagascar’s health sector.

    • COVID-19 and Molecular Diagnostics

      The majority of deaths in Madagascar are due to infectious diseases and most cases go undiagnosed. The COVID-19 pandemic has underscored the need to implement new diagnostic assays in rural health districts. Pivot has partnered with the Ministry of Public Health and Centre ValBio to develop the first molecular diagnostics laboratory for COVID-19 outside of the capital city, which provides both diagnostic capacity as well as a platform for scientific innovation at the intersection of biomedicine and planetary health.

       

      Learn more: 

      Integrating Health Systems and Science to Respond to COVID-19 in a Model District of Madagascar, Rakotonanahary, R.J.L., et al., 2021, Frontiers in Public Health

      Reconciling model predictions with low reported cases of COVID-19 in Sub-Saharan Africa: Insights from Madagascar, Evans, M.V., et al., 2020, Global Health Action

    • Eco-Epidemiology & Surveillance

      There have been major advances in the science of infectious disease dynamics. But there is inadequate application of these advances at local scales to inform health interventions. We combine environmental information, spatially granular health system data, and population surveys, with mathematical models to understand and forecast local disease dynamics (such as malaria, measles, schistosomiasis, lymphatic filariasis and diarrheal disease) to improve service delivery. 

       

      Learn more:

      Estimating the local spatio‐temporal distribution of malaria from routine health information systems in areas of low health care access and reporting, Hyde, E, et al., 2021, International Journal of Health Geographics

      Reconciling model predictions with low reported cases of COVID-19 in Sub-Saharan Africa: Insights from Madagascar, Evans, M.V., et al., 2020, Global Health Action

      Towards elimination of lymphatic filariasis in southeastern Madagascar: Successes and challenges for interrupting transmission, Garchitorena, A., et al., 2018, PLOS Neglected Tropical Diseases

    • Operational Research for UHC

      Operational research provides insights into how programs are implemented with an emphasis on quality and fidelity to organizational, national, and international standards. Our operational research priority areas include UHC financing, health care quality, patient satisfaction, and health worker performance. This is a high priority area for growth across clinical and data teams at Pivot. 

       

      Learn more:

      Evaluation of a novel approach to community health care delivery in Ifanadiana District, Madagascar, Razafinjato, B., et al., 2020, medRxiv

      Rapid response to a measles outbreak in Ifanadiana District, Madagascar, Finnegan, K.E., et al., 2020 medRxiv

      Networks of Care in Rural Madagascar for Achieving Universal Health Coverage in Ifanadiana District, Cordier, L.F., 2020, Health Systems & Reform

      In Madagascar, Use Of Health Care Services Increased When Fees Were Removed: Lessons For Universal Health Coverage, Garchitorena, A., et al., 2017, Health Affairs

    • Geography & Community Health

      We are advancing new methods that combine granular health system data with a massive GIS dataset containing over 100,000 structures and 15,000 miles of footpaths in the district. This is used to identify geographic barriers and improve the design of the health system to reach everyone. To overcome geographic barriers, we have piloted a model of proactive community health for Madagascar, and are evaluating its impact and feasibility.

       

      Learn more:

      Estimating the local spatio‐temporal distribution of malaria from routine health information systems in areas of low health care access and reporting, Hyde, E, et al., 2021, International Journal of Health Geographics

      Improving geographical accessibility modeling for operational use by local health actors, Ihantamalala, F.A, et al., 2020, International Journal of Health Geographics

      Evaluation of a novel approach to community health care delivery in Ifanadiana District, Madagascar, Razafinjato, B., et al., 2020, medRxiv

    • Population-Level Impact Evaluation

      We measure the impact of our work through the analysis of a district-representative longitudinal cohort. Our quasi-experimental design allows for some of the most rigorously evaluated analysis of health systems change on population health in Africa. These analyses show improvements in nearly every major health indicator, including infant mortality, under-five mortality, vaccine coverage, access to and quality of health care services, and health equity.

       

      Learn more:

      District-level health system strengthening for universal health coverage: evidence from a longitudinal cohort study in rural Madagascar, 2014-2018, Garchitorena, A., et al. 2020, BMJ Global Health

      Early changes in intervention coverage and mortality rates following the implementation of an integrated health system intervention in Madagascar, Garchitorena, A., et al. 2018, BMJ Global Health

      Assessing trends in the content of maternal and child care following a health system strengthening initiative in rural Madagascar: A longitudinal cohort study, Ezran, C., et al. 2019, PLOS Medicine

      Baseline Population Health Conditions Ahead of a Health System Strengthening Program in Rural Madagascar, Miller, A., et al. 2017, Global Health Action

    • Data Systems

      PIVOT is fully integrated with Madagascar’s Health Management information System. By combining many kinds of data – outcomes, programmatic, geographic and more – we gain insights to inform our approaches in an ongoing improvement cycle.

    • Quality of Care

      Care is of no use to our patients’ health unless it is high-quality. We focus on bringing the best possible care to all levels of the health system, whether offered at a patient’s doorstep, at health centers, or at the hospital.

    • Supply Chain & Equipment

      Constant collaboration and integration with Madagascar’s national supply chain helps maintain adequate stocks of more than 40 essential medicines and supplies across all levels of care while steadily closing the gaps on stockout rates and saving lives.

    • Finance

      PIVOT is working with the government’s National Health Solidarity Fund to create a transparent system for patients, providers, donors, and government officials, building on success in removing financial barriers to care through patient reimbursements.

    • HR & Recruitment

      Skilled, well trained and compassionate people are the most vital components of any health system. To meet the unusual challenge of a setting where posting may be truly remote, PIVOT teamed up with the Ministry of Public Health to design and implement a joint recruitment and retention strategy.

    • Infrastructure

      A public health system needs dignified and durable spaces – a difficult task in a mountainous rainforest environment. The district’s most remote facilities are a priority, where partnership with local communities and contractors ensure that rehabilitated spaces are maintained over time.

    • Emergency Transport

      74% of the people in Ifanadiana District live more than a 5-kilometer walk to the nearest health center. Our district-wide public ambulance referral system is the only one of its kind in Madagascar, operating 24/7 since 2014 to bring urgent cases in for treatment at no cost to the patient.

    • Patient Accompaniment

      PIVOT accompagnateurs welcome people into a system that may be unfamiliar. They explain the process, assist in navigating the system,address the need for food and lodging when needed, and check on the kids back home. This service, along with the improved availability and quality of care, has helped quadruple the use of outpatient health services in Pivot’s catchment area.

    • Maternal Health

      Our programs successfully address high rates of maternal deaths among the women in Ifanadiana District, who give birth an average of seven times during their reproductive lifetime, with four in every five deliveries occurring at home. In the first two years of fully supported obstetric services, access to family planning, and facility-based deliveries, the maternal mortality rate dropped by 20%.

    • Malnutrition

      More than half of the children under 5 in Madagascar are chronically malnourished. We are piloting a national program in Ifanadiana District that combines screening, treatment, and prevention across all levels of the health system to address this major cause of child mortality.

    • Tuberculosis

      TB is so prevalent in Madagascar that there are an estimated 500 new cases in our district every year, most undetected and untreated. In partnership with the National Tuberculosis Program,  we launched a program to control TB in 2017 to upgrade basic resources and capacity, and to ensure that diagnosis and treatment are available in Ifanadiana District.

    • Child Health

      Children under 5 are those most likely to die from preventable causes like malaria, pneumonia and diarrhea. We implement protocols for the Integrated Management of Childhood Illness to guide health workers in diagnosis and treatment and to assess nutrition and vaccine status in low-resource settings.