Data Series 1: Our Motivation
This series is written by Patrick Aylward, Chief Operating Officer at Integrate Health. To stay up to date on this series, follow our blog and social media accounts on Twitter (@integratehealth), Instagram (@integratehealthglobal), Facebook, and LinkedIn. To hear more about Integrate Health, follow IH’s CEO Jennifer Schechter, Integrate Health
In October, I had the pleasure of shadowing Lare Koutom, a Community Health Worker who supports the Integrated Primary Care Program in the Bassar district of Togo. As we went from house to house, I witnessed Lare engage families and identify, diagnose, and treat very sick children with the utmost professionalism.
In one morning, Lare diagnosed and treated six malaria cases, and it seemed like every other child was affected. During each visit, Lare asked families about their use of bednets, and we learned that while the families generally understood the importance of bednet use, many children weren’t sleeping underneath them because the families didn’t have access due to stock-outs, the bednets were torn or worn out, or families rationed them because they didn’t have enough to cover everyone.
I followed up the next day with our program leadership at Integrate Health and was told that bednet distribution was the responsibility of the Ministry of Health (MOH) and therefore, Integrate Health didn’t track it or follow up on bednet usage. The percentage of children sleeping under a bednet was not included in our list of indicators; instead, we only tracked the number of malaria cases treated.
To me, this represented a few failures:
1. First, as an organization, we weren’t thinking about the entire chain of care delivery. While treating malaria cases is clearly important and saves lives, it also represents a failure to prevent an infection in the first place.
2. Second, our team was taking the perspective of the program rather than the patient, which showed in our indicators. By measuring our own activities, we were potentially missing an opportunity to intervene sooner and strengthen or reinforce the activities of our partners. We might not deliver bednets, but we could assist in notifying our colleagues in the MOH and assisting them with supply chain management, distribution campaigns, or simple advocacy.
3. Finally, there’s the old management cliché, “You don’t manage what you don’t measure.” Because the percentage of children sleeping under a bednet wasn’t in our list of indicators, our team had no visibility on the problem whatsoever.
This experience weighed on me. This fall, Integrate Health launched a concentrated effort to upgrade our data systems and indicators.
While Integrate Health has always used data to inform programs and demonstrate impact, we recognized that as we have grown, our needs have outpaced many of the systems that were developed to support our initial pilot launched in 2015.
Those systems have begun to encounter challenges as we scaled to cover thirteen clinic catchment areas with a population of 140,000 people, and we know that the challenges will only increase as we scale to cover 25 clinic catchment areas and 250,000 people within the next 18 months. In addition, the MOH systems have also developed and changed over the past four years. As a result, many of our indicators are no longer aligned to the definitions, systems, and targets used by our colleagues in the MOH. Given that our ultimate goal is to scale through government adoption, this was problematic.
Recognizing these challenges, Integrate Health has embarked on a journey to upgrade our systems and indicators. We have four objectives:
1. Measure the right things: Make sure that our team is thinking about care from the perspective of the patient and investing in the right steps across the care delivery value chain from prevention to diagnosis, treatment, and the management of complications.
2. Facilitate continuous improvement: Think control charts and trend lines. We believe that continuous improvement tools developed over the last half century can be critical in ensuring that our programs achieve high quality at scale and at a price point that our government partners can afford.
3. Demonstrate impact: We need systems that allow us to easily compare the results of interventions against baselines (e.g. performance before the intervention) and benchmarks (e.g. performance relative to areas without the intervention or national targets). We need to do this with both program data and data collected from our annual household surveys.
4. Embed in the national health system: When we prioritize alignment with the MOH, we can achieve better results together. We’ve learned that the challenge of scaling innovation often lies more with embedding change and best practices in the system rather than developing something truly novel.
Our journey is likely to cover everything from developing a new indicator framework, to automating data preparation, to improving the tools used by our Community Health Workers, while addressing data quality issues throughout. We are embarking on this journey with humility and with the knowledge that our journey is not unique. That is why we decided to document this journey along the way through a series of blog posts in order to solicit feedback and share our lessons learned.
Some topics that we think we’ll cover include the following:
· The challenges of applying a logframe to primary care delivery and why we think we may have a better approach
· The importance (and challenges) of harmonizing indicators with national systems and targets
· How we set indicator targets
· How we used Alteryx to connect directly to Togo’s National District Health Information System (DHIS2) so we could analyze data offline in Tableau…in under a week
· How we’re using data from DHIS2 to benchmark results and demonstrate impact
· And more!
Whether you are an implementer, a funder, or just really interested in the use of data in a highly practical way, we believe this series will provide you with some valuable lessons learned, behind-the-scenes processes, and resources to help you build your data system. We also hope that this series can start a broader conversation around this topic. We, too, seek advice, and we hope that, together, we can learn how to improve our data systems to best serve our communities. If you would like to stay up to date, follow our blog and social media accounts on Twitter (@integratehealth), Instagram (@integratehealthglobal), Facebook, and LinkedIn. @integratehealth
