A Walk in My Shoes: Developing trust in the EMR system through capacity building at the health facility level

Author: Chimango Munthali, MSc, Data.FI Malawi Senior Technical Advisor, November 27 2020 - I'm a curious person with an insistent impetus to inquire and dig. I say it all the time: "Let's find out."

I call this "shoe-leather informatics" - with apologies to shoe-leather epidemiologists who have been out in the field with notebooks and digital tablets for years, digging into the causes of disease outbreaks. This is shoe-leather data inquiry - finding out what is going on behind the numbers in the electronic systems by travelling to sites to learn what the numbers are saying. This is a crucial approach to build more trust in electronic data collection systems, which leads to better use, and ultimately, better data for decision making. I have two examples of this approach in action in my role supporting HIV electronic medical record (EMR) scale-up as a senior technical advisor on the Data.FI project in Malawi.

In the first instance, we saw a discrepancy in the EMR data at a health facility in one of the clinics in the southern part of Malawi. The number of clients newly initiated on antiretroviral treatment (ART) was steady, but the number of people continuing treatment was going down. That meant we were losing people on treatment - our best weapon to stop HIV transmission.

I went to the facility with some staff from the Ministry of Health and an implementing partner monitoring and evaluation (M&E) manager to figure out what was happening.

At the site level, we talked with providers and data handlers and dug deeper into the data. Through these conversations and our exploration of the data we realized that there was a gender discrepancy - it was almost entirely pregnant clients who were not continuing HIV treatment after being initiated.

ART services for pregnant women are chiefly provided in the antenatal care (ANC) clinic, which is supposed to have its own computer for logging ART clients, both new clients and those continuing treatment. Visiting the site, we found out they didn't have a computer, which told me it wasn't a question of more or fewer women on ART, it was a workflow issue. We had to be there on site to discover that. Without a workstation with trained staff, pregnant women on ART were being managed using manual documentation and their data were not entered retrospectively in the EMR system. The EMR data seemed to say that pregnant women were missing ART appointments, when - in fact - it was that attendance at appointments wasn't captured in the electronic system. We developed a workaround so they could capture the data going forward. Now decision makers have an accurate picture of the numbers of people on treatment at that site. To investigate the problem, we had to add some shoe leather to the data.

The second example of shoe-leather informatics has to do with human motivation to use the power of data. It's often a relationship issue, as it was in the case of Malawi's EMR scale-up.

Our implementing partners working at health facilities had been tallying data by hand and then reporting it, despite having the same data in the EMR system. At first, they said they didn't see value in the EMR. In fact, they kept tallying by hand even after the EMR was introduced - a process that could take days at high-volume sites! After the hand tally, they then had to spend time entering the data into the electronic system. That double-entry was overburdening them and didn't allow them time to look at their data and understand trends to support performance improvement. My job was to support the implementing partners in developing trust in the EMR system through capacity building on EMR use, data access and cleaning, and data use for reporting and programming at the facility level.

Working closely with M&E leads of implementing partners, we quickly found discrepancies in the data being entered into the EMRs, like the example at the ANC clinic, and worked one-on-one with members of the facility-based teams to address those. We also took the opportunity to further explain how use of the EMR system could lessen their workload and allow them to better pinpoint performance issues. I think a lot of the shift in providers' confidence in the EMR system was thanks to us working side by side with them and providing a helping hand on how to use these new systems.

I mentored partners and health facility staff on how to access EMR data, validate it, and use it for both reporting and programming. I made myself accessible by phone and email for any help and kept on checking on selected users with major challenges to help them turn the corner. They appreciated that, and when we visited their site, they would say, "We have this issue, what should we do?" Or they would solve a problem and ask us, "What do you think?" Where they expected judgment, they saw help and their trust grew, as they felt that we were not checking up on them but were empowering them.

Over time, across partners and sites, we started talking one language or 'jumping into one car,' so to speak - focusing on the same things. The partners are independently doing a lot around EMRs and Data.FI continues to augment the partners' efforts to institutionalize the EMRs in their supported facilities. When it comes to the EMR, facility teams cannot run away from it anymore. They see the value of these systems - it has become so key.

I don't emphasize having off-site training session on EMRs. I visit a facility and I ask, "What do you want to achieve?" We figure out how to do it together, and then the next day we demonstrate it. This is hands-on and more effective, I think. Shoe leather.

As with all of the blogs posted on our website, the content above does not imply the endorsement of The CI or its Partners and is from the perspective of the writer alone. We do not check facts and strive to retain the writer's voice, as is detailed in our Editorial Policy.