Picture of the old DataPall Home Screen (A). Our updates have slightly altered this, but the visual appearance is pretty similar.
If I had to point to a single encompassing experience for our time at St. Gabriel’s, it would definitely be palliative care data collection. DataPall, as I mentioned earlier, began as an on-site project by a group of interns in 2012. The expansive need for palliative care, spanning from high burdens of HIV/AIDS and Cancer, and the cumbersome method of manual recording, inspired the students to create a centralized database. With the goals of user-friendliness and efficiency, they used Microsoft Access to provide a platform to chronicle patient appointment, diagnosis, symptoms, medications, and considerable other information. (Note: DataPall can be found on SourceForge, if you’re interested in checking it out.) Even today, the palliative care staff truly appreciates the innovation. As government reporting is a major impetus for data collection, this system, with its simple reporting functions, saves tremendous time! As the interns did before us, we were tasked to understand and continue system updates based on user and personal experience. A pervasive lesson in global health initiatives is that implementation is never quick or simple; instead, it requires long-term commitment and understanding to attain its real influence. We were the third group to experience this lesson firsthand.
CORRECT data collection is VERY hard
The current methods of collecting and recording palliative care data allow for many errors. The potential for mishaps begins at the onset of the doctor-patient interaction. The nurse/clinician handwrites the patient information, such as name, village, diagnosis, etc., in a giant journal. Sometimes this data can be copied from the patients’ health passport, but often the patient must verbally provide such information. Error 1. Verbal communication – spelling spoken words is not easy. Error 2. Often a patient doesn’t know his exact birthday or name’s spelling. Then the nurse, once all the outpatient visits are completed, comes to the DataPall office to retroactively input the data. However, often the scribe and reader are different. Error 3. Illegibility. Moreover, DataPall allows for significant typing. Error 4. Unplanned Typos. Together, all of these errors, had led to large amounts of duplicate records and misspellings of common diagnoses/drugs. While drop-down menus were used, they did not provide all the options necessary to be singularly effective. Thus, human input was greatened, as they typed the diagnosis, drug, etc. in the “Other” blank. Overall, the data was not as correct as it could be.
Our first task was to improve the drop down menus and reduce the amount of typing required. Our reasoning was less typing = less errors. Joao created a program to go through all the data (over 6,000 appointments and 1,500 patients) to find the common misspellings and “other” inputs. Then, we manually corrected the spelling errors and added the additional options to the drop down features. We also converted certain original typing features into drop-down/click menus. We hope that these corrections will make the data more useful. Having consist methods of inputting a certain diagnosis or drug will allow the program to lump the total and report correct amounts, which were otherwise lost with misspellings and variable spellings.
Second, we compiled the duplicate records. Joao created an algorithm based on name, village, age, diagnosis, etc. to identify possible repeat patients. Then, we went through and ensured the matches. This step concluded the “cleaning up” phase. Like the drop-down menus, we wanted to create an on-going method to prevent future errors. This includes a new, highly technical searching method (so glad to have Joao’s programming skills), which will alert the inputter in an efficient/comprehensive manner to possible matches.
Our final step to completing these updates is better tailoring the reports to the needs of the hospital. This is something we will soon work on. We want to move beyond just numbers and incorporate visual diagrams to convey this information.
As you can tell, our work isn’t necessarily the most glorious, but we believe it can go a long way in improving care and reporting. Proper data goes beyond just analysis. The hospital is required to report such values to the Ministry of Health. This allows the government as well as the hospital to better allocate and request resources. We are happy to assist in this process.
Beyond DataPall
Our time in the Palliative Care ward opened us to another great opportunity.to improve data collection and resource allocation – an electronic medical record focused on monitoring Morphine usage and stock. In fact, today, we had a chance to briefly showcase our project to two government officials including the national coordinator for morphine distribution. There is considerable enthusiasm and we are truly excited for its impact not only at St. Gabriel’s, but Malawi at-large. As the end draws closer, our list of things to complete seems to incessantly rise. However, we couldn’t ask for it another way; we are honored to contribute in a lasting manner. More about this and my adventures outsourcing the Ob-Gyn Stirrups to local companies next time!