Improving Health Data Quality in Mozambique
November 7th, 2016 | Viewpoint
November 7th, 2016 | Viewpoint
More than ever before, people living with HIV in Mozambique are accessing the care they need, including Antiretroviral Therapy (ART). In some areas, facilities built to provide basic health services to the community now see three or four times as many clients as usual, most of whom receive ART.
The evidence is in the numbers as well. At the end of 2014, Mozambique had 646,312 people on ART compared with 308,578 at the end of 2012. These numbers could double again based on the current situation, and the Ministry of Health’s (MOH) plan to follow the international test and start strategy and have 80% of eligible people living with HIV on treatment.
While this is undoubtedly excellent news, the challenges that Mozambique’s health system faces in caring for the growing patient load could ultimately jeopardize the initiative’s success. As patients flood the system for HIV care, it becomes more difficult for overburdened clinics to ensure that all patients receive their regular follow-up and ART doses, and to facilitate continuous availability of needed medicines. Both processes are reliant on effective data systems, yet maintaining these is often the lowest priority during a hectic day at the health clinic. Although Mozambique is moving towards a unified health date electronic system, many clinics still rely on manual data collection and recording systems that are prone to error, making data collection and data-based decision making even more challenging.
In 2013, JSI began work on the USAID-funded Mozambique Monitoring Strategic Information Project (M-SIP) to assess the quality of the data collected on six key indicators related to HIV by performing data quality assessments (DQAs) at health facilities. These assessments evaluate data collected at the facility level and compare recorded data to data captured at the national level in order to determine discrepancies and improve overall data quality.
The core of JSI’s strategy—to transfer knowledge on data quality improvement to the MOH—is based on a participatory and capacity-building approach. As a result, the MOH was involved in preparing, implementing, and reporting DQA results. After the first two DQA rounds (2014-2015), the team and stakeholders determined that the MOH would lead the third round with support from JSI staff under M-SIP, as needed.
The results of the two DQA rounds showed striking results at the central, province, and district levels. Not only did the quality of data collected and reported improve at the facilities where a DQA was completed the previous year, but the data also improved at the rest of the province’s facilities, as seen in figure one below. One of the reasons for these improvements is a standard DQA practice: the team performing the assessment always leaves each facility with a written findings and recommendations report. Once all the DQAs in a province are done, the team prepares a comprehensive findings and recommendations presentation for the district and province authorities.
As the above graphic shows, there was an increase in the percentage of health facilities that fell into the “good” category for the ART indicator, demonstrating an improvement in the quality of data collected.
The M-SIP team determined that provinces which had improved data quality during the second DQA round had health managers who prepared action plans based on the results of the first DQA round.
These action plans cannot solve the infrastructure and human resource challenges related to data quality. However, they support the implementation of better filing systems, determine the best data flows within a facility, strengthen communications between the facilities and district and province offices to reduce data inconsistencies, and identify data quality champions within facilities who serve as resources for staff.
The action plans ensured the success of the second DQA round and increased the local leadership’s investment and buy-in for the whole process.
One of the best examples is Cabo Delgado Province which had higher HIV data deviations than the national average in 2014. After the DQA, it achieved a higher improvement on data quality than the national average in 2015 as shown below.
The MOH’s involvement in the process has been invaluable in binding together all components of Mozambique’s health system, emphasizing the importance of data quality and taking ownership of the DQA intervention.
Written by Merce Gasco