Over the past year and a half in Benin, VillageReach has supported the Agence de Médecine Préventive (AMP) in their work to pilot and scale Logivac, an informed push distribution system for immunization commodities.
Starting as a pilot in a single health zone in 2014, this system will be deployed in approximately one-third of the 34 health zones nationwide under the name Logivac+ by July, 2016. Working with AMP, VillageReach has helped deploy and adapt OpenLMIS, locally called the Systeme Informatisé d’Information Logistique (SIIL), collecting data and providing the information needed to improve the informed push system. Over the past weeks, VillageReach and AMP have been working to determine program needs and ensure that SIIL is up-to-date as Logivac+ moves to scale.
The informed push system for health products is based on the same principle. Logisticians fill up their truck (often called a “mobile warehouse”) with quantities of medicines or vaccines they predict the health centers will need based on historical and/or population data. Then, they head out and visit each health center, physically count the available stock, remove any products that have expired or gone bad, and fill the shelves or refrigerators back up so that the health center can meet demand until the next visit. Throughout the distribution, the logistician also assists health agents in logistics-related activities, passing along best practice supply chain management lessons to health workers.
In reality, the informed push system for vaccines (and other medical commodities) is a little more complicated than has been described. While the idea of a vaccine vending machine is not a new concept, it’s not available yet. So, how does the logistician know how much of each product to deliver? Where does the information system fit in? What data is used to determine the quantities needed? What if the quantity required is greater or less than what will fit in the vaccine fridge?
Overstocking can lead to losses of expensive, lifesaving products, while understocking results in unmet needs for critical health services. The ideal stock amount can be determined in a number of ways, but for vaccines in particular it is determined using population data. Using this data helps estimate the number of children that a specific health center should vaccinate each month; this calculated amount is called the Ideal Stock Amount, or ISA.
In some cases, however, population estimates may not accurately represent the number of children who actually come to the health center to be vaccinated each month. For example, during the initial pilot phase of the Logivac+ project in Benin, some health centers used up to three times the quantity of vaccines predicted by their ISA!
There are a few reasons why this can happen. First, the census data used for the population estimates may not be extremely accurate. Second – and perhaps more importantly – just because administrators draw lines on maps and assign people to certain health centers to estimate populations doesn’t mean those people will actually go to that health center. The health centers in the Benin example are close to the border with the country of Togo; for people living in that region of Togo, the most convenient health center may be just across the (very porous) border in Benin, causing higher-than-estimated consumption of vaccines at that facility. Another theory is that service quality impacts demand. Mothers and caregivers often prefer to bring their children to the health facilities they know, from experience, are likely to have vaccines in stock, and avoid those that are typically stocked out, or carry what they perceive to be low-quality product. The ultimate goal of the Logivac+ informed push system, and consequently of OpenLMIS, is to increase the availability of high-quality vaccines at the last mile, so the intervention itself could impact demand for vaccinations.
With an information system like OpenLMIS, the logistician can overcome challenges like these to ensure that each health center can meet the actual demand for vaccines. The Logivac+ system is designed not only ensure that enough vaccines are available to meet demand, but also that those vaccines are stored in the right conditions to remain potent and effective for longer. As a platform, OpenLMIS is mobile and works offline, so the logistician is able to enter data about real consumption and losses he observes while he is in the storeroom of a health facility, typically without internet connectivity, rather than relying on paper reports sent to a district office. These calculations are done automatically in real time, so he can assess whether the actual demand the health center experienced in the past month corresponds with the estimated ISA quickly and easily. Using that information, he can make a rapid, informed decision to distribute the right amount of commodities, based on the most up-to-date data.
However, Logivac+ is still in the process of being scaled nationally in Benin. Current national policy for vaccine distribution is based fully on a push system where quantities to be delivered are determined mainly by a population-based ISA instead of real-time consumption data. This type of system, while adequate, can result in misleading data for delivery indicators, further demonstrating the need for a system like OpenLMIS. By empowering logisticians to make informed decisions about the quantities they should deliver through an electronic information system, data quality is improved and verified, and logisticians can make informed, justified decisions.
In Benin, OpenLMIS collects critical supply chain data at the “last mile” of healthcare, where data visibility and accuracy is most important, providing a solid evidence base to inform decision makers on future national policies and support the data needs of Logivac+. When the data used for supply chain decisions matches the actual demand at the health center level, the “right” indicators will drive the distribution system, ensuring more children have access to the vaccines they need, when they need them.
To learn more about the OpenLMIS Initiative and Community, please visit www.openlmis.org, the OpenLMIS Wiki Page, or write to info@openlmis.org.