From Semera to Seattle…with Real Time Medicine Logistics Data
December 8th, 2016 | Viewpoint
December 8th, 2016 | Viewpoint
On a recent work trip to Seattle, I found myself in a late opening Walgreens pharmacy counter to pick up a prescription for my daughter. A stock out (yes, they happen in the US also!) was mitigated as the helpful pharmacist checked his online inventory system to see if any nearby Walgreens had my item. Sure enough, it was available in another location near the airport, so I could stop there the next morning before my departure.
But he wanted to call first to check. “Oh, you don’t have real time stock data?” I asked. “No,” he explained, “often there are errors, so it’s good to check over the phone.” My pharmacist wasn’t quite sure why the stock level in the other location might not always be accurate. I am assuming they have instant connectivity, that the sales system is integrated with the inventory system, and that staff would never issue or sell an item on paper without doing it in the point-of-sale information system; all issues we deal with in our work on inventory management systems in Ethiopia. Probably any discrepancy in his system is just a matter of a few minutes or at most hours for the system to “update.”
My pharmacist was curious about my interest in how the system worked, so I explained how I work on medicine inventory management systems in Ethiopia. There, our USAID-funded AIDSFree project and Bill & Melinda Gates Foundation-funded work on vaccines have developed information systems (the Health Commodity Management Information System) in use for all health commodities across the public sector healthcare system including the central warehouse, 18 regional warehouses, nearly 600 health centers and (for vaccines) 50 districts. We are currently installing the system in 18 refugee camp pharmacies in border areas, and planning to expand the district level deployment of a mobile version for vaccines to 800 districts.
In Ethiopia, we don’t exactly have real time data either, and the lag for us is usually more than hours, but is often only days. Decision makers at PFSA (the government supply chain agency) have dashboards with close to real time data for thousands of items including ARVs, HIV test kits, condoms, and vaccines. That’s pretty good given connectivity challenges.
Back in my hotel, I checked to see how “live” our data was. In Semera, the capital of the remote Afar region, a two-day drive from Addis, our data was only 2 minutes old; our system syncs and uploads data to a central datamart whenever an internet connection is established. But like my Walgreens pharmacist, I knew our real-time data there probably wasn’t “real time,” even when the connectivity is working.
For us, it’s not just the connectivity but also the process and the people. Warehouse workers have to manually enter their sales/issues (no scanning!) and sometimes they make mistakes. Or the printed forms may stay on a desk for a few days before they are entered. Or maybe the form goes missing. Or they may issue an urgently needed item which hasn’t been actually “received” because of some bureaucratic delay. So our work is not just developing and implementing automated systems, it’s working to understand the underlying business processes and the behavior of the person handling the medicines. Truly it’s a people + process + technology activity. Sometimes we would like to start from scratch and redesign the entire process, but most of the time all we can do is suggest gradual and incremental improvements.
When we first started this work about seven years ago, when customers came to a warehouse in search of their medicines, the pharmacist had to run down to the warehouse to check if they had an item (it took too long to find the stock card). Now they can access the data the same way as my Walgreens pharmacist can. And they can look to see if the stock is available at any of the other warehouses (even if the nearest location might be a day away). The data might be a few days old, but it’s an improvement from the old days of quarterly reports on paper. And as soon as we can we fix the business process and modify behavior, it will truly be real time data…so long as the internet is not down!
And I did get my prescription the next day. My hope is that, thanks to our work, more Ethiopians can get theirs, too.
**This post was originally published on December 7, 2016 on Impatient Optimists.