I feel strongly that we should discuss the what first, before getting distracted by the how. To that end I created a different Talk topic about the how at Technical approaches to measuring OpenMRS impact.
I think these are a great start. And while I would like to eventually get to the clinical impact measures suggested by @jteich, I think we need to crawl before we can walk, and I would suggest focusing on the MVP of just the very broad statistics.
I think that this is useful information for us to know, but I think our KPI should be number of patients records managed in the OpenMRS platform (optionally/eventually: “in a good-quality installation”). For purposes of judging if we are doing well or poorly we shouldn’t be judging that KenyaEMR is better/worse than the OpenMRS Reference Application.
I do think it’s also worthwhile to have a second-priority KPI like “% of known OpenMRS patient records that are stored in a modern OpenMRS version” (meaning it’s a supported version of the OpenMRS platform, rather than an EOL version).
I agree with Burke that we shouldn’t try to get too clever with “active”. I would look at something as simplistic as “# of patient records with an encounter in the last 12 months” (and I would make this lower-priority than “# of patient records”).