The state of Minnesota is not making this easy, intentionally I think. They are trying to minimize to the public how significant the number of breakthrough cases is. Dave Dixon’s notes below further set forth some of the difficulties. The basic problem is that a breakthrough case is not initially identified as such. So the positive test comes in and it gets called a case and gets assigned to the appropriate date in the table of cases by date of specimen collection. Then the state starts running all cases against their vaccinated database (and who knows how accurate that is or complete–what about people vaccinated out-of-state) and when they determine a person was fully vaxed (here is another thing they should report–all cases in anyone who is even partially vaxed), they add that person to the total of breakthrough cases to be reported on the next Monday. Then you have to subtract that total from the prior Monday’s total to know how many new breakthroughs are being reported. But unlike the other cases, you have no idea what day the breakthroughs actually came from. Best estimate, and the state gave a little info on this, is two to four weeks lag. So that is what we use. But I suspect we are underestimating the proportion of cases and other events that are breakthrough. And I also wonder how many cases are in the partially vaxed.
In any event, the charts below are best guess and likely in the rough range, which I again think is likely higher. Here are Dave’s notes:
1. Breakthrough positive cases, hospital admissions, and deaths are published every Monday by the Minnesota Department of Health: https://www.health.state.mn.us/diseases/coronavirus/stats/vbt.html
2. Breakthrough cases are defined as anyone who tests positive for covid 14 days or more after completing their vaccination shots.
3. Breakthrough hospital admissions include any vaccinated person who is admitted to a hospital for any reason, and tests positive for covid up to 14 days prior to admission.
4. Breakthrough deaths are vaccinated people who have a positive covid test and COVID-19 is listed on the death certificate or clinical history provides evidence for COVID-19.
5. DOH also reports the number of vaccinated Minnesotans as of a date 4 weeks prior to the date that breakthrough cases, hospital admissions, and deaths are published.
6. Matching up the breakthrough cases to the total cases in a week that ended 4 weeks prior to the breakthrough report gives strange results. For example, on 8/2/2021 the total announced breakthrough cases were 591 cases higher than the announced total the week before. However, 4 weeks before this date only 506 cases were announced for the week ending 7/5/2021. It is likely that either breakthrough cases are more recent than 4 weeks prior to publication, or they occur in a time period longer than 1 week.
7. News reports in the Star-Tribune have mentioned that breakthrough cases represent approximately 30% of total cases, for example here https://www.startribune.com/vaccine-urged-amid-rise-in-minnesota-breakthrough-infections-mayo-study-says/600094706/
8. Not fully understanding the timing of breakthrough cases we have chosen to present charts assuming that the breakthrough cases occurred up to 14 days prior to the date of publication, and that announced breakthrough cases essentially occur within one week. Charts generated with these assumptions generally match the 30% proportion of breakthrough cases that have been stated in the Star-Tribune.