Pennsylvania

We noticed a case increase of about 1,500 cases in long-term care this week and wondered if this is due to some data reconciliation? It doesn't seem to fit with the pattern and I was wondering if there's an annotation we should add to explain this increase.

This increase is a result of continued and regular reconciliations of the data to ensure we provide a comprehensive view of the impact of COVID-19 in those facilities.   We have engaged additional staff to formally link these past COVID-19 cases reported during the winter wave to long-term care facilities, which may increase our cumulative long-term care facility case count over the next few weeks. These are not recent cases.

This information about NCHS data is good to have, but I want to clarify that my question is specifically about the data in the CDC's COVID Data Tracker, which is distinct from the NCHS data. For example, for Pennsylvania, the NCHS reports 'All Deaths involving COVID-19' at 22918, while the CDC's COVID Data Tracker reports 'Total number of deaths' at 23,126. (see screenshots attached.)

We understand that there will be differences between the NCHS data and the COVID Data Tracker: Our question is not about these discrepancies. Rather, we're interested in knowing more about the information for Pennsylvania used by the CDC's COVID Data Tracker. Specifically:

1. Where is the CDC getting the information it reports about Pennsylvania from: Does the state report this information to the CDC? Does it come from a different process?

2. In looking at the CDC’s COVID Data Tracker's Cases and Deaths dataset (United States COVID-19 Cases and Deaths by State over Time) we notice that Pennsylvania has values reported for new probable deaths (‘pnew_death’) however no value is reported for probable deaths (‘probdeath’). For example, on 2/15, the value for ‘pnew_death’ was 16 while the field for ‘probdeath’ was blank.

Can you confirm that Pennsylvania is reporting probable deaths information to the CDC? And, can share with us the most up-to-date information you have regarding probable deaths over time?

The data source you are referring to appears to be coming from data submitted each day from our epidemiologists.   The department is using deaths among confirmed cases as the number to report as confirmed deaths. The number of probable deaths shows the number of deaths reported through our Electronic Death Registration System that are not among confirmed cases.   We are submitting this data daily, and are unsure why the probable death column is blank for Pennsylvania.   The best data to use is what the CDC is presenting, since it will change daily.

We have two questions about the information Pennsylvania shares with the CDC regarding COVID-19.

1. In looking at the CDC’s Cases and Deaths dataset (United States COVID-19 Cases and Deaths by State over Time. link: https://data.cdc.gov/Case-Surveillance/United-States-COVID-19-Cases-and-Deaths-by-State-o/9mfq-cb36) we notice that Pennsylvania has values reported for new probable deaths (‘pnew_death’) however no value is reported for probable deaths (‘probdeath’). For example, on 2/15, the value for ‘pnew_death’ was 16 while the field for ‘probdeath’ was blank.

Can you confirm that Pennsylvania is reporting probable deaths information to the CDC? And, can share with us the most up-to-date information you have regarding probable deaths over time?

2. More generally, can you tell us about the different mechanisms of reporting cases and deaths to the federal government (voluntary NNDS, aggregate counts process, and line level reporting from the state health department)? We are particularly interested in the aggregate count process as it appears to be the source of the state-level data published by the federal government regularly.

Hello,   I received the below information from our vital records team:   “NCHS Death Counts are based on death certificates that use information reported by physicians, medical examiners, or coroners in the cause-of-death section of each certificate. Data from each of these pages are considered provisional (not complete and pending verification) and are therefore subject to change. Counts from previous weeks are continually revised as more records are received and processed. Because not all jurisdictions report counts daily, counts may increase at different intervals.”   Here's good reference to share back with the individual to understand how states report death data to NCHS: https://www.cdc.gov/nchs/nvss/covid-19.htm#understanding-the-numbers

Will PA consider publishing facility-level vaccine dose data?

It is something that is planned for the next rollout of our dashboard. Stay tuned!

We are wondering about ICU metrics—if PA might be able to begin providing the number of COVID-19 patients currently in an ICU?

“Yes, we can look to do that ... We can look to make that available.”

Writing to ask if you can provide any further context to DOH's note about today's data. We're wondering whether you can give us a sense of how many cases reported in today's 2,063 count in fact should have been added yesterday. Are we talking, like, 3 or 30 or 300? And should we consider backfilling these to yesterday's count, or would you recommend against that? Really, any context you can provide would help.

We do not know exactly how many, no. It is likely a hundred to a few hundred, but we do not know exactly how many. Those cases would be assigned to yesterday, October 21, but since we do not know exactly how many, we do not recommend backfilling, but rather keeping with the data as reported.

Do you have plans to release total testing values, separately by the total number of unique individuals tested and the total number of specimens tested?

The total number of test results received for the state, as of today, August 24, is 2,116,551. That number is not available anywhere currently, but if you wish to ask for it periodically I can provide it.

Do you perform any deduplication to reach the "Negative" ? If so, how? (i.e. instances swabbed per day/week; recording only one positive test per person, but multiple negative tests for the same person; etc.)? If the same person gets different test results on different days, would you report raw results as part of the daily positives or do you apply any kind of logic such as reporting only the first negative or positive result per individual?

For our positive case counts, we only count an individual once. So, of the 130,035 total cases as of today, they are 130,035 unique individuals. If some has tested negative, they would only count once. However, if we are announcing the total number of negative results, that is the total number of negative test results received.

Will you report testing encounters?

At this time, we do not intend to release data specifically on testing encounters.