sacundim
commented
4 years ago While this introduces some deviation from the reporting elsewhere in the US, we include Puerto Rico's probable cases as part of our count in line with the Departamento de Salud's reporting.
Your practice of summing the confirmed and probable case numbers is not in line with Puerto Rico's Department of Health' reporting, which discontinued the practice of reporting such sums on June 10th:
The Department of Health will now report confirmed and probable COVID-19 cases
The Department of Health's secretary, Lorenzo González Feliciano, informed today, Wednesday, that starting tomorrow, Thursday, the agency will report confirmed cases and probable cases of COVID-19, according to guidelines established by the Council for State and Territorial Epidemiologists.
And the actual implementation of the change, as can be from the front pages I attach below, was that daily case reports changed from listing these categories:
- Unique positive cases [sum of molecular + serological]
- By molecular test
- By serological test
...to these:
- Confirmed cases
- Probable cases
...which eliminates the sums thereof that you report. Along with this, also, a number of summary tables and graphics and data tables that were based on such sums were also eliminated from the daily reports.
It is notable as well that the PRDoH appealed to the CSTE's statement to justify its elimination of the summed figures and content based on them and emphasize its distinction between confirmed and probable cases, whereas you appeal to the very same statement to report only the sum of these two categories. The CDC, as I illustrated in #2900, appeals to the same statement to report both the disaggregated figures and their sums. These statements do not unambiguously support your choices like your appeal to them tends to imply.
Sources:
- Puerto Rico Department of Health COVID-19 cases report published on June 10 with data up to June 9
- Puerto Rico Department of Health COVID-19 cases report published on June 11 with data up to June 10
- "El Departamento de Salud ahora reportará casos confirmados y probables de COVID-19." El Nuevo Día, June 10, 2020.
Front page of report published on June 11 (which says data until June 10). Note that it doesn't say anywhere the 5,352 figure that you give in this corresponding file of yours, but that is the sum of its 1,403 confirmed and 3,949 probable figures:
Front page of report published on June 10 (which says data until June 9). This is the last one that reported the summed figures, which was 5,319 on that date (as your data files record).
This report was also the last one to include this chart of confirmed and probable cases by sample date, and several other similar summary tables and charts that were based on summed figures:
CSSEGISandData
Since April 14th, the CDC's recommendation for counting COVID-19 cases has been to include both laboratory confirmed and probable cases in line with the agency's case definition [see here]. We have applied this standard both in the US and globally, including probable cases in our counts* where available based on the local reporting criteria. As pointed out in #2900, the probable case counts for Puerto Rico include serology test results as well as cases that meet clinical or epidemiological criteria without a positive PCR test, but the counts are not disaggregated between these two categories. While this introduces some deviation from the reporting elsewhere in the US, we include Puerto Rico's probable cases as part of our count in line with the Department de Salud's reporting.
* The "confirmed" field, as explained in the field description includes both confirmed and probable cases. The naming of this field as "confirmed" rather than "total cases" is an issue of backwards compatibility and a desire not to introduce breaking changes into the code of the numerous users of this data set.