Stockholm City’s Elderly Care and Covid19: Interview with Barbro Karlsson

The data for Stockholm city comes from: https://​platz.​se/​coronavirus/​?​city=​Stockholm&​lang=​en, showing 669 deaths as of May 11, 2020. The data for Sweden comes from: https://​www.​worldometers.​info/​coronavirus/​#countries, showing 3256 deaths as of May 11, 2020.

We say “at least” because there would also be additional 70+ deaths (though small in number) of those in elderly-care services other than nursing homes and home-care, but we do not have data for them.

70% = 79.6% * 88%.

Most of the data used in this article derives from the National Board of Health and Welfare (Socialstyrelsen), which is the officially designated authority on deaths and their causes. Meanwhile, the Public Health Agency of Sweden (Folkhälsomyndigheten) also reports deaths of people with Covid19 lab-confirmed infection, on a quicker basis, and its numbers are used, for example, at Worldometers (often used in journalistic reporting to make international comparisons). We have reviewed the Board’s official Covid19 guidelines issued to doctors about reporting cause of death, the description of how the Board uses that report to arrive at its determination, and a description of the two methods of the two organizations. We have also communicated by email with someone from the Board and with someone from the Agency about the matter. For the Board, the doctor’s report largely determines whether the death is reported as a Covid19 death. The form the doctor completes asks him or her to indicate, in part 1, a primary progression of ailments leading to death, and then, in part 2, possibly contributory ailments. When the doctor identifies Covid19 as primary, the death clearly is counted as a Covid19 death. But, according to the email we received, when the doctor lists Covid19, not as primary but only as a possibly contributory cause, the death would not be counted as a Covid19 death. (And, of course, the death would not be counted as a Covid19 death if the doctor did not even list it as a possibly contributory cause, even though the patient had Covid19.) For the Agency, a Covid19 death is a death of anyone who was lab-confirmed with Covid19. We have examined Covid19 death counts by the two organizations, and they are roughly equal. The Board includes some deaths that were not lab-confirmed cases of Covid19, while the Agency includes some deaths that the doctor did not deem to have been caused by Covid19 even though the deceased was a lab-confirmed case. Apparently the two differentials roughly match and cancel out.

The regional (as opposed to municipal) authorities deal chiefly with hospitals and public transportation, not elderly care, so the focus here is more on municipal authorities, not regional.

Stockholm City’s Äldreförvaltningen also has another department (Administrativa avdelningen) which tends the internal administration of Äldreförvaltningen, so that too would not present nosocomial problems.

“Sheltered housing” seems to be the expression in the United Kingdom, whereas in the United States one might say “assisted living;” we’ve opted here for sheltered living.

On immigrant populations in Stockholm showing disproportionately high Covid incidence, see here.

There are forms of elderly care aside from nursing homes and home-care (such as the trigger alarm, the day-centers, the sheltered living homes, and so on); however most people receiving such other services would also be receiving home-care, so this 20 to 1 ratio is probably close to reality.

As of May 14, 2020 the National board reported 357 individuals dead by Covid-19 in nursing homes, of 5268 living in nursing homes (as reported in the Elderly care in Stockholm time-series) making 6.8%. For home-care the numbers are 180 dead by Covid-19, of 12,901 receiving home-care, or 1.4%.

The numerators of those two percentages come from the first paragraph of this subsection, while the denominators come from Stockholm City 2020. Note that the denominator for the nursing homes is actually the number of places (which are essentially always filled), not the number of residents served over the approximately two months during which the Covid19 deaths occurred, which would be somewhat more than 5047. Thus the 7.1% is a bit of an overstatement of the nursing-home death rate.

There had been a controversy over mandating face masks in elderly care, see Swedish reporting here.

Stockholms City 2020.

An internal document of Stockholm city, where Karlsson works, reports that as of May 5, 2020, both private-sector and public-sector had rates 0.6 to 0.7% Covid19 cases among home-care recipients and 6.4% among nursing home residents (Stockholm City 2020).

We are grateful to Benny Carlson for pressing the issue with us.