by revere, cross-posted from Effect Measure
Swine flu infection of health care workers (or as CDC refers to them, health care personnel or HCP) was of interest early in the pre-pandemic phase for at last two reasons. One was the obvious goal of estimating the risk to front line workers and devising best practices for their protection. Another was the belief, reinforced by the SARS outbreak in 2003, that spread to HCP was an early warning that the virus was easily transmissible from person to person. SARS is a disease where patients are most infectious in the later stages when they are extremely ill, and HCP were among the hardest hit groups. Most flu is transmitted in the community, but the SARS model still seems appropriate for flu viruses like H5N1 (“bird flu”) where transmission is rare. Any report that a health care worker has been infected from a bird flu patient is viewed with alarm, possibly suggesting that the probability of transmission has increased. And bird flu was the template upon which pandemic planning was based. So within a few weeks of the outbreak (the first case of novel H1N1 was diagnosed in mid-April), CDC asked state health departments to report any cases of novel H1N1 among HCP. Yesterday they reported the first results in CDC’s Morbidity and Mortality Weekly Reports (MMWR):
As of May 13, CDC had received 48 reports of confirmed or probable infections with novel influenza A (H1N1) virus; of these, 26 reports included detailed case reports with information regarding risk factors that might have led to infection. Of the 26 cases, 13 (50%) HCP were deemed to have acquired infection in a health-care setting, including one instance of probable HCP to HCP transmission and 12 instances of probable or possible patient to HCP transmission. Eleven HCP had probable or possible acquisition in the community, and two had no reported exposures in either health-care or community settings. Among 11 HCP with probable or possible patient to HCP acquisition and available information on PPE [Personal Protective Equipment] use, only three reported always using either a surgical mask or an N95 respirator. (CDC, MMWR [cites omitted])
This is information from 18 states during the earliest phase of the outbreak, with illnesses that had onset between April 23 and May 4. Additional information was available for only 26 of the 48 HCP cases, so this is a small sample making generalizations risky. The HCP reporting was of special interest and not representative of all cases and we know many cases were missed in the general population. On the basis of this evidence, however, HCP don’t seem to have been a greater proportion of the cases than their percentage in the general population (“Among confirmed and probable cases in adults aged 18–64 years and reported to CDC as of May 13, approximately 4% have occurred in HCP; approximately 9% of working adults in the United States are employed in health-care settings”). In other words, infection of HCW was not a good sentinel for transmissibility in general for this virus.
What do we learn from the 26 cases with detailed information on risk factors? 50% appear to have contracted their disease in a health care setting, with at least 5 directly from a patient. In the early weeks of the outbreak hospitalizations were uncommon, so most of the HCP contracted their disease in outpatient settings. Only 3 had used a surgical or N95 mask, but of these, one, a physician used an N95 mask all the time but it had not been fit tested. In general, CDC’s recommendations for PPE during this outbreak were not followed, although masks and gloves were worn sometimes.
Like anyone else, HCP also contracted their infections in community settings. So they are a possible vector between health care and the community in both directions. CDC emphasized in its Thursday press briefing that ongoing investigations of HPC clusters was suggesting two things related to this: (a) infectious patients should be identified as soon after they come in the door as possible so that HCP can use the necessary protective measures; and, (b) HCPs should not come to work sick. At least one of the cases was a probable HCP to HCP transmission and a sick HCP can infect vulnerable patients. We can expect to see additional guidance and better data as more data is collected and analyzed.
It is perhaps somewhat counterintuitive that HCP, with a high potential for exposure, have no higher risk. Perhaps we will revise this judgment as more reliable and larger amounts of data come available. At this point it seems that when there is lots of this virus in the community, you are as likely to get it outside of as inside a health care institution. That would seem to apply both to patients there for other reasons and to people who work there.
We’ll have to see if this holds as we learn more.