From Feb 12 to Apr 9, a total of 9,282 US healthcare professionals (HCPs) were infected with COVID-19, according to a paper published yesterday in the US Centers for Disease Control and Prevention's (CDC's) Morbidity and Mortality Weekly Report.
In the same journal, researchers published a case report on three HCPs infected with the novel coronavirus after contact with the first patient with community-acquired COVID-19 in a Solano County, California, hospital in February.
Also, a study published today in Nature Medicine of 94 COVID-19 patients in Guangzhou, China, estimated that 44% of secondary case-patients were infected before the index patient experienced symptoms.
Healthcare workers made up 3% to 11% of cases
The CDC researchers analyzed preliminary data on confirmed cases of the novel coronavirus reported from 50 states, 4 territories and islands, and Washington D.C., excluding returned travelers from Wuhan, China, and the Diamond Princess cruise ship in January and February.
Only 3% of the 315,531 cases involved HCPs, but they accounted for 1,689 (11%) of 15,194 cases in states with more complete reporting of occupational status.
Occupational status was available for only 49,370 (16%) of the 315,531 cases, 19% of which were identified as HCPs. Of the workers reporting contact with COVID-19 patients in the 14 days before symptom onset, 780 (55%) reported contact at work, 384 (27%) had contact at home, 187 (13%) had contact in the community, and 72 (5%) had contact in more than one setting.
Of the infected HCPs with data on age, sex, and underlying diseases, median age was 42 years, 6,603 (73%) were women, and 1,779 (38%) had at least one underlying disease.
Of the 3,801 (41%) infected HCPs with data on race, 2,743 (72%) were white, 801 (21%) were black, 199 (5%) were Asian, and 58 (2%) were other or multiracial. Of 3,624 (39%) with data on ethnicity, 3,252 (90%) were non-Hispanic/Latino, and 372 (10%) were Hispanic/Latino.
The vast majority (6,760 [90%]) were not hospitalized, but 184 (2% to 5%) were admitted to an intensive care unit, and 27 (0.3% to 0.6%) died. Death rates were highest in HCPs 65 years and older (10 in that age-group [37%] died).
A total of 4,336 (92%) of infected HCPs said they had one or more symptoms, including fever, cough, or shortness of breath. Two-thirds (3,122) reported muscle pain, while 3,048 (65%) said they had a headache. Loss of taste or smell was added to the patient record for 750 (16%) for "other" symptoms.
"It is critical to ensure the health and safety of HCP, both at work and in the community," the authors wrote. "Improving surveillance through routine reporting of occupation and industry not only benefits HCP, but all workers during the COVID-19 pandemic."
Noting that contact tracing alone is unlikely to identify many at-risk workers, they called for screening all HCPs for fever and respiratory symptoms before shifts, prioritizing them for testing, having flexible and nonpunitive leave policies to discourage working while sick, provision of personal protective equipment (PPE), and assignment of older HCPs to telehealth, administration, or clinics reserved for non-coronavirus patients.
The authors noted that the number of infected HCPs is likely an underestimate because of the low percentage of records reporting occupation and the high likelihood that some workers with mild or asymptomatic disease did not get tested.
In the California case report, two of the three infected HCPs were present during aerosol-generating procedures and were not wearing PPE because transmission-based precautions were not yet in use.
Substantial transmission before symptom onset
In the Nature Medicine study, Chinese researchers identified patterns of coronavirus shedding from throat swabs and modeled contagiousness profiles from a separate sample involving 77 pairs of infector-infectee transmission.
They collected 414 swabs from the 94 patients from symptom onset to 32 days later. In 44% of the secondary cases (95% confidence interval [CI], 25% to 69%), viral load was highest at symptom onset, indicating that COVID-19 may be most infectious before or at symptom onset.
Viral loads gradually decreased until about day 21, when they were no longer detectable, with no obvious differences between sexes, age-groups, or illness severity.
Of the secondary case-patients, 47 (50%) were male, and median age was 47 years. Sixty-one (66%) had moderate illness characterized by fever and/or respiratory symptoms and evidence of pneumonia on radiographs; none had severe or critical illness on hospitalization.
The results suggest that COVID-19's infectiousness profile is closer to that of the flu than that of severe acute respiratory syndrome (SARS), which is most infectious about 7 to 10 days after symptom onset, improving the ability to stop disease spread through isolation and quarantine.
"Significant presymptomatic transmission would probably reduce the effectiveness of control measures that are initiated by symptom onset, such as isolation, contact tracing, and enhanced hygiene or use of face masks for symptomatic persons," the authors wrote, adding that social distancing would likely be a key strategy.
Using data on the 77 infector-infectee pairs, the researchers estimated the mean time from symptom onset in the index patient to symptom onset in the secondary patients, or the serial interval, at 5.8 days (95% CI, 4.8 to 6.8 days) and the median serial interval at 5.2 days (95% CI, 4.1 to 6.4)
"Assuming an incubation period distribution of mean 5.2 days from a separate study of early COVID-19 cases, we inferred that infectiousness started from 2.3 days (95% CI, 0.8 – 3.0 days) before symptom onset and peaked at 0.7 days (95% CI, −0.2 – 2.0 days) before symptom onset," they said. They added that contagiousness declined quickly over 7 days and pointed out that isolating most patients after symptoms began prevented some post-symptomatic transmission.
Sensitivity analysis showed that holding constant the start of contagiousness from 1 to 7 days before symptoms began, infectiousness peaked 0 to 2 days before symptoms appeared, and 46% to 55% of transmissions occurred in the presymptomatic period.
Their simulation revealed that the percentage of serial intervals shorter than 2 days would be larger if infectiousness was assumed to begin before symptoms emerged. Based on the 7.6% negative serial intervals estimated from the infector-infectee pairs, infectiousness at least 2 days before symptom onset and peak infectiousness 2 days before to 1 day after is likely.
"More inclusive criteria for contact tracing to capture potential transmission events 2 to 3 days before symptom onset should be urgently considered for effective control of the outbreak," they said.
