Intended for healthcare professionals

  1. Alicia L Rauh, assistant professor of medicine1,
  2. Jeffrey A Linder, professor of medicine2
  1. 1Division of Hospital Medicine, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL, USA
  2. 2Division of General Internal Medicine and Geriatrics, Department of Medicine, Northwestern University Feinberg School of Medicine, 750 North Lake Shore Drive, 10th Floor, Chicago, IL 60611, USA
  1. jlinder{at} (or @jeffreylinder on Twitter)

It’s time to shift the research focus to studies on living with this disease

By the third week of May 2020, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the virus that causes coronavirus disease 2019 (covid-19), had infected about 4.7 million people worldwide and over 300 000 had died.1 Without a vaccine or disappearance of the virus, we could be living with SARS-CoV-2 and covid-19 for the foreseeable future, possibly years.

In the initial phase of the pandemic, descriptive studies of patients in hospital have been invaluable in understanding the epidemiology, populations at risk, and outcomes for patients with severe covid-19. Peer reviewed studies and preprints have described patients admitted to hospital in China, the United States, and Korea, ranging in size from 69 to 5700 patients.2345 Other studies, notably several from Italy, describe those in intensive care.67

In a linked study, Docherty and colleagues (doi:10.1136/bmj.m1985) describe what appears to be the largest cohort of patients in hospital so far.8 Using a standardized data collection protocol, the investigators collected information on 20 133 patients with covid-19 from 208 hospitals, representing 34% of patients admitted with this disease in England, Scotland, and Wales.

Beyond the size of the cohort, the researchers had several notable findings. This cohort is among the oldest reported (median age 73). The proportion of intensive care unit admissions (17%) was slightly lower than many other studies. The mortality (26%) at the time of reporting was higher than almost all other studies. Independent predictors of mortality included older age, male sex, obesity, and several chronic conditions. The authors reason that the high mortality in this cohort was due to older age, health system differences (for example, proportion of intensive care unit beds), and practices for advanced care planning. Indeed, guidance from the National Institute for Health and Care Excellence regarding use of critical care services emphasized early advanced care planning in addition to consideration of patient frailty, judicious use of potentially limited resources of the UK’s health service, and protection of NHS staff.9

Observational studies of hospital admissions cover only part of the healthcare response to the pandemic: we need better information and studies of care before, during, and beyond the hospital. Before admission, physicians need guidance about prevention, transmission, monitoring, home care, integration with primary care, risk factors for clinical worsening, and criteria for escalating patients to emergency or hospital care. Observational studies in community or home care settings could serve as platforms for interventional studies of prevention, treatment, or supportive strategies. Patients, families, and care givers are desperate for better, clearer information on effective home care.

Two large studies, published as preprints, link community with hospital care. The OpenSAFELY Collaborative is using electronic health record data from 17.4 million English patients in primary care in a study that has integrated exposure, incidence, and 5683 deaths in hospital.10 Independent risk factors for death included black and Asian ethnicity (versus white ethnicity) and deprivation, similar to findings from the United States.11 An integrated health delivery plan in California and Washington states has used data from 9.6 million enrollees in a study that reported 1277 hospital admissions and 15% mortality in hospital to examine community transmission dynamics.12

In the absence of “game changing” antiviral or immunomodulating pharmacotherapies, more controlled studies of supportive treatment strategies are needed for patients in hospital. Clinical experience and research evidence so far suggest that key components of good supportive care include oxygenation; management of symptoms, fluid balance, and chronic conditions; judicious use of antibacterial drugs; prophylactic anticoagulation; closer monitoring of frail or immunosuppressed patients; multidisciplinary management among hospital medicine, infectious diseases, pulmonary critical care, palliative care, and medical ethics teams; and advanced care planning with patients and care givers. Establishing criteria for discharge and arranging follow-up are challenging, particularly for patients who might still have symptoms or are infectious.

Beyond the hospital, long term outcomes of covid-19 are unknown. Anecdotal reports have included symptoms that last for weeks or months. Sequelae of covid-19 infection might include worsening of chronic conditions and profound needs for rehabilitation. And we still have much to learn about transmission, immunity and its durability, and, as with other coronaviruses, the potential for reinfection.131415

At the outset of the covid-19 pandemic, it was natural to focus first on the people with severe disease who might need potentially scarce resources in hospital and intensive care. Cohort studies of such patients important, and the work described by Docherty and colleagues is a testament to good planning and preparation before, and implementation of data collection during a pandemic. If we are going to be managing covid-19 for the next several years, however, we need to understand and optimize care before, during, and beyond the hospital.


  • Research, doi: 10.1136/bmj.m1985
  • Competing interests: The BMJ has judged that there are no disqualifying financial ties to commercial companies. The authors declare the following other interests: JAL is supported by a contract from the Agency for Healthcare Research and Quality (HHSP233201500020I), and is also supported by grants from the National Institute on Ageing (R21AG057400, R21AG057396, R33AG057383), National Institute on Drug Abuse (R33AG057395), and the Agency for Healthcare Research and Quality (R01HS026506), the Peterson Center on Healthcare. The BMJ policy on financial interests is here:

  • Provenance and peer review: Commissioned; not peer reviewed.