This retrospective study conformed with all legal guidelines and the protocol was approved by the Ethical Committee of our institute (reference BC-07671).
The retrospective study cohort was defined as all (anonymized) residents at 2 care homes with COVID-19 diagnosis based on clinical grounds and/or polymerase chain reaction (PCR) laboratory testing (nasopharyngeal swabs) from March 1 to April 16. Both care homes for older individuals experienced COVID-19 outbreaks during this period. Although several diagnostic tools are currently available, none gives high specificity and sensitivity on its own. However, it is generally accepted that the clinical diagnosis is a key herein.
Clinical presentation of COVID-19: A systematic review focusing on upper airway symptoms.
Moreover, some techniques like computed tomography are not feasible in the context of nursing homes. To determine the day of disease onset, structured and unstructured diagnostic records were analyzed for symptoms suggesting COVID-19 infection. The first day of suggestive symptoms on 2 of 3 consecutive days was considered as the day of disease onset. For the PCR-diagnosed residents, the suggestive symptoms used for disease onset were cough, shortness of breath (dyspnea), sore throat, runny nose, general weakness, headache, confusion, muscle pain, arthralgia, diarrhea, abdominal pain, vomiting, fever (T° > 37.6°C), increased oxygen need, or low oxygen saturation (SpO2 ≤92%). In cases in which no symptoms were mentioned (while still being PCR COVID-19–positive diagnosed), the date of nasopharyngeal sampling was used as the day of disease onset. For the clinically diagnosed residents without a confirmatory PCR laboratory test, the symptoms used for determining disease onset were defined more strictly, that is, respiratory complaints (cough, shortness of breath, sore throat, runny nose), fever (T° > 37.6°C), increased oxygen need, or low oxygen saturation (SpO2 ≤92%).
The primary outcomes were (1) serious COVID-19, that is, long-stay hospital admission (length of stay ≥7 days) or death (at nursing home or hospital) within 14 days of disease onset, and (2) asymptomatic, that is, no disease symptoms as defined previously throughout the whole study period while still being PCR diagnosed.
Information about daily drug intake was available through the medication administration records at nursing home level. All residents were stratified according to drug exposure to ACEi or ARB within 7 days before the day of disease onset or during the disease (before an outcome being reached). Specifically, we considered as treated all residents taking for ≥2 days an ACEi (ramipril, lisinopril, enalapril, captopril, quinapril, imidapril, fosinopril, trandolapril) or ARB (candesartan, irbesartan, losartan, olmesartan, telmisartan, valsartan) up to 7 days before or 14 days after disease onset. Only 1 day of drug intake (eg, for acute hypertension) was not considered as (chronic) treatment. An identical protocol was used to stratify according to drug exposure to statins (atorvastatin, fluvastatin, pravastatin, rosuvastatin, simvastatin).
As no direct diagnosis codes were available, we developed a mapping table based on clinical drug prescriptions to determine the diabetic and hypertension status of all residents. It was designed by a specialist in care of older individuals and validated by 2 independent physicians, one a general physician and the other a cardiologist.
The functional status of all residents was a dichotomous variable (high vs low functioning). This definition was based on the available Katz scale for residents before the day of disease onset. The Katz scale is a measure of independent activity of daily living.
Data Collection and Quality Control
Data about residents were available at the level of nursing homes. After anonymization, these data were merged in a relational database for further processing by making use of Extract, Transform, and Load techniques. All received anonymized data were then evaluated on basic data quality attributes such as completeness (ie, the extent of missing data) and accuracy (ie, whether or not suspicious outliers were present in the individual attributes). Data were enriched with Anatomical Therapeutic Chemical codes for the included drugs. Suggestive symptoms were searched for based on biometrical measurements as well as indications in text. For the latter, basic Natural Language Processing (NLP) techniques were used. For the residents still in the hospital at the moment of data extraction, median imputation was used to estimate length of hospital duration. Two independent physicians manually verified all recorded symptoms as well as all data for a random subsample.
We calculated the distributions for dependent and independent variables for the total cohort using appropriate measures of central tendency and dispersion. For our main analysis, we investigated the association between ACEi/ARB and/or statin treatment and (1) serious disease, measured as long-stay hospital admission or death, or (2) asymptomatic disease using a series of logistic regressions applying Firth's correction. Unless otherwise indicated, the logistic analyses were conducted on the whole cohort (n = 154).
This procedure has been used previously by our group and shown to be robust for low-prevalence events and low-dimensional settings.
- Bean D.
- Kraljevic Z.
- Searle T.
- et al.
Treatment with ACE-inhibitors is associated with less severe SARS-Covid-19 infection in a multi-site UK acute Hospital Trust.
Firth logistic regression for rare variant association tests.
Bias reduction of maximum-likelihood-estimates.
We first explored the independent association between ACEi/ARB and both outcomes, as well as the association between statins and the same outcomes. Then we adjusted the models stepwise for age, sex, functional status, hypertension, and diabetes mellitus. Also, the method of diagnosis (PCR or clinical) was evaluated as possible confounder. All statistical analyses were performed using SAS 9.4 (SAS Institute, Cary, NC) and RStudio 3.5.2 (R Foundation for Statistical Computing, Vienna, Austria).