Identifying common pharmacotherapies associated with reduced COVID-19 morbidity using electronic health records

Abstract

Objective

Absent a vaccine or any established treatments for the novel and highly infectious coronavirus-19 (COVID-19), rapid efforts to identify potential therapeutics are required. We sought to identify commonly prescribed medications that may be associated with lesser risk of morbidity with COVID-19 across 6 Eastern Massachusetts hospitals.

Design

In silico cohort using electronic health records from individuals evaluated in the emergency department between March 4, 2020 and July 12, 2020.

Setting

Emergency department and inpatient settings from 2 academic medical centers and 4 community hospitals.

Participants

All individuals presenting to an emergency department and undergoing COVID-19 testing.

Main Outcome or Measure

Inpatient hospitalization; documented requirement for mechanical ventilation.

Results

Among 7,360 individuals with COVID-19 positive test results by PCR, 3,693 (50.2%) were hospitalized in one of 6 hospitals. In models adjusted for sociodemographic features and overall burden of medical illness, medications significantly associated with diminished risk for hospitalization included ibuprofen and sumatriptan. Among individuals who were hospitalized, 962(26.0%) were admitted to the intensive care unit and 608 (16.5%) died; ibuprofen and naproxen were also more commonly prescribed among individuals not requiring intensive care.

Conclusions

These preliminary findings suggest that electronic health records may be applied to identify medications associated with lower risk of morbidity with COVID-19, but larger cohorts will be required to address the substantial problem of confounding by indication, such that extreme caution is warranted in interpreting nonrandomized results.

Trial Registration

None

Summary Boxes

Section 1: What is already known on this topic

Absent a vaccine or any established treatments for the novel and highly infectious coronavirus-19 (COVID-19), rapid efforts to identify potential therapeutics are required.

Section 2: What this study adds

This cohort study across 6 hospitals identified medications enriched among individuals positive for COVID-19 who are less likely to experience adverse outcomes including hospitalization, intensive care, or death.

SciScore for 10.1101/2020.04.11.20061994: (What is this?)

Please note, not all rigor criteria are appropriate for all manuscripts.

Table 1: Rigor

Institutional Review Board Statement	IRB: The Partners HealthCare Human Research Committee approved the study protocol. Consent: As no participant contact was required in this study based on secondary use of data arising from routine clinical care, the committee waived the requirement for informed consent as detailed by 45 CFR46.116.
Randomization	not detected.
Blinding	not detected.
Power Analysis	not detected.
Sex as a biological variable	not detected.

Table 2: Resources

No key resources detected.

Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:

Two key limitations must be emphasized. First, as in any nonrandomized study, the risk for bias - and particularly for confounding by indication - is high. That is, other differences between the groups being compared, such as differences in comorbidity, may be proxied by the medication exposure; causal relationships cannot be inferred. Indeed, the association with greater adverse outcomes with renal disease and diabetes treatments, and with bowel regimens commonly seen in skilled nursing facilities, likely represents such residual confounding. Second, the power afforded by this cohort, despite incorporating a network of hospitals, is modest. In particular, absence of effect does not preclude potential efficacy, particularly as our approach will exclude rarer medications, such as those typically prescribed for time-limited periods such as short-term antibiotics. Further limitations include the reliance on medications electronically prescribed, rather than filled, in the 12 months prior to hospitalization. The lag in availability of current medication data precludes characterization of current medications on admission. This lag, as well as the possibility that medications may be discontinued or not filled, would tend to bias results toward the null hypothesis, by introducing additional heterogeneity. While sensitivity analysis requiring multiple prescriptions for definition of exposure did not meaningfully change results, further studies with more precise exposure characterizat...

Results from TrialIdentifier: No clinical trial numbers were referenced.

Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

Results from JetFighter: We did not find any issues relating to colormaps.

Results from rtransparent:

Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
No protocol registration statement was detected.

Read the original source