Prevalence of diabetes, hypertension among COVID-19 patients likely lower than reported
Risks for ICU admission and death from COVID-19 are higher for adults with diabetes and hypertension, but the number of COVID-19 patients with diabetes and hypertension may be lower than previously reported.
In conducting a systematic review and meta-analysis of reports on COVID-19 and comorbidities, Fady Hannah-Shmouni, MD, FRCPC, principal investigator of endocrine genetics and hypertension disorders for the Stratakis Laboratory’s Section on Endocrinology and Genetics at the Eunice Kennedy Shriver National Institute of Child Health and Human Development at NIH, and colleagues found high risk for bias and low confidence in the literature, according to a study published in the Journal of the Endocrine Society.
The researchers suspected overreporting of data and employed a method to avoid repetition of data in their analyses, and then used directed acyclic graphing to identify confounders.
“Our confidence was still low because there is so much heterogeneity in these initial reports. ... But our report, I would say, is closer to the prevalence rates than previous reports given the rigorous methodology we employed,” Hannah-Shmouni told Healio. “The biggest finding is that the overall prevalence for diabetes was 12% and 17% for hypertension, which is lower than previously reported.”
However, adults with diabetes or hypertension remain at an elevated risk for severe COVID-19, ICU admission and death.
“The message is very clear: People with these two comorbidities should be aware of this heightened risk, and appropriate patient education and resource allocation should be in place,” Hannah-Shmouni said.
The researchers analyzed data from databases featuring peer-reviewed, English-language manuscripts published from Dec. 1, 2019, to April 6, 2020. Researchers looked for observational and interventional studies reporting on the frequency of diabetes and hypertension among adults with COVID-19, as well as associations between diabetes and hypertension with severe cases, ICU admission and mortality. Studies with duplicated cases, high risk for bias or lack of overall quality of evidence were filtered out of the findings, leaving 65 studies with 15,794 patients.
Overall, hypertension was reported among 17% of patients (95% CI, 13-22; I2 = 95%), lower than the 26% to 30% reported in previous studies, according to the researchers. Diabetes prevalence was 12% among COVID-19 patients, (95% CI, 10-15; I2 = 89%), lower than the estimated 14.3% reported in a previous study. Co-occurring diabetes and hypertension was reported for 12% of patients (95% CI, 6-22; I2 = 0%).
Adults with diabetes had elevated relative risks for severe COVID-19 (RR = 1.5; 95% CI, 0.9-2.5; n = 1,991), ICU admission (RR = 1.96; 95% CI, 1.19-3.22; n = 8,890) and death (RR = 2.78; 95% CI, 1.39-5.58; n = 2,058). COVID-19 patients with hypertension also had elevated risk for severe disease course (RR = 1.48; 95% CI, 0.99-2.23; n = 2,023), ICU admission (RR = 2.95; 95% CI, 2.18-3.99; n = 1,737) and death (RR = 2.39; 95% CI, 1.54-3.73; n = 3,107).
“We believe that there is reverse causality in the reporting of hypertension prevalence in COVID-19,” Hannah-Shmouni said. “For instance, if you have an 80-year-old population where the prevalence of hypertension is about 80%, and if they get COVID-19, published data will link hypertension to COVID-19 and COVID-19 to hypertension. It is very important to distinguish what is regarded as a true risk factor for COVID-19 and not to introduce confounders and bias.”
Hannah-Shmouni said the study still had several limitations. Researchers were unable to categorize diabetes and hypertension into subtypes. Other important variables, such as BMI, smoking status and ethnicity, were also not included.
“We’re now getting more refined data, estimates and studies factoring various variables that were not available in those initial reports,” Hannah-Shmouni said. “Collectively, moving forward, we should include data in a live systematic review and meta-analysis so we can get better estimates.”
For more information:
Fady Hannah-Shmouni, MD, FRCPC, can be reached at the National Institutes of Health, 10 Center Drive, Room 1-3150, Bethesda, MD 20892; e-mail: firstname.lastname@example.org; Twitter: @DrShmouni.