HERSHEY, Pa. — COVID-19 survivors have a 66% higher risk of developing type 1 or type 2 diabetes following their diagnosis compared to those not diagnosed with COVID-19, according to a study by Penn State College of Medicine researchers. Their findings are among the latest evidence suggesting that people diagnosed with COVID-19 may experience a range of health issues in the time period following their illness.

While prior research has indicated that COVID-19 can lead to increased incidence of diabetes in survivors, compared to the general population, the causes are not well understood. According to the researchers, SARS-CoV-2 — the virus that causes COVID-19 — binds to an enzyme receptor found on the surface of many organs and tissues – including cells found in the pancreas, small intestine and kidneys. Other research teams have found that the virus affects insulin levels and induces the death of pancreatic beta cells, which produce insulin.

“We can’t definitively conclude that COVID-19 causes diabetes, and more research on whether there is a biological cause to explain this association is needed,” said Paddy Ssentongo, an internal medicine resident physician at Penn State Health. “But we know that other viruses like mumps virus, rotavirus and cytomegalovirus are associated with the development of diabetes, so it’s not implausible that SARS-CoV-2, which has been shown to affect multiple systems in the human body, can also do the same.”

Ssentongo and Djibril Ba, assistant professor of public health sciences, completed one of the largest meta-analyses, or comprehensive review of existing studies, of the association between COVID-19 and diabetes. They pulled data from eight eligible studies out of more than 850 that took place from December 2019 until mid-October 2022. Their final analysis included data from more than 4 million COVID-19 patients and 43 million control patients who were not diagnosed with the illness.

While each of the eight studies reported a different level of risk, the team used statistical modeling to find an average, or pooled, risk ratio, which is a measure of how much higher a risk a COVID-19 survivor would be of developing diabetes than someone who was not diagnosed with the virus. They found a pooled risk ratio of 1.66 — implying that COVID-19 survivors have a 66% higher risk of developing new onset diabetes. The risk did not vary by age, sex or the quality of the study used. Through further statistical analysis, the team studied whether any one study was disproportionately influencing their results and found that none of the included studies was swaying the results more than the others. The results were published in Nature Scientific Reports on Nov. 23.

“To date, more than 660 million people have tested positive for COVID-19,” Ba said. “It is important that survivors and their health care providers are aware of this trend so they can be on the lookout for development of diabetes.”

Ba said their results are similar to previous studies examining the relationship between COVID-19 and diabetes, but that their analysis is one of the largest to date. However, he noted it’s possible that some people in the control group could have had undetected, mild asymptomatic COVID-19 because they had not been tested. As a result, misclassification could have led to an underestimate of the strength of the association between the two diseases. The researchers were also unable to study whether vaccination status affected outcomes as those data were not available to them.

The research team said future studies should examine the social determinants of health associated with new onset diabetes so effective public health prevention and management strategies can be developed. They also said more research is needed on whether there are biological causes to explain this increased risk of diabetes, and that genomics data could be used to identify COVID-19 survivors at most risk for developing the disease.

Yue Zhang, Lisa Witmer and Vernon Chinchilli of Penn State College of Medicine also contributed to this research. The authors declare no conflicts of interest or specific funding for this research.