Diabetes and risk of hospitalization due to infection in northeastern Thailand: Retrospective cohort study using population-based healthcare service data

Abstract

Background

Population-based studies describing the association between diabetes and increased risk of infection have largely been based in high-income countries. There is limited information describing the burden of infectious disease attributable to diabetes in low and middle-income countries.This study aimed to describe the burden and risk of infectious disease hospitalisation in people with diabetes compared to those without diabetes in northeastern Thailand.

Methods

In a retrospective cohort study using electronic health record data for 2012-2018 for 3.8 million people aged ≥20 years in northeastern Thailand, hospitalization rates for any infectious diseases (ICD-10 codes A00-B99) were estimated and negative binomial regression used to estimate rate ratios (RR) for the association between diabetes and infectious disease hospitalization adjusted for age, sex and area of residence.

Results

164,177 people had a diagnosis of diabetes mellitus at any point over the study period. Infectious disease hospitalization rates per 1,000 person-years (95%CI) were 71.8 (70.9, 72.8), 27.7 (27.1, 28.3) and 7.5 (7.5, 7.5) for people with prevalent diabetes, incident diabetes and those without diabetes respectively. Diabetes was associated with a 4.6-fold higher risk of infectious disease hospitalization (RR (95%CI) 4.59 (4.52, 4.66)). RRs for infectious disease hospitalization were 3.38 (3.29, 3.47) for people with diabetes managed by lifestyle alone and 5.29 (5.20, 5.39) for people receiving prescriptions for diabetes drugs.

Conclusions

In this Thai population, diabetes was associated with substantially increased risk of hospitalisation due to infectious diseases and people with diabetes who were on pharmacological treatment had a higher risk than those receiving lifestyle modification advice alone.

Novelty statement

What is already known?

• Diabetes was associated with an increased risk of serious infection as defined by need for hospitalization, but the evidence is largely from high-income countries.

• Evidence to describe the burden and risk of infectious disease attributable to diabetes in low and middle-income countries is scarce.

What has this study found?

• Infectious disease hospitalization rates in people with and without diabetes were lower than that of high income countries

• Diabtetes was associated with a 4.6-time increased risk of hospitalization due to infection, the magnitude of association being higher than previously reported.

• Risk of infection-related hospitalization was differential by diabetes treatment, with people with diabetes receiving prescriptions for diabetes drugs having a considerably higher risk than those managed by lifestyle alone.

What are the implications of this study?

• People with diabetes, particularly those receiving pharmacological treatment, are at high risk of serious infection and further actions are needed to identify factors underlying the excess risk of infection due to diabetes and its treatment to inform appropriate preventive strategies