Estimation of the incidence of MRSA patients: evaluation of a surveillance system using health insurance claim data.

S Tanihara,S Suzuki

doi:10.1017/s0950268816000674

Abstract

Because sentinel surveillance systems cannot obtain information about patients who visit non-sentinel medical facilities, the characteristics of patients identified by these systems may be biased. In this study, we evaluated the representativeness of a methicillin-resistant Staphylococcus aureus (MRSA) surveillance system using health insurance claim (HIC) data, which does not depend on physician notification. We calculated the age-specific incidence of MRSA patients using data from the Japan Nosocomial Infections Surveillance (JANIS) programme, which is based on sentinel surveillance systems, and inpatient HICs submitted to employee health insurance organizations in 2011, and then computed age-specific incidence ratios between the HIC and JANIS data. Age-specific MRSA incidence in both datasets followed J-shaped curves with similar shapes. For all age groups, the ratios between HIC and JANIS data were around 10. These findings indicate that JANIS notification of MRSA cases was not affected by patients' age.

Highlights

If a patient was hospitalized for 52 months without interruption, all of their inpatient health insurance claim (HIC) were collected; we identified all HICs issued on behalf of the patient and investigated injection of antibiotics for methicillin-resistant Staphylococcus aureus (MRSA) during the hospitalization
The number of insured persons peaked in the 30–39 years age group, which accounted for 21·4% of the total
We calculated and compared the incidences of MRSA patients based on HIC data for those insured by health insurance organizations, as well as data reported to Japan Nosocomial Infections Surveillance (JANIS)

Summary

SUMMARY

Because sentinel surveillance systems cannot obtain information about patients who visit nonsentinel medical facilities, the characteristics of patients identified by these systems may be biased. We evaluated the representativeness of a methicillin-resistant Staphylococcus aureus (MRSA) surveillance system using health insurance claim (HIC) data, which does not depend on physician notification. We calculated the age-specific incidence of MRSA patients using data from the Japan Nosocomial Infections Surveillance (JANIS) programme, which is based on sentinel surveillance systems, and inpatient HICs submitted to employee health insurance organizations in 2011, and computed age-specific incidence ratios between the HIC and JANIS data. Age-specific MRSA incidence in both datasets followed J-shaped curves with similar shapes. The ratios between HIC and JANIS data were around 10. These findings indicate that JANIS notification of MRSA cases was not affected by patients’ age

INTRODUCTION

METHODS

RESULTS

DISCUSSION

CONCLUSION