A typical investigation on the status of diagnosis and reporting of hepatitis B inpatients in non-surveillance hospitals in three provinces in China, 2015 / 中华预防医学杂志

ABSTRACT

Objective@#To review the consistency of diagnosis and reporting of hepatitis B (HB) patient in non-surveillance hospitals in three provinces and analyze the influencing factors.@*Methods@#In 2016, using typical survey methods, we carried out a hospital-based pilot study in three provinces Fujian, Hainan and Gansu. In each province, we chose two hospitals with grade 3 and grade 2 respectively in each province, using the following criteria (1) in 2015, the hospital reported a greater number of hepatitis B cases compared the hospital-based provincial mean; (2) the hospital had an advanced laboratory information system (LIS) with access to HBsAg test results; (3) the hospital had an electronic hospital information system (HIS) which linked to the LIS via the inpatient medical record number; (4) general hospital; (5) non-surveillance hospitals for hepatitis B. Using national notifiable infectious disease reporting system (NNDRS), we chose all HB patients who were reported by the investigated hospitals in 2015, and we linked NNDRS HBV case-reports with patient-data from hospital information systems (HIS) to review the diagnosis, and then to compare the consistency of reviewed diagnosis and NNDRS report diagnosis, which we made a descriptive analysis. We used multivariable logistic regression to examine factors associated with misclassification of case-reports to NNDRS.@*Results@#We found the NNDRS report accuracy was 47.11% (669) among 1 420 eligible inpatient hepatitis B inpatients. Of the 352 reported acute HBV cases, 6.53% (23) were consistent with our medical record review, the accuracy rate for level 2 hospitals and level 3 hospitals was 9.42% (21) and 1.55% (2), respectively. Of the1 068 reported chronic HBV cases, 60.49% (646) were consistent with our medical record review, the accuracy rate for level 2 hospitals and level 3 hospitals was 57.92% (106) and 60.02% (540), respectively. Compared to primary diagnosis of HB patients, the OR(95%CI) for mis-report was 29.36 (19.21-44.76) in non-primary diagnosis of HB patients. Compared to Fujian Province, the mis-report risk was higher in Hainan province and Gansu Province, with the values of OR (95%CI) being 2.33 (1.58-3.44) and 20.38 (11.29-36.78), respectively; compared to level 3 hospitals, the OR (95%CI) for mis-report was 2.38 (1.66-3.42) for level 2 hospitals; compared to HB related wards, the OR (95%CI) for mis-report was 1.45 (1.04-2.01) in non-HB-related wards.@*Conclusion@#In some non-surveillance areas of China, the consistency between hepatitis B diagnosed in hospital and reported in NNDRS was low. Factors affecting the accuracy of HB surveillance data in NNDRS were level 2 hospitals, non-liver disease departments and nonprimary diagnosis of HB.