Completeness of Reporting of Chronic Hepatitis B and C Virus Infections — Michigan, 1995–2008

Chronic hepatitis B virus (HBV) and chronic hepatitis C virus (HCV) infections are leading causes of death from cirrhosis and hepatocellular carcinoma in the United States (1). Because underreporting has complicated the understanding of disease burden, in 2010 the Institute of Medicine requested that CDC perform a comprehensive evaluation of national viral hepatitis surveillance (1). Hepatitis surveillance data rely on local and state estimates, and a better understanding of reporting at these levels can inform strategies to improve national data quality. As an initial assessment, CDC partnered with the Michigan Department of Community Health (MDCH) and an urban health-care system in southeastern Michigan to evaluate the completeness of reporting (including case status, demographic, and risk factor information) of cases of chronic HBV and HCV infection among persons who were enrolled in a multicenter chronic hepatitis cohort study (2) to the MDCH viral hepatitis registry. This report summarizes the results of that assessment. Among clinically confirmed chronic hepatitis infections, 82% of HBV infections and 65% of HCV infections were reported. Completeness of reporting of chronic HBV and HCV infections was significantly improved for those with more recent clinical diagnoses, but reporting still remained incomplete. The completeness of reporting varied significantly by demographic characteristics of patients with HCV infection. Few reports of either HBV or HCV infection included risk factors. Improving surveillance of chronic hepatitis in Michigan will require exploration of more efficient methods for the transfer of laboratory and clinical data and evaluation of the most appropriate sources for risk factor information to aid in the prevention of viral hepatitis transmission. Similar collaborations with health-care institutions that use electronic International Classification of Diseases, Ninth Revision (ICD-9) codes and laboratory data can provide local and state health departments with insight into the challenges to case reporting in their jurisdictions.

Reporting of chronic HBV and HCV infections became mandatory in Michigan in 2004 and 2000, respectively. In 2004, electronic dual reporting of both infections by laboratories and by health-care providers began with the launch of the Michigan Disease Surveillance System (MDSS). As part of an ongoing, multicenter, chronic hepatitis cohort study, investigators compiled clinical data from patients suspected to have chronic HBV or chronic HCV infection at any time during 2006–2008 and who resided in Michigan and sought care within the health-care network, which was comprised of several hospitals and clinics serving approximately 1 million Michigan residents (2). As of 2011, this health-care network had been documented by MDCH as a reporting institution for approximately 8% of all state reports of chronic HBV infection and 6% of all reports of chronic HCV infection since the advent of MDSS. According to U.S. Census data, the health-care network's patient population has a higher representation of blacks (37% versus 25%), females (57% versus 51%), persons aged ≥65 years (20% versus 13%), and persons aged <17 years (24% versus 20%) than the surrounding regional population.

Clinically confirmed cases of chronic HBV and chronic HCV infection were identified in the cohort study by methods that have been described previously (2) (Box). To evaluate the completeness of reporting of chronic HBV and HCV infections, all clinically confirmed cohort cases found in the health-care system during the study period of 2006–2008 were matched to cases reported to MDSS by first name, last name, and date of birth, using probabilistic record-linkage software. The year of initial diagnosis (i.e., the year of the first written diagnosis or laboratory evidence of infection) among cohort cases found in the health-care system ranged from 1995 to 2008.

Cases from the cohort study that matched in MDSS were queried for case classification in MDSS (acute, chronic, or both) and for the presence of age, sex, and risk factor data in MDSS. For cohort patients coinfected with HBV and HCV who met confirmation criteria for only one infection, only the diagnosis meeting definitive inclusion criteria was considered a case. Age, sex, and year of initial diagnosis were examined for their association with completeness of reporting. Differences between the proportions of confirmed cases reported, by age, sex, and race/ethnicity, were tested for statistical significance by chi-square test. Year-to-year differences in the proportions of cases reported were assessed for trend using the Cochrane-Armitage trend test and year of initial diagnosis.

In the cohort of 4,393 persons, 14% had HBV infections, 85% had HCV infections, and 1% had coinfections, yielding a total of 670 HBV and 3,796 HCV infection cases. Of the HBV infection cases, 597 (89%) met clinical confirmation criteria for chronic HBV infection (29 by physician diagnosis alone and 568 by laboratory criteria with or without a physician diagnosis). Of the HCV infection cases, 3,036 (80%) met clinical confirmation criteria for chronic HCV infection (115 by physician diagnosis alone and 2,921 by laboratory criteria with or without a physician diagnosis). A total of 490 (82%) of the 597 confirmed cases with chronic HBV infection were matched to MDSS, and 1,967 (65%) of the 3,037 confirmed cases with chronic HCV infection were matched. Of the cases matched to MDSS, sex was reported in 99.6% (488 of 490) of HBV infection cases and 99.0% (1,947 of 1,967) of HCV cases. Race/ethnicity was reported for 75.1% and 66.7% of cases, respectively. Risk factor data were reported for <5% of HCV infection cases because of inadequate health department resources for case follow-up. HBV infection risk factor data were not recorded because of the absence of risk factor–related questions in case questionnaires. Of the 597 chronic HBV infections, 463 (78%) were appropriately classified as chronic in MDSS. Of the 3,036 chronic HCV infections, 1,918 (64%) were appropriately classified as chronic in MDSS (Table 1).

Completeness of reporting of chronic HBV and HCV infection consistently improved over time and varied significantly by the year of diagnosis, with more complete reporting among cases with more recent diagnoses (p<0.001). Reporting of confirmed cases of HCV infection varied significantly by age group (p=0.001), sex (p=0.049), and race/ethnicity (p=0.024); reporting of these cases was more complete among persons aged 0–30 years, among males, and among non-Hispanic whites and Asians/Pacific Islanders (Table 2).

Reported by

Kim Kirkey, PhD, Karen MacMaster, MPA, Michigan Dept of Community Health. Anil Suryaprasad, MD, Fujie Xu, MD, Monina Klevens, DDS, Henry Roberts, PhD, Anne Moorman, MPH, Scott Holmberg, MD, Div of Viral Hepatitis, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention; Jenna Webeck, DVM, EIS officer, CDC. Corresponding contributor: Anil Suryaprasad, [email protected], 404-718-8571.

Editorial Note

This initial evaluation of viral hepatitis surveillance in Michigan showed that reporting of chronic HBV and HCV infections was incomplete. However, reporting has improved over time, with more recently diagnosed cases significantly more likely to be reported and included in state surveillance data, particularly after dual reporting by laboratories and health-care providers began in 2004. Incomplete reporting and demographic disparities in reporting of chronic HCV infections should be considered when using surveillance data to estimate actual disease burden. Information on risk factors for infection, which could inform prevention efforts, were seldom reported because of the constrained resources for case follow-up of HCV infections and the absence of risk factor–related questions in HBV case forms.

Case reporting of notifiable infectious conditions is intended to describe disease burden, facilitate case management, ascertain risk factors to prevent transmission, identify and curtail outbreaks, and monitor implementation and impact of public health recommendations (3). Several challenges complicate chronic hepatitis surveillance efforts. The number of cases eligible for reporting is large, and data management is burdened by the reporting of numerous laboratory tests meeting the case definition per individual case (4). A previous comprehensive evaluation of viral hepatitis surveillance programs underscored the need for additional resources to achieve better investigation and case management of reported chronic viral hepatitis infections (5). Because of the challenges of case reporting, few states reported chronic HBV cases (11 through passive surveillance and eight through active surveillance) and chronic HCV cases (eight through passive surveillance and eight through active surveillance) to CDC in 2010 (6).

Chronic viral hepatitis cases might not be reported for several reasons. First, many cases, particularly before 2008, were reported to health departments by fax, which has made the completeness subject to the limitations of manual entry of cases by health departments. Second, older cases might have been diagnosed before mandatory reporting was implemented. Third, cases are often diagnosed at outside institutions and referred to the health-care system; in the transfer of care, case reports might not have been made by the diagnosing institution.

The findings in this report are subject to at least four limitations. First, the matching of actual cases with MDSS is subject to potential misclassification of clinically confirmed cases by study investigators. Second, detection of cases in MDSS is subject to the limitations of matching, and at least some cases might be missed by changes in names or changes in residence. Third, reporting at the participating health-care system's facilities might not be representative of reporting at other clinical-care or testing centers. Finally, in some cases, the year of diagnosis might be different from the year of the report, so the observed trends in reporting should be interpreted with caution.

This evaluation was possible because the state health department and hospital officials were willing to take a critical look at reporting for purposes of quality improvement. As an important response to the Institute of Medicine report and action plan (1), this evaluation serves as a model for similar efforts in other states and, in fact, will be replicated in other states with facilities participating in the cohort study.

The improvements in reporting of chronic HBV and HCV infections in Michigan coincide with improvements statewide in automated laboratory reporting, and a more detailed investigation of the association between the two factors is warranted. The persisting gaps in reporting highlight the need for more efficient means of transferring and interpreting reportable data. In previous studies, electronic reporting has been shown to improve the reporting of notifiable diseases, including hepatitis (7–8), and might be a method for improving the quality of reporting. For example, investigators at the U.S. Department of Veterans Affairs found that ICD-9 codes for HCV infections were highly predictive of actual HCV infection in their administrative databases (9).

Given the complexity of chronic hepatitis surveillance and the limited resources available, public health authorities should explore new strategies to improve reporting, such as wider adoption of electronic reporting. This report offers a roadmap for using large datasets from clinical institutions to provide state and local health departments with insight into the disease burden represented by chronic viral hepatitis case reports. The findings suggest the need for exploration of additional data sources for risk factor information, especially because data in chronic viral hepatitis case reports might not reflect the current risk for secondary transmission. Such a critical evaluation of surveillance data can help inform efforts to improve linkages to care and to prevent viral hepatitis transmission.

References

1. Institute of Medicine. Hepatitis and liver cancer: a national strategy for prevention and control of HBV and C 2010. Washington, DC: National Academies Press; 2010. Available at http://www.iom.edu/reports/2010/hepatitis-and-liver-cancer-a-national-strategy-for-prevention-and-control-of-hepatitis-b-and-c.aspx.
2. Moorman AC, Gordon SC, Rupp LB, et al. Baseline characteristics and mortality among people in care for chronic viral hepatitis: the chronic hepatitis cohort study. Clin Infect Dis 2013;56:40–50.
3. Thacker SB. Historical development. In: Lee LM, Teutsch SM, Thacker SB, eds. Principles and practice of public health surveillance. New York, NY: Oxford University Press; 2010.
4. Klevens RM, Miller J, Vonderwahl C, et al. Population-based surveillance for hepatitis C virus, United States, 2006–2007. Emerg Infect Dis 2009;15:1499–502.
5. Fleming DT, Zambrowski A, Fong F, et al. Surveillance programs for chronic viral hepatitis in three health departments. Public Health Rep 2006;121:23–35.
6. CDC. Viral hepatitis surveillance—United States, 2009. Atlanta, GA: US Department of Health and Human Services, CDC; 2011. Available at http://www.cdc.gov/hepatitis/statistics/2009surveillance/index.htm.
7. Heisey-Grove DM, Church DR, Haney GA, DeMaria A Jr. Enhancing surveillance for hepatitis C through public health informatics. Public Health Rep 2011;126:13–8.
8. Overhage JM, Grannis S, McDonald CJ. A comparison of the completeness and timeliness of automated electronic laboratory reporting and spontaneous reporting of notifiable conditions. Am J Public Health 2008;98:344–50.
9. Kramer JR, Davila JA, Miller ED, Richardson P, Giordano P, El-Serag HB. The validity of viral hepatitis and chronic liver disease diagnoses in Veterans Affairs administrative databases. Aliment Pharm Therap 2008;
   27:274–82.