Estimating High-Priority Patients in the US With Chronic HCV
Estimating High-Priority Patients in the US With Chronic HCV
We used data from the 2003 to 2010 National Health and Nutrition Examination Survey (NHANES)—the most recent data available—to determine a national estimate of the number of people with chronic HCV infection in the United States. The NHANES is a nationally representative survey of the US civilian noninstitutionalized population conducted by the National Center for Health Statistics to assess the health and nutrition status of the US population. About 5000 noninstitutionalized civilians living in the 50 states and the District of Columbia are sampled each year. The NHANES includes interviews and a physical examination component. Information on demographics and other characteristics is collected through in-home interviews, and biological specimens are taken for laboratory analyses as part of the physical examination, which is conducted in a specially designed mobile examination center. Since 1999, this cross-sectional survey has been conducted continuously, with data releases every 2 years. Because of the low prevalence of HCV infection, we combined data from 4 cycles of the NHANES to produce reliable estimates.
For our analysis, we used these 4 cycles of NHANES to estimate the number of HCV-infected people (those with either diagnosed or undiagnosed HCV infection) according to stage of liver disease and comorbid conditions. In the NHANES, having chronic HCV infection is defined as being HCV RNA positive. The blood tests necessary for the calculation of FIB-4 score were performed as part of the NHANES, along with urinary test results and selected self-reported comorbid conditions. We used SAS-Callable SUDAAN software (release 11.0; RTI International, Research Triangle Park, NC) for analyses because of the complex survey design used in NHANES, and we incorporated the survey examination weights to make our estimates representative of the noninstitutionalized civilian US population. We adjusted the examination weights to account for missing HCV testing results for some participants and multiple NHANES cycles.
On the basis of the 2003 to 2010 NHANES data, we estimated the number of people with chronic HCV infection in the United States to be 2.7 million. Because of the low awareness about and asymptomatic nature of HCV infection, many people who may otherwise qualify for therapy have not been tested and identified. We assumed that about 50% of HCV-infected people remain undiagnosed.
We used data from the Chronic Hepatitis Cohort Study (CHeCS), an observational cohort of HCV-infected patients, which has been described in previous publications, to estimate the percentages of patients with diagnosed HCV infection meeting highest or high priority for treatment. Briefly, this cohort was drawn from among more than 2.7 million patients aged 18 years or older who had a clinical service visit (i.e., outpatient or inpatient, emergency department, or laboratory test) between January 1, 2006, and December 31, 2012, at 1 of 4 sites: Geisinger Health System, Danville, Pennsylvania; Henry Ford Health System, Detroit, Michigan; Kaiser Permanente-Northwest, Portland, Oregon; and Kaiser Permanente-Honolulu, Hawaii.
In CHeCS, patients were initially identified principally on the basis of laboratory criteria and secondarily on International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) criteria. Electronic health record data and administrative data were collected for each patient and supplemented with individual chart review by trained data abstractors, who also verified chronic HCV infection. Data collected included patient demographics, medical encounters, treatment data, and laboratory, radiology, and biopsy results. Because the purpose of this analysis was to describe characteristics of patients who are still infected with HCV and potentially eligible for therapy, we assessed liver disease and comorbidities on the basis of the latest clinical data for each cohort patient through December 31, 2012 (the most recent data available).
In estimating the percentages of patients with diagnosed HCV infection meeting highest and high priority for treatment, we excluded from our analysis CHeCs patients who had died, had achieved sustained virologic response, or had received a liver transplant. Although liver transplant recipients qualify for the highest priority for treatment under current guidance, the national number of liver transplant patients alive with HCV infection is relatively small. In 2012, approximately 65 000 individuals in the United States were living with transplanted livers, of whom, from 2008 to 2012, only 30% exhibited serologic evidence of HCV infection.
Liver fibrosis has traditionally been staged by biopsy; however, biopsy data are not collected by the NHANES. For NHANES participants and CHeCS patients who had not received a recent biopsy (limited to biopsy in 2004 or later), we approximated fibrosis stage by using the most current FIB-4 score. FIB-4 score has been found to increase with successive fibrosis levels and can be useful in monitoring patients who may need therapy or additional histologic evaluations. We calculated FIB-4 score with the following formula:
where AST = aspartate aminotransferase and ALT = alanine aminotransferase.
An FIB-4 score threshold of 2.5 or more was assumed to indicate a fibrosis level of METAVIR F3 or higher. We chose this threshold value on the basis of the upper limit of the 95% confidence interval (CI) for the mean FIB-4 score in approximately 400 HCV-infected patients who were staged at METAVIR F3 using biopsies (mean FIB-4 score = 2.32, 95% CI = 2.17, 2.47). Because the distribution of FIB-4 scores in HCV patients is skewed, we used the upper limit of the mean, which was rounded to 2.5. Similarly, a threshold of 1.6 was assumed to indicate a fibrosis level of METAVIR F2 or higher. For this analysis, because the METAVIR fibrosis scale categorizes HCV-infected people into mutually exclusive groups of METAVIR F2 or METAVIR F3, but not both, we assumed that only people whose FIB-4 score was 1.6 or higher but less than 2.5 were at the METAVIR F2 stage.
In a comparison scenario, we also included estimates when the METAVIR F3 threshold was changed from an FIB-4 score of 2.5 or more to 3.25 or more. Using the threshold value of 3.25 has a higher specificity for advanced fibrosis or cirrhosis but is not sensitive enough to rule out all people below this threshold who have substantial fibrosis; thus, using this higher cutoff yielded a conservative estimate (or underestimate) of people with moderate or worse fibrosis meeting the highest priority criteria. We calculated the FIB-4 score for patients in CHeCS from routine laboratory data available in patients' medical records; the clinician did not necessarily order the tests to obtain an FIB-4 score to assess a patient's need for treatment.
We identified HCV-infected individuals who met the highest or high-priority criteria for treatment on the basis of kidney damage or other relevant comorbid conditions, using laboratory tests or ICD-9-CM codes (Appendix A, available as a supplement to the online version of this article at http://www.ajph.org). Because of the absence of reliable data, we could not assess all conditions meeting the treatment criteria.
To simplify presentation, we rounded all estimates to the nearest thousand.
Methods
We used data from the 2003 to 2010 National Health and Nutrition Examination Survey (NHANES)—the most recent data available—to determine a national estimate of the number of people with chronic HCV infection in the United States. The NHANES is a nationally representative survey of the US civilian noninstitutionalized population conducted by the National Center for Health Statistics to assess the health and nutrition status of the US population. About 5000 noninstitutionalized civilians living in the 50 states and the District of Columbia are sampled each year. The NHANES includes interviews and a physical examination component. Information on demographics and other characteristics is collected through in-home interviews, and biological specimens are taken for laboratory analyses as part of the physical examination, which is conducted in a specially designed mobile examination center. Since 1999, this cross-sectional survey has been conducted continuously, with data releases every 2 years. Because of the low prevalence of HCV infection, we combined data from 4 cycles of the NHANES to produce reliable estimates.
For our analysis, we used these 4 cycles of NHANES to estimate the number of HCV-infected people (those with either diagnosed or undiagnosed HCV infection) according to stage of liver disease and comorbid conditions. In the NHANES, having chronic HCV infection is defined as being HCV RNA positive. The blood tests necessary for the calculation of FIB-4 score were performed as part of the NHANES, along with urinary test results and selected self-reported comorbid conditions. We used SAS-Callable SUDAAN software (release 11.0; RTI International, Research Triangle Park, NC) for analyses because of the complex survey design used in NHANES, and we incorporated the survey examination weights to make our estimates representative of the noninstitutionalized civilian US population. We adjusted the examination weights to account for missing HCV testing results for some participants and multiple NHANES cycles.
On the basis of the 2003 to 2010 NHANES data, we estimated the number of people with chronic HCV infection in the United States to be 2.7 million. Because of the low awareness about and asymptomatic nature of HCV infection, many people who may otherwise qualify for therapy have not been tested and identified. We assumed that about 50% of HCV-infected people remain undiagnosed.
People With Diagnosed HCV Infection
We used data from the Chronic Hepatitis Cohort Study (CHeCS), an observational cohort of HCV-infected patients, which has been described in previous publications, to estimate the percentages of patients with diagnosed HCV infection meeting highest or high priority for treatment. Briefly, this cohort was drawn from among more than 2.7 million patients aged 18 years or older who had a clinical service visit (i.e., outpatient or inpatient, emergency department, or laboratory test) between January 1, 2006, and December 31, 2012, at 1 of 4 sites: Geisinger Health System, Danville, Pennsylvania; Henry Ford Health System, Detroit, Michigan; Kaiser Permanente-Northwest, Portland, Oregon; and Kaiser Permanente-Honolulu, Hawaii.
In CHeCS, patients were initially identified principally on the basis of laboratory criteria and secondarily on International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) criteria. Electronic health record data and administrative data were collected for each patient and supplemented with individual chart review by trained data abstractors, who also verified chronic HCV infection. Data collected included patient demographics, medical encounters, treatment data, and laboratory, radiology, and biopsy results. Because the purpose of this analysis was to describe characteristics of patients who are still infected with HCV and potentially eligible for therapy, we assessed liver disease and comorbidities on the basis of the latest clinical data for each cohort patient through December 31, 2012 (the most recent data available).
In estimating the percentages of patients with diagnosed HCV infection meeting highest and high priority for treatment, we excluded from our analysis CHeCs patients who had died, had achieved sustained virologic response, or had received a liver transplant. Although liver transplant recipients qualify for the highest priority for treatment under current guidance, the national number of liver transplant patients alive with HCV infection is relatively small. In 2012, approximately 65 000 individuals in the United States were living with transplanted livers, of whom, from 2008 to 2012, only 30% exhibited serologic evidence of HCV infection.
Fibrosis Staging
Liver fibrosis has traditionally been staged by biopsy; however, biopsy data are not collected by the NHANES. For NHANES participants and CHeCS patients who had not received a recent biopsy (limited to biopsy in 2004 or later), we approximated fibrosis stage by using the most current FIB-4 score. FIB-4 score has been found to increase with successive fibrosis levels and can be useful in monitoring patients who may need therapy or additional histologic evaluations. We calculated FIB-4 score with the following formula:
where AST = aspartate aminotransferase and ALT = alanine aminotransferase.
An FIB-4 score threshold of 2.5 or more was assumed to indicate a fibrosis level of METAVIR F3 or higher. We chose this threshold value on the basis of the upper limit of the 95% confidence interval (CI) for the mean FIB-4 score in approximately 400 HCV-infected patients who were staged at METAVIR F3 using biopsies (mean FIB-4 score = 2.32, 95% CI = 2.17, 2.47). Because the distribution of FIB-4 scores in HCV patients is skewed, we used the upper limit of the mean, which was rounded to 2.5. Similarly, a threshold of 1.6 was assumed to indicate a fibrosis level of METAVIR F2 or higher. For this analysis, because the METAVIR fibrosis scale categorizes HCV-infected people into mutually exclusive groups of METAVIR F2 or METAVIR F3, but not both, we assumed that only people whose FIB-4 score was 1.6 or higher but less than 2.5 were at the METAVIR F2 stage.
In a comparison scenario, we also included estimates when the METAVIR F3 threshold was changed from an FIB-4 score of 2.5 or more to 3.25 or more. Using the threshold value of 3.25 has a higher specificity for advanced fibrosis or cirrhosis but is not sensitive enough to rule out all people below this threshold who have substantial fibrosis; thus, using this higher cutoff yielded a conservative estimate (or underestimate) of people with moderate or worse fibrosis meeting the highest priority criteria. We calculated the FIB-4 score for patients in CHeCS from routine laboratory data available in patients' medical records; the clinician did not necessarily order the tests to obtain an FIB-4 score to assess a patient's need for treatment.
Comorbid Conditions
We identified HCV-infected individuals who met the highest or high-priority criteria for treatment on the basis of kidney damage or other relevant comorbid conditions, using laboratory tests or ICD-9-CM codes (Appendix A, available as a supplement to the online version of this article at http://www.ajph.org). Because of the absence of reliable data, we could not assess all conditions meeting the treatment criteria.
To simplify presentation, we rounded all estimates to the nearest thousand.
