Hepatitis C–Related Mortality Among American Indian/Alaska Native Persons in the Northwestern United States, 2006-2012

Abstract

Objective:

American Indian and Alaska Native (AI/AN) persons are commonly misclassified in epidemiologic and administrative data sets. The race-corrected hepatitis C virus (HCV)–related mortality rate among AI/AN persons in the Northwest United States (Idaho, Oregon, and Washington State) is unknown. We quantified the disparity in HCV-related mortality between AI/AN persons and non-Hispanic white (NHW) persons in the Northwest during 2006-2012 after correcting misclassified AI/AN race.

Methods:

After conducting probabilistic record linkage between death records and the Northwest Tribal Registry, we calculated HCV-related mortality rates for AI/AN and NHW persons by using National Center for Health Statistics bridged-race estimates standardized to the US 2000 standard population.

Results:

The 2006-2012 aggregate age-adjusted HCV-related mortality rate per 100 000 population in the Northwest was 19.6 (95% confidence interval [CI], 17.3-22.2) for AI/AN persons and 5.9 (95% CI, 5.7-6.1) for NHW persons (rate ratio [RR] = 3.3; 95% CI, 3.0-3.8). The disparity was larger among females (RR = 4.6; 95% CI, 3.8-5.5) than among males (RR = 2.9; 95% CI, 2.5-3.4).

Conclusion:

The disproportionate rate of HCV-related mortality among AI/AN persons in the Northwest highlights the need to expand HCV education, screening, and treatment among this population.

Chronic hepatitis C virus (HCV) infection can be fatal, leading to cirrhosis in approximately 10% to 20% of patients; chronic HCV-infected patients with cirrhosis have a 1% to 5% annual risk for hepatocellular carcinoma.¹ Chronic HCV infection increases the risk of mortality from all causes, and decedents with hepatitis B virus (HBV) or HCV infection listed on their death certificates die approximately 20 years earlier than decedents without HBV or HCV infection listed on their death certificates.^2,3 HCV-related mortality is increasing in the United States; in 2012, it exceeded the number of deaths caused by 60 other nationally notifiable infectious conditions combined, including HIV.⁴ Among HCV-positive participants in the 2001-2008 National Health and Nutrition Examination Survey, 49.7% knew their HCV status before testing.⁵

Morbidity and mortality among American Indian and Alaska Native (AI/AN) persons are often underestimated because of misclassification of AI/AN race/ethnicity in epidemiologic and administrative data sets. Racial and ethnic information on the death certificate is obtained by the funeral director or coroner, who might obtain information on race/ethnicity from the decedent’s next of kin, from the medical record, or based on subjective observation. Several studies documented that information on race/ethnicity in vital statistics data sets shows excellent agreement with self-reported race/ethnicity for the white population but poor agreement for the AI/AN population, which can introduce bias toward underestimating AI/AN death rates.^6,7

Disparities in HCV-related mortality exist in the United States. Nationally, the HCV-related mortality rate among AI/AN persons was more than 2.5 times that of non-Hispanic white (NHW) persons in 2016.⁸ A national analysis that corrected for misclassification of AI/AN race also reported that the 1999-2009 age-adjusted HCV-related cirrhosis mortality rate among non-Hispanic AI/AN persons nationally was 2.5 times that of NHW persons, but the authors of that study were unable to report HCV-related cirrhosis mortality results by Indian Health Service (IHS) region or state of residence.⁹

Substantial regional differences exist in AI/AN health outcomes and health disparities.^10-12 In Oregon during 2009-2013 and Washington State during 2010-2014, the HCV-related mortality rate among AI/AN persons was twice that of NHW persons.^13,14 However, the race-corrected HCV-related mortality rate among AI/AN persons in the Northwest (which consists of Idaho, Oregon, and Washington State) is unknown. Hepatitis C–related mortality is affected by comorbidities that can differ by age and sex, such as coinfection with HIV or hepatitis A or B, metabolic diseases, and alcohol abuse.^15,16

The Northwest Portland Area Indian Health Board and the Portland Area Indian Health Service serve the 43 federally recognized tribes of Idaho, Oregon, and Washington State. Approximately 362 000 AI/AN persons live in the Northwest, representing 2.9% of the total population in the Northwest and 6.8% of the total AI/AN population in the United States. Compared with the total population in the Northwest, AI/AN persons in the Northwest are younger, and a larger proportion (27%) live in rural areas.^17,18 Given the heterogeneity in health outcomes of AI/AN persons by region and the regional organization of clinical and public health services for AI/AN persons, regional analyses are useful in guiding public health practice.

To address these gaps in knowledge on race-corrected HCV-related mortality among AI/AN persons in the Northwest, we conducted a region-specific, descriptive study of HCV-related mortality. We corrected for misclassification of AI/AN race to produce a more accurate estimate of HCV-related mortality among AI/AN residents of the Northwest during 2006-2012, compared with HCV-related mortality estimates that do not correct for misclassification of AI/AN race. We describe disparities in HCV-related mortality between AI/AN persons and NHW persons in the Northwest by age and sex because risk factors for hepatitis C-related infection and mortality differ by age and sex.

Methods

The Northwest Tribal Registry (NTR) is a demographic roster of patients who have received care at IHS, tribal, and urban Indian health facilities in the Northwest. It is a unique data source maintained by the Northwest Portland Area Indian Health Board in collaboration with the Portland Area Indian Health Service. To receive care at IHS and tribal facilities as an AI/AN beneficiary, patients provide proof of tribal membership or descendancy from a tribal member of a federally recognized tribe. At urban Indian health facilities, many of which are federally qualified health centers, data on race are generally collected by self-report. Records indicating a race other than AI/AN are excluded from the NTR. AI/AN race is a political designation that, unlike for other racial/ethnic groups in the United states, confers treaty obligations of the US government and eligibility for health care services; therefore, patients included in the NTR are considered AI/AN. Not all AI/AN persons who are eligible seek care at IHS, tribal, and urban Indian health facilities. As such, the NTR is not a complete list of all AI/AN persons in the Northwest, nor is it a compilation of tribal enrollment data; it represents an estimated 76% of the AI/AN census population in the Northwest.¹⁹

We obtained data on death certificates from the Idaho Bureau of Vital Records and Health Statistics and the Oregon and Washington State Centers for Health Statistics. Northwest Portland Area Indian Health Board statisticians corrected death records from each state for AI/AN misclassification through probabilistic record linkage with the NTR using Link Plus.²⁰ We matched death records to the NTR by using social security number, date of birth, name (last, first, middle), and sex. Two Northwest Portland Area Indian Health Board staff members reviewed all potential matches to identify true matches between the 2 data sets.⁷ We classified decedents as AI/AN if they were classified as AI/AN on the death certificate or if the record matched with the NTR. We combined death records from each of the 3 states into a single data set for analysis. Death certificate data contain death records of residents and nonresidents whose deaths occurred in state and death records of residents whose deaths occurred out of state. We included deaths of residents of Idaho, Oregon, and Washington State (including resident deaths that occurred out of state) during 2006-2012 and excluded deaths of nonresidents.

An HCV-related death was defined as a record with International Classification of Diseases, Tenth Revision²¹ codes for acute hepatitis C (B17.1) or chronic hepatitis C (B18.2) as the underlying cause of death or among the contributing causes of death. We calculated mortality rates by using the National Center for Health Statistics bridged-race estimates (vintage 2014 postcensal series) as population denominators.²² Mortality rates were age adjusted to the US year 2000 standard population by using groups aged ≤44, 45-54, 55-64, and ≥65; 95% confidence intervals (CIs) were calculated using the gamma distribution.^23,24 We calculated rate ratios (RRs) and 95% CIs to compare age-adjusted mortality rates between AI/AN persons and NHW persons. We also calculated crude age-specific rates and RRs. We did not tabulate age-specific mortality rates for Idaho because cell counts were <5. We stratified age-adjusted mortality rates and RRs by sex and comorbidities (eg, HIV or hepatitis B coinfection). We used negative binomial regression to examine the significance of the association between HCV-related mortality and race, after adjusting for age, with race and sex as an interaction term in the model. We also used negative binomial regression to investigate the trend in age-adjusted mortality rates and RRs over time. We calculated P values of individual regression coefficients by using the Wald χ² method, with P < .05 considered significant. We completed age-adjusted and age-specific analyses by using SAS version 9.4.²⁵ We computed age-specific 95% CIs by using the Byar approximation of Poisson in OpenEpi version 3.01.²⁶ The Centers for Disease Control and Prevention (CDC) reviewed this project for human subjects protection and deemed it to be nonresearch.

Results

HCV-Related Mortality

We identified 286 HCV-related deaths among AI/AN residents and 5358 HCV-related deaths among NHW residents in the Northwest during 2006-2012, resulting in age-adjusted mortality rates per 100 000 population of 19.6 (95% CI, 17.3-22.2) for AI/AN persons and 5.9 (95% CI, 5.7-6.1) for NHW persons (Table 1). The 2006-2012 aggregate age-adjusted HCV-related mortality rate per 100 000 population was highest among AI/AN residents of Washington State (22.3; 95% CI, 19.0-26.1), followed by Oregon (18.7; 95% CI, 14.9-23.5) and Idaho (8.0; 95% CI, 4.1-14.8). Overall, the age-adjusted HCV-related mortality rate for AI/AN persons in the Northwest was 3.3 (95% CI, 3.0-3.8) times that of NHW persons. The disparity in HCV-related mortality between AI/AN persons and NHW persons was largest in Washington State (RR = 3.9; 95% CI, 3.3-4.5), followed by Idaho (RR = 3.2; 95% CI, 1.9-5.6) and Oregon (RR = 2.5; 95% CI, 2.0-3.1). Thirty-three of the 286 (11.5%) HCV-related deaths among AI/AN residents were originally classified as non-AI/AN in death certificates but were identified as AI/AN after linkage with the NTR.

Table 1.

Age-adjusted hepatitis C–related mortality rates and rate ratios for American Indian and Alaska Native (AI/AN) and non-Hispanic white (NHW) persons, by region, state, and sex, Northwestern United States, 2006-2012^a

Variable	AI/AN		NHW		AI/AN: NHW Rate Ratio (95% CI)
	No.	Rate per 100 000 Population (95% CI)	No.	Rate per 100 000 Population (95% CI)
Northwestb
Female	116	15.5 (12.6-18.8)	1559	3.4 (3.2-3.6)	4.6c (3.8-5.5)
Male	170	24.1 (20.4-28.4)	3799	8.2 (8.2-8.8)	2.9c (2.5-3.4)
Total	286	19.6 (17.3-22.2)	5358	5.9 (5.7-6.1)	3.3 (3.0-3.8)
Idahoc
Total	13	8.0 (4.1-14.8)	289	2.5 (2.2-2.8)	3.2 (1.9-5.6)
Oregon
Female	39	16.9 (11.7-23.9)	649	4.2 (3.9-4.6)	4.0 (2.9-5.5)
Male	51	20.6 (15.1-27.9)	1622	10.8 (10.3-11.4)	1.9 (1.4-2.5)
Total	90	18.7 (14.9-23.5)	2271	7.4 (7.1-7.7)	2.5 (2.0-3.1)
Washington State
Female	73	16.9 (13.1-21.6)	812	3.3 (3.1-3.6)	5.1 (4.0-6.5)
Male	110	28.5 (23.0-35.0)	1986	8.3 (7.9-8.7)	3.4 (2.8-4.2)
Total	183	22.3 (19.0-26.1)	2798	5.8 (5.5-6.0)	3.9 (3.3-4.5)

^aData source: Death records from the Idaho Bureau of Vital Records and Statistics, Oregon Center for Health Statistics, and Washington Center for Health Statistics. Death records were linked to the Northwest Tribal Registry to correct misclassified AI/AN records by the Northwest Portland Area Indian Health Board.

^bIdaho, Oregon, and Washington State.

^cSex-specific data suppressed because of cell counts <5.

Age- and Sex-Stratified HCV-Related Mortality

Age-adjusted mortality rates differed by sex (P for interaction = .01). Overall, the age-adjusted HCV-related mortality rate was higher among males than females; however, the RR comparing AI/AN persons with NHW persons was higher among females (RR = 4.6; 95% CI, 3.8-5.5) than among males (RR = 2.9; 95% CI, 2.5-3.4; Table 1). The small number of HCV-related deaths indicating HIV or hepatitis B coinfection prevented a robust analysis of age-adjusted mortality rates among decedents with HCV and HIV or hepatitis B coinfection.

In the Northwest overall and in each state, most HCV-related deaths among AI/AN and NHW persons occurred among adults aged 45-54 and 55-64 (Table 2). The HCV-related mortality rate was lowest among AI/AN and NHW persons aged ≤44, but the disparity in HCV-related mortality between AI/AN persons and NHW persons was largest in this age category (RR = 4.5; 95% CI, 3.1-6.6).

Table 2.

Age-stratified number and aggregate rate^a and rate ratios (RRs)^b of hepatitis C virus–related deaths among American Indian and Alaska Native (AI/AN) and non-Hispanic white (NHW) persons in the Northwest,^c Oregon, and Washington State, 2006-2012^d

Region and Age, y	AI/AN			NHW			AI/AN: NHW, RR (95% CI)
	No.	Rate (95% CI)	RR (95% CI)	No.	Rate (95% CI)	RR (95% CI)
Northwest
≤44	29	2.3 (1.5-3.2)	1.0 [Reference]	216	0.5 (0.4-0.6)	1.0 [Reference]	4.5 (3.1-6.6)
45-54	107	48.6 (39.9-58.8)	21.6 (14.3-32.6)	1604	14.9 (14.1-15.6)	29.8 (25.9-34.4)	3.3 (2.7-4.0)
55-64	108	72.1 (59.2-87.1)	32.0 (21.3-48.3)	2675	28.3 (27.2-29.4)	56.6 (49.3-65.0)	2.6 (2.1-3.1)
≥65	42	41.7 (30.0-56.4)	18.5 (11.6-29.8)	863	8.6 (8.1-9.2)	17.3 (14.9-20.0)	2.3 (1.7-3.1)
Oregon
≤44	13	3.3 (1.7-5.6)	1.0 [Reference]	19	1.0 (0.6 -1.5)	1.0 [Reference]	3.4 (1.7-6.9)
45-54	34	49.0 (33.9-68.5)	15.1 (8.0-28.6)	93	19.6 (15.8-24.0)	20.6 (12.6-33.8)	2.5 (1.7-3.7)
55-64	33	68.7 (47.3-96.5)	21.1 (11.1-40.2)	202	40.5 (35.1-46.5)	42.6 (26.6-68.2)	1.7 (1.2-2.5)
≥65	10	31.8 (15.2-58.4)	9.8 (4.3-22.3)	60	10.8 (8.3-13.9)	11.4 (6.8-19.0)	2.9 (1.5-5.8)
Washington State
≤44	15	2.0 (1.1-3.4)	1.0 [Reference]	10	0.3 (0.1-0.6)	1.0 [Reference]	6.8 (3.0-15.1)
45-54	67	53.1 (41.2-67.5)	26.2 (14.9-45.8)	104	12.6 (10.3-15.3)	42.0 (22.0-80.4)	4.2 (3.1-5.7)
55-64	71	83.1 (64.9-104.9)	40.9 (23.5-71.5)	247	31.7 (27.9-36.0)	105.7 (56.2-198.8)	2.6 (2.0-3.4)
≥65	30	52.1 (35.1-74.5)	25.7 (13.8-47.7)	107	13.0 (10.7-15.7)	43.3 (22.7-82.9)	4.0 (2.7-6.0)

^aCrude rate per 100 000 population.

^bCalculated using the Byar approximation Poisson method using Open Epi.²⁶

^cIdaho, Oregon, and Washington State. Data for Idaho are not reported because cell counts in age categories were <5.

^dData source: Death records from the Idaho Bureau of Vital Records and Statistics, Oregon Center for Health Statistics, and Washington Center for Health Statistics. Death records were linked to the Northwest Tribal Registry to correct misclassified AI/AN records by the Northwest Portland Area Indian Health Board.

HCV-Related Mortality Over Time

The age-adjusted HCV-related mortality rate among AI/AN persons in the Northwest increased significantly during 2006-2012 (P = .049; Figure). We found a nonsignificant increase in the age-adjusted HCV-related mortality rate for NHW persons. The disparity in HCV-related mortality between AI/AN persons and NHW persons persisted but did not change significantly over time.

Figure.

Age-adjusted hepatitis C–related mortality rate among American Indian and Alaska Native (AI/AN) and non-Hispanic white (NHW) persons, by year, Idaho, Oregon, and Washington State, 2006-2012. Error bars indicate 95% confidence intervals. Data source: Death records from the Idaho Bureau of Vital Records and Statistics, Oregon Center for Health Statistics, and Washington Center for Health Statistics. Death records were linked to the Northwest Tribal Registry to correct misclassified AI/AN records by the Northwest Portland Area Indian Health Board.

Discussion

The disparity in HCV-related mortality between AI/AN persons and NHW persons in the Northwest was larger than previously reported. We found that the race-corrected 2006-2012 aggregate age-adjusted mortality rate for AI/AN persons residing in Idaho, Oregon, or Washington State was 19.6 per 100 000 population, which exceeded the 2016 national HCV-related mortality rate among AI/AN persons of 10.75 per 100 000 population.⁸ The disparity in HCV-related mortality between AI/AN persons and NHW persons in the Northwest during 2006-2012 also exceeded the disparity observed at the national level during 2012.⁸ National rates are not corrected for misclassification of AI/AN race. It is unclear if the discrepancy between our estimate and national estimates resulted from differences in how AI/AN persons are identified in regional and national estimates or if they reflect real differences in the burden of HCV infection, comorbidities, or screening and treatment.

For both AI/AN and NHW residents of the Northwest, the crude HCV-related mortality rate was highest among adults aged 45-54 and 55-64. However, the disparity in HCV-related mortality was largest among decedents aged ≤44, although the estimate was imprecise because of few deaths in this age category. Both AI/AN and NHW males had a higher mortality rate than females, but the disparity between AI/AN persons and NHW persons was significantly larger among females. This finding is consistent with other race-corrected analyses that reported that AI/AN disparities in chronic liver disease mortality were largest among younger age groups and females.⁹

The overall HCV-related mortality disparity between AI/AN persons and NHW persons and the age- and sex-specific disparities might reflect differences in risk factors between groups; these could include risk factors for transmission, such as health care–related exposures and injection drug use, or risk factors for rapid disease progression and death, such as access to care, alcohol abuse, or comorbidities. Among persons with chronic HCV infection, disease progression is nonlinear; the disease course and response to treatment can be affected by comorbidities, such as coinfection with HIV or hepatitis A or B, metabolic diseases, and alcohol abuse.^15,16,27 Risk factors were identified in only 958 of 2967 (32%) case reports from newly identified HCV-infected persons nationally in 2016.²⁸ We attempted to quantify risk factors such as HIV or hepatitis B coinfection among AI/AN and NHW decedents, but the small number of HCV-related deaths limited our analyses to overall, age-specific, and sex-specific differences between AI/AN persons and NHW persons. Although the small number of HCV-related deaths makes it difficult to determine the possible cofactors contributing to the elevated HCV-related mortality rate for the younger age group, the elevated disparity in HCV-related mortality between AI/AN and NHW persons aged ≤44, compared with the disparity in HCV-related mortality between AI/AN and NHW persons in other age groups, likely represents infection acquired via injection drug use and rapid disease progression related to comorbidities. This, in turn, demonstrates the need for early prevention education and expanded screening and treatment in AI/AN communities. The greater disparity in HCV-related mortality between AI/AN and NHW females compared with the disparity in HCV-related mortality between AI/AN and NHW males might reflect sex-related differences in health care–seeking behavior, comorbidities, or behavioral risk factors. However, we were unable to examine the reasons for the observed difference in disparity by age or sex in this study.

We found that the HCV-related mortality rate among AI/AN persons in the Northwest increased by approximately 88% from 2006 to 2012 (from 13.7 to 25.7 per 100 000 population), and the HCV-related mortality disparity between AI/AN and NHW persons persisted over time. This finding differs from a 23.2% increase in the national HCV-related mortality rate for AI/AN persons during 2009-2013.²⁹ The persistence of this disparity highlights the importance of comprehensive screening and treatment efforts at IHS, tribal, and urban Indian health facilities in the Northwest. Others have documented the role that socioeconomic status and access to health care play in HCV screening, diagnosis, and treatment for racial/ethnic minority groups.³⁰ Linkage to specialized care is important in this region because of the lack of a regional IHS or tribally operated hospital where specialized care might be available.

Consideration of HCV genotype is important when interpreting HCV-related outcomes. In an analysis of Kaiser Permanente patients, genotype distribution was similar between AI/AN persons and NHW persons. However, among those with genotype 1, AI/AN persons were more likely than NHW persons to have genotype subtype 1a than genotype subtype 1b.³¹ The standard treatment for HCV infection before direct-acting antiviral medications resulted in a better sustained virologic response among genotype subtype 1a patients.³² Therefore, the disparity is unlikely to be explained by differences in the prevalence of specific genotypes between the 2 populations.

Differences in screening practices at IHS facilities may also explain the difference in HCV-related mortality rates between AI/AN persons and NHW persons, because increased screening and diagnosis might lead to more complete HCV reporting on death certificates. An electronic health record reminder was developed for birth cohort testing in IHS facilities in August 2011; screening and testing strategies were implemented at IHS facilities after CDC recommendations in August 2012. These efforts quadrupled the number of unique birth cohort patients tested during 2012-2015.³³ However, because adoption of the CDC recommendations and the widespread use of screening reminders and other strategies to increase birth cohort testing did not occur until late 2012, they did not affect the mortality disparity observed in our study.

Limitations

This study had several limitations that may have affected our estimate of HCV-related mortality. First, HCV infection is likely underreported on death certificates. Among a cohort of confirmed and well-characterized chronic HCV-infected decedents, only 19% had HCV listed as a cause of death.³⁴ Second, we used contributing and multiple cause-of-death fields to assess HCV-related mortality; it is possible that some HCV-related deaths identified in the multiple cause-of-death fields did not directly contribute to death but were still noted on the death certificate, which would lead to an overestimation of HCV-related mortality. Third, the NTR might underestimate the AI/AN population, because it includes only AI/AN persons who accessed services at Northwest IHS, tribal, or urban Indian health facilities and is limited to AI/AN persons who are members or descendants of federally recognized tribes. An evaluation of the 2011 NTR reported that it represented approximately 76% of the census population in the Northwest, but urban AI/AN persons might be less represented.¹⁹ Therefore, it is uncertain to what degree our estimates represented the true HCV-related mortality rate among AI/AN persons in the Northwest. Despite this limitation, probabilistic linkage between the NTR and the Idaho, Oregon, and Washington State death certificates resulted in the identification of 33 additional HCV-related deaths among AI/AN persons, increasing the estimated AI/AN HCV-related age-adjusted mortality rate from 17.5 to 19.6 deaths per 100 000 population.

Fourth, we were unable to correct for misclassification of NHW race/ethnicity; however, previous studies documented that the quality of data on race for white persons is excellent in vital statistics data sets.⁶ Misclassification of death records for NHW persons, with the exception of AI/AN records misclassified as NHW, was unlikely to have a major effect on our estimates, especially given the relative population size of NHW persons compared with AI/AN persons. Finally, the small number of HCV-related deaths indicating HIV or hepatitis B coinfection prevented a robust analysis of certain risk factors for HCV-related mortality; as a result, we were not able to assess factors that might contribute to the observed disparity in HCV-related mortality among AI/AN females and persons aged ≤44.

Conclusions

This AI/AN race-corrected, regional analysis improves our understanding of HCV-related mortality among AI/AN persons in the Northwest. Implementation of similar race-corrected morbidity and mortality estimates for AI/AN persons might be beneficial for other regions where AI/AN persons are commonly misclassified in administrative and epidemiologic data sets. The disparity in HCV-related mortality and its persistence highlight the need for improved prevention, diagnosis, and treatment services for AI/AN persons in the Northwest. At least 1 federally recognized tribe in the United States has embarked on a plan to eliminate HCV infection, including expanded screening of adults aged 20-69 and implementation of community-based interventions for persons who inject drugs, with a goal of treating 85% of their patients during a 3-year period.³⁵ If widely adopted, this approach might narrow the disparity in HCV-related mortality for AI/AN persons. For this approach to succeed, a national strategy that includes increased access to treatment for the approximately 123 000 AI/AN persons living with HCV infection in the United States would be needed.^36,37

These results can serve as a baseline for evaluating the effectiveness of increased HCV screening and treatment efforts at IHS, tribal, or urban Indian health facilities. Continued surveillance of HCV-related mortality among AI/AN persons in the Northwest is needed, and it is important to also include efforts to improve the identification of AI/AN persons in state and national epidemiologic and administrative data sets. Patients born after 1965 are usually screened only if they develop symptoms or have identified risk factors associated with increased risk for acquiring HCV infection. Expanded screening to identify and treat those acquiring infection at younger ages and more research to understand the increased disparity in HCV-related mortality among AI/AN women are needed.³⁸ In May 2019, IHS passed a policy to expand universal screening for HCV infection to include all adults aged 18 and older at least once, regardless of risk factors.³⁹