Alcoholism, Hepatitis B and C Viral Infections, and ...

/ffl

American Journal of Epidemiology Copyright O 1996 by The Johns Hopkins University School of Hygiene and Public Health All rights reserved

Vol. 143, No. 9 Printed In U.S.A.

Alcoholism, Hepatitis B and C Viral Infections, and Impaired Liver Function among Taiwanese Aboriginal Groups

Chen-Yang Shen,1 Hsuan-Shu Lee,2 Lu-Chin Huang,3 Keh-Sung Tsai,4 Ding-Shinn Chen,2 and Andrew T. A. Cheng 1

alanine aminotransferase; alcoholism; aspartate aminotransferase; hepatitis B virus; hepatitis C viruses

The hepatopathogenic effects of excessive alcohol consumption are well recognized. In Western societies, the prevalence of liver disease (mostly cirrhosis) is highest in those countries that have the highest per capita consumption of alcohol (reviewed in ref. 1). Alcoholic liver disease is the most common medical complication of alcoholism, accounting for a majority of the cases of cirrhosis of the liver in industrialized countries (2). However, in Asia and Africa the prevalence of alcoholism is much lower, and, by contrast, most liver diseases and cirrhosis are due to chronic infection with hepatitis viruses (2). The acquisition rates of chronic hepatitis B virus and hepatitis C virus infections among the general population in Taiwan can reach as high as 15-20 percent and 1-5 percent, respectively (3). These viral factors are thought to con-

tribute to the high morbidity and mortality from liver diseases in the Taiwanese population (4). Apart from the dominant Han Chinese who migrated from Mainland China from the early 18th century until the end of World War U, there are nine aboriginal groups in Taiwan, all of which are descended from Malayo-Polynesian stock and speak languages stemming from the Austronesian family. The current combined population of these nine tribes is about 330,000, constituting 1.7 percent of the total population of Taiwan (5). Appreciable health disparities exist among the different ethnic groups in Taiwan, and the health status of the Taiwanese aborigines is relatively poor. Among the various health problems suffered by the aboriginal minority, liver diseases including hepatocellular carcinoma, liver cirrhosis, and hepatitis appear to be some of the most serious. Although official ethno-specific death rates for aboriginal populations are not available, the counties inhabited predominantly by aborigines have the highest death rates due to liver diseases (about 80-120/ 100,000), and we estimated that these rates are 3-8 times higher than those in the whole population of Taiwan (calculated based on ref. 6).

Received for publication May 10,1995, and infinalform February 15, 1996. Abbreviation: Cl, confidence interval. 1 1nstitute of Biomedlcal Sciences, Academia Sinica, Taipei, Taiwan. 2 Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan. 3 Department of Internal Medicine, Buddhist Tzu-Chi General Hospital, Hualien, Taiwan. 4 Department of Laboratory Medicine, National Taiwan University Hospital, Taipei, Taiwan. Reprint requests to Dr. Andrew Cheng, Institute of Biomedlcal Sciences, Academia Sinica, Taipei 11529, Taiwan.

Two major causes possibly account for the high prevalence of liver diseases among the Taiwanese aborigines. First, a high prevalence of infection by 936

Downloaded from aje.oxfordjournals.org by guest on July 13, 2011

Viral hepatitis and alcoholism prevail in four major Taiwanese aboriginal groups. To study the relative importance of the acquisition of hepatitis B virus or hepatitis C virus infection and alcoholism to the presence of impaired liver function in these groups, the authors conducted a semistructured clinical interview for alcoholism and test for seromarkers for viral hepatitis among 993 cohort members enrolled in 1990-1992 in an ongoing prospective study (Taiwan Aboriginal Study Project). The subjects' blood specimens were tested for serum alanine aminotransferase/aspartate aminotransferase levels and for the presence of hepatitis B surface antigen and anti-hepatitis C virus antibody. The prevalence of a combination of an alanine aminotransferase level of >35 lU/lrter and an aspartate aminotransferase level of >40 ILJ/liter, implying impaired liver function or advanced liver disease, was 4.3% overall. Univariate and multiple logistic regression analysis showed that, rather than chronic hepatitis B virus infection, hepatitis C virus infection and alcoholism were the two dominant risk factors that signalled the risk of liver damage among these Taiwanese aborigines. In addition, these two contributing factors were able to act synergistically to cause impaired liver function. Am J Epidemiol 1996;143: 936-42.

Alcoholism, Viral Infection, and Impaired Liver Function

different hepatitis viruses, including those transmitted by body fluids or the fecal-oral route, has recently been consistently reported in these aboriginal populations (7-9). Of these viruses, hepatitis B virus and hepatitis C virus appear to be of particular importance, because they can cause persistent infection, resulting in prolonged and chronic health effects. Second, very high prevalences of alcoholism (44.5-54.5 percent) among four major aboriginal groups in Taiwan have been reported in a recent epidemiologic survey (5). The coexistence of high prevalences of both viral hepatitis and alcoholism among Taiwanese aborigines provides a unique opportunity for scientific investigation of their individual and combined effects on the risk of liver diseases. This study reports such an investigation. MATERIALS AND METHODS Study populations

Am J Epidemiol

Vol. 143, No. 9, 1996

during the follow-up period. Only four subjects were lost to the follow-up, with the remaining subjects participating in the phase II survey {n = 929). A total of 915 members received laboratory tests for blood chemistry, liver function, and infection with hepatitis B and C viruses. Case definition and case ascertainment

In this study, we used alcoholic status (normal or alcoholic) to represent high or low exposure to alcohol among study subjects. The design and standardization of the clinical interview section for alcoholism were based on our preliminary observations of drinking attitudes and behavior among the aborigines. This section covered all the symptoms, and their duration, of alcohol use disorders listed in both the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (12), and the Diagnostic and Statistical Manual of Mental Disorders: DSM1I1R (13). Useful information concerning respondents' drinking problems was also collected from their family members and significant others. In a reliability study of our semistrucrured interview, the generalized kappa for the diagnoses of normal, harmful, and dependent use of alcohol was 0.80 among psychiatrists and physicians and 1.0 between Cheng and the two research assistants (both the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, and the Diagnostic and Statistical Manual of Mental Disorders: DSMIIIR criteria) (5). On the basis of possible damage to liver cells by alcohol and subsequent influence on liver function, the term "alcoholism" in this study includes only those who were currently in the categories of "harmful use of alcohol" and "alcohol dependence" (in the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision); the term does not include those who previously belonged to these two categories, i.e., formerly alcohol dependent but abstinent for more than 6 months before the investigation. Serum specimens were collected a few weeks before the interview. All participants' fresh sera were tested by radioimmunoassay to detect their hepatitis B virus carrier status (identified by the presence of hepatitis B viral surface antigen; Abbott Laboratories, Abbott Park, Illinois) and by enzyme-linked immunosorbent assay for the presence of anti-hepatitis C virus antibody (the second generation hepatitis C virus kit; Abbott Laboratories). On the basis of our own experience (14,15), the performance of this anti-hepatitis C virus antibody assay was satisfactory. The presence of serum antibodies to hepatitis C virus indicates acquisition of hepatitis C virus infection. Fresh blood sam-


This study is part of an ongoing prospective study (the Taiwan Aboriginal Study Project) that has been in operation since 1986. The original cohort of the Taiwan Aboriginal Study Project consists of 993 members randomly selected from four dominant aboriginal groups in Taiwan. The four groups are the Ami, Atayal, Bunun, and Paiwan. We selected these groups because: 1) they make up nearly 90 percent of the total aboriginal population in Taiwan; 2) they represent ethnic groups that have different degrees of acculturation; and 3) they represent different lineages in family/social structure. The Taiwan Aboriginal Study Project focuses mainly on morbidity risks, clinical manifestations, the course and outcome of major psychoses, depression, alcohol use disorders, suicide and accidental death, and biologic and sociocultural risk factors of these morbidities. Detailed descriptions of the Taiwan Aboriginal Study Project and its epidemiologic findings on alcoholism and other mental disorders in phase I cross-sectional surveys (1986-1988) appear elsewhere (5, 10, 11). The field work at phase I consisted of continuing ethnographic observations of the surveyed communities and detailed interviews of study subjects. The latter covered sociodemographic information, health status, and the extent of acculturation. Psychiatric assessment using a semistructured clinical interview that covers alcoholism was conducted by Cheng with the help of local public health workers (response rates, 96.5-99.2 percent). The phase II survey, conducted in 1990-1992, applied the same clinical interview for alcoholism and a follow-up inventory covering changes in various life domains since phase I. Two well-trained research assistants also participated in the interview. Sixty members died

937

938

Shen et a).

pies were also tested to determine their levels of serum alanine aminotransferase (which is also called serum glutamic pyruvic transaminase) and serum aspartate aminotransferase (which is also called serum glutamic oxaloacetic transaminase) by a commercial chemical analyzer (Model 7450; Hitachi, Tokyo, Japan). In this study, we defined the presence of an "elevated level of alanine aminotransferase/aspartate aminotransferase" as alanine aminotransferase >35 IU/liter and aspartate aminotransferase >40 IU/liter, and this definition was considered to indicate impaired liver function or advanced liver disease on the basis of routine diagnostic criteria in Taiwan. All of the tests were performed in the same laboratory at the National Taiwan University Hospital. Statistical analysis

RESULTS

The prevalence (406/915, 44 percent) of alcoholism was appreciably high in each of our four aboriginal groups: 48.6 percent in Atayal, 41.7 percent in Ami, 44.8 percent in Bunun, and 42.8 percent in Paiwan. The frequency distributions of alanine aminotransferase levels of 50 IU/liter among 915 study participants were 857 (93.7 percent), 32 (3.5 percent), and 26 (2.8 percent), respectively. The frequency distributions of aspartate aminotransferase levels of < 4 0 , 4 0 - 4 9 , 50-100, and >100 were 810 (88.5 percent), 26 (2.8 percent), 62 (6.8 percent), and 17 (1.9 percent). Taken together, the proportion of subjects

TABLE 1. Risk factors In relation to elevated serum alanine aminotransferase/aspartate aminotransferase (ALT/AST) level* in four aboriginal groups In Taiwan, 1990-1992* No. with elevated ALT/AST level per no. tested

Prevalence ratio

Sex Male Female

29/462 (6.3)t 10/453(2.2)

2.8 1.0

1.4-5.8

Age (years) 50

33/643(5.1) 6/272 (2.2)

2.3 1.0

1.0-5.5

Group Ami Atayal Bunun Paiwan

7/232 (3.0) 18/210(8.6) 8/239 (3.4) 6/234 (2.6)

1.2 3.3 1.4 1.0

0.4-3.5 1.4-8.3 0.5-3.8

HBsAgt Positive Negative

15/215 (7.0) 24/700 (3.4)

2.0 1.0

1.1-3.8

HCVf antibody Positive Negative

7/26 (26.9) 32/889 (3.6)

7.5 1.0

3.6-15.4

Current alcohol use Alcoholic Nonalcoholic§

29/406(7.1) 10/509 (2.0)

3.6 1.0

1.8-7.4

Factor

95% Clt

* An elevated ALT/AST level was defined as alanine aminotransferase > 35 IU/liter and aspartate aminotransferase > 40 IU/liter. t Cl, confidence Interval; HBsAg, hepatitis B surface antigen; HCV, hepatitis C virus. $ Numbers In parentheses, percentage. § Current nonalcoholic includes "ex-alcohol dependent" (formerly alcohol dependent but had been abstinent for over 6 months") as well as lifetime nonalcoholics.

Am J Epidemiol

Vol. 143, No. 9, 1996


Univariate analysis and multiple logistic regression analysis were done to determine the risk factors for an elevated level of alanine aminotransferase/aspartate aminotransferase. Univariate analysis was done on all suspected risk factors using chi-square tests, and odds ratios and corresponding 95 percent confidence intervals were estimated. The consideration of biologic plausibility was the most important criterion for inclusion of variables in the logistic model. A backward elimination strategy was used to select the optimal model (16). In addition, a logistic model containing a set of dummy variables to define various combinations of anti-hepatitis C virus positivity and the status of alcoholism was used to evaluate the possible interaction between an elevated level of alanine aminotransferase/aspartate aminotransferase and both hepatitis C virus infection and alcohol consumption among our study participants. Mantel's chi-square test (17) for trend was also used to examine the dose-response relation for the risk estimates of elevated alanine aminotransferase/aspartate aminotransferase caused by such combinations.

with elevated levels of alanine aminotransferase/aspartate aminotransferase was significantly higher in the Atayal (8.57 percent) than in the other three groups (3.02 percent in Ami, 3.02 percent in Bunun, and 2.56 percent in Paiwan). As a whole, 39 (4.3 percent) study participants demonstrated elevated levels of alanine aminotransferase/aspartate aminotransferase, which is higher than the proportion (1.5 percent) found in the general Taiwanese population (18). Individual risk factors correlated with elevated levels of alanine aminotransferase/aspartate aminotransferase (predicting impaired liver function or advanced liver disease) are shown in table 1. Being male and of a younger age were two significant risk factors. As expected, an elevated level of alanine aminotransferase/aspartate aminotransferase was highly correlated with alcoholism, hepatitis C virus infection, and hepatitis B virus carrier status. Notably, among these


TABLE 2. Logistic regression analysis of multiple risk factors of elevated serum alanine aminotransferase/aspartate aminotransferase (ALT/AST) among the 915 members of the aboriginal community in Taiwan, 1990-1992*

Risk factor

Sex (male vs. female) Group (Atayal vs. others):): Current alcoholism§ (yes vs. no) HBVt carrierll (yes vs. no) HCVf infecttonH (yes vs. no)

Muttlvariatead] usted odds ratio

95%Clt

2.46 2.59 2.49 1.73 8.94

1.08-5.64 1.30-5.16 1.13-5.42 0.85-3.56 3.16-25.3

* An elevated ALT/AST level was defined as alanine aminotransferase > 35 lU/lrter and aspartate aminotransferase > 40 Ill/liter. t CI, confidence interval; HBV, hepatitis B virus; HCV, hepatitis C virus. t Others include members of the Ami, Bunun, and Paiwan groups. § Current alcoholics include both current alcohol dependents and alcohol harmful users. II HBV carrier is defined as being hepatitis B surface antigen positive. U HCV infection status is defined by the presence of antihepatitis C virus antibody.

Am J Epidemiol

Vol. 143, No. 9, 1996

vated level of alanine aminotransferase/aspartate aminotransferase (95 percent CI 0.85-3.56). Conversely, being male was found to be associated with an elevated level of alanine aminotransferase/aspartate aminotransferase. Subjects from different groups independently showed different degrees of influence on elevated levels of alanine aminotransferase/aspartate aminotransferase. Compared with subjects of the Ami, Bunun, and Paiwan groups combined, the Atayal subjects exhibited a significantly higher risk of having elevated levels of alanine aminotransferase/aspartate aminotransferase, with a multivariate-adjusted odds ratio of 2.59 (95 percent CI 1.30-5.16). Alcoholism remained significantly associated with elevated levels of alanine aminotransferase/aspartate aminotransferase, and the risk of alcoholics having elevated levels of alanine aminotransferase/aspartate aminotransferase was estimated to be over twice that in lifetime nonalcoholics and ex-alcoholics. Hepatitis C virus infection played by far the most important role in causing elevated levels of alanine aminotransferase/aspartate aminotransferase in our aboriginal subjects, and the multivariate-adjusted odds ratio was as high as 8.94 for those who were positive for anti-hepatitis C virus as compared with those who were negative. We also examined the possible interactive effect of acquisition of hepatitis C virus and alcohol use on the presence of elevated levels of alanine aminotransferase/aspartate aminotransferase. Data in table 3 show that a significantly synergistic effect and combined risk were posed by hepatitis C virus infection combined with alcoholism. A strikingly increased risk of elevated levels of alanine aminotransferase/aspartate aminotransferase resulted when anti-hepatitis C virus was present in alcoholics, roughly more than 20 times higher than in the lowest risk group of current nonalcoholics not infected with hepatitis C virus. The Mantel chi-square test for trend also indicated that the TABLE 3. Combined risk of elevated serum alanine aminotransferase/aspartate aminotransferase (ALT/AST) associated with alcoholism and hepatitis C virus infection In four aboriginal groups in Taiwan, 1990-1992* HCVf Wection/current alcoholism status

Not Infected/nonalcoholic Not infected/alcoholic Infected/nonalcoholic Infected/alcoholic

%of elevated ALT/AST

Odds ratio*

1.8 5.9 10.0 37.5

1.0 2.4 5.9 22.6

95%Clt

1.7-5.5 0.7-54.6 6.5-73.7

* An elevated ALT/AST level was defined as alanine aminotransferase > 35 lU/iiter and aspartate aminotransferase > 40 IU/Dter. t HCV, hepatitis C virus; CI, confidence interval. X The odds ratios have been adjusted for the effects of both group and sex


well-known risk factors, hepatitis C virus infection showed the strongest association with an elevated level of alanine aminotransferase/aspartate aminotransferase in our study subjects, followed by alcoholism and hepatitis B virus carrier status. Of those who had acquired hepatitis C virus infection, more than 25 percent did not have normal levels of alanine aminotransferase/aspartate aminotransferase. This association between anti-hepatitis C virus status and an elevated level of alanine aminotransferase/aspartate aminotransferase was highly significant, with a risk ratio of 7.5 (95 percent confidence interval (CI) 3.615.4). Finally, an important public health implication was found in our ex-alcoholism group, in which the proportion with an elevated level of alanine aminotransferase/aspartate aminotransferase dropped to 2.1 percent, equivalent to that in the lifetime nonalcoholic group (2.2 percent). In contrast, the proportions of subjects having an elevated level of alanine aminotransferase/aspartate aminotransferase in the groups with harmful use of alcohol and alcohol dependence remained as high as 5.5 percent and 9.1 percent, respectively. Results of logistic regression analysis of multiple risk factors with elevated levels of alanine aminotransferase/aspartate aminotransferase are shown in table 2. When the effects of individual risk factors were considered simultaneously, age was no longer a risk factor. In addition, the hepatitis B virus carrier status played only a minor role (odds ratio = 1.73) and was not found to be significantly associated with an ele-

939

940

Shen et al.

relative risk of correlating with elevated levels of alanine aminotransferase/aspartate aminotransferase posed by hepatitis C virus infection combined with alcoholism increased in a dose-response pattern (p < 0.01). DISCUSSION

Am J Epidemiol

Vol. 143, No. 9, 1996


The present study investigates the effects of both alcoholism and viral infection on liver function among representative study subjects from four major Taiwanese aboriginal groups who have high prevalences of these two risk factors. Previous studies involving the effect of alcohol intake on health have mainly relied on a very simple, less reliable check of frequency and/or amount of alcohol consumption. The present study has applied a standardized psychiatric interview to assess drinking behavior and alcohol use disorders in an intensive prospective community survey. Subjects with a clinical diagnosis of alcoholism clearly represent those with a high exposure to alcohol for years. Thus, information bias concerning alcohol intake can largely be reduced. We found that, except for possible genetic differences, hepatitis C virus infection and alcoholism are the two major risk factors that signal a risk of liver damage among these aboriginal groups. In addition, these two factors can act synergistically to correlate with elevated levels of alanine aminotransferase/aspartate aminotransferase, which might represent the presence of impaired liver function or advanced liver disease. The prevalence odds ratios from our cross-sectional study shed some light on suspected risk factors for the presence of elevated levels of alanine aminotransferase/aspartate aminotransferase. However, the estimates are subject to the well-known bias of "selective survival" and may not theoretically serve well as estimates of relative risk for "developing" elevated levels of alanine aminotransferase/aspartate aminotransferase. Such bias indicates that the study subjects we observed were actually more "healthy" than the original cohort members, since those who possessed more risk factors (such as acquisition of hepatitis C virus infection or being an alcoholic) would be more likely to die because of poor liver function or other diseases and would be excluded from this cross-sectional study, if the hypothesized relation is true. This suspected bias may be present in this study, given that >50 percent of the mortality in our cohort during the follow-up period was caused by accidents, suicide, liver cirrhosis, and liver cancer, which are mostly alcohol or virus related. Consequently, such failure to include the dead members in our data analysis might make the odds ratios underestimates, and therefore these odds ratios for

elevated levels of alanine aminotransferase/aspartate aminotransferase would be conservative. The major limitation of the present study is that our definition of elevated levels of alanine aminotransferase/aspartate aminotransferase was based on a single measurement of alanine aminotransferase/aspartate aminotransferase. The alanine aminotransferase/aspartate aminotransferase values may vary and, sometimes, may revert to normal. In addition, there are multiple causes of elevated levels of alanine aminotransferase/aspartate aminotransferase, which were not properly measured in our study. However, for a largescale epidemiologic research study such as ours, it appears to be impractical (or sometimes impossible) to perform repeated measurement for all our cohort members, who resided in different mountain areas in Taiwan. In our ongoing follow-up study, we will determine levels of alanine aminotransferases/aspartate aminotransferases at the second time point, and it may provide further information to verify our current findings. Since the viral genome was first isolated a few years ago (19), accumulating observations have been Unking acquisition of hepatitis C virus infection to the occurrence of various liver diseases including impaired liver function, chronic liver diseases, and hepatocellular carcinoma (20-22). In Taiwan, the relative importance of hepatitis C virus infection in hepatitis B surface antigen-positive patients with those liver diseases has also been defined (3). Our finding indicates that the acquisition of hepatitis C virus infection appears to be the dominant factor predicting liver damage and impaired liver function (manifested by elevated levels of liver enzymes) in the aboriginal community in Taiwan. The mechanism relating abnormal levels of alanine aminotransferase/aspartate aminotransferase to hepatitis C virus infection possibly involves a common viral lytic infection, causing destruction of liver cells and leakage of liver enzymes. The pathogenesis of a viral lytic infection's causing the death (lysis) of liver cells is totally contradictory to the cell immortalization and transformation that occur in liver cancer. However, these chronic impairments and damage to the liver are believed to be highly correlated with the subsequent occurrence of hepatocellular carcinoma. Recent epidemiologic studies have also indicated an association between hepatitis C virus and hepatocellular carcinoma in Taiwan (23). Therefore, a higher infection rate with hepatitis C virus in our aboriginal population (2.8 percent vs. 0.5-2 percent in the general Taiwanese population), especially in the Atayal group whose seroprevalence of anti-hepatitis C virus antibody is 6.8 percent, and its relation to elevated levels of alanine aminotransferase/aspartate amino-


ACKNOWLEDGMENTS

This work was supported by grants from the National Science Council, Taipei, for 6 years (NSC 75, 76-0301H002-28, NSC 78, 79-0301-H002-40, NSC 83-0203B001-102, NSC 84-2331-B001-046) and by a grant from the Department of Health, Taiwan, Republic of China (DOH-80-03; DOH-81-111) for 2 years. The authors acknowledge the excellent performance of our research assistants, J. C. Chang, C. L. Cheng, J. Y. Kao, and J. Y. Sun. Am J Epidemiol

Vol. 143, No. 9, 1996

REFERENCES 1. Saunders JB, Latt N. Epidemiology of alcoholic liver disease. Baillieres Clin Gastroenterol 1993,7:555-79. 2. Rubin E, Farber JL. Pathology. 2nd ed. Philadelphia, PA: J. B. Lippincott Company, 1994. 3. Chen DS, Kuo GC, Sung JL, et al. Hepatitis C virus infection in an area hyperendemic for hepatitis B and chronic liver disease: the Taiwan experience. J Infect Dis 1990;162:817-22. 4. Public health in Taiwan area, Republic of China. Taipei, Taiwan: Department of Health, 1993. 5. Cheng ATA, Chen WJ. Alcoholism among four aboriginal groups in Taiwan: high prevalences and their implications. Alcohol Clin Exp Res 1995;19:1-11. 6. Health statistics: 1. In: General health statistics, 1991. Taipei, Taiwan: Department of Health, 1991. 7. Wu JS, Lu CF, Wu LZ, et al. Changing seroepidemiology of hepatitis A virus infection between two regions in Taiwan differing in socioeconomic status. J Formos Med Assoc 1993; 92:812-15. 8. Wu JS, Lu CF, Chou WH, et al. High prevalence of hepatitis C virus infection in aborigines in Taiwan. Jpn J Med Sci Biol 1992;45:165-74. 9. Lin HH, Wu JS, Lu CF, et al. Hepatitis B virus infection among aboriginal children in eastern Taiwan. J Formos Med Assoc 1992;91:691-3. 10. Cheng TA, Hsu M. A community study of mental disorders among four aboriginal groups in Taiwan. Psychol Med 1992; 22:255-63. 11. Cheng TA, Hsu M. Sex differences in minor psychiatric morbidity among three aboriginal groups in Taiwan: the effects of lineage. Psychol Med 1993;23:949-56. 12. World Health Organization. International statistical classification of diseases and related health problems. Tenth Revision. Geneva: World Health Organization, 1993. 13. American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM UIR. 3rd ed., rev. Washington, DC: American Psychiatric Association, 1987. 14. Wang JT, Wang TH, Sheu JC, et al. Hepatitis C virus infection in volunteer blood donors in Taiwan. Evaluation by hepatitis C antibody assays and the polymerase chain reaction. Arch Pathol Lab Med 1993;117:152-6. 15. Lin HH, Kao JH, Leu JH, et al. Comparison of three different immunoassays and PCR for the detection of hepatitis C virus infection in pregnant women in Taiwan. Vox Sang 1993;65: 117-21. 16. Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic research: principles and quantitative methods. New York: Van Nostrand Reinhold Company, 1982:419-56. 17. Mantel N. Chi-square tests with one degree of freedom: extensions of the Mantel-Haenszel procedure. J Am Stat Assoc 1963;58:690-700. 18. Ho MS, Chao WH, Wu CW, et al. A prospective study of chronic diseases in civil servants in Taiwan: annual report (In Chinese). Taipei, Taiwan: Department of Health, 1993. 19. Choo QL, Kuo G, Weiner AJ, et al. Isolation of a cDNA clone derived from a blood-bome non-A, non-B viral hepatitis genome. Science 1989;244:359-62. 20. Makris M, Preston FE, Triger DR, et al. Hepatitis C antibody and chronic liver diseases in haemophilia. Lancet 1990;335: 1117-19. 21. Kew MC, Houghton M, Choo QL, et al. Hepatitis C virus antibodies in southern African blacks with hepatocellular carcinoma. Lancet 1990;335:873-4. 22. Colombo M, Kuo G, Choo QL, et al. Prevalence of antibodies to hepatitis C virus in Italian patients with hepatocellular carcinoma. Lancet 1989;2:1006-8. 23. Yu MW, You SL, Chang AS, et al. Association between hepatitis C virus antibodies and hepatocellular carcinoma in Taiwan. Cancer Res 1991;51:5621-5. 24. Akerman PA, Cote PM, Yang SQ, et al. Long-term ethanol


transferase deserve immediate public health action to explore the relative importance of different routes of hepatitis C virus transmission. Similar to hepatitis C virus infection, excessive alcohol intake is also known to induce hepatocellular carcinoma, although its direct role at a cellular genetic level is not known. Therefore, it is reasonable to expect a close linkage between alcohol-related impaired liver function as found in this study and the relatively high rates of chronic liver diseases in our aboriginal communities. However, the observed association between alcohol drinking and the risk of elevated alanine aminotransferase/aspartate aminotransferase should also focus more attention on our finding that the ex-alcoholic group had a significantly reduced risk, resembling that of lifetime nonalcoholics. Our data suggest that the presence of hepatitis C virus infection exerts a multiplicative synergistic effect for alcohol-related impaired liver function and that chronic hepatitis C virus hepatitis may increase the liver damage in alcoholic cases with liver disease. One possibility may be that hepatic cells are damaged by early infection with hepatitis viruses, and then an impaired wound healing response to liver regeneration is induced by long-term ethanol consumption (24). Furthermore, it was recently shown that habitual alcoholic drinking was able to increase hepatitis C virus RNA levels in serum (which indicates the proliferation of hepatitis C virus), subsequently resulting in the progression of liver damage (25). The extremely high prevalence of alcoholism and relatively high rate of hepatitis C virus infection in our aboriginal communities make it likely that the contributions of both factors to the risks of elevated levels of alanine aminotransferase/aspartate aminotransferase and other liver diseases are much more important than in other populations. We are conducting prospective studies to understand the subtle interactions between excessive alcohol drinking (alcoholism) and viral hepatitis and the genetic and environmental factors that may contribute to impaired liver function and liver diseases in these aboriginal populations.

941

942

Shen et al. consumption alters the hepatic response to the regenerative effects of tumor necrosis factor-alpha. Hepatology 1993; 17: 1066-73.

25. Oshita M, Hayashi N, Kasahara A, et al. Increased serum hepatitis C virus RNA levels among alcoholic patients with chronic hepatitis C. Hepatology 1994;20:1115-20.


Am J Epidemiol

Vol. 143, No. 9, 1996