comparison of hepatitis b virus and hepatitis c virus ... - CiteSeerX

Am. J. Trop. Med. Hyg., 66(4), 2002, pp. 389–393 Copyright © 2002 by The American Society of Tropical Medicine and Hygiene

COMPARISON OF HEPATITIS B VIRUS AND HEPATITIS C VIRUS PREVALENCE AND RISK FACTORS IN A COMMUNITY-BASED STUDY CHONG-SHAN WANG, TING-TSUNG CHANG, WEI-JEN YAO, AND PESUS CHOU A-Lein Community Health Center, Kaohsiung County, Taiwan; Institute of Public Health, National Cheng Kung University, Tainan, Taiwan; Department of Internal Medicine, Division of Gastroenterology, National Cheng Kung University, Tainan, Taiwan; Department of Nuclear Medicine, National Cheng Kung University, Tainan, Taiwan; Community Medicine Research Center and Institute of Public Health, National Yang-Ming University, Taipei, Taiwan

Abstract. We performed a community-based study of 12 villages of southern Taiwan’s A-Lein Township to investigate the epidemiology of hepatitis B and hepatitis C virus (HCV) infections. Of 6,095 patients, 13.8% were positive for hepatitis B surface antigen positive (HBsAg(+)) and 17.0% were positive for anti-HCV (anti-HCV(+)). Infection was found to be inversely related to educational level and to be directly related to the frequency of the receipt of parenteral injection for medical purposes. Risk factors for HBsAg positivity were male sex, age ⱕ 50 years, and a family history of hepatocellular carcinoma. Risk factors for HCV seropositivity were lower education level, frequent parenteral injections, blood transfusion, menial occupations, smoking, and age > 50 years. Therefore, risk factors for HBsAg(+) and antiHCV(+) were different in these Taiwanese communities. Safe medical injections and improved health education for high-risk groups are imperative for preventing HCV transmission. between patients, was ubiquitous because of poverty and the scarcity of medical resources in Taiwan, especially in rural communities, until about a decade ago. From long-term observations, most injections are unnecessary because alternatives exist that can be delivered orally, and the indications for injections include several nonspecific symptoms, such as colds, fatigue, dizziness, myofacial pain, diarrhea, abdominal pain, and fever. In this study, we also analyzed the risk factors for frequent medical injections among those with HBV and those with HCV infections. We hoped that our results would help prevent the spread of HCV and other diseases transmitted by frequent, unsafe, and unnecessary medical injections. Many studies have shown that socioeconomic status (in the present study, this was assessed as a combination of educational level and occupational status) affects health status, morbidity, and mortality, especially for cardiovascular diseases.12,13 Few studies have reported on the relationship between socioeconomic status and the HBV or HCV carrier state.14,15 In addition, most previous epidemiological studies of hepatitis have focused exclusively on one virus, HBV or HCV, but have not considered the relationship between them. Accordingly, the present study population provided a unique opportunity to simultaneously observe and compare the distribution and risk factors of the HBsAg carrier state and HCV infection. These results may provide insight into strategies to prevent hepatitis virus infection.

INTRODUCTION Hepatitis B virus (HBV) and hepatitis C virus (HCV) are major causes of chronic liver diseases worldwide, particularly cirrhosis and hepatocellular carcinoma. Although the incidence and HBV infection has been markedly reduced after mass vaccination programs,1 HCV infection remains a worldwide public health concern.2 Taiwan, still a developing country, is located in subtropical east Asia and has a high mortality rate due to hepatocellular carcinoma.3 Southern Taiwan’s ALein Township has a hepatocellular carcinoma mortality rate more than double that of Taiwan as a whole. A previous study in A-Lein showed that the prevalences of HCV antibodies (anti-HCV(+)) and HBV surface antigen (HBsAg(+)) among residents aged ⱖ 65 years were 30.5 and 11.4%, respectively.4 The national prevalence rates among adults in Taiwan for anti-HCV(+) is 1–3%5 and for HBsAg(+) 15–20%.6 Therefore, we suspected that HCV was hyperendemic in this community. The main purpose of this community-wide population survey among residents aged ⱖ 35 years was to investigate the prevalence of hepatitis and its risk factors for infection. Both HBV and HCV are blood-borne hepatotropic viruses but have distinct routes of transmission. Most commonly, HBV is acquired by vertical transmission from an HBsAg(+) mother or via horizontal transmission in childhood.6–8 However, HCV is primarily transmitted parenterally in adulthood by intravenous drug use, blood transfusion, or medically related parenteral exposures, but rarely through the placenta, breast-feeding, or sexual contact.4,9–11 Previous studies have showed that frequent medical parenteral injections that used nondisposable needles were highly associated with HCV seropositivity and that most persons with anti-HCV(+) did not have a history of blood transfusion or intravenous drug use,4,9 the most commonly documented risk factors for HCV infection.10 In this community, patients and health care providers, including unlicensed practitioners, pharmacists, and even some doctors, popularly considered medical injections a routine service provided during health care visits; consequently, medical injections were overused.4,9 Unsafe injections, including the reuse of syringes or needles without sterilization

PATIENTS AND METHODS A-Lein Township has a population of ∼ 30,000 residents. Of these, ∼ 8,800 are aged ⱖ 35 years. The age distribution in this study was 42.9% for those 35–50 years old, 33.4% for those 50–65 years old, and 23.7% for those ⱖ 65 years old (mean age ⳱ 54.0 years; standard deviation, ± 13.0). Although there are only 12 villages in A-Lein Township, for this study, we created a 13th comprising the residents of temples and the township’s single nursing home because their daily activities are different from those of the general population (Figure 1). Our study lasted from January 1996 to January 1998. Intensive health promotion activities for hepatitis and hepatocellular carcinoma screening were held before blood samples and questionnaires were collected. Villagers came of their

389

390

WANG AND OTHERS

TABLE 1 Hepatitis prevalence for those ⱖ 35 years old in each village of ALein, Taiwan, 1996–1997* Village

Total no.

1 1,024 2 625 3 492 4 761 5 372 6 370 7 442 8 521 9 199 10 296 11 379 12 249 13† 365 Total 6,095

HBV+/HCV− Prevalence (%)

HBV−/HCV+ Prevalence (%)

HBV+/HCV+ Prevalence (%)

HBV−/HCV− Prevalence (%)

13.2 10.2 9.3 10.2 10.9 9.3 13.0 8.7 8.8 6.8 9.4 13.8 17.3 11.8

8.4 11.2 11.0 7.2 35.7 37.6 14.5 12.6 12.6 9.4 20.2 9.5 9.3 15.0

1.2 1.5 1.2 0.7 6.2 4.0 3.2 2.1 1.4 1.0 1.0 0.4 1.6 2.0

68.3 68.1 73.7 76.7 39.5 42.4 62.2 69.0 69.7 78.6 63.0 66.9 71.8 71.2

* + ⳱ positive; − ⳱ negative; anti-HCV ⳱ hepatitis C virus antibodies; HBC, hepatitic C virus; HBsAG ⳱ hepatitis B surface antigen; HBV, hepatitis B virus; HBV+/HCV−: HBsAG(+) and anti-HCV(−); HBV−/HCV+: HBsAG(−) and anti-HCV(+); HBV+/HCV+: HBsAG(+) and anti-HCV(+); HBV−/HCV− ⳱ HBsAG(−) and anti-HCV(−). † Although there are only 12 villages in A-Lein Township, we created a 13th comprising the residents of temples and the single nursing home in A-Lein.

FIGURE 1. Map of A-Lein Township showing the location of the 12 study villages.

own volition and provided informed consent. A total of 6,095 residents attended this hepatitis screening program, a response rate of 69.3% (6,095 of 8,800). We recruited only those aged ⱖ 35 years for screening for financial and feasibility reasons. Specifically, the second portion of this program was to screen for hepatocellular carcinoma in those patients found to be HBsAg(+) and anti-HCV(+), or patients with elevated serum alanine aminotransferase. This study was approved by the A-Lein Health Promotion Committee. The questionnaire consisted of age, education levels, occupation, family history of hepatocellular carcinoma, daily consumption of cigarettes and alcohol, and medical history with special emphasis on blood transfusion and parenteral therapy. Alcohol drinking was subclassified as frequent, occasional, or rare. Smoking was subclassified as ⱖ 1 pack per day, < 1 pack per day, abstained (i.e., quit smoking at least 6 months before this study), or never smoked. A “frequent injection” recipient was someone who had had > 6 injections per year before the introduction of disposable needles and syringes ∼ 10 years ago. The township had no reported cases of human immunodeficiency virus and only a negligible number of reported and suspected intravenous drug users among those ⱖ 35 years old. Serum samples were sent to the Tainan Blood Center of the Chinese Blood Service Foundation to test for HBsAg and anti-HCV markers. The Murex (London, UK) HBsAg (version 1) enzyme immunoassay method was used to test for HBsAg. Anti-HCV was tested by the third-generation Murex anti-HCV enzyme immunoassay, which contains the antigen from the HCV core, nonstructural 3 (NS3), NS4, and NS5 regions. A chi-square test was used to determine the difference of

risk factors in seropositive hepatitis status with seronegative persons. Odds ratios (OR) and 95% confidence intervals (CI), which used seronegative persons as the reference group, were applied in univariate analysis. Multiple logistic regression was used for multivariate analysis to determine the independent risk factors for exclusively HBsAg(+), exclusively anti-HCV(+), and frequent medical injections. RESULTS Of the 6,095 blood samples collected, 71.2% were negative, 13.8% HBsAg(+), 17.0% anti-HCV(+), and 2.0% were both HBsAg(+) and anti-HCV(+). The prevalence of hepatitis virus infection varied from village to village. The seronegative rates ranged 39.5–78.6%, the HBsAg(+) 6.8–17.3%, and the anti-HCV(+) 7.2–37.6% (a 5.2-fold difference) (Table 1). By use of seronegative people as the reference group in univariate analysis, we found those who tested HBsAg(+) were more likely to be men, to be aged < 50 years, to have higher educational levels, to have a family history of hepatocellular carcinoma, and to habitually drink alcohol and smoke cigarettes (P < 0.05). Those who tested anti-HCV(+) were more likely to be ⱖ 50 years old, to have lower educational levels, to have menial employment, to be smokers, to have had a blood transfusion, and to have had frequent medical injections with nondisposable needles and syringes. The prevalence of anti-HCV(+) was inversely related to educational levels and positively related to the number of medical injections. There was no significant difference between the prevalence of anti-HCV(+) in men and women (Table 2). By use of seronegative people as the reference group in multiple logistic regression, after controlling covariates, the independent risk factors for HBsAg(+) carriers were male sex, being aged < 50 years, and having a family history of hepatocellular carcinoma. Among these factors, a family history of hepatocellular carcinoma (OR, 2.38; 95% CI, 1.70– 3.35) was the most significant. In the multivariate logistic analysis, the independent risk factors for anti-HCV(+) were exactly the same as in the univariate analysis. Among these factors, lower educational level (OR, 1.95; 95% CI, 1.46–2.61)

391

HBV AND HCV EPIDEMIOLOGY

TABLE 2 Univariate analysis of risk factors for HBV and HCV in A-Lein, Taiwan, 1996–1997* HBV+/HCV− versus HBV−/HCV−

HBV−/HCV+ versus HBV−/HCV−

OR

95% CI

OR

95% CI

Demographic data Age (ⱖ 50 versus < 50 years) Sex (male versus female)

0.61 1.41

0.52–0.71 1.20–1.65

2.41 1.05

2.05–2.83 0.91–1.21

Occupation (menial versus not menial)

1.11

0.86–1.42

0.41

0.30–0.57

Education Primary school versus illiterate Middle school versus illiterate High school versus illiterate College versus illiterate

1.41 1.68 1.70 1.36

1.16–1.71 1.32–2.15 1.32–2.19 0.95–1.95

0.61 0.35 0.29 0.17

0.51–0.72 0.26–0.47 0.21–0.41 0.10–0.31

Family history of hepatocellular carcinoma (yes versus no)

2.56

1.95–3.36

1.05

0.76–1.45

Lifestyle Drinking (yes versus no) Smoking (yes versus no)

1.36 1.35

1.14–1.63 1.10–1.56

0.90 1.33

0.76–1.08 1.13–1.55

Medical Experience Blood transfusion (yes versus no)

0.93

0.71–1.21

1.67

1.37–2.04

Medical injections Frequently versus rarely Occasionally versus rarely

0.86 0.91

0.64–1.14 0.75–1.10

2.93 1.37

2.35–3.65 1.14–1.65

Risk factor

* + ⳱ positive; − ⳱ negative; CI, confidence interval; HBV ⳱ hepatitis B virus; HCV ⳱ hepatitis C virus; HBsAG ⳱ hepatitis B surface antigen; HBV+/HCV− ⳱ HBsAG(+) and anti-HCV(−); HBV−/HCV+ ⳱ HBsAG(−) and anti-HCV(+); HBV−/HCV− ⳱ HBsAG(−) and anti-HCV(−); OR, odds ratio.

and frequent medical injections (OR, 1.93; 95% CI, 1.54–2.40) were highly associated (Table 3). In multivariate analysis of risk factors for frequent medical injections among HBsAg(+) and anti-HCV(+) by multivariate logistic models, we found that lower educational levels and blood transfusion history were independent factors for both strains of hepatitis and that smoking was highly associated with anti-HCV(+) (Table 4). DISCUSSION This community-wide population study produced several important findings. First, the prevalence of anti-HCV(+) in each village (range, 7.9–41.9%) was more variable than that of HBsAg(+) (range, 7.8–17.1%). Second, although HBV and HCV transmission is blood-borne, the risk factors for HBsAg(+) subjects and anti-HCV(+) subjects are different. The HBsAg(+) subjects were more likely to be young (aged < 50 years) men and to have a family history of hepatocellular carcinoma, but anti-HCV(+) subjects were more likely to be

older (aged ⱖ 50 years), to have had a blood transfusion, to have had frequent medical injections, to have a lower level of education and a menial job, and to habitually smoke cigarettes. This shows that HBV infection is more associated with familial transmission and occurs earlier than HCV infection. Most HCV infections were acquired later in adulthood, and transmission was associated with lower socioeconomic status and iatrogenic routes. There is therefore no significant association between anti-HCV and a family history of hepatocellular carcinoma in this study. Third, the major risk factors of HCV infection, except blood transfusion, are different from those found in developed countries. This population survey demonstrates that the main source of hyperendemic HCV infections in Taiwan—hitherto a mystery—is primarily unsafe and unnecessary medical injections associated with contaminated needle use. This rarely occurs in developed countries, except among users of illicit parenteral drugs10,16,17 To prevent further transmission of HCV infection in the world, this iatrogenic route should be regulated and international travelers should be warned to avoid

TABLE 3 Multiple logistic regression analysis of risk factors for HBV and HCV in A-Lein, Taiwan, 1996–1997* HBV+/HCV− versus HBV−/HCV−

HBV−/HCV+ versus HBV−/HCV−

Risk factor

OR

95% CI

OR

95% CI

Sex (male versus female) Age (ⱖ 50 versus < 50 years) Education (ⱕ primary school versus ⱖ junior high) Occupation (menial versus not menial) Drinking (yes versus no) Smoking (yes versus no) Family history of hepatocellular carcinoma (yes versus no) Blood transfusion (yes versus no) Frequent medical injections (yes versus no)

1.43 0.65 0.97 1.23 0.70 1.14 2.38 0.89 0.97

1.11–1.84 0.52–0.81 0.76–1.25 0.88–1.71 0.43–1.11 0.86–1.51 1.70–3.35 0.65–1.24 0.71–1.33

1.12 1.75 1.95 1.60 1.15 1.34 1.08 1.62 1.93

0.88–1.42 1.41–2.17 1.46–2.61 1.05–2.45 0.77–1.72 1.03–1.74 0.71–1.65 1.26–2.09 1.54–2.40

* + ⳱ positive; − ⳱ negative; CI, confidence interval; HBV ⳱ hepatitis B virus; HCV ⳱ hepatitis C virus; HBsAG ⳱ hepatitis B surface antigen; HBV+/HCV− ⳱ HBsAG(+) and anti-HCV(−); HBV−/HCV+ ⳱ HBsAG(−) and anti-HCV(+); HBV−/HCV− ⳱ HBsAG(−) and anti-HCV(−); OR, odds ratio.

392

WANG AND OTHERS

TABLE 4 Multiple logistic regression analysis of risk factors for frequent medical injections among HBsAG(+) and anti-HCV(+) subjects in A-Lein, Taiwan* HBsAG(+)

Anti-HCV(+)

Medical injections (frequent versus infrequent)

OR

95% CI

OR

95% CI

Sex (male versus female) Education (ⱕ primary school versus ⱖ junior high) Occupation (menial versus not menial) Drinking (yes versus no) Smoking (yes versus no) Blood transfusion (yes versus no)

0.8 5.3 1.4 1.1 1.5 2.3

0.4–1.5 2.6–10.9 0.5–4.3 0.5–2.1 0.8–2.8 1.3–4.1

0.8 1.8 0.9 1.5 1.9 2.2

0.5–1.2 1.1–3.2 0.4–1.8 0.6–3.9 1.2–3.1 1.5–3.2

* + ⳱ positive; − ⳱ negative; CI ⳱ confidence interval; HBsAG ⳱ hepatitis B surface antigen; HCV ⳱ hepatitis C virus; HBsAG(+) ⳱ HBsAG(+) and anti-HCV(−); anti-HCV(+) ⳱ HBsAG(−) and anti-HCV(+); OR ⳱ odds ratio.

such injections if possible. Overuse of medical injections has been another cause of HCV transmission that occurs in a dose-response relationship with the number of unsafe injections.18 Although unsafe and unnecessary injections have decreased substantially over the past few decades in Taiwan, they are still routinely used in outpatient visits in many developing countries.2 They spread not only HBV and HCV, but also human immunodeficiency virus, which will result in more morbidity and mortality and generate an increase in economic burdens from these diseases several years later.2 Although the risk for HCV transmission by blood transfusion has been greatly decreased since an anti-HCV test was introduced in the early 1990s,19 an increase in the prevalence of hepatitis C virus infection in the world might occur through cross transmission from the unsafe and unnecessary injections that are routine in outpatient visits and ubiquitous for international travelers in developing countries.2 In this study, those subjects who habitually smoked had a higher prevalence of HCV infection and were at significant risk for frequent medical injections. This might be so because smoking is highly associated with an increased use of outpatient physician and hospital services.18 In past decades, unfortunately, health care services in Taiwan usually administered unnecessary injections with reusable and inadequately sterilized needles and syringes. The mechanism of transmitting HCV through repeated use of the same needles and other medical instruments is similar to the sharing of needles by intravenous drug users, where the probability of HIV infection has a clear correlation with the period and frequency of injection.20 Socioeconomic status also plays an important, although indirect, role in the transmission of HCV infection. In this study, educational levels were inversely related to the prevalence of anti-HCV(+) and the frequency of medical injections independent of HBsAg(+) or anti-HCV(+) (Tables 2 and 4). This might be because people of lower socioeconomic status are more likely to have an unhealthy lifestyle and lack access to health care, and be less well informed about the prevention of diseases.21 Consequently, their rates of morbidity for diseases and medical interventions were higher than average, with each incidence exposing them to infection via reused and virus-contaminated instruments.10,22 Two villages had an antiHCV(+) prevalence nearing 42%, 3 times higher than the average of the other villages in this community and 20 times higher than the national average. These higher levels are related to the existence of unlicensed practitioners who, for several decades, had taken advantage of a lack of consumer information in the medical services market and urged patients

to take unnecessary injections. This is truly iatrogenic transmission. This type of transmission likely exists elsewhere, especially in developing countries. These findings present a call for cooperation by the community, health care systems, government, and world health organizations. They must agree on and emphasize the use of disposable medical supplies, the sterilization of reusable supplies and equipment, and the strict adherence to and enforcement of safe and standard medical practices to prevent further transmission of HBV and HCV. Acknowledgments: We thank the staff and volunteers of the A-Lein Community Health Center for their assistance in data collection and the Tainan Blood Center of the Chinese Blood Service Foundation for assistance with laboratory analyses. We thank Bill Franke for revising the English in this article. Financial support: This study was supported by the A-Lein Community Health Promotion Committee, Kaohsiung County, Taiwan, and the C. T. Hsu Cancer Research Foundation, Taipei City, Taiwan. Authors’ addresses: Chong-Shan Wang, Tainan, Taiwan 704. TingTsung Chang, Tainan, Taiwan 704. Wei-Jen Yao, Tainan, Taiwan 704. Pesus Chou, National Yang Ming University, Institute of Public Health, Shih-Pai, Taipei, 112, Taiwan. Reprint requests: Pesus Chou, National Yang Ming University, Institute of Public Health, Shih-Pai, Taipei, 112, Taiwan, Telephone: 8862-2822-9695, Fax: 886-2-2820-1461, E-mail: [email protected].

REFERENCES 1. Chang MH, Chen CJ, Lai MS, Hsu HM, Wu TC, Kong MS, Liang DC, Shau WY, Chen DS, 1997. Universal hepatitis B vaccination in Taiwan and the incidence of hepatocellular carcinoma in children. N Engl J Med 336: 1855–1859. 2. Simonsen L, Kane A, Lloyd J, Zaffran M, Kane M, 1999. Unsafe injections in the developing world and transmission of bloodborne pathogens: a review. Bull World Health Organ 77: 789– 800. 3. Wang CS, Chou P, 1997. An analysis of the number of deaths and the years of potential life lost in A-Lein, Taiwan, in 1990–1994 (in Chinese). Public Health (Taiwan) 24: 51–60. 4. Wang CS, Chang TT, Chou P, 1998. Differing characteristics of hepatitis B and C risk factors among elders in a rural area in Taiwan. J Gerontol A Biol Sci Med Sci 53A: M107–M111. 5. Wang JT, Wang TH, Sheu JC, Tsai SJ, Hsieh YS, Lin DT, Wang CY, Chen DS, 1993. 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 117: 152–156. 6. Sung JL, 1997. Prevention of hepatitis B and C virus infection for prevention of cirrhosis and hepatocellular carcinoma. J Gastroenterol Hepatol 12: S370–S376. 7. Stevens CE, Beasley RP, Tsui J, Lee WC, 1975. Vertical trans-

HBV AND HCV EPIDEMIOLOGY

8. 9.

10. 11. 12.

13. 14. 15.

mission of hepatitis B antigen in Taiwan. N Engl J Med 292: 771–774. Beasley RP, Hwang LY, Lin CC, Leu ML, Stevens CE, Szmuness W, Chen KP, 1982. Incidence of hepatitis B virus infections in preschool children in Taiwan. J Infect Dis 146: 198–204. Wang CS, Chang TT, Chou P, 1998. Differences in risk factors for being either a hepatitis B carrier or anti-hepatitis C+ in a hepatocellular carcinoma-hyperendemic area in rural Taiwan. J Clin Epidemiol 51: 733–738. Di Bisceglie AM, 1998. Hepatitis C. Lancet 351: 351–355. Dienstag JL, 1997. Sexual and perinatal transmission of hepatitis C. Hepatology 26(Suppl 1): 66S–70S. Lantz PM, House JS, Lepkowski JM, Williams DR, Mero RP, Chen J, 1998. Socioeconomic factors, health behaviors, and mortality: results from a nationally representative prospective study of US adults. JAMA 279: 1703–1708. Kind P, Dolan P, Gudex C, Williams A, 1998. Variations in population health status: results from a United Kingdom national questionnaire survey. BMJ 316: 736–741. Stuver SO, Boschi-Pinto C, Trichopoulos D, 1997. Infection with hepatitis B and C viruses, social class and cancer. IARC Sci Publ 138: 319–324. Murphy EL, Bryzman S, Williams AE, Co-Chien H, Schreiber

16. 17. 18. 19. 20. 21.

22.

393

GB, Ownby HE, Gilcher RO, Kleinman SH, Matijas L, Thomson RA, Nemo GJ, 1996. Demographic determinants of hepatitis C virus seroprevalence among blood donors. JAMA 275: 995–1000. Trepo C, Pradat P, 1999. Hepatitis C virus infection in Western Europe. J Hepatol 31(Suppl 1): 80–83. Alter MJ1999. Hepatitis C virus infection in the United States. J Hepatol 31(Suppl 1): 88–91. Nina HN, Seppo M, Ilkka V, Matti P, Pekka O, 1999. The impact of smoking, alcohol consumption, and physical activity on use of hospital services. Am J Public Health 89: 691–698. Van der Poel CL, 1999. Hepatitis C virus and blood transfusion: past and present risks. Hepatology 31(Suppl 1): 101–106. Thomas DL, Vlahov D, Solomon L, 1995. Correlates of hepatitis C virus infections among injection drug users. Medicine (Baltimore) 74: 212–220. Kawachi I, 1998. Social inequalities in health: new frontiers of research. The 1998 Conference on Health and Social Service Policy: Health for the 21st Century. Kaohsiung: Kaohsiung Medical College, Taiwan, 39–90. Buckingham K, Freeman PR, 1997. Sociodemographic and morbidity indicators of need in relation to the use of community health services: observational study. BMJ 315: 994–996.