Demographic characteristics of the 500 participants tested for HCV are summarized in the Table. In 2000, more than half of the participants were 55–64 years of age (54.2%), 34.4% were 65–74 years, and 11.4% were 75–84 years. This age distribution reflects that participants were aged 40–69 years when enrolled in 1985. Our study included 200 (40.0%) men and 300 (60.0%) women. One hundred seventy (34.0%) participants were from Yaocun, 95 (19.0%) from Rencun, 116 (23.2%) from Donggang, and 119 (23.8%) from Hengshui.

Of the 500 participants studied, 63 (12.6%) were positive for HCV antibody by EIA. When the specimens from these 63 participants were tested by HCV RIBA, 48 were positive, 7 were indeterminate, and 8 were negative. Therefore, the estimated prevalence of HCV infection (defined as positive by both HCV EIA and HCV RIBA) in this population in 2000 was 9.6% (95% CI 7.0%–12.2%). We tested all 23 samples that were positive for HCV RIBA and had sufficient remaining plasma for HCV core antigen. HCV core antigen was found in 16 (69.6%) of 23 specimens.

The prevalence of HCV infection did not differ meaningfully by age or sex (Table), but varied significantly among the four communes. Donggang, the most geographically isolated commune, had a lower prevalence of HCV infection (2.6%) than Yaocun (11.8%, p < 0.01), Rencun (14.7%, p < 0.01), or Hengshui (9.2%, p = 0.03; p = 0.02, overall chi-square test). No other significant differences in HCV prevalence were found between the communes.

Among the 48 participants who were positive by HCV RIBA in 2000, 42 had specimens available from 1985 for HCV antibody testing. Of these 42, 16 (38.1%) were positive by both HCV EIA and HCV RIBA, indicating that most participants who were infected with HCV in 2000 had acquired the virus since 1985.

Antibodies to HBV core antigen (HBc), an indicator of past exposure to HBV, were found in 273 (54.6%) of 500 participants. The prevalence of anti-HBc did not vary significantly by age, sex, or commune (Table). Participants who were infected with HCV had a slightly higher prevalence of anti-HBc than those who were not (64.6% vs. 53.5%, p =0.14). Participants who were positive for both anti-HBc and hepatis B surface antigen (HBsAg), a marker for chronic HBV infection, comprised 6.4% of the population. The prevalence of HBsAg differed relatively little by age, sex, or commune (Table).

A separate group of 200 participants was randomly selected for anonymous testing for antibodies to HIV-1. These participants included 77 men (38.5%) and 123 women (61.5%). Age as of the year 2000 was distributed as follows: <59 years, 78 (39.0%); 60–69 years, 76 (38.0%); >70 years, 46 (23.0%). Seventy-two participants (36.0%) were from Yaocun, 29 (14.5%) were from Rencun, 48 (24.0%) were from Donggang, and 51 (25.5%) were Hengshui. None tested positive for antibodies to HIV-1. The 95% CI for HIV-1 prevalence in this population was 0%–0.03%.

In this population-based study of 55- to 84-year-old persons living in a rural area of Henan Province, HCV prevalence in 2000 was 9.6%. HCV was present in the Linxian population in 1985, but most of the HCV-infected participants in this study likely acquired the virus between 1985 and 2000. In contrast to the high rate of HCV infection, HIV-1, which also is transmitted through infected blood, was not found in this population. Evidence of resolved or chronic HBV infection, which is endemic in China, was found in about 55% of the participants.

The observed prevalence of HCV in Linxian is higher than in most previous population-based studies from China. In the nationwide cross-sectional study conducted in 1992, HCV prevalence was 3.10% in residents of rural China and 3.96% in the group aged 50 to 59 years, the eldest in that study (4). However, because HCV prevalence in the nationwide study was determined by HCV EIA alone, estimates from that study are higher than would have been obtained with confirmation by HCV RIBA. The RIBA-confirmed HCV prevalence in Linxian was, therefore, considerably higher than the national rate.

Around the world, the prevalence of HCV infection appears to be low in most populations (13–15), but areas of high prevalence have been found. Perhaps the most notable example is Egypt, where >15% of the population may be infected with HCV (16,17). Transmission of HCV occurred in Egypt from the 1960s through the 1980s when a campaign against schistosomiasis involved mass parenteral injections, and unsterilized needles were used (17). Evidence also exists that HCV infection may have been transmitted in Egypt through other types of medical care (18). Iatrogenic transmission may have contributed as well to high rates of HCV infection that have been reported from Taiwan (19), Japan (20), and Italy (21).

We could not determine how HCV spread in Linxian because the nutritional trial did not collect information on potential exposure to contaminated blood. Widespread HCV infection in a population generally results from iatrogenic transmission or sharing of recreational drug injection equipment. HCV spreads rapidly among injection drug users, but, to our knowledge, reports of injection drug use in China are limited to younger age groups in border provinces and large cities (22,23), not older residents of inland provinces. It seems unlikely, therefore, that our participants acquired HCV infection through injection drug use. Transmission of HCV has been linked to paid blood and plasma donations in Central Chinese provinces, including Henan, during the 1980s and 1990s (24,25). Reuse of needles and equipment without proper sterilization and reinfusion of pooled red blood cells from multiple donors reportedly led to outbreaks of HCV (24,25). These reports, along with evidence from our study that many HCV-infected participants acquired the virus between 1985 and 2000, suggest that HCV may have been transmitted among the Linxian population through blood or plasma collection activities. Our finding of the lowest HCV prevalence (2.6%) in the most isolated commune seems consistent with this explanation, since geographic isolation may have limited the opportunity to contract HCV infection through these means.

Outbreaks of HIV-1 have been reported among paid plasma donors in central Chinese provinces, including Henan (8–10), but we found no evidence of HIV-1 infection among 200 randomly selected participants. Our analysis yielded an upper 95% CI of 0.03% for the prevalence of HIV-1 in this population, which indicates that HIV-1 infection is, at worst, extremely rare among older residents of Linxian. The absence of HIV-1 in the presence of a relatively high prevalence of HCV is not surprising because the entry of either virus into a community may depend to some degree on chance.

HBV infection is endemic in China (26), and most transmission occurs during the perinatal period when the risk of chronic infection is much higher than in adulthood. Among our participants, 54.6% had HBc, and 6.4% had both anti-HBc and HBsAg (which indicates chronic infection). Most or all of the participants who were chronically infected with HBV likely became infected early in life. More recent bloodborne transmission may have contributed some participants with resolved HBV infection, as suggested by the trend toward higher anti-HBc prevalence among participants who were also infected with HCV.

The accuracy with which our estimates reflect the prevalence of viral infections among older Linxian residents in the year 2000 depends on 2 factors: the test characteristics of the assays we used and how well our study population represents the target population. The third generation HCV EIA we used has high sensitivity and specificity. With confirmation of HCV EIA-positive samples by a highly specific RIBA, we are unlikely to have overestimated the prevalence of HCV antibodies among our participants. We have similar confidence in the assays that we used for HIV-1 and HBV testing.

For practical reasons, we used specimens that had been collected as part of a population-based research effort that began in 1985. The limitations of our approach for determining the prevalence of viral infections among Linxian residents in 2000 should be considered. First, the age criterion of the original study prevented us from ascertaining infection prevalence among persons <55 years of age. HCV prevalence may be higher or lower among younger residents of Linxian. Second, the participants may not represent older adults residing in Linxian in 2000 with regard to viral prevalence. About 60% of eligible residents enrolled in the trial in 1985 and about 70% of surviving enrollees participated in the 2000 follow-up. If surviving enrollees who were infected with one of the viruses that we studied were less likely to participate in the 2000 survey due to poor health, we may have underestimated the true prevalence of infection. However, usually a long period elapses from infection with these viruses to disease, and our results likely provide a reasonable estimate of viral prevalence among older residents of Linxian.

In summary, HCV is now common in this rural Chinese community, at least among its older residents. In contrast, we found no evidence of HIV-1 infection in the population. The public health impact of the high prevalence of HCV infection in Linxian may be substantial. HCV is an important cause of end-stage liver disease and hepatocellular carcinoma, and it can act synergistically with HBV infection (27), which is endemic in China. Future studies should examine the prevalence of bloodborne viruses in other parts of China, how these viruses are transmitted, and the resulting health effects. Efforts to halt the transmission of HCV and other bloodborne viruses in rural China should be a top public health priority.