Risk Factors for Hepatitis C Transmission in the Victorian Population: A Telephone Survey

A study was conducted to measure knowledge and prevalence of risk factors for hepatitis C infection in the Victorian community. A telephone survey of 757 Victorian householders aged 15+ years was carried out in March 1996. The study found a very small percentage of Victorians are injecting drug users and therefore at highest risk of hepatitis C infection; thus transmission control programs can be efficiently focused on this group. The Victorian community needs to be better educated about risk factors for hepatitis C, in particular that transfusions and blood donations are safe.

Objective: To measure knowledge and prevalences of risk factors for hepatitis C infection in the Victorian community.

Method: Telephone survey of 757 Victorian householders aged 15+ years, March 1996.

Results: An estimated 2.2% of Victorians have injected illicit drugs; 4.8% have tattoos and 49.6% have skin piercings; 11.4% received a blood transfusion before screening for HCV was introduced; 10.9% have had a test for HCV antibody. A majority of respondents perceived sexual contact and receipt of a transfusion to be risks for HCV transmission.

Conclusions: Risk factors associated with the spread of HCV are widespread in the Victorian community; however, the most important risk factor - injecting drug use - has low prevalence. Considerable uncertainty exists about risk factors for hepatitis C.

Implications: A very small percentage of Victorians are injecting drug users and therefore at highest risk of hepatitis C infection, therefore transmission control programs can be efficiently focused on this group. The Victorian community needs to be better educated about risk factors for hepatitis C, in particular that transfusions and blood donations are safe. (Aust NZJ Public Health 1999; 23:622-6)

Introduction

Hepatitis C virus (HCV) spreads between humans through the transfer of infected blood. The most important risk factor is injecting drug use (IDU) (1,2,3) although infection via blood transfusion affected a significant number prior to the introduction of blood screening in 1990. (1) Other reported routes of transmission of hepatitis include tattooing, (4,5) skin piercing and scarification (6,7) and occupational exposure. (8,9) Vertical transmission has been documented but is a relatively low-risk route, (10,11) and household and nosocomial transmission have also been reported; (12,13) as with sexual transmission, they are uncommon routes. (14,15) A low but significant number of HCV cases have no definite or identifiable risk factor. (2,3) In 1996, 4,544 cases of hepatitis C were newly diagnosed in Victoria; (16) however, the asymptomatic nature of many hepatitis C infections means that an unknown number of cases are not recorded, and the overall prevalence of HCV in the Victorian population remains unknown. Moreover, the prevalences of risk factors for HCV infection are not known with any precision, information which is valuable for planning and targeting both screening and prevention campaigns.

To make estimates of the prevalences of risk factors for HCV infection in Victorians, we interviewed a sample of adults (15 years and over) about their participation in intravenous drug use, tattooing, skin piercing and other behaviours. The respondent's knowledge of the risk of HCV associated with these and other potential sources of exposure was evaluated, and we have estimated the proportion of the Victorian population that has been tested for HCV.

Methods

A cross-sectional sample of Victorians aged 15 years and over was interviewed using a telephone survey in early 1996. The selection process was designed to achieve a balance between urban and rural localities. A stratified sampling frame of Victorian telephone numbers was created from a commercially available CD-Rom telephone directory (which excluded mobile telephones, business and unlisted numbers). (17) Australian Bureau of Statistics labour force regions (18) were used to stratify the population into 14 geographical subdivisions and a random selection of telephone numbers was made from each stratum (choosing a randomly selected number between the first and 400th, and every 400th number thereafter). Within each household, the individual with the most recent birthday was identified and invited to participate. (19) If the selected household member was temporarily unavailable, or if the call was not answered or if an answering machine was reached, contact was re-attempted at a later date.

Interviews were performed by trained personnel over a three-week period, with calls made Monday to Thursday between 4pm and 8pm, and call-backs outside these times. The questionnaire covered demographic details, perceived risk of HCV infection and seven potential risk factors for infection. After completion of the interview, respondents were provided with the facts about hepatitis C risk if their knowledge was judged to be incomplete, and were offered information brochures and the HCV Help-line telephone numbers.

Data were analysed using EpiInfo and SPSS computer packages. Results were corrected to account for bias in the sample's demographic characteristics by reference to population projections for Victoria, 1995. (20) To identify predictors of hepatitis C testing, odds ratios were calculated for unequivocal responses using W contingency tables and variables subsequently found to be significant or considered potentially confounding were used in a logistic regression to obtain adjusted odds ratios. Conference intervals for proportions were calculated using a standard method. (21)

Results

From a total of 2,808 calls made to 1,744 Victorian telephone numbers, 1,391 eligible householders were contacted and asked to participate. Seven hundred and fifty-seven complete interviews were obtained, a positive response rate of 54.4% (757/1391); 39.0% (543/1391) declined to participate, 6.1% (85/1391) could not take part because of language difficulties and 0.4% (6/1391) of contacted households gave incomplete data. Of the 353 telephone numbers that did not yield contact with an eligible household, 102 were disconnected numbers, 32 non-residential numbers, 13 facsimile numbers and 206 numbers remained unanswered after five calls or at the end of the survey period.

Four hundred and forty-nine (59.3%) of the interviewees were resident in the Melbourne metropolitan area. Most respondents were born in Australia (77.4%) and spoke English at home (93.5%); 452 (59.7%) were women.

Perceptions of risk

Respondents were asked if they believed there was a risk of HCV infection from a range of behaviours, environments and situations (Table 1). Most knew that injecting drug use posed a risk for HCV infection, and three-quarters identified tattooing as carrying a risk, but a majority thought that receipt of a blood transfusion and sexual contact with an HCV-infected person posed a meaningful risk for HCV transmission. There was considerable uncertainty, with high proportions answering 'don't know' for each risk factor. For 'contaminated food or water' and 'living in the same household as an HCV-infected person' responses were almost evenly distributed between 'yes', 'no' and 'don't know'.

Intravenous drug use

Fifteen of the respondents (2.0%, 95% CI 1.0-3.0%) reported having ever injected drugs. Of these, three stated they had used in 1996 (meaning within the previous two months, as the survey was conducted in February 1996), one in 1995, and six prior to 1995; five gave no answer. Five ever-injectors had been tattooed and 10 had one or more skin piercing (Table 2).

Ten (66.7%) of the injectors had been tested for HCV; only one did not know there was a risk of HCV infection from intravenous drug use. Nine identified tattooing as an infection risk.

Tattooing

Of the total sample of 757 adults, 33 (4.4%, 95% CI 2.9-5.9%) reported they had been tattooed at least once. This group included 22 men and 11 women. The youngest age at which a respondent had been tattooed was 13 years, the oldest was 50 years (mean 20.9 years, SID 7.2 years). Number of tattoos per person ranged from one to 20 (mode and median = 1); one respondent said half his body was covered in tattoos.

Twenty-five of the 33 tattooed respondents (75.8%) had all their tattoos professionally applied; the remaining eight (24.2%) had acquired tattoos from amateurs and/or on premises other than registered tattoo parlours. Of these, one had been tattooed at a friend's house by a professional tattooist using standard equipment; another reported being tattooed by a friend using sewing needles; a third was given a tattoo on a public footpath with a broken razor blade. Two of the eight reported tattooing being performed in a group setting.

Eight tattooed individuals (24.2%) reported having been tested for HCV. Twenty-five (75.8%) recognised that there was a risk of HCV infection associated with tattooing. Five of the tattooed respondents also self-reported injecting drug use (Table 2).

Skin piercing

Piercing of the ears, face and body was very common; 407 respondents (53.8%, 95% CI 50.2-57.3%) reported at least one form of piercing. Three hundred and thirty-one women (73.2%) were pierced compared to 76 men (24.9%). Ear piercings were overwhelmingly the most popular form; of 402 individuals with pierced ears, 58.7% had two holes pierced, and there was a range of one to 11 holes. Sixty per cent had ear piercing performed at a pharmacy, 15.9% at a hairdressing or beauty salon, 6.7% at a jeweller's shop; 12.4% were pierced by a relative or friend, 11.4% by a doctor or nurse and 0.5% by a tattooist (multiple piercings mean these figures sum to more than 100%); 7.2% had pierced themselves.

Nine respondents (1.2%) reported face piercing, all of whom had one hole. Face piercing was most commonly performed at a pharmacy (44.4%). Six respondents had body piercings (1 -3 holes); body piercing was most often performed by a tattooist (50.0%).

Of the 407 respondents with any form of skin piercing, 74 (14.7%) reported that their piercings were performed by a friend or relative, by themselves, or elsewhere than a registered piercing business. Sixty (13.3%) reported illnesses probably associated with the piercing procedure, mostly inflammation or local infections. One person reported hepatitis (type unknown), saying that a doctor had advised that the piercing was the most likely source of the infection.

Table 1: Knowledge of HCV risks.

Table 2: Multiple risk factors for HCV infection

Note:
(a) Chi squared significance of proportion compared to rest of sample, p<0.05.
(b) Chi squared significance of proportion compared to rest of sample, p<0.001.

Medical and occupation exposure

Eighty-four respondents (11. 1 %) claimed to have had a blood transfusion prior to 1990, when blood donations began to be screened for hepatitis C virus. (This figure seems somewhat high, but is lower than the 14.3% reported in a study of Australian blood donors.) (22) Respondents were asked if they had undergone medical treatment other than a transfusion which put them at risk of hepatitis C infection, and if their occupations exposed them to HCV risk. Twenty-three (3.0%) identified some form of potentially risky medical procedure. Another 153 individuals (20.2%) believed their occupation put them at risk of HCV. The largest group with occupational hazards were health workers (69) including nurses, doctors, dentists, domestic and orderly staff, and cleaning staff.

Hepatitis C testing

Eighty-six respondents (11.4%, 95% CI 9.1-13.6%) reported having been tested for hepatitis C. These respondents reported a range of risk factors for HCV; skin piercing (57), occupational risk (27), IDU (10), blood transfusion before 1990 (9), tattooing (8) and medical procedures (4).

Crude odds ratios were calculated for the association of HCV testing with the demographic and other risk factor variables. Respondents aged 25-44 years were more likely to have had a hepatitis C test, as were those with a history of piercing, acupuncture, a previous test for hepatitis B, hepatitis B vaccination, a tattoo, blood donation or injecting drug use. Respondents who believed their occupation put them at risk of HCV infection were more likely to have bad a test.

A logistic regression equation was computed using the listed variables. Missing data meant 221 (29%) of the 757 subjects had to be excluded from the regression. As Table 3 shows, neither age nor occupation was a predictor of hepatitis C testing. Statistically significant associations exist for gender, a history of donating blood, of hepatitis B tests, of hepatitis B vaccination and of injecting drug use.

Prevalence of risk factors in the community

Age and sex-standardised prevalences for the three risk practices and the number who have been tested for HCV were estimated for the Victorian population. Our estimates are shown in Table 4.

Discussion

This is the first population-based estimate of the prevalence of risk factors for HCV infection other than injecting drug use in the Australian community. In interpreting our results, it should be noted that the telephone survey methodology, the non-participation rate and the high refusal rate are potential sources of bias. Telephone surveys obviously exclude households without telephones, and thereby under-represent people in lower socioeconomic groups, older people, the homeless, and people living in institutional accommodation such as hospitals, hotels, prisons and other non-private dwellings. (23,24)

Other sampling biases were the exclusion of people because of language difficulties and those not contacted within five telephone calls. Biases identifiable in the final sample included an over-representation of women, an under-representation of the 15-24 year age group and an overrepresentation of the 35-44 year age group. A likely outcome of such biases is an under-estimate of involvement in intravenous drug use (and to a lesser extent, tattooing) due to the relatively high proportion of illicit drug use in younger age groups and by men.

At 2.2%, the prevalence of IDU calculated in this study is significantly higher than that identified in the 1995 Victorian Drug Strategy Household Survey, which estimated that 31,900 Victorians aged 15 years and over had ever injected illicit drugs, equivalent to 0.9% of this population group. (25) It is notable that a relatively high percentage was recorded in a telephone survey, a method normally regarded as ineffective for collecting data on sensitive issues. (24)

Table 3: Associations with testing for hepatitis C virus infection

As noted above, both tattooing and skin piercing are identified as possible forms of transmission for HCV but their contribution to the overall spread of the virus is undoubtedly significantly less than that attributed to IDU. Clearly, this is due to the relative inefficiency of these procedures in transmitting the virus (which, in turn, must be related to the volume of potentially infected blood involved), but other characteristics of the behaviours and frequency of involvement may also be important. Our data imply that most tattooing and skin piercing in Victoria takes place in regulated settings in which infection control procedures are legally required. In addition, the average frequency of involvement in tattooing and skin piercing is very low, with the majority of tattooed respondents reporting one tattoo, and most pierced respondents having only one or two piercings and being pierced on only one occasion. This means that even when an infection is transmitted, the newly infected person is unlikely to participate in these practices again, and thus the opportunity for the spread of the virus by these procedures is constrained.

By contrast, injecting drug use often occurs in unhygienic circumstances and is frequently performed by individuals with minimal knowledge of infection control. Injecting often takes place in groups, and individuals may inject multiple times per day. Under these circumstances HCV can spread very rapidly, and prevalences of HCV in Australian IDUs of 60% or greater are testament to the efficiency of this method of transmission. (26)

While the contribution of tattooing and skin piercing to the spread of HCV in the general community is likely to be low, it does not mean that they do not pose significant individual risk if proper infection control is not in place. In addition, even if IDU is identified as the practice of primary concern, there is significant crossover between the different groups. In our sample, tattooing and skin piercing were common among IDU, the former being at significantly higher prevalence than in the general population.

We have estimated that more than 389,000 Victorians 15 years or over have been tested for HCV but this figure may be inflated, and may include people aware of HCV testing associated with blood donations. Of the 86 in our sample who reported being tested for HCV, 48 (46.5%) had also donated blood. The multivariate analysis confirms the link between testing and donating blood and, in accordance with the Victorian surveillance data, women are more likely to be tested. An association with prior testing for HBV was also identified, probably indicating a common cause for concern. The increased likelihood of IDUs being tested compared to the general population confirms that there is heightened awareness of the risk of HCV among this group.

It is of concern that a high proportion of Victorians still believe that receipt of a blood transfusion is a risk for HCV infection, and that almost a fifth believe that donating blood is a risk. Similarly, the confusion over transmission routes, with many unsure about specific modes, may imply that there is confusion in the minds of many between the different types of viral hepatitis. This may be the explanation for the high proportions reporting or unsure of risk from sexual transmission or contaminated food or water. Unhappily, such uncertainty over the ways in which HCV is transmitted contributes to the widespread stigmatisation and discrimination against HCV-infected individuals. (27) Community education and information programs focusing on risk factors for hepatitis C should also highlight what is not a risk, in particular that transfusion is now safe.

Conclusions

This study shows that the risk factors associated with the spread of HCV are widespread in the community. Although tattooing and skin piercing represent less risk than injecting drug use, non-negligible percentages of Victorians are having these procedures carried out in ways that put them at increased risk. Greater efforts may be required to encourage people to be tattooed and pierced in registered premises where safe operating procedures and infection control are more likely to be practised. Nevertheless, on the whole there is limited potential for tattooing and skin piercing to contribute to the ongoing spread of the hepatitis C virus across the community. Despite the low prevalence of injecting drug use found in our survey, there can be little doubt that it will remain the main mode of transmission for the hepatitis C virus for the foreseeable future.

Table 4: Estimated proportion and numbers of the victorian population (aged 15+) with risk for HCV, and who have been tested for HCV.

References

1 . Alter MI, Margolis HS, Krawczynski K, Judson FN, et al. The natural history of community-acquired hepatitis C in the
United States. N Engl 3 Med 1992; 327(27): 1899-905.

2. Andrews R, Curran M. Enhanced surveillance for incident cases of hepatitis C in Australia. 1995. Commun Dis Intell 1996;
20: 384-8.

3. Strasser SL Risk factors and predictors of outcome in an Australian cohort with hepatitis C infection. Med JAust 1995; 162:
355-8.

4. Ko YC, Ho MS, Chiang TA, et al. Tattooing as a risk of hepatitis C virus infection. I Med Virol 1992; 38(4): 288-91.

5. Thompson SC, Hemberger F, Wale E, Crofts N. Hepatitis C transmission through tattooing: a case report. Aust NZJ Public Health 1996; 20(3): 3178.

6. Abdool Karim SS, Thejpal R, Singh B. High prevalence of hepatitis B virus infection in rural black adults in Mselemni, South
Africa. Am 3 Public Health 1989; 79(7): 893-4.

7. Conry-Cantilena C, VanRaden M, Gibble 3, et al. Routes of infection, viremia, and liver disease in blood donors found to
have hepatitis C virus infection. N Engl J Med 1996; 334(26): 1691-6.

8. Cooksley WG, Butterworth LA. Hepatitis C virus infection in health care workers referred to a hepatitis clinic. Med JAust
1996; 164(11): 656-8.

9. Mihaly I, Lukacs A, Telegdy L, lbranyi E. Screening for hepatitis C of hospital personnel at the Szent Laszlo Hospital of
Budapest. Orv Hetil 1996; 137(50): 2791-4.

10. Granovsky MO, Minkoff HL, Tess BH, et al. Hepatitis C virus infection in the mothers and infants cohort study. Pediatrics
1998; 102(2 Pt 1): 355-9.

11. Resti M,Azzari C, Mannelli Fet al. Mother to child transmission of hepatitis C virus: Prospective study of risk factors and timing of infection in children born to women seronegative for HIV-1 Br Med] 1998; 317(7156): 437-41.

12. Kobayashi M, Tanaka E, Oguchi H, et al. Prospective follow-up study of hepatitis C virus infection in patients undergoing maintenance haemodialysis: Comparison among haemodialysis units. 3 Gastroenterol Hepatol 1998@ 13(6): 604-9.

13. Saltoglu N. Tasova Y, Burgut R, Dundar IH. Sexual and non-sexual intrafamilial spread of hepatitis C virus: Intrafamilial
transmission of HCV Eur J Epiderniol 1998; 14(3): 225-8.

14. Osella AR, Massa MA, Joekes S, et al. Hepatitis B and C virus sexual transmission among homosexual men. Am J
Gastroenterol 1998, 93(1): 49-52.

15. Wyld R, Robertson JR, Brettle RE, et al. Absence of hepatitis C virus transmission but frequent transmission of HIV-1 from sexual contact with doublyinfected individuals. J In/cr 1997; 35(2): 163-6.

16. Stevenson E, Rodger A, editors. Surveillance of sexually transmissible diseases in Victoria 1996. Melbourne: Infectious
Diseases Unit, Department of Human Services, 1997.

17. Brylor, Oz on Disk [CD-ROM]. Telephone directory Sydney (NSW): Brylor Pty Ltd, 1995.

18. Australian Bureau of Statistics. Labour Force, Victoria. Canberra@ ABS, 1995. Catalogue No.: 3207.2.

19. Velu R, Naidu GM. A review of current sampling methods in market research. In: Krishnaiah RR, Krishnaiah MAC, Rao CR,
editors. Handbook of Statistics. New York: Elsevier, 1988.

20. Australian Bureau of Statistics. Estimated resident population by age and sex in statistical local areas, Victoria. Canberra:
ABS, 1996. Catalogue No.: 3207.2.

21. Daly LE, Bourke GI, McGilvray J. Interpretation and uses of medical statistics. Oxford: Blackwell Scientific, 1991.

22. Kaldor JM, Archer GT, Buring ML, et al. Risk factors for hepatitis C virus infection in blood donors: A case-control study
Med J Aust 1992; 157(4): 227-30.

23. Groves RM, Lyberg L. An overview of non-response issues in telephone surveys. In: Groves RM, editor. Survey errors and
survey costs. New York: Wiley, 1988.

24. Steel D, Vella 3, Harrington P. Quality-issues in telephone surveys: Coverage, non-response and quota sampling. Aust J
Slat 1996; 380): 15-34.

25. AGB McNair/Department of Health and Community Services. Victorian drug strategy household survey report, 1995.
Melbourne: Government Printer, 1995.

26. Crofts N, Jolley D, Kaldor 3, et al. The epidemiology of Hepatitis C virus infection among injecting drug users in Australia.
3 Epidemiol Communit - V Health 1997; 51(6): 692-7.

27. Crofts N. Louie R, Loff B. The next plague: Stigmatisation and discrimination related to hepatitis C virus infection in
Australia. 3 Health Hum Rights 1997; 2(2): 86-97.
4/13/2000 2:53