- Email: [email protected]
An opportunity lost: HIV infections associated with lack of a national needle-exchange programme in the USA
THE LANCET
An opportunity lost: HIV infections associated with lack of a national needle-exchange programme in the USA
Peter Lurie, Ernest Drucker
Summary Background Our aim was to estimate the number of HIV infections that could have been prevented had needleexchange programmes been implemented during the early stages of the AIDS epidemic in the USA. We also estimated t he c ost t o t he US healt h-c are sy st em t o t reat t hese preventable HIV infections. Methods The formula w e used to calculate the annual number of preventable HIV infections accounted for the effec t i v eness and l ev el of use of needl e- ex c hang e programmes, as well as sex ual transmission to injection drug users ( IDUs) and secondary transmission to their sex ual partners and children. Data for the model w ere obtained from epidemiological and mathematical studies in peer-reviewed published research, government reports, and consultations with experts. Using data from Australia as a model, w e calculated the number of HIV infections that could have been prevented by a national needle-ex change programme in the USA betw een 1987 and 1995. Cost calculations w ere based on the current US government estimate of the discounted lifetime cost of treating an HIV infection (US$55 640). Findings Our conservative calculation of the number of HIV infections that could have been prevented ranged from 4394 (15% incidence reduction due to needle ex changes) to 9666 (33% incidence reduction). The cost to the US healt h-c are sy st em of t reat ing t hese prevent able HIV infections is between US$ 244 million and US$ 538 million, respectively. If current US policies are not changed, we estimate that an additional 5150–11 329 preventable HIV infections could occur by the year 2000. Interpretation The failure of the federal government in the USA to implement a national needle-exchange programme, despite six government-funded reports in support of needle exchanges, may have led to HIV infection among thousands of IDUs, their sexual partners, and their children. Revoking the US government ban on funding for needle-ex change pr og r ammes and ac c el er at i ng t he g r ow t h of suc h programmes in the USA are urgent public-health priorities.
Lancet 1997; 349: 604–08
Center for AIDS Prevention Studies, University of California, Suite 600, San Francisco, California 94105, USA (P Lurie MPH), and Department of Epidemiology and Social Medicine, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, New York (Prof E Drucker PhD) Correspondence to: Dr Peter Lurie
604
Introduction In the USA, HIV infection is the leading cause of death among individuals aged 25–44 years. According to Holmberg,1 from the Centers for Disease Control and Prevention (CDC), most of the 41 000 new HIV infections each year occur among injection drug users (IDUs) and their sexual partners and children. Thus, the prevention of HIV transmission among IDUs should be a cornerstone of any attempt to stem the HIV epidemic in the USA. Because only about 15% of the estimated 1–1·5 million IDUs in the USA are in drug treatment on any given day, and because hypochlorite bleach has only limited efficacy for the disinfection of injection equipment, the provision of sterile syringes to IDUs seems to be the most viable method for reducing the transmission of HIV among active IDUs who continue to inject.2,3 With only limited empirical research in hand, public-health officials in countries other than the USA relied on standard precepts of infection control—for example, “If the reservoir or source [of an infectious disease] is an inanimate object, then it may be controlled by either decontamination procedures or by using disposable materials”4—to formulate their initial responses to the AIDS epidemic. In Amsterdam, a needle-exchange programme opened as early as 1984, in an effort to reduce the transmission of blood-borne infections such as hepatitis B. Thus, when evidence on the efficacy of such programmes began to accumulate, most countries in western Europe, as well as in Australia, New Zealand, and Canada, were able to move rapidly to provide sterile syringes to IDUs through a combination of needle-exchange programmes and increased availability of sterile injection equipment through pharmacies. By contrast, in the USA, opposition to needle-exchange programmes arose from some drug-treatment providers, ethnic minority communities, law-enforcement officials, politicians, local business people, and residents. These groups asserted that programmes would lead to increased drug use and would also divert public funds from already under-funded drug-treatment facilities. Consequently, needle-exchange programmes in the USA are limited in number and generally small in scale. Furthermore, about three dozen staff who work for such programmes have been arrested.3 Since 1988, US law has banned the use of federal funds for needle-exchange programmes. The bans generally require that, before federal funds can be used, the programmes must be shown to reduce the transmission of HIV infection and not to lead to increased illicit drug use.5 In maintaining a ban on national funding for these programmes, the USA is unique in the world. There is much evidence that needle-exchange programmes can reduce the incidence of HIV infection as part of a comprehensive strategy of HIV prevention that includes increased availability of drug treatment, programmes of public education about drugs, and the
Vol 349 • March 1, 1997
THE LANCET
provision of syringes at pharmacies. Needle-exchange programmes are associated with substantial reductions in the sharing of syringes, the referral of large numbers of IDUs to drug-treatment facilities, and a six-fold and seven-fold reduction in the transmission of hepatitis B and C, respectively.3,6 Des Jarlais and colleagues7 showed that needle-exchange programmes were also associated with stable HIV seroprevalences among IDU populations in cities that had implemented the programmes. However, the low incidence of HIV among IDUs who agree to be followed up in prospective studies requires very large sample sizes to show significant intervention effects. Logistical and ethical impediments to randomised trials further preclude such “definitive” evidence of an association between reduced seroincidence and needleexchange programmes. Therefore, mathematical models have been used to obtain quantitative estimates of the effectiveness of needle-exchange programmes in reducing HIV seroconversion rates.3,8 Six government-sponsored reviews of needle-exchange programmes2,3,5,9–11 concluded that such programmes reduce the incidence of HIV infection among IDUs and do not lead to an increase in rates of drug use—ie, the criteria that must be met before the ban on federal funding for needle-exchange programmes can be lifted. Moreover, four of the reviews made policy recommendations that the federal funding ban be revoked and that statutes extant in ten states that require a physician’s prescription to obtain or sell a syringe be rescinded.2,3,9,10 Such laws, and other state restrictions upon the sale and possession of syringes (paraphernalia laws), restrict the availability of sterile syringes and potentially lead to the sharing of syringes and the transmission of HIV. The public-health community involved with HIV and AIDS is, therefore, almost unanimous in its judgment that needle-exchange programmes are one of the most effective ways to reduce the incidence of HIV infection and the burden of mortality and morbidity associated with AIDS among IDUs. The wide discrepancy between US government policies on needle-exchange programmes and those of most other industrialised countries prompted us to conduct this study. We report conservative estimates of the number of HIV infections that could have been prevented by a national policy of needle-exchange programmes in the USA between 1987 and 1995, and of the cost of the failure to implement such a policy.
Methods We used the following formula to estimate the number of HIV infections that could have been prevented by the implementation of needle-exchange programmes between 1987 and 1995 in the USA. 1995 p=∑it⫻(1–x)⫻ut⫻e⫻s t=1987 where: p=the number of preventable infections; it=the number of incident HIV infections among IDUs in each year; x=the proportion of incident HIV infections among IDUs that are from sexual transmission; ut=the proportion of IDUs who could have used a needle-exchange programme in each year; e=the proportionate reduction in HIV incidence among IDUs because of needle-exchange programmes; s=the ratio of primary HIV infections among IDUs plus secondary infections among the sexual partners and children to primary HIV infections among IDUs. This formula assumes the conditional independence of the variables. The model predicts that incident HIV infections will be reduced by the product of the proportion of IDUs who use
Vol 349 • March 1, 1997
needle-exchange programmes and the effectiveness of the programmes in reducing HIV transmission among IDUs who take part in the programmes. This product is then modified to account for sexual transmission to IDUs as a result of unsafe sex (not syringe-sharing) and secondary transmission from IDUs to their sexual partners and children. The annual number of incident HIV infections (it) is difficult to calculate, because risk-group-specific incidence rates in the USA were not published until 1996. For at least 6 years, the CDC estimated the total annual number of incident HIV infections at 40 000–80 000.12 In 1987, 17% of AIDS cases in the USA occurred among IDUs, excluding men who were IDUs and also had sex with men.2 Using the lower estimate of 40 000 HIV infections and this proportion, we calculated that 6800 HIV infections occurred among IDUs in 1987. This figure is probably an underestimate, because HIV infection precedes clinical AIDS by a median of 10 years and the proportion of AIDS cases among IDUs has increased since the beginning of the epidemic. 2 Holmberg1 used a components model with data available through 1994 to update the incidence estimates, and calculated that there are about 38 100 incident HIV infections annually in the 96 largest metropolitan areas in the USA, of which 19 000 occur among IDUs. We used linear interpolation to calculate the number of incident HIV infections between 1987 and 1994. This approach underestimates the number of incident infections in the intervening period, because the epidemic curve for HIV among IDUs characteristically shows a saturation phenomenon. This levelling off of HIV seroprevalence among IDUs was already evident in New York City by 1984. 13 For 1995, we used the 1994 estimate of 19 000 incident HIV infections among IDUs. We reduced the number of HIV infections among IDUs to account for the proportion of HIV infections that are the result of sexual behaviour (x). This proportion is difficult to determine because the risk factors for HIV infection among most IDUs include drug use and sexual behaviour. Based on consultation with four researchers who have conducted prospective studies of IDUs and Kaplan’s study14 that used a range of 30–40% for sexrelated transmission of HIV among IDUs, we estimated that 35% of HIV infections among IDUs were caused by sexual transmission in each year of our study (x=0·35). We used data from Loxley and colleagues’ 1994 crosssectional study15 of 872 IDUs in four Australian cities to calculate the proportion of IDUs who could have used a needle-exchange programme at least once in any given year (ut) in the USA. Since 1987, Australia has implemented needle-exchange programmes with the guidance and support of national and regional health authorities; indeed, Australia can be viewed as a model for the feasibility of a national policy of needle-exchange programmes. Furthermore, national data on the use of needle-exchange programmes is more complete for Australia than for any other country. Loxley and colleagues15 found that 49·2% of IDUs had used a needle-exchange programme in the previous month, even though Australia, unlike the USA, has a substantial programme of syringe provision at pharmacies. The proportion of IDUs who use needle exchanges in a given month is less than the proportion who use the exchanges at least once during a given year. However, we used data on monthly exchange users to be consistent with definitions of needle-exchange-programme effectiveness in the mathematical models.3,8 We assumed that the USA could have initiated a national needle-exchange programme in 1987, and that the programme would have grown at the same annual rate to cover similar proportions of the IDU population. Thus, we assumed that 0% of IDUs in the USA could have used needle exchanges in 1987 and that 49·2% could have used exchanges in 1994, and estimated the use of needle exchanges for the intervening years by linear interpolation. We believe that this technique introduces a conservative bias into our calculations. The proportion of Australian IDUs who identified needleexchange programmes as their usual source of syringes rose rapidly to 22·0% in 1989.16 Thereafter, the programme expanded more slowly: in 1994, 43·1% of IDUs identified needle exchanges as their main source of syringes. 15 This pattern of
605
Cumulative number of preventable HIV infections
THE LANCET
25 000 20 000 15 000
Higher estimate of NEP effectiveness Lower estimate of NEP effectiveness
1991 (9) 1993 (5) 1993 (3) 1993 (10) 1995 (2) 1995 (11)
10 000 5000 0
00 20 9 9 19 8 9 19 7 9 19 6 9 19 5 9 19 4 9 19 3 9 19 2 9 19 1 9 19 0 9 19 9 8 19 8 8 19 7 8 19 Year Cumulative number of NEP-preventable HIV infections in the USA between 1987 and 2000 NEP=needle-ex change programmes. Arrows represent year of publication of six government-funded studies of efficacy of NEPs (see table).
exchange use in Australia suggests that the curve describing the use of needle exchanges over time shows evidence of saturation, and that a straight line connecting the curve’s two ends would underestimate IDUs’ use of needle exchanges. For 1995, we assumed that potential needle exchange use remained at 49·2% of IDUs. Because it is difficult to calculate the percentage decline in HIV incidence among IDUs who use needle exchanges (e) from prospective studies, we relied on data from two studies3,8 that depended on the integration of detailed empirical data with mathematical modelling. The model, developed for the needleexchange programme in New Haven, Connecticut, USA, was based on needle-circulation theory, which assumes that the proportion of IDUs who share syringes remains constant, and that only the circulation pattern of the syringes is altered by the needle exchange. In this model, an IDU was defined as a needleexchange participant if he or she used the exchange at least once in a given month. The circulation-theory model estimated that participants in the needle-exchange programme had a 33% lower HIV incidence than non-participants.8 We are aware of no data that the effectiveness of needle-exchange programmes increases or decreases over time, or that the New Haven programme is more or less effective than other needle-exchange programmes. In the second study of different cities in the USA, three different models (one based on needle-circulation theory, one on behaviour changes by IDUs, and one on a combination of these two theories) were constructed and yielded IDU incidence reduction estimates that ranged from 15% to 70%.3 In the behavioural models, a broad definition of needle-exchange participation was used that ranged from ever use to regular use of the needle exchange. For this study, we used 15% as a lower limit and 33% as an upper limit for e, and assumed that this incidence reduction was constant over time. To estimate the ratio of primary plus secondary HIV infections to primary infections (s), we used data on AIDS cases reported to the CDC.17 In 1994, there were 21 717 AIDS cases reported among IDUs, and 3853 among men who were IDUs and also had sex with men; inclusion of this latter group drives the estimate of s downwards. In that same year, 2952 non-IDUs were reported to have developed AIDS from sex with IDUs, and a further 417 paediatric cases were attributed to perinatal transmission from a mother who either was an IDU herself or had had sex with one. Thus, in 1994, s=1·13. This value, which we assumed remained constant over time, is probably an underestimate because these data are based on AIDS cases, not HIV infections, and the proportion of secondary transmissions can be expected to rise over time as infections spread from IDUs to their sex partners and children. We calculated the direct medical costs of HIV infections that could have been prevented had a national needle-exchange programme been implemented (preventable HIV infections) by multiplying the number of preventable HIV infections by the lifetime cost of treating an infection, discounted into present
606
Year of publication (ref no)
Conclusions and recommendations of studies Reduction in HIV transmission
No increase in drug use
Revoke federal funding ban
Revoke state prescription and paraphernalia laws
Yes* Yes† Yes Yes Yes‡ Yes
Yes Yes Yes Yes Yes Yes
Yes N/A Yes Yes Yes N/A
Yes N/A Yes Yes Yes N/A
NA=not available because studies reviewed data without making policy recommendations. *Legal barriers that preclude needle exchange lead to increased HIV transmission. †Research suggests promise as an AIDS prevention strategy. ‡Reducing proportion of contaminated syringes in circulation will reduce HIV transmission.
Findings of US government-funded studies of efficacy of needle-exchange programmes value. Indirect costs were not included. We used the lowest available estimate, US$55 640, for the entire study period.18
Results The figure shows the cumulative number of HIV infections that could have been prevented had a national needle-exchange programme been implemented in the USA in 1987, and then expanded at the same rate as the Australian programme. At the lower estimate of the effectiveness of such a programme (15% incidence reduction), 4394 HIV infections could have been prevented between 1987 and 1995. At the higher estimate of effectiveness (33%), 9666 HIV infections could have been prevented. 88% of these infections would have been prevented among IDUs, with the remainder occurring among their sexual partners and children. The findings of the six US government-funded studies of needle-exchange programmes were published between 1991 and 1995 (table).2,3,5,9–11 Thus, even as evidence of the effectiveness of needle-exchange programmes accumulated, the cumulative number of preventable HIV infections continued to rise exponentially (figure). We also estimated preventable infections in the USA for the period 1996–2000, by assuming that all model variables, including it and ut, remained unchanged at their 1995 values. This analysis showed that, between 1996 and 2000, 5150–11 329 additional HIV infections could still be prevented if the US government were to immediately expand needle-exchange coverage to current Australian levels. We found that the costs of treating preventable HIV infections between 1987 and 1995 ranged from US$244 million, at the lower estimate of effectiveness of a national needle-exchange programme, to US$538 million at the higher estimate. The corresponding costs for 1996–2000 range from US$287 million to US$630 million.
Discussion Our data show that the absence of a national needleexchange programme in the USA has already contributed to an estimated 4000–10 000 preventable HIV infections, and to societal costs for treating these infections of between a quarter and half a billion dollars. The costs in terms of human suffering and loss are, of course, impossible to quantify. Our analysis was dependent upon the availability of appropriate data. Except for the data on the use of needleexchange programmes in Australia, all model inputs were obtained from US sources. We believe that all our variable
Vol 349 • March 1, 1997
THE LANCET
estimates are conservative. For example, we used the lower end of the CDC’s estimate of 40 000–80 000 incident HIV infections to estimate 1987 IDU incidence. Similarly, we used rates of AIDS cases to estimate ratios of secondary HIV transmission. For any given year, the ratio of secondary transmission AIDS cases to AIDS cases among IDUs will underestimate that ratio for incident HIV infection if secondary transmission represents an increasing proportion of incident HIV infections relative to incident HIV infections among IDUs. In addition, because the CDC’s estimates of incident HIV infections implicitly takes the existence of US needle-exchange programmes into account, the numbers of HIV infections preventable by needle exchanges reported here are an underestimate. Similarly, the estimate of the lifetime costs of HIV treatment is the lowest available, and was formulated before the approval of the expensive protease inhibitors. Finally, Des Jarlais and colleagues’ 1996 study,19 which compared HIV incidence among IDUs who used needle exchanges with those who did not, suggests that needle exchanges may have a stronger protective effect against HIV seroconversion than we assumed in our study, a hazard ratio of 3·35 (95% CI 1·3–9·1) among IDUs who did not use needle exchanges. If we apply those estimates of needle-exchange efficacy to our model, 20 504 HIV infections could have been prevented by a national needle-exchange programme in the USA between 1987 and 1995 and an additional 24 032 will occur between 1996 and 2000 if the programmes do not expand. We believe that it was feasible for a national needleexchange programme to have been established in the USA in 1987. In 1986, the Municipal Health Service in Amsterdam had already taken over the needle-exchange programme from the drug users’ union, which had set up the programme in 1984.3 In 1987, the UK initiated its own pilot needle-exchange programme.3 Similarly, by 1987, France had revoked its prescription law, which means that the only country in western Europe to retain a prescription law is Sweden.3 Indeed, as early as 1985, some cities in the USA, such as New York, were considering the introduction of needle-exchange programmes.20 We also believe that the high proportion of IDUs who use needle exchanges in Australia can be achieved elsewhere. In Amsterdam, 52·3% of IDUs obtained all their syringes from needle exchanges in 1988,21 and in San Francisco, where needle exchanges are considered illegal by state officials but run, in part, with city funds, 61% of IDUs who were interviewed in 1992 reported having used needle exchanges during the previous 12 months.22 Even if needle-exchange programmes had been expanded at the rate of the Australian programme only after the first government-funded study was published in 1991,9 between 1417 and 3118 HIV infections could have been prevented in the USA by the end of 1995. The high costs of treating preventable HIV infections must be weighed against the cost of providing needle exchanges. In 1992, the average budget of US and Canadian needle-exchange programmes was about US$168 650 per programme.3 Thus, for 1987–95, the cost of treating preventable HIV infections (US$244–538 million) is equivalent to operating between 161 and 354 needle-exchange programmes. At a cost of US$1·35 per syringe3 these programmes could have distributed between
Vol 349 • March 1, 1997
181 and 398 million syringes between 1987 and 1995 with the funds. Despite the federal funding ban, the frequent and continuing arrests of needle-exchange-programme staff and users, and chronic shortages in funding, needle exchanges have expanded in the USA. There were 37 programmes in the USA in 19933 and 68 in 1994.23 Nonetheless, the US needle-exchange programme is far smaller relative to its IDU population than its Australian counterpart. In 1994, needle exchanges in the USA distributed 7·9 million syringes23 to a population of about 1·5 million IDUs—ie, less than one syringe per IDU per month. By contrast in Fiscal Year 1993–94, Australian needle exchanges distributed 6 million syringes to an IDU population that included 172 000 heroin users, with an additional 4·3 million syringes provided to IDUs through pharmacies.24 The large scale of the preventable tragedy of HIV infection among IDUs in the USA requires that we learn from it and redouble our efforts to initiate and sustain needle-exchange programmes. Leading government investigators of substance misuse have already advocated just such a course; in December, 1993, they recommended that the ban on federal funding for needleexchange programmes be lifted.25 When asked to review the University of California report,3 the CDC consulted with all four agencies within the Department of Health and Human Services with responsibility for substance misuse issues; their recommendation was clear: “We conclude that the ban on Federal funding of NEPs should be lifted to allow communities and States to use Federal funds to support [needle-exchange programmes] as components of comprehensive HIV prevention programs”.10 However, rather than implement the recommendations of its own scientists, the government elected to suppress the report, which only became public when obtained by the Washington Post on Feb 16, 1995. 47% of preventable HIV infections during our study period (between 2060 and 4532 infections) have occurred since the CDC made its recommendation. While government officials continue to obstruct needle-exchange programmes, public opinion is shifting in favour of the programmes. A 1996 US national telephone poll found that 66% of respondents supported the provision of sterile syringes to IDUs, including majorities in both political parties.26 The failure of the US federal government to initiate a national programme of needle exchanges has shifted the emphasis of HIV prevention among IDUs to the states. Some states have risen to the challenge: Connecticut, Hawaii, Maryland, Massachusetts, New York, and Rhode Island have allocated funds to needle-exchange programmes. In October, 1993, Maine revoked its law requiring a prescription to possess or sell a syringe. Similar legal changes in Connecticut, in July, 1992, led to a 39% relative reduction in the sharing of syringes.27 Although some states and local communities have responded appropriately to the void created by the federal opposition to needle-exchange programmes, the role of the federal government remains critical. First, federal funding would allow a substantial expansion of current needle-exchange programmes. Second, such funding would allow better communication between needle exchanges and other government-sponsored services for IDUs, such as drug treatment programmes. Third, federal 607
THE LANCET
government funding would send a clear signal to the states that needle exchanges have an important role in the prevention of HIV infection among IDUs; some states have used the federal opposition to justify their own inaction. The dangers of the US government approach have particular relevance for other countries only now coming to terms with HIV infection among IDUs, such as countries in eastern Europe and southeast Asia, which should not view US policy as a model to be emulated. Australia has made pharmacy-based programmes that increase the availability of sterile syringes a central part of its HIV prevention programme among IDUs. Clearly, some IDUs prefer the anonymity afforded by pharmacy purchase of syringes, whereas others opt for the free syringes and access to other public-health services more readily available through needle exchanges. The Australian experience seems to indicate that the more choices we offer IDUs, the more likely they are to reduce their risk behaviours. In the USA, however, access to sterile syringes through needle exchanges and pharmacies remains limited. By revoking laws that ban federal funding for needle exchanges as well as those state laws that impede pharmacy access to sterile syringes, and by educating pharmacists about the importance of selling sterile syringes to IDUs, we can greatly reduce the number of HIV infections that would otherwise occur. Our study shows that rapid expansion of needle-exchange programmes in the USA could prevent between 5000 and 11 000 HIV infections by the end of the millenium. We thank Edward H Kaplan and Alex Wodak for their comments and assistance.
7
8 9
10
11 12
13
14 15
16
17 18
19
20 21
References 1 2
3
4
5
6
Holmberg SD. The estimated prevalence and incidence of HIV in 96 large US metropolitan areas. Am J Public Health 1996; 86: 642–54. Normand J, Vlahov D, Moses LE, eds. Preventing HIV transmission: the role of sterile needles and bleach. National Research Council/Institute of Medicine. Washington, DC: National Academy Press, 1995. Lurie P, Reingold AL, eds. The public health impact of needle exchange programs in the United States and abroad, volume I. University of California, 1993. Brachman PS. Transmission and principles of control. In: Mandell GL, Douglas RG, Bennett JE, eds. Principles and practice of infections diseases, 2nd edn. New York: John Wiley & Sons, 1985. General Accounting Office. Needle exchange programs: research suggests promise as an AIDS prevention strategy (GAO/HRD-93-60). Washington, DC: US Government Printing Office, 1993. Hagan H, Des Jarlais DC, Friedman SR, Purchase D, Alter MJ.
608
22
23 24
25 26 27
Reduced risk of hepatitis B and hepatitis C among injection drug users in the Tacoma Syringe Exchange Program. Am J Public Health 1995; 85: 1531–37. Des Jarlais DC, Hagan H, Friedman SR, et al. Maintaining low HIV seroprevalence in populations of injecting drug users. JAMA 1995; 274: 1226–31. Kaplan EH. A method for evaluating needle exchange programmes. Stats Med 1994; 13: 2179–87. National Commission on Acquired Immune Deficiency Syndrome. The twin epidemics of substance use and HIV. Washington, DC, 1991. Satcher D. Note to Jo Ivey Boufford, Dec 10, 1993. Available from the Drug Policy Foundaton, 4455 Connecticut Avenue, NW, Suite B-500, Washington, DC, 20008, USA. Office of Technology Assessment. The effectiveness AIDS prevention efforts. Washington, DC, 1995. Centers for Disease Control. HIV prevalence estimates and AIDS case projections for the United States: report based upon a workshop. MMWR 1990; 39 (no 16): 8. Des Jarlais DC, Friedman SR, Novick DM, et al. HIV-1 infection among intravenous drug users in Manhattan, New York City, from 1977 through 1987. JAMA 1989; 261: 1008–12. Kaplan EH. Probability models of needle exchange. Op Res 1995; 43: 558–69. Loxley W, Carruthers S, Bevan J. In the same vein: first report of the Australian study of HIV and injecting drug use. National Centre for Research into the Prevention of Drug Abuse, Curtin University of Technology. Perth, 1995. Australian National AIDS and Injecting Drug Use Study. Neither a borrower nor a lender be: first report of the Australian National AIDS and Injecting Drug Use Study: Sydney, 1991. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report 1994; 6: 10. Guinan ME, Farnham PG, Holtgrave DR. Estimating the value of preventing a human immunodeficiency virus infection. Am J Prev Med 1994; 10: 1–4. Des Jarlais DC, Marmor M, Paone D, et al. HIV incidence among injecting drug users in New York City syringe-exchange programmes. Lancet 1996; 348: 987–91. Waldholz M. New York City’s health unit urges easier syringe rule. Wall Street Journal Sept 3, 1985: A-14. Van den Hoek JA, van Haastrecht HJ, Coutinho RA. Risk reduction among intravenous drug users in Amsterdam under the influence of AIDS. Am J Public Health 1989; 79: 1355–57. Watters JK, Estilo MJ, Clark GL, Lorvick J. Syringe and needle exchange as HIV/AIDS prevention for injection drug users. JAMA 1994; 271: 115–20. Paone D, Des Jarlais DC, Clark J, et al. Syringe exchange programs— United States, 1994–1995. MMWR 1995; 44: 684–85. Feachem RGA. Valuing the past. . .investing in the future. Commonwealth Department of Human Services and Health. Canberra: Australian Government Publishing Service, 1995. Lurie P. When science and politics collide: the federal response to needle-exchange programs. Bull NY Acad Med 1995; 72: 380–96. Anon. The Kaiser survey on Americans and AIDS/HIV. Henry J Kaiser Family Foundation, Menlo Park, CA, March 26, 1996. Groseclose SL, Weinstein B, Jones TS, Valleroy LA, Fehrs LJ, Kassler WJ. Impact of increased legal access to needles and syringes on practices of injecting-drug users and police officers—Connecticut, 1992–1993. J Acquir Immune Defic Syndr Hum Retrovirol 1995; 10: 82–89.
Vol 349 • March 1, 1997
An opportunity lost: HIV infections associated with lack of a national needle-exchange programme in the USA
Peter Lurie, Ernest Drucker
Summary Background Our aim was to estimate the number of HIV infections that could have been prevented had needleexchange programmes been implemented during the early stages of the AIDS epidemic in the USA. We also estimated t he c ost t o t he US healt h-c are sy st em t o t reat t hese preventable HIV infections. Methods The formula w e used to calculate the annual number of preventable HIV infections accounted for the effec t i v eness and l ev el of use of needl e- ex c hang e programmes, as well as sex ual transmission to injection drug users ( IDUs) and secondary transmission to their sex ual partners and children. Data for the model w ere obtained from epidemiological and mathematical studies in peer-reviewed published research, government reports, and consultations with experts. Using data from Australia as a model, w e calculated the number of HIV infections that could have been prevented by a national needle-ex change programme in the USA betw een 1987 and 1995. Cost calculations w ere based on the current US government estimate of the discounted lifetime cost of treating an HIV infection (US$55 640). Findings Our conservative calculation of the number of HIV infections that could have been prevented ranged from 4394 (15% incidence reduction due to needle ex changes) to 9666 (33% incidence reduction). The cost to the US healt h-c are sy st em of t reat ing t hese prevent able HIV infections is between US$ 244 million and US$ 538 million, respectively. If current US policies are not changed, we estimate that an additional 5150–11 329 preventable HIV infections could occur by the year 2000. Interpretation The failure of the federal government in the USA to implement a national needle-exchange programme, despite six government-funded reports in support of needle exchanges, may have led to HIV infection among thousands of IDUs, their sexual partners, and their children. Revoking the US government ban on funding for needle-ex change pr og r ammes and ac c el er at i ng t he g r ow t h of suc h programmes in the USA are urgent public-health priorities.
Lancet 1997; 349: 604–08
Center for AIDS Prevention Studies, University of California, Suite 600, San Francisco, California 94105, USA (P Lurie MPH), and Department of Epidemiology and Social Medicine, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, New York (Prof E Drucker PhD) Correspondence to: Dr Peter Lurie
604
Introduction In the USA, HIV infection is the leading cause of death among individuals aged 25–44 years. According to Holmberg,1 from the Centers for Disease Control and Prevention (CDC), most of the 41 000 new HIV infections each year occur among injection drug users (IDUs) and their sexual partners and children. Thus, the prevention of HIV transmission among IDUs should be a cornerstone of any attempt to stem the HIV epidemic in the USA. Because only about 15% of the estimated 1–1·5 million IDUs in the USA are in drug treatment on any given day, and because hypochlorite bleach has only limited efficacy for the disinfection of injection equipment, the provision of sterile syringes to IDUs seems to be the most viable method for reducing the transmission of HIV among active IDUs who continue to inject.2,3 With only limited empirical research in hand, public-health officials in countries other than the USA relied on standard precepts of infection control—for example, “If the reservoir or source [of an infectious disease] is an inanimate object, then it may be controlled by either decontamination procedures or by using disposable materials”4—to formulate their initial responses to the AIDS epidemic. In Amsterdam, a needle-exchange programme opened as early as 1984, in an effort to reduce the transmission of blood-borne infections such as hepatitis B. Thus, when evidence on the efficacy of such programmes began to accumulate, most countries in western Europe, as well as in Australia, New Zealand, and Canada, were able to move rapidly to provide sterile syringes to IDUs through a combination of needle-exchange programmes and increased availability of sterile injection equipment through pharmacies. By contrast, in the USA, opposition to needle-exchange programmes arose from some drug-treatment providers, ethnic minority communities, law-enforcement officials, politicians, local business people, and residents. These groups asserted that programmes would lead to increased drug use and would also divert public funds from already under-funded drug-treatment facilities. Consequently, needle-exchange programmes in the USA are limited in number and generally small in scale. Furthermore, about three dozen staff who work for such programmes have been arrested.3 Since 1988, US law has banned the use of federal funds for needle-exchange programmes. The bans generally require that, before federal funds can be used, the programmes must be shown to reduce the transmission of HIV infection and not to lead to increased illicit drug use.5 In maintaining a ban on national funding for these programmes, the USA is unique in the world. There is much evidence that needle-exchange programmes can reduce the incidence of HIV infection as part of a comprehensive strategy of HIV prevention that includes increased availability of drug treatment, programmes of public education about drugs, and the
Vol 349 • March 1, 1997
THE LANCET
provision of syringes at pharmacies. Needle-exchange programmes are associated with substantial reductions in the sharing of syringes, the referral of large numbers of IDUs to drug-treatment facilities, and a six-fold and seven-fold reduction in the transmission of hepatitis B and C, respectively.3,6 Des Jarlais and colleagues7 showed that needle-exchange programmes were also associated with stable HIV seroprevalences among IDU populations in cities that had implemented the programmes. However, the low incidence of HIV among IDUs who agree to be followed up in prospective studies requires very large sample sizes to show significant intervention effects. Logistical and ethical impediments to randomised trials further preclude such “definitive” evidence of an association between reduced seroincidence and needleexchange programmes. Therefore, mathematical models have been used to obtain quantitative estimates of the effectiveness of needle-exchange programmes in reducing HIV seroconversion rates.3,8 Six government-sponsored reviews of needle-exchange programmes2,3,5,9–11 concluded that such programmes reduce the incidence of HIV infection among IDUs and do not lead to an increase in rates of drug use—ie, the criteria that must be met before the ban on federal funding for needle-exchange programmes can be lifted. Moreover, four of the reviews made policy recommendations that the federal funding ban be revoked and that statutes extant in ten states that require a physician’s prescription to obtain or sell a syringe be rescinded.2,3,9,10 Such laws, and other state restrictions upon the sale and possession of syringes (paraphernalia laws), restrict the availability of sterile syringes and potentially lead to the sharing of syringes and the transmission of HIV. The public-health community involved with HIV and AIDS is, therefore, almost unanimous in its judgment that needle-exchange programmes are one of the most effective ways to reduce the incidence of HIV infection and the burden of mortality and morbidity associated with AIDS among IDUs. The wide discrepancy between US government policies on needle-exchange programmes and those of most other industrialised countries prompted us to conduct this study. We report conservative estimates of the number of HIV infections that could have been prevented by a national policy of needle-exchange programmes in the USA between 1987 and 1995, and of the cost of the failure to implement such a policy.
Methods We used the following formula to estimate the number of HIV infections that could have been prevented by the implementation of needle-exchange programmes between 1987 and 1995 in the USA. 1995 p=∑it⫻(1–x)⫻ut⫻e⫻s t=1987 where: p=the number of preventable infections; it=the number of incident HIV infections among IDUs in each year; x=the proportion of incident HIV infections among IDUs that are from sexual transmission; ut=the proportion of IDUs who could have used a needle-exchange programme in each year; e=the proportionate reduction in HIV incidence among IDUs because of needle-exchange programmes; s=the ratio of primary HIV infections among IDUs plus secondary infections among the sexual partners and children to primary HIV infections among IDUs. This formula assumes the conditional independence of the variables. The model predicts that incident HIV infections will be reduced by the product of the proportion of IDUs who use
Vol 349 • March 1, 1997
needle-exchange programmes and the effectiveness of the programmes in reducing HIV transmission among IDUs who take part in the programmes. This product is then modified to account for sexual transmission to IDUs as a result of unsafe sex (not syringe-sharing) and secondary transmission from IDUs to their sexual partners and children. The annual number of incident HIV infections (it) is difficult to calculate, because risk-group-specific incidence rates in the USA were not published until 1996. For at least 6 years, the CDC estimated the total annual number of incident HIV infections at 40 000–80 000.12 In 1987, 17% of AIDS cases in the USA occurred among IDUs, excluding men who were IDUs and also had sex with men.2 Using the lower estimate of 40 000 HIV infections and this proportion, we calculated that 6800 HIV infections occurred among IDUs in 1987. This figure is probably an underestimate, because HIV infection precedes clinical AIDS by a median of 10 years and the proportion of AIDS cases among IDUs has increased since the beginning of the epidemic. 2 Holmberg1 used a components model with data available through 1994 to update the incidence estimates, and calculated that there are about 38 100 incident HIV infections annually in the 96 largest metropolitan areas in the USA, of which 19 000 occur among IDUs. We used linear interpolation to calculate the number of incident HIV infections between 1987 and 1994. This approach underestimates the number of incident infections in the intervening period, because the epidemic curve for HIV among IDUs characteristically shows a saturation phenomenon. This levelling off of HIV seroprevalence among IDUs was already evident in New York City by 1984. 13 For 1995, we used the 1994 estimate of 19 000 incident HIV infections among IDUs. We reduced the number of HIV infections among IDUs to account for the proportion of HIV infections that are the result of sexual behaviour (x). This proportion is difficult to determine because the risk factors for HIV infection among most IDUs include drug use and sexual behaviour. Based on consultation with four researchers who have conducted prospective studies of IDUs and Kaplan’s study14 that used a range of 30–40% for sexrelated transmission of HIV among IDUs, we estimated that 35% of HIV infections among IDUs were caused by sexual transmission in each year of our study (x=0·35). We used data from Loxley and colleagues’ 1994 crosssectional study15 of 872 IDUs in four Australian cities to calculate the proportion of IDUs who could have used a needle-exchange programme at least once in any given year (ut) in the USA. Since 1987, Australia has implemented needle-exchange programmes with the guidance and support of national and regional health authorities; indeed, Australia can be viewed as a model for the feasibility of a national policy of needle-exchange programmes. Furthermore, national data on the use of needle-exchange programmes is more complete for Australia than for any other country. Loxley and colleagues15 found that 49·2% of IDUs had used a needle-exchange programme in the previous month, even though Australia, unlike the USA, has a substantial programme of syringe provision at pharmacies. The proportion of IDUs who use needle exchanges in a given month is less than the proportion who use the exchanges at least once during a given year. However, we used data on monthly exchange users to be consistent with definitions of needle-exchange-programme effectiveness in the mathematical models.3,8 We assumed that the USA could have initiated a national needle-exchange programme in 1987, and that the programme would have grown at the same annual rate to cover similar proportions of the IDU population. Thus, we assumed that 0% of IDUs in the USA could have used needle exchanges in 1987 and that 49·2% could have used exchanges in 1994, and estimated the use of needle exchanges for the intervening years by linear interpolation. We believe that this technique introduces a conservative bias into our calculations. The proportion of Australian IDUs who identified needleexchange programmes as their usual source of syringes rose rapidly to 22·0% in 1989.16 Thereafter, the programme expanded more slowly: in 1994, 43·1% of IDUs identified needle exchanges as their main source of syringes. 15 This pattern of
605
Cumulative number of preventable HIV infections
THE LANCET
25 000 20 000 15 000
Higher estimate of NEP effectiveness Lower estimate of NEP effectiveness
1991 (9) 1993 (5) 1993 (3) 1993 (10) 1995 (2) 1995 (11)
10 000 5000 0
00 20 9 9 19 8 9 19 7 9 19 6 9 19 5 9 19 4 9 19 3 9 19 2 9 19 1 9 19 0 9 19 9 8 19 8 8 19 7 8 19 Year Cumulative number of NEP-preventable HIV infections in the USA between 1987 and 2000 NEP=needle-ex change programmes. Arrows represent year of publication of six government-funded studies of efficacy of NEPs (see table).
exchange use in Australia suggests that the curve describing the use of needle exchanges over time shows evidence of saturation, and that a straight line connecting the curve’s two ends would underestimate IDUs’ use of needle exchanges. For 1995, we assumed that potential needle exchange use remained at 49·2% of IDUs. Because it is difficult to calculate the percentage decline in HIV incidence among IDUs who use needle exchanges (e) from prospective studies, we relied on data from two studies3,8 that depended on the integration of detailed empirical data with mathematical modelling. The model, developed for the needleexchange programme in New Haven, Connecticut, USA, was based on needle-circulation theory, which assumes that the proportion of IDUs who share syringes remains constant, and that only the circulation pattern of the syringes is altered by the needle exchange. In this model, an IDU was defined as a needleexchange participant if he or she used the exchange at least once in a given month. The circulation-theory model estimated that participants in the needle-exchange programme had a 33% lower HIV incidence than non-participants.8 We are aware of no data that the effectiveness of needle-exchange programmes increases or decreases over time, or that the New Haven programme is more or less effective than other needle-exchange programmes. In the second study of different cities in the USA, three different models (one based on needle-circulation theory, one on behaviour changes by IDUs, and one on a combination of these two theories) were constructed and yielded IDU incidence reduction estimates that ranged from 15% to 70%.3 In the behavioural models, a broad definition of needle-exchange participation was used that ranged from ever use to regular use of the needle exchange. For this study, we used 15% as a lower limit and 33% as an upper limit for e, and assumed that this incidence reduction was constant over time. To estimate the ratio of primary plus secondary HIV infections to primary infections (s), we used data on AIDS cases reported to the CDC.17 In 1994, there were 21 717 AIDS cases reported among IDUs, and 3853 among men who were IDUs and also had sex with men; inclusion of this latter group drives the estimate of s downwards. In that same year, 2952 non-IDUs were reported to have developed AIDS from sex with IDUs, and a further 417 paediatric cases were attributed to perinatal transmission from a mother who either was an IDU herself or had had sex with one. Thus, in 1994, s=1·13. This value, which we assumed remained constant over time, is probably an underestimate because these data are based on AIDS cases, not HIV infections, and the proportion of secondary transmissions can be expected to rise over time as infections spread from IDUs to their sex partners and children. We calculated the direct medical costs of HIV infections that could have been prevented had a national needle-exchange programme been implemented (preventable HIV infections) by multiplying the number of preventable HIV infections by the lifetime cost of treating an infection, discounted into present
606
Year of publication (ref no)
Conclusions and recommendations of studies Reduction in HIV transmission
No increase in drug use
Revoke federal funding ban
Revoke state prescription and paraphernalia laws
Yes* Yes† Yes Yes Yes‡ Yes
Yes Yes Yes Yes Yes Yes
Yes N/A Yes Yes Yes N/A
Yes N/A Yes Yes Yes N/A
NA=not available because studies reviewed data without making policy recommendations. *Legal barriers that preclude needle exchange lead to increased HIV transmission. †Research suggests promise as an AIDS prevention strategy. ‡Reducing proportion of contaminated syringes in circulation will reduce HIV transmission.
Findings of US government-funded studies of efficacy of needle-exchange programmes value. Indirect costs were not included. We used the lowest available estimate, US$55 640, for the entire study period.18
Results The figure shows the cumulative number of HIV infections that could have been prevented had a national needle-exchange programme been implemented in the USA in 1987, and then expanded at the same rate as the Australian programme. At the lower estimate of the effectiveness of such a programme (15% incidence reduction), 4394 HIV infections could have been prevented between 1987 and 1995. At the higher estimate of effectiveness (33%), 9666 HIV infections could have been prevented. 88% of these infections would have been prevented among IDUs, with the remainder occurring among their sexual partners and children. The findings of the six US government-funded studies of needle-exchange programmes were published between 1991 and 1995 (table).2,3,5,9–11 Thus, even as evidence of the effectiveness of needle-exchange programmes accumulated, the cumulative number of preventable HIV infections continued to rise exponentially (figure). We also estimated preventable infections in the USA for the period 1996–2000, by assuming that all model variables, including it and ut, remained unchanged at their 1995 values. This analysis showed that, between 1996 and 2000, 5150–11 329 additional HIV infections could still be prevented if the US government were to immediately expand needle-exchange coverage to current Australian levels. We found that the costs of treating preventable HIV infections between 1987 and 1995 ranged from US$244 million, at the lower estimate of effectiveness of a national needle-exchange programme, to US$538 million at the higher estimate. The corresponding costs for 1996–2000 range from US$287 million to US$630 million.
Discussion Our data show that the absence of a national needleexchange programme in the USA has already contributed to an estimated 4000–10 000 preventable HIV infections, and to societal costs for treating these infections of between a quarter and half a billion dollars. The costs in terms of human suffering and loss are, of course, impossible to quantify. Our analysis was dependent upon the availability of appropriate data. Except for the data on the use of needleexchange programmes in Australia, all model inputs were obtained from US sources. We believe that all our variable
Vol 349 • March 1, 1997
THE LANCET
estimates are conservative. For example, we used the lower end of the CDC’s estimate of 40 000–80 000 incident HIV infections to estimate 1987 IDU incidence. Similarly, we used rates of AIDS cases to estimate ratios of secondary HIV transmission. For any given year, the ratio of secondary transmission AIDS cases to AIDS cases among IDUs will underestimate that ratio for incident HIV infection if secondary transmission represents an increasing proportion of incident HIV infections relative to incident HIV infections among IDUs. In addition, because the CDC’s estimates of incident HIV infections implicitly takes the existence of US needle-exchange programmes into account, the numbers of HIV infections preventable by needle exchanges reported here are an underestimate. Similarly, the estimate of the lifetime costs of HIV treatment is the lowest available, and was formulated before the approval of the expensive protease inhibitors. Finally, Des Jarlais and colleagues’ 1996 study,19 which compared HIV incidence among IDUs who used needle exchanges with those who did not, suggests that needle exchanges may have a stronger protective effect against HIV seroconversion than we assumed in our study, a hazard ratio of 3·35 (95% CI 1·3–9·1) among IDUs who did not use needle exchanges. If we apply those estimates of needle-exchange efficacy to our model, 20 504 HIV infections could have been prevented by a national needle-exchange programme in the USA between 1987 and 1995 and an additional 24 032 will occur between 1996 and 2000 if the programmes do not expand. We believe that it was feasible for a national needleexchange programme to have been established in the USA in 1987. In 1986, the Municipal Health Service in Amsterdam had already taken over the needle-exchange programme from the drug users’ union, which had set up the programme in 1984.3 In 1987, the UK initiated its own pilot needle-exchange programme.3 Similarly, by 1987, France had revoked its prescription law, which means that the only country in western Europe to retain a prescription law is Sweden.3 Indeed, as early as 1985, some cities in the USA, such as New York, were considering the introduction of needle-exchange programmes.20 We also believe that the high proportion of IDUs who use needle exchanges in Australia can be achieved elsewhere. In Amsterdam, 52·3% of IDUs obtained all their syringes from needle exchanges in 1988,21 and in San Francisco, where needle exchanges are considered illegal by state officials but run, in part, with city funds, 61% of IDUs who were interviewed in 1992 reported having used needle exchanges during the previous 12 months.22 Even if needle-exchange programmes had been expanded at the rate of the Australian programme only after the first government-funded study was published in 1991,9 between 1417 and 3118 HIV infections could have been prevented in the USA by the end of 1995. The high costs of treating preventable HIV infections must be weighed against the cost of providing needle exchanges. In 1992, the average budget of US and Canadian needle-exchange programmes was about US$168 650 per programme.3 Thus, for 1987–95, the cost of treating preventable HIV infections (US$244–538 million) is equivalent to operating between 161 and 354 needle-exchange programmes. At a cost of US$1·35 per syringe3 these programmes could have distributed between
Vol 349 • March 1, 1997
181 and 398 million syringes between 1987 and 1995 with the funds. Despite the federal funding ban, the frequent and continuing arrests of needle-exchange-programme staff and users, and chronic shortages in funding, needle exchanges have expanded in the USA. There were 37 programmes in the USA in 19933 and 68 in 1994.23 Nonetheless, the US needle-exchange programme is far smaller relative to its IDU population than its Australian counterpart. In 1994, needle exchanges in the USA distributed 7·9 million syringes23 to a population of about 1·5 million IDUs—ie, less than one syringe per IDU per month. By contrast in Fiscal Year 1993–94, Australian needle exchanges distributed 6 million syringes to an IDU population that included 172 000 heroin users, with an additional 4·3 million syringes provided to IDUs through pharmacies.24 The large scale of the preventable tragedy of HIV infection among IDUs in the USA requires that we learn from it and redouble our efforts to initiate and sustain needle-exchange programmes. Leading government investigators of substance misuse have already advocated just such a course; in December, 1993, they recommended that the ban on federal funding for needleexchange programmes be lifted.25 When asked to review the University of California report,3 the CDC consulted with all four agencies within the Department of Health and Human Services with responsibility for substance misuse issues; their recommendation was clear: “We conclude that the ban on Federal funding of NEPs should be lifted to allow communities and States to use Federal funds to support [needle-exchange programmes] as components of comprehensive HIV prevention programs”.10 However, rather than implement the recommendations of its own scientists, the government elected to suppress the report, which only became public when obtained by the Washington Post on Feb 16, 1995. 47% of preventable HIV infections during our study period (between 2060 and 4532 infections) have occurred since the CDC made its recommendation. While government officials continue to obstruct needle-exchange programmes, public opinion is shifting in favour of the programmes. A 1996 US national telephone poll found that 66% of respondents supported the provision of sterile syringes to IDUs, including majorities in both political parties.26 The failure of the US federal government to initiate a national programme of needle exchanges has shifted the emphasis of HIV prevention among IDUs to the states. Some states have risen to the challenge: Connecticut, Hawaii, Maryland, Massachusetts, New York, and Rhode Island have allocated funds to needle-exchange programmes. In October, 1993, Maine revoked its law requiring a prescription to possess or sell a syringe. Similar legal changes in Connecticut, in July, 1992, led to a 39% relative reduction in the sharing of syringes.27 Although some states and local communities have responded appropriately to the void created by the federal opposition to needle-exchange programmes, the role of the federal government remains critical. First, federal funding would allow a substantial expansion of current needle-exchange programmes. Second, such funding would allow better communication between needle exchanges and other government-sponsored services for IDUs, such as drug treatment programmes. Third, federal 607
THE LANCET
government funding would send a clear signal to the states that needle exchanges have an important role in the prevention of HIV infection among IDUs; some states have used the federal opposition to justify their own inaction. The dangers of the US government approach have particular relevance for other countries only now coming to terms with HIV infection among IDUs, such as countries in eastern Europe and southeast Asia, which should not view US policy as a model to be emulated. Australia has made pharmacy-based programmes that increase the availability of sterile syringes a central part of its HIV prevention programme among IDUs. Clearly, some IDUs prefer the anonymity afforded by pharmacy purchase of syringes, whereas others opt for the free syringes and access to other public-health services more readily available through needle exchanges. The Australian experience seems to indicate that the more choices we offer IDUs, the more likely they are to reduce their risk behaviours. In the USA, however, access to sterile syringes through needle exchanges and pharmacies remains limited. By revoking laws that ban federal funding for needle exchanges as well as those state laws that impede pharmacy access to sterile syringes, and by educating pharmacists about the importance of selling sterile syringes to IDUs, we can greatly reduce the number of HIV infections that would otherwise occur. Our study shows that rapid expansion of needle-exchange programmes in the USA could prevent between 5000 and 11 000 HIV infections by the end of the millenium. We thank Edward H Kaplan and Alex Wodak for their comments and assistance.
7
8 9
10
11 12
13
14 15
16
17 18
19
20 21
References 1 2
3
4
5
6
Holmberg SD. The estimated prevalence and incidence of HIV in 96 large US metropolitan areas. Am J Public Health 1996; 86: 642–54. Normand J, Vlahov D, Moses LE, eds. Preventing HIV transmission: the role of sterile needles and bleach. National Research Council/Institute of Medicine. Washington, DC: National Academy Press, 1995. Lurie P, Reingold AL, eds. The public health impact of needle exchange programs in the United States and abroad, volume I. University of California, 1993. Brachman PS. Transmission and principles of control. In: Mandell GL, Douglas RG, Bennett JE, eds. Principles and practice of infections diseases, 2nd edn. New York: John Wiley & Sons, 1985. General Accounting Office. Needle exchange programs: research suggests promise as an AIDS prevention strategy (GAO/HRD-93-60). Washington, DC: US Government Printing Office, 1993. Hagan H, Des Jarlais DC, Friedman SR, Purchase D, Alter MJ.
608
22
23 24
25 26 27
Reduced risk of hepatitis B and hepatitis C among injection drug users in the Tacoma Syringe Exchange Program. Am J Public Health 1995; 85: 1531–37. Des Jarlais DC, Hagan H, Friedman SR, et al. Maintaining low HIV seroprevalence in populations of injecting drug users. JAMA 1995; 274: 1226–31. Kaplan EH. A method for evaluating needle exchange programmes. Stats Med 1994; 13: 2179–87. National Commission on Acquired Immune Deficiency Syndrome. The twin epidemics of substance use and HIV. Washington, DC, 1991. Satcher D. Note to Jo Ivey Boufford, Dec 10, 1993. Available from the Drug Policy Foundaton, 4455 Connecticut Avenue, NW, Suite B-500, Washington, DC, 20008, USA. Office of Technology Assessment. The effectiveness AIDS prevention efforts. Washington, DC, 1995. Centers for Disease Control. HIV prevalence estimates and AIDS case projections for the United States: report based upon a workshop. MMWR 1990; 39 (no 16): 8. Des Jarlais DC, Friedman SR, Novick DM, et al. HIV-1 infection among intravenous drug users in Manhattan, New York City, from 1977 through 1987. JAMA 1989; 261: 1008–12. Kaplan EH. Probability models of needle exchange. Op Res 1995; 43: 558–69. Loxley W, Carruthers S, Bevan J. In the same vein: first report of the Australian study of HIV and injecting drug use. National Centre for Research into the Prevention of Drug Abuse, Curtin University of Technology. Perth, 1995. Australian National AIDS and Injecting Drug Use Study. Neither a borrower nor a lender be: first report of the Australian National AIDS and Injecting Drug Use Study: Sydney, 1991. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report 1994; 6: 10. Guinan ME, Farnham PG, Holtgrave DR. Estimating the value of preventing a human immunodeficiency virus infection. Am J Prev Med 1994; 10: 1–4. Des Jarlais DC, Marmor M, Paone D, et al. HIV incidence among injecting drug users in New York City syringe-exchange programmes. Lancet 1996; 348: 987–91. Waldholz M. New York City’s health unit urges easier syringe rule. Wall Street Journal Sept 3, 1985: A-14. Van den Hoek JA, van Haastrecht HJ, Coutinho RA. Risk reduction among intravenous drug users in Amsterdam under the influence of AIDS. Am J Public Health 1989; 79: 1355–57. Watters JK, Estilo MJ, Clark GL, Lorvick J. Syringe and needle exchange as HIV/AIDS prevention for injection drug users. JAMA 1994; 271: 115–20. Paone D, Des Jarlais DC, Clark J, et al. Syringe exchange programs— United States, 1994–1995. MMWR 1995; 44: 684–85. Feachem RGA. Valuing the past. . .investing in the future. Commonwealth Department of Human Services and Health. Canberra: Australian Government Publishing Service, 1995. Lurie P. When science and politics collide: the federal response to needle-exchange programs. Bull NY Acad Med 1995; 72: 380–96. Anon. The Kaiser survey on Americans and AIDS/HIV. Henry J Kaiser Family Foundation, Menlo Park, CA, March 26, 1996. Groseclose SL, Weinstein B, Jones TS, Valleroy LA, Fehrs LJ, Kassler WJ. Impact of increased legal access to needles and syringes on practices of injecting-drug users and police officers—Connecticut, 1992–1993. J Acquir Immune Defic Syndr Hum Retrovirol 1995; 10: 82–89.
Vol 349 • March 1, 1997