Given the explosive and ongoing epidemics of syphilis and gonorrhea among men who have sex with men (MSM) and, more recently, outbreaks of Lymphogranuloma venereum and hepatitis C among the HIV-infected MSM in industrialized countries, a possible increase in HIV incidence has been suggested [1–3]. However, findings on HIV incidence have been inconsistent among MSM. In the past decade, the UK, Sydney and San Francisco reported stable trends, whereas Amsterdam, Italy and Canada reported rising trends [4–9]. Reliable measurement of the HIV incidence in a population is difficult . Indirect methods, such as interpreting trends based on the registering of new HIV diagnoses, are subject to bias due to changes in screening policy, for example. Sources that yield direct estimates are scarce, being mainly prospective cohort studies and anonymous HIV surveillance studies in which a less sensitive (LS) HIV assay is used to determine HIV incidence . Their limitation is that they may comprise only a selected, namely high risk, part of the population under investigation.
In The Netherlands, as in many industrialized countries, male homosexual contact is still the largest transmission category, accounting for 52% of all new infections according to the Dutch HIV monitoring foundation . The western part of the Netherlands, including Rotterdam and Amsterdam, harbors the majority (76%) of the known HIV-infected MSM population. In this region, there are the three national sources of monitoring HIV incidence among MSM; two prospective cohort studies [Rotterdam Cohort Study (ROHOCO) and Amsterdam Cohort Study (ACS)] [13–15] and one HIV surveillance study in which MSM are tested with a LS HIV assay (Amsterdam STI clinic) . The ethics committees of the Rotterdam Erasmus MC and the Amsterdam GGD/Academic Medical Center approved the study protocols.
We have previously demonstrated an increase in HIV incidence up to 2001 among older MSM attending the Amsterdam STI clinic, while finding no increase in the younger population of both the sexually transmitted infections (STI) clinic and the ACS up to 2001 [7–15]. Herein updated HIV incidence rates up to 2005 are presented, together with rates from Rotterdam. In particular, the time-trends of the three study groups have been investigated and the relationship between HIV incidence and age assessed. As MSM in The Netherlands historically have a markedly low rate of HIV testing, namely 54% in 2003 , the HIV incidence in relation to HIV testing behavior has also been investigated.
Materials and methods
General procedures and population
Amsterdam STI clinic
The STI clinic of the Health Service of Amsterdam offers anonymous and free-of-charge examination for STI in an outpatient setting. Over the past 5 years, there were approximately 20 000 new consultations annually. From 1991 to 2005, except for 1993, anonymous and unlinked cross-sectional HIV prevalence surveys were conducted at the clinic twice a year (or once a year in 2004) . In this ongoing program, attendees are considered eligible for these surveys when they visit the clinic for evaluation of a possible new STI episode during a survey period. Each year (in spring and autumn), approximately 2000 clinic attendees are surveyed. Attendees who consent to participate are interviewed on sociodemographic and sexual characteristics, HIV test history and use of HAART, and they are anonymously tested for HIV antibodies. Participants can refuse participation in the study. All clinic attendees are routinely screened for gonorrhea, syphilis and chlamydia. For the present study, data were used from male participants in the HIV prevalence surveys in 1991–2005, who identified themselves as being gay (approximately 15% of all participants) and who consented to blood testing for HIV (over 95% of all MSM participants), totaling 4401 MSM . Data on current diagnoses of infectious syphilis and gonorrhea were available for analyses from 1998 onwards.
Amsterdam Cohort Study
The ACS among MSM is an open cohort that started in 1984 among HIV-negative and positive men, of all age groups, who had had at least one male sexual partner in the previous year and lived in or near Amsterdam . In the period 1995–2004 only men aged ≤ 30 years could enter the study and older HIV-negative individuals were no longer followed after October 1996. Recruitment has entailed ‘convenience sampling’ (using brochures at the STI clinic and advertisements in the gay scene) and ‘chain referral sampling’ (participants recruited by other participants). At each semi-annual follow-up visit, blood was drawn for HIV antibody testing, a structured behavioral questionnaire was taken and data on self-reported STI was collected. Of 1677 HIV-negative men who were recruited into the ACS between 1984 and 2005, 1498 had at least one follow-up visit.
Rotterdam Cohort Study
From February 1999 to February 2000, 286 MSM were recruited to participate in the 5-year ROHOCO in Rotterdam, as a closed cohort [13,14,19]. Both HIV-positive and HIV-negative men who had had sex with at least one male partner in the preceding year were recruited by trained volunteers at gay meeting-places such as bars and saunas in Rotterdam. At each semi-annual visit, demographic and sexual behavior information was collected using structured questionnaires, and the men were tested for STI, including gonorrhea and infectious syphilis. The ROHOCO ended in 2003. Of the 286 recruits, all of the 265 HIV-negative men with at least one follow-up visit were included.
Definition of recent HIV infection
At all three study sites, HIV antibodies were tested by commercially available recombinant HIV-1/-2 enzyme assays, mainly enzyme immunoassay (EIA; Abbott Laboratories, Abbott Park, Illinois, USA) and micropartide enzyme immunoassay AxSym HIV-1/2 reagens (Abbott, Santa Clara, California, USA) with confirmation by the western blot test (mainly Diagnostics Biotechnology, Herent, Belgium and Innogenetics, Gent, Belgium). In the ACS and ROHOCO participants, the date of HIV seroconversion was estimated as the midpoint between the last HIV-negative and the first HIV-positive test. For the Amsterdam STI clinic participants, a different procedure was used to estimate HIV incidence, as described earlier . Briefly, stored HIV-positive samples from which sufficient material was available were tested with both a sensitive and a less sensitive EIA (LS assay) (Organon Teknika Vironostika HIV enzyme immunoassay; Organon Teknika, Durham, North Carolina, USA) to determine which persons were recently infected. Persons who tested reactive with the sensitive HIV assay and non-reactive with the LS assay were considered to be in the period of early HIV infection [i.e., less than 170 days, 95% confidence interval (CI), 162–183 days], when antibody titer is increasing but has not yet peaked .There was insufficient material for testing 33 (4.6%) of the 721 HIV-infected participants. All seropositive persons reporting a known HIV-positive status were considered as having longstanding infections and were excluded to avoid possible misclassification of recent infections among persons longer infected with HIV or among persons using HAART [11,20].
In ACS and ROHOCO, HIV incidence was calculated per calendar year as the number of new infections divided by total person years (PY) under observation. For the Amsterdam STI clinic, annual HIV incidence was calculated as the prevalence of persons with recent infection (n) among the susceptible population (all HIV-negative plus recently infected men, N), divided by duration (T) of the mean time between seroconversion on the assays (here: 170 days), thus I = [(n/N)/(T/365.25)] × 100. To determine the 95% CI for the incidences, we used the Bonferroni principle as described earlier [7,11,21].
First, to evaluate trends in HIV incidence over time, HIV incidence was modeled separately for each study site with calendar year as a continuous variable using restricted cubic splines (for STI clinic and ACS). For ROHOCO, a linear model was applied as here the study duration was only 5 years. As we were a priori interested in investigating trends for different years of age, age was included as a covariate, using restricted cubic splines, and allowed for interaction between calendar year and age. Second, HIV incidence was calculated for three continuous time periods. Their cut-off values were defined based on the start of the three studies, namely 1984–1990, 1991–1998 and 1999–2003/5. Third, we investigated whether high-risk populations could be identified by exploring HIV incidence in different age groups, in Dutch and non-Dutch MSM, in men with and without an STI co-infection, and in those who did or did not report a history of HIV testing. Here, analyses were restricted to the period 1999–2003/5 in order to yield results of interest for current prevention strategies and as in that period, data were available for all three study settings. Additionally, in the ACS and STI clinic population differences in risk factors between the most recent period and the preceding periods were assessed; therefore data from the complete study period was included and allowance for interaction between time period (in three categories) and the factor of interest was included. We calculated crude and adjusted (for age, nationality, STI, and HIV testing history when appropriate) odds ratios (OR) or relative risks (RR) and their corresponding 95% CI. We used a logistic regression model to analyze data from the Amsterdam STI clinic, as described earlier [7,11,21], and a Poisson regression model for the two cohort studies. A P-value of < 0.05 was considered to be statistically significant. The statistical package ‘R’ was used for all analyses [22,23].
The basic characteristics of the three studies are shown in Table 1. Notably, the Rotterdam group (ROHOCO) is somewhat older, while the ACS and STI clinic populations have the largest sample size. Furthermore, the STI clinic has the highest proportion of MSM infected with STI and the highest proportion of non-Dutch MSM.
HIV incidence over time (1984–2005)
The crude HIV incidence for time periods that are comparable for the three study sites are shown in Table 2. In 1999–2003/5 incidence was significantly higher compared to 1991–1998 at the Amsterdam STI clinic. Likewise, in the ACS, a slightly higher incidence was observed in this time period, although it was not statistically significant. HIV incidences in the ACS and ROHOCO were comparable.
Figure 1 depicts the estimated HIV incidence by way of linear or spline-modeled regression analyses, with calendar time as the explaining variable and age as the effect modifier. For the Amsterdam STI clinic, an overall increase in HIV incidence over time was observed (P = 0.0334). However, time trends differed across age groups (P = 0.0135). The HIV incidence increased only in older men and did not change in the young. Consequently, HIV incidence in older MSM surpassed that in the younger men over time (Fig. 1a).
For ROHOCO, similar time trends and age effects were observed. However, the time trends differed with only a marginal statistical significance between age groups (P = 0.0831); no overall time-trend was present (P = 0.2227), but numbers were small (Fig. 1c).
The ACS covers the longest study period, and Fig. 1b shows the observed decrease in incidence from 1984 to 1990 followed by a period of relative stable incidence. The recent increase since 1999 was not statistically significant (Fig. 1b) Trends did not differ across years of age (P = 0.6247). Of note, only in the ACS, HIV incidence could in recent years be calculated only for younger MSM, as follow-up of MSM aged older than 30 years was terminated in 1997.
HIV-testing behavior, demographics and sexually transmitted co-infection in relation to current HIV incidence (1999–2005)
To explore the factors that are associated with acquisition of HIV infection, we assessed for each site whether HIV incidence differed according to history of a HIV test and certain demographic characteristics or STI co-infection for the years 1999–2005 (Table 3).
Previous HIV testing in relation to HIV incidence was explored only for MSM attending the Amsterdam STI clinic, since all ACS and ROHOCO participants had been tested before and were largely aware of their HIV serostatus. After 1999, comparable HIV incidences were observed in ever (negative) tested and in never tested men. However, before 1999, HIV incidence was remarkably lower in never tested MSM compared with men who had tested negative previously (P for interaction of testing and time period: 0.06). Of the 49 men who participated in anonymous HIV testing at the STI clinic and appeared to have a recent HIV infection, only 28% (i.e. 11 of the 39 men with available data on current testing) also accepted named HIV testing at that time. The proportion of those men with recent infection who accepted named HIV testing increased from 21% before 2002 to 40% after 2002, the year the STI clinic started to actively promote HIV testing.
In the period 1999–2005, HIV incidence for the STI clinic population was lowest in the age group below 30 years of age, whereas no statistically significant age effect was observed for ACS and ROHOCO. However, in earlier time periods, MSM aged less than 30 years had the highest HIV incidence, as shown by the STI clinic and ACS population (P for interaction of age categories and time period, < 0.05). In 1999–2005, incidence was comparable in Dutch and non-Dutch MSM, although in the ACS, the risk of having Dutch nationality changed over time (P for interaction of nationality and time period, < 0.10). Incidence was highest among those with a concurrent STI, namely infectious syphilis or gonorrhea. Of note, in the ACS, the effect of having an STI co-infection on the HIV incidence is currently less marked than in 1984–1998 (P for interaction of STI and time period, 0.02).
The present study has demonstrated a continuing HIV epidemic in MSM in The Netherlands, being between one and four infections per 100 PY in the period 1999 to 2005. We confirm an increase in HIV incidence among older MSM who visit the Amsterdam STI clinic. The study employed three major sources of incidence data in the two largest cities in The Netherlands The ROHOCO, which had a small sample size, demonstrated a marginally significant age pattern in incidence trends, similar to that found at the Amsterdam STI clinic. Notably, in the ACS, due to changing entry criteria based on age, the follow-up of older (≥ 40 years of age) HIV-negative individuals is limited. This has resulted in reduced power to detect differences in incidence between older and younger MSM in recent years in the ACS. A striking finding of our study is that, whereas the young were at highest risk for HIV infection as the epidemic started, they currently do not comprise the group at highest risk, corroborating findings from the UK .
Remarkably, HIV incidence rates in the two cohort studies were similar, but both were lower than those found at the Amsterdam STI clinic. This may be explained by the fact that men who come for an STI check are likely to represent a higher risk group than participants in cohort studies who are willing to engage in long-term follow-up. Furthermore, it is known that the use of the LS methodology can cause persons with long-term infection and persons treated with HAART to be falsely classified as recently infected. Moreover, the LS-assay may not be completely valid for non-B HIV subtypes [11–20]. These considerations on the LS-assay are unlikely to have seriously biased the results however, as all known HIV seropositives (including those who reported or had serological evidence for HAART ) were excluded and as almost exclusively HIV subtype B is circulating among recently infected MSM in Amsterdam .
As mentioned earlier, international trends on HIV incidence are inconsistent [4–9]. Many factors may explain why the observed HIV incidence increased in one geographical area but not another while seemingly sharing facilitating cofactors such as surges in unsafe anal sex and ongoing STI epidemics. Here we speculate that the HIV testing rate in a country greatly influences differences in reported HIV incidence trends. Using mathematical modeling, we previously demonstrated that a higher testing rate can reduce the observed HIV incidence . In The Netherlands, testing has long been discouraged and was only actively promoted after HAART was introduced in 1996. The Netherlands have one of the lowest testing rates in the industrialized world, although the situation is changing. For example, the proportion ever tested increased from 41% in 2000 to 54% in 2005 among MSM who participated in a national health monitor [17,27]. In Amsterdam, this rate among MSM is considerably higher ranging from 71 to 78% in 2003 [28–30]. Yet, based on STI clinic data in Amsterdam we estimate that at least 25% of men with recent or even longstanding HIV infection are still unaware of their infection at the moment of their STI clinic visit. In addition, of men found recently infected in the anonymous HIV surveys at the Amsterdam STI clinic, only 40% were willing to accept named testing. Such a large proportion of MSM who are unaware of their HIV infection may prevent a decrease in HIV incidence among MSM, or even facilitate an increase-for two main reasons. First unaware HIV-infected persons will not use treatment that can lower viral load and decrease infectiousness . Among MSM in The Netherlands who know they are infected, current care-seeking behavior and the proportion treated with HAART is relatively high (approximately 80%), thereby limiting potential spread. Second, we currently find a relatively high HIV prevalence (10%, data not shown) and incidence (3.8 infections per 100 PY) in the never- tested group. Knowing ones HIV status enables the practice of safe sex and sexual harm reduction strategies based on ones own serostatus and the serostatus of the partner (serosorting). Effective serosorting may thus lead to a reduction in HIV transmission, although it may paradoxically facilitate STI spread . The two factors linked to testing, namely HAART use and sexual harm reduction practices, may prevent a rise in incidence in cities with high testing rates such as San Francisco and Sydney, where over 95% of MSM know their HIV status, but not in countries with lower testing rates, such as The Netherlands and, for example, in the UK [32–34].
Prevention programs should continue to target untested MSM. Furthermore, we previously showed that frequent testers; that is, MSM who tested for HIV at least twice in the past year, had low risky behavior. This group may thus comprise the worried well who regularly test irrespective of their behavior . The highest HIV incidence rates are found among men diagnosed with STI, but only a fraction of the recently HIV infected MSM accept named testing for HIV at their STI clinic visit.
Given the considerations mentioned above, just targeting to increase the testing rate may not be enough. MSM should be encouraged to test more efficiently, namely soon after having practiced high-risk sexual behavior or when suspecting STI acquisition. It is important to detect HIV infection as early as possible for the benefit of the individual as well as the population, since persons in their primary infection may contribute disproportionately to HIV spread . We recommend an opting-out HIV testing strategy, applied routinely to all MSM who are screened for STI. The opting-out approach to HIV testing, introduced in 2004 into sentinel screening among pregnant women in The Netherlands, has greatly increased the HIV testing rate in women, especially those at risk for HIV .
In conclusion, the HIV epidemic continues among MSM in The Netherlands. Among older, but not young, MSM who attend the Amsterdam STI clinic, HIV incidence has increased. Moreover, there is still a high rate of unawareness of HIV infection among MSM. Prevention should be developed specifically for older MSM. Further, more effective HIV testing behavior and approaches are needed such as including standard HIV testing of MSM when they are screened for STI.
The authors thank the personnel of the Amsterdam STI clinic, the personnel of the ACS, and the personnel of the ROHOCO for their contribution in data collection; the staff of the Public Health Laboratory of the Health Service, the staff of the Department of Retrovirology, Academic Medical Center, and the staff of the Departments of Virology and Medical Microbiology and Infectious Diseases of the Erasmus Medical Center for laboratory support; and Lucy Phillips for editing the final manuscript.
Sponsorship: The Amsterdam Cohort Studies on HIV infection and AIDS, a collaboration between the Amsterdam Health Service, the Academic Medical Center of the University of Amsterdam, Sanquin Blood Supply Foundation and the University Medical Center Utrecht, are part of the Netherlands HIV Monitoring Foundation and financially supported by the Netherlands National Institute for Public Health and the Environment. ROHOCO was sponsored by the Department of Dermatology and Venereology of the Erasmus Medical Center, Rotterdam.
1. Stolte IG, Dukers NHTM, de Wit JBF, Fennema JSH, Coutinho RA. Increase in sexually transmitted infections among men who have sex with men in Amsterdam in relation to HAART. Sex Transm Infect 2001; 77:184–186.
2. Centers for Disease Control and Prevention. Consultation on recent trends in STI and HIV morbidity and risk behaviors among MSM [Meeting report].
Atlanta: Centers for Disease Control and Prevention; 2000.
3. Macdonald N, Dougan S, McGarrigle CA, Baster K, Rice BD, Evans BG, Fenton KA. Recent trends in diagnoses of HIV and other sexually transmitted diseases in England and Wales. Sex Transm Infect 2004; 80:492–497.
4. Anon. San Francisco serosorting may explain odd HIV data. STIs have risen, but not new HIV infections.AIDS Alert
5. National Centre in HIV Epidemiology and Clinical Research. HIV/AIDS, viral hepatitis and sexually transmissible infections in Australia. Annual Surveillance Report 2004: National Centre in HIV Epidemiology and Clinical Research, Sydney, NSW
. Canberra: Australian Institute of Health and Welfare; 2005. http://web.med.unsw.edu.au/nchecr/Downloads/05_ansurvrp.pdf
Accessed: 8 November 2006.
6. Murphy G, Charlett A, Jordan LF, Osner N, Gill ON, Parry JV. HIV incidence appears constant in men who have sex with men despite widespread use of effective antiretroviral therapy. AIDS 2004; 18:265–272.
7. Dukers NH, Spaargaren J, Geskus RB, Beijnen J, Coutinho RA, Fennema HS. HIV incidence on the increase among men who have sex with men attending an Amsterdam sexually transmitted disease clinic: using a novel approach for detecting recent infections.AIDS
:F19–24. Erratum in: AIDS
8. Giuliani M, Di Carlo A, Palamara G, Dorrucci M, Latini A, Prignano G, et al
. Increased HIV incidence among men who have sex with men in Rome. AIDS 2005; 19:1429–1431.
9. Calzavara L, Burchell AN, Major C, Remis RS, Corey P, Myers T, et al
. Increases in HIV incidence among men who have sex men with undergoing repeat diagnostic HIV testing in Ontario, Canada. AIDS 2002; 16:1655–1661.
10. Ghys PD, Kufa E, George MV. Measuring trends in prevalence and incidence of HIV infection in countries with generalised epidemics. Sex Transm Infect 2006; 82(suppl 1):i52–i56.
11. Janssen RS, Satten GA, Stramer SL, Rawal BD, O'Brien TR, Weiblen MS, et al
. New testing strategy to detect early HIV infection for use in incidence estimates and for clinical and prevention purposes. JAMA 1998; 280:42–48.
13. Van der Snoek EM, De Wit JBF, Götz HM, Mulder PGH, Van't Hof AJ, Verkooyen RP, Van der Meijden WI. Demographics, sexual behaviour and STD/HIV prevalence in two groups of men who have sex with men, in Rotterdam, The Netherlands. Acta Derm Venereol 2004; 84:145–150.
14. Van der Snoek EM, De Wit JBF, Mulder PGH, Van der Meijden WI. Incidence of STDs and HIV infection related to perceived HIV/AIDS threat since HAART and PEP availability in men who have sex with men: Dutch MSM-cohort study. Sex Transm Dis 2005; 32:170–175.
15. van der Bij AK, Stolte IG, Coutinho RA, Dukers NH. Increase of sexually transmitted infections, but not HIV, among young men who have sex with men in Amsterdam: are STIs still reliable markers for HIV transmission? Sex Transm Infect 2005; 81:34–37.
16. Fennema JSA, van Ameijden EJC, Coutinho RA, van Doornum GJJ, Cairo I, van den Hoek JAR. HIV surveillance among sexually transmitted disease clinic attenders in Amsterdam, 1991-1996. AIDS 1998; 12:931–938.
18. van Griensven GJ, Tielman RA, Goudsmit J, van der Noordaa J, de Wolf F, de Vroome EM, Coutinho RA. Risk factors and prevalence of HIV antibodies in homosexual men in the Netherlands. Am J Epidemiol 1987; 125:1048–1057.
19. Van der Snoek EM. STD and HIV infections among men who have sex with men. Rotterdam Cohort Study [in Dutch]. Thesis 2004. Rotterdam: Erasmus MC.
20. Centers for Disease Control HIV/AIDS Information sheet. Using the BED HIV-1 capture EIA Assay to estimate incidence using STARHS in the context of surveillance in the US
. Atlanta: Centers for Disease Control; 2006.
21. Centers for Disease Control. Guide to calculating HIV incidence using STARHS results from Organon Teknika Corporation (OTC) Vironostika Less Sensitive HIV EIA
. Atlanta: Centers for Disease Control; 2001.
23. R Development Core Team. R: A language and environment for statistical computing
. Vienna, Austria: R Foundation for Statistical Computing; 2005. URL http://www.R-project.org
Accessed: 8 November 2006.
24. Murphy G, Charlett A, Jordan LF, Osner N, Gill ON, Parry JV. HIV incidence appears constant in men who have sex with men despite widespread use of effective antiretroviral therapy. AIDS 2004; 18:265–272.
25. Bezemer D, Jurriaans S, Prins M, van der Hoek L, Prins JM, de Wolf F, et al
. Declining trend in transmission of drug-resistant HIV-1 in Amsterdam. Municipal Health Service Amsterdam, The Netherlands. AIDS 2004; 18:1571–1577.
26. Xiridou M, Geskus R, De Wit J, Coutinho R, Kretzschmar M. The contribution of steady and casual partnerships to the incidence of HIV infection among men who have sex with men in Amsterdam. AIDS 2003; 17:1029–1038.
29. Stolte IG, de Wit JBF, Kolader M, Fennema JSA, Coutinho RA, Dukers NHTM. Low HIV-testing rates among younger high-risk men who have sex with men in Amsterdam.
In: Thesis. Stolte, Amsterdam, 2005.
30. Dukers NHTM, Raymond HF, Page-Shafer K, Stolte IG, Davidovich U, Fennema HAS et al.
, HIV testing rates among men who have sex with men in San Francisco and Amsterdam: implications for HIV spread [abstract TO-005]. Amsterdam, The Netherlands, July 2005.
31. Porco TC, Martin JN, Page-Shafer KA, Cheng A, Charlebois E, Grant RM, Osmond DH. Decline in HIV infectivity following the introduction of highly active antiretroviral therapy. AIDS 2004; 18:81–88.
32. Jin FY, Prestage G, Law MG, Kippax S, Van de Ven P, Rawsthorne P, et al
. Predictors of recent HIV testing in homosexual men in Australia. HIV Med 2002; 3:271–276.
33. Chen SY, Gibson S, Weide D, McFarland W. Unprotected anal intercourse between potentially HIV-serodiscordant men who have sex with men. San Francisco J Acquir Immune Defic Syndr 2003; 33:166–170.
34. McGarrigle CA, Mercer CH, Fenton KA, Copas AJ, Wellings K, Erens B, Johnson AM. Investigating the relationship between HIV testing and risk behaviour in Britain: National Survey of Sexual Attitudes and Lifestyles 2000. AIDS 2005; 19:77–84.
35. Xiridou M, Geskus R, de Wit J, Coutinho R, Kretzschmar M. Primary HIV infection as source of HIV transmission within steady and casual partnerships among men who have sex with men. AIDS 2004; 18:1311–1320.
36. Mulder-Folkerts DK, van den Hoek JA, van der Bij AK, Boer K, Schutte MF, Coutinho RA. Less refusal to participate in HIV screening among pregnant women in the Amsterdam region since the introduction of standard HIV screening using the opting-out method.Ned Tijdschr Geneeskd
:2035–2037 [Article in Dutch].