Recently the world has seen a growing number of "transformative
programs" engaging men in eradicating gender-based violence and reducing the spread of HIV/AIDS. The global Men as Partners Program (MAP) implemented by EngenderHealth in 20 countries since 1998, primariliy, targets various groups of men through ecological model to transform male gender norms. The purpose of MAP is twofold: 1) to challenge t...
Recently the world has seen a growing number of "transformative
programs" engaging men in eradicating gender-based violence and reducing the spread of HIV/AIDS. The global Men as Partners Program (MAP) implemented by EngenderHealth in 20 countries since 1998, primariliy, targets various groups of men through ecological model to transform male gender norms. The purpose of MAP is twofold: 1) to challenge the
attitudes, values, and behaviors of men that compromise their own health and safety as well as the health and safety of women and children and 2)to encourage men to become actively involved in preventing Gender Based
Violence as well as HIV/AIDS-related prevention, care and support activities.
Results from various pre-workshop and post-workshop evaluations have indicated that men's participation in the MAP workshops have resulted in some attitudinal changes in South Africa. More scientific evaluations in 1). Nepal resulted in increased utilization of Family planning services by men and women and increased utilization of STI services by men. 2.)In Swaziland and Botswana, promoting positive male behavior norms through participatory theatre, resulted in male participants/audience members expressing to be challenged by messages. More than 60% of the audience
believed the elimination of gender-based violence is both critical and urgent.
The need lies with demonstrating the linkages between these programs that target men with the women focused programs for the common cause.
In his first paragraph Gersovitz mis-represents the aim of our paper¹ by suggesting that we set out to “..investigate adoption of the ABC approach..”. In fact, the focus of that paper is on how one should best measure and summarise age at first sex, in the context of the intense interest that HIV prevention efforts have generated. We discussed the relative merits of the different measures and concluded th...
In his first paragraph Gersovitz mis-represents the aim of our paper¹ by suggesting that we set out to “..investigate adoption of the ABC approach..”. In fact, the focus of that paper is on how one should best measure and summarise age at first sex, in the context of the intense interest that HIV prevention efforts have generated. We discussed the relative merits of the different measures and concluded that, for the most informative and unbiased measures, one should use survival analysis to describe age at first sex; a conclusion that Gersovitz dismisses with “A proportion is easier to calculate and sidesteps knowing whether the median is within the age range of the data.”
A life table (survival) approach is not limited to the calculation of medians, it leaves open the choice of summary statistic(s) to be used (we present inter-quartile range as well as medians), and certainly does not preclude the reporting of proportions who have become sexually active by a given age. What it does ensure is that if the analyst needs to combine reports from respondents of different ages this is done with due allowance for censoring. Simpler analytical approaches have their uses, such as in cross-sectional surveys limited to very young respondents or for the accessible presentation of results to non-specialist audiences. However, if we have access to high quality, rich data sources such as the DHS, where we can make comparisons between age groups, cohorts and time periods, and challenge the data to detect reporting errors or selection biases it behoves the research scientist to go beyond the calculation of the simple indicators such as those used for Millennium Development Goals.
The proportion who have had sex by a particular age correlates with the median age at first sex, and other summary measures of this transition, but describes only one aspect. This is illustrated by recent data on age at first sex for women in Kenya and Tanzania (Figure 1). In both countries about 30% of women report first sex before age 16 but in Kenya the median is, at 17.7 years, 6 months later than in Tanzania. Three-quarters of women in Tanzania have had sex by 18.9 years of age, a point reached almost a year later in Kenya.
Figure 1: Cumulative proportion who had had sex by age, women aged 15-49.
In one respect we agree with Gersovitz's standpoint: age at first sex is of critical importance. For effective HIV prevention we must know if age at first sex has changed and whether this has affected HIV transmission. It has become a hot political topic and if science is not to be hijacked by ideology we must honestly and accurately measure and describe what has taken place in Uganda and elsewhere over the last 30 years. The value and feasibility of abstinence, partner reduction and condom use, and the relative importance of each of these in different contexts defies simple packaging and is not well served by promulgation of a simple one size fits all approach2-4.
In our paper we deliberately set out to search for reporting errors and/or selection biases, and we found plenty of evidence of these in many of the series of national surveys that we examined. In the case of Uganda we actually used data from three surveys, going back to the 1988/1989 survey for women. Looking at the experience of different birth cohorts of women (those born in 1971-75, 1976-80, 1981-85) captured at different interview ages in the three surveys (table 4 in our paper) we found no consistent evidence of misreporting by the oldest cohort, for which median age at first sex was calculated as 16.3, 16.0 and 16.4 as they aged from 15-19, to 20-24 and finally 25-29; the middle cohort reported the same median age (16.3) at 15-19 and 20-24 – no different to the older cohort, whereas the youngest cohort, which could only be observed at the last survey when they were aged 15-19 had an older median age at first sex of 17.1, which cannot be assessed for age-related reporting bias until they are questioned again in another survey of this sort. For men we observed a small increase in the reported median age at first sex as the cohorts aged: from 17.4 to 17.9 for the oldest cohort as it moved from 20-24 to 25-29; and from 17.8 to 18.2 for the middle cohort as it aged from 15-19 to 20-24 – these inconsistencies are of the same order of magnitude as the apparent increase in age at first sex between the cohorts, which apparently continues to the youngest cohort which reported a median age at first sex of 18.4 when seen at survey aged 15-19 – again there is no way of checking the consistency of the reports of these youngest male respondents.
Interestingly, when Gersovitz focuses on the group that we refer to as the “middle” birth cohort, he concludes: “Clear significant positive bias is observed only for men for this 5-year group” which does not contradict our conclusion (no bias for women, positive bias for men which is not statistically significant) given that his comparison was confined to transitions to sexual activity occurring before age 16, whereas we considered 50% of all the transitions that could be observed for the cohort.
We allowed for survey design in our analysis of DHS data, and discussed problems that may arise in comparisons between successive sample surveys of the same population which do not use the same panel of respondents. Unlike Gersovitz, we make an attempt to quantify how changes in composition of the sample (by residence and education) and the possible association between age at interview and reported age at sexual debut affect our estimates of trend (table 5 in our paper). In some populations apparent trends disappear when structural factors and reporting biases are allowed for (males in Zambia and Kenya, females in Tanzania and Zimbabwe); in others observed significant differences persist (females in Ghana, Kenya and Uganda) and in a few populations trends that seemed trivial attain statistical significance when these structural and reporting factors are allowed for (males in Tanzania and Uganda).
We do not make any claims, in the case of Uganda, that these findings indicate successes for abstinence campaigns. Indeed, in the case of Ugandan males the significant change appears to be between the middle and eldest cohorts. The majority of the eldest cohort made their sexual debut between 1988 and 1992, and the majority of the middle cohort made their sexual debut between 1994 and 1999; in other words our analysis implies that the biggest changes took place in the early 90s before the “ABC” campaigns really got under way. If we were invited to speculate on the cause of this apparent rise we might put it down to a long-term secular rise in age at marriage, which in Uganda appears to be more closely associated with sexual debut than in other African populations.
Since we undertook similar analyses for Ghana, Kenya, Tanzania, Zambia and Zimbabwe we were able to observe that a rise in the reported age of sexual debut with age at interview (the bias identified by Gersovitz) was quite a common finding for men, whereas the opposite was generally true for women (women from the same birth cohort generally reported earlier sexual debut in later surveys). Although Ugandan women do not follow this pattern, it is worth bearing in mind that “social desirability” biases that may induce misreporting of age at sexual debut in response to abstinence campaigns are not the only source of reporting error. Total denial of sexual activity by young women interviewed at an age when they are reluctant to discuss sexual behaviour, but who are more relaxed about admitting their age at sexual debut when interviewed at a later age, generate reporting inconsistencies of this type, which can also be detected by cohort analyses.
One difference between our findings and those of Gersovitz is his detection of a bias in the women’s reports which we did not identify. This difference arises because our analysis was restricted to respondents born from 1971 onwards. Extending our analysis to the older age groups included by Gersovitz does show evidence of a change in the age at first sex reported in the most recent of the three surveys (Table 1). Reporting bias alone does not seem the most likely explanation for this discrepancy. The differences become apparent when respondents are in their late 30s or older. These cohorts first reported at an age where one would expect them to have already overcome any youthful embarrassment about their early sexual behaviour. In the later survey, when a change is apparent, the respondents are at the age when HIV associated mortality would have taken effect for those who were infected at a young age: very likely those who first had sex at a young age. In other words, mortality selection effects (rather than social desirability bias) may be responsible for much of the apparent difference in reported age at first sex for older women - this should not affect reports by younger women.
Median AFS (approx age at survey)
Pooled
Surveyed in:
1988
1995
2000/2001
Women born in:
1981-1985
17.1 (<20)
17.1
1976-1980
16.3 (<20)
16.3 (˜20-24)
16.3
1971-1975
16.3 (<18)
16.0 (˜20-24)
16.4 (˜25-29)
16.2
1966-1970
15.9 (˜18-22)
15.8 (˜25-29)
16.0 (˜30-34)
15.9
1961-1965
15.5 (˜23-27)
15.6 (˜30-34)
16.1 (˜35-39)
15.7
1956-1960
15.3 (˜28-32)
15.3 (˜35-39)
15.9 (˜40-44)
15.4
1951-1955
15.4 (˜33-37)
15.4 (˜40-44)
16.2 (˜45-49)
15.6
Table 1: Median age at first sex by birth cohort and survey.
There are some trivial reasons for minor differences between our results and Gersovitz's. We used true birth cohorts in our analysis instead of estimated age at survey and have therefore more accurately identified common cohorts of respondents. DHS datasets contain two variables describing age at first sex: one as reported by respondents and an imputed version that has been edited for consistency with answers to other questions. The 2000/2001 data for men were heavily edited using criteria that cannot be applied to the 1995 survey in which age at first birth was not collected. Gersovitz used the imputed variable as provided by Macro and excluded from the analysis 13% of the male respondents from 2000/2001. 7% of these respondents appear to have been wrongly classified as inconsistent. Exclusion of these men from the 2000/2001 analysis results in a slight overestimate of the proportion who had sex by age 16 and makes the comparison with the earlier data set problematic. This does not undermine Gersovitz's conclusions but makes it difficult to compare his results with ours.
In short: we cannot understand why Gersovitz repeatedly asserts that his conclusions are opposite to ours. Where the focus of our study and his are similar the results are not contradictory; however our analytical scope is wider in the sense that we examine several countries, we look at the first 50% of transitions in the cohorts of interest (as opposed to transitions occurring before age 16) and consider not just the transition from virginity to sexual activity, but also the transition to first marriage.
References
1. Zaba, B., Pisani, E., Slaymaker, E. & Boerma, J. T. Age at first sex: understanding recent trends in African demographic surveys. STI80, ii28-ii35 (2004).
2. Barnett, T. & Parkhurst, J. HIV/AIDS: sex, abstinence, and behaviour change. Lancet Infectious Diseases 5, 590-593 (2005).
3. Boerma, J. & Weir, S. Integrating demographic and epidemiologic approaches to research on HIV/AIDS: the proximate determinants framework. Journal of Infectious Diseases 191, S61-S67 (2005).
4. Wellings, K. et al. Sexual behaviour in context: a global perspective. The Lancet 368, 1706-1728 (2006).
In their recent review of gonorrhoea testing published in STI, Bignell et al state that whether or not to perform anal screening for gonorrhoea in men who have sex with men (MSM) should be guided by a sexual history [1]. Both UK [2] and US [3,4] guidelines recommend screening for anal gonorrhoea and chlamydia among men who have sex with men (MSM). US guidelines recommend anal screening only for MSM who re...
In their recent review of gonorrhoea testing published in STI, Bignell et al state that whether or not to perform anal screening for gonorrhoea in men who have sex with men (MSM) should be guided by a sexual history [1]. Both UK [2] and US [3,4] guidelines recommend screening for anal gonorrhoea and chlamydia among men who have sex with men (MSM). US guidelines recommend anal screening only for MSM who report receptive anal sex, while UK guidelines suggest screening based on sexual history, noting that oro-anal transmission of gonorrhoea may occur without penetrative
anal intercourse [2]. However, we believe that current screening guidelines do not acknowledge the possible contribution of a wider variety of receptive anal sexual practices to anal infections among this population.
Our recently published study in Sexually Transmitted Infections [5] examined risk factors for anal gonorrhoea and chlamydia among MSM participants in the community-based Health in Men (HIM) cohort in Sydney, Australia. Among participants with casual male partners who reported no
unprotected receptive anal intercourse (UAI), we found a strong independent association between anal gonorrhoea and receptive oro-anal sex, (p for trend 0.001) which has also been described among MSM in cross sectional studies in clinical settings [6]. In addition, we also found a
strong independent association of anal gonorrhoea with receptive fisting (p for trend 0.001) and receptive anal fingering (p for trend 0.002). Receptive fingering (p for trend 0.038) and being receptive of dildos (p
for trend 0.029) were also independently associated with anal chlamydia infection.
Not only do these anal practices predict anal infection with chlamydia and gonorrhoea, but they are very common among MSM. In the HIM study, among those with regular and casual partners at baseline, most participants reported receptive rimming (66.2% and 69.1% respectively) and
receptive fingering (71% and 66.8% respectively) over the previous 6-month period. Being receptive of a dildo was reported by 27.1% of those with regular partners, and 13.0% of those with casual partners. Receptive fisting was less commonly reported.
The common and diverse nature of receptive anal sexual practices among MSM, and their association with anal gonorrhoea and chlamydia infections, suggests that anal screening for these infections should be performed in all sexually active MSM, not just those reporting receptive
penile- or oro-anal sexual practices. Offering anal STI screening uniformly to all MSM, may also overcome difficulties of risk assessment, given varying depths of sexual history taking skills among clinicians. On
the basis of these data, Australian screening guidelines for MSM [7] are currently being updated to recommend universal anal screening for all sexually active MSM irrespective of their reported sexual practices (Personal communication, Dr Christopher Bourne, 11th May 2007).
David J Templeton 1, 2
Fengyi Jin 1
Basil Donovan 1, 3
Andrew E Grulich 1
1. National Centre in HIV Epidemiology and Clinical Research, The University of New South Wales, Sydney, Australia
2. Sexual Health Service, Sydney South West Area Health Service, Sydney, Australia
3. Sydney Sexual Health Centre, Sydney, Australia
Competing interests: none
References
[1] Bignell C, Ison CA, Jungmann E. Gonorrhoea. Sex Transm Infect 2006;82(Suppl 4):iv6-9.
[2] UK national screening and testing guidelines for sexually transmitted infections, 2006 [accessed 10th May 2007]; Available from:
http://www.bashh.org/guidelines/2006/sti_screening_guidelines_v14_0806.pdf
[3] Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines, 2006. MMWR Recomm Rep 2006;55(RR-11):1-94.
[4] Public Health Seattle and King County Sexually Transmitted Diseases Program. STD/HIV Screening Guidelines for men who have sex with men, 2005
[accessed 10th May 2007]; Available from:
http://www.metrokc.gov/health/apu/std/msmstd.htm
[5] Jin F, Prestage GP, Mao L, Kippax SC, Pell CM, Donovan B, et al. Incidence and risk factors for urethral and anal gonorrhoea and chlamydia
in a cohort of HIV-negative homosexual men: the Health in Men Study. Sex Transm Infect 2007;83(2):113-9.
[6] McMillan A, Young H, Moyes A. Rectal gonorrhoea in homosexual men:
source of infection. Int J STD AIDS 2000;11(5):284-7.
[7] Sexually transmitted infections in gay men action group. Sexually transmitted infection testing guidelines for men who have sex with men, 2005 [accessed 10th May 2007]; Available from:
http://www.racp.edu.au/public/SH_MSMguidelines.pdf
We would like to correct the inaccurate comments made in the commentary paper “The National Chlamydia Screening Programme and the NICE guidance on one-to-one interventions: remember the under 25s” from Simms et al, in relation to the National Institute for Health and Clinical Excellence (NICE) guidance ‘Prevention of sexually transmitted infections
and under 18 conceptions’ (Sexually Transmitted Infect...
We would like to correct the inaccurate comments made in the commentary paper “The National Chlamydia Screening Programme and the NICE guidance on one-to-one interventions: remember the under 25s” from Simms et al, in relation to the National Institute for Health and Clinical Excellence (NICE) guidance ‘Prevention of sexually transmitted infections
and under 18 conceptions’ (Sexually Transmitted Infections 2007;83:171, http://sti.bmj.com/cgi/content/full/83/2/171).
The paper states that the draft guidance misses out under 25 year olds as reflected in the paper’s title and in further references within the text in relation to chlamydia screening. This view is not correct – the draft and final NICE guidance identified that those who have ‘early
onset of sexually activity’ and who have ‘unprotected sex and frequent change of and/or multiple sexual partners are at risk’ - under 25 year olds fall into these two categories. These are key risk behaviours for
chlamydia infection. The guidance recommendations for vulnerable young people under 18 also relate to this group.
A second comment “the guidelines appeared to focus exclusively on consultations in which patients are seeking care for an issue related to sexual health” does not represent the guidance. The guidance actually
states that risk assessment can be 'during routine care and or travel immunisation’. The final NICE guidance also states that the recommendations should be implemented alongside the National Chlamydia Screening Programme.
The full NICE guidance ‘‘Prevention of sexually transmitted infections and under 18 conceptions’ is available at http://guidance.nice.org.uk/phi003 .
Yours faithfully,
Dr Catherine Law
Chair
Public Health Interventions Advisory Committee at NICE
Professor Mike Kelly
Director
Centre for Public Health Excellence, NICE
NICE
MidCity Place
71 High Holborn
London
WC1V 6NA
Drs Shamanesh and Radcliffe on the one hand and Dr Horner on the other provide finely balanced arguments for and against screening of asymptomatic men for non-specific urethritis (NSU). Drs Shamanesh and Radcliffe, in addition, slip in their recommendation to abandon examining asymptomatic men and this needs to be challenged. Dr Ross suggests that individual departments may chose whether to continue with...
Drs Shamanesh and Radcliffe on the one hand and Dr Horner on the other provide finely balanced arguments for and against screening of asymptomatic men for non-specific urethritis (NSU). Drs Shamanesh and Radcliffe, in addition, slip in their recommendation to abandon examining asymptomatic men and this needs to be challenged. Dr Ross suggests that individual departments may chose whether to continue with the urethral smear (US) but that may result in some difficulties and it is the pragmatic implications of the change in practice which I would like to address.
It is hard to understand the logic that suggests that a finding in an asymptomatic patient should be interpreted differently in a symptomatic patient. If a man has a discharge and his tests for N gonorrhoeae and C
trachomatis are negative but he has a urethritis, as evidenced by a significant number of pus cells on a US, we would call this NSU. We might believe that M genitalium could be a cause, but clearly according to current data, that infection cannot account for all symptomatic patients.
We would treat him, and in the past would have treated his partner. Our rationale for doing this is to remove his symptoms and to prevent re-infection from his partner and possible harm to her. If he is minimally symptomatic, presumably the same argument applies. At what point does the
argument no longer apply? Can we assume that truly asymptomatic men who do not have a visible urethral discharge but have urethritis on a US do not have M genitalium or any other pathogen? Clearly that is not the case, although the likelihood is relatively small. The pathogenic potential of M genitalium is becoming increasingly apparent and it seems wrong to abandon
the only routine test, albeit a surrogate marker, we have. Thinking back to the early to mid 1980s, we might have made a similar decision about C trachomatis then when we had relatively little evidence of the potential
pathogenicity of C trachomatis. The widespread availability of testing for M genitalium would make a huge difference in progressing this argument but we seem to be someway away from that at the moment.
However the pragmatic problem is how we deal with patients who have had a urethral smear in the past and have been told they had NSU. Are we now to tell them that we no longer believe in that diagnosis, that their
treatment and partner assessment was unnecessary (unless of course they had symptoms, in which case we still believe in the diagnosis). What new evidence can we point to, to enable us to reject our previously held views or are we just seeking to abandon a difficult diagnosis to push people
through our clinics faster and meet DH targets of doubtful value?
If the US is abandoned, clinicians have the option of diagnosing gonorrhoea by taking a urethral culture or using a NAAT urine test. If we continue to test for male gonorrhoea using urethral culture, an examination will be necessary and this will therefore make little
difference to the clinic throughput. Using the “pee and go” approach, men will not be examined. Dr Horner points out that it may be wrong to assume that men who say there are asymptomatic, do not have a urethral discharge. We are all aware of examining men with obvious discharges who are unaware that a discharge is abnormal. How should we manage such men- clearly we should not risk missing a diagnosis of gonorrhoea, so a US is necessary. If we find only NSU, do we ignore the findings?
Furthermore, some men, particularly young men, may be reluctant to express their concerns about particular issues during an interview or in completing a questionnaire and want the opportunity to explore these at the time of examination. Sometimes they don’t have the words to explain
what they perceive might be a problem. While it may not be important in public health terms to reassure a young man that his pearly penile papules are normal or that the dimensions of his penis are perfectly adequate, a
professional opinion may alleviate months of anxiety. Conversely, men with significant disease such as lichen sclerosus, often claim to be asymptomatic. Genital warts may be little more than a cosmetic problem, but the virus is sexually transmitted and we surely have a responsibility
to make the diagnosis, if only to provide reassurance and advice about risk reduction. Finally the opportunity for health promotion, for example teaching testicular self examination, is lost. While “pee and go” may be
ideal for outreach sessions especially in areas of high N gonorrhoeae prevalence where the positive predictive value of the test is likely to be more acceptable, I believe an examination is an integral part of a level 3 specialist service.
Another issue which we may need to consider is the implications of some clinics continuing to test for NSU when other clinics have abandoned the practice as the following recent case illustrates. A young man and his partner presented for a screen before staring a new relationship. He had previously had a negative screen in another clinic a
few weeks after the end of his previous relationship and was only attending at the request of his girl friend who had asked that they should do the tests together. His urethral smear showed more than 20 pus cells. Tests for chlamydial infection and gonorrhoea were negative. We assumed
that the first clinic had used a “pee and go” screening procedure thus missing the diagnosis of asymptomatic NSU.While it could be argued that it would have been better if we had not done a urethral smear and the diagnosis not made, my concern is that inconsistency in practice will make
it very difficult to sustain the confidence of patients in such scenarios. I think we will all need to agree on what constitutes a sexual health screen.
Drs Shamanesh and Radcliffe dismiss the small delay likely to occur if asymptomatic chlamydia positive patients are not detected at their first visit. In most real life settings this is likely to be at least a week, allowing for specimen transport, test performance, result generation, result transmission to the clinician and then the patient and the return of the patient for treatment. During this time the patient may transmit the infection to others. It should be noted that we (and the Department of Health) believe that more than two days is too long to wait
for a GUM appointment in case the patient transmit to others while he is waiting for the appointment. Is there a little hypocrisy here? It should also be noted that the second visit will generate another PBR fee, making the service not only more costly but less cost effective per infection
diagnosis made. The US is our oldest near patient test (apart from the 2 glass test) and at a time when others are seeking to develop near patient testing in order to incorporate treatment into a single visit (one stop
shop) it seems strange that we should be moving away from the concept.
Finally, are we producing “urethral cripples” by diagnosing NGU? A lot depends on what we say to our patients. Clearly in the first instance we do not know whether the patient has chlamydial infection or possibly
gonorrhoea and this needs to be stated. But we could move away from the practice of stating that NGU is almost always sexually transmitted and advise men that if their tests for GC and NG are negative they have a urethritis which may or may not be sexually transmitted. We would then
advise a screen for the partner (which would pick up those cases of NG and CT missed in the index case) and offer treatment, explaining that we are unable to test for M genitalium. This of course, takes a bit more time
than “pee and go”, but seems to me a better way forward than abandoning a diagnosis without firm evidence to do so.
In her recent editorial, Judith Stephenson highlighted the lack of trial evidence supporting the opportunistic approach used by the National Chlamydia Screening Programme (NCSP) and calls for a randomised control trial (RCT) to compare opportunistic versus a register based call/recall
approach to screening1. What this editorial and the National Institute for Health and Clinical Evidence (NICE...
In her recent editorial, Judith Stephenson highlighted the lack of trial evidence supporting the opportunistic approach used by the National Chlamydia Screening Programme (NCSP) and calls for a randomised control trial (RCT) to compare opportunistic versus a register based call/recall
approach to screening1. What this editorial and the National Institute for Health and Clinical Evidence (NICE) commissioned rapid review, of which she is a co-author, inadequately highlight and therefore underplay, are the practical and ethical difficulties, time requirement and expense of undertaking such a trial2. Senok et al. undertook a pilot exercise addressing this and concluded that a definitive trial would be resource intensive and need to be of a substantial size3.
The NCSP uses a multifaceted opportunistic approach to screening which is intended to make a sustainable impact within the local health economy and is not restricted to health care settings. Those RCTs that have been undertaken in relation to chlamydia screening measured the success of achieving a single screen and were restricted to specific
settings, such as general practice, or a single method, such as postal invitation. This underestimates the potential benefit of the broader approach employed by the NCSP. The constraints of a RCT design would also restrict the ability of the programme to be a continuous public health
intervention incorporating new technology to promote accessibility and acceptability to young persons, such as screening requests via the Internet, and texting results.
We agree with Stephenson that good quality research into the natural history of chlamydial infection and related complications is to be encouraged. The evolution of the NCSP research and development agenda and associated performance indicators is well underway. This is integral to the development of the programme, as are more focussed investigations including RCTs in addressing ways of improving programme effectiveness and modernising sexual health services for young people.
We do not support the pursuit of the holy grail of RCT evidence in relation to the specific question of opportunistic versus register call/recall. Further debate here takes the focus away from addressing the
real problem in hand, that of the rising number of young persons with chlamydial infection and the associated risk of serious long term morbidity and potential infertility.
References
1 Stephenson J. NICE guidance and the National Chlamydia Screening Programme. Sex Transm Infect 2007;83:170.
2 National Institute for Health and Clinical Excellence. Revised rapid review of evidence for the effectiveness of screening for genital chlamydia infection in sexually active young women and men.
http://nice.org.uk/page.aspx?o=371768. Accessed 19 April 2007.
3 Senok A, Wilson P, Reid M, Scoular A, Craig N, McConnachie A, et al.. Can we evaluate population screening strategies in UK general practice? A pilot randomized controlled trial comparing postal and opportunistic screening for genital chlamydial infection. J Epidemiol
Community Health 2005;59:198-204.
In their manuscript, Utility of the Determine TP Rapid Syphilis Test in Commercial Sex Venues in Peru, Campos et al. report poor validity and utility of a point-of-care rapid syphilis assay in an outreach facility for commercial sex workers in urban Peru [1]. We have identified
potential limitations in their study design, which might have contributed to their conclusion that the TP Determine Rapid Syphilis Te...
In their manuscript, Utility of the Determine TP Rapid Syphilis Test in Commercial Sex Venues in Peru, Campos et al. report poor validity and utility of a point-of-care rapid syphilis assay in an outreach facility for commercial sex workers in urban Peru [1]. We have identified
potential limitations in their study design, which might have contributed to their conclusion that the TP Determine Rapid Syphilis Test (Abbott Laboratories, Tokyo, Japan) is not a suitable screening test.
One possible methodological shortcoming was the direct application of unheparanized whole blood onto the TP Determine testing device. In a combined laboratory and clinical study of the TP Determine test, we previously reported a poor sensitivity (11/17, 64.7%) with the rapid-test as compared to TP-PA using whole blood directly applied to the testing pad [2]. In response, we continued our study with the use of heparinized capillary tubes for collection of whole blood and deposition onto the testing pad. This alteration resulted in 100% sensitivity (52/52) of the TP Determine for the remaining specimens. The authors here report an almost identical sensitivity of 64% when comparing directly applied whole blood to TP Determine tests compared to rapid plasmin reagin (RPR) confirmed with treponema pallidum particle agglutination (TPHA). When the
authors repeated the use of rapid tests in laboratory using sera from the RPR and TPHA samples, they noted a 97% sensitivity of the rapid tests. They conclude that the assays do not perform well with whole blood
specimens. We suggest that this shortcoming is potentially limited to unheparainzed whole blood.
Furthermore, the poor performance of the Determine TP test reported in terms of positive predictive value (71.4%) is possibly due to use of sequential RPR and TPHA testing as a gold standard to evaluate the TP Determine Test. The authors do not specifically describe the analysis of
specimens that tested positive by TP Determine but negative by the RPR gold standard. We assume that these specimens were considered false positives and therefore contributed to the poor positive predictive value. There are at least two cases where specimens with a positive TP Determine
result might be expected to test negative by RPR: 1) in subjects with a history of treated syphilis because non-treponemal antibodies diminish following proper treatment and 2) in very early and late-stage syphilitic
infection when non-treponemal tests have poorer sensitivity [3]. A more appropriate gold standard might be the direct comparison of the TP Determine test, a treponemal antibody assay, to another treponemal antibody assay (e.g. TPHA or TP-PA alone), in the absence of a preliminary non-treponemal assay (RPR). To minimize false positives, we use the TP
Determine assays in San Francisco as a point-of-care preliminary test in outreach centers only on patients who deny a history of syphilis treatment.
Many previous studies have documented strong validity of rapid syphilis tests with the use of sera [4-6]. Because of the need for point-of-care testing for syphilis without complex laboratory methods, there is a corresponding need for analysis of these same assays with the use of
whole blood. Whether it be the TP Determine or another rapid test for the diagnosis of syphilis, a valid and simple test will be invaluable to control programs in stemming the worldwide morbidity associated with syphilitic infection.
References
1. Campos PE, Buffardi AL, Chiappe M, et al. Utility of the DetermineTM Syphilis TP Rapid Test in Commercial Sex Venues in Peru. Sex Transm Infect 2006
2. Siedner M, Zapitz V, Ishida M, De La Roca R and Klausner JD. Performance of rapid syphilis tests in venous and fingerstick whole blood specimens. Sex Transm Dis 2004;31:557-60
3. Young H. Syphilis: new diagnostic directions. Int J STD AIDS 1992;3:391-413
4. WHO Special Programme for Research and Training in Tropical Diseases (TDR) STD Diagnostics Initiative (SDI). Laboratory-based evaluation of rapid syphilis diagnostics. Diagnostics Evaluation Series 2003;1:1-31
5. Diaz T, Almeida M, Georg I, Maia S, Souza RD and Markowitz L. Evaluation of the Determine rapid syphilis TP assay using sera. Clin Diagn Lab Immunol 2004;11:98-101
6. Herring AJ, Ballard RC, Pope V, et al. A Multi-centre Evaluation of Nine Rapid Point of Care Syphilis Tests Using Archived Sera. Sex Transm Infect 2006
The article by Prost el al.[1] entitled ‘There is such a thing as asking for trouble’: Taking rapid HIV testing to gay venues is fraught with challenges” is an important addition to the growing literature focusing on HIV testing at entertainment venues catering to men who have sex with men (MSM). Their ethnographic approach to exploring the
complexity of such programs adds further dimension to this topic,...
The article by Prost el al.[1] entitled ‘There is such a thing as asking for trouble’: Taking rapid HIV testing to gay venues is fraught with challenges” is an important addition to the growing literature focusing on HIV testing at entertainment venues catering to men who have sex with men (MSM). Their ethnographic approach to exploring the
complexity of such programs adds further dimension to this topic, however their experience should not discourage others from integrating HIV (and other STI) screening in such venues.
In their article, Prost et al [1] identified confidentiality, the apparent disconnect between a serious medical issue and gay life, post-test support, and venue-specific concerns as potentially damaging to venue
based testing for MSM. Since February 2006, the NYU School of Medicine Center for AIDS Research has been conducting a study in New York City bathhouses to assess the feasibility of HIV testing in these venues; to date we performed 406 tests. Many of the points that the authors have made regarding implementation of such a program influenced our study design well before we ever performed our first test. Privacy, risk-reduction counseling, and clear paths connecting those found to be HIV positive to care were central to our design and the design of any successful HIV testing program at bathhouses, saunas, and other commercial
venues. Before we approached the management of the two remaining bathhouses in Manhattan, we studied models described by similar programs successfully functioning in Seattle and Los Angeles.[2]
All of these observations are valid issues facing non-traditional venue HIV rapid testing. It is the charge of the organization developing a venue-specific program to predict these issues rather than to respond to
them. Each issue can be dealt with to create a program that is acceptable and effective.
• Confidentiality
In creating our program, we followed many of the dictums laid out in a successful bathhouse testing program in Seattle.[2] To create an effective program, community ties and the relationship with venue management need to be developed. The testing program becomes a service that is provided by a healthcare or research group in close association with the staff and management of the venues. Maintaining a sustainable program requires working within the culture and environment of the venue, not against it. With the support of the venue management, we converted
unused space into discreet, soundproof, confidential testing rooms.
• HIV testing is “too serious” for an entertainment venue.
In our survey of the men attending one Manhattan bathhouse, nearly 80% of men responded that HIV testing was appropriate in the bathhouse setting. As demonstrated in one study, bathhouse testing may impact behaviors
associated with HIV risk.[3] Part of HIV stigma is the
compartmentalization of sexuality from health. We believe that HIV testing at the bathhouse in a model acceptable to venue staff and patrons makes sexual health an important reality for these MSM. Rather than too serious, our service is just serious enough. This is evidenced by the
fact that many men who choose not to be tested through our program approach testing staff to ask questions about safer sex and HIV exposure. Furthermore, the testing staff report overwhelmingly positive feedback from venue patrons, staff, and management.
• Post-test support/Venue Specific Concerns
No HIV testing program should exist without a good post-test plan. When we discussed our program with venue management, they expressed concern about connection to care and the risk for self-destructive behavior for men with preliminarily positive tests. Ninety-three percent of our
positives have been successfully connected to HIV primary care. Additionally, we have in place a clear and discreet plan to remove anyone from the bathhouse upset by a result. Despite a 5.1% seroprevalence rate among testers there have been no such “evacuations” necessary [4].
Additionally there is no data to support that people newly diagnosed with HIV are at increased risk of suicidal behavior.
The data presented in the study by Prost et al1 confirms many of the assumptions we made in designing a testing protocol in bathhouses in New York City. This valuable addition to the literature should be used to help guide and design such testing programs rather than be used as evidence for why they should not exist. A close understanding of the community of clients such programs will reach as well as a close working relationship with venue management is central in a lifestyle-competent
testing program. Connection to care and safety concerns need to be assumed and planned for in the design of any non-traditional rapid testing program.
The challenge of venue based testing programs is that each city and venue represents its’ own culture. What works in one venue, may not translate well to a different venue. Prost et al [1] identified several critical components in their evaluation of the acceptability of HIV
testing in commercial sex venues in the UK. However, for every venue based HIV testing program that is not accepted by its clients, there are ones that succeed. Our bathhouse screening program in New York has been not only well received by the clients its serves, but also by the larger
prevention community. We hope that readers of Prost et al see opportunities to improve and expand venue based testing programs, and not reasons to abandon them all together.
References
1. Prost A, Chopin M, McOwan A, et al. "There is such a thing as asking for trouble": Taking rapid HIV testing to gay venues is fraught with challenges. Sex Transm Infect 2007
2. Spielberg F, Branson BM, Goldbaum GM, Kurth A and Wood RW. Designing an HIV counseling and testing program for bathhouses: the Seattle experience with strategies to improve acceptability. J Homosex 2003;44:203-20
3. Huebner DM, Binson D, Woods WJ, Dilworth SE, Neilands TB and Grinstead O. Bathhouse-based voluntary counseling and testing is feasible and shows preliminary evidence of effectiveness. J Acquir Immune Defic Syndr
2006;43:239-46
4. Daskalakis DRSRHKB, PhD; Richard Hutt, RN; Dylan Stein; Timothy Neithercott; Ancil Brown; Fred Valentine, MD; Michael Marmor, PhD. Correlates of Incident HIV Diagnosis, Knowledge of Chemoprophylaxis, and
Burden of Acute and Recent Infection among Patrons of new York City Bathhouses, a Pilot Study. In: Conference on Retroviruses and Opportunistic Infections. Los Angeles, CA, 2007
As the editor of the recently published NAM Manual of HIV Prevention and a person living with HIV, I think serosorting is on balance a positive influence on the course of the HIV epidemic and, as Robert Reinhardt says, say, it’s ‘natural selection’. If I’m honest, if I was HIV
negative, I’m not sure if I’d want to have sex with someone with HIV.
But I don’t think serosorting is a good thing in itsel...
As the editor of the recently published NAM Manual of HIV Prevention and a person living with HIV, I think serosorting is on balance a positive influence on the course of the HIV epidemic and, as Robert Reinhardt says, say, it’s ‘natural selection’. If I’m honest, if I was HIV
negative, I’m not sure if I’d want to have sex with someone with HIV.
But I don’t think serosorting is a good thing in itself, and I agree that campaigning specifically for serosorting risks promoting 'viral apartheid'. You cannot legislate against human nature; HIV negative and positive people are going to continue to be attracted to each other and have sex and relationships and this should not become a socially shunned activity.
No, I think serosorting is a good thing because in order to serosort you have to do two things: a) Know your status b) Disclose it.
It’s not serosorting we should be encouraging, it’s the behaviours that make it possible.
Gus Cairns,
Board of Directors
UKC - UK Coalition of People Living with HIV and AIDS
The authors of the cited article on serosorting in San Francisco, including members of the city’s Public Health Department, have created more problems than they have solved due, in part, to the premature rollout in San Francisco of a campaign to actively “encourage” serosorting as a HIV prevention strategy. That campaign was initiated coincidentally when their article was published in STI by means of...
The authors of the cited article on serosorting in San Francisco, including members of the city’s Public Health Department, have created more problems than they have solved due, in part, to the premature rollout in San Francisco of a campaign to actively “encourage” serosorting as a HIV prevention strategy. That campaign was initiated coincidentally when their article was published in STI by means of a press release, a web based campaign and use of other media. 1-2
“Serosorting” has been variously and vaguely defined, even by the authors themselves. 2-4 It is also subject to radically different public perceptions by the at risk community. Although the STI article refers to it as a rather recent phenomenon in terms of effect on HIV incidence, the campaign acknowledges that this new fangled term describes an obvious behavior, in my view one that is much like natural selection, operating over the last 25 years. Most likely its effect on incidence has always
been a background phenomenon. The web campaign states:
"Since the beginning of the epidemic some gay men have been deciding that they prefer to be with someone of the same HIV status. Many men just feel more comfortable having sex with same status men."
Although the term has a clinically-sounding soothing appeal, cool rational “preference” decision making based on considerate comfort levels seems to be stretching the boundaries to describe the deep fears, emotional anxieties, ignorance of transmission and discrimination characteristic of the epidemic then and now. The authors have not evaluated any psychological or emotional determinants of this practice as a prelude to rollout of public prevention program.
Of great concern is that with the health department’s recognition that the study has significant limitations and “remains a hypothesis that needs validation,” the agency’s Prevention Section moved forward on a serosorting based campaign publicly. The admitted limitations include
lack of clear causality, uncertainty of temporal sequence trends in the different ways data are collected, lack of key biological or behavioral information, alternative hypotheses for explaining HIV incidence and the nature of ecological modeling based studies. To those could be added lack of information about comparative circumcision rates recently shown to affect transmission and suspected to apply also in MSM and differentially in racial or ethnic populations.5 The article data do not describe any population influenced differences.
But there are more serious problems with a rollout of a serosorting prevention campaign. None of the epidemiological or ecological studies have evaluated social and other harms that may be expected. These include
legitimizing stigma against HIV infected individuals and ratifying a “pariah” status, unfairly criticizing committed serodiscordant couples because the campaign message is tantamount to telling them they are in the wrong relationship or engaged in practices that may not work, confusing messaging about the viability of other safe sex practices or the benefits of a responsible healthy sexual relationship with discordant partners. The campaign is also illogical in that the agency did not study or embark
on similar campaigns for other serodiscordant risks such as those for heterosexual HPV status.
Other messages described in this public health initiative are welcome and useful. Increasing knowledge, testing, discussion and disclosure are key elements of a prevention program. Encouraging serosorting – as it is
called today –raises a serious risk of return to objectionable, discriminatory and harmful social behaviors based not on what a person does but simply on who they are.
Sincerely,
Robert Reinhard
Community Advisory Board Member, San Francisco Clinical Trials Unit
Email: Robert_Reinhard@worldnet.att.net
References
1. G. Newsom and M. Katz. AIDS Office to Launch new HIV Prevention Campaign. Press release, City and County of San Francisco. November 1, 2006.
2. http://www.disclosehiv.org/
3. Mao L, Crawford JM, Hospers HJ, Prestage GP, Grulich AE, Kaldor JM, Kippax SC.. Serosorting" in casual anal sex of HIV-negative gay men is noteworthy and is increasing in Sydney, Australia. AIDS. 2006 May 12;20(8):1204-6
4. Parsons JT, Severino J, Nanin J, Punzalan JC, von Sternberg K, Missildine W, Frost D. Positive, negative, unknown: assumptions of HIV status among HIV-positive men who have sex with men. AIDS Educ Prev. 2006 Apr;18(2):139-49
5. Centers for Disease Control. Male Circumcision and Risk for HIV Transmission: Implications for the United States. December, 2006.
http://www.cdc.gov/hiv/resources/factsheets/PDF/circumcision.pdf
Dear Editor,
Recently the world has seen a growing number of "transformative programs" engaging men in eradicating gender-based violence and reducing the spread of HIV/AIDS. The global Men as Partners Program (MAP) implemented by EngenderHealth in 20 countries since 1998, primariliy, targets various groups of men through ecological model to transform male gender norms. The purpose of MAP is twofold: 1) to challenge t...
Dear Editor,
In his first paragraph Gersovitz mis-represents the aim of our paper¹ by suggesting that we set out to “..investigate adoption of the ABC approach..”. In fact, the focus of that paper is on how one should best measure and summarise age at first sex, in the context of the intense interest that HIV prevention efforts have generated. We discussed the relative merits of the different measures and concluded th...
Dear Editor,
In their recent review of gonorrhoea testing published in STI, Bignell et al state that whether or not to perform anal screening for gonorrhoea in men who have sex with men (MSM) should be guided by a sexual history [1]. Both UK [2] and US [3,4] guidelines recommend screening for anal gonorrhoea and chlamydia among men who have sex with men (MSM). US guidelines recommend anal screening only for MSM who re...
Dear Editor,
We would like to correct the inaccurate comments made in the commentary paper “The National Chlamydia Screening Programme and the NICE guidance on one-to-one interventions: remember the under 25s” from Simms et al, in relation to the National Institute for Health and Clinical Excellence (NICE) guidance ‘Prevention of sexually transmitted infections and under 18 conceptions’ (Sexually Transmitted Infect...
Dear Editor,
Drs Shamanesh and Radcliffe on the one hand and Dr Horner on the other provide finely balanced arguments for and against screening of asymptomatic men for non-specific urethritis (NSU). Drs Shamanesh and Radcliffe, in addition, slip in their recommendation to abandon examining asymptomatic men and this needs to be challenged. Dr Ross suggests that individual departments may chose whether to continue with...
Dear Editor,
In her recent editorial, Judith Stephenson highlighted the lack of trial evidence supporting the opportunistic approach used by the National Chlamydia Screening Programme (NCSP) and calls for a randomised control trial (RCT) to compare opportunistic versus a register based call/recall approach to screening1. What this editorial and the National Institute for Health and Clinical Evidence (NICE...
In their manuscript, Utility of the Determine TP Rapid Syphilis Test in Commercial Sex Venues in Peru, Campos et al. report poor validity and utility of a point-of-care rapid syphilis assay in an outreach facility for commercial sex workers in urban Peru [1]. We have identified potential limitations in their study design, which might have contributed to their conclusion that the TP Determine Rapid Syphilis Te...
Dear Editor,
The article by Prost el al.[1] entitled ‘There is such a thing as asking for trouble’: Taking rapid HIV testing to gay venues is fraught with challenges” is an important addition to the growing literature focusing on HIV testing at entertainment venues catering to men who have sex with men (MSM). Their ethnographic approach to exploring the complexity of such programs adds further dimension to this topic,...
Dear Editor,
As the editor of the recently published NAM Manual of HIV Prevention and a person living with HIV, I think serosorting is on balance a positive influence on the course of the HIV epidemic and, as Robert Reinhardt says, say, it’s ‘natural selection’. If I’m honest, if I was HIV negative, I’m not sure if I’d want to have sex with someone with HIV.
But I don’t think serosorting is a good thing in itsel...
Dear Editor,
The authors of the cited article on serosorting in San Francisco, including members of the city’s Public Health Department, have created more problems than they have solved due, in part, to the premature rollout in San Francisco of a campaign to actively “encourage” serosorting as a HIV prevention strategy. That campaign was initiated coincidentally when their article was published in STI by means of...
Pages