Family Practice Vol. 21, No. 5 © Oxford University Press 2004, all rights reserved.
Randomized trial in family practice of a brief intervention to reduce STI risk in young adults
a School of Public Health, University of Sydney, b (formerly) Needs Assessment and Health Outcomes Unit, Central Sydney Area Health Service, c School of Medical Practice and Population Health, University of Newcastle and d Division of Population Health, South Western Sydney Area Health Service, Australia.
Correspondence to Dr Elizabeth Proude, Drug Health Services, Level 5, Page Building, Royal Prince Alfred Hospital, Missenden Rd, Camperdown, NSW 2050, Australia; E-mail: elizabeth.proude{at}email.cs.nsw.gov.au
Received 15 March 2004; Accepted 17 May 2004.
Proude EM, D'Este C and Ward JE. Randomized trial in family practice of a brief intervention to reduce STI risk in young adults. Family Practice 2004; 21: 537–544.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
Background. Young adults represent a high-risk group for sexually transmissible infections (STIs). No randomized controlled trials of interventions in family practice to reduce sexual risk have been reported.
Objectives. We evaluated the impact of brief advice initiated in routine consultations in 20 family practices to modify young adults' risk perception and self-reported risk behaviour for STIs, particularly human immunodeficiency virus (HIV) and hepatitis.
Methods. Patients aged 18–25 years completed self-administered confidential questionnaires in the waiting-room before seeing participating family physicians (FPs). Patients were unobtrusively randomized to receive advice about safe sex and complementary resources from their FP (INT group) or usual care (CONTROL group). Three months later, patients completed mailed follow-up questionnaires to assess changes in risk perception and behaviour.
Results. Three hundred and twelve patients completed baseline questionnaires. Of 237 who agreed to follow-up, 156 (68%) returned self-administered follow-up questionnaires. At post-test, we found equivalent self-reported use of questions to assess any new sexual partner's risk (33 versus 36%) and equivalent self-reported use of condoms on first occasion of sex with a new partner (73 versus 77%). While self-reported behaviour did not change significantly, we demonstrated significant change from pre- to post-test for knowledge of hepatitis risk, for both intervention and control groups. Multiple regression showed that the odds of risk perceptions were significantly higher at post-test than at pre-test for the CONTROL group but not the INT group.
Conclusion. FPs remain to be convinced that allocation of time within consultations to opportunistic health promotion is worthwhile. Indeed, there are many competing demands for such time. Our results do not yet provide compelling evidence for the impact of a brief intervention about STI risk by FPs. As primary prevention efforts for young adults remain important, randomized controlled trials of larger scope are needed.
Keywords. Family practice, HIV, primary prevention, sexually transmissible infections.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
Reduction of mortality and morbidity due to sexually transmissible infections (STIs) such as human immunodeficiency virus (HIV) is a global priority. In Australia, young adults have been nominated as a priority group for prevention initiatives.1 Furthermore, there has been a shift in the route of HIV infection in Australia, with a steady increase in the proportion of new cases attributed to heterosexual contact, from 4% in 1988–1990 to 25% in 1998, which stabilized at 23% over the period 1999–2003.2
Previous research has demonstrated a sufficiently high prevalence of STI risk behaviour among heterosexual patients visiting family physicians (FPs) to consider their potential as agents for preventive interventions.3 While there is a substantial body of evidence from randomized controlled trials (RCTs) proving that FPs are effective when they initiate discussion about cervical screening4 and smoking cessation,5 there is little rigorous evidence that they can make such a difference in respect of their patients' STI risk. US research suggests that FPs are much less likely to ask about and discuss HIV than they are to talk about smoking or excessive use of alcohol.6 They are unlikely to discuss injecting drug use unless they perceive the patient to be at risk.7,8 However, it is possible to improve the rate of HIV/STI risk counselling of adults in out-patient clinics by intensive training of health care providers to implement a brief systemized intervention.9 Further, two RCTs have shown video-based interventions are effective with young adolescent girls at a hospital STI clinic,10 and with high-risk men attending STI clinics.11 The impact of STI counselling by FPs is untested.
Hence, we evaluated the impact of brief advice initiated in routine consultations in family practice for patients aged 18–25 years to modify risk perception and self-reported risk behaviour for STIs including chlamydia, herpes, genital warts, HIV and hepatitis B or C, using an RCT design. Our age group of 18–25 years was chosen because younger adults are likely to be at an experimental stage of their sexual career12–14 and to have more than one sexual partner over any 12 month period.15
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
FP and patient recruitment
FPs were recruited to participate in our research through advertisements in professional media. One of the authors (EP) discussed the aims of the study during a personal visit and provided training in the brief intervention. On a specified commencement date, patients aged 18–25 years with sufficient levels of English to give informed consent were asked by receptionists on our behalf to read a subject information sheet about the study. Patients who did not wish to participate in follow-up (and therefore to sign the form) were asked only to tick ‘yes’ to signify consent. Patients then completed self-administered questionnaires before seeing their FP. Those who agreed to a 3 month follow-up questionnaire were asked to provide contact details and to sign their consent. All patients sealed their completed questionnaires in envelopes attached to their questionnaire. They were asked to give this sealed envelope to the FP. All FPs knew from an unobtrusive mark on sealed envelopes which intervention the patient had been randomized to receive: either the safe sex intervention for those allocated to the intervention group (INT) or, for those allocated to the control group (CONTROL), a question about smoking as a health risk factor. Envelopes were pre-randomized and distributed consecutively by receptionists blind to study design.
Intervention
For those patients allocated to receive the safe sex intervention, the FP first asked, "Do you think you are at risk for a sexually transmitted disease? For an unwanted pregnancy? For hepatitis B or C?" According to the patient's response, the FP offered brief behavioural advice and gave the patient a discreetly packaged set of complementary resources (see Box 1). Once the patient had left, FPs completed a checklist for each patient, indicating the reason for the visit and whether the intervention was given as allocated. Three months after this ‘index’ consultation, patients in both groups who had agreed to follow-up were mailed confidential questionnaires with reply-paid envelopes. Non-responders were given up to three reminders by post or telephone.
| BOX 1 Complementary resources distributed as part of the brief advice ‘safe sex’ pack A packet of two condoms and two sachets of water-based lubricant A set of pamphlets from a Sexual Health Clinic about genital wart virus infection, chlamydia and genital herpes; Pamphlets giving locations of relevant clinics and their services A pamplet about hepatitis B, describing risks and availability of vaccination A pamphlet about safe sex, HIV and AIDS A postcard with information about alcohol and drug information services A pamphlet about condoms, including written suggestions about how to discuss their use with a partner
|
Baseline measures
Our 18 page baseline survey incorporated items from previous research.3,16–21 It comprised four sections as follows. (i) Socio-demographics: demographic data, the length of time the respondent had been a patient of this FP and the reason for the visit. (ii) Risk perception: self-assessment of risk for STIs, HIV, unwanted pregnancy, hepatitis B and C; perceived likelihood of becoming infected with HIV. (iii) Risk behaviour: use of drugs including tobacco, alcohol, marijuana and injected drugs; sexual orientation, activity and lifetime occurrence of STI; number of sexual partners in the past 3 months, the event of a new partner in that time, whether their partners had concurrent partners; if condoms were used; whether they asked a new sexual partner about other partners, drugs and HIV status; likelihood of using a condom with a new partner; lifetime occurrence of anal sex; and whether the patient had had an HIV test, and its result. (iv) Other related issues: knowing where to buy condoms; possible objections to their current FP asking about sexual orientation, sexual practices and number of partners and giving information on safe sex or advice on illicit drugs.
Follow-up measures
In addition to identical replication in our follow-up questionnaire of items asked previously at baseline about risk perception and risk behaviour, we also asked about: (i) recall of the previous ‘index’ consultation; whether the FP had asked or given any advice about safe sex; recall of any complementary resources given at this visit; and (ii) ease with which the patient could discuss sex problems with their FP, and from whom they would prefer to obtain information about sex matters.
Sample size estimation
We aimed to recruit a sample size of 300 patients at baseline to estimate a prevalence of 
25% for risk behaviours with a confidence interval within ± 5% of the point estimate. It also was assumed (‘worst case’) that 33% could be lost to follow-up, leaving 100 in each group to detect a between-group difference at post-test of an intervention effect on risk behaviours of 15–20% with a significance level of 5% and 80% power.
Statistical methods
Frequencies were produced for all baseline characteristics using SPSS version 9.22 Baseline characteristics of the INT and CONTROL groups were then compared using the chi-square statistic (since all variables of interest were categorical). In order to assess attrition and loss to follow-up, baseline characteristics of those completing and not completing follow-up were compared.
The primary outcome of interest was ‘any risk’, classified as ‘yes’ if a respondent reported any perceived risk for an STI, HIV, hepatitis B or C, and unwanted pregnancy, and ‘no’ if a respondent reported ‘no’ to all risks. Baseline and follow-up data for participants with measures at both times were included in logistic regression analyses. All socio-demographic variables as well as smoking, alcohol and drug use were included initially in the regression models, and backward stepwise regression was undertaken to exclude variables. As well as the intervention group, time period and the interaction term between these two variables, age and sex were retained in the models even if not statistically significant, because of their importance in this type of research. Other variables were included in the final model if they had a P-value of <0.10. For prevalence estimates, clustering of patients within FPs was taken into account using the svylogit command in STATA version 6.0.23 The svylogit options allow for analysis of data from studies with a range of complex sampling schemes, including cluster sampling. The FP was defined as the primary sampling unit. Because this analysis does not produce likelihood functions, the Wald test was used to assess significance rather than the likelihood ratio test. Risk behaviour (in terms of STI) was defined as those patients who had a new sexual partner in the previous 3 months and had not used condoms. Finally, frequencies for responses to items about recall and acceptability of the FP intervention, the ease of discussing sex with their FP and their preferred sources of information about sex were produced.
Ethical approval
This study protocol was approved by the Central Sydney Area Health Service Ethics Review Committee (RPAH zone).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
Twenty-three FPs in 20 practices participated in the study. Fourteen (61%) were female and nine (39%) male. The majority (83%) had been trained in Australia, and 14 (61%) had additional postgraduate qualifications.
Patient characteristics
Of 383 patients aged 18–25 years seen during the study period, 41 were ineligible because they were too ill or had already participated. Of 342 eligible patients, 30 refused (91% response rate). There was no evidence of response bias by age group, but females were significantly more likely than males to refuse to participate (11 versus 3%) (chi-squared = 4.91, P = 0.027). Of those 312 who completed baseline questionnaires, 156 were allocated to the INT group (46 males, 110 females) and 156 to the CONTROL group (44 males, 112 females). The number of patients recruited per FP ranged from two to 34 (mode 16, median 19).
Table 1 summarizes the socio-demographic characteristics of the sample. Characteristics were similar between groups. Twenty-five percent of our sample were daily cigarette smokers (24% of females and 29% of males), 33% had smoked marijuana in the previous 3 months (59% ‘ever’) and 5% had ‘ever’ injected drugs (Table 1). Sixty-four percent had consumed alcohol in the previous week.
|
Baseline risk perceptions and behaviour
At baseline, the majority (256, 83%) of the sample had ever had heterosexual sex. Almost 92% of males and 87% of females, respectively, were heterosexual (89% overall). In the previous 3 months, 221 (71%) had had heterosexual sexual intercourse. Nine (3%) had had sex in the previous 3 months with someone of the same sex (one man and eight women; one in the INT group and eight in the CONTROL group). The majority of sexually active patients (61%) had only one partner in the previous 3 months, 26% none, and only 14% had had more than one partner. In total, 53 (17%) had a new sexual partner in the previous 3 months.
Ninety-eight patients (38%) had been tested for HIV (Table 1). Being born in Australia (43% of Australian-born, versus 27% of those from other countries; P = 0.017) and being in the older age group of 22–25 years, versus the group aged 18–21 years (48 versus 26%; P < 0.001), were positively associated with HIV testing (data not shown). Further, 10 of the 14 (71%) patients who were homosexual or bisexual had been tested for HIV (P = 0.011).
Other knowledge and beliefs
When asked if they would object to being asked by their FP about their sexual orientation, 91% of participants indicated ‘no’; similarly, number of sex partners (81%, no); sexual practices (79%, no); or drug taking (89%, no). In addition, 83% indicated in their pre-consultation questionnaire that they would not object to being given information on safe sex and 85% would not object to practical advice about illicit drugs from their FP. All patients knew where to obtain condoms.
FP compliance with safe sex intervention
For 156 patients in the INT group, FPs completed 151 (97%) of 156 checklists. For 156 patients in the CONTROL group, FPs completed 146 of 156 (96%) checklists (chi-squared = 1.75, df 1, P = 0.186). For the 151 checklists completed for the INT group, 140 (92%) confirmed that the safe sex intervention had been delivered.
Follow-up
Of 156 patients in the INT group, 116 (74%) agreed to a follow-up questionnaire in 3 months' time. Of 156 in the CONTROL group, 121 (78%) agreed. There was no difference in follow-up rates by intervention group, sexual orientation or age, although males were significantly less likely to agree to follow-up than females (chi-squared = 13.07, df 1, P < 0.001). Hence, 237 participants agreed to follow-up. The follow-up rate for the INT group was 66%, and for the CONTROL group 69%. A higher proportion of females than males returned their follow-up questionnaires (70 versus 61%; chi-squared = 1.754, df 1, P = 0.185). Further, the fully employed (chi-squared = 6.485, df 1, P = 0.011) and current smokers (chi-squared = 13.32, df 1, P = 0.001) who agreed to be followed-up were significantly less likely to complete and return their questionnaires.
Changes in risk perception
Perception of risk for HIV, other sexually transmitted diseases or unwanted pregnancy did not change significantly from baseline to follow-up. Only with respect to perception of risk for hepatitis was there a significant increase in both groups (44–59% in the INT group; 31–51% in the CONTROL group). Using McNemar's test for dependent samples, the P-value for the CONTROL group was 0.005, and for the INT group, 0.001.
Results of multiple regression analysis using the combined variable ‘any risk’ show that the odds of risk perceptions were significantly higher at post-test than at pre-test for the CONTROL group but not for the INT group (see Table 2). Males, those in the younger age group, people with post-school qualifications, smokers, those who had used either marijuana or injected drugs and those who were not definitely heterosexual had higher odds of perceiving themselves at risk for HIV or other sexually transmitted diseases and conditions (Table 2).
|
Changes in risk behaviour
At post-test, there was an equivalent proportion of respondents with new partners in the previous 3 months in each group. Self-reported behaviours to minimize sexual risk are summarized in Table 3. There was an equivalent self-reported use of questions by the INT and CONTROL groups to assess partner's risk, and an equivalent proportion self-reporting that they used condoms on the first occasion of sex with this new partner (Table 3).
|
Recall of FP advice by patients in the intervention group
Table 4 shows recall by the INT group of the safe sex intervention at the index consultation. While over two-thirds (68%) recalled advice about safe sex, only 45 (58%) of the INT group remembered receiving complementary resources. Table 4 also shows the INT group's reported ease of talking about sex with their FP and where they preferred to get information about sex. Small numbers precluded analyses to determine whether variables such as sexual orientation or baseline risk were associated with these views. Males were significantly more likely than females to cite friends or family as sources of information about sex, however (25 versus 7%; chi-squared = 4.385, df 1, P = 0.036).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
If FPs are to be encouraged to provide STI risk reduction counselling, they need evidence acquired from rigorous research to justify confidence in its acceptability and impact. This study was an attempt to generate such evidence. It is useful to know that, at baseline, the majority of participants had no objection to the FP asking about their sexual orientation, their sexual practices, numbers of partners or about drug taking. In addition, 83 and 85%, respectively, raised no objection to being given information on safe sex or practical advice about drugs. Given this favourable orientation, it seems that FPs would be well placed to provide such counselling.
In terms of the impact of their counselling, however, very little objective scientific evidence has been gained through this largely negative trial. While perceptions of overall risk increased in both groups, some changes were paradoxical. Logistic regression analysis showed that younger patients, males, smokers, ‘ever’ drug users, patients with post-school qualifications and those who were not exclusively heterosexual or who were unsure about their sexuality had higher odds of perceiving themselves as being ‘at risk’ of sexually transmitted diseases or conditions, although this risk was not always borne out by their stated behaviour. The cause of the CONTROL group's significantly greater change in risk perception cannot be determined, except that completing the initial questionnaire may have prompted respondents who did not receive the intervention to find out more about these conditions, especially if they had a greater interest in the subject. That control groups also modify their behaviour to the same extent as experimental groups has been noted in other studies.24–26 Indeed, the frequency of specific behaviours at baseline may act as an effect modifier in trials of this type.27
Low levels of risk behaviour in our study are reassuring, although they increased the methodological risk of a type II error. Eleven (14%) of the INT group and 13 (15%) of the CONTROL group reported at follow-up that they had a new sexual partner in the previous 3 months. Only three respondents in each group had not used condoms at the first occasion of sex with this new partner; however, no questions had been asked about partner status, and only very speculative inferences can be made about the degree of risk.
The methodological limitations of the study include the relatively small group of FPs (23 in 20 practices) who participated in the study, the small ensuing patient sample size (312 initial respondents) and the consequent loss to follow-up, resulting in only half those recruited at baseline providing follow-up data. A larger sample size would have minimized the risk of a type II error. Also, because males attend FPs less often than females, there were fewer males in the study. Further, only indirect measures were possible to corroborate that FPs adhered to the randomization. While FPs reported that 92% of INT group patients received the safe sex intervention, patient recall in the INT group was surprisingly low. Randomization at the individual level (by patient) reduces the potential for ‘practice effect’ which could occur in randomization at practice level; however, it introduces the possibility of contamination of the control group. Measures taken to prevent this included the marked envelope being the trigger for the intervention (CONTROL group patients had unmarked envelopes), while patients were not consecutive and were recruited over several weeks.
Nonetheless, our RCT is a novel and useful contribution to the evidence base for STI counselling. This study suggests that an opportunistic discussion about sexual risk behaviour in the context of a routine consultation, e.g. for a respiratory condition, is acceptable both to FPs and to young patients, even though individual risk may be low. We recommend a larger study in order to test a more intensive intervention which specifically targets subjects ‘at risk’. The need to employ an RCT remains compelling.28
|
Declaration |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
Funding: At the time the research was conducted, EP was supported by a CARG (Commonwealth AIDS Research Grant) PhD scholarship.
Ethical Approval: Central Sydney Area Health Service Ethics Review Committee (RPAH Zone).
Conflicts of interest: None.
|
Acknowledgments |
|---|
We thank the Newtown Needle Exchange for providing condoms and lubricant packages; NUAA for postcards; Livingstone Road Sexual Health Clinic, Royal North Shore Hospital and Smith Kline Beecham for donating brochures; all FPs and their patients who took part in the study; and the Royal Australian College of General Practitioners for awarding professional development points to participating FPs.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionDeclarationReferences |
|---|
1 CDHFS. Partnerships in Practice: National HIV/AIDS Strategy 1996–7 to 1998–9. CDHFS: Canberra; 1996.
2 National Centre in HIV Epidemiology and Clinical Research (NCEHCR). HIV/AIDS, Viral Hepatitis & Sexually Transmissible Infections in Australia: Annual Surveillance Report 2003. Sydney: NCEHCR; 2003.
3 Ward J, Sanson-Fisher R. Prevalence and detection of HIV risk behaviour in primary care: implications for clinical preventive services. Am J Prev Med 1995; 11: 224–230.[ISI][Medline]
4 Ward JE, Boyle K, Redman S, Sanson-Fisher RW. Increasing women's compliance with opportunistic cervical cancer screening in general practice consultations: a randomised trial. Am J Prev Med 1991; 7: 285–291.[ISI][Medline]
5 Slama K, Redman S, Perkins J, Reid AL, Sanson-Fisher RW. The effectiveness of two smoking cessation programmes for use in general practice: a randomised clinical trial. Br Med J 1990; 300: 1707–1709.[ISI][Medline]
6 Stryker J, Coates T, DeCarlo P, Haynes-Sanstad K, Shriver M, Makadon H. Prevention of HIV infection. Looking back, looking ahead. J Am Med Assoc 1995; 273: 1143–1148.[Abstract]
7 Kerr SA, Valdiserri RO, Loft J et al. Primary care physicians and their HIV prevention practices. AIDS Patient Care STDs 1996; 10: 227–235.[ISI][Medline]
8 Quirk ME, Godkin MA, Schwenzfeier E. Evaluation of two AIDS prevention interventions for inner-city adolescent and young adult women. Am J Prev Med 1993; 9: 21–26.[ISI][Medline]
9 Dodge WT, BlueSpruce J, Grothaus L et al. Enhancing primary care HIV prevention. Am J Prev Med 2001; 20: 177–183.[CrossRef][ISI][Medline]
10 Shrier LA, Ancheta R, Goodman E et al. Randomised controlled trial of a safer sex intervention for high-risk adolescent girls. Arch Pediatr Adolesc Med 2001; 155: 73–79.
11 O'Donnell CR, O'Donnell L, San Doval A, Duran R, Labes K. Reductions in STD infections subsequent to an STD clinic visit. Using video-based patient education to supplement provider interactions. Sex Transmitted Dis 1998; 25: 161–168.
12 Cochran SD, Peplau LA. Sexual risk reduction behaviours among young heterosexual adults. Soc Sci Med 1991; 33: 25–36.[CrossRef][ISI][Medline]
13 Seidman SN, Mosher WD, Aral SO. Women with multiple sexual partners: United States, 1988. Am J Public Health 1992; 82: 1388–1394.
14 Seal DW, Minichiello V, Omodei M. Young women's sexual risk taking behaviour: re-visiting the influences of sexual self-efficacy and sexual self-esteem. Int J STD AIDS 1997; 8: 159–165.
15 Bauman A, McLellan L, Rissel C, Wise M, Lesjak M, Stoker L. NSW Health Promotion Survey. Unpublished report; 1994.
16 Rickert EJ, Rickert DL. Different HIV risk profiles in samples of college students and homeless persons. Psychol Rep 1995; 76: 1123–1132.[ISI][Medline]
17 Simkins LD. Risk of HIV transmission in sexual behaviours of college students. Psychol Rep 1995; 76: 787–789.[ISI][Medline]
18 Simkins LD. Update on AIDS and sexual behaviour of college students: seven years later. Psychol Rep 1994; 74: 208–210.[ISI][Medline]
19 Holtedahl KA, Doumenc M, Steinert S, Roghell PK. Patients with sexually transmitted disease: a well-defined HIV risk group in general practice? Fam Pract 1991; 8: 42–46.
20 Darke S, Hall W, Heather N, Ward J, Wodak A. The reliability and validity of a scale to measure HIV risk-taking behaviour among intravenous drug users. AIDS 1991; 5: 181–185.[ISI][Medline]
21 Johnson AM, Wadsworth J, Wellings K, Field J. Sexual Attitudes and Lifestyles. Oxford: Blackwell Scientific Publications; 1994.
22 SPSS for Windows [computer program]. Release 9.0; SPSS Inc.; 1999.
23 StataCorp. Stata Statistical Software: Release 7.0. College Station (TX): Stata Corporation; 2001.
24 Calsyn D, Saxon A, Freeman G. Ineffectiveness of AIDS education and HIV antibody testing in reducing high-risk behaviours among injecting drug users. Am J Public Health 1992; 82: 573–575.
25 Dengelegi L, Weber J, Torquato S. Drug users' AIDS-related knowledge, attitudes, and behaviours before and after AIDS education sessions. Public Health Rep 1990; 105: 504–510.[ISI][Medline]
26 Gibson DJ, Lovelle-Drache M, Young M. Does brief counseling reduce HIV risk in IV drug users? Final result from a randomised clinical trial. Paper presented at the VIIth International Conference on AIDS, Florence, 1991.
27 Warner BD, Leukefeld CG. Assessing the differential impact of an HIV prevention intervention: Who's putting the message into practice? AIDS Educ Prev 2001; 13: 479–494.[CrossRef][ISI][Medline]
28 Whitlock EP, Orleans T, Pender N. Allan J. Evaluating primary care behavioural counseling interventions. An evidence-based approach. Am J Prev Med 2002; 22: 267–284.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. J Raveney and P. Oakeshott Randomized trial in family practice of a brief intervention to reduce STI risk in young adults Fam. Pract., October 1, 2005; 22(5): 578 - 579. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||