The Effect of Pre-Travel Advice on Sexual Risk Behavior Abroad: A Systematic Review

Authors

  • Mieke Croughs MD,

    Corresponding author
    1. Department of General Health, GGD Hart voor Brabant,'s-, Hertogenbosch, The Netherlands
    2. Department of Clinical Sciences, Institute of Tropical Medicine, Antwerp, Belgium
    • Corresponding Author: Mieke Croughs, MD, GGD Hart voor Brabant, Postbus 3166, 5203 DD 's-Hertogenbosch, The Netherlands. E-mail: m.croughs@ggdhvb.nl

    Search for more papers by this author
  • Roy Remmen MD,

    1. Department of Primary and Interdisciplinary Care, University of Antwerp, Antwerp, Belgium
    Search for more papers by this author
  • Jef Van den Ende MD

    1. Department of Clinical Sciences, Institute of Tropical Medicine, Antwerp, Belgium
    Search for more papers by this author

Abstract

Background

Travelers often have casual sex abroad and the risk of acquiring a sexually transmitted infection (STI) associated with casual travel sex is considered to be threefold higher compared to the risk of casual sex in the home country. Consequently, international guidelines recommend including STI advice in the pre-travel consultation. We performed a systematic review on the effect of a pre-travel STI intervention on sexual risk behavior abroad.

Methods

In September 2012, a systematic analysis and meta-analysis of peer reviewed literature were performed on the relation between pre-travel STI advice for travelers and sexual risk behavior abroad. Primary outcome measure consisted of the number of travelers with a new sexual partner abroad; secondary outcome measure entailed the proportion of consistent condom use.

Results

Six studies were identified for inclusion in the review, of which three clinical trials on the effect of a motivational intervention compared to standard pre-travel STI advice qualified for the meta-analysis. Two of these trials were performed in US marines deployed abroad and one in visitors of a travel clinic. The extensive motivational training program of the marines led to a reduction in sexual risk behavior, while the brief motivational intervention in the travel clinic was not superior to standard advice. The meta-analysis established no overall effect on risk behavior abroad. No clinical trials on the effect of a standard pre-travel STI discussion were found, but a cohort study reported that no relation was found between the recall of a nonstructured pre-travel STI discussion and sexual risk behavior, while the recall of reading the STI information appeared to be related to more consistent condom use.

Conclusions

Motivational pre-travel STI intervention was not found to be superior to standard STI advice, while no clinical trials on the effect of standard pre-travel STI advice were found.

A recent review on travel and sexually transmitted infections (STIs) found that one in five international travelers is prone to have a new sexual partner abroad. Moreover, half of those who have casual sex abroad do not consistently use condoms.[1] There is not much data on the risk of acquiring a STI associated with casual travel sex; based on three studies, however, it was estimated to be threefold higher compared to the risk accompanying casual sex in the home country.[1] The Canadian Committee to Advise on Tropical Medicine and Travel (CATMAT) even claims that among the infectious risks associated with travel, the STIs probably come second to malaria in terms of potential for serious morbidity and mortality.[2] Moreover, the number of overseas acquired human immunodeficiency virus (HIV) infections appears to increase steadily.[3, 4] Therefore, numerous reviews and international guidelines recommend that health care professionals should counsel on STI risks in the pre-travel consultation and commend to be abstinent or to use barrier protection.[5-11] It has been shown that STI interventions and in particular motivational interventions in nontraveling high-risk populations could successfully reduce sexual risk behavior and subsequent STIs.[12-15] However, it is not clear if pre-travel counseling and motivational interventions reduce sexual risk behavior abroad. General campaigns on safe sex and preventive HIV interventions at airports do not appear to influence risk behavior.[16, 17] A specific problem of travelers is that, although they usually have adequate knowledge of STIs and HIV, there is no perceived personal risk of STI acquisition.[18-20] Moreover, casual travel sex is often unexpected as the unfamiliar surroundings appear to change normal attitude and lead to unplanned sexual activity.[21-27]

It is difficult to establish a relation between a STI diagnosis and a risk abroad; some STIs have a long incubation time and travelers who have unsafe sex on a journey often also have unsafe sex in their home country.[22, 28] Therefore most studies use sexual risk behavior abroad as a measure of the effect of pre-travel STI interventions, in spite of the fact that only self-reported data are available. This systematic review focuses on the effect of a pre-travel STI intervention on sexual risk behavior abroad.

Methods

Study Criteria

All studies addressing the relation between pre-travel STI advice for travelers and sexual risk behavior abroad were potentially eligible for inclusion. It was anticipated that not many controlled trials would be available, so cohort studies and surveys were also eligible for inclusion. Case reports, qualitative reports, and opinion articles were excluded from the review. Studies among all kinds of international travelers, including special travelers like expatriates and military personnel deployed to foreign countries were included. Pre-travel STI advice was defined as any intervention by health services with the intention to prevent international travelers from engaging in casual sex abroad. The primary outcome measure consisted of the number of travelers with a new sexual partner abroad, while the secondary outcome measure entailed the proportion of consistent condom use.

Search Strategy

The methodology outlined in the Prisma Statement was used as a guideline for this systematic review.[29] In September 2012, a search of scientific literature was carried out in the databases of Medline, Cochrane, Embase, and CINAHL (Cumulative Index to Nursing and Allied Health Literature). We decided to perform the broadest possible search and to include all publications that might touch upon travel and sexual behavior without necessarily being the article's main subject. We did not filter on publication date; the only limitation we used was language: only articles in Dutch, English, French, Spanish, and German were included. Through the “all fields” search the keyword “travel” was combined (through “AND”) with “sexual behavior” or “sexual behaviour” or “HIV risk” or “HIV” or “HIV infection,” depending on the definition of the search terms used by the database. In January 2013 the Medline search was updated. Additionally, the reference lists of 14 reviews on sexual risk behavior abroad were screened for studies that met the criteria for inclusion.

Data Extraction

The first screening of titles and abstracts was performed by the principal researcher, who also discarded reviews, letters, comments, and guidelines. The abstracts of the remaining studies were screened by the principal researcher and a second author independently. The principal author developed a data extraction sheet and extracted the data from included studies while the second author checked these data. Disagreements were resolved by discussion between the three authors. The following information was extracted: study design, participants and setting, the intervention, the comparator, and the outcomes for casual travel sex and the use of condoms.

Assessment of Risk of Bias

Two researchers independently reviewed the included studies, paying attention to the following criteria: the research question, study design, and methods. Sample size, response rate, randomization, risk of selection bias, multivariate control for confounders, significance, effect size, and confidence intervals were identified from the presented data. The Cochrane risk-of-bias tool was used to assess the risk of bias in the included studies.[30] A low risk of bias is unlikely to alter the results, while an unclear risk raises doubt about them and high risk may alter the results seriously.

Statistical Analysis

Two researchers independently assessed whether the outcome measurements were adequate for inclusion in a meta-analysis, ie, if new sexual partners or condom use were reported. Only studies with a low risk of bias were included. To ascertain the effect of an STI intervention on the rate of casual travel sex and condom use we calculated odds ratios (OR) with a 95% confidence interval (CI). The analyses were conducted using the Cochrane Review Manager 5.2. Forest plots were used for graphical representation of the meta-analyses. This type of representation shows the estimate from each individual study with 95% CI, together with the pooled estimate. It also illustrates the amount of variation between the studies and the significance at a glance.

Results

Study Selection

The search of Medline, Embase, Cochrane, and CINAHL databases provided a total of 1,465 citations (Table 1). After adjusting for duplicates 1,259 remained. Another 102 citations were discarded as these articles clearly did not meet the criteria. Two studies that met the criteria for inclusion were identified by screening reference lists of 14 reviews on sexual risk behavior abroad. Eight articles were added through updates of Medline, none of which was eligible (Figure 1). Four clinical trials, two of which were in US marines, an observational cohort study in visitors of a travel clinic, and a survey in expatriates were identified for inclusion in the review.[20, 22, 31-34] Thus a total of six studies were included in the systematic review.

Table 1. Search strategy
DatabaseSearch strategyHits
Medline#1 “travel” AND “sexual behavior”(“travel”[MeSH Terms] OR “travel”[All Fields]) AND (“sexual behavior”[MeSH Terms] OR (“sexual”[All Fields] AND “behavior”[All Fields]) OR “sexual behavior”[All Fields] OR (“sexual”[All Fields] AND “behavior”[All Fields]) OR “sexual behavior”[All Fields]) AND (English[lang] OR French[lang] OR Dutch[lang] OR German[lang])553
 #2 “travel” AND “hiv risk” NOT #1((“travel”[MeSH Terms] OR “travel”[All Fields]) AND ((“hiv”[MeSH Terms] OR “hiv”[All Fields]) AND (“risk”[MeSH Terms] OR “risk”[All Fields]))) NOT #1 AND (English[lang] OR French[lang] OR Dutch[lang] OR German[lang])263
Embase#10 “travel”/exp AND “sexual behavior”/exp AND [embase]/lim277
 #11 “travel”/exp AND “human immunodeficiency virus”/exp AND [embase]/lim306
Cochrane#1 “travel” AND “sexual behavi*r”7
 #2 “travel” AND “hiv risk”10
CINAHL#1 “travel” AND “sexual behavi*r”23
 #2 “travel” AND “hiv infection”31
Figure 1.

Study selection flow diagram.

Characteristics of Included Studies

Booth and colleagues and Boyer and colleagues studied the effect of an extensive cognitive-behavioral intervention program in predominantly male US marines (Tables 2 and 3).[31, 32] Booth and colleagues found no difference in casual sex, but condom use was higher in the intervention group, while Boyer and colleagues saw less casual sex and less high sexual risk behavior in the intervention group. The latter also established that the intervention group had a higher STI/HIV symptom and treatment knowledge but had a more negative attitude toward the use of condoms. In the cohort study in visitors of a travel clinic of Croughs and colleagues the sexual risk discussion was not structured or formalized to mimic a routine consultation.[22] Although those who recalled a STI discussion more often had casual sex, no relation was found with the use of condoms. On the other hand, there was a significant relation between the recall of reading the STI information in the received travel health brochure and the use of condoms, which was confirmed in the logistic regression analysis (OR 3.3, 95% CI 10.0–1.1). Croughs and colleagues also found that risk behavior in visitors who traveled with their steady partner was very low (0.3%). Hamer and colleagues saw in an observational survey of expatriate corporate workers with casual sex in Ghana, that those who recalled to have received pre-travel HIV education more often used condoms consistently than those who did not.[33] However, this study lacks power. Gagneux and colleagues compared having sex with a local partner between those who received and read a brochure on “sex tourism problems” and those who did not, but they did not report on the relation between the different interventions and the defined outcomes.[20] Senn and colleagues studied the effect of a motivational brief intervention and free condoms in visitors of a travel clinic who planned to travel without their partner. They did not find a significant effect on casual sex or on condom use.[34] The most important analytic characteristics are displayed in Table 3.

Table 2. Characteristics of included studies
 Booth 2001Boyer 2000Croughs 2008Hamer 2008Gagneux 1996Senn 2011
  1. STI = sexually transmitted infection; HIV = human immunodeficiency virus; MBI = motivational brief intervention; CPR = cardiopulmonary resuscitation; BMI = body mass index.
Study designTrialTrialCohort studyObservational surveyTrialRandomized controlled trial
ParticipantsMarines in military school, 96% men, 6 months deployment abroadMale marines, employment abroad, data of previous 6 months abroadVisitors of travel clinic, 45% men, 39% traveled without steady partnerMale expatriates, deployed abroad who had casual sex abroadVisitors of travel clinic, 44% menVisitors of travel clinic, 18–44 years, 47% men, traveling without regular partner
InterventionPre-departure STI/HIV Intervention Program, 6 hours (N = 96)Pre-departure 8-hour multicomponent interactive group sessions and a 2-hour booster (N = 296)Recall of pre-travel STI discussion (N = 955)Recall of HIV education before departure after casual sex abroad (N = 5)Reading a brochure on “sex tourism problems” (N = 338)Standard pre-travel STI advice and free condoms (N = 240) or MBI and free condoms (N = 394)
ComparatorStandard pre-departure STI advice (N = 80)Standard pre-departure STI advice and 8-hour CPR-training and booster mid-deployment (N = 288)No recall of pre-travel STI discussion (N = 948)No recall of HIV education before departure after casual sex abroad (N = 5)Did not read the brochure on sex tourism (N = 1,524)Standard pre-travel STI advice (N = 481)
Results casual sexIntervention 83%—Control 80%Intervention 35%—Control 54%Intervention 6%—Control 4%  BMI 14%—Condoms only 18%—Control 18%
Results consistent condom useIntervention 59%—Control 42% Intervention 68%—Control 62%Intervention 20% (N = 5)—Control 80% (N = 5) BMI 72%—Condoms only 76%—Control 76%
Other outcomes High sexual risk behavior: Intervention 16%—Control 26%  Casual sex with local partner: Intervention 8%—Control 4% 
Table 3. Analytic characteristics of trials in meta-analysis
 Booth 2001Boyer 2000Senn 2011
  1. BMI = body mass index.
Multivariate controlled?Yes, covariance analysisYesYes, logistic regression
Sample size96 intervention + 80 controls275 intervention + 288 controls394 BMI and condoms + 240 condoms + 481 controls
Response rate94%–92%88%–94%66%–67%–65%
Randomized?No, historical cohort as controlNo, quasi experimental (different ships)Yes
Bias riskLow riskLow riskLow risk
Outcome 1New sexual partnerNew sexual partnerNew sexual partner
Outcome 2Inconsistent condom useHigh sexual risk behavior (>1 partner per port and/or inconsistent condom use)Inconsistent condom use
Effect size outcome 13% more casual sex19% less casual sex3% less casual sex
Effect size outcome 217% more condom use19% less high risk behavior5% less condom use
Confidence interval calculatedNoYesYes

Risk of Bias

Three studies had a low overall bias risk (Table 4). The other studies had a high risk of bias and were therefore not eligible for inclusion in a meta-analysis.

Table 4. Risk of bias
 Booth 2001Boyer 2000Croughs 2008Hamer 2008Gagneux 1996Senn 2011
Selection bias++
Performance bias??++
Detection bias?
Attrition bias??+?
Reporting bias??+
Overall bias risk+++

Meta-Analysis: Effect on Casual Sex Abroad

The trials of Booth and colleagues, Boyer and colleagues, and Senn and colleagues qualified for use in the meta-analysis on the effect of a STI intervention on casual sex. The Forest plot of these trials established no difference in sexual behavior abroad between travelers with a motivational pre-travel STI intervention and those who got standard pre-travel advice [OR 0.70 (0.41, 1.21)] (Figure 2).

Figure 2.

Forest plot of comparison: the effect of a pre-travel STI intervention on casual sex abroad. STI = sexually transmitted infection.

Meta-Analysis: Effect on Condom Use

The studies of Booth and colleagues. and Senn and colleagues qualified for the meta-analysis of the effect of STI intervention on consistent use of condoms. The Forest plot of the two eligible controlled trials showed no difference in condom use between travelers with a motivational pre-travel STI intervention and those who got standard pre-travel advice [OR 1.49 (0.91, 2.44)] (Figure 3).

Figure 3.

Forest plot of comparison: the effect of a pre-travel STI intervention on condom use. STI = sexually transmitted infection.

Interpretation

Motivational intervention was not found to be superior to standard STI advice, even though an extended STI training in a subgroup of marines led to significantly lower sexual risk behavior. However, these predominantly young men staying abroad without a steady partner form an unusual risk group, which is reflected in the high number of them having casual sex abroad, compared to visitors of a travel clinic. Probably marines also expect more often to have sex, making them conceivably more receptive for STI advice. Moreover, the marines were subjected to a comprehensive training program of several sessions with slide presentations, interactive educational games, and group discussions, while the visitors of a travel clinic received a brief motivational intervention. Senn and colleagues compared a brief motivational intervention with standard pre-travel advice; standard pre-travel advice was defined as the STI information normally given in a pre-travel consultation without specific instructions for the health care worker in regard to STI prevention. Croughs and colleagues found no effect of this “usual” kind of advice on sexual risk behavior, although reading the STI information was significantly related with more consistent condom use.

Limitations

Few studies on the relation between pre-travel STI advice and risk behavior were found and the only three controlled trials showed a marked variability in participants and intervention. Besides, the controlled trials only used indirect comparisons: an STI intervention was only compared with standard pre-travel STI information. Obviously only self-reported data on sexual risk behavior are available, which may influence validity of the data.

Conclusion

Not more than six articles qualified for inclusion in this review, while only three clinical trials qualified for inclusion in the meta-analysis. Although a comprehensive training program in a high risk sub-group of marines appeared to be superior to standard advice, a brief motivational STI intervention in visitors of a travel clinic was not found to be superior. No clinical trials on the effect of a standard pre-travel STI discussion were found. One cohort study reported on the relation between the recall of a nonstructured pre-travel STI discussion and sexual risk behavior. In this study no protective effect of the recall of such a discussion was found, but the recall of reading the STI information appeared to be related to more consistent condom use.

Implications for Practice

Although a specific pre-travel STI motivational intervention did not prove to be superior to standard pre-travel STI advice, there are no clinical trials on the effect of standard pre-travel STI information. Therefore we still recommend including some STI information for all travelers at risk, namely those traveling without a steady partner. We also would inform them that casual sex abroad often happens, even when this is not expected.[22, 35, 36] In addition, we would advise them to take condoms along, as this was found to be an independent predictor of protected casual sex.[22] Furthermore, we would recommend handing over written information to all travelers at risk, as it has been demonstrated that they usually read this information.[20, 22] This folder should not only focus on prevention but also on post-exposition management.[36]

Implications for Research

Further research should focus on the effect of STI interventions that are feasible in an average pre-travel consultation. We would recommend robust qualitative studies with large samples on the effect of written and digital information. In particular, we would suggest an exploration of the possibilities of using new communicational means like social media and apps. The formation of ethically acceptable control groups remains a challenge, as well as the seemingly unavoidable self-reported data.

Declaration of Interests

The authors state they have no conflicts of interest to declare.

Ancillary