When developing a program for sti prevention, which action would need to be done first?

Infect Dis Clin North Am. Author manuscript; available in PMC 2009 Dec 1.

Published in final edited form as:

PMCID: PMC2610536

NIHMSID: NIHMS79117

Synopsis

Sexually transmitted infections (STIs) are a major public health concern in the Unites States and worldwide. Many differing types of initiatives have been attempted to reduce the annual incidence of STIs including systematic initiatives (e.g. public health policy changes) and community based programs (e.g. targeted educational campaigns). Since acquisition of STIs is based on individual sexual risk behaviors, the most successful STI prevention initiatives have been behavioral interventions seeking to change those individual risk behaviors. Each of the types of STI prevention initiatives described above as well as the steps in development and validation of behavioral interventions to reduce STIs will be explored. Examples of key interventions developed thus far and the trials in which they were proven effective will be discussed in greater detail. Although physicians have historically been uncomfortable with the concept of incorporating behavioral intervention into their prevention toolkit, it is hoped that gaining greater understanding of the background behind behavioral interventions for STI prevention will allow us to better accept these interventions as additional tools to prevent disease and suffering among those we serve.

Scope of the Problem

Sexually transmitted infections including human immunodeficiency virus (HIV) contribute greatly to the global health burden each year. In 2007 it was estimated that there were 2.5 million new cases of HIV worldwide including 2.1 million adults and 420,000 children [1]. Furthermore there are estimated to be more than 300 million new cases of treatable STIs each year including gonorrhea, chlamydia, syphilis and trichomonas [2]. Although the situation in the United States is somewhat different from that of the developing world, sexually transmitted infections continue to be a major public health concern. Despite targeted prevention efforts, there were almost 37,000 new diagnoses of syphilis in 2006 along with over 350,000 cases of gonorrhea and over 1 million cases of chlamydia [3]. Although the overall rate of new HIV diagnoses has stabilized in the U.S. at about 35–40,000 per year since 2001 (based on the 33 states and US dependent areas with confidential name-based HIV infection reporting), the number of persons living with HIV/AIDS increases annually, estimated to be over 430,000 by the end of 2005 [4].

Both globally and in the United States, women are disproportionately impacted by STIs and HIV. Globally, women now comprise 48% of adults living with HIV [5]. In the US, the incidence of gonorrhea and chlamydia are higher in young women than in men of the same age group (20–24 year olds: chlamydia 2797 versus 857 per 100,000, gonorrhea 606 versus 454 per 100,000) [3]. Minority women in the U.S. are impacted to an even greater extent than their non-Hispanic White counterparts. Gonorrhea rates among African-American women (618 per 100,000) far exceed those of both non-Hispanic White and Hispanic women (44 and 85 per 100,000). Chlamydia, syphilis and HIV are also more commonly diagnosed in African-American and Hispanic women than they are in non-Hispanic White women in the U.S. [3].

STI Prevention via Systematic Initiatives

Due to the significant public health impact of STIs and HIV, numerous attempts have been made to reduce the incidence and prevalence of these infections. These efforts can be grouped into several major categories including: systematic initiatives, community level programs and individual or small group interventions. Systematic initiatives may occur at the local, regional, national or even international level and have included items such as implementing a program to encourage rapid diagnosis and treatment of curable STIs, public health education efforts, identification of clusters of new infections and enacting targeted interventions in those cluster communities and identification of barriers to acquisition of health care with targeted interventions to remove or reduce those barriers. Challenges faced by systematic initiatives include: accessibility to those at highest risk, timeliness, comprehensiveness, ability to be implemented through functional health care systems, and ability to be monitored for efficacy.

Researchers in the United Kingdom recently reviewed the effectiveness of a series of systematic initiatives targeting STI/HIV prevention on a national level [6]. They focused on 5 specific initiatives enacted between 1999 and 2004 including universal offering of an HIV test during prenatal care, offering voluntary counseling and testing (VCT) for HIV to all first time STI clinic attendees, offering Hepatitis B vaccine to all men who have sex with men (MSM) at their first STI clinic attendance, reducing waiting times at STI clinics and implementing programs for early detection of asymptomatic chlamydia. For the two initiatives on which they had prior data, significant improvements were noted in the selected outcome metrics. With regard to prenatal HIV testing, the proportion of HIV infected women diagnosed prior to delivery increased from 40% in 1998 (prior to the 1999 implementation) to 88% in 2003. Rates of acceptance of VCT for STI clinic attendees increased from 40% in 2001 to 56% in 2003 after implementation of that initiative in 2002. Most of these systematic initiatives, however, focus on secondary prevention of STIs and HIV via early detection and treatment to prevent secondary spread (curable STIs) or early detection to reduce secondary transmission via pre-exposure prophylaxis (prevention of mother-to-child transmission of HIV). Although systematic initiatives allow broad implementation and potentially widespread impact, in the absence of a preventive vaccine, systematic prevention initiatives are simply inadequate to stem the current STI/HIV epidemic.

STI Prevention via Community Based Programs

Primary prevention of STIs including HIV requires interventions targeted toward high risk groups on a small group or individual basis as well as different types of interventions for those at lower (but still not negligible) risk. Many of the interventions aimed at lower risk individuals are implemented via community based programs. These include educational campaigns, assuring that condoms are widely available (and affordable), and providing access to STI testing/treatment and counseling. An example of a common community based program, would be the following based on the community of a college campus: improving access to condoms by installing dispensers in campus restrooms, providing STI education (via posters, flyers, email information, web-based information and/or seminars) and making students aware of access to STI care and counseling at the Student Health Center. Another example would be provision of condoms at community venues commonly used for commercial sex work.

Analysis of the effectiveness of community based programs for STI prevention is unfortunately often problematic. Increased outreach efforts often initially result in an increase in diagnosis of STIs and the true baseline rate in the community prior to the intervention is frequently unknown. Condom distribution may result in increased availability of condoms but does not assure their correct and consistent use. Educational campaigns are only of value if the information is put into action via risk reduction or more rapid entry to the health care system for diagnosis and treatment. Although the intervention occurs in a more widespread fashion, community based programs essentially create their impact through encouraging individuals to change their behavior or access health care earlier for diagnosis and treatment to prevent secondary spread. The intervention may be on the community, but the impact must occur at the individual level.

Background on Behavioral Interventions for STI Prevention

Since acquisition of STIs is based on sexual risk behavior, any primary prevention intervention for STIs must impact individual risk behavior. Whereas public health initiatives (such as the systematic initiatives and community based programs described above) are usually at least minimally addressed in the American medical education curriculum, the term “behavioral intervention” is bound to strike fear into the hearts of many physicians because it is a subject matter on which they feel ill prepared to understand or comment. Even though many physicians are becoming more comfortable with the concept of behavioral interventions targeting weight loss or smoking cessation, behavioral interventions targeting such sensitive areas as sexual risk behaviors may still be outside their comfort zone. Training medical students in the counseling skills needed to effectively obtain sensitive sexual information and provide effective STI prevention counseling has been long identified as a significant gap in the medical education system, but with many competing priorities for the limited classroom and clinical hours, this gap often goes unfilled [7]. Within the limited time set aside for STI training in most medical school curricula, the majority goes to diagnosis and treatment with precious little time left for prevention as a whole and specifically for the concept of behavioral interventions for STI prevention. This lack of emphasis in the medical education system along with the resulting lack of knowledge of and experience with behavioral intervention (and behavioral science as a whole) has created a sense among many physicians that these subjects are somewhat mystical or incomprehensible and as such should not be considered solid science.

Providers of STI preventive care regard the behavioral science theory behind STI prevention interventions as being vague, abstract, boring and irrelevant [8]. Interestingly, when a group of STI prevention providers (participating in training sessions at a CDC sponsored STD/HIV Prevention Training Center) were asked about factors affecting STI risk behavior, they consistently independently identified most of the concepts that form the basis for the behavioral science behind STI prevention interventions (recognition and acknowledgement of individual risk and recognition of internal and external factors that influence an individual’s ability to alter their risk status) [8]. Using the Theoretical Domains model, instructors assisted the participants in categorizing the factors they identified as important for influencing behavioral change into 5 theoretical domains: risk appraisal, self-perceptions, emotion and arousal, relationships and social influence, and structural and environmental factors. Even providers with no prior formal education in behavioral science had an intuitive understand of factors that influence behavioral change. Prior to completing this exercise many of the same providers would have assumed that behavioral science was simply beyond their comprehension, clearly that is not a valid assumption. There remains a need for better communication between individuals performing research on behavioral interventions to prevent STIs and those in the public health field currently engaged in STI prevention activities in order to move the field forward at the most expeditious pace.

Development of Behavioral Interventions for STI Prevention

Why is behavioral science and behavior change theory an essential part of intervention development? Even though having a care provider say the words “You should stop having risky sex” is technically a STI prevention intervention, it is likely to be no more effective than saying “You should stop smoking” or “You should lose weight”. In order to have much impact, a STI prevention intervention needs to address multiple aspects of sexual risk behavior. Theoretical models of behavior change provide the framework around which to address risk reduction in a more holistic fashion. A number of different theoretical models have been applied in STI prevention interventions, and there are common themes among them (as described above) including recognition of risk, desire to reduce risk, recognition of barriers to change and identification/implementation of ways to eliminate or reduce those barriers to facilitate change.

One example of adaptation of a specific theoretical model into a successful STI prevention intervention is Project SAFE [9]. The Project SAFE intervention is based on an adaptation of the AIDS Risk Reduction Model. The three stages of behavioral change incorporated into the Project SAFE intervention included: 1) recognition of one’s individual risk, 2) commitment to making changes to reduce that risk, and 3) acquiring the skills needed to implement changes to reduce risk. Indeed, the flow of the Project SAFE three session intervention (one 3 hour session per week for 3 consecutive weeks) is based on spending each of the weekly sessions on one of the previously mentioned aspects of behavioral change. Session 1 focuses on recognition of risk including: raising awareness that minorities are disproportionately impacted by STIs and HIV, addressing commonly held beliefs/myths about STIs (for example one’s ability to determine which men “look safe”) and provision of information on STI acquisition, symptoms (or lack thereof), treatment and long-term impact (including the impact of untreated STIs on fertility and the impact of STI/HIV on families). The second session focuses on commitment to change including: information on prevention of future STIs, emphasis on the importance of early and complete treatment for both partners, teaching how to ask prospective partners about risk behaviors and erotic condom placement and discussing individual empowerment to make decisions and overcome barriers to reducing sexual risk. The third session focuses on acquisition of the skills needed to make changes to reduce risk including: increasing communication skills regarding condom negotiation and other sexual issues (via use of videos, group discussion and role playing), increasing skills in erotic condom application, goal setting and encouragement in building a support network for risk reduction. In a randomized controlled trial, minority women who received this culture and gender specific behavioral intervention were significantly less likely to have a new infection with gonorrhea or chlamydia over 12 months of follow-up compared to the control group women who received standard (15–20 minute) STI prevention counseling (see extended description under Major Trials: Project SAFE below) [9].

A variety of differing types of intervention have been applied to STI prevention. Some, such as Project SAFE, utilized small group interventions targeted toward a single gender or gender/ethnicity group. Some have used small group interventions with diverse participants. Others have employed individual one-on-one counseling either inside a health care setting (doctor’s office or clinic) or out in the community. The intervention facilitator also varied greatly from physicians in private practice, academic researchers, extensively trained/experienced research staff to relatively minimally trained/experienced peer educators selected from the target community. Use of each type of facilitator has advantages and disadvantages. Physicians may have an established relationship and typically hold a position of respect, but have very limited time and (as discussed above) frequently have little training or comfort in behavioral intervention. Academic researchers understand the behavioral theory and its application to STI prevention, but may have little in common with the individuals they are intervening upon and thus have difficulty connecting with them. Trained, experienced research staff probably have the greatest advantage/disadvantage ratio since they have the advantage of understanding the development of the intervention and the theory behind it but may also have experience working with those at high risk and understanding the challenges they face in daily life. Peer educators clearly have the greatest comprehension of the social and economic pressures faced by community members at risk, but unfortunately also have the least understanding of the intervention process. Facilitator demographics could also theoretically impact intervention success. Project SAFE chose to use Hispanic facilitators for Hispanic group sessions and African-American facilitators for African American sessions based on community input prior to implementation. Other studies have chosen not to match facilitators with participants and have described no impact of facilitator demographics on intervention outcome [10].

Trial Design for the Testing of STI Prevention Interventions

Once an intervention has been designed to reduce STI acquisition, it must be tested to determine whether or not it has any impact. Although some practitioners have long held the belief that any intervention must do some good and should be implemented without testing, there have occasionally been well-intentioned interventions that resulted in harm and many others that were not harmful but had no benefit. Transition of an intervention from research/development into widespread implementation is quite costly. It is thus essential to select the most efficacious interventions in which to invest the necessary human and financial resources for this transition to occur. Factors to be considered in testing of an intervention include selection of a target population, length and type of follow-up to be utilized and facilities/staffing required to implement and assess the efficacy of the intervention. The target population may be determined based on the intervention design (only women, only men, only certain types of high-risk individuals, etc.) or may be based on convenience, availability or most efficient use of resources. In order to show efficacy of the intervention, background rates of the desired outcome must be high enough in the targeted population to allow a reasonable chance of finding a significant reduction in the intervention group as compared to the control group. Interventions to reduce STIs frequently recruit from STI clinics because this ensures that individuals who are at higher than average risk for future STI acquisition will be enrolled.

Another key factor in trial design is determining the duration of the follow-up period. Numbers of new infections detected are clearly impacted by the duration of follow-up, but a more important issue is the potential for diminution or acceleration of intervention impact over time. Intervention participants may be very motivated to make risk reduction changes immediately following the intervention, but that enthusiasm may wane as time passes. Conversely, it may take time for participants (particularly women) to disengage themselves from a risky situation and move toward lower risk. The follow-up period needs to be long enough to detect these trends, but must be balanced by the cost and difficulty of retaining high-risk participants over extended periods of time.

The type of follow-up may vary based on the outcome measures employed and whether or not they require examinations for specimen collection (see below). Use of STI acquisition as an outcome requires the ability to perform STI testing in a timely fashion whenever the participants suspects infection, not just at the time of routine visits. This requirement has extensive implications for facilities and staffing needed to perform the study. The resources needed to collect behavioral data via either face-to-face interview or by a computer assisted technique must also be considered. All totaled, large scale behavioral intervention trials are quite costly, complex and difficult to implement regardless of the type of behavior targeted by the intervention. STI prevention interventions have some factors (such as the potential need for pelvic exams to collect specimens) that may make their trials even more complicated than those of other types of interventions.

Outcome Measures for Testing of STI Prevention Interventions

When comparing trials of behavioral interventions to reduce STIs, another key feature to evaluate is the outcome measures selected. Primary outcome measures in these trials can typically be categorized as either biologic outcomes or behavioral outcomes. Biologic outcomes include documented new infection with STI organisms and presence of biomarkers for unprotected intercourse. Behavioral outcomes commonly utilized include self-reported condom use (measured multiple ways), number of new partners, risk status of partners, and use of alcohol or drugs in conjunction with sexual activity.

As mentioned above, condom use can be collected and analyzed in many different ways. The time frame queried may be the last sex act, the last 10 sex acts, the last 3 months or the last 6 months. The context may be with all partners, with the main partner, with casual partners or collection of data separately on each partner in the last 3 or 6 months. Other important aspects of condoms use are consistency and correct use (placement prior to any genital contact, covering the entire penis, use of the correct size with appropriate lubricants, and securing the condom during withdrawal). Common problems with condom use are often related to incorrect use. Slippage or dislodgement into the vagina is often a result of only applying the condom partway up the penis and failure to secure the condom during withdrawal. Breakage may be due to failure to leave room at the tip of the condom, use of the wrong size condom or use of an inappropriate lubricant that accelerates degradation (oils or petroleum jelly). Infection may also result from genital contact prior to condom application.

There has been a great deal of discussion over the past 15 years as to whether use of a biologic outcome was preferable to use of a behavioral outcome. Experts in the field have eloquently argued for one or the other, in the literature and at symposia or conferences. In the end, if the goal of the intervention is to reduce STI acquisition, direct measurement of STI acquisition rates is logically the most appropriate outcome measure. This is most commonly the goal of an interventional trial, but requires the ability to collect samples for STI testing at routine and problem-oriented visits to allow complete ascertainment. Until recently, this required frequent pelvic examinations or urethral swabs to collect specimens, creating a significant burden for both participants and research staff. With the advent of urine based STI testing via nucleic acid amplification (NAATs), biologic outcomes could be obtained without requiring a clinician examination. This advance made collection of a biologic outcome much more feasible in large-scale trials or studies conducted outside of a clinic setting.

In a situation where the goal of the intervention is primarily to alter risk behavior (with or without an impact on STI acquisition), measurement of self-reported behavior would seem to be the most valid primary outcome. Currently, however, behavioral measures are more commonly used as a secondary outcome to help explain the mechanisms by which the intervention impacted STI acquisition rates. Unfortunately, there is no direct measure of risk behaviors beyond the use of self-report, which may lack precision for several reasons, one of which is the dependence on participants remembering details of sexual activity that occurred days or months earlier. Recall bias can be reduced through use of diaries or other methods to improve memory or by limiting questions to recent events (last act or last 10 acts, as opposed to the last 3 or 6 months). Even so, recall bias remains a significant problem. Diaries often go unused, or are filled out on the way to their appointment (commonly while sitting in the waiting room). The highest risk participants are those with many partners and high frequency of sexual activity, making them the least likely to remember when and with whom they used (or did not use) condoms and other details of sexual activity.

Another limitation of self-reported data is social desirability bias (giving the answer you think the interviewer would like to hear, even if it is not a valid response to the question). Any time face-to-face interviews are utilized to collect sensitive information, social desirability is an important confounding factor. Use of self-administered pen and paper interviews was previously sometimes utilized to limit this bias, but required respondents to have adequate literacy to read and respond to the questions. Recently, attempts have been made to utilize information technology advances to minimize desirability bias in collection of sensitive information. Use of computer based interviewing with the inclusion of audio for the questions (ACASI) has been widely adopted for collection of sensitive data and allows collection even in lower literacy populations. ACASI can also be used in settings with multiple language needs as the questions can be pre-recorded in multiple languages and selected as needed. Responses can also be displayed pictorially on the computer screen for selection by the participant. Other uses of technology to improve behavioral data collection include use of phone diaries where participants call frequently to report events to computerized databases as they occur instead of waiting until visits to report them. Research is currently underway to attempt to develop devices to detect sexual intercourse and determine whether or not a condom was used. Obviously this would be quite invasive and could never replace self report in the large scale trial setting.

STI Prevention Interventions: What has been done thus far

It would be impossible in the scope of this review to discuss every behavioral intervention that has been studied for STI prevention in women worldwide. Instead the focus will be on differing types of interventions studied (with illustrative examples) and the major intervention trials that are most commonly quoted in the literature. As mentioned above, behavioral interventions can vary from the simple words “You should stop having risky sex” to multi-session interventions using audio-visual aides, role-playing and skills-building. An illustrative example of a simpler form of intervention utilizing face-to-face brief advice from a physician was studied in Australia [11]. It consisted of brief behavioral advice provided at the time of a routine visit to a family practitioner’s office. The specific advice was based on the patient’s response to the questions: “Do you think you are at risk for a sexually transmitted disease? For an unwanted pregnancy? For hepatitis B or C?” Each participant randomized to receive brief advice was also given a “safe sex” pack that included condoms and educational materials. Individuals who agreed to participate in the study completed a survey of sexual risk behaviors while waiting to be seen by the doctor. Those randomized to brief advice received the advice and “safe sex” pack as described above. Those randomized to control conditions were asked about smoking as a health risk factor. All participants were requested to complete a follow-up survey in 3 months. There were 312 participants enrolled (156 each intervention and control) of whom 237 agreed to receive a follow-up survey in 3 months and 156 actually completed and returned their questionnaire (50% of enrollees). At the time of the follow-up survey, 68% of brief advice participants recalled having received information on safe sex from their doctor and only 58% remembered having been given a package of educational material. A major focus of the questionnaire was sexual behaviors with a new partner. Only 11 intervention participants and 13 control participants had acquired a new partner in the 3 months since their visit. Although obviously limited by small numbers, equivalent proportions in both groups had discussed safe sex with their new partner (36% versus 38%) and had used a condom for their first sexual encounter with the new partner (73% versus 77%). Interestingly, perceptions of risk seemed to increase more in the control group than the intervention group between enrollment and follow-up.

Major Trials: Project RESPECT

Project RESPECT is an excellent example of a more intense individual face-toface intervention and is one of the major trials from the last 15 years [12]. Project RESPECT was conducted at 5 public health STD clinics in large US cities between July 1993 and September 1996. It enrolled 5758 heterosexual men and women who presented to the clinics for STD testing and randomized them to one of three interventions. Arm 1 received 4 individual intervention sessions, the first lasting 20 minutes followed by three 60 minute sessions. Arm 2 received two 20 minute intervention sessions, the first of which was the same as session 1 of arm 1. Arm 3 received two 5 minute sessions of didactic messages similar to those commonly provided in a STD clinic. Arms 1–3 were requested to return for follow-up at 3, 6, 9, and 12 months with testing for gonorrhea and chlamydia. A fourth arm was included that received the same didactic messages as arm 3 but had no scheduled follow-up. Arm 4 was an attempt to evaluate the intervention effect of repeated clinic visits and contact with the research staff. An attempt was made to contact arm 4 participants after 12 months to ask about condom usage. Any time participants from any arm presented to the clinic (whether for a scheduled visit or for a problem) they underwent testing for gonorrhea and syphilis as per the usual clinic routine. This allowed collection of infection data on the arm 4 participants and for more complete ascertainment of infections in the other arms. All interventions were provided by counselors and clinicians in the STD clinics.

The primary outcome measure for Project RESPECT was acquisition of gonorrhea, chlamydia, syphilis or HIV during follow-up. The secondary outcome was condom use measured as 100% condom use with vaginal intercourse during the past 3 months (no unprotected vaginal sex). Of the 4328 participants randomized to arms 1–3, only 51% returned for all 4 follow-up visits but 81% returned for at least 1 follow-up visit and 66% returned for the 12 month visit. Through the 12 month follow-up visit, acquisition of a new STI was less common in arms 1 (11.5%) and 2 (12%) as compared to arm 3 (14.6%). Comparison of the unscheduled (problem) visits between arms 3 and 4 revealed no significant differences in rates of acquisition of gonorrhea or syphilis (3.3% versus 4.1%). With regard to the secondary outcome, having had no unprotected vaginal intercourse was reported more commonly by arm 1 and 2 participants (as compared to arm 3) at 3 and 6 months, but condom use increased for all groups over time and differences between intervention arms had disappeared by 9 and 12 months follow-up.

Major trials: Project SAFE

Project SAFE (which was introduced above in the discussion of intervention design) is a good example of a small group intervention [9]. The Project SAFE intervention has been used in 2 major trials conducted in San Antonio, Texas. The first, Project SAFE enrolled 617 minority women (424 Mexican-American and 193 African-American) with an active, curable STI of whom 313 were randomized to receive the 3 session small group intervention and 304 received standard brief risk reduction counseling. African-American and Mexican-American women randomized to the intervention group returned for 3 weekly 3 hour sessions with women of their own ethnicity. The facilitator was an experienced member of the research staff of the same ethnicity of the participants. Ninety percent of the intervention group attended at least 1 session and 75% attended all 3 sessions. All participants were scheduled for follow-up at 6 and 12 months and were requested to return to the research clinic for testing any time they had suspicion of infection. Retention rates at 6 and 12 months were 82% and 89% and did not differ between intervention and control groups.

The primary outcome measure for Project SAFE was acquisition of a new infection with gonorrhea or chlamydia over 12 months of follow-up. Rates of new infection were significantly lower in the intervention group in both the 0–6 and 6–12 months visit windows [11.3% versus 17.2%, p=0.05 and 9.1% versus 17.7%, p=0.008]. The 12 month new infection rates were 16.8% for the intervention group as compared to 26.9% for the control group (p=0.004) which was a 38% reduction. Secondary outcome measures included behavioral factors addressed by the intervention. Intervention group women were less likely to be noncompliant with their STI treatment (including avoidance of intercourse until the partner was treated), to have multiple partners or to engage in high-risk sex. A subsequent more extensive analysis of risk behaviors addressed by the intervention that may have contributed to the intervention effect identified 5 factors that were more common in women who acquired a new infection during follow-up and also were less common in the intervention group [13]. These were: 1) having unprotected sex with an untreated or incompletely treated partner (noncompliance), 2) waiting less than 3 months to acquire a new sexual partner (rapid partner turnover), 3) not practicing mutual monogamy, 4) never having used condoms with a casual partner or a combination of 5 or more unprotected acts in 3 months and problematic condom use (unsafe sex), 5) douching after intercourse.

Major trials: Project SAFE 2

The second trial of the Project SAFE intervention (Project SAFE 2) utilized the standard intervention and control arms from Project SAFE but also added a third enhanced intervention arm of women who received the intervention and were offered the option of attending monthly support group meetings for 6 months following the intervention [14]. In addition to confirming the results of the prior study and determining the impact of offering support group meetings, Project SAFE 2 sought to evaluate the duration of intervention effect for a longer follow-up period with follow-up initially scheduled for 2 years and subsequently extended to 5 years (years 2–5 results still in analysis). A total of 775 minority women were enrolled (585 Mexican-American and 190 African-American women) with 262 assigned to the enhanced group (intervention plus support group meetings), 237 to the standard intervention and 276 to the control group. Of the 2 arms assigned to receive the intervention, 96% of women attended at least 1 session and 86% attended all 3 sessions. Of the women assigned to be offered support groups, 63% chose not to attend, 37% attended one meeting and 26% attended more than one meeting. Retention rates at 12 and 24 months were 91.4% and 91.2% and did not differ between groups.

The primary outcome measure for Project SAFE 2 was acquisition of a new infection with gonorrhea or chlamydia during follow-up. Both intervention groups had significantly lower rates of new infection than the control group when adjusted for drug and alcohol abuse (Table 1). Women who attended the support group meetings had the lowest rate of new infections in the first year, the second year and for the cumulative 2 year follow-up period with a 63% reduction in the first year as compared to the control group. Factors addressed in the intervention that were associated with infection rates and differed between the study groups included unprotected sex with an untreated or incompletely treated partner, and more than one partner in either year one or year two.

Table 1

Primary outcome analysis of Project SAFE 2: Adjusted rates of acquisition of a new infection with gonorrhea and/or chlamydia over 12 and 24 months of follow-up in the enhanced (intervention plus support groups) and standard (intervention alone) groups as compared to the control (brief counseling only) group. Project SAFE 2 enrolled 775 minority women in San Antonio, Texas between March 1996 and June 1998. Results are presented as adjusted infection rates and odds ratio (95% confidence interval).

This was the first trial of a behavioral intervention to prevent STIs that continued follow-up for a second year. It was previously reported that retention rates and intervention effects diminished over time. In contrast, Project SAFE 2 was able to maintain excellent retention rates in both years (91% each year) and the intervention effect remained significant up to 24 months after intervention. Interestingly, infection rates in all 3 randomized study groups were lower in the second year than the first, but both intervention groups remained significantly reduced as compared to the controls.

Major trials: NIMH multisite trial

The last of the major intervention studies in the literature is the NIMH multisite HIV prevention trial which was also a small group intervention study performed at 37 inner-city community-based clinics in seven states [15]. Participants were randomized to either the control group (a one-hour AIDS education session including a videotape and question and answer period) or the intervention group. The intervention consisted of seven 90–120 minute HIV risk reduction sessions conducted twice weekly for single gender groups of 5–15 participants. The intervention sessions were facilitated by experienced male and female staff members who underwent extensive training and certification prior to study initiation. The participants were recruited from STD clinics (both men and women) and primary care clinics (women only). The NIMH study enrolled 3706 participants between January 1994 and February 1996 with 1851 assigned to the intervention group and 1855 to the control group. All participants were interviewed at baseline, 3, 6, and 12 months of follow-up with urine testing for gonorrhea and chlamydia at 12 months only. For the 2426 participants enrolled at STD clinics (65% of total), an attempt was made to identify new STI acquisition by chart abstraction in addition to the 12 month urine screen. Participants assigned to the intervention group attended an average of 5.2 of the 7 sessions with 63% attending either 6 or 7 sessions. Ninety percent of the participants attended at least 1 follow-up visit and 79% attended the 12 month visit at which STI screening was performed.

The NIMH trial had 5 primary endpoints and 2 secondary endpoints listed in the main study publication including: 1) self-reported unprotected intercourse in 90 days prior to visit, 2) self-reported consistent condom use in last 90 days prior to visit, 3) self-reported proportion of sexual acts in which a condom was used, 4) STD reinfection rate from chart abstraction over the 12 months study period (male STD clients only), 5) point prevalence of gonorrhea and chlamydia at the 12 month visit. The secondary endpoints were self-reported STD symptoms in the last 90 days prior to visit and incidence of gonorrhea over 12 month study period from chart abstraction for STD clients. Although the self-reported sexual risk behaviors declined in both groups, the intervention group reported less risky behavior during follow-up than the control group (more frequent condom use, higher proportion of condom use and more likely to report consistent condom use). STI symptoms were reported more commonly by the control group than the intervention group over follow-up (35% versus 28%). In contrast, neither the rates of new STD acquisition by chart review (9.4% versus 9.1%) nor by urine screening (chlamydia 2.8 versus 2.9, gonorrhea 1.5 versus 0.9) differed between the intervention and control groups. This study is an example of the conflict between those who favor behavioral outcomes and those who argue for biologic outcomes. Even though self-reported sexual risk behaviors were successfully reduced in the intervention group, there was no impact on STI acquisition as assessed in this study. The key issue, however, was the suboptimal detection of new infections utilized in this study.

Next steps: Translation from research trials to public health impact

Behavioral interventions to prevent STI acquisition in heterosexual women have been developed and implemented in research trials. Many trials have resulted in reductions in self-reported sexual risk behaviors and several have resulted in reductions in STI acquisition during follow-up. In order to have a significant public health impact, these interventions must be translated for widespread use. The U.S. Centers for Disease Control and Prevention (CDC) have been at the forefront of the effort to translate successful behavioral interventions from the research arena to the public health domain for broad implementation.

In the mid-nineties, CDC sponsored development of a collection of STI prevention interventions aimed at adolescents in their Program Archive on Sexuality, Health and Adolescence (PASHA). The PASHA collection of 13 behavioral interventions to reduce STIs in adolescents was published in 1996 [16]. To build upon that, CDC sponsored a second project to select a group of successful behavioral interventions for adults and to provide all of the materials necessary to implement each of those interventions in a single package (one package per intervention). The results of the HIV/AIDS Prevention Program Archive (HAPPA) were published in 2001 [17]. The HAPPA selection process included review by an expert scientific panel based on the intervention having had a demonstrated salutary effect on HIV-related risk behavior, in at least one subgroup of adults, in at least one site in the United States. The scientific panel reviewed 32 intervention programs and selected 18 of those for potential inclusion in their “promising prevention programs in a box”. Following discussions with the developers or current holders of the 18 selected interventions, HAPPA was able to acquire 10 of the programs for packaging and distribution. [Of the other eight programs selected, seven were already available from their developer and the eighth was deemed by it’s developer to be obsolete.] Following development of the intervention packages, each was field tested in settings that differed from those in which it was initially tested to verify usability, clarity and appeal to practitioners. When package development was completed, CDC further assisted in dissemination of these interventions through provision of training for local STI prevention providers at CDC-funded Behavioral Intervention Training Centers in California, Denver, Rochester and Dallas. These four Centers are a part of the National Network of STD/HIV Prevention Training Centers funded by CDC.

One important aspect of translation from the research setting to diverse implementation is adaptation to fit the community of interest. Solomon and colleagues recently reviewed the importance of maintaining a scientific basis for the necessary adaptation in order to avoid loss of efficacy [18]. The five steps to research-based adaptation of an intervention include: 1) identify the desired target population and community context, 2) select an intervention program that best matches the population and context, 3) retain fidelity to the “core” of the intervention program, 4) identify and reduce conflicts between the selected program and the target context, and 5) document the changes made and evaluate the outcome of the adapted intervention to allow further adaptation/fine tuning as needed. Utilizing this approach, an intervention designed for Mexican-American women in San Antonio could feasibly be adapted for Latino women in other parts of the country based on community assessment and use of appropriate cultural context.

Summary

Alteration of individual sexual risk behavior is key to prevention of sexually transmitted infections including HIV. Individual or small group behavioral interventions are the most efficacious approach to altering individual sexual risk behavior. Successful behavioral interventions have been designed, developed and tested in their target populations resulting in reductions in risk behavior and/or acquisition of STIs. Efforts are underway to implement those successful interventions more broadly via translation for use in diverse public health settings. This process of development, validation in randomized controlled trials and subsequent dissemination to populations in need will be completed when the ongoing implementations result in feedback to the developers to allow development of second generation interventions which have been improved based on “real world” experience. At that point, the scientific development process for behavioral interventions to prevent sexually transmitted infections will have come full circle. Hopefully, behavioral interventions can soon be recognized by all medical practitioners as another scientifically validated tool to prevent disease and suffering among the populations we serve.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

1. World Health Organization website. 2007.

2. World Health Organization Fact Sheet #110, Sexually Transmitted Infections, revised October. 2007.

3. Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 2006. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention: Atlanta; 2007. [Google Scholar]

4. Centers for Disease Control and Prevention. HIV/AIDS Surveillance Report, 2005. Vol 17. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention: Atlanta; 2007. [Google Scholar]

5. 2006 Report on the Global AIDS Epidemic. World Health Organization. 2006.

6. Brown AE, Tomkins SE, Logan LE, et al. Monitoring the effectiveness of HIV and STI prevention initiatives in England, Wales, and Northern Ireland: where are we now? Sex Transm Infect. 2006;82:4. [PMC free article] [PubMed] [Google Scholar]

7. Steinberg JK, Wellman J, Melrod J. A proposal for strengthening medical school training in STD prevention techniques. Public Health Rep. 1991;106:196. [PMC free article] [PubMed] [Google Scholar]

8. Gandelman AA, Vogan SA, Dolcini MM. Assessing HIV prevention provider knowledge of behavior science theory: building on existing intuitive experience. Health Promot Pract. 2005;6:299. [PubMed] [Google Scholar]

9. Shain RN, Piper JM, Newton ER, et al. A randomized, controlled trial of a behavioral intervention to prevent sexually transmitted disease among minority women. N Engl J Med. 1999;340:93. [PubMed] [Google Scholar]

10. Pealer LN, Peterman TA, Newman DR, et al. Are counselor demographics associated with successful human immunodeficiency virus/sexually transmitted disease prevention counseling? Sex Transm Dis. 2004;31:52. [PubMed] [Google Scholar]

11. Proude EM, D'Este C, Ward JE. Randomized trial in family practice of a brief intervention to reduce STI risk in young adults. Fam Pract. 2004;21:537. [PubMed] [Google Scholar]

12. Kamb ML, Fishbein M, Douglas JM, Jr, et al. Efficacy of risk-reduction counseling to prevent human immunodeficiency virus and sexually transmitted diseases: a randomized controlled trial. Project RESPECT Study Group. Jama. 1998;280:1161. [PubMed] [Google Scholar]

13. Shain RN, Perdue ST, Piper JM, et al. Behaviors changed by intervention are associated with reduced STD recurrence: the importance of context in measurement. Sex Transm Dis. 2002;29:520. [PubMed] [Google Scholar]

14. Shain RN, Piper JM, Holden AE, et al. Prevention of gonorrhea and Chlamydia through behavioral intervention: results of a two-year controlled randomized trial in minority women. Sex Transm Dis. 2004;31:401. [PubMed] [Google Scholar]

15. The NIMH Multisite HIV Prevention Trial: reducing HIV sexual risk behavior. The National Institute of Mental Health (NIMH) Multisite HIV Prevention Trial Group. Science. 1998;280:1889. [PubMed] [Google Scholar]

16. Card JJ, Niego S, Mallari A, et al. The program archive on sexuality, health & adolescence: promising "prevention programs in a box". Fam Plann Perspect. 1996;28:210. [PubMed] [Google Scholar]

17. Card JJ, Benner T, Shields JP, et al. The HIV/AIDS Prevention Program Archive (HAPPA): a collection of promising prevention programs in a box. AIDS Educ Prev. 2001;13:1. [PubMed] [Google Scholar]

18. Solomon J, Card JJ, Malow RM. Adapting efficacious interventions: advancing translational research in HIV prevention. Eval Health Prof. 2006;29:162. [PubMed] [Google Scholar]