Sexual health interventions among early adolescents : an integrative review

Purpose – Adolescent pregnancy and sexually transmitted infections including HIV are important public health issues in many countries around the world. The trend is escalating among adolescents of younger ages. Many programmes aim to prevent adolescent sexual risk behaviours associated with these health outcomes. The purpose of this paper is to describe and analyse methodological and substantive features of research on interventions to delay the initiation of sexual intercourse and prevent other sexual risk behaviours among early adolescents. Design/methodology/approach – The review includes studies published between 2006 and 2017 that appear in Thai and international databases. Findings – A total of 33 studies met specified review criteria. Of the studies reviewed, 14 used randomized control designs, 16 used quasi-experimental designs and three used a pre-test, post-test design. Sample sizes ranged from 64 to 4,776 subjects; nearly two-thirds of the studies (n 1⁄4 22) included multiple follow-up surveys. The programmes evaluated in these studies can be grouped into three categories: abstinence-based sexuality education programmes (n 1⁄4 12), comprehensive sexuality education programmes (n 1⁄4 13) and youth development programmes designed for early adolescents (n 1⁄4 10). In total, 22 programmes reviewed were effective in changing targeted adolescent psychosocial and/or behavioural outcomes. Research limitations/implications – One of the limitations encountered in this study involved the search in library databases published only in Thai and English. Due to the limitation of searching library databases included in the review that reported the effects and differences among the included studies. Moreover, the search included publications consisted of heterogeneous designs and purpose with reports of different types of outcomes made it impossible to compare effect sizes. Practical implications – The knowledge gained can be used to design and implement effective sexual health promotion programmes for early adolescents. Moreover, the findings can be applied as a guideline for health provider working in a schools, communities and adolescent health services. Recommendations for future studies are needed to consider methodological and substantive characteristics as well as the people who are involved in youth sexual health outcomes. Social implications – It is necessary for government to develop a policy for encouraging parents to be aware and participate in solving the problems. Originality/value – The review indicated that the positive effects of parental involvement and technology media as a programme material on the psychosocial and behavioural outcomes.


Introduction
Sexual risk behaviours among young adolescents appear to be on the rise [1].Meanwhile, the average age of the first sexual intercourse is declining.The average age of first sexual intercourse in male and female students was 13.2 and 13.3 years old, respectively [2].Globally, 1m of these girls being 15 years old or younger give birth every year [3].Correspondingly, in Thailand, the number of births among extremely young adolescents age 10-14 is also increasing to twice the rate report ten years ago [4].
Throughout the world, countries are taking steps to ensure that adolescents access high-quality sexual health education and prevention services.A variety of sexual risk prevention programmes based on the principles of nursing and public health have been implemented to achieve reductions in teen pregnancy and other sexually transmitted infections (STIs) including HIV [5].A number of publications and programmes have described several models for engaging adolescents in sexual health through computer programme or social media [6,7], or interactive activities [8].Some interventions explicitly involve parents [9], peers [10] and teachers [11] as an important resource of influence on adolescents' attitudes, norms, self-efficacy, and sexual behaviours.
A need for reviewing the literature specific to types of characteristics, content and study designs is required to examine the literature regarding young adolescents in particular.A very few studies have evaluated sexual health interventions specifically designed for the early adolescent with age between 10 and 13 years [2].Moreover, specific support on what features or variables influence the effects of these programmes is unclear, especially in the early adolescent.Thus, this is an opportune time to offer developmentally appropriate, evidence-based sexual health programmes [3].
During the past decade, research has been conducted to examine the effects of a sexual promoting intervention.However, the methodological and substantive features of these existing studies have never been reviewed in Thailand to identify gaps in the studies that, in turn, will affect the effectiveness of the sexual health interventions to delay the initiation of sexual intercourse and prevent other sexual risk behaviours particularly early adolescents.

Search strategy for identification of studies
The integrative review was to provide information on the design and methodology of a programme to prevent sexual risk behaviours among early adolescents and construct from the analytic framework of the previous study [12].The programme may be designed for adolescents and/or people who influence adolescents' sexual behaviours.The literature search included different databases: PubMed, CINAHL, Scopus, Science Direct, Web of Science, ThaiJo and TCI.Keywords are the following: "adolescent sexual" (health AND prevention), "early adolescent" AND "sexual intervention", and "sexual promoting programme" OR "sexual abstinence programme".All of the studies identified during the database search were assessed for relevance to the review based on the information provided in the titles and abstracts.The applicability of these full-text articles to the inclusion criteria was assessed in order to determine the relevance of the review objectives.For all articles that appearing to meet the inclusion criteria is shown below.
Types of studies.This review included randomized controlled trials (RCTs), quasi-experimental designs and pre-post-test designs that examined the effects of s interventions on early adolescents' sexual behaviours.Studies were written in Thai or English because sexual risk behaviour in adolescent is a serious public health issue in Thailand.Studies reported as abstracts only were excluded.
Types of participants.Research works that included early adolescents aged between 10 and 13 years were included in this review.This review focussed on general populations of early adolescents, vs highly vulnerable populations.In addition to early adolescents, some of the studies in this review explicitly included important persons influencing the attitudes, beliefs and behaviours of adolescents, such as parents or guardians, peers and teachers.
Types of interventions.The interventions of interest were abstinence-based sexuality education programmes, comprehensive sexuality education programmes and youth development programmes [13] designed for early adolescents.The intervention activities had to be mainly geared towards adolescents and/or their parents, peers and teachers.
Types of outcomes.To be included in this review, the outcomes of interest had to be collected from adolescents.This review is focussed on adolescent sexual behaviour outcomes including initiation of sexual activity, condom use and other contraceptive use.Other outcomes to be considered include intermediate or psychosocial outcomes such as adolescents' attitudes, self-efficacy and intentions related to sexual behaviour.
Studies were excluded from this review, if the intervention targeted a group other than early adolescents or if the study did not include outcome data from adolescents.The first author screened and selected the articles.Then, all authors independent reviewers for their eligibility.Full-text articles were reviewed for acceptability; any disagreement among them was resolved by consensus.

Results
The titles and abstracts of 195 articles were screened from 218 articles reviewed during the searches, 144 records were excluded because they were not intervention studies or full-text articles, 51 published full-text articles, 15 articles were excluded because participants were not early adolescents and 3 articles were excluded because data were not collected from adolescents.A total of 33 articles were retained for this review.For the intervention part of each study, the text was read several times with the intention of finding common patterns and further relevant content in terms of similarities and differences.The methodological characteristics and the substantive features of the studies included in this review are summarised in Tables I and II and described in the following paragraphs.

Methodological characteristics
Research design.This review included studies utilising research designs ranging from those considered to be the most rigorous (individual and group RCTs) to moderately rigorous designs (quasi-experimental designs), and less rigorous pre-test, post-test research designs [14].Of the 33 studies in this review, 14 used a randomized control design, 16 used a quasi-experimental design and 3 used a pre-test, post-test design, No Thai study with randomized control trial design was found.
Sample size.The studies included in this review had sample sizes ranging from 64 to 4,776 participants.
Follow-up data collection points.Nearly two-thirds of studies (n ¼ 22) had more than one follow-up data collection point.In most studies, the first follow-up was conducted immediately after the intervention programme had ended.The final follow-up data collection point ranged from immediately after the intervention to seven years after the intervention had ended.

Sexual health interventions adolescent and their parent into the programme) and only one study evaluated parent-only interventions (including only parent into the programme).
Type of programmes.The studies can be grouped into three categories employed by the US Department of Health and Human Services Teen Pregnancy Prevention Evidence Review [13].A total of 12 studies evaluated abstinence-based sexuality education programmes, 14 studies evaluated comprehensive sexuality education programmes and 9 studies evaluated youth development programmes.Five studies evaluated programmes designed to prevent multiple sexual and non-sexual risk behaviours, and two studies compared abstinence-based sexuality education and comprehensive sexuality education programmes.
Intervention settings.The majority of the studies conducted interventions in school settings (n ¼ 27).Five studies conducted interventions in community settings and one study perform the intervention in home settings.

Sexual health interventions
Programme delivery.As noted in Table II, these studies can be classified as primary programme delivery model.In five studies, technology was implemented to deliver the intervention including two studies with online interventions, and three studies using computer game based interventions.The game provided colourful animation with audio and subtitles.Through real-life scenarios involving cartoon characters participants learned decision making skill in sexual situations.In one study, a set of audio CDs was mailed to parents with emphasis on improving parents' awareness and parenting practices to help their early adolescent daughters avoid alcohol use and sex.In 27 studies, face-to-face interactive activities were used to deliver the intervention including, group discussions, brainstorming, role-playing, demonstrations, games, parent-child homework activities and journaling.
Programme length and length of sessions.Over 75 per cent (n ¼ 25) of programmes included fewer than ten sessions.Four programmes included in this review consisted of 10-20 sessions and four programmes included more than 20 sessions.The length of individual sessions ranged from 30 to 180 min.
Booster sessions.Most of the studied programmes (n ¼ 26) included no programme booster sessions during the follow-up period.Of the seven programmes that included booster sessions, six had 1-3 booster sessions and one programme had 4 booster sessions.
Programme facilitators.In total, 11 studies explicitly involved people in adolescents' daily lives as programme facilitators, while 22 studies involved only specially-trained research staff as programme facilitators.Of programmes facilitated by persons in adolescents' daily lives, three studies used peers as role models or educators for leading a curriculum.Two studies concern parents and family members in adolescent homework assignments.Three studies involved regular classroom teachers in delivering the programme, intervention and one study involved teachers in delivering the programme and parents in sending a postcard to their adolescent, encouraging sexual abstinence in children.Two studies involved teachers and peers as educators.

Adolescent outcomes
To be included in this review, studies had to include at least one psychosocial or behavioural measure of adolescent sexual health as depicted in Table I.A total of 14 studies measured only adolescent psychosocial outcomes related to sexual behaviour.A total of 17 studies measured the outcomes of sexual initiation, while 18 studies measured other sexual risk behaviours such as recent sexual activity (six studies), a number of sexual partners (three studies) and contraception and/or condom use (nine studies).Other adolescent outcomes reported by some studies (not summarised in this review) included risk behaviours, such as violence, alcohol and other substance use.
Positive outcomes.The studies that had statistically significant ( p o 0.05) differences between intervention groups in adolescent sexual health outcomes included seven studies measuring psychosocial outcomes only (the psychosocial outcomes in this reviews are knowledge, attitude, self-efficacy, intention and parent communication), 12 studies measuring sexual initiation, 4 studies measuring recent sexual activity, 1 study measuring the number of sexual partners and 5 studies measuring condom and/or other contraceptive use.Various types of interventions indicated statistically significant reductions in adolescents' sexual risk behaviours ( p o 0.05).For example, interventions delivered via technology (n ¼ 3), CDs mailed to parents (n ¼ 1) and face-to-face interactive activities (n ¼ 13) were all associated with significant reductions in adolescents' sexual risks.In addition, interventions that were designed for adolescents only (n ¼ 6), parents and teens (n ¼ 9) and interventions that involved other important people in adolescents' lives (n ¼ 4) were associated with significant reductions in sexual risks.

Discussion
Early adolescence, a period of rapid physical, social, cognitive and emotional development, is an ideal time for health promotion and risk prevention [4].Early and repeated exposure to messages about sexual risk reduction and sexual health promotion during this period increases the likelihood of positive sexual health outcomes.This integrative review describing substantive and methodological characteristics and adolescent sexual health outcomes of selected studies provide an overall picture of the nature of research on interventions to prevent sexual risk behaviours during early adolescence.
This review highlights important substantive considerations including intervention activities designed to address important predictors of sexual behaviours; interventions tailoring to the age, sexual experience and cultural context of the adolescents involved including parents and other caring adults who can support adolescents' healthy sexual decision making and behaviour.The following paragraph details each of these considerations.
In all of the studies reviewed, intervention activities were clearly designed to influence important predictors or mediators of adolescent sexual risk behaviour, an important step in creating health behaviour programmes.Across the studies reviewed, intervention activities were consistently engaging and interesting such as using media and computer games as instruments [15][16][17].This review included studies that targeted adolescents of different age ranges (i.e.10-13 years vs 12-16 years), and differing levels of sexual experience (i.e.sexually inexperienced or sexually experienced groups of youth).Across the studies in this review, intervention content was consistently tailored to the age and sexual history of the target population of youths [15,18].In addition to age and prior sexual experience, cultural contexts can affect the sexual norms, beliefs and behaviours of adolescents.Importantly, the interventions included in this review were tailored to particular cultural contexts [19][20][21][22][23]. Prior research across cultural contexts has found connectedness to caring adults to be a consistent protective factor against a range of adolescent risk behaviours [22][23][24][25][26][27].Thus, interventions that involve adults who support young adolescents in making healthy decisions regarding their sexual behaviour can be considered the best practice.The finding can be applied in sexual health promotion programme for early adolescents.In other words, future research should evaluate interventions involving supportive adults [28][29][30].
This review also highlights important methodological considerations including research designs: assessment of intervention exposure, length of study follow-up period, measurement of outcomes and conclusions regarding the effectiveness of interventions.Each of these considerations is detailed below.
In total, 14 studies included in this review used a RCT study design.RCT is considered a gold standard in evaluating the effects of an intervention [31].In terms of adolescents' exposure, the dose of intervention is an important consideration in research on the effectiveness of interventions.Across the studies in this review, variations were observed in the following three aspects of dose: length of individual programme sessions, and duration of overall programmes [32].For example, the number of interventions ranged in duration from 2 days to 35 weeks.Thus, the depth and coverage of sexual risk information, the number of skills practiced and the levels of social or family support provided were different across the studies.
In terms of study follow-up period, the studies in this review involved between one and five follow-up surveys, over a time period that ranged from an immediate post-programme follow-up survey to a follow-up survey at 84 months after the intervention had ended.In total, 66 per cent of the studies in this review involved two or more follow-up surveys.Conducting multiple follow-up surveys enable the researchers to test short-term programme impacts such as psychosocial outcomes and longer-term programme impacts such as behaviour and health outcomes as well as the mechanisms or pathways through which programmes work [23,33].Methods that proved successful for participant and parental

Sexual health interventions
recruitment need to be concerned with studies including minimising the number of intervention sessions, using booster calls and mailings, and allowing flexible scheduling [30].
Further empirical work is needed to determine the time points at which sexual health outcomes should be measured.Conceptual or theoretical frameworks and intervention contents of the programme are also important.Intervention contents communicated to participants must be related to expected outcomes.Many of the reviewed studies demonstrated impacts on short-term outcomes, such as knowledge, attitudes, perception and intention.Both behaviour modification theory and previous studies have indicated that these psychosocial variables are strong predictors of behaviour [34][35][36][37][38]. Delay in the initiation of sexual intercourse, the sexual behaviour most commonly measured by studies in this review, was seen in 12 of 17 studies evaluating this outcome.There were differences in other sexual behaviour outcomes measured by individual studies in this review.Some studies evaluated moderate-term outcomes including recent sexual activity or contraceptive use and others considered long-term outcomes including pregnancy and STIs [39].
Research is also needed to identify evidence-based approaches to promoting sexual health among young Thai adolescents at home, school and in community settings.Most young adolescents spend a substantial amount of time in each of these settings.Ideally, young teens receive consistent messages about their sexuality and sexual health across this setting from parents and other family members, teachers and other school staffs, friends and peer groups, counsellors and youth workers [40][41][42].Sexuality education programmes should be tailored to the ages of the youths involved and to the cultural beliefs and practices of the communities in which these programmes are implementeds [21].
Furthermore, future studies must strive to employ rigorous methodological characteristics such as RCT study designs with at least two follow-up surveys.The measurement of a common set of adolescent sexual behaviour outcomes would allow for a direct comparison of the outcomes associated with different types of programmes, programmes delivered in different settings and programmes using different delivery modes.Some outcomes measuring particular sexual behaviours are sensitive and prone to socially desirable response errors.Staff concerned with professional development may be needed to ensure comfort with sex-related content and sensitivity when they approach participants in the programmes.

Conclusions
This integrative review included 33 intervention studies.The reviewed studies reveal that several types of studies are commonly targeted in programmes and interventions to improve adolescent sexual health outcomes, especially in the early adolescent.The interventions were different in terms of participants, intervention instrumentation, dose of intervention, measurement times, intervention instruments and boosters given.Recommendations for future studies are needed to consider methodological and substantive characteristics as well as the people who are involved in youth sexual health outcomes.Moreover, content and intervention type should be concerned with age, sexual experience, context of each area, persons involved in adolescent behaviours, intervention instruments and need for more studies with strong designs to increase the weight of the evidence.Moreover, it is necessary for the government to develop a policy for encouraging parents to be aware and participate in solving this problem.

Limitations
One of the limitations encountered in this study involved the search in library databases published only in Thai and English.Due to the limitation of searching library databases included in the review that reported the effects and differences among the included studies.Moreover, the search included publications consisted of heterogeneous designs and purpose with reports of different types of outcomes made it impossible to compare effect sizes.
Substantive features Programme participants.Nearly all of the studies (n ¼ 31) targeted adolescents aged 13 years old and younger.Two studies included adolescents between the ages of 12 and 16 years (average 13-14 years old).The majority of studies (n ¼ 27) included both boys and girls.Most of the studies tested adolescent only interventions (including only adolescent into the programme) (n ¼ 20); 12 studies evaluated parent-teen interventions (including 469 Positive results for multiple psychosocial variables; b results vary, depending on type of programmes and participant characteristics [39]17],[26],[39]Notes: n ¼ 33. a [41]]Notes: n ¼ 33. a Comparative intervention study involving abstinence-based sex education and comprehensive sex