Adolescent Intimate Heterosexual Relationships: Measurement Issues

Abstract

During adolescence, individuals develop increased ability to have emotionally and physically intimate relationships. The type of intimate relationship will make a difference as to whether the adolescent engages in protective behaviors or avoids risky behaviors. However, in reviewing the literature on the association of relationship type and sexual risk and protective behaviors, four methodological challenges were noted. These challenges limit our understanding of the impact of relationship type. These four challenges are: a) inconsistent definitional frameworks; b) lack of adolescent voices; c) lack of consideration of gender; and d) poor differentiation between individual and relationship risk. These issues direct the course of future work in this area and are necessary to advance the field of adolescent sexual health, particularly with regard to the development and testing of appropriate interventions designed to reduce untoward outcomes of adolescent sexuality.

Introduction

Adolescence is a time period during which individuals develop relationships that are increasingly emotionally and physically intimate.¹ During the course of this developmental period, adolescents’ relationships change over time, moving from those focused on “doing things together” to behaviors that are increasingly “intimate.” ². Romantic relationships are not uncommon, with over 70% of adolescents reporting having had one in the last 18 months. ¹ As these relationships develop, sexual behaviors will occur, ranging from kissing and touching to sexual intercourse. Non-penile-vaginal intercourse behaviors are common; for example, in a study of 14 -16 year olds, 68% had engaged in kissing, 54% genital touching, and 28% had received oral sex; the frequency of these behaviors increases with age.³ Sexual intercourse also becomes increasingly common with 65% of adolescents having had sexual intercourse by age 18, and 99% by 26 years of age. ⁴ These intimate relationships allow adolescents to explore their sexuality and hopefully, develop responsible sexual behavior. Despite sustained efforts focused on helping adolescents make wise sexual and relationship decisions, adverse outcomes continue to occur at unacceptable rates. Overall, US rates for teenage pregnancy, teenage births, sexually transmitted infections (STIs), and abortion continue to be high and most of these rates are among the highest for western industrialized nations.⁵ For example, it is estimated that one in four sexually experienced adolescents will acquire an STI. ⁶ Further, over 400,000 infants were born to adolescent girls in 2009, with a rate of 39.1 per 1000 adolescent girls.⁷

There is clear evidence that the type of relationship makes a difference for sexual risk behaviors and emotional experiences. Studies have found that sexual behaviors (i.e., frequency of oral sex and intercourse) differ by relationship type, even though the terms for relationship type differ. The associations between relationship and risk behaviors and emotional experience are summarized below, but in doing so, the difficulties with the inconsistent use of language become apparent. Thus, there is no doubt that relationship type matter, since studies keep finding differences, as demonstrated by the studies below. However, these consistencies make hard to compare these results across studies, since they do not use the same language. For example, adolescents with exclusive partners have more frequent sex than adolescents with non-exclusive partners.⁸ With regard to oral sex, one study found that females were more likely to have oral sex with main partners as compared to casual partners.⁹ When examining use of condoms and contraception, adolescents in romantic relationships as opposed to those in non-romantic relationships are more likely to report ever using contraception,¹⁰ using contraception in first relationships,¹¹ and using contraception at first intercourse.¹² However, adolescents in romantic relationships are less likely than adolescents in non-romantic relationships to report consistently using contraception or always using contraception.^10,13 Similarly, adolescents are more likely to use condoms with casual partners than with steady partners.^9,14-17 Lastly, with regard to adverse outcomes, one study found that adolescents who had casual-only partners were less likely to ever be pregnant than those with main partners.¹⁸ With regard to STI/HIV risk, one study found that risk perception is higher in casual partners than in main partners.¹⁹

Relationship types also make a difference in the emotional experiences of adolescent relationships. More specifically, it appears that one's experience of the relationship (i.e. desire, wanting, and commitment) differ by relationship type (i.e. friends with benefits (FWB), serious, hook-ups, and dating). One study found that three contextual variables (desire, wanting, and pleasure) were highest among FWB and lowest among hook-ups.²⁰ In another study, adolescents rated aspects of emotional intimacy, shared history and shared commitment (e.g., caring about each other or their feelings about personal matters, lasting a long time, can talk about everything, can talk about sex and condoms, and doing things together) as more important in steady relationships than in other types of relationships (i.e., casual, friends, and one-night stands). Further, these same characteristics were more important in casual and FWB than one-night stands.¹⁴ Finally, Plichta ⁸ found that nonexclusive partners were less likely to be considered as someone whose feelings were important as compared to exclusive partners.

Yet, despite the overwhelming evidence that relationship types matter, our understanding of these issues is still limited (note the variety of terms used in the previous two paragraphs), since there are methodological challenges in this literature. Overall, there are four key methodological challenges that occur across this literature, including: a) inconsistent definitional frameworks; b) lack of adolescent voices; c) lack of consideration of gender; and d) poor differentiation between individual and relationship risk.

Given these methodological issues, understanding the results across studies is difficult. While it is not recommended that all studies be methodologically identical, a lack of cohesiveness renders it difficult to form a consensus regarding the effects of relationship type on sexual behaviors. For example, it is unclear whether there are meaningful differences between casual partners and FWB. One study separated out “main/steady”, “casual”, and “one-night stands” but ultimately in the analysis, they were all meaningfully different than steady partners but not different than each other, i.e., there were no discriminating differences between “casual” and “one-night stands.”¹⁴ Until there is a consistent way of approaching relationship types in adolescents, we will not be able to tell which types of relationships are meaningful and to what extent we should “lump” types together or “split” them up. Once there is a better understanding of the issues, more consistent and standardized ways of measuring relationship types can be developed; thereby allowing us to fully maximize our understanding of the impact of relationship type and utilize it appropriately to develop interventions. Thus, we have reviewed the methodological issues in the hopes that future research will be sensitive to these issues.

Inconsistent definitional frameworks

Studies vary in terms of the “labels,” number of categories, and the variables used to define relationships. Most studies divide relationships into two categories, but use different labels. For example, studies may label partners as main or steady; whereas, others may label partners as secondary or casual or other.^{15,17-19, 21, 22} Other studies have used the terms romantic versus non-romantic.^{11, 13,23} However, it is not clear that these labels refer to the same relationships. For example, one adolescent's idea of what “main” means may not be consistent with another adolescent's classification, and a “main” partner may not have the same connotation to an adolescent as a “steady” one? Similarly, would all secondary relationships be considered “casual” or “non-romantic”? In addition, few studies have examined whether those types have the same meaning as adolescents mature or gain more experience. For example, is the casual partner of a 14 year old, the same as the casual partner of an 18 year old?

Some studies have included more than two types (e.g., steady, casual, friends, and one-night stands),^12,14,20 but even then studies do not use the same number of categories or the same levels. Another difficulty is that in research, adolescents may only be queried about further relationships if they indicate having the first type of relationship. For example, in one study adolescents only were queried about casual relationships if they listed a main relationship;⁹ thus, it is possible that relationships were not completely captured.

In addition to differences in number of categories, there are differences in the key aspects by which relationships are defined. Sometimes researchers define relationships in terms of sexual behavior (e.g., having sexual intercourse with only one person in the last 6 months is defined as “exclusive”) ⁸ or by perception/meaning (when you like someone and s/he likes you back) ²⁴ or a combination (After the main partner, is there anyone you have sexual intercourse with more often than others?).⁹ Few studies focus on defining relationships by contextual variables (e.g., satisfaction, commitment) although this may be influential for the likelihood of engaging in risk behaviors. In addition, these contextual variables may vary within the same type of relationship, i.e., some FWB may be relationships with high trust and commitment while others may not. Thus, there may be an interaction which warrants further exploration.

Lack of voices of adolescents

In nearly all studies in this body of literature, researchers have developed the number and definition of relationship types a priori and without data from adolescents. Only one study actually conducted a pilot study to have adolescents develop the different categories of relationships. Those relationships were then used in two larger studies.^{14, 19} It is possible that there are relationships that are important to adolescents which are not being captured by adult perceptions. In our recent study, we found that adolescents labeled sexual partnerships in ways that often have not been captured such as “baby's father” or an ex-boyfriend. We also noted in our interviews with adolescents that there seemed to be a difference between a “sexual partner,” a partnership in which the sexual behaviors are primary and a “friend with benefits” a partnership in which the emotional relationship seemed primary.²⁵ It is unknown whether these differences lead to different levels of risk, but that could be tested empirically.

Lack of consideration of gender

Some studies include both male and female participants, but many studies focus on only females.^{8, 10, 12, 17, 20-22} However, a few studies examined differences by gender in the prevalence of different types of relationships. For example, females were more likely to report partners as romantic, main, or steady as compared to males. ^{13, 23} One study found that males were more likely to have a non-steady partner than females.²⁶ In another study, women were more likely to receive “booty calls” than men. ²⁷ On the other hand, one study found no gender differences across frequency of partner types⁹, and another study found no gender differences in the frequency of FWB relationships.²⁸ There is only one study which examined whether adolescents viewed the same “type” differently based on gender. In this study, females rated “doing things together” and “can talk about anything” to be less important in one-night stands than males did, and males rated these attributes as less important for casual relationships and FWB than females. ¹⁴

Poor differentiation of individual versus relationship risk

Although many studies find an association between relationship type and risk behaviors, they have not done an adequate job of differentiating whether the increased risk truly comes from the context of that relationship, or from individual characteristics that are associated both with having that type of relationship and engaging in risk behaviors. For example, careful analysis of the association between substance use and sexual risk behaviors suggests that adolescents who use substances are also likely to engage in high risk sex, rather than adolescents engage in more risky sexual behavior only when they are drunk.²⁹ For relationship type, the associations are more complicated as some riskier behaviors increase in more steady, stable relationships (i.e., lack of contraception), and in general some types of relationships are considered to be risky just by their nature (e.g., one-night stands).

Next Steps

In order to increase communication among health care providers and researchers and between professionals and adolescents, there are immediate and long-term steps that can be taken. Immediately, clinicians and researchers can be more sensitive to potential confusion of terms, and “check in” when collecting information or asking for information. Thus, if the adolescent generates the type, the clinician could ask for clarification, or if the researcher provides the label, they should provide the adolescent a definition. However, long term, a comprehensive definition or measures of relationship types needs to be developed. These measures need to be based on the “voices” and experiences of both male and female adolescents and designed to create a classification system, which could be tested with a variety of populations. This will require studies, in which the samples are large enough to have the potential power, to detect differences between more than two levels of relationship type and to allow for the assessment of differences in gender.

Conclusions

The initiation of intimate, romantic relationships is a key developmental task during adolescence. Although most adolescents master this well, an unacceptable number of them experience negative outcomes despite the investment in research and programs that attempt to mitigate these outcomes. In reviewing the literature addressing the association between type of relationship and risk, there exists a clear association, yet the interpretation of it is rather unclear, in large part due to methodological challenges. A consistent classification system is needed in order to meaningfully compare and contrast results across studies and enumerate changes across age and developmental experience.

These advances, in turn, may help in the development of interventions that better support adolescents as they explore intimate relationships and sexuality, while engaging in behaviors that reduce their risk of untoward consequences. Further, understanding the relative contribution of the individual factors associated with choosing to have certain types of relationships and differences in the behaviors of the same individual across relationships will help develop interventions.