Cluster analysis of caregiver and adolescent emotion regulation and its relation to sexual health and dating communication

Abstract

Objective

Adolescent emotion regulation (ER) has been positively linked to caregiver–adolescent sexual communication. With ER becoming increasingly conceptualized as an interpersonal process, it is likely that both adolescent and caregiver ER impact communication patterns to some extent; thus, each must be accounted for in scientific inquiry and intervention approaches. We aimed to identify distinct profiles of caregiver and adolescent ER and examine how each profile is differentially related to caregiver–adolescent communication about sexual health and relationships.

Methods

Participants included adolescent males (n = 117; M_age = 13.06; SD = 0.72) and their caregivers (M_age = 42.33; SD = 0.72) who were recruited as part of a dyadic, web-based dating violence prevention intervention trial for middle school boys.

Results

Using a self-report measure, three clusters were identified: families with Moderate ER strategy use (by both parents and adolescents), families with Low ER strategy use (by both parents and adolescents), and families with Mixed ER strategy use (moderate adolescent but low caregiver ER strategy use). Caregivers in the Moderate ER strategy use cluster reported discussing the greatest total number of sexual health and relationship topics. Adolescents in the Mixed ER strategy use cluster indicated the greatest perceived caregiver openness during discussions.

Conclusions

Findings suggest that adolescent ER may more strongly influence positive patterns of communication regarding sexual health and relationships than parent ER. Future research should compare the efficacy of adolescent sexual health and dating interventions with adolescents alone versus dyadic interventions to determine whether there is a clinically significant additive effect of including caregivers or if adolescent engagement alone may suffice.

Current trends in adolescent sexual behavior suggest that, despite an overall decline in sexual activity among adolescents in the United States from 2013 to 2017, a considerable amount still report engaging in high-risk sexual behaviors (Witwer et al., 2018). For example, there was evidence of an overall decrease in condom and contraceptive use from 2013 to 2017 (Witwer et al., 2018). This is problematic and may partially explain the high (though declining) rates of unintended pregnancies in the United States, especially among minoritized youth (Brindis et al., 2020), as well as the increasing rates of sexually transmitted infections found among U.S. adolescents (Scott-Sheldon & Chan, 2020). These consequences of high-risk adolescent sexual behaviors can contribute to suboptimal long-term functioning, including an observed overall increased risk of mental health problems among adolescent mothers, especially those from disadvantaged backgrounds (Xavier et al., 2018). Moreover, adolescents who engage in high-risk sexual behaviors are also more likely to engage in various other health risk behaviors (e.g., substance use) that can adversely impact their health (Bozzini et al., 2021; Cho & Yang, 2023). Given that sexual and relationship development is a typical part of adolescence, there is a critical need to better understand the processes that foster its healthy development and those that precede the initiation of high-risk behaviors. Such information can inform intervention and prevention programs aimed at reducing high-risk adolescent sexual behaviors and its correlates across the lifespan.

Research underscores the importance of positive caregiver–adolescent sexual communication in reducing adolescent engagement in high-risk sexual behaviors. For example, enhanced caregiver–adolescent communication regarding sexual health is associated with reduced high-risk sexual behaviors in adolescents, including later sexual initiation, lower frequency of intercourse, and lower rates of unprotected sex and sexually transmitted disease infection, such as HIV (Coakley et al., 2017; Dilorio et al., 2006; Guilamo-Ramos et al., 2011; Murry et al., 2011). Despite this well-established link between caregiver–adolescent communication and decreased high-risk sexual behaviors, caregivers may be reluctant to initiate these conversations due to discomfort in discussing sexual health topics in a developmentally appropriate manner (Jerman & Constantine, 2010; Randolph et al., 2017). This may be especially true when trying to navigate sexual health conversations when adolescents are emotionally dysregulated.

Emotion regulation (ER) is broadly defined as the process through which an individual intentionally modulates the occurrence and intensity of their emotional experience. This process typically includes the identification, selection, and successful implementation of at least one strategy to achieve this self-regulatory goal (Gross, 2015), and it has been positively linked to caregiver–adolescent sexual communication. For example, one study conducted with seventh graders exhibiting mental health symptoms found that when adolescents displayed adaptive ER behaviors more often, parents were more likely to have discussions around sexual health topics (Shuster et al., 2021). Within the same study, the researchers also found that adolescents who endorsed discussing sexual health topics with a caregiver also endorsed using significantly more ER strategies than their counterparts who did not endorse sexual health communication with their caregivers. It may be that caregivers are more reluctant to have sexual health conversations with adolescents if they are unsure of their child’s emotional responses (Afifi et al., 2008). In addition, adolescents with suboptimal ER may avoid having uncomfortable conversations about sex and dating/relationships to evade their own emotional discomfort (Grossman et al., 2018).

Suboptimal ER among caregivers may also contribute to a caregiver’s decisions of whether to broach sexual health and dating topics that are perceived as more difficult to discuss, as caregivers tend to shy away from having direct conversations with their children and often communicate messages through innuendo and at a superficial level (Hyde et al., 2013). Indeed, it has been suggested that caregivers of adolescents with limited access to ER skills may also struggle with ER themselves and thus may be more likely to avoid conversations to prevent emotional discomfort among both parties (Pluhar et al., 2008). This dyadic avoidance due to worries and fears is consistent with the ever-evolving definition of ER, which is becoming increasingly conceptualized as an interpersonal process, whereby one’s emotional experience is impacted by those with whom they engage in reciprocal interactions (Butler, 2017; Niven, 2017). This bidirectional relationship of ER suggests that both caregiver and adolescent ER must be accounted for within scientific inquiry and therapeutic approaches (Campos et al., 2011; Zaki & Williams, 2013). In the context of adolescent sexual health promotion, given that more open communication is associated with safer sex and healthier outcomes for adolescents (Schalet, 2011), research is needed to better characterize unique patterns of ER in caregiver–adolescent dyads. Such research can be used to inform how those patterns are linked to the occurrence of discussions around sexual health and relationships, as well as the overall process that occurs during such conversations.

Person-centered cluster analysis offers a distinctive method of classifying unique patterns of adolescent and caregiver ER that might exist and differentially relate to caregiver–adolescent sexual communication. Particularly, person-centered cluster analysis aims to distinguish groups of people, based on distinct patterns of common characteristics, such that the groups are created in a way that augments within-group homogeneity and between-group heterogeneity. In the context of ER, this method can be used to empirically derive unique profiles of dyadic (adolescent and caregiver) ER; that is, dyads with matched caregiver–adolescent ER strategy use (e.g., both the caregiver and adolescent demonstrate higher or lower ER strategy use) can be distinguished from dyads with disparate use of ER strategies (e.g., only the caregiver uses ER skills frequently but the adolescent does not). These unique dyadic ER profiles may then potentially be used to differentially predict caregiver–adolescent communication about sexual health and relationships.

Though person-centered cluster analyses offer a more robust conceptualization of the dyad that is consistent with evolving conceptualizations of ER within family systems, researchers have yet to leverage the benefits of cluster analysis in this research context. Instead, current literature on sexual health interventions often focuses on enhancing either the parent’s or the adolescent’s emotional regulation skills as the primary target of intervention to improve sexual health communication, rather than the dyad together (Houck et al., 2016; Shuster et al., 2021). Given the importance of caregiver–adolescent communication in preventing and reducing high-risk sexual health and dating/relationship behaviors, as well as the importance of ER in increasing caregiver–adolescent communication, identifying dyadic profiles that may be associated with more optimal communication patterns is essential.

This study sought to address this critical gap in the literature by using person-centered cluster analyses to identify profiles of caregiver and adolescent ER that may uniquely relate to communication about sexual health and relationships within these dyads. Specifically, the purpose of the current study was (1) to identify distinct profiles of caregiver and adolescent ER in families of adolescent boys, and (2) to examine how each profile is differentially related to caregiver–adolescent communication pertaining to topics related to sex and relationships. Given the dearth of research on the relationship of adolescent and caregiver ER in the context of adolescent sexual health, no a priori hypotheses were made. Instead, the study aims were exploratory and intended to inform whether future research assessing dyadic ER in this context might be clinically informative.

Methods

Participants

Participants (N = 119 adolescents and their caregivers) were recruited as part of a pilot trial examining a dyadic, web-based dating violence preventive intervention for young adolescent males (Rizzo et al., 2021). Due to missing data for our key study variables (caregiver and adolescent ER), data from a final sample of 117 adolescents and their caregivers are used in the present study. Participants were recruited from middle schools in the Providence, RI area between June 2015 and November 2017. Eligibility included (1) identifying as male, (2) being enrolled in the seventh or eighth grade, and (3) speaking English (as the intervention was only available in English at the time).

Adolescent males were an average of 13.06 years old (SD = 0.72); caregivers were an average age of 42.33 years old (SD = 6.84). Most caregivers identified as female (n = 105; 90%) and the biological mother (n = 101; 86%) of the adolescent male. Fifty percent (n = 58) of the adolescents identified as White, 16% (n = 19) identified as Black/African American/Haitian, 15% (n = 18) identified as multiracial, 2% (n = 2) identified as Asian, and 1% (n = 1) identified as American Indian/Alaskan Native. Notably, 16% (n = 19) indicated their racial background did not fall into any of the prior categories. Seventy-one percent (n = 83) of the caregivers identified as White, 10% (n = 12) identified as Black/African American/Haitian, 3% (n = 3) identified as multiracial, 2% (n = 2) identified as Asian, 1% (n = 1) identified as American Indian/Alaskan Native, and 1% (n = 1) identified as Native Hawaiian/Pacific Islander. Twelve percent (n = 14) indicated their racial background did not fall into any of the prior categories. Latino ethnicity was endorsed by 22% (n = 26) of the adolescents and 18% (n = 21) of the caregivers. Ninety-eight percent (n = 115) of adolescents reported having a mother figure at home; 65% (n = 77) endorsed having a father figure in the home. Regarding socioeconomic status, 49% (n = 58) of adolescents reported qualifying for free or reduced lunch and 27% (n = 31) of parents reported a family income under $30,000. Last, 18% (n = 21) of adolescents were receiving psychological treatment at the time of the study and 44% (n = 54) had sought psychological treatment in the past per caregiver report.

Procedures

The current study is a secondary analysis of data from a clinical trial aimed at teaching ER and communication strategies to caregiver–adolescent dyads to reduce dating violence behaviors in adolescent boys (see Rizzo et al., 2021 for the primary study analyses). All procedures were approved by the Rhode Island Hospital Institutional Review Board (Project number 681701). Project staff presented the nature of the study to students in classroom presentations and student assemblies, where consent to contact forms were distributed. Principals also emailed parents with a link to this form and study staff attended school events (e.g., open houses) to recruit. After receiving permission to contact, study staff met with interested families to describe the project and obtain informed consent and assent. See Rizzo et al., 2021 for information regarding participant flow. The current study used data from the baseline assessment, for which families were compensated $30 after completing audio computer-assisted structured interviews on laptop computers provided by study staff during meetings at locations convenient to the family, including their home, school, or local libraries.

Measures

Demographics questionnaires were administered to both adolescents and caregivers, which included questions pertaining to participant age, race (checkboxes with the following options: American Indian/Alaskan Native, Asian, Black/African American/Haitian, Native Hawaiian/Pacific Islander, White, Other/Not Listed), and ethnicity (Hispanic/Latino vs non-Hispanic/Latino). Additional data were gathered regarding multiple indices of socioeconomic status (household income per caregiver report and free or reduced lunch per adolescent-report), family composition (having a mother and/or father figure in the home per adolescent-report), and whether the adolescent had a history of psychological treatment (per caregiver report).

Emotion Regulation Behavior Scale (ERBS; Houck et al., 2016) is an eight-item self-report scale measuring use of the three types of ER strategies taught in the trial. The strategies assessed were selected based on prior qualitative work that characterized the most commonly used ER strategies as reported by adolescents (Lansing et al., 2019). Importantly, the strategies mapped onto four of the five “families” of ER strategies outlined in Gross’s original ER process model (Gross, 2015): “Get Out” (i.e., removing oneself, either physically or cognitively, from emotional triggers), “Let It Out” (i.e., adaptive verbal or physical expression of an emotional experience), and “Think It Out” (i.e., use of cognitive restructuring or other manipulation of one’s mental appraisals of an emotional trigger). Adolescents and their caregivers completed the ERBS about themselves. Participants were instructed to rate how often they used the ER techniques (i.e., “get out,” “let it out,” and “think it out”) in the last week to calm strong feelings (“In the last week, when you were having a strong feeling [e.g., really mad, really sad, or really excited] how often did you… get away from whatever was causing your feeling? …take deep breaths to calm down? …think about the situation differently?”). Responses were rated on a five-point Likert scale from 1 = all the time to 5 = never, reverse coded, and summed and averaged to create a total score, with higher scores indicating more frequent use of the ER strategies. The ERBS total score demonstrated adequate internal consistency for the present study (α = 0.82 and α = 0.70 for caregivers). The measure also showed sensitivity to intervention in the current study, as a moderate effect size (0.32) was observed over 9-month follow-up (Rizzo et al., 2021). The ERBS also detected similar change over time (d = 0.30) in a separate study teaching these ER skills to early adolescents (Houck et al., 2016).

The Sexual Communication Scale (SCS; Miller et al., 1998) was adapted from the Miller SCS and was completed by both caregivers and sons. The six-item Open Sexual Communication subscale was used to assess openness of sexual communication between parents and adolescents (e.g., “My parent or caregiver wants to know my questions about topics regarding sex” and “My parent/caregiver and I talk openly and freely about topics regarding sex” [adolescent version] and “I want to know my daughter/son’s questions about topics regarding sex” and “My teen and I talk openly and freely about topics regarding sex” [parent version]; Miller et al., 1998). The six items were scored on a 1 (very untrue) to 7 (very true) scale and then summed together and averaged, with higher scores representing greater perceived caregiver openness and ability to discuss sex topics. The subscale demonstrated adequate internal consistency for the present study per adolescent report (α = 0.78) and caregiver report (α = 0.74).

Next, two items from the original six item Parent–Adolescent Sexual Topic Discussion subscale were used to assess the content of sexual communication between parents and adolescents. To capture the emphasis of content in the intervention, the remaining sexual health items were replaced with five items that were added to explore topics explicitly discussed in the intervention (how to handle problems in relationships, how to manage emotions, appropriate behavior when communicating through technology [text, social media], values with romantic relationships and sex, and overall sexual health [using condoms, diseases, sexual behaviors]). Respondents indicated whether or not they discussed the topic with their parent/adolescent (yes/no; e.g., Have you and your parent/caregiver EVER talked about when it is appropriate to start having sex?). For the purposes of the current study, a sum score of the number of topics discussed (seven items total) was created.

Data analysis plan

Data were examined for normality and outliers prior to all analyses, and all assumptions were met. All statistical analyses were performed using the Statistical Package for the Social Sciences for MacOS. Person-centered analysis was conducted through a hierarchical cluster analysis using Ward’s method and a dendrogram to visualize and determine the number of clusters in the sample based on caregiver and adolescent ER (i.e., ERBS scores). Next, a k-mean’s cluster analysis and one-way ANOVA were conducted to assign cluster membership to families; clusters were then described and labeled based on these results. Based on the possible range of the ERBS total score (1-5) and average scores found within the full sample, average scores of 1.00-2.00 (“never” to “once in a while”) were labeled as low ER strategy use, 2.01-4.00 (“once in a while” to “most of the time”) were labeled as moderate ER strategy use, and 4.01-5.00 (“most of the time” to “all the time”) were labeled as high ER strategy use. Post-hoc t-tests were used to further characterize the clusters based on caregiver and adolescent ER. Last, one-way ANOVAs and post-hoc t-tests were used to characterize significant differences in caregiver- and adolescent-reported total number of sexual health and relationship topics discussed and caregiver openness during such discussions.

Transparency and openness

Data, analysis code, and research materials are available upon request. The current study’s analysis was not pre-registered, as it was a secondary analysis of data from a pilot clinical trial.

Results

Hierarchical and k-means cluster analysis indicated three clusters in the sample. The first cluster (n = 37), labeled as Moderate ER, was characterized by moderate self-reported frequency of using ER strategies by both adolescents and caregivers. The second cluster (n = 47), labeled as Low ER, was characterized by less frequent self-reported ER strategy use for both adolescents and caregivers. The third cluster (n = 33), labeled as Mixed ER, was characterized by moderate ERBS scores for adolescents but less frequent use for caregivers. There were no differences in demographics (i.e., caregiver age, family income, adolescent race, caregiver race, adolescent ethnicity, and caregiver ethnicity) by cluster, with the exception of adolescent age. Specifically, the Low ER cluster was comprised of mainly 13- and 14-year-old boys, whereas the Moderate and Mixed ER clusters included relatively younger boys (i.e., mainly 12- and 13-year-old boys). See Table 1 for the mean and standard deviations of caregiver and adolescent ER and demographic breakdown for each cluster. Further supporting the identified clusters as being distinct profiles, one-way ANOVAs yielded significant differences in adolescent ERBS scores (F(2, 114) = 93.53, p < .001, partial η² = 0.62) and caregiver ERBS scores (F(2, 114) = 54.59, p < .001, partial η² 0 = 0.49) across clusters. Post-hoc t-tests indicated that the Moderate ER cluster reported having higher caregiver (t(77.73) = 9.40, p < .001, Cohen’s d = 0.49) and adolescent (t(81.24) = 7.67, p < .001, Cohen’s d = 0.51) ERBS scores than adolescents and caregivers in the Low ER cluster. The Moderate ER cluster also reported having more frequent caregiver (t(66.11) = 8.20, p < .001, Cohen’s d = 0.44), but less frequent adolescent (t(61.63) = −6.24, p < .001, Cohen’s d = 0.52) use of ER strategies than the Mixed ER cluster. The Mixed ER cluster reported having more frequent adolescent ER strategy use (t(66.11) = −12.79, p < .001, Cohen’s d = 0.55) than the Low ER cluster, but similar frequency of ER strategy use for caregivers (t(77.71) = −1.79, p > .05, Cohen’s d = 0.44).

One-way ANOVAs demonstrated that cluster membership was significantly associated with caregiver reports of the number of sexual health and relationship topics discussed (F(2, 114) = 3.13, p = .047, partial η² = 0.05) and adolescent-reported caregiver openness of sexual communication (F(2, 114) = 3.46, p = .035, partial η² = 0.06; see Table 2). No significant differences emerged in relation to parent-reported caregiver openness of sexual communication or adolescent report of the number of sexual health and relationship topics discussed by cluster (p’s > .05).

Post-hoc t-tests indicated that caregivers in the Moderate ER cluster reported more sexual health and relationship topics discussed with their adolescents than the Low ER cluster (t(74.89) = 2.40, p = .019, Cohen’s d = 0.53); there were no significant differences in number of sexual health topics discussed between the Moderate ER cluster and the Mixed ER cluster (t(67.86) = 0.77, p > .05, Cohen’s d = 0.19). There were no significant differences in caregiver-reported topics discussed when comparing the Low ER cluster to the Mixed ER cluster (t(69.30) = −1.58, p > .05, Cohen’s d = −0.36). Additionally, the Mixed ER cluster was reported as having higher adolescent-reported caregiver openness of sexual communication than the Low ER cluster (t(57.33) = −2.33, p = .023, Cohen’s d = −0.55), but not the Moderate ER cluster (t(54.93) = −1.14, p > .05, Cohen’s d = −0.28). See Table 2 for descriptive statistics. Interestingly, there were no differences in adolescent-reported caregiver openness of sexual communication when comparing the Low ER cluster to the Moderate ER cluster (t(81.15) = 1.58, p > .05, Cohen’s d = 0.34).

Discussion

Within the current study, we explored profiles of ER strategy use by caregivers and adolescents, as well as how each distinct profile was uniquely related to caregiver–adolescent sexual health and relationship communication. Three clusters were identified within the sample, labeled as Moderate ER (moderate caregiver- and adolescent-reported frequency of use of ER strategies), Low ER (low caregiver- and adolescent-reported frequency of use of ER strategies), and Mixed ER (moderate adolescent-reported frequency of use of ER strategies but low frequency of use of ER strategies for caregivers). Interestingly, no cluster emerged in which parents reported moderate use of ER strategies while their adolescents reported low use. This may suggest that parents who use ER strategies frequently are likely to have adolescents who will do the same to some extent.

Notably, there were no significant differences in total number of sexual health and relationship topics discussed or caregiver openness of sexual communication between the Moderate and Mixed ER clusters. However, findings indicated that the Moderate ER cluster demonstrated the highest average caregiver ER strategy use, whereas the Mixed ER cluster demonstrated the highest use among adolescents. Caregivers in the Moderate ER cluster reported the greatest total number of sexual health and relationship topics discussed between caregivers and adolescents, and the Mixed ER cluster demonstrated the greatest adolescent-reported caregiver openness of sexual communication. In each case, participants reported about one additional topic being discussed relative to their counterparts in the Low ER cluster, which may be clinically relevant in the context of adolescent decision-making and health. Study findings may have important clinical and research implications in sexual health promotion and prevention of dating violence in adolescence.

In this sample, three distinct patterns of ER strategy use within families emerged, supporting the use of a person-centered approach toward identifying relative strengths and weaknesses in ER within families. When measuring ER, researchers sometimes use only caregiver or adolescent report to index overall family functioning; however, this neglects to center potential differences in ER within the family system and across families. Developmentally, it is expected that adolescents will struggle with ER, especially in early adolescence, but that ER will consistently improve across adolescence and into adulthood (McRae et al., 2012). However, there is wide variability in ER across persons and across the lifespan, especially when examining specific subsets of ER strategies (Zimmermann & Iwanski, 2014). Accordingly, it may be beneficial to gather information about ER directly from all parties involved in an interaction of interest, in this case being discussions about sexual health and relationships. Through doing so, we can garner a more nuanced understanding of distinct patterns of ER within the family, which may yield important implications for treatment planning and intervention optimization.

Moreover, the three clusters were further distinguished based on their relationship with sexual health and relationship communication, whereby the two clusters with moderate adolescent ER strategy use reported discussing the greatest range of topics and greater perceived caregiver openness during these conversations. First, per parent-report, the Moderate ER strategy use cluster discussed the greatest range of sexual health and relationships topics, though the Mixed ER strategy use cluster reported discussing a similar number of topics. There are several potential reasons for this relationship. In middle school, discussing sex and romantic relationships can serve as a significant stressor for both adolescents and their caregivers (e.g., Afifi et al., 2008). It is possible that adolescents using ER strategies more frequently feel more capable of (and are thus more willing to attempt) tolerating the distress that comes with this conversation, naturally reinforcing caregivers to continue the conversation and cover more topics in ongoing discussions. The Moderate ER cluster was also the cluster with the highest caregiver-reported ER strategy use. It may be that adolescents who observe their caregivers using strategies may have greater confidence that their caregiver can tolerate these discussions and thus engage in these conversations. This healthy dyadic ER pattern then reduces the likelihood that either partner will abruptly end a discussion. Whereas, if there is a mismatch in ER or both dyad members demonstrate limited access to ER strategies, family members may avoid discussing other sex and relationship topics in the future, thereby increasing risk for adolescent engagement in high-risk sexual behaviors and their associated adverse outcomes (e.g., contracting STIs and having unintended pregnancies). This co-regulation of ER is outlined in temporal interpersonal emotion systems theory and can occur both intentionally and more automatically (Butler, 2017).

At the same time, there were no significant differences in the range of topics discussed between the Moderate and Mixed ER strategy use clusters. Given both clusters demonstrated relative strengths in adolescent ER strategy use, especially when compared to the Low ER strategy use cluster, this finding suggests that, as long as an adolescent perceives themselves as being well-regulated, the dyad is more likely to cover a broader range of topics. Combined, the findings support that interventions must continue to target ER as a means of promoting healthy patterns of communication within family systems, particularly in facilitating important sexual health and relationship discussions during early adolescence.

Adolescents in the Mixed ER cluster reported the greatest perceived caregiver openness about sex and relationship discussions. Here, adolescents reported moderate ER strategy use, while caregivers reported low use. Interestingly, this cluster reported the most openness, even more so than the cluster with moderate ER reported by both dyad members, though the difference between the Moderate and Mixed ER clusters was non-significant but had a small-to-moderate effect size. Since this cluster demonstrated the highest use of ER strategies for adolescents among all clusters, the findings suggest adolescent ER may serve a stronger role than caregiver ER in facilitating greater caregiver openness during discussions around sex, at least from the lens of the adolescent. Additionally, research has found that adolescents may avoid discussions about sex with caregivers due to fear of feeling awkward or receiving a negative reaction (Grossman et al., 2018). However, our finding suggests that adolescents who perceive themselves as being able to implement ER strategies frequently may negate those fears and approach these conversations because they perceive that their caregivers will approach the conversations with an open mind, regardless of their caregivers’ actual use of ER strategies. Accordingly, these findings suggest that, at a minimum, adolescent ER must be an intervention target to promote positive communication about sexual health and relationships with their caregivers, which can be theorized to support more optimal adolescent sexual health in the short term and across the lifespan.

Limitations

The findings of the current study should be interpreted in the context of its limitations. First, the sample was limited to adolescent boys and their caregivers. It is possible that our findings may not generalize when examining these patterns among girls, younger boys, or non-binary/genderqueer youth and their caregivers. These data also may not translate to nontraditional households or adolescents who seek their sexual health and dating knowledge from other sources (e.g., siblings or other extended family members). Relatedly, most caregivers participating in the study were biological mothers of the boys, and data regarding ER among fathers of adolescent boys whose mothers participated in the study were not available. Therefore, findings may not reflect the impact of ER on communication with fathers, who may be an important source of information about sex and relationships for young adolescent boys. For example, it is possible that adolescent boys might experience exacerbated ER difficulties when attempting to discuss sexual health topics with their mothers, whereas this may occur to a lesser degree when attempting these conversations with father figures, resulting in more communication around these topics.

Second, the two key study variables were assessed via self-report questionnaires. It is unknown whether the findings would replicate when using behavioral observations of ER and caregiver–adolescent communication patterns. Relatedly, it is possible that more frequent use of strategies may mean the adolescent and/or caregiver are more emotionally volatile. Behavioral measures can shed light on this question. Third, the data presented in the current study were cross-sectional. It is unknown whether the relationships explored in the current study might evolve over time, especially across adolescence as the parent–adolescent relationship naturally changes. Fourth, while there were data collected for standard demographic variables, some additional contextual variables that might have impacted the key study relationships were missing (e.g., medical/chronic illness diagnosis status, mental health diagnosis). Fifth, the measure of ER used in the current study indexed caregiver- and adolescent-reported use of a range of ER strategies. Clusters were identified based on average strategy use and did not examine differences based on the types of strategies used or differences based on the number or range of strategies preferred. Therefore, our findings lack nuance regarding how these features of ER may be differentially related to communication about sexual health and relationships among caregiver–adolescent dyads. Sixth, within the sample, there was no cluster that emerged that was characterized by high/moderate caregiver ER and low adolescent ER, which limits the ability to fully understand the role of caregiver ER in shaping communication patterns within the dyads. Particularly, it is possible that caregivers who demonstrate a relative strength in ER could overcome adolescent ER problems and contribute to optimal sexual health and dating communication. However, the lack of this fourth cluster may also reflect the relative infrequence of this familial pattern. Last, while many of our findings were statistically significant, it should not be assumed these findings are clinically significant, especially considering the current study did not examine the full mechanistic pathway linking caregiver–adolescent ER to adolescent sexual health behavior via sexual health and dating communication.

Conclusions and future directions

The findings of the study yield important preliminary implications for clinical work and future research. First, they highlight that families indeed demonstrate differing profiles of how often individuals use various ER strategies. Accordingly, adolescent ER research must continue to take a person-centered approach to account for such variability. In the context of clinical interventions that target adolescent ER, it may be useful to take an individualized approach to identify these strengths and weaknesses, then match treatment based on those profiles (i.e., if both the caregiver and adolescent demonstrate difficulties with ER, they could both benefit from treatment; whereas if only one dyad member does, they might receive the treatment alone to reduce patient burden). Through doing so, we can ensure interventions are optimized to only deliver the necessary “dose” of treatment required to yield clinically significant effects on ER development.

Second, the current study provides preliminary evidence that adolescent ER is more strongly related to positive sexual health and dating communication patterns than caregiver ER. Therefore, while it may be ideal to include caregivers within sexual health interventions, it may be possible to still find positive outcomes if only the adolescent is actively engaged in interventions. Nonetheless, this is speculative, and future research should compare the efficacy of interventions with adolescents alone versus dyadic interventions to determine whether there is a clinically significant additive effect of including caregivers. This has been done in some prior work (e.g., Brown et al., 2014), but continued research in this area is needed, especially in the context of sexual and relationship health promotion. This includes a need to include mothers, fathers, and other common sources of sexual and relationship health advice to better characterize the impact of dyad’s ER on health communication with adolescents, as this has important implications for future adolescent health promotion efforts. Importantly, such work must be done longitudinally to examine whether changes in ER (both individually and dyadically) promote more optimal dyadic communication and prevent high-risk sexual behaviors and unhealthy relationships across adolescence and into young adulthood. This must also include a consideration of the psychosocial factors (e.g., executive functioning, mental health problems, neurodevelopmental disorders, chronic illness status) that might further impede or promote optimal sexual health communication and its associated behavioral outcomes (e.g., sexual behavior initiation, contracting of sexually transmitted infections), especially among adolescents and their caregivers who struggle with ER and may face the greatest risk. This is crucial given high-risk sexual behaviors during adolescence are often linked to engagement in other common adolescent health risk behaviors (e.g., substance use; Bozzini et al., 2021; Cho & Yang, 2023). Accordingly, identifying the transdiagnostic factors that are linked to ER, sexual health communication, high-risk sexual behaviors, and other common adolescent health risk behaviors can inform the future development of a unified protocol to holistically shape broader adolescent health and development. Last, future research should aim to identify ways to disseminate interventions targeting ER and sexual health communication within the context of routine preventative care visits among adolescents, as there is an expressed need for such work per adolescent and caregiver reports.

Author contributions

Caroline Cummings (Conceptualization [lead], Formal analysis [lead], Software [lead], Visualization [lead]), Christie Rizzo (Data curation [lead], Funding acquisition [lead], Investigation [lead], Methodology [lead], Project administration [lead]), Christopher Houck (Conceptualization [lead], Data curation [lead], Funding acquisition [lead], Investigation [lead], Methodology [lead], Project administration [lead])

Funding

This work was supported by the National Institute of Justice (NIJ; 2014-MU-CX-4002) and the National Institute of Mental Health (T32 MH019927).

Conflicts of interest: The authors declare that they have no conflict of interest.

Registration: The authors have no conflicts of interest to report. The authors confirm that the research presented in this article met the ethical guidelines, including adherence to the legal requirements, of the United States of American and received approval from the Institutional Review Board of Rhode Island Hospital.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

References

Author notes