Effect of Primary Care Parent-Targeted Interventions on Parent-Adolescent Communication About Sexual Behavior and Alcohol Use

This randomized clinical trial evaluates the efficacy, feasibility, and acceptability of a parent-targeted intervention for parent-adolescent communication about sexual health and alcohol use.


Introduction
Engaging in sexual behaviors and drinking alcohol during the second decade of life is common. [1][2][3][4] Community-, school-, and home-based interventions involving direct contact between staff and parents or caregivers can favorably affect parent-adolescent communication (PAC) and a wide range of adolescent risk-associated behaviors. 4,5 Parents can influence adolescents' risk of unwanted pregnancy, sexually transmitted infection, and alcohol-related injury that cause substantial morbidity and mortality among adolescents and young adults. [5][6][7][8] Enthusiasm about these interventions must be balanced with acknowledgment that such programs can be difficult to scale and sustain.
Delivering effective PAC interventions in conjunction with annual adolescent well care visits in primary care clinics could provide a recurring mechanism to systematically reach sizable portions of adolescents and their parents or guardians. 9 This would also align with goals of patient-centered and family-centered care because adolescents and parents or guardians are interested in receiving information from primary care clinicians to facilitate increased PAC about a variety of issues, including sexual health and alcohol use. 10,11 To our knowledge, few primary care clinic-based PAC interventions exist. 12,13 We tested the efficacy, feasibility, and acceptability of delivering parent-targeted interventions in a primary care pediatric practice for PAC about sexual health or alcohol use. Our study was not sufficiently large or long enough to test influences of PAC on adolescent behaviors, but we adapted and tested interventions shown to improve both PAC and behavior (ie, initiation of first sexual intercourse, condom use, 14 and alcohol-associated injury 15 ) in other settings. Another study arm involved older adolescents and focused on PAC about safe driving; that study is reported elsewhere. 16 The study protocol (Supplement 1) was approved by the Children's Hospital of Philadelphia Institutional Review Board. Parents provided verbal informed consent, and adolescents provided verbal assent. This study is reported following the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline. community based and does not include trainees, and clinician salaries are linked to patient volume.

Overview
Parents were mailed an introductory letter and invited to contact the study team; telephone calls were placed to all parents who did not contact the study team. Families interested in participating were screened for eligibility. The Figure presents a diagram of participant recruitment. To be eligible, adults had to be the parent or legal guardian of the adolescent scheduled for the well care visit, planning to attend the appointment, and fluent in written and spoken English. For adolescents to be eligible, they had to be aged 14 to 15 years at the well care visit, an established practice patient, fluent in written and spoken English, able to complete study procedures, and not pregnant.

Procedures
Eligible parent-adolescent dyads were enrolled after providing informed parental consent and adolescent assent, and they each privately completed a telephone survey before the well care visit.
The dyads were randomized into 1 of 3 groups: sexual health intervention group, alcohol prevention intervention group, or control group. Randomization was performed using a computer-generated random listing of the arms using a prespecified seed. Intervention group parents were instructed to arrive at the clinic 15 minutes early, taken to a quiet area in the waiting room away from their adolescent, and given intervention materials based on group assignment (ie, sexual health or alcohol prevention). Parents later joined their adolescent in the examination room. Parents in the control group received usual care.
Two weeks after the well care visit, intervention group parents were contacted via telephone call. Research assistants (RAs) covered material that was not completed during the in-person health coaching visit, if applicable, and administered a short survey. If parents had not had a conversation with their adolescent about the intervention topic in the preceding 2 weeks, barriers were identified and RAs discussed how the materials might be used to overcome them.

Interventions
Intervention materials were adapted from parent-targeted interventions shown to influence PAC and adolescent sexual behavior or alcohol use. Interventions selected were informed by extensive theoretically grounded research identifying parental attitudes and beliefs affecting communication with adolescents about sexual health or alcohol use and adolescent attitudes and beliefs affecting sexual behaviors or alcohol use. 14,15,17,18 Further, selected interventions could be pragmatically adapted for use in busy primary care settings. The intervention process included a health coach discussing written materials with parents in the clinic lobby during a well care visit, conveying key messages, and encouraging PAC within 2 weeks (Box). This was followed by a brief direct verbal and written endorsement of the intervention from the adolescent's clinician and a 2-week follow-up telephone call from the health coach to the parent. Health coaches were college graduates without specific health care training.
Written materials were adapted to parents of our targeted age group, provided local data on sexual behaviors or alcohol use, and showed visual images reflecting the racial/ethnic composition of

Sexual Health Intervention Clinic Intervention
Health coach talks with parent in lobby to discuss: Sexual health brochure and handbook a (provided in written form, also available in electronic form) General communication handbook (provided in written form, also available in electronic form) Physician or nurse practitioner provides a direct endorsement and written prescription reinforcing the key messages in the clinic room with the parent and adolescent at the end of the visit.

Follow-up Telephone Call
Follow-up telephone call to the parent at 2 weeks to ask about PAC about sexual health, inquire about barriers, and review how materials can be used to overcome barriers.

Alcohol Prevention Intervention Clinic Intervention
Health coach talks with parent in lobby to discuss: Alcohol prevention brochure and handbook b (provided in written form, also available in electronic form) General communication handbook (provided in written form, also available in electronic form)  the local community and of gender-nonconforming adolescents and parents. Materials included colorful spiral-bound handbooks with discussion guides and activity sheets on general PAC and a similarly formatted handbook and brochure focused on PAC about sexual health or alcohol use prevention; written materials were also available electronically.

Measures Sociodemographic Characteristics
Adolescents' age was calculated based on date of birth; sex at birth, race, and ethnicity were measured by self-report. Parental age, marital status, and highest level of education were measured by self-report.

Adolescent Behaviors
Adolescent behaviors were measured by self-report at baseline. Sexual behavior was assessed by 4 items: "In your lifetime, have you ever engaged in vaginal sexual intercourse? …had anal sexual intercourse? …given oral sex? …received oral sex?" Alcohol use was assessed with the question, "Have you ever had a drink of alcohol, other than a few sips?" Adolescents were coded as having engaged in adolescent risk behavior if they responded yes to any of these items.

Communication
There

Intervention Feasibility and Acceptability
Feasibility was assessed by measuring length of time the RA spent with the parent (RA report) and checklists to document content delivered (ie, RA report, parent report at 2-week follow-up, and clinician report). Acceptability was assessed by final surveys. Parents were asked about general helpfulness and intentions to use materials over the next 12 months (ie, use with the adolescent participant, use with other children, or give to other parents). Adolescents were asked how helpful they thought the materials were to their parent. Clinicians received an anonymous web-based survey asking open-ended questions about implementing the study in their clinic and using these interventions in real life if they were found to be effective.

Statistical Analysis
We tested for differences in sociodemographic characteristics between participants who completed the follow-up survey and those who were lost to follow-up. We examined frequencies of parent-and adolescent-reported PAC and evaluated bivariate associations with adolescent sex, age, race/ethnicity, and risk behaviors using analysis of variance. Unadjusted and adjusted models were estimated. Covariates in the adjusted models were adolescent age, sex, race/ethnicity, and sexual behavior or alcohol use.
Generalized linear models were conducted to compare differences between each intervention group vs the control group in quality of PAC and frequency of PAC about sex or alcohol. Depending on the distribution of the outcome variable being analyzed, different distributions and link functions were specified for the generalized linear models: linear to estimate outcome means, bivariate and adjusted, with 95% CIs or log-binomial to estimate unadjusted and adjusted risk ratios (RRs) with 95% CIs. We conducted similar analyses specifically for safety plan PAC content.

JAMA Network Open | Pediatrics
Analyses were performed separately for adolescents and parents. Data were analyzed using an intent-to-treat principle and multiple imputation. To account for missing outcomes, multiple imputation with 100 imputations was used, analyses were conducted on each of the 100 imputed data set, and the parameter estimates across the data sets were combined to produce a unique point estimate and SE taking into account the uncertainty of the imputation process. Statistical analysis was performed using SAS statistical software version 14.2 (SAS Institute). Our sample size of 40 participants per group had 80% power to detect a difference in 2 population means corresponding to a Cohen d of 0.625, using a 2-sided P value of less than .05 for statistical significance. Further details on the trial protocol and statistical analysis can be found in Supplement 1.

Sample
A total of 118 parents and 118 adolescents participated. Among parents, 112 were women (94.9%), and the mean (SD) age was 45.8 (6.9) years ( Sociodemographic and behavior characteristics were similar between dyads who completed and those who did not complete the study.

Bivariate Associations Between Demographic Characteristics and Communication
Responses to quality of PAC were summed into an index ranging from 41 to 96 for parents (α = .84) and 43 to 96 for adolescents (α = .87). Adolescent-reported quality of PAC varied by adolescent age; younger adolescents reported a significantly higher mean (SD) score for quality of PAC than older

Intervention Influence on Reported Qualtiy of PAC and Topic-Specific PAC
Neither intervention influenced parent-reported quality of PAC, frequency of PAC about sex, or frequency of PAC about alcohol at the 4-month follow-up survey (

Feasibility and Acceptability
Parents spent a median of 10 (range, 7-24) minutes in coaching sessions. Among 78 parents in the 2 intervention groups, 75 (96%) received an entire health coaching session in the clinic and 74 (95%) reported receiving an in-person clinician endorsement, which was consistent with clinician reports.
We were able to contact 66 intervention parents (85%) for the 2-week follow-up call.
Sixty-two parents in the intervention groups rated the intervention materials as moderately or very helpful (80%), 66 parents (85%) reported they would probably or definitely refer to materials  approximately 2-fold as many parents and adolescents in the intervention groups reported increased PAC about strategies for adolescents to avoid getting into a car with someone who has been drinking alcohol, although this specific content was only included in alcohol prevention intervention materials. This finding suggests that there may be broader beneficial effects of topic-specific parenttargeted interventions on PAC, which warrants further study.
It is important to note that we found these interventions feasible within the context of a research study, which supplied health coach staffing resources to provide the interventions as well as the 2-week follow-up telephone call to parents. Clinicians reported that resources for health coach staffing would be needed to implement similar strategies in real-life conditions. Our health coaches were trained in the use of materials and protocols but were not trained in health care, suggesting that it may be possible to consider a range of staffing models. Our study was intentionally designed to Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: This information or content and conclusions are those of the authors and should not be construed as the official position or policy of, nor should any endorsements be inferred by HRSA, HHS or the US government. Sexual health and alcohol prevention intervention materials were modified with permission from Dr Jaccard.