The Effects of a Romantic Relationship Treatment Option for Adults With Autism Spectrum Disorder

Social Impacts and Treatment Options

Adults with ASD present with mild, moderate, and severe social communication impairments and limited interests or repetitive behaviors (American Psychological Association, 2013). Only 2–10% of all adults with ASD report experience with romantic relationships, including dating, cohabitation, and marriage (Eaves & Helena, 2008). Those with mild ASD report better social outcomes than those moderately or severely affected, with 58% reporting current or past intimate partnerships (Howlin et al., 2004).

Even though adults with mild ASD have more opportunities for social relationships, they often experience negative social effects such as isolation, bullying, rejection, and poor-quality relationships (Balfe & Tantam, 2010; Tani et al., 2012). Parents or caregivers of individuals with ASD report their family members demonstrate sexual interest and behaviors but often do so inappropriately (e.g., stalking), which leads to more rejection and isolation (Hellemans, Colson, Verbraeken, Vermeiren, & Deboutte, 2007; Mehzabin & Stokes, 2011). These challenges often result in a cycle of failed attempts that result in negative social experiences or limited opportunities for romantic relationships.

Research on adults with ASD in romantic relationships is limited but yields important social and clinical implications. Byers, Nichols, Voyer, and Reilly (2012) report that males with mild deficits in social communication reported significantly greater sexual satisfaction and desire in their romantic relationships than females; females with ASD reported more sexual knowledge and fewer solitary activities. In a study of married couples in which one partner was diagnosed with ASD, the ASD partner presented an avoidant style of attachment. Their spouses without ASD presented secure attachments, which led to high-marital satisfaction among both partners (Lau & Peterson, 2011). One explanation for this may be that when the ASD partner required autonomy (avoidant), their non-ASD partner provided healthy distance without feeling rejected (secure), which ultimately satisfied both partners.

These studies present a mixed profile of the issues that adults with ASD face in romantic relationships. Recent research in adult treatment for social and dating relationships was conducted by the Program for the Education and Enrichment of Relational Skills (PEERS^®) at University of California Los Angeles (UCLA). In two randomized control trials (RCT), the UCLA PEERS program resulted in improvements in social skills, autistic traits, social get-togethers, and social communication 1–5 years after treatment (Gantman, Kapp, Orenski, & Laugeson, 2012; Laugeson, Gantman, Kapp, Orenski, & Ellingsen, 2015). This research demonstrates both immediate and long-term gains for young adults with autism who participated group treatment. Some of the skills included (a) how to start a conversation, (b) asking someone on a date, and (c) dating etiquette (Gantman et al., 2012). This research identified social initiation skills essential for romantic relationships with positive treatment outcomes.

Romantic Relationship Variables

Adults with autism could benefit from additional romantic relationship skills. These include (a) social skills, (b) empathy, and (c) social support.

Relationship Enhancement^® (RE)

The RE program entitled “Ready for Love” was developed by Ortwein and Guerney in 2008. RE is a psychoeducational treatment and brief therapy model with 20 years of empirical research validating its effectiveness with various populations (Accordino & Guerney, 2003). One reason RE was selected for this study was its use with special populations, including personality disorders, schizophrenia, and intellectual disabilities (Accordino & Guerney, 2001; Ginsberg, 1984). Additionally, when compared to other couples counseling programs, RE yielded the best outcomes in interpersonal communication, expressive empathy, problem-solving, marital adjustment, self-esteem, adaptability, marital happiness, and self-differentiation (Accordino & Guerney, 2001, 2003).

RE assumes a history of relationship experience and level of competency in its program materials. Due to the limited exposure to romantic relationship experiences that adults with ASD report, we developed a supplemented version of RE, entitled RE-ASD, for this study. Supplements include three lessons associated with relationship initiation, such as flirting, assessing someone’s interest in you, and asking someone on a date. These supplements were selected based on their inclusion in other ASD treatment research (Gantman et al., 2012). We hypothesized that social skills, empathy, and social support would improve for participants across treatment conditions. We also hypothesized that participants in the RE-ASD condition would have more significant improvements than in the RE condition.

Participants

Participants were recruited from Florida’s Centers for Autism and Related Disabilities (CARD), a support service agency for individuals with ASD and their family members. Therefore, the sample was derived from a nonprobability, convenience sample, and the primary researcher was employed by CARD. Adults with ASD were determined eligible for this study based on the following criteria: 18 years of age and above (no age limit); preexisting ASD diagnosis (i.e., mild specifier or history of Asperger’s syndrome); age and legal ability to consent to treatment (at least 18 years old and no legal guardianship); average verbal communication skills, a 10th-grade reading level, and interest in participating in treatment for romantic relationships.

A G*power a priori power analysis was conducted to determine the number of participants needed to achieve an α error probability of .05, effect size of .96, and power of .80 (Faul, Erdfelder, Buchner, & Lang, 2009). Recruitment resulted in 69 interested applicants and 38 of who met criteria, consented to participate, and received treatment. The number of participants met the a priori analysis criterion.

Participants included 30 males and 8 females. A majority of the participants identified as single, Caucasian, between 18- and 29-year-old, with some college education. Table 1 summarizes participant and group demographics. Participants were volunteers who did not receive incentive or compensation and who were legally and ethically able to consent to participate in treatment (American Counseling Association [ACA], 2014). A computer-generated randomization program was used to assign a treatment condition to the participants.

OPEN IN VIEWER

Table 1.

Participant Demographics and Group Membership.

Variable	n	Percentage	RE n	RE-ASD n
Gender
Male	30	78.9	14	16
Female	8	21.1	5	3
Race
Caucasian	25	65.8	11	14
Hispanic	8	21.1	5	3
African American/ Black	1	2.6	0	1
Asian/Pacific  Islander	2	5.3	2	0
Other	2	5.3	1	1
Age
18–29	29	76.3	14	14
30–39	4	10.6	1	3
40–49	2	5.3	1	1
50 and over	3	7.8	2	1
Education
Some high school	2	5.3	1	1
High school  graduate	9	23.6	7	2
Some college	21	55.3	8	13
College graduate	2	5.3	1	1
Graduate degree	4	10.5	2	2
Marital status
Single	35	92.1	16	19
Married	3	7.9	3	0

Note. RE = Relationship Enhancement; ASD = autism spectrum disorder.

Procedure

Data were collected during pre- and posttreatment, which occurred during the eligibility interview and at the end of the final treatment session. All groups followed the same treatment format, meeting 2 hr weekly for a total of 8 weeks. The same licensed mental health counselor (primary researcher) facilitated all groups. The RE-ASD condition is differentiated by three supplements that were added to the original RE program to include topics of relationship initiation. The components included in RE-ASD included how to flirt, assessing if someone likes you, and asking someone out on a date (Gantman et al., 2012). These components were specifically incorporated into three different sessions at the third, fourth, and fifth sessions during the 8-week treatment.

There were two primary differences between the RE and RE-ASD conditions. First, lessons were introduced at different times. The RE group received treatment as usual, two lessons per session, and the RE-ASD group received three lessons in the sessions identified above. Second, the time allotted for rehearsal of these components was adjusted for these groups to fit within the 2-hr time frame. These additional lessons did not extend treatment time but were integrated into the treatment as usual protocol by modifying time spent engaging in behavior rehearsal of standard RE skills.

Self-report measures were used to determine the effects of treatment on social skills, empathy, and social support. Three instruments were used to determine levels of social skill acquisition for participants including the Social Responsiveness Scale 2 (SRS-2; Constantino, 2012), the Autism Spectrum Quotient (AQ; Baron-Cohen et al., 2001), and the Dating and Assertion Questionnaire (DAQ; Levenson & Gottman, 1978). For sensitivity to the population being treated, social skills were measured by two autism sensitive instruments, the SRS-2 and AQ, which determined changes in autism diagnostic categories, subscales, and overall symptomology. The DAQ was used to measure the social skills of dating and assertion. The Empathy Quotient (EQ; Baron-Cohen & Wheelwright, 2004) was used to measure empathy, and the Social Provisions Scale (SPS; Cutrona & Russell, 1987) was used to measure social support.

Measures

Social Skills

For the first set of social skills, improvements included decreases in total and subscale scores related to symptoms of autism on the SRS-2. Outcomes in the subscale categories of social communication, social motivation, and RRBIs demonstrated improvements in each of these areas. It is noteworthy that treatment focused on teaching principles of self-expression (communication) and practicing these skills. It is likely that this contributed to the increase in social communication. During treatment, the counselor delivered corrective feedback and coaching that targeted behaviors that interfere with romantic relationships, and this likely influenced the improvement in RRBIs.

Interestingly, participants did not report improvements in social awareness or cognition. Likewise, there were nonsignificant reductions in ASD symptomology as measured by the AQ which does not reflect the same results of the SRS-2. One reason for this may be that the AQ was not specifically designed to measure treatment outcomes, as is the SRS-2, and therefore may not have accurately reflected the effects of treatment.

The DAQ measured the social skills of dating and assertion which produced mixed results, with improvements in dating but not assertion. Since the intervention specifically addressed communication and functioning in a romantic relationship, an increase in dating social skills is encouraging. The specific strategies of dating initiation were only addressed in the RE-ASD condition which led to our hypothesis that participants in this group would have better results in dating; yet, participants in both conditions demonstrated effects in dating. The implication of this is that simply participating treatment contributed to improved self-reported dating skills. The social skill of assertion was not effected by treatment.

Empathy

An increase in empathy is an important result of this study. The foundational skill of showing empathy was a focus in RE and repeatedly practiced in treatment. Therefore, the increase in this variable may be attributed to the emphasis of empathy in this protocol.

Due to previously researched and reported differences between genders in the domain of empathy, it was important to explore the potential for gender differences in this study. Descriptively (mean scores), females did report higher mean scores in empathy than their male counterparts. However, those differences did not demonstrate effects indicating that empathy improved in similar intervals for both genders. Interestingly, RE has been shown to balance certain variables between genders which may have had an effect on this variable (Accordino & Guerney, 2001, 2003).

Clinical and Research Implications

The demographic and cultural factors of those who participated in this research are an important consideration. A majority of participants were single, Caucasian males, between the ages of 18 and 29 years with some college education. Gender and racial minority participants in this study fell within the majority age range and education. These characteristics were addressed to those in the minority as a part of the consent process (ACA, 2014, G.2.a.). The only demographic that was formally assessed in this study was for empathy differences according to gender which did not produce effects. One question that arose during treatment were from married participants who were interested in receiving treatment with their partner (instead of solo as the protocol dictated). Therefore, tailoring treatment for romantic relationship skills based on relationship status could be preferable.

The ethical and legal implications of the rights of research participants were particularly important for this study (ACA, 2014, G.2.a.). First, to ensure that participants could benefit from treatment, we determined cognitive functioning through verbal and written responses to questions at a 10th-grade reading level. Those who were not able to benefit from verbal questions or written material at this level would likely become frustrated or discouraged by participating in such a group and therefore were deemed ineligible during the screening process. The researcher provided referrals to other service providers for these individuals. A legal and ethical factor considered in participant selection was the individual’s determined ability to consent to treatment. Individuals who were under the legal guardianship of a caregiver, which allows the caregiver to make medical or legal decisions on the person’s behalf, were not eligible to participate in this study (ACA, 2014, G.2.e.). This was a protection put in place to ensure that participants were legally and ethically consenting.

There were several outcomes from this study that can help inform future research for adults with ASD. One is the need to explore ways in which social support can be better acquired by adults with ASD. A recommendation to address this would be to incorporate individuals from the adult’s life in treatment through partner, caregiver, or peer mediation.

Another outcome of this study was that social skills did not improve in all measured domains. Individuals with ASD do benefit from repetition of instruction, therefore extending treatment may have led to consistent improvements in social skills. A modification of treatment from its current format to extend time in treatment may result in greater effects.

The utilization of gathering such qualitative feedback from adults with ASD through focus groups and semistructured interviews would likely enhance research in this area. Given that the majority of participants in this study were single, individuals in this demographic could identify topics of need for treatment. Another recommendation for future qualitative research is to conduct interviews with ASD individuals who have had success in romantic relationships to identify what has worked for this group.

More quantitative research needs to be done to build upon the existing outcomes for romantic relationship treatment. Firstly, RCTs should be conducted to compare the effects of counseling interventions for adults with ASD to those not in treatment. In addition to conducting RCTs to assess the differences for adults who are and are not receiving treatment, research could be expanded to compare results of different romantic relationship skills treatment protocols.

We also suggest that there are significant variables beyond those included in this study that would be important to measure quantitatively. Outside of self-report and caregiver data, observational data such as the frequency and nature of romantic encounters like dates, hookups, and sexual activity could be gathered before, during, and after treatment. Future research could also explore predictors of relationship success through facial expressions, communication style, and personality traits at individual, couple, and group levels.

Limitations

The utilization of a convenience sample with 38 participants minimizes the ability to generalize the results of this study. Participants mostly represented single, Caucasian males between the ages of 18 and 29; therefore, results cannot be applied to other relationship statuses, races, genders, or age ranges. This limitation could have been minimized by including a more diverse sample of participants through wider recruitment strategies in different community agencies and through social media.

This study provides guidelines on utilizing self-report measures. Although self-report is a helpful metric for measuring treatment effects, it is important to note that third-party reports would have added to the value of the research. We attempted to gather data from caregivers, partners, and others but did not receive a rate of return to satisfy a priori power analysis requirements. Effectively, gathering data from third-party reporters could have added to the treatment results and discussion of this research.

Lastly, no follow-up or long-term data were collected from participants to determine lasting treatment effects. Data collected after a period of time without treatment (e.g., 1- to 6-months posttreatment) would have helped inform if the gains made posttreatment were lasting. In future research, follow-up data collection procedures should be implemented in order to determine any long-term effects.