Primary Health Professionals’ Beliefs, Experiences, and Willingness to Treat Minor-Attracted Persons

There is a desire and need among minor-attracted persons (MAPs) to access support within the community, and this often begins with an approach to healthcare providers working in general medical/mental health settings. However, little is known about the experiences of these non-specialist professionals in relation to their beliefs, knowledge, and decision-making processes when working with patients who disclose sexual attractions to children. Using an online survey, this study explored the knowledge, comfort, competence, and treatment willingness of 220 non-specialist healthcare providers when faced with patients who disclose sexual attractions to children. We investigated how often such disclosures were made, clinician stigma, treatment priorities, and professionals’ willingness to report MAPs to external agencies because of their sexual attractions. Some key differences were found when comparing primary medical vs. mental health professionals, including increased likelihood to view MAPs as dangerous, unable to control behaviors and that sexual attractions are an avoidable choice, in the former group. Both groups prioritized mental health treatment targets above controlling attractions and living with stigmatized attractions, although controlling or changing attractions were still relatively high priorities. Results indicated a need for further training, focusing on increasing comfort around working with MAPs, as this was associated with a greater willingness to work with this group. We identify current gaps in service provision for MAPs seeking professional support and discuss recommendations for professional training. Supplementary Information The online version contains supplementary material available at 10.1007/s10508-021-02271-7.

After conducting exploratory analyses of the dimensionality of this measure, no meaningful factors underpinned the data, and as such we presented between-groups comparisons of each item. However, a reviewer of this earlier version suggested removing this owing to the degree of overlap between the attribution scale's items and those contained within the SPS (Imhoff, 2015). As such, these analyses are not reported within the main body of this paper. In the interest of transparency, though, we present them within this supplement.

Results
We subjected the attributions measure to an exploratory factor analysis to explore whether participants made attributions about MAPs in a way that could be conceived of as belief clusters. This analysis yielded adequate outcomes for the Kaiser-Meyer-Olkin coefficient (0.71) and Bartlett's test for sphericity was significant, χ 2 (136) = 436.62, p < .001.
Taken in combination, these tests show that the data we collected were suitable for factor analysis. However, the rotated pattern matrix (using an oblique 'direct oblimin' rotation method and extracting factors with eigenvalues greater than 1 using the maximum likelihood method of factor extraction) produced seven factors that each contained either one (four factors) or two (three factors) significantly-loading items when considering Field's (2005) cutoff item loading value of 0.40 (see Table S1). As this does not meaningfully reduce the measure into a smaller number of distinct domains, we present descriptive statistics for each item, broken down by professional group, in Table S2.  Higher scores indicate an answer that equates to a 'very' or completely' response to the question. For questions containing two options, higher scores indicate responding in the affirmative to the latter option. Effect sizes show how those working in primary medical contexts scored in comparison to those working within mental health settings.
In comparison to professionals working in mental health, those working in a primary medical setting (e.g., general practitioners, nurses, and medical physicians) were more likely than mental health professionals to view patients with sexual attractions to children as being dangerous, attribute blame to patients for their attractions, and see having sexual attractions to children as being avoidable. They were also more likely to think that MAPs would have issues with work, view clinical symptoms as being confined to the sexual domain, and support the use of psychotherapy as the most effective treatment option. Medical professionals were also less likely than mental health professionals to say that MAPs could control their behavior.

Supplementary Material -Professionals' Knowledge about Pedophilia
We recoded each myth / knowledge statement within our dataset to be scored 0 = incorrect and 1 = correct. Following this, we computed a 'proportion correct' score for each participant (labeled 'knowledge accuracy'). This outcome ranged from 0 to 1 where higher scores indicated greater factual knowledge about pedophilia. Descriptive statistics for each statement, broken down by professional group, are presented in Table   S3.  (1)  Values range from 0-1 and represent mean levels of accuracy within each group with standard deviations in parentheses. Scores closer to 1 indicate greater levels of accuracy (i.e., multiplying by 100 will indicate an average % correct score). The 'overall accuracy' test was computed by converting the % correct score for each participant into a continuous z-score and comparing the two groups using this z-scores as the dependent variable. Statements in italics are coded as false based on the current empirical evidence. Effect sizes show how those working in primary medical contexts scored in comparison to those working within mental health settings.

Measure
Because the treatment priorities measure has not been used in peer-reviewed research before, we initially conducted an exploratory factor analysis to examine its dimensionality.
Within this analysis we used the maximum likelihood method of factor extraction and an oblique (direct oblimin) rotation to allow any underlying factors to correlate with each other.
We retained factors with an eigenvalue greater than one, with participants with missing values excluded in a pairwise manner. The Kaiser-Meyer-Olkin measure of sampling adequacy was calculated as 0.76, and Bartlett's test of sphericity was significant, χ 2 (55) = 682.20, p < .001.
These statistics indicate that the data collected on the treatment priorities scale were suitable for factor analysis.
Three factors were retained (Table S4). The first factor was comprised of two items and relate to general mental health concerns. The second factor was much larger, consisting of five items related to the forensic control of sexual attractions. The final factor contained three items that were related to living within society with a stigmatized sexual attraction pattern. When considering Field's (2005) factor loading cut-off value of 0.40, one item ("To help the patient deal with sexual frustration") loaded significantly on to both factor two and factor three. As such, we excluded this item from the measure. An average score for each factor was computed and used in subsequent exploratory analyses. The three treatment target factors were all positively correlated with each other, though to a small-to-moderate degree (see Table S5). This differentiation between treatment targets is supportive of the view that there are different clusters of needs that might be targeted by healthcare professionals when working with MAPs. A summary of the descriptive statistics indicating treatment prioritization scores between the two professional groups is presented in Table S6.