Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Effect of Gonadotropin-Releasing Hormone Antagonist on Risk of Committing Child Sexual Abuse in Men With Pedophilic Disorder

Effect of Gonadotropin-Releasing Hormone Antagonist on Risk of Committing Child Sexual Abuse in... Research JAMA Psychiatry | Original Investigation Effect of Gonadotropin-Releasing Hormone Antagonist on Risk of Committing Child Sexual Abuse in Men With Pedophilic Disorder A Randomized Clinical Trial Valdemar Landgren, MD; Kinda Malki, MD; Matteo Bottai, ScD; Stefan Arver, MD; Christoffer Rahm, MD Editorial page 893 IMPORTANCE Evidence-based treatments from randomized clinical trials for pedophilic Supplemental content disorder are lacking. OBJECTIVE To determine whether a gonadotropin-releasing hormone antagonist reduces dynamic risk factors for committing child sexual abuse. DESIGN, SETTING, AND PARTICIPANTS This academically initiated, double-blind, placebo-controlled, parallel-group, phase 2 randomized clinical trial was conducted at the ANOVA center in Stockholm, Sweden, from March 1, 2016, to April 30, 2019. Individuals who contacted PrevenTell, the national telephone helpline for unwanted sexuality, were recruited. Eligible participants were men seeking help aged 18 to 66 years with a pedophilic disorder diagnosis and no contraindications to the intervention. The primary end point was assessed by intent-to-treat analysis. INTERVENTIONS Randomization to receive either 2 subcutaneous injections of 120 mg of degarelix acetate or equal volume of placebo. MAIN OUTCOMES AND MEASURES The primary end point was the mean change between baseline and 2 weeks in the composite risk score of 5 domains of child sexual abuse ranging from 0 to 15 points; each domain could be rated from 0 to 3 points. Secondary end points included efficacy at 2 and 10 weeks as measured by the composite score, each risk domain, quality of life, self-reported effects, and adverse events. RESULTS A total of 52 male participants (mean [SD] age, 36 [12] years) were randomized to receive either degarelix (n = 25; with 1 withdrawal) or placebo (n = 26). At 2 weeks, the composite risk score decreased from 7.4 to 4.4 for participants in the degarelix group and from 7.8 to 6.6 for the placebo group, a mean between-group difference of –1.8 (95% CI, –3.2 to –0.5; P = .01). A decrease was seen in the composite score at 10 weeks (−2.2 [95% CI, −3.6 to −0.7]) as well as in the domains of pedophilic disorder (2 weeks: −0.7 [95% CI, −1.4 to 0.0]; 10 weeks: −1.1 [95% CI, −1.8 to −0.4]) and sexual preoccupation (2 weeks: −0.7 [95% CI, −1.2 to −0.3]; 10 weeks: −0.8 [95% CI, −1.3 to −0.3]) in the degarelix group compared with the placebo group. No difference was seen for the domains of self-rated risk (2 weeks: −0.4 [95% CI, −0.9 to 0.1]; 10 weeks: −0.5 [95% CI, −1 to 0.0]), low empathy (2 weeks: 0.2 [95% Author Affiliations: Institute of CI, −0.3 to 0.6]; 10 weeks: 0.2 [95% CI, −0.2 to 0.6]), and impaired self-regulation (2 weeks: Neuroscience and Physiology, −0.0 [95% CI, −0.7 to 0.6]; 10 weeks: 0.1 [95% CI, −0.5 to 0.8]), or quality of life (EuroQol 5 Gothenburg University, Gothenburg, Sweden (Landgren); Karolinska Dimensions questionnaire index score, 2 weeks: 0.06 [95% CI, −0.00 to 0.12], and 10 weeks: University Hospital, Stockholm, 0.04; 95% CI, −0.02 to 0.10; EuroQol visual analog scale, 2 weeks: 0.6 [95% CI, −9.7 to 10.9], Sweden (Malki); Institute of and 10 weeks: 4.2 [95% CI, −6.0 to 14.4]). Two hospitalizations occurred from increased Environmental Medicine, Karolinska Institutet, Stockholm, Sweden suicidal ideation, and more injection site reactions (degarelix: 22 of 25 [88%]; placebo: 1 of 26 (Bottai); Department of Medicine [4%]) and hepatobiliary enzyme level elevations were reported by participants who received Huddinge, Karolinska Institutet, degarelix (degarelix: 11 of 25 [44%]; placebo: 2 of 26 [8%]). Among the 26 participants Stockholm, Sweden (Arver); randomized to receive degarelix, 20 (77%) experienced positive effects (eg, improved Centre for Psychiatry Research, Department of Clinical Neuroscience, attitude or behavior) on sexuality and 23 (89%) reported adverse effects on the body. Karolinska Institutet, Stockholm, Sweden (Rahm); Stockholm Health CONCLUSION AND RELEVANCE This trial found that degarelix reduced the risk score for Care Services, Region Stockholm, committing child sexual abuse in men with pedophilic disorder 2 weeks after initial injection, Stockholm, Sweden (Rahm). suggesting use of the drug as a rapid-onset treatment option. Further studies are warranted Corresponding Author: Christoffer into the effects and long-term adverse effects of hormone deficiency. Rahm, MD, Stockholm Health Care Services, Region Stockholm, TRIAL REGISTRATION EU Clinical Trials Register Identifier: 2014-000647-32 Norra Stationsgatan 69, SE-113 64 JAMA Psychiatry. 2020;77(9):897-905. doi:10.1001/jamapsychiatry.2020.0440 Stockholm, Sweden Published online April 29, 2020. (christoffer.rahm@ki.se). (Reprinted) 897 Research Original Investigation Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder hild sexual abuse affects 1 in 5 girls and 1 in 10 boys worldwide. It is accompanied by adverse psychoso- Key Points 1,2 C cial outcome across the life span. Preventive mea- Question Can a gonadotropin-releasing hormone antagonist sures have been advocated, but to date evidence for inter- rapidly reduce the risk for committing child sexual abuse in men 4,5 ventions has been limited. with pedophilic disorder who are seeking help? The estimated proportion of child sexual offenders who Findings In this randomized clinical trial of 52 men with 1,4 are prosecuted is 1%. Of those who were prosecuted, up to pedophilic disorder, treatment with degarelix statistically 95% were first-time offenders and half of them had pedo- significantly reduced the risk for committing child sexual abuse philic disorder, defined as recurrent sexual attraction to pre- 2 weeks after the initial injection. pubescent children associated with distress or negative con- Meaning This finding suggests that degarelix may serve as a sequences. Studies have found that not all men with pedophilic rapid-onset, risk-reducing medication for men with pedophilic disorder commit a sexual offense, but those who do gener- disorder. ally report struggling with their sexual urges for 10 years be- 6,8 fore committing a sexual crime. Consequently, an opportu- nity for prevention exists in treating high-risk individuals lenges associated with conducting a trial in a hard-to-reach popu- without prior convictions. Effective treatment could prevent lation along with the issues of tolerability of previous therapies, child sexual abuse and reduce psychosocial stress for the in- we performed ancillary interviews of self-reported experiences dividual with pedophilic disorder. of treatment. We regarded using sexually arousing material and Currently recommended interventions include psycho- measuring participants’ penile responses as too intrusive. therapy and antidepressants as well as testosterone-suppressing We hypothesized that men with pedophilic disorder who medicationsforhigh-riskindividuals. Opinionsabouttreatment were randomized to receive degarelix compared with pla- are vehement. In many countries, an informed consent proce- cebo would have a substantial reduction in risk of commit- dure is required for chemical castration. However, chemical cas- ting child sexual abuse after 2 weeks. Furthermore, we hy- tration is used as a compulsory legal sentence for child sexual pothesized that degarelix would be sufficiently tolerated offenders in some jurisdictions in the US, Asia, and Europe but by the participants and thus would be a safe and effective is prohibited in other countries owing to ethical concerns of rapid-onset treatment option. 10,11 coercion and uncertain efficacy. Because they lower testosterone through receptor desen- sitization, gonadotropin-releasing hormone agonists are con- Methods sidered effective in reducing paraphilic symptoms. How- ever, the use of these agonists is limited to supervised This randomized clinical trial was approved by the Swedish correctional settings because of the lack of randomized clini- Central Ethical Review Board and the Swedish Medical Prod- cal trials; their metabolic adverse effects; and the initial flare-up ucts Agency (trial protocol in Supplement 1). All participants of testosterone, which may be associated with increased ag- signed an informed consent agreement and were offered treat- gression and libido that require concurrent antiandrogen ment after the study. To ensure adherence to good clinical medication. Experience from the Swedish helpline Preven- practice, the Karolinska Trial Alliance, a research center that Tell suggests a need for both rapid-acting short-term treat- supports clinical trials, provided independent monitoring and ment (eg, in critical phases of the disorder to quickly control recommended a longer follow-up period to monitor the safety sexual impulses or reduce a high degree of struggling) and of the intervention; hence, a second follow-up visit with the long-term treatment. Degarelix acetate is a gonadotropin- same outcome measures was conducted at 10 weeks. The study releasing hormone antagonist that was approved by the US was conducted without any kind of collaboration with the phar- Food and Drug Administration in 2008 for treating advanced maceutical industry. This trial followed the Consolidated Stan- prostate cancer. Degarelix decreases testosterone to castra- dards of Reporting Trials (CONSORT) reporting guideline. tion levels within 3 days without testosterone flare ; there- fore, the drug could serve as a rapid-onset treatment for indi- Trial Design and Participants 9,14,15 viduals seeking help in outpatient settings. This academically initiated, double-blind, placebo- For individuals without a prior conviction, no validated controlled, parallel-group phase 2 trial with balanced random- measures of risk exist. Therefore, the PRIOTAB (Pedophilia at ization (1:1) was conducted at the ANOVA center, a highly Risk–Investigations of Treatment and Biomarkers) project, of specialized center for andrology and sexual medicine in Stock- which this present trial is a part, included the construction holm, Sweden, from March 1, 2016, to April 30, 2019. The trial of a composite score based on previous observational studies was conducted conjointly with a case-control study, for which of dynamic (ie, potentially changeable over time) risk factors participants underwent magnetic resonance imaging and for sexual offense recidivism. One such risk factor is deviant additional testing at the same time and venue. The ANOVA sexual interest (eg, pedophilic disorder). Other factors are center hosts PrevenTell, a national telephone helpline for self- sexual preoccupation, impaired self-regulation, and low identified unwanted sexuality, through which trial partici- empathy. Because these 4 factors are all possibly mitigated pants were recruited. In Sweden, the legal obligation to by testosterone suppression, we believed that degarelix could report suspected child abuse supersedes professional confi- 17-20 have a risk-reducing effect. Bearing in mind the chal- dentiality, which may make individuals with pedophilic dis- 898 JAMA Psychiatry September 2020 Volume 77, Number 9 (Reprinted) jamapsychiatry.com Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Original Investigation Research order reluctant to seek help. To minimize compromises in re- Figure. CONSORT Diagram cruitment rates and to attain authenticity in self-reports, we provided participants a temporary identification number. This 623 Individuals called PrevenTell helpline identification number kept participants anonymous to trial as- 523 Consented to answering a structured questionnaire and screened for eligibility sessors (S.A., C.R.), although they were still informed of the professionals’ obligation, according to the Swedish Social Ser- 468 Excluded vices Act, to send a notification of concern to the authorities 404 Had no sexual attraction to children when they suspect a child is at risk of abuse or maltreatment 54 Had sexual attraction and the legal possibility of reporting to the police suspected to children but did not fulfill inclusion criteria perpetrators of crimes against children. and/or met exclusion criteria (most commonly Participants were recruited from March 1, 2016, to Janu- hebephilia) ary 31, 2019. A flow diagram of the participants through each stage of the trial is shown in the Figure. Eligible participants 65 Interviewed by telephone were help-seeking, self-identified men aged 18 to 66 years with a pedophilic disorder diagnosis, as defined in the Diagnostic 10 Excluded 3 Had negative result for and Statistical Manual of Mental Disorders (Fifth Edition). A full pedophilic disorder list of inclusion and exclusion criteria is provided in eAppen- 6 Declined 1 Underwent recent dix1in Supplement 2. Participants recruited through Preven- antiandrogen therapy Tell were offered transportation costs for study visits and financial compensation of Sk 1000 (Swedish krona) (US $100 55 Underwent screening at ANOVA center before taxation) after study completion. 3 Excluded 1 Had severe psychotic Outcome Measures symptoms The primary end point was the mean change in composite 1 Had liver abnormalities risk score (range, 0-15 points) between baseline and 2 weeks 1 Withdrew consent after randomization. The composite risk score consisted of 5 domains: the 4 empirically derived dynamic risk factors 52 Randomized (pedophilic disorder, sexual preoccupation, impaired self- regulation, and low empathy) and self-rated risk, each of 25 Received degarelix 26 Received placebo which could be rated from 0 to 3 points (Table 1; eAppendix 1 1 Withdrew consent before injection and declined follow-up in Supplement 2). Each domain was weighted equally in the score. Three risk groups were prespecified (risk class 1: 0-5 23 Included in analysis of primary 26 Included in analysis of primary points; risk class 2: 6-10 points; risk class 3: 11-15 points). end point at 2 weeks end point at 2 weeks 1 Did not receive intervention We used the prespecified analysis for the primary end 2 Had incomplete baseline point. This analysis included only participants with complete self-ratings data at baseline and 2 weeks after injection. The secondary end point analyses excluded no data. Sec- 24 Completed follow-up at 10 weeks 26 Completed follow-up at 10 weeks 1 Did not receive intervention ondary end points were efficacy at 2 and 10 weeks in terms of 1 Hospitalized for suicidal ideation reduction in the composite risk score and its 5 domains, effi- cacy in participants in risk class 3 (11-15 points), quality of life, 26 Included in analyses 26 Included in analyses of secondary end points of secondary end points adverse events, and self-reported effects. Quality of life was measured at all time points, using the EuroQol 5 Dimensions (EQ-5D) questionnaire, which consists of a preference-based health status measure convertible into index scores ranging improved attitudes or behaviors) and negative (eg, adverse events) effects of the injection and willingness to maintain the from 0 to 1 (with the higher score indicating better health sta- tus), and a EuroQol visual analog scale (EQ-VAS) of 0 to 100, experienced effects with further injections. Interviews were 27,28 analyzed using qualitative descriptive content analysis, as de- by which higher scores represent better health status. Participants were given a study diary and instructed to note scribed in eAppendix 2 in Supplement 2. Neither the asses- sors nor the participants were aware of treatment randomiza- any adverse events between assessments and to call the study nurse if they perceived serious harm from treatment. At follow- tion at the time of the interviews or qualitative analysis. up, physical adverse events were registered by the study nurse using both participant responses to open-ended questions Trial Intervention about current health status and notes in the diary. Adverse On the first visit, participants underwent eligibility and base- events were documented by the assessing psychiatrist (C.R.) line evaluations and received the study drug (2 subcutane- also, and all events were coded according to the MedDRA ous injections of 120 mg degarelix acetate or equal volume of (Medical Dictionary for Drug Regulatory Affairs). Self- placebo). A computer-generated sequence with permuted block reported experiences were collected in a structured inter- randomization was provided by the Karolinska Trial Alliance. view, including open-ended questions about the positive (eg, The randomization sequence was concealed from one of us jamapsychiatry.com (Reprinted) JAMA Psychiatry September 2020 Volume 77, Number 9 899 Research Original Investigation Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Table 1. Composite Score of Dynamic Risk for Committing Child Sexual Abuse Score definition Risk domain 01 2 3 Pedophilic disorder No pedophilic Pedophilic attraction Pedophilic attraction + distress or Pedophilic attraction + distress + attraction negative consequences negative consequences Sexual preoccupation Hyposexual according Not hyposexual according to Not hyposexual + hypersexual Not hyposexual + hypersexual + to the SDI the SDI according to the HBI; no ongoing ongoing abusive behavior abusive behavior according to the SChiMRA-B Impaired Normal CCPT II result 1 Abnormal CCPT II domain out 2 Abnormal CCPT II domains of the All 3 abnormal CCPT II domains self-regulation of the inattention, impulsivity, inattention, impulsivity, and of inattention, impulsivity, and and vigilance domains vigilance domains vigilance Low empathy No abnormality 1 of RAADS-14 mentalizing 2 of RAADS-14 mentalizing domain RAADS-14 mentalizing domain domain >10, RMET <22, or >10, RMET <22, or current >10, + RMET<22, + current current antisocial behavior antisocial behavior antisocial behavior according to the MINI Self-rated risk Normal SChiMRA-A 1 Domain of SChiMRA-A watch, 2 Domains of SChiMRA-A watch, SChiMRA-A watch + socialize + result socialize, or interact domain socialize, or interact domain interact domains c 21 Abbreviations: CCPT II, Conners Continuous Performance Test; Hyposexuality is defined as a score of <45 on the SDI. Hypersexuality is HBI, Hypersexual Behavior Inventory (score range: 19-95, with higher scores defined as an HBI score of53. The SChiMRA-B (eAppendix 1 in indicating more severe hypersexual behavior); MINI, Mini International Supplement 2) assesses self-reported frequency of sexually abusive behavior Neuropsychiatric Interview; RAADS-14, Ritvo Autism and Asperger Diagnostic in the past week (never, several days, more than half of days, or almost every Scale, 14 Screen (score range: 0-42, of which the mentalizing domain ranges day) regarding watching of, socializing with, and sexual interaction with from 0 to 21, with higher scores indicating more autistic features); children, in which occurrence of any sort is scored as positive. RMET, Reading the Mind in the Eyes Test (score range: 0-36, with higher scores d 23 CCPT II for inattention, impulsivity, and vigilance. indicating better capacity for emotion recognition); SChiMRA-A, Sexual Child e 24 25 Self-ratings on the mentalization domains of the RAADS-14, RMET scores, Molestation Risk Assessment part A; SDI, Sexual Desire Inventory (score range: and current antisocial behavior as reported in the MINI. 12-104, with higher scores indicating increased sexual desire); VAS, visual analog The SChiMRA-A (eAppendix 1 in Supplement 2) consists of VAS ratings to the scale (score range: 0%-100%, with higher scores indicating increased self-rated question, “How likely is it that you would do any of the following, if there was risk). an easy way to do it without being caught?” regarding watching of, socializing For a full description of the composite risk score, see eAppendix 1 in with, and sexual interaction with children. A rating of 40% or higher on the Supplement 2. VAS was interpreted as a substantial risk. The 3 criteria for pedophilic disorder according to the DSM-5 are pedophilic interest, significant distress, and significant negative consequences. (C.R. and S.A.) who enrolled and assessed participants and then dition, we evaluated the trajectories of the secondary end transferred onto cards placed in sequentially numbered, points in the 2 treatment groups at the 3 visits, including all opaque envelopes stored in a locked cabinet in the dispen- randomized participants. The mean of the numeric second- sary and accessible to only 1 independent study nurse, who in ary end points was estimated using linear random-effects re- turn informed the nurse responsible for drug administration. gression models, including the treatment indicator (binary co- A detailed description of the injection procedure is provided variate), indicator variables for the 2 follow-up visits (binary), in eAppendix 1 in Supplement 2. Except for the study nurse and the 2 interaction terms between the treatment indicator who was responsible for treatment randomization and who also and the 2 visit indicators (binary). The models also included a registered the physical adverse effects, all of the assessors participant-specific random intercept, which was assumed to involved in the repeated outcome assessments were blinded follow a normal distribution. The random intercept was in- to the treatment allocation. cluded to consider the potential dependence in the repeated observations on each participant. Statistical Analysis We examined the differences in the time trajectories be- The sample size of this trial was determined from the experi- tween the treatment groups by testing the composite hypoth- ence of similar, previously published studies of biomarkers esis that the interaction terms were jointly equal to zero using of treatment response. We calculated that the planned sample Wald-type tests. We used the estimates from the models to es- size of 60 would give a power greater than 90% to detect a clini- timate the mean of the numeric end points. The SEs used to cally significant difference in the primary end point. We had calculate their CIs were obtained with the delta method. All no prespecified expected dropout rate. statistical analyses were performed using Stata, version 15 The primary end point was assessed by intent-to-treat (StataCorp LLC). analysis, using the unequal-variance, 2-sided t test at interim When 20 participants had completed the study, the analysis and at 2 weeks. To preserve the overall significance planned interim analysis of efficacy for the primary end point level of the test at .05, we chose a 2-sided P = .0294 (exact value and a review of the safety of the treatment were performed by given per the prespecified statistical plan). Participants for the sponsor (S.A.), who could find no statistically significant whom the score change could not be calculated (ie, missing differences between the groups. Because no evidence of sub- values for the risk score at baseline or 2 weeks) were excluded stantial harm from active treatment was found that would mo- from analysis of the primary end point under the assumption tivate the premature termination of the study, the study was that the missing-data generating process was at random. In ad- permitted to continue with the planned sample size of 60. 900 JAMA Psychiatry September 2020 Volume 77, Number 9 (Reprinted) jamapsychiatry.com Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Original Investigation Research Table 2. Demographic and Clinical Characteristics of Participants at Baseline No. (%) Variable Degarelix acetate (n = 26) Placebo (n = 26) Demographic characteristics Abbreviations: ASRS, Adult ADHD (attention-deficit/hyperactivity Age, median (IQR) [range], y 36 (25-39) [19-54] 35 (28-47) [18-66] disorder) Self-Report Scale screener Highest completed educational level (with a minimum of 4 of 6 questions Primary school, 9 y 2 (8) 3 (12) rated above cutoff indicating ADHD); IQR, interquartile range; MADRS, Secondary education, 12 y 12 (46) 13 (50) Montgomery-Åsberg Depression Postsecondary education 12 (46) 10 (38) Rating Scale (score range, 0-48 Unemployed 12 (46) 9 (35) points, with higher scores indicating increased depression severity); Living status MINI, Mini International Caregiver of child 7 (27) 12 (46) Neuropsychiatric Interview); Living without partner 17 (65) 17 (65) RAADS-14, Ritvo Autism and Asperger Diagnostic Scale, 14 Screen. Ever lived with a partner ≥2 y 10 (38) 15 (58) Two participants in the placebo Self-reported prior criminal conviction group and 2 in the degarelix group Noncontact sexual offense 4 (15) 4 (15) stated that the attraction had Contact sexual offense 2 (8) 3(12) “always been present.” Nonsexual offense 2 (8) 5 (19) Based on the Wechsler Adult Intelligence Scale 4 (with the score Sexuality of 100 indicating the mean Attraction primarily to boys 4 (15) 4 (15) intelligence of the population), Attraction primarily to girls 19 (73) 21 (81) MINI, MADRS-S self-rating version, Alcohol Use Disorder Attraction to boys and girls 3 (12) 1 (4) Identification Test (score of8; Attraction exclusively to prepubescent children 2 (8) 9 (35) score range: 0-40, with higher Age at discovery of attraction to minors, median (IQR), y 16 (14-23) 16 (13-25) scores indicating more alcohol b abuse) and the Drug Use Disorder Psychiatric characteristics Identification Test (score of3; Any psychiatric disorder 18 (69) 25 (96) score range: 0-44, with higher Ongoing depression 7 (27) 12 (46) scores indicating more drug 33,34 abuse), ASRS Self-Report Scale MADRS-S score in patients with depression, median (IQR) 25 (24-29) 24 (21-27) screener, and RAADS-14. Current psychotic symptoms 0 2 (8) As indicated by MINI, ASRS, and Previous manic episode 0 1 (4) RAADS-14 scores. Hazardous drug or alcohol use 4 (17) 11 (42) Other medications included sleep medications, antihistamines, Full-scale IQ, median (IQR) 103 (93-118) 99 (96-116) stimulants, and mood stabilizers. Psychoactive medication Composite score ranged from 0 to Antidepressants 7 (27) 8 (31) 15 points, with higher scores Other 7 (27) 5 (19) indicating higher risk. In an age-matched comparison sample of Testosterone, median (IQR), nmol/L 16.5 (11-18.5) 14.0 (10.8-18.3) men (n = 55), the median (IQR) Composite risk score, median (IQR) 7.5 (6.0-8.0) 8 (6.8-9.0) composite risk score was 2 (1-2.5). risk class 1 (n = 6) or risk class 2 (n = 44), but 3 individuals in each treatment group had a baseline score of 10 points and were Results therefore analyzed as a high-risk subgroup. The median (IQR) number of days from baseline to visits was 15 (14-18) days at Participants and Intervention Of the 52 male participants (mean [SD] age, 36 [12] years), 26 2 weeks and 73 (73-74) days at 10 weeks for those in the de- (50%; with 1 withdrawal) were randomized to receive degare- garelix group and 14 (14-14) days at 2 weeks and 73 (73-77) days lix and 26 (50%) to placebo. Of these 52 participants, 39 (75%) at 10 weeks for those in the placebo group. The mean (SD) cas- lived in urban areas of more than 50 000 inhabitants. The travel tration levels for serum testosterone were ascertained at distance from home to the study center was more than 100 km 2 weeks (0.7 [0.2] nmol/L) and 10 weeks (0.6 [0.2] nmol/L) for 28 participants (54%), 6 of whom (12%) traveled more than in all participants randomized to receive degarelix (eTable 4 400 km. Participant characteristics are described in Table 2. in Supplement 2). Minor differences in demographic and clinical characteristics Because the retention rate of participants at 10 weeks was were observed between the 2 treatment groups at baseline, but high (96%; n = 50), analysis of secondary end points was con- the median (interquartile range [IQR]) composite risk score was sidered feasible without reaching the planned sample size of about the same (degarelix: 7.5 [6.0-8.0] vs placebo: 8 [6.8- 60. This finding, along with a temporarily slow inclusion rate, 9.0]) (eTable 1 in Supplement 2). A high prevalence of depres- ended the trial enrollment after randomization of 52 partici- sion (35%) was found at baseline. All participants were in either pants, a decision we made before unblinding. jamapsychiatry.com (Reprinted) JAMA Psychiatry September 2020 Volume 77, Number 9 901 Research Original Investigation Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Table 3. Primary and Secondary End Points Degarelix acetate Placebo Difference (95% CI) Baseline 2wk 10 wk Baseline 2wk 10 wk P End points (n = 24) (n = 25) (n = 24) (n = 26) (n = 26) (n = 26) Baseline 2 wk 10 wk value Primary end points, mean (SE) Composite risk score 7.4 (0.3) 4.4 (0.6) NA 7.8 (0.3 6.6 (0.5) NA NA −1.8 NA .01 (−3.2 to −0.5) Secondary end points [all data], mean (delta method SE) Composite risk score 7.3 (0.5) 4.4 (0.5) 3.6 (0.5) 7.8 (0.5) 6.6 (0.5) 6.2 (0.5) −0.5 −1.8 −2.2 .01 (−1.8 to 0.8) (−3.2 to −0.3) (−3.6 to −0.7) High risk group 10 (1.2) 6.7 (1.2) 2.3 (1.2) 10 (1.2) 10 (1.2) 8.3 (1.2) 0.0 −3.3 −6 .04 [ = 10 points] (n=3+3) (−3.3 to 3.3) (−7.9 to 1.2) (−10.6 to −1.4) Risk domain Pedophilic disorder 2.4 (0.2) 0.9 (0.2) 0.8 (0.2) 2.6 (0.2) 1.8 (0.2) 2.1 (0.2) −0.2 −0.7 −1.1 .01 (−0.7 to 0.3) (−1.4 to 0.0) (−1.8 to −0.4) Sexual 1.6 (0.1) 0.7 (0.1) 0.4 (0.1) 1.6 (0.1) 1.4 (0.1) 1.2 (0.1) 0.0 −0.7 −0.8 .001 preoccupation (−0.4 to 0.4) (−1.2 to −0.3) (−1.3 to −0.3) Impaired 1.5 (0.2) 1.2 (0.2) 1.4 (0.2) 1.4 (0.2) 1.2 (0.2) 1.2 (0.2) 0.0 −0.0 0.1 .82 self-regulation (−0.5 to 0.6) (−0.7 to 0.6) (−0.5 to 0.8) Low empathy 0.8 (0.2) 1.1 (0.2) 0.8 (0.2) 1 (0.2) 1.1 (0.2) 0.8 (0.2) −0.2 0.2 0.2 .61 (−0.6 to 0.3) (−0.3 to 0.6) (−0.2 to 0.6) Self-rated risk 1.0 (0.2) 0.5 (0.2) 0.2 (0.2) 1.2 (0.2) 1.1 (0.2) 0.9 (0.2) −0.2 −0.4 −0.5 .16 (−0.8 to 0.3) (−0.9 to 0.1) (−1 to 0.0) Quality of life EQ-5D index score 0.79 0.83 0.82 0.86 0.83 0.85 −0.06 0.06 0.04 .16 (0.03) (0.03) (0.03) (0.0.3) (0.03) (0.03) (−1.37 to 0.01) (−0.00 to 0.12) (−0.02 to 0.10) EQ-VAS 59.8 59.0 61.3 60.5 59.1 57.8 −0.6 0.6 4.2 .68 (4.5) (4.4) (4.5) (4.5) (4.5) (4.5) (−13.0 to 11.7) (−9.7 to 10.9) (−6.0 to 14.4) Abbreviations: EQ-5D, EuroQol 5 Dimensions questionnaire (score range, P value is presented for a 2-sample t test with unequal variances for the 0-1, with a higher score indicating better health status); EQ-VAS, EuroQol visual primary end point and for a test for different time trajectories between groups analog scale questionnaire (score range, 0-100, with higher scores indicating in the random-effects regression models for all secondary end points. better health status); NA, not applicable. c Complete data for calculating the difference (baseline and 2 weeks) were Differences indicate the status of the participants randomized to receive available for 23 participants randomized to degarelix and 26 participants degarelix. randomized to placebo (Figure). 4.2 [95% CI, −6.0 to 14.4]) (eTable 6 in Supplement 2). Themes Outcomes Primary and secondary end points of efficacy in reducing the and categories from the ancillary interviews are displayed in composite risk score at 2 and 10 weeks differed substantially, Table 4. In the degarelix group, positive attitudes toward sexu- decreasing from 7.4 to 4.4 points for participants in the de- ality (20 of 26 [77%]) and adverse effects on the body (23 of garelix group and from 7.8 to 6.6 points for the placebo group 26 [89%]) were the most common self-reported experiences. (mean difference: –1.8 [95% CI, –3.2 to –0.5]; P = .01) (Table 3). Analysis of specific participant quotes is provided in eAppen- In the regression models of secondary end points, statisti- dix2in Supplement 2. cally significant differences were observed in composite risk Post hoc analyses revealed that 15 of the 26 participants scores (2 weeks: –1.8 [95% CI, –3.2 to –0.5]; 10 weeks: −2.2 [95% (58%) in the degarelix group and 3 of the 26 participants (12%) CI, −3.6 to −0.7]) (eFigure 1 and eFigure 2 in Supplement 2) and in the placebo group denied sexual attraction to minors at high-risk group (2 weeks: −3.3 [95% CI, −7.9 to 1.2]; 10 weeks: 10 weeks. −6 [95% CI, −10.6 to−1.4]), between the pedophilic disorder (2 weeks: −0.7 [95% CI, −1.4 to 0.0]; 10 weeks: −1.1 [95% CI, Adverse Events −1.8 to −0.4]) and sexual preoccupation (2 weeks: −0.7 [95% No serious adverse events occurred in the placebo group. A se- CI, −1.2 to −0.3]; 10 weeks: −0.8 [95% CI, −1.3 to −0.3]) (eFig- rious adverse event of increased suicidal ideation was re- ure 3 and eFigure 4 in Supplement 2) domains. No statisti- ported by 2 of 25 participants (8%) in the degarelix group, which cally significant differences in scores were seen in the do- led to hospitalization. One participant reported the event at mains of low empathy (2 weeks: 0.2 [95% CI, −0.3 to 0.6]; 10 weeks, when his condition had improved; the other re- 10 weeks: 0.2 [95% CI, −0.2 to 0.6]), impaired self-regulation ported suicidal ideation at baseline but was lost to follow-up (2 weeks: −0.0 [95% CI, −0.7 to 0.6]; 10 weeks: 0.1 [95% CI, at 10 weeks (purportedly also owing to a travel distance of >500 −0.5 to 0.8]), and self-rated risk (2 weeks: −0.4 [95% CI, −0.9 km) and was contacted by telephone. In post hoc analyses, rates to 0.1]; 10 weeks: −0.5 [95% CI, −1 to 0.0]) or in either mea- of suicidality (2 weeks: −0.3 [95% CI, −0.5 to 0.1]; 10 weeks: sure of quality of life (EQ-5D index score, 2 weeks: 0.06 [95% −0.1 [95% CI, −0.5 to 0.3]; P = .33), depression (2 weeks: 2.3 CI, −0.00 to 0.12], and 10 weeks: 0.04 [95% CI, −0.02 to 0.10]; [95% CI, 0.3-20.7]; 10 weeks: 2.0 [95% CI, 0.2-19.5]; P =.74), EQ-VAS, 2 weeks: 0.6 [95% CI, −9.7 to 10.9], and 10 weeks: or depression severity (2 weeks: −3 [95% CI, −10 to 4]; 10 weeks: 902 JAMA Psychiatry September 2020 Volume 77, Number 9 (Reprinted) jamapsychiatry.com Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Original Investigation Research Table 4. Questions and Categories from Self-reported Treatment Experiences No. (%) Questions/Categories Degarelix acetate (n = 26) Placebo (n = 26) “What positive/negative effects do you experience from the injection?” Positive effects of treatment Positive effects on sexuality 20 (77) 11 (42) Improved mental health 4 (15) 1 (4) Changed perspective 4 (15) 3 (12) Improved cognitive ability 1 (4) 0 Improved self-control 1 (4) 1 (4) Positive effects on relationship 1 (4) 0 Improved physical health 0 1 (4) Negative effects of treatment Negative effects on body 23 (89) 11 (42) Negative effects on sexuality 11 (42) 4 (15) Relationship problems 4 (15) 2 (8) Mental health issues 1 (4) 4 (15) Decreased cognitive ability 1 (4) 1 (4) Participants were interviewed with Negative emotions 1 (4) 0 open-ended questions. Themes and Negative effects on work 1 (4) 0 categories were abstracted from the “Would you like a repeated injection maintaining the effects for another 10 weeks? Please explain your answer.” answers through qualitative content analysis (eAppendix 2 in Reasons for continuing treatment Supplement 2). Overall positive effects 3 (12) 0 Categories are displayed in Positive emotions 2 (8) 0 descending order of frequency in Positive effects on sexuality 1 (4) 0 the degarelix group. Combining answers from 2 and It’s necessary 1 (4) 0 10 weeks. The question was asked Legal matter 0 1 (4) in 2 steps: (1) “What positive effects To achieve effect 0 1 (4) do you experience from the injection?” and then (2) “What Positive effect on relationship 0 1 (4) negative effects do you experience Reasons for discontinuing treatment from the injection?” Positive effects Negative effects on sexuality 2 (8) 0 refers to improvement in attitudes, behaviors, thinking, and Negative effects on body 2 (8) 1 (4) relationships. Negative effects Achieved effect 2 (8) 0 refers to adverse events. Cautiousness 1 (4) 0 Question asked only at 10 weeks. No effect 1(4) 9 (35) Three themes were abstracted. In the degarelix vs placebo groups, Attitudes about treatment 15 vs 9 responded “yes” and 9 vs Positive attitude 6 (23) 1 (4) 17 responded “no.” −4 [95% CI, −12 to 4]; P = .55) did not differ significantly be- adverse events among help-seeking men with pedophilic tween groups (eTable 5 in Supplement 2). disorder, both in the short (2-week) and in the medium The most commonly reported moderate adverse event was (10-week) terms. The drug was also effective among high- transient injection site reactions at 2 weeks (degarelix: 22 of risk participants (Table 3). The rapid onset of degarelix ap- 25 [88%]; placebo: 1 of 26 [4%]), and the most commonly re- pears to have a crucial advantage compared with earlier medi- ported minor adverse event was hepatobiliary enzyme level cations for paraphilic disorders, which hada1to3months’ elevations, the largest being 3.5 times the upper bound of the lag in exerting their effects on sexuality. normal range (degarelix: 11 of 25 [44%]; placebo: 2 of 26 [8%]). The self-reports provided an empirical basis for the A full description of adverse events and blood sample results patient side of shared decision-making and may facilitate pa- is found in eTable 2 and eTable 3 in Supplement 2. tient-centered care for pedophilic disorder. In weighing the benefits and harms of the drug, we found that the partici- pants self-reported a more positive than negative attitude to- ward treatment, specifically regarding the effects on sexual- Discussion ity. Thus, participants expressed relief of symptoms for which In this phase 2 randomized clinical trial, a single dose (240 mg) they sought help, in addition to experiencing the treatment aim of degarelix acetate statistically significantly reduced the of risk reduction. This finding may also be reflected in the re- dynamic risk factor scores for sexual offense with minimal duced risk scores for pedophilic disorder and sexual preoccu- jamapsychiatry.com (Reprinted) JAMA Psychiatry September 2020 Volume 77, Number 9 903 Research Original Investigation Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder pation (Table 3). Only 1 participant was lost to follow-up, and between groups indicated a decrease (10 weeks: −4 [95% CI, 58% of those randomized to receive degarelix wished to con- −12 to 4]) among participants who were randomized to re- tinue treatment (Table 4), which indicates to us a potential role ceive degarelix. When including the MADRS-S scores of all par- for long-term treatment along with psychosocial support in ticipants, we found that the difference at 10 weeks increased most participants. Ultimately, the treatment decision be- (data not shown). longs to the physician, who should take into consideration the In addition, future studies need to identify the risk fac- risk for abuse, patient preferences, and drug benefits and tors that benefit from other kinds of intervention, such as psy- harms. In view of the participant wishes and effects ex- chotherapy. The impaired self-regulation and low empathy pressed in the self-reports, we believe degarelix should be con- domains in the present trial highlighted the residual risk not sidered for help-seeking individuals with pedophilic disor- amenable to degarelix treatment. Abnormal results in these der. The maintained motivation among participants for such measures in a clinical setting are usually associated with potent therapy and the low EQ-VAS ratings reflect the sever- attention-deficit/hyperactivity disorder or autistic features. We think that if these conditions are adequately addressed, the ity and associated distress of the condition. However, a mod- est effect on quality of life associated with treating the core potential exists to both improve health and reduce the risk for symptoms of pedophilic disorder indicates the need to better committing child sexual offense. understand the reasons for the struggles experienced by those with this condition. Strengths and Limitations Future research needs to address the effects and predict- This trial has some strengths. The high inclusion rates, able long-term adverse effects of hormone deficiency as well diverse geographic origin of participants, and comorbidity as the sometimes excessive effects on sexuality, as reported suggest the generalizability of study results to other help- by the participants (Table 4; eFigures 3 and 4 in Supple- seeking populations. Although few participants reported prior ment 2). Incremental add-back therapy of hormones could be contact offenses, the results are in line with findings of pre- considered in this regard. Given that 2 participants re- vious unblinded cohorts of convicted offenders, which indi- ported severe adverse events of suicidal ideation, vigilance for cated efficacy of testosterone suppression in reducing sexual 12,14 the risk of exacerbating suicidality in predisposed individu- symptoms in this population. als is warranted. Although depressive symptoms sometimes However, this study also has some limitations. Its results are manifestations of hypogonadism, a 12-month open-label pertain to only men, the risk measure relies mainly on self- trial of degarelix in patients with prostate cancer (n = 409) re- reports, and the findings have not yet been validated against ported depression as an adverse event in only 1 individual. actual abuse rates. Therefore, the number of child sexual abuse The high baseline prevalence of depression in the present cases that the reduced risk factor scores have prevented re- trial (35%) may indicate an enriched sample of individuals sus- mains to be determined. Furthermore, the prespecified clini- ceptible to depressive deterioration from gonadotropin- cally significant reduction of 5 points was not reached, with releasing hormone antagonist treatment. However, most par- the exception of participants who were classified as high risk ticipants may not be susceptible. In post hoc comparisons of (Table 3). the treatment groups over the study period (eTable 5 in Supple- ment 2), no differences in change of Montgomery-Åsberg De- pression Rating Scale self-rating (MADRS-S) version scores were Conclusions observed in depression severity, incidence of dysthymia, de- pression, or suicide risk. Although uncertainty from insuffi- Treatment with degarelix appeared to decrease the scores of cient power and post hoc analysis remained, most estimates risk factors for child sexual abuse after 2 weeks of administra- within the CI of the difference in MADRS-S scores at 10 weeks tion for help-seeking men with pedophilic disorder. ARTICLE INFORMATION Statistical analysis: Bottai, Rahm. collection, management, analysis, and Obtained funding: Arver, Rahm. interpretation of the data; preparation, review, or Accepted for Publication: February 9, 2020. Administrative, technical, or material support: Arver, approval of the manuscript; and decision to submit Published Online: April 29, 2020. Rahm. the manuscript for publication. doi:10.1001/jamapsychiatry.2020.0440 Supervision: Arver, Rahm. Data Sharing Statement: See Supplement 3. Open Access: This is an open access article Conflict of Interest Disclosures: None reported. Additional Contributions: Pia Jaensen and distributed under the terms of the CC-BY License. Funding/Support: This study was funded by the Susanne Jarlvik Alm, both from Karolinska © 2020 Landgren V et al. JAMA Psychiatry. Gothenburg Society of Medicine; the Swedish University Hospital, assisted with trial organization Author Contributions: Dr Rahm was the principal Society of Medicine; the Söderström König and data collection; Anna Fenander Hedin, investigator. Dr Rahm had full access to all of the Foundation; the Fredrik and Ingrid Thuring Karolinska University Hospital, assisted with data data in the study and takes responsibility for the Foundation; the Center for Psychiatry Research, collection; and Niklas Juth, Karolinska Institutet, integrity of the data and the accuracy of the data Department of Clinical Neuroscience, Karolinska assisted with data analysis. These individuals analysis. Institutet; the Research Unit of Skaraborg Hospital; received no additional compensation, outside of Concept and design: Arver, Rahm. and nonpharmaceutical academic sources, as well their usual salary, for their contributions. Acquisition, analysis, or interpretation of data: as by grants under the ALF agreement between the All authors. Swedish government and the country councils. REFERENCES Drafting of the manuscript: Landgren, Malki, Rahm. Role of the Funder/Sponsor: The funders had no 1. Collin-Vézina D, Daigneault I, Hébert M. Lessons Critical revision of the manuscript for important role in the design and conduct of the study; learned from child sexual abuse research: intellectual content: Landgren, Bottai, Arver, Rahm. 904 JAMA Psychiatry September 2020 Volume 77, Number 9 (Reprinted) jamapsychiatry.com Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Original Investigation Research prevalence, outcomes, and preventive strategies. blockers? Urology. 2008;71(6):1001-1006. doi:10. Swedish version. February 2009. Accessed Child Adolesc Psychiatry Ment Health. 2013;7(1):22. 1016/j.urology.2007.12.070 December 13, 2018. http://www.viss.nu/Global/ doi:10.1186/1753-2000-7-22 Blanketter/MINI_vers_6.pdf 14. Rösler A, Witztum E. Treatment of men with 2. Stoltenborgh M, van Ijzendoorn MH, Euser EM, paraphilia with a long-acting analogue of 27. EuroQol Group. EuroQol–a new facility for the Bakermans-Kranenburg MJ. A global perspective on gonadotropin-releasing hormone. N Engl J Med. measurement of health-related quality of life. child sexual abuse: meta-analysis of prevalence 1998;338(7):416-422. doi:10.1056/ Health Policy Amst Neth. 1990;16(3):199-208. around the world. Child Maltreat. 2011;16(2):79-101. NEJM199802123380702 doi:10.1016/0168-8510(90)90421-9 doi:10.1177/1077559511403920 15. Sorrentino R. Degarelix: an antagonist to 28. Burström K, Sun S, Gerdtham U-G, et al. 3. Walsh K, Zwi K, Woolfenden S, Shlonsky A. GnRH–theoretical and treatment considerations in Swedish experience-based value sets for EQ-5D School-based education programmes for the paraphilia. J Sex Med. 2012;9(1):327-329. doi:10.1111/ health states. Qual Life Res. 2014;23(2):431-442. prevention of child sexual abuse. Cochrane j.1743-6109.2011.02540.x doi:10.1007/s11136-013-0496-4 Database Syst Rev. 2015;(4):CD004380. doi:10. 16. Hanson RK, Morton-Bourgon KE. 29. Wood KL. The Medical Dictionary for Drug 1002/14651858.CD004380.pub3 The characteristics of persistent sexual offenders: Regulatory Affairs (MEDDRA) project. 4. Khan O, Ferriter M, Huband N, Powney MJ, a meta-analysis of recidivism studies. J Consult Clin Pharmacoepidemiol Drug Saf. 1994;3(1):7-13. doi:10. Dennis JA, Duggan C. Pharmacological Psychol. 2005;73(6):1154-1163. doi:10.1037/0022- 1002/pds.2630030105 interventions for those who have sexually offended 006X.73.6.1154 30. Sandelowski M. Whatever happened to or are at risk of offending. Cochrane Database of 17. Jordan K, Fromberger P, Stolpmann G, qualitative description? Res Nurs Health. 2000;23 Systematic Reviews. Intervention version published Müller JL. The role of testosterone in sexuality and (4):334-340. doi:10.1002/1098-240X(200008)23: February 18, 2015. Accessed September 9, 2019. paraphilia–a neurobiological approach. Part I: 4<334::AID-NUR9>3.0.CO;2-G 5. Långström N, Enebrink P, Laurén E-M, testosterone and sexuality. J Sex Med. 2011;8(11): 31. Pocock SJ. Group sequential methods in the Lindblom J, Werkö S, Hanson RK. Preventing sexual 2993-3007. doi:10.1111/j.1743-6109.2011.02394.x design and analysis of clinical trials. Biometrika. abusers of children from reoffending: systematic 18. van Honk J, Schutter DJ, Bos PA, Kruijt A-W, 1977;64(2):191-199. doi:10.1093/biomet/64.2.191 review of medical and psychological interventions. Lentjes EG, Baron-Cohen S. Testosterone 32. Svanborg P, Asberg M. A comparison between BMJ. 2013;347:f4630. doi:10.1136/bmj.f4630 administration impairs cognitive empathy in the Beck Depression Inventory (BDI) and the 6. Knack N, Winder B, Murphy L, Fedoroff JP. women depending on second-to-fourth digit ratio. self-rating version of the Montgomery Asberg Primary and secondary prevention of child sexual Proc Natl Acad SciUSA. 2011;108(8):3448-3452. Depression Rating Scale (MADRS). J Affect Disord. abuse. Int Rev Psychiatry. 2019;31(2):181-194. doi:10.1073/pnas.1011891108 2001;64(2-3):203-216. doi:10.1016/S0165-0327(00) doi:10.1080/09540261.2018.1541872 19. Aluja A, García LF, Martí-Guiu M, et al. 00242-1 7. Seto MC. Pedophilia and Sexual Offending Interactions among impulsiveness, testosterone, 33. Bergman H, Källmén H. Alcohol use among Against Children: Theory, Assessment, and sex hormone binding globulin and androgen Swedes and a psychometric evaluation of the Intervention. 2nd ed. American Psychological receptor gene CAG repeat length. Physiol Behav. alcohol use disorders identification test. Alcohol. Association; 2018. 2015;147:91-96. doi:10.1016/j.physbeh.2015.04.022 2002;37(3):245-251. doi:10.1093/alcalc/37.3.245 8. Beier KM, Ahlers CJ, Goecker D, et al. Can 20. Jordan K, Fromberger P, Stolpmann G, 34. Nesvåg R, Lange EH, Faerden A, et al. The use pedophiles be reached for primary prevention of Müller JL. The role of testosterone in sexuality and of screening instruments for detecting alcohol and child sexual abuse? First results of the Berlin paraphilia–a neurobiological approach. Part II: other drug use disorders in first-episode psychosis. Prevention Project Dunkelfeld (PPD). J Forensic testosterone and paraphilia. J Sex Med. 2011;8(11): Psychiatry Res. 2010;177(1-2):228-234. doi:10.1016/ Psychiatry Psychol. 2009;20(6):851-867. doi:10. 3008-3029. doi:10.1111/j.1743-6109.2011.02393.x j.psychres.2010.01.007 1080/14789940903174188 21. Spector IP, Carey MP, Steinberg L. The sexual 35. Kessler RC, Adler L, Ames M, et al. The World 9. Thibaut F, De La Barra F, Gordon H, Cosyns P, desire inventory: development, factor structure, Health Organization Adult ADHD Self-Report Scale Bradford JM; WFSBP Task Force on Sexual and evidence of reliability. J Sex Marital Ther. 1996; (ASRS): a short screening scale for use in the Disorders. The World Federation of Societies of 22(3):175-190. doi:10.1080/00926239608414655 general population. Psychol Med. 2005;35(2):245- Biological Psychiatry (WFSBP) guidelines for the 22. Reid RC, Garos S, Carpenter BN. Reliability, 256. doi:10.1017/S0033291704002892 biological treatment of paraphilias. World J Biol validity, and psychometric development of the 36. Barry MJ, Edgman-Levitan S. Shared decision Psychiatry. 2010;11(4):604-655. doi:10.3109/ Hypersexual Behavior Inventory in an outpatient making–pinnacle of patient-centered care. N Engl J sample of men. Sex Addict Compulsivity. 2011;18(1): Med. 2012;366(9):780-781. doi:10.1056/ 10. Lee JY, Cho KS. Chemical castration for sexual 30-51. doi:10.1080/10720162.2011.555709 NEJMp1109283 offenders: physicians’ views. J Korean Med Sci. 23. Conners CK. Conners’ Continuous Performance 37. Carr BR, Stewart EA, Archer DF, et al. Elagolix 2013;28(2):171-172. doi:10.3346/jkms.2013.28.2.171 Test II (CPT II V.5). Multi-Health Systems, Inc; 2000. alone or with add-back therapy in women with 11. Turner D, Petermann J, Harrison K, Krueger R, 24. Eriksson JM, Andersen LM, Bejerot S. heavy menstrual bleeding and uterine leiomyomas: Briken P. Pharmacological treatment of patients RAADS-14 Screen: validity of a screening tool for a randomized controlled Trial. Obstet Gynecol. with paraphilic disorders and risk of sexual autism spectrum disorder in an adult psychiatric 2018;132(5):1252-1264. doi:10.1097/AOG. offending: an international perspective. World J Biol population. Mol Autism. 2013;4(1):49. doi:10.1186/ 0000000000002933 Psychiatry. 2019;20(8):616-625. doi:10.1080/ 2040-2392-4-49 38. Klotz L, Boccon-Gibod L, Shore ND, et al. 15622975.2017.1395069 25. Baron-Cohen S, Wheelwright S, Hill J, Raste Y, The efficacy and safety of degarelix: a 12-month, 12. Turner D, Briken P. Treatment of paraphilic Plumb I. The “Reading the Mind in the Eyes” Test comparative, randomized, open-label, disorders in sexual offenders or men with a risk revised version: a study with normal adults, and parallel-group phase III study in patients with of sexual offending with luteinizing adults with Asperger syndrome or high-functioning prostate cancer. BJU Int. 2008;102(11):1531-1538. hormone-releasing hormone agonists: an updated autism. J Child Psychol Psychiatry. 2001;42(2):241- doi:10.1111/j.1464-410X.2008.08183.x systematic review. J Sex Med. 2018;15(1):77-93. 251. doi:10.1111/1469-7610.00715 doi:10.1016/j.jsxm.2017.11.013 26. Sheehan D, Lecrubier YMINI. Mini 13. van Poppel H, Nilsson S. Testosterone surge: Internationell Neuropsykiatrisk Intervju 6.0.0. rationale for gonadotropin-releasing hormone jamapsychiatry.com (Reprinted) JAMA Psychiatry September 2020 Volume 77, Number 9 905 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JAMA Psychiatry Pubmed Central

Effect of Gonadotropin-Releasing Hormone Antagonist on Risk of Committing Child Sexual Abuse in Men With Pedophilic Disorder

JAMA Psychiatry , Volume 77 (9) – Apr 29, 2020

Loading next page...
 
/lp/pubmed-central/effect-of-gonadotropin-releasing-hormone-antagonist-on-risk-of-qTrGrDXc9f

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Pubmed Central
Copyright
Copyright 2020 Landgren V et al. JAMA Psychiatry.
ISSN
2168-622X
eISSN
2168-6238
DOI
10.1001/jamapsychiatry.2020.0440
Publisher site
See Article on Publisher Site

Abstract

Research JAMA Psychiatry | Original Investigation Effect of Gonadotropin-Releasing Hormone Antagonist on Risk of Committing Child Sexual Abuse in Men With Pedophilic Disorder A Randomized Clinical Trial Valdemar Landgren, MD; Kinda Malki, MD; Matteo Bottai, ScD; Stefan Arver, MD; Christoffer Rahm, MD Editorial page 893 IMPORTANCE Evidence-based treatments from randomized clinical trials for pedophilic Supplemental content disorder are lacking. OBJECTIVE To determine whether a gonadotropin-releasing hormone antagonist reduces dynamic risk factors for committing child sexual abuse. DESIGN, SETTING, AND PARTICIPANTS This academically initiated, double-blind, placebo-controlled, parallel-group, phase 2 randomized clinical trial was conducted at the ANOVA center in Stockholm, Sweden, from March 1, 2016, to April 30, 2019. Individuals who contacted PrevenTell, the national telephone helpline for unwanted sexuality, were recruited. Eligible participants were men seeking help aged 18 to 66 years with a pedophilic disorder diagnosis and no contraindications to the intervention. The primary end point was assessed by intent-to-treat analysis. INTERVENTIONS Randomization to receive either 2 subcutaneous injections of 120 mg of degarelix acetate or equal volume of placebo. MAIN OUTCOMES AND MEASURES The primary end point was the mean change between baseline and 2 weeks in the composite risk score of 5 domains of child sexual abuse ranging from 0 to 15 points; each domain could be rated from 0 to 3 points. Secondary end points included efficacy at 2 and 10 weeks as measured by the composite score, each risk domain, quality of life, self-reported effects, and adverse events. RESULTS A total of 52 male participants (mean [SD] age, 36 [12] years) were randomized to receive either degarelix (n = 25; with 1 withdrawal) or placebo (n = 26). At 2 weeks, the composite risk score decreased from 7.4 to 4.4 for participants in the degarelix group and from 7.8 to 6.6 for the placebo group, a mean between-group difference of –1.8 (95% CI, –3.2 to –0.5; P = .01). A decrease was seen in the composite score at 10 weeks (−2.2 [95% CI, −3.6 to −0.7]) as well as in the domains of pedophilic disorder (2 weeks: −0.7 [95% CI, −1.4 to 0.0]; 10 weeks: −1.1 [95% CI, −1.8 to −0.4]) and sexual preoccupation (2 weeks: −0.7 [95% CI, −1.2 to −0.3]; 10 weeks: −0.8 [95% CI, −1.3 to −0.3]) in the degarelix group compared with the placebo group. No difference was seen for the domains of self-rated risk (2 weeks: −0.4 [95% CI, −0.9 to 0.1]; 10 weeks: −0.5 [95% CI, −1 to 0.0]), low empathy (2 weeks: 0.2 [95% Author Affiliations: Institute of CI, −0.3 to 0.6]; 10 weeks: 0.2 [95% CI, −0.2 to 0.6]), and impaired self-regulation (2 weeks: Neuroscience and Physiology, −0.0 [95% CI, −0.7 to 0.6]; 10 weeks: 0.1 [95% CI, −0.5 to 0.8]), or quality of life (EuroQol 5 Gothenburg University, Gothenburg, Sweden (Landgren); Karolinska Dimensions questionnaire index score, 2 weeks: 0.06 [95% CI, −0.00 to 0.12], and 10 weeks: University Hospital, Stockholm, 0.04; 95% CI, −0.02 to 0.10; EuroQol visual analog scale, 2 weeks: 0.6 [95% CI, −9.7 to 10.9], Sweden (Malki); Institute of and 10 weeks: 4.2 [95% CI, −6.0 to 14.4]). Two hospitalizations occurred from increased Environmental Medicine, Karolinska Institutet, Stockholm, Sweden suicidal ideation, and more injection site reactions (degarelix: 22 of 25 [88%]; placebo: 1 of 26 (Bottai); Department of Medicine [4%]) and hepatobiliary enzyme level elevations were reported by participants who received Huddinge, Karolinska Institutet, degarelix (degarelix: 11 of 25 [44%]; placebo: 2 of 26 [8%]). Among the 26 participants Stockholm, Sweden (Arver); randomized to receive degarelix, 20 (77%) experienced positive effects (eg, improved Centre for Psychiatry Research, Department of Clinical Neuroscience, attitude or behavior) on sexuality and 23 (89%) reported adverse effects on the body. Karolinska Institutet, Stockholm, Sweden (Rahm); Stockholm Health CONCLUSION AND RELEVANCE This trial found that degarelix reduced the risk score for Care Services, Region Stockholm, committing child sexual abuse in men with pedophilic disorder 2 weeks after initial injection, Stockholm, Sweden (Rahm). suggesting use of the drug as a rapid-onset treatment option. Further studies are warranted Corresponding Author: Christoffer into the effects and long-term adverse effects of hormone deficiency. Rahm, MD, Stockholm Health Care Services, Region Stockholm, TRIAL REGISTRATION EU Clinical Trials Register Identifier: 2014-000647-32 Norra Stationsgatan 69, SE-113 64 JAMA Psychiatry. 2020;77(9):897-905. doi:10.1001/jamapsychiatry.2020.0440 Stockholm, Sweden Published online April 29, 2020. (christoffer.rahm@ki.se). (Reprinted) 897 Research Original Investigation Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder hild sexual abuse affects 1 in 5 girls and 1 in 10 boys worldwide. It is accompanied by adverse psychoso- Key Points 1,2 C cial outcome across the life span. Preventive mea- Question Can a gonadotropin-releasing hormone antagonist sures have been advocated, but to date evidence for inter- rapidly reduce the risk for committing child sexual abuse in men 4,5 ventions has been limited. with pedophilic disorder who are seeking help? The estimated proportion of child sexual offenders who Findings In this randomized clinical trial of 52 men with 1,4 are prosecuted is 1%. Of those who were prosecuted, up to pedophilic disorder, treatment with degarelix statistically 95% were first-time offenders and half of them had pedo- significantly reduced the risk for committing child sexual abuse philic disorder, defined as recurrent sexual attraction to pre- 2 weeks after the initial injection. pubescent children associated with distress or negative con- Meaning This finding suggests that degarelix may serve as a sequences. Studies have found that not all men with pedophilic rapid-onset, risk-reducing medication for men with pedophilic disorder commit a sexual offense, but those who do gener- disorder. ally report struggling with their sexual urges for 10 years be- 6,8 fore committing a sexual crime. Consequently, an opportu- nity for prevention exists in treating high-risk individuals lenges associated with conducting a trial in a hard-to-reach popu- without prior convictions. Effective treatment could prevent lation along with the issues of tolerability of previous therapies, child sexual abuse and reduce psychosocial stress for the in- we performed ancillary interviews of self-reported experiences dividual with pedophilic disorder. of treatment. We regarded using sexually arousing material and Currently recommended interventions include psycho- measuring participants’ penile responses as too intrusive. therapy and antidepressants as well as testosterone-suppressing We hypothesized that men with pedophilic disorder who medicationsforhigh-riskindividuals. Opinionsabouttreatment were randomized to receive degarelix compared with pla- are vehement. In many countries, an informed consent proce- cebo would have a substantial reduction in risk of commit- dure is required for chemical castration. However, chemical cas- ting child sexual abuse after 2 weeks. Furthermore, we hy- tration is used as a compulsory legal sentence for child sexual pothesized that degarelix would be sufficiently tolerated offenders in some jurisdictions in the US, Asia, and Europe but by the participants and thus would be a safe and effective is prohibited in other countries owing to ethical concerns of rapid-onset treatment option. 10,11 coercion and uncertain efficacy. Because they lower testosterone through receptor desen- sitization, gonadotropin-releasing hormone agonists are con- Methods sidered effective in reducing paraphilic symptoms. How- ever, the use of these agonists is limited to supervised This randomized clinical trial was approved by the Swedish correctional settings because of the lack of randomized clini- Central Ethical Review Board and the Swedish Medical Prod- cal trials; their metabolic adverse effects; and the initial flare-up ucts Agency (trial protocol in Supplement 1). All participants of testosterone, which may be associated with increased ag- signed an informed consent agreement and were offered treat- gression and libido that require concurrent antiandrogen ment after the study. To ensure adherence to good clinical medication. Experience from the Swedish helpline Preven- practice, the Karolinska Trial Alliance, a research center that Tell suggests a need for both rapid-acting short-term treat- supports clinical trials, provided independent monitoring and ment (eg, in critical phases of the disorder to quickly control recommended a longer follow-up period to monitor the safety sexual impulses or reduce a high degree of struggling) and of the intervention; hence, a second follow-up visit with the long-term treatment. Degarelix acetate is a gonadotropin- same outcome measures was conducted at 10 weeks. The study releasing hormone antagonist that was approved by the US was conducted without any kind of collaboration with the phar- Food and Drug Administration in 2008 for treating advanced maceutical industry. This trial followed the Consolidated Stan- prostate cancer. Degarelix decreases testosterone to castra- dards of Reporting Trials (CONSORT) reporting guideline. tion levels within 3 days without testosterone flare ; there- fore, the drug could serve as a rapid-onset treatment for indi- Trial Design and Participants 9,14,15 viduals seeking help in outpatient settings. This academically initiated, double-blind, placebo- For individuals without a prior conviction, no validated controlled, parallel-group phase 2 trial with balanced random- measures of risk exist. Therefore, the PRIOTAB (Pedophilia at ization (1:1) was conducted at the ANOVA center, a highly Risk–Investigations of Treatment and Biomarkers) project, of specialized center for andrology and sexual medicine in Stock- which this present trial is a part, included the construction holm, Sweden, from March 1, 2016, to April 30, 2019. The trial of a composite score based on previous observational studies was conducted conjointly with a case-control study, for which of dynamic (ie, potentially changeable over time) risk factors participants underwent magnetic resonance imaging and for sexual offense recidivism. One such risk factor is deviant additional testing at the same time and venue. The ANOVA sexual interest (eg, pedophilic disorder). Other factors are center hosts PrevenTell, a national telephone helpline for self- sexual preoccupation, impaired self-regulation, and low identified unwanted sexuality, through which trial partici- empathy. Because these 4 factors are all possibly mitigated pants were recruited. In Sweden, the legal obligation to by testosterone suppression, we believed that degarelix could report suspected child abuse supersedes professional confi- 17-20 have a risk-reducing effect. Bearing in mind the chal- dentiality, which may make individuals with pedophilic dis- 898 JAMA Psychiatry September 2020 Volume 77, Number 9 (Reprinted) jamapsychiatry.com Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Original Investigation Research order reluctant to seek help. To minimize compromises in re- Figure. CONSORT Diagram cruitment rates and to attain authenticity in self-reports, we provided participants a temporary identification number. This 623 Individuals called PrevenTell helpline identification number kept participants anonymous to trial as- 523 Consented to answering a structured questionnaire and screened for eligibility sessors (S.A., C.R.), although they were still informed of the professionals’ obligation, according to the Swedish Social Ser- 468 Excluded vices Act, to send a notification of concern to the authorities 404 Had no sexual attraction to children when they suspect a child is at risk of abuse or maltreatment 54 Had sexual attraction and the legal possibility of reporting to the police suspected to children but did not fulfill inclusion criteria perpetrators of crimes against children. and/or met exclusion criteria (most commonly Participants were recruited from March 1, 2016, to Janu- hebephilia) ary 31, 2019. A flow diagram of the participants through each stage of the trial is shown in the Figure. Eligible participants 65 Interviewed by telephone were help-seeking, self-identified men aged 18 to 66 years with a pedophilic disorder diagnosis, as defined in the Diagnostic 10 Excluded 3 Had negative result for and Statistical Manual of Mental Disorders (Fifth Edition). A full pedophilic disorder list of inclusion and exclusion criteria is provided in eAppen- 6 Declined 1 Underwent recent dix1in Supplement 2. Participants recruited through Preven- antiandrogen therapy Tell were offered transportation costs for study visits and financial compensation of Sk 1000 (Swedish krona) (US $100 55 Underwent screening at ANOVA center before taxation) after study completion. 3 Excluded 1 Had severe psychotic Outcome Measures symptoms The primary end point was the mean change in composite 1 Had liver abnormalities risk score (range, 0-15 points) between baseline and 2 weeks 1 Withdrew consent after randomization. The composite risk score consisted of 5 domains: the 4 empirically derived dynamic risk factors 52 Randomized (pedophilic disorder, sexual preoccupation, impaired self- regulation, and low empathy) and self-rated risk, each of 25 Received degarelix 26 Received placebo which could be rated from 0 to 3 points (Table 1; eAppendix 1 1 Withdrew consent before injection and declined follow-up in Supplement 2). Each domain was weighted equally in the score. Three risk groups were prespecified (risk class 1: 0-5 23 Included in analysis of primary 26 Included in analysis of primary points; risk class 2: 6-10 points; risk class 3: 11-15 points). end point at 2 weeks end point at 2 weeks 1 Did not receive intervention We used the prespecified analysis for the primary end 2 Had incomplete baseline point. This analysis included only participants with complete self-ratings data at baseline and 2 weeks after injection. The secondary end point analyses excluded no data. Sec- 24 Completed follow-up at 10 weeks 26 Completed follow-up at 10 weeks 1 Did not receive intervention ondary end points were efficacy at 2 and 10 weeks in terms of 1 Hospitalized for suicidal ideation reduction in the composite risk score and its 5 domains, effi- cacy in participants in risk class 3 (11-15 points), quality of life, 26 Included in analyses 26 Included in analyses of secondary end points of secondary end points adverse events, and self-reported effects. Quality of life was measured at all time points, using the EuroQol 5 Dimensions (EQ-5D) questionnaire, which consists of a preference-based health status measure convertible into index scores ranging improved attitudes or behaviors) and negative (eg, adverse events) effects of the injection and willingness to maintain the from 0 to 1 (with the higher score indicating better health sta- tus), and a EuroQol visual analog scale (EQ-VAS) of 0 to 100, experienced effects with further injections. Interviews were 27,28 analyzed using qualitative descriptive content analysis, as de- by which higher scores represent better health status. Participants were given a study diary and instructed to note scribed in eAppendix 2 in Supplement 2. Neither the asses- sors nor the participants were aware of treatment randomiza- any adverse events between assessments and to call the study nurse if they perceived serious harm from treatment. At follow- tion at the time of the interviews or qualitative analysis. up, physical adverse events were registered by the study nurse using both participant responses to open-ended questions Trial Intervention about current health status and notes in the diary. Adverse On the first visit, participants underwent eligibility and base- events were documented by the assessing psychiatrist (C.R.) line evaluations and received the study drug (2 subcutane- also, and all events were coded according to the MedDRA ous injections of 120 mg degarelix acetate or equal volume of (Medical Dictionary for Drug Regulatory Affairs). Self- placebo). A computer-generated sequence with permuted block reported experiences were collected in a structured inter- randomization was provided by the Karolinska Trial Alliance. view, including open-ended questions about the positive (eg, The randomization sequence was concealed from one of us jamapsychiatry.com (Reprinted) JAMA Psychiatry September 2020 Volume 77, Number 9 899 Research Original Investigation Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Table 1. Composite Score of Dynamic Risk for Committing Child Sexual Abuse Score definition Risk domain 01 2 3 Pedophilic disorder No pedophilic Pedophilic attraction Pedophilic attraction + distress or Pedophilic attraction + distress + attraction negative consequences negative consequences Sexual preoccupation Hyposexual according Not hyposexual according to Not hyposexual + hypersexual Not hyposexual + hypersexual + to the SDI the SDI according to the HBI; no ongoing ongoing abusive behavior abusive behavior according to the SChiMRA-B Impaired Normal CCPT II result 1 Abnormal CCPT II domain out 2 Abnormal CCPT II domains of the All 3 abnormal CCPT II domains self-regulation of the inattention, impulsivity, inattention, impulsivity, and of inattention, impulsivity, and and vigilance domains vigilance domains vigilance Low empathy No abnormality 1 of RAADS-14 mentalizing 2 of RAADS-14 mentalizing domain RAADS-14 mentalizing domain domain >10, RMET <22, or >10, RMET <22, or current >10, + RMET<22, + current current antisocial behavior antisocial behavior antisocial behavior according to the MINI Self-rated risk Normal SChiMRA-A 1 Domain of SChiMRA-A watch, 2 Domains of SChiMRA-A watch, SChiMRA-A watch + socialize + result socialize, or interact domain socialize, or interact domain interact domains c 21 Abbreviations: CCPT II, Conners Continuous Performance Test; Hyposexuality is defined as a score of <45 on the SDI. Hypersexuality is HBI, Hypersexual Behavior Inventory (score range: 19-95, with higher scores defined as an HBI score of53. The SChiMRA-B (eAppendix 1 in indicating more severe hypersexual behavior); MINI, Mini International Supplement 2) assesses self-reported frequency of sexually abusive behavior Neuropsychiatric Interview; RAADS-14, Ritvo Autism and Asperger Diagnostic in the past week (never, several days, more than half of days, or almost every Scale, 14 Screen (score range: 0-42, of which the mentalizing domain ranges day) regarding watching of, socializing with, and sexual interaction with from 0 to 21, with higher scores indicating more autistic features); children, in which occurrence of any sort is scored as positive. RMET, Reading the Mind in the Eyes Test (score range: 0-36, with higher scores d 23 CCPT II for inattention, impulsivity, and vigilance. indicating better capacity for emotion recognition); SChiMRA-A, Sexual Child e 24 25 Self-ratings on the mentalization domains of the RAADS-14, RMET scores, Molestation Risk Assessment part A; SDI, Sexual Desire Inventory (score range: and current antisocial behavior as reported in the MINI. 12-104, with higher scores indicating increased sexual desire); VAS, visual analog The SChiMRA-A (eAppendix 1 in Supplement 2) consists of VAS ratings to the scale (score range: 0%-100%, with higher scores indicating increased self-rated question, “How likely is it that you would do any of the following, if there was risk). an easy way to do it without being caught?” regarding watching of, socializing For a full description of the composite risk score, see eAppendix 1 in with, and sexual interaction with children. A rating of 40% or higher on the Supplement 2. VAS was interpreted as a substantial risk. The 3 criteria for pedophilic disorder according to the DSM-5 are pedophilic interest, significant distress, and significant negative consequences. (C.R. and S.A.) who enrolled and assessed participants and then dition, we evaluated the trajectories of the secondary end transferred onto cards placed in sequentially numbered, points in the 2 treatment groups at the 3 visits, including all opaque envelopes stored in a locked cabinet in the dispen- randomized participants. The mean of the numeric second- sary and accessible to only 1 independent study nurse, who in ary end points was estimated using linear random-effects re- turn informed the nurse responsible for drug administration. gression models, including the treatment indicator (binary co- A detailed description of the injection procedure is provided variate), indicator variables for the 2 follow-up visits (binary), in eAppendix 1 in Supplement 2. Except for the study nurse and the 2 interaction terms between the treatment indicator who was responsible for treatment randomization and who also and the 2 visit indicators (binary). The models also included a registered the physical adverse effects, all of the assessors participant-specific random intercept, which was assumed to involved in the repeated outcome assessments were blinded follow a normal distribution. The random intercept was in- to the treatment allocation. cluded to consider the potential dependence in the repeated observations on each participant. Statistical Analysis We examined the differences in the time trajectories be- The sample size of this trial was determined from the experi- tween the treatment groups by testing the composite hypoth- ence of similar, previously published studies of biomarkers esis that the interaction terms were jointly equal to zero using of treatment response. We calculated that the planned sample Wald-type tests. We used the estimates from the models to es- size of 60 would give a power greater than 90% to detect a clini- timate the mean of the numeric end points. The SEs used to cally significant difference in the primary end point. We had calculate their CIs were obtained with the delta method. All no prespecified expected dropout rate. statistical analyses were performed using Stata, version 15 The primary end point was assessed by intent-to-treat (StataCorp LLC). analysis, using the unequal-variance, 2-sided t test at interim When 20 participants had completed the study, the analysis and at 2 weeks. To preserve the overall significance planned interim analysis of efficacy for the primary end point level of the test at .05, we chose a 2-sided P = .0294 (exact value and a review of the safety of the treatment were performed by given per the prespecified statistical plan). Participants for the sponsor (S.A.), who could find no statistically significant whom the score change could not be calculated (ie, missing differences between the groups. Because no evidence of sub- values for the risk score at baseline or 2 weeks) were excluded stantial harm from active treatment was found that would mo- from analysis of the primary end point under the assumption tivate the premature termination of the study, the study was that the missing-data generating process was at random. In ad- permitted to continue with the planned sample size of 60. 900 JAMA Psychiatry September 2020 Volume 77, Number 9 (Reprinted) jamapsychiatry.com Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Original Investigation Research Table 2. Demographic and Clinical Characteristics of Participants at Baseline No. (%) Variable Degarelix acetate (n = 26) Placebo (n = 26) Demographic characteristics Abbreviations: ASRS, Adult ADHD (attention-deficit/hyperactivity Age, median (IQR) [range], y 36 (25-39) [19-54] 35 (28-47) [18-66] disorder) Self-Report Scale screener Highest completed educational level (with a minimum of 4 of 6 questions Primary school, 9 y 2 (8) 3 (12) rated above cutoff indicating ADHD); IQR, interquartile range; MADRS, Secondary education, 12 y 12 (46) 13 (50) Montgomery-Åsberg Depression Postsecondary education 12 (46) 10 (38) Rating Scale (score range, 0-48 Unemployed 12 (46) 9 (35) points, with higher scores indicating increased depression severity); Living status MINI, Mini International Caregiver of child 7 (27) 12 (46) Neuropsychiatric Interview); Living without partner 17 (65) 17 (65) RAADS-14, Ritvo Autism and Asperger Diagnostic Scale, 14 Screen. Ever lived with a partner ≥2 y 10 (38) 15 (58) Two participants in the placebo Self-reported prior criminal conviction group and 2 in the degarelix group Noncontact sexual offense 4 (15) 4 (15) stated that the attraction had Contact sexual offense 2 (8) 3(12) “always been present.” Nonsexual offense 2 (8) 5 (19) Based on the Wechsler Adult Intelligence Scale 4 (with the score Sexuality of 100 indicating the mean Attraction primarily to boys 4 (15) 4 (15) intelligence of the population), Attraction primarily to girls 19 (73) 21 (81) MINI, MADRS-S self-rating version, Alcohol Use Disorder Attraction to boys and girls 3 (12) 1 (4) Identification Test (score of8; Attraction exclusively to prepubescent children 2 (8) 9 (35) score range: 0-40, with higher Age at discovery of attraction to minors, median (IQR), y 16 (14-23) 16 (13-25) scores indicating more alcohol b abuse) and the Drug Use Disorder Psychiatric characteristics Identification Test (score of3; Any psychiatric disorder 18 (69) 25 (96) score range: 0-44, with higher Ongoing depression 7 (27) 12 (46) scores indicating more drug 33,34 abuse), ASRS Self-Report Scale MADRS-S score in patients with depression, median (IQR) 25 (24-29) 24 (21-27) screener, and RAADS-14. Current psychotic symptoms 0 2 (8) As indicated by MINI, ASRS, and Previous manic episode 0 1 (4) RAADS-14 scores. Hazardous drug or alcohol use 4 (17) 11 (42) Other medications included sleep medications, antihistamines, Full-scale IQ, median (IQR) 103 (93-118) 99 (96-116) stimulants, and mood stabilizers. Psychoactive medication Composite score ranged from 0 to Antidepressants 7 (27) 8 (31) 15 points, with higher scores Other 7 (27) 5 (19) indicating higher risk. In an age-matched comparison sample of Testosterone, median (IQR), nmol/L 16.5 (11-18.5) 14.0 (10.8-18.3) men (n = 55), the median (IQR) Composite risk score, median (IQR) 7.5 (6.0-8.0) 8 (6.8-9.0) composite risk score was 2 (1-2.5). risk class 1 (n = 6) or risk class 2 (n = 44), but 3 individuals in each treatment group had a baseline score of 10 points and were Results therefore analyzed as a high-risk subgroup. The median (IQR) number of days from baseline to visits was 15 (14-18) days at Participants and Intervention Of the 52 male participants (mean [SD] age, 36 [12] years), 26 2 weeks and 73 (73-74) days at 10 weeks for those in the de- (50%; with 1 withdrawal) were randomized to receive degare- garelix group and 14 (14-14) days at 2 weeks and 73 (73-77) days lix and 26 (50%) to placebo. Of these 52 participants, 39 (75%) at 10 weeks for those in the placebo group. The mean (SD) cas- lived in urban areas of more than 50 000 inhabitants. The travel tration levels for serum testosterone were ascertained at distance from home to the study center was more than 100 km 2 weeks (0.7 [0.2] nmol/L) and 10 weeks (0.6 [0.2] nmol/L) for 28 participants (54%), 6 of whom (12%) traveled more than in all participants randomized to receive degarelix (eTable 4 400 km. Participant characteristics are described in Table 2. in Supplement 2). Minor differences in demographic and clinical characteristics Because the retention rate of participants at 10 weeks was were observed between the 2 treatment groups at baseline, but high (96%; n = 50), analysis of secondary end points was con- the median (interquartile range [IQR]) composite risk score was sidered feasible without reaching the planned sample size of about the same (degarelix: 7.5 [6.0-8.0] vs placebo: 8 [6.8- 60. This finding, along with a temporarily slow inclusion rate, 9.0]) (eTable 1 in Supplement 2). A high prevalence of depres- ended the trial enrollment after randomization of 52 partici- sion (35%) was found at baseline. All participants were in either pants, a decision we made before unblinding. jamapsychiatry.com (Reprinted) JAMA Psychiatry September 2020 Volume 77, Number 9 901 Research Original Investigation Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Table 3. Primary and Secondary End Points Degarelix acetate Placebo Difference (95% CI) Baseline 2wk 10 wk Baseline 2wk 10 wk P End points (n = 24) (n = 25) (n = 24) (n = 26) (n = 26) (n = 26) Baseline 2 wk 10 wk value Primary end points, mean (SE) Composite risk score 7.4 (0.3) 4.4 (0.6) NA 7.8 (0.3 6.6 (0.5) NA NA −1.8 NA .01 (−3.2 to −0.5) Secondary end points [all data], mean (delta method SE) Composite risk score 7.3 (0.5) 4.4 (0.5) 3.6 (0.5) 7.8 (0.5) 6.6 (0.5) 6.2 (0.5) −0.5 −1.8 −2.2 .01 (−1.8 to 0.8) (−3.2 to −0.3) (−3.6 to −0.7) High risk group 10 (1.2) 6.7 (1.2) 2.3 (1.2) 10 (1.2) 10 (1.2) 8.3 (1.2) 0.0 −3.3 −6 .04 [ = 10 points] (n=3+3) (−3.3 to 3.3) (−7.9 to 1.2) (−10.6 to −1.4) Risk domain Pedophilic disorder 2.4 (0.2) 0.9 (0.2) 0.8 (0.2) 2.6 (0.2) 1.8 (0.2) 2.1 (0.2) −0.2 −0.7 −1.1 .01 (−0.7 to 0.3) (−1.4 to 0.0) (−1.8 to −0.4) Sexual 1.6 (0.1) 0.7 (0.1) 0.4 (0.1) 1.6 (0.1) 1.4 (0.1) 1.2 (0.1) 0.0 −0.7 −0.8 .001 preoccupation (−0.4 to 0.4) (−1.2 to −0.3) (−1.3 to −0.3) Impaired 1.5 (0.2) 1.2 (0.2) 1.4 (0.2) 1.4 (0.2) 1.2 (0.2) 1.2 (0.2) 0.0 −0.0 0.1 .82 self-regulation (−0.5 to 0.6) (−0.7 to 0.6) (−0.5 to 0.8) Low empathy 0.8 (0.2) 1.1 (0.2) 0.8 (0.2) 1 (0.2) 1.1 (0.2) 0.8 (0.2) −0.2 0.2 0.2 .61 (−0.6 to 0.3) (−0.3 to 0.6) (−0.2 to 0.6) Self-rated risk 1.0 (0.2) 0.5 (0.2) 0.2 (0.2) 1.2 (0.2) 1.1 (0.2) 0.9 (0.2) −0.2 −0.4 −0.5 .16 (−0.8 to 0.3) (−0.9 to 0.1) (−1 to 0.0) Quality of life EQ-5D index score 0.79 0.83 0.82 0.86 0.83 0.85 −0.06 0.06 0.04 .16 (0.03) (0.03) (0.03) (0.0.3) (0.03) (0.03) (−1.37 to 0.01) (−0.00 to 0.12) (−0.02 to 0.10) EQ-VAS 59.8 59.0 61.3 60.5 59.1 57.8 −0.6 0.6 4.2 .68 (4.5) (4.4) (4.5) (4.5) (4.5) (4.5) (−13.0 to 11.7) (−9.7 to 10.9) (−6.0 to 14.4) Abbreviations: EQ-5D, EuroQol 5 Dimensions questionnaire (score range, P value is presented for a 2-sample t test with unequal variances for the 0-1, with a higher score indicating better health status); EQ-VAS, EuroQol visual primary end point and for a test for different time trajectories between groups analog scale questionnaire (score range, 0-100, with higher scores indicating in the random-effects regression models for all secondary end points. better health status); NA, not applicable. c Complete data for calculating the difference (baseline and 2 weeks) were Differences indicate the status of the participants randomized to receive available for 23 participants randomized to degarelix and 26 participants degarelix. randomized to placebo (Figure). 4.2 [95% CI, −6.0 to 14.4]) (eTable 6 in Supplement 2). Themes Outcomes Primary and secondary end points of efficacy in reducing the and categories from the ancillary interviews are displayed in composite risk score at 2 and 10 weeks differed substantially, Table 4. In the degarelix group, positive attitudes toward sexu- decreasing from 7.4 to 4.4 points for participants in the de- ality (20 of 26 [77%]) and adverse effects on the body (23 of garelix group and from 7.8 to 6.6 points for the placebo group 26 [89%]) were the most common self-reported experiences. (mean difference: –1.8 [95% CI, –3.2 to –0.5]; P = .01) (Table 3). Analysis of specific participant quotes is provided in eAppen- In the regression models of secondary end points, statisti- dix2in Supplement 2. cally significant differences were observed in composite risk Post hoc analyses revealed that 15 of the 26 participants scores (2 weeks: –1.8 [95% CI, –3.2 to –0.5]; 10 weeks: −2.2 [95% (58%) in the degarelix group and 3 of the 26 participants (12%) CI, −3.6 to −0.7]) (eFigure 1 and eFigure 2 in Supplement 2) and in the placebo group denied sexual attraction to minors at high-risk group (2 weeks: −3.3 [95% CI, −7.9 to 1.2]; 10 weeks: 10 weeks. −6 [95% CI, −10.6 to−1.4]), between the pedophilic disorder (2 weeks: −0.7 [95% CI, −1.4 to 0.0]; 10 weeks: −1.1 [95% CI, Adverse Events −1.8 to −0.4]) and sexual preoccupation (2 weeks: −0.7 [95% No serious adverse events occurred in the placebo group. A se- CI, −1.2 to −0.3]; 10 weeks: −0.8 [95% CI, −1.3 to −0.3]) (eFig- rious adverse event of increased suicidal ideation was re- ure 3 and eFigure 4 in Supplement 2) domains. No statisti- ported by 2 of 25 participants (8%) in the degarelix group, which cally significant differences in scores were seen in the do- led to hospitalization. One participant reported the event at mains of low empathy (2 weeks: 0.2 [95% CI, −0.3 to 0.6]; 10 weeks, when his condition had improved; the other re- 10 weeks: 0.2 [95% CI, −0.2 to 0.6]), impaired self-regulation ported suicidal ideation at baseline but was lost to follow-up (2 weeks: −0.0 [95% CI, −0.7 to 0.6]; 10 weeks: 0.1 [95% CI, at 10 weeks (purportedly also owing to a travel distance of >500 −0.5 to 0.8]), and self-rated risk (2 weeks: −0.4 [95% CI, −0.9 km) and was contacted by telephone. In post hoc analyses, rates to 0.1]; 10 weeks: −0.5 [95% CI, −1 to 0.0]) or in either mea- of suicidality (2 weeks: −0.3 [95% CI, −0.5 to 0.1]; 10 weeks: sure of quality of life (EQ-5D index score, 2 weeks: 0.06 [95% −0.1 [95% CI, −0.5 to 0.3]; P = .33), depression (2 weeks: 2.3 CI, −0.00 to 0.12], and 10 weeks: 0.04 [95% CI, −0.02 to 0.10]; [95% CI, 0.3-20.7]; 10 weeks: 2.0 [95% CI, 0.2-19.5]; P =.74), EQ-VAS, 2 weeks: 0.6 [95% CI, −9.7 to 10.9], and 10 weeks: or depression severity (2 weeks: −3 [95% CI, −10 to 4]; 10 weeks: 902 JAMA Psychiatry September 2020 Volume 77, Number 9 (Reprinted) jamapsychiatry.com Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Original Investigation Research Table 4. Questions and Categories from Self-reported Treatment Experiences No. (%) Questions/Categories Degarelix acetate (n = 26) Placebo (n = 26) “What positive/negative effects do you experience from the injection?” Positive effects of treatment Positive effects on sexuality 20 (77) 11 (42) Improved mental health 4 (15) 1 (4) Changed perspective 4 (15) 3 (12) Improved cognitive ability 1 (4) 0 Improved self-control 1 (4) 1 (4) Positive effects on relationship 1 (4) 0 Improved physical health 0 1 (4) Negative effects of treatment Negative effects on body 23 (89) 11 (42) Negative effects on sexuality 11 (42) 4 (15) Relationship problems 4 (15) 2 (8) Mental health issues 1 (4) 4 (15) Decreased cognitive ability 1 (4) 1 (4) Participants were interviewed with Negative emotions 1 (4) 0 open-ended questions. Themes and Negative effects on work 1 (4) 0 categories were abstracted from the “Would you like a repeated injection maintaining the effects for another 10 weeks? Please explain your answer.” answers through qualitative content analysis (eAppendix 2 in Reasons for continuing treatment Supplement 2). Overall positive effects 3 (12) 0 Categories are displayed in Positive emotions 2 (8) 0 descending order of frequency in Positive effects on sexuality 1 (4) 0 the degarelix group. Combining answers from 2 and It’s necessary 1 (4) 0 10 weeks. The question was asked Legal matter 0 1 (4) in 2 steps: (1) “What positive effects To achieve effect 0 1 (4) do you experience from the injection?” and then (2) “What Positive effect on relationship 0 1 (4) negative effects do you experience Reasons for discontinuing treatment from the injection?” Positive effects Negative effects on sexuality 2 (8) 0 refers to improvement in attitudes, behaviors, thinking, and Negative effects on body 2 (8) 1 (4) relationships. Negative effects Achieved effect 2 (8) 0 refers to adverse events. Cautiousness 1 (4) 0 Question asked only at 10 weeks. No effect 1(4) 9 (35) Three themes were abstracted. In the degarelix vs placebo groups, Attitudes about treatment 15 vs 9 responded “yes” and 9 vs Positive attitude 6 (23) 1 (4) 17 responded “no.” −4 [95% CI, −12 to 4]; P = .55) did not differ significantly be- adverse events among help-seeking men with pedophilic tween groups (eTable 5 in Supplement 2). disorder, both in the short (2-week) and in the medium The most commonly reported moderate adverse event was (10-week) terms. The drug was also effective among high- transient injection site reactions at 2 weeks (degarelix: 22 of risk participants (Table 3). The rapid onset of degarelix ap- 25 [88%]; placebo: 1 of 26 [4%]), and the most commonly re- pears to have a crucial advantage compared with earlier medi- ported minor adverse event was hepatobiliary enzyme level cations for paraphilic disorders, which hada1to3months’ elevations, the largest being 3.5 times the upper bound of the lag in exerting their effects on sexuality. normal range (degarelix: 11 of 25 [44%]; placebo: 2 of 26 [8%]). The self-reports provided an empirical basis for the A full description of adverse events and blood sample results patient side of shared decision-making and may facilitate pa- is found in eTable 2 and eTable 3 in Supplement 2. tient-centered care for pedophilic disorder. In weighing the benefits and harms of the drug, we found that the partici- pants self-reported a more positive than negative attitude to- ward treatment, specifically regarding the effects on sexual- Discussion ity. Thus, participants expressed relief of symptoms for which In this phase 2 randomized clinical trial, a single dose (240 mg) they sought help, in addition to experiencing the treatment aim of degarelix acetate statistically significantly reduced the of risk reduction. This finding may also be reflected in the re- dynamic risk factor scores for sexual offense with minimal duced risk scores for pedophilic disorder and sexual preoccu- jamapsychiatry.com (Reprinted) JAMA Psychiatry September 2020 Volume 77, Number 9 903 Research Original Investigation Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder pation (Table 3). Only 1 participant was lost to follow-up, and between groups indicated a decrease (10 weeks: −4 [95% CI, 58% of those randomized to receive degarelix wished to con- −12 to 4]) among participants who were randomized to re- tinue treatment (Table 4), which indicates to us a potential role ceive degarelix. When including the MADRS-S scores of all par- for long-term treatment along with psychosocial support in ticipants, we found that the difference at 10 weeks increased most participants. Ultimately, the treatment decision be- (data not shown). longs to the physician, who should take into consideration the In addition, future studies need to identify the risk fac- risk for abuse, patient preferences, and drug benefits and tors that benefit from other kinds of intervention, such as psy- harms. In view of the participant wishes and effects ex- chotherapy. The impaired self-regulation and low empathy pressed in the self-reports, we believe degarelix should be con- domains in the present trial highlighted the residual risk not sidered for help-seeking individuals with pedophilic disor- amenable to degarelix treatment. Abnormal results in these der. The maintained motivation among participants for such measures in a clinical setting are usually associated with potent therapy and the low EQ-VAS ratings reflect the sever- attention-deficit/hyperactivity disorder or autistic features. We think that if these conditions are adequately addressed, the ity and associated distress of the condition. However, a mod- est effect on quality of life associated with treating the core potential exists to both improve health and reduce the risk for symptoms of pedophilic disorder indicates the need to better committing child sexual offense. understand the reasons for the struggles experienced by those with this condition. Strengths and Limitations Future research needs to address the effects and predict- This trial has some strengths. The high inclusion rates, able long-term adverse effects of hormone deficiency as well diverse geographic origin of participants, and comorbidity as the sometimes excessive effects on sexuality, as reported suggest the generalizability of study results to other help- by the participants (Table 4; eFigures 3 and 4 in Supple- seeking populations. Although few participants reported prior ment 2). Incremental add-back therapy of hormones could be contact offenses, the results are in line with findings of pre- considered in this regard. Given that 2 participants re- vious unblinded cohorts of convicted offenders, which indi- ported severe adverse events of suicidal ideation, vigilance for cated efficacy of testosterone suppression in reducing sexual 12,14 the risk of exacerbating suicidality in predisposed individu- symptoms in this population. als is warranted. Although depressive symptoms sometimes However, this study also has some limitations. Its results are manifestations of hypogonadism, a 12-month open-label pertain to only men, the risk measure relies mainly on self- trial of degarelix in patients with prostate cancer (n = 409) re- reports, and the findings have not yet been validated against ported depression as an adverse event in only 1 individual. actual abuse rates. Therefore, the number of child sexual abuse The high baseline prevalence of depression in the present cases that the reduced risk factor scores have prevented re- trial (35%) may indicate an enriched sample of individuals sus- mains to be determined. Furthermore, the prespecified clini- ceptible to depressive deterioration from gonadotropin- cally significant reduction of 5 points was not reached, with releasing hormone antagonist treatment. However, most par- the exception of participants who were classified as high risk ticipants may not be susceptible. In post hoc comparisons of (Table 3). the treatment groups over the study period (eTable 5 in Supple- ment 2), no differences in change of Montgomery-Åsberg De- pression Rating Scale self-rating (MADRS-S) version scores were Conclusions observed in depression severity, incidence of dysthymia, de- pression, or suicide risk. Although uncertainty from insuffi- Treatment with degarelix appeared to decrease the scores of cient power and post hoc analysis remained, most estimates risk factors for child sexual abuse after 2 weeks of administra- within the CI of the difference in MADRS-S scores at 10 weeks tion for help-seeking men with pedophilic disorder. ARTICLE INFORMATION Statistical analysis: Bottai, Rahm. collection, management, analysis, and Obtained funding: Arver, Rahm. interpretation of the data; preparation, review, or Accepted for Publication: February 9, 2020. Administrative, technical, or material support: Arver, approval of the manuscript; and decision to submit Published Online: April 29, 2020. Rahm. the manuscript for publication. doi:10.1001/jamapsychiatry.2020.0440 Supervision: Arver, Rahm. Data Sharing Statement: See Supplement 3. Open Access: This is an open access article Conflict of Interest Disclosures: None reported. Additional Contributions: Pia Jaensen and distributed under the terms of the CC-BY License. Funding/Support: This study was funded by the Susanne Jarlvik Alm, both from Karolinska © 2020 Landgren V et al. JAMA Psychiatry. Gothenburg Society of Medicine; the Swedish University Hospital, assisted with trial organization Author Contributions: Dr Rahm was the principal Society of Medicine; the Söderström König and data collection; Anna Fenander Hedin, investigator. Dr Rahm had full access to all of the Foundation; the Fredrik and Ingrid Thuring Karolinska University Hospital, assisted with data data in the study and takes responsibility for the Foundation; the Center for Psychiatry Research, collection; and Niklas Juth, Karolinska Institutet, integrity of the data and the accuracy of the data Department of Clinical Neuroscience, Karolinska assisted with data analysis. These individuals analysis. Institutet; the Research Unit of Skaraborg Hospital; received no additional compensation, outside of Concept and design: Arver, Rahm. and nonpharmaceutical academic sources, as well their usual salary, for their contributions. Acquisition, analysis, or interpretation of data: as by grants under the ALF agreement between the All authors. Swedish government and the country councils. REFERENCES Drafting of the manuscript: Landgren, Malki, Rahm. Role of the Funder/Sponsor: The funders had no 1. Collin-Vézina D, Daigneault I, Hébert M. Lessons Critical revision of the manuscript for important role in the design and conduct of the study; learned from child sexual abuse research: intellectual content: Landgren, Bottai, Arver, Rahm. 904 JAMA Psychiatry September 2020 Volume 77, Number 9 (Reprinted) jamapsychiatry.com Effect of GnRH Antagonist vs Placebo on Committing Child Sexual Abuse in Men With Pedophilic Disorder Original Investigation Research prevalence, outcomes, and preventive strategies. blockers? Urology. 2008;71(6):1001-1006. doi:10. Swedish version. February 2009. Accessed Child Adolesc Psychiatry Ment Health. 2013;7(1):22. 1016/j.urology.2007.12.070 December 13, 2018. http://www.viss.nu/Global/ doi:10.1186/1753-2000-7-22 Blanketter/MINI_vers_6.pdf 14. Rösler A, Witztum E. Treatment of men with 2. Stoltenborgh M, van Ijzendoorn MH, Euser EM, paraphilia with a long-acting analogue of 27. EuroQol Group. EuroQol–a new facility for the Bakermans-Kranenburg MJ. A global perspective on gonadotropin-releasing hormone. N Engl J Med. measurement of health-related quality of life. child sexual abuse: meta-analysis of prevalence 1998;338(7):416-422. doi:10.1056/ Health Policy Amst Neth. 1990;16(3):199-208. around the world. Child Maltreat. 2011;16(2):79-101. NEJM199802123380702 doi:10.1016/0168-8510(90)90421-9 doi:10.1177/1077559511403920 15. Sorrentino R. Degarelix: an antagonist to 28. Burström K, Sun S, Gerdtham U-G, et al. 3. Walsh K, Zwi K, Woolfenden S, Shlonsky A. GnRH–theoretical and treatment considerations in Swedish experience-based value sets for EQ-5D School-based education programmes for the paraphilia. J Sex Med. 2012;9(1):327-329. doi:10.1111/ health states. Qual Life Res. 2014;23(2):431-442. prevention of child sexual abuse. Cochrane j.1743-6109.2011.02540.x doi:10.1007/s11136-013-0496-4 Database Syst Rev. 2015;(4):CD004380. doi:10. 16. Hanson RK, Morton-Bourgon KE. 29. Wood KL. The Medical Dictionary for Drug 1002/14651858.CD004380.pub3 The characteristics of persistent sexual offenders: Regulatory Affairs (MEDDRA) project. 4. Khan O, Ferriter M, Huband N, Powney MJ, a meta-analysis of recidivism studies. J Consult Clin Pharmacoepidemiol Drug Saf. 1994;3(1):7-13. doi:10. Dennis JA, Duggan C. Pharmacological Psychol. 2005;73(6):1154-1163. doi:10.1037/0022- 1002/pds.2630030105 interventions for those who have sexually offended 006X.73.6.1154 30. Sandelowski M. Whatever happened to or are at risk of offending. Cochrane Database of 17. Jordan K, Fromberger P, Stolpmann G, qualitative description? Res Nurs Health. 2000;23 Systematic Reviews. Intervention version published Müller JL. The role of testosterone in sexuality and (4):334-340. doi:10.1002/1098-240X(200008)23: February 18, 2015. Accessed September 9, 2019. paraphilia–a neurobiological approach. Part I: 4<334::AID-NUR9>3.0.CO;2-G 5. Långström N, Enebrink P, Laurén E-M, testosterone and sexuality. J Sex Med. 2011;8(11): 31. Pocock SJ. Group sequential methods in the Lindblom J, Werkö S, Hanson RK. Preventing sexual 2993-3007. doi:10.1111/j.1743-6109.2011.02394.x design and analysis of clinical trials. Biometrika. abusers of children from reoffending: systematic 18. van Honk J, Schutter DJ, Bos PA, Kruijt A-W, 1977;64(2):191-199. doi:10.1093/biomet/64.2.191 review of medical and psychological interventions. Lentjes EG, Baron-Cohen S. Testosterone 32. Svanborg P, Asberg M. A comparison between BMJ. 2013;347:f4630. doi:10.1136/bmj.f4630 administration impairs cognitive empathy in the Beck Depression Inventory (BDI) and the 6. Knack N, Winder B, Murphy L, Fedoroff JP. women depending on second-to-fourth digit ratio. self-rating version of the Montgomery Asberg Primary and secondary prevention of child sexual Proc Natl Acad SciUSA. 2011;108(8):3448-3452. Depression Rating Scale (MADRS). J Affect Disord. abuse. Int Rev Psychiatry. 2019;31(2):181-194. doi:10.1073/pnas.1011891108 2001;64(2-3):203-216. doi:10.1016/S0165-0327(00) doi:10.1080/09540261.2018.1541872 19. Aluja A, García LF, Martí-Guiu M, et al. 00242-1 7. Seto MC. Pedophilia and Sexual Offending Interactions among impulsiveness, testosterone, 33. Bergman H, Källmén H. Alcohol use among Against Children: Theory, Assessment, and sex hormone binding globulin and androgen Swedes and a psychometric evaluation of the Intervention. 2nd ed. American Psychological receptor gene CAG repeat length. Physiol Behav. alcohol use disorders identification test. Alcohol. Association; 2018. 2015;147:91-96. doi:10.1016/j.physbeh.2015.04.022 2002;37(3):245-251. doi:10.1093/alcalc/37.3.245 8. Beier KM, Ahlers CJ, Goecker D, et al. Can 20. Jordan K, Fromberger P, Stolpmann G, 34. Nesvåg R, Lange EH, Faerden A, et al. The use pedophiles be reached for primary prevention of Müller JL. The role of testosterone in sexuality and of screening instruments for detecting alcohol and child sexual abuse? First results of the Berlin paraphilia–a neurobiological approach. Part II: other drug use disorders in first-episode psychosis. Prevention Project Dunkelfeld (PPD). J Forensic testosterone and paraphilia. J Sex Med. 2011;8(11): Psychiatry Res. 2010;177(1-2):228-234. doi:10.1016/ Psychiatry Psychol. 2009;20(6):851-867. doi:10. 3008-3029. doi:10.1111/j.1743-6109.2011.02393.x j.psychres.2010.01.007 1080/14789940903174188 21. Spector IP, Carey MP, Steinberg L. The sexual 35. Kessler RC, Adler L, Ames M, et al. The World 9. Thibaut F, De La Barra F, Gordon H, Cosyns P, desire inventory: development, factor structure, Health Organization Adult ADHD Self-Report Scale Bradford JM; WFSBP Task Force on Sexual and evidence of reliability. J Sex Marital Ther. 1996; (ASRS): a short screening scale for use in the Disorders. The World Federation of Societies of 22(3):175-190. doi:10.1080/00926239608414655 general population. Psychol Med. 2005;35(2):245- Biological Psychiatry (WFSBP) guidelines for the 22. Reid RC, Garos S, Carpenter BN. Reliability, 256. doi:10.1017/S0033291704002892 biological treatment of paraphilias. World J Biol validity, and psychometric development of the 36. Barry MJ, Edgman-Levitan S. Shared decision Psychiatry. 2010;11(4):604-655. doi:10.3109/ Hypersexual Behavior Inventory in an outpatient making–pinnacle of patient-centered care. N Engl J sample of men. Sex Addict Compulsivity. 2011;18(1): Med. 2012;366(9):780-781. doi:10.1056/ 10. Lee JY, Cho KS. Chemical castration for sexual 30-51. doi:10.1080/10720162.2011.555709 NEJMp1109283 offenders: physicians’ views. J Korean Med Sci. 23. Conners CK. Conners’ Continuous Performance 37. Carr BR, Stewart EA, Archer DF, et al. Elagolix 2013;28(2):171-172. doi:10.3346/jkms.2013.28.2.171 Test II (CPT II V.5). Multi-Health Systems, Inc; 2000. alone or with add-back therapy in women with 11. Turner D, Petermann J, Harrison K, Krueger R, 24. Eriksson JM, Andersen LM, Bejerot S. heavy menstrual bleeding and uterine leiomyomas: Briken P. Pharmacological treatment of patients RAADS-14 Screen: validity of a screening tool for a randomized controlled Trial. Obstet Gynecol. with paraphilic disorders and risk of sexual autism spectrum disorder in an adult psychiatric 2018;132(5):1252-1264. doi:10.1097/AOG. offending: an international perspective. World J Biol population. Mol Autism. 2013;4(1):49. doi:10.1186/ 0000000000002933 Psychiatry. 2019;20(8):616-625. doi:10.1080/ 2040-2392-4-49 38. Klotz L, Boccon-Gibod L, Shore ND, et al. 15622975.2017.1395069 25. Baron-Cohen S, Wheelwright S, Hill J, Raste Y, The efficacy and safety of degarelix: a 12-month, 12. Turner D, Briken P. Treatment of paraphilic Plumb I. The “Reading the Mind in the Eyes” Test comparative, randomized, open-label, disorders in sexual offenders or men with a risk revised version: a study with normal adults, and parallel-group phase III study in patients with of sexual offending with luteinizing adults with Asperger syndrome or high-functioning prostate cancer. BJU Int. 2008;102(11):1531-1538. hormone-releasing hormone agonists: an updated autism. J Child Psychol Psychiatry. 2001;42(2):241- doi:10.1111/j.1464-410X.2008.08183.x systematic review. J Sex Med. 2018;15(1):77-93. 251. doi:10.1111/1469-7610.00715 doi:10.1016/j.jsxm.2017.11.013 26. Sheehan D, Lecrubier YMINI. Mini 13. van Poppel H, Nilsson S. Testosterone surge: Internationell Neuropsykiatrisk Intervju 6.0.0. rationale for gonadotropin-releasing hormone jamapsychiatry.com (Reprinted) JAMA Psychiatry September 2020 Volume 77, Number 9 905

Journal

JAMA PsychiatryPubmed Central

Published: Apr 29, 2020

There are no references for this article.