Older Individuals Convicted of Sexual Offenses: A Literature Review

The population of older individuals convicted of sexual offenses (OSOs) is rapidly increasing. However, we have little understanding of their characteristics (e.g., demographic, psychological, individual, offense, and risk) and needs. To identify any similarities or differences that are unique to older individuals convicted of sexual offending, it is important to compare such characteristics across the adult lifespan. Therefore, the aim of this systematic review was to specify and synthesize the current knowledge of characteristics across the adult lifespan of the population of individuals convicted of sexual offenses. Five databases were searched and 10,680 results were screened, resulting in 100 studies included in the final review. The findings were grouped into four emergent themes: age of onset and prevalence; offender and offense characteristics; age and the risk of reoffending; and treatment. Implications of the findings from this review are discussed in relation to future research and clinical practice.

445,127 registered sex offenders Sourced from registries from 49 States Mean age, accounting for invalid and unknown cases, was 44.8 (SD = 13.32) with a range of 12-99 (n = 449,534) This is interesting, as more people are places on registries for long durations (or life) with little attrition, the mean age will continue to grow older. This contradicts research indicating decline in risk with aging; sex offence recidivism is rare with advanced age. The sexual aggressors against women as a group were significantly younger than the offenders against girls and the offenders against pubertal females. There were no significant differences in terms of age between the aggressors against girls and the aggressors against pubertal females

Canada
The Sexual Violence  was scored for each offender and the relationship between age-atrelease and SVR-20 item and total scores was examined.
Quantitative Analysis of case file data 468 sex offenders 175 rapists; 155 child molesters, 93 familial offenders; 45 mixed offenders; 5 sexual offenders with adult male victims, and 3 offenders with noncontact sexual offenses Significant difference among the mean ages-at release, released on average in the following rank order from youngest to oldest: rapists, mixed offenders, nonfamilial child molesters, and incest offenders. Rapists were released at a significantly younger age than the other groups, and mixed offenders were released at a significantly younger age than the incest offenders. Risk factors prevalent in rapists and mixed offenders (e.g., antisocial behavior) more prevalent in younger offenders, whereas risk factors more prevalent in child molesters (e.g., paraphilia) more prevalent in older offenders.
Aging and risk of offending Most actuarial items in sex offender assessment instruments were correlated with the age at release from custody. Items reflecting aspects of antisocial behavior were negatively correlated with age at release, whereas items reflecting sexual deviance were positively correlated. Antisocial traits more prevalent among rapists who were released from custody at a younger age, whereas aspects of sexual deviance were more prevalent in child molesters who were released from custody at an older age.
Aging and risk of offending Specialist Sex Offenders (SOs) younger than 40.5 years with an introverted personality most likely to confess; Specialist SOs older than 40.5 years with an extroverted personality less likely to confess More mature SOs seem to cope better with the unfamiliarity and demands of police interrogation As offenders grow older, they become increasingly resistant to interrogation. Age has a paradoxical effect: the older the individuals, the more they see the causes of their behaviour as controllable, whilst considering that the causes cannot be changed. The more age increased, the more the scores on the Self-Esteem scale increased. The higher the age at first incarceration, the higher the respondents' self-esteem score. Individuals considered experts in the sample started their deviant sexual activities at an early age. As they aged, their sexual preferences become fixed into a specific age range; their sexual preferences did not develop age appropriately as they matured. Numerous salient behavioral traits were acquired by expert sex offenders in their early life and were utilized in the commission of their sexual offenses. Thus, in later life they were able to manipulate others and viewed them as objects to be exploited to meet their own needs. Relative family and social stability in early life and adulthood. 3 participants disrupted upbringing by loss events or multiple care-takers before age 11. 10 were married at the time of offences; all subjects had been married at some point. 12 gave a history of steady and continuous employment. 10 were retired at time of current offence and in 2 cases an onset of SO was closely related to retirement from full time work. 2 of the subjects had a history of excessive alcohol use, none admitted to illegal drug use. 8 had medical problems. 5 faced charges for SO of family members, 8 offended against kids outside of family but were acquainted with (i.e. neighbours). 9 offended against females, 2 against both sexes, and 2 against boys only. Victim age range 3-11. Penetrative abuse occurred in eight cases; 5 involved fellatio, and one of the victims subjected to buggery. 10 of the subjects had been offending against the current victim for more than year, and for more than 10 years in four of these cases. Sentences ranged from 2-10 years.

Characteristics
Average age of the low-risk group was found to be significantly lower than the high-risk group.
Aging and risk of offending Sex offenders more likely to be white, unemployed at the time of their offence, most recently worked as a car or truck driver, and have a past psychiatric history recorded in their medical notes; Age of conviction for the index offence of the sex offenders was significantly older' No significant differences were found between the sex offenders and other offenders for diagnoses of psychotic illness, major depressive episode, organic disorder (dementia) and personality disorder. SOs less likely to be substance abusers; Most of the sex offenders were the father, stepfather, grandfather, or an acquaintance of their victim; and had committed crimes on several girls, and usually alone in the offender's home. There were no differences in the rates of psychiatric illness between these two groups apart from being assigned a diagnosis of personality disorder. Incest offenders were found to be significantly older than the rapists and child molesters, and child molesters were significantly older than rapists Incest perpetrators tend to be older as their sexual offenses are more opportunistic in nature. They do not offend until after they are married and have children.

Characteristics
Child molesters tend to be more sexually aroused to children than incest offenders, are more predatory in their offense patterns and often are younger when they start their sexual offenses. Offenders able to avoid a conviction, on average, for 7.5 years (SD = 9.3) after the start of their sex offending. Sex offenders in this study start offending in their early thirties. Official data reveals that their first sex crime conviction occurs in late thirties; Typical adult first-time convicted sex offender is in his late thirties at the start of his sentence, has limited education, is married or in a common-law relationship, and may or may not have a criminal record for a non-sex offence. often unemployed at the time of their sex offence, and have an average of two victims of who they offended against approximately ten times.

Canada
To examine the life of an especially persistent and exceptionally older offender to determine the extent to which these characteristics apply.

Qualitative Case study
A case study of Atypical Offender (AO); man committed as a sexually violent person in his mid-90s Typical protective factors were absent in this case; AO was in good health, both physically and mentally; denial of offending and non-compliance with treatment.
Concluded that it is "not unreasonable to view older individuals who are exceptionally healthy as also having a higher risk of recidivism than sameaged but less healthy peers"

Canada
To provide data with respect to the outcomes and characteristics of a cohort of offenders who were over 50 at age of release and examined the probability and severity of violent and sexual recidivism in this group. Age at release demonstrated inverse relationship to the probability of sexual and/or violent recidivism. Rates of sexual and violent recidivism declined in older age, with very low rates of recividim in older offender groups.
Overall, age at release was negatively correlated with BARS score. Older offenders appeared to be a lower risk group in general, with higher risk older offenders being rare. Age tends to mitigate overall risk, but not all older offenders are low risk.
Aging and risk of reoffending;

Nilsson et al. (2014) ***** [74]
Sweden The aims of this study are: (i) to establish the rate of recidivism among child sexual offenders, (ii) to compare intra-and extrafamilial offenders and (iii) to test the predictive ability of common demographic, criminological and clinical characteristics in a prospective, long-term clinical follow-up study of convicted perpetrators of child sexual abuse sentenced to correctional treatment or forensic psychiatric care. Age at first conviction was tested with Receiver Operating Characteristic (ROC) analyses for its ability to predict sexual, non-sexual violent and any criminal recidivism in both main groups. This variable showed a good predictive ability in both groups, especially for violent reconvictions, indicating that the lower the offender's age at first conviction, the higher the risk for recidivism Aging and risk of reoffending

Olver, Nicholaichuk, Gu, & Wong (2012) *** [75]
Canada An examination of sex offender treatment outcome in a large national cohort of Canadian Federally incarcerated sex offenders followed up an average of 11.7 years post-release.

Quantitative
Analysis of case file data and actuarial risk assessment 732 sex offenders 625 were in the treated group; 107 were in the comparison group Younger offenders had significantly higher sexual recidivism failure rates; No differences between older treated and untreated offenders in sexual recidivism failure rate; Younger offenders reoffended at higher rates irrespective of treatment condition; Significant differences in outcome observed only among younger offenders and treatment completion uniquely predicted reductions in subsequent violence; Differences in outcome and the lack of an interaction explained by the fact that the older offenders were already a lower risk group, both in terms of their actuarial risk as well as older age. Rapists: proportion of sexual reoffence was stable at 28% through to late 30s, reducing to 22% in the 40s, declining in 50s, and reaching 0 (no detected re-offence) in the oldest group aged 60+ Child molesters: Pattern quadratic not linear. Youngest group (18-29) had lower sexual re-offence rate, increasing for those released in their 30s, declining slightly for those released in their 40s, and dropping for those released in their 50s and 60s. Decline does not appear until age of 50.
Aging and risk of reoffending

Proulx et al. (1997) *** [78]
Canada To verify whether the recidivism rate in a large sample of sexual aggressors is a function of static or dynamic predictors of the following categories: criminal history, demographic characteristics, and psychometric and phallometric data Quantitative 382 men with an official record of at least one sexual offence For violent offences, those who reoffended had a significantly lower mean age than those who did not reoffend violently. Those who reoffended had a significantly lower mean age than those who did not reoffend Reconvicted rapists were younger and had more previous convictions than those who were not reconvicted.
Aging and risk of reoffending Older men more likely to be recommended for an extended supervision order due to having a history of repeated sexual deviance; The older the offender, the more time he has had to incur further sexual and general offences, which can elevate his score on dynamic risk measures; Older men with higher ASRS scores tend to demonstrate a greater number of dynamic risk factors than their younger counterparts. Younger age groups re-offending at a higher rate in comparison to offenders over age 50. Older age-at-release groups began sexually offending at a progressively older age. Except the youngest and oldest age groups, those who sexually re-offend had a significantly younger age at first sexual offence compared to those who did not sexually re-offended While sexual re-offences became less frequent with age in general, those with the highest static risk accounted for a disproportionate number of these reoffences until after age 60.
Aging and risk of reoffending

Smallbone & Wortley (2004) *** [91]
Australia To examine onset and persistence of sexual and nonsexual offending among those offenders in the original sample who (a) admitted to at least one sexual offense against a child, and (b) provided all required self-report data (n = 207). Moderate to strong positive correlations were obtained between the three onset variables (self-reported age at first sexual contact with a child, age at first conviction for a sexual offences, and age at first conviction for any offence) Small negative correlations between age at first sexual contact with a child and number of sexual offence victims, number of sexual offence convictions, and number of nonsexual offence convictions. A moderate negative correlation was found between age at first conviction for any offence and the number of nonsexual convictions, and a small negative correlation was found between age at first sexual offence conviction and the number of nonsexual offence convictions.
The mean age at first sexual contact with a child was 32.3 years (median = 31). Onset ages were normally distributed across a wide range (10-63). Extrafamilial offenders reported an earlier onset than intrafamilial offenders. Intrafamilial were significantly older at when first convicted of a CSO than both extrafamilial and mixed-type offenders Offenders were significantly older at the time of their first sexual contact with a child than they were at the time of their first conviction of any offence, and significantly older when first convicted of a SO than time at first conviction for any offence. Age positively correlated with number of prior sexual sentencing occasions, and negatively with a history of nonsexual offending; those released at a younger age tended to be general offenders, whereas those released at a later age were more likely to be sexual specialists.
The odds of sexual reconviction declined by 0.02 with each year of increasing age. Confirms inverse relationship between age of release and sexual reconviction. For those with one prior sexual sentencing, the effect of age was characterized as gradual linear decline in the odds of sexual recidivism. For those with no prior sexual sentencing occasions, age on release and sexual recidivism were unrelated. For those with two or more prior sex offences, there was a significant cubic trend; the graph changed slope at two points. High sexual recidivism rate between 18 and 25, a reduction (from 80% to 50%) for offenders released after that age, with no decline in sexual recidivism until the ago of 60 when the rate falls by approx. 40% UK To identify the best predictors of sexual and violent recidivism in a group of treated sexual offenders.

Quantitative
Risk assessment data (RM 2000) 3,773 sexual offenders Age at release, and younger age, was a predictor of recidivism.
Aging and risk of reoffending

Willis & Ward (2011) **** [98]
New Zealand To explore the practical utility of the GLM with a sample of released child molesters, and investigate the relationship between primary goods attainment and overall re-entry conditions (in terms of accommodation, social support and employment) Quantitative Semi-structured interviews 16 child molesters interviewed at three different timepoints after release (1,3 and 6 months) Significant positive correlations between offender age and good lives ratings at one month and three months post-release; older participants tended to have higher good lives ratings. Significant negative correlation between offender age and Stable-2007 scores, meaning that older participants tended to have lower assessed levels of dynamic risk factors

Canada
To answer a number of important questions pertaining to high-risk sexual offenders. Specifically, we sought to answer the following: (1)What is the prevalence of deviant sexual fantasy among high-risk offenders?; (2) Is there a relationship between offenders' fantasy themes and the offences they perpetrate?; (3) Is there a relationship between level of psychopathy and the sexual fantasies that offenders report?; (4) What is the prevalence of sexual paraphilia among high-risk offenders?; (5)  Australia To apply a criminal careers model to child sexual abusers Quantitative Review of official data and self-report data on personal and offending histories 362 convicted offenders, 213 provided self-report data Limited/specialized offenders: relatively older at the time of first sexual contact with child; Waited until early 30s before first offense; abuse of power/trust; already have access to victim. Persistent/versatile offenders: Involvement in criminal justice system from an early age and earlier age for first sexual contact with a child. Persistent/specialist: earlier sexual contact with child; involvement in criminal justice system begins at later age (earlier than limited specialists).
The number of * after the Author denotes the 'Mixed Methods Appraisal Tool' (MMAT;Hong et al., 2018) rating that was given to each study;1 and 2 stars are referred to as low quality studies, and 4 and 5 stars are higher quality studies.