Hostname: page-component-78c5997874-xbtfd Total loading time: 0 Render date: 2024-11-11T01:08:21.019Z Has data issue: false hasContentIssue false

A meta-analysis of the prevalence of attention deficit hyperactivity disorder in incarcerated populations

Published online by Cambridge University Press:  07 April 2014

S. Young*
Affiliation:
Centre for Mental Health, Division of Brain Sciences, Department of Medicine, Imperial College London, UK Broadmoor Hospital, West London Mental Health Trust, London, UK
D. Moss
Affiliation:
Caudex Medical, Oxford, UK
O. Sedgwick
Affiliation:
Department of Psychology, Institute of Psychiatry, King's College London, UK
M. Fridman
Affiliation:
AMF Consulting, Inc., Los Angeles, CA, USA
P. Hodgkins
Affiliation:
Global HEOR, Vertex Pharmaceuticals, Boston, MA, USA
*
*Address for correspondence: Dr S. Young, Centre for Mental Health, Division of Brain Sciences, Department of Medicine, Imperial College London, 37 Claybrook Road, London W6 8LN, UK. (Email: susan.young1@imperial.ac.uk)
Rights & Permissions [Opens in a new window]

Abstract

Background

Studies report the variable prevalence of attention deficit hyperactivity disorder (ADHD) in incarcerated populations. The aim of this meta-analysis was to determine the prevalence of ADHD in these populations.

Method

Primary research studies reporting the prevalence (lifetime/current) of ADHD in incarcerated populations were identified. The meta-analysis used a mixed log-binomial model, including fixed effects for each covariate and a random study effect, to estimate the significance of various risk factors.

Results

Forty-two studies were included in the analysis. ADHD prevalence was higher with screening diagnoses versus diagnostic interview (and with retrospective youth diagnoses versus current diagnoses). Using diagnostic interview data, the estimated prevalence was 25.5% and there were no significant differences for gender and age. Significant country differences were noted.

Conclusions

Compared with published general population prevalence, there is a fivefold increase in prevalence of ADHD in youth prison populations (30.1%) and a 10-fold increase in adult prison populations (26.2%).

Type
Review Articles
Creative Commons
Creative Common License - CCCreative Common License - BY
The online version of this article is published within an Open Access environment subject to the conditions of the Creative Commons Attribution licence .
Copyright
Copyright © Cambridge University Press 2014

Introduction

Attention deficit hyperactivity disorder (ADHD) is defined in DSM-IV (APA, 1994) and ICD-10 (WHO, 1992) as high levels of hyperactive, impulsive and/or inattentive behaviours beginning in early childhood. A level of impairment in at least two areas of life needs to be evident for at least 6 months' duration for a diagnosis of ADHD to be made. ADHD is a common disorder that often persists into adulthood (Faraone et al. Reference Faraone, Biederman and Mick2006), affecting 3–7% of youths (Asherson, Reference Asherson2004; Asherson et al. Reference Asherson, Kuntsi and Taylor2005; Polanczyk et al. Reference Polanczyk, de Lima, Horta, Biederman and Rohde2007; Willcutt, Reference Willcutt2012) and 1–5% of adults (Asherson, Reference Asherson2004; Asherson et al. Reference Asherson, Kuntsi and Taylor2005, Reference Asherson, Akehurst, Kooij, Huss, Beusterien, Sasane, Gholizadeh and Hodgkins2012; Kessler et al. Reference Kessler, Adler, Barkley, Biederman, Conners, Demler, Faraone, Greenhill, Howes, Secnik, Spencer, Ustun, Walters and Zaslavsky2006; Simon et al. Reference Simon, Czobor, Balint, Meszaros and Bitter2009; Willcutt, Reference Willcutt2012).

There is no generally accepted estimate of the prevalence of ADHD in incarcerated populations; research studies have consistently reported a disproportionately higher rate of individuals with ADHD in the criminal justice system compared with general population prevalence. These individuals, who are often untreated, seem to have greater judicial contact and pose higher risk within the criminal justice system. Compared with other offenders, they present with younger age at first contact, greater recidivism and institutional behaviour disturbance (Young et al. Reference Young, Gudjonsson, Ball and Lam2003, Reference Young, Gudjonsson, Wells, Asherson, Theobald, Oliver, Scott and Mooney2009b , 2011). A Scottish prison study found that inmates with ADHD symptoms were involved in up to eight times more institutional aggression; this strong association was maintained even when controlling for antisocial personality characteristics, with aggressive incidents being six times greater compared with non-ADHD peers (Young et al. Reference Young, Gudjonsson, Wells, Asherson, Theobald, Oliver, Scott and Mooney2009b ). These problems are likely to be associated with underlying deficits in executive dysfunction (Young et al. Reference Young, Morris, Toone and Tyson2007; Bramham et al. Reference Bramham, Ambery, Young, Morris, Russell, Xenitidis, Asherson and Murphy2009; Rose et al. Reference Rose, Bramham, Young, Paliokostas and Xenitidis2009); in particular, behavioural disinhibition and emotional dysregulation (Gudjonsson et al. Reference Gudjonsson, Sigurdsson, Young, Newton and Peersen2009, Reference Gudjonsson, Sigurdsson, Adalsteinsson and Young2012).

However, prevalence varies widely among studies and methodologies differ. Hence, this study reviewed the data obtained from a systematic literature search performed to ascertain the lifetime/current prevalence of ADHD in prison populations in both youths and adults. The study aimed to specify (1) the prevalence of ADHD in incarcerated youth and adult offenders, including differences by gender; (2) prevalence from studies applying screening tools and diagnostic clinical interviews; (3) differences in prevalence obtained from retrospective diagnoses from adults and youth prevalence obtained using a current diagnoses methodology; and (4) geographical differences.

Method

Eligibility criteria

The systematic review was performed in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA; Liberati et al. Reference Liberati, Altman, Tetzlaff, Mulrow, Gotzsche, Ioannidis, Clarke, Devereaux, Kleijnen and Moher2009) guidelines. Initial searches were carried out in a variety of databases and websites to gain an understanding of the amount of information available. Reports published since 1980 and in English were included.

Data sources

Searches were conducted in OvidSP Medline (1948 to present) and EMBASE (1988 to present segment), Datastar PsycINFO (unrestricted) and Social SciSearch (1972 to date; limited to English and added since 1 January 1980), including the literature published between 1 January 1980 and 3 May 2011; a citations update was performed on 21 June 2011 and extended to 31 August 2012 in October 2012.

Search

Search terms were developed, refined and tested for relevance by cross-checking results against a list of known relevant articles provided by the lead author. The following descriptors were used in OvidSP EMBASE (1988 to present segment): ADHD; attention deficit disorder; [EMTREE] crime; criminals; criminology; criminal behaviour; criminal justice; criminal law; court; criminal psychology; delinquency; juvenile delinquency; gang; legal evidence; legal procedure; police; legal liability; mandatory programs; violence; prisons; prisoner; probation; law enforcement; recidivism; jurisprudence; punishment; offender; drug abuse; drug misuse.

Study selection

Articles obtained from the final searches were first de-duplicated, then an inclusion/exclusion process was undertaken based on the following exclusion criteria: non-English language articles, articles published before 1980, animal studies, articles that were not peer reviewed (e.g. theses), and articles that obviously did not hold relevance (e.g. they did not focus on ADHD or crime, or they focused on substance misuse but not from the criminology perspective). Review articles pre-2006 were excluded, and post-2006 reviews were kept with the sole purpose to examine bibliographies to check for any other articles not identified in the search; these review articles were not included in the final prevalence calculation; only primary research articles were included. Articles with no abstract (including initial PsycINFO and Social SciSearch search outputs) were also excluded unless the title or other information (e.g. key terms) suggested they may hold relevance. The inclusion/exclusion review was first completed based on title/abstract/key words by four researchers, and if the relevance of an article was unclear, the full text was retrieved before a final decision was made. Once the initial bulk of the inclusion/exclusion process was completed, abstracts were reviewed independently by the lead author before full-text articles were retrieved. Any uncertainties over including/excluding articles were discussed among the researchers and a final decision was made once the full-text articles were reviewed.

The full texts of included articles were retrieved for detailed evaluation against eligibility criteria. Only the publications that focused on incarcerated/prison populations and those that reported ADHD prevalence were selected. Studies that reported results on mixed gender populations were excluded as we aimed to separate the gender effect in modelling.

Data collection process

A data extraction sheet was developed in Microsoft Excel and pilot tested on 13 randomly selected studies and refined accordingly. The publications were divided among four researchers (including two authors: D.M. and O.S.) who performed the data extraction independently. Data were reviewed for consistency and any queries were resolved by discussion among the researchers and the lead author. The lead author also made the final decision whether to include/exclude data by reviewing the identified publications.

The authors of 11 studies were contacted directly in personal communications to ascertain missing information relating to age of the study population (Black et al. Reference Black, Arndt, Hale and Rogerson2004; Sanz-Garcia et al. Reference Sanz-Garcia, Dueñas, Burges, Muro and Perez2010a , Reference Sanz-Garcia, Dueñas, Domenech, Ramón, Muro and Perez b ), gender (Vitelli, Reference Vitelli1996), method of assessment of ADHD (Gordon & Moore, Reference Gordon and Moore2005), study population (Cahill et al. Reference Cahill, Coolidge, Segal, Klebe, Marle and Overmann2012; Colins et al. Reference Colins, Vermeiren, Vahl, Markus, Broekaert and Doreleijers2012) and verification of the incarcerated status of the sample (Langevin, Reference Langevin2003; Rosler et al. Reference Rosler, Retz, Yaqoobi, Burg and Retz-Junginger2009; Retz & Rosler, Reference Retz and Rosler2010; Lindsay et al. Reference Lindsay, Carson, Holland, Michie, Taylor, Bambrick, O'Brien, Wheeler and Steptoe2012). All responded with additional information that had not been included in the original publication and, if meeting inclusion criteria, was included in this meta-analysis (indicated in Supplementary Tables S1 and S2).

Some cohorts of prison populations were published more than once. To avoid double counting data, multiple reports of the same cohort were pieced together by juxtaposing author names, treatment comparisons, sample sizes and outcomes. An exception to this rule occurred in relation to the Young et al. (Reference Young, Gudjonsson, Misch, Collins, Carter, Redfern and Goodwin2010a , Reference Young, Misch, Collins and Gudjonsson2011) reports; whereas Young et al. (Reference Young, Gudjonsson, Misch, Collins, Carter, Redfern and Goodwin2010a ) used diagnostic interviews, Young et al. (Reference Young, Misch, Collins and Gudjonsson2011) used screening data to provide ADHD prevalence. In the interest of obtaining meaningful data on diagnostic differences in ADHD, these reports were considered as separate studies. In addition, following author communication, more recent data from Cahill et al. (Reference Cahill, Coolidge, Segal, Klebe, Marle and Overmann2012) were included rather than data from an earlier publication of the same population (Coolidge et al. Reference Coolidge, Segal, Klebe, Cahill and Whitcomb2009).

Data items

For each included study, data were extracted into a data set containing: (1) study location (country of origin), (2) study sample size, (3) study population (including age and gender), (4) diagnostic criteria (screening or diagnostic interview), (5) study design (retrospective, that is adult studies reporting a prevalence of childhood ADHD, or current diagnoses), (6) ADHD prevalence and (7) treatment (including prison management). We have used the term ‘prison’ but included all individuals who were incarcerated, jailed or imprisoned, or similar descriptors. For the analysis, studies were grouped by age for youth and adult offenders (see Supplementary Tables S1 and S2). It was noted that there was no clear definition across publications on the age of an adult versus a youth. For the purposes of this meta-analysis, we designated 18 years to be the cut-off point for a youth becoming an adult (i.e. youths were ⩽18 years and adults were >18 years); papers reporting youth data often cited 18 years as the upper limit of the age range of participants. In those papers where an age range was given that spanned this cut-off point (e.g. 15–28 years), the mean (or median if mean was not provided) was used to define whether the study population should be listed as ‘youth’ or ‘adult’. For studies reporting on (1) both genders separately, (2) both diagnostic criteria or (3) both study designs, information was recorded as two separate observations (‘study strata’) linked by the study number. None of the studies reported separate results on more than one of these three characteristics and none reported on both age groups.

For a diagnosis of ADHD to be given in adulthood, clinically significant symptoms must have been present in childhood. Therefore, when assessing adults for ADHD, it is necessary to establish whether they meet criteria for childhood ADHD, requiring them to retrospectively report on their symptoms (if a diagnosis did not already exist). Some adult studies (n = 13) included in this meta-analysis also reported a prevalence of childhood ADHD for the cohort (referred to as a retrospective diagnosis in this study).

Quality control

To ascertain the validity of eligible publications, each researcher checked and independently reviewed a random sample of each others' papers for data extraction and interpretation consistency. Disagreements were resolved by reviewing the data source and by discussion between the three reviewers (one of whom is an author).

Statistical methods

Summary prevalence estimates were calculated for the meta-analysis. Observed prevalence per study stratum was reported by age and gender groups. Model-predicted prevalence was reported by the various model covariates together with their corresponding 95% confidence intervals (CIs).

A mixed log-binomial model was fitted for the observed prevalence in the study strata to estimate the significance of the various risk factors (model 1). Covariates collected for predicting ADHD prevalence included gender, age group, diagnostic method, study design and country. The model included fixed effects for each of the covariates and a random study effect, and can be defined as follows:

$$\eqalign{& y_{ij} \sim {\rm binomial}(n_{ij,} p_{ij} ) \cr &\hskip-2 \log (\,p_{ij} ) = {\rm study}_i + {\rm gender}_{ij} + {\rm age}_{ij} + {\rm diagnostic}\;{\rm method}_{ij} \cr & \quad \quad \quad \quad\hskip1 + {\rm study}\;{\rm design}_{ij} + {\rm country}_i ,} $$

where y ij is the number of patients with the response, p ij is the ADHD prevalence and n ij is the total number of inmates included in study i and on stratum j. As implemented here, the model made the following assumptions:

  1. (1) study i represented a random study effect, assumed to follow a normal (0, σ 2) distribution. The study effect was considered to be the within-study logarithmic baseline prevalence, thus eliminating the study-specific effect from the other comparison estimates.

  2. (2) gender ij represented the adjusted gender effect common to all patients in study i and stratum j on the logarithm of prevalence. Similarly for age ij , diagnostic method ij , study design ij and country i . Countries were also grouped into regions in an attempt to obtain more stable estimates (North America, Europe and Other).

Given the observed level of interaction between gender, age group and diagnostic method, these three covariates were fully interacted in the model and were represented by eight indicator variables (one omitted as a reference group). Other interactions were also tested for significance. Marginal estimates of prevalence (least-square means) for each of the covariates (e.g. males, adults, etc.) and for their combinations were calculated by applying the inverse logarithmic transformation to model predictors. To estimate the sensitivity of results to the diagnostic method, the model was refitted to a subset of data using diagnostic interview strata only (model 2). For the sensitivity analysis model (model 2), the aforementioned eight indicator variables for the interaction of gender, age group and diagnostic method were collapsed to four to capture the interaction of age-by-gender group.

The models were fitted adopting a frequentist estimation approach using PROC GLIMMIX in SAS version 9.2 (SAS Institute, USA).

Results

In total, 8520 publications were identified following the OvidSP Medline, EMBASE, Datastar, PsycINFO and Social SciSearch database search. Publications not specifically relating to ADHD and criminality and duplicates between databases were excluded electronically, leaving 325. Once further duplicates had been identified manually, 311 publications remained. A further 218 publications were excluded after further review as they did not report on samples composed entirely of incarcerated/prison populations. Of the 93 articles remaining, a further 40 articles were excluded as they did not provide prevalence data on ADHD; 53 publications included ADHD prevalence data. Three studies (Young et al. Reference Young, Chesney, Sperlinger, Misch and Collins2009a ; Ginsberg & Lindefors, Reference Ginsberg and Lindefors2012; Tidefors & Strand, Reference Tidefors and Strand2012) were excluded as they used samples composed solely of participants who had ADHD, hence giving a prevalence rate of 100%. A further eight studies were excluded as they only provided mixed-gender prevalence (Milin et al. Reference Milin, Halikas, Meller and Morse1991; Fulwiler et al. Reference Fulwiler, Forbes, Santangelo and Folstein1997; Rasmussen et al. Reference Rasmussen, Almvik and Levander2000, Reference Rasmussen, Almvik and Levander2001; Black et al. Reference Black, Arndt, Hale and Rogerson2004, Reference Black, Gunter, Loveless, Allen and Sieleni2010; Anckarsater et al. Reference Anckarsater, Nilsson, Stahlberg, Gustafson, Saury, Rastam and Gillberg2007; Einat & Einat, Reference Einat and Einat2008; Gudjonsson et al. Reference Gudjonsson, Sigurdsson, Young, Newton and Peersen2009), leaving 42 studies included in the meta-analysis, reporting prevalence by gender (Fig. 1).

Fig. 1. Flow diagram of the manual screening process for eligible literature inclusion. ADHD, attention deficit hyperactivity disorder.

Prevalence

Observed ADHD prevalence per study stratum by age group (youths ⩽18 years and adults >18 years) and gender showing all data (42 studies, 61 strata) and interview diagnosis strata (21 studies, 30 strata) are presented in Fig. 2. Bubble plots show the ADHD prevalence per study stratum for the full sample and the subset of study strata where a diagnostic interview was administered. The plots demonstrate that some of the high prevalence outliers are study strata where screening was used for ADHD diagnosis. Thus, all other data used diagnostic data only. No other differences were observed between the gender and age groups. Table 1 and Figs 3 and 4 (and Supplementary Figs S1 and S2) summarize the results obtained from model 1 (all data, that is both screening and diagnostic interview) and model 2 (diagnostic interview subset only). There were no significant differences between genders in both models.

Fig. 2. Observed attention deficit hyperactivity disorder prevalence per study stratum by age group (youths ⩽18 years and adults >18 years) and gender showing all data (42 studies, 61 strata) and interview diagnosis (Dx) strata (21 studies, 30 strata). Each bubble/square in the plot represents a study stratum. Bubble/square areas are proportional to stratum sample size. The numbers of study strata per column are shown in parentheses.

Fig. 3. Model-predicted attention deficit hyperactivity disorder (ADHD) prevalence by country (model 2: diagnostic interview only). a Upper bounds of 95% confidence intervals (CIs) are truncated to 1.0.

Fig. 4. Model-predicted attention deficit hyperactivity disorder (ADHD) prevalence by gender group and age groups (youths ⩽18 years and adults >18 years; model 2: diagnostic interview only). CI, Confidence interval.

Table 1. Summary of prevalence results (%) from fitted models (youths ⩽18 years and adults >18 years)

F, Female; M, male; n.a., not applicable.

a The number of study strata included are listed in parentheses for each covariate (number of strata in models 1 and 2 are provided for covariates in both models, respectively). The numbers of studies in each country are: Austria (n = 1), Brazil (n = 1), Canada (n = 1), Germany (n = 3), Greece (n = 1), Iceland (n = 2), Iran (n = 1), Ireland (n = 1), South Korea (n = 1), Netherlands (n = 2), Nigeria (n = 1), Spain (n = 2), Sweden (n = 6), UK (n = 3), USA (n = 16). All studies are listed in Supplementary Tables S1 and S2.

Overall, the observed ADHD prevalence in prison populations in all 42 studies (61 study strata) was 21.3% (total number of inmates was 26641, of whom 5677 were reported to be diagnosed with ADHD). The observed ADHD prevalence in prison populations for inmates with an interview diagnosis (21 studies, 30 study strata) was 20.5% (total number of inmates was 19575, of whom 4008 were reported to be diagnosed with ADHD).

Screening versus diagnostic clinical interview methodology

Studies using screening for diagnosis had a significantly higher estimated ADHD prevalence of 43.3% (95% CI 33.2–56.4) compared with 25.5% (95% CI 20.0–32.4) obtained from the subset of studies using a diagnostic clinical interview (p = 0.001). This suggested a high rate of false positives was being identified by screening tools, and thus subsequent reported analysis used prevalence obtained only from diagnostic clinical interview data (i.e. model 2).

Retrospective versus current youth diagnoses methodology

ADHD prevalence was estimated to be significantly lower (p < 0.001) in studies with current diagnoses; ADHD prevalence for current diagnoses was 21.7% (95% CI 17.5–26.9), compared with 36.4% (95% CI 25.6–51.8) for retrospective diagnoses.

Geographical differences

There were significant differences in ADHD prevalence estimated across countries (p < 0.0001). ADHD prevalence ranged from 6.6% in Brazil (Ponde et al. Reference Ponde, Freire and Mendonca2011) to 65.2% in Sweden (Table 1 and Fig. 2; model 2 diagnostic interview). When aggregated into three regions (North America, Europe and Other countries), differences were not significant when fitting models similar to model 2, replacing countries with regions (p = 0.240); Europe had the highest estimated prevalence (32.1%), followed by North America (26.9%) and Other countries (17.6%).

Males versus females

Prevalence estimates for males with ADHD were not significantly different from estimates for females with ADHD: 30.3% (95% CI 23.9–38.4) versus 26.1% (95% CI 19.3–35.2).

Youths versus adults

Youth prevalence estimates were not significantly different from estimates for adults: 30.1% (95% CI 22.1–41.1) versus 26.2% (95% CI 18.4–37.5).

Gender-by-age interaction

Gender-by-age (Table 1 and Fig. 4) comparisons reported that female adults had the lowest predicted ADHD prevalence: 22.1% (95% CI 13.6–35.8). By contrast, the prevalence estimate for male adults was 31.2% (95% CI 22.7–42.9). Female and male youth estimates were 30.8% (95% CI 22.4–42.2) and 29.5% (95% CI 21.6–40.1), respectively.

There were no significant differences found within the clinical interview diagnostic groups among the four gender-by-age groups.

Discussion

Prevalence

This review presents a meta-analysis of 42 studies conducted in 15 countries that provides a reliable estimate of 25.5% as the prevalence of ADHD in incarcerated populations from studies using diagnostic clinical interview. Heterogeneity is likely to be present in most meta-analyses as individual studies with the ‘same’ population and study characteristics are never identical with respect to both measured and unmeasured factors, which can lead to differences in outcome. For this reason, models that allow for heterogeneity are recommended (DerSimonian & Laird, Reference DerSimonian and Laird1986; van Houwelingen et al. Reference van Houwelingen, Arends and Stijnen2002). We fitted models that assumed a fixed effect (equal across studies) of measurable covariates, such as gender or age group, on ADHD prevalence and a random effect for study that allowed for the same types of subjects (e.g. male adult with a diagnostic interview and current diagnoses design) to have different prevalence in different studies, and also provided for a mechanism to allow prevalence correlation in the same study strata (intra-cluster correlation).

In recent years, a high prevalence of psychopathology (depression, psychotic illness, ADHD, conduct disorder) has been recognized in prison inmates (Fazel et al. Reference Fazel, Langstrom, Grann and Fazel2008a , Reference Fazel, Doll and Langstrom b ; Bradley, Reference Bradley2009), and although more comprehensive mental health screens are being developed for use at prison reception to develop effective care strategies (Senior et al. Reference Senior, Birmingham, Harty, Hassan, Hayes, Kendall, King, Lathlean, Lowthian, Mills, Webb, Thornicroft and Shaw2012), a possible explanation for the results of the meta-analysis may be that there is a high rate of false positives identified by ADHD screens in youths/adults compared with the diagnostic interview-only subset. Retrospective screening and interview accounts of childhood symptoms given by adult inmates were estimated to be significantly higher than those given by a current diagnoses cohort (36.4% v. 21.7%). It seems important that all inmates are considered for mental health screening, including ADHD, and those with positive screens be referred for a clinical assessment. What is unknown, however, is the rate of false-negative screens in this population.

The predominantly reported analysis used prevalence obtained only from diagnostic clinical interview data. This is also consistent with the diagnosis of ADHD in the community and enables a fairer comparison when comparing rates with the general population.

Consistent with prevalence and patterns of remission in the general population (Polanczyk et al. Reference Polanczyk, de Lima, Horta, Biederman and Rohde2007; Young & Gudjonsson, Reference Young and Gudjonsson2008; Simon et al. Reference Simon, Czobor, Balint, Meszaros and Bitter2009), we would expect adult prevalence to have been lower than youth prevalence; yet when applying a diagnostic interview (model 2), these were very similar (26.2% v. 30.1%). It has been reported that the onset of offending is at a younger age for youth with ADHD, even as young as 10 years of age (Langley et al. Reference Langley, Fowler, Ford, Thapar, van den Bree, Harold, Owen, O'Donovan and Thapar2010; Young et al. Reference Young, Wells and Gudjonsson2010b ), and perhaps these young offenders are less likely to be given a custodial sentence and/or are diverted out of the criminal justice system (e.g. to residential homes as ‘at-risk’ youths). However, an arbitrary classification was determined in the current review to stratify youth and adult offenders (i.e. if the mean/median age was above 18 years, a study was classified as an ‘adult study’). This might have led to some youth diagnoses being included as an adult diagnosis, and possibly skewed the data.

Geographical differences

Large differences were reported between countries. However, when grouped into three regions (North America, Europe and Other countries), these differences were not significant. Europe had the highest estimated rate, followed by North America and then Other countries. These variations may be due partly to study sample sizes: 70% (seven of 10) of the diagnostic interview studies carried out in North America had sample sizes >200, compared with only 22% (two of nine) of European studies. The North American prevalence (26.9%) is closer to the prevalence obtained overall (by diagnostic interview) for offenders with youth and adult ADHD (30.1% and 26.2%, respectively). Hence, these larger samples may give an accurate estimate of prevalence.

The differential may also reflect differences in the North American and European criminal justice systems. Following the introduction of tougher laws and penalties surrounding drug crimes by the US government, the number of inmates convicted of a drug crime rose by 1300% in the USA between 1980 and 2001 (Jensen et al. Reference Jensen, Gerber and Mosher2004). This compares with only 12% of a Scottish prison population incarcerated for drug-related crimes (Young et al. Reference Young, Wells and Gudjonsson2010b ). Nevertheless, just under half of this Scottish sample self-reported drug dependence, and hierarchical multiple regressions indicated that previous drug use was the most powerful predictor of the total extent of offending, whereas a history of ADHD symptoms was the most powerful predictor of violent crimes. Thus, US prison populations may have an over-representation of those who have been incarcerated for drug crimes, which may not necessarily be related to ‘reactive, impulsive offences' (e.g. robbery and property crimes) thought to be associated with ADHD (Retz & Rosler, Reference Retz and Rosler2009).

Gender differences

Female adults (with diagnostic interview) had the lowest predicted ADHD prevalence when comparing by age and gender (22.1% v. 31.2%, male adults). However, female youths had similar prevalence to male youths (30.8% and 29.5%, respectively). The general population prevalence indicates a 4:1 ratio of boys to girls with ADHD (Cuffe et al. Reference Cuffe, Moore and McKeown2005), yet in the prison population this ratio seems much reduced. Hence, the protective mechanisms that may usually keep females out of prison (e.g. sentencing policies, family considerations) seem to be overridden for female offenders with ADHD. However, only a few data sets reporting on female prevalence were included in the ‘diagnostic interview’ analysis (n = 8) compared with male prevalence (n = 22), which may have given a restricted representation of the position. A recent study from Konstenius et al. (Reference Konstenius, Larsson, Lundholm, Philips, Glind, Jayaram-Lindstrom and Franck2012) sampled 56 adult incarcerated women who were assessed by diagnostic interview in a Swedish prison, and reported that 29% met DSM-IV criteria for ADHD. This study, not identified during our initial literature search, supports a similar ADHD prevalence in female adults to this meta-analysis. In the present review, male adult diagnostic interview data are the most reliable because of the quantity and hence high power available: 14752 male youth interview subjects and 1978 male adult interview subjects compared with 2623 female youth and 222 female adult subjects.

The prevalence rate of males to females in adults with ADHD (~1.5:1) is much lower than the observed ratio of 4:1 in youths with ADHD (Montejano et al. Reference Montejano, Sasane, Hodgkins, Russo and Huse2011; Willcutt, Reference Willcutt2012; McCarthy et al. Reference McCarthy, Wilton, Murray, Hodgkins, Asherson and Wong2012). Male adults diagnosed by clinical interview showed a similar prevalence to youths (31.2% v. 30.8% female or 29.5% male). This suggests that incarcerated youths may remain in the criminal justice system because of repeat offending. Analysis of data from the Swedish National Register found that 36.6% males with ADHD were convicted of crime compared with 15.4% of females. Importantly, it was found that the use of ADHD medication reduced the risk of criminality among probands with ADHD: by 32% in men and 41% in women (Lichtenstein et al. Reference Lichtenstein, Halldner, Zetterqvist, Sjolander, Serlachius, Fazel, Langstrom and Larsson2012).

Limitations

Two important caveats to this review should be noted. First, publication bias is always an issue in systematic reviews, but efforts to address this were made in several ways. Data were obtained from all available sources, including those from electronic databases and authors, and review papers were checked (post-2006). Bias in study selection was addressed by delivering training on how to extract data from a random sample of publications, and each researcher checked and independently reviewed a random sample of each others’ papers for data interpretation consistency. Second, factors such as the reliability of ADHD diagnoses and differences in the criminal justice system between and within countries could have contributed to the broad ranges of prevalence observed in this review. The heterogeneity of samples used is also a potential source of variability and was the reason for using a fitted model that assumed a fixed effect (equal across studies) of measurable covariates but random study effects in the meta-analysis. In addition, only those publications since 1980 and in English were included.

Conclusions

Overall, the estimated prevalence of ADHD in incarcerated populations is 25.5% based on diagnostic clinical interviews. The ADHD prevalence was estimated to be significantly lower in studies with current diagnoses compared with retrospective diagnoses. There was a large variation in ADHD prevalence between countries, but this was no longer evident when grouped regionally. This systematic review and meta-analysis found, on average, a fivefold increased prevalence of ADHD in youth prison populations (30.1%) and a 10-fold increase in adult prison populations (26.2%) compared with published general population prevalence (3–7% and 1–5%, respectively). Female adults had a lower prevalence of ADHD, when compared by age and gender, than male adults. However, the risk for females was almost as high as that for males.

Supplementary material

For supplementary material accompanying this paper visit http://dx.doi.org/10.1017/S0033291714000762.

Acknowledgements

We thank E. Watts (Fishawack Communications Ltd, UK, funded by Shire AG, Switzerland) for performing the initial searches; G. Stoudt (Shire, USA) for performing the database literature searches; and J. Wright and K. Lay (Caudex Medical, UK, funded by Shire AG, Switzerland) for extracting relevant data from the studies. Editorial assistance in formatting, proofreading, copy-editing and fact-checking was also provided by Caudex Medical, funded by Shire AG, Switzerland. We also thank A. Elsner (Shire AG, Switzerland) for constructive review.

Declaration of Interest

S. Young has received speakers’ fees and travel honoraria from Janssen, Eli Lilly, Shire, Novartis and Flynn Pharma, and received research grants from Janssen, Eli Lilly and Shire. D. Moss is a professional medical writer employed by Caudex Medical, UK, whose contribution to this manuscript was funded by Shire AG, Switzerland. O. Sedgwick has no conflicts of interest to disclose. However, she receives funding support from the National Institute for Health Research (NIHR) Mental Health Biomedical Research Centre at South London and Maudsley National Health Service (NHS) Foundation Trust and King's College London. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. M. Fridman is a paid statistical consultant to Shire. P. Hodgkins was an employee of Shire and owns stock/stock options in the company. He is now Vice President, Global HEOR, Vertex Pharmaceuticals, USA. Shire develops and markets drugs to treat psychiatric disorders, including ADHD.

References

Anckarsater, H, Nilsson, T, Stahlberg, O, Gustafson, M, Saury, JM, Rastam, M, Gillberg, C (2007). Prevalences and configurations of mental disorders among institutionalized adolescents. Developmental Neurorehabilitation 10, 5765.CrossRefGoogle ScholarPubMed
APA (1994). Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. DSM-IV. American Psychiatric Association: Washington, DC.Google Scholar
Asherson, P (2004). Attention-deficit hyperactivity disorder in the post-genomic era. European Child and Adolescent Psychiatry 13 (Suppl. 1), I50I70.CrossRefGoogle ScholarPubMed
Asherson, P, Akehurst, R, Kooij, JJ, Huss, M, Beusterien, K, Sasane, R, Gholizadeh, S, Hodgkins, P (2012). Under diagnosis of adult ADHD: cultural influences and societal burden. Journal of Attention Disorders 16, 20S38S.CrossRefGoogle ScholarPubMed
Asherson, P, Kuntsi, J, Taylor, E (2005). Unravelling the complexity of attention-deficit hyperactivity disorder: a behavioural genomic approach. British Journal of Psychiatry 187, 103105.Google Scholar
Black, DW, Arndt, S, Hale, N, Rogerson, R (2004). Use of the Mini International Neuropsychiatric Interview (MINI) as a screening tool in prisons: results of a preliminary study. Journal of the American Academy of Psychiatry and Law 32, 158162.Google Scholar
Black, DW, Gunter, T, Loveless, P, Allen, J, Sieleni, B (2010). Antisocial personality disorder in incarcerated offenders: psychiatric comorbidity and quality of life. Annals of Clinical Psychiatry 22, 113120.Google Scholar
Bradley, K (2009). The Bradley Report. Lord Bradley's review of people with mental health problems or learning disabilities in the criminal justice system. Department of Health: London.Google Scholar
Bramham, J, Ambery, F, Young, S, Morris, R, Russell, A, Xenitidis, K, Asherson, P, Murphy, D (2009). Executive functioning differences between adults with attention deficit hyperactivity disorder and autistic spectrum disorder in initiation, planning and strategy formation. Autism 13, 245264.CrossRefGoogle ScholarPubMed
Cahill, BS, Coolidge, FL, Segal, DL, Klebe, KJ, Marle, PD, Overmann, KA (2012). Prevalence of ADHD and its subtypes in male and female adult prison inmates. Behavioral Sciences and the Law 30, 154166.Google Scholar
Colins, O, Vermeiren, R, Vahl, P, Markus, M, Broekaert, E, Doreleijers, T (2012). Parent-reported attention-deficit hyperactivity disorder and subtypes of conduct disorder as risk factor of recidivism in detained male adolescents. European Psychiatry 27, 329334.CrossRefGoogle ScholarPubMed
Coolidge, FL, Segal, DL, Klebe, KJ, Cahill, BS, Whitcomb, JM (2009). Psychometric properties of the Coolidge Correctional Inventory in a sample of 3,962 prison inmates. Behavioral Sciences and the Law 27, 713726.Google Scholar
Cuffe, SP, Moore, CG, McKeown, RE (2005). Prevalence and correlates of ADHD symptoms in the National Health Interview Survey. Journal of Attention Disorders 9, 392401.CrossRefGoogle ScholarPubMed
DerSimonian, R, Laird, N (1986). Meta-analysis in clinical trials. Controlled Clinical Trials 7, 177188.CrossRefGoogle ScholarPubMed
Einat, T, Einat, A (2008). Learning disabilities and delinquency: a study of Israeli prison inmates. International Journal of Offender Therapy and Comparative Criminology 52, 416434.CrossRefGoogle ScholarPubMed
Faraone, SV, Biederman, J, Mick, E (2006). The age-dependent decline of attention deficit hyperactivity disorder: a meta-analysis of follow-up studies. Psychological Medicine 36, 159165.Google Scholar
Fazel, M, Langstrom, N, Grann, M, Fazel, S (2008 a). Psychopathology in adolescent and young adult criminal offenders (15–21 years) in Sweden. Social Psychiatry and Psychiatric Epidemiology 43, 319324.CrossRefGoogle Scholar
Fazel, S, Doll, H, Langstrom, N (2008 b). Mental disorders among adolescents in juvenile detention and correctional facilities: a systematic review and metaregression analysis of 25 surveys. Journal of the American Academy of Child and Adolescent Psychiatry 47, 10101019.Google Scholar
Fulwiler, C, Forbes, C, Santangelo, SL, Folstein, M (1997). Self-mutilation and suicide attempt: distinguishing features in prisoners. Journal of the American Academy of Psychiatry and Law 25, 6977.Google ScholarPubMed
Ginsberg, Y, Lindefors, N (2012). Methylphenidate treatment of adult male prison inmates with attention-deficit hyperactivity disorder: randomised double-blind placebo-controlled trial with open-label extension. British Journal of Psychiatry 200, 6873.Google Scholar
Gordon, J, Moore, P (2005). ADHD among incarcerated youth: an investigation on the congruency with ADHD prevalence and correlates among the general population. American Journal of Criminal Justice 30, 8797.CrossRefGoogle Scholar
Gudjonsson, G, Sigurdsson, JF, Young, S, Newton, AK, Peersen, M (2009). Attention deficit hyperactivity disorder: how do ADHD symptoms relate to personality among prisoners. Personality and Individual Differences 47, 6468.CrossRefGoogle Scholar
Gudjonsson, GH, Sigurdsson, JF, Adalsteinsson, TF, Young, S (2012). The relationship between ADHD symptoms, mood instability, and self-reported offending. Journal of Attention Disorders 17, 339346.Google Scholar
Jensen, EL, Gerber, J, Mosher, C (2004). Social consequences of the war on drugs: the legacy of failed policy. Criminal Justice Policy Review 15, 100121.CrossRefGoogle Scholar
Kessler, RC, Adler, L, Barkley, R, Biederman, J, Conners, CK, Demler, O, Faraone, SV, Greenhill, LL, Howes, MJ, Secnik, K, Spencer, T, Ustun, TB, Walters, EE, Zaslavsky, AM (2006). The prevalence and correlates of adult ADHD in the United States: results from the National Comorbidity Survey Replication. American Journal of Psychiatry 163, 716723.CrossRefGoogle ScholarPubMed
Konstenius, M, Larsson, H, Lundholm, L, Philips, B, Glind, GV, Jayaram-Lindstrom, N, Franck, J (2012). An epidemiological study of ADHD, substance use, and comorbid problems in incarcerated women in Sweden. Journal of Attention Disorders. Published online: 13 July 2012 . doi: 10.1177/1087054712451126.Google ScholarPubMed
Langevin, R (2003). A study of the psychosexual characteristics of sex killers: can we identify them before it is too late? International Journal of Offender Therapy and Comparative Criminology 47, 366382.Google Scholar
Langley, K, Fowler, T, Ford, T, Thapar, AK, van den Bree, M, Harold, G, Owen, MJ, O'Donovan, MC, Thapar, A (2010). Adolescent clinical outcomes for young people with attention-deficit hyperactivity disorder. British Journal of Psychiatry 196, 235240.CrossRefGoogle ScholarPubMed
Liberati, A, Altman, DG, Tetzlaff, J, Mulrow, C, Gotzsche, PC, Ioannidis, JP, Clarke, M, Devereaux, PJ, Kleijnen, J, Moher, D (2009). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Medicine 6, e1000100.Google Scholar
Lichtenstein, P, Halldner, L, Zetterqvist, J, Sjolander, A, Serlachius, E, Fazel, S, Langstrom, N, Larsson, H (2012). Medication for attention deficit-hyperactivity disorder and criminality. New England Journal of Medicine 367, 20062014.CrossRefGoogle ScholarPubMed
Lindsay, WR, Carson, D, Holland, A, Michie, AM, Taylor, JL, Bambrick, M, O'Brien, G, Wheeler, JR, Steptoe, L (2012). A comparison of sex offenders and other types of offenders referred to intellectual disability forensic services. Psychiatry, Psychology and Law 19, 566576.CrossRefGoogle Scholar
McCarthy, S, Wilton, L, Murray, ML, Hodgkins, P, Asherson, P, Wong, IC (2012). The epidemiology of pharmacologically treated attention deficit hyperactivity disorder (ADHD) in children, adolescents and adults in UK primary care. BMC Pediatrics 12, 78.Google Scholar
Milin, R, Halikas, JA, Meller, JE, Morse, C (1991). Psychopathology among substance abusing juvenile offenders. Journal of the American Academy of Child and Adolescent Psychiatry 30, 569574.Google Scholar
Montejano, L, Sasane, R, Hodgkins, P, Russo, L, Huse, D (2011). Adult ADHD: prevalence of diagnosis in a US population with employer health insurance. Current Medical Research Opinion 27 (Suppl. 2), 511.CrossRefGoogle Scholar
Polanczyk, G, de Lima, MS, Horta, BL, Biederman, J, Rohde, LA (2007). The worldwide prevalence of ADHD: a systematic review and metaregression analysis. American Journal of Psychiatry 164, 942948.CrossRefGoogle ScholarPubMed
Ponde, MP, Freire, AC, Mendonca, MS (2011). The prevalence of mental disorders in prisoners in the city of Salvador, Bahia, Brazil. Journal of Forensic Sciences 56, 679682.CrossRefGoogle ScholarPubMed
Rasmussen, K, Almvik, R, Levander, S (2000). ADHD in prisoners: an unfavourable index of criminal careers and a forensic challenge. Nordic Journal of Psychiatry 54 (Suppl. 43), 47.Google Scholar
Rasmussen, K, Almvik, R, Levander, S (2001). Attention deficit hyperactivity disorder, reading disability, and personality disorders in a prison population. Journal of the American Academy of Psychiatry and Law 29, 186193.Google Scholar
Retz, W, Rosler, M (2009). The relation of ADHD and violent aggression: what can we learn from epidemiological and genetic studies? International Journal of Law and Psychiatry 32, 235243.CrossRefGoogle ScholarPubMed
Retz, W, Rosler, M (2010). Association of ADHD with reactive and proactive violent behavior in a forensic population. Attention Deficit Hyperactivity Disorder 2, 195202.Google Scholar
Rose, E, Bramham, J, Young, S, Paliokostas, E, Xenitidis, K (2009). Neuropsychological characteristics of adults with comorbid ADHD and borderline/mild intellectual disability. Research in Developmental Disabilities 30, 496502.CrossRefGoogle ScholarPubMed
Rosler, M, Retz, W, Yaqoobi, K, Burg, E, Retz-Junginger, P (2009). Attention deficit/hyperactivity disorder in female offenders: prevalence, psychiatric comorbidity and psychosocial implications. European Archives of Psychiatry and Clinical Neuroscience 259, 98105.CrossRefGoogle ScholarPubMed
Sanz-Garcia, O, Dueñas, RM, Burges, V, Muro, A, Perez, F (2010 a). Comorbidity in patients with attention deficit hyperactivity disorder in psychiatric population in prison. European Neuropsychopharmacology 20 (Suppl. 3), S616.Google Scholar
Sanz-Garcia, O, Dueñas, RM, Domenech, N, Ramón, R, Muro, A, Perez, F (2010 b). How do we treat patients with undiagnosed attention deficit hyperactivity disorder? European Neuropsychopharmacology 20 (Suppl. 3), S623.CrossRefGoogle Scholar
Senior, J, Birmingham, L, Harty, MA, Hassan, L, Hayes, AJ, Kendall, K, King, C, Lathlean, J, Lowthian, C, Mills, A, Webb, R, Thornicroft, G, Shaw, J (2012). Identification and management of prisoners with severe psychiatric illness by specialist mental health services. Psychological Medicine 43, 15111520.CrossRefGoogle ScholarPubMed
Simon, V, Czobor, P, Balint, S, Meszaros, A, Bitter, I (2009). Prevalence and correlates of adult attention-deficit hyperactivity disorder: meta-analysis. British Journal of Psychiatry 194, 204211.Google Scholar
Tidefors, I, Strand, J (2012). Life history interviews with 11 boys diagnosed with attention-deficit/hyperactivity disorder who had sexually offended: a sad storyline. Journal of Trauma and Dissociation 13, 421434.Google Scholar
van Houwelingen, HC, Arends, LR, Stijnen, T (2002). Advanced methods in meta-analysis: multivariate approach and meta-regression. Statistics in Medicine 21, 589624.CrossRefGoogle ScholarPubMed
Vitelli, R (1996). Prevalence of childhood conduct and attention-deficit hyperactivity disorders in adult maximum-security inmates. International Journal of Offender Therapy and Comparative Criminology 40, 263271.CrossRefGoogle Scholar
WHO (1992). The ICD-10 Classification of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines. World Health Organization: Geneva.Google Scholar
Willcutt, EG (2012). The prevalence of DSM-IV attention-deficit/hyperactivity disorder: a meta-analytic review. Neurotherapeutics 9, 490499.Google Scholar
Young, S, Chesney, S, Sperlinger, D, Misch, P, Collins, P (2009 a). A qualitative study exploring the life-course experiences of young offenders with symptoms and signs of ADHD who were detained in a residential care setting. Criminal Behaviour and Mental Health 19, 5463.Google Scholar
Young, S, Gudjonsson, G, Ball, S, Lam, J (2003). Attention deficit hyperactivity disorder (ADHD) in personality disordered offenders and the association with disruptive behavioural problems. Journal of Forensic Psychiatry and Psychology 14, 491505.CrossRefGoogle Scholar
Young, S, Gudjonsson, G, Misch, P, Collins, P, Carter, P, Redfern, J, Goodwin, E (2010 a). Prevalence of ADHD symptoms among youth in a secure facility: the consistency and accuracy of self- and informant-report ratings. Journal of Forensic Psychiatry and Psychology 21, 238246.CrossRefGoogle Scholar
Young, S, Gudjonsson, GH (2008). Growing out of ADHD: the relationship between functioning and symptoms. Journal of Attention Disorders 12, 162169.CrossRefGoogle ScholarPubMed
Young, S, Gudjonsson, GH, Wells, J, Asherson, P, Theobald, D, Oliver, B, Scott, C, Mooney, A (2009 b). Attention deficit hyperactivity disorder and critical incidents in a Scottish prison population. Personality and Individuality Differences 46, 265269.Google Scholar
Young, S, Misch, P, Collins, P, Gudjonsson, G (2011). Predictors of institutional behavioural disturbance and offending in the community among young offenders. Journal of Forensic Psychiatry and Psychology 22, 7286.CrossRefGoogle Scholar
Young, S, Morris, R, Toone, B, Tyson, C (2007). Planning ability in adults with attention-deficit/hyperactivity disorder. Neuropsychology 21, 581589.Google Scholar
Young, S, Wells, J, Gudjonsson, GH (2010 b). Predictors of offending among prisoners: the role of attention deficit hyperactivity disorder (ADHD) and substance use. Journal of Psychopharmacology 25, 15241532.Google Scholar
Figure 0

Fig. 1. Flow diagram of the manual screening process for eligible literature inclusion. ADHD, attention deficit hyperactivity disorder.

Figure 1

Fig. 2. Observed attention deficit hyperactivity disorder prevalence per study stratum by age group (youths ⩽18 years and adults >18 years) and gender showing all data (42 studies, 61 strata) and interview diagnosis (Dx) strata (21 studies, 30 strata). Each bubble/square in the plot represents a study stratum. Bubble/square areas are proportional to stratum sample size. The numbers of study strata per column are shown in parentheses.

Figure 2

Fig. 3. Model-predicted attention deficit hyperactivity disorder (ADHD) prevalence by country (model 2: diagnostic interview only). a Upper bounds of 95% confidence intervals (CIs) are truncated to 1.0.

Figure 3

Fig. 4. Model-predicted attention deficit hyperactivity disorder (ADHD) prevalence by gender group and age groups (youths ⩽18 years and adults >18 years; model 2: diagnostic interview only). CI, Confidence interval.

Figure 4

Table 1. Summary of prevalence results (%) from fitted models (youths ⩽18 years and adults >18 years)

Supplementary material: File

Young Supplementary Material

Figure S1

Download Young Supplementary Material(File)
File 804.1 KB
Supplementary material: File

Young Supplementary Material

Figure S2

Download Young Supplementary Material(File)
File 821.5 KB
Supplementary material: File

Young Supplementary Material

Table S1

Download Young Supplementary Material(File)
File 87 KB
Supplementary material: File

Young Supplementary Material

Table S2

Download Young Supplementary Material(File)
File 104.4 KB
Supplementary material: File

Young Supplementary Material

Figure legends

Download Young Supplementary Material(File)
File 24.6 KB