Impulsive Lifestyle Counselling versus treatment as usual to reduce offending in people with co-occurring antisocial personality disorder and substance use disorder: a post hoc analysis | BMC Psychiatry

Design and settings

A pragmatic randomized trial was conducted between January 2012 and June 2014 in 13 sites in Denmark [29]. Patients enrolled in free-of-charge community outpatient treatment for people with SUDs were approached by the clinical staff and assessed using the ASPD module from the Mini International Neuropsychiatric Interview, version 5 [31].

Inclusion criteria were being between 18–65 years old, seeking or currently receiving treatment for an SUD, meeting lifetime and last-year criteria for ASPD according to DSM-IV criteria, and being able to provide informed consent. The DSM-IV criteria for ASPD met by the patients were reviewed and explained individually to each patient before patients consented to participate in the study. Patients were excluded from the trial if they were participating in group therapy with another patient enrolled in the trial, had acute psychosis or severe brain damage, did not speak Danish, or had plans that would interfere with study participation over the next three months. All participating patients were informed that according to the MINI International Neuropsychiatric Interview [31], they met the criteria for ASPD.

Participants

In total, 176 patients were randomized in the trial, and, of these, 165 (93%) could be identified in the registers (ILC: n = 91; 54%; TAU: n = 74, 44.9%). In all, 12.7% were women, the median age was 31.5, and the interquartile range was 25.5–38 years (Table 1). Nearly half (46.7%) were undergoing medication-assisted treatment (MAT) for opioid dependence. In the year prior to randomization, nearly half of both groups had committed a crime leading to conviction (ILC: n = 44, 48%; TAU: n = 36, 49%). There were minor differences in specific offending types between groups, but none approached statistical significance (ps > 0.10).

Psychiatric diagnoses from the Danish Psychiatric Central Research Register are summarized in Table 1. In total, 71 (43%) had at least one psychiatric diagnosis, and 34 (20.6%) had been in inpatient psychiatric care in the past 10 years. The most common diagnoses were mood or anxiety disorders (n = 52, 31.5%) followed by attention deficit/hyperactivity disorder (n = 11, 6.7%). No difference was found in the prevalence rate of diagnoses between the TAU group (n = 35, 47.3%) and the ILC group (36, 39.6%), χ²(1) = 1.00, p = 0.318.

Randomization

Random assignment was conducted in blocks of varying sizes at the Centre for Alcohol and Drug Research. Blocks varied between four and six patients per block within each site. Clinicians were informed of the results of the randomization only after the baseline assessment had been completed.

Interventions

Treatment as usual (TAU)

Patients in both treatment conditions had access to counselling and medication for drug use disorders in Denmark under the Act of Social Services § 101 and for alcohol use disorders under the Healthcare Act § 141. When patients were randomly assigned to the TAU condition, clinicians were explicitly asked to ensure that the patients got the highest possible level of care, based on mutual agreement between the counsellor and the patient. TAU always included access to MAT for patients with opioid use disorders, psychosocial support in the form of casework and counselling, as well as referral to residential rehabilitation if this was deemed relevant. At some clinics, a liaison psychiatrist saw patients onsite, whereas patients in other clinics were referred to an off-site psychiatrist for diagnosis and treatment of other psychiatric conditions, such as attention-deficit/hyperactivity disorder, anxiety, or depression.

Impulsive Lifestyle Counselling (ILC)

ILC is a six-session psychoeducational add-on module to usual care that focuses on raising awareness of maladaptive antisocial behaviors. In brief, the program is inspired by the Lifestyle Theory and the Lifestyle Change Program developed by Glenn D. Walters [32, 33]. The ILC program contains a number of elements from the Lifestyle Change Program, such as introducing the patient to the concepts of behavioral styles, actions, choices, and consequences in relation to crime and impulsive behaviors. However, in order to reach a wider group of people with comorbid SUDs outside of prison, the ILC program was adjusted to an individual format, and the lifestyle approach was labeled ‘impulsive lifestyle’ rather than ‘criminal lifestyle’, addressing impulsive behavior related to ASPD, including problems with substance use and conflicts with others. The six sessions in the ILC program cover topics related to antisocial behavior, including a simplified Triggers-Actions-Consequences model, streetwise pride, values that increase or decrease impulsive actions, how social contacts may support or challenge lifestyle changes, and a booster session in which the patient and the clinician summarize the sessions and discuss future work with lifestyle changes. The content of the ILC program has been elaborately discussed in previous papers [29, 34], with the workbook available as supplementary material for this article.

Clinicians who delivered the ILC sessions at the participating sites attended a one-and-a-half-day workshop, where they were introduced to ASPD and the ILC program and practiced using the workbook by role-playing the sessions. The clinicians were encouraged to try to complete the sessions in the ILC program on a weekly basis, except for the booster session, which was to be delayed for six weeks. Sessions were planned to last 45–60 min, and the median number of sessions completed was two [29].

Registers and data linkage

Baseline data from the trial were linked with register data, including date of death, socio-demographic data, and criminal justice data on a secure Statistics Denmark server. Further, we combined data from different registers to obtain information on psychiatric and alcohol-related diagnoses registered over a 10-year period prior to randomization.

The Central Criminal Register was established in 1978 and contains information on offenses and offenders in criminal cases for use in criminal procedures. The information is updated on a regular basis by the police districts in Denmark and the departments of the National Commissioner of Police [35]. The register was used in this study to obtain information on convictions up to one year after randomization.

The Danish Psychiatric Central Research Register contains diagnoses given by a medical doctor based on ICD-10 codes, as well as dates of treatment onset and termination [36]. While validation studies have been limited to specific diagnoses, the register is almost complete for hospital-based care, and thus most patients with moderate to severe mental health problems are likely to be included [36].

The National Patient Register was used to obtain alcohol-related diagnoses and covers all hospital contacts for somatic conditions [37].

The Danish Prescription Register was used to obtain information about prescription drugs received for alcohol use disorders [38]. This register contains information about all prescriptions filled by residents in Denmark, and each record contains the ATC code for the drug, the date of prescription filling, and the patient’s individual identification number.

Outcomes

The primary outcome was the total number of crimes committed during the first year after study randomization leading to a conviction (i.e., not a warning or charges dropped). Secondary outcomes included number of specific offences that could be directly linked to antisocial behavior: property offences, violent offences, drug-related offences (excluding simple possession of drugs for own use), and driving under the influence of alcohol and drugs (DUI). In order to avoid small cells in the analyses, violent offences included sexual offences and weapons offences, as they both involve aggressive behavior towards others [39].

For both the primary and secondary outcomes, we considered only the first year after randomization as the observation period. This timeframe allowed enough time for the patients to commit an offence, while at the same time being able to observe treatment effects that may otherwise lose strength over a longer period due to external factors, such as life events or relapse to severe substance use [40]. The date of crime was the date at which the police believed that the criminal activity was initiated according to the recorded charge(s).

Control variables

In all analyses, models adjusted for age, gender, and MAT at baseline, similar to previous reports from this trial [29], as well as previous offending of the same type in the year prior to randomization. All of these variables could potentially influence offending behavior. Crime rates differ by gender [41] and decline with age [5], and both male gender and lower age are associated with crime within samples of patients treated for substance use disorders [42].

Demographic and clinical variables

To better characterize our sample, we assessed the presence of severe psychiatric illness, mood or anxiety disorder, attention deficit/hyperactivity disorder, substance induced psychoses, or alcohol- related problems, all within a time frame of 10 years prior to randomization (3650 days).

Severe mental illness was defined as the presence of a schizophrenia spectrum disorder (F2X) or a bipolar disorder (F30-F319) diagnosis in the Danish Psychiatric Central Research Register. Mood or anxiety disorder was defined as the presence of either a unipolar mood disorder (F32-F99X) or an anxiety disorder (F4X). A personality disorder was defined as the presence of a personality disorder regardless of type (F6X), and hyperkinetic disorder was defined as the presence of an F900X diagnosis.

We defined alcohol-related problems as any hospital contact, inpatient or outpatient, involving an alcohol use disorder diagnosis (ICD-10 code F10X) identified in the National Patient Register or the Psychiatric Research Register, or the filling of a prescription for a drug used in the treatment of alcohol dependence (ATC code N07BB) identified in the Danish Prescription Register.

Statistical analyses

Since the outcomes were count variables, we first explored models appropriate for this type of variable (i.e., Poisson, negative binomial, and zero-inflated models). To select the most parsimonious model, we relied on the Bayesian Information Criterion (BIC, [43]). The BIC takes on lower values as the model becomes more parsimonious, taking both model fit and model complexity into consideration.

Analyses for the best model were conducted with ILC randomization status as the variable of interest. In additional steps, we included the control variables listed above. According to the BIC, the best-fitting model was simple negative binomial regression for all outcomes in this study.