top of page
Search

Recidivism, Mental Health and RMV Evaluations in Massachusetts

Writer: Joshua Silver
Joshua Silver
Oct 10
17 min read

Impaired driving in Massachusetts is a public-safety problem and, for many drivers, a clinical one. Alcohol and drug involvement in serious crashes persists year after year, and research on repeat offenders shows that untreated substance use disorders and co-occurring psychiatric conditions are common among them. This article summarizes official crash and enforcement data, the evidence on psychiatric conditions and medications as they relate to driving, the legal framework that brings drivers to a clinician, and what "recidivism" means when a clinician is asked to assess it.


Key findings

  • Deaths remain high. Massachusetts recorded 363 traffic deaths in 2024. NHTSA estimates that 100 of them (27%) involved a driver at or above a 0.08 blood alcohol concentration (BAC), compared with 11,904 deaths (30%) nationally (NHTSA, DOT HS 813 813).

  • The alcohol share is partly estimated. Only 226 of 518 Massachusetts drivers in fatal crashes in 2024 (44%) had a known BAC, so the rest of the picture rests on NHTSA's estimates (same source).

  • Drug involvement is under-measured. Police-reported drug-suspected crashes were 787 in federal fiscal year 2023, 794 in 2024 and 800 in 2025 (preliminary) in the state's latest annual report, which revised the higher figures in the prior year's report, and the state says it cannot distinguish drug types (EOPSS FY2025 Annual Report; FY2024 Annual Report).

  • Repeat offenders carry a heavy clinical load. In a Massachusetts inpatient program for 729 repeat OUI offenders, 40.7% met criteria for lifetime alcohol dependence, against 7.1% in a general-population comparison group (Shaffer et al., 2007).

  • Other psychiatric conditions are reported across a wide range. Studies of convicted offenders report non-substance psychiatric disorders in roughly one third to about one half, and alcohol use disorders from about 40% lifetime dependence to nearly all, depending on sample and method. No systematic review pools these studies, so the comorbidity table below gives each source and its limits instead of one number.

  • Treatment and monitoring work, in combination. NHTSA rates alcohol problem assessment and treatment, and ignition interlocks, as five-star countermeasures. Median interlock re-arrest risk fell to a ratio of 0.25 while the device was installed but returned to 0.93 after removal (Elder et al., 2011; NHTSA, Countermeasures That Work, 11th ed.).

  • Psychiatric diagnosis alone is a weak predictor. A systematic review of 24 studies found no psychiatric disorder category consistently linked to higher crash risk, and the authors did not support blanket driving restrictions (Rapoport et al., 2023).


Two cautions apply throughout. First, official figures use different definitions ("alcohol-impaired," "alcohol suspected," "OUI arrest"), so they should not be added together or compared across tables without checking the definition. Second, statistics describe populations, not individuals. A driver's own history, test results and treatment engagement are what a clinician evaluates, as later sections explain.



Crash, OUI and drug-impaired driving data in Massachusetts

Deaths and serious injuries

Massachusetts recorded 434 traffic deaths in 2022 (MassDOT) and 363 in 2024 (NHTSA). NHTSA's alcohol-impaired category (a driver at a BAC of 0.08 or higher) covers a minority of those deaths, but a persistent one, and the alcohol-impaired shares rest on estimation because only 226 of 518 Massachusetts drivers in fatal crashes in 2024 (44%) had a known BAC. Serious injuries fell from 3,017 in 2022 to 2,859 in 2023 and 2,751 in 2024 (the 2024 figure is a draft) (MassDOT 2026 HSIP implementation plan). Nationally, CDC reports 13,524 deaths in alcohol-impaired-driver crashes in 2022, or 32% of traffic deaths (CDC).


Enforcement and convictions

The Massachusetts Sentencing Commission's fiscal year 2023 survey estimates 2,098 OUI convictions, or 10.8% of an estimated 19,504 convictions (Survey of Sentencing Practices FY2023). Arrests, charges and convictions are different counts, and the gap between them reflects dismissals, continuances, reduced charges and continued-without-a-finding dispositions. We could not retrieve a verified statewide OUI arrest count for this article. We found no published RMV totals for interlock installations or OUI-related suspensions.


Drugs and medications in crashes

Police reports flag drug-suspected crashes at 787 in federal fiscal year 2023, 794 in 2024 and 800 in 2025 (preliminary); the FFY2024 report had given higher figures (920 and 953 for 2023 and 2024), so the series has been revised. The state had 102 certified Drug Recognition Experts in September 2023 and 83 in September 2024 (EOPSS FY2025; FY2024). The state report says Massachusetts cannot distinguish different types of drugs in crash data, so these counts cannot be split into cannabis, opioids, stimulants or prescribed psychiatric medication. Nationally, a CDC study of 4,243 seriously injured drivers at seven trauma centers found 54% positive for alcohol or drugs: alcohol 22%, cannabis 25%, opioids 9%, stimulants 10% and sedatives 8% (categories overlap; the page does not say whether drugs were prescribed). Self-report surveys add context: in 2018, 4.7% of Americans aged 16 and older reported driving under the influence of marijuana in the past 12 months and 8.0% under the influence of alcohol (CDC MMWR).


What these data mean for clinicians

These figures show the scale of the problem and its measurement gaps. They cannot show how many offenders have a diagnosable mental health or substance use disorder; the clinical literature in the next section addresses that. For town-level crash counts, see the local statistics on each of our RMV evaluation pages.


Mental health, psychiatric medication and driving risk

Co-occurring psychiatric conditions among OUI/DUI offenders: a range, not a single figure

How common are psychiatric conditions among people convicted of impaired driving? Published estimates differ widely, and no single number can honestly be offered as fact. Studies use different samples (court-mandated inpatient programs, community probation, treatment admissions), different diagnostic methods (structured interviews versus chart diagnoses), different time frames (lifetime versus past year) and different comparison groups. We found no systematic review or meta-analysis that pools these studies, so the table below lists individual findings with their limits. Read it as a range of reported values, not a prevalence rate for Massachusetts drivers.


Source and sample

What was reported

Main limits

Shaffer et al., 2007, 729 repeat DUI offenders in a Massachusetts inpatient treatment-sentencing program

40.7% lifetime alcohol dependence versus 7.1% in a general-population comparison. Lifetime alcohol use disorder in about 98%. Higher-than-comparison rates of drug dependence, PTSD, generalized anxiety, bipolar disorder and conduct disorder; no difference for major depression, dysthymia, intermittent explosive disorder or ADHD. About 44% had a lifetime non-substance psychiatric diagnosis in addition to substance use disorders (Responsibility.org summary; Nelson et al., 2015, which says approximately 45%). NHTSA's Countermeasures That Work page gives a figure above 60% and attributes it to this study, which does not match the paper

Offenders sentenced to a residential treatment program; lifetime diagnoses; single program

Mullen et al., 2015, 119 adults with at least one DWI (residential correctional and community)

96.6% met criteria for an alcohol use disorder

Small sample (59 residential, 60 community)

Older convicted-offender studies (Lapham and colleagues, 2000 to 2001), as summarized by Responsibility.org

About two-thirds alcohol dependent; lifetime alcohol abuse or dependence in 91% of men and 83% of women; among offenders with an alcohol use disorder, 33% of men and 50% of women had another psychiatric disorder

Secondary, advocacy-organization summary; original papers not reviewed here; data are two decades old

Freeman et al., 2010, 36,372 Texas DUI treatment clients

Depression was the most common psychiatric diagnosis; elevated risk of a bipolar diagnosis compared with the general population

Abstract only; no percentages published

McMillan et al., 2007, 233 repeat DUI offenders in mandated treatment

Clinic providers missed comorbid illness that research screening found (adjusted estimates): about 97% of bipolar disorder (only 14 cases), 68% of major depression and 37% of drug use disorders; all 6 OCD cases were missed

Measures under-diagnosis, not prevalence; small counts for bipolar disorder and OCD


Across these sources, the alcohol use disorder figures cluster at the high end (from roughly 40% lifetime dependence in one repeat-offender program to nearly all of the offenders in two other samples), while the share with an additional non-substance psychiatric disorder is reported from about one in three to about one in two in the sources above, depending on sample and method. Three consistent themes survive the variation. Substance use disorders are the norm rather than the exception. Anxiety, trauma-related and bipolar disorders appear more often than in comparison groups in the Massachusetts study. And these conditions are frequently unrecognized without structured assessment, which is the central argument for a comprehensive evaluation such as our RMV substance abuse evaluation with risk of recidivism assessment.


Comorbidity does not by itself predict who will offend again with an OUI. In a prospective follow-up of 743 repeat DUI offenders, Nelson et al. (2015) found a 7.5% DUI re-offense rate and a 15% motor-vehicle-related re-offense rate within five years; ADHD predicted motor-vehicle re-offending, and alcohol, drug and conduct disorders predicted general criminal re-offending, but specific mental disorders did not predict DUI re-offending. The low base rate and a single court-mandated inpatient sample limit what can be concluded. Conditions such as anxiety, depression, ADHD, bipolar disorder and PTSD are treatable, and treating them is part of reducing risk.


Conditions that tend to surface in these evaluations

A good evaluation does not stop at the alcohol or drug question. It asks what else is going on, because untreated conditions are often part of why someone is back in front of the RMV. The conditions below are the ones the literature above keeps returning to, and they are all part of the conditions we treat.


Depression and anxiety are common companions of heavy drinking. The Texas treatment-admissions abstract found depression to be the most common psychiatric diagnosis among DUI clients, although the Massachusetts repeat-offender study did not find major depression more common than in the comparison group, while generalized anxiety was. People often describe drinking as a way to manage low mood or worry, which is why treating the underlying condition matters as much as addressing the alcohol.


PTSD stood out in the same Massachusetts study as more common among repeat offenders than in the general population. Trauma, sleep disruption and alcohol feed each other, so an evaluation that asks only about drinking can miss the driver of the pattern.


Bipolar disorder deserves special mention because it is easy to overlook. Clinic providers in the under-diagnosis study missed nearly all of the bipolar disorder that structured research screening identified, and an elevated rate of bipolar diagnoses appeared among Texas DUI clients. A manic or mixed episode can bring impulsivity and substance use together, and the treatment plan changes completely once it is recognized.


ADHD is the condition with the clearest link to driving outcomes: it predicted motor-vehicle re-offending in the prospective follow-up, and registry and insurance-claims studies found fewer crashes during medicated than unmedicated periods. Mood and emotional regulation problems, including irritability and impulsive anger, overlap with this picture and are worth discussing openly in an evaluation.


Sleep disturbance belongs in the conversation for a practical reason: some sleep medications can impair driving the next morning, which is why the FDA changed zolpidem dosing. OCD was among the conditions that clinics missed in the under-diagnosis study, although the numbers were small. And for people living with schizophrenia, the systematic review above found no category of disorder that consistently predicted crashes, so fitness to drive turns on current symptoms, stability and treatment rather than on the diagnosis itself, which is the question the MAB118 form asks the clinician to address.


Psychiatric conditions and crash risk

The evidence that a psychiatric diagnosis raises crash risk is mixed. A systematic review of 24 studies (Rapoport et al., 2023) found that 15 reported increased risk and 9 did not, with most studies of fair or poor quality, and concluded that no disorder category was consistently linked to higher risk. ADHD is the best-studied exception: a meta-analysis of 16 studies reported a relative crash risk of 1.36 unadjusted and 1.23 after controlling for driving exposure, and Swedish registry and US claims studies found lower crash risk in medicated than unmedicated periods (Chang et al., 2014; Chang et al., 2017). These are observational findings and do not prove causation.


Psychiatric medication and driving

Medication effects depend on drug class, dose, timing and the person. A meta-analysis of benzodiazepine studies found pooled odds ratios of about 1.6 for crashes in both case-control and cohort studies. A Danish registry study of about 130,000 drivers in injury crashes in Denmark found elevated odds for filled prescriptions of benzodiazepines and Z-drugs, antidepressants and ADHD stimulants, with the largest increases for single-vehicle crashes; it could not separate medication effects from the illness being treated. By contrast, the NHTSA Virginia Beach case-control study in a single city found no statistically significant crash risk for sedatives, narcotic analgesics, stimulants or antidepressants after adjusting for alcohol and demographics, while a breath alcohol level of 0.05 or more with no drugs present carried an odds ratio of 6.75. The FDA's 2013 zolpidem label change lowered recommended starting doses because of next-day impairment risk.


The practical point is that having a drug detected is not the same as being impaired, and prescribed treatment is not the same as misuse. Massachusetts law prohibits operating under the influence of intoxicating liquor or of marijuana, narcotic drugs, depressants or stimulant substances (M.G.L. c. 90, § 24); the statute lists categories of substances, not prescriptions by name. Patients should review driving safety with their prescriber whenever a medication is started or changed, which is part of our psychiatric medication management and online psychiatric evaluation services.


View through a car windshield of a quiet two-lane Massachusetts road at dusk in autumn, hands resting on the steering wheel

RMV evaluations: process and legal framework

Most drivers reach a psychiatric or substance use evaluation by one of three routes: a court order after an OUI charge, a Registry of Motor Vehicles (RMV) request tied to license reinstatement, or an RMV medical review of whether a driver is fit to drive. The statutes and regulations below are the formal framework; this article describes them in general terms and is not legal advice.


Court-ordered assessment and education

Under M.G.L. c. 90, § 24, penalties for operating under the influence escalate with prior offenses, and license-loss periods run from one year for a first offense to two years for a second, eight for a third and ten for a fourth, with lifetime loss for a fifth or later offense. Under c. 90, § 24Q, a court-ordered alcohol assessment is mandatory at sentencing when the evidence shows a blood alcohol level of 0.20 or higher or the person has a prior assignment or conviction of this kind. Section 24D lets the court place a consenting first offender on probation for up to two years; the Department of Public Health (DPH) program used for this, now called the Massachusetts Impaired Driving program, is 38 hours in length. Under what is commonly called Melanie's Law, people with two or more convictions generally must install an ignition interlock device, and some first offenders with a breath result of 0.15 or higher must do so for a hardship license.


RMV reinstatement and the Immediate Threat process

The RMV may require DPH-approved alcohol education or counseling as a condition of reinstatement. Separately, under c. 90, § 22 the RMV can suspend a license without a prior hearing where continued operation is an immediate threat to public safety, sending written notice, and the driver is entitled to a hearing within 30 days of the suspension. Hearings are governed by 540 CMR 9, which, among other circumstances, defines an unfit driver to include one with a serious mental or physical functional incapacity, naming addiction to alcohol or to a controlled substance scheduled under M.G.L. c. 94C. Drivers facing this process often need current clinical documentation to show that a condition is diagnosed, treated and stable.


Form MAB118 and the psychiatric evaluation

The RMV's medical affairs process uses Form MAB118 for conditions that may affect driving. The form is completed by a Massachusetts-licensed physician or psychiatric nurse practitioner and sent to the RMV Driver Control Unit. It covers diagnosis, symptoms, medications and the clinician's opinion on fitness to drive. Our clinicians complete these through the RMV psychiatric evaluation (Form MAB118) service; you can review who performs them on the team page, including Caryn Metzger and Thomas Smith.


What the RMV does not specify

We did not find an RMV page that defines a "risk of recidivism" evaluation or maintains an approved-evaluator scheme for one. When an RMV, court, attorney or employer asks for such an assessment, the content and standard are set by the requesting party and by the clinician's professional judgment. That is why the next section explains how the term is used clinically and why any report should state its method, sources and limits. Our RMV substance abuse evaluation with risk of recidivism assessment is built around that principle.


Recidivism: what it means and how clinicians assess risk

What recidivism means in this context

In criminal justice research, recidivism is the return to prohibited behavior after a sanction or intervention. For impaired driving it is measured in different ways: a new arrest, a new conviction, a new alcohol-related crash, or a new license action. It is also measured over different windows, commonly one, two or five years. The Massachusetts Sentencing Commission's 2002 recidivism study, for example, used a new arrest or an incarcerating technical violation within one year, a general definition that is not specific to OUI. Because definitions vary, recidivism figures from different studies cannot be compared directly. In a clinical report, "risk of recidivism" should be read as an estimate of the likelihood of repeating impaired driving, given a stated definition and time frame, and not as a prediction about any particular day.


What the numbers say

Repeat offending is substantial but not universal. NHTSA reports that about 25% of arrested drivers had a prior DWI arrest and about 30% of convicted drivers had prior convictions (state look-back periods vary); a 1994 CDC analysis put the figure near one third; and a 1999 Bureau of Justice Statistics report (1995 to 1997 survey data) found prior DWI sentences for about a third of DWI probationers, about 61% of jail inmates and about 62% of state prisoners. We found no Massachusetts-specific repeat-OUI rate. Follow-up of 743 repeat offenders found a DUI re-offense rate of 7.5% and a motor-vehicle re-offense rate of 15% within five years (Nelson et al., 2015). These figures describe different populations and definitions and should be read as a range.


Factors clinicians consider

Research and practice guidance point to a set of factors that bear on risk. None is decisive alone.


  • Prior offenses and blood alcohol level at arrest. A California Department of Motor Vehicles study found that blood alcohol concentration at arrest was statistically related, though not linearly, to the likelihood of reoffending within a year, and that a driver's prior two-year traffic convictions could raise predicted recidivism about as much as a large increase in BAC (Marowitz, 1996).

  • Severity of the substance use disorder. DSM-5 grades alcohol use disorder by the number of criteria met: 2 to 3 mild, 4 to 5 moderate, 6 or more severe. Level of care is commonly matched using the ASAM criteria.

  • Co-occurring psychiatric conditions. Untreated anxiety, depression, bipolar disorder, PTSD and ADHD complicate treatment, although in the available prospective data specific disorders did not reliably predict DUI re-offense, while ADHD predicted motor-vehicle re-offense.

  • Treatment engagement and sobriety. Completion of assessment and treatment, documented abstinence and relapse-prevention planning.

  • Monitoring. Ignition interlocks reduce re-arrest while installed (risk ratio about 0.25) but the benefit largely disappears after removal (about 0.93), and treatment combined with an interlock was associated with about one-third lower recidivism after removal, compared with interlock alone, in Florida data (Elder et al., 2011; Voas et al., 2016, as summarized by NHTSA).


Instruments and their limits

Screening tools such as the AUDIT identify risky drinking but do not diagnose a disorder. Instruments designed for DUI populations, including the Research Institute on Addictions Self-Inventory (RIASI), have evidence mainly for first-time offenders and little for repeat offenders. In a 1999 NHTSA-sponsored study of 609 DWI program participants in Pennsylvania and Massachusetts, Lacey, Jones and Wiliszowski compared several screening instruments with a criterion interview; the instruments correctly classified roughly 72% to 92% of problem drinkers, and greater sensitivity came at the cost of specificity. That study measured problem drinking, not recidivism. NHTSA's review reports, citing Chang and colleagues (2002), that no assessment instrument studied correctly identified more than 70% of the offenders likely to recidivate. In short, tests inform judgment but cannot replace a structured clinical interview, collateral records and laboratory or monitoring data.


How a clinician may reach a risk opinion

A defensible assessment is typically transparent about each of the following steps.


  • A structured clinical interview and review of the driving, legal and treatment history.

  • Diagnosis using DSM-5 criteria, with attention to co-occurring psychiatric conditions and medications.

  • Standardized screening results, interpreted as screening and not as proof.

  • Collateral information, such as court records, prior treatment records and, where appropriate, laboratory results.

  • A stated risk formulation (for example lower, moderate or higher) with the factors that support it and the factors that would change it.

  • Specific recommendations: level of treatment, medication management, monitoring and follow-up.


Caveats

Risk opinions are probabilistic, depend on information that may be incomplete, and can change with treatment. NHTSA's Countermeasures That Work (11th edition) rates assessment and treatment as a five-star countermeasure, but also acknowledges the limits of prediction. A clinician should not promise an outcome with the RMV or a court. For evaluation or treatment, see our RMV evaluation and psychiatric medication management services.


Care pathways, clinicians and insurance

Evidence-based care for drivers with OUI-related concerns usually combines several elements: a thorough assessment, treatment of any substance use disorder, treatment of co-occurring psychiatric conditions, medication review, and monitoring. NHTSA rates alcohol problem assessment and treatment as a five-star countermeasure, and combining treatment with an ignition interlock was associated with lower recidivism after the device was removed.


Services at Greater Boston Psychiatric Services


Common co-occurring concerns are addressed on our condition pages for anxiety, depression, ADHD, bipolar disorder, PTSD, OCD, sleep disturbance and mood and emotional regulation.


Clinicians

Evaluations and medication management are provided by licensed clinicians on our team, including Caryn Metzger, Thomas Smith, Susan Ouellette, Jill A. Forte, Jessica Crawford-Grey, Kristen Baker and Mia Copeland-Brock. Which clinician performs a particular type of evaluation is confirmed when you schedule.


Insurance and costs

Coverage for psychiatric evaluation and medication management varies by plan, and RMV or court-requested evaluations may not be covered. See our insurance overview and plan pages for Blue Cross Blue Shield, Aetna, Harvard Pilgrim Health Care and Optum, and our good faith estimate information for self-pay patients.


Limits of this article

This article is educational and is not medical or legal advice. Statistics describe groups and sources use differing definitions; the comorbidity findings above are ranges drawn from studies with differing samples and methods, and some figures come from secondary summaries as noted. If you are under an RMV or court deadline, confirm requirements with the RMV, the court or an attorney.


Frequently asked questions

What is an RMV psychiatric evaluation (Form MAB118)?

Form MAB118 is the RMV medical affairs form used for conditions that may affect driving. A Massachusetts-licensed physician or psychiatric nurse practitioner completes it and sends it to the RMV Driver Control Unit. It covers diagnosis, symptoms, medications and the clinician's opinion on fitness to drive.


What does "risk of recidivism" mean in an RMV or OUI evaluation?

Recidivism is a return to the prohibited behavior after a sanction or intervention. For impaired driving it is measured as a new arrest, conviction, alcohol-related crash or license action within a stated window. A clinician's "risk of recidivism" opinion is an estimate of that likelihood based on stated factors, not a prediction about any particular day.


Does a psychiatric diagnosis mean I cannot drive?

No. A systematic review of 24 studies found that no psychiatric disorder category was consistently linked to higher crash risk, and the authors did not support blanket driving restrictions. Fitness to drive turns on current symptoms, stability and treatment.


Can prescribed psychiatric medication lead to an OUI?

Massachusetts law prohibits operating under the influence of intoxicating liquor or of marijuana, narcotic drugs, depressants or stimulant substances. Having a medication detected is not the same as being impaired, but dose and timing matter, so review driving safety with your prescriber whenever a medication is started or changed.


Does insurance cover an RMV or court-requested evaluation?

Coverage varies by plan, and RMV or court-requested evaluations may not be covered. Check with your plan, and see our insurance and good faith estimate pages for self-pay information.


Sources


Comments

bottom of page