Reinforcing Helpful Changes

If a person finds it hard to sustain change, can a program notice and support a specific step, such as attending treatment or a period without a particular substance? Researchers call this approach contingency management (CM). Its results deserve attention, but they do not turn care into buying obedience or give relatives a right to monitor someone. To understand its value and limits, ask what was compared, for how long, and what happened when the program ended.

If an overdose is suspected, life is in immediate danger, someone is thinking of suicide, violence is occurring, or harm seems imminent, seek local emergency medical help and a safe place. Do not wait for a reward, a routine appointment, or the questions at the end of this article. If serious harm is already known, a person can seek help earlier even without an emergency or a settled diagnosis.

1. What the programs study

In a professional CM program, a clearly specified action is linked to a clearly explained material incentive. Clinical studies have targeted treatment attendance, adherence to prescribed treatment, or objectively verified absence of a particular substance. Researchers compare that link with another condition for a limited period. This allows them to study the added effect of the reinforcement structure. It does not sort people into those who “have motivation” and those who do not.

The name describes an approach, not a ready-made home recipe. Participants, goals, accompanying treatment, verification methods, and funding differ. A negative sample during a program reports one measured outcome in one period; it does not describe a person's whole health, relationships, or future. Attendance is not the same as recovery, either. Useful change can require medical, psychological, and social support together.

2. A comparison in a community outpatient service

The randomized trial by McDonell and colleagues involved 176 outpatients at a Seattle service who had stimulant dependence and serious mental illness. One group received CM alongside usual treatment. The other received the same usual treatment and incentives for study participation that were not conditional on abstinence. Both groups retained clinical and social care. This is more informative than simply comparing “money with nothing”: the trial examined the role of a specified condition.

During the three-month program, the CM group was about 2.4 times more likely to submit a stimulant-negative urine sample than the comparison group. This was a sample outcome under trial conditions, not a claim that every participant was “cured.” At the same time, 42% of CM participants completed the program, compared with 65% in the comparison group. A gain on one target cannot automatically be called success at keeping everyone engaged. For a person facing several serious difficulties, the demands of participation matter too.

In the following three months, CM participants reported fewer days of stimulant use. Analyses of sustained objectively verified abstinence did not give a uniform result, however, and many follow-up observations were unavailable. The authors also noted the limits of one treatment site. This trial supports a short-term comparative benefit for a particular group. It neither promises a lasting outcome for every person nor supplies a household plan for relatives.

3. What a larger set of trials shows

Bolívar and colleagues brought together 74 experimental reports involving 10,444 adults receiving medication for opioid use disorder; 60 reports could be included in numerical meta-analyses. Targets varied, including stimulant and other substance use, treatment attendance, and adherence to prescribed medication. Their results chiefly concern CM added to care in this clinical setting. They do not establish that every person facing any kind of problem will have the same response.

At the end of treatment, the comparative effect favored CM for stimulant abstinence: standardized Cohen's d = 0.70, with a 95% confidence interval of 0.49–0.92. The estimate for treatment attendance was smaller: d = 0.43, interval 0.22–0.65. These figures compare groups of studies; they are not an individual's chance of recovery or an amount to offer as an incentive. The individual trial above also shows that attendance or completion can move differently from substance-use outcomes.

4. What is known after incentives end

In Bolívar's review, a post-CM effect could be calculated for only seven of the 74 reports. The pooled estimate, d = 0.02 with an interval from −0.16 to 0.21, did not show a clear difference between groups. The authors identify sustained benefit as a research gap. This does not mean that every helpful change disappears when a program ends: the available data cannot support that sweeping claim.

Another meta-analysis by Ginley and colleagues specifically selected 23 randomized trials with objective indicators of substance use measured up to a year after incentives ended. In its official abstract, the odds ratio for abstinence was 1.22 in favor of CM, with a 95% interval of 1.01–1.44. This is a modest comparative advantage among the selected trials. An odds ratio does not mean that every person's chance of recovery rose by 22 percentage points. The full paper was unavailable for this preparation, so we do not draw more detailed conclusions about study quality or applicability.

The two reviews asked different questions and selected follow-up data differently. Their findings should not be treated as a simple vote “for” or “against” the method. A useful, honest conclusion is that effects during a program are better studied, while the picture after it ends depends on the available trials and outcomes. Alongside a short-term target, it makes sense to discuss what support and real sources of stability will remain when incentives stop.

5. Reinforcement does not replace other care

In McDonell's trial, usual treatment and social support continued in both groups. In Bolívar's review, participants were already receiving medication for opioid use disorder. These data therefore concern an addition to care, not a choice between a reward and treatment. The CDC explains that substance use disorders are treatable and that no single treatment works for everyone. A person should not stop prescribed medication on their own to take part in an incentive program; medication changes belong in a discussion with the prescribing clinician.

When choosing a goal, more than a test result or attendance can matter: safety, sleep, work, relationships, housing, and what the person finds meaningful. One mark cannot contain all these parts of life. It is also wrong to assume that any support from relatives “weakens motivation.” Warmth, conversation, and help reaching care are not the intervention compared in CM trials.

6. Questions for a voluntary program

If a local service offers CM, an adult should be able to understand what is checked, how and by whom the outcome is verified, which incentives are possible, what happens after a missed visit, who can access the information, and whether ordinary care continues when things are difficult. Terms should be clear in advance, participation based on consent, and sensitive results kept private. These are questions for a respectful choice, not a promise that every particular service is well run.

It is also reasonable to ask who funds the program, whether places are available, whether it fits the person's age and health needs, and what comes next. Access and fairness matter: the approach must not be used to withhold basic treatment, food, or housing because of a test result. Numerical funding rules in particular US programs cannot be turned into worldwide law or a household payout schedule.

7. What relatives can do without becoming monitors

Relatives can notice helpful steps already taken, offer to accompany a person to a professional with that person's consent, ask what actually helps, and protect their own money, time, and safety. This is ordinary support within a relationship. It does not require buying test kits, arranging covert checks, keeping a family rewards-and-penalties chart, or monitoring an adult's phone or prescriptions. CM trials give a family no right to make basic necessities conditional on results.

If an adult declines discussion or treatment, relatives can still seek support for themselves and set their own boundaries without violence. Children must not become observers, intermediaries, or monitors of adults' substance use. When threats or violence occur, protecting the affected person and contacting local safety services is necessary independently of any incentive plan.

8. Finding care that is available locally

A person can start with a local doctor, mental health service, or addiction professional and ask whether professionally run CM is available for their situation at all. Then the practical conditions matter: whether new people are accepted, age requirements, language, cost and coverage, wait time, data privacy, accompanying treatment, and a plan for after the program. If CM is unavailable or unsuitable, other forms of help may still be available. An appointment is not the same as care received; if the situation worsens, seek a faster local route.

A person can discuss goals and consent with a qualified professional without promising “perfect motivation.” Love, faith, or a sense of meaning may support someone personally, but none is an entry requirement for treatment. A person's ability to ask for help does not depend on whether they met an incentive condition yesterday.

9. What can be carried into everyday decisions

Professionally organized reinforcement can help achieve some measured goals in particular clinical settings. In one trial, better sample outcomes coexisted with lower program completion. The reviews agree that short-term effects are more clearly studied; the question of how changes last after incentives end is more complex. These facts can guide questions to a service and planning for continued support, but they offer no universal scheme of money, tests, or penalties.

If a program is offered, consider it alongside other care and the person's actual living conditions. If it is not offered, dignified help can still be possible. No one has to become “ideal” or lose support before their efforts to change deserve attention.

Five optional questions about what is already known

If it feels safe and useful, these questions may help prepare a conversation with a professional. They are not a self-diagnosis and require no new substance use, home test, or change to prescribed medication. A person can seek help without answering them.

1. What change would the person want to support, and which already-known circumstances make it matter?
2. What does the local program actually offer alongside usual treatment, and for how long?
3. Are its terms, consent, result privacy, and response to missed visits or difficulties clear?
4. What support will remain after the program, and which boundaries do relatives want to keep?
5. What is already known about local access by age, language, cost, waiting time, and urgency?

Sources and limits

1. McDonell M. G. et al. A randomized controlled trial of contingency management for psycho-stimulant use in community mental health outpatients with co-occurring serious mental illness. American Journal of Psychiatry, January 2013. Single-site trial of 176 people: CM plus usual care versus usual care with noncontingent incentives; three months of treatment and three months of follow-up, with missing observations limiting later conclusions.
2. Bolívar H. A. et al. Contingency Management for Patients Receiving Medication for Opioid Use Disorder: A Systematic Review and Meta-analysis. JAMA Psychiatry, August 4, 2021. Review of 74 experimental reports involving 10,444 adults receiving medication for opioid use disorder; 60 in numerical analyses and only seven with calculable post-treatment effects.
3. Ginley M. K. et al. Long-term efficacy of contingency management treatment based on objective indicators of abstinence from illicit substance use up to 1 year following treatment: A meta-analysis. Journal of Consulting and Clinical Psychology, January 2021. Official abstract for a meta-analysis of 23 randomized trials with post-program outcomes; the full paper was not read in this preparation.
4. Centers for Disease Control and Prevention. Treatment of Substance Use Disorders. US public-health guidance dated April 25, 2024, about SUD treatment; neither a CM effectiveness trial nor a directory of local services outside the US.

Topics

contingency management; helpful change; addiction treatment; stimulants; opioid use disorder; clinical trials; treatment retention; duration of effects; consent; privacy; family support; access to care.

Authors

Andrey Zarinsh — author and founder of United Wisdom. He explores inner transformation, the unity of traditions, intercultural dialogue, and contemporary spiritual discipline. Instagram: @andrey.zarinsh.

Maria Evilkut — co-author of United Wisdom. Alongside her research work, she seeks to test these ideas in practice and observe how principles of unity and mutual understanding work in real life between people and cultures. Instagram: @ilkutki.

Content checked: October 14, 2026. Language edition: EN.