Family and Couple Work: When It Helps and When Safety Must Come First

When substance use affects a relationship, joint help can seem like the natural next step: bring everyone together and agree on how to live from here. For some adults, this can be useful. Yet couple therapy, support for one family member, and an ordinary conversation at home address different questions. Having someone close does not imply their consent to joint treatment, and wanting to preserve a relationship does not oblige anyone to tolerate threats or coercion.

If a person or children face immediate danger, threats of violence, or an acute health crisis, seek local urgent help and a way to speak with a professional separately before arranging a joint meeting or considering the optional questions at the end. In an abusive situation, disclosing personal information in a shared session can increase danger. A person can seek protection and support before anyone agrees to therapy or receives a diagnosis.

I. Three formats that are easily confused

In structured couple or family work, a professional helps adults discuss treatment goals, the effects of substance use, communication, and supportive actions. This is more than a relative attending an appointment: it requires voluntary participation, professional training, confidentiality arrangements, and a safety assessment. UK clinical guidance on alcohol treatment distinguishes work involving both people, help for a close person that may support entry into treatment, and help addressing the relative's own needs. The distinction is useful even though services and rules outside the UK may differ.

CRAFT is an example of work with a concerned close person when the other person is not yet participating. A common research outcome is whether the other person starts treatment. It is neither couple therapy nor a promise that a relative can “make someone sober.” Separate support may address the relative's stress, boundaries, and safety even without a goal of changing another adult's decision. An ordinary family conversation may help people exchange requests when it is safe, but results from clinical programs cannot automatically be applied to it.

II. What couple therapy research does and does not show

In a meta-analysis by Song and colleagues, published in 2023, structured behavioural couple therapy for alcohol or other drug use disorder did better on average than comparison conditions on frequency of use, related consequences, and relationship satisfaction after treatment and at 12-month follow-up. This is an average across different trials and comparisons, not a particular couple's chance of success. It does not show that any joint conversation works the same way or that couple work is safe when abuse is present.

An earlier 2008 meta-analysis by Powers and colleagues pooled 12 randomized trials involving 754 participants. The authors found a clearer advantage in relationship satisfaction immediately after therapy and in several outcomes at later follow-up. Their conclusion primarily concerned married or cohabiting people already seeking help for alcohol or drug dependence; evidence for less severe problem drinking was limited. The findings should not be extended to every person who uses substances, to someone without a willing partner, or to a couple in which one person is afraid to speak honestly.

The potential value of this format does not make relationships a medical test of “willpower.” Prescribed medicines and individual medical and psychological care have their own roles. A family should not require someone to change prescribed treatment or infer a diagnosis solely because a partner is tired or worried. In its communication about benzodiazepines, the FDA distinguishes therapeutic benefit from risks related to physical adaptation and warns that abrupt withdrawal from this drug class can be dangerous. Physical adaptation to a prescribed medicine is not, by itself, a substance use disorder; changes to a medicine belong in a discussion with the prescribing clinician, not a family vote.

III. What programs for one close person have found

A 2017 randomized trial by Kirby and colleagues enrolled 115 concerned close people whose relatives or partners were refusing treatment. It compared full CRAFT, a narrower program focused on talking about treatment entry, and another support approach, with follow-up through nine months. The close people reported treatment entry for 62% and 63% in the first two groups, compared with 37% in the comparison group. Full CRAFT and the narrower program did not differ significantly on that outcome, and the groups did not differ in the patient's substance use as reported by the close people. These figures describe study groups, not an individual relative's chance of “persuading” someone.

A 2020 systematic review by Archer and colleagues included 691 concerned close people across 14 studies with different designs. Entry into treatment varied substantially by program, delivery format, and the problem studied. Participants were predominantly female spouses and parents, so the review does not establish the same result for every family. Starting treatment is also different from completing it, sustaining recovery, or being safe at home. CRAFT is professional support for a close person, not a set of pressure tactics, phone surveillance, or financial punishments to apply independently.

IV. When joint work may be suitable

A starting question is what each adult wants from help and whether each can freely decline a joint format. If both agree, a professional can clarify goals, interests, possible harms, and rules for discussion, including in separate conversations. Boundaries around personal information matter from the outset: what the person in treatment consents to share, what each may discuss without the other, and how confidentiality works in the particular service. Closeness or marriage does not automatically grant access to a diagnosis, prescription, or private appointment.

Joint work may help when it is safe enough to speak, a trained professional is accessible, and the participants can discuss feasible changes without appointing anyone a “supervisor.” UK psychosocial treatment guidance describes behavioural couple therapy as a structured approach for a partner who is willing to participate and stresses practitioner training. In the alcohol-care context it addresses, the guidance does not recommend this format with current domestic abuse or when the partner also drinks problematically. This is the boundary of a specific clinical recommendation, not a reason for readers to diagnose themselves or label every family difficulty a contraindication.

V. With threats or coercion, address safety separately first

If someone fears what may happen after speaking, or a partner controls contacts, money, or movement or threatens the person or children, a shared session does not automatically become a “safe space.” UK guidance for alcohol services on domestic abuse puts risk assessment, private contact, and specialist support first. In that UK service setting, a person experiencing abuse and the person causing harm should not be seen together; confidentiality matters because disclosed information can increase danger. A change in substance use does not prove that violence has ended.

The second-edition 2026 Home Office standards for England and Wales say joint work between the affected person and the person causing harm is rarely appropriate in domestic abuse. In the specialist perpetrator-intervention setting, it should be considered only after the person causing harm has completed behaviour-change work and both people have had separate suitability assessments; any consideration must be led by the affected person's safety and wishes. These are standards for specialist services in England and Wales, not a legal rule for every country. The practical point is to seek local specialist support separately when threatened. There is no need to stage a family confrontation, disclose details in front of a partner, or attend couple therapy to prove love. Contact local emergency services for an immediate threat. Adults and professionals must address children's safety; children are not go-betweens, substance-use monitors, or keepers of medication.

VI. A relative can get help without the other person's participation

Someone close may face exhaustion, anxiety, and financial effects even if the person with problematic use does not want treatment. They need a way to speak with a professional, consider which costs and responsibilities they can carry, protect their own resources, and receive support. Caring alone is not a diagnosis of “codependency,” nor an obligation to remain in a relationship. UK guidance explicitly addresses support for a relative who is not part of the patient's treatment while maintaining each adult's information boundaries.

A relative can seek general advice about available support and their own wellbeing without receiving or sharing another person's confidential medical details without consent. It helps to ask a particular service who can see its records, how separate conversations are protected, and whether local law allows exceptions to confidentiality. Those rules cannot be promised to be the same in every country. Support for one close person should not turn them into a round-the-clock monitor or make a child track an adult's condition or persuade them into treatment.

VII. One realistic next step

If there is no immediate danger, one feasible step is to identify a local service where someone can describe what is already known and ask what form of help it offers. Does it see the person who uses substances, a close person alone, or both? Is a separate meeting available to discuss safety? Who conducts couple work, and how do they assess whether it is suitable? Are the relevant age group, language, cost, waiting time, and private contact workable? If that service is inaccessible, ask for another route. A booking is not the same as care actually received.

Help remains possible if family is absent, unwilling, or unsafe. Support from friends, professionals, and local community does not have to pass through family; meaning, love, and perfect reconciliation are not entrance requirements for treatment. Worsening health, threats of violence, or another immediate danger call for local urgent help before a routine appointment or the questions below. The aim is an accessible contact and a response to real safety needs, not a promise that one conversation will repair an entire life.

Appendix: five optional points for a conversation

These questions concern only circumstances already known. They are not a home screening scale, a task to test substance use, or an invitation to arrange a joint conversation under threat. An immediate danger calls for local urgent help and, when abuse is involved, specialist support before these questions or a routine booking.

Format. Which adults already want to participate themselves, and who needs a separate conversation now?
Safety. What do I already know about the ability to speak freely and privately; is separate safety support needed first?
Purpose. Which real need do I already see: support for myself, access to treatment, or joint work with both adults' consent?
Access. Which local service actually fits the age, language, cost, wait, and confidentiality conditions?
Boundaries. Which responsibilities and resources is each adult already willing to discuss voluntarily without making a child the intermediary?

Not having answers is no reason to delay help. Any adult can start with a separate contact.

Sources and limits of application

Song Y et al. The Effect of Behavior Couples Therapy on Alcohol and Drug Use Disorder: a Systematic Review and Meta-Analysis. Meta-analysis, journal publication 2023; group results after couple therapy for alcohol or drug use disorder, not an individual guarantee or an assessment of safety amid abuse.
Powers MB, Vedel E, Emmelkamp PMG. Behavioral couples therapy (BCT) for alcohol and drug use disorders: a meta-analysis. Meta-analysis, 2008; mainly help-seeking married or cohabiting couples, with limited applicability to less severe problem drinking.
Kirby KC et al. Analyzing components of Community Reinforcement and Family Training (CRAFT): Is treatment entry training sufficient?. Randomized study, 2017; compares support for a close person on patient treatment entry, not proof of stopped use or safety for every family.
Archer M et al. Community reinforcement and family training and rates of treatment entry: a systematic review. Systematic review, 2020; varied programs and close-person samples, with treatment entry rather than lasting recovery as the principal outcome.
UK Department of Health and Social Care. Clinical guidelines for alcohol treatment — 5. Psychosocial interventions. UK professional guidance, published 2025 and updated 2026; alcohol-care formats, consent, and separate help for close people, not worldwide service availability.
UK Department of Health and Social Care. Clinical guidelines for alcohol treatment — 22. People experiencing or perpetrating domestic abuse. UK professional guidance, published 2025 and updated 2026; safety, privacy, and specialist support for abuse in alcohol services, not a universal legal process.
UK Home Office. Standards for stalking and domestic abuse perpetrator interventions (accessible). England and Wales specialist perpetrator-intervention standard, second edition April 2026, page updated 7 August 2026; limits on joint work amid abuse, not a substance-use-treatment trial.
U.S. Food and Drug Administration. FDA requiring Boxed Warning updated to improve safe use of benzodiazepine drug class. US regulatory communication, 2020; therapeutic benefit and abrupt withdrawal risk for benzodiazepines specifically, not every medicine.

About the 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, 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 review date: 14 October 2026. Language version: EN.