Research on Self-Help/Mutual Aid Groups
For the past few decades, researchers have been evaluating the effects of self-help/mutual aid groups on participants. Most research studies of self-help groups have found important benefits of participation. Unfortunately, few of these studies has gotten into the hands of self- help group members, clearinghouse staff and others who wish to advocate for self-help/mutual aid. The purpose of this chapter is to help correct this problem by summarizing the best research on the effectiveness of self-help groups in a brief and clear fashion.
As we read over research on the effects of mutual help groups, we noticed a common confusion. Many studies that claim to study self-help groups are actually studies of psychotherapy or support groups solely led by a professional who does not share the condition addressed by the group. We excluded such studies from this review. Instead, we focused on groups where the participants all shared some problem or condition and ran the group on their own. In a very few cases, we included studies where a group was co-led by a professional and by a self-helper. Professional involvement in an advisory or assistance capacity did not rule a study out of consideration, because in the real world, many member-run self-help groups use professional advisors.
We have been selective about the methodological strengths of the studies we chose to summarize. Many studies have demonstrated that if the current members of any self-help group are surveyed at any given time, the members will respond positively about the group and say that it helps them. Such studies (which are sometimes called "single-group cross-sectional surveys") have some value, but they do not tell us much about how members change over time, or whether members change more than non-members. For this reason, we focus here primarily on studies that compared self-help participants to non-participants, and/or gathered information on mutliple occasions over time (that is, "longitudinal" studies). Because we focus primarily on such studies, the following is only a subset of research on self-help effectiveness. At the same time, it is a methdological stronger subset of studies and thus should be more convincing to people outside of the self-help movement.
In the brief summaries below, we have tried to use as little jargon as possible. One excpetion to this rule is to use the scientific convention of using the letter "N" to refer to the number of people participating in each research project. For the sake of space and simplicity, we have generally omitted most details about how the study was conducted and about secondary findings. Readers who wish to have further details about any particular study can use the reference information provided to locate the original sources.
If you know of an article on self-help group effectiveness that is not on this page, or have comments, please contact Keith Humphreys.
Mental Health Groups
- Edmunson, E. D., J. R. Bedell, et al. (1982). Integrating Skill Building and Peer Support in Mental Health Treatment: The Early Intervention and Community Network Development Projects. Community Mental Health and Behavioral Ecology. A. M. Jeger and R. S. Slotnick. New York: Plenum Press: 127-139.
- After ten months of participation in a patient-led, professionally supervised social network enhancement group, one-half as many former psychiatric inpatients (N=40) required rehospitalization as did non-participants (N=40). Participants in the patient-led network also had much shorter average hospital stays (7 days vs. 25 days). Furthermore, a higher percentage of members than non-members could function with no contact with the mental health system (53% vs. 23%).
- Galanter, M. (1988). Zealous Self-Help Groups as Adjuncts to Psychiatric Treatment: A Study of Recovery, Inc. American Journal of Psychiatry 145(10): 1248-1253.
- This study surveyed 356 members of Recovery, Inc., a self-help group for nervous and former mental patients, and compared them to a 195 community residents of similar age and sex. Although about half of the Recovery Inc. members had been hospitalized before joining, only 8% of group leaders and 7% of recent members had been hospitalized since joining. Members used more outpatient non-psychiatric resources than did the community sample.
- Kennedy, M. (1990). Psychiatric Hospitalizations of GROWers. Paper presented at the Second Biennial Conference on Community Research and Action, East Lansing, Michigan.
- This study found that 31 members of GROW, a self-help organization for people with chronic psychiatric problems, spent significantly fewer days in a psychiatric hospital over a 32-month period than did 31 former psychiatric patients of similar age, race, sex, marital status, number of previous hospitalizations and other factors. Members also increased their sense of security and self-esteem, decreased their existential anxiety, broadened their sense of spirituality, and increased their ability to accept problems without blaming self or others for them.
- Kurtz, L. F. (1988). Mutual Aid for Affective Disorders: The Manic Depressive and Depressive Association. American Journal of Orthopsychiatry 58(1): 152-155.
- This study found that 82% of 129 members of the Manic Depressive and Depressive Association reported coping better with their illness since joining the self-help group. The longer they were members and the more intensely they were involved with the group, the more their coping had improved. Further, the percentage of members reporting being admitted to a psychiatric hospital before joining the group was 82%, but the percentage reporting hospital admission after joining was only 33%.
- Lieberman, M. A., Solow, N. et al. (1979). "The psychotherapeutic impact of women's consciousness-raising groups." Archives of General Psychiatry 36: 161-168.
- 32 participants in women=s consciousness-raising groups were studied over a 6 month period. Over the course of the study, participants reported decreased distress about their target problem, increased self-esteem, and greater self-reliance. They also reported greater identification with feminist values and politics.
- Raiff, N. R. (1984). "Some Health Related Outcomes of Self-Help Participation." Chapter 14 in The Self-Help Revolution, edited by Alan Gartner and Frank Riessman. New York: Human Sciences Press.
- Highly involved members of Recovery, Inc. (N=393, mostly female and married), a self-help group for former mental patients, reported no more anxiety about their health than did the general population. Members who had participated for two years or more had the lowest levels of worry and the highest levels of satisfaction with their health. Members also rated their life satisfaction levels as high or higher than did the general public. Members who had participated less than two years, were still on medication, lived below the poverty level, or lacked social-network involvements also appeared to benefit from group participation, although to a lesser degree.
Weight Loss Groups
- Grimsmo, A., G. Helgesen, et al. (1981). Short-Term and Long-Term Effects of Lay Groups on Weight Reduction. British Medical Journal 283: 1093-1095.
- These researchers conducted three studies of mostly female participants in 8-week peer-led weight-loss groups in Norway (Grete Roede Slim-Clubs). The first study gathered information from 33 women before, during, immediately after, and 1 year after participation. Participants lost an average of 14.3 pounds while they were in the group, and had kept almost all of it from coming back by the end of the year (they had an average of 12.1 pounds less weight). The second study surveyed 1000 people who had completed the group from 1 to 5 years previously, and found that average weight loss remained stable for the first couple of years and was still 5 - 6% below starting weight after 5 years. The third study surveyed more than 10,000 participants before and immediately after participation, and found an average weight loss of 15.2 pounds.
- Peterson, G., D. B. Abrams, et al. (1985). Professional Versus Self-Help Weight Loss at the Worksite: The Challenge of Making a Public Health Impact. Behavior Therapy 16: 213- 222.
- This study compared 30 employees assigned to a professionally-led weight-loss group with 33 employees assigned to a peer-led group. Both groups used "Learn to Be Lean" workbooks based on behavioral therapy principles. Members of both groups lost weight in equal amounts over a six-month period. The peer-led group was only half as costly as the professional-led group.
Addiction-Related Recovery Groups
- Alemi, F., Mosavel, M. Stephens, R. et al. (1996). "Electronic Self-Help and Support Groups." Medical Care 34(Supplement): OS32-OS44.
- This was a study of 53 pregnant women who had a history of drug use. Participants, most of whom were African-American, were assigned either to attend face-to-face biweekly self-help group meetings (N=25) or to participate in self-help meetings operated over a voice bulletin board accessed by phone (N=28). In the bulletin board group, participants could leave voice mail messages for the entire group to hear. Significantly more women participated in the voice mail group (96% of those assigned) than in the face-to-face self-help groups (32% of those assigned). Bulletin board participants made significantly fewer telephone calls and visits to health care clinics than did individuals assigned to participate in the face-to-face group. Both groups has similar health status and drug use at the end of the study.
- Christo, G. and S. Sutton (1994). Anxiety and Self-Esteem as a Function of Abstinence Time Among Recovering Addicts Attending Narcotics Anonymous. British Journal of Clinical Psychology 33: 198-200.
- Members of Narcotics Anonymous (NA) self-help groups (N=200) who stayed off drugs for three years or more while they were members showed no more anxiety and no less self-esteem than a comparison group of 60 never-addicted students. The longer people remained members while staying off drugs, the less anxiety and the more self-esteem they experienced.
- Emrick, C. D., J. S. Tonigan, et al. (1993). Alcoholics Anonymous: What is Currently Known? In Research on Alcoholics Anonymous: Opportunities and Alternatives, edited by Barbara S. McCrady and William R. Miller. New Brunswick, NJ: Rutgers Center of Alcohol Studies, pp. 41-75.
- Using meta-analysis of more than 50 studies, these authors report that AA members stayed sober more if they (1) had an AA sponsor, (2) worked the "twelfth step" of the program, (3) led a meeting, (4) increased their degree of participation over time, or (5) sponsored other AA members. The study also found that professionally treated alcoholic patients who attend AA during or after treatment are somewhat more likely to reduce drinking than are those who do not attend AA. Membership in AA was also found to reduce physical symptoms and to improve psychological adjustment.
- Hughes, J. M. (1977). Adolescent Children of Alcoholic Parents and the Relationship of Alateen to These Children. Journal of Consulting and Clinical Psychology 45(5): 946-947.
- This study compared 25 Alateen members with 25 non-members who had an alcoholic parent and 25 non-members with no alcoholic parent. Adolescents with an alcoholic parent who were members of Alateen experienced significantly fewer negative moods, significantly more positive moods and higher self-esteem than those who were not members. In fact, Alateen members had self-esteem and mood scores similar to those of adolescents who did not have an alcoholic parent.
- Humphreys, K., B. E. Mavis, and B. E. Stoffelmayr (1994). Are Twelve Step Programs Appropriate for Disenfranchised Groups? Evidence from a Study of Posttreatment Mutual Help Involvement. Prevention in Human Services 11(1): 165-179.
- One year after being admitted to a public substance abuse treatment agency, Caucasion- and African-Americans were attending mutual help (Narcotics Anonymous, Alcoholics Anonymous) groups at the same rate. African-American participants (N=253) in NA and AA self-help groups showed significant improvements over twelve months in six problem areas (employment, alcohol, drug, legal, psychological, and family). African-American self-help group participants had significant more improvement in their medical, alcohol, and drug problems than did African- American patients who did not participate in self-help groups after treatment.
- Humphreys, K. and R. H. Moos (1996) Reduced Substance-Abuse-Related Health Care Costs among Voluntary Participants in Alcoholics Anonymous. Psychiatric Services, 47, 709-713.
- Over a period of three years, alcoholics who initially chose to attend AA were compared to those who sought help from a professional outpatient treatment provider (total N=201). Those who chose to attend AA had 45% ($1826) lower average per-person treatment costs than did those who chose outpatient treatment. Despite the lower costs, AA attenders also experienced significant improvements in alcohol consumption, dependence symptoms, adverse consequences, days intoxicated and depression. These outcomes did not differ significantly from those of alcoholics who chose professional treatment. This was true both at one year and at three years after the beginning of the study.
- Jason, L. A., C. L. Gruder, et al. (1987). Work Site Group Meetings and the Effectiveness of a Televised Smoking Cessation Intervention. American Journal of Community Psychology 15: 57-77.
- This study compared the effects of two smoking cessation programs at work. One hundred and ninety-two workers viewed a television program and used a self-help manual, while 223 workers had these materials supplemented by 6 self-help group meetings. Group meetings were led by recruited smoking employees who had been given a three-hour training session in how to lead groups. The two programs were implemented at 43 companies. Initial rates of quitting smoking were significantly higher for the 21 companies that used self-help groups (average of 41% vs. 21% of participants). Group participants also smoked significantly fewer cigarettes per day, with lower tar, nicotine and carbon monoxide content. Three months later, an average of 22% of group participants had continued not to smoke, compared to 12% in companies with no self-help groups.
- McAuliffe, W. E. (1990). A Randomized Controlled Trial of Recovery Training and Self- Help for Opiod Addicts in New England and Hong Kong. Journal of Psychoactive Drugs 22(2): 197-209.
- This study randomly assigned volunteer graduates from substance abuse treatment programs (N=168) to participate in RTSH (Recovery Training and Self-Help), an aftercare program that combined professionally led recovery-training sessions with peer-led self-help sessions. Participants in the recovery program significantly reduced their' likelihood of relapse into opiod addiction compared to those who received only referrals to other programs and crisis- intervention counseling. The RTSH program helped unemployed participants find work and reduced criminal behavior.
- McKay, J. R., A. I. Alterman, et al. (1994). Treatment Goals, Continuity of Care, and Outcome in a Day Hospital Substance Abuse Rehabilitation Program. American Journal of Psychiatry 151(2): 254-259.
- Male substance abuse patients (N=180, 82% African American, mostly low income) who participated in self-help groups (Alcoholics Anonymous, Narcotics Anonymous) after treatment significantly reduced their frequency of alcohol and cocaine use by the 7-month followup. Participants with high self-help attendance rates used alcohol and/or cocaine less than half as much as did those with low self-help attendance. This was true regardless of previous substance use and whether or not they completed a 4-week hospital rehabilitation program. Hence, the effects of self-help groups were not simply due to motivation or other characteristics of the individuals who participated.
- Pisani, V. D., J. Fawcett, et al. (1993). The Relative Contributions of Medication Adherence and AA Meeting Attendance to Abstinent Outcome for Chronic Alcoholics. Journal of Studies on Alcohol 54: 115-119.
- A group of 122 mostly male, White alcoholic patients admitted to short-term hospital treatment programs participated in this study. In the 18 months following treatment, the more days the patient attended Alcoholics Anonymous self-help meetings, the longer their abstinence lasted. AA meeting attendance improved abstinence considerably more than did adherence to prescribed medication.
- Tattersall, M. L. and C. Hallstrom (1992). Self-Help and Benzodiazepine Withdrawal. Journal of Affective Disorders 24(3): 193-198.
- This study followed members (N=41) of TRANX (Tranquilizer Recovery and New Existence), a British self-help organization that provided telephone counseling and support groups to its members. Members were mostly White women who had been addicted to tranquilizers for an average of 12 years. During a 9-month period, members of the group were more likely to stop using tranquilizers than were individuals (N=76) who made an initial telephone contact but did not become a member. Most members (73%) also reported that the symptoms for which they had initially been prescribed tranquilizers improved, and 65% reported that they were at least moderately satisfied with their withdrawal in terms of its effects on their subjective quality of life.
- Walsh, D. C., R. W. Hingson, D. M. Merrigan, et al. (1991). A Randomized Trial of Treatment Options for Alcohol-Abusing Workers. The New England Journal of Medicine 325(11): 775-782.
- Workers assigned to participate in Alcoholics Anonymous self-help groups reduced their drinking problems over a two-year period. Furthermore, compulsory AA groups (n=83) did not significantly differ from compulsory inpatient treatment (N=73) in their effects on job-related outcomes of participants. Costs of inpatient treatment averaged 10 percent less for AA participants than for hospital rehabilitation participants.
- Caserta, M. S. and Lund, D. A. (1993). Intrapersonal Resources and the Effectiveness of Self-Help Groups for Bereaved Older Adults. Gerontologist 33(5): 619-629.
- Widows and widowers over age 50 who participated in bereavement self-help groups (N=197) experienced less depression and grief than nonparticipants (N=98) if their initial levels of interpersonal and coping skills were low. Those with initially high interpersonal skill levels also benefitted from participation if they participated in the groups for longer than eight weeks.
- Lieberman, M. A. and L. Videka-Sherman (1986). The Impact of Self-Help Groups on the Mental Health of Widows and Widowers. American Journal of Orthopsychiatry 56(3): 435-449.
- This study followed 36 widowers and 466 widows, 376 of whom were members of the bereavement self-help group THEOS. Over a period of one year, THEOS members who formed social relationships with other group members outside group time experienced less psychological distress (depression, anxiety, somatic symptoms) and improved more in psychological functioning (well-being, mastery, self-esteem) than did non-members and members who did not form such relationships.
- Marmar, C. R., M. J. Horowitz, et al. (1988). A Controlled Trial of Brief Psychotherapy and Mutual-Help Group Treatment of Conjugal Bereavement. American Journal of Psychiatry 145(2): 203-209.
- Bereaved women who sought treatment for grief after the death of their husband were randomly assigned to either professional psychotherapy (N=31) or self-help groups (N=30). Self-help groups worked just as well as the therapy. Participants and non-participants in the self-help groups reduced stress-specific and general psychiatric symptoms such as depression equally. They also experienced similar improvements in social adjustment and work functioning.
- Vachon, M. L. S., W. A. L. Lyall, et al. (1980). A Controlled Study of Self-Help Intervention for Widows. American Journal of Psychiatry 137(11): 1380-1384.
- Women (N=162) whose husbands had died within the past month were studied over a two-year period. Half of these women were assigned to participate in a "widow-to-widow" program. After 6 months in the program, participants were more likely than non-participants to feel more healthy and to feel "better," and less likely to anticipate a difficult adjustment to widowhood. After 12 months, participants were more likely than non-participants to feel "much better," to have made new friends, and to have begun new activities, and were less likely to feel constantly anxious or to feel the need to hide their true emotions. Participation facilitated adjustment both inside the person (in their relationship with themselves) and outside the person (in their relationships with others).
- Videka-Sherman, L. and M. Lieberman (1985). The Effects of Self-Help and Psychotherapy Intervention on Child Loss: The Limits of Recovery. American Journal of Orthopsychiatry 55(1): 70-82.
- This study compared White, mostly female bereaved parents who had received psychotherapy (N=120) to those who attended a Compassionate Friends (CF) bereavement self-help group sporadically (N=81), actively (N=25) or actively with social involvement with group members outside the group (N=97). Active participation in the self-help group accompanied by involvement with group members outside the group increased bereaved parents' comfort in discussing their bereavement with others and reduced parents' self-directed anger. Psychotherapy did not have these effects. CF members reported that group involvement had increased their self-confidence, sense of control, happiness, and freedom to express feelings, and decreased their depression, anxiety, guilt, anger, and isolation.
- Gilden, J. L., Hendryx, M. S., et al. (1992). Diabetes Support Groups Improve Health Care of Older Diabetic Patients. Journal of the American Geriatrics Society 40: 147-150.
- Male diabetic patients were randomly divided into three groups. The first group (N=8) received no intervention. The second group (N=13) received a six-session education program on diabetes self-care. The third group (N=11) received the education program plus 18 meetings of a patient- led self-help group. The patient-led group focused on coping skills, group discussions, structured social activities, and continuing diabetes education. At the end of the study, those who participated in both the education program and the patient-led group had better diabetes knowledge and quality of life and lower depression than non-participants. The participants in the peer-led group also reported less stress, greater family involvement, and better glycemic control than the patients who recieved no intervention.
- Simmons, D. (1992). Diabetes Self Help Facilitated by Local Diabetes Research: The Coventry Asian Diabetes Support Group. Diabetic Medicine 9: 866-869.
- Researchers assessed members of a self-help group for South Asian diabetics in England (N=53) for levels of glycated haemoglobin and knowledge about diabetes. Those who attended the group twice or more during a year had a significantly greater drop in glycated haemoglobin levels and a significantly greater increase in knowledge about diabetes. Although professionals helped start the group, it continues to operate independently, emphasizing education, mutual support, information sharing, and family social activities.
- Minde, K., N. Shosenberg, et al. (1980). Self-Help Groups in a Premature Nursery--a Controlled Evaluation. Behavioral Pediatrics 96(5): 933-940.
- Parents of premature infants were randomly assigned to participate in support groups in a hospital. The weekly groups (1.5 to 2 hours long) focused on coping and were co-led by a mother who had had a premature infant and by a nurse. Speakers were also brought in from outside periodically. Compared to 29 parents who did not participate, the 28 participants visited their infants in the hospital significantly more often, and touched, talked to, and gazed at their infants more often during visits. Participants also rated themselves more competent at infant care. Three months after their babies were discharged, group participants continued to show more involvement with their infants during feedings and were more concerned about their infants' general development.
- Toseland, R. W., Rossiter, C. M., and Labrecque, M. S. (1989). The Effectiveness of Two Kinds of Support Groups for Caregivers. Social Service Review, September: 415-432.
- This study divided 103 adult women caring for frail older relatives into three conditions: participation in a peer-led self-help group, participation in a professional-led support group, and no participation in either group. Groups met for eight weekly two-hour sessions. Both groups focused on enhancing coping skills. Compared to non-participants, women who participated in either type of group experienced significantly greater (1) increases in the size of their support network, (2) increases in their knowledge of community resources, (3) improvement in their interpersonal skills and ability to deal with the problems of caregiving, (4) improvement in their relationships with their care receivers, and (5) decreases in pressing psychological problems.
Groups for Elderly People
- Lieberman, M. A. and Bliwise, N. G. (1985). Comparisons Among Peer and Professionally Directed Groups for the Elderly: Implications for the Development of Self-Help Groups. International Journal of Group Psychotherapy 35(2): 155-175.
- This study compared participants (86 women and 22 men) in peer-led and professionally-led SAGE (Senior Actualization and Growth Explorations) self-help groups for the elderly to those who were on a waiting list to join the groups. Members of both types of SAGE groups felt they achieved their desired goals to a greater extent than those in the waiting-list group. Participation in either SAGE group also reduced psychological problems, such as nervousness and depression.
- Maisiak, R., M. Cain, et al. (1981). Evaluation of TOUCH: An Oncology Self-Help Group. Oncology Nursing Forum 8(3): 20-25.
- This study surveyed 139 members of TOUCH, a self-help group for cancer patients in Alabama. TOUCH focuses on teaching its members about cancer and training them to be peer counselors to help other patients. The longer members participated in a group, the more they improved their knowledge of cancer, their ability to talk with others, their friendships, their family life, their coping with the disease, and their following of doctors' orders. The percentage of people indicating their coping was very good after TOUCH was 59%, more than double the percentage indicating it was very good before TOUCH (28%).
- Spiegel, D., Bloom, J. R., Kraemer, H.C. and Gottheil, E. (1989). "Effect of psychosocial treatment on survival of patients with metastatic breast cancer." The Lancet October 14: 888-891.
- Participants in this study were 86 women undergoing treatment for metastatic breast cancer. A subset of these women (N=50) were randomly assigned to have their oncologic care supplemented with a weekly support group. The support groups were co-facilitated by a therapist who had breast cancer in remission and a psychiatrist or social worker. The sessions focused on living life fully, improving communication with family members and doctors, facing death, expressing emotions such as grief, and controlling pain through self-hypnosis. On average, support group participants lived twice as long as controls (an average of almost 18 months longer).
- Becu, M., Becu, N., Manzur, G. and Kochen, S. (1993). Self-Help Epilepsy Groups: An Evaluation of Effect on Depression and Schizophrenia. Epilepsia 34(5): 841-845.
- A group of Argentine researchers conducted a 4-month longitudinal study of 67 epileptic patients who participated in weekly self-help group meetings. Epileptic patients trained by psychologists led the groups. Group participants had decreased depression and other psychological problems over the course of the study.
- Hinrichsen, G. A., T. A. Revenson, et al. (1985). Does Self-Help Help? An Empirical Investigation of Scoliosis Peer Support Groups. Journal of Social Issues 41(1): 65-87.
- Adults with scoliosis who had undergone bracing or surgery and participated in a Scoliosis Association self-help group (N=33) were compared to adults with similar treatment who did not participate in the group (N=67). Compared to non- participants, group participants reported (1) a more positive outlook on life, (2) greater satisfaction with the medical care they received, (3) reduced psychosomatic symptoms, (4) increased sense of mastery, (5) increased self-esteem, and (6) reduced feelings of shame and estrangement.
- Nash, K. B. and K. D. Kramer (1993). Self-Help for Sickle Cell Disease in African American Communities. Journal of Applied Behavioral Science 29(2): 202-215.
- This study focused on 57 African Americans who had been members of self-help groups for sickle-cell anemia. The members who had been involved the longest reported the fewest psychological symptoms and the fewest psychosocial interferences from the disease, particularly in work and relationship areas.
- Sibthorpe, B., D. Fleming, et al. (1994). Self-Help Groups: A Key to HIV Risk Reduction for High-Risk Injection Drug Users? Journal of Acquired Immune Deficiency Syndromes 7(6): 592-598.
- Injection drug users (N=234) who had shared a dirty needle in the previous 30 days were followed over six months. Those who attended self-help groups (mostly Narcotics Anonymous and Alcoholics Anonymous) during that time were almost twice as likely to report reducing or eliminating their risk of exposure to HIV compared to those who did not attend such groups.
If you know of an article on self-help group effectiveness that is not on this page, or have comments, please contact Keith Humphreys.
This paper is in the public domain. Preparation of this paper was supported by National Institute of Alcohol Abuse and Alcoholism Grant #10652 and the Department of Veterans Affairs Mental Health Strategic Health Group.
The HTML copy of this paper was produced free-of-charge as a public service to the research community by Mental Help Net. Please inform MHN of other Web reproductions of this paper.