CoChat
Sign Up
Cited by 15Homelessness and Social IssuesSubstance Abuse Treatment and OutcomesHIV, Drug Use, Sexual Risk

The effectiveness of abstinence-based and harm reduction-based interventions in reducing problematic substance use in adults who are experiencing homelessness in high-income countries: A systematic review and meta-analysis: A systematic review

Open access
O'Leary, Chris, Ralphs, Rob, Stevenson, Jennifer, Smith, Andrew, Harrison, Jordan, Kiss, Zsolt, Armitage, Harry·2024·Campbell Systematic Reviews: Better Evidence for a Better World
DOI 10.1002/cl2.1396PMC PMC11032639
Open this paper in CoChat — free accountOpen PDF

Abstract

Background: Homelessness is a traumatic experience, and can have a devastating effect on those experiencing it. People who are homeless often face significant barriers when accessing public services, and have often experienced adverse childhood events, extreme social disadvantage, physical, emotional and sexual abuse, neglect, low self-esteem, poor physical and mental health, and much lower life expectancy compared to the general population. Rates of problematic substance use are disproportionately high, with many using drugs and alcohol to deal with the stress of living on the street, to keep warm, or to block out memories of previous abuse or trauma. Substance dependency can also create barriers to successful transition to stable housing. Objectives: To understand the effectiveness of different substance use interventions for adults experiencing homelessness. Search Methods: The primary source of studies for was the 4th edition of the Homelessness Effectiveness Studies Evidence and Gaps Maps (EGM). Searches for the EGM were completed in September 2021. Other potential studies were identified through a call for grey evidence, hand-searching key journals, and unpacking relevant systematic reviews. Selection Criteria: Eligible studies were impact evaluations that involved some comparison group. We included studies that tested the effectiveness of substance use interventions, and measured substance use outcomes, for adults experiencing homelessness in high income countries. Data Collection and Analysis: Descriptive characteristics and statistical information in included studies were coded and checked by at least two members of the review team. Studies selected for the review were assessed for confidence in the findings. Standardised effect sizes were calculated and, if a study did not provide sufficient raw data for the calculation of an effect size, author(s) were contacted to obtain these data. We used random-effects meta-analysis and robust-variance estimation procedures to synthesise effect sizes. If a study included multiple effects, we carried out a critical assessment to determine (even if only theoretically) whether the effects are likely to be dependent. Where dependent effects were identified, we used robust variance estimation to determine whether we can account for these. Where effect sizes were converted from a binary to continuous measure (or vice versa), we undertook a sensitivity analysis by running an additional analysis with these studies omitted. We also assessed the sensitivity of results to inclusion of non-randomised studies and studies classified as low confidence in findings. All included an assessment of statistical heterogeneity. Finally, we undertook analysis to assess whether publication bias was likely to be a factor in our findings. For those studies that we were unable to include in meta-analysis, we have provided a narrative synthesis of the study and its findings. Main Results: = 25, or 52%). By far the most common reason for studies being rated as low confidence was high rates of attrition and/or differential attrition of study participants, that fell below the What Works Clearinghouse liberal attrition standard. Eleven of the included studies were rated as medium confidence and 12 studies as high confidence. The interventions included in our analysis were more effective in reducing substance use than treatment as usual, with an overall effect size of -0.11 SD (95% confidence interval [CI], -0.27, 0.05). There was substantial heterogeneity across studies, and the results were sensitive to the removal of low confidence studies (-0.21 SD, 95% CI [-0.59, 0.17] - 6 studies, 17 effect sizes), the removal of quasi-experimental studies (-0.14 SD, 95% CI [-0.30, 0.02] - 14 studies, 41 effect sizes) and the removal of studies where an effect size had been converted from a binary to a continuous outcome (-0.08 SD, 95% CI [-0.31, 0.15] - 10 studies, 31 effect sizes). This suggests that the findings are sensitive to the inclusion of lower quality studies, although unusually the average effect increases when we removed low confidence studies. The average effect for abstinence-based interventions compared to treatment-as-usual (TAU) service provision was -0.28 SD (95% CI, -0.65, 0.09) (6 studies, 15 effect sizes), and for harm reduction interventions compared to a TAU service provision is close to 0 at 0.03 SD (95% CI, -0.08, 0.14) (9 studies, 30 effect sizes). The confidence intervals for both estimates are wide and crossing zero. For both, the comparison groups are primarily abstinence-based, with the exception of two studies where the comparison group condition was unclear. We found that both Assertative Community Treatment and Intensive Case Management were no better than treatment as usual, with average effect on substance use of 0.03 SD, 95% CI [-0.07, 0.13] and -0.47 SD, 95% CI [-0.72, -0.21] 0.05 SD, 95% CI [-0.28, 0.39] respectively. These findings are consistent with wider resear

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1002/cl2.1396

Cites

235 of 235 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Psychosocial and pharmacological treatments can be effective in reducing substance use and improving physical and mental health.

    “as psychosocial and pharmacological treatments can be effective in reducing substance use and improving physical and mental health [5,6,7,8,9,10].”

    From Treatment Use Among U.S. Adults with a Substance Use Disorder: Associations with Symptom Severity, Problem Self-Perception, Comorbid Mental Illness, and Mental Health Treatment (2025) · cites this paper as [10]

    The Discussion (Summary of main results; page unavailable) reports that contingency management reduced substance use among adults experiencing homelessness, supporting the claim that some psychosocial treatments can be effective. The review says it focused only on substance-use outcomes, so it does not establish improvements in physical or mental health; its intervention typology lists medication-based treatments but does not establish their effectiveness. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 12

  1. Ballard, C. (2002). Counseling outcome research: The use of the Addiction Severity Index in a homeless population [Thesis. PhD Counselling].
  2. Bernad, R., & Yunca, R. (2016). Introducing the Housing First model in Spain: First results of the habitat programme. European Journal of Homelessness, 10(1), 53– 82.
  3. Chung, T. E., Gozdzik, A., Lazgare, L. I. P., To, M. J., Aubry, T., Frankish, J., Hwang, S. W., & Stergiopoulos, V. (2018). Housing First for older homeless adults with mental illness: A subgroup analysis of the At Home/Chez Soi randomized controlled trial. International Journal of Geriatric Psychiatry, 33(1), 85– 95. 28206715
  4. Consoli, T., Cortese, C., Molinari, P., & Zenarolla, A. (2016). The Italian network for implementing the ‘housing first’ approach. European Journal of Homelessness, 10(1), 83– 98.
  5. Kirst, M., Zerger, S., Misir, V., Hwang, S., & Stergiopoulos, V. (2015). The impact of a Housing First randomized controlled trial on substance use problems among homeless individuals with mental illness. Drug and Alcohol Dependence, 146, 24– 29. 25465295
  6. McGlave, C., Kruse, G. R., Yaqubi, A., Chang, Y. C., Rigotti, N. A., & Baggett, T. P. (2017). SmokefreeTXT for homeless smokers: A pilot RCT with a mixed‐methods analysis. Journal of General Internal Medicine, 32, S316– S317.
  7. Morse, G. A., Calsyn, R. J., Klinkenberg, W. D., Trusty, M. L., Gerber, F., Smith, R., Tempelhoff, B., & Ahmad, L. (1997). An experimental comparison of three types of case management for homeless mentally ill persons. Psychiatric Services, 48(4), 497– 503. 9090733
  8. Bretherton, J. (2017). Reconsidering gender in homelessness. European Journal of Homelessness, 11(1), 1– 21.
  9. Klee, H., & Reid, P. (1998b). Drug use among the young homeless: Coping through self‐medication. health – An interdisciplinary journal for the social study of health. Illness and Medicine, 2(2), 115– 134.
  10. National Alliance to End Homelessness. (2022). State of Homelessness 2022 edition. Report.
  11. National Health Care for the Homeless Council. (2017). Addressing the opioid epidemic how the opioid crisis affects homeless populations (pp. 1– 5).
  12. Singh, S., & White, H. (2022). Studies of the effectiveness of interventions to improve the welfare of those experiencing or at risk of homelessness: An evidence and gap map (4th edition). Report.

Cited by

Sorted by citations

Related

Authors
O'Leary, C., Ralphs, R., Stevenson, J., Smith, A., Harrison, J., Kiss, Z., & Armitage, H.
Affiliations
  • Manchester Metropolitan University, Manchester, UK
  • Independent Consultant, Cambridge, UK
  • Manchester Metropolitan University (GB)
  • ZK Analytics, Oxford, UK
ORCID
Smith, Andrew
Venue
Campbell Systematic Reviews: Better Evidence for a Better World
Publisher
SAGE Publications
Published
2024-04-21
Volume
20
Issue
2
Pages
e1396-e1396
ISSN
1891-1803 (ISSN-L)
DOI
10.1002/cl2.1396
PMC
PMC11032639
OpenAlex
W4394983549
Cited by
15
References
232
Citations by year
2026: 6 · 2025: 7 · 2024: 2
FWCI
1.35
Type
journal article
Language
en
Open access
Licence
Locations
  • Campbell Systematic Reviews (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • e-space (Manchester Metropolitan University) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Centre for Homelessness Impact
Topics
  • Health Sciences › Health Professions › General Health Professions › Homelessness and Social Issues
  • Health Sciences › Medicine › Epidemiology › Substance Abuse Treatment and Outcomes
  • Health Sciences › Medicine › Epidemiology › HIV, Drug Use, Sexual Risk
Keywords
homelessness, problematic substance use, harm reduction, substance use interventions, alcohol consumption, drug use, high-income countries, impact evaluation, publication bias, heterogeneous effects, robust variance estimation, non-randomized studies

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review