Shelter, Skills, and Cognition: Evaluating Integrated Strategies to Reduce Intimate Partner Violence in Rural Tanzania

Last registered on September 21, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Shelter, Skills, and Cognition: Evaluating Integrated Strategies to Reduce Intimate Partner Violence in Rural Tanzania
RCT ID
AEARCTR-0019573
Initial registration date
September 02, 2026

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
September 21, 2026, 6:35 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country
Tanzania, United Republic of
Region
Mara Region

Primary Investigator

Name
Alexis Wang
Affiliation
UC Santa Barbara
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Rachel Heath
PI Affiliation
University of Washington
Contact Investigator
PI Name
Zixu Chen
PI Affiliation
University of California, Berkeley
Contact Investigator
PI Name
Ziyue Chen
PI Affiliation
University of California, Berkeley
Contact Investigator

Additional Trial Information

Status
In development
Start date
2026-10-15
End date
2027-12-31
Keywords
Crime, Violence, & Conflict, Gender, Health, Labor, Welfare
Additional Keywords
intimate partner violence, domestic violence, violence against women, shelter, safe house, economic empowerment, vocational training, cognitive behavioral therapy, feasibility pilot, Tanzania
JEL code(s)
J12, I31, K42, O12, D13
Secondary IDs
CVI-25-04550 (J-PAL Crime and Violence Initiative); S7127 (MIT subaward)
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
There is almost no causal evidence on whether shelters reduce intimate partner violence.

This 16-month pilot randomized controlled trial tests the feasibility, safety and acceptability of integrated shelter-based support for women experiencing severe intimate partner violence in rural Mara Region, Tanzania, in preparation for a full-scale trial. Forty women are randomized 1:1:1:1 to four arms: delayed-shelter control; shelter for up to six months; shelter plus an economic empowerment program; and shelter plus a non-violent conflict management program for husbands. Women are interviewed at baseline and at 2, 7 and 12 months.

The pilot is not powered to detect treatment effects. It is judged against pre-specified feasibility thresholds and estimates outcome variance for the full trial.
External Link(s)

Registration Citation

Citation
Chen, Ziyue et al. 2026. "Shelter, Skills, and Cognition: Evaluating Integrated Strategies to Reduce Intimate Partner Violence in Rural Tanzania." AEA RCT Registry. September 21. https://doi.org/10.1257/rct.19573-1.0
Sponsors & Partners

Sponsors

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
All arms receive safety planning and referrals.

Arm 1 (control)

Arm 2: shelter for up to six months.

Arm 3: shelter plus an economic empowerment program (vocational and business skills, coaching).

Arm 4: shelter plus a non-violent conflict management program for the husband, using cognitive behavioral methods, offered only with the woman's permission and after a safety review.
Intervention Start Date
2026-10-15
Intervention End Date
2027-10-15

Primary Outcomes

Primary Outcomes (end points)
Women's outcomes, measured at baseline and at 2, 7 and 12 months:

1. Physical and sexual intimate partner violence.
2. Emotional abuse, controlling behaviour and economic abuse.
3. Depressive symptoms.
4. Perceived safety.
5. Earnings and hours worked.
6. Decision-making in the household, control over assets, and livelihoods self-efficacy.

Feasibility outcomes, measured over the pilot:

7. Women enrolled, and weeks taken to reach the target.
8. Shelter take-up among women assigned to a shelter arm.
9. Participation and completion in the economic empowerment program (Arm 3).
10. Share of participating husbands attending three or more NVCM sessions (Arm 4).
11. Retention at 2, 7 and 12 months.
12. Adverse events, and share of participants reporting intervention-related safety concerns.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Husbands of women in Arm 4, measured at baseline and after the program: attitudes toward conflict and violence, and self-reported conflict behaviors.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
A 16-month individually randomized pilot feasibility trial: four arms, 1:1:1:1, 40 women in one rural district of Mara Region, 10 per arm.

Eligible women are 18 or older, married, cohabiting or recently separated, not pregnant, have experienced intimate partner violence in the past year and need shelter for safety, and can be interviewed privately. Danger level is not a ground for exclusion.

The baseline survey comes before randomization. Women are interviewed privately by trained female enumerators, never at home. Husbands in Arm 4 are interviewed separately.
Experimental Design Details
Not available
Randomization Method
Randomization done by computer in office, off site: block randomization in a 1:1:1:1 ratio with randomly varying block sizes.
Randomization Unit
Individual woman.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
40 women. The design is not clustered.
Sample size: planned number of observations
40 women, interviewed up to four times each (baseline and 2, 7 and 12 months), so up to 160 woman-wave observations. Up to 10 husbands in Arm 4, interviewed up to twice each; actual enrollment may be lower.
Sample size (or number of clusters) by treatment arms
10 women delayed-shelter control, 10 women shelter only, 10 women shelter plus economic empowerment, 10 women shelter plus husband NVCM. 40 women in total, plus up to 10 husbands in the fourth arm.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
No minimum detectable effect size is set for the pilot: with 10 women per arm it is not powered to detect treatment effects. The precision it is designed for is on the feasibility rates. The pilot estimates outcome variance for the planned full trial.
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
University of Washington Human Subjects Division, Institutional Review Board Committee A
IRB Approval Date
2026-06-11
IRB Approval Number
STUDY00025705
IRB Name
Lake Zone Institutional Review Board, National Institute for Medical Research (NIMR), Mwanza Centre, Tanzania
IRB Approval Date
2026-08-31
IRB Approval Number
Certificate pending issuance; number to be added on receipt
Analysis Plan