Evaluating the Impact of Emergency Maternity Housing on Mother and Child Well-being, Housing, and Employment

The control group will be referred to other services in the home’s locality – this could include other maternity homes, other general homeless centers, agencies that could help them access Medicaid or WIC, etc. Some of the homes might also have the ability to provide diapers or women’s clothing to women in the control group.

The treatment group moms will be given the option of moving into the home and provided with a bed, prenatal care, food, etc. They typically have a caseworker or advocate with whom they will set goals, such as finding employment or childcare. The caseworker will guide her in achieving these goals. The moms also have access to group or individual counseling, parenting classes, support for advancing their education, some basic transportation services, etc. Moms may remain in the home typically up to one year post-birth of the child. Some mothers are unable to live under the house rules and are asked to leave. Others leave voluntarily before the birth of the child for a variety of reasons. We discuss how these departures impact the estimates of treatment effects below.

2022-05-30

2025-12-01

Pregnancy Outcomes, Life Satisfaction, Life Stability

Pregnancy Outcomes:
1. Primary measure: Baby carried to term
- Follow up survey question asking baby’s gestational age at birth
- Hypothesis: The intervention may increase the client's confidence in her ability to carry the child to term and give her access to health services that may improve pregnancy health outcomes. We will test the null hypothesis that there is no effect of the intervention on carrying the baby to term.
2. Alternative measure: Mom retains custody of baby
- Follow-up survey will ask whether the mother retained custody of the baby immediately after birth and whether she currently has custody of the baby.
- Hypothesis: Many of the mothers have previous interactions with state Child protective Services (CPS) and have lost custody of previous children. It is also the case that the underlying reason for their homelessness during pregnancy (e.g., drug use) may increase the chance they will lose custody of their baby to CPS once born. We will test that there is no effect of the intervention on retaining custody of the baby.


Life Satisfaction:
Primary measure: : Overall life satisfaction
- Baseline and follow-up survey asking where the client feels that his/her life is on a scale of one to ten
- Hypothesis: The intervention is designed to increase the client’s overall life satisfaction. We will test the null hypothesis that there is no effect of the intervention on life satisfaction.
Alternative measure: Sense of self-efficacy
- Baseline and follow-up survey asking about mother's sense of self-efficacy
- Hypothesis: The intervention is designed to increase the client’s sense of self-efficacy. We will test the null hypothesis that there is no effect of the intervention on self-efficacy ratings.
Alternative measure: Isolation + Loneliness scale + Depression
- Baseline and follow-up survey questions asking whether the client feels lonely, feels like they have people who support them, has experienced depression, etc.
- Hypothesis: Applicants to maternity homes indicate a high degree of loneliness and isolation and the intervention is designed to reduce this by creating a sense of community and family. We will test the null hypothesis that there is no effect of the intervention on loneliness.

Life Stability
Primary Measure: Address History
- Constructed using follow-up survey questions asking her independent living status and whether she has a mortgage or lease in her own name. Additionally, we will explore using address history information from Infutor Data Solutions, comparing information there with the housing statuses self-reported in the follow-up survey.
- Hypothesis: The intervention is designed to improve the client’s housing stability. We will test the null hypothesis that there is no effect of the intervention on the number of address changes.
Alternative Measure: Continuum of Care Utilization
- HMIS captures a mother’s interactions with other local services agencies. LEO is in the process of establishing data sharing relationships with HMIS agencies in each locality. This will capture her movements outside of having permanent addresses.
- Hypothesis: The intervention is designed to improve the client’s housing stability. We will test the null hypothesis that there is no effect of the intervention on the number of interactions with the continuum of care services.
Alternative Measure: Employment
- Mothers will be asked to report their employment in the baseline and follow-up surveys, including whether they are currently employed, how many jobs they have, how many hours per week they work, their income and the industry of their primary job.
- Hypothesis: The intervention is designed to improve the client’s employment. We will test the null hypothesis that there is no effect of the intervention on employment.
Alternative Measure: Safety Net utilization
- Mothers will be asked to report all sources of income, including welfare, in the follow-up survey.
- Hypothesis: The intervention is designed to improve the client’s access to social safety nets while simultaneously improving their employment. It is unclear whether the treatment will increase their welfare utilization as access is increased, or decrease their welfare as self-sufficiency and employment increase. We will test the null hypothesis that there is no effect of the intervention on welfare utilization.
Alternative Measure: Education
- Mothers will be asked to report their education in the baseline and follow-up surveys, including their highest level of education, vocational training, active school status, etc.
- Hypothesis: The intervention is designed to improve the client’s education levels. We will test the null hypothesis that there is no effect of the intervention on education.

Moms will call and agencies will determine if she is eligible based on basic criteria. If she is eligible, the agency will send her an intake application link to complete, which includes a consent form. After completing the survey, if she is still eligible, she will be placed on a waitlist for services. When a bed becomes available in an individual agency, the agency will call two moms at the top of the waitlist, conduct in-person interviews to confirm eligibility and good fit. They will use a computer to randomly assign one to treatment and one to control. Agency will let moms know of their status and will allow the treatment mom to move in. Moms in treatment and control will be surveyed approximately one year after the estimated due date of the child (as reported on the baseline survey).

Not available

On computer

Individual

No

0

1048

534 treatment, 534 control

Assuming a baseline housing stability rate of 50.3%, initial power calculations suggest that we will be able to measure a 16.3 pp effect on housing stability after one year of enrollment, 11.53 pp effect after two years of enrollment, and a 9.41 pp effect after three years. We anticipate that we will need to enroll for at least two years. This is assuming approximately 60% of mothers assigned to treatment actually take up the program.