Anisfeld, E., Sandy, J., & Guterman, N. B. (2004). Best Beginnings: A randomized controlled trial of a paraprofessional home visiting program: Technical report. Report to the Smith Richardson Foundation and New York State Office of Children and Family Services. New York: Columbia University School of Social Work.
| Screening decision | Screening conclusion | HomVEE procedures and standards version |
|---|---|---|
| Passes screens | Eligible for review | Version 1 |
| Rating | Design | Attrition | Baseline equivalence | Compromised randomization | Confounding factors | Valid, reliable measure(s) |
|---|---|---|---|---|---|---|
| Moderate | Randomized controlled trial | High | Established on race/ethnicity, SES, and feasible outcomes. Statistical controls for feasible baseline outcomes included. | Yes for some outcomes | None | Not assessed in manuscripts reviewed before 2021 |
Moderate rating applies to outcomes where the analytic sample size is at least 90 percent as large as the sample used to demonstrate baseline equivalence. Other outcomes receive a low rating. The moderate rating applies to the following outcomes: Increase in mother’s highest grade completed (baseline to 24 months); receipt of public assistance (up to 24 months postpartum); number of pediatric emergency room visits (up to 36 months of age); ASQ communication, gross motor, fine motor, social, and composite score (6 months); perceived social support (6, 12, and 24 months); service referrals (up to 24 months postpartum) for daycare/babysitting, English as a second language, job skills/search, pediatric primary care, AFDC or food stamps, housing assistance, Medicaid, immigration services, GED prep, early intervention program, family planning, food pantry, counseling, adult primary care, child health plus, WIC, parenting education or training, and domestic violence services.
| Study participants | Women who were pregnant or had an infant under 2 months old and who lived in two eligible census tracts were screened using a checklist for risk factors for child abuse and neglect. Women who were deemed at risk were further screened using the Kempe Family Screening Inventory (FSI). Consenting women (n = 672) who received a score of 25 or higher on the FSI were assigned to the program group (n = 329) or the comparison group (n = 343). From November 1994 to November 1995, all women in one tract were assigned to the program group and participants from the other tract were enrolled in the comparison group. If Family Assessment Workers determined during the course of screening that a member of the family was affected by substance use, the family was provided additional Best Beginnings services (regardless of program or comparison group membership). Starting in November 1995, women were randomly assigned to the program group or comparison group within their census tract. Of the women who were randomly assigned, 535 were enrolled in Best Beginnings (273 in the program group, and 262 in the comparison group). Among participating women, 89.9% were born outside the United States and 88.3% were of Dominican ethnicity. On average at baseline, mothers were 26.3 years of age, 19.1% were employed, and 61.4% were receiving public assistance. |
|---|---|
| Setting | Eligible women were recruited from one of two contiguous census tracts in Washington Heights, New York. Most women were receiving health care from the New York Presbyterian Hospital (NYPH) Ambulatory Care Network Corporation (ACNC). |
| Intervention services | Families in the program group received home visits modeled after Healthy Families New York, but modified to meet specific goals. Goals for the visits included (1) assessing family strengths and needs and connecting the families with needed services, (2) improving maternal psychosocial functioning and life course, (3) improving parent-child interactions, and (4) promoting child growth and development. |
| Comparison conditions | Participants in the comparison group received 1-2 home visits every six months until the child was age 5, and were provided with educational materials and information about community services. Assessment, screening and referrals for needed services were provided. |
| Subgroups examined |
This field lists subgroups examined in the manuscript (even if they were not replicated in other samples and not reported on the summary page for this model’s report). • Child gender (boy or girl) |
| Funding sources | New York State Office of Children and Family Services; the U.S. Department of Health and Human Services Children’s Bureau Abandoned Infants Assistance Program; and the Smith Richardson Foundation. |
| Author affiliation | None of the study authors are developers of this model. |
| Peer reviewed | Peer reviewed status is not listed for manuscripts reviewed before 2021. |
| Rating | Outcome measure | Effect | Sample | Timing of follow-up | Sample size | Intervention group | Comparison group | Group difference | Effect size | Statistical significance | Notes |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Moderate | ASQ Communication score | FavorableUnfavorable or ambiguousNo Effect |
Program group children not affected by substance abuse, NY Best Beginnings trial |
6 months | 359 children | Mean = 55.70 | Mean = 55.30 | Mean difference = 0.40 | HomVEE calculated = 0.07 | Statistical significance not reported | |
| Moderate | ASQ Composite score | FavorableUnfavorable or ambiguousNo Effect |
Program group children not affected by substance abuse, NY Best Beginnings trial |
6 months | 359 children | Mean = 54.80 | Mean = 54.00 | Mean difference = 0.80 | HomVEE calculated = 0.15 | Statistical significance not reported | |
| Moderate | ASQ Fine Motor score | FavorableUnfavorable or ambiguousNo Effect |
Program group children not affected by substance abuse, NY Best Beginnings trial |
6 months | 359 children | Mean = 55.10 | Mean = 54.20 | Mean difference = 0.90 | HomVEE calculated = 0.12 | Statistical significance not reported | |
| Moderate | ASQ Gross Motor score | FavorableUnfavorable or ambiguousNo Effect |
Program group children not affected by substance abuse, NY Best Beginnings trial |
6 months | 359 children | Mean = 52.90 | Mean = 51.80 | Mean difference = 1.10 | HomVEE calculated = 0.12 | Statistical significance not reported | |
| Moderate | ASQ Social score | FavorableUnfavorable or ambiguousNo Effect |
Program group children not affected by substance abuse, NY Best Beginnings trial |
6 months | 359 children | Mean = 53.70 | Mean = 54.20 | Mean difference = -0.50 | HomVEE calculated = -0.06 | Statistical significance not reported |
| Rating | Outcome measure | Effect | Sample | Timing of follow-up | Sample size | Intervention group | Comparison group | Group difference | Effect size | Statistical significance | Notes |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Moderate | Number of pediatric emergency department visits | FavorableUnfavorable or ambiguousNo Effect |
Program group children not affected by substance abuse, NY Best Beginnings trial |
36 months | 271 children | Not available | Not available | Beta = 0.12 | Not available | Statistically significant, p < 0.05 |
| Rating | Outcome measure | Effect | Sample | Timing of follow-up | Sample size | Intervention group | Comparison group | Group difference | Effect size | Statistical significance | Notes |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Moderate | Increased education by year or more since baseline | FavorableUnfavorable or ambiguousNo Effect |
Program group families not affected by substance abuse, NY Best Beginnings trial |
24 months | 512 families | % = 18.40 | % = 7.40 | OR = 2.50 | HomVEE calculated = 0.63 | Statistically significant, p < 0.05 |
|
| Moderate | Receipt of public assistance | FavorableUnfavorable or ambiguousNo Effect |
Program group families not affected by substance abuse, NY Best Beginnings trial |
24 months | 512 families | Not available | Not available | Beta = 0.02 | Not available | Not statistically significant, p ≥ 0.05 | footnote62Study reports results using a standardized partial regression coefficient. |
| Rating | Outcome measure | Effect | Sample | Timing of follow-up | Sample size | Intervention group | Comparison group | Group difference | Effect size | Statistical significance | Notes |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Moderate | Referral to AFDC or food stamps | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 33.00 | % = 26.00 | = 7.00 | HomVEE calculated = 0.20 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to GED prep | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 29.00 | % = 26.00 | = -7.00 | HomVEE calculated = 0.09 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to WIC | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 19.00 | % = 12.00 | = 7.00 | HomVEE calculated = 0.33 | Not statistically significant, p > 0.10 |
footnote61In contrast to the study-reported results, HomVEE calculations showed this difference to be statistically significant (p ≤ 0.05). The HomVEE tests of statistical significance are based on the HomVEE calculated effect sizes, whereas authors may have used other techniques to determine statistical significance, such as regression models or analyses of variance (ANOVA). |
| Moderate | Referral to adult primary care | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 20.00 | % = 17.00 | = 3.00 | HomVEE calculated = 0.12 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to child health plus | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 18.00 | % = 16.00 | = 2.00 | HomVEE calculated = 0.09 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to counseling | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 28.00 | % = 21.00 | = 7.00 | HomVEE calculated = 0.23 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to daycare/babysitting | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 52.00 | % = 50.00 | = 2.00 | HomVEE calculated = 0.05 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to domestic violence services | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 10.00 | % = 5.00 | = 5.00 | HomVEE calculated = 0.45 | Not statistically significant, p > 0.10 | footnote61In contrast to the study-reported results, HomVEE calculations showed this difference to be statistically significant (p ≤ 0.05). The HomVEE tests of statistical significance are based on the HomVEE calculated effect sizes, whereas authors may have used other techniques to determine statistical significance, such as regression models or analyses of variance (ANOVA). |
| Moderate | Referral to early intervention program | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 28.00 | % = 24.00 | = 4.00 | HomVEE calculated = 0.13 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to English as a Second Language | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 44.00 | % = 36.00 | = 8.00 | HomVEE calculated = 0.20 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to family planning | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 25.00 | % = 10.00 | = 15.00 | HomVEE calculated = 0.67 | Statistically significant, p < 0.01 |
|
| Moderate | Referral to food pantry | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 25.00 | % = 19.00 | = 6.00 | HomVEE calculated = 0.21 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to housing assistance | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 33.00 | = 37.00 | = -4.00 | HomVEE calculated = -0.11 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to immigration services | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 30.00 | % = 23.00 | = 7.00 | HomVEE calculated = 0.22 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to job skills/search | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 43.00 | % = 39.00 | = 4.00 | HomVEE calculated = 0.10 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to Medicaid | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 30.00 | % = 36.00 | = -6.00 | HomVEE calculated = -0.16 | Not statistically significant, p > 0.10 |
|
| Moderate | Referral to parenting education or training | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 14.00 | % = 31.00 | = -17.00 | HomVEE calculated = -0.62 | Statistically significant, p < 0.001 |
|
| Moderate | Referral to pediatric primary care | FavorableUnfavorable or ambiguousNo Effect |
Full sample | 24 months postpartum | 289 families | % = 41.00 | % = 32.00 | = 9.00 | HomVEE calculated = 0.24 | Not statistically significant, p > 0.10 |
footnote61In contrast to the study-reported results, HomVEE calculations showed this difference to be statistically significant (p ≤ 0.05). The HomVEE tests of statistical significance are based on the HomVEE calculated effect sizes, whereas authors may have used other techniques to determine statistical significance, such as regression models or analyses of variance (ANOVA). |
| Rating | Outcome measure | Effect | Sample | Timing of follow-up | Sample size | Intervention group | Comparison group | Group difference | Effect size | Statistical significance | Notes |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Moderate | Perceived social support (MSSI) | FavorableUnfavorable or ambiguousNo Effect |
Program group members not affected by substance abuse, NY Best Beginnings trial |
6 months | 354 mothers | Mean = 6.20 | Mean = 7.20 | Mean difference = -1.00 | HomVEE calculated = -0.31 | Statistical significance not reported | |
| Moderate | Perceived social support (MSSI) | FavorableUnfavorable or ambiguousNo Effect |
Program group members not affected by substance abuse, NY Best Beginnings trial |
24 months | 274 mothers | Mean = 6.20 | Mean = 6.60 | Mean difference = -0.40 | HomVEE calculated = -0.12 | Statistical significance not reported | |
| Moderate | Perceived social support (MSSI) | FavorableUnfavorable or ambiguousNo Effect |
Program group members not affected by substance abuse, NY Best Beginnings trial |
12 months | 325 mothers | Mean = 6.50 | Mean = 6.60 | Mean difference = -0.10 | HomVEE calculated = -0.03 | Statistical significance not reported |
