Healthy Families New York (HFNY)

OVERVIEW

Healthy Families New York (HFNY) is a home visitation program for mothers who are at risk for developing abusive or neglectful behaviors towards their children. The goals of HFNY are to promote positive parenting, prevent child abuse and/or neglect, support prenatal care and child health and development, and improve parental self-sufficiency. An experimental evaluation of HFNY found that the program was effective in reducing the amount of child abuse and neglect in the experimental group. Also, among parents in "prevention" and "psychologically vulnerable" subgroups, those parents in the experimental condition were less likely to report severe abuse and/or neglect than mothers in the control group. A second evaluation found that the program reduced the prevalence of low birth weight among participants' children.

DESCRIPTION OF PROGRAM

Target population: Pregnant women or mothers with newborns younger than 3 months who are at risk for engaging in child abuse and neglect

Healthy Families New York was developed as a prevention effort against child abuse and neglect. The program is geared specifically towards young, first-time mothers who join the program before the birth of their children. The program is centered on home visitation services from trained paraprofessionals and home visitors from the community who help promote positive parenting, prevent child abuse or neglect, support prenatal care and child health and development, and improve parent self-sufficiency. Parents who are evaluated as being potentially eligible for the program are given a brief screening to determine final eligibility for the program. Once eligibility is determined, the women in the program receive home visits every other week during the mother's pregnancy. Once the child is born, visits are conducted weekly until the newborn is six months old. Then, the visitation schedule is slowly decreased until the child enrolls in a school program, Head Start, or reaches the age of 5 years. The home visits are designed to improve the parent-child relationship, teach about child development, encourage optimal growth, address family problems, and develop individual family plans for self-sufficiency and functioning. Visits are individually suited to the child and use varied curricula selected by staff (example curricula used include "Partners for a Healthy Baby" and "Parents as Teachers"). Children are also screened using instruments that measure developmental progress and referrals to other programs are available and recommended to parents as needed. The cost per family of the HFNY program ranged from $3000-3500 per year.

EVALUATION(S) OF PROGRAM

Study 1

Mitchell-Herzfeld, S. D., Izzo, C., Greene, R., Lee, E., & Lowenfels, A. A. (2005). Evaluation of Healthy Families New York (HFNY): First year program impacts. Rensselaer, New York: New York State Office of Children & Family Services.

Evaluated Population: Women from three sites (N=1,173) who were determined eligible for participation in the program based on scores from a screener, the Family Stress Checklist, were evaluated.To be eligible, women had to score above 25 points which defined those who were at risk of engaging in child abuse and neglect. Women were interviewed and determined eligible (N=1,254) through screener interviews, and 1,173 of those women completed a baseline interview. This study adopted an intent-to-treat (ITT) analysis whereby these 1,173 women were included in final analyses even if they dropped out of the program. Thirty-four percent of mothers in the study were white, 45 percent were African-American, and 18 percent were Latina. Thirty-one percent of mothers were under 19, 54 percent were first time mothers, 53 percent had never completed high school, and 82 percent had never been married.

Approach: Mothers were identified using an initial screening as being potential study participants. A Family Assessment Worker (FAW) then visited the home of potential participants and described the program and evaluation. After agreeing to take part in the program, the FAW administered the Family Stress Checklist (FSC) to determine if the woman was eligible to participate in the study. Those deemed eligible (greater than 25 points on the FSC) were randomly assigned to either an experimental condition (N=579), which received all of the HFNY services, or a control condition (N=594), which received referrals to other programs based on a needs assessment given at the beginning of the interview.

The variables measured at baseline and year 1 include: parents' demographic characteristics and background, marital status and living arrangements, housing and neighborhood, parenting attitudes and practices, childhood history of abuse and neglect, prenatal care and birth outcomes, child's and parents' health and health care, child's safety, employment, dependence on government assistance, financial hardship, parents' education and training, domestic violence, parents' mental health and coping, social support, parental use of substances, and family planning.

Parenting practices included a previous abuse/neglect composite of behaviors, including: hit with fist, hit with object, threw/knocked down, shook, grabbed neck, beat up, burned, threatened with gun, left child home alone, not able to make sure child got necessary medical attention, and so drunk/high had problem taking care of child.

Results:

Parenting Attitudes and Practices

For the entire sample, there were no significant impacts on parenting attitudes regarding physical punishment, inappropriate expectations, empathy, role reversal, or power/independence. However, certain subgroups did have impacts in one category. For physical punishment, mothers in the treatment group who were under age 18 and mothers who scored in the lowest 25 percent on the depression scale were significantly less likely to be in favor of physical punishment than control mothers in the same sub-groups (38 versus 37, and 39 versus 37, respectively).

Parent-reported neglect was the only parenting practice that was significantly different between the treatment and control group (5 versus 9 percent, respectively). The Latina subgroup also had a significantly smaller percentage of parents who had neglected their children (2 percent versus 13 percent). For minor physical aggression, mothers in a situation with no domestic violence showed a significantly smaller percentage of any aggressive behaviors than the control group (37 percent versus 51 percent). No significant differences were found for the other prevalence outcomes, including the previous abuse/neglect composite, severe/very severe physical abuse, minor physical aggression, and CPS reports.

The same variables were examined, with one additional, psychological aggression, to assess how often the parenting practices took place,. HFNY parents had significantly fewer instances in the previous abuse/neglect composite (0.2 versus 0.5), of minor physical aggression (2.3 versus 3.3), and of psychological aggression (3.2 versus 4.9). The subgroups with significant impacts were 18 or older (0.2 versus 0.7) and no domestic violence (0.1 versus 0.9) for serious abuse/neglect composite, Latinas (0 versus 1.2) for the neglect variable, no domestic violence (1.8 versus 3.3) for the minor physical aggression variable, and highest 25 percent on depression scale (4.0 versus 9.3) and no domestic violence (3.1 versus 5.1) for psychological aggression.

Prenatal Care and Birth Outcomes

For prenatal care and birth outcomes, the only significant impact was for low birth weight. A smaller percentage of HFNY group mothers randomized at least two months before their birth compared with control group mothers in the same situation (3 versus 8 percent) had low birth weight babies. No significant impacts were found for number of prenatal care visits in the third trimester, premature birth, or receipt of neonatal intensive care.

Child Health and Safety

For child health and safety, there was one significant impact for the entire sample. HFNY group children were more likely to have health insurance (94 versus 90 percent). Among parents who had already had another child at intake, the HFNY group mothers were more likely to have breast-fed the child selected for this study than control group mothers (50 versus 40 percent). No significant impacts were found for whether the mother had health insurance, whether the mother had a primary care provider, child having primary care provider, child ever went without needed medical care, number of child emergency room visits, whether the family received WIC, number of well baby visits, whether all immunizations were complete, number of months breast-fed, child safety checklist, or any emergency room visits involving injury or ingestion.

Mother's Life Course Development

There were two significant impacts for mothers' life course development variables, one for the entire sample and one for mothers under age 18. HFNY group mothers were less likely to be employed at year 1 compared with control mothers (41 versus 48 percent). Smoking frequency was significant for mothers under age 18; HFNY group mothers smoked fewer cigarettes per day than control group mothers (an average of 1.8 versus 3.3). There were no significant impacts for: TANF receipt, receipt of less than 50 percent of income from work, having education appropriate for age, achieving more education in the past year, depression score, being above the depression cutoff, having a sense of personal mastery, smoking status, alcohol abuse score, scoring above the alcohol abuse cutoff, amount of drinking, average number of drinks when drinking, having any drug use, consistently using birth control, and having a pregnancy or childbirth after the birth of the target child or within a year prior to the year 1 interview.

Study 2

DuMont, K., Mitchell-Herzfeld, S., Greene, R., Lee, E., Lowenfels, A., & Rodriguez, M. (2006). Healthy Families New York (HFNY) randomized trial: Impacts on parenting after the first two years. Working Paper Series: Evaluating Healthy Families, OCFS Working Paper #1.

Evaluated population: The evaluated population is the same sample as in Study 1.

Approach: For randomization procedures, see Study 1. Baseline measures were taken two weeks after the intake interview. Then, women were assessed again when their child was born and on the child's first and second birthdays.

Mothers were given a self-report measure of abusive and neglectful parenting and data was taken from the Office of Children and Family Services (OCFS), which keeps records of child abuse and neglect. Mothers were paid $40 for participation at each of the three waves of data collection.

For analysis, two subgroups were designed to measure the program's impacts on different populations. The first subgroup, the "prevention subgroup", consisted of 170 first-time mothers under the age of 19 years who had been assigned to the evaluation at a gestational age of 30 weeks or less. The second subgroup consisted of 122 mothers determined to be "psychologically vulnerable" based on the presence of depressive symptoms and mastery, measured respectively by the Center for Epidemiologic Studies - Depression scale and the Mastery of Psychological Coping Resources Scale. These variables were combined into an index of psychological vulnerability and women scoring in the top 10 percent were categorized as psychologically vulnerable.

To measure parenting behaviors, parents were given the Conflict-Tactics Scale (Parent-Child) to determine how often in the past year mothers engaged in abusive or neglectful behaviors. The researchers also included substantiated reports from Child Protective Services (CPS) as a measure of child abuse and neglect.

Results: At the first birthday follow-up, mothers in the HFNY program reported fewer acts of very serious physical abuse compared with mothers in the control group. Mothers in the HFNY program also reported fewer cases of minor physical aggression, psychological aggression, and harsh parenting behaviors compared with mothers in the control group. At year 2, mothers in the HFNY program reported fewer cases of serious physical abuse compared with mothers in the control condition. There were no differences found between groups at any of the follow-up assessments with respect to substantiated reports from CPS.

Within the "prevention subgroup," mothers in the HFNY program were less likely to report minor physical aggression compared to their counterparts in the control condition (51 compared with 70 percent). Likewise, "prevention subgroup" mothers in HFNY were less likely to report harsh parenting behaviors than "prevention subgroup" mothers in the control condition (41 versus 62 percent).

Within the "psychologically vulnerable subgroup", mothers in the experimental program reported fewer acts of serious abuse or neglect at year one compared with their counterparts in the control group (two versus nine incidents of abuse or neglect). At year one follow up, "psychologically vulnerable" mothers reported less psychological aggression and, at year two, fewer instances of serious abuse and neglect than mothers in the control condition (0.02 versus 0.6 incidents).

Study 3

DuMont, K. A., Mitchell-Herzfeld, S. D., Greene, R., Lee, E., Lowenfels, A. A., Rodriguez, M., et al. (2008). Healthy Families New York (HFNY) randomized trial: Effects on early child abuse and neglect. Child Abuse & Neglect, 32, 295-315.

Evaluated Population: The evaluated population is the same as in Study 1.

Approach: See Study 1 for randomization procedures. Mothers reported 27 of their own parenting behaviors in the past year and five in the past week in an interview. The questions covering the past year fell under six categories: neglect, psychological aggression, nonviolent discipline, minor physical aggression, severe abuse, and very severe abuse. The five questions for the past week covered harsh parenting. In addition, official data on substantiated reports of abuse and neglect (CPS reports) came from the Office of Children and Family Services (OCFS) database, called CONNECTIONS. The variables taken into account when examining the associations between program assignment and child abuse and neglect are the following: race, random assignment at gestational age 30 weeks versus earlier, site of participation, history of substantiated abuse or neglect, first-time mother versus not, mother's age, whether the mother was born in the U.S., welfare receipt in the past year, psychological vulnerability, and a single item capturing the mother's history of being abused as a child.

Results: The HFNY treatment group did not differ significantly in prevalence from the control group on any of the six categories or number of substantiated reports of abuse and neglect based on official records. Based on maternal report at year 1, however, the HFNY group mothers reported fewer incidents of very serious physical abuse (0.01 versus 0.08 times), minor physical aggression (2.4 versus 3.5), psychological aggression (3.3 versus 4.7), and harsh parenting in the past week (1.2 versus 1.8) compared with control group mothers. At year 2, the frequency of serious physical abuse was significantly lower in the HFNY group mothers than the control group mothers (0.01 versus 0.04).

The prevention subgroup, first-time mothers under age 19 who were randomly assigned at gestational age 30 weeks or less, were less likely to report minor physical aggression against their children in the past year (51 versus 70 percent) or harsh parenting behaviors in the past week (41 versus 62 percent) at year 2. No significant impacts were found for the prevalence of any other self-report parenting behaviors or the frequency of self-report parenting behaviors.

A second subgroup is called psychologically vulnerable; these are mothers with a low sense of mastery and high levels of depressive symptoms. One significant impact was found at year 1 for self-reported parenting behaviors in frequency, and none for prevalence. HFNY mothers reported a significantly lower number of acts of psychological aggression compared with the control mothers (2 versus 8.6). At year 2 HFNY group mothers were less likely to report serious abuse or neglect (5 versus 19 percent). The frequency of serious abuse and neglect behaviors at year 2 was similar, 0.02 for HFNY mothers and 0.6 for control mothers. No impacts were found based on substantiated CPS reports.

Study 4

DuMont, K. A., Mitchell-Herzfeld, S. D., Kirkland, K., Rodriguez, M., Walden, N., Greene, R., et al. (2008). Effects of Healthy Families New York on maternal behaviors: Observational assessments of positive and negative parenting. Rensselaer, New York: New York State Office of Children and Family Services.

Evaluated Population: The base sample for this evaluation comes from the same sample in Study 1. However, this study had special selection criteria for this observational study embedded within its year 3 follow-up. Mothers who had completed year 2 interviews and who lived in a certain range of the interviewers were eligible. Of the 643 eligible women, 522 agreed to participate. The race makeup of the sample was 42 percent white, 39 percent black, and 16 percent Hispanic. Twenty-nine percent were under 19, and 12 percent were at least 30. Fifty-five percent of participants were first-time mothers. Twenty-one percent of mothers had been involved with CPS prior to baseline and 8 percent had at least one substantiated report. The children involved in the year 3 observations were an average of three years old, and 66 percent were at least three.

Approach: The 522 mothers were in treatment (N=255) and control (N=267) groups, and had not been randomized a second time. Three interaction tasks were video recorded: a puzzle problem-solving task, a delay of gratification task, and a cleanup task. In each task, the mother was observed for variables related to positive parenting and negative parenting (harsh parenting and role-reversed parenting). Positive parenting consisted of stimulating and engaging the child's cognitive skills, being sensitive to the child's needs, and creating a structured environment in which the child can explore.

Results: A higher percentage of HFNY group mothers used positive parenting techniques in all three tasks compared with mothers in the control group. Impact on the puzzle task (97 versus 93 percent), delay task (17 versus 11 percent), and cleanup task (85 versus 78 percent) differences were statistically significant.

Impacts for subgroups were also examined for positive parenting and for negative parenting. No significant impacts on positive parenting were found for mothers in the prevention subgroup (first-time mothers under age 19 randomly assigned at gestational age 30 weeks or less).

No significant impacts were found for negative parenting overall, but there were when looking at the prevention subgroup of young first-time mothers for the puzzle task (5 versus 22 percent) and the delay task (5 versus 24 percent), with the HFNY mothers in the prevention subgroup being less likely to use negative parenting behaviors. In contrast, no significant impacts were found for the non-prevention subgroup on negative parenting.

Study 5

Lee, E., Mitchell-Herzfeld, S. D., Lowenfels, A. A., Greene, R., Dorabawila, V., & DuMont, K. A. (2009). Reducing low birth weight through home visitation: A randomized controlled trial. American Journal of Preventive Medicine, 36(2), 154-160.

Evaluated Population: From the sample in Study 1, 501 mothers who had a single birth and were randomized prior to 30 weeks gestational age were evaluated. Sixty-six percent of participants were black or Hispanic and 90 percent were unmarried.

Approach: Participants were randomly assigned to receive bi-weekly home-visitation services (treatment group, N=236) or to a control group (N=265) at 30 weeks or less of gestation. Low birth weight (less than 2,500 grams) was the only outcome examined in this paper. Other variables taken into account were smoking during pregnancy, TANF enrollment (Temporary Assistance for Needy Families), race, Hispanic origin, program site, marital status, and number of previous pregnancies.

Results: Births to mothers in the treatment group compared with the control group were statistically significantly less likely to be under 2,500 grams (five percent versus ten percent, respectively). The difference in low birth weight births in the treatment group did not vary by gestational age at randomization; rather, within all subgroups (randomized at 30 weeks or fewer, 24 weeks or fewer, or 16 weeks or fewer) the treatment group mothers had significantly fewer low birth weight babies than the control group mothers.

In racial subgroups, the black mothers in the treatment group were the only ones to be significantly different from the control group. The difference between HFNY mothers and control mothers were non-significant for Hispanics and whites.

SOURCES FOR MORE INFORMATION

References

DuMont, K., Mitchell-Herzfeld, S., Greene, R., Lee, E., Lowenfels, A., & Rodriguez, M. (2006). Healthy Families New York (HFNY) randomized trial: Impacts on parenting after the first two years. Working Paper Series: Evaluating Healthy Families, OCFS Working Paper #1.

DuMont, K. A., Mitchell-Herzfeld, S. D., Greene, R., Lee, E., Lowenfels, A. A., Rodriguez, M., et al. (2008). Healthy Families New York (HFNY) randomized trial: Effects on early child abuse and neglect. Child Abuse & Neglect, 32, 295-315.

DuMont, K. A., Mitchell-Herzfeld, S. D., Kirkland, K., Rodriguez, M., Walden, N., Greene, R., et al. (2008). Effects of Healthy Families New York on maternal behaviors: Observational assessments of positive and negative parenting. Rensselaer, New York: New York State Office of Children and Family Services.

Lee, E., Mitchell-Herzfeld, S. D., Lowenfels, A. A., Greene, R., Dorabawila, V., & DuMont, K. A. (2009). Reducing low birth weight through home visitation: A randomized controlled trial. American Journal of Preventive Medicine, 36(2), 154-160.

Mitchell-Herzfeld, S. D., Izzo, C., Greene, R., Lee, E., & Lowenfels, A. A. (2005). Evaluation of Healthy Families New York (HFNY): First year program impacts. Rensselaer, New York: New York State Office of Children & Family Services.

Link to program curriculum:  http://healthyfamiliesnewyork.org/

KEYWORDS: Adolescents; Youth; Young Adults; Infants; Toddlers; Children; High-Risk; Adolescent Mothers; Female Only; White/Caucasian; Black/African American; Hispanic/Latino; Home-based; Home Visitation; Clinic-based; Child Care; Early Childhood Education; Case Management; Public Assistance; Parent-Child Relationship; Child Maltreatment; Depression/Mood Disorders; Other Physical Health; Manual is Available; Parent or Family Component; Tobacco Use; Illicit Drug Use; Alcohol Use; Contraception; Births.

Program information last updated on 11/9/11.