Family Connects Meets HHS Criteria

Model effectiveness research report last updated: 2023

Model overview

Theoretical approach

The Family Connects* model aims to intervene with families early to help them address health concerns and access support to reduce potential disparity gaps later, especially among families that are part of historically marginalized groups. Families in need might be unaware of local resources and how to navigate various systems to access services. Thus, registered nurses working as Family Connects home visitors provide free check-ups to mothers and babies during the early postpartum period and help connect families to local resources.

Family Connects is a universal model built on the premise that all families can benefit from postpartum nursing support or referrals. The model is designed to serve all families in a community, including families with surrogate, adoptive, and bereaved parents. The model’s aim to achieve population-level results is informed by a socioecological framework in which communities’ beliefs, practices, and environmental structures influence the choices, and thereby the actions, of the communities’ families. Similarly, communities are influenced by their state’s context, and each state is nested within a broader setting of national influences. Thus, Family Connects International (FCI) works at the national, state, community, and family levels to support newborns and their families.

*The model began in 2008 as Durham Connects—a pilot that served Durham County, North Carolina. As part of subsequent replication, the model was renamed Family Connects to reflect its larger service area. The model has not changed between the pilot and replication.

View Revisions

Model services

Family Connects nurses assess newborns and their mothers. The nurses then use the assessment results to speak with families about concrete next steps to address opportunities and concerns. For example, they might advise a family to seek immediate medical care when necessary. Family Connects nurses consider the needs of the entire family and might recommend mental health services or medical care for other family members as needed. Program staff follow up with families to ensure their needs are met.

Please contact the model developer for information about the model’s curriculum.

View Revisions

Intended population

The Family Connects model is a universal model in which all parents of newborns in a community are offered a home visit with a nurse within about three weeks of birth.

View Revisions

Where to find out more

Kimberly Friedman, Managing Director, External Resources
Family Connects International
3710 University Drive, Suite 310
Durham, NC 27707

Email: info@familyconnects.org
Website: http://www.familyconnects.org/

View Revisions

Effects shown in research

Maternal health

Findings rated moderate

Family Connects
Show findings details
Outcome measure Effect Follow-up timing Sample Sample size Intervention group Comparison group Group difference Effect size Statistical significance
Mother possible anxiety disorder
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted proportion = 0.21 Unadjusted proportion = 0.30 OR = 0.65 HomVEE calculated = -0.27 Statistically significant, p<0.05
Mother possible clinical depression disorder
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted proportion = 0.08 Unadjusted proportion = 0.12 OR = 0.69 HomVEE calculated = -0.29 Not statistically significant,
p = 0.25
Mother possible substance use problems
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 530 families Unadjusted proportion = 0.05 Unadjusted proportion = 0.06 OR = 0.80 HomVEE calculated = -0.19 Not statistically significant,
p = 0.58
View Revisions

Child health

Findings rated high

Family Connects
Show findings details
Outcome measure Effect Follow-up timing Sample Sample size Intervention group Comparison group Group difference Effect size Statistical significance Notes

Inpatient hospital overnights (component of total child emergency medical care)

FavorableUnfavorable or ambiguousNo Effect

Age 12 - 24 months

Children born in Durham County, North Carolina during the 18-month study period

531 children Adjusted mean = 0.02 Adjusted mean = 0.11 Mean difference = -0.09 Study reported = 0.12

Not statistically significant, p= 0.06

Inpatient hospital overnights (component of total child emergency medical care)

FavorableUnfavorable or ambiguousNo Effect

Through age 24 months

Children born in Durham County, North Carolina during the 18-month study period

531 children Adjusted mean = 0.13 Adjusted mean = 0.86 Mean difference = -0.73 Study reported = 0.24

Statistically significant, p= <0.001

Outpatient ER visits (component of total child emergency medical care)

FavorableUnfavorable or ambiguousNo Effect

Age 12 - 24 months

Children born in Durham County, North Carolina during the 18-month study period

531 children Adjusted mean = 0.71 Adjusted mean = 0.72 Mean difference = -0.01 Study reported = 0.01

Not statistically significant, p= 0.88

Outpatient ER visits (component of total child emergency medical care)

FavorableUnfavorable or ambiguousNo Effect

Through age 24 months

Children born in Durham County, North Carolina during the 18-month study period

531 children Adjusted mean = 1.39 Adjusted mean = 1.55 Mean difference = -0.16 Study reported = 0.08

Not statistically significant, p= 0.47

Total child emergency medical care

FavorableUnfavorable or ambiguousNo Effect

Age 12 - 24 months

Children born in Durham County, North Carolina during the 18-month study period

531 children Adjusted mean = 0.73 Adjusted mean = 0.84 Mean difference = -0.11 Study reported = 0.07

Not statistically significant, p= 0.73

Total child emergency medical care

FavorableUnfavorable or ambiguousNo Effect

Through age 24 months

Children born in Durham County, North Carolina during the 18-month study period

531 children Adjusted mean = 1.52 Adjusted mean = 2.41 Mean difference = -0.89 Study reported = 0.24

Statistically significant, p= <0.001

Findings rated moderate

Family Connects
Show findings details
Outcome measure Effect Follow-up timing Sample Sample size Intervention group Comparison group Group difference Effect size Statistical significance
No. of emergency department visits, birth to 12 months (hospital records)
FavorableUnfavorable or ambiguousNo Effect
birth to 12 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.68 Unadjusted mean = 0.83 Mean difference = -0.15 Not available Not statistically significant,
p = 0.22
No. of overnights in hospital, birth to 12 months (hospital records)
FavorableUnfavorable or ambiguousNo Effect
birth to 12 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.11 Unadjusted mean = 0.74 Mean difference = -0.63 Study reported = 0.27 Statistically significant, p<0.001
No. of total emergency medical care episodes, 6 to 12 months (hospital records)
FavorableUnfavorable or ambiguousNo Effect
6 to 12 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.36 Unadjusted mean = 0.52 Mean difference = -0.16 Study reported = 0.14 Statistically significant,
p = 0.03
No. of total emergency medical care episodes, birth to 12 months (hospital records)
FavorableUnfavorable or ambiguousNo Effect
birth to 12 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.78 Unadjusted mean = 1.57 Mean difference = -0.79 Study reported = 0.28 Statistically significant, p<0.001
Family Connects
Show findings details
Outcome measure Effect Follow-up timing Sample Sample size Intervention group Comparison group Group difference Effect size Statistical significance
Did not have emergency medical care episodes (hospital records), proportion
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted proportion = 0.74 Unadjusted proportion = 0.72 Mean difference = 0.02 HomVEE calculated = 0.06 Not statistically significant, p=0.50
Did not have emergency medical care episodes (parent report) proportion
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted proportion = 0.50 Unadjusted proportion = 0.51 Mean difference = -0.01 HomVEE calculated = -0.03 Not statistically significant, p=0.70
Most recent well-baby visit within past month
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted proportion = 0.70 Unadjusted proportion = 0.67 OR = 1.12 HomVEE calculated = 0.07 Not statistically significant,
p = 0.54
No. of emergency department visits (hospital records)
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.32 Unadjusted mean = 0.39 Mean difference = 0.14 HomVEE calculated = -0.09 Not statistically significant,
p = 0.36
No. of emergency medical visits (parent report)
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.82 Unadjusted mean = 0.98 Mean difference = 0.17 HomVEE calculated = -0.12 Not statistically significant,
p = 0.07
No. of overnights in hospital (parent report)
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.08 Unadjusted mean = 0.40 Mean difference = 1.60 Study reported = 0.20 Statistically significant, p<0.001
No. of total emergency medical care episodes (hospital records)
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.43 Unadjusted mean = 1.05 Mean difference = 0.91 Study reported = 0.26 Statistically significant, p<0.001
No. of total emergency medical care episodes (parent report)
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.89 Unadjusted mean = 1.37 Mean difference = 0.40 Study reported = 0.21 Statistically significant, p<0.001
Three or more emergency medical care episodes (hospital records), proportion
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted proportion = 0.03 Unadjusted proportion = 0.09 Mean difference = -0.05 HomVEE calculated = -0.65 Statistically significant, p<0.01
Three or more emergency medical care episodes (parent report), proportion
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted proportion = 0.09 Unadjusted proportion = 0.13 Mean difference = -0.04 HomVEE calculated = -0.26 Statistically significant, p<0.01
Two or more emergency medical care episodes (hospital records), proportion
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted proportion = 0.10 Unadjusted proportion = 0.15 Mean difference = -0.06 HomVEE calculated = -0.31 Statistically significant, p<0.01
Two or more emergency medical care episodes (parent report), proportion
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted proportion = 0.21 Unadjusted proportion = 0.25 Mean difference = -0.04 HomVEE calculated = -0.13 Not statistically significant, p=0.13
View Revisions

Positive parenting practices

Findings rated moderate

Family Connects
Show findings details
Outcome measure Effect Follow-up timing Sample Sample size Intervention group Comparison group Group difference Effect size Statistical significance
Father–infant relationship quality
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 524 families Unadjusted mean = 2.09 Unadjusted mean = 1.92 Mean difference = 0.11 HomVEE calculated = 0.21 Not statistically significant,
p = 0.08
Mother knowledge of infant development
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.75 Unadjusted mean = 0.77 Mean difference = 0.02 HomVEE calculated = -0.11 Not statistically significant,
p = 0.10
Mother negative parenting behaviors
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 0.32 Unadjusted mean = 0.33 Mean difference = 0.02 HomVEE calculated = -0.03 Not statistically significant,
p = 0.59
Mother positive parenting behaviors
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 4.12 Unadjusted mean = 4.01 Mean difference = 0.10 Study reported = 0.25 Statistically significant, p<0.01
Mother sense of parenting competence
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 4.62 Unadjusted mean = 4.63 Mean difference = 0.01 HomVEE calculated = -0.02 Not statistically significant,
p = 0.83
View Revisions

Linkages and referrals

Findings rated moderate

Family Connects
Show findings details
Outcome measure Effect Follow-up timing Sample Sample size Intervention group Comparison group Group difference Effect size Statistical significance
No. of community connections
FavorableUnfavorable or ambiguousNo Effect
Infant Age 6 months Interviewed subsample, 7/2009-12/2010 births 531 families Unadjusted mean = 5.02 Unadjusted mean = 4.31 Mean difference = 0.86 Study reported = 0.28 Statistically significant, p<0.001
View Revisions

Reductions in child maltreatment

Findings rated high

Family Connects
Show findings details
Outcome measure Effect Follow-up timing Sample Sample size Intervention group Comparison group Group difference Effect size Statistical significance Notes

Emergency medical care for accidents or injuries

FavorableUnfavorable or ambiguousNo Effect

Through age 24 months

Children born in Durham County, North Carolina during the 18-month study period

531 children Adjusted mean = 0.27 Adjusted mean = 0.30 Mean difference = -0.03 Study reported = 0.07

Not statistically significant, p= 0.45

Maltreatment-related injuries

FavorableUnfavorable or ambiguousNo Effect

Through age 24 months

Children born in Durham County, North Carolina during the 18-month study period

531 children Adjusted mean = 0.02 Adjusted mean = 0.02 Mean difference = 0.00 Not available

Not statistically significant, p= <0.30

Multiple emergency visits for accidents and injuries

FavorableUnfavorable or ambiguousNo Effect

Through age 24 months

Children born in Durham County, North Carolina during the 18-month study period

531 children Adjusted mean = 0.03 Adjusted mean = 0.06 Mean difference = -0.03 Not available

Not statistically significant, p= 0.10

View Revisions

In brief

Evidence of model effectiveness

Title General population Tribal population Domains with favorable effects
Family Connects Yes, Meets HHS Criteria Meets HHS criteria Does not meet HHS criteria for tribal population because the model has not been evaluated with a tribal population.
  • Child health,
  • Linkages and referrals,
  • Maternal health,
  • Positive parenting practices,

Model description

Family Connects is a universal nurse home visiting model available to all families with newborns residing within a defined service area. The model aims to help families enhance maternal and child health and well-being and reduce rates of child abuse and neglect. It consists of one nurse home visit within three weeks of birth and follow-up contacts to confirm families’ successful linkages with community resources. During the initial home visit, a nurse conducts assessments of the mother and newborn. Nurses may make one or two follow-up visits or calls to provide additional focused support. If a family has a significant risk or need, the nurse might advise the family to seek immediate medical care. The model began as a pilot under the name Durham Connects and is being replicated under the name Family Connects.

View Revisions

Extent of evidence

Results of search and review
Number of manuscripts
At least one finding was eligible for review…
9
  …and at least one finding rated high
1
  …and at least one finding rated moderate (but none rated high)
2
  …and all findings that were eligible for review rated low or indeterminate2
3
  …but manuscript is additional source3
3

For more information, see the research database. For more information on the criteria used to rate research, please see details of HomVEE’s methods and standards.

View Revisions

Summary of findings

View Revisions

Criteria established by the U.S. Department of Health and Human Services

Information based on comprehensive review of all high- and moderate-rated manuscripts
CriterionCriterion descriptionCriterion met?
1High- or moderate-quality impact study?Yes
2Across high- or moderate-quality studies, favorable impacts in at least two outcome domains within one sample OR the same domain for at least two non-overlapping samples?Yes
3Favorable impacts on full sample?Yes
4Any favorable impacts on outcome measures sustained at least 12 months after model enrollment?
Reported for all research but only required for RCTs.
Yes
5One or more favorable, statistically significant impact reported in a peer-reviewed journal?
Reported for all research but only required for RCTs.
Yes
View Revisions