A note on this page's publication date
The content we created in 2010 appears below. This content is likely to be no longer fully accurate, both with respect to the research it presents and with respect to what it implies about our views and positions.
In a nutshell
- The Program: The Nurse-Family Partnership program offers home visits from trained registered nurses to low-income, first-time mothers. Nurses advise mothers on prenatal health, child care, and personal life management such as birth spacing and employment (more).
- Track record: The program has been the subject of several formal, rigorous evaluations that imply measurable impacts on both mothers (particularly on how long they go between births) and children (particularly cognitive and behavioral measures) (more).
- Cost-effectiveness: Based on information from the Nurse-Family Partnership National Service Office, we estimate a total cost per family served (for approximately 2.4 years) of about $11,000 (more).
- Bottom line: The Nurse-Family Partnership program has one of the strongest track records among the developed-world programs we've reviewed.
Table of Contents
According to the Coalition for Evidence-Based Policy:
Evidence of effectiveness
The primary evidence for the Nurse-Family Partnership (NFP) model's effectiveness consists of three randomized controlled trials (RCTs), in which low-income mothers were (or were not) assigned nurses based on lottery, and then researchers followed both those who did and those did not receive nurse visits, collecting a broad range of data. We find the evidence from these studies to provide a compelling case that the model in question improves life outcomes for children and mothers, across a variety of regions and population profiles.
A thorough literature review of home visitation programs points to NFP as a standout among such programs in terms of the methodological strength and encouraging implications of its studies.2 In addition, NFP is one of the seven social programs that meet the Coalition for Evidence-Based Policy's Top Tier evidence standard.3
What were the results?
In our discussion below, we only report results that are statistically significant at the .05 level.
Study 1: Elmira
The first RCT of the NFP model, in Elmira County, NY, followed families for 19 years after birth and found that youth who had participated in the NFP program were about half as likely to ever have been arrested or convicted as youth in the control group.5 The follow up 15 years after birth (but not the one 19 years after) measured child abuse and neglect and found that the families who received nurse-visits had about half as many reports of child abuse and neglect as the control group.6
Study 2: Memphis
As of August 2010, this evaluation has published follow ups from 3, 6, and 9 years after birth. The 6-year follow up found that children of visited mothers had better vocabulary and intellectual functioning, and were reported by their mothers to have fewer behavioral problems compared to the control group.7
The follow up at age 9 found no statistically significant effects on children for the whole sample,8 but it did find effects for mothers in the treatment group. Compared to the control group, nurse-visited mothers had 13% fewer live births, had been with their current partner 18% longer, and had spent 12% less time on welfare and 10% less time on food stamps during the nine years after their child's birth.9
Study 3: Denver
As of August 2010, the Denver study had followed children and mothers for 4 years post-birth. The study found no effects on the full sample of children.10 For mothers, the study found that about half as many nurse-visited mothers reported domestic violence from partners in the 6-month period prior to the interview as did mothers in compared to the control group. It also found that nurse-visited mothers had greater intervals between the births of their first and second children, and reported enrolling their children less frequently in preschool.11
We are in general concerned about researchers' conflicts of interests, which could bias results or lead to publication bias (the possibility that studies are more likely to be published when they show positive results).12
In the case of the NFP program, David Olds is the co-author of all publications from the three main studies of NFP discussed above, and also initiated development of the program.13
According to the NFP National Service Office, the cost to a community/agency averages $4,500 per family per year served in the NFP program, and ranges from $2,914 to $6,463 per family per year.14 These figures imply a typical cost of $10,800 per family served in the NFP program.15 We do not know the details of what costs were included in these figures.
- Coalition for Evidence-Based Policy. Nurse-Family Partnership. http://evidencebasedprograms.org/wordpress/?page_id=57 (accessed June 23, 2010). Archived by WebCite® at http://www.webcitation.org/5qht5bgHG.
- Coalition for Evidence-Based Policy. Social programs that work. http://evidencebasedprograms.org/wordpress/ (accessed October 5, 2010). Archived by WebCite® at http://www.webcitation.org/5tFgngQZO.
- Eckenrode, John, et al. 2010. Long-term effects of prenatal and infancy nurse home visitation on the life course of youths: 19-year follow-up of a randomized trial. Archives of Pediatric and Adolescent Medicine 164: 9-15.
- GiveWell. Summary of Nurse-Family Partnership study Results (XLS).
- Gomby, Deanna S. 2005. Home visitation in 2005: Outcomes for children and parents (PDF).
- Nurse-Family Partnership National Service Office. A cornerstone of Nurse-Family Partnership (PDF). Invest in Kids Working Paper 7. Washington, DC: Committee for Economic Development Invest in Kids Working Group.
- Nurse Family Partnership National Service Office. Program history. http://www.nursefamilypartnership.org/about/program-history (accessed November 11, 2010). Archived by WebCite® at http://www.webcitation.org/5u9Sz3b6V.
- Olds, David L., et al. 1997. Long-term effects of home visitation on maternal life course and child abuse and neglect: Fifteen-year follow-up of a randomized trial. Journal of the American Medical Association 278: 637-643.
- Olds, David L, et al. 1998. Long-term effects of nurse home visitation on children's criminal and antisocial behavior: 15-year follow-up of a randomized controlled trial. Journal of the American Medical Association 280: 1238-1244.
- Olds, David L., et al. 2004b. Effects of home visits by paraprofessionals and by nurses: Age 4 follow up results of a randomized trial. Pediatrics 114: 1560-1568.
- Olds, David L., et al. 2007. Effects of Nurse Home Visiting on Maternal and Child Functioning: Age-9 Follow-up of a Randomized Trial. Pediatrics 120: e832-e845.
- Stapleton, Michelle. NFP Fund Development Associate. Email to GiveWell, November 16, 2010.
Coalition for Evidence-Based Policy, "Nurse-Family Partnership."
“The following are the main conclusions of this review:
”¢ The popularity of home visiting has been driven by the results of a few studies of programs such as the Nurse-Family Partnership that demonstrate long-term benefits for parents and children.
”¢ However, there are very few long-term studies of home visiting programs.” Gomby 2005, Pg 2.
“The popularity of home visiting has been propelled by the findings of large and long-term benefits in a few studies (most notably, the studies of the Nurse-Family Partnership).” Gomby 2005, Pg 9.
“A hallmark of this program [Nurse-Family Partnership] is its use of research to determine program effectiveness and to improve services. Program implementation is monitored carefully in each site, using a management information system that is integral to the program. Longitudinal follow ups are being conducted in Elmira, Memphis, and Denver.” Gomby 2005, Pg 65.
“The Nurse-Family Partnership meets the Top Tier Evidence standard, defined per recent Congressional legislation to include: Interventions shown in well-designed and implemented randomized controlled trials, preferably conducted in typical community settings, to produce sizeable, sustained benefits to participants and/or society.” Coalition for Evidence-Based Policy, “Nurse-Family Partnership.”
As of October 5, 2010, there were 7 interventions labeled "Top Tier" on the web site Coalition for Evidence-Based Policy, "Social Programs that Work."
GiveWell, "Summary of Nurse-Family Partnership Study Results."
The publication for the follow up when the children are 19 years old is Eckenrode et al. (2010). When considering the whole sample, it finds the following statistically significant (at 5% significance level) effects of NFP on the 19-year olds:
- Fewer youth ever arrested (risk ratio 0.57 [95% CI: 0.35-0.93]; 21% in treatment vs 37% in control)
- Fewer youth ever convicted (risk ratio 0.42 [95% CI: 0.20-0.90]; 12% in treatment vs 28% in control)
- Fewer arrests in lifetime (incidence risk ratio 0.43 [95% CI: 0.23-0.80]; mean in treatment group 0.37, mean in control group 0.86)
- Fewer convictions in lifetime (incidence risk ratio 0.34 [CI:0.15-0.79]; mean in treatment group 0.20, mean in control group 0.58)
The publication from follow up when the children are 15 years old is Olds et al. (1997). When considering the full sample, it finds only one statistically significant (at 5% significance level) effects on mothers and child abuse and neglect of NFP: the families with who got nurse-visits had fewer substantiated reports of child abuse and neglect. The average incidence was 0.29 in the treatment group and 0.54 in the control group (P<0.01).
The publication for the follow up when the children were 6 years old is Olds et al. (2004a). When considering the whole sample, it finds the following statistically significant (at 5% significance level) effects of NFP on the 6-year olds:
- Higher scores on mental processing composite (KABC) (effect size 0.18; P=0.03; mean in treatment group 92.34, mean in control group 90.24)
- Higher scores on receptive vocabulary (PPVT-III) (effect size 0.17; P=0.04; mean in treatment group 84.32, mean in control group 82.13)
- Less total behavioral problems (borderline/clinical) based upon Achenbach Child Behavioral Problems Checklist (odds ratio 0.32; P=0.04; 1.8% in treatment group vs 5.4% in control)
"Nurse-visited children had higher scores on tests of intellectual functioning and receptive language (92.34 vs 90.24, P = .03, and 84.32 vs 82.13, P = .04, respectively) and were reported by their mothers to have fewer problems in the borderline or clinical range of the CBCL Total Problems scale (1.8% vs 5.4%, P = .04)." Olds et al. 2004a, Pgs 1554, 1556.
"Effects on the first-born children of nurse-visited women at age 9 (versus the control group):
- Lower mortality rate (0.4% of the children in the nurse-visited group died before age 9 vs. 1.9% of children in the control group). This effect was statistically significant at the .10 level, but not the .05 level. Most excess deaths in the control group were attributable to causes that are often preventable (e.g., injury, SIDS).
- The subsample of children whose mothers had low psychological resources prior to program participation (i.e. were in the lower half of the sample in intelligence, mental health, and self-confidence) made sizeable gains in academic performance. These children:
- Scored 9 percentile points higher on Tennessee state reading and math achievement tests in grades 1-3 than their counterparts in the control group (the nurse-visited group scored in the 45th percentile, versus the 36th percentile for their control group counterparts).
- Had 10% higher reading and math grade point averages in grades 1-3 than their control group counterparts (2.68 vs. 2.44).
- There were no significant effects on these children's conduct and behavior.
- For the whole sample of children (as opposed to the above subsample), there were no significant effects on academic performance (e.g. test scores, grade retentions, special education placements), or conduct or behavioral outcomes."
Coalition for Evidence-Based Policy, "Nurse-Family Partnership."
For the full sample, the 9-year follow up found (at least) the following statistically significant (at 5% significance level) effects on mothers:
"12% less time on welfare during the nine years (5.2 months per year for the nurse-visited women vs. 5.9 months per year for control group women). 10% less time on food stamps during the nine years (7.0 months per year vs. 7.8 months per year). 13% fewer subsequent live births (an average of 0.81 births vs. 0.93).... 18% more time with their current partner (an average of 61.6 months vs. 52.4 months)." Coalition for Evidence-Based Policy, "Nurse-Family Partnership."
Olds et al. 2004b, Table 4, Pg 1565.
The publication for the follow up when children were 4 years old is Olds et al. (2004b). When considering the full sample, it found the following statistically significant (at 5% significance level) effects on mothers:
- "nurse-visited women, compared with control subjects, had greater intervals between the births of their first and second children when a second birth occurred (24.51 vs 20.39 months, P = .01)." Effect size 0.32.
- "Nurse-visited women also reported less domestic violence from partners during the 6-month interval before the 4-year interview (6.9% vs 13.6% [had experience any domestic violence in the past six months], P = .05)" Odds ratio 0.47.
- "Nurse-visited mothers reported enrolling their children less frequently in preschool, Head Start, or licensed day care (P = .03)." (54.4% vs 65.9%) odds ratio 0.62.
Olds et al. 2004b, Pgs 1563 and 1565.
For more, see our blog post on publication bias.
"Nurse-Family Partnership is founded on the pioneering work of David Olds, professor of pediatrics, psychiatry, and preventive medicine at the University of Colorado Denver. While working in an inner-city day care center in the early 1970s, Olds was struck by the endemic risks and difficulties in the lives of low-income children. He realized the children needed help much earlier—at home, with their mothers, when they were infants, and even before they were born. Olds' determination to help young children and families get a better start in life led to the development of a nurse home visitation program for first-time, low-income moms and their children.
Over the next 30 years, he tested the program in randomized controlled trials with three different populations: Elmira, New York, in 1977; Memphis, Tennessee, in 1988; and Denver, Colorado, in 1994."
Nurse Family Partnership National Service Office, "Program History."
"Built upon the pioneering work of Professor David Olds, Nurse-Family Partnership's model is based on more than 30 years of evidence from randomized, controlled trials that prove it works. Beginning in the early 1970s, Dr. Olds initiated the development of a nurse home visitation program that targeted first-time mothers and their children. Over the next three decades, he and his colleagues continued to test the program in three separate, randomized, controlled trials with three different populations in Elmira, N.Y., Memphis, Tenn., and Denver, Colo." Nurse Family Partnership National Service Office, "A Cornerstone of Nurse-Family Partnership."
"Following is information about the average cost of a family in the NFP program to a community/agency and is approved for public dissemination:
Nurse-Family Partnership typically costs $4,500 per family per year with a range throughout the country of $2,914 - $6,463 per family per year based primarily on variations in costs of living." Michelle Stapleton, email to GiveWell, November 16, 2010.
$4,500 * 2.4 = $10,800
"The program costs approximately $11,200 per woman over the three years of visits (in 2008 dollars)." Coalition for Evidence-Based Policy, "Nurse-Family Partnership."