Objective. Fathers make important contributions to many aspects of children’s well-being, but relatively few studies have evaluated father involvement in their child’s health care. The objective of this study was to explore the extent to which fathers are involved in their children’s health care and identify those factors that are associated with greater involvement.
Methods. A convenience sample of 104 English-speaking, urban fathers with children younger than 7 years were recruited to complete an anonymous, structured interview. Fathers self-reported the number of well-child visits (WCVs) that they had ever attended and which factors had influenced their attendance. Bivariate and multivariate analyses were used to identify those factors that predicted greater involvement.
Results. Eighty-nine percent of the fathers had attended at least 1 WCV. Fifty-three percent had high involvement, ie, had attended ≥40% of the American Academy of Pediatrics recommended visits for their child’s age. In multivariate modeling, factors that were significantly associated with high involvement in attending WCVs included attendance at the child’s delivery (odds ratio [OR]: 7.3 [1.7–30.4]), younger child age (OR: 0.96 [0.94–0.99]), older father age (OR: 1.2 [1.2–1.3]), the child’s having health insurance (OR: 4.1 [1.3–12.0]), and having >1 child (OR: 0.22 [0.06–0.72]).
Conclusions. The factors identified suggest ways that pediatric providers can support fathers’ involvement in their children’s health care. Providers should focus on encouraging greater involvement early, especially for younger fathers and those with older children. In addition, support of universal health coverage for children might, in addition to other obvious benefits, enhance a father’s engagement with his child’s health care.
Despite previously held beliefs that fathers were important to children only as economic providers and disciplinarians,1 fathers have been increasingly shown to be important contributors to the overall development and well-being of their children. Various studies have shown higher father involvement to be associated with improved receptive language skills,2–4 higher cognitive skills5,6 higher infant weight gain for preterm infants,7 improved social and adaptive behavior,7 higher academic achievement,8,9 fewer behavioral problems,1,2,3,8 and a lower likelihood of neglect.10 The definition of father involvement ranges from participation in caregiving tasks to the demonstration of nurturance. Factors associated with greater involvement have included younger child age, firm control in child-rearing practices (ie, setting and monitoring limits), maternal employment, paternal education, and paternal employment.11–13
Although researchers14,15 have included health and medical issues as part of their framework for thinking about father involvement, to date, little research has evaluated the extent of a father’s involvement in his child’s preventive health care and which factors influence that involvement. Among intact, middle-class, white families, fathers were more involved in taking the child to the doctor or the dentist when the mother was employed.16 In urban London, researchers found fathers to be 2.3 times more likely to bring their children to a health clinic visit during the evening (45%) compared with daytime hours (20%).17 To our knowledge, these are the only 2 studies that address potential factors that influence fathers’ involvement in their children’s health care in the outpatient setting, and neither describes the barriers to such involvement that fathers may encounter.
Another limitation of the current literature is that relatively few studies use father’s self-report of involvement.10 Most studies use the maternal report of the father’s behavior,3,5,8,18–20 which many researchers believe may not always accurately reflect the father’s behavior. Although 1 study suggested that a mother’s report of a father’s behavior is relatively reliable, absolute agreement between mother’s and father’s report was not established.21 Furthermore, there is no evidence to support the reliability of a mother’s report about a father’s perceptions.
This research study sought to explore the extent of fathers’ involvement in their children’s health care (specifically focusing on their attendance at well-child visits [WCVs]), the primary motivators and barriers to their involvement, and which factors are associated with higher versus lower involvement.
During an 8-month study period, a trained research assistant conducted 20-minute, structured interviews with a convenience sample of 108 English-speaking men who had children younger than 7 years. Fathers with children younger than 7 years were chosen as the target group because there are a clustering of WCVs during the infancy and early preschool years. Fathers were recruited from 3 locations: an urban hospital’s pediatric and adult primary care practices serving a large number of urban, low-income families of culturally diverse backgrounds; community health centers, serving similar populations; and selected nonmedical settings (eg, recreation facilities, community events) in the surrounding urban neighborhoods served by the hospital.
In the hospital practices and community health centers, men were approached directly and asked whether they would agree to participate. At the community sites, trained interviewers were placed at tables with a poster requesting participation in a survey of fathers who had children younger than 7 years. Alternatively, the interviewer approached men directly and asked whether they would like to participate. When men agreed to participate, verbal informed consent was obtained for anonymous participation. The protocol was approved by the Boston University Medical Center Institutional Review Board. Participants received a $10 gift certificate to a local merchant after completion of the interview questionnaire.
The designation of father was determined by the participant. Of the 276 men approached, 195 were fathers with children younger than 7 years. Thirty-five of the fathers were not eligible because they did not meet the inclusion criteria of being fluent in English. Of the remaining 160 fathers, 108 agreed to participate, resulting in a 68% response rate for the aggregate. The response rate for nonhealth, community sites was slightly lower (63%) than that for the aggregate. Four of the surveys were incomplete and were not analyzed, for a total of 104 surveys analyzed. Seventy-five percent of the fathers were recruited at the hospital’s primary care sites, 17% were recruited from community sites, and 8% were recruited from community health centers.
Fathers were asked whether they had ever attended a WCV and how many they had attended. They were also asked ∼22 possible barriers to their attendance at WCVs. The barriers focused on factors related to work, the child’s mother, finances, the pediatric office practice, personal interest in their child, family encouragement or support, beliefs about self-competency, and culture. Fathers who reported having attended a WCV were also asked ∼18 possible motivators for their attendance at WCVs. The motivators focused on similar areas as the list of barriers. Motivator and barrier questions were derived from a review of the literature. Additional items included were derived from focus group discussions with fathers who were not included in the interview cohort. The focus group was part of a father support group at the urban hospital where study participants were recruited. Demographic variables and information about fathering behaviors (presence at child’s delivery, currently living with child, and financial support of the child) were also obtained from the father (Table 1). Race was self-identified by the father. When fathers had >1 child younger than 7 years, a single index child was randomly chosen as the focus for questions about attendance at WCVs. The first names of the children were elicited, and the child whose name began with the earliest letter in the alphabet was designated the index child.
Fathers were divided into 2 groups on the basis of the percentage of the American Academy of Pediatrics (AAP) recommended number of WCVs (of the total possible for their age) that they had attended.22 Because the index children were of different ages, percentage of recommended visits were used rather than absolute number of visits. A natural split was observed (in the distribution of percentage of visits attended) at ∼40%. High attenders were categorized as those fathers who had attended 40% or more of the AAP recommended number of visits (based on their child’s age over their child’s lifetime). Low attenders were categorized as those who had attended <40% of the recommended number of visits.
For descriptive purposes, we computed simple frequencies of barriers and motivators. All motivator and barrier questions were initially asked using a 4-point Likert scale but were subsequently collapsed into yes/no responses for analyses. After dividing the sample into the high and low attender groups, demographic variables, fathering behaviors, barriers, and motivators were assessed as possible factors to distinguish among these groups. The χ2 test for categorical variables was used in the bivariate analysis. Fisher exact test was used when appropriate. Multivariate analysis was then used to determine the best set of independent predictors of being a high attender. Variables that were found to have a significance level of ≤.10 in the bivariate analysis were entered into the multivariate model. The covariates were marital status, education, number of children, race, father’s age, age at which the father had his first child, child’s age, presence in the delivery room, living with child, the child’s having health insurance, feeling respected by medical staff, having vacation/sick time, and mother’s role to take child. Stepwise selection with α set at ≤.10 was used to enter and remove terms from the logistic regression model. Because of missing variables, the model included only 92 respondents. Eleven of the respondents had missing values because they had never attended a WCV and thus did not answer the motivator questions (an a priori skip pattern established in the survey). The fit of the final model, predicting high versus low attenders among fathers who had ever attended a WCV, controlling for covariates, was assessed by the Hosmer-Lemeshow statistic, the pseudo-R2, and the overall significance. The SAS software package (version 8e) was used to conduct the analysis (SAS Institute Inc, Cary, NC).
Characteristics of Respondents
Table 1 summarizes the characteristics of the respondents. The mean and median ages of the fathers surveyed were 34.7 years and 35.0 years (range: 22–60 years), respectively. Forty-eight percent had a high school education, and 44% had completed some college or higher. The overwhelming majority were employed at the time of the survey. Most fathers lived with their child, but 32% did not. The mean and median ages at which the father had his first child were 26.1 years and 25.0 years (range: 13–42 years), respectively. The majority of fathers were black (86%) and included respondents who identified themselves as African American, Caribbean/West Indian, and African. Fifty-four percent of the fathers reported their child’s health insurance to be Medicaid. Nearly all of the fathers reported giving financial support to their child (98%; data not shown). The mean and median ages of the index child were 39.5 months and 38.5 months, respectively.
Attendance at WCVs
Eighty-nine percent of fathers reported attending at least 1 WCV for the index child. Although not statistically significant, a higher percentage of fathers who were recruited from a health site reported attending at least 1 WCV as compared with fathers who were recruited from a nonhealth site (91% vs 82%, respectively). The mean and median number of visits attended by all fathers was 4 and 3, respectively (range: 0–14). Among fathers who had attended a visit, nearly one half reported attending all visits with another adult. Forty-nine fathers were high attenders, ie, had attended at least 40% of the recommended AAP visits. On average, fathers attended 55% of the recommended WCVs (based on the child’s age) for children 2 weeks to 24 months of age, 40% of the recommended WCVs for children 25 to 48 months of age, and 30% of the recommended WCVs for children older than 48 months. There were no significant differences between the mean number (P = .7) or mean percentage (P = .3) of visits attended among fathers who were recruited at the health site (4.0 [42%]) versus the nonhealth site (3.7 [35%]).
Barriers and Motivators
Tables 2 and 3 list the barriers and motivators identified by fathers in order of decreasing frequency within a category. Work-related reasons were the most commonly reported barriers, as indicated by 46% of fathers. For example, approximately one quarter of fathers indicated that their boss would not give them time off or that they did not have flexible working hours. Twenty-three percent indicated a reason related to the child’s mother, such as the child lives with the mother and it is more convenient for her to take the child to a WCV. Nearly one fifth indicated that factors related to the pediatric office were barriers, such as a long wait time before being seen. Overall, only a few of the barriers were highly endorsed.
Fathers more positively endorsed many of the motivators. Virtually all of the fathers reported that personal interest in their child’s life, health, or development as motivators. Three quarters of fathers indicated a family-related reason as a motivator, such as a family member’s encouraging them. Sixty percent of fathers indicated a work-related reason as a motivator in helping them to attend WCVs, such as their boss’s being supportive or having vacation or sick time or flexible working hours.
Factors Associated With Being a High Attender
Table 4 summarizes factors that were associated with being a high attender at WCVs in bivariate analyses. Fathers were more likely to be high attenders when they had more than a high school education, had younger children, or were of nonblack race. As a trend, fathers who had >1 child were less likely to be high attenders (P = .10). There were no differences based on income, employment status, child’s gender, or child’s health insurance type (all P > .10; data not shown). Fathering behaviors were associated with being high attenders. Fathers who attended the delivery of their child or lived with their child were more likely to be high attenders. Among the fathers who did not live with their child (31%), those who had contact with their child 3 or more times per week were more likely to be high attenders (P = .02; data not shown).
Among the list of motivators and barriers, only a few were highly associated with fathers’ being high attenders at WCVs. Fathers who identified their child as having health insurance or feeling respected by pediatric office staff as motivators were more likely to be high attenders. As a trend, having vacation or sick time was associated with being a high attender (P = .08), whereas believing that it was the mother’s role to accompany the child to a WCV was associated with being a low attender (P = .10). There were no other significant differences found among high and low attenders for the remaining motivators and barriers at the ≤.10 significance level (data not shown).
Of the 13 factors significant in bivariate analyses, only 5 of them remained statistically significant in the multivariate model (Table 5). Significant predictors in the final model included 1) presence in the delivery room, 2) the index child’s having health insurance, 3) younger child age, 4) older father age, and 5) having >1 child. We further explored the multivariate model using only black respondents and obtained similar findings.
The overwhelming majority of urban fathers in our sample have attended at least 1 WCV, yet those who have ever attended did so at relatively low rates. Slightly fewer than half have attended >40% of the recommended AAP WCVs (based on their child’s age). These are missed opportunities for fathers. The WCV provides a unique opportunity for parents to ask questions about their child’s growth, development, and health and seek advice about parenting issues.23 The need for both parents to attend WCVs increases as more mothers enter the workforce and fathers share more of the day-to-day caregiving responsibilities for their children.24,25
The percentage of visits attended drops substantially for older children in comparison with younger children. The finding of a lower level of father involvement at older child ages is similar to findings in other studies.12,13 In this study, fathers who had >1 child were also less likely to be high attenders. A sense of wonderment or newness that many first-time fathers experience may lead to greater attendance at WCVs initially. This feeling may fade over time, resulting in lower attendance rates, especially as the child becomes older. Another explanation may be that fathers are needed at home to care for other children while the mother is taking 1 child to a WCV, or perhaps fathers were previously discouraged from coming to visits with other children. Enabling both parents to attend may require pediatric practices to modify their approach to families with multiple children and offer multiple well-child appointments to a family on a given day.
The positive motivators to attendance seemed most salient to the father, as a large percentage of them endorsed multiple motivators. Reasons related to family, work, the pediatric office practice, finances, and provider encouragement were identified as motivators, yet only a select few of the motivators seemed to have a differential influence on their attendance. In the bivariate analyses, fathers who identified their child as having health insurance or feeling respected by medical staff as motivators were more likely to be high attenders; however, only health insurance as a motivator remained significant in the multivariate analysis, controlling for covariates. Thus, fathers may feel less embarrassed to attend visits and that they garner more respect by office staff when their children have insurance. The unexpected association between a child’s having health insurance and a father’s being a high attender at WCVs cannot be fully explained by this data set but does suggest an area worthy of additional investigation.
One third of the fathers reported that encouragement by their child’s doctor to attend a visit was a motivator. Although this was not associated with being a high attender, it did reveal that only a small percentage of providers are successfully encouraging fathers to attend the WCVs. What is not known is whether only one third of the providers encouraged the fathers or more of the providers encouraged the fathers and only one third of them viewed the encouragement as a motivator. Would this low rate be viewed as acceptable if the discussion were about encouragement of mothers?
Being involved with your child at birth (as measured by delivery-room presence) was highly associated with being a high attender, suggesting that early involvement in a child’s life is associated with later involvement, specifically in attending WCVs. Although delivery-room attendance may simply reflect the a priori decision of the father to be involved with his child, it may well be an especially powerful time to support father involvement.
This study suggests several ways in which pediatric providers can foster greater involvement of fathers in their children’s health care. Overall, more providers should encourage men to be involved. An opportune time to do this may be in the newborn nursery with follow-up at later contacts. In the nursery, efforts should particularly be targeted to younger fathers (<25 years of age), who, these data suggest, are less likely to be involved in WCVs. In subsequent visits, fathers should be encouraged to remain involved beyond infancy, because fathers of older children in this sample were less likely to attend visits. In addition, informing mothers and fathers of the important benefits of father involvement to their child’s development is critical.
Pediatric providers should support national and local policies that promote father involvement, such as universal health insurance coverage for children and more family-friendly employment policies. In addition to the medical benefits that health insurance provides for children, it strongly motivates fathers to be involved in their child’s health care. Work-related reasons were the most common barriers to attending WCVs. Getting time off from work (having vacation or sick time) was particularly influential in bivariate analysis, although when controlling for related covariates (eg, the child’s having health insurance), it was less influential. However, it is an area that can be addressed on a policy as well as an individual level. The federal Family Medical Leave Act26 allows individuals to take up to 12 weeks off to care for family members with a serious medical condition without fear of losing their job. It does not, however, include taking time off from work to meet the needs of family members regarding preventive health care. There is a little-known law in Massachusetts that does addresses this issue. The Small Necessities Leave Act (modeled after the Family Medical Leave Act and applies to companies with 50 or more employees) allows employees to have 24 hours of unpaid time off per year to accompany an elderly person or a child to a routine medical or dental appointment or to participate in a child’s school-related activities. In the state of Massachusetts, where this study was conducted, providers should inform fathers of this law. Nationally, providers can support the creation of such a law in other states as a means of promoting father involvement in their child’s health care.
This is an exploratory study using a convenience sample of urban, primarily minority fathers, and the findings may not generalize to all fathers of other backgrounds. However, it did reveal issues that have important clinical and policy implications for all fathers and pediatric practice. Selection bias could have influenced the findings, because a majority of fathers were recruited from a health care site and may, therefore, be more child health care oriented and be more likely to attend WCVs regularly. However, the data do not bear this out, because there was no statistically significant difference between the mean number of visits or the mean percentage of WCVs attended among fathers who were recruited from a health site in comparison with those who were recruited from a nonhealth site, and roughly one half the fathers in both groups had attended <40% of the visits based on their child’s age. We believe that this is an important and valuable sample as an initial first study in understanding how to get potentially higher motivated fathers to attend WCVs more regularly. Future studies many want to look at more diverse samples of fathers to enhance generalizability and to identify barriers and motivators for potentially less motivated fathers—those who have never attended a WCV.
Another limitation of this study, which is a simultaneous strength, is the use of father’s self-report. Much of the father involvement literature relies on a mother’s report of a father’s behavior, which may lead to underreporting of involvement. However, the accuracy of the father’s report of the number of visits attended cannot be confirmed. Fathers who viewed attendance at WCVs as a positive attribute may have tended to overreport their attendance. Recall bias may have been introduced, as fathers with older children may not accurately recall the number of visits attended because they have a longer time period to recall in comparison with fathers with younger children. We attempted to minimize recall bias by limiting the study to fathers with relatively younger children. Finally, we may not have adequately assessed all potential motivators and barriers as they were based on single questionnaire items that were derived from focus groups and existing literature.
Contrary to widely held stereotypes, most urban fathers in this sample were involved in attending at least some of their children’s WCVs, suggesting the potential to enhance fathers’ participation in children’s health care. To our knowledge, this is the first study to evaluate factors that influence fathers’ self-reported involvement in attending WCVs and which factors influence involvement. The factors identified in this study can help to guide pediatric providers in supporting greater father involvement. Notably, a special emphasis should be placed on fathers who are less likely to attend WCVs—younger fathers and fathers with older children. Because health insurance and work factors were other key issues identified by fathers as influencing their attendance at WCVs, attention to national and local policies that support universal health coverage for children and allows parents of either gender to take time off from work to attend WCVs is important. These findings provide a beginning framework for pediatric providers to enhance fathers’ participation in children’s preventive health care and their overall development.
This study was supported by HRSA MCHB 5T77MC00015-10, MCHB T76MC00017, and the Joel and Barbara Alpert Endowment for Children of the City Grant.
We thank Dr Howard Cabral for statistical support.
- Received May 19, 2003.
- Accepted August 11, 2003.
- Reprint requests to (T.M.) St Louis University School of Medicine, Cardinal Glennon Children’s Hospital, 1465 S Grand Blvd, St Louis, MO 63104-1095. E-mail:
- ↵Amato PR. More than money? Men’s contributions to their children’s lives. In: Booth A, Crouter AC, eds. Men in Families. When Do They Get Involved? What Difference Does It Make? Mahwah, NJ: Lawrence Erlbaum Associates; 1998:241–277
- ↵Radin N. Primary caregiving and role-sharing fathers. In: Lamb ME, ed. Nontraditional Families: Parenting and Child Development. Hillsdale, NJ: Lawrence Erlbaum Associates; 1982:173–204
- ↵Levy-Shiff R, Hoffman MA, Mogilner S, Levinger S, Mogilner MB. Fathers’ hospital visits to their preterm infants as a predictor of father-infant relationship and infant development. Pediatrics.1990;86 :289– 293
- ↵McLanahan SS, Seltzer JA, Hanson TL, Thomson E. Child support enforcement and child well-being: greater security or greater conflict? In: Garkinfel I, McLanahan SS, Robins PK, eds. Child Support and Well-Being. Washington, DC: The Urban Institute Press; 1995:239–256
- ↵Lamb ME, Pleck JH, Charnov EL, Levine JA. Paternal behavior in humans. Am Zool.1985;25 :883– 894
- ↵Gadsen V, Fagan J, Ray A, Davis JE. The Fathering Indicators Framework: A Tool for Quantitative and Qualitative Analysis. Philadelphia, PA: National Center on Fathers and Families; 2001
- ↵Bailey WT. Fathers’ involvement in their children’s healthcare. J Genet Psychol.1990;152 :289– 293
- ↵Murray A, Sandquist K. Father absence and children’s achievement from age 13–21. Scand J Educ Res.1990;34 :3– 28
- ↵Committee on Practice and Ambulatory Medicine. Recommendations for Preventive Pediatric Health Care (RE9939). Available at: www.aap.org/policy/re9939.html. Accessed June 5, 2001
- ↵Family Medical Leave Act. Pub L No. 103-3 (1993)
- Copyright © 2004 by the American Academy of Pediatrics