Abstract
Fathers of young children may face health challenges. In this article, the authors sought to characterize health-related challenges among fathers of young children (aged 4 years or younger) through the use of an Internet survey. Questions covered an unusually large battery of outcomes related to men’s health, including general health, affect, and sexual function. A total of 126 adult fathers provided usable data. The most notable findings were that scores on the General Health Questionnaire, Kessler Psychological Distress Scale, and negative affect indicated poor general health and high levels of distress and negative affect compared with normative data. Relationship satisfaction, sexual desire, and sexual activity were similar to normative data, although sexual enjoyment with and without a partner were lower among fathers participating in this study. Some variables were associated with each other (e.g., relationship satisfaction and affect) but most were not. This study also indicates that fathers can be successfully recruited through Internet-based survey research.
Introduction
Fathers of young children may face health challenges (Bartlett, 2004; Garfield, Clark-Kauffman, & Davis, 2006; Gray & Anderson, 2010). Men’s relationships with their wives may suffer, as attention shifts to some degree from the couple to caring for a needy child. Accordingly, in many cross-sectional and longitudinal studies, couples’ relationship quality tends to diminish after the birth of a child (Doss, Rhoades, Stanley, & Markman, 2009). To illustrate these effects, in a meta-analysis of 90 studies, largely drawn from North American and Western European samples, both fathers and mothers experienced lower marital satisfaction than nonparents, and these effects were not related to the age of their children though they were inversely related to socioeconomic status (Twenge, Campbell, & Foster, 2003).
Couples’ sexual function also declines. Although mothers’ physiology and attention shift toward offspring care at the expense of ongoing sexuality, several studies suggest that fathers also report lower libido within the year after having a child (Gray & Anderson, 2010). In a review of 59 sexuality studies, largely drawn from the United States and Western Europe, von Sydow (1999) reports, “Compared with the prepregnancy period, coital frequency is reduced in most couples during the first year after the birth” (p. 36). Moreover, a variety of factors contribute to lower rates of sexual behavior among parents (De Judicibus & McCabe, 2002). Marital quality is positively correlated with earlier resumption of sexual activity postpartum.
Other health parameters adjust while fathering young children. Sleep disruption and deprivation may contribute to mood and metabolic adjustments, although few quantitative data on men’s postpartum sleep are available to adequately characterize them (Germo, Chang, Keller, & Goldberg, 2007; Gray & Anderson, 2010). Men’s moods appear to change, both in positive (e.g., validation as a father) and negative (e.g., anxiety) directions. As an illustration, a study of Wisconsin men found that fathers exhibited more fatigue, irritability, headaches, difficulties concentrating, insomnia, nervousness, and restlessness, but no differences in backaches, depression, or colds compared with nonfather controls (Clinton, 1987, p. 65). Recent attention given to postpartum depression suggests that depression is fairly common. In a meta-analysis of 20 relevant studies, all from North America, Western Europe, or Australia, rates of postpartum male depression varied from 1% to 26% (Goodman, 2004). Between 24% and 50% of men reported postpartum depression if their partner also had postpartum depression, making this the strongest predictor of men’s postpartum depression. In the United States, a nationally representative study of approximately 5,000 U.S. families revealed that 10% of fathers with 9-month-old infants exhibited clinically significant rates of depressive symptomology; this rate was twice the expected rate among nonfather controls (Paulson, Dauber, & Leiferman, 2006).
Young children, particularly those in day care, can also serve as disease vectors (e.g., Hillis, Miranda, McCann, Bender, & Weigle, 1992), presenting a father with greater exposure to transmissible diseases. The outcome of factors such as these is that fathers of young children may have compromised health. Yet as those same children age, men’s health may rebound as the strains of parenting young children wane. Accordingly, both in the United States and Australia, younger biological fathers with younger children rated their health as worse than older fathers with older children (Bartlett, 2004; Hewitt, Baxter, & Western, 2006). A nascent literature on the physiology of fatherhood helps unravel some of the mechanisms through which fatherhood affects men’s cognition, behavior, and ultimately health (Gray & Anderson, 2010; Gray, Parkin, & Samms-Vaughan, 2007; Swain, Lorberbaum, Kose, & Strathearn, 2007). Among the relevant findings from brain imaging studies, men responded differently when presented with cries of their own infants compared with unrelated infants, and a structure activated by exposure to infant stimuli is the anterior cingulate cortex, which rests at the intersection between the emotional (limbic) and higher cognitive (neocortex) functions. Among the hormonal correlates of paternal care, lower testosterone levels have been commonly found, raising the possibility that these may be associated with men’s shift in social orientation toward children and also sleep disruptions, mood alterations, and decreased sexual function.
The objectives of the present study were (a) to better characterize health-related outcomes among a sample of fathers of young children; (b) to do so through the use of an Internet-based platform, to determine if this was feasible; and (c) to test for associations between variables, because some associations (e.g., a positive association between relationship quality and sexual function) would be expected on the basis of previous empirical research. We attempt to measure a wider battery of health-related variables than is customary in research on fatherhood (e.g., self-reported health, moods, relationship satisfaction, sexual function).
Method
Participants
An Internet survey was constructed that consisted of 147 questions drawn from multiple instruments as specified below. To ensure data quality, we used multiple methods to screen responses prior to data analysis. Multiple submissions from an identical Internet Protocol (IP) address were narrowed down to one sole response to eliminate connection/reconnection errors. Respondents who did not answer at least 70% of the questions in our survey were also not included in our data analysis.
Materials and Procedure
Questions covered six facets of fatherhood with an overall health orientation: paternal involvement, overall health, distress, affect, relationship satisfaction, and sexual function. Questions measured sociodemographic variables such as male age and education. Items concerning health-related behaviors (e.g., sleep, alcohol consumption, and exercise) and contextual factors (e.g., whether the couple had additional family support available) were also included. The questions were retrospective and the original instruments’ scales (which varied) were used. Percentages in the “Results” section do not cover the entire sample, but only fathers who answered that particular item.
Links to the survey were posted on fatherhood websites as well as academic online study sites (e.g., http://www.parentingscience.com/; http://www.hbes.com). Responses were collected over approximately a 17-month period from October 1, 2007, to May 31, 2009, and stored using a secure university server and 128-bit encryption. The study specified that fathers needed to have a child aged 4 years or younger to participate. This study was approved by the University of Nevada, Las Vegas Institutional Review Board and all participants were provided an electronic informed consent, which required the participant to click “Yes, I am at least 18 years old and agree to participate in this study” before entering the survey site.
Six individual items (i.e., gave encouragement and emotional support to their children’s mother, cooperated with their children’s mother, provided for their children’s basic needs, spent time with their child doing things the child likes to do, showed physical affection toward their children, and were involved in their children’s daily routine) assessing Paternal Involvement were employed to yield an overall measure (i.e., aggregate of the above six items gave the variable, Overall Paternal Involvement). The six items covered primary characteristics of parental involvement; support of spouse, child, and physical involvement in child’s life. The General Health Questionnaire (Goldberg & Williams, 1988) provides a global measure of self-reported health. The General Health Questionnaire consists of 12 questions that cover a range of psychological (worry, self-confidence), cognitive (concentration), and emotional (happiness) attributes and gives an aggregate score that can reveal severe problems or distress.
Psychological distress was measured using the Kessler Psychological Distress Scale (Kessler et al., 2002). The Kessler Scale is a 10-question screening scale. (We supplemented the original 10 with 5 follow-up questions.) Responses were retrospective and limited to experiences within the previous 30 days. The scale assesses self-reported feelings of fatigue, nervousness, hopelessness, restlessness, and worthlessness.
To assess affect, we employed the Positive and Negative Affect Schedule Short Form (PANAS; Thompson, 2007). The PANAS short form consists of 10 questions, 5 items each of the positive and negative subscales to measure the individual’s trait affect. The two subscales ask questions primarily concerning attention and mood.
To measure relationship quality, we used the Dyadic Satisfaction Subscale from the Dyadic Adjustment Scale (Spanier, 1976). This scale assesses the current level of physical and emotional intimacy, the perceived future of the relationship, and the overall quality of the two-person relationship. The scale consists of 10 questions (this experiment used only 9 of the original 10 questions because of a clerical error; we imputed an average of the 9 questions for the missing one).
To assess sexual function, we relied on items from the Psychosexual Daily Questionnaire (Lee et al., 2003). This self-reporting instrument uses 15 questions to assess sexual function within three domains: sexual desire, enjoyment, and performance; sexual activity; and mood. For this experiment, we used only the questions pertaining to the first two domains.
Raw data were calculated as mean (±SD). To test for associations between measures, we relied on Pearson’s correlation coefficient with alpha set to .05 and using two-tailed tests. For tests of effects of fatherhood on alcohol consumption and exercise, we relied on paired t tests.
Results
A total of 126 men completed the online survey. Demographic results are given in Table 1. The sample runs across a wide age range but is more highly educated than the general U.S. population.
Demographic Results (N = 126)
Respondents answered nine items concerning their child-raising duties and responsibilities. These data help provide context to the focal health-related outcomes. The majority (80.2%) of fathers did not have extra help, such as in-laws or a nanny, in raising their child. A small percentage of our respondents reported that each week they were responsible for one to two nighttime feedings (20%), and some were responsible for one to four nighttime diaper changes (30%). Concerning daytime duties, the majority (57%) responded that they performed between one and four feedings and diaper changes each week. A slight majority of dads had sole responsibility of their child less than 4 hours per day (52%) and less than 21 hours per week (56%). On average, dads claimed to cook 3.59 dinners per week, or, put another way, 65.5% of fathers cooked dinner at least 4 nights a week. Similarly, dads helped clean dishes after dinner 4.11 nights a week on average. They were also “burdened” with other nightly chores on an average of 4.97 nights a week.
Fathers’ health-related behaviors appeared to be affected by having young children. The majority (60.8%) of fathers reported less than 6 hours of uninterrupted sleep per night with 57% of dads being awakened by their children (10% were awakened 3 or more times per night). Overall, fathers reported that the average number of drinks consumed per night decreased from 1.98 before having children to 1.44 postpartum; this difference represented a statistical trend (t = 1.807, p = .077). More than 80% of our sample reported having at least 3 drinks a night with a minority of 15% having between 4 and 10 drinks a night prior to the birth of their child. After the birth of their child, 94% of our sample reported having at least 3 drinks a night, but the heavy drinkers (4 to 6 drinks) dropped to 5%. As alcohol consumption tended to decrease after the birth of the child, men also reported that physical exercise decreased from a mean of 7.34 hours per week before children to 5.02 hours per week after having children (t = 3.52, p < .001).
Table 2 contains descriptive data for the focal health-related variables assessed in this study: paternal involvement, overall health, distress, affect, relationship satisfaction, and sexual function. Also reported in the table are descriptive statistics for specific items that comprise the overall measure of paternal involvement, distinct scores for positive and negative affect, and individual sexual function items.
Descriptive Statistics for Primary Health-Related Outcomes
For paternal involvement items, responses were on a 5-point Likert-type scale ranging from 1 = never to 5 = always.
Psychosexual Daily Questionnaire used a 7-point Likert-type scale ranging from 1 = none to 7 = very high.
Many of these fathers’ scores differ from normative data. For the General Health Questionnaire, population scores of 11 to 12 are typical, scores >15 show signs of distress, and scores >20 suggest severe problems. Among the fathers who responded to this Internet survey, 21.2% were in the distressed category, and 74.1% were in the severely distressed category. For the Kessler Distress Scale, scores of 10 to 19 are considered low psychological distress, scores of 20 to 24 are considered mild psychological distress, scores of 25 to 29 are considered moderate psychological distress, and scores of 30 to 50 are considered severe psychological distress. Among respondents in our Internet study, only 1 (1%) qualified as having low distress, whereas 38 (32.8%) were mildly distressed, 53 (45.7%) were moderately distressed, and 24 (20.7%) were severely distressed. Normative U.S. data for the positive affect scores were as follows: M = 19.73 (SD = 2.58) for positive affect and M = 11.27 (SD = 2.66) for negative affect. The average scores among fathers in our Internet study were comparable for positive affect, but were notably higher for negative affect.
Dyadic satisfaction scores for a normative U.S. population were as follows: M = 35.0 (SD = 11.8). Thus, relationship satisfaction scores in our Internet study were similar to these normative data. Average weekly sexual function scores from two samples of normal men aged 19 to 50 were sexual desire of M = 4.28 (SE = 0.19) and 4.59 (SE = 0.20) in the two normative samples, respectively; sexual enjoyment without a partner (M = 3.34, SE = 0.29; M = 3.17, SE = 0.35); sexual enjoyment with a partner (M = 3.05, SE = 0.67; M = 3.78, SE = 0.39); and sexual activity scores (M = 5.51, 5.07; Lee et al., 2003). These data indicate that the men in our Internet study expressed fairly comparable sexual desire and sexual activity, but lower sexual enjoyment without a partner and lower sexual enjoyment with a partner compared with normative data. Also with respect to sexual behavior, half of our respondents (58.5%) reported having begun sexual relations within 2 months after the birth of their child. More than half (57%) of our dads reported having sexual relations once a week, which was the most frequent response as well.
To test for associations between the primary health-related variables, bivariate correlations among them are given in Table 3. As indicated in Table 3, there were some significant correlations between variables. For example, dyadic satisfaction was significantly correlated with positive (r = .407) and negative (r = .388) affect. The psychological distress scale was significantly and negatively correlated with positive (r = −.270) and negative (r = −.304) affect. However, many variables were not associated with each other; for example, the general health score was not significantly correlated with any of the other measures.
Correlations Between Main Health-Related Measures
p < .05. **p < .01.
Separate partial correlations were performed on the variables in Table 3 (i.e., paternal involvement, general health, Kessler Distress Scale, positive affect and negative affect, relationship satisfaction, sexual desire, sexual enjoyment without partner and with partner, and sexual activity) with age, education level, personal income, and marital status individually partialed out. No significant changes in the correlation results presented in Table 3 were found when taking into account these demographic variables.
Discussion
Research into the impact of fatherhood on men’s health continues to grow. This is important because most postpartum parental health focus has been on mothers, and yet mothers and fathers experience a different set of postpartum challenges (Fagerskiold, 2008). The present study contributes to an understanding of health and fatherhood in several respects. First, by several measures, fathers in the present study appear to be faring poorly. They exhibited extremely poor scores on the General Health Questionnaire and Kessler Distress Scale as well as high negative affect compared with normative data. Second, although some of the health variables measured in the present study were associated with each other (e.g., relationship satisfaction with affect), most of the associations were not statistically significant. We consider possible reasons for these largely null associations below because some might be surprising in light of earlier findings. Third, the present study indicates the feasibility of conducting research on health and fatherhood through an Internet study. A total of 126 men provided usable data during the 17 months of Internet study recruitment.
The central findings were that scores on the General Health Questionnaire, on the Kessler Distress Scale, and on the negative affect scale indicated remarkably poor general health and high levels of distress and negative affect. Although researchers have devoted considerable attention to maternal postpartum distress and affect, less attention has been given to the mental health of postpartum men. Elevated levels of postpartum depression among fathers have been noted in recent years, and the present findings contribute to a growing recognition of the mental health challenges that fathers may face.
Several health-related behaviors appear to be affected by fatherhood, potentially contributing to the mental health challenges identified in the present study. Fathers reported poor sleep patterns and lower exercise rates (consistent with a handful of recent studies on the effects of parenting on activity levels; Bellows-Riecken & Rhodes, 2008), and also a trend toward lower rates of alcohol consumption after having children. Interestingly, relationship satisfaction scores were similar to normative data, suggesting that this element of respondents’ relationship may have been less dramatically affected than what previous research might have anticipated. Sexual desire and sexual activity were similar to normative data, but levels of sexual enjoyment without a partner (e.g., masturbation) and with a partner (e.g., with a man’s wife) were markedly lower among men in this study of fathers of young children.
Previous research has highlighted some of the challenges posed by parenting, but most of that scholarship tended to isolate either a single health-related mental health outcome (e.g., depression) or contributed to a very large literature highlighting reduced marital quality or satisfaction. As one of the most integrative U.S. longitudinal studies on the effects of parenting on men, women, and relationships, Cowan and Cowan (1992) documented from interviews and questionnaire outcomes the challenges negotiating sleep, conflicts over household tasks, reduced marital quality, and decreases in sexual function in the wake of having an infant or toddler. As children command parents’ attention, this means less time for parents to share emotional or sexual intimacy (Cowan & Cowan, 1992). Combined with other studies, finding that fathers have less time to spend in leisure activities with other men or beneficial behaviors such as exercise and sleep, the fact that fathers of young children might fare poorly on the measures assessed in the present study is not surprising (Gray & Anderson, 2010).
Why do men then become fathers, or even repeat the experience, given this pattern of health-related challenges? Men commonly state that their lives are enriched or more validated by becoming fathers (Gray & Anderson, 2010). The immediate health-related challenges may be outweighed by the legacy they leave, also consistent with an evolutionary perspective in which reproductive success, not health, serves as the ultimate bottom line (Gray & Anderson, 2010). Furthermore, those health-related challenges may also be relaxed as the children grow older, as studies both in the United States and Australia have found (e.g., Bartlett, 2004).
The ambivalence men may face as fathers of young children appears to be manifest in some of the correlations between health-related measures investigated in the present study. Positive and negative affect were significantly, and positively, correlated with each other. Negative affect was also positively correlated with relationship satisfaction. One possible interpretation of these associations is that many features of men’s emotional lives and family relationships are exaggerated postpartum, resulting in both the experience of considerable negative affect alongside positive affect and relationship satisfaction. A recent review of men’s postpartum psychological adjustment highlighted these kinds of mixed emotions (Genesoni & Tallandini, 2009), suggesting that positive and negative dimensions of fatherhood are closely intertwined.
The results of this study demonstrate the feasibility of Internet-based research with fathers. As our demographics show, we were able to capture a very diverse sample in terms of age, education level, and income, although the education attainment of participants tended to be relatively high. We were able to target a population (postpartum fathers) that is normally very difficult and costly to collect data on. Furthermore, men responded to items on some sensitive topics such as relationship satisfaction, mental health, and sexual function, perhaps in part because of the anonymous nature of this Internet survey. The feasibility of reaching fathers through Internet-based means may be important for future research as well as for parenting outreach programs (Fletcher, Vimpani, Russell, & Keatinge, 2008).
In tests of associations between variables, some were significantly associated with each other, but most were not. Some of these observed associations were consistent with expectations based on previous research. As examples, relationship satisfaction and affect were correlated, as were relationship satisfaction and sexual enjoyment with a partner. Several reasons may account for the null associations generally observed, including cases (e.g., paternal involvement and sexual activity) where relationships might have been expected. These Internet-based data may be “noisier” than data collected through more targeted face-to-face surveys; the questions could have been better selected to measure some of the desired constructs; and the lower scores on many of the outcomes narrowed the ranges with which to test for correlations. Additional post hoc tests for contextual factors (male education, male age, length of relationship with child’s mother, whether other help was available to care for children) revealed no notable associations with the core health-related variables under focus in this study.
In conclusion, results of this Internet study revealed that (a) fathers exhibited poor scores on the General Health Questionnaire, Kessler Distress Scale, and high negative affect compared with normative data; (b) sleep, exercise, and alcohol consumption patterns appeared to be influenced by fatherhood; (c) fathers reported comparable relationship satisfaction, sexual desire, and sexual activity, but markedly lower sexual enjoyment without and with a partner compared with normative data, indicative of some alterations in the dynamics of their sociosexual relationships; and (d) fathers can be successfully recruited through an Internet study. Future research on fathers’ mental health could benefit by expanding the cross-cultural scope of such investigations, continuing to specify the physiological substrates underlying these outcomes, and refining the instruments employed in such studies.
Footnotes
The author(s) declared no potential conflicts of interests with respect to the authorship and/or publication of this article.
The author(s) received no financial support for the research and/or authorship of this article.