We used a daily diary approach to test the linkages between knowledge and youth physical symptoms and HPA functioning. Most studies of parental knowledge rely on global reports, whereby parents and/or youth rate parents’ average level of knowledge, usually over the past month (Lippold et al., 2014; Racz & McMahon, 2011). Although such global reports have been linked to youth outcomes, they can be limited by response biases and memory demands. Further, both knowledge and youth health may fluctuate on a day-to-day basis (Almeida, 2005). That is, parents may be more knowledgeable and youth may experience more health symptoms on some days than on others. Diary data can be used to investigate two types of associations. Between-person associations test whether or not average levels of knowledge are linked to better youth health outcomes. Within-person associations test whether, on days parents are more knowledgeable than usual (i.e., relative to their own cross-day average), youth have more positive health outcomes than usual that same day. Such within-person analyses control for stable between-person differences, or selection effects, and thereby allow for stronger inferences about the links between knowledge and health. Thus, in this study we used multi-level modeling (MLM) to analyze links between average levels of parental knowledge (between-person associations) and youth health, as well as to assess whether day-to-day variations in parents’ knowledge were linked to same day variations in youth health (within-person associations).

As noted, we assessed youth's physical health in terms of both health symptoms and HPA functioning. We used salivary cortisol as a biomarker of youth's stress-related physiology. Healthful patterns of cortisol—the hormonal byproduct of the HPA system—are marked by a diurnal rhythm wherein cortisol levels peak shortly after waking and then decline over the course of the day (McEwen, 1998). To the extent that parental knowledge is linked to better HPA functioning, the cortisol levels of youth with more knowledgeable parents should exhibit a healthful pattern of rising sharply in the morning, and then declining across the course of the day, resulting in low levels of cortisol before dinner and at bedtime. Dysregulated HPA functioning has been linked to disrupted immune processes and an increased risk of disease (Miller, Chen, & Parker, 2011; Neeck, Federlin, Graef., Rusch Schmidt, 1990; Repetti, Robles, & Reynolds, 2011). However, the causal process that may link HPA functioning and illness and disease are not yet clear (Miller et al., 2011). Thus, we assessed both physical health symptoms and salivary cortisol in this study.

Our study is grounded in prior research and theory suggesting that parental knowledge may have important linkages to health behaviors and youth stress, with positive implications for their HPA functioning and fewer physical health symptoms. In this study, we assessed the direct associations between parental knowledge and indices of HPA functioning and youth physical health.

Parental knowledge may be linked to youth stress-related physiology and physical health symptoms and at both daily and average levels. If parents are knowledgeable about their youth experiences on a particular day, they may be more likely to be aware of and promote behaviors and routines that improve physical health and reduce youth stress-- such as exercising, sleeping, and eating well (Buckley & Schatzberg, 2005; Cohen, 2004; Meerlo, Koehl, van der Borght, & Turek, 2002; Pedersen & Hoffman-Goetz, 2000; Traustadóttir, Bosch, & Matt, 2005). In addition, knowledgeable parents may be more aware of their children's physical well-being and thus more likely to intervene in ways that prevent illness and promote more rapid recovery when their children do become ill. For example, parents may limit their children's exposure to friends who are ill or provide medication and arrange doctors’ visits. Some of the linkages between knowledge and youth health may occur on the same day: For example, parents who know that a child is not feeling well may encourage more restful activities, resulting in fewer physical symptoms and healthier HPA functioning on that same day. Yet, some of the associations between parental knowledge and youth health may be cumulative, and it may be that average parental knowledge, rather than knowledge on particular day, is most closely linked to physical health and HPA functioning. For example, the effects of parenting practices such as enforcing curfews and bedtimes, encouraging a balanced diet, and promoting physical activity may not be immediate, but build up over time (Cohen, 2004; Umberson, Crosnoe, & Reczek, 2010; Kitzman-Ulrich et al., 2010). Therefore, average levels of knowledge may also be important for promoting healthy HPA functioning and physical health.

Furthermore, when their parents are more knowledgeable about their activities, whereabouts and companions, youth may develop a sense of safety and security, which in turn, buffers youth against the effects of stressors (Cummings & Davies, 1996). Youth with knowledgeable parents may feel more comfortable communicating with their parents and therefore may experience greater parental support and advice. Social support has been linked to improved physical health and HPA functioning among adults (Lutgendorf et al., 2005; Turner-Cobb, Sephton, Koopman, Blake-Mortimer, & Spiegel, 2000), as well as fewer physiological responses when faced with a stressor (Uchino, 2006). Similarly, because of improved communication and problem solving support, youth with knowledgeable parents may experience fewer stressors and may appraise stressors are less threatening with weaker physiological responses to those stressors (Bai & Repetti, 2015; Power, 2004). Therefore, on a particular day, youth with parents who are knowledgeable may experience fewer physical symptoms and healthier cortisol patterns that same day. However, the effects of parental knowledge on youth health may also be cumulative. It may be that only when parents have high knowledge across many days that youth may develop a strong sense of safety and support. Thus, average levels of knowledge may also have associations with youth physical symptoms and HPA functioning.

The linkages between parental knowledge and health may also be due to child effects: Children's health may influence the level of knowledge that their parents obtain. Parental knowledge is a transactional process, with parents providing support and seeking information and youth making decisions about what information to share with parents (Racz & McMahon, 2011). Thus, youth with better health and HPA functioning may be more likely to share information with their parents, resulting in more knowledge (Stattin & Kerr, 2000). In contrast, those who experience stressors and physical symptoms may withdraw and fail to disclose information to their parents. Parents may also track the activities of their children who are ill more than children who are healthy. Perceptions also may matter: Youth with better health may perceive their parents as more engaged and helpful, leading to greater disclosure.

For these and other reasons, higher levels of both daily and average parental knowledge may be linked to fewer youth physical health symptoms such as headaches and stomachaches as well as more healthful HPA functioning. As such, an important step is to examine the links between parental knowledge and indices of youth's physical health. Establishing linkages between knowledge and youth health is an important first step that may lay the groundwork for future studies that unpack the mechanisms through which knowledge exerts its effects on youth health.

The associations between parental knowledge and youth health may vary as a function of youth characteristics, in particular, gender and age. Moderation analysis may shed light on subgroups to target in family-based interventions aimed to improve child health. First, theory and research suggest that parents monitor girls more than boys and that girls disclose more information to their parents (Waizenhofer, Buchanan, & Jackson-Newsom, 2004). Further, girls tend to be more relationship-oriented than boys (Leaper, 2002), with the result that they may be more strongly affected by parental knowledge, and knowledge may be more predictive of health for girls than boys.

Parental knowledge may also have a stronger impact on the health of younger as compared to older adolescents. With age, youth spend less time with their parents and more unsupervised time with their peers, making it more difficult for parents to obtain knowledge about their activities (Larson, Richards, Moneta, Holmbeck, & Duckett, 1996), and several studies documented declines in parental knowledge across adolescence (Laird, Pettit, Bates, & Dodge, 2003; Masche, 2010). Further, youth become more self-directed around health behaviors with age. Food choices, decisions to engage in physical activities and bedtimes are increasingly viewed by adolescents as their own, rather than their parents’, jurisdiction (Smetana, 2011). Thus, as children become older, knowledgeable parents may be less likely to promote specific health behaviors due to youth's increasing autonomy (Wray-Lake, Crouter, & McHale, 2010). Further, as youth become older, they may share less information with their parents (Keijsers & Poulin, 2013). Parents may be less aware of stressors that youth experience and therefore have fewer opportunities to provide support and guidance. Further, older adolescents may be more likely to spend time with and seek support from peers rather than parents in dealing with stressors (Larson et al., 1996). As such, the relation between parental knowledge and health may be stronger for younger than older adolescents.

In sum, we addressed two goals. First, we assessed the links between youth reports of parental knowledge and HPA functioning and physical health symptoms, testing two kinds of hypotheses: (a) At the between-person level, youth who reported lower levels of parental knowledge, on average, will also report more physical health symptoms and exhibit less healthful levels of cortisol (e.g., high levels of cortisol before dinner and at bedtime, flatter evening slopes, and blunter next-day morning rise); and (b) at the within-person level, controlling for average knowledge, on days when youth reported lower parental knowledge than usual, they also will report more physical health symptoms and exhibit less healthful cortisol patterns. Our second goal was to test whether the associations between parental knowledge and youth health were moderated by youth gender or age. We expected to find stronger associations between knowledge and youth health outcomes in girls and younger adolescents.

High parental knowledge has been linked to a lower levels of a host of youth risky behaviors, ranging from delinquency and substance use to risky sexual behavior (Racz & McMahon, 2011). We built on this literature to document that parental knowledge may also be linked to youth's physical health and HPA functioning. Youth who reported that their parents were more knowledgeable about their experiences, on average, exhibited lower (more healthful) bedtime cortisol levels, and younger adolescents reported fewer physical health symptoms. Further, on days when parents were more knowledgeable than usual, youth also had lower levels of cortisol before dinner. The results contributed to the literature that links parental knowledge to youth risky behavior (Racz & McMahon, 2011) by extending analyses to youth physical health outcomes, including physical symptoms and HPA functioning. Our findings suggests that interventions aimed to improve parental knowledge may have benefits not only for behavioral health, but for physical health as well. We found no prior studies that tested links between parental knowledge and youth physical health. As such, although the results require replication, this study breaks new ground and suggests directions for future research and application.

Although it was beyond the scope of this study to identify underlying processes, there are several reasons why parental knowledge may be linked to healthful HPA functioning and fewer physical health symptoms. Parents who know about youth activities may be more aware of whether or not youth engage in healthful behaviors such as physical activity, eating well, and getting adequate rest (Cohen, 2004). In turn, they may be more likely to promote healthful behaviors in their children, with positive implications for youth's physical health and stress-related physiology (Buckley& Schatzberg, 2005; Meerlo et al., 2002; Pederson & Hoffman-Goetz, 2000; Traustadóttir et al., 2005). Parents who are more knowledgeable also may know more about youth stresses and anxieties and thus can offer advice and support (Stattin & Kerr, 2000; Lippold et al., 2013). As such, youth with knowledgeable parents may experience fewer stressors and their reactions to stressors may be tempered, resulting in more healthful HPA axis functioning as indexed by lower levels of salivary cortisol at the end of the day. As noted, future studies are needed to identity the processes underlying these associations, such as how parents’ knowledge may be linked to orchestration of their children's healthful behaviors and routines and the role of knowledge as a protective factor in youth's stress and coping. Such studies would further illuminate targets for interventions aimed to improve youth health.

Our study also built on prior work by assessing the implications of daily knowledge for youth's daily health. Analyses revealed a significant link at the within-individual level: On days when parents have more knowledge than usual, youth exhibited lower than usual cortisol levels before dinner. Such within-person effects are notable given that they rule out stable third variable explanations, whether measured or not (Almeida, 2005). Perhaps the period before dinner, when parents and youth have recently returned home from work or school, is a time when youth are especially inclined to process the day's stressful experiences and look to parents for support and advice. Thus, interventions may need to focus on ways to encourage parent-youth communication at the end of the school day. Consistent with this possibility, there was more variance at the within-person level for cortisol before dinner than at bedtime, which may help to account for the pattern of findings. The lack of within-person associations with physical health, bedtime cortisol, and next day CAR was surprising, given that there were substantial day-to-day fluctuations in parental knowledge and most of the youth health measures. It may be that average levels of knowledge reflect parental practices, such as promoting positive health behaviors, whose effects accumulate over days (Cohen, 2004; Umberson et al., 2014; Kitzman-Ulrich et al., 2010). Because the effects of promoting healthful behavior, such as eating a balanced diet, adequate physical activity, regular bedtimes, and adequate sleep accumulate over time, average, rather than daily levels of knowledge may be more closely linked to health symptoms. Average knowledge may also reflect a family climate of connectedness, such that youth feel a sense of security--that their parents will be available when they need them-- even if parents are less knowledgeable on a particular day.

Interestingly, although we found linkages between knowledge and evening and bedtime levels of cortisol, there were no significant linkages between knowledge and the CAR. Daily parental knowledge may have stronger effects on same-day cortisol levels because they are more proximal in time; next day effects may be too distal to detect effects given our measurement approach. Further, there is some evidence that the CAR is more endogenous and less likely to be affected by contextual factors than evening levels (Fries, Dettenborn, & Kirschbaum, 2009). Therefore, even average levels of parental knowledge may have fewer linkages to morning rise than to other indicators of HPA functioning.

As predicted, the link between knowledge and physical health symptoms was moderated by age and was significant for younger but not older adolescents. These findings may reflect increased self-direction, including around health behaviors, as youth become older and issues of bedtime, food choice, and activity involvement become part of the personal domain of decision-making (Smetana, 2011). Thus, interventions aimed to increase parental knowledge may have stronger impacts on younger adolescents’ health. As noted research is needed to illuminate the processes underlying the links between parental knowledge and youth health, including the role of youth autonomy. In contrast, age moderation was not evident for salivary cortisol outcomes, however, suggesting that parental knowledge may be linked to HPA functioning across adolescence. There also was no evidence of gender moderation in face of mixed findings of gender differences in parental knowledge and cortisol (Hankin & Abramson, 2001; Hardie, Moss, Vanyukov, Yao, & Kirillovac, 2002). This may be due to our small sample size and findings must be replicated, but they are suggestive that parental knowledge is significant for the physical health of both girls and boys.

Our findings should be interpreted in light of this study's limitations. First, our correlational design does not allow for conclusions about direction of effects: parental knowledge may promote youth health, but it also may be that youth with better health are more likely to see their parents as more knowledgeable. Experimental studies that, for example, test the effects of interventions designed to promote parental knowledge, would provide evidence of its causal role in youth health. Further, the sample included youth with generally well-educated parents who were employed in one industry; thus, the findings require replication on other groups of youth. We also did not measure potential processes underlying parental knowledge—health outcome linkages, a next important step in this line of research. Parental knowledge may reflect parents’ solicitation or behavioral control or youth disclosure (Racz & McMahon, 2011) and its influence may be moderated by contextual factors such as the affective quality of the parent-child relationship, parental support of youth autonomy, or parenting style (Lippold et al, 2014), additional directions for future research. Further, traditional estimates of effect sizes do not apply to multi-level models (Holden, Kelley, & Agarwal, 2008). Our estimates of the percent change in our outcome variables associated with a one-unit increase in knowledge were modest, especially for our cortisol models (ranging from 9.5 to 11 percent). Clearly other factors not measured here also influence youth cortisol levels. Finally, we relied on youth self-reports of parental knowledge. Parental reports of knowledge might yield a different pattern of effects (Lippold et al., 2011).

In the face of its limitations, this study provided several important contributions to our understanding of the role of parental knowledge in youth health. We focused on physical health, assessing the linkages between knowledge and youth physical health symptoms and cortisol. The consistency of results across indicators of physical health and cortisol is noteworthy and suggests that, beyond its documented links with youth behavioral adjustment, parental knowledge also has implications for youth's HPA functioning and physical health symptoms. We used a daily diary approach, asking youth to report on parental knowledge and health symptoms across eight days. Such approaches minimize recall bias and allow us to understand how both average and daily knowledge may be linked to youth physical health. We also moved beyond self-reports of health by including a biomarker of HPA axis functioning. We know little about parental behaviors that are linked to children's physical health, but our findings suggest parental knowledge may be one means by which parents impact youth physical well-being. At the most general level, they imply that interventions aimed to improve parental knowledge may have positive implications for adolescent physical health.