Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2010 Mar 1.
Published in final edited form as: J Health Soc Behav. 2009 Mar;50(1):1–15. doi: 10.1177/002214650905000101

Psychosocial and Biological Markers of Daily Lives of Midlife Parents of Children with Disabilities*

Marsha Mailick Seltzer 1, David M Almeida 2, Jan S Greenberg 3, Jyoti Savla 4, Robert S Stawski 5, Jinkuk Hong 6, Julie Lounds Taylor 7
PMCID: PMC2679175  NIHMSID: NIHMS88260  PMID: 19413131

Abstract

Using daily telephone interviews, 82 midlife parents (mean age = 57.4) of children with disabilities (mean age = 29.9) were compared with a closely matched sample of unaffected parents (n = 82) to elucidate the daily experience of nonnormative parenting. In addition, salivary cortisol samples were obtained to examine whether parents of children with disabilities had dysregulated diurnal rhythms and the extent to which the amount of time spent with children was associated with divergent patterns of cortisol expression. We found that parents of children with disabilities had similar patterns of daily time use and similar likelihood of positive daily events as the comparison group, but they had elevated levels of stress, negative affect, and physical symptoms, all reported on a daily basis. In addition, their diurnal rhythm of cortisol expression differed significantly from the comparison group, a pattern that was strongest for parents of children with disabilities on days when they spent more time with their children.

Introduction

When parents reach midlife, they look back on their accomplishments and life challenges (Han and Moen 1999). By this time, it is becoming clear whether their children have lived up to their potential and successfully negotiated the transition to adulthood. However, some children have limited success in life, experience social difficulties, or remain dependent on parents. Whereas the first scenario is the one that all parents hope for and most experience, the latter occurs as a result of life disruptions experienced by children, including disability.

We have found that parenting a child with a disability has deleterious effects on parents’ health and psychological well-being during midlife (Seltzer et al. 2001). Parents of children with disabilities face a range of stressors associated with their child’s behavior problems, including stigma, cost of care, and having to negotiate a fragmented service system, but personal and social buffers can reduce the negative effects of stress (Cook et al. 1992; Glidden, Billings, and Jobe 2006). Parents of children with disabilities experience both acute intense stress and long-term chronic strain persisting over many decades. Most past research has focused on the long-term effects of non-normative parenting, bringing a global perspective to the investigation of the stress process and assessing functioning over a period of years or decades (Floyd and Gallagher 1997; Glidden et al. 2006).

In contrast, in this study, we focus on the acute stressors that such parents experience, bringing a daily life perspective to the study of caregiving stress. We assess the extent to which features of day-to-day experiences (such as time use, frequency and severity of stressors, mood, and physical symptoms) differ between midlife parents who have a child with a developmental or mental health problem and those whose children are healthy and non-disabled. We also examine divergent patterns in stress physiology by assessing differences between parents of children with disabilities and parents of typical children with respect to diurnal rhythms of salivary cortisol.

Biological Effects of Stress

Although parent caregivers have been shown to have elevated levels of mental and physical health problems, the biological mechanisms by which caregiving takes its toll remain relatively unexplored for this population. Previous research has focused largely on self-reported health and stressors (Seltzer et al., 2001). Results are commonly qualified by discussions of possible response biases and questions concerning the validity of self-reported health measures. Thus, questions remain regarding the direct relation between daily caregiving and physiological functioning. The study of biomarkers thus has the potential to add to our understanding of daily stress by corroborating and validating self-report measures. Biomarkers also give insight into the mechanisms by which stress takes a toll on health and well-being. The only study on parenting a child with a disability that has used biomarkers was conducted by Epel and colleagues (2004), who reported that life stress is associated with accelerated telomere shortening, reflecting cellular aging. Longer duration of caregiving was associated with greater telomere shortening, controlling for maternal age.

Other pathways between stress and physical/mental health problems have been demonstrated in past research, including disruption of the hypothalamic-pituitary-adrenocortical (HPA) axis. Cortisol is a biological marker of activity of the HPA axis, which plays a vital role in linking stress exposure to health problems (McEwen 1998). However, this pathway has not been examined in parents of children with disabilities.

The present study examines diurnal patterns in salivary cortisol in midlife parents (mean age = 57) of adolescent and adult children (mean age = 29) with disabilities, and how this pattern is affected by daily stressors. Cortisol typically peaks shortly after waking in the morning and then gradually declines throughout the rest of the day. Diurnal cortisol provides a window into individuals’ chronobiology (Keenan, Licinio, and Veldhuis 2001). The early morning and evening levels of cortisol reflect daily engagement and disengagement of the brain with peripheral physiology, and hence the external environment. Failure to deactivate the HPA axis in the evening may indicate difficulty in disengaging from external demands, leading to inhibition of restoration and recovery processes (Sapolsky, Krey, and McEwen 1986).

Short-term increases in cortisol are thought to reflect a ‘normal’ physiological response to stressor exposure (Sapolsky et al., 1986). However, individual differences as well as variation in the nature of stressors may influence the magnitude of such responses, leading to exaggerated (hyper) or diminished (hypo) responsiveness. The impact of variation in cortisol stress reactivity is thought to cumulate over time, in response to repeated or chronic stressor exposure, leading to persistent high or low levels of circulating cortisol (which in turn can influence multiple aspects of physiological functioning). Hyper- and hypo-responsive cortisol stress reactivity are symptomatic of poor physical health, generally interpreted as wear-and-tear on the HPA-axis (Kiecolt-Glaser et al. 1986; Segerstrom and Miller 2004). The measurement of daily cortisol rhythms may provide the best window into cortisol physiology, providing information about overall levels and fluctuations in cortisol levels across the day, and the association of these characteristics of cortisol with exposure to stressful experiences and individual/contextual factors. Studying cortisol expression in parents of children with disabilities offers an exciting opportunity to examine how daily life experiences influence daily physiology and associations with indicators of health and well-being.

Research has shown that acute stressors are associated with elevations in cortisol levels at waking and 30 minutes after waking as compared to individuals who do not experience stress. For example, caregivers of family members with dementia have a more pronounced elevation in daytime level of cortisol than the norm (Bauer et al. 2000, Cacioppo et al. 2000; De Vugt et al. 2005). However, a different pattern of cortisol dysregulation is evident in other examples of chronic stress. Hypoactivity of the HPA axis has been documented in chronic unemployment, bereavement, environmental disasters, chronic fatigue syndrome, and fibromyalgia (Baum, Schaeffer, and Lake 1985; Demitrack et al. 1991; Griep, Boersma, and de Kloet 1993; Jacobs et al. 1987; Ockenfels et al. 1995; Scott and Dinan 1998). Pruessner, Hellhammer, and Kirschbaum (1999) found that teachers scoring high on burnout showed lower overall cortisol secretion. Of particular relevance to the present study, Adam and Gunnar (2001) found that mothers who worked more hours and had more children at home had lower morning cortisol values and a less pronounced decline in cortisol levels across the day than mothers working fewer hours and having fewer children. One purpose of the present study is to examine the extent and specific pattern of cortisol dysregulation in parents of children with disabilities as compared with closely matched controls.

Daily versus Global Indicators of Stress, Health, and Psychological Well-Being

Global measures of health and mental health problems in parents of children with disabilities provide valuable insight about the long-term toll taken by non-normative parenting. Examination of global outcomes indicates that such parents have divergent patterns of social participation, employment, health, and psychological functioning (Avison et al. 1993; Seltzer et al. 2001). However, little is known about their daily life and whether the stress of enacting this caregiving role can be linked directly with daily measures of health and psychological well-being. There is an emerging literature on daily stressors in the general population, suggesting that sources of interpersonal stress such as conflicts with spouse, work-family conflicts, arguments with friends and family, and the like, take a toll on health and well-being (Almeida 2005; Serido, Almeida, and Wethington 2004).

We hypothesize that spending more time with a child with a disability will be experienced as stressful due to the greater demands for care and supervision that such children present, as well as their behavior problems and behavioral unpredictability (Abbeduto et al. 2004; Floyd and Gallagher 1997). Our design permits us to assess if parents of children with disabilities report greater psychological distress and display greater dysregulation of the diurnal rhythm of cortisol on days when they spend more time with their children. Thus, a second purpose of the present paper is to determine the extent to which daily measures of time use are associated with daily measures of physical symptoms, psychological health, and dysregulation of the HPA axis. This analysis builds on a recent study by Adam and colleagues (2006) showing the association between daily fluctuations in affective states and daily fluctuations in diurnal cortisol rhythm. We extend these analyses by assessing if daily variation in the amount of time spent in a stressful role is associated with daily variation in both affective states and diurnal cortisol rhythm. Thus, the present paper seeks to establish a more direct link between daily stressors and HPA axis dysregulation.

Research Questions

We address three research questions. First, we ask whether midlife parents of children with disabilities differ from their counterparts whose children have no disabilities with respect to the amount of time they spend daily with their children, on routine chores, at work, and on leisure activities; the type and frequency of stressors they experience on a daily basis; and their experience of positive daily events. In addition, we ask whether they differ in their levels of positive affect, negative affect, and physical health symptoms. We hypothesize that, due to the special needs of their son or daughter with the disability, these parents will spend more time with their children and less time on leisure activities and work, but will not differ from the comparison group with respect to time spent on chores. We further predict that parents of children with disabilities will experience more stressors on a daily basis and fewer positive events. We also hypothesize that parents of children with disabilities will have higher levels of negative affect, lower levels of positive affect, and more physical health symptoms than their counterparts whose children do not have disabilities.

Second, we focus on the diurnal rhythms of salivary cortisol by fitting individual growth curves based on four daily samples on four consecutive days for each participant, using multilevel modeling. We hypothesize that parents of children with disabilities will evidence a pattern of cortisol dysregulation, with a steeper morning rise but a less pronounced decline than parents in the comparison group, suggesting less of a deactivation of the HPA axis.

Third, we examine whether parents of children with disabilities differ from the comparison group in the association between the amount of time spent with children during the day and positive affect, negative affect, physical symptoms, and cortisol dysregulation. We hypothesize that, due to the stress of responding to the needs of the son or daughter with the disability, the more time parents of children with disabilities spend with their child on a given day, the lower the degree of positive affect, and the greater the degree of negative affect, physical symptoms, and cortisol dysregulation. We predict that these effects will be significantly less pronounced in the comparison group.

Methods

The data for this analysis come from the National Study of Daily Experiences (NSDE), one of the projects that comprise the National Survey of Midlife in the United States (MIDUS; Carol Ryff, PI). MIDUS is a national probability sample of English-speaking, non-institutionalized adults aged 25 to 74 in 1994 (MIDUS I; Brim, Ryff, and Kessler 2004). Follow-up data were collected from 2003 to 2005 (MIDUS II). At MIDUS II, all parents were asked if any of their children had a developmental or a mental health problem, and if so, which child had the condition and the name of the diagnosis. Approximately one in ten (10.5%) responded affirmatively, of whom nearly half (46.3%) had a child with a developmental problem, about the same number (42.7%) had a child with a mental health problem, and the remaining 11% had a child with another type of neurological disability.

A subset of MIDUS II sample members was also included in the second wave of the National Study of Daily Experiences (NSDE II; David Almeida, PI), which is the source of data for the present analyses. The NSDE II, which is currently ongoing, consists of short telephone interviews at the end of each of eight consecutive days, with salivary cortisol samples obtained four times a day on days 2 through 5. The NSDE daily telephone interview includes questions about daily experiences in the past 24 hours concerning time use, number of stressors and positive events, daily measures of positive and negative affect, and daily measures of health symptoms and their severity (Almeida et al. 2002).

Sample

For this analysis, two groups were selected from the 806 MIDUS II members who had participated in the NSDE II prior to the time of the present data analyis. The first group consisted of 82 parents of children of disabilities, and the second group consisted of a matched comparison group of equal size. Disabilities in children included attention deficit hyperactivity disorder (15.9%), bipolar disorder (12.2%), schizophrenia (9.8%), depression (7.3%), Down syndrome (6.1%), and others (48.7%). If there was more than one child with a disability in the family, we designated the oldest child with a disability as the target child for this investigation because the oldest child defines the duration of non-normative parenting.

The comparison group was drawn from MIDUS II respondents who have participated in the NSDE II and have at least one living child, but no child with a disability or chronic health condition, and who never provided care to a family member. Of the 806 NSDE participants, 54 were excluded from this analysis because they were not parents, an additional 254 were excluded because they provided care to another family member (e.g., parent, spouse), and 82 had a child with a disability (the target sample). Thus, there were 416 potential matches. We selected the 82 closest matches with respect to the following variables (in this order): parent gender, parent age, number of children in the household, child age, whether the target child lives with the parent, parent marital status, and parent educational attainment.

The 82 members of the comparison group were similar to the other potential comparison group cases with respect to virtually all background characteristics and outcome variables. The only significant difference was the number of children in the family (3.21 for the selected comparison group cases versus 2.73 for the non-selected cases). As parents of children with disabilities tend to have larger families, this difference reflects one of the matching criteria.

As shown in Table 1, the matching procedures were successful in producing highly comparable groups. They were similar in parental age, gender, race, marital status, employment status, years of education, income, and number of children. The target child averaged 29 years of age, with a wide age range (from 3 to 63). However, the vast majority of the children (87.7%) were adolescents or adults. There were more sons (60%) than daughters (40%). Whereas 41% of the children with disabilities lived with their parents, only 32% of the comparison group still lived in the parents’ household. The analysis of the third research question focuses on the coresident sub-samples only, in order to be confident of daily contact between parent and child.

Table 1.

Descriptive statistics (mean with standard deviation in parenthesis) of parents of children with disabilities (n = 82) and comparison group parents (n = 82).

Variables Parents of children with disabilities Comparison group
Parent’s Characteristics
  Age 57.4 (13.0) 57.4 (13.1)
  Gender (1 = female/ 0 = male) .59 (.50) .59 (.50)
  Race (1 = non-Hispanic white / 0 = others) .96 (.19) .97 (.16)
  Marital status (1 = married / 0 = not married) .79 (.41) .84 (.37)
  Employment status (1 = employed; 0 = not employed) .57 (.50) .61 (.49)
  Years of education 14.4 (2.65) 14.5 (2.35)
  Total household income $74,400 (49,800) $78,300 (50,100)
  Number of children 3.29 (1.91) 3.21 (1.26)
Child’s Characteristics
  Age 29.3 (13.4) 29.9 (13.4)
  Gender (1 = female/ 0 = male) .40 (.49) .40 (.49)
  Living with parents (1 = yes/ 0 = no) .41 (.50) .32 (.47)
Open in a new tab

Measures

The 8-day daily telephone interview included questions about daily experiences in the previous 24 hours concerning time use, daily stressors, positive events, mood, and physical symtoms.

Time Use

Time spent with children was coded as the amount of time in minutes that the parent spent with children, including taking them places or doing things with them. Time spent on routine chores assessed the amount of time the parent spent on routine chores in the house as well as yard work. Time spent on work was coded as the amount of time spent on activities related to business, paid work or school, which included time traveling and thinking about work. Finally, time on leisure activities included time spent relaxing, engaging in leisure activities, sleeping or taking a nap, or physical exercise. These variables were significantly skewed and thus log transformed in the analysis.

Stressors

Daily stressors were assessed through the Daily Inventory of Stressful Events (Almeida et al. 2002), which is available from the second author upon request. The inventory consists of questions asking whether certain types of daily stressors had occurred in the past 24 hours: arguments, avoided arguments, work stressors, home stressors, and network stressors (defined as stressors that occurred in the life of close friends and family members) and a final question that captured other miscellaneous stressors (e.g., traffic). For each daily interview, individuals who responded affirmatively to each question received a code of 1 for the relevant stressor domain. Codes of 0 were assigned to domains where no stressors were experienced on that day. The stressors were summed to create number of stressors per day. Additionally, a dummy coded variable was created to indicate if the participant experienced any stressors on the study day. Subjective severity of stressors was based on respondents’ answers to questions about perceived severity of stressors and was coded from 1 (not at all stressful) to 4 (very stressful).

Positive Events

In a similar manner to stressful events, parents reported positive events that occurred in the previous 24 hours using five questions: a particularly positive interaction with somebody, a positive interaction at work or volunteering, at home, reporting that a positive event occurred for a close friend or relative, or anything else that was particularly positive. The positive events were summed to create number of positive events per day. Additionally, a dummy coded variable was created to reflect if the participant experienced any positive events on the study day.

Mood

Negative and positive emotions were measured each day using an adapted inventory of 20 emotions from the Non-Specific Psychological Distress and Positive Emotions Scale (Kessler et al. 2002). The negative affect scale (10 items) measured anxiety, hostility and depression. The positive affect scale (10 items) measured enthusiasm, alertness, and vitality. Each day, respondents indicate how frequently they felt each emotion over the past 24 hours on a 5-point scale from “none of the time” to “all of the time”. Averaging reliability analyses conducted on each diary day, the mean Cronbach’s alpha for negative and positive affect were .94 and .93, respectively.

Physical Symptoms

Symptoms were assessed each day using an adapted version of the Larsen and Kasimatis (1991) symptom checklist. We omitted items that overlapped with the psychological distress scale (e.g., “urge to cry”). This version is a checklist that combines the occurrence and severity of 19 physical symptoms including aches (headaches, backaches, and muscle soreness), gastrointestinal symptoms (poor appetite, nausea/upset stomach, constipation/diarrhea), and upper respiratory symptoms (sore throat, runny nose) as well as an open-ended item. Each day the respondents indicated if they experienced each symptom and the degree of severity on a 10-point scale from “very mild” to “very severe”.

Salivary Cortisol

Salivary cortisol samples were collected four times on each of four days in the NSDE II. Respondents received a Home Saliva Collection Kit one week prior to their initial phone call. Saliva was obtained using salivette collection devices (Sarstedt). Sixteen numbered and color-coded salivettes were included in the collection kit, each containing a small absorbent wad, about 3/4 of an inch long, as well a detailed instruction sheet. In addition to written instructions, telephone interviewers reviewed the collection procedures and answered any questions. Respondents provided 4 saliva samples per day on days 2 through 5 of the 8-day period to be assayed for cortisol: one upon wakening, one 30 minutes after getting out of bed, one before lunch, and one at bed time. Data on the exact time respondents provided each saliva sample were obtained from the nightly telephone interviews and on a paper-pencil log sent with the collection kit. In addition, approximately 25% of the respondents received a “smart box” to store their salivettes. These boxes contain a computer chip that records the time respondents open and close the box. The correlations between self-reported times and the times obtained from the “smart box” range from .75 for the evening occasion to .95 for the morning.

Measures of salivary cortisol include three parameters of diurnal rhythm: morning rise, daily decline and acceleration of daily decline. Morning rise is an indicator of how high an individual’s cortisol rises, measured from awakening to 30 minutes after awakening. Daily decline refers to the slope from the typically highest point in the day, measured at 30 minutes after awakening, through the collection before bed. The acceleration of daily decline was calculated as the quadratic form of daily decline.

Results

Group Differences in Daily Experiences and Well-Being

Table 2 presents the data for the first research question which examined differences between parents of children with disabilities and comparison group parents on daily experiences and well-being. Parents of children with disabilities and comparison parents did not differ in the time they spent on a daily basis with their children, on routine chores, at work, or in pursuit of leisure activities.

Table 2.

Mean comparisons between parents of children with disabilities (n = 82) and comparison group parents (n = 82) on type and severity of stressors, mood, and symptoms.

Variables Parents of children with disabilities Comparison group
M SD M SD t-test
Time Use:
  Time spent with children (logged) 1.33 1.75 1.27 1.90 0.23
  Time spent on routine chores (logged) 3.13 1.45 2.82 1.44 1.37
  Time spent on work (logged) 2.59 1.98 2.55 2.19 0.13
  Time spent on leisure activities (logged) 4.20 1.16 4.27 1.08 −0.35
Stressors:
  Arguments 0.13 0.15 0.08 0.12 2.36*
  Avoided Arguments 0.18 0.17 0.13 0.13 2.21*
  Number of Stressors 0.74 0.64 0.52 0.42 2.60**
  Days with any Stressors 0.50 0.26 0.40 0.25 2.49**
  Work Stressors 0.07 0.10 0.08 0.14 −0.57
  Home Stressors 0.13 0.14 0.09 0.11 2.30*
  Network Stressors 0.02 0.07 0.01 0.03 1.28
  Severity of Stressors 2.51 1.32 2.09 1.00 2.27*
Positive Events:
  Number of Positive Events 1.09 0.66 1.04 0.63 0.49
  Days with any Positive Event 0.69 0.28 0.69 0.26 −0.13
Affect:
  Negative Affect 0.20 0.18 0.14 0.15 2.17*
  Positive Affect 2.57 0.73 2.78 0.66 −1.88+
Symptoms:
  Number of Symptoms 2.20 2.11 1.55 1.28 2.38*
  Any Symptoms 0.74 0.30 0.65 0.35 1.69
  Severity of Health Symptoms 7.53 9.09 5.59 5.42 1.66
Open in a new tab
+

Note: p=.06

*

p < .05

**

p < .01.

However, consistent with our hypothesis, parents of children with disabilities experienced a significantly greater number of stressors and more days on which they experienced at least one stressor. On average, parents of children with disabilities reported experiencing at least one stressor on 50% of the study days (versus 40% for the comparison group). They also reported significantly more arguments and tense moments with others (i.e., avoided arguments) than the comparison group. The elevation in their measures of daily stressors reflected significantly more stressors at home, but they did not differ from the comparison group in the stressors they experienced at work or with their social network. They rated their stressors as significantly more severe than the comparison group.

Counter to our prediction, parents of children with disabilities experienced a similar number of positive events as the comparison group and a similar number of days in which they experienced at least one positive event. Both sets of parents experienced at least one positive event on 69% of the study days. However, parents of children with disabilities had significantly higher levels of negative affect and a greater number of physical symptoms than the comparison group, and marginally lower levels of positive affect, as we predicted.

Daily Rhythms of Salivary Cortisol

Next, we used multilevel modeling to assess both inter- and intra-individual variation of diurnal cortisol (Adam et al. 2006). These models estimated variation in the morning rise, daily decline and the acceleration of daily decline. The inclusion of the acceleration parameter allowed us to assess whether deactivation of the HPA system occurs closer to the end of the day, as failure to deactivate HPA by the end of the day is a marker of dysregulation.

As shown in Table 3, both groups showed a significant morning rise in cortisol level, but counter to our prediction, parents of children with disabilities and comparison parents did not significantly differ in the slope of the morning rise (.245 vs. .199, n.s). Although both groups declined in cortisol level during the day, parents of children with disabilities exhibited significantly less pronounced daily decline slopes, a pattern consistent with our prediction. The linear component of the daily decline parameter was substantially less pronounced for parents of children with disabilities (44% less pronounced than for the comparison group). Furthermore, there was a significant quadratic trend in the daily decline slopes, indicating that the linear decline in cortisol abates throughout the day, and this abatement is much stronger (by 100%) for parents of children with disabilities than for the comparison parents. This pattern indicates that parents of children with disabilities are significantly less likely to deactivate the HPA axis at the end of the day than their counterparts in the comparison group, suggesting inhibition of restoration and recovery processes.

Table 3.

Parameter estimates for parents of children with disabilities (n = 82) and comparison group parents (n = 82) for diurnal cortisol rhythm.

Parents of children with disabilities Comparison group parents Difference Interpretation
Fixed Effects Estimate (SE) Estimate (SE) Estimate (SE)
Morning Rise .245 (.063)** .199 (.069)** −.046 (.094)
Daily Decline (L) −.038 (.010)** −.068 (.010)** −.030 (.015)* 44% flatter for parents of children with disabilities
Daily Decline (Q) .004 (.001)** .002 (.001)** −.002 (.001)* 100% greater for parents of children with disabilities
Average Bedtime Cortisol Level 1.314 (.067)** 1.241 (.067)** −.073 (.095)
Variance Components Between-Person (Level 3)
    Var(Intercept) .261** .239**
    Var(Morning Rise) .067† .109**
    Var(Daily Decline (L)) .002** .001**
    Cov(Morning Rise.Daily Decline(L)) .016** .009**
Within-Person Across Day (Level 2)
    Var(Intercept) .103** .102**
    Var(Dialy Decline (L)) .001** .001**
    Cov(Intercept.Dialy Decline(L)) .007** .008**
Within-Person Within-Day (Level 1)
Residual .246** .218**
Open in a new tab
†

Note: p = .07

*

p < .05

**

p < .01. L = Linear, Q = Quadratic. Var = Variance, Cov = Covariance.

The bottom portion of Table 3 shows the variance components of the diurnal rhythm of cortisol. At the between-person level, both groups of parents exhibited significant individual differences in the each of the parameters of diurnal rhythm. Furthermore, the between-person covariance suggests that individuals with a steeper morning rise had a less pronounced daily decline in cortisol. At the within-person level, there was also evidence of significant day-to-day variation in diurnal patterns of cortisol among both groups of parents.

We also examined group differences in the diurnal rhythm of cortisol, controlling for co-residence status of the child. The significant differences remained, with the magnitude of the group difference remaining virtually unchanged for either the linear (estimate = −.03, SE = .01, p < .05) or the quadratic (estimate = −.002, SE = .0008, p < .05) components.

Time with children and daily well-being

The goal of the final set of analyses was to use the amount of time parents spend with their children on a given day to predict within-person variation in diurnal pattern of cortisol and other indicators of daily well-being. Specifically, we investigated if there were within-person associations between time spent with children and negative affect, physical symptoms, and the cortisol measures among parents with children with disabilities living in the household. Although only 34 of the 82 children with disabilities were co-resident with the parent at the time of data collection, we focused on the co-resident subgroup in this final analysis to ensure a closer association between daily contact with children and well-being; it was only in these 34 cases that we could be confident that there was daily contact between parent and the child with the disability. Since our sample of co-resident children with disabilities was small, we used a multilevel modeling approach with a time-varying covariate (time spent with children) with standard error estimates based on 1000 bootstrap samples to obtain bias-corrected confidence intervals by repeated re-estimations of the parameter estimates using random samples with replacement from the original panel data.

As shown in Table 4, there was a significant interaction between parental status (having a child with a disability versus unaffected children) and time spent with co-resident children with respect to parental well-being outcomes. On days when they spent more time with their children, parents of children with disabilities reported higher levels of negative affect compared to days they spent less time with their children, whereas parents in the comparison group did not evidence a difference in negative affect based on the amount of time they spent with their children (see Figure 1). Figure 2 shows that parents of children with disabilities reported fewer physical symptoms on days when they spent more time with their children, whereas parents in the comparison group showed the opposite pattern. Finally, parents of children with disabilities had a (marginally) less pronounced daily decline of cortisol expression on days when they spent more time with their children as compared to days they spent less time, whereas again the opposite pattern was evident for the parents in the comparison group (see Figure 3), suggesting that parents of children with disabilities were less likely to deactivate the HPA axis during higher stress days than the comparison group.

Table 4.

Multilevel model parameter estimates for daily variation in time with children predicting daily well-being in parents with co-residing children with (n = 34) and without (n = 26) disabilities

Negative Affect Positive Affect Physical Symptoms Cortisol Morning Rise Cortisol Dialy Decline
Fixed Effects Est. (BSE) Est. (BSE) Est. (BSE) Est. (BSE) Est. (BSE)
Intercept .16 (.03)** 2.66 (16)** 1.36 (.23)** .79 (.31)** −.16 (.03)**
Time Spent with Children (Log) .01 (0.01) −.01 (.01) 0.03 (.04) −.05 (.06) .00 (.01)
Parenting Status (disabilities vs control) .01 (0.04) .06 (.19) 1.18 (.44)** −.05 (.35) −.03 (.04)
Time with Children × Parenting Status .02 (0.01)* −.01 (.02) −0.10 (.05)* −.04 (.08) .01 (.01)†
Variance Components
Between-Person (Level 2) .02** .44** 2.72** 0.30** 0.01
Within-Person .04** .12** 1.44** 1.21** 0.01**
Across Day (Level 1)
Open in a new tab
†

Note: p = .06

*

p<.05

**

p<.01.

BSE = Bootstrapped Standard Error

Parenting status is coded as 0 = control and 1 = parents of children with disabilities.

Technical Note. Since Table 4 presents a multi-level model, the analysis is based on two sample sizes at each level, the number of subjects and the number of person-days (negative affect, positive affect, and physical symptoms were collected over an eight day period and cortisol over a four day period). Thus, for negative affect, positive affect, and physical symptoms, there were 34 parents of children with disabilities and 26 comparison group parents, with the total number of days in the analysis being 453 (258 days in the disabilities sample and 195 days in the comparison sample). With respect to the morning rise in cortisol, because of missing data, there were 22 comparison group parents and 28 parents of children with disabilities, representing a total of 177 days (97 parent days for parents of children with disabilities and 80 comparison group parent days). The analysis predicting the morning decline was based on 28 parents of children with disabilities and 22 comparison group cases, representing a total of 172 parent days (96 parent days for the disability group and 76 days in the comparison group).

Figure 1.

Figure 1

Open in a new tab

Daily negative affect in parents as a function of time spent with co-residing children with and without disabilities

Figure 2.

Figure 2

Open in a new tab

Daily reported physical symptoms in parents as a function of time spent with co-residing children with and without disabilities

Figure 3.

Figure 3

Open in a new tab

Daily decline in cortisol in parents as a function of time spent with co-residing children with and without disabilities

Discussion

Although a considerable amount of past research has provided global assessments of the toll of parenting a child with disabilities (Glidden et al. 2006; Seltzer et al., 2001), the present study extends this research. First, we focused on the daily level, providing data about these parents’ use of time, the sources of stress and positive events that comprise their days, and the daily experience of positive affect, negative affect, and physical symptoms. The comparison between parents of children with disabilities and their counterparts whose children are unaffected presents a deeper understanding of the specific ways that parenting a child with a disability does and does not take a toll on daily life.

Counter to our expectations, this nonnormative parenting role does not dysregulate parents’ patterns of time use, as we found that they spent the same amount of time as the comparison group with children, doing chores, at work, and on leisure activities, on average. This finding supports prior research on the amount of time spent on a daily basis by parents with their young children with developmental delays (Konstantareas, Homatidis, and Plowright 1992), which did not differ from an unaffected comparison group. Combined with our findings regarding stress and negative affect, the comparability in time use suggests that it is the qualitative aspects of daily life, rather than the schedule, that differentiates the lives of parents of children with disabilities.

Parents of children with disabilities report having (and avoiding) more arguments, and experiencing more frequent and severe stressors, especially at home, than the comparison group. Similarly, they report more negative and less positive affect on a daily basis, and experience a greater number of physical health symptoms. Although the present study focuses on the daily lives of these parents, the patterns we observed during this eight-day window accumulate over decades. The average age of the children in this study was 29 years and the average duration since the onset of the child’s disability was 18.23 years. Thus, it is not surprising that daily stress translates in the long run into chronic strain.

We collected these data from an unselected sample of parents of children with disabilities who were drawn from a national probability sample. Nearly all previous studies of the impact of parenting a child with a disability were based on self-selected samples and thus it is possible that they over- or under-estimated the degree of stress experienced. Much in the same way that Thoits (2005) used a national community sample to include both identified and unidentified persons with a mental illness, the present sample was not based on self-selection from caregiving support groups, service utilization, or professional referral.

A limitation of this study is that the MIDUS participants have higher than average levels of socioeconomic status, and thus generalizations should be made cautiously. An additional limitation is that we were not able to analyze sub-groups based on age of the child or type of disability (due to the small size of the sample) or severity of disability (as the MIDUS study did not collect data on severity). Our previous research, which focused on global rather than daily processes, has shown that there are differences in the impact of nonnormative parenting according to the type of disability of the child (Seltzer et al. 2001). This type of contrast should be pursued in future studies of daily stress. In addition, future studies should examine the specific aspects about the daily lives of these parents that contribute to poor outcomes, such as the perception that the child with the disability can control his or her own behavior, which has been shown to increase distress (Greenley 1986; Hooley and Campbell 2002).

Another limitation of our study is that we were unable to control for parental well-being separate from the child’s disability. There is considerable evidence that first degree relatives of persons with mental illness suffer from high rates of their own mental health problems (Laurent et al. 2002; Wilens et al. 2007). Thus, we cannot rule the possibility that some of these parents had poorer well-being prior to onset of their child’s disability. However, in a recent study (Achenbrenner, Greenberg, and Seltzer in press), we found that parents of adults with bipolar disorder had poorer well-being even after controlling for the parents’ pre-existing mental health problems.

A strength of this study is that we included biomarkers of stress as well as self-report measures, and the data show a converging pattern. We found that parents of children with disabilities experience significantly more stress and have dysregulated cortisol rhythms. Follow-up analysis showed that, among parents who live in the same household as the child with the disability, more time spent with children is associated with more psychological distress and more dysregulated cortisol expression, in contrast to the comparison group. Thus, the study provides strong new evidence that the stress reported by these parents leaves a biological signature.

The specific pattern of cortisol dysregulation was somewhat different than expected. Although we hypothesized that parents of children with disabilities would show a steeper morning rise than the comparison group, there was no group difference in this regard. A steeper morning rise is typical of the response to acute stress. Rather, we observed the expected differences in the daily decline parameter, with parents of children with disabilities showing a substantially less pronounced decline than the comparison group. This diurnal pattern is associated with chronic strain more than with acute stress (Adam and Gunnar 2001; Cacioppo et al. 2000), consistent with the chronic nature of these parents’ caregiving responsibilities, extending over decades. Our focus on midlife parents made it possible to detect effects of chronic strain rather than acute stress which would be more likely in the early years of nonnormative parenting (Ha et al. 2008).

Although we found that parenting a child with a disability was associated with a number of negative outcomes (e.g., elevated levels of stressors and negative affect, and dysregulated cortisol expression), there were other measures on which they were similar to the norm, including time spent with children, on chores, at work and on leisure activities; stress at work and with friends and family; and the occurrence of positive events. Furthermore, on days when parents of children with disabilities spent more time with their children, even though they had elevated negative affect, they did not differ in their level of positive affect. These findings reveal the ways in which these families have normative daily lives as well as the ways in which their nonnormative parenting role takes a toll on well-being.

A surprising finding was that on days when parents of children with disabilities spent more time with their children they reported fewer physical symptoms. Although this might be an anomalous finding, it also might suggest the reverse causal order, namely that parents spend less time with their children, and even make other arrangements for child care, on days when they are not feeling well. Such an explanation would cast these parents into the role of “psychological activists” (Thoits, 1994) who head off problems as a strategy for stress reduction.

In conclusion, the present study provides a new understanding of the daily lives of midlife parents of children with disabilities, benchmarked by a normative comparison group, and offers implications for future research as well as for service delivery for parents and their children with disabilities. The findings show how daily stress accumulates over the months and years of parenting a child with a disability, leading to a pattern of dysregulation of cortisol that is more consistent with chronic than acute stress. The association between time spent by the parent with the child during the day and dysregulation of parental cortisol underscores the need for respite services and other sources of family supports that may buffer the effects of daily stress. This study provides insight into one mechanism by which parenting a child with a disability takes a toll on parental health and psychological well-being and points the way for future research to determine the extent to which parenting children with different types of disabilities may lead to distinct profiles of daily stress processes.

Acknowledgements

This research was supported by grants from the National Institute on Aging (P01 AG020166 and R01AG019239) to conduct a longitudinal follow-up of the MIDUS (Midlife in the US) investigation. The original study was supported by the John D. and Catherine T. MacArthur Foundation Research Network on Successful Midlife Development. Support was also obtained from P01 AG21079 from the National Institute on Aging and P30 HD03352 from National Institute of Child Health and Human Development to the Waisman Center at the UW-Madison.

Contributor Information

Marsha Mailick Seltzer, University of Wisconsin-Madison.

David M. Almeida, Pennsylvania State University

Jan S. Greenberg, University of Wisconsin-Madison

Jyoti Savla, Virginia Polytechnic Institute and State University.

Robert S. Stawski, Pennsylvania State University

Jinkuk Hong, University of Wisconsin-Madison.

Julie Lounds Taylor, Vanderbilt University.

References

  1. Abbeduto, Leonard, Mailick Seltzer Marsha, Shattuck Paul, Wyngaarden Krauss Marty, Orsmond Gael, Murphy Melissa M. Psychological Well-Being and Coping in Mothers of Youths with Autism, Down Syndrome, or Fragile X Syndrome. American Journal of Mental Retardation. 2004;109:237–254. doi: 10.1352/0895-8017(2004)109<237:PWACIM>2.0.CO;2. [DOI] [PubMed] [Google Scholar]
  2. Adam Emma K, Gunnar Megan R. Relationship Functioning and Home and Work Demands Predict Individual Differences in Diurnal Cortisol Patterns in Women. Psychoneuroendocrinology. 2001;26:189–208. doi: 10.1016/s0306-4530(00)00045-7. [DOI] [PubMed] [Google Scholar]
  3. Adam Emma K, Hawkley Louise C, Kudielka Brigitte M, Cacioppo John T. Day-to-Day Dynamics of Experience-Cortisol Associations in a Population-Based Sample of Older Adults. Proceedings of the National Academy of Sciences. 2006;103:17058–17063. doi: 10.1073/pnas.0605053103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Almeida David M. Resilience and Vulnerability to Daily Stressors Assessed via Diary Methods. Current Directions in Psychological Science. 2005;14:64–68. [Google Scholar]
  5. Almeida David M, Elaine Wethington, Kessler Ronald C. The Daily Inventory of Stressful Events: An Interview-Based Approach for Measuring Daily Stressors. Assessment. 2002;9:41–55. doi: 10.1177/1073191102091006. [DOI] [PubMed] [Google Scholar]
  6. Aschbrenner, Kelly, Greenberg Jan S, Marsha Mailick Seltzer. “Parenting an Adult Child with Bipolar Disorder in Later Life” Forthcoming. Journal of Nervous and Mental Disease. doi: 10.1097/NMD.0b013e3181a206cc. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Avison William R, Jay Turner R, Noh Samuel, Kathy Nixon Speechley. The Impact of Caregiving: Comparisons of Different Family Contexts and Experiences. In: Zarit SH, Pearlin LI, Schaie KW, editors. Caregiving Systems: Informal and Formal Helpers. Hillsdale, NJ: Lawrence Erlbaum Associates; 1993. pp. 75–105. [Google Scholar]
  8. Bauer, Moises E, Kav Vedhara, Paula Perks, Wilcock Gordon K, Lightman Stafford L, Shanks Nola. Chronic Stress in Caregivers of Dementia Patients is Associated with Reduced Lymphocyte Sensitivity to Glucocorticoids. Journal of Neuroimmunology. 2000;103:84–92. doi: 10.1016/s0165-5728(99)00228-3. [DOI] [PubMed] [Google Scholar]
  9. Baum, Andrew, Schaeffer Marc A., C Lake Raymond. Psychological and Endocrinological Correlates of Chronic Stress at Three Mile Island. In: Williams RB, editor. Perspectives on Behavioral Medicine, Vol. 2: Neuroendocrine Control and Behavior. New York: Academic Press; 1985. pp. 201–217. [Google Scholar]
  10. Brim G, Orville G, Ryff Carol D, Kessler Ronald C. The MIDUS National Survey: An Overview. In: Brim OG, Ryff CD, Kessler RC, editors. How Healthy Are We?: A National Study of Well-Being at Midlife. Chicago, IL: University of Chicago Press; 2004. pp. 1–36. [Google Scholar]
  11. Cacioppo John T, Burleson Mary H, Poehlmann Kristen M, Malarkey William B, Kiecolt-Glaser Janice K, Berntson Gary G, Uchino Bert N, Ronald Glaser. Autonomic and Neuroendocrine Responses to Mild Psychological Stressors: Effects of Chronic Stress on Older Women. Annals of Behavioral Medicine. 2000;22:140–148. doi: 10.1007/BF02895778. [DOI] [PubMed] [Google Scholar]
  12. Cook Judith A, Hoffschmidt Sara, Cohler Bertram J, Pickett Susan A. Marital Satisfaction Among Parents of the Severely Mentally Ill Living in the Community. American Journal of Orthopsychiatry. 1992;62:552–563. doi: 10.1037/h0079370. [DOI] [PubMed] [Google Scholar]
  13. Demitrack Mark A, Dale Janet K, Straus Stephen E, Laue Louisa, Listwak Samuel J, Kruesi Markus JP, Chrousos George P, Gold Philip W. Evidence for Impaired Activation of the Hypothalamic-Pituitary-Adrenal Axis in Patients with Chronic Fatigue Syndrome. Journal of Clinical Endocrinology and Metabolism. 1991;73:1224–1234. doi: 10.1210/jcem-73-6-1224. [DOI] [PubMed] [Google Scholar]
  14. De Vugt, Marjolein E, Nicolson Nancy A, Aalten Pauline, Lousberg Richel, Jolle Jelle, Verhey Frans RJ. Behavioral Problems in Dementia Patients and Salivary Cortisol Patterns in Caregivers. Journal of Neuropsychiatry and Clinical Neurosciences. 2005;17:201–207. doi: 10.1176/jnp.17.2.201. [DOI] [PubMed] [Google Scholar]
  15. Epel Elissa S, Blackburn Elizabeth H, Lin Jue, Dhabhar Firdaus S, Adler Nancy E, Morrow Jason D, Cawthon Richard M. Accelerated Telomere Shortening in Response to Life Stress. Proceedings of the National Academy of Sciences. 2004;101:17312–17315. doi: 10.1073/pnas.0407162101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Floyd Frank J, Gallagher Erin M. Parental Stress, Care Demands, and Use of Support Services for School-Age Children with Disabilities and Dehavior Problems. Family Relations. 1997;46:359–371. [Google Scholar]
  17. Glidden Laraine M, Billings James F, Jobe Brian M. Personality, Coping Style and Well-Being of Parents Rearing Children with Developmental Disabilities. Journal of Intellectual Disability Research. 2006;50:949–962. doi: 10.1111/j.1365-2788.2006.00929.x. [DOI] [PubMed] [Google Scholar]
  18. Greenley James R. Social Control and Expressed Emotion. Journal of Nervous and Mental Disease. 1986;174:24–30. doi: 10.1097/00005053-198601000-00004. [DOI] [PubMed] [Google Scholar]
  19. Griep Eduard N, Boersma Johannes W, de Kloet Ronald E. Altered Reactivity of the Hypothalamic-Pituitary-Adrenal Axis in the Primary Fibromyalgia Syndrome. Journal of Rheumatology. 1993;20:469–474. [PubMed] [Google Scholar]
  20. Ha Jung-Hwa, Jinkuk Hong, Seltzer Marsha Mailick, Greenberg Jan. Age and Gender Differences in the Well-Being of Midlife and Aging Parents with Children with Mental Health Problems or Developmental Problems: Report of a National Study. Journal of Health and Social Behavior. 2008;49:301–316. doi: 10.1177/002214650804900305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Han Shin-Kap, Moen Phyllis. Clocking Out: The Changing Temporal Pattern of Retirement. American Journal of Sociology. 1999;105:191–136. [Google Scholar]
  22. Hooley Jill, Campbell Christine. Control and Controllability: Beliefs and Behaviour in High and Low Expressed Emotion Relatives. Psychological Medicine. 2002;32:1091–1099. doi: 10.1017/s0033291702005779. [DOI] [PubMed] [Google Scholar]
  23. Jacobs Selby C, Mason John, Kosten Thomas R, Kasl Stanislav V, Ostfeld Adrian M, Wahby Victor. Urinary Free Cortisol and Separation Anxiety Early in the Course of Bereavement and Threatened Loss. Biological Psychiatry. 1987;22:148–152. doi: 10.1016/0006-3223(87)90225-3. [DOI] [PubMed] [Google Scholar]
  24. Keenan Daniel M, Licinio Julio, Veldhuis Johannes D. A Feedback-Controlled Ensemble Model of the Stress-Responsive Hypothalamo-Pituitary-Adrenal Axis. Proceedings of the National Academy of Sciences USA. 2001;98:4028–4033. doi: 10.1073/pnas.051624198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kessler Ronald C, Gavin Andrews, Colpe Lisa J, Hiripi Eva, Mroczek Daniel K, Normand Sharon-Lise, Walters Ellen E, Alan Zavalasvsky. Short Screening Scales to Monitor Population Prevalence and Trends in Nonspecific Psychological Distress. Psychological Medicine. 2002;32:959–976. doi: 10.1017/s0033291702006074. [DOI] [PubMed] [Google Scholar]
  26. Kiecolt-Glaser Janice K, Glaser Ronald, Strain Eric C, Stout Julie C, Tarr Kathleen L, Holliday Jane E, Speicher Carl E. Modulation of Cellular Immunity in Medical Students. Journal of Behavioral Medicine. 1986;9:5–21. doi: 10.1007/BF00844640. [DOI] [PubMed] [Google Scholar]
  27. Konstantareas Mary M, Homatidis Soula, Plowright CMS. Assessing Resources and Stress in Parents of Severely Dysfunctional Children through the Clarke Modification of Holroyd's Questionnaire on Resources and Stress. Journal of Autism and Developmental Disorders. 1992;22:217–234. doi: 10.1007/BF01058152. [DOI] [PubMed] [Google Scholar]
  28. Larsen Randy J, Kasimatis Margaret. Day-to-Day Physical Symptoms: Individual Differences in the Occurrence, Duration, and Emotional Concomitants of Minor Daily Illnesses. Journal of Personality. 1991;59:387–423. doi: 10.1111/j.1467-6494.1991.tb00254.x. [DOI] [PubMed] [Google Scholar]
  29. Laurent, Annie, Gilvarry Catherine, Russell Ailsa, Robin Murray. Personality Dimensions and Neuropsychological Performance in First-Degree Relatives of Patients with Schizophrenia and Affective Psychosis. Schizophrenia Research. 2002;55:239–248. doi: 10.1016/s0920-9964(01)00280-8. [DOI] [PubMed] [Google Scholar]
  30. McEwen Bruce S. Stress, Adaptation, and Disease. Allostasis and Allostatic Load. Annals of the New York Academy of Sciences. 1998;840:33–44. doi: 10.1111/j.1749-6632.1998.tb09546.x. [DOI] [PubMed] [Google Scholar]
  31. Ockenfels Marget C, Porter Laura, Smyth Joshua, Kirschbaum Clemens, Hellhammer Dirk H, Stone Arthur S. Effect of Chronic Stress Associated with Unemployment on Salivary Cortisol: Overall Cortisol Levels, Diurnal Rhythm, and Acute Stress Reactivity. Psychosomatic Medicine. 1995;57:460–467. doi: 10.1097/00006842-199509000-00008. [DOI] [PubMed] [Google Scholar]
  32. Pruessner Jens C, Hellhammer Dirk H, Kirschbaum Clemens. Burnout, Perceived Stress, and Cortisol Responses to Awakening. Psychosomatic Medicine. 1999;61:197–204. doi: 10.1097/00006842-199903000-00012. [DOI] [PubMed] [Google Scholar]
  33. Sapolsky Robert M, Krey Lewis C, McEwen Bruce S. The Neuroendocrinology of Stress and Aging: The Glucocorticoid Cascade Hypothesis. Endocrine Reviews. 1986;7:284–301. doi: 10.1210/edrv-7-3-284. [DOI] [PubMed] [Google Scholar]
  34. Scott Lucinda V, Dinan Timothy G. Urinary Free Cortisol Excretion in Chronic Fatigue Syndrome, Major Depression and in Healthy Volunteers. Journal of Affective Disorders. 1998;47:49–54. doi: 10.1016/s0165-0327(97)00101-8. [DOI] [PubMed] [Google Scholar]
  35. Segerstrom Suzanne C, Miller Gregory E. Psychological Stress and the Human Immune System: A Meta-Analytic Study of 30 Years of Inquiry. Psychological Bulletin. 2004;130:601–630. doi: 10.1037/0033-2909.130.4.601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Seltzer Marsha M, Greenberg Jan S, Floyd Frank J, Pettee Yvette, Hong Jinkuk. Life Course Impacts of Parenting a Child with a Disability. American Journal of Mental Retardation. 2001;106:265–286. doi: 10.1352/0895-8017(2001)106<0265:LCIOPA>2.0.CO;2. [DOI] [PubMed] [Google Scholar]
  37. Serido, Joyce, Almeida David M, Wethington Elaine. Chronic Stressor and Daily Hassles: Unique and Interactive Relationships with Psychological Distress. Journal of Health and Social Behavior. 2004;45:17–33. doi: 10.1177/002214650404500102. [DOI] [PubMed] [Google Scholar]
  38. Thoits Peggy A. Stressors and Problem-Solving: The Individual as Psychological Activist. Journal of Health and Social Behavior. 1994;35:143–159. [PubMed] [Google Scholar]
  39. Thoits Peggy A. Differential Labeling of Mental Illness by Social Status: A New Look at an Old Problem. Journal of Health and Social Behavior. 2005;46:102–119. doi: 10.1177/002214650504600108. [DOI] [PubMed] [Google Scholar]
  40. Wilens Timothy E, Biederman Joseph, Adamson Joel, Monuteaux Michael, Henin Aude, Sgambati Stephanie, Santry Alison, Faraone Stephen V. Association of Bipolar and Substance Use Disorders in Parents of Adolescents with Bipolar Disorder. Biological Psychiatry. 2007;62:129–134. doi: 10.1016/j.biopsych.2006.11.022. [DOI] [PubMed] [Google Scholar]

RESOURCES