The Adolescent Response to Parental Illness: The Influence of Dispositional Gratitude
- First Online:
- Cite this article as:
- Stoeckel, M., Weissbrod, C. & Ahrens, A. J Child Fam Stud (2015) 24: 1501. doi:10.1007/s10826-014-9955-y
- 465 Downloads
Existing literature suggests that the children of ill parents are susceptible to anxiety and depression. However, some may respond to threatening circumstances such as chronic illness or grief with positive emotions and finding benefits, including appreciation of life and greater emotional strength. We explored the college-age child response to parental illness by examining the relationship of dispositional gratitude, family variables, and characteristics of parental illness to college-age child depression and anxiety. College students completed self-report measures examining dispositional gratitude, family quality of life, depression, and anxiety. In comparison to the healthy parent group, the ill parent group experienced lower family quality of life. There were no significant between group differences in depression or anxiety levels. Dispositional gratitude moderated the relationship between parental health status and college-age child anxiety and depression. This suggests that dispositional gratitude may have served as a buffer against internalizing symptoms within the college-age children of ill parents.
KeywordsChild response to parental illness Dispositional gratitude Family quality of life Positive psychology Protective factors
In addition to the usual obstacles of childhood and adolescence, many children must cope with the stress of growing up with an ill parent. Past research and conventional wisdom suggest that the children of physically or psychologically ill parents are significantly more susceptible to a number of psychosocial adjustment issues, including anxiety, depression, low self-esteem, and decreased social competence (Armistead et al. 1995; Siegel et al. 1992). In a review of the relevant research, however, Korneluk and Lee (1998) found that the presence of parental illness does not always lead to child maladjustment. Therefore, one might wonder whether there are specific conditions or traits that make certain children of ill parents either more or less vulnerable to depression and anxiety. For example, existing findings in the field of positive psychology suggest that certain individuals may respond to negative life events, such as death, traumas, or chronic illness, with positive emotions, including hope, optimism, and gratitude (Fredrickson et al. 2003; Nolen-Hoeksema and Davis 2002; Oltjenbruns 1991).
In many instances, when parents are ill, children may become the de facto leaders of the household, bringing on great responsibility and risks to development and psychological well-being. One potential risk to the children of ill parents is that these children will be “parentified,” taking on more responsibility than appropriate in order to compensate for their parents’ deficiencies (Bauman et al. 2009). Growing up with an ill parent, regardless of whether or not the child takes on caregiving responsibilities, may also lead to higher levels of depression, anxiety, and behavioral problems, as well as lower levels of self-esteem and social competence than the children of healthy parents (Armsden and Lewis 1994; Korneluk and Lee 1998; Siegel et al. 1992). Other studies suggest that the children of parents with chronic pain are generally maladjusted and have poorer social skills and higher reports of somatic concerns and headaches relative to the children of non-ill parents (Mikail and von Baeyer 1990). Existing work in this area also suggests that the child’s perceptions of stress and illness severity could be significant predictors of child psychosocial functioning in response to parental illness (Compas et al. 1994; Howes et al. 1994; Kotchick et al. 1997). Additionally, the child’s perception of certain family variables, such as affective responsiveness, affective involvement, roles, and communication, has been found to be associated with adolescent child psychological symptoms (Stanescu and Romer 2011). Therefore, understanding how children react to parent vulnerability would inform strategies to support these children’s psychosocial functioning.
While difficult life events have clear downsides, research shows that certain individuals manage to turn these events into positive changes (Fredrickson et al. 2003; Nolen-Hoeksema and Davis 2002; Oltjenbruns 1991). In a discussion of benefit finding in adverse events, Affleck and Tennen (1996) state that many people report responding to threatening circumstances, such as chronic illness, with the development of patience, tolerance, empathy, and courage. More specifically, bereavement studies suggest that in times of loss, some adults may experience developmental changes for the better by increasing empathy for others, building relationships with family members, and finding the positive in loss.
Research findings suggest that this ability to find benefits in loss occurs across many different kinds of events, ranging from traumas to natural disasters to personal losses. Thus, it appears that the ability to find benefits exists within certain individuals and is not isolated to specific situations (Nolen-Hoeksema and Davis 2002). Similar results have been observed in studies of adolescents’ response to grief. Oltjenbruns (1991) found that ninety-three percent of adolescents identified positive benefits from experiencing grief, such as having a deeper appreciation of life, strengthened emotional bonds with others, and greater emotional strength.
In this light, research has been done on the role of positive emotions in response to the September 11th attacks. Findings suggest that in addition to negative emotions, many people responded to this crisis by experiencing various positive emotions. For example, Fredrickson et al. (2003) found that while there were considerable reports of distress, many experienced feelings of gratitude, interest, and love following the attacks. Participants with high trait resilience tended to more frequently experience these positive emotions. Additionally, further analyses suggest that the experience of positive emotions was a key factor in allowing resilient individuals to thrive following September 11th. Therefore, it appears that certain individuals respond to negative events with positive emotions, such as gratitude. Consequently, these positive responses can lessen an individual’s vulnerability to psychopathology. The work of Fredrickson et al. (2003) is informative in that it not only investigates positive emotions in response to a negative life event but also explores the relationship between these positive emotions and internalizing symptoms.
Research examining the potentially positive responses of children to parental illness is limited. However, the existing literature suggests that some children and adolescents may experience benefits as a result of growing up with an ill parent. Consistent with past research, Pakenham et al. (2006) found that young caregivers between the ages of 10 and 25 years reported higher somatic complaints and lower life satisfaction in comparison to the children of healthy parents. However, the authors also found that the children of ill parents reported higher levels of perceived maturity on a self-report assessment of their experience with parental illness. Therefore, increased perceived maturity may be one potential positive outcome for children coping with parental illness.
In an examination of the effects of long-term parental illness, Johnston et al. (1992) suggest that the children of chronically ill parents are forced to become more independent, more tolerant, and more helpful. Responses of this nature may allow children to properly adapt to frightening or confusing changes within the family structure due to a parent’s condition. Families who focus on small gains and specific positive events may be able to cope more positively with parental illness.
The general consensus in existing work is that in the face of difficult life events, individuals can have a variety of responses, ranging from negative to positive. Those individuals who respond positively may be less vulnerable to the deleterious effects of potentially damaging circumstances, such as parental illness. The question still remains as to whether dispositional gratitude serves a protective role against internalizing symptoms in the children of ill parents.
Dispositional gratitude is a relatively new construct in the field of positive psychology. While gratitude is sometimes described as an emotion or a moral affect (Ahrens and Forbes 2014; McCullough et al. 2001; Wood et al. 2008), dispositional gratitude refers specifically to a life orientation towards noticing and appreciating the positive (Wood et al. 2010). This construct can be broken down into three components: simple appreciation, sense of abundance, and appreciation for others (Watkins et al. 2003). Research suggests that dispositional gratitude may be associated with a reduction in depressive symptoms (Wood et al. 2010). More specifically, in a study of the criterion validity of the Gratitude, Resentment, and Appreciation Test (GRAT), dispositional gratitude was negatively correlated with depression, as measured by the Beck Depression Inventory (BDI) (Watkins et al. 2003). Similarly, in a study of adolescents, dispositional gratitude was negatively correlated with depression, as measured by the Center for Epidemiologic Studies Depression Scale for Children (CES-DC) (Froh et al. 2011). Other work suggests that dispositional gratitude is correlated with lower anxiety scores, as measured by the state-trait anxiety inventory (STAI) (Krumrei and Pargament 2008).
Dispositional gratitude is one of many potentially protective psychological constructs and is of particular interest in our work. Existing literature suggests that heightened levels of gratitude may be associated with lower internalizing symptoms. Furthermore, dispositional gratitude has recently been directly incorporated into treatment through the use of gratitude journals. Studies suggest that gratitude can be cultivated (e.g., through gratitude journals), and that individuals who participate in gratitude interventions report greater life satisfaction and positive affect (Emmons and McCullough 2003).
In addition, the experience of dispositional gratitude has been shown to play a uniquely protective role for individuals responding to difficult circumstances. For example, as described above, Fredrickson et al. (2003) found that certain individuals responded to the September 11th terrorist attacks with gratitude. Those who reported this positive emotion following this event endorsed fewer depressive symptoms. The results of this investigation are relevant to our study in that they suggest that dispositional gratitude could serve as a buffer against depression and anxiety within the college-age children of ill parents. There are clear distinctions between these two distressing experiences, such as their chronicity, but some resilience factors buffer against a variety of stressors. To our knowledge, no studies to date have explored the role of dispositional gratitude in the child response to parental illness.
Our major aim was to examine the association of dispositional gratitude, parental illness characteristics, and family variables with depression and anxiety in college-age children of ill parents versus those with healthy parents. Given the population studied, participants provided a distal point of view of the child experience of growing up with a parent who is ill. More specifically, all participants were young adult college students living outside of the home, providing a retrospective report of their experiences growing up with a parent who was either ill or healthy. We explored two major questions in the present study. First, how will the college-age children of ill parents compare to the college-age children of healthy parents in regard to depression, anxiety, and perception of family quality of life? We hypothesized that the ill parent group would have higher levels of depression and anxiety and report lower family quality of life than the healthy parent group. Second, what is the impact of dispositional gratitude on the association between parental health status and internalizing symptoms? In addition, what is the impact of dispositional gratitude on the association between parental health status and family quality of life? We hypothesized that there would be a significant main effect of dispositional gratitude on internalizing symptoms, with increased gratitude associated with decreased depression and anxiety. We also hypothesized that dispositional gratitude would moderate the relationship between parental health status and depression and anxiety. In particular, we expected that parental health status would be more closely related to depression and anxiety among those lower in gratitude. Gratitude might play a particularly important role in buffering against depression and anxiety in the wake of stresses such as parental illness. No specific hypothesis was made in regard to family quality of life given the lack of existing literature to date.
The current study included a total of 136 predominantly Caucasian undergraduates with a mean age of 19 years (72 males, 64 females; 72.1 % Caucasian, 4.4 % African–American, 6.6 % Hispanic or Latino(a), 0.7 % Native American, 6.6 % Asian or Pacific Islander, 8.8 % Multi-racial). In order to be eligible for the study, participants were required to be at least 18 years of age and speak English.
Participants were divided into two groups: those with healthy parents and those will ill parents. Parents were designated as either healthy or ill according to participant categorization of parental illness on Question # 20 of the Demographics and Parental Health Information Questionnaire, developed for this study. The incapacitation scale used for Question # 20 was based on existing work examining the assessment of severity of diseases and the corresponding treatment implications (Tallarida et al. 1979). This scale, ranging from 1 to 7, provides an assessment of the degree of incapacitation and disruption of normal activity caused by parental illness. Participants who categorized their parent’s illness as three (a condition which may interfere with normal activity and which is incapacitating for intermittent periods of time) or higher were placed in the ill parent group. Participants who reported that their parents were healthy or categorized their parent’s illness as two (a condition whose symptoms do not occur with regularity and do not interfere with normal activity) or lower were placed in the healthy parent group. According to this classification system, the current study included 72 participants (39 males, 33 females) with healthy parents and 64 participants (33 males, 31 females) with either one or two ill parents. There were no significant differences in the major outcome variables between participants with one ill parent versus those with two ill parents.
Demographics and Parental Health Information
This questionnaire was created by the first author. In the first section, participants responded to questions related to age, sex, ethnicity, and family income. In the second section, participants responded to questions related to various aspects of parental illness.
The Gratitude Questionnaire (GQ-6) (McCullough et al. 2002)
The GQ-6 is a 6-item questionnaire used to measure an individual’s level of dispositional gratitude. The GQ-6 has demonstrated strong internal consistency in a number of past studies, with Cronbach’s α values from .76 to .84 (McCullough et al. 2002). The GQ-6 has shown significant correlations with positive affect (r = .31), life satisfaction (r = .53), negative affect (r = −.31) and depression (r = −.30), while also proving to be distinct from measures of other constructs such as happiness, optimism, and hope (Froh et al. 2011; McCullough et al. 2002). In the current study, GQ-6 demonstrated respectable internal consistency, with a Cronbach’s α of .75.
The Gratitude, Resentment, and Appreciation Test Revised Short Form (GRAT) (Watkins et al. 2003)
The GRAT is a 16-item scale measuring dispositional gratitude. The GRAT has demonstrated sound internal consistency in existing work, both overall (Cronbach’s α = .92) and for each of the three subscales [sense of abundance (Ab) = .88; simple appreciation (SA) = .90; appreciation of others (AO) = .76]. The GRAT has also been shown to be positively correlated with positive affect and negatively correlated with negative affect and depression (Watkins et al. 2003). In the present study, the GRAT appeared to have acceptable internal consistency (overall Cronbach’s α = .86; Ab = .85; SA = .78; AO = .78). In the current study, the GQ-6 and the GRAT were highly correlated with one another (r = .655, p < .001).
We elected to employ both the GQ-6 and the GRAT in the current study because they are the most commonly used measures of dispositional gratitude. Given the limited research in this area, we sought to maximize the amount of information gathered by using two rather than one measure of the construct of interest. Use of both measures also allows us to speak to their convergent validity.
The Positive and Negative Affect Schedule (PANAS) (Watson et al. 1988)
Studies have found dispositional gratitude to be correlated with positive affect. In the current study, positive affect was controlled for in order to account for potential overlap between this construct and dispositional gratitude. Participants were asked to report the extent to which they experienced a series of positive and negative emotions over the past week. The PANAS has had high internal consistency in past studies (Watson et al. 1988). The current study showed similar results, with a Cronbach’s α level of .84 for the Positive Affect Scale and .87 for the Negative Affect Scale.
Family Quality of Life Scale (FQOL) (Hoffman et al. 2006)
This scale was used to assess various facets of functioning and cohesion in families with and without disabled family members (Summers et al. 2007; Zuna et al. 2009). Four questions related to disability-related support were removed from the scale because they would not necessarily be applicable to all participants. The validity and reliability of this form of the FQOL has been confirmed in studies of families without disabilities (Zuna et al. 2009). The FQOL has been shown to have strong psychometric properties (Hoffman et al. 2006). The current study demonstrated sound internal consistency (Cronbach’s α = .93).
Center for Epidemiologic Studies Depression Scale (CES-D) (Radloff 1977)
The CES-D is a 20-item self-report measure used to assess depressive symptoms. This scale has been shown to be a reliable measure across a number of types of depressive symptoms (Radloff 1977). The current study showed strong internal consistency, with a Cronbach’s α of .89.
Beck Anxiety Inventory (BAI) (Beck and Steer 1990)
The BAI is a self-report questionnaire composed of 21 common symptoms of anxiety, such as “numbness or tingling,” “feelings of choking,” and “scared.” Respondents are asked to indicate, on a four-point Likert Scale ranging from “not at all” to “severely,” how much they have been bothered by each of the symptoms during the past month. The BAI has sound psychometric properties (Beck and Steer 1990). In the current study, it had strong internal consistency, with a Cronbach’s α of .92.
Approval to recruit participants was obtained from the university’s Institutional Review Board. Participants were recruited through the psychology department’s website, as well as via fliers posted throughout campus. Students wishing to participate were asked to contact the first author to set up an individual 30-min appointment to come into a laboratory in the psychology building. The entire study was completed on a computer via the Survey Monkey website. After the informed consent process, participants completed a series of self-report questionnaires. All participants completed the seven self-report measures in the following order: Demographics and Parental Health Information Questionnaire, the GQ-6, the GRAT, the PANAS, the CES-D, and the BAI. As compensation, participants were entered into a raffle to win a $50 gift card and received course credit if desired.
Intercorrelations of major outcome variables
Scores on the major outcome variables did not vary significantly according to race/ethnicity or combined average yearly income of primary household. In comparison to male participants, female participants reported higher scores on the BAI, F(1,134) = 9.873, p < .01, and on the GQ-6, F(1,134) = 6.384, p = .01. Given these differences, gender was controlled for in the main analyses.
With three main outcome variables (CES-D, BAI, and FQOL) and two main predictors (GQ-6 and GRAT), six setwise regressions were performed. In each, gender was entered as a control variable in the first step. Parental health status and dispositional gratitude (GQ-6 or GRAT) were entered in the second step, and so analyses of these two control for the other. Finally, the interaction of parental health status and dispositional gratitude (GQ-6 or GRAT) was entered in the third step. All variables were centered prior to the calculation of interaction terms. We will first report on effects of parental health status, then on gratitude, then on the interaction of the two.
Parental Health Status
We hypothesized that the college-age children of ill parents would exhibit higher levels of depression and anxiety than the college-age children of healthy parents. However there were no differences between the two groups in levels of depression, GQ-6: β = .236, t(132) = .150, p = .881, f2 = .19; GRAT: β = −.676, t(132) = −.456, p = .649, f2 = .36, or anxiety, GQ-6: β = 2.857, t(132) = 1.717, p = .088, f2 = .17; GRAT: β = 2.093, t(132) = 1.307, p = .193, f2 = .28.
In the analysis involving the GQ-6, the college-age children of ill parents reported lower FQOL than the college-age children of healthy parents, β = −4.776, t(132) = −2.240, p < .05, f2 = .43. However, when the GRAT was the measure of dispositional gratitude, there was no significant group difference on the FQOL: β = −3.111, t(132) = −1.633, p = .105, f2 = .83.
Dispositional gratitude (as measured by both the GQ-6 and the GRAT) was associated with depression, GQ-6: β = −.767, t(132) = −4.840, p < .01, f2 = .19; GRAT: β = −.293, t(132) = −6.784, p < .01, f2 = .267, anxiety, GQ-6: β = −.471, t(132) = −2.808, p < .01, f2 = .17; GRAT: β = −.217, t(132) = −4.655, p < .01, f2 = .29, and family quality of life, GQ-6: β = 1.504, t(132) = 6.992, p < .01, f2 = .43; GRAT: β = .552, t(132) = 9.935, p < .01, f2 = .83. More grateful people had less depression and anxiety and greater family quality of life.
Dispositional gratitude moderated the relationship between parental health status and depression, GQ-6: β = −.738, t(131) = −2.357, p < .05, f2 = .25. We examined this interaction of parental health status and gratitude by using procedures described by Cohen and Cohen (1983). We looked at predicted CES-D scores for those with ill and healthy parents who were one standard deviation above or below the mean on gratitude. Among those who were low in gratitude, parental illness was associated with higher CES-D scores (predicted values of 20.72 for the ill-parent group, but 17.3 for the healthy parent group). In contrast, among those high in gratitude, the tendency was for having an ill parent to be associated with lower predicted CES-D scores (9.88) than having a healthy parent (12.86). Therefore, parental illness is less associated with a negative outcome among those who are more grateful, consistent with a buffering effect. The interaction effect of the GRAT and parental health status on depression was not statistically significant, GRAT: β = −.080, t(131) = −.909, p = .365, f2 = .39.
Dispositional gratitude (as measured by both the GQ-6 and the GRAT) also moderated the relationship between parental health status and anxiety, GQ-6: β = −1.064, t(131) = −3.258, p < .01, f2 = .27; GRAT: β = −.243, t(131) = −2.594, p < .01, f2 = .36. We used the same procedure as for the CES-D to examine the interaction. For those low in gratitude, having an ill parent was associated with higher predicted anxiety (17.39) than having healthy parents (9.33). In contrast, for those high in gratitude, having an ill parent was actually associated with lower predicted anxiety (7.62) than having healthy parents (10.01). The pattern was similar for the GRAT. Once again, illness is less associated with a negative outcome for those who are most grateful, consistent with a buffering effect.
The interaction effect of dispositional gratitude and parental health status on family quality of life was trending toward significance, GQ-6: β = .812, t(131) = 1.915, p = .058, f2 = .477; GRAT: β = .218, t(131) = 1.941, p = .054, f2 = .52. Using the same procedure as described earlier, we decomposed the interaction. For those low in gratitude, having an ill parent was related to lower predicted family quality of life (73.13) than having healthy parents (81.57). This relationship was weaker for those high in gratitude, as those with an ill parent had an expected FQOL score of 91.35, whereas those with healthy parents had an expected score of 93.24. We found similar effects when examining results from the GRAT. Once again, illness matters less for those high in gratitude, suggesting a buffering effect.
Additional Post-hoc Analyses
Existing literature illuminates the potential impact of various parental illness factors on child internalizing symptoms in response to this life stressor (Compas et al. 1994; Howes et al. 1994; Kotchick et al. 1997). Therefore, we examined the relation of a number of parental illness factors to college-age child depression and anxiety. The following variables were not significantly correlated with college-age child depression and anxiety: parental illness type (physical vs. mental illness), child age at onset of parental illness, child report of stress experienced due to parental illness, child report of feeling upset due to parental illness, child report of parental illness severity, and whether or not the ill parent was the primary caregiver. There was a significant correlation between course of onset of parental illness (chronic vs. acute) and college-age child depression, r = −.330, p < .01, with chronic parental illness associated with greater college-age child depression than acute parental illness. Therefore, a series of setwise regressions were conducted to examine the potential interaction of course of onset of parental illness and dispositional gratitude on depression and anxiety.
Results of the setwise regressions indicate that in comparison to the college-age children of parents with an acute illness, college-age children of parents with a chronic illness reported higher scores on the CES-D, GQ-6: β = −5.175, t(61) = −2.014, p < .05, f2 = .51; GRAT: β = −5.368, t(61) = −2.179, p < .05, f2 = .62. Furthermore, the interaction of course of illness onset (chronic vs. acute) with dispositional gratitude on depression was significant, GQ-6: β = 1.178, t(60) = 2.249, p < .05, f2 = .64; GRAT: β = .377, t(60) = 3.016, p < .01, f2 = .87.
To decompose this interaction, we once again looked at predicted CES-D scores for those one standard deviation above or below the mean in gratitude, as well as the impact of being in the acute or chronic illness group. For those low in gratitude, having a parent with a chronic illness was associated with much higher predicted CES-D scores (25.94) than having a parent with an acute illness (14.43). In contrast, among those high in gratitude, it made little difference whether one’s parent had a chronic illness (predicted score of 10.50) or an acute illness (10.53). Once more, we found similar effects for the GRAT. These results are consistent with the possibility that gratitude is a particularly strong buffer for those whose parents have had a chronic illness.
The main effect of course of illness onset (chronic vs. acute) on anxiety was not significant, GQ-6: β = −2.933, t(61) = −1.021, p = .311, f2 = .34; GRAT: β = −2.942, t(61) = −1.085, p = .282, f2 = .49. However, the interaction of course of illness onset with dispositional gratitude on anxiety was significant, GQ-6: β = 1.328, t(60) = 2.269, p < .05, f2 = .46; GRAT: β = .360, t(60) = 2.566, p = .01, f2 = .65.
For those low in gratitude, having a parent with a chronic illness was associated with much higher anxiety (predicted BAI of 22.05) than having a parent with an acute illness (11.82). In contrast, for those high in gratitude, having a parent with a chronic illness was, if anything, associated with lower predicted BAI scores (6.81) than having a parent with an acute illness (9.60). Once again, we found similar results for the GRAT. As with depression, this suggests that the effect of having a parent with a chronic illness on anxiety is buffered by gratitude.
Past research suggests that growing up with an ill parent can leave children vulnerable to a variety of psychosocial adjustment issues, including heightened levels of depression and anxiety (Armistead et al. 1995; Siegel et al. 1992). Other work, however, describes that certain individuals may respond to negative events by experiencing positive emotions, such as hope, optimism, and gratitude (Fredrickson et al. 2003; Nolen-Hoeksema and Davis 2002; Oltjenbruns 1991). The purpose of our study was to explore the college-age child response to parental illness by examining the potential association of dispositional gratitude, parental illness characteristics, and family variables with college-age child vulnerability to depression and anxiety.
We hypothesized that in comparison to the college-age children of healthy parents, the college-age children of ill parents would exhibit lower family quality of life, as well as increased levels of depression and anxiety. Consistent with this hypothesis, the college-age children of ill parents reported significantly lower FQOL scores than the college-age children of healthy parents. This finding is in line with prior work suggesting that the child’s perception of family variables may be correlated with child psychosocial functioning in response to parental illness (Stanescu and Romer 2011). Surprisingly, there were no significant group differences in depression or anxiety. These results are inconsistent with existing literature suggesting that the children of ill parents are vulnerable to a number of psychological concerns, including heightened depression and anxiety (Armistead et al. 1995; Hirsch et al. 1985).
Despite these unexpected findings, the results pertaining to gratitude revealed information that contributes to the understanding of factors that may influence internalizing symptoms in college students. First, there was a significant main effect of dispositional gratitude on depression, anxiety, and family quality of life across the entire sample. This is consistent with our hypothesis as well as with past research suggesting that dispositional gratitude is inversely correlated with internalizing symptoms (Froh et al. 2011; Krumrei and Pargament 2008; Watkins et al. 2003).
One of our major goals was to explore the potential impact of dispositional gratitude on the association between parental health status (ill vs. healthy) and depression, anxiety, and family quality of life. A series of regressions revealed the moderating effect of dispositional gratitude on the relation of parental health status to depression, anxiety, and, marginally, family quality of life. For five out of six analyses, the relation of parental illness to higher internalizing symptoms and lower family quality of life was stronger, or marginally stronger, among those lower in gratitude. There are a number of possible explanations for this interaction effect. First, existing literature suggests that in periods of heightened stress or in coping with significant life stressors (i.e., death of a loved one, traumatic events), the experience of positive emotions and finding benefits may buffer against internalizing symptoms (Fredrickson et al. 2003; Nolen-Hoeksema and Davis 2002; Oltjenbruns 1991). However, individuals low in positive emotions, such as gratitude, may be particularly vulnerable to heightened internalizing symptoms as they attempt to cope with a difficult life event (i.e., parental illness). The college-age children of ill parents with lower levels of dispositional gratitude may not have experienced the protective nature of this construct, thereby resulting in higher depression and anxiety.
Second, some college-age children may have responded to the potentially negative effects of growing up with an ill parent (i.e., heightened depression and anxiety and decreased family quality of life) in adaptive ways. One of these adaptive responses may have been the experience of dispositional gratitude. This explanation is consistent with prior work suggesting that certain individuals responded to the September 11th terrorist attacks with positive emotions, such as gratitude, hope, and love (Fredrickson et al. 2003).
Finally, there could be a third variable explanation for the moderating effect of gratitude on the association of parental health status with depression, anxiety, and family quality of life. Fredrickson et al. (2003) found that individuals high in trait resiliency were most likely to respond to the September 11th terrorist attacks with positive emotions. Consequently, these individuals were able to thrive following this negative life event. Although not directly assessed in the current study, it is possible that a third variable, such as resiliency, could be contributing to the apparent interaction effect of gratitude and parental health status.
Post-hoc analyses revealed that in comparison to the college-age children of parents with an acute illness, the college-age children of parents with a chronic illness reported higher depressive symptoms. The main effect of onset of illness on anxiety was not statistically significant. However, there was a significant interaction effect of gratitude and onset of illness on both depression and anxiety. Specifically, the association of chronic parental illness (vs. acute parental illness) with higher depression and anxiety was stronger among those lower in gratitude. These findings are consistent with existing literature suggesting that increased illness demands and disruption to family functioning can lead to lower psychological well-being and greater behavioral, academic, and social problems in the children of chronically ill parents (Anderson and Hammen 1993; Armistead et al. 1995; Lewis et al. 1993). Perhaps gratitude is particularly important for coping with chronic strains. It is also possible that chronic illnesses, which are typically longer lasting, provided children with more of an opportunity to reflect on parental illness in positive, meaningful ways (i.e., with gratitude). The one study we know of that has examined gratitude as a stress buffer (Fredrickson et al. 2003) explored a more acute stressor but nevertheless found gratitude to be a buffer. Taken together with our study, this suggests that gratitude can be associated with resilience to both chronic and acute stress.
We elected to employ both the GQ-6 and the GRAT because they are the most commonly used measures of dispositional gratitude in the field. Given that investigations of positive emotions in the children of ill parents are limited, at best, we aimed to maximize the amount of information gathered in regard to dispositional gratitude. Furthermore, the convergent validity of the two measures can only be determined by employing them both in the same study. Our preliminary analyses indicate that the GQ-6 and the GRAT were highly correlated and largely produced similar but not always identical results. Specifically, the GQ-6 (but not the GRAT) predicted lower family quality of life. In addition, the GQ-6 (but not the GRAT) moderated the relationship between parental health status and depression. It will be important in future work on dispositional gratitude to use multiple measures so as to further assess their relative explanatory power.
The current study’s reliance on non-experimental methods and measures collected at a single time makes it difficult to establish the definitive direction of causality. In their study of the September 11th terrorist attacks, Fredrickson et al. (2003) were better equipped to establish temporal precedence because they gathered pre-crisis assessments of trait resiliency and post-crisis assessments of positive emotions and depression. In the context of our study, it is unclear whether variables of interest manifested themselves in response to parental illness or existed within the college-age children irrespective of negative life events.
In addition, the difference in participant age in the current sample as compared to child samples used in previous studies of parental illness may make it difficult to draw parallels between existing work and our study (and may partly explain the lack of significant group differences in depression and anxiety). We explored the child response to parental illness from a distal perspective, as all participants were currently enrolled in college. Existing literature, however, has typically relied on samples of children living in the home, either due to young age or an obligation to take on the caregiving role (Armistead et al. 1995; Dellmann-Jenkins and Blankemeyer 2009). Also, perhaps adolescent children experiencing the detrimental effects described in existing literature are less likely to be found in a university sample. Nonetheless, the results perhaps provide information about how the experience of having grown up with a parent who is ill affects children who are now away at college.
Despite these differences from earlier studies and potential limitations, our study represents a step in better understanding the college-age child experience of parental illness. Certain variables, such as increased levels of dispositional gratitude, may have served as buffers against internalizing symptoms, suggesting that the child response to parental illness is not inherently negative. Furthermore, there may be certain traits that allow certain children of ill parents to thrive more than others. These findings are promising and speak to the need to continue to explore the potentially protective factors that can influence children who experience family stressors.