Abstract
Objective
To determine the differential impact of potentially traumatic events (PTEs) and other stressful life events on psychological functioning in two groups of children: those with cancer, and those without history of serious illness.
Methods
Children with cancer aged 8–17 (n=254) and age-, sex-, and race/ethnicity-matched controls (n=142) completed self-report measures of stressful life events, and psychological functioning. Stressful life events included those that may meet DSM-IV A1 criteria (PTEs; 9 events) and others that would likely not (other events; 21 events).
Results
Children with cancer endorsed significantly more PTEs than control children. There were no differences between groups in number of other events experienced. Hierarchical regression analyses revealed that number of other events accounted for significant variance in psychological functioning, above and beyond group status, demographic factors (age and SES) and number of PTEs.
Discussion
The number of cumulative other events experienced is a significant predictor of psychological functioning in both youth with serious illness and controls. In contrast, cumulative PTEs appear to have a minor (albeit significant) impact on children’s psychological functioning. Assessment of psychological functioning would benefit from a thorough history of stressful life events, regardless of their potential traumatic impact.
Keywords: life events, psychological functioning, potentially traumatic events, childhood cancer
Stress is frequently highlighted as a cause of or contributor to mental and physical health problems across the life span. The stress reaction can come in many forms, but is often precipitated by a life event – albeit one that can be positive or negative, traumatic or non-traumatic. Past research has suggested that negative life events are particularly indicative of the development or intensification of mental health concerns (Cameron, Palm, & Follette, 2010; Furniss, Beyer, & Müller, 2009), with some debate regarding the necessary severity of those events (Lancaster, Melka, & Rodriguez, 2009; Robinson & Larson, 2010). For example, the diagnosis of post-traumatic stress disorder (PTSD) requires a traumatic precipitating negative life event, and in particular, one that “involved actual or threatened death or serious injury, or a threat to the physical integrity of self or others” (DSM-IV A1 criteria; American Psychiatric Association, 2000, p. 467). In DSM-5, the diagnosis of PTSD is slightly more stringent and requires “exposure to actual or threatened death, serious injury or sexual violence” (DSM-5 A criteria; American Psychiatric Association, 2013, p. 271). However, recent research has suggested that negative life events that would not meet DSM-IV/ DSM-5 criteria may be just as, if not more, indicative of mental health concerns than those that do (Copeland, Keeler, Angold, & Costello, 2010; Furniss et al., 2009; Gold, Marx, Soler-Baillo, & Sloan, 2005; Van Hoof, McFarlane, Baur, Abraham, & Barnes, 2009; Verlinden et al., 2013).
Children with cancer have experienced a DSM-IV PTSD-qualifying negative life event, and as such, are assumed to meet A1 criteria. (Diagnosis of a medical illness such as cancer may no longer qualify under DSM-5). They have been diagnosed with an illness that is potentially life-threatening and, as part of their treatment, will likely undergo numerous procedures that pose a threat to their physical integrity (e.g., surgery, chemotherapy, radiation therapy). As such, a significant body of research has suggested that they are at risk for the development of PTSD and related symptoms (see Bruce, 2006 for a review). However, numerous studies have revealed that children diagnosed with cancer demonstrate no more symptoms of PTSD than controls (Barakat et al., 1997; Phipps, Jurbergs, & Long, 2009; Phipps et al., 2014). Indeed, these studies suggest that children with cancer demonstrate increased resilience in the face of this potential trauma, and may even exhibit fewer mental health symptoms than their non-affected peers (Howard Sharp, Rowe, Russell, Long, & Phipps, in press; Maurice-Stam et al., 2009; Phipps et al., 2014; Wechsler & Sánchez-Iglesias, 2013).
Past research has suggested that the interplay between the diagnosis of cancer and the accumulation of stressful life events may predispose children with cancer to experience symptoms of post-traumatic stress (Currier, Jobe-Shields, & Phipps, 2009). Specifically, in a study of 121 pediatric patients with cancer, the number of stressful life events experienced was a unique and significant predictor of symptoms of post-traumatic stress related to the cancer diagnosis. Notably, however, the authors did not attempt to determine the impact of events of varying severities on psychological functioning. Rather, they combined their assessment of events that were both potentially traumatic (e.g., the death of a parent), as well as those that were not (e.g., birth of a sibling). Additionally, Currier and colleagues’ (2009) sample was limited to children with cancer and they did not have a comparison group. As such, it is difficult to determine whether their findings are reflective of all children, or just those affected by a serious illness, and thus those who have experienced a potentially traumatic event.
The objective of the current paper was to determine the unique and differential impact of potentially traumatic events that could meet DSM-IV A1 criteria and other events (non-traumatic, non A-1) on the psychological functioning (PTSD/PTSS, anxiety and depression) of children with cancer as compared to community controls. As such, we aimed to both replicate and extend the work of Currier and colleagues (2009). Based on the extant literature, it was hypothesized that non-traumatic significant life events would have a similar impact on psychological functioning as potentially traumatic events. Secondly, it was hypothesized that there would be a greater impact of events, both PTEs and other events, on the psychological functioning of children with cancer in comparison to their healthy peers.
Methods
Participants
Cancer
Children with cancer (n=254) were recruited from the outpatient clinics of a large children’s cancer center in the mid-South to participate in the current study. Eligibility criteria included: 8–17 years old, able to read and speak English, ≥1 month from diagnosis of a malignancy, and no significant cognitive deficits that would prevent completion of measures. Of 378 patients approached, 258 (68%) agreed to participate and 254 provided usable data. Those who declined cited lack of interest or lack of time. There were no differences between those who participated and those who declined with regards to age, gender, race/ethnicity, and diagnosis.
Community Controls
Children without a history of a chronic or life threatening illness were recruited from local schools via a two-step process. First, permission slips were provided to children to create a database of possible participants. Then, children from this database were contacted on the basis of demographic match on age, sex, and race/ethnicity. Similar to cancer patients, community controls were required to be between the age of 8–17, able to read and write English, and to have no significant cognitive deficits that would prevent completion of measures. Of 169 potential participants contacted, 142 (84%) agreed to participate and completed study measures.
Demographic information for all participants and diagnostic and treatment information for the cancer group is available in Table 1. There were no differences between the two groups on age, race/ethnicity or gender. However, there was a significant difference in SES (χ2=12.54, p<.01), such that there were more participants in the cancer group in the low SES category.
Table 1.
Demographic and cancer-specific characteristics
| Cancer (n = 254) | Control (n = 142) | |||
|---|---|---|---|---|
| M ± SD | N (%) | M ± SD | N (%) | |
| Age (years) | 13.09 ± 2.89 | 12.68 ± 2.89 | ||
| Gender | ||||
| Male | 132 (52.0) | 74 (52.1) | ||
| Female | 122 (48.0) | 68 (47.9) | ||
| Race/Ethnicity | ||||
| Caucasian | 184 (72.4) | 109 (76.8) | ||
| African-American | 58 (22.8) | 29 (20.4) | ||
| Other | 12 (4.8) | 4 (2.8) | ||
| Socioeconomic Statusa | ||||
| Groups I & II | 70 (27.5) | 58 (40.8) | ||
| Group III | 81 (31.9) | 50 (35.2) | ||
| Groups IV & V | 103 (40.6) | 34 (24.0) | ||
| Cancer-Specific Characteristics | ||||
| Age at diagnosis (years) | 9.29 ± 4.74 | |||
| Months since diagnosis | 45.06 ± 51.57 | |||
| Diagnostic Category | ||||
| ALL | 61 (24.0) | |||
| AML or APL | 18 (7.1) | |||
| Lymphoma | 34 (13.4) | |||
| Solid Tumor | 99 (39.0) | |||
| Brain Tumor | 42 (16.5) | |||
| Treatment Intensity | ||||
| Least intensive | 14 (5.5) | |||
| Moderately intensive | 83 (32.7) | |||
| Very intensive | 83 (32.7) | |||
| Most intensive | 58 (22.8) | |||
| Treatment Status | ||||
| On Therapy | 147 (57.9) | |||
| Off Therapy | 107 (42.1) | |||
Abbreviations: ALL = acute lymphoblastic leukemia; AML = acute myeloid leukemia; APL = acute promyelocytic leukemia
Socioeconomic status was determined by the Barratt Simplified Measure of Social Status (Barratt, 2006).
Procedure
Participants and an adult caregiver provided informed consent/assent using IRB-approved methods. Participants from the cancer group completed measures during a routine hospital visit. Control group participants also completed questionnaires during an individual appointment at the hospital; evening and weekend hours were available as necessary. All participants were generally able to complete questionnaires independently; however, trained research assistants were available to answer questions and to read aloud the forms if necessary. Youth were separated from their parent during questionnaire completion to prevent discussion. All participants and their caregivers were provided with a small monetary incentive for their time and participation.
Measures
Life Events Scale for Children (LESC)
The LESC is a 30-item measure that assesses the experience of different stressful life events. The scale is a modified version of the Coddington Life Events Questionnaire (Coddington, 1972) which includes both positive and negative life events. It was first amended by Johnston and colleagues (2003) to focus exclusively on 22 negative life events. An additional 8 items were added in this iteration. Children are asked to report on whether they had ever experienced the event, with the primary variable of interest being the total number of life events experienced. The 30 events include both normative significant stressful life events (e.g., argument with a parent, death of a pet, parental divorce), as well as those that would potentially meet DSM criteria (e.g., witnessed an act of violence, involved in a car accident). As such, additional variables of interest for the current study included the number of potentially A1 events, referred to as “PTEs” (9 possible) and number of non-A1 events, referred to as “other events” (21 possible). While the possibility of weighting individual events by severity of the event was considered, we elected to count events. Notably, in the Currier and colleagues (2009) study, analyses were completed with both events that were weighted by severity and with those that were simply summed. Results were very similar, suggesting that weighting events for severity did not add significant additional information. Moreover, the simple summing of events is a technique also used by the Adverse Childhood Experiences (ACE) study (Felitti et al., 1998) and other large cohort studies (e.g., Dube, Felitti, Dong, Giles, & Anda, 2003).
UCLA PTSD Reaction Index for DSM-IV (UCLA PTSDI; Pynoos, Rodriguez, Steinberg, Stuber, & Frederick, 1998; Steinberg, Brymer, Decker, & Pynoos, 2004)
The UCLA PTSDI is a 22-item self-report measure of how frequently in the past month children experienced the re-experiencing, avoidance, and arousal symptoms that characterize PTSD. Children are asked to identify their most stressful or traumatic event and then are directed to respond to all questions based on this event on a 4-point Likert scale. Children with cancer were not prompted to choose cancer as their most stressful event, but were allowed to independently identify an event of their choosing (as were controls; see Table 2 for events chosen by participants). The primary variable of interest was the total score; scores above 38 are considered clinically significant (Steinberg et al., 2004).
Table 2.
List of traumatic stressors spontaneously identified and described by participants on the UCLA PTSDI. Events are delineated by group.
| Cancer Group (n = 254) | Control Group (n = 142) |
|---|---|
| Different Events | |
| Cancer* | Fight with a parent |
| Bike wreck | Getting lost |
| Death of a parent | Strep Throat |
| Witnessed a shooting | Shots |
| Fear of the future | Finding out needed to have surgery |
| Parental deployment | Hearing about an accident |
| Watching a scary movie | Items stolen / Break-in |
| Haunted House | |
| Accused of wrongdoing when innocent | |
| Punishment (being spanked, threat of punishment) | |
| Moving / new school | |
| Mission trip | |
| Report made to Children and Family Services | |
| Siblings quirks | |
| Potential parent divorce | |
| Foreclosure | |
| Divorce of family members (aunt/uncle, grandparents) | |
| Feeling that couldn’t breathe (from coughing or falling) | |
| Similar Events | |
| Death of a family member (cousin, aunt, grandparent, great-grandparent) | |
| Death of a friend/family friend | |
| Injury or illness in a family member | |
| Natural disaster / severe weather (severe thunderstorm, hurricane) | |
| Car wreck | |
| Injuries/illness that required hospital visit (stitches, broken bones, asthma attack) | |
| Death of a pet | |
| School issues (testing, homework, bad grades, not being prepared, unfriendly school) | |
| Separation of family (sibling moved out, lived with grandparents for short time) | |
| Specific phobias (clowns, snakes, dogs, monsters) | |
| Parental divorce | |
| Difficulties with a relationship (break-up, fight with a friend) | |
Cancer (or a related topic) was spontaneously identified by 133 (52.4%) participants in the cancer group
Children’s Depression Inventory (CDI; Kovacs, 1992)
The CDI is a frequently used measure of children’s depressive symptoms. Respondents choose one of three items that best applies across 27-items, with scores above 16 indicating clinically-significant symptomatology (Timbremont, Braet, & Dreessen, 2004). The total score was used for the current study.
Screen for Child Anxiety Related Emotional Disorders (SCARED; Birmaher et al., 1997)
The SCARED is a 41-item measure of anxiety, with items rated using a 3-point Likert scale. Subscales assess five specific anxiety disorders (e.g., Generalized Anxiety Disorder, School Avoidance) and the measure demonstrates sound psychometric properties (Birmaher et al., 1999). A total score was used, with scores above 25 considered clinically significant (Birmaher et al., 1999).
Analytical Plan
To assess group differences in life events, chi-square analyses were used to determine statistical differences in the proportion of youth from each group who endorsed each event. Independent sample t-tests were conducted to determine group differences in cumulative life events, including PTEs, other events, and total events. To assess possible influences of demographic factors on life events, a multiple regression analysis was conducted to determine the impact of group status (cancer vs. control) and theoretically relevant demographic factors (age, gender, SES, race).
The differential impact of PTEs and other events on psychological functioning was assessed via a two-step approach. First, Pearson correlations were computed to determine the association between frequency of life events (PTEs, other events, total events) and psychological functioning. Analyses were completed both with the sample as a whole and separately for each group (cancer and control). Fisher’s z comparisons were conducted to determine the difference in magnitude for correlations with PTEs and other events. Second, a series of hierarchical regression analyses were conducted to determine the impact of PTEs versus other events on psychological functioning. Order of entry was consistent for each outcome measure: 1) group status (cancer vs. control); 2) significant demographic factors from the above multiple regression; 3) number of PTEs; and 4) number of other events. Exploratory analyses assessed the potential moderation of group status on the impact of events on psychological functioning.
Results
Preliminary Analyses
Preliminary analyses were completed to characterize the psychological functioning of our sample with descriptive analyses for the UCLA PTSDI, SCARED, and CDI in Table 3. Mean scores are well below the clinical cutoff and within the normative range for both groups. There was a range of levels of distress, with 6 to 33% of the sample (depending on the measure) scoring above clinical cutoffs. There were no group differences across measures, both with regards to means as well as the proportion of the sample that was above the clinical cutoff.
Table 3.
Descriptive statistics of psychological functioning.
| Cancer (n = 254) | Control (n = 142) | |
|---|---|---|
| UCLA PTSD Reaction Index for DSM-IV | ||
| Mean | 18.29 | 18.73 |
| Standard Deviation | 13.83 | 14.67 |
| Range | 0 – 64 | 0 – 76 |
| Number (Percent) above clinical cutoffa | 28 (11%) | 16 (11.3%) |
| Screen for Child Anxiety Related Emotional Disorders | ||
| Mean | 18.67 | 20.99 |
| Standard Deviation | 11.89 | 12.04 |
| Range | 0 – 82 | 1 – 57 |
| Number (Percent) above clinical cutoffb | 72 (28.3%) | 48 (33.8%) |
| Children’s Depression Inventory | ||
| Mean | 6.45 | 6.73 |
| Standard Deviation | 5.40 | 5.48 |
| Range | 0 – 28 | 0 – 25 |
| Number (Percent) above clinical cutoffc | 16 (6.3%) | 10 (7.0%) |
Note. There were no significant differences in psychological functioning between groups.
Clinical cutoff = 38
Clinical cutoff = 25
Clinical cutoff = 16
Frequency of Events
The frequency of life events is listed in Table 4. Out of 30 possible events, children with cancer indicated an average of 8.12 events (SD = 3.84, range 2–22), whereas control children reported an average of 7.25 events (SD = 3.42, range 0–18). Importantly, one of the events listed is “serious illness.” Since all children with cancer could endorse this event (79.5% did), and no comparison children should have endorsed this event (9.9% did), we elected to temporarily remove this item from the scale. With that item removed, the average number of events endorsed by children with cancer (7.33 ± 3.76) and control children (7.15 ± 3.36) was comparable.
Table 4.
Frequency of identified life events, delineated by cancer vs. control
| Cancer (n = 254) | Control (n = 142) | ||
|---|---|---|---|
| N (%) | N (%) | Χ2 | |
| Potentially Traumatic Events | |||
| A parent died | 11 (4.3) | 0 (0.0) | 6.32* |
| A grandparent died | 149 (58.7) | 93 (65.5) | 1.79 |
| A relative/close friend died | 158 (62.2) | 83 (58.5) | 0.54 |
| Abused (and reported) | 10 (3.9) | 6 (4.2) | 0.02 |
| Witnessed shooting/act of violence | 21 (8.3) | 5 (3.5) | 3.34 |
| Involved in natural disaster | 97 (38.2) | 38 (26.8) | 5.29* |
| Car crash or serious accident | 67 (26.4) | 31 (21.8) | 1.01 |
| Injured and had to go to ER | 89 (35.0) | 38 (26.8) | 2.86 |
| Serious illness | 202 (79.5) | 14 (9.9) | 178.30** |
| Total Events [M ± SD (Range)]a | 3.16 ± 1.51 (0 – 7) | 2.17 ± 1.17 (0 – 5) | 6.82** |
| Total Events (no illness) [M ± SD (Range)]a | 2.37 ± 1.38 (0 – 6) | 2.07 ± 3.36 (0 – 5) | 2.22* |
| Other Events | |||
| A brother or sister was born | 155 (61.0) | 92 (64.8) | 0.55 |
| Your parents separated | 89 (35.0) | 38 (26.8) | 2.86 |
| Your parents divorced | 66 (26.0) | 36 (25.4) | 0.02 |
| Your parent remarried | 52 (20.5) | 23 (16.2) | 1.08 |
| A parent became sick, hospitalized | 56 (22.0) | 33 (23.2) | 0.07 |
| A sibling became sick, hospitalized | 38 (15.0) | 18 (12.7) | 0.39 |
| Learned you were adopted | 6 (2.4) | 1 (0.7) | 1.44 |
| Parent lost a job and was unemployed | 79 (31.1) | 40 (28.2) | 0.37 |
| Parent’s job changed, less time at home | 44 (17.3) | 38 (26.8) | 4.94* |
| Parents began fighting | 32 (12.6) | 25 (17.6) | 1.85 |
| Parent went to jail | 41 (16.1) | 8 (5.6) | 9.28** |
| Moved to a new city | 66 (26.0) | 49 (34.5) | 3.21 |
| Parents worrying more about money | 90 (35.4) | 67 (47.2) | 5.25* |
| Drug/alcohol problems in family | 50 (19.7) | 15 (10.6) | 5.52* |
| Bullied at school or threatened | 43 (16.9) | 38 (26.8) | 5.41* |
| Pet died | 173 (68.1) | 82 (57.7) | 4.27* |
| Failed an important exam | 45 (17.7) | 35 (24.6) | 2.71 |
| Suspended or expelled | 22 (8.7) | 13 (9.2) | 0.03 |
| Attend court for wrongdoing | 6 (2.4) | 3 (2.1) | 0.03 |
| Boy/girlfriend broke up with you | 70 (27.6) | 38 (26.8) | 0.03 |
| Serious argument with parents | 36 (14.2) | 29 (20.4) | 2.59 |
| Total Events [M ± SD (Range)]a | 4.96 ± 2.99 (0 – 16) | 5.08 ± 2.84 (0 – 15) | 0.39 |
| All Events Combined | |||
| Total Events [M ± SD (Range)]a | 8.12 ± 3.84 (2 – 22) | 7.25 ± 3.42 (0 – 18) | 2.26* |
| Total Events (no illness) [M ± SD (Range)]a | 7.33 ± 3.76 (1 – 21) | 7.15 ± 3.36 (0 – 18) | 0.47 |
Independent sample t-test
p < .05
p < .01
Given our interest in the influence of potentially traumatic events (PTEs), the 30 life events were subdivided into events that could meet A1 criteria (PTEs; 9 events) and those that would likely not (other events; 21 events). Children with cancer endorsed significantly more PTEs than control children (3.16 ± 1.51 vs. 2.17 ± 1.17, t = 6.815, p <.001). This difference remained when “serious illness” was removed (t = 2.22, p <.03). Items more commonly endorsed by children with cancer included: death of a parent, witnessing an act of violence, involvement in a natural disaster, and emergency room visits. No PTEs were more commonly endorsed by control children. See Table 4 for all items.
In contrast, there were no significant differences in the number of other events reported by children with cancer and control children, though there was some variability in the individual events endorsed by each group (Table 4). Specifically, children with cancer were more likely to endorse: parental separation, parent jailed, drug/alcohol problems in the family, and death of a pet. In contrast, control children were more likely to endorse: change in a parent’s job, parents’ increased worry about money, and bullying at school.
Given the significant difference in SES between the two participant groups, questions were raised regarding the potential impact of demographic variables on cumulative life events. As such, a multiple regression was computed with demographic variables – group status (cancer vs. control), age, gender, SES, and race – as predictors of cumulative life events. Results revealed that the overall model was significant [F(5, 390) = 8.94, p <.001, R2 = .091]; however only two individual predictors reached significance. Specifically, there was a significant impact of age at evaluation (Std β = .22, SE = .06, t = 4.50, p <.001) and SES (Std β = −.20, SE = .02, t = −4.08, p <.001) such that older age and lower SES were associated with a greater number of events. When regressions were run with PTEs and other events separately, age remained a significant predictor in both models. In contrast, SES was only a significant predictor of cumulative other events, not PTEs.
Impact of Life Events
To explore the relationship between life events and ratings of psychological functioning (PTSS, depression, anxiety), Pearson correlations were computed. As seen in Table 5, correlations were generally significant (r’s = .32–.36) between the three psychological functioning measures and cumulative life events. However, when correlations were analyzed separately for PTEs and other events, there was only a minimal association between cumulative incidence of PTEs and psychological functioning (r’s = .11–.16). In contrast, the magnitude of the correlations between other events and psychological functioning was significantly stronger (r’s = .35–.37; Fisher’s z = −3.19–−3.70; Table 5). The same pattern emerged when partial correlations were computed to control for group differences in SES (data not shown).
Table 5.
Pearson correlations between cumulative life events (Total Events, PTEs, and Other Events) and measures of psychological functioning.
| Total Events | PTEs | Other Events | ||
|---|---|---|---|---|
| r | r | r | Fisher’s za | |
| All Patients (n = 396) | ||||
| UCLA PTSDI | .36*** | .16** | .37*** | −3.19*** |
| CDI | .33*** | .12* | .36*** | −3.70*** |
| SCARED | .32*** | .11* | .35*** | −3.57*** |
| Cancer Only (n = 254) | ||||
| UCLA PTSDI | .33*** | .17* | .33*** | −1.82* |
| CDI | .26** | .05 | .29*** | −2.83** |
| SCARED | .31*** | .13 | .32*** | −2.27* |
| Control Only (n = 142) | ||||
| UCLA PTSDI | .39*** | .18** | .40*** | −2.00* |
| CDI | .38*** | .17** | .40*** | −2.13* |
| SCARED | .34*** | .16* | .36*** | −1.88* |
Abbreviations: PTE = potentially traumatic event; UCLA PTSDI = UCLA PTSD Reaction Index for DSM-IV; CDI = Children’s Depression Inventory; SCARED = Screen for Child Anxiety Related Emotional Disorders
Fisher’s exact test to evaluate differences in magnitude of correlations between associations with PTEs and other events
p < .001
p < .01
p < .05
A similar picture emerged when correlations were computed separately for the participant groups (cancer and control). Specifically, there was a minimal and largely non-significant association between PTEs and measures of psychological functioning in children with cancer (r’s ranged from .05 to .17). In contrast, correlations were larger and moderate when psychological functioning was associated with other events (r’s = .29–.33; Table 5). Fisher’s z analyses confirmed that correlations with other events were significantly stronger (z’s = −1.82–−2.83). The same pattern emerged when correlations were computed with only the control group (Table 5).
To determine whether the associations between number of events and psychological functioning varied by group status, Fisher’s z analyses were computed. Results revealed no significant differences between correlations computed separately with the cancer and control group (z’s = 0.1–1.22).
Predictors of Psychological Functioning
Hierarchical regression analyses were conducted to determine the unique impact of PTEs vs. other events on psychological functioning. Order of entry into the model was consistent across the three models, with variables entered at separate steps: 1) group status (cancer vs. control); 2) SES and age at evaluation; 3) number of PTEs; and 4) number of other events. The results of each regression are presented in Table 6.
Table 6.
Hierarchical regression analyses to evaluate the impact of PTEs versus other events on psychological functioning
| UCLA PTSD Reaction Index for DSM-IV (UCLA PTSDI) | ||||||
| Std β | SE | t | F | ΔR2 | ΔF | |
| Step 1 | 0.09 | -- | -- | |||
| Cancer vs. Control | .02 | 1.48 | 0.29 | |||
| Step 2 | 5.85** | .04 | 8.73*** | |||
| SES | −.21 | 0.06 | −4.13*** | |||
| Age | −.04 | 0.24 | −0.81 | |||
| Step 3 | 7.72*** | .03 | 12.79*** | |||
| PTEs | .19 | 0.50 | 3.97*** | |||
| Step 4 | 15.94*** | .10 | 45.32*** | |||
| Other Events | .35 | 0.25 | 6.73*** | |||
| Children’s Depression Inventory (CDI) | ||||||
| Std β | SE | t | F | ΔR2 | ΔF | |
| Step 1 | 0.25 | -- | -- | |||
| Cancer vs. Control | .02 | 0.57 | 0.50 | |||
| Step 2 | 2.20 | .02 | 3.18* | |||
| SES | −.13 | 0.02 | −2.49* | |||
| Age | .01 | 0.09 | 0.29 | |||
| Step 3 | 3.18* | .02 | 6.03* | |||
| PTEs | .13 | 0.20 | 2.46* | |||
| Step 4 | 12.44*** | .11 | 47.96*** | |||
| Other Events | .36 | 0.10 | 6.92*** | |||
| Screen for Child Anxiety Related Emotional Disorders (SCARED) | ||||||
| Std β | SE | t | F | ΔR2 | ΔF | |
| Step 1 | 3.43 | -- | -- | |||
| Cancer vs. Control | .09 | 1.25 | 1.85 | |||
| Step 2 | 7.66*** | .05 | 9.70*** | |||
| SES | −.17 | 0.05 | −3.36** | |||
| Age | −.15 | 0.20 | −2.98** | |||
| Step 3 | 8.56*** | .02 | 10.69** | |||
| PTEs | .17 | 0.42 | 3.27** | |||
| Step 4 | 17.42*** | .10 | 48.65*** | |||
| Other Events | .36 | 0.21 | 6.98*** | |||
Abbreviations. SES = socioeconomic status; PTE = potentially traumatic event
p < .001
p < .01
p < .05
Results of the regression models were generally consistent across the three measures. First, there was no significant impact of group status (cancer vs. control) on psychological functioning. Second, demographic factors – SES and age – did generally significantly impact psychological functioning, accounting for approximately 4% of the variance (UCLA PTSDI and SCARED; no impact for the CDI). Third, the impact of cumulative PTEs was small, but significant, with R2 values ranging from .02 (CDI) to .03 (UCLA PTSDI). Finally, the impact of cumulative other events was much larger, with R2 > .10.
Finally, given our interest in the differential impact of life events on psychological functioning based on group status (cancer vs. control), we ran exploratory analyses to test for moderation. Regression models had similar outcomes regardless of the group analyzed, thus demonstrating that group status does not moderate the relationship between life events and psychological functioning.
Discussion
The objective of the current paper was to examine the differential impact of potentially traumatic events and other significant life events on psychological functioning in children with cancer and community controls. Both groups demonstrated psychological functioning within normative limits; however, a proportion evidenced mean scores suggestive of clinical levels of distress. Results revealed that children with cancer endorsed more cumulative life events than controls (approximately one event). However this difference disappeared when the “serious illness” item was removed from the scale. Moreover, when life events were distinguished between PTEs and other stressful life events, children with cancer endorsed experiencing significantly more PTEs than controls. There were no group differences in other events. The overall number of cumulative life events was significantly associated with psychological functioning; however, in contrast to expectations, the magnitude of this association was dependent on the type of event evaluated. Specifically, the association between psychological functioning and other events was significantly stronger than that with PTEs. Hierarchical regression analyses confirmed this pattern, with other events accounting for approximately ten percent of the variance in psychological functioning, over and above the influence of group status, demographic factors, and cumulative PTEs. This would suggest that while cumulative incidence of negative life events is associated with psychological distress, this relationship is driven by incidence of events that would not meet DSM-IV A1 criteria (e.g., family or school problems), rather than more potentially traumatic events. Moreover, this relationship is similar for both children with cancer and community controls, as exploratory analyses revealed that group status does not moderate this relationship. Our results both replicate and extend the prior work of Currier and colleagues (2009) who demonstrated that more cumulative life events was associated with greater PTSS in children with cancer.
The current findings are consistent with past research that has indicated that significant life events do not have to meet DSM-IV A1 criteria in order to be associated with psychological distress (Gold et al., 2005; Van Hoof et al., 2009; Verlinden et al., 2013). Indeed, our results suggest that, even for those youth who have experienced an A1 event (e.g., our cancer patients) the impact of other events on psychological functioning is stronger than that of PTEs. Such findings are supported by recent research that suggests that a child’s perception of an event (e.g., DSM-IV A2 criteria) is more important for the development of PTSS/PTSD than the severity or traumatic nature of the event (e.g., DSM-IV A1 criteria) (Verlinden et al., 2013). This is perhaps ironic, given the elimination of A2 criteria from the DSM-5 (American Psychiatric Association, 2013), ostensibly due to a lack of predictive validity. Although A2 criteria was not explicitly measured in our study, that ‘other events’ were more indicative of psychological functioning than PTEs could be explained by children’s perceptions of those events (e.g., A2 criteria). For example, a child could have very different perceptions of the severity of an ‘other event’ as compared to a PTE (e.g., the death of a pet may have a significantly greater impact on their psychological distress than the death of a grandparent they have never met). However, without knowing more about these events, we cannot explicitly evaluate this hypothesis.
Our life events measure generally captured significant events that have the potential to have a lasting impact on a child’s world and thus psychological functioning as well (e.g., parental divorce, moving to a new city). These events, while not likely to meet DSM-IV or 5 criteria, certainly have the potential to precipitate psychological distress, and this was demonstrated in the current study. In contrast, a smaller proportion of the events captured were those that may have less of a lasting impact, but may also occur with more frequency and have the potential to repeat (e.g., failing a test, death of a pet, break-up of a romantic relationship). While still significant, these latter events may be better assessed as part of a measure of daily stress or frequent stress, rather than stressful life events per se. This does not suggest that these events are not equally important, but rather begs the question of the impact of daily/more frequent stressors on psychological functioning (Chamberlain & Zika, 1990).
In addition to the intensity of events, there remains the question of alternate ways to classify events that may be more predictive of or relevant to psychological functioning. One such category would be events that involve an interpersonal relationship and those that do not. Specifically, past research (Lancaster et al., 2009) has suggested that individuals who experience traumatic events that are interpersonal in nature evidence higher levels of PTSS than those who experience non-interpersonal traumatic events. Similarly, a study of preschool-age children suggested that interpersonal events, and particularly those that involved a change to the structure of the family (e.g., parental job change, divorce, moving), were indicative of mental health outcomes (Furniss et al., 2009). Although beyond the scope of this study, examination of the LESC suggested that a large percentage of the events directly involve or affect a friend or family member. Given the importance of the family in child development, and in particular for children undergoing treatment for a serious illness like cancer, these effects warrant further attention.
Limitations to consider include lack of method variance, as all measures were self-report in nature. However, we felt it was most appropriate in our design that both life events and psychological functioning be based on youth’s perceptions of these factors. Relatedly, our measures of psychological distress focused exclusively on internalizing concerns and not externalizing behaviors. While children are often reliable reporters of their internal thoughts and feelings, they may be less dependable with regards to ratings of their behavior. Future studies may wish to examine concordance of self- and parent-report of stressful life events and/or psychological functioning, and would benefit from the inclusion of measures of externalizing behaviors as well. Past studies using the LESC have demonstrated significant differences between parent and child report of negative life events, though adequate agreement regarding specific events was also found (Currier et al., 2009; Johnston et al., 2003). Relatedly, no additional information was gathered regarding the experiences of our sample’s nine PTEs. As such, it is quite possible that many endorsed PTEs would not meet DSM-IV (or 5) A criteria. However, given that our findings confirmed some prior research that non-PTEs are more impactful with regard to psychological functioning (Copeland et al., 2010; Gold et al., 2005; Verlinden et al., 2013), this reduced our concern regarding this limitation. Similarly, we did not collect information about how often each event occurred. It is certainly possible that many of the other events occurred more than once, and this may be true of some PTEs as well. Finally, our sample was not highly distressed, with mean scores falling within normative limits. However, we were still able to account for significant variance in psychological functioning.
Ultimately, our findings highlight the impact of stressful life events on psychological functioning in children, regardless of illness history. More specifically, psychological functioning appears to be most affected by likely non-traumatic events that, while still significant, would not necessarily qualify an individual for a diagnosis of PTSD. Given this, it would be prudent for clinicians to assess a wide range of negative life events as a potential cause of or contributor to difficulties with psychological functioning. Moreover, assessment of interpersonal life events may be particularly important, though this will require confirmation through additional research.
Acknowledgments
Funding: This work was supported in part by a grant from the National Cancer Institute to S. Phipps (R01 CA136782) and by ALSAC.
References
- American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders DSM-IV-TR. 4th ed. Washington, DC: American Psychiatric Association; 2000. [Google Scholar]
- American Psychiatric Association. Diagnositc and Statistical Manual of Mental Disorders (DSM-5) 5th ed. Arlington, VA: American Psychiatric Publishing; 2013. [Google Scholar]
- Barakat LP, Kazak AE, Meadows AT, Casey R, Meeske K, Stuber ML. Families surviving childhood cancer: A comparison of posttraumatic stress symptoms with families of healthy children. Journal of Pediatric Psychology. 1997;22:843–859. doi: 10.1093/jpepsy/22.6.843. [DOI] [PubMed] [Google Scholar]
- Barratt W. The Barratt Simplified Measure of Social Status (BSMSS) measuring SES. Indiana State University; 2006. Retrieved from http://wbarratt.indstate.edu/socialclass/Barratt_Simplifed_Measure_of_Social_Status.pdf. [Google Scholar]
- Birmaher B, Brent DA, Chiappetta L, Bridge J, Monga S, Baugher M. Psychometric properties of the Screen for Child Anxiety Related Emotional Disorders (SCARED): A replication study. Journal of the American Academy of Child and Adolescent Psychiatry. 1999;36:545–553. doi: 10.1097/00004583-199910000-00011. [DOI] [PubMed] [Google Scholar]
- Birmaher B, Khetarpal S, Brent D, Cully M, Balach L, Kaufman J, Neer SM. The Screen for Child Anxiety Related Emotional Disorders (SCARED): Scale construction and psychometric characteristics. Journal of the American Academy of Child and Adolescent Psychiatry. 1997;36:545–553. doi: 10.1097/00004583-199704000-00018. [DOI] [PubMed] [Google Scholar]
- Bruce M. A systematic and conceptual review of posttraumatic stress in childhood cancer survivors and their parents. Clinical Psychology Review. 2006;26:233–256. doi: 10.1016/j.cpr.2005.10.002. [DOI] [PubMed] [Google Scholar]
- Cameron A, Palm K, Follette V. Reaction to stressful life events: What predicts symptom severity? Journal of Anxiety Disorders. 2010;24:645–649. doi: 10.1016/j.janxdis.2010.04.008. [DOI] [PubMed] [Google Scholar]
- Chamberlain K, Zika S. The minor events approach to stress: Support for the use of daily hassles. British Journal of Psychology. 1990;81:469–481. doi: 10.1111/j.2044-8295.1990.tb02373.x. [DOI] [PubMed] [Google Scholar]
- Coddington RD. The significance of life events as etiologic factors in the diseases of children. II. A study of a normal population. Journal of Psychosomatic Research. 1972;16:205–213. doi: 10.1016/0022-3999(72)90045-1. [DOI] [PubMed] [Google Scholar]
- Copeland WE, Keeler GK, Angold A, Costello EJ. Posttraumatic stress without trauma in children. American Journal of Psychiatry. 2010;167:1059–1065. doi: 10.1176/appi.ajp.2010.09020178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Currier JM, Jobe-Shields LE, Phipps S. Stressful life events and posttraumatic stress symptoms in children with cancer. Journal of Traumatic Stress. 2009;22:28–35. doi: 10.1002/jts.20382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dube SR, Felitti VJ, Dong M, Giles WH, Anda RF. The impact of adverse childhood experiences on health problems: Evidence from four birth cohorts dating back to 1900. Preventive Medicine. 2003;37:268–277. doi: 10.1016/s0091-7435(03)00123-3. [DOI] [PubMed] [Google Scholar]
- Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, Marks JS. Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: The Adverse Childhood Experiences (ACE) Study. American Journal of Preventive Medicine. 1998;14:245–258. doi: 10.1016/s0749-3797(98)00017-8. [DOI] [PubMed] [Google Scholar]
- Furniss T, Beyer T, Müller JM. Impact of life events on child mental health before school entry at age six. European Child and Adolescent Psychiatry. 2009;18:717–724. doi: 10.1007/s00787-009-0013-z. [DOI] [PubMed] [Google Scholar]
- Gold SD, Marx BP, Soler-Baillo JM, Sloan DM. Is life stress more traumatic than traumatic stress? Journal of Anxiety Disorders. 2005;19:687–698. doi: 10.1016/j.janxdis.2004.06.002. [DOI] [PubMed] [Google Scholar]
- Howard Sharp KM, Rowe AE, Russell K, Long A, Phipps S. Predictors of psychological functioning in children with cancer: Disposition and cumulative life stressors. Psycho-Oncology. doi: 10.1002/pon.3643. (in press). [DOI] [PMC free article] [PubMed] [Google Scholar]
- Johnston CA, Steele RG, Herrera EA, Phipps S. Parent and child reporting of negative life events: Discrepancy and agreement across pediatric samples. Journal of Pediatric Psychology. 2003;28:579–588. doi: 10.1093/jpepsy/jsg048. [DOI] [PubMed] [Google Scholar]
- Kovacs M. Children's Depression Inventory, CDI Manual. North Tonawanda, NY: Multi-Health Systems, Inc.; 1992. [Google Scholar]
- Lancaster SL, Melka SE, Rodriguez BF. An examination of the differential effects of the experience of DSM-IV defined traumatic events and life stressors. Journal of Anxiety Disorders. 2009;23:711–717. doi: 10.1016/j.janxdis.2009.02.010. [DOI] [PubMed] [Google Scholar]
- Maurice-Stam H, Oort FJ, Last BF, Brons PPT, Caron HN, Grootenhuis MA. School-aged children after the end of successful treatment of non-central nervous system cancer: Longitudinal assessment of health-related quality of life, anxiety and coping. European Journal of Cancer Care. 2009;18:401–410. doi: 10.1111/j.1365-2354.2008.01041.x. [DOI] [PubMed] [Google Scholar]
- Phipps S, Jurbergs N, Long A. Symptoms of post-traumatic stress in children with cancer: Does personality trump health status? Psycho-Oncology. 2009;18:992–1002. doi: 10.1002/pon.1496. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Phipps S, Klosky JL, Long A, Hudson MM, Huang Q, Zhang H, Noll RB. Posttraumatic stress and psychological growth in children with cancer: Has the traumatic impact of cancer been overestimated? Journal of Clinical Oncology. 2014;32:641–646. doi: 10.1200/JCO.2013.49.8212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pynoos R, Rodriguez N, Steinberg N, Stuber M, Frederick C. UCLA PTSD Index for DSM-IV. UCLA Trauma Psychiatry Service; 1998. unpublished manual. [Google Scholar]
- Robinson JS, Larson C. Are traumatic events necessary to elicit symptoms of posttraumatic stress? Psychological Trauma: Theory, Research, Practice, and Policy. 2010;2:71–76. [Google Scholar]
- Steinberg AM, Brymer MJ, Decker KB, Pynoos RS. The University of California at Los Angeles Posttraumatic Stress Disorder Reaction Index. Current Psychiatry Reports. 2004;6:96–100. doi: 10.1007/s11920-004-0048-2. [DOI] [PubMed] [Google Scholar]
- Timbremont B, Braet C, Dreessen L. Assessing depression in youth: Relation between the Children's Depression Inventory and a structured interview. Journal of Clinical Child and Adolescent Psychology. 2004;33:149–157. doi: 10.1207/S15374424JCCP3301_14. [DOI] [PubMed] [Google Scholar]
- Van Hoof M, McFarlane AC, Baur J, Abraham M, Barnes DJ. The stressor Criterion-A1 and PTSD: A matter of opinion? Journal of Anxiety Disorders. 2009;23:77–86. doi: 10.1016/j.janxdis.2008.04.001. [DOI] [PubMed] [Google Scholar]
- Verlinden E, Schippers M, Van Meijel EPM, Beer R, Opmeer BC, Olff M, Lindauer RJL. What makes a life event traumatic for a child? The predictive values of DSM-Criteria A1 and A2. European Journal of Psychotraumatology. 2013;4:20436. doi: 10.3402/ejpt.v4i0.20436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wechsler AM, Sánchez-Iglesias I. Psychological adjustment of children with cancer as compared with healthy children: A meta-analysis. European Journal of Cancer Care. 2013;22:314–325. doi: 10.1111/ecc.12031. [DOI] [PubMed] [Google Scholar]