Cancer as a stressful life event: Perceptions of Children with Cancer and their Peers

Abstract

Background

The medical traumatic stress model is commonly applied to childhood cancer, assuming that the diagnosis of cancer is a traumatic event. However, little is known regarding what specifically children perceive as stressful about cancer or how it compares with other stressful events more commonly experienced by children.

Method

Children with cancer (N=254) and demographically similar peers without history of serious illness (N=202) identified their most stressful life event as part of a diagnostic interview assessing for symptoms of posttraumatic stress disorder (PTSD). The events identified as most stressful were categorized thematically, with categories established separately for cancer-related and non-cancer related events. Events were also examined to assess whether they met Diagnostic and Statistical Manual of Mental Disorders (DSM) Criterion-A for PTSD.

Results

In the cancer group, 54% of children described a cancer-related event as the most stressful they had experienced. Six distinct categories of cancer-related events and 10 categories of non-cancer related events were identified. The same non-cancer events were identified by children in both groups, and occurred in similar frequencies. The proportion of cancer-related events that met PTSD Criterion-A differed dramatically depending on which version of the DSM was applied.

Conclusion

Children do not necessarily view their cancer experience as their most stressful life event. These findings suggest that the diagnosis of cancer might be better viewed as a manageable stressor, rather than a major trauma, and are consistent with the change in the DSM-V to eliminate diagnosis of a life-threatening illness as a qualifying trauma for PTSD.

Over the last two decades, adjustment to childhood cancer has frequently been discussed and studied through the lens of the medical traumatic stress model.^1–3 According to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV),⁴ a child’s perception of and response to a potentially traumatic event was a critical component to the diagnosis of post-traumatic stress disorder (PTSD). More specifically, children’s experience of a potentially traumatic event would need to meet the following criteria: (a) be perceived as an event that involves the potential of death, injury, or threat to physical integrity (A1); and (b) respond with intense fear, helplessness, or horror (A2).⁴ Aligned with the medical traumatic stress model, the diagnosis of childhood cancer has commonly been assumed to satisfy both of these criteria.

Within the lens of the medical traumatic stress model, researchers have typically assumed that diagnosis is the event that is most stressful for children – as this is in keeping with the inclusion of “diagnosis of a life-threatening illness” as a qualifying event necessary for a PTSD diagnosis.⁴ However, there is an inherent understanding that other aspects – such as the need for intensive treatments, invasive procedures, hospital stays, and separation from the normal aspects of day-to-day life – are also stressful and may have a significant impact on child adjustment.^5,6 As such, a child’s perception of the event – A2 criteria – is a crucial component when considering a child’s adjustment to the cancer experience. Indeed, within non-cancer populations, recent research finds that A2 Criteria is more predictive of children meeting full criteria for PTSD than is A1 Criteria.^7–9 That is, whether an event results in trauma symptoms for children seems to depend more on their subjective experience of the event than the characteristics of the event itself. For children with cancer, their subjective experience of cancer continues to hold important predictive value and is at the crux of whether and when it is appropriate to consider PTSD as a possible outcome. A greater understanding of how children with cancer perceive the cancer experience – and what aspects they find stressful – will be critical for both the study and conceptualization of children’s adjustment. Such questions are particularly important with the elimination of A2 criteria from the most recent revision of the DSM, the DSM-5.¹⁰

In addition to elimination of the A2 criteria, another significant change in the DSM-5 relates to the A1 criteria regarding a life-threatening illness. Specifically, the new manual now states, “A life-threatening illness or debilitating medical condition is not necessarily considered a traumatic event. Medical incidents that qualify as traumatic events involve sudden, catastrophic events (e.g., waking during surgery, anaphylactic shock).”¹⁰ What one considers catastrophic may also be subject to interpretation, further highlighting the importance of understanding what specific aspects of the cancer experience children find most stressful.

Several studies have explicitly assessed what parts of the cancer experience youth find stressful or distressing.^11–13 In each study, youth (and their parents¹¹ and/or nurses^11,13) were interviewed regarding the most difficult aspects of their cancer treatment. Commonly identified stressors included physical side effects such as fatigue, nausea, pain (from the disease or procedures), and anxiety (about procedures or death), as well as more social stressors such as appearance changes (e.g., hair loss) missing home and confinement in the hospital, and lack of support.^11–13 This research suggests that many of the specific aspects of the cancer experience deemed most difficult for patients do not correspond to the A1 criteria of the DSM. However, a limitation of these studies was that children were asked to focus explicitly and solely on cancer, precluding them from reporting on other non-cancer events that they might consider equally or more stressful. This also limits appreciation of the impact and magnitude of cancer-related stressors relative to other commonly experienced stressful life events for children (e.g., parental divorce).

The identification of what children perceive as stressful about cancer may also help to explain the low rates of PTSD/PTSS in children with cancer, which are typically no higher than those observed in their healthy peers.^1,14–16 Notably, PTSS levels in the child cancer patient also tend to be lower than those found in their parents, suggesting that parents and children may perceive different aspects of the cancer experience as stressful.^1,16–18 Perhaps children are concerned with social aspects such as hair loss or missing school, while parents are more focused on the life-threatening nature of the diagnosis.

This study reflects an extension and secondary analysis of previously published work that assessed the prevalence of PTSD in children with and without cancer.¹⁴ In that study, 52.6% of children with cancer identified a cancer-related event as their most traumatic; however further analysis regarding the specific aspects of cancer described by these children was not conducted. As such, the objective of the current study was to build upon previous work by identifying those aspects of the disease and its treatments that children with cancer find the most stressful. Specifically, without orientation to cancer, we asked children with cancer to identify what life event they consider their most traumatic. For those children who identified a cancer-related event, we also sought to identify what specific aspects of the cancer experience they perceived to be most stressful or difficult. For those children who identified events that were not cancer-related, we compared these events to those reported by healthy children to ascertain whether there were significant differences between children with and without cancer regarding the events they perceive as most stressful.

Method

Participants

Patients

Patients with a history of cancer were recruited from St. Jude Children’s Research Hospital in four strata based on time since diagnosis: a) 1–6 months, b) 6–24 months, c) 2–5 years, and d) >5 years. Eligibility criteria included: a) age 8–17 years; b) at least one month from diagnosis; c) able to read and speak English; d) no history of significant cognitive or sensory deficits that would preclude completion of measures; and e) availability of a parent/caregiver who was willing to provide consent and to participate in the study. A total of 258 (68% of those approached for recruitment) patients consented to participate; 254 completed all study measures. Demographic and diagnostic information is presented in Table 1. Patients who agreed to participate did not differ from those who declined with regard to age, gender, race/ethnicity, or diagnostic category.

Table 1.

Demographic Information

	Patient Group n=254	Comparison Group n=202
Age
Mean (SD)	12.6 (2.9)	12.2 (2.89)
Range	8–17	8–17
Gender (%)
Female	48.4	50.7
Male	51.6	49.3
Race (%)
Caucasian	72.5	75.2
African American	22.7	21.4
Other	4.8	3.4
SES (%)
Groups I & 2	28.0	34.8
Group III	31.5	31.6
Groups IV & V	40.5	33.6
Diagnosis (%)
Acute Lymphoblastic Leukemia	23.6
Acute Myeloid Leukemia	7.1
Hodgkin’s & Non-Hodgkin’s Lymphoma	13.4
Solid Tumor	38.6
Brain Tumor	17.3
Time since diagnosis (%)
<6 months	25.2
6 months to 2 years	24.4
2 years to 5 years	25.6
>5 years	24.8

Note. SES = socioeconomic status, obtained with the BSMSS¹⁹. SES groups are ordered highest to lowest, with Group I reflecting highest SES strata and Group V indicating lowest SES strata.

Healthy Comparison Participants

Children were recruited from schools within three states neighboring the hospital. Control participants met the same eligibility criteria as patients, with the exception that they reported no history of significant medical illnessOf those contacted for participation, 89% (n=211) agreed to participate and 202 completed all study measures with evaluable interviews. Demographic variables are in Table 1. Patients and healthy comparison children did not differ in age, gender, race/ethnicity, or socioeconomic status (obtained using the Barratt Simplified Measure of Social Status).¹⁹

Procedures

Patients were recruited during routine medical appointments at the hospital. Control participants were recruited from area schools in a two phase process. First permission slips were distributed at schools, requesting permission to contact the family subsequently if the student was a good demographic match for a participating patient. Parents provided demographic information relevant to the study (child age, gender, race/ethnicity; parent education/occupation). Parents of children who were an appropriate match for an enrolled patient were contacted via telephone and invited to participate. The fact that the student was identified as matching a specific patient who was already enrolled on study helped to increase the participation rate. All participants, cancer and control, received a 25 dollar gift card for their participation. All study procedures were conducted at the outpatient Psychology Clinic of the hospital. Following IRB-approved consent procedures and completing a battery of paper-and-pencil measures, children completed a diagnostic interview. The focus of the current analysis is on data from the child’s diagnostic interview, which is described below.

Measure

Clinician-Administered PTSD Scale for Children and Adolescents (CAPS-CA).²⁰

The CAPS-CA is a structured diagnostic interview that corresponds with the DSM-IV criteria for diagnosis of PTSD. It is administered by a trained clinician to assess for PTSD symptoms in youth ages 8–17 years. The CAPS-CA identifies whether the child’s event meets Criteria A1 and A2 and examines the frequency, intensity, and impact of PTSD symptoms associated with the DSM-IV criteria. (Note: the A2 criteria was dropped from DSM-5,¹⁰ an issue which will be addressed further in the discussion). It is a widely-used diagnostic assessment tool and has strong psychometric properties. Prior to beginning the interview, the child was provided with a list of stressful life events and asked to identify which ones they had experienced. They were then asked to identify the one they considered the most stressful or traumatic event during their lifetime, and to answer all subsequent questions with this identified event in mind. It is important to note that both patient and control participants self-identified their most stressful event rather than being oriented to a particular event (i.e., cancer).

Data Analyses

Data were analyzed by five individuals with experience related to the research topic. The events identified as “most stressful” were reviewed across the full sample of participants (both cancer patients and controls) by two team members. Thematic categories were subsequently generated using principles of structured Q-sort²¹ and thematic analysis²² to capture these events into broader themes, with cancer-related events further subdivided. Children’s cancer-related events were categorized according to the specific aspects of the cancer experience that children reported as most stressful. Because some patients noted multiple specific aspects of the cancer experience, cancer-related responses were first broken into discrete meaning units.

For all types of events, two initial coders reviewed all events independently and generated broad primary themes to characterize the stressful events. They then reviewed each event independently and assigned each event to a category. Any discrepancies were resolved by consensus. Three additional members of the research team then coded the data to assess reliability. The reliability coders were provided definitions of the categories generated by the primary coders and instructed to code each meaning unit using one of these categories. Reliability was excellent, with Kappa’s ranging from.94-.96 for cancer events and .94-.97 for non-cancer events.

Results

Cancer-Related Stressful Events

About half of the children in the patient group (54%, n=137) described a cancer-related event as their most stressful. Children who identified a cancer event differed from those who identified a non-cancer event according to time since diagnosis (based on strata: χ²(3,254)=19.20, p<.001): children 5+ years from diagnosis were significantly less likely to identify a cancer-related event than children in the other three strata. However, it should be noted that children who were further from diagnosis were also younger at diagnosis; children in strata 4 were significantly younger at diagnosis that those in strata’s 1–3. There were no differences based on current age, gender, treatment status, or relapse history.

The manner in which these individuals identified cancer varied. Some participants (n=20, 14.6% of the 137 who identified a cancer-related event) identified the overall cancer experience as their most stressful event, and when further probed were unable to identify a specific aspect that was most stressful. The majority of patients identified particular aspects of their cancer diagnosis, including: (a) diagnosis/adjustment to diagnosis, (b) treatment/acute side effects of treatment, (c) family/social impact, (d) pre-diagnosis period, (e) late/permanent effects of treatment, and (f) the possibility of death. The following describes the frequency of each cancer-related sub-category from among only the children who described a cancer-related event. Because some patients described multiple specific cancer-related events, meaning units were coded rather than events, with some patients receiving codes for two sub-categories. As such, the percentages below exceed 100%.

Diagnosis/adjustment to the cancer diagnosis

The most frequently described cancer event was learning about the cancer diagnosis and/or initial adjustment to diagnosis. Approximately half of the 137 patients identifying a cancer-related event (n=64; 46.7%) described this aspect of the cancer experience as the most stressful.

Treatment/acute side effects of treatment

A second subcategory included medical treatment(s) and the acute side effects of medical treatment(s). Side effects included in this category were restricted to effects that were concurrent with treatment but would not be expected to continue after the patient had completed treatment. Commonly-identified treatment experiences included receiving radiation, chemotherapy, and invasive procedures, as well as side effects such as hair loss, weight loss, and “feeling sick due to chemo.” Thirty-eight patients (27.7% of those identifying a cancer-related event) described events from this subcategory.

Family/Social impact of cancer

A small proportion of patients (n=9, 6.6%) indicated that the negative impact of cancer on their family and/or social relationships was the most stressful aspect. These children described such stressors as being away from friends and family when living away from home for medical treatment and the impact of their cancer experience on their family and/or friends.

Pre-diagnosis period

Other patient participants (n=7, 5.1%) identified the period of time when they were ill and experiencing symptoms of cancer but had not yet been diagnosed as most stressful. For example, one participant described his/her most stressful event as “the time when no one knew what was making me so sick,” prior to diagnosis.

Late effects/permanent effects of treatment

In contrast to acute side effects, some patients (n=6, 4.4%) identified side effects that appeared after the illness/treatment(s) had ceased (e.g., cognitive late effects) and/or were more permanent and lasting effects of their treatment. For example, participants described such experiences as having had treatment procedures with lasting effects (e.g., enucleation, limb sparing), reduced growth, and experiencing neurocognitive late effects (e.g., concentration and memory difficulties) as the most stressful aspects of the cancer experience.

Possibility of death

Finally, six patients (4.4%) identified the possibility of dying from cancer as the most stressful aspect of cancer.

Non-Cancer Events as the Most Stressful

Approximately half of the cancer patients spontaneously reported a non-cancer stressful event (46%; n=117). Ten distinct non-cancer categories were identified, and are further described below. Events are described for children in both the cancer group and the healthy comparison group, and Table 2 includes Chi-square comparisons of the frequency of these events between the two groups.

Table 2.

Frequency of Non-Cancer Events Identified in the Patient vs. Healthy Control Groups

Category	Frequency		Pearson Chi-Square
	Patient group n=117	Comparison group n=202
Death of a Family Member/Friend/Pet	66 (56.4%)	73 (36.1%)	11.57**
Family Member	54 (46.2%)	65 (32.1%)	5.60*
Friend	9 (5.1%)	5 (2.5%)	3.68
Pet	3 (2.6%)	3 (1.5%)	0.07
Family-Related Stressor/Conflict	12 (10.3%)	28 (13.9%)	0.59
Natural Disaster/Severe Weather/Fire	7 (6.0%)	12 (5.9%)	<0.01
Violence/Threat of Violence to Self-Others/Sexual Abuse	7 (6.0%)	11 (5.4%)	0.04
Fears/Phobias	5 (4.3%)	10 (4.9%)	<0.01
Non-Cancer Illness/Injury to Self	5 (4.3%)	16 (7.9%)	1.10
Vehicle Accident	5 (4.3%)	13 (6.4%)	0.30
Illness or Injury of Other	5 (4.3%)	19 (9.4)%	2.12
Peer Stress/Conflict	3 (2.6%)	14 (6.9%)	2.00
School/Academic Stressor	2 (1.7%)	8 (3.9%)	0.61

^*p<.05

^**p<.001

Note. Percentages in the cancer group are taken just from the children who reported non-cancer events.

Death of a family member/friend/pet

The majority of both patient and healthy comparison participants (n=66, 56.4% of the 117 patients reporting a non-cancer event and n=73, 36.1% of healthy control participants) identified the death of a family member, friend, or pet as their most stressful event. The deaths reported ranged from expected deaths of elderly or terminally ill family members (e.g., great-grandparents, family members with terminal cancer) to sudden, violent deaths (e.g., watching a family member be murdered, a friend dying in a car accident). Additionally, the deaths described also varied with regard to the child’s level of involvement in the event, with some children hearing about the death from a family member and other children reporting that they were physically present during the event. As seen in Table 2, significantly more children from the cancer group identified the death of a friend or loved one as their most stressful life event, χ² (1,319) = 11.57, p < .001. Given this difference, we examined the type of loss more closely, and although children in the cancer group reported a greater frequency of all losses, only loss of a family member was significantly higher in the cancer group (χ² (1,319) = 5.60, p < .05).

Family-related stressor/conflict

Family-related stressors included such events as family conflict (e.g., parents fighting, children’s arguments with parents, siblings fighting), divorce, separation from family members (e.g., parents in the military, sibling running away), and punishment. In the patient group, 12 (10.3%) participants identified a family stressor as their most stressful event, compared with 28 (13.9%) participants in the healthy comparison group.

Natural disaster/severe weather/fire

Seven participants in the patient group (6.0%) and 12 in the comparison group (5.9%) identified a natural disaster as their most stressful event (e.g., tornado, hurricane, fire). Children’s experience ranged from being displaced by the event (e.g., evacuating due to hurricane flooding and losing property) to seeing the aftermath of disasters in others’ property (e.g., the remains of a friend’s house fire).

Violence/threat of violence to self-others/sexual abuse

Some children identified an act of violence as their most stressful event, including such violence as assault, abuse, theft, and witnessing violence (e.g., witnessing a shooting or fighting in the streets). In the patient group, 7 (6.0%) identified acts of violence, with 10 children from the comparison group (5.0%) identifying such an event. Consistent with other event categories, children’s exposure to the violence ranged from experiencing violence to witnessing it.

Fears/phobias

Some children identified fears or phobias as their most stressful experience, such as nightmares, fear of monsters, and fear of heights. Five patients (4.3%) and 9 comparison participants (4.5%) identified fears or phobias.

Non-cancer illness/injury to self

Non-cancer illnesses or injuries identified ranged from those requiring hospitalization (e.g., pneumonia, broken bone) to minor illnesses/injuries. In the patient group, 5 (4.3%) participants identified a non-cancer illness or injury, with 16 (7.9%) participants in the healthy comparison group.

Vehicle accident

The vehicle accidents that children identified were primarily car accidents. These events ranged from children learning about family member(s) being in an accident to personally experiencing the accident. Five (4.3%) participants in the patient group and 13 (6.4%) in the healthy comparison group identified a vehicle accident as their most stressful event.

Illness or physical injury of another person

Events in this category ranged from serious life threatening illnesses (e.g., parental cancer) to minor injuries. Five (4.3%) children from the patient group and 19 (9.4%) children from the comparison group identified such an event.

Peer stress/conflict

Some children (3 children in the patient group (2.6%), 14 (6.9%) in the comparison group) identified interpersonal stressors with peers, including bullying, romantic break-ups, non-physical fights with friend(s), and feeling abandoned by a friend.

School/academic stressor

A small proportion of participants identified school related stressors: 2 children in the patient group (1.7%) and 8 children in the comparison group (4.0%). Children reported such events as standardized tests, anxiety about grades, and feeling pressured by parents to perform well in school.

DSM Criteria for PTSD

In our prior report,¹⁴ children referring to any type of cancer-related event were credited as having met A1 criteria. However, with more specific delineation of cancer-related stressors, the percentage of those who met A1 criteria according to the DSM-IV was reduced to 70.1% (n=96). This did not differ significantly from the rate of patients who identified a non-cancer event (n=88, 75.2%, χ²(1,254)=0.60). Likewise, patients who reported cancer vs. non-cancer related events did not differ in their frequency of meeting A2 criteria [79.5% vs. 72.6%, χ²(1,254)=1.31]. However, when we applied the new DSM-5 guidelines regarding life-threatening illness, only 16 (11.7%) of those identifying a cancer event met A1 criteria, which differed significantly from those identifying a non-cancer event [χ²(1,254)=90.7, p<.001].

When comparing children identifying a cancer event to healthy comparison children using DSM-IV there was no difference in those meeting A1 criteria [70.1% vs. 72.3%, χ²(1,339)=1.01, p = .75]. However, in applying DSM-5 criteria, a significantly higher percentage in the healthy comparison group met A1 than did those identifying cancer-related events [65.3% vs. 11.7%, χ²(1,339)=93.4, p<.001]. In contrast, for children in the cancer group who identified a non-cancer event, there was no difference in the frequency of those meeting A1 regardless of whether DSM-IV [χ²(1,319)=0.19, p = .66] or DSM-5 criteria [χ²(1,319)=0.82, p = .36] were applied.

Discussion

Despite considerable evidence of children’s positive adjustment and resilience to this challenge,^23,24 childhood cancer continues to be viewed predominantly from a trauma model.^1–3 While most would agree that cancer presents numerous stressors for a child, it is not clear that it is experienced as a trauma by most children. In this study we sought to clarify how the cancer experience is perceived by children, and to place this in context relative to other commonly experienced childhood stressors. To achieve this, we interviewed children both with and without a history of cancer, and without any orientation to, or suggestion that cancer should be the focus of their report. By allowing children to spontaneously describe whatever they consider to be their most stressful events, we avoid a focusing effect,²⁵ and can obtain an unbiased depiction of how cancer is perceived relative to other stressful life events, as well as an understanding of the specific aspects of cancer that children find most stressful.

As we have previously reported,¹⁴ slightly more than half of cancer patients identified a cancer-related event as their most stressful life experience. In the current analysis, we examined the specific aspects of the cancer experience perceived as most stressful, with most children spontaneously identifying particular issues or events (e.g., diagnosis, medical procedures, physical sequelae, social impacts). It is important to note that many of the cancer-related stressors identified (e.g., concern over hair loss, missing out on school activities), would not meet established A1 criteria for a PTSD diagnosis. When the DSM-IV A1 criteria were re-examined in light of these more specifically described cancer-related stressors, the rate of those meeting A1 was reduced from 100% (previously credited for all cancer events) to 70%. This rate did not differ significantly from those children in the patient group who identified a non-cancer event. With introduction of the DSM-5,¹⁰ the language regarding criterion A1 and medical events was changed such that a life-threatening illness did not automatically constitute a trauma, but only sudden catastrophic medical events. When we re-examined cancer-related events with this new guideline, the rate meeting A1 criteria was dramatically reduced to 11.7%, significantly lower than that for both patient-reported non-cancer events, and those of healthy comparison children. The utility of the DSM A1 criteria has been criticized from a number of perspectives, and its validity has been questioned.^26,27 Moreover, despite the elimination of the A2 criterion from the DSM-5, considerable data points to the greater importance of the subjective perception of the event as a predictor of stress response, a finding which appears particularly true in children.^7–9 The current data raise additional questions about the application of Criterion A1, as well as the usefulness of a trauma model applied to childhood cancer. Certainly these findings highlight the importance of clarifying what it is that children find stressful in the cancer experience.

For those patients who identified a non-cancer event as most stressful, the identified events were quite varied in content and severity, and occurred with a frequency generally consistent with the events reported by their healthy peers. The lone exception was regarding the experience of death of a friend/family member/pet. This was the most common non-cancer event reported by children in both groups, but was endorsed significantly more frequently by those in the cancer group. One possible explanation for this is that children in the cancer group simply experienced deaths of significant others more frequently, and our earlier report provides some evidence for this, at least in regard to parental loss and death of a pet.²⁸ Perhaps a more likely explanation is that children who have experienced a life-threatening illness may become sensitized, and perceive the death of others as more salient and distressing. However, our finding that death of a loved one was the most commonly reported non-cancer type of event in both groups is consistent with the literature on stressful events in children.²⁹ The events reported by the cancer and comparison groups were also comparable to stressors reported by other general samples of children and adolescents in previous investigations.^29–31 Aside from death of a loved one, there is striking similarity in the rate of most significant stressful events endorsed in both groups.

Further comparison of patients who identified non-cancer events with their healthy peers revealed that they endorsed DSM-IV A1 and A2 Criteria for PTSD (as well as DSM-5 A1) at the same rate. These findings emphasize that one cannot assume that the cancer experience is the most significant stressor in the patient’s life or that children with a cancer history only identify non-cancer events as more stressful when identifying severe traumas. Although some of the non-cancer events identified by children with cancer are severe, many are less severe stressors commonly experienced by children that would clearly not meet A1 criteria. Beyond severity, it is important to consider the impacts that a child’s history and development have on their experience and interpretation of potentially stressful events.³² The child’s subjective appraisal will likely be moderated by many factors, such as premorbid disposition or parental adjustment.^33,34 Further, examination of a child’s non-cancer stressor in addition to the cancer experience can be important in determining a child’s level of distress during cancer treatment. Cumulative stressful life events have been shown to be an important predictor of PTSS and other distress outcomes in children with cancer.^28,35

Much of the prior literature that has applied a medical traumatic stress model to childhood cancer has assumed that diagnosis was the most stressful aspect of the experience. Indeed, we found that diagnosis was the most commonly identified stressor, endorsed by nearly half of those children who identified a cancer-related event. However, many other aspects of the cancer experience were identified as most stressful, and far more patients identified non-cancer events than did those specifically endorsing the cancer diagnosis. Importantly for clinicians working with children with cancer, particularly in the post-diagnosis period, although it may be reasonable to assume that patients have been stressed and challenged by their experience, one need not assume they have been traumatized, nor should one assume that any distress measured or observed is specifically due to cancer. Rather, it is appropriate to obtain a history that includes other stressful life events and to screen for psychosocial distress caused by these experiences in addition to those related to cancer and its treatment. Those patients who have experienced other significant life stressors could be identified as at risk and targeted for supportive intervention. Unfortunately, the medical traumatic stress model is one that implicitly assumes psychopathology, and may lead both researchers and clinicians working with this population to be more vigilant for signs of pathology than of healthy adjustment, and of the factors that promote positive outcomes.

Some study limitations must be acknowledged. First, this is a single-site study, and it is not clear how this may impact the generalizability of the findings. Replication of these findings at other sites or multi-site studies should be conducted. Although our methodology to avoid focusing on cancer was intentional and is a strength of the study, we did not examine cancer-related stressors in those children who identified a non-cancer event. This would allow for a more comprehensive assessment of the relative frequency of cancer-related events perceived as stressful. Finally, we focused solely on the perceptions of the child, and did not examine the perceptions of the parent regarding either what they felt was most stressful for their child, or for themselves. This would be an appropriate target of future research, and could shed light on differences observed in levels of PTSS and other distress measures between parent and child.

In summary, using a methodology that avoids focusing on cancer, roughly half of children with cancer identify cancer-related events as the most difficult or stressful they have experienced. The remaining children identify a range of other events as having been more stressful than any aspect of their cancer experience. The non-cancer events reported are very similar to those reported by demographically-matched children without a cancer history, and range from the severe to milder and more common everyday events. Viewed in the context of prior findings pointing to low levels of PTSD in the pediatric oncology population,¹⁴ the current findings suggest that the diagnosis of cancer might be better viewed as a difficult, but manageable stressor, rather than a major trauma, and are consistent with the change in the DSM-5 to eliminate diagnosis of a life-threatening illness as a qualifying trauma for PTSD.