Assessing the Experience of Medically Ill Youth Participating in Psychological Research: Benefit, Burden, or Both?

Abstract

Few data exist pertaining to the perceived burdens or benefits of medically ill children participating in psychological research studies, particularly in outpatient pediatric settings. As part of a larger study that involved completing self-report questionnaires, this study assessed whether participation was burdensome and/or beneficial to 271 children undergoing treatment for cancer, NF1, sickle cell, HIV, primary immune deficiencies, and Li Fraumeni and to their caregivers.

The majority of patients (83%) and their caregivers (93%) did not find participating burdensome. Moreover, the majority of patients (85%) and caregivers (95%) found at least some benefit to participation, including finding it helpful to be asked about issues that affect their life and feeling good about helping others. The data suggest that resistance to psychological research based on the belief that such research is intrusive and potentially harmful overestimates the negative aspects and potentially impedes progress studying positive psychosocial outcomes in outpatient pediatric research.

Introduction and Background

Limited empirical data exist that examine a child or caregivers’ reactions to research participation of children in medical settings. Even less data are available pertaining to the actual or perceived burdens or benefits of medically ill children participating in psychological research studies. While behavioral research is generally minimal risk, some hospital Institutional Review Boards (IRBs) have been reluctant to approve psychosocial studies due to the perceived vulnerability of medically ill children and concern over minimal benefits and potential for harm by research participation.^{1, 2} With limited empirical knowledge available to guide the process, IRBs are often left to make decisions based on professional experience and informed conjecture about the likely effects of psychosocial research participation.^3–5

Studies that have systematically investigated the risks or benefits to ill patients, and the impact of their participation in psychological research, have primarily been conducted with adults. Reported psychological risks include altered self-concept, increased anxiety, receiving information about oneself that is unpleasant, boredom, and inconvenience.^6,7 A study designed to investigate data collection procedures frequently associated with psychological research, such as audiotaping, questionnaires and interviews, found minimal negative reactions to the procedures and participation somewhat useful.^8,9 The risk of emotional distress resulting from psychological research in adults has been specifically investigated in the area of trauma and little empirical evidence was found that recall and discussion of traumatic events re-traumatize the individual; moreover, the data suggest that studies on violence and trauma may provide some benefit to participants.^5,10 Only one study was found that investigated the risks or benefits of ill patients’ participation in psychosocial research with patients who were receiving palliative care.¹¹ The majority of adult patients reported no burden associated with participation and found the experience as moderately to highly beneficial. Research participants reported a sense of contributing to society and appreciated the opportunity to interact with others and discuss their illness.

The few studies that have provided empirical findings regarding children’s appraisal of research participation or those of their caregivers are limited by homogeneous samples (e.g. those who have experienced a trauma, witnessed violence, or are affected by family alcoholism) and varied methodological assessment strategies between studies (e.g. qualitative versus quantitative, telephone versus mail-in survey). ^{3, 12–14} Within a feasibility study that consisted of conducting structured diagnostic interviews with preadolescents, risks were not associated with direct interviewing of children about their own psychopathology.¹⁵ One study that assessed both child and caregiver reactions in a pediatric medical setting included a sample of children that had recently incurred a pediatric injury, assessing for acute posttraumatic stress reactions.¹⁶ Participation had little risk of generating distress for children or caregivers. Feeling good about helping others was the most commonly reported positive aspect of research participation. A more recent study interviewed 177 adolescents participating in clinical research for a medical or psychiatric illness, or as a healthy volunteer, and their caregiver. Overall, 90% of the adolescents and caregivers were willing to have the adolescent undergo a few extra blood draws, and 65% were willing to have the adolescent undergo an extra skin biopsy for research purposes. The vast majority felt that the adolescents were making an important contribution to help others.^{17, 18}

No empirical data have been found on whether children undergoing care for a serious illness and their caregivers would perceive participating in psychosocial research as burdensome or beneficial during their child’s outpatient medical care. Therefore, the objective of this study was to determine whether participation in research that involved approximately 20 minutes of completing self-report questionnaires in an outpatient medical setting was burdensome and/or beneficial to the child and to the caregivers. We also explored whether the level of burden would be related to specific demographic and psychosocial variables such as child’s age, gender, race/ethnicity, disease, self-reported distress level and symptoms such as fatigue, as well as caregiver self-reported distress symptoms.

Methods

Participants

The study was approved by the NIMH and Georgetown University IRBs. A convenience sample of patients between the ages of 7–21 years was recruited as part of a larger cross sectional study assessing distress and its correlates. Eligible patients were already enrolled in a treatment trial for cancer, HIV, neurofibromatosis type 1 (NF1), a primary immune deficiency, sickle cell disease, or Li Fraumeni syndrome. One caregiver (primarily the parent) for each patient was also enrolled.

Procedures

A study investigator approached caregivers of children under the age of 18 and patients ages 18–21, described the study and obtained written consent. The study was described to all minor children who were asked if they would feel comfortable participating. If so, they completed a written assent, and were given several questionnaires. When all of the study measures were complete, the patient completed a scale¹¹ measuring the benefits and burdens of participating in the research.

While the patient was completing study measures, the child’s primary caregiver was administered the same measures in order to assess their perception of their child’s degree of overall distress, pain and fatigue, anxiety, and depression. Caregivers also completed a demographic background information sheet, a self-report measure designed to assess their own psychological distress symptoms, and the same scale measuring benefits and burdens of participating in the study. The depression, anxiety and distress measures were monitored by the principal or associate investigators immediately after administration for clinically significant symptoms. If a patient or caregiver disclosed information during completion of the assessment form that suggested a danger to themselves or others, or suggested a need for psychiatric intervention, an immediate psychiatric assessment was conducted and when needed, a referral for further assessment or services was made with a provider in their home community.

The consent/assent process was conducted over 5 to 10 minutes. Most caregivers completed their study instruments in approximately 20–30 minutes while most patients completed their study instruments in 15–25 minutes.

Measures

Both caregivers and patients completed several validated self-report questionnaires including the State Trait Anxiety Inventory (STAI), Brief Depression Inventory (BDI) or Children’s Depression Inventory (CDI), Childhood Cancer Fatigue Scale (CCFS), and the Burden and Benefit Scale. These study instruments are described in Table 1. The outcome of focus for this report was the analysis of the Burden and Benefit Scale described in detail below.

Table 1.

Measures Administered

MEASURE	DESCRIPTION	COMPLETED BY	APPROXIMATE TIME TO COMPLETE (minutes)
Burden & Benefit Scale11	Measures patient reactions to research participation.	Caregivers and Patients	3
Demographic Face Sheet	Provides information on child’s age, race, diagnosis and date of diagnosis, as well as if the child has a psychiatric diagnosis or learning disability. This form also collects information on household income, relationship of the caregiver to the child.	Caregivers	3
Distress Thermometer26	Patients rate their emotional distress on a visual analog scale. The screen also includes a problem checklist where raters < age 12 indicate reasons for distress from 5 separate domains.	Caregivers and Patients	5
Brief Symptom Inventory27	Identifies psychological distress and psychiatric disorders.	Caregivers and Patients aged 18–21	4
Children’s Depression Inventory28	Identifies psychological distress and psychiatric disorders.	Patients aged 7–17	4
State Anxiety Inventory / State Inventory for Children29	Assesses anxiety and tension.	Caregivers and Patients	5
Wong-Baker FACES Pain Scale30	Assesses pain.	Caregivers and Patients	1
Childhood Fatigue Scales31	Measures fatigue in pediatric oncology patients.	Caregivers and Patients	5

Burden and Benefit Scale

The Burden and Benefit Scale¹¹ was based on a literature review that listed common benefits and burdens of research with palliative care patients.¹⁹ It is a brief clinician-administered scale designed to measure patient reactions to research participation. Patients and caregivers each completed their own scale. Each participant was asked, “Was participation burdensome to you in any way?” and was measured on a five-point Likert scale (0–4) with the following response options: “Not at all,” “A little bit”, “Somewhat”, “Quite a bit”, and “Very much”. In our analysis, we collapsed the last 4 categories (see Table 2). Each person who endorsed any burden was provided with six follow up questions to identify what part of participation was burdensome. Examples included, “Interfered with my other activities”, “The questions were too personal”, “The questions were confusing”. A second section assessed whether participation in this study provided any benefit to the participant with the same five-point Likert anchors. If any benefit was endorsed, participants were provided six possible benefits including “Addressed issues the health care team doesn’t often ask me (or my child) about”, “Helped to facilitate treatment for my/my child’s emotional distress”, “Helped pass the time/keep my mind busy”. Participants had the option of endorsing multiple items for both perceived burden and benefit.

Table 2.

Patient and Caregiver Perceived Burden and Benefit

	Patient n (%)	Caregiver n (%)
Burden
Not at all1	205 (83.0)	243 (93.1)
A little bit	31 (12.6)	14 (5.4)
Somewhat/Quite a bit/Very Much	11 (4.4)	4 (1.5)
Benefit
Not at all	31 (12.6)	12 (4.8)
A little bit	65 (24.0)	38 (15.3)
Somewhat/Quite a bit/Very Much	151 (61.2)	199 (79.9)

¹In the analyses, burden and benefit were dichotomized, with ‘not at all’ representing one category and ‘a little bit’, ‘somewhat’, ‘quite a bit’ and ‘very much’ collapsed into the other.

Analyses

Analyses were conducted using SPSS 12.0.1 (Chicago, IL). Descriptive statistics are reported for the primary outcomes (levels of burden and benefit reported by patients and caregivers). Bivariate analyses were conducted using chi-square and Pearson product moment correlations to assess associations between demographic factors, patient distress and perceived burden/benefit.

Results

Between 2009 and 2014, 271 children at 2 hospital centers met study eligibility and completed study enrollment. Of those enrolled, 53% of the patient population was male, 59% Caucasian and 75% of caregivers were mothers. Comprehensive demographic information can be found in Table 3. Five others refused participation, for an overall consent rate of 98%. Interviews were scheduled as soon as possible at the child and caregiver’s convenience; in most cases interviews took place the same day.

Table 3.

Demographics

Characteristic	N (%)
Child gender
Male	143 (52.8)
Female	128 (47.2)
Relationship to child
Mom	202 (74.5)
Dad	52 (19.2)
Aunt/Uncle- Legal Guardian	6 (2.2)
Grandparent	1 (0.4)
Other	1 (0.4)
Race/Ethnicity
White	161 (59.4)
Black or African American	65 (24.0)
Biracial	17 (6.3)
Latino	15 (5.5)
Asian/Pacific Islander	6 (2.2)
Other	4 (1.5)
Missing	3 (1.1)
Household Income
Less than $1,000 per month	23 (8.5)
$1,000–$1,999 per month	38 (14.0)
$2,000–$2,999 per month	34 (12.5)
$3,000–$3,900 per month	41 (15.1)
$4,000–$4,900 per month	31 (11.4)
$5,000–$5,999 per month	22 (8.1)
$6,000–$6,999 per month	71 (26.2)
Missing	11 (4.1)

The mean age of the children in the study was 14.5 (range 7 – 21 years). Reported race/ethnicity for most children in the study was Caucasian (59%) or African American (24%). The majority (75%) of the participating caregivers were mothers, 19% were fathers, and 3% were other legal guardians. Nine participants over the age of 18 presented to their appointment without a guardian present (3%). The breakdown of ages and other demographic data are presented in Table 3.

Burden

The majority of patients (83%) and their caregivers (93%) did not find participating in a psychosocial study that consisted of completing a number of self-report measures burdensome. The most commonly reported factors that caused burden included “questions being confusing or difficult to answer” (4% for patients, 2% for caregivers), “too many questions/took too long” (4% for patients, 3% for caregivers), “questions were too personal” (3% for patients, <1% for caregivers) and “interfered with other activities” (3% for patients, <1% for caregivers) (see Table 4). Of those who did report burden, 81% of patients and 78% of caregivers also reported that they found participating in the study to be beneficial.

Table 4.

Factors Associated with Perceived Burden and Benefit

What did you find burdensome	Patient n (%)	Caregiver n (%)
Too weak/feeling too ill	4 (1.5)	1 (0.4)
Too many questions/took too long	11 (4.1)	7 (2.6)
Interfered with other activities	8 (3.0)	2 (0.7)
Questions were upsetting	9 (3.3)	0 (0.0)
Questions were confusing/difficult to answer	12 (4.4)	6 (2.2)
Questions were too personal	9 (3.3)	1 (0.4)
What did you find beneficial	Patient N (%)	Caregiver N (%)
Helpful/relief to be asked about issues that affect my/my child’s life	118 (43.5)	171 (63.1)
Addressed issues I often don’t have an opportunity to discuss with others	71 (26.2)	113 (41.7)
Addressed issues the health care team doesn’t often ask me about	61 (22.5)	99 (36.5)
Helped facilitate treatment of my/my child’s emotional distress	40 (14.8)	78 (28.8)
Helped pass the time/keep my mind busy	142 (52.4)	63 (23.2)
Made me feel good to help others/contribute to society	160 (59.0)	189 (69.7)

Benefit

The majority of patients (85%) and their caregivers (95%) found at least some benefit to participation. Patient reasons for finding benefit in participation included feeling good about helping others or contributing to society (59%), providing a task to help patients pass the time/keep their mind busy (52%), and finding it helpful/relief to be asked about issues that affect their life (44%). Almost three quarters of the caregivers also indicated feeling good about helping others (70%) and finding it helpful/relief to be asked about issues that affect their child’s life (63%). Thirty seven percent of caregivers found benefit in that the study allowed them to address issues the health care team do not often ask about (see Tables 2 and 4).

Differences in Burden

A minimal number of differences were found related to patient burden. Caregivers of children with HIV found the study to be more burdensome than caregivers of patients with other conditions (15.5% vs. 5–13% for others, X² = 8.9, p < .05). Caregivers who reported that their children were anxious on the STAI reported greater levels of burden than caregivers of patients who did not find their children to be anxious (r=.17, p<.01). Likewise, caregivers who reported greater fatigue for their children on the CCFS reported greater levels of burden (r = .17, p < .01). There was no association between caregiver report of SES, caregiver or patient report of anxiety or fatigue and reported level of burden. Patients whose caregivers reported that their own depression was in the clinical range on the BSI (‘caseness’, t-scores ≥ 63) were more likely to find the study burdensome (X² = 7.1, p < .01). For all other reported measures, no differences were found in caregiver or patient reported burden (treatment center, patient age, caregiver gender, patient race or patient distress (DT)).

Differences in Benefit

There were a small number of differences reported in benefit based on demographic and psychosocial variables. Patients between 18–21 years of age were significantly more likely to find the study beneficial than the patients under the age of 18 (X²=14.4, p<.001); caregivers did not report this difference. African American patients were less likely to report benefit than Caucasian patients (X² = 17.1, p<.01); caregivers did not report a difference in benefit by race/ethnicity. Caregivers of female patients found the study less beneficial than caregivers of male patients (X²=7.0, p<.01). No differences were found in caregiver or patient reported benefit between specific diseases or on all other reported measures.

Discussion

The current study provides evidence for low perceived burden by medically ill children and their caregivers participating in a study that involves answering standardized behavioral assessments about mood, pain, fatigue, and overall psychological distress. Participants and family members did not report negative consequences as a function of their child’s participation in this study. On the contrary, participants and family members reported appreciation for being asked about difficult issues and benefited from having something to do while waiting at the hospital for medical treatment.

Overall, the participants in this study reported less burden and more benefit with increasing age which may be due to improved cognitive understanding of the abstract concepts of burden and benefit. The data suggest that caregiver perceptions of burden of participating in this type of research may be illness specific or due to high symptom burden in their children or themselves. We found caregivers of children with HIV and those who reported that their children were anxious on the STAI or having greater fatigue on the CCFS endorsed greater levels of burden than other comparable caregivers. Interestingly there was no association between patient report of anxiety or fatigue and burden. Patients whose caregiver reported that their own depression was high on the BSI were more likely to find the study burdensome which is consistent with the literature on perception of poorer child functioning associated with increased caregiver problems. Clearly, additional research on patient-caregiver interactions with regard to understanding benefit and burden are indicated.

Worth noting is the 70% of caregivers and 59% of youth from the ages of 7 to 21 who reported that they felt good about helping others by participating in this research study. This lends support to the idea that altruism can be a source of perceived benefit from (and a motivation for) research participation among children¹⁶ and those who do not anticipate clinical benefit for themselves.^17–18 Even for those who did express some burden, many often also expressed benefit particularly in the areas of feeling good about helping others and being relieved about being asked about issues their/their child’s life suggesting the perceived benefit could outweigh risk when both are present.

The National Research Council (NRC) has pointed out that there are limited data on how much IRBs over-or underestimate risks. It has been noted that IRB members generally make assessments of risk based on their own experience, with almost no information on subjects’ perceptions of the probability, magnitude, duration, and consequences of these risks.²¹ It is understandable that IRB members are concerned that asking a chronically or seriously ill child about psychological issues such as depression or anxiety may provoke negative reactions. Another potential risk of psychological research is concern by individuals who share personal information.²² Yet 44% of patients and 63% of caregivers found it helpful or a relief to be asked about issues related to their child’s life, and the few that found the questions too personal (3% of patients, less than 0.5% of caregivers) suggests sensitivity to psychosocial evaluation was not a major concern. These data imply that negative reactions are minimal and IRB concerns regarding negative reactions may be unwarranted in straightforward survey research using standardized instruments. Furthermore, an Institute of Medicine panel reexamined low-risk human research protections and concluded that IRBs are placing unnecessary work burdens on themselves by not exempting social and behavioral research when federal regulations permit them to do so.²³ While federal regulation does not permit exemption of minimal risk research involving minors, these data demonstrating low burden and high benefit with pediatric psychosocial research suggests that when human subjects protection and the integrity of research methods are not of concern, IRBs may consider utilizing expedited review.

There are several important limitations to note. First, this study assessed the burden and benefit of participating in a specific cross-sectional study in an outpatient research setting. Burden and benefit associated with participating in psychosocial research might differ with longitudinal studies, where measures are repeated over time. Follow-up assessment might augment our understanding of any longer lasting benefits or perceived burden from participation in psychosocial research. Second, our study included specific self-report measures. Future investigations should ascertain whether burden or benefit changes with different study methodologies, including studies that involve interviews in addition to questionnaires. Third, our participants reported less burden and more benefit associated with increasing age. Empirically-based understanding of risks associated with psychological wellbeing and social behavior at different developmental stages is needed. Fourth, our study participants were already participating in clinical research and therefore may have been more likely to find less burden and more benefit participating in psychosocial studies. Lastly, the findings also suggest that one could both feel burden and benefit by research participation. Future studies might ask whether participants are likely to participate in similar research again to better ascertain the degree of perceived burden.

A strength of this study was our ability to address limitations noted in earlier studies²⁴ including small sample size, narrow age limit, and focus on one particular condition (e.g. cancer), each of which limits generalizability. We focused on youth across a 14-year age span, living with many types of chronic and acute illnesses. The NRC recommended researchers build questions into their studies in order to collect perceptions of risk and harm in behavioral research.²⁵ By investigating the burdens and benefits of psychosocial research participation, we sought to better understand the nature of risks associated with this type of study. Additionally, earlier studies had not attempted to address potential differences in burden and benefit as a function of race/ethnicity. We were able to note some small differences that researchers should remain attentive to.

Conclusion

Medically ill children and their caregivers participating in concurrent psychological and medical studies that involve answering questionnaires about psychological symptoms report low burden and relative benefit. Understanding the experience of being a participant in a behavioral science or psychological study is necessary for developing guidelines that ensure both the protection of human subjects and the integrity of research methods while avoiding unnecessary and burdensome procedural obstacles to clinical studies. This study suggests that resistance to psychological research based on a belief that such research is intrusive and potentially harmful overestimates the negative aspects and potentially impedes progress in studying positive psychosocial outcomes in outpatient pediatric research.