Shared spiritual beliefs between adolescents with cancer and their families

Abstract

Background:

FAmily CEntered (FACE) Advance Care Planning helps family decision makers to understand and honor patients’ preferences for future health care, if patients cannot communicate. Spiritual well-being is a key domain of pediatric oncology care and an integral dimension of pediatric advance care planning.

Procedure:

As part of four-site randomized controlled trial of FACE for teens with cancer, the functional assessment of chronic illness therapy-spiritual well-being-version 4 (FACIT-Sp-EX-4) was completed independently by 126 adolescents with cancer/family dyads. The prevalence-adjusted and bias-adjusted kappa (PABAK) measured congruence on FACIT-Sp-EX-4.

Results:

Adolescents (126) had mean age of 16.9 years, were 57% female and 79% White. Religious/spiritual classifications were: Catholic (n = 18), Protestant (n = 76), Mormon (n = 3), none/atheist (n = 22), other (n = 5), and unknown (n = 2). Agreement at item level between spiritual well-being of adolescents and families was assessed. Three items had ≥90% agreement and Excellent PABAK: “I have a reason for living,” “I feel loved,” “I feel compassion for others in the difficulties they are facing.” Three items had <61% agreement and Poor PABAK: “I feel a sense of harmony within myself,” “My illness has strengthened my faith or spiritual beliefs,” “I feel connected to a higher power (or God).” Dyadic congruence was compared by social-demographics using median one-way analysis. Male family members (median = 72%) were less likely to share spiritual beliefs with their adolescent than female family members (median = 83%), P = .0194.

Conclusions:

Family members may not share spiritual beliefs with adolescents and may be unaware of the importance of spiritual well-being for adolescents.

1|. INTRODUCTION

Spiritual well-being is a key domain of pediatric oncology care and an integral dimension of pediatric advance care planning (pACP). Spiritual well-being has been positively correlated with overall quality of life in adolescents living with cancer and their family members.¹ As end-of-life decisions are often made with patient and families together, it is important to understand the family dynamic of critically ill people. In a pediatric context, end-of-life decisions for children and teens under 18 or 19 are made by parents or legal guardians; for those age 18/19 years or older, ideally a patient-chosen adult makes decisions as the adolescents’ surrogate decision maker if the adolescent is legally or medically unable to do so themselves.

Many health care providers agree that good spiritual care can dramatically improve health outcomes.^2–4 As an update to its clinical practice guidelines, the American Society of Clinical Oncology now recommends that those living with cancer in both inpatient and outpatient cancer care settings should have access to comprehensive psychosocial care, which includes spiritual supportive services, yet this is still not standard of practice.⁵ The standards for psychosocial care for children with cancer further recommends appropriate care for both persons living with cancer and their families to improve quality of life.^6,7

Data on spirituality in pediatric cancer are limited. Understanding adolescent-family agreement in spiritual well-being is important. In clinical practice, we have observed differences in spiritual well-being can lead to alienation between an adolescent with cancer and a parent at a time when support and connectedness are so critical. For example, a mother may believe that the cancer is “God’s will” while the adolescent finds this idea enraging. Or, an adolescent may feel strengthened in his belief in God since receiving a cancer diagnosis, while the father is contemptuous of this belief.

FAmily CEntered (FACE) pACP for teens with cancer (FACE-TC) facilitates conversations between family decision makers and adolescents with cancer about the adolescents’ preferences for future medical decisions, regardless of current disease status, if adolescents cannot speak for themselves.⁸ Analyzed here is a cross-sectional analysis of baseline spiritual well-being data from the ongoing FACE-TC trial.

2|. METHODS

2.1|. Participants and setting

This secondary analysis is part of the ongoing trial of the FACE-TC pACP, which is a longitudinal, intent-to-treat, single-blinded, randomized controlled trial for adolescents diagnosedwith cancer and their family decision makers. Participants were enrolled from four quaternary pediatric hospital-based settings in urban locations of the United States. Enrollment was from July 16, 2016 through April 30, 2019. Methods of this study and the protocol have been published elsewhere.⁸ Eligibility criteria for adolescents were ≥14 and <21 years old at the time of enrollment, have a cancer diagnosis, English speaking, and without developmental delays. Family decision makers had to be 18 years of age or older and know the adolescent’s diagnosis. If the adolescent was under the age of 18, the decision maker had to be their legal guardian. If they were between age 18 and 21, they could decide who was their decision maker. Teens with cancer and family members were excluded if on secondary screening they scored a 29 or greater on the Beck Depression Index,⁹ had active suicidal and/or homicidal ideation, or demonstrated psychosis. Participants were enrolled in the study by trained Research Assistant-Assessors (RA-Assessor). Randomized intent-to-treat intervention assignment occurred postbaseline assessment. All the data described in this study were from baseline assessment of enrolled participants, prior to randomization in the trial. All the participants who failed secondary screening received appropriate referrals and were not enrolled in the trial.

This study received Institutional Review Board approval at all participating sites. Participants with cancer ages 18 years up to 21 years provided informed written consent. All participants under the age of 18 provided written informed assent with parent or guardian consent as well.⁸ A Safety Monitoring Committee monitored the protocol yearly. There were no recorded adverse or serious adverse events related to the study procedure. There were participant deaths that occurred, which were unrelated to the study procedures.

2.2|. Measures

2.2.1|. Demographics survey

Trained and blinded RA-Assessors collected both patient and family decision maker’s self-reported socio-demographics at baseline, post enrollment into the study. Data were verified through chart abstraction. These data were used to identify the social determinants of health examined in the analysis, which were gender, race, education, and income.

2.2.2|. Functional assessment of chronic illness therapy-spiritual well-being-expanded- version 4 (FACIT-Sp-EX-4)

During baseline assessment, adolescents and families were administered the FACIT-Sp-EX-4 face-to-face by a trained RA-Assessor as part of the set of outcome questionnaires. The FACIT-Sp-EX-4 has demonstrated factorial, convergent, and concurrent validation among patients with cancer.^10–13 This 23-question survey analyzes aspects of participants’ general spiritual well-being and in the context of illness. Answer choices are on a 5-point Likert scale: “Not at all,” “A little bit,” “Somewhat,” “Quite a bit,” and “Very much.” For the present analysis, responses were collapsed into three categories: “Not at all,” “A little bit/Somewhat,” and “Quite a bit/Very much” due to small sample size.

2.2.3|. Brief multidimensional measurement of religiousness/spirituality (BMMRS)

A validated survey^14–16 was also collected at baseline to identify religious and spiritual beliefs for both patient and family decision makers. This 41-item survey includes questions involving religious history, support, coping, and other religious/spiritual practices. For this analysis, data from the question “What is your current religious preference?” were collected. This question was open-ended, so participants were able to reply as was appropriate for them.

2.3|. Procedure

Clinicians identified potentially eligible patients, who were then approached by the trained RA-Assessor during a clinic visit. Eligible dyads (patient and family) were administered assessment measures independently at baseline (prior to randomization), immediately post-Session 2, and at 3, 6, 12, and 18 months postintervention. Follow-up assessments are ongoing. RA-Assessors reviewed medical records to confirm cancer status, health care utilization, and comorbidities. A second, blinded RA-Assessor, who was not an interventionist, administered nine surveys to the adolescent patients and six surveys to the family decision makers at baseline after consent/assent was obtained. All questionnaires were read aloud to all participants to control for literacy or uncorrected vision, which also served as an engagement strategy and prevented missing data. Responses were directly entered by the RA-Assessor into the research electronic data capture system (REDCap) database. Of these surveys, the FACIT-Sp-Ex-4 spiritual well-being data were assessed from family decision makers and adolescents with cancer independently as well as total family congruence on spiritual well-being.

2.4|. Data analysis

Data were collected and stored in REDCap at all sites. Analysis of congruence at item level was completed using a kappa statistic, the prevalence-adjusted and bias-adjusted kappa (PABAK).^17,18 To adjust for bias in kappa statistics caused by imbalanced responses, also known as the “kappa paradox” (eg, high prevalence of “very much” responses), the PABAK was calculated for assessment of agreement.^19,20 PABAK scores range from 0 to 1, wherein the level of congruence varies according to the following scores: 1.00–0.75, Excellent; 0.74–0.60, Good; 0.59–0.40, Fair; and <0.40, Poor.^17,18 In tandem with the PABAK, percentage of agreement between adolescent and family responses of “Quite a bit/Very much” to the 23 FACIT-Sp-EX-4 questions was considered as a measure of overall dyadic congruence. This percentage was compared by socio-demographic group, specifically race, gender, education, and poverty level. As the percentage was highly skewed, median one-way analysis was conducted for comparisons. SAS 9.2 was used for the analyses (SAS Institute Inc., SAS 9.2 Enhanced Logging Facilities, Cary, NC; 2008).

3|. RESULTS

Adolescents with cancer (N = 126) enrolled in this study had a mean age of 16.9 years (range 14–20), were 57% female, and were 79% non-Hispanic White. Religious classifications were categorized into the following groups: Catholic (n = 18), Protestant (n = 76), Mormon (n = 3), none/atheist (n = 22), other (n = 5), and unknown (n = 2) (Table 1). Family demographics revealed that 58% of this population has less than a Bachelor’s degree and approximately 56% have a family income at or below 200% of the Federal poverty line (Table 2). Surrogate decision makers for the adolescents were predominantly biological mother (75%), followed by biological father (15%). All other surrogate decision makers were other family members (such as siblings, grandmothers, aunts, uncles, adoptive parents, etc), significant others, or legal guardians.

TABLE 1.

Demographics of adolescents with cancer

Description	Statistics
Age
Mean (SD)	16.9 (1.9)
Range	(14.0, 20.0)
Gender	N (%)
Male	54 (42.9)
Female	72 (57.1)
Race
Asian	3 (2.4)
Black or African American	17 (13.5)
White	100 (79.4)
More than one race	5 (4.0)
Declined	1 (0.8)
Ethnicity
Not Hispanic or Latino	116 (92.1)
Hispanic or Latino	5 (4.0)
Declined	5 (4.0)
What is your current religious preference?
Catholic	18 (14.3)
Protestant	76 (60.3)
Mormon	3 (2.4)
None/atheist	22 (17.5)
Other	5 (4.0)
Unknown	2 (1.6)
Hospital participants
Children’s National	3 (2.4)
Akron Children’s	77 (61.1)
St Jude Children’s	11 (8.7)
University of Minnesota	35 (27.8)

TABLE 2.

Demographics of family decision makers

Description	Statistics
Age
Mean (SD)	45.9 (8.3)
Range	(19.0, 67.0)
Gender	N (%)
Male	22 (17.5)
Female	104 (82.5)
Race
American Indian or Alaska Native	1 (0.8)
Asian	3 (2.4)
Black or African American	14 (11.1)
White	103 (81.7)
More than one race	5 (4.0)
Ethnicity
Not Hispanic or Latino	121 (96.0)
Hispanic or Latino	4 (3.2)
Declined	1 (0.8)
Education
No high school diploma or GED	2 (1.6)
High school or GED equivalency	23 (18.3)
Some college but no Bachelor’s	48 (38.1)
Bachelor’s degree	30 (23.8)
Master’s degree	20 (15.9)
Doctorate (PhD, etc)	1 (0.8)
Professional degree	2 (1.6)
Employment Status
Full-time student	2 (1.6)
Part-time employed or self-employed	15 (11.9)
Full-time employed or self-employed	79 (62.7)
Retired	3 (2.4)
Unemployed	5 (4.0)
Disability/SSI or SSD	7 (5.6)
Other, such as homemaker	15 (11.9)
Income
≤ Federal poverty line	33 (26.2)
101–200% of Federal poverty line	37 (29.4)
201–300% of Federal poverty line	19 (15.1)
>300% of Federal poverty line	33 (26.2)
Declined	4 (3.2)
Relationship with adolescent
Biological mother	94 (74.6)
Biological father	19 (15.1)
Adoptive mother	5 (4.0)
Biological grandmother	1 (0.8)
Step-grandmother	1 (0.8)
Aunt	1 (0.8)
Uncle	1 (0.8)
Boy/girl friend	3 (2.4)
Legal guardian	1 (0.8)
Hospital participants
Children’s National	3 (2.4)
Akron Children’s	77 (61.1)
St Jude Children’s	11 (8.7)
University of Minnesota	35 (27.8)

Our results showed that agreement between family decision makers and patients substantially vary by the FACIT-Sp-EX-4 items. The 5-point Likert scale was collapsed into three categories, which were used to calculate the PABAK. Overall agreement for each item may vary between answer choices, meaning that overall agreement can be “Not at all,” “A little bit/Somewhat,” or “Quite a bit/Very much.” Items “I have a reason for living,” “I feel loved,” and “I feel compassion for others in the difficulties they are facing” all had greater than 94% agreement and an Excellent PABAK score (Table 3). The highest recorded congruence was for item “I have a reason for living” with an overall congruence between patient and families of 97% and 0.95 PABAK. Following this, “I feel loved” had an overall congruence between patient and families of 94% and 0.92 PABAK. Lastly, “I feel compassion for others in the difficulties they are facing” had an overall congruence between patient and families of 94% and 0.90 PABAK. On the other hand, items “I feel a sense of harmony within myself,” “My illness has strengthened my faith or spiritual beliefs,” and “I feel connected to a higher power (or God)” had equal to or less than 60% agreement and had a Poor PABAK score (Table 3). “I feel a sense of harmony within myself” had an overall congruence between dyads of 60% and 0.40 PABAK. Majority of family decision makers reported (82%) a sense of harmony within themselves as did adolescents with cancer (72%). “My illness has strengthened my faith or spiritual beliefs” received an overall congruence between dyads of only 56% and 0.33 PABAK. Lastly at the lowest reported congruence, “I feel connected to a higher power (or God)” received an overall congruence between dyads of only 56% and 0.34 PABAK. All other items are within an overall agreement of 61–93% and within Fair to Excellent PABAK scores.

TABLE 3.

Congruence of FACIT-Sp-EX-4 between adolescents with cancer and family decision makers

Question	Agreement by each answer
	Not at all	A little bit/Somewhat	Quite a bit/Very much	Overall agreement	PABAKa
	N (%)	N (%)	N (%)	N (%)
1.1 feel peaceful	0 (0)	8 (6.3)	80 (63.5)	88 (69.8)	0.55
2. I have a reason for living	0 (0)	0 (0)	122 (96.8)	122 (96.8)	0.95
3. My life has been productive	0 (0)	0 (0)	101 (80.2)	101 (80.2)	0.7
4. I have trouble feeling peace of mind (reversed)	1 (0.8)	5 (4.0)	76 (60.3)	82 (65.1)	0.48
5. I feel a sense of purpose in my life	0 (0)	1 (0.8)	111 (88.1)	112 (88.9)	0.83
6. I am able to reach deepdown into myself for comfort	0 (0)	7 (5.6)	76 (60.3)	83 (65.9)	0.49
7. I feel a sense of harmony within myself	0 (0)	3 (2.4)	73 (57.9)	76 (60.3)	0.4
8. My life lacks meaning and purpose (reversed)	1 (0.8)	1 (0.8)	110 (87.3)	112 (88.9)	0.83
9. I find comfort in my faith or spiritual beliefs	0 (0)	2 (1.6)	78 (61.9)	80 (63.5)	0.45
10. I find strength in my faith or spiritual beliefs	0 (0)	4 (3.2)	73 (57.9)	77 (61.1)	0.42
11. My illness has strengthened my faith or spiritual beliefs	4 (3.2)	6 (4.8)	60 (47.6)	70 (55.6)	0.33
12. I know that whatever happens with my illness, things will be okay	0 (0)	10 (7.9)	85 (67.5)	95 (75.4)	0.63
13. I feel connected to a higher power (or God)	1 (0.8)	9 (7.1)	61 (48.4)	71 (56.3)	0.34
14. I feel connected to other people	0 (0)	4 (3.2)	83 (65.9)	87 (69.1)	0.54
15. I feel loved	0 (0)	0 (0)	119 (94.4)	119 (94.4)	0.92
16. I feel love for others	0 (0)	0 (0)	117 (92.9)	117 (92.9)	0.89
17. I am able to forgive others for any harm they have ever caused me	0 (0)	8 (6.3)	71 (56.3)	79 (62.6)	0.44
18. I feel forgiven for any harm I may have ever caused	0 (0)	7 (5.6)	75 (59.5)	82 (65.1)	0.48
19. Throughout the course of my day I feel a sense of thankfulness for my life	0 (0)	3 (2.4)	96 (76.2)	99 (78.6)	0.68
20. Throughout the course of my day I feel a sense of thankfulness for what others bring to my life	0 (0)	0 (0)	108 (85.7)	108 (85.7)	0.79
21. I feel hopeful	0 (0)	1 (0.8)	111 (88.1)	112 (88.9)	0.83
22. I feel a sense of appreciation for the beauty of nature	0 (0)	1 (0.8)	114 (90.5)	115 (91.3)	0.87
23. I feel compassion for others in the difficulties they are facing	0 (0)	0 (0)	116 (93.5)	116 (93.5)	0.9

Abbreviation: PABAK, prevalence-adjusted bias-adjusted kappa.

^aLess than 0.40: Poor; between 0.40 and 0.59: Fair; between 0.60 and 0.74: Good; between 0.75 and 1.00: Excellent.

Measures of social determinants of health, in essence adolescent gender, family member race, family member education level, and household income, were not statistically associated with overall congruence on spiritual well-being, as shown in Table 4. The percentage of agreement on the 23 FACIT-Sp-EX-4 items was not normally distributed, so instead of mean percentage, the median percentage of agreement was compared by socio-demographics. Male family members were less likely to have the same spiritual beliefs as their adolescent compared to female family members (median 72% vs 83%, respectively; P = .0194).

TABLE 4.

Descriptive statistics of median percentage of agreement on 23 FACIT-Sp-EX-4 items by social demographics (N = 126 dyads)

Variable	Category	N	Median (%)	% Range (minimum, maximum)	P-valuea
Adolescent gender	Male	54	78	(35, 100)	.2820
	Female	72	83	(30, 100)
Family member gender	Male	22	72	(57, 100)	.0194
	Female	104	83	(30, 100)
Family member race	Non-White	23	87	(30, 100)	.1078
	White	103	78	(35, 100)
Family member education	No high school diploma or GED equivalency or high school or GED equivalency	25	78	(61, 100)	.3417
	Some college but no Bachelor’s degree	48	83	(30, 100)
	Bachelor’s degree, Master’s degree, Doctorate (PhD, etc)	53	78	(39, 100)
Household income	Equal to or below the Federal poverty line	33	83	(30, 100)	.4736
	Above 100% of Federal poverty line	89	78	(39, 100)
	Declined	4	85	(57, 100)

^aComparing social-demographics using median one-way analysis.

4|. DISCUSSION

We identified important areas of concordance and discordance in shared spiritual well-being between adolescents with cancer and their family members, Adolescent/family dyads showed excellent agreement on having a reason for living, feeling loved, and feeling compassion for others. Conversely, dyads showed an impressive difference between key concordance topics for adolescents and their adult decision makers, specifically in strengthened spiritual beliefs since cancer diagnosis, connection to a higher power, and harmony within self. Highest levels of concordance were more “humanistic” in that they are related to person-to-person connections, while lowest levels of concordance between adolescents and family members were more “religious” in that they referred to a “higher power (or God)” or “felt cancer strengthened their spiritual beliefs.” Contrary to our hypotheses, social determinants of health were not associated with congruence between adolescents and family decision makers on spiritual well-being in this sample. Families may be unaware of the importance of spiritual well-being for adolescents. Clinicians may expect that families experience the same level of spiritual connection as adolescents. In either of these cases, many of the psychosocial needs associated with the spiritual beliefs, or lack thereof, may be overlooked.

The finding of discordance in key topics of spiritual well-being may inform interventions that improve religious and spiritual communication. Previous research indicates that these differences have clinical implications, including contributing to stress through the medical decision-making process.²¹ Spirituality is important to teens suffering from severe illness and their families, especially when nearing end of life.^22,23 When people suffer spiritually, this can result in additional physical suffering for those who are already critically ill.²⁴ Often patients do not report a lack of spiritual well-being due to concern that this will negatively affect other family members’ moods and emotional state,²⁵ inadvertently adding burden onto those who are already critically ill.²⁶ Conversations in which adolescents with cancer are able to express potential spiritual concerns with their family decision makers may relieve this burden and continue to improve quality of life. Assessment of discordance in spiritual well-being is potentially an opportunity to reflect on questions of meaning amid the context of suffering associated with cancer. Future research should explore how to strengthen adolescent/family spiritual well-being or meaning-making for more nonreligious persons.

Study findings support the need for early spiritual assessment with the offering of continued support for the spiritual functioning of the 83% of adolescents with cancer who identified as spiritual or religious. In the Psychosocial Standards of Care for AYAs with cancer, there was just one mention of spirituality in one standard and none of the standards used “spirit” or “spiritual” as a search term in developing recommendations.⁶ Our study findings support the call in a separate paper later published in Pediatric Blood & Cancer for a very specific spiritual standard.²⁷ Thus findings are consistent with the recommendations of this recent systematic review.²⁷

Seventeen percent of adolescents (22/126) self-identified as having no religion or as an atheist. Nationwide survey data of young adults (ages 18–39 years) in the United States indicate 15% report no religious affiliation, while only 10% of families report no religious affiliation.²⁸ Adolescents who identified as having no religion or as atheists may still have existential or meaning-making needs. Support or interventions to meet these needs through meaning-making interventions may contribute to quality-of-life outcomes, as has been found in adults.^29–31 Discordance in spiritual beliefs between female adolescents and male family members is consistent with research by the Pew Forum that shows White Protestant females are typically more religious than White Protestant males.^32,33 This finding merits further research on the role that male parents play in an adolescent’s (with cancer) spiritual well-being.

Adolescents are often at a stage in their development where they begin to question³⁴ and distance themselves from the traditional religious views in which they were raised.^35–37 Roughly one-in-five US adults were raised with a mixed religious background, according to a Pew Research Center study.³⁸ This includes about one-in-10 who say they were raised by two people, both of whom were religiously affiliated but with different religions, such as a Protestant mother and a Catholic father, or a Jewish mother and a Protestant stepfather. An additional 12% say they were raised by one person who was religiously affiliated and another person who was religiously unaffiliated (atheist, agnostic, or “nothing in particular”).³⁸ People who live with chronic advanced illness, such as cancer, and their families may face spiritual struggles that, if addressed, could alleviate the burden of illness. Given the consensus that the spiritual needs of children with cancer and their families should be identified and addressed from the onset of diagnosis and throughout disease progression,^27,39,40 the first step could be identifying the religious affiliation of all family members who are involved in caregiving and decision making for adolescents with cancer.

Study findings support recommendations for assessment of spirituality through hospital-based chaplaincy services. Although spiritual and religious aspects of a person’s critical illness are understood as important, chaplaincy care is not often integrated into standard oncology care.^41–43 A study of adults with cancer showed that the majority of patients believed health care providers’ consideration of spiritual beliefs was important for care,⁴⁴ which supports the integration of spiritual care as a standard component of cancer care and, further, of palliative care medicine. To date, much of the interest in quality of life has focused solely on physical and psychological aspects of care.⁴⁵ Pediatric ACP may also facilitate a culture of integration of spiritual perspectives directly from children and teens themselves in pediatric health care. Future studies should examine longitudinal changes in adolescent/family congruence in religious beliefs and well-being as they apply to psychosocial care of adolescents living with cancer. It is recommended by the Standards for the Psychosocial Care of Children with Cancer^6,7 to incorporate routine assessments of children’s psychosocial needs, including religious and spiritual support,⁴⁶ to ensure quality of life and care are maintained. More research is needed to determine if a more “generalist” approach to spiritual assessment is appropriate, or if a specialist approach using chaplains has better outcomes for adolescents with cancer and their families.⁴⁷

Strengths of this study include the design and completion of surveys in four real-world hospital-based oncology settings increases the generalizability of findings to clinical practice. The targeted enrollment of 130 adolescents with cancer/family dyads were accomplished. Analyzed data were controlled for gender, so the results were more robust, and gender was relatively equal across the adolescents enrolled in this study. All patients and families self-identified that they feel loved in some capacity, which may be earthly love or an omnipresent love as described within an attachment to a higher being.⁴⁸ The study sample was economically diverse with the majority of the participants living in low-income households. Geographical diversity also existed because of the quaternary nature of the four pediatric hospitals. Lastly, over 95% of the data collected were complete due to good fidelity to the research protocol by research staff and the face-to-face administration of study questionnaires.

Limitations of this study include the cross-sectional design, which prevents conclusions about causality. We also do not know the stability of spiritual well-being over time. Additionally, we did not collect data on religious variability within households. Findings may not generalize beyond the population of adolescents 14–20 years old, with a diagnosis of cancer, who are English speaking, and reside in the United States. Our population did not have a great degree of racial diversity, as it represented predominantly White adolescents and families, two-thirds of whom identified as Protestant, which does not provide us with as much religious/spiritual variability as is seen in other racial classifications.⁴⁹ According to a report from the Pew Research Center, over 80% of adults who identify as Protestant are White/Caucasian, whereas many other religions (ie, Buddhism, Hindu, Muslim, historically Black Protestant, etc) have much greater minority populations.²⁴ Participant responses were collected in secular settings, so the scores calculated may represent individual self-representation biases. Limitations may also include social desirability bias, which may have influenced responding, if participants said what they thought the facilitator wanted to hear. This study is subject to selection bias, with adolescent/family dyads who participated in a pACP trial being more likely to be comfortable talking about death and dying.

Adolescents with cancer and their families in this study felt they had a reason for living, felt loved, and felt compassion for others. However, adolescents were less likely than their family members to experience a feeling in harmony within themselves or to believe that cancer had strengthened their faith or spiritual beliefs. Adolescents were also less likely than their family members to feel connected to a higher power (or God). Future research should examine the importance of these discordant beliefs for adolescents’ quality of life and cancer-related symptom profile over time, as both adolescents and family decision makers have stated this to be an important aspect of care for them.^42,46

Abbreviations:

FACE-TC pACP

FAmily CEntered pediatric advanced care planning for teens with cancer

FACIT-Sp-EX-4

functional assessment of chronic illness therapy-spiritual well-being-expanded - version 4

PABAK

prevalence-adjusted and bias-adjusted kappa

RA-Assessor

Research Assistant-Assessor

REDCap

research electronic data capture system