Are We Meeting the Informational Needs of Cancer Patients and Families? Perception of Physician Communication in Pediatric Oncology

Abstract

Background:

High quality oncology care is marked by skillful communication, yet little is known about patient and family communication perceptions or content preferences. Our study sought to elicit pediatric oncology patient and parent perceptions of early cancer communication to establish if informational needs were met and identify opportunities for enhanced communication throughout cancer care.

Methods:

An original survey instrument was developed, pretested, and administered to 129 patients, age 10–18, and their parents, at three cancer centers from 2011–2015. Statistical analysis of survey items about perceived communication, related associations, and patient/parent concordance, was performed.

Results:

A greater percentage of participants reported “a lot” of discussion about the physical impact of cancer (patients-58.1%(N=75), parents-69.8%(N=90)) compared to impact on quality of life (QOL) (patients-44.2%(N=57), parents-55.8%(N=72)) or emotional impact (patients-31.8%(N=41), parents-43.4%(N=56)). One fifth of patients (20.9%(N=27)) reported they had no up-front discussion about the emotional impact of cancer treatment. Parents indicated a desire for increased discussion regarding impact on family life (27.9%(N=36)), long-term QOL (27.9%(N=36)), and daily activities (20.2%(N=26)). Patients more frequently than parents indicated a desire for increased physician/patient discussion around the impact on daily activities (patients-40.3%(N=52)/parents-21.7%(N=28); P<0.001), long-term QOL (patients-34.9%(N=45)/parents-16.3%(N=21); P<0.001), pain management (patients-23.3%(N=30)/parents-7%(N=9); P<0.001), physical symptom management (patients-24%(N=31)/parents-7.8%(N=10); P<0.001), short-term QOL (patients-23.3%(N=30)/parents-9.3%(N=12); P=0.001), and curative potential (patients-21.7%(N=28)/parents-8.5%(N=11); P=0.002, P-values calculated using McNemar’s test).

Conclusion:

Oncologists may not be meeting the informational needs of many patients and some parents/caregivers. Communication could be enhanced through increased direct physician/patient communication, as well as proactive discussion of emotional symptoms and impact of cancer on QOL.

Precis:

This multi-institutional survey study of pediatric oncology patients and parents describes the perceptions of physician/parent and physician/patient communication in oncology to establish if informational needs are met. The results highlight key differences in perceived patient and parent communication experiences, content and preferences, and identifies opportunities for enhanced communication beginning early in cancer care.

Introduction

High-quality communication is a central pillar of patient- and family-centered care in oncology. Ineffective communication in the context of cancer care is a source of distress and dissatisfaction for patients and families,^1–4 while skillful communication can lead to improved health outcomes, better compliance, and enhanced patient/family satisfaction with care^1,5. From diagnosis, beginning with the “day one talk”, the oncologist can lay the foundation for a therapeutic alliance by providing clear information about diagnosis, prognosis, and treatment.⁶ Clinicians can work toward improving their effective communication by extending beyond imparting information, and honing skills such as utilizing non-technical language, open-ended questions, active listening, expression of empathy, responding to emotion, building trust, and making time and space for discussion.^6,7,8 Even when cancer communication is medically accurate and informative, and delivered with great caring and skill, it is also important that clinicians elicit and address the primary concerns of patients and families. Research has suggested that oncologists may not be meeting the informational needs of adult cancer patients especially with regard to psychological well-being and health-related quality of life.⁹

In pediatric oncology, the majority of parents of children with cancer wish to receive as much information as possible about diagnosis, treatment, and short- and long-term side effects.¹⁰ Children with cancer often have predominant concerns about the impact of cancer and its treatment on their quality of life (QOL) including, day-to-day activities, social life, and short and long term future.¹¹ Both pediatric oncology patients and parents value QOL-focused care from the beginning of cancer treatment,¹² yet little is known about how much information is effectively communicated about the potential impact of cancer therapy on QOL. Further, there is little in the existing literature with regard to physician/patient/parent communication content around the time of cancer diagnosis that would inform if we are meeting the informational needs of patients and families, especially with regard to the potential impact of cancer and its treatment.

Importantly, physician-patient/family communication in pediatric oncology is characterized as triadic, whereby the medical team must negotiate the informational and relational needs of both patients and parents,¹¹ and encompasses physician/patient, physician/parent and patient/parent relationships. Pediatric patients often receive information from both their medical providers and parents. Increased openness between parents and children with cancer can lead to improved outcomes,¹³ which is especially relevant as parents may act as gatekeepers of information. Pediatric oncology patients as young as eight years old desire direct communication from their medical team,¹¹ however many parents feel that children should not be included in physician discussions.¹⁰ Inclusion of age appropriate pediatric patients in discussions with the oncology team can help ensure that patients receive accurate information and improves patient trust and partnership in care,^6,11 yet it is important to recognize that communication with pediatric oncology patients may stem from discussions with clinicians and/or parents.

Our study sought to elicit patient and parent perceptions of communication regarding the impact of cancer and its treatment in pediatric oncology, around the time of diagnosis, as well as their preferences regarding the content of cancer communication, in order to fill an important knowledge gap and ensure that we are best meeting the informational and relational needs of patients and families. To this aim we surveyed pediatric oncology patients and parents to identify potential communication content deficiencies and/or opportunities for enhanced communication in oncology from the onset of cancer care.

Methods

Survey Development

Details of the study methodology have been previously published.¹³ In brief, original parent and patient survey instruments were created in a step-wise process including comprehensive literature review,^14,15 draft instrument development, expert review, cognitive pre-testing, behavioral pre-testing, multiple revision, and generation of final surveys.

Participant Enrollment

Pediatric oncology patients and their parents were approached for paired participation in the study at the National Institute of Health (NIH), Johns Hopkins (JH) and St. Jude Children’s Research Hospital (St. Jude), between 2011 and 2015. Patients were eligible to participate if they were between the ages of 10–18, had been diagnosed with cancer between the ages of 10–17, and were 1 month to 1 year from initial diagnosis. Enrollment required the ability to speak and understand English and informed consent/assent of both patient and parent.

Survey Administration

Surveys were administered by trained study staff to the paired participants, verbally to patients and in written form to parents, simultaneously in separate rooms, to avoid biasing results within dyads. Upon survey completion, each participant was offered a gift card in appreciation of their participation. Self-reported demographic characteristics and QOL metrics were collected via multiple choice survey questions.

To assess physician/patient/parent communication, patients and parents were asked the extent to which the following topics were discussed at the beginning of cancer therapy: physical impact of cancer treatment, emotional impact of cancer treatment, and impact of cancer treatment on QOL. Recognizing that information sharing with pediatric patients may include both the oncologist and parents, patients were asked about the extent to which these topics were discussed with them by their doctor and/or parents, and results herein represent patient perception of discussion from either or both sources without differentiation. To identify gaps in communication content, participants were asked what items they wished members of the oncology care team had discussed with them more or given them more information about, in a multiple choice/select all that apply format.

Statistical Analysis

Demographic, socioeconomic, health characteristics and questionnaire responses reported by participants were summarized using descriptive statistics. Demographic, socioeconomic, and health characteristics were compared between participants by enrollment site (NIH/JH versus St. Jude) using the Fisher’s exact test for categorical variables and the Wilcoxon rank sum test for continuous variables. Agreement between patient and parent questionnaire responses was assessed using concordance and differences in the distribution of questionnaire responses were examined by the McNemar’s or Bhapkar test. Intra-dyad concordance was defined as complete agreement in response categories, including nonresponse.

Mantel-Haenszel Chi-square test (exact p-values reported) or Fisher’s exact test were used to test for associations between patient and parent characteristics and their respective answers to the following questions: 1) the frequency of communication about the physical, emotional, and QOL effects of cancer treatments (“not at all” to “a lot”) and 2) topic areas that patients wished had increased discussion with them and parents wished had increased discussion with them or with their children (select all that apply). Patient and parent characteristics included age, gender, race, educational level, patient’s cancer diagnosis, time since patient’s diagnosis, parent’s income level, parent’s marital status, parent’s current overall health, patient’s current QOL, and patient’s projected health in one year.

Raw p-values are reported for the above associations but were adjusted for multiple testing using the false discovery rate to determine statistical significance. Statistical analyses were conducted using SAS software, Version 9.4 (SAS Institute Inc., Cary, NC). A two-sided significance level of P<0.05 was used for all statistical tests.

Results

From 2011–2015, 280 participants were approached for inclusion in the study, of which a total of 258 participants (129 patient/parent pairs, 92.1% positive response rate) were enrolled from the three participating sites. See Table 1 for a summary of demographic characteristics. The majority of participants (N=200) were enrolled at St Jude. Demographic characteristics of the participants did not differ by enrollment site except for time since diagnosis (P=0.007); a larger percentage of patient participants were enrolled within 6 months after diagnosis at St Jude compared to participants at the other sites (76%(N=76) vs. 48%(N=14)). Both patient and parent participants were asked to rate the patient’s current QOL and expected health status in one year, while parents were also asked to rate their own overall health, in a five-point Likert scale using multiple choice format. See Table 2 for a summary of self-rated health characteristics. Parent and patient survey responses for self-reported QOL and perceived communication were not found to have significant correlation with time from diagnosis. (Supplemental Tables 3 &4).

Table 1.

Patient and parent study participant demographics, N=129 patient/parent dyads

Characteristics	Patient	Parent
Age in years
10–12	34 (26%)
13–15	64 (50%)
16–18	31 (24%)
25–34		22 (17%)
35–44		63 (49%)
45–65		43 (33%)
Unknown		1 (1%)
Mean (SD)	14.0 (2.2)	41.6 (7.5)
Median (range)	14 (10–18)	40 (26–65)
Gender
Female	61 (47%)	114(88%)
Male	68 (53%)	15 (12%)
Race
White	85 (66%)	90 (70%)
Black or African American	28 (22%)	31 (24%)
Other	16 (12%)	7 (5%)
Unknown	0 (0%)	1 (1%)
Highest grade level completed
Elementary school (1–6)	39 (30%)
Junior high school (7–9)	62 (48%)
High school (10–12)	28 (22%)
Highest level of education
High school and below		57 (44%)
College and above		70 (54%)
Unknown		2 (2%)
Household income
<$25,000		38 (29%)
$25,000-$49,999		25 (19%)
$50,000-$99,999		30 (23%)
≥$100,000		35 (27%)
Unknown		1 (1%)
Marital status
Married or partnered		86 (67%)
Divorced or separated		23 (18%)
Single or widowed		20 (16%)
Cancer type (parent-reported)
Brain Tumor	20 (16%)
Leukemia	38 (29%)
Lymphoma	26 (20%)
Solid Tumor	41 (32%)
Other	1 (1%)
Unknown	3 (2%)
Time since diagnosis at survey completion
1–3 months	55 (43%)
4–6 months	35 (27%)
7–9 months	18 (14%)
10–12 months	21 (16%)

Table 2.

Patient and parent study participant self-reported health characteristics and expectations, N=129 patient/parent dyads

Characteristics	Patient	Parent
Patient’s current quality-of-life
Fair or poor	26 (20%)	32 (25%)
Good	41 (32%)	46 (36%)
Very good	32 (25%)	33 (26%)
Excellent	30 (23%)	18 (14%)
Patient’s health one year from now
The same or worse	4 (3%)	6 (5%)
Somewhat better	23 (18%)	24 (19%)
A lot better	102 (79%)	99 (77%)
Parent’s current overall health
Fair		13 (10%)
Good		39 (30%)
Very good		51 (40%)
Excellent		26 (20%)

The majority of participants (patients N=75, 58.1% and parents N=90, 69.8%) stated that they had “a lot” of discussion about how cancer treatments might make the patient feel physically (Figure 1). Older patients more frequently indicated they had “a lot” of discussion of physical effects of treatment than younger patients (26 patients aged 16–18 (83.9%) vs 33 (51.6%) patients aged 13–15 and 16 (47.1%) patients aged 10–12, P=.004). A greater proportion of patients who felt that their current QOL was “excellent” (N=25, 83.3%), reported “a lot” of discussion of physical impact compared to those who felt their current QOL was not as high, indicating “very good” (N=15, 46.9%), “good” (N=25, 61%) or “fair/poor” QOL (N=10, 38.5%, P=0.006). A greater proportion of patients who predicted their health to be “a lot better” in one year also reported “a lot” of physical impact discussion (N=65, 63.7%) compared to those who predicted their health would be “somewhat better” (N=9, 39.1%) or “the same or worse” in one year (N=1, 25%, P=0.005). Concordance within dyads was 51.2% for exact agreement in perceived physical impact communication and there was not a significant difference in patient/parent response distribution (P=0.063).

Figure 1.

Perceived physician/patient and physician/parent communication by way of response to the question: Before starting cancer treatment, did your/your child’s doctors (and or parents) discuss with you how the cancer treatments might....

a) make you/your child feel physically

b) make you/your child feel emotionally

c) affect your/your child’s quality-of-life

Options: “yes, a lot”, “yes, a moderate amount”, “yes, a little”, “no”

Note: Overall percentages may not sum to 100% because missing data is not shown in the figure.

Parents and patients both reported less discussion about the emotional impact of cancer therapy before starting cancer treatment compared to discussion of physical impact, with 31.8% of patients (N=41) and 43.4% of parents (N=56) reporting “a lot” of discussion of emotional impact. (Figure 1). Before starting treatment, 20.9% of patient respondents (N=27) reported having no discussion about how their cancer treatment might make them feel emotionally. (Figure 1). Concordance within dyads was 34.1% for exact agreement in self-reported extent of communication about the emotional impact of cancer, and a significant difference in overall distribution between patient and parent responses was observed (P=0.006). A larger percentage of older patients, 16–18 years old (N=15, 48.4%), reported “a lot” of discussion about emotional impact, as compared to younger patients (sixteen patients aged 13–15 (25%) and 10 (29.4%) patients aged 10–12, P=0.007). No patients who thought that their health would be “the same or worse” in one year reported “a lot” of emotional impact discussion, as compared to more than one-third (N=36, 35.3%) of patients who predicted “a lot better” health in one year and more than one-fifth who predicted health to be “somewhat better” in one year (N=5, 21.7%, p=0.007).

Fifty-seven (44.2%) patients and 72 (55.8%) parents perceived that they had “a lot” of discussion about how cancer treatments might affect the patient’s QOL before the start of cancer therapy (Figure 1). Concordance within dyads was 37.2% for exact agreement in self-reported communication about the impact of cancer on QOL, and there was not a significant difference in in patient/parent response distribution (P=0.067).

In reference to physician/parent communication, the three most common items parents indicated they wish had been discussed with them further were the impact of cancer and its treatment on: family life (27.9%, N=36), QOL after cancer treatment (27.9%, N=36), and daily activities (20.2%, N=26). (Figure 2). Addressing physician/patient communication preferences, patient participants indicated several topics that they wished their oncologist would have discussed further, including: the impact of cancer and its treatment on daily activities (40.3%, N=52), QOL after cancer treatment (34.9%, N=45), identifying fears and hopes about illness (28.7%, N=37), physical symptom management (24%, N=31), pain management (23.3%, N=30), impact of different available therapies on QOL (23.3%, N=30), the possibility of not being cured (21.7%, N=28), emotional symptom management (20.9%, N=27), and impact on family life (20.9%, N=27). Similar responses were observed among parent participants, when asked which topics they wish had increased discussion with their children, in the areas of emotional symptom management (19.4%, N=25) and the impact on family life (19.4%, N=25) In contrast, parents less frequently indicated an interest in their children having increased communication regarding the majority of these topics, including: the impact of cancer and its treatment on daily activities (40.3%(N=52) vs. 21.7%(N=28), P<0.001); the impact of cancer and its treatment on QOL after cancer treatment (34.9%(N=45) vs. 16.3%(N=21), P<0.001); how to treat pain (23.3%(N=30) vs. 7%(N=9), P<0.001); how to treat physical symptoms (24%(N=31) vs. 7.8%(N=10), P<0.001); how available cancer treatments differ in the ways they could affect QOL (23.3%(N=30) vs. 9.3%(N=12), P=0.001); and the possibility of not being cured (21.7%(N=28) vs. 8.5%(N=11), P=0.002).

Figure 2.

Perceived gaps in communication/opportunities for enhanced discussion by way of response to the question: (Select all that apply)

Patient:

What things do you wish that members of your oncology care team had discussed with you more or given you more information about?

Parent:

a) Parent for Child: What things do you wish that members of your child’s oncology care team had discussed with your child more or given your child more information about?

b) Parent for Self: What things do you wish that members of your child’s oncology care team had discussed with you more or given you more information about?

Discussion

Quality communication is a cornerstone of excellent clinical care and proactive discussion of cancer and its impact may reduce anxieties that arise from unanswered questions or unexpected side effects during therapy. Both patient and parent participants reported the greatest degree of communication regarding the physical impact of cancer therapy, however nearly a quarter of patients indicated a desire for more physician discussion regarding the treatment of pain or other physical symptoms during cancer therapy. This finding indicates a potentially unmet need among pediatric oncology patients regarding physical symptom communication, which may also go unrecognized by their parents as desire for increased discussion of pain and physical symptoms amongst parent participants was significantly lower.

Patient participants who reported that they had “a lot” of discussion about the physical impact of cancer therapy more frequently rated their current QOL as “excellent” (P=0.006) and were more likely to predict “a lot better” health in one year (P=0.005) suggesting that enhanced physical symptom communication is linked to improved patient QOL and future outlook. While causation cannot be established in this context, these data underscore the importance of utilizing patient-reported outcome measures and providing anticipatory guidance for physical symptoms throughout cancer therapy. Our finding is consistent with prior studies which demonstrated that improvements in provider communication lead to enhanced QOL scores¹⁶ and that increased information-sharing leads to improved outcomes¹⁷.

Our study identified that patients and parents perceive less frequent communication regarding the emotional impact of cancer therapy as compared to physical impact or impact on QOL. Similarly, a recent review of communication studies in pediatric oncology identified physician discussion of emotions to be less frequent than other aspects of care, with physicians deferring these discussions to others¹⁷. The fact that greater than one fifth of patients reported having no discussion about the impact of cancer on their emotional well-being, fewer than one-third reported talking “a lot” about emotional impact, and nearly one-quarter indicated a desire for increased physician discussion about emotional symptoms, suggests that discussion about the emotional impact of cancer therapy is a potentially unmet need of many children and adolescents with cancer. Importantly, no patients who thought their health would be the same or worse in one year reported “a lot” of discussion about their emotions.

Children and adolescents with cancer may be at an increased risk for a major depressive disorder, depressive symptoms, or anxiety.^18,19 In fact, about half of our survey participants self-reported suffering from depression (49.6%) and anxiety (59.7%) in the first month of cancer therapy¹². Given the paucity of discussion between oncologists and children with cancer about emotional health, it is possible that mental health needs may be underrecognized or undertreated, and increased communication may lead to improved outcomes. Importantly, research has demonstrated in adult oncology that while physicians and patients both express the desire and willingness to discuss the emotional and social impact of cancer, very few discuss these topics in practice.²⁰ Further, parents of children with cancer tend to demonstrate a strong desire to discuss the emotional impact of their experience²¹, yet even when patients and caregivers initiate these conversations clinicians may miss the cues and/or avoid discussions of emotional and social topic areas.^20,21 Similarly, our findings suggest that pediatric patients’ could benefit from increased provider initiated discussion of the emotional impact of cancer therapy.

Children and adolescents wish to know how treatments will affect their daily life now and in the future,¹¹ and our findings suggest that we may not be meeting their informational needs regarding current and future QOL, as both patients and parents most frequently indicated these topics as areas about which they desired further physician discussion. Providers could enhance communication in this area by initiating open-ended discussions with pediatric oncology patients and families about their goals, values, and concerns, especially regarding the impact of cancer treatment on QOL throughout the illness trajectory¹⁹. Improved communication around self-perceived QOL is a starting point that also may open doors to conversations about ongoing symptoms requiring clinical intervention. Further, more than one-quarter of patient participants indicated that they wished for increased provider discussion concerning their fears and hopes about their illness. Initiating open-ended physician/patient communication may also create opportunities to explore patients’ fears and hopes, which may lead to fostering enhanced therapeutic alliance and patient care. Observed pediatric oncology communication in the early cancer treatment period has demonstrated that providers may be missing opportunities to elicit patient and parent concerns²² and clinicians can better meet the informational and relational needs of patients and families by proactively assessing their concerns and opening the conversation.

Patient age was identified as a significant factor in patient perception of physical and emotional communication, with older patients aged 16–18 reporting more frequent discussion about these topics than did younger children aged 10–15. Research has demonstrated that children as young as 8 years desire communication about their illness and treatment and many prefer direct communication from their clinicians¹¹. Further, our data demonstrate that perceptions around illness communication are not well aligned between pediatric oncology patients and their parents, and patients were more likely than parents to indicate a desire for increased physician/patient discussion of multiple topic areas. Since parents may serve as gatekeepers of information, patients’ informational and relational needs may not be met if they do not have directed communication with their medical providers and caregivers may not be adequate proxies. Further when including pediatric patients in discussions, providers should ensure that the patient is not merely present but is included to the degree they desire, has information in terms they can understand, and has their concerns assessed and addressed. Whether discussions are solely between physician and patient, or occur in a triadic manner, providers should prioritize communication with pediatric oncology patients of all ages, in a developmentally appropriate manner^23,24 throughout the disease trajectory.

Multiple topic areas for expanded discussion were identified by parent and patient participants in our study. These areas represent opportunities for enhanced physician/patient and physician/parent discussion. Greater than one third of patients indicated a desire for increased physician/patient discussion about the impact of cancer on current and future QOL, and more than one-fifth indicated a desire for increased physician/patient discussion around fears and hopes about their illness, pain management, physical and emotional symptom management, prognosis, decision-making, and impact on family. Opportunities identified for increased physician/parent communication, by around one fifth of parents and beyond, included discussion about the impact of cancer and its treatment on family life, patients’ current and future QOL and emotional symptoms. Enhanced physician communication in these critical topic areas can aid clinicians in better meeting the informational and relational needs of cancer patients and families. While this data can help guide clinicians, it is important to elicit the communication needs and preferences of each patient and family and create time and space for expanded discussion of the areas that are most important to them, to improve individualized patient and family-centered care.

Limitations

We acknowledge several limitations of this study. The survey study design intrinsically risks participant recall bias, which could affect accuracy of results. The survey tools were developed and tested in English, and as such non-English speakers were excluded from this study, which might introduce cultural biases. Further, patient/parent dyads for children under age 9 were not included in the study and this population might have different informational needs. The survey utilized was not a validated instrument and participant responses may have been influenced by questions that lacked clarity; however, rigorous survey development and extensive pre-testing mitigate this concern. Lastly, while our study reveals potentially unmet informational needs, it is not discernable if this is a function of communication structure (for example patient inclusion/exclusion) or the process by which information is delivered; an important distinction that would benefit from future investigation.

Conclusion

Based on observed parent and patient perceptions of communication in pediatric oncology, opportunities for enhanced communication have been identified which could lead to improved dialogue and relationship-based cancer care throughout the disease trajectory. Specific areas highlighted as opportunities for increased proactive discussion include the impact of cancer and its treatment on emotional well-being and on current and future QOL. The informational needs of many cancer patients are not being met and clinicians should make every effort to include patients in discussions whenever feasible and create a space for increased information sharing with both patients and caregivers, to optimize cancer communication and care.

Protection of Human Subjects.

This multi-institutional survey study was performed after approval by the institutional review board at each participating site to assure the ethical conduct of research and protection of human subjects, in accordance with the principles embodied in the Declaration of Helsinki and in compliance with the U.S. Department of Health and Human Services. The investigators obtained informed consent from each competent adult participant. For each pediatric participant informed consent was obtained from the pediatric participant’s guardian as well as assent from the pediatric participant.