A pilot study of the Hopetimize program for pediatric oncology healthcare professionals

Abstract

Background

Research suggests that burnout and work-related quality of life are concerns for pediatric oncology professionals, with limited supportive resources. This study aimed to address this resource gap through an intervention promoting hope and targeting burnout and quality of life in this group.

Methods

A single-arm pilot study of the intervention was conducted at Rady Children’s Hospital San Diego. The intervention was delivered as a single-session workshop, followed by reinforcement through a mobile application. Feasibility was assessed through recruitment and retention metrics, application usage data, and informal participant acceptability feedback. Exploratory efficacy analyses were conducted on measures of hope and perceived control (intervention mechanisms), occupational burnout and professional quality of life (primary outcomes), and empathy (secondary outcome) at baseline and 1, 4, and 12 weeks post-intervention.

Results

Twenty participants were consented and 18 participated, with 13 (72%) completing 12-week surveys. A total of 15 participants (83%) opted to receive daily push notifications, but only 3 (20%) engaged in daily booster questions. Positive feedback suggested acceptability. Exploratory findings suggested changes in emotional exhaustion (baseline to 1 week, P = .02; baseline to 4 weeks, P = .01), a sub-construct of burnout, and secondary traumatic stress (baseline to 4 weeks, P = .01; baseline to 12-week scores, P = .01), a sub-construct of professional quality of life.

Conclusions

The Hopetimize intervention was feasible for pediatric oncology health-care professionals, although daily application usage was limited. Exploratory findings suggested that a randomized controlled trial would be beneficial to delineate intervention effects on the proposed mechanisms and outcomes.

Introduction

Pediatric oncology—a discipline focusing on cancers that arise among infants, children, adolescents, and young adults—is predicated on the commitment and teamwork of an array of dedicated professionals.¹ Due to the demands of caring for this population, staff members bear high risks of sustaining emotional consequences including burnout and deterioration of work-related quality of life.² Burnout is a work-related syndrome composed of emotional exhaustion, depersonalization, and a diminished sense of personal accomplishment.³ The latter often reflects compromised professional self-efficacy. The limited available data on pediatric oncology workforce burnout suggest that approximately 60% of the professionals in this field suffer from this debilitating condition.^2,4 The related, broader phenomenon of psychological health at work recognizes the criticality of professional quality of life for those employed in the modern medical center.⁵

Hope theory is a cognitive construct proposed by Snyder nearly 4 decades ago.⁶ Snyder’s theory is predicated on an activist view of hopefulness that emphasized that 3 components must be deployed for hope to thrive: goals (aspirations that are not only meaningful but also plausible to achieve), pathways (strategies to reach those goals), and agency (drive and motivation to set out on the aforementioned pathways). Hope is theorized to promote motivation and positive emotions, enhance problem-solving, and lead to higher quality of life, while protecting from stressful life situations. Three decades of research have supported Snyder’s theory and demonstrated that enhancing hope can have broad positive effects, especially when goals are aligned with core values.^7-10

Research suggests that interventions designed to enhance hope may bring about measurable improvement in hope and related outcomes, including burnout.^11-15 Accordingly, our group adopted and applied the Hopetimize intervention, a program based on Hope theory.^16-18 This program has demonstrated feasibility and acceptability as a psychosocial intervention for oncology professionals to navigate work–life balance in the face of emotionally taxing roles.^17,18 A previous single-arm Hopetimize study of patients with breast cancer and oncologists demonstrated increased levels of hopefulness and decreased levels of burnout among the oncologist cohort from baseline to 3 months post-workshop.¹⁸ The present study aimed to expand the assessment of intervention feasibility to pediatric oncology professionals and to explore potential mechanisms and outcomes of the intervention through exploratory efficacy analyses. Figure 1 depicts intervention mechanisms, components, and outcomes.

Figure 1.

Intervention mechanisms, components, and outcomes, (see Figure 2, provided by Life's Door).

Methods

Recruitment

Hospital staff who interact with patients with pediatric cancer were eligible to participate in the intervention. Eligible participants were informed about the study through staff email notifications, announcements at meetings, and fliers that were placed in staff workspaces and distributed at work between May 2023 and October 2023. Potential participants were instructed to contact the study consenter (S.O.) individually by email or phone for discussion of the protocol and consenting. The study consenter (S.O.) was a graduate student at San Diego State University and conducted recruitment activities remotely to reduce the risk for conflicts of interest or coercion. The consenting was done with an online consent form in REDCap, a HIPAA-compliant database. Each participant received a gift card (up to $100) as compensation for their participation, with compensation prorated to the portion of the study completed. Given the pilot nature of the study, all participants received the intervention and there was no control group.

Intervention

The Hopetimize intervention consisted of a 2-hour workshop and a mobile application that was available to the participants during the workshop and for 12 weeks following the workshop.¹⁴ The workshop was conducted on a virtual conference platform (ZOOM). Participants were given a choice of 5 available time slots to attend the workshop.

The Hopetimize intervention was delivered by a mental health professional (D.C.) trained in the standardized workshop model, who received 30-hour standardized, manualized instruction accompanied by a certification process and led all 5 sessions to maintain fidelity.^17,18 A co-facilitator trained in the Hopetimize intervention provided additional support during sessions. To promote intervention fidelity, the program was implemented using the standardized 2-hour Hopetimize workshop format, consisting of a structured slide deck and scripted facilitation guidelines that define the sequence and core content of each session.

During the workshop, a didactic component summarizing the principles of Hope theory was followed by interactive exchanges to bolster the strength of participants in nominating goals and considering their obstacles through a tool called “hope mapping” (see Figure 2). Hope mapping, previously a paper-and-pencil exercise, was conducted for the first time via a smartphone application known as Hopetimize. Following hope mapping, participants participated in “mental rehearsal” (a guided-imagery exercise) in which participants envisioned successful overcoming of obstacles en route to the respective goals selected. An open forum was then provided for discussion among participants and provision of feedback in the workshop. Further information about the intervention is described in a previously published article on the feasibility and acceptability of the Hopetimize intervention for oncology professionals.¹⁷

Figure 2.

Hope map.

During the workshop, the participants were instructed in how to use the Hopetimize mobile application and were encouraged to track their progress on their application hope maps for 12 weeks. The application was programmed to provide encouragement and reinforcement through optional text, email, and push notifications. Participants could use the application at their own pace to add steps toward completing their goals and could indicate when these steps had been completed. For those who “opted in,” the application sent daily push notifications.

Measures

Baseline demographic (eg, age, gender, race, ethnicity, marital status, education) and employment information (eg, profession, years of employment) about the participants were collected. To understand the impact of the Hopetimize intervention, 2 exploratory mechanisms were measured (hope and perceived control/self-efficacy), as well as 2 exploratory primary outcomes (occupational burnout and professional quality of life). Furthermore, empathy was included as an exploratory secondary outcome, given its close relationship with burnout.¹⁷ The impact of the intervention on these mechanisms and outcomes was evaluated via surveys administered through REDCap at baseline (within 4 weeks before workshop attendance), 1 week post-workshop, 4 weeks post-workshop, and 12 weeks post-workshop.

All measure descriptions are included in Table S1. Hope was measured through the State Hope Scale’s Pathways Subscale, Agency Subscale, and Total Score, in which higher scores indicate greater hope.¹⁹ Perceived control (self-efficacy) was measured through the Perceived Control Scale’s Perceived Constraints and Perceived Mastery Subscales, in which higher scores indicate greater perceived control.^20,21 Occupational burnout was measured through the Maslach Burnout Inventory’s (MBI) Emotional Exhaustion, Depersonalization, and Personal Accomplishment Subscales, in which higher scores indicate a higher level of each construct.^22-24 Professional quality of life was measured through the Professional Quality of Life Scale’s (ProQOL) General Burnout, Compassion Satisfaction, and Compassion Fatigue/Secondary Trauma Subscales, in which higher scores indicate a higher level of each construct.^25,26 Because the MBI was more dedicated to occupational burnout explicitly, as well as its sub-constructs, it was used to assess the construct of burnout rather than the ProQOL. Empathy was measured through the Toronto Empathy Questionnaire, in which higher scores indicate greater empathy.^27,28

Statistical analyses

Repeated-measures analyses of variance were used to evaluate overall effects, including effect sizes, and differences between marginal mean scores between timepoints (baseline, 1 week, 4 weeks, 12 weeks). To expand upon these findings, Wilcoxon signed rank tests were conducted to compare baseline and post-intervention scores (baseline and 1 week, baseline and 4 weeks, baseline and 12 weeks).²⁹ This allowed for casewise, rather than listwise deletion of missing data, and accounted for non-normality.³⁰

Expected outcomes

We hypothesized that the intervention would have suitable feasibility, demonstrated through retention, application usage, survey completion, and informal patient feedback that would suggest that the intervention was positively received.³¹ Given the exploratory nature of efficacy analyses, we did not anticipate the ability to make robust statistical inferences on changes in exploratory constructs over time.

Results

Out of approximately 190 staff members, 20 (11%) consented to participate in the intervention over the 5-month recruitment period. A total of 18 health-care providers (90% of those consented) participated in the intervention, with 2 withdrawing due to logistical timing difficulties related to the workshop. See Table 1 for participant demographics, including professions.

Table 1.

Participant demographics.

Characteristic	N = 15 a  # (%), unless otherwise specified
Age: mean (SD)	40.7 (13.5)
Gender
Female	14 (93.3)
Male	1 (6.7)
Self-reported race/ethnicity
Caucasian	10 (66.7)
Black/African American	1 (6.7)
Asian	1 (6.7)
European	1 (6.7)
Latino/Hispanic	2 (13.3)
Education
Bachelor’s degree	10 (66.7)
Master’s degree	3 (20.0)
PhD or higher	2 (13.3)
Years at hospital (N = 12)	5.5 (5.4)
Employment status
Full-time	12 (80.0)
Part-time	3 (20.0)
Profession
Nurse	4 (22.2)
Research professionals	3 (16.7)
Child life specialists	2 (11.1)
Nurse practitioner	1 (5.6)
Front desk scheduler	1 (5.6)
Psychologist	1 (5.6)
Declined to provide	6 (33.3)
Relationship status
Married	13 (86.7)
Single	2 (13.3)
Living situation
Alone	1 (6.7)
With friends	1 (6.7)
With spouse	13 (86.7)
Religious status
Not religious	7 (46.7)
Slightly religious	3 (20.0)
Moderately religious	3 (20.0)
Very religious	2 (13.3)
Religious identification
Atheist	1 (6.7)
Catholicism/Christianity	8 (53.3)
Prefer not to say	6 (40.0)

^aThree participants did not report demographics.

All participants created hope maps during the workshop, with 16 creating hope maps in the application and 2 completing hope maps using paper and pencil. Examples of participant goals included reducing stress, improving fitness, spending more time with family, creating more balance in one’s life, deepening friendships, saving money, and improving sleep. Of the 18 participants who participated in the intervention, 15 (83%) opted into receiving push notifications. Three participants (20%) engaged in daily booster questions, and 15 participants (83%) outlined all 3 steps of their goal in the application. While participants had the option to document if steps were successfully completed, only 4 participants documented successful completion of step 1 (22%), 1 participant documented completion of step 2 (6%), and no participants documented successful completion of step 3.

All participants completed baseline surveys, 17 completed 1-week surveys (94%), 14 completed 4-week surveys (78%), and 13 completed 12-week surveys (72%). Representative remarks that were shared by participants following completion of the mapping exercise are included in Table S2. These remarks illustrate the core components of Hope theory. Health-care professionals described how identifying multiple pathways reduced cognitive rigidity and increased perceived attainability of goals, particularly in complex clinical contexts. Simultaneously, articulating small, actionable steps appeared to enhance agency.

Repeated measures analysis findings

While least significant mean differences between timepoints, including confidence intervals, are reported in Tables S3-S14, exploratory findings reporting focused on overall effects and Wilcoxon signed rank test findings due to the small, underpowered, 1-armed sample size, and non-normality of the data. Estimated marginal means at each timepoint for every measure are shown in Figures S1-S10. Figures 3-5 display estimated marginal means of measures with notable least significant differences (ie, with P-value of .05 or less). Large overall effect sizes are reported below, as well as findings with P-value of .05 or less, with the caveat that the study’s sample size and lack of control group precluded the reliable estimation of effect size magnitude and was underpowered to detect statistically significant effects. All within-subjects overall effect sizes are reported in Table 2.

Figure 3.

Perceived Control Scale Perceived Mastery Subscale Score Marginal Means.

Table 2.

Within-subjects overall effect sizes.

Scale	Effect size (partial Eta squared)	Interpretationa
State Hope Scale: Pathways Score	0.106	Medium
State Hope Scale: Agency Score	0.145	Large
State Hope Scale: Total Score	0.125	Medium
Perceived Control Scale: Perceived Mastery Score	0.098	Medium
Perceived Control Scale: Perceived Constraints Score	0.058	Small
Professional Quality of Life Scale: Compassion Satisfaction Score	0.070	Medium
Professional Quality of Life Scale: Burnout Score	0.065	Medium
Professional Quality of Life Scale: Traumatic Stress Score	0.381	Large
Maslach Burnout Inventory: Emotional Exhaustion Score	0.104	Medium
Maslach Burnout Inventory: Depersonalization Score	0.052	Small
Maslach Burnout Inventory: Personal Accomplishment Score	0.031	Small
Toronto Empathy Scale	0.051	Small

^aBased on the following interpretation guidelines: 0.01-0.059 = small effect, 0.06-0.139 = medium effect, 0.14 or higher = large effect.²⁹

Baseline descriptive scores

Descriptives of baseline, 1, 4, and 12 weeks findings are reported in Table 3. Based on scoring criteria (see Table S1) participants’ average emotional exhaustion and depersonalization were moderate on the MBI at baseline. Study participants demonstrated a mean score of 25.06 (SD = 11.05) at baseline, suggesting high levels of emotional exhaustion nearing clinically significant burnout (score of 27).^22-24 Sense of personal accomplishment was low but bordered on moderate.^22,23 Based on scoring criteria, participants scored in the average range for all subscales of the ProQOL at baseline.^25,26

Table 3.

Efficacy outcome descriptives.

	Baseline	1-week	4-week	12-week
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)
State Hope Scale: Pathways Score Range = 3-24	18.11 (2.85)	18.76 (2.71)	19.43 (2.31)	19.31 (2.36)
State Hope Scale: Agency Score Range = 3-24	17.56 (3.00)	16.53 (3.45)	17.21 (3.58)	18.85 (2.97)
State Hope Scale: Total Score Range = 6-48	35.67 (5.09)	35.29 (5.48)	36.64 (5.39)	38.15 (5.00)
Perceived Control Scale: Perceived Mastery Score Range = 1-7	5.92 (0.59)	5.92 (0.46)	5.95 (0.56)	6.13 (0.55)
Perceived Control Scale: Perceived Constraints Score Range = 1-7	2.77 (1.06)	2.62 (0.80)	3.09 (1.08)	3.01 (0.93)
Professional Quality of Life Scale:Compassion Satisfaction Score Range = 0-50	36.50 (4.46)	36.27 (5.24)	35.44 (5.68)	36.29 (5.21)
Professional Quality of Life Scale: Burnout Score Range = 0-50	23.50 (5.29)	23.13 (5.44)	23.25 (5.58)	22.79 (6.55)
Professional Quality of Life Scale: Secondary Traumatic Stress Score Range = 0-50	25.50 (5.46)	24.47 (5.41)	24.31 (6.13)	23.29 (5.92)
Maslach Burnout Inventory: Emotional Exhaustion Score Range = 0-54	25.06 (11.05)	21.57 (12.26)	23.13 (11.82)	22.50 (13.04)
Maslach Burnout Inventory: Depersonalization Score Range = 0-30	6.39 (4.57)	5.57 (4.48)	5.75 (4.33)	6.07 (6.01)
Maslach Burnout Inventory: Personal Accomplishment Score Range = 0-48	33.33 (8.33)	34.57 (8.03)	33.00 (7.31)	34.14 (7.55)
Toronto Empathy Scale Range = 0-64	49.83 (6.54)	50.27 (7.19)	50.71 (6.14)	50.36 (7.77)

Hope and perceived control (intervention mechanisms)

On the State Hope Scale, while there was a large within-subjects overall effect for the Agency subscale score (${\eta }_p^2$ = 0.145), there were no notable effects across or between timepoints for the overall score, Agency subscale score, or Pathways subscale score. On the Perceived Mastery subscale of the Perceived Control Scale (see Figure 3), while there was a large quadratic effect across timepoints (${\eta }_p^2$ = 0.341, P = .05), the within-subjects overall effect size was medium (${\eta }_p^2$ = 0.098) and there were no differences between timepoints. There were no notable effects across or between timepoints for the perceived constraints subscale.

Professional burnout and professional quality of life (primary outcomes)

On the Maslach Burnout Scale, while there was not an overall effect across timepoints for any of the subscales, Wilcoxin signed rank tests suggested differences on the Emotional Exhaustion subscale (see Figure 4) between baseline and 1-week scores (Z = −2.358, P = .02) and baseline and 4-week scores (Z = −2.521, P = .01). On the ProQOL, there were no notable effects across or between timepoints for the Compassion Satisfaction subscale and the Burnout subscale. There was a large linear effect (${\eta }_p^2$ = 0.555, P = .01; overall effect ${\eta }_p^2$ = 0.381) across timepoints for secondary traumatic stress (see Figure 5). Wilcoxon signed rank tests suggested differences between baseline and 4-week post-intervention scores (Z = −2.579, P = .01) and baseline and 12-week scores (Z = −2.805, P = .01).

Figure 4.

Maslach Burnout Inventory Emotional Exhaustion Subscale Score Marginal Means.

Figure 5.

Professional Quality of Life Scale Secondary Traumatic Stress Subscale Score Marginal Means.

Empathy (secondary outcome)

On the Toronto Empathy Scale, there were no notable effects across or between timepoints for the overall score or subscale scores.

Discussion

Limited attention has been devoted toward the well-being of staff members entrusted with the care of patients with pediatric cancer. Accordingly, our team committed to programmatic deployment of an intervention that could have benefit for this professional cohort. While enrollment rates were low over the course of the recruitment period, participation and survey completion rates were high. Given that the study team used passive recruitment strategies in a population that is historically over-burdened with workplace responsibilities, and in light of restrictions preventing healthcare professionals from attending the workshop during worktime hours, a low consent rate was expected.^3,4,32 Additionally, none of the participants self-reported that they were practicing medical doctors. The recruitment difficulties emphasize the importance of allotting dedicated worktime toward supporting employee wellbeing. In a larger scale randomized controlled trial (RCT), it would also be beneficial to recruit more broadly and use techniques that increase the likelihood of accruing professionals engaged in diverse clinical roles.

Notably, the high participation and survey completion rates in those who consented, as well as the positive feedback, were encouraging. Participant quotes post-intervention demonstrate intervention acceptability. Participant feedback suggests that structured hope-mapping processes may strengthen both cognitive flexibility and motivational engagement, while supporting resilience within demanding clinical environments.

Previous research has indicated the feasibility and acceptability of this intervention among oncology professionals.^17,18 The present study’s rates suggest feasibility of creating a hope map and outlining steps, though participants did not document in the application whether steps were successfully completed. While it appears that participants benefited from workshop and application engagement and successfully worked toward their goals, it would be helpful to obtain further stakeholder feedback on feasible application usage prior to initiating a larger-scale RCT of the intervention in a pediatric oncology professional population. Study feasibility findings ultimately demonstrate that further research of this intervention is feasible and warranted.

Exploratory findings

Hope and perceived control (intervention mechanisms)

Several exploratory findings were intriguing, but effect size magnitudes should be interpreted with caution and as preliminary in nature, due to the small, underpowered single-arm sample size. Total and subscale scores for hope descriptively changed in the anticipated direction with a potentially notable overall effect for the Agency subscale. In informal feedback collection post-workshop, all professionals indicated a desire to incorporate the hope-enhancement techniques that they acquired during the session into their clinical practice. It would be worthwhile to conduct an RCT study in a larger sample of pediatric oncology professionals and explore more definitively if hope is indeed a mechanism promoted by the intervention. Further, exploratory analyses suggested a potential quadratic effect of the intervention on perceived mastery, a subscale of perceived control. This demonstrates the potential utility of including perceived control and its sub-constructs as outcomes in a future RCT. Per the literature, perceived mastery and self-efficacy have significant overlap as constructs, and professional self-efficacy has been associated with hope.³³

Professional burnout and professional quality of life (primary outcomes) and empathy (secondary outcome)

In our limited study sample, findings were similar to previous literature identifying emotional exhaustion as a significant impediment for oncology professionals.³² Although our study sample was underpowered, score changes from baseline to later timepoints suggest the potential benefit of future studies targeting emotional exhaustion in oncology professionals. Notably, the burnout subscale of the ProQOL did not change across timepoints. A larger RCT could further investigate whether experiences of occupational emotional exhaustion vs general burnout differ.

Furthermore, secondary traumatic stress, or being affected by patient trauma, decreased over time, with a potential linear effect size. Research has indicated the presence of secondary traumatic stress in oncology and hospice nurses, as well as limited studies targeting this outcome.³⁴ While findings were not notable for empathy, a future RCT study could be a promising option to address the understudied burden of secondary traumatic stress.

Limitations

The study had notable limitations. Recruitment challenges, largely attributed to the absence of protected work time for the participants, led to limited enrollment. Future studies should consider how to deal with the inherent irony of increasing the accessibility of the intervention for providers experiencing high levels of burnout and limited time and bandwidth to enroll in psychosocial interventions.^3,4,32 Furthermore, limited data were collected on application usage acceptability or number of times that participants used the application, warranting continuing conversations with stakeholders as the application is amended to provide participants with a flexible, accessible tool to combat burnout. Given the unique burdens experienced by pediatric oncology professionals, the generalizability of study findings may be limited to this professional group, but the diversity of professions in the study population suggests that the intervention may be more widely beneficial, potentially warranting exploration of Hopetimize’s feasibility in a broader range of medical professions.

Furthermore, it must also be acknowledged that the overwhelming majority of participants in the current study were women. Multiple previous studies on burnout-related interventions have also primarily pertained to women.^35-40 While the predominance of women in the current study aligns with the work of previous investigators, the generalizability of our findings is limited by gender. Finally, the small sample size (N = 18) was especially sensitive to missing data effects at the postintervention assessments and underpowered for imputation strategies. The casewise deletion of data further increased missingness and diminished power.

Conclusion

Overall, the Hopetimize study demonstrated feasibility in a small sample of pediatric oncology professionals, with qualitative and exploratory quantitative data suggesting intervention benefits. Furthermore, our intervention, as well as other hope-related interventions, have demonstrated feasibility and acceptability in oncology providers and patients.^7-18 Therefore, there is potential for establishing future preliminary efficacy in a larger-scale RCT study in this population, pending adjustments to promote application usage. Currently, the study team is adapting the application for improved self-directed use without a workshop, thereby eliminating scheduling burden. Our team aims to continue to increase the accessibility of the intervention, to prepare for an RCT, and to explore and promote burnout-reduction measures for medical professionals in demanding roles.

Author contributions

Chelsea S. Rapoport (Formal analysis, Validation, Visualization, Writing—original draft, Writing—review & editing), Benjamin W. Corn (Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Software, Supervision, Writing—original draft, Writing—review & editing), Phyllis Dvora Corn (Conceptualization, Methodology, Project administration, Software, Writing—original draft, Writing—review & editing), Alyssa K. Choi (Writing—original draft, Writing—review & editing), Shea O’Donnell (Data curation, Project administration, Writing—original draft, Writing—review & editing), Michal Manasseh (Conceptualization, Methodology, Project administration, Software, Writing—original draft, Writing—review & editing), Vanessa L. Malcarne (Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing—original draft, Writing—review & editing), and Dennis J. Kuo (Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing—original draft, Writing—review & editing)

Supplementary material

Supplementary material is available at JNCI Cancer Spectrum online.

Funding

This work was supported by the T. Denny Sanford Institute for Empathy and Compassion at UCSD, Center for Empathy and Technology Seed Grant Program. The funder had no role in the design of the study; the collection, analysis, or interpretation of the data; or the writing of the article and decision to submit it for publication.

Conflicts of interest

The authors report there are no competing interests to declare.

Data availability

The data underlying this article cannot be shared publicly due to the privacy of individuals that participated in the study. The data will be shared on reasonable request to the corresponding author.