Use of an Electronic Journal Club to Increase Access to and Acceptance of Palliative Care Literature across General Pediatricians and Pediatric Subspecialties

Introduction

Primary pediatric palliative care is recognized as the basic skill set and universal competencies required of all pediatricians. ¹ Investment in primary palliative care education ideally translates into improved knowledge of palliative principles and ultimately in enactment of palliative principles at the bedside. Many palliative care-relevant articles are contained in palliative care journals and it is unclear whether general pediatricians or subspecialty pediatricians access this literature on a regular or routine basis. To foster a shared knowledge base, our study team (consisting of a pediatric palliative care physician, nurse and social worker, and a research librarian) sought to find and strategically distribute articles targeted to quality of life, symptom management, communication, goals of care, care coordination, interdisciplinary collaboration, team dynamics, care for caregiver, psychosocial domains, and ethics for healthcare providers in our free-standing children's hospital in the Midwest. The goal of this Palliative Care Electronic Journal Club (Pal Care Club) was to increase familiarity with pediatric palliative care literature base across our free-standing children's hospital. The Pal Care Club did not include the face-to-face interaction associated with a traditional journal club, but instead consisted of electronically sharing a reading list summary of palliative care articles and full-text article content. A single cohort, predesign–post-design was utilized to explore the impact of a monthly palliative care electronic journal club.

Methods

All members of seven pediatric academic divisions (neonatology, intensive care, cardiology and cardiac surgery, hematology/oncology, hospitalist, and pulmonary) and all pediatric residents at our free-standing children's hospital were invited to particulate in the Pal Care Club. To personalize the Pal Care Club per subspecialty group, a preintervention survey was used to assess baseline beliefs and behaviors. The Gehlbach Survey Check-List ² and the Tailored Method of Survey Design were used for questionnaire development. ³ Questions were reviewed by two medical librarians and an educational liaison as content experts for clarity, relevance, and topic coverage.^4,5 The survey was independently piloted, revised, and repiloted by an interdisciplinary team from the target population (a behavioral health specialist, three pediatric generalists, and three pediatric subspecialists). Review by the behavioral health specialist resulted in the addition of attitude- and feelings-based questions that were missing from the original survey draft. Cognitive interviewing with three hospitalists before survey distribution occurred to ensure that respondents understand the questions in the intended way.

The survey format and content qualified as exempt from full Institutional Review Board review. The survey instrument (available as Supplementary Appendix 1; Supplementary Data are available online at www.liebertpub.com/jpm) consisted of 13 questions administered on SurveyMonkey©. Questions were categorized according to demographics, comfort level (comfort was defined in the survey as “belief and confidence in personal ability”) relevant to palliative care, interest in palliative care topics, baseline behaviors in regard to reading and discussing palliative topics, perceived barriers to accessing palliative knowledge, and emotive response to proposed journal club intervention. A link to the SurveyMonkey© questionnaire was sent in an introductory e-mail inviting survey completion, with three reminder e-mails sent in two-week intervals.

The information gathered from this pretest guided article selection and distribution for the Pal Care Club. Journal articles were gathered by searching four databases each month (PubMed, EBSCO, SCOPUS, and PsychInfo) for articles specific to MeSH search terms identified by topical interest of the subspecialty group. Articles were additionally hand-selected from the PedPalASCNet (Accessible, Sustainable, Collaborative Network in Pediatric Palliative Care) monthly citation list. ⁶ Inclusion of topics relevant to shared active patient cases between the subspecialty team and palliative care physician was further selected. The Pal Care Club intervention included distribution of 5–10 full-text articles monthly with narrative summary. There was no in-person interaction. The Pal Care Club consisted of sharing information electronically. Each e-mail included the article name and author; a link to the full-text abstract; the full-text article as an attachment; and a link to a shared common-access intranet folder containing all cumulative articles organized by medical subspecialty and topic. By the end of Year 1, each subspecialty group had cumulatively received between 104 and 120 shared articles.

At study conclusion (12 months after initial survey distribution), the survey was readministered as a post-test. For measures taken only at baseline, counts and percentages summarize categorical responses, while means/standard deviations or medians/interquartile ranges summarize quantitative responses. Precomparisons to post-comparisons for Likert responses were conducted using Wilcoxon's signed-rank test. Comparisons of Likert responses across roles were accomplished through the use of the Freeman-Halton extension of Fisher's exact test after collapsing responses of 1 and 2 into one group, and 4 and 5 into another group. Ambivalent responses (3) were ignored. A two-sided p-value of <0.05 was used to denote statistical significance. All analyses were completed using SAS software.

Results

A total of 71 pretest and post-test surveys were complete. Initial survey response rate was 78/113 = 69% with presurvey and post-survey completion rate 71/78 = 91% (meaning that 71 providers completed both the before-intervention and after-intervention survey and 7 providers only completed the before-intervention survey). The presurveys and post-surveys were matched by respondent code. Demographics and baseline interest are available as Table 1.

Initial interest

Before initiation of Pal Care Club, survey respondents identified barriers to reading palliative care-relevant articles as time n = 49 (69%), access n = 42 (59%), interest n = 3 (4.2%), and perceived relevance of topic n = 1 (0.06%). Respondents were allowed to choose more than one barrier. Preferred frequency of Pal Care Club was monthly n = 42 (59.2%), quarterly n = 27 (38%), and never n = 2 (2.8%). Half (50.7%) of respondents preferred to also have an accompanying in-person format journal club with n = 28 (39.4%), indifferent to in-person discussion inclusion, n = 7 (9.9%), preferring only electronic format. Summary of emotive responses to the Pal Care Club idea before intervention onset is depicted in Figure 1 with size of the word congruent with frequency of word selection. Quality of life, pain and complex symptom management, end-of-life care needs/bereavement care, communication about goals to include code status and advance care planning, and medical ethics were noted to be article topics of highest interest in the preintervention survey. Rank scores of article topics in order of personal/professional interest are available as Table 2.

OPEN IN VIEWER

FIG. 1.

Words survey respondents used to describe journal club offering.

OPEN IN VIEWER

Table 2.

Summary Ranking of Personal/Professional Interest in Topics

Table values are count (%) a
Article topics	Interest level			Response
	High (1–3)	Moderate (4–7)	Low (8–10)	N out of 65
Quality of life for patients	41 (64.1)	22 (34.4)	1 (1.5)	64
Pain or other complex symptom management unique to my patient population	29 (45.3)	24 (37.5)	11 (17.2)	64
End-of-life/comfort care or bereavement care	25 (39.7)	28 (44.4)	10 (15.9)	63
Communication about code status or advanced directives	25 (39.7)	23 (36.5)	15 (23.8)	63
Medical ethics	20 (30.8)	28 (43.0)	17 (26.2)	65
Longitudinal goals of care	18 (28.6)	28 (44.4)	17 (27.0)	63
Care for the caregiver/staff resiliency	13 (20.6)	28 (44.4)	22 (35.0)	63
Integrative therapies (nonpharmaceutical interventions)	12 (19.0)	33 (52.4)	18 (28.6)	63
Cultural, spiritual/existential, or psychosocial topics relevant to my patient population	7 (10.9)	25 (39.1)	32 (50.0)	64
Art/philosophy relevant to my field (the “art” of the “science”)	5 (7.9)	16 (25.4)	42 (66.7)	63

a

Six out of 71 did not select any article topic.

Discussion

The Pal Care Club pilot intervention served as a feasible, acceptable, and implementable educational innovation to foster shared learning between palliative care providers and pediatric colleagues. The preintervention survey identified time and access as the primary barriers to engagement with palliative care literature. Overcoming these two primary barriers by proactively partnering to provide a narrative summary overview and ready access to full-text articles resulted in quantifiable increase in the number of articles read (by 2.56 articles per month; p < 0.0001) and content discussion by colleagues. The intervention was specifically designed to test acceptance and impact of an electronic journal club and thus did not include an in-person journal club format. Half of preintervention survey respondents would have wanted an in-person journal discussion partnered. Review of the first author's calendar year revealed invitation to provide palliative care education at four times more journal clubs and subspecialty staff meeting than in the previous year—most likely motivated by shared literature fostering dialogue across teams. This increase in invitation to in-person journal club presentations may reveal improving integration and collaboration as an unintended benefit.

Topical interest for palliative care principles includes an emphasis on quality of life and symptom management. Topical prioritization in the presurvey portrayed a clear leaning toward the practical and tangible bedside applications more than the philosophical aspects of the palliative field.

The finding that residents and fellows were more likely to have a prescore of “very uncomfortable” or “uncomfortable” regarding palliative care integration or introduction than all other professional roles combined compels development of a trainee-specific palliative care curriculum and mentorship. This finding reminds us that the training years are formative times for palliative care educational integration for enhanced comfort and competency in primary palliative care skill set. Overall, professional role correlated with palliative integration or introduction comfort level with physicians being more comfortable than advanced practice clinicians. This finding warrants attentiveness to empowerment and targeted education of nurse practitioners and physician assistants in palliative care educational interventions. Interestingly, years in practice did not correlate with preintervention-perceived comfort level.

Palliative care reading and discussion (knowledge) did not result in increased personal comfort with palliative care integration or introduction (self-efficacy). This finding is consistent with the recognition that didactic medical education rarely leads to skill-based improvements. This finding speaks to the reality that palliative care is ultimately a mentored, relational shared practice. While knowledge is foundational groundwork, book insight does not replace mentored and partnered practice at a bedside to foster comfort, competence, and confidence in a provider. The literature-based program introduced by this study may then be further augmented by face-to-face workshop teaching and skills-based practice (role modeling, rotations with interdisciplinary team members, and simulation events) to translate knowledge into practice.

Nonresponse bias and recall bias are study limitations due to nature of survey methodology, ⁷ although these were mitigated by high response rate and study duration limited to one-year pilot. A strength of this study was the high completion rate of post-intervention surveys. The use of validated survey design instruments and interdisciplinary process of piloting, revising, pretesting, and cognitively interviewing before survey distribution likely maximized content outcomes ⁸ and minimized measurement error. ⁹

This innovative pilot initiative at a single pediatric institution revealed that an electronic journal club is an acceptable way to increase the number of palliative care articles read and discussed across the institution. Analyzing for clinical significance within our institution would be a meaningful future inquiry. Next step study would be to explore how an electronic distribution of palliative articles and palliative knowledge base could be further grown across institutions. This study lays the groundwork for larger scale approaches toward continuation of interventions to now foster increased self-perceived comfort and competency in applying palliative care principles within pediatric settings with additional skill-based learning modalities.