Abstract
As demand for palliative care (PC) services rise, there are insufficient numbers of PC specialists to provide PC for the US population. “Primary palliative care” refers to PC services that are administered by non-specialist PC providers. Educating trainees in graduate medical education (GME) programs is 1 strategy for expanding primary palliative care, though questions remain regarding the impact of PC education for GME trainees and where additional education is needed. This study is a multicenter, cross-sectional, web-based survey study of GME trainees assessing the needs for and impacts of primary palliative care education. The survey assessed the implementation of and participants’ confidence with fundamental PC skills. The survey also asked about prior exposure to PC education and for participants’ beliefs regarding areas that would be particularly helpful for future education. 170 residents and fellows from diverse training backgrounds participated in the survey out of 851 potential participants (response rate 19.98%). Exposure to PC education was associated with higher confidence and increased frequency of implementation of fundamental PC skills. Of the forms of education that were assessed, clinical/experiential education was associated most often with higher confidence and higher frequency of use of PC skills. Discussing goals of care, pain management for seriously ill patients, and communicating difficult information were those skills most frequently identified as important for additional training. This study demonstrates that by improving existing PC education or increasing access to PC education for GME trainees, it may be possible to improve primary palliative care.
Keywords
Introduction
Palliative care (PC) teams provide essential supportive services for patients who are diagnosed with serious illnesses. Though specialist PC providers are increasing in numbers, the demand for PC services far outstrips this limited supply of specialists.1,2 This gap between available specialist providers and patients’ needs for PC will only continue to grow as time passes,3,4 highlighting the importance of expanding access to PC services for patients. Increasing the supply of specialist PC providers alone is unlikely to keep up with the growing demand for services, 3 and many hospitals currently do not meet national staffing recommendations for PC teams or rely on unfunded positions. 5 The development of primary palliative care (PPC) represents an alternative strategy to address this shortfall. 6 In PPC, fundamental PC practices are implemented broadly by a diverse range of providers in a variety of settings, with the goal of increasing access to high quality PC for patients. This is in contrast to specialist PC, which is provided by those clinicians who have undergone particular training to specialize in palliative medicine. Given that the population of specialist PC providers alone is unlikely to successfully deliver all of the palliative care that is needed, the development of PPC education has been recognized as critical for increasing access to palliative care for over 2 decades.7,8 Particularly in the last 10 years, there has been increased interest in studying the role of PPC education in graduate medical education (GME) trainees.9,10 Multiple primary medical specialties have put forth core competencies for their trainees which are directly related to key palliative care skills.11,12 GME programs and trainees represent particularly impactful targets for PPC education programs for a variety of reasons, including: the opportunity for trainees to develop PPC skills early on in their career, a setting which encourages acquiring and refining new skills, and ample opportunity and time allotted for structured teaching opportunities. 10
Several studies have sought to describe the landscape and effectiveness of PPC education for GME trainees. Most frequently, studies describe individual educational interventions implemented within single hospital systems.8,13-19 Primary palliative care educational interventions have been developed and evaluated for residents in family medicine,14,16,17 internal medicine,13,16 emergency medicine, 15 and radiation oncology 20 among others. Additionally, multiple studies in the last several years have explored the impact of PPC education in fellows including those from cardiology,21,22 hematology/oncology,19,23-25 nephrology, 26 and pulmonary/critical care; 27 though many training programs remain underrepresented in the literature. 10 Program directors from a diverse group of fields also describe that PPC skills are important for their trainees,28-31 and in general program directors recognize that current PPC curricula need ongoing improvement. 32 Oncology trainees recognize the importance of developing PPC skills, as Wong et al. describe that two-thirds of those surveyed believe that a mandatory palliative care rotation during training is important, 33 though oncology trainees are also likely to report the quality of PC training in fellowship is poor relative to that of oncology training. 23 The variety of educational interventions described in the literature reflects both the creativity of individual programs in providing education for trainees 34 and the absence of standardized methods of providing such education, assessing the impact of education, 35 or following up longitudinally after education is completed.36,37 Didactic training alone has had mixed effectiveness, with more consistently positive results noted in studies with a clinical/experiential or simulation component to the education provided.18,19
Despite being frequently advocated, PPC education is provided inconsistently, the modalities which are used to provide education are highly variable, and the impacts of incorporating formal PPC education (that is, PPC education provided in a structured manner using didactics/clinical rotations/electronic modules/simulations or role-playing exercises) are incompletely understood. Additionally, there has been limited exploration of which PC skills graduate medical trainees consider most valuable to their practices, and for which skills they prefer more education. The primary aims of this study are to determine what the impacts of and needs for PPC education are among a diverse group of graduate medical trainees. The secondary aims of the study are to understand what forms of PC education these trainees have been exposed to, if any particular forms of education are more impactful than others, and what PPC skills represent particularly salient targets for additional education from the perspective of trainees.
Methods
A survey was developed by the research team to assess the need for primary PC education across 2 large health systems. The survey was constructed based on a review of the available literature and with input from content experts in palliative care and survey development.
Qualtrics was used to both develop and administer the web-based electronic survey (Qualtrics, Provo, UT). The populations targeted by the study were selected based on a review of the literature and input from PC experts on the research team to identify those specialties which might obtain the greatest benefit from additional PC training.10,11 Several PPC review articles were analyzed to determine content areas frequently covered in PPC education programs.10,11,36 The most common PPC skills targeted for education in the literature are patient/family communication skills in the context of serious illness (including goals-of-care conversations), 36 symptom management,10,11 end-of-life care,10,11,36 transitions of care and community resources. 10 Following this review, the research team identified 12 fundamental palliative care skills that were deemed necessary to provide effective primary PC based both on expert consensus and prior studies. These skills are: managing pain for seriously ill patients, managing dyspnea for seriously ill patients, managing symptoms other than pain/dyspnea for seriously ill patients, identifying surrogate decision makers, discussing goals of care or advanced care planning, communicating difficult information, discussing code status, leading family meetings, assessing prognosis, discussing prognosis, determining eligibility for hospice, and discussing end of life care options including hospice. The survey collected demographic information from participants (age, gender, racial/ethnic background, year of residency/fellowship training, residency program type, and fellowship type where relevant). Additionally, the survey asked about exposure to formal PC education and asked participants to rate their confidence with the 12 fundamental PPC skills, along with how frequently these 12 skills were implemented in their practices.
This survey study took place within 2 large hospital systems in southeastern Pennsylvania: Jefferson Health and Main Line Health. The survey was distributed via email to a diverse group of resident and fellow graduate medical trainees in each institution, and participants were recruited via the aid of residency/fellowship program directors (PDs). A description of the survey was provided to 28 PDs across these institutions, with requests that they assist with distributing the electronic survey among their trainee populations. Trainee email addresses were collected from PDs when possible, though some programs instead elected to distribute the survey internally. Both residency programs (internal medicine, family medicine, emergency medicine, physical medicine and rehabilitation, neurology, psychiatry, general surgery, otolaryngology, and anesthesia) and fellowship programs (cardiology, pulmonary/critical care, hematology/oncology, nephrology, and gastroenterology/hepatology) were invited to participate in the study. The surveys were administered from April 2021 to June 2021. Initial survey emails were sent out with 2 reminders over the subsequent 3 weeks. The survey and study procedures were approved by the Thomas Jefferson University Institutional Review Board.
The data was exported from Qualtrics directly into IBM SPSS Statistics which was used for data analysis. SPSS was used to generate descriptive statistics, and chi square statistics were used to identify significant associations between participant characteristics and responses to survey questions. Given the potential for multiple comparisons in the process of statistical analysis, a partial Bonferroni correction was applied to the analysis and the threshold of statistical significance was set at a P-value of ≤.01. Data tables and figures were generated in Microsoft Excel.
Results
Characteristics of Survey Participants.
Palliative Care Education/Experience.
* percentages add to greater than 100% as participants were able to select multiple options in the survey
** percentages here are calculated with a denominator of 132
Chi-square testing was implemented to determine if there were any significant associations between demographic variables and whether or not participants had been exposed to formal PC education. Residency training background (χ2 [10, N = 164] = 34.63, P <.01) was found to be significantly associated with differences in exposure to formal PC education. Age (χ2 [2, N = 164] = .44, P = .80), gender (χ2 [2, N = 161] = 4.22, P = .12), race/ethnic background (χ2 [5, N = 164] = 6.27, P = .28), and PGY year (χ2 [7, N = 164] = 15.03, P = .04) were not found to be associated with significant differences in exposure to formal PC education.
Most participants reported high confidence with incorporating fundamental palliative care skills, as well as having a high frequency of implementing these fundamental skills (Figure 1). Those participants who had formal PC education reported significantly higher confidence with using 5/12 (41.7%) of surveyed PC skills, compared with those who did not receive formal PC education (Figure 2). The skills for which significant differences were documented were: discussing goals of care or advanced care planning (χ2 [4, N = 147] = 19.76, P <.01), discussing end-of-life care options including hospice (χ2 [4, N = 149] = 22.45, P <.01), leading family meetings (χ2 [4, N = 148] = 13.05, P = .01), managing dyspnea for seriously ill patients (χ2 [4, N = 150] = 14.31, P <.01), and managing symptoms other than pain/dyspnea for seriously ill patients (χ2 [4, N = 149] = 14.65, P <.01) (Figure 2). Similarly, those who had formal PC education reported that they implemented 5/12 (41.7%) of fundamental PC skills more frequently (Figure 3). The skills that were more frequently implemented by those trainees who reported receiving formal PC education were: communicating difficult information (χ2 [4, N = 149] = 13.59, P <.01), discussing goals of care or advanced care planning (χ2 [4, N = 149] = 17.07, P <.01), assessing prognosis (χ2 [4, N = 149] = 15.02, P <.01), discussing end-of-life care options including hospice (χ2 [4, N = 149] = 15.71, P <.01), and determining eligibility for hospice (χ2 [4, N = 149] = 15.37, P <.01) (Figure 3). Caption: Confidence with fundamental palliative care skills (A), and frequency that fundamental palliative care skills are used (B). Overall, participants reported in aggregate that they were fairly/very confident with incorporating the assessed fundamental palliative care skills, and also reported frequent use of many fundamental palliative care skills. Caption: Confidence of graduate medical trainees with 12 fundamental palliative care skills. Pairs of bars clustered together represent those trainees who had (above) and did not have (below) formal palliative care education. Those pairs with an asterisk adjacent are significantly different at a significance level of p ≤ .01. Caption: Frequency of implementation of 12 fundamental palliative care skills by of graduate medical trainees. Pairs of bars clustered together represent those trainees who had (above) and did not have (below) formal palliative care education. Those pairs with an asterisk adjacent are significantly different at a significance level of p ≤ .01.
Experiential/Clinical Rotations and Confidence with, Frequency of Implementation of, and Helpfulness in Learning Fundamental PC Skills.
*Significant at the level of P ≤ .01
Comparison of those participants who had formal palliative care education in the form of experiential/clinical rotations against those who had not had this form of education (but who still had another form of formal palliative care education). Chi-square statistics are reported with accompanying significance (p) values. With the exception of “assessing prognosis,” trainees who had experiential palliative care education rated their education as more helpful in increasing confidence in fundamental palliative care skills when compared to those who had not had this form of education.
Participants reported that discussing goals of care or advanced care planning, managing pain for seriously ill patients, and communicating difficult information were the fundamental PC skills that would benefit most frequently (in descending order) from additional training (Figure 4). There were no significant differences between those who had received palliative care education vs those who had not regarding which skills would be most important for additional training. Caption: Frequency of ranking each fundamental palliative care skill as first, second, or third most important for additional training.
Discussion
This study demonstrates that PPC education for GME trainees is associated with increases in self-reported confidence with and frequency of implementation of fundamental PC skills. Furthermore, this analysis demonstrates that there are certain forms of PC education which may be more helpful in increasing confidence/frequency of implementation of these skills, with a PC rotation/clinical experience highlighted in particular as being helpful in increasing confidence in nearly all of the assessed fundamental PC skills. These results support the utility and effectiveness of PPC educational interventions for graduate medical trainees, and provide evidence that implementing PPC education is an impactful strategy for improving the ability of trainees to provide PPC.
Though didactic education was the most frequently reported type of formal PC education, the limits of didactic education alone have been noted previously, 13 a view which was also described in a survey of program directors. 28 Though this survey data was not able to analyze differences between groups of trainees who had received didactics and those who had not, the data was used to describe differences between those who had other types of formal education. The finding that the experiential rotation could be shown to be more effective in increasing confidence is an important finding and can serve as a justification for the expansion of this particular form of PPC education to many different GME trainees.
Tactful and effective communication skills in the setting of serious illness are a hallmark of high-quality PC, and the literature on PPC education also frequently highlights these skills as important for training.10,11,36 Notably, this study suggests that PPC education is associated with increased confidence and frequency of use of several fundamental communication PC skills including the discussion of end-of-life options and goals of care. Additionally, among those who had formal PC education, those who had experiential education demonstrated increased confidence in discussing goals of care or advanced care planning, and also reported using this skill more frequently when compared to those who had not had this type of education. Though skills in symptom management (such as management of dyspnea) and care coordination (such as determining eligibility for hospice) were also associated with higher confidence/frequency of use for those with palliative care education, communication skills were most consistently reported to be used more frequently and with more confidence by those trainees with PC education.
In addition, this study describes particular skills identified by graduate medical trainees that are important for additional training. While there are varied strategies for designing the most effective PPC curricula, acknowledging and engaging the trainees themselves as stakeholders in this effort may increase the uptake and success of planned educational efforts in the future. Interestingly, though there were not significant differences in the rankings of fundamental PC skills between those who had formal education and those who had not, 2 of the skills most frequently identified as important for additional training (discussing goals of care or advanced care planning, and communicating difficult information) were also skills that were used more frequently by those who had formal PC education. This suggests that trainees identify these communication skills as important to their practices, and also continue to recognize the need for additional communication training even with exposure to formal PC education. Pain management for seriously ill patients, a frequently employed skill for both specialist palliative and primary palliative care providers, was also highly ranked as a skill that would be important for additional training. Were the skills of a diverse group of graduate medical trainees to improve in these 3 areas, the impact on the quality of PPC provided by the entire healthcare system would likely improve as well.
Residency training background was associated with significant differences in self-reported exposure to formal PC education. This may reflect differential institutional and programmatic priorities and policies for providing formal PC education. For instance at 1 participating institution (Thomas Jefferson University Hospital), PC clinical rotations are frequently incorporated into several GME training programs (including family medicine, internal medicine, and pulmonary/critical care) whereas other GME training programs may rely on purely elective rotations or trainee interest.
There are several limitations to this data. The study took place in a limited number of health systems in 1 state in the US, the participants were largely homogenous in racial/ethnic background, and the trainees were mostly from internal and family medicine backgrounds; all of these affect the generalizability of these results. Though response rates for similar studies reviewed are variable (14%-63%),22-24,26 the response rate in this study of nearly 20% is lower than many similar studies; this also effects the generalizability of these results. There was a significant association with residency training background and exposure to formal PC education, which may indicate that residency training background is a confounding variable when analyzing this data in aggregate. Using a survey as the primary data collecting tool necessitates reliance on self-reporting of measures such as confidence and frequency of use of fundamental PC skills, which is an indirect measure of actual practices. Nearly all participants who reported exposure to palliative care education also reported receiving didactic education, limiting the ability of the study to provide meaningful conclusions about the most accessible form of PC training available to graduate medical trainees. Additionally, with a cross-sectional study design, conclusions can only be drawn on associations rather than causal relationships.
Despite these limitations, these data support the conclusion that PPC education is associated with increased confidence in providing palliative care and with increased frequency of implementing fundamental palliative care skills by GME trainees. Future directions for this field of study should focus on implementation of curricula which focus on training valuable fundamental PC skills to a wide variety of trainees and implementing validated measurements for assessing the impacts of individual PPC educational interventions. Additionally, future studies should implement more direct measurements of the quality of primary palliative care provided by trainees before and after PC education, and should also incorporate the perspective of patients and families on PPC provided by trainees. The COVID-19 pandemic has also demonstrated the importance of flexible education models which can readily accommodate learning at a distance should social distancing precautions preclude in-person learning; the expansion of asynchronous or self-directed resources like electronic modules could be a viable method for providing such education.
In conclusion, GME trainees report that PPC education is helpful, increases confidence, and increases frequency of PC skill use. PPC education efforts should continue and should expand to accommodate the greatest number and variety of GME trainees as is feasible. Educators should engage their trainees as stakeholders in determining the most important targets for PPC education, though this study may provide a starting place for those efforts.
Footnotes
Acknowledgments
The research team would like to thank the palliative care team and Department of Family and Community Medicine at Thomas Jefferson University Hospital for their support and engagement with this work, particularly: John Liantonio MD MBA HDMC, Kathleen Mechler MD, Margaret Kreher MD, Randa Sifri MD, Amy Cunningham PhD MPH, and Howard Rabinowitz MD.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by VITAS Healthcare.
