Serious Illness Communication Training in Family Medicine Residencies: A National Study of Program Director Perspectives

Given their close, long-term relationships with patients and families, family physicians are uniquely well-positioned to provide primary palliative care. This includes essential pain and symptom management as well as initiating serious illness conversations with patients and their families. The benefits of such care are well-established, including improved quality of life for patients and families,^1,2 increased satisfaction with care,³ and decreased care costs.³ Since palliative care became formally recognized as a medical subspecialty in 2008, the demand for specialty palliative care has steadily increased. This increase has been attributed partly to the aging population and better understanding and acceptance of palliative care.⁴ However, the number of palliative care specialists is insufficient to respond to this increased demand alone.⁵ In turn, family physicians have an essential role to play in meeting this demand and supporting patients and families facing serious illness and the end of life.

The term “serious illness” refers to conditions with a high risk of mortality and negative impact on daily function or quality of life, or significant strain on caregivers.⁶ Serious illness conversations are discussions among patients with serious illness, their families, and clinicians that elicit prognostic understanding, emotions, and care preferences and that ideally occur longitudinally over the course of a patient’s illness trajectory.⁷ These conversations are central to primary and specialty palliative care practice. The value of these conversations has been well-documented.^1,8,9 Studies have shown that conversations about end-of-life care between patients and clinicians lead to care more aligned with patient preferences.⁸ Meanwhile, patients who do not have such conversations are more likely to receive aggressive medical care at the end of life, which has been associated with decreased quality of life for both patients and their caregivers.¹ Primary care physicians often struggle with these conversations, as reflected in a 2016 review of serious illness communication in primary care. This review highlighted deficits in clinician knowledge and skills regarding serious illness communication and uncertainty about disease prognosis and appropriate conversation timing.⁹ Also in 2016, a survey study of resident training in end-of-life communication skills at one institution found that despite these conversations being common in practice, they were often unsupervised, and most residents had inadequate preparation for them in residency and medical school.¹⁰

While studies have explored palliative care training for medical students¹¹ and residents and fellows in other specialties, including internal medicine,¹² pulmonology,¹³ emergency medicine,¹⁴ and surgery,¹⁵ less is known about how family medicine residents are taught primary palliative care skills, including serious illness communication, during their training. This study aimed to explore the landscape of serious illness communication training in family medicine residencies across the United States, including the frequency, methods, barriers to training, and trainers’ expertise, as well as whether the presence of family medicine faculty with hospice and palliative care (HAPC) expertise improves the quality of serious illness communication training. To this end, we explored associations among the presence of faculty with HAPC expertise, program director ranking of resident preparedness to lead conversations, and training characteristics.

For this study, we surveyed family medicine program directors about serious illness communication training in residency. Survey questions were included in a larger omnibus survey of Accreditation Council for Graduate Medical Education (ACGME)-accredited US family medicine program directors conducted by the Council of Academic Family Medicine Educational Research Alliance (CERA). Details of the CERA program director survey methodology have been described elsewhere.¹⁶ The survey was distributed by email using the online program SurveyMonkey (SurveyMonkey Inc). Data was collected from September 26 to October 30, 2023. The authors used the change theory in education to develop the specific serious illness communication survey questions,¹⁷ which were pretested with local family medicine educators who were not part of the sampling population, but who had educational theory expertise and a clinical perspective. Pretesting allowed face validity, interpretation of the language, and clarity of the questions’ purpose to be refined. The question-and-answer formats were subsequently edited to improve readability, flow, and timing. The American Academy of Family Physicians Institutional Review Board approved the survey. The survey is available in the Supplement.

The CERA survey covers program director characteristics (including gender, age, race, and years as program director) and program characteristics (including location and size). The survey included 10 questions specific to serious illness communication education. These questions explored curriculum content, including the frequency and methods of serious illness communication training; qualifications of educators, including whether the program had family medicine faculty with expertise in HAPC; outcome measures, including program director confidence in resident abilities to conduct conversations and the frequency of residents leading their own discussions; and barriers to training. Though the term has been defined more broadly elsewhere as suggested earlier,⁷ in the survey, “serious illness communication/conversations” was defined as discussing prognosis and care goals with patients with serious illness. Survey questions prompted program directors to consider training in discussions about prognosis and goals of care rather than advance care planning (including discussions about advance directives), which has been more extensively studied and integrated into quality metrics.¹⁸ Though an important element of serious illness communication, assessment of patient understanding was not included as part of the survey definition. In the questions, “expertise” referred to any person with notable training and experience with HAPC, including fellowship training or obtaining a certificate of added qualifications.

We used logistic regressions to explore demographic predictors of having family medicine faculty with HAPC expertise. We used χ² tests to investigate the associations among preparedness to lead serious illness conversations, the presence of faculty with HAPC expertise, and training characteristics. For significant associations, logistic regressions were adjusted for program director gender, years in the program director role, number of residents in the program, and residency type (university-affiliated vs. nonuniversity affiliated). Statistical analyses were conducted using SAS version 9.4 (SAS Institute Inc).

The overall response rate for the September 2023 CERA program director survey was 37.9% (271/715). Of 271 respondents, 258 (95%) responded to all questions about serious illness communication, and six responded to at least 70% of topic-specific questions. Most program directors (56.5%) came from community-based, university-affiliated residency programs. Program directors came from every region of the country and a range of community sizes, with ^{Characteristics (N=271)} most (60.5%) serving in their roles for less than 5 years. Respondents were majority female (53.5%) and majority White (73.4%). Only 37 (13.7%) program directors identified as underrepresented in medicine (Table 1).

Most programs (71.6%) provide serious illness communication training at least once per year of residency. One-third (33.6%) of program directors reported that residents primarily learn serious illness communication skills through didactic sessions, while 23.6% reported that skills are taught mainly through HAPC rotations. About half (49.8%) of program directors reported that the most important skill for residents to learn was discussing goals of care. Only eight programs (3%) did not provide training. The training method and frequency did not vary by program size, location, or university- versus community-based.

Ninety-seven program directors (35.8%) reported no barriers to serious illness communication training. Of those who cited barriers, the most common were lack of resident learning time (29.5% of cited barriers) and lack of faculty expertise (24.7%).

Educator qualifications differed by program size and director characteristics. Program size was a significant predictor of having family medicine faculty with HAPC expertise. While most programs (60%) had no faculty member with HAPC expertise, larger family medicine residency programs with more than 31 residents had higher odds of having faculty with HAPC expertise than small programs with fewer than 19 residents (OR 3.13, 95% CI 1.48–6.60). Programs with program directors who identified as underrepresented in medicine were less likely to have faculty with HAPC expertise (OR 0.35, 95% CI 0.15–0.82). Other demographic characteristics (i.e., program director gender, race/ethnicity, years of experience, community size, location, or program type) were not associated with faculty expertise.

Characteristics of HAPC expertise are presented in Table 2. Over half (52.4%) of programs reported the involvement of HAPC experts in serious illness communication training (including either family medicine faculty with HAPC expertise or palliative care faculty not affiliated with family medicine). About one-third (34.0%) of program directors reported that serious illness communication training is primarily taught by faculty who do not have HAPC expertise.

Programs with family medicine faculty with HAPC expertise had significantly higher odds of having one or more serious illness communication training sessions per year of residency compared to programs that did not have faculty with HAPC expertise (adjusted OR: 2.41, 95% CI: 1.24–4.68), adjusted for program director gender, years in program director role, number of residents in the program, and residency type (university-affiliated vs nonuniversity affiliated; Table 3). Likewise, programs with HAPC faculty also had higher odds of residents leading their own serious illness conversations (rather than referring to a specialist; adjusted OR: 2.81, 95% CI: 1.34–5.87), adjusting for the same demographics (Table 4).

The presence of faculty with HAPC expertise was not associated with whether programs had required didactic sessions to help residents understand how to navigate patients’ death anxiety and questions about life purpose/meaning. Faculty HAPC expertise also was not associated with whether residents primarily received serious illness communication training during rotations (vs. by other means), or program director ranking of resident preparedness to lead conversations.

Despite fewer reported serious illness discussions among residents at programs without HAPC expertise, 86.0% of all program directors perceived that graduates of their programs were prepared to lead effective serious illness conversations independently. Programs where program directors rated residents as prepared to lead serious illness conversations had higher odds of residents leading their own discussions (as reported by program directors) than referring to specialists (adjusted OR 4.87, 95% CI: 2.08–11.43; Table 4), adjusting for program director gender, years in a program director role, number of residents in the program, and residency type. The program director’s ranking of resident preparedness was not associated with the training method, training frequency, or HAPC faculty’s presence.

Serious illness communication training is essential for residents to become well-rounded family physicians and meet the needs of patients who are aging and living longer with serious illness. This study explored the landscape of serious illness communication training in family medicine residencies and avenues for improving training in this crucial skill set. We showed that programs with family medicine faculty with HAPC expertise had higher odds of having more frequent training and of residents leading conversations, suggesting that exposure to faculty with HAPC expertise promotes training in serious illness communication.

By 2050, 23% of the United States population will be 65 or older.¹⁹ At the same time, the prevalence of chronic conditions is rising.^20,21 As more of the population ages and chronic disease burden worsens, the demand for palliative care is expected to increase. That said, palliative care specialists cannot meet this increased demand alone. According to one estimate, by 2030, just one palliative care physician will be available for every 23,000 patients with palliative needs.⁵ Furthermore, access to specialty palliative care, which traditionally has been concentrated in academic hospital settings, remains limited in significant portions of the United States, particularly in southern and central regions of the country and in rural communities. The Center to Advance Palliative Care found that only about 17% of rural hospitals with 50 or more beds have palliative care programs.²²

Family physicians are essential in filling these access gaps by providing primary palliative care services, making family medicine residency training in serious illness communication essential. While family physicians face numerous demands and residency training time is limited, serious illness communication training should be prioritized as a foundational skill in both inpatient and outpatient settings. Fortunately, communication skills are already recognized by ACGME as a critical competency for family medicine residents. ACGME family medicine training milestones include the ability to “sensitively and compassionately deliver medical information, managing patient/family values, goals, preferences, uncertainty, and conflict” and use shared decision-making to “align patient/family values, goals, and preferences with treatment options.”²³ Integrating serious illness communication training in family medicine residency training (with even one session per year) will prepare residents for the realities of caring for patients with serious illness, particularly in the outpatient setting and in the context of an aging population and the increasing prevalence of chronic conditions. Furthermore, these skills can promote interprofessional collaboration²⁴ and may even benefit physicians personally with evidence showing that participation in communication skills training can increase empathy and reduce burnout.²⁵

Although our findings indicate the benefits of involving faculty with HAPC expertise in resident training, access to such faculty may be a barrier for some programs. For instance, our study indicated that smaller programs were less likely to have faculty with HAPC expertise. For such programs, one solution may be the use of established serious illness communication training programs such as VitalTalk, which offers virtual training courses for clinicians and faculty development to train the trainer.²⁶ Inviting experts from outside institutions, either in person or virtually, is another option when faculty expertise is lacking. Additionally, defining specific competencies for family medicine residents within serious illness communication training would enhance existing milestones and help to align training across the country.

Our survey strength was the participation of programs across the country with various community sizes and program types. The strength of the CERA survey mechanism is that it provides significant insight into resident curricula as medicine advances and our populations age.

Our limitations were those commonly reported for CERA surveys, including lack of data from nonresponding programs. Data also relied on the program director’s perceptions of resident preparedness to lead conversations and frequency of conversations, which may be prone to bias and inaccuracies. Indeed, even well-meaning program directors are unlikely to directly observe residents’ serious illness conversations to assess their competencies in most cases. Likewise, program directors are unlikely to know precisely how often residents lead serious illness conversations in the outpatient setting.

Multiple different directions for further exploration in this topic are possible. Future studies must assess family medicine residents’ perceived comfort and competencies in conducting serious illness conversations. Specifically, studies that elicit residents’ comfort with leading serious illness communication at graduation and perceptions of different training techniques are needed. Furthermore, in a world of multiple competing demands on primary care providers and limited time in training, care must be taken to ensure that training is efficient and practical. Finally, exploring further whether unique benefits exist to having trainings conducted by family medicine faculty with HAPC expertise versus other specialists with HAPC expertise would be valuable.

This national survey provides an initial overview of the landscape of serious illness communication training in family medicine residencies and indicates important areas for improvement and further exploration. Notably, equipping family medicine faculty with HAPC skills will likely enhance the frequency and quality of this critical training. In turn, in addition to allocating time for structured training sessions throughout residency, programs should consider investing in family medicine faculty development in this important skill set. The ability to lead effective serious illness conversations provides an opportunity to connect more deeply with patients and families, promotes care aligned with patient goals, and aligns directly with family medicine’s goal of providing high-quality primary care across the lifetime.

Poster presented at the Society of Teachers of Family Medicine Annual Spring Conference, Los Angeles, California, May 5, 2024.

This research was supported by the National Clinician Scholars Program and the Department of Family Medicine at UCLA and by Health Resources and Services Administration (HRSA) Ruth L. Kirschstein Institutional National Research Service Award (NRSA), grant no. T32-HP-19001.