Too Close for Comfort? Attitudes of Gynecologic Oncologists Toward Caring for Dying Patients

Purpose To assess gynecologic oncologists’ attitudes relating to palliative care referrals among advanced cancer patients. Methods Gynecologic oncologists were surveyed using validated measures to assess stigmatizing attitudes toward palliative care, anticipated stigma of palliative care, acceptance of palliative care, and willingness to refer to palliative care. Descriptive statistics were calculated. Analysis was performed using linear regression. Results 1200 physicians received the survey and 108 (9%) completed it. Most were female (69.4%) and white (82.4%). Most practiced in academics (64.8%) in urban environments (71.3%). Respondents did not have anticipated stigma surrounding palliative care referral (mean score 1.89, range 1-7, higher score indicating more stigma), were accepting of palliative care (mean score 1.45, range 1-7, higher score indicating less acceptance), and were willing to refer patients to palliative care (mean score 5.75, range 1-7, higher score indicating more willingness to refer). Linear regression demonstrated females had less anticipated stigma surrounding palliative care (B = −.213, P = .04) and higher acceptance of palliative care (B = −.244, P = .01). Most surveyed derived satisfaction from work with advanced cancer patients (83%). Nineteen percent were depressed by managing advanced cancer patients. One fourth felt emotionally burned out by dealing with too many deaths. Conclusions Most gynecologic oncologists did not exhibit stigma surrounding palliative care and derive satisfaction from their work. Some gynecologic oncologists experience depression and burnout related to their profession. This close connection with patients as they transition to the end of life may take a toll on providers.


Introduction
Early integration of palliative care into the management of advanced cancer patients is recommended by the National Comprehensive Cancer Network (NCCN) and American Society of Clinical Oncology (ASCO). 1,2][5] Several randomized controlled trials have shown early involvement of palliative care services improves patient quality of life, decreases caretaker burden, and decreases overall cost of care when compared to standard therapy alone.Despite these benefits, palliative care services remain underutilized.When patients are referred, it is often at the late stages of their disease. 6wo underexplored barriers to palliative care utilization are physician stigma surrounding palliative care and the personal/ professional impact that palliative care discussions (inclusive of symptom management, formal palliative care referral, and endof-life care discussions) cause providers managing these patients.
For example, caring for patients with terminal cancer presents challenges in terms of addressing the complex emotional, physical, and social needs of both patients and their families.These situations may put gynecologic oncologists at risk for burnout or avoidance of palliative care interventions given the emotionally exhaustive nature of this work.
Stigma is defined as society's negative evaluation of a particular individual or entity. 7Palliative care stigma consists of both the perceived and experienced stigma from patients and caregivers, as well as stigmatizing attitudes from health care providers. 8,9Palliative care stigma may occur because of palliative care's association with death and dying which can provoke fear, anxiety, avoidance, and hopelessness in patients, caretakers, and providers. 10,113][14][15] In these contexts, stigmas related to these health issues reduced rates of medical care seeking.
Stigma associated with palliative care may lead to both: (1) lack of acceptance of referrals to or utilization of palliative care by patients and (2) lack of referrals to palliative care by providers.Both patients who transition to palliative care and their health care providers frequently experience feelings of guilt and shame for "giving up." 8,16 Patients may also experience devaluation and fear of abandonment/loss of care. 17his fear and avoidance of palliative care among advanced cancer patients and the reduced/delayed referrals to palliative care by health care providers suggests that palliative care stigma may impact not only patients but also providers.
Another component that may act as a barrier for palliative care referral could be the impact that caring for terminally ill patients has on the personal and professional life of the gynecologic oncologist.While guiding patients in crisis through the complex decision making associated with end-of-life care, gynecologic oncologists may often feel an emotional and physical toll.This toll is seen in other physician groups with palliative care clinicians experiencing burnout rates as high as 62%. 18This stress may add up over time in their practice leading to feelings of depression, burnout, dissatisfaction, and depersonalization. 19Ultimately, this may contribute to decreased utilization of palliative care services resulting from physician avoidance of addressing end of life care in order to minimize their own negative feelings surrounding caring for terminally ill patients.
Although most gynecologic oncologists are directly involved in the end-of-life care of their patients, there is a paucity of data pertaining specifically to the attitudes of gynecologic oncologists regarding palliative care referrals for their patients.It is important to better understand gynecologic oncologists' attitudes relating to palliative care so that potential stigmas or other barriers to palliative care referrals among advanced cancer patients may be explored.This understanding may inform new and impactful interventions that could improve utilization of palliative care services among advanced cancer patients.The primary objective of this study was to assess gynecologic oncologists' attitudes relating to palliative care referrals among advanced cancer patients.

Study Design
This study was an Institutional Review Board (IRB) approved online survey of gynecologic oncologists who were Society of Gynecologic Oncology (SGO) members regarding their attitudes toward palliative care for advanced cancer patients.By completing the online survey, participants consented to be in the study.
SGO members categorized as full member gynecologic oncologists were identified using the SGO email mailing list.Study data were collected and managed using REDCap electronic data capture tools.Eligible SGO members were then sent an anonymous electronic survey via RedCap.The survey took approximately 20-25 minutes to complete and potential participants received up to 2 email reminders to complete the survey.
The survey consisted of questions assessing respondents' basic demographics which included respondents age, race, gender, type of practice, time since completion of fellowship and presence of palliative care at their institution.Participants were then asked to complete the following surveys: (1) Attitudes of Medical Oncologists toward Palliative Care for Patients with Advanced and Incurable Cancer, a 24-item survey that assesses oncologists' attitudes toward palliative care but does not directly measure stigma. 20It is rated on a Likert-type scale of 1 (disagree strongly) to 5 (agree strongly).( 2) Anticipated Stigma of Palliative Care scale, an 7item adaptation of prior versions of the Anticipated Stigma Scale and Chronic Illness Anticipated Stigma Scale, which are both well-validated and reliable scales (Cronbach's α = .95,test-retest ICC=.82, respectively). 21,22This scale measures the degree to which individuals anticipate feeling stigma within an encounter.All items are rated on a Likert-type scale of 1 (not at all likely) to 7 (very likely).(3) Acceptance of Palliative Care scale, a 4-item measure which assesses the degree to which oncologists are able to refer to palliative care and continue normal care (Acceptance of Palliative Care).All items are rated on a Likert-type scale of 1 (strongly disagree) to 7 (strongly agree).( 4) Willingness to Refer to Palliative Care.7-item measure developed by Shen and Wellman to assess in which scenarios oncologists would be willing to refer patients to palliative care.All items are rated on a Likert-type scale of 1 (very unwilling) to 7 (very willing).
See Supplemental Figure 1 for scales

Analysis
IBM SPSS Statistics version 27 was used for statistical analysis.Descriptive statistics were calculated.Scores for each scale were calculated per coding instructions for each scale.Associations between provider sociodemographic characteristics and scores on the validated measures were analyzed using linear regression.

Results
There were 1200 surveys sent, and 108 completed the survey; the response rate was 9%.Most respondents were female (n = 75, 69.4%) and white (n = 89, 82.4%).Most described their practice as academic (n = 70, 64.8%) and in an urban environment (n = 77, 71.3%).See The majority (n = 103, 95.4%) of respondents had palliative care providers at their institution and stated they were often directly involved with discussing end of life care preferences with their patients (n = 95, 88%).See Table 1 for practice characteristics.The majority believed that oncologists should coordinate the care of cancer patients at all stages of disease, including end of life care (n = 67, 66%).Most believed that all advanced cancer patients should receive concurrent palliative care along with antitumor therapies (n = 87, 81%).Regarding attitudes of gynecologic oncologists toward caring for dying patients, the majority derived satisfaction from their work managing patients with advanced cancer (n = 91, 83%).Nineteen percent (n = 20) stated that managing patients with advanced cancer and dying patients depresses them.Twentyfive percent (n = 27) stated they felt emotionally burned out by having to deal with too many deaths.For the above questions, linear regression analyses demonstrated that there were no significant associations between sociodemographic characteristics and responses (P > .05).Sociodemographic characteristics included age, race, gender, academic vs non-academic practice and urban vs non-urban environment.
Scores on the validated measures demonstrated that respondents reported low levels of anticipated stigma surrounding referral (mean score 1.89, range 1-7 with higher score indicating more stigma).Respondents were accepting of palliative care for their patients (mean score 1.45, range 1-7 with higher score indicating less acceptance) and were willing to refer patients to palliative care (mean score 5.75, range 1-7 with higher score indicating more willingness to refer).Linear regression demonstrated that female providers had less stigma surrounding palliative care for their patients (B = À.213,P = .04)and had higher acceptance of palliative care for their patients (B=À.244,P = .01).See Table 2 for scores.

Discussion
The results of our survey demonstrate that most gynecologic oncologists surveyed did not exhibit stigma toward palliative care for their patients.They were generally accepting of and willing to refer to these services.Male providers had more stigma surrounding palliative care and were less accepting of palliative care for their patients when compared to their female colleagues.Most gynecologic oncologists are involved in the care of their patients at the end of life and derive satisfaction from their work, however, this close connection with patients as they transition to the end-of-life may take a toll on physicians.One fourth of respondents felt emotionally burned out by managing death, and about 1 in 5 surveyed felt depressed by end-of-life care work.Although this study is limited in its small sample size and low response rate, it is the first to explore the impact of palliative care stigma on the field of gynecologic oncology.Strengths of our study include the use of validated scales.Limitations of this study include our low response rate and the demographics of our respondents.Unfortunately, unincentivized physician surveys are often characterized by low response rates.Reasons for poor response rates includes lack of time, length of the survey, perceived salience of the study, concerns about confidentiality of the results or if the respondent feels questions appear biased or do not allow the full range of choices on the subject. 23There is not a universally accepted response rate for an unincentivized survey, however response rate usually falls between 6-20%. 23,24Our response rate was low at 9%.This may be due to the topic matter or length of the survey.
Another limitation of our study was the demographics of our population.Our respondents were primarily young white women in academics.While this demographic is similar to the demographics of the recent SGO 2020 State of the Society Survey (54% female, 70% white), there is significant work needed to enhance diversity, equity, and inclusion within the specialty of Gynecologic Oncology. 25This study may also be affected by selection bias of survey responders vs non responders since those responding to the survey may disproportionately represent gynecologic oncologists with an interest in palliative care.
Our study highlights the toll caring for terminally ill patients may take on physicians.With 1 in 4 respondents reporting emotional burnout by managing death and nearly a fifth surveyed feeling depressed by end-of-life care, many gynecologic oncologists may experience secondary trauma from their work and have challenges coping with death of their patients.Although the majority surveyed derive satisfaction in their work with end-of-life care, further investigation of factors leading to burnout, coping mechanisms and stress management strategies utilized by gynecologic oncologists is warranted.Better understanding of the prevalence and predictors of burnout may help to mitigate the burden gynecologic oncologists feel managing patients with advanced cancer.Given that this study also demonstrates a potential gender bias on the overall positive view of palliative care, further studies could also explore if there is a relationship between provider gender and palliative care referrals.
Managing patients with advanced cancer is central to the work of gynecologic oncologists and as evidenced by this study, the majority of gynecologic oncologists do not exhibit stigma regarding palliative care for their patients.Most gynecologic oncologists surveyed recognize the importance of palliative care and its role in treating their patients with advanced cancer.Despite this sentiment, there does continue to be low or late utilization of palliative care among advanced cancer patients. 6Perhaps there is a disconnect between physicians' perceived acceptance of palliative care services and their actual usage of it.There may also be patient factors that limit palliative care utilization.Although our survey did not assess this, future studies may evaluate a physician's perceived acceptance of palliative care and their actual usage of palliative care services for their patients.A focus on educational interventions may also help physicians feel more comfortable incorporating palliative care in clinical practice. 26For instance, among those surveyed, nearly a third of respondents disagreed that they received good training in palliative care during their fellowship, and nearly a third of respondents endorsed that they do not feel comfortable managing the physical and psychological symptoms of advanced cancer.This may present an opportunity to improve palliative care education during fellowship training.
This study enhances understanding around gynecologic oncologists' attitudes toward utilizing palliative care services for their patients.It also highlights the challenges that gynecologic oncologists may have in coping with the death of their patients.Future work should evaluate ways to improve provider education regarding palliative care, increase utilization of palliative services and develop interventions to assist gynecologic oncologists who may be experiencing a negative personal impact from their rewarding but challenging work.

Table 1 .
Demographics of Respondents.

Table 2 .
Scores on Validated Scales.
aSurveys not completed in their entirety by all participants.