Gynecologic Oncology 133 (2014) 319–325

Contents lists available at ScienceDirect

Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno

Predictors of palliative care consultation on an inpatient gynecologic oncology service: Are we following ASCO recommendations?☆ Carolyn Lefkowits a,⁎, Anna B. Binstock b, Madeleine Courtney-Brooks a, Winifred G. Teuteberg c, Janet Leahy c, Paniti Sukumvanich a, Joseph L. Kelley a a Department of Obstetrics, Gynecology & Reproductive Sciences, Division of Gynecologic Oncology, Magee-Womens Hospital of the University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA b Department of Obstetrics, Gynecology & Reproductive Sciences, Magee-Womens Hospital of the University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA c Department of Medicine, Division of General Internal Medicine, Section of Palliative Care & Medical Ethics, University of Pittsburgh Medical Center, Pittsburgh, PA 15213, USA

H I G H L I G H T S • Gynecologic oncology patients referred to inpatient palliative care have higher disease and symptom burden and poor prognosis. • Referral patterns to inpatient palliative care of patients meeting ASCO recommendations for early palliative care lack consistency and timeliness.

a r t i c l e

i n f o

Article history: Received 17 December 2013 Accepted 23 February 2014 Available online 1 March 2014 Keywords: Inpatient palliative care Gynecologic cancer Early palliative care integration Palliative care referral patterns

a b s t r a c t Objective. Determine predictors of inpatient palliative care (PC) consultation and characterize PC referral patterns with respect to recommendations from the American Society of Clinical Oncology (ASCO). Methods. Women with a gynecologic malignancy admitted to the gynecologic oncology service 3/2012–8/ 2012 were identified. Demographic information, disease and treatment details and date of death were abstracted from medical records. Student's t-test, Fischer's exact test or χ2-test was used for univariate analysis. Binomial logistic regression was used for multivariate analysis. Results. Of 340 patients analyzed, 82 (24%) had PC consultation. Univariate predictors of PC consultation included race, cancer type and stage, recurrent disease, admission frequency, admission for symptom management or malignant bowel obstruction (MBO), discharge to skilled nursing facility (SNF) and number of lines of chemotherapy. On multivariate analysis, significant predictors of PC consultation were recurrent disease (OR 2.4, 95% CI 1.1–5.3), number of admissions (≥3, OR 10.9, 95% CI 3.4–34.9), admission for symptom management (OR 19.4, 95% CI 7.5–50.1), discharge to SNF (OR 5, 95% CI 1.9–13.5) and death within 6 months (OR 16.5, 95% CI 6.9–39.5). Of patients considered to meet ASCO guidelines, 53% (63/118) had PC referral. Of patients referred to PC, 51.2% (42/82) died within 6 months of last admission. Conclusions. Patients referred to inpatient PC have high disease and symptom burden and poor prognosis. High-risk patients, including those meeting ASCO recommendations, are not captured comprehensively. We continue to use PC referrals primarily for patients near the end of life, rather than utilizing early integration as recommended by ASCO. © 2014 Elsevier Inc. All rights reserved.

Introduction The American Society of Clinical Oncology (ASCO) defines palliative care as the “integration into cancer care of therapies that address the multiple issues that cause suffering for patients and their families and impact their quality of life” [1]. Unlike hospice services, which are

☆ This paper will be presented as an oral presentation at the SGO 45th Annual Meeting on Women's Cancer, Tampa, FL, March 2014. ⁎ Corresponding author at: Magee-Womens Hospital of UPMC, Division of Gynecologic Oncology, 300 Halket St, Pittsburgh, PA 15213-3180, USA. Fax: +1 412 641 54170. E-mail address: [email protected] (C. Lefkowits).

http://dx.doi.org/10.1016/j.ygyno.2014.02.031 0090-8258/© 2014 Elsevier Inc. All rights reserved.

available only to patients with a projected survival of less than six months, palliative care (PC), including symptom management, can be offered from the time of diagnosis until death and should, per National Comprehensive Cancer Network (NCCN) guidelines “be delivered concurrently with disease-directed, life-prolonging therapies” [2]. Palliative care can be delivered to cancer patients by the oncology team (often referred to as “generalist” or “primary palliative care”) or by providers with specialty training in palliative care (“specialist palliative care”), or a combination of both [3]. Palliative care has been shown to be associated with benefits including improved clinical outcomes (quality of life, symptom control and patient and caregiver satisfaction), as well as improved healthcare

320

C. Lefkowits et al. / Gynecologic Oncology 133 (2014) 319–325

utilization outcomes (decreased hospital costs, inpatient and ICU admissions) [4–10]. A landmark randomized controlled trial comparing integration of palliative care from the time of diagnosis to usual care in patients with metastatic non-small-cell lung cancer found improved quality of life, less depression and longer median survival [4]. The literature regarding benefits of palliative care in the gynecologic oncology population specifically is limited to retrospective studies, largely of decedents. Demonstrated benefits of palliative care services for gynecologic oncology patients, including end-of-life discussions, hospice care and palliative care consultation, have included shorter hospital length of stay, less aggressive end of life care and less resource utilization, all without detriment to survival [11–16]. As evidence has mounted for its benefits, palliative care has become more widely accepted as a standard component of cancer care, recognized by the World Health Organization (WHO) as one of the four main components of a national cancer control program [17] and required in the United States for accreditation of cancer hospital programs [18]. As we seek to utilize palliative care integration into oncology care most effectively, we must ask the following questions: (1) Which providers (oncologists versus specialty palliative care providers) should provide which components of palliative care services? (2) When in the disease course should palliative care services be integrated? and (3) Which patients are likely to gain the greatest benefit from palliative care integration? Recently released recommendations from the Society of Gynecologic Oncology suggest that “for women with advanced or relapsed gynecologic cancer, basic level palliative care should not be delayed, and when appropriate, referral should be provided for specialty palliative medicine” [19]. This recommendation includes direction on which patients might be targeted for palliative care (patients with advanced or relapsed disease), and recognition that not all palliative care needs to be delivered by specialists. It does not specify when referral to specialists would be considered “appropriate.” The general oncology literature does include some recommendations about when integration of palliative care consultation might be appropriate including NCCN guidelines and expert opinion [2,3]. However, there has been minimal evaluation of these guidelines, none of which has included gynecologic oncology patients [20,21]. In 2012 ASCO released a Provisional Clinical Opinion entitled "The Integration of Palliative Care Into Standard Oncology Care", recommending consideration of combined standard oncology care and palliative care “early in the course of illness for any patient with metastatic cancer and/or high symptom burden” [22]. This recommendation does not address the question of who (oncologist versus specialist) should provide the palliative care. It does provide some general advice regarding two components of appropriate circumstances for palliative care integration: timing and patient characteristics. It also leaves room for interpretation; there is no consensus regarding the definition of “early” palliative care integration, or what constitutes “high symptom burden”. There has been no experimental evaluation of implementation of these recommendations. In order to ultimately answer the questions outlined above regarding the who (provider), when, and to whom (patient characteristics) of optimal palliative care integration, we must begin by describing the current patterns of palliative care integration in gynecologic oncology. In the absence of data to describe what is optimal, we can only assess whether our current referral patterns adhere to ASCO recommendations, which has not been previously described. The existing literature regarding predictors of non-hospice palliative care in gynecologic oncology patients specifically is comprised of a single cohort study limited to decedents [23]. Therefore we do not know which patients in the general gynecologic oncology population are receiving palliative care consultation, particularly early in the disease course, nor do we know whether our existing referral patterns are in line with ASCO recommendations. The objective of our study is to describe the current patterns of utilization of non-hospice palliative care consultation services among

gynecologic oncology inpatients in order to (1) determine predictors of non-hospice palliative care referral and (2) characterize referral patterns with respect to ASCO recommendations. By describing existing practice patterns regarding referral to palliative care, we hope to set the stage for the larger discussion of what might constitute optimal palliative care integration. Materials and methods Study approval was obtained from the University of Pittsburgh Institutional Review Board for a retrospective chart review of patients with a gynecologic malignancy admitted to the gynecologic oncology inpatient service between March 1st, 2012 and August 31st, 2012. Data was abstracted by two authors (C.L. and A.B.); one author performed spot checks of charts reviewed by the second author in order to promote consistency in data abstraction. Abstracted data included patient age, marital status, race, cancer type and stage, and identity of primary oncologist. For each unique patient admitted during the study time period, we recorded total number of admissions, whether the patient had a palliative care consultation, whether the patient developed recurrent disease, total number of chemotherapy regimens administered by 8/2012, and whether the patient died within six months of her last admission during that time period. The palliative care consultation variable reflects only those patients actually seen by palliative care, as we did not have record of patients referred but not seen. We recorded whether or not each patient was discharged to a skilled nursing facility (SNF), had an admission for malignant bowel obstruction (MBO), or an admission primarily for symptom management. An admission was considered to be primarily for symptom management if the primary reason for admission was management of symptoms rather than evaluation for etiology of symptoms and the patient did not have other indications for admission, such as infection or planned procedure. We did not have access retrospectively to any patient reports of symptom intensity. Age, marital status, race and cancer type were chosen as possible predictive variables based on prior studies in non-gynecologic oncology populations and the single cohort study of gynecologic oncology decedents previously described [23–29]. The other variables were exploratory and intended to be hypothesis generating. We considered the following patients to represent a subset of patients meeting ASCO recommendations for incorporation of early palliative care: patients with admission primarily for symptom management or for MBO were designated as having “high symptom burden” and patients with stage IV disease or recurrent disease were designated as having “metastatic disease.” These categories were not intended to comprehensively capture all patients who met ASCO recommendations, rather to capture an easily identified subset of patients meeting ASCO guidelines. The inpatient PC consult service began seeing patients at our institution in 2001. In 2012 the PC consult service was consulted by the gynecologic oncology service during 212 patient admissions (this total may include more than one consultation on a given patient). The PC consult service currently includes a full-time nurse practitioner and physician and a dedicated social worker. The following services are available part-time: behavioral health, massage therapy, pet therapy, and music therapy. Patient care at our institution is done in a multidisciplinary manner with all palliative care services provided by the PC consult service. Statistical analysis was performed using IBM SPSS Statistics 21 software (SPSS Inc.). Baseline demographics were summarized using descriptive statistics including medians, frequencies, and percentages. Univariate analysis was performed using Student's t-test for continuous variables and either χ2-test or Fisher's exact test for categorical variables. Prior to performing univariate analysis on continuous variables, Levene's test was used for equal variance between the two populations. To identify independent factors associated with a palliative care referral, all variables with a p-value b 0.05 on univariate analysis were included

C. Lefkowits et al. / Gynecologic Oncology 133 (2014) 319–325

in a multivariate logistic regression using backward selection. A p value of b0.05 was considered statistically significant. Results There were a total of 615 admissions to the gynecologic oncology service during the study period, representing 509 unique patients. The 340 patients with a gynecologic malignancy were included in the analysis. Demographic and clinical information for the entire cohort is presented in Table 1. The patients were primarily white (89.4%) and about half were married (50.3%). Most had ovarian (36.2%) or endometrial (45%) cancer. Almost a quarter had recurrent disease and a total of 16% died within six months of their last admission during the study period. Overall, 24.1% of our cohort (82/340) was referred to PC. On univariate analysis, patients who were non-white, had ovarian or cervical cancer, higher stage disease, recurrent disease, greater frequency of admissions, admission for symptom management or MBO, discharge to a SNF, and more lines of chemotherapy were more likely to be referred to palliative care (Tables 2 & 3). Patients dying within 6 months of their last admission during the study period were also more likely to be referred to PC (Table 2). There were no differences by age, marital status, or identity of primary oncologist. All variables which had a p value of b0.05 on univariate analysis were included in a multivariate logistic regression using backward selection. The initially included variables were: race, disease site, stage at diagnosis, number of lines of chemotherapy, recurrent disease, death within 6 months of last admission within the study time period, number of admissions, admission primarily for symptom management, admission for malignant bowel obstruction, and discharge to a SNF. On multivariate logistic regression, significant predictors of PC referral were recurrent disease (p = 0.03), ≥ 3 admissions (p b 0.001), admission for symptom management (p b 0.001), discharge to a SNF (p = 0.001), and death within 6 months of last admission (p b 0.001) (Table 4). Of patients referred to PC, 51.2% (42/82) died within 6 months of their last admission during the time period in question. Fig. 1 illustrates the following rates of PC consultation among patients considered to meet ASCO recommendations for consideration of early PC integration: stage IV 59% (23/39), recurrent disease 54.4% (43/79), MBO 72.7% (8/11) and admission for symptom management

Table 1 Patient characteristics among 340 unique patients with gynecologic malignancy admitted over a six-month period. n Median age (years) Median number admissions Race African American White Other Unknown Married Yes No Disease site Ovarian Endometrial Cervical Vulvar/vaginal Stage at diagnosis I II III IV N/A Recurrent diseasea Death within 6 months of last admissionb a b

%

62.0 1.0 21 304 6 9

6.2 89.4 1.8 2.6

163 171

47.9 50.3

123 153 45 15

36.2 45 13.2 4.4

156 26 103 39 16 79 55

45.9 7.6 30.3 11.5 4.7 23.2 16.2

Recurrent disease by end of study time period (8/2012). Death within 6 months of last admission during study time period (3/2012–8/2012).

321

79.2% (38/48). Of the overall group of patients included in at least one of our categories for meeting ASCO recommendations, 53.3% (63/118) were referred to palliative care. Of patients who died within 6 months of their last admission, 74.5% (41/55) were seen by PC. Discussion Having reached the point where palliative care is recognized as a key component of comprehensive cancer care, we must address the following issues in order to optimize its integration into the care of gynecologic oncology patients: who should provide palliative care, when in the disease course it should be initiated, what the intervention should entail and which patients are most likely to benefit. Our project provides a description of the “to whom” and “when” of current utilization of non-hospice PC on an inpatient gynecologic oncology service. Overall, 24% of our population was referred to inpatient PC. Glare et al. describe the rate of palliative care consultation on an inpatient gastrointestinal oncology service at about 10% prior to an intervention designed to standardize screening for palliative care needs [21]. Prior to a pay-forperformance intervention designed to increase palliative care consultation, Bernacki et al. described rates of palliative care consultation among patients considered “high risk” at 15% and 16% at two different academic hospitals [30]. Otherwise, description in the literature of rates of palliative care consultation among cancer patients is generally limited to decedents. Hui et al. described an inpatient palliative care consultation rate among a cohort of decedents at MD Anderson Cancer Center of 45% [24]. In the gynecologic oncology literature, Fauci et al. described a rate of 70% of referral to hospice or palliative care among a cohort of 268 decedents, but did not distinguish between hospice and nonhospice palliative care [15]. Most recently, Nevadunsky et al. described a non-hospice palliative care consultation rate of 40% among a cohort of 100 gynecologic oncology decedents [16]. Given that only 16% of our cohort died within 6 months of last admission within the study period, our overall rate of PC consultation of 24% is likely higher than what has been previously described. Anecdotally we believe that contributors to our relatively high consultation rate may include (1) the long-standing nature of the palliative care consultation service at our institution, which has been in place since 2001, (2) the palliative care team being well integrated into the gynecologic oncology floor, with the palliative care nurse practitioner sitting in on nursing rounds every morning Monday through Friday to familiarize herself with the inpatient service, (3) the cordial working relationship between the palliative care nurse practitioner and the gynecologic oncology midlevel providers is such that the palliative care nurse practitioner is encouraged to bring to the attention of the gynecologic oncology team patients that might be appropriate for a palliative care consultation, (4) the palliative care team having demonstrated over the past decade good consultation etiquette, including limiting the scope of consultation to what has been requested (i.e. not addressing goals of care unless specifically invited to do so) and (5) the comprehensive nature of the available palliative care services (physician available in house or on call 24 hours a day, 7 days a week, full-time nurse practitioner Monday through Friday who provides continuity when the physicians rotate, full-time social work and other interdisciplinary services available part-time including behavioral health, massage therapy, pet therapy and music therapy). Our future research plans include leveraging some of these perceived strengths in a project using clinical triggers to increase consultation rates further among certain subgroups of patients. In terms of the “to whom” of current referral practices to PC, patients referred to PC were more likely to be of non-white race, have ovarian or cervical cancer, advanced stage disease, recurrent disease, a greater number of hospital admissions, an admission for symptom control or MBO, higher number of previous lines of chemotherapy, and were more likely to die within 6 months of last admission. On multivariate logistic regression, significant independent predictors of PC consultation

322

C. Lefkowits et al. / Gynecologic Oncology 133 (2014) 319–325

Table 2 Demographic & disease characteristics by palliative care (PC) referral group (n = 340). Not referred to PC (n = 258) No Age (mean, years) Age categorical b50 years 50–74 years ≥75 years Race White Non-White Married Yes No Primary oncologist 1 2 3 4 5 6 7 Disease site Ovarian Endometrial Cervical Vulvar/vaginal Stage at diagnosis I/II III/IV Number of lines of chemotherapyd 0 1–2 ≥3 Recurrent diseasee Y N Death within 6 months of last admission Y N

Referred to PC (n = 82) %

No

61.2

p % 0.41a 0.64b

59.7

45 169 44

71.4 77.2 75.9

18 50 14

28.6 22.8 24.1

238 15

78.3 55.6

66 12

21.7 44.4

124 129

76.1 75.4

39 42

23.9 24.6

16 34 76 50 28 40 14

61.5 70.8 79.2 84.7 71.8 78.4 66.7

10 14 20 9 11 11 7

38.5 29.2 20.8 15.3 28.2 21.6 33.3

78 133 30 14

63.4 86.9 66.7 93.3

45 20 15 1

36.6 13.1 33.3 6.7

161 84

88.5 59.2

21 58

11.5 40.8

139 101 15

88.5 70.6 40.5

18 42 22

11.5 29.4 59.5

37 221

46.8 84.7

42 40

53.2 15.3

14 244

25.5 85.9

41 40

74.5 14.1

0.02c

0.90c

0.22b

b0.001b

b0.001c

0.001b

b0.001c

b0.001c

PC = palliative care. a Independent samples t-test. b Chi-square test. c Fisher's exact test. d Number of lines of chemotherapy by end of study period (8/2012). e Recurrent disease by end of study period (8/2012).

were recurrent disease, number of admissions, admission for symptom management, discharge to SNF and death within 6 months of last admission. We interpreted discharge to SNF as an indicator of poorer functional status. Overall our results suggest that patients referred to palliative care have high disease burden, high symptom burden, high admission rates, poor functional status, and poor prognosis compared to those not referred. Our study is the first to describe predictors of non-hospice palliative care referral in a gynecologic oncology cohort not limited to decedents. In the cohort of 100 gynecologic oncology decedents described by Nevadunsky et al., there were no difference in age, race, cancer site, stage, grade or insurance between patients referred (49%) and not referred to PC prior to death [31]. The literature on this question is also limited in the general oncology population. One retrospective study of non-gynecologic cancer patients found predictors of utilization of nonhospice PC services to include older age, married and certain types of cancer (including gynecologic, lung and head & neck) [24]. Other studies have had similar findings regarding marital status [25,26,31], but several have found older patients to be less likely to utilize nonhospice palliative care [25–27]. In the limited number of studies addressing race, findings have been mixed, with one study finding lower

rates of utilization of palliative care among ethnic minorities [28] and others finding no racial differences [24,29,31]. ASCO recommendations regarding palliative care integration suggest integration of palliative care for “any patient with metastatic cancer and/or high symptom burden” [22]. As described above, we considered patients with stage IV disease or recurrent disease to have “metastatic cancer” and patients with an admission for symptom management or malignant bowel obstruction to have “high symptom burden.” Given that patients with PC consultation were more likely to have advanced stage or recurrent disease and admission for symptom management or MBO, we conclude that our referral patterns to PC are capturing a group of patients characterized by higher symptom burden and rates of advanced or metastatic disease, compared to those not referred. Regarding the comprehensiveness with which patients meeting ASCO recommendations are being referred, we found the referral rates for individual categories to range from 59% (stage IV disease) to 79% (symptom management admission) and the overall referral rate for patients in ≥1 of these categories to be 53%. While these referral rates are higher than what has been described in the literature, even in cohorts consisting entirely of decedents [21,24,30,31], these subgroups of patients meeting ASCO recommendations are still not being captured

C. Lefkowits et al. / Gynecologic Oncology 133 (2014) 319–325

323

Table 3 Admission information by palliative care (PC) referral group (n = 340). Not referred to PC (n = 258)

Referred to PC (n = 82)

p

No

%

No

%

251 7

78.7 33.3

68 14

21.3 66.7

10 247

20.8 84.9

38 44

79.2 15.1

3 255

27.3 77.5

8 74

72.7 22.5

18 240

48.6 79.2

19 63

51.4 20.8

b0.001a

Number of admissions 1–2 ≥3 Admission primarily for symptom management Y N Admission for malignant bowel obstruction Y N Discharge to SNF Y N

b0.001b

0.001b

b0.001b

PC = palliative care. SNF = skilled nursing facility. a Chi-square test. b Fisher's exact test.

comprehensively, representing potential missed opportunities for these women to benefit from PC. The question of how many missed opportunities for palliative care consultation are represented in these referral percentages remains unanswered; there is no literature base or best practice standard to dictate what the “right” palliative care referral percentage should be, in our categories or any other categories. The ASCO guidelines appropriately leave room for interpretation, advising that combining “palliative care” and standard oncology care should be “considered” “early” in the disease course, without further defining either “palliative care” or “early.” In our estimation, the ASCO guidelines certainly do not suggest that 100% of patients in the categories we defined should receive inpatient palliative care consultation (the only measure of palliative care used in this study). And while it would seem appropriate that patients in these categories all receive some form of palliative care prior to being hospice eligible, that could come in various forms (inpatient or outpatient specialized palliative care or palliative care provided by the oncology team or primary care provider, at a specialty center or in the community), many of which were not measured in this study. Therefore the referral percentage among patients in these categories that would reflect optimal utilization of inpatient specialized palliative care consultation likely lies somewhere between the 53% that we observed and 100%. A project using clinical triggers to provide some standardization to palliative care referral patterns is ongoing at our institution. In terms of the “when” of referral practices to PC in our cohort, 51% (42/82) of the patients referred to PC died within 6 months of their last admission in the study period. There is no consensus on the optimal timing for palliative care integration or what constitutes “early” integration of non-hospice palliative care. The randomized controlled trial by Temel et al. incorporated palliative care from the time of diagnosis in a

Table 4 Multivariate logistic regression analysis for predictors associated with palliative care (PC) referral. Variable

Odds ratio

95% CI

p-Value

≥3 admissions during time period Admission primarily for symptom management Death w/in 6 months end of time period Discharge to a SNF Recurrent disease

10.9 19.4 16.5 5.0 2.4

3.4–34.9 7.5–50.1 6.9–39.5 1.9–13.5 1.1–5.3

b0.001 b0.001 b0.001 0.001 0.03

SNF = skilled nursing facility.

population with median overall survival of b12 months [4]. One study of outpatient palliative care defined “early referral” as patients with expected survival ≥2 years or those receiving treatment with curative intent. Other definitions or “early” or “timely” palliative care integration have included 3–6 months before cessation of chemotherapy, when prognosis for survival is “a matter of months”, while patient is still receiving chemotherapy and 2 weeks before death [2,27,32,33]. In the gynecologic oncology population, Nevadunsky et al. defined “timely” PC consultation as ≥30 days before death. Rate of PC consultation in that cohort of 100 decedents was 49% and rate of “timely” consultation was 18% [16]. In their analysis of the cost-effectiveness of early palliative care in recurrent platinum-resistant ovarian cancer, Lowery et al. defined early palliative care consultation as 6 months prior to death [34]. In recommending palliative care integration “at diagnosis, when disease recurs and when disease becomes advanced/incurable,” Ramchandran & Von Roenn take a different approach to recommendations regarding timing of palliative care integration, focusing on clinical events in disease course rather than prognosis for duration of survival [35]. This may be more practical for use prospectively, given the documented phenomenon of physicians' overestimation of prognosis [36]. Though the ASCO recommendations do not define “early in the course of illness,” given that the majority of patients receiving PC consultation in our cohort did so within a time frame in which they would have been eligible for hospice care (prognosis b 6 months), we conclude that we continue to miss opportunities for early integration of non-hospice PC [22]. Shortcomings of our study include its retrospective nature, the exclusion of patients admitted to services other than the gynecologic oncology service and the lack of consideration of outpatient PC services. Based on the fact that during the study period availability of outpatient PC services was limited, we presumed that the number of patients seen by outpatient PC who were not seen by inpatient PC would be low. This was confirmed when we reviewed the outpatient records of the 55 patients meeting ASCO recommendations who were not referred to inpatient PC and found that only 1 of them (0.8% of the 118 patients meeting ASCO recommendations) was seen by outpatient PC during the study period. Another limitation is that our referral rates do not account for either (1) patients referred to but not seen by PC or (2) patients who may have been seen by PC either inpatient or outpatient prior to the start of the study period and thus could be considered to have already had PC integration into their oncology care. Possible scenarios in which patients may have been referred to PC but not seen include if PC consultation

C. Lefkowits et al. / Gynecologic Oncology 133 (2014) 319–325

Percent of Patients Referred to PC

324

100 90 80 70 60 50 40 30 20 10 0 Stage IV

Recurrent disease

Malignant bowel

Symptom

(n=39)

(n=79)

obstruction

management

(n=11)

admission (n=48)

Fig. 1. Palliative care consultation rates among selection of patients meeting ASCO recommendations for early integration of palliative care.

was not available at the time, payment or insurance constraints or patient refusal. In our experience, all of these scenarios are exceedingly rare. We hypothesized that only a very small number of patients would have been seen by PC prior to the start of the study period. This was confirmed by review of the 54 patients meeting ASCO recommendations and not seen by inpatient or outpatient PC during the study period; a total of only 4 (3.3% of the 118 meeting ASCO recommendations) were seen by PC prior to the start of the study period (2 inpatient and 2 outpatient). We recognize that the patterns of palliative care at our single institution may not be generalizable, particularly given the findings of Wentlandt et al. that predictors of oncologists' willingness to refer patients to PC centered around characteristics of the PC service and the relationship between the oncologist and that service [27]. And though we didn't find differences in palliative care consultation rates by race, the generalizability of our findings may be limited by our population's racial homogeneity, being almost 90% Caucasian. A final limitation of the study is the lack of outcome data — whether PC consultation actually resulted in improved quality of life, impacted survival or affected health care spending. Our study suggests that current practices of utilizing inpatient PC referral in gynecologic oncology leave opportunities for improvement in both capturing the patient population (the “to whom”) and adhering to the timing (the “when”) of the ASCO recommendations regarding incorporation of palliative care into standard oncologic care. Efforts to close these gaps will need to include initiatives in research, healthcare delivery systems, and education. Research should be directed toward identifying the specific components of PC services that are beneficial, the patients most likely to benefit from, and the ideal timing for integration. As research informs best practices for palliative care integration, healthcare delivery systems can be created to standardize their implementation. Existing descriptions of initiatives to standardize delivery of PC services to patients considered most likely to benefit including clinical triggers for PC referral [21] and pay for performance initiatives [30] have resulted in increased rates of referral to PC. Clinical outcomes of such systems have not been evaluated [21,30]. Additional education of both training and practicing gynecologic oncologists may also help to close these gaps, by ensuring that all gynecologic oncologists possess a basic skill set in palliative care that renders them capable and comfortable acting as primary palliative care providers and improving providerrelated barriers to PC utilization, including misperception of palliative care as being synonymous with end-of-life care [37,38]. The above initiatives will help us address the gaps identified in our study and take

further steps toward optimal integration of palliative care into oncology care and the provision of truly comprehensive cancer care to our patients.

Conflict of interest statement The authors have no conflicts of interest to report.

References [1] Ferris FD, Bruera E, Cherny N, Cummings C, Currow D, Dudgeon D, et al. Palliative cancer care a decade later: accomplishments, the need, next steps — from the American Society of Clinical Oncology. J Clin Oncol 2009;27:3052–8. [2] National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology: Palliative Care (Version 2.2013); 2013. [3] Quill TE, Abernethy AP. Generalist plus specialist palliative care — creating a more sustainable model. N Engl J Med 2013;368:1173–5. [4] Temel JS, Greer JA, Muzikansky A, Gallagher ER, Admane S, Jackson VA, et al. Early palliative care for patients with metastatic non-small-cell lung cancer. N Engl J Med 2010;363:733–42. [5] Bakitas M, Lyons KD, Hegel MT, Balan S, Brokaw FC, Seville J, et al. Effects of a palliative care intervention on clinical outcomes in patients with advanced cancer: the Project ENABLE II randomized controlled trial. JAMA 2009;302:741–9. [6] Meyers FJ, Carducci M, Loscalzo MJ, Linder J, Greasby T, Beckett LA. Effects of a problem-solving intervention (COPE) on quality of life for patients with advanced cancer on clinical trials and their caregivers: simultaneous care educational intervention (SCEI): linking palliation and clinical trials. J Palliat Med 2011;14:465–73. [7] Ciemins EL, Blum L, Nunley M, Lasher A, Newman JM. The economic and clinical impact of an inpatient palliative care consultation service: a multifaceted approach. J Palliat Med 2007;10:1347–55. [8] Penrod JD, Deb P, Luhrs C, Dellenbaugh C, Zhu CW, Hochman T, et al. Cost and utilization outcomes of patients receiving hospital-based palliative care consultation. J Palliat Med 2006;9:855–60. [9] Morrison RS, Dietrich J, Ladwig S, Quill T, Sacco J, Tangeman J, et al. Palliative care consultation teams cut hospital costs for Medicaid beneficiaries. Health Aff 2011;30:454–63. [10] Morrison RS, Penrod JD, Cassel JB, Caust-Ellenbogen M, Litke A, Spragens L, et al. Cost savings associated with US hospital palliative care consultation programs. Arch Intern Med 2008;168:1783–90. [11] Keyser EA, Reed BG, Lowery WJ, Sundborg MJ, Winter III WE, Ward JA, et al. Hospice enrollment for terminally ill patients with gynecologic malignancies: impact on outcomes and interventions. Gynecol Oncol 2010;118:274–7. [12] Lewin SN, Buttin BM, Powell MA, Gibb RK, Rader JS, Mutch DG, et al. Resource utilization for ovarian cancer patients at the end of life: how much is too much? Gynecol Oncol 2005;99:261–6. [13] Lopez-Acevedo M, Havrilesky LJ, Broadwater G, Kamal AH, Abernethy AP, Berchuck A, et al. Timing of end-of-life care discussion with performance on end-of-life quality indicators in ovarian cancer. Gynecol Oncol 2013;130:156–61. [14] Doll KM, Stine JE, Van Le L, Moore DT, Bae-Jump V, Brewster WR, et al. Outpatient end of life discussions shorten hospital admissions in gynecologic oncology patients. Gynecol Oncol 2013;130:152–5.

C. Lefkowits et al. / Gynecologic Oncology 133 (2014) 319–325 [15] Fauci J, Schneider K, Walters C, Boone J, Whitworth J, Killian E, et al. The utilization of palliative care in gynecologic oncology patients near the end of life. Gynecol Oncol 2012;127:175–9. [16] Nevadunsky NS, Gordon S, Spoozak L, Van Arsdale A, Hou Y, Klobocista M, et al. The role and timing of palliative medicine consultation for women with gynecologic malignancies: association with end of life interventions and direct hospital costs. Gynecol Oncol 2013;131:3–7. [17] World Health Organization. National Cancer Control Programmes; 2013. [18] Cancer ACoSCo. Cancer Program Standards 2012: ensuring patient-centered care; 2012 [Chicago, IL]. [19] ABIM. Choosing wisely campaign: Society of Gynecologic Oncology's 5 things physicians and patients should question; 2013. [20] Glare PA, Semple D, Stabler SM, Saltz LB. Palliative care in the outpatient oncology setting: evaluation of a practical set of referral criteria. J Oncol Pract/Am Soc Clin Oncol 2011;7:366–70. [21] Glare P, Plakovic K, Schloms A, Egan B, Epstein AS, Kelsen D, et al. Study using the NCCN guidelines for palliative care to screen patients for palliative care needs and referral to palliative care specialists. J Natl Compr Canc Netw 2013;11:1087–96. [22] Smith TJ, Temin S, Alesi ER, Abernethy AP, Balboni TA, Basch EM, et al. American Society of Clinical Oncology provisional clinical opinion: the integration of palliative care into standard oncology care. J Clin Oncol 2012;30:880–7. [23] Nevadunsky NS, Gordon S, Spoozak L, Van Arsdale A, Hou Y, Klobocista M, et al. The role and timing of palliative medicine consultation for women with gynecologic malignancies: association with end of life interventions and direct hospital costs. Gynecol Oncol 2014;132:3–7. [24] Hui D, Kim SH, Kwon JH, Tanco KC, Zhang T, Kang JH, et al. Access to palliative care among patients treated at a comprehensive cancer center. Oncologist 2012;17:1574–80. [25] Burt J, Raine R. The effect of age on referral to and use of specialist palliative care services in adult cancer patients: a systematic review. Age Ageing 2006;35:469–76. [26] Rosenwax LK, McNamara BA. Who receives specialist palliative care in Western Australia—and who misses out. Palliat Med 2006;20:439–45. [27] Wentlandt K, Krzyzanowska MK, Swami N, Rodin GM, Le LW, Zimmermann C. Referral practices of oncologists to specialized palliative care. J Clin Oncol 2012;30:4380–6.

325

[28] Walshe C, Todd C, Caress A, Chew-Graham C. Patterns of access to community palliative care services: a literature review. J Pain Symptom Manage 2009;37:884–912. [29] Fadul N, Elsayem A, Palmer JL, Zhang T, Braiteh F, Bruera E. Predictors of access to palliative care services among patients who died at a Comprehensive Cancer Center. J Palliat Med 2007;10:1146–52. [30] Bernacki RE, Ko DN, Higgins P, Whitlock SN, Cullinan A, Wilson R, et al. Improving access to palliative care through an innovative quality improvement initiative: an opportunity for pay-for-performance. J Palliat Med 2012;15:192–9. [31] Nevadunsky NS, Spoozak L, Gordon S, Rivera E, Harris K, Goldberg GL. End-of-life care of women with gynecologic malignancies: a pilot study. Int J Gynecol Cancer 2013;23:546–52. [32] Gonsalves WI, Tashi T, Krishnamurthy J, Davies T, Ortman S, Thota R, et al. Effect of palliative care services on the aggressiveness of end-of-life care in the Veteran's Affairs cancer population. J Palliat Med 2011;14:1231–5. [33] Vergo MT, Cullinan AM. Joining together to improve outcomes: integrating specialty palliative care into the care of patients with cancer. J Natl Compr Canc Netw 2013;11(Suppl. 4):S38–46. [34] Lowery WJ, Lowery AW, Barnett JC, Lopez-Acevedo M, Lee PS, Secord AA, et al. Costeffectiveness of early palliative care intervention in recurrent platinum-resistant ovarian cancer. Gynecol Oncol 2013;130:426–30. [35] Ramchandran K, Von Roenn JH. Palliative care always. Oncology 2013;27:13–6 [27–30, 2–4 passim]. [36] Glare P, Virik K, Jones M, Hudson M, Eychmuller S, Simes J, et al. A systematic review of physicians' survival predictions in terminally ill cancer patients. BMJ 2003;327:195–8. [37] Fadul N, Elsayem A, Palmer JL, Del Fabbro E, Swint K, Li Z, et al. Supportive versus palliative care: what's in a name?: a survey of medical oncologists and midlevel providers at a comprehensive cancer center. Cancer 2009;115:2013–21. [38] Enguidanos S, Housen P, Goldstein R, Vesper E, Allen J, Braun W. Physician and nurse perceptions of a new inpatient palliative care consultation project: implications for education and training. J Palliat Med 2009;12:1137–42.