- Email: [email protected]
Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian cancer
YGYNO-977229; No. of pages: 7; 4C: Gynecologic Oncology xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian cancer Shayan Dioun a,⁎, Jennifer R. Jorgensen b, Eirwen M. Miller b, Joan Tymon-Rosario a, Xianhong Xie c, Xiaonan Xue c, Dennis Yi-Shin Kuo a,d, Nicole S. Nevadunsky a,d a
Department of Obstetrics & Gynecology and Women’s Health, Albert Einstein College of Medicine, Montefiore Medical Center, United States of America Division of Gynecologic Oncology, Department of Obstetrics & Gynecology and Women's Health, Albert Einstein College of Medicine, Montefiore Medical Center, United States of America c Department of Epidemiology & Population Health, Albert Einstein College of Medicine, United States of America d Albert Einstein Cancer Center, Albert Einstein College of Medicine, Bronx, NY, United States of America b
H I G H L I G H T S • Hospital readmission rates may be higher in diverse populations. • Body mass index and hypertension are predictors for readmission. • Readmission may be a predictor for overall survival and disease-free survival.
a r t i c l e
i n f o
Article history: Received 23 February 2018 Received in revised form 14 August 2018 Accepted 15 August 2018 Available online xxxx Keywords: Hospital readmission after chemotherapy Epithelial ovarian cancer Racially and ethnically diverse population
a b s t r a c t Objectives. Unplanned hospital admission following chemotherapy is a measure of quality cancer care. Large retrospective datasets have shown admission rates of 10–35% for women with ovarian cancer receiving chemotherapy. We sought to evaluate the prevalence and associated risk factors for hospital admission following chemotherapy in our racially diverse urban population. Methods. After IRB approval, clinicopathologic and treatment data were abstracted from all patients with newly diagnosed epithelial ovarian cancer who received chemotherapy at our institution from 2005 to 2016. Two-sided statistical analyses and Cox regression analysis were performed using Stata. Results. Of 217 evaluable patients, 87 (40%) had unplanned admissions following chemotherapy: adjuvant 64 (74%) and neoadjuvant 23(26%). Thirty (14%) had more than one admission. In total, there were 1314 days of hospitalization. The median readmission duration was 3 days. Body mass index and hypertension were predictive of readmission (p b 0.05). When comparing those readmitted more than once to those admitted once, both race and aspirin use were predictive of readmission (p b 0.05). Of those admitted more than once the self-identified race and ethnicity was 12 (40%) Hispanic, 8 (27%) White, 8 (27%) Black and 2 (7%) other. There was a significant difference in disease free (p = 0.01) and overall survival (p = 0.004) for patients with unplanned admission after chemotherapy as compared to those without admission. Conclusions. Readmission rates in our racially diverse patient population were higher than previously reported in the literature. Identifying patients at risk of readmission may play a role in chemotherapy decisionmaking, and resource allocation including patient care navigators. © 2018 Elsevier Inc. All rights reserved.
1. Background
⁎ Corresponding author at: Division of Gynecologic Oncology, Montefiore Medical Center, Albert Einstein College of Medicine, Department of Obstetrics, Gynecology and Women's Health, 3332 Rochambeau Ave, Bronx, New York 10467, United States of America. E-mail address: sdioun@montefiore.org (S. Dioun).
Chemotherapy has contributed to improved survival of ovarian cancer patients seen over the last several decades [1]. Following data supporting the benefit of platinum based chemotherapy agents in ovarian cancer, paclitaxel with cisplatin was adopted as the gold standard of chemotherapy treatment. Subsequently, it was found that carboplatin
https://doi.org/10.1016/j.ygyno.2018.08.021 0090-8258/© 2018 Elsevier Inc. All rights reserved.
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
2
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
in combination with paclitaxel had decreased side effects compared to cisplatin and paclitaxel, and is non-inferior [8–10]. Paclitaxel and Carboplatin have become the standard of care for epithelial ovarian cancers [2–4]. Studies show that patients often do not receive the treatments that demonstrate the best efficacy as recommended by the National Cancer Care Network (NCCN) Guidelines [1]. A potential cause for this is that while treatments may be efficacious in highly selected clinical trial participants, these findings may not be generalizable and the toxicities may be greater in clinical practice [5,6]. With the advent of the Affordable Care Act, the accountable care organization (ACO) introduced the concept of rewards based on savings. The ACO developed the oncology care model, which creates a discrete instance of care associated with chemotherapy and the six-months that follow. Providers receive a combination of fee-for-service payments, monthly payments for additional care and performance-based payments [11,12]. Unplanned hospitalizations are a performancebased measure that is used as a metric of quality care by the Oncology Care Model. While there are limited studies looking at unplanned hospitalization readmission rates in patients receiving chemotherapy, the literature suggests rates between 10 and 35% [12,13]. The purpose of our study was to describe toxicity of adjuvant chemotherapy for women with epithelial ovarian cancer as measured by unplanned hospital admission in an ethnically, racially and socioeconomically diverse urban population at our tertiary care medical center.
2. Methods After institutional review board approval, all patients with newly diagnosed primary epithelial ovarian cancer from 2005 to 2016 who received at least one cycle of chemotherapy were identified from our tumor registry. The cohort included both patients who received neoadjuvant chemotherapy and those who received primary debulking surgery followed by adjuvant chemotherapy. Clinicopathologic and treatment data were recorded from the electronic medical record and included: age, race, parity, menopausal status, body mass index, diabetes, hypertension, hyperlipidemia, aspirin use, hormone replacement, tobacco use, histology, stage, grade, optimal debulking status, date of chemotherapy cycles, type and dosing of chemotherapy, and date of hospitalizations. The Charlson comorbidity index score was also calculated for each patient [19]. During the study time period, three chemotherapy regimens were used in either the adjuvant or neoadjuvant setting at our institution. The first was carboplatin area under the curve (AUC) of 6 and paclitaxel 175 mg/m2 every 3 weeks. The second was dose dense chemotherapy consisting of carboplatin area under the curve (AUC) of 6 every 3 weeks and paclitaxel 80 mg/m2 weekly. Lastly, intravenous and intraperitoneal chemotherapy was administered as intravenous paclitaxel 135 mg/m2 on day 1, intravenous cisplatin 75–100 mg/m2 on day 2 and intraperitoneal paclitaxel 60 mg/m2 on day 8. At the time of chemotherapy, patients received palonosetron and dexamethasone for chemotherapy induced nausea and vomiting prophylaxis. Patients were prescribed prochlorperazine and ondansetron for at home management of nausea and vomiting. The primary outcome of the analysis was hospitalization for the management of a chemotherapy associated complication. Hospitalization was defined as an unplanned admission to one of the three hospitals affiliated with our institution within thirty days of day one of a chemotherapy cycle. The chemotherapy regimens prescribed to our patients had 21-day cycles. Admissions were attributed to cycles that happened in the antecedent 30 days, provided the next cycle had not been given on day 21. Readmissions were not counted twice. In instances where additional doses of chemotherapy were administered after day one of the cycle, hospitalization was still defined as within thirty days after day one of the cycle.
The secondary outcome of the study included risk factors associated with unplanned hospitalizations and reasons for admission. Definitions for each category of reason for readmission were as follows: neutropenic fever defined as fever with absolute neutrophil count less than 1500 neutrophils per microliter of blood. Hematologic toxicity included anemia with a hemoglobin less than 12.3 g/dL and thrombocytopenia with platelets less than 150 k/uL. Gastrointestinal toxicity included nausea and vomiting, abdominal pain, constipation, diarrhea, small bowel obstruction, lower tract bleeding, colostomy malfunction, and bowel perforation. Venous thromboembolism encompassed both deep vein thromboses and pulmonary emboli. Neurologic toxicities were defined as syncope, numbness and tingling and mental status change not related to electrolyte disturbances or infection. Cardiovascular included chest pain and atrial fibrillation. Poor nutrition was encompassed electrolyte disturbances and malnutrition. The remaining categories were infection with a known source, pleural effusion and other. Summary statistics were used to report the data. Two-sided statistical analyses were performed using Stata. A p value of b0.05 was considered statistically significant. Analyses were performed comparing cohorts of unplanned admission versus not admitted, as well as comparing patients who were not admitted, admitted one time, and admitted more than once during chemotherapy treatment. Univariable analyses were calculated with each clinical and pathologic factor. Multivariable logistic regression analysis was performed and included all variables with statistical significance (p b 0.05) on univariate analyses in addition to variables with clinical significance regardless of statistical significance. Kaplan Meier survival analysis was performed between cohorts of patients who were not admitted versus those who were admitted. 3. Results A total of 217 evaluable patients were identified. The mean age for the entire cohort was fifty-nine years (Table 1). The self-described race and ethnicity of the cohort was White (34%), Black (29%), Hispanic non-white (18%), and other (19%). The cohort included 50 (23%) patients who received neoadjuvant chemotherapy and 167 (77%) patients who received primary surgical debulking followed by adjuvant chemotherapy. For the adjuvant chemotherapy group, the median days between surgery and initial chemotherapy cycle was thirty-five days. Of those treated with adjuvant chemotherapy, 110 (66%) received IV carboplatin and paclitaxel every 3 weeks, 52 (31%) received IV/IP cisplatin and paclitaxel every 3 weeks and 5 (3%) received IV carboplatin and weekly paclitaxel. Forty-seven (94%) patients treated with neoadjuvant chemotherapy were given IV carboplatin and paclitaxel every 3 weeks, 2 (4%) were treated with IV carboplatin and weekly paclitaxel and 1 (2%) was given IV carboplatin and paclitaxel every 3 weeks followed by IV/IP cisplatin and paclitaxel after secondary debulking surgery. There was a total of 141 admissions in the entire cohort with a corresponding 1314 days of hospitalization. The median readmission duration was 3 days. Eighty-seven (40%) of the patients had an admission within thirty days of their chemotherapy dose: adjuvant 64 (74%) and neoadjuvant 23 (26%). Within the adjuvant group, 41 (34%) of the patients had an admission after adjuvant IV chemotherapy and 23 (46%) of the patients had an admission after adjuvant IV/IP chemotherapy. Cycle one of chemotherapy was the most common (41, 31%) cycle for hospitalizations. Cycle two through six of chemotherapy had roughly the same number of hospitalizations (range 16 to 22). The demographics for those hospitalized were: 27 White (31%), 28 Black (32%), 20 Hispanic non-white (23%) and 12 other (14%) (Table 1). When looking over multiple cycles of chemotherapy, 30 (14%) of the patients had an admission after more than one cycle. Of those admitted more than once, 12 (40%) were Hispanic, 8 (27%) were White, 8 (27%) were Black and 2 (7%) self-identified as other (Table 2). The most common reasons for readmission were for gastrointestinal complaints (32%), neutropenia (18%) and hematologic disturbances not
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
3
Table 1 Demographic and histopathological characteristics of study population. Variable
Entire cohort (N = 217)
Readmitted (N = 87)
No (N = 130)
P-value
Age, mean (SD) Age, n (%) b50 50–59 60–69 70–79 ≥80 Race, n (%) White Black Hispanic Other Parity, n (%) Nulliparity Multiparity Menopausal status, n (%) Yes No BMI, n (%) ≤20 20–25 25–30 N30 Diabetes, n (%) Yes No Hypertension, n (%) Yes No Hyperlipidemia, n (%) Yes No Aspirin, n (%) Yes No Hormone replacement, n (%) Yes No Tobacco, n (%) Yes No Histology, n (%) Serous Endometrioid Other Stage, n (%) 1 2 3 4 Grade, n (%) 1 2 3 Not reported Management, n (%) Primary surgery Neoadjuvant Optimally debulked, n (%) Yes No Charlson comorbidity, median (IQR) Excluding age & cancer Excluding cancer Including age & cancer Type of chemotherapy, n (%) Adjuvant IV/IP cisplatin & paclitaxel Adjuvant IV carboplatin & paclitaxel every 3 weeks Adjuvant IV carboplatin & weekly paclitaxel Neoadjuvant IV carboplatin & paclitaxel every 3 weeks Neoadjuvant IV carboplatin & weekly paclitaxel
59.2 (12.6)
60.2 (11.8)
58.6 (13.2)
0.37 0.59
53 (24) 62 (29) 51 (24) 39 (18) 12 (6)
19 (22) 24 (28) 24 (28) 17 (20) 3 (3)
34 (26) 38 (29) 27 (21) 22 (17) 9 (7)
73 (34) 63 (29) 40 (18) 41 (19)
27 (31) 28 (32) 20 (23) 12 (14)
46 (35) 35 (27) 20 (15) 29 (22)
52 (24) 163 (76)
21 (24) 66 (76)
31 (24) 97 (76)
153 (73) 58 (27)
60 (71) 24 (29)
93 (73) 34 (27)
13 (6) 57 (27) 65 (30) 79 (37)
2 (2) 18 (21) 25 (30) 39 (46)
11 (8) 39 (30) 40 (31) 40 (31)
49 (23) 166 (77)
23 (27) 63 (73)
26 (20) 103 (80)
117 (54) 100 (46)
59 (68) 28 (32)
58 (45) 72 (55)
87 (40) 129 (60)
40 (46) 47 (54)
47 (36) 82 (64)
30 (14) 186 (86)
11 (13) 75 (87)
19 (15) 111 (85)
6 (4) 160 (96)
2 (3) 66 (97)
4 (4) 94 (96)
62 (29) 151 (71)
23 (27) 61 (73)
39 (30) 90 (70)
123 (61) 22 (11) 58 (29)
55 (70) 5 (6) 19 (24)
68 (55) 17 (14) 39 (31)
51 (24) 25 (12) 115 (53) 26 (12)
13 (15) 10 (11) 55 (63) 9 (10)
38 (29) 15 (12) 60 (46) 17 (13)
14 (6) 19 (9) 170 (78) 14 (6)
4 (5) 5 (6) 72 (83) 6 (7)
10 (8) 14 (11) 98 (75) 8 (6)
167 (77) 50 (23)
64 (74) 23 (26)
103 (79) 27 (21)
143 (86) 23 (14)
56 (89) 7 (11)
87 (84) 16 (16)
0 (0–1) 2 (1–3) 7 (5–9)
0 (0–1) 2 (1–3) 8 (6–9)
0 (0–1) 2 (1–3) 7 (4–9)
52 (24) 110 (51) 5 (2) 47 (22) 2 (1)
24 (28) 39 (45) 1 (1) 21 (24) 2 (2)
28 (22) 71 (55) 4 (3) 26 (20) 0 (0)
0.22
0.99
0.77
0.048
0.26
0.001
0.16
0.70
1.00
0.65
0.08
0.052
0.44
0.33
0.42
0.20 0.45 0.051 0.21
Note: menopausal status = yes corresponds to post-menopausal; tobacco = yes corresponds to ever smoked.
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
4
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
Table 2 Demographic and histopathological characteristics of patients readmitted once, those readmitted more than once and those not readmitted Variable
=1 readmitted (N = 57)
N1 readmission (N = 30)
No (N = 130)
P-value 1
P-value 2
Age, mean (SD) Age, n (%) b50 50–59 60–69 70–79 ≥80 Race, n (%) White Black Hispanic Other Parity, n (%) Nullip Multip Menopausal status, n (%) Yes No BMI, n (%) ≤20 20–25 25–30 N30 Diabetes, n (%) Yes No Hypertension, n (%) Yes No Hyperlipidemia, n (%) Yes No Aspirin, n (%) Yes No Hormone replacement, n (%) Yes No Tobacco, n (%) Yes No Histology, n (%) Serous Endometrioid Other Stage, n (%) 1 2 3 4 Grade, n (%) 1 2 3 Not reported Management, n (%) Primary surgery Neoadjuvant Optimally debulked, n (%) Yes No Charlson comorbidity, median (IQR) Excluding age & cancer Excluding cancer Including age & cancer Type of chemotherapy, n (%) Adjuvant IV/IP cisplatin & paclitaxel Adjuvant IV carboplatin & paclitaxel every 3 weeks Adjuvant IV carboplatin & weekly paclitaxel Neoadjuvant IV carboplatin & paclitaxel every 3 weeks Neoadjuvant IV carboplatin & weekly paclitaxel
60.9 (12.1)
58.8 (11.2)
58.6 (13.2)
0.51 0.79
0.43 0.67
12 (21) 15 (26) 17 (30) 10 (18) 3 (5)
7 (23) 9 (30) 7 (23) 7 (23) 0 (0)
34 (26) 38 (29) 27 (21) 22 (17) 9 (7) 0.03
0.04
19 (33) 20 (35) 8 (14) 10 (18)
8 (27) 8 (27) 12 (40) 2 (7)
46 (35) 35 (27) 20 (15) 29 (22) 0.92
0.69
13 (23) 44 (77)
8 (27) 22 (73)
31 (24) 97 (76) 0.77
0.51
42 (74) 15 (26)
18 (67) 9 (33)
93 (73) 34 (27) 0.13
0.46
1 (2) 14 (26) 14 (26) 25 (46)
1 (3) 4 (13) 11 (37) 14 (47)
11 (8) 39 (30) 40 (31) 40 (31) 0.45
0.60
16 (29) 40 (71)
7 (23) 23 (77)
26 (20) 103 (80) 0.003
0.52
40 (70) 17 (30)
19 (63) 11 (37)
58 (45) 72 (55) 0.16
0.21
29 (51) 28 (49)
11 (37) 19 (63)
47 (36) 82 (64) 0.02
0.01
3 (5) 53 (95)
8 (27) 22 (73)
19 (15) 111 (85) 0.86
0.55
2 (4) 43 (96)
0 (0) 23 (100)
4 (4) 94 (96) 0.78
0.59
14 (25) 41 (75)
9 (31) 20 (69)
39 (30) 90 (70) 0.24
0.79
35 (67) 4 (8) 13 (25)
20 (74) 1 (4) 6 (22)
68 (55) 17 (14) 39 (31) 0.08
0.28
10 (18) 5 (9) 38 (67) 4 (7)
3 (10) 5 (17) 17 (57) 5 (17)
38 (29) 15 (12) 60 (46) 17 (13) 0.87
0.96
3 (5) 4 (7) 46 (81) 4 (7)
1 (3) 1 (3) 26 (87) 2 (7)
10 (8) 14 (11) 98 (75) 8 (6) 0.62
0.97
42 (74) 15 (26)
22 (73) 8 (27)
103 (79) 27 (21) 0.36
0.23
38 (93) 3 (7)
18 (82) 4 (18)
87 (84) 16 (16)
0 (0–1) 2 (1–3) 8 (6–9)
1 (0–1) 3 (1–4) 8 (6–10)
0 (0–1) 2 (1–3) 7 (4–9)
0.12 0.64 0.12 0.41
0.11 0.55 0.50 0.83
17 (30) 24 (42) 1 (2) 13 (23) 2 (4)
7 (23) 15 (50) 0 (0) 8 (27) 0 (0)
28 (22) 71 (55) 4 (3) 26 (20) 0 (0)
Note: P-value 1 was testing the overall difference across all 3 groups; p-value 2 was testing the difference between =1 readmitted and N1 readmitted.
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
related to neutropenia (15%) (Table 3). Twenty-nine (64%) of forty-five patients admitted for gastrointestinal complaints had nausea and vomiting; other gastrointestinal causes for admission were constipation (4%), diarrhea (7%), abdominal pain (9%) and other (16%). The number of days from chemotherapy administration to readmission for nausea and vomiting ranged from 1 to 29 days. The median number of days was 6; 55% (16/29) readmissions for nausea and vomiting occurred within 1–6 days of chemotherapy. When comparing all reasons for readmission for patients receiving neoadjuvant chemotherapy to those undergoing adjuvant chemotherapy (all regimens combined), there was a statistically significant difference. Thirty-three patients (15%) received granulocyte colony stimulating factors during their chemotherapy course: 9 in the adjuvant IV cohort, 11 in the adjuvant IV/IP cohort and 13 in the neoadjuvant cohort. There was a significant difference between those with unplanned admissions and those without in body mass index (BMI, p = 0.048) and hypertension (p = 0.001) (Table 1) on univariable analysis. When the entire cohort was separated into those with one readmission, those with more than one readmission and those without any readmissions (Table 2), there was a significant difference between all three groups in race (p = 0.03), hypertension (p = 0.003) and aspirin use (p = 0.02). There was also a significant difference between those with one readmission and those with more than one readmission in race (p = 0.04) and aspirin use (p = 0.01). Although aspirin use was included in our data because of reports suggesting an association between aspirin use and ovarian cancer risk and mortality [21], testing for collinearity revealed aspirin use was highly associated with hypertension (p = 0.0005); therefore, aspirin use was excluded from the logistic regression model. There was no significant difference in the readmission rates when comparing different adjuvant chemotherapy regimens and neoadjuvant chemotherapy. The logistic regression model demonstrated that BMI greater than thirty was associated with increased odds of readmission of 3.32 (95% CI: 1.14–9.65, p = 0.03) (Table 4). There was an increased odds for readmission for patients with hypertension 3.72 (95% CI: 1.58–8.79, p = 0.003). Additional logistic regression modeling grouping all neoadjuvant chemotherapy regimens versus all regimens of primary debulking surgery plus adjuvant therapy showed increased odds of readmission for patients with BMI greater than thirty and hypertension (Appendix 1). Kaplan-Meier estimates for the entire cohort stratified by unplanned hospital admission showed a significant difference in overall survival (p = 0.004) and progression-free survival (p = 0.01) between those with no hospitalizations, those with one hospitalization and those with more than one hospitalization (Fig. 1, Fig. 2). 4. Discussion Unplanned hospital admission rates in our ethnically and racially diverse population of women with epithelial ovarian cancer were higher
5
Table 4 Logistic regression model for readmission. Variable Race White (ref) Black Hispanic Other Age b50 (ref) 50–59 60–69 70–79 ≥80 BMI ≥20 20–25 (ref) 25–30 N30 Hypertension No (ref) Yes Histology Serous (ref) Endometrioid Other Stage 1 (ref) 2 3 4 Tobacco No (ref) Yes Type of chemotherapy Adjuvant IV/IP cisplatin & paclitaxel Adjuvant carboplatin and paclitaxel every 3 weeks (ref) Adjuvant carboplatin and weekly paclitaxel Neoadjuvant chemotherapy Optimally debulked No (ref) Yes
Odds ratio
95% CI
P value
1 1.29 0.95 0.77
0.48–3.47 0.33–2.79 0.27–2.22
0.61 0.93 0.63
1 1.43 1.19 1.18 1.10
0.51–4.00 0.37–3.85 0.34–4.11 0.12–9.93
0.50 0.77 0.80 0.94
1.54 1 2.33 3.32
0.22–10.90 0.66 0.80–6.84 1.14–9.65
0.12 0.03
1 3.72
1.58–8.79
0.003
1 0.34 0.45
0.09–1.29 0.18–1.15
0.11 0.10
1 1.63 3.99 0.46
0.45–5.91 0.46 1.25–12.78 0.02 0.08–2.62 0.38
1 1.09
0.46–2.59
0.85
0.50 1
0.19–1.33
0.16
0.17 0.73
0.01–2.51 0.20 0.02–26.32 0.86
1 1.34
0.34–5.25
0.68
The odds ratio with 95% CI for neoadjuvant vs primary surgery was 1.65 (0.04–60.58), pvalue = 0.79.
PFS: No readmission 1 readmission >1 readmission
Table 3 Reason for unplanned hospital admission for patients receiving chemotherapy for epithelial ovarian cancer
Neutropenic fever Gastrointestinal toxicity Poor nutrition Hematologic toxicity Venous thromboembolism Infection Neurologic toxicity Cardiovascular toxicity Pleural effusion Other
Entire cohort (%)
Neoadjuvant (%)
Adjuvant: IV (%)
Adjuvant IV/IP (%)
25 (18) 45 (32) 11 (11) 23 (15) 5 (4)
5 (13) 14 (35) 0 (0) 5 (13) 0 (0)
11(15) 18 (25) 9 (13) 18 (25) 1 (1)
9 (30 13 (43) 2 (2) 0 (0) 4 (13)
12 (9) 5 (4) 3 (2) 4 (3) 8 (4)
6 (15) 3 (8) 1 (3) 2 (5) 4 (10)
5 (7) 2 (3) 2 (3) 1 (1) 3 (4)
1 (3) 0 (0) 0 (0) 1 (3) 0 (0)
P value comparing all reasons for readmission for neoadjuvant vs combined adjuvant chemotherapy = 0.03.
Fig. 1. Kaplan-Meier curves for PFS vs readmission. Log-rank test p-value = 0.01 (overall comparison).
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
6
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
OS: No readmission 1 readmission >1 readmission
Fig. 2. Kaplan-Meier curves for OS vs readmission. Log-rank test p-value = 0.004 (overall comparison).
(40%) than the previously reported rates in the literature. Hypertension and obesity were predictive of a single unplanned readmission. Race was associated with multiple unplanned readmissions. Poorer diseasefree survival and overall survival estimates in women with unplanned hospital readmissions were seen. Identification of obesity and hypertension as a risk factor may help identify patients who would benefit from care navigators and outpatient resources to identify preventable problems before admission. Patient care navigators could aid patients in coordinating appointments with primary care physicians as well as their oncologists to optimize treatment of comorbidities; patients with other cancer types have significantly fewer problems with care coordination when given support from patient care navigators [22]. We also identified nausea and vomiting as the most common reason for readmission. From the data available in this retrospective review, it is unclear why nausea and vomiting were the most common reasons for readmission. A quality improvement initiative prospectively examining anti-emetic prophylaxis at time of chemotherapy administration and anti-emetic use in the outpatient setting is warranted to further characterize this finding. The addition of a neurokinin-1 receptor antagonist to the pre-medication regimen may lead to reduced admissions for nausea and vomiting in our population with delayed phase chemotherapy induced nausea and vomiting. In a systematic review, dos Santos et al. demonstrated that the addition of a neurokinin-1 receptor antagonist to the standard regimen of a 5-HT3 receptor antagonist and glucocorticoid significantly improved effectiveness in both eliminating or reducing nausea in the acute and delayed phases of chemotherapy induced nausea and vomiting as compared to the standard regimen [23]. Available retrospective studies in the literature looking at hospital readmission rates after initiating treatment for ovarian cancer including surgery and chemotherapy suggest rates of readmission from 10 to 35% [12,13]. These studies largely focused on postoperative readmissions. Clark et al. found a readmission rate of 12% following surgical cytoreduction in their patients in Boston with stages II–IV ovarian cancer. Perioperative complications placed the patients at highest risk while age, medical comorbidities, surgical approach did not affect the likelihood of readmission [15]. In another study, Fauci et al demonstrated a hospital readmission rate of 16% within 30 days of primary surgery in Alabama [16]. Eskander et al looked at readmission rates using the SEER database and found a rate of 19.5% readmission rate. The increased readmission rates may be partially secondary to the limitation of the database, which only includes patients 65 years or older [14].
Duska et al, focused on both surgery and chemotherapy and demonstrated a readmission rate of 12% after primary debulking and adjuvant chemotherapy with body mass index, residual tumor after debulking and surgical stage being predictive of readmission [17, 18]. Similarly, we found that body mass index was a risk factor for readmission. Duska et al also suggested that patients with advanced disease may have lower readmission rates if they had undergone neoadjuvant chemotherapy prior to debulking. The unplanned admission rate in the cohort of patients receiving intraperitoneal therapy reported by Wright et al. most closely approximates the unplanned admission rate identified in our study (35%) [20]. The dose dense and standard intravenous cohorts from that study at unplanned admission rates of 25% and 21% respectively. Patients with breast cancer are often treated with taxane containing chemotherapy regimens; rates of hospital admission related to chemotherapy in breast cancer patients are reported as 8–13%. These rates are lower than the readmission rate that our study found [24,25]. Our study showed a difference in readmission rates as compared with previously reported literature (10–35%). While there has been much focus on hospital readmission after primary debulking and adjuvant chemotherapy there has been less on readmission after neoadjuvant chemotherapy. It has been suggested that patients at higher risk of readmission after primary surgery and chemotherapy may benefit from neoadjuvant chemotherapy instead [18]. Our study demonstrates no significant difference in readmission rates after neoadjuvant chemotherapy and primary debulking surgery followed by adjuvant chemotherapy. We found a statistically significant difference when combining reasons for readmission when comparing neoadjuvant chemotherapy to all regimens of adjuvant chemotherapy; however, the number of patients in each group was too small to further delineate which specific causes for readmission were significantly different. This would suggest that future approaches to reduce readmission would be applicable in neoadjuvant patients as well. Unique to our study, we examined the progression free and overall survival rates from the prospective of unplanned hospital readmissions. Prior studies have focused on treatment regimens (primary debulking versus neoadjuvant, IV versus IV/IP adjuvant chemotherapy) when reporting these outcomes. When combining all treatment regimens, we found a significant difference in both progression free and overall survival rates for those with unplanned hospital readmissions. Patients with more than one hospital readmission have the worst survival outcomes in our study population. Effects of unplanned hospital readmission rates on both progression free and overall survival warrants further research in patients treated for ovarian cancer. In contrast to the previously reported studies on readmission, our study population included an ethnically diverse population in one of the poorest urban counties in America. Previous studies have focused on a more racially homogenous and economically affluent population than our study population with rates of white patients between 75 and 90% while in our study the percentage of white patients was 34%. Reasons for unplanned readmissions (gastrointestinal 32%, neutropenic fever 18%, and hematological 15%) also differed in our study population as compared to the reasons described by Wright et al (infection 15%, gastrointestinal 13%, and electrolyte disorders 12%) [13]. These differences may have been related to variation in use and access to supportive therapies such as colony stimulating factor and anti-emetics [13]. The limitations of this study include the possibility of underestimating the total number of unplanned hospitalizations since patients may have been hospitalized at other hospitals not affiliated with our institution. The monetary consequences and impact on quality of life of over one thousand days of hospital admission were beyond the scope of this study. There is also the possibility of under-capturing toxicity by selecting only one reason for admission. These complications that we report may not have been caused by the chemotherapy administered and instead caused by other factors, such as disease progression or surgery. While planned treatment regimens were included in the analysis, the lack of accommodation for variations between actual doses prescribed
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
and delivered as well as dose reductions is a limitation. In comparison of the unplanned hospitalizations for our patients as compared to previous studies we did not evaluate compulsory use of granulocyte colony stimulating factors or use of these factors in subsequent cycles after unplanned admission. Our data suggests that unplanned hospitalizations in our racially diverse patient population are higher than those reported in the literature. Unplanned hospitalization may be a predictor for overall survival and disease-free survival. Identifying patients that are more likely to be readmitted may play a role in chemotherapy decision-making, and resource allocation including care navigators. Going forward, further studies to stratify patients to optimize adjuvant chemotherapy for women with ovarian cancer are needed. Supplementary data to this article can be found online at https://doi. org/10.1016/j.ygyno.2018.08.021. Conflict of interest statement None of the authors have any conflicts of interest to disclose
Author contribution Shayan Dioun: Study conception and design, data collection, analysis and interpretation, manuscript writing. Jennifer R. Jorgensen: Analysis and interpretation, manuscript writing, critical revision. Eirwen M. Miller: Data collection, critical revision. Joan Tymon-Rosario: Data collection and critical revision. Xianhong Xie: Analysis and interpretation of data, critical revision. Xiaonan Xue: Analysis and interpretation of data, critical revision. Dennis Yi-Shin Kuo: Study conception and design, critical revision. Nicole S. Nevadunsky: Study conception and design, analysis and interpretation, manuscript writing, critical revision. References [1] J.D. Wright, L. Chen, A.I. Tergas, S. Patankar, W.M. Burke, J.Y. Hou, et al., Trends in relative survival for ovarian cancer from 1975 to 2011, Obstet. Gynecol. 125 (2015) 1345–1352. [2] G.A. Omura, B.N. Bundy, J.S. Berek, S. Curry, G. Delgado, R. Mortel, Randomized trial of cyclophosphamide plus cisplatin with or without doxorubicin in ovarian carcinoma: a Gyneco- logic Oncology Group Study, J. Clin. Oncol. 7 (1989) 457–465. [3] W.P. McGuire, W.J. Hoskins, M.F. Brady, P.R. Kucera, E.E. Partridge, K.Y. Look, et al., Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer, N. Engl. J. Med. 334 (1996) 1–6. [4] M.J. Piccart, K. Bertelsen, K. James, J. Cassidy, C. Mangioni, E. Simonsen, et al., Randomized intergroup trial of cisplatin- paclitaxel versus cisplatin-cyclophosphamide in women with advanced epithelial ovarian cancer: three-year results, J. Natl. Cancer Inst. 92 (2000) 699–708. [5] E.A. McGlynn, S.M. Asch, J. Adams, J. Keesey, J. Hicks, A. DeCristofaro, et al., The quality of health care delivered to adults in the United States, N. Engl. J. Med. 348 (2003) 2635–2645. [6] J.M. Unger, W.E. Barlow, D.P. Martin, S.D. Ramsey, M. Leblanc, R. Etzioni, et al., Comparison of survival outcomes among can- cer patients treated in and out of clinical trials, J. Natl. Cancer Inst. 106 (2014), dju002.
7
[8] J.P. Neijt, S.A. Engelholm, M.K. Tuxen, et al., Exploratory Phase III study of paclitaxel and cisplatin vs paclitaxel and carboplatin in advanced ovarian cancer, J. Clin. Oncol. 18 (2000) 3084–3092. [9] A. du Bois, H. Lück, W. Meier, et al., A randomized clinical trial of cisplatin/paclitaxel vs carboplatin/paclitaxel as first-line treatment of ovarian cancer, J. Natl. Cancer Inst. 95 (2000) 1320–1329. [10] N.K. Aaronson, S. Ahmedzai, B. Bergman, et al., The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology, J. Natl. Cancer Inst. 85 (1993) 365–376. [11] Jeffrey Clough, D. Kamal, H. Arif, Oncology care model: short- and long-term considerations in the context of broader payment reform, J. Oncol. Pract. 11 (4) (2015) 319–321. [12] Blase Polite, N. Miller, D. Harold, Medicare innovation center oncology care model, J. Oncol. Pract. 11 (2) (2015) 117–119. [13] A.A. Wright, A. Cronin, D.E. Milne, M.A. Bookman, R.A. Burger, D.E. Cohn, et al., Use and effectiveness of intraperitoneal chemotherapy for treatment of ovarian cancer, J. Clin. Oncol. 33 (2015) 2841–2847. [14] P. Basch, Quality of health care delivered to adults in the United States, N. Engl. J. Med. 349 (19) (2003) 1866–1868. [15] R.M. Clark, W.B. Growdon, A. Wiechert, D. Boruta, M. Del Carmen, A.K. Goodman, et al., Patient, treatment and discharge factors associated with hospital readmission within 30 days after surgical cytoreduction for epithelial ovarian carcinoma, Gynecol. Oncol. 130 (3) (2013) 407–410. [16] J.M. Fauci, K.E. Schneider, P.J. Frederick, G. Wilding, J. Consiglio, A.L. Sutton, et al., Assessment of risk factors for 30-day hospital readmission after surgical cytoreduction in epithelial ovarian carcinoma, Int. J. Gynecol. Cancer 21 (5) (2011) 806–810. [17] R.N. Skander, J. Chang, A. Ziogas, H. Anton-Culver, R.E. Bristow, Evaluation of 30-day hospital readmission after surgery for advanced-stage ovarian cancer in a Medicare population, J. Clin. Oncol. 32 (36) (2014) 4113–4119. [18] L.R. Duska, J.J. Java, D.E. Cohen, R.A. Burger, Risk factors for readmission in patients with ovarian, fallopian tube, and primary peritoneal carcinoma who are receiving front- line chemotherapy on a clinical trial (GOG 218): an NRG oncology/gynecologic oncology group study (ADS-1236), Gynecol. Oncol. 139 (2) (2015) 221–227. [19] M.E. Charlson, P. Pompei, K.L. Ales, C.R. MacKenzie, A new method of classifying prognostic comorbidity in longitudinal studies: development and validation, J. Chronic Dis. 40 (1987) 373–383. [20] J.D. Wright, J.Y. Hou, W.M. Burke, A. Tergas, L. Chen, J. Hu, C. Ananth, A. Neuget, D. Hershman, Utilization and toxicity of alternative delivery methods of adjuvant chemotherapy for ovarian cancer, Obstet. Gynecol. 127 (6) (2016) 985–991. [21] B. Trabert, R.B. Ness, W.H. Lo-Ciganic, M.A. Murphy, E.L. Goode, E.M. Poole, L.A. Brinton, P.M. Webb, C.M. Nagle, S.J. Jordan, Australian Ovarian Cancer Study Group, Australian Cancer Study (Ovarian Cancer), H.A. Risch, M.A. Rossing, J.A. Doherty, M.T. Goodman, G. Lurie, S.K. Kjaer, E. Hogdall, A. Jensen, D.W. Cramer, K.L. Terry, A. Vitonis, E.V. Bandera, S. Olson, M.G. King, U. Chandran, H. Anton-Culver, A. Ziogas, U. Menon, S.A. Gayther, S.J. Ramus, A. Gentry-Maharaj, A.H. Wu, C.L. Pearce, M.C. Pike, A. Berchuck, J.M. Schildkraut, N. Wentzensen, Ovarian Cancer Association Consortium, Aspirin, nonaspirin nonsteroidal anti-inflammatory drug, and acetaminophen use and risk of invasive epithelial ovarian cancer: a pooled analysis of the Ovarian Cancer Consortium, J. Natl. Cancer Inst. 106 (2) (2014), djt431. [22] E.H. Wagner, E.J. Ludman, E.J. Aiello Bowles, R. Penfold, R.J. Reid, C.M. Rutter, J. Chubak, R. McCorkle, Nurse navigators in early cancer care: a randomized, controlled trial, J. Clin. Oncol. 32 (1) (2014) 12–18. [23] L.V. dos Santos, F.H. Souza, A.T. Brunetto, A.D. Sasse, J.P. da Silveira Nogueira Lima, Neuorkinin-1 receptor antagonists for chemotherapy-induced nausea and vomiting: a systematic review, J. Natl. Cancer Inst. 104 (17) (2012) 1280. [24] N.M. Pittman, W.M. Hopman, M. Mates, Emergency room visits and hospital admission rates after curative chemotherapy for breast cancr, J. Oncol. Pract. 11 (2) (2015) 120–125. [25] J.M. Beana-Canada, S. Estalella-Mendoza, P. Rosado-Varela, I. Exposito-Alvarez, M. Gonzalez-Guerrero, M.C. Diaz-Blanco, C. Cortes-Carmona, P. Ramirez-Daffos, E. Arriola-Arellano, A. Reuda-Ramos, L. Solana-Grimaldi, E. Benitez-Rodriguez, Use of health-care services during chemotherapy for breast cancer, Eur. J. Cancer 48 (18) (2012) 3328–3334.
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian cancer Shayan Dioun a,⁎, Jennifer R. Jorgensen b, Eirwen M. Miller b, Joan Tymon-Rosario a, Xianhong Xie c, Xiaonan Xue c, Dennis Yi-Shin Kuo a,d, Nicole S. Nevadunsky a,d a
Department of Obstetrics & Gynecology and Women’s Health, Albert Einstein College of Medicine, Montefiore Medical Center, United States of America Division of Gynecologic Oncology, Department of Obstetrics & Gynecology and Women's Health, Albert Einstein College of Medicine, Montefiore Medical Center, United States of America c Department of Epidemiology & Population Health, Albert Einstein College of Medicine, United States of America d Albert Einstein Cancer Center, Albert Einstein College of Medicine, Bronx, NY, United States of America b
H I G H L I G H T S • Hospital readmission rates may be higher in diverse populations. • Body mass index and hypertension are predictors for readmission. • Readmission may be a predictor for overall survival and disease-free survival.
a r t i c l e
i n f o
Article history: Received 23 February 2018 Received in revised form 14 August 2018 Accepted 15 August 2018 Available online xxxx Keywords: Hospital readmission after chemotherapy Epithelial ovarian cancer Racially and ethnically diverse population
a b s t r a c t Objectives. Unplanned hospital admission following chemotherapy is a measure of quality cancer care. Large retrospective datasets have shown admission rates of 10–35% for women with ovarian cancer receiving chemotherapy. We sought to evaluate the prevalence and associated risk factors for hospital admission following chemotherapy in our racially diverse urban population. Methods. After IRB approval, clinicopathologic and treatment data were abstracted from all patients with newly diagnosed epithelial ovarian cancer who received chemotherapy at our institution from 2005 to 2016. Two-sided statistical analyses and Cox regression analysis were performed using Stata. Results. Of 217 evaluable patients, 87 (40%) had unplanned admissions following chemotherapy: adjuvant 64 (74%) and neoadjuvant 23(26%). Thirty (14%) had more than one admission. In total, there were 1314 days of hospitalization. The median readmission duration was 3 days. Body mass index and hypertension were predictive of readmission (p b 0.05). When comparing those readmitted more than once to those admitted once, both race and aspirin use were predictive of readmission (p b 0.05). Of those admitted more than once the self-identified race and ethnicity was 12 (40%) Hispanic, 8 (27%) White, 8 (27%) Black and 2 (7%) other. There was a significant difference in disease free (p = 0.01) and overall survival (p = 0.004) for patients with unplanned admission after chemotherapy as compared to those without admission. Conclusions. Readmission rates in our racially diverse patient population were higher than previously reported in the literature. Identifying patients at risk of readmission may play a role in chemotherapy decisionmaking, and resource allocation including patient care navigators. © 2018 Elsevier Inc. All rights reserved.
1. Background
⁎ Corresponding author at: Division of Gynecologic Oncology, Montefiore Medical Center, Albert Einstein College of Medicine, Department of Obstetrics, Gynecology and Women's Health, 3332 Rochambeau Ave, Bronx, New York 10467, United States of America. E-mail address: sdioun@montefiore.org (S. Dioun).
Chemotherapy has contributed to improved survival of ovarian cancer patients seen over the last several decades [1]. Following data supporting the benefit of platinum based chemotherapy agents in ovarian cancer, paclitaxel with cisplatin was adopted as the gold standard of chemotherapy treatment. Subsequently, it was found that carboplatin
https://doi.org/10.1016/j.ygyno.2018.08.021 0090-8258/© 2018 Elsevier Inc. All rights reserved.
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
2
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
in combination with paclitaxel had decreased side effects compared to cisplatin and paclitaxel, and is non-inferior [8–10]. Paclitaxel and Carboplatin have become the standard of care for epithelial ovarian cancers [2–4]. Studies show that patients often do not receive the treatments that demonstrate the best efficacy as recommended by the National Cancer Care Network (NCCN) Guidelines [1]. A potential cause for this is that while treatments may be efficacious in highly selected clinical trial participants, these findings may not be generalizable and the toxicities may be greater in clinical practice [5,6]. With the advent of the Affordable Care Act, the accountable care organization (ACO) introduced the concept of rewards based on savings. The ACO developed the oncology care model, which creates a discrete instance of care associated with chemotherapy and the six-months that follow. Providers receive a combination of fee-for-service payments, monthly payments for additional care and performance-based payments [11,12]. Unplanned hospitalizations are a performancebased measure that is used as a metric of quality care by the Oncology Care Model. While there are limited studies looking at unplanned hospitalization readmission rates in patients receiving chemotherapy, the literature suggests rates between 10 and 35% [12,13]. The purpose of our study was to describe toxicity of adjuvant chemotherapy for women with epithelial ovarian cancer as measured by unplanned hospital admission in an ethnically, racially and socioeconomically diverse urban population at our tertiary care medical center.
2. Methods After institutional review board approval, all patients with newly diagnosed primary epithelial ovarian cancer from 2005 to 2016 who received at least one cycle of chemotherapy were identified from our tumor registry. The cohort included both patients who received neoadjuvant chemotherapy and those who received primary debulking surgery followed by adjuvant chemotherapy. Clinicopathologic and treatment data were recorded from the electronic medical record and included: age, race, parity, menopausal status, body mass index, diabetes, hypertension, hyperlipidemia, aspirin use, hormone replacement, tobacco use, histology, stage, grade, optimal debulking status, date of chemotherapy cycles, type and dosing of chemotherapy, and date of hospitalizations. The Charlson comorbidity index score was also calculated for each patient [19]. During the study time period, three chemotherapy regimens were used in either the adjuvant or neoadjuvant setting at our institution. The first was carboplatin area under the curve (AUC) of 6 and paclitaxel 175 mg/m2 every 3 weeks. The second was dose dense chemotherapy consisting of carboplatin area under the curve (AUC) of 6 every 3 weeks and paclitaxel 80 mg/m2 weekly. Lastly, intravenous and intraperitoneal chemotherapy was administered as intravenous paclitaxel 135 mg/m2 on day 1, intravenous cisplatin 75–100 mg/m2 on day 2 and intraperitoneal paclitaxel 60 mg/m2 on day 8. At the time of chemotherapy, patients received palonosetron and dexamethasone for chemotherapy induced nausea and vomiting prophylaxis. Patients were prescribed prochlorperazine and ondansetron for at home management of nausea and vomiting. The primary outcome of the analysis was hospitalization for the management of a chemotherapy associated complication. Hospitalization was defined as an unplanned admission to one of the three hospitals affiliated with our institution within thirty days of day one of a chemotherapy cycle. The chemotherapy regimens prescribed to our patients had 21-day cycles. Admissions were attributed to cycles that happened in the antecedent 30 days, provided the next cycle had not been given on day 21. Readmissions were not counted twice. In instances where additional doses of chemotherapy were administered after day one of the cycle, hospitalization was still defined as within thirty days after day one of the cycle.
The secondary outcome of the study included risk factors associated with unplanned hospitalizations and reasons for admission. Definitions for each category of reason for readmission were as follows: neutropenic fever defined as fever with absolute neutrophil count less than 1500 neutrophils per microliter of blood. Hematologic toxicity included anemia with a hemoglobin less than 12.3 g/dL and thrombocytopenia with platelets less than 150 k/uL. Gastrointestinal toxicity included nausea and vomiting, abdominal pain, constipation, diarrhea, small bowel obstruction, lower tract bleeding, colostomy malfunction, and bowel perforation. Venous thromboembolism encompassed both deep vein thromboses and pulmonary emboli. Neurologic toxicities were defined as syncope, numbness and tingling and mental status change not related to electrolyte disturbances or infection. Cardiovascular included chest pain and atrial fibrillation. Poor nutrition was encompassed electrolyte disturbances and malnutrition. The remaining categories were infection with a known source, pleural effusion and other. Summary statistics were used to report the data. Two-sided statistical analyses were performed using Stata. A p value of b0.05 was considered statistically significant. Analyses were performed comparing cohorts of unplanned admission versus not admitted, as well as comparing patients who were not admitted, admitted one time, and admitted more than once during chemotherapy treatment. Univariable analyses were calculated with each clinical and pathologic factor. Multivariable logistic regression analysis was performed and included all variables with statistical significance (p b 0.05) on univariate analyses in addition to variables with clinical significance regardless of statistical significance. Kaplan Meier survival analysis was performed between cohorts of patients who were not admitted versus those who were admitted. 3. Results A total of 217 evaluable patients were identified. The mean age for the entire cohort was fifty-nine years (Table 1). The self-described race and ethnicity of the cohort was White (34%), Black (29%), Hispanic non-white (18%), and other (19%). The cohort included 50 (23%) patients who received neoadjuvant chemotherapy and 167 (77%) patients who received primary surgical debulking followed by adjuvant chemotherapy. For the adjuvant chemotherapy group, the median days between surgery and initial chemotherapy cycle was thirty-five days. Of those treated with adjuvant chemotherapy, 110 (66%) received IV carboplatin and paclitaxel every 3 weeks, 52 (31%) received IV/IP cisplatin and paclitaxel every 3 weeks and 5 (3%) received IV carboplatin and weekly paclitaxel. Forty-seven (94%) patients treated with neoadjuvant chemotherapy were given IV carboplatin and paclitaxel every 3 weeks, 2 (4%) were treated with IV carboplatin and weekly paclitaxel and 1 (2%) was given IV carboplatin and paclitaxel every 3 weeks followed by IV/IP cisplatin and paclitaxel after secondary debulking surgery. There was a total of 141 admissions in the entire cohort with a corresponding 1314 days of hospitalization. The median readmission duration was 3 days. Eighty-seven (40%) of the patients had an admission within thirty days of their chemotherapy dose: adjuvant 64 (74%) and neoadjuvant 23 (26%). Within the adjuvant group, 41 (34%) of the patients had an admission after adjuvant IV chemotherapy and 23 (46%) of the patients had an admission after adjuvant IV/IP chemotherapy. Cycle one of chemotherapy was the most common (41, 31%) cycle for hospitalizations. Cycle two through six of chemotherapy had roughly the same number of hospitalizations (range 16 to 22). The demographics for those hospitalized were: 27 White (31%), 28 Black (32%), 20 Hispanic non-white (23%) and 12 other (14%) (Table 1). When looking over multiple cycles of chemotherapy, 30 (14%) of the patients had an admission after more than one cycle. Of those admitted more than once, 12 (40%) were Hispanic, 8 (27%) were White, 8 (27%) were Black and 2 (7%) self-identified as other (Table 2). The most common reasons for readmission were for gastrointestinal complaints (32%), neutropenia (18%) and hematologic disturbances not
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
3
Table 1 Demographic and histopathological characteristics of study population. Variable
Entire cohort (N = 217)
Readmitted (N = 87)
No (N = 130)
P-value
Age, mean (SD) Age, n (%) b50 50–59 60–69 70–79 ≥80 Race, n (%) White Black Hispanic Other Parity, n (%) Nulliparity Multiparity Menopausal status, n (%) Yes No BMI, n (%) ≤20 20–25 25–30 N30 Diabetes, n (%) Yes No Hypertension, n (%) Yes No Hyperlipidemia, n (%) Yes No Aspirin, n (%) Yes No Hormone replacement, n (%) Yes No Tobacco, n (%) Yes No Histology, n (%) Serous Endometrioid Other Stage, n (%) 1 2 3 4 Grade, n (%) 1 2 3 Not reported Management, n (%) Primary surgery Neoadjuvant Optimally debulked, n (%) Yes No Charlson comorbidity, median (IQR) Excluding age & cancer Excluding cancer Including age & cancer Type of chemotherapy, n (%) Adjuvant IV/IP cisplatin & paclitaxel Adjuvant IV carboplatin & paclitaxel every 3 weeks Adjuvant IV carboplatin & weekly paclitaxel Neoadjuvant IV carboplatin & paclitaxel every 3 weeks Neoadjuvant IV carboplatin & weekly paclitaxel
59.2 (12.6)
60.2 (11.8)
58.6 (13.2)
0.37 0.59
53 (24) 62 (29) 51 (24) 39 (18) 12 (6)
19 (22) 24 (28) 24 (28) 17 (20) 3 (3)
34 (26) 38 (29) 27 (21) 22 (17) 9 (7)
73 (34) 63 (29) 40 (18) 41 (19)
27 (31) 28 (32) 20 (23) 12 (14)
46 (35) 35 (27) 20 (15) 29 (22)
52 (24) 163 (76)
21 (24) 66 (76)
31 (24) 97 (76)
153 (73) 58 (27)
60 (71) 24 (29)
93 (73) 34 (27)
13 (6) 57 (27) 65 (30) 79 (37)
2 (2) 18 (21) 25 (30) 39 (46)
11 (8) 39 (30) 40 (31) 40 (31)
49 (23) 166 (77)
23 (27) 63 (73)
26 (20) 103 (80)
117 (54) 100 (46)
59 (68) 28 (32)
58 (45) 72 (55)
87 (40) 129 (60)
40 (46) 47 (54)
47 (36) 82 (64)
30 (14) 186 (86)
11 (13) 75 (87)
19 (15) 111 (85)
6 (4) 160 (96)
2 (3) 66 (97)
4 (4) 94 (96)
62 (29) 151 (71)
23 (27) 61 (73)
39 (30) 90 (70)
123 (61) 22 (11) 58 (29)
55 (70) 5 (6) 19 (24)
68 (55) 17 (14) 39 (31)
51 (24) 25 (12) 115 (53) 26 (12)
13 (15) 10 (11) 55 (63) 9 (10)
38 (29) 15 (12) 60 (46) 17 (13)
14 (6) 19 (9) 170 (78) 14 (6)
4 (5) 5 (6) 72 (83) 6 (7)
10 (8) 14 (11) 98 (75) 8 (6)
167 (77) 50 (23)
64 (74) 23 (26)
103 (79) 27 (21)
143 (86) 23 (14)
56 (89) 7 (11)
87 (84) 16 (16)
0 (0–1) 2 (1–3) 7 (5–9)
0 (0–1) 2 (1–3) 8 (6–9)
0 (0–1) 2 (1–3) 7 (4–9)
52 (24) 110 (51) 5 (2) 47 (22) 2 (1)
24 (28) 39 (45) 1 (1) 21 (24) 2 (2)
28 (22) 71 (55) 4 (3) 26 (20) 0 (0)
0.22
0.99
0.77
0.048
0.26
0.001
0.16
0.70
1.00
0.65
0.08
0.052
0.44
0.33
0.42
0.20 0.45 0.051 0.21
Note: menopausal status = yes corresponds to post-menopausal; tobacco = yes corresponds to ever smoked.
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
4
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
Table 2 Demographic and histopathological characteristics of patients readmitted once, those readmitted more than once and those not readmitted Variable
=1 readmitted (N = 57)
N1 readmission (N = 30)
No (N = 130)
P-value 1
P-value 2
Age, mean (SD) Age, n (%) b50 50–59 60–69 70–79 ≥80 Race, n (%) White Black Hispanic Other Parity, n (%) Nullip Multip Menopausal status, n (%) Yes No BMI, n (%) ≤20 20–25 25–30 N30 Diabetes, n (%) Yes No Hypertension, n (%) Yes No Hyperlipidemia, n (%) Yes No Aspirin, n (%) Yes No Hormone replacement, n (%) Yes No Tobacco, n (%) Yes No Histology, n (%) Serous Endometrioid Other Stage, n (%) 1 2 3 4 Grade, n (%) 1 2 3 Not reported Management, n (%) Primary surgery Neoadjuvant Optimally debulked, n (%) Yes No Charlson comorbidity, median (IQR) Excluding age & cancer Excluding cancer Including age & cancer Type of chemotherapy, n (%) Adjuvant IV/IP cisplatin & paclitaxel Adjuvant IV carboplatin & paclitaxel every 3 weeks Adjuvant IV carboplatin & weekly paclitaxel Neoadjuvant IV carboplatin & paclitaxel every 3 weeks Neoadjuvant IV carboplatin & weekly paclitaxel
60.9 (12.1)
58.8 (11.2)
58.6 (13.2)
0.51 0.79
0.43 0.67
12 (21) 15 (26) 17 (30) 10 (18) 3 (5)
7 (23) 9 (30) 7 (23) 7 (23) 0 (0)
34 (26) 38 (29) 27 (21) 22 (17) 9 (7) 0.03
0.04
19 (33) 20 (35) 8 (14) 10 (18)
8 (27) 8 (27) 12 (40) 2 (7)
46 (35) 35 (27) 20 (15) 29 (22) 0.92
0.69
13 (23) 44 (77)
8 (27) 22 (73)
31 (24) 97 (76) 0.77
0.51
42 (74) 15 (26)
18 (67) 9 (33)
93 (73) 34 (27) 0.13
0.46
1 (2) 14 (26) 14 (26) 25 (46)
1 (3) 4 (13) 11 (37) 14 (47)
11 (8) 39 (30) 40 (31) 40 (31) 0.45
0.60
16 (29) 40 (71)
7 (23) 23 (77)
26 (20) 103 (80) 0.003
0.52
40 (70) 17 (30)
19 (63) 11 (37)
58 (45) 72 (55) 0.16
0.21
29 (51) 28 (49)
11 (37) 19 (63)
47 (36) 82 (64) 0.02
0.01
3 (5) 53 (95)
8 (27) 22 (73)
19 (15) 111 (85) 0.86
0.55
2 (4) 43 (96)
0 (0) 23 (100)
4 (4) 94 (96) 0.78
0.59
14 (25) 41 (75)
9 (31) 20 (69)
39 (30) 90 (70) 0.24
0.79
35 (67) 4 (8) 13 (25)
20 (74) 1 (4) 6 (22)
68 (55) 17 (14) 39 (31) 0.08
0.28
10 (18) 5 (9) 38 (67) 4 (7)
3 (10) 5 (17) 17 (57) 5 (17)
38 (29) 15 (12) 60 (46) 17 (13) 0.87
0.96
3 (5) 4 (7) 46 (81) 4 (7)
1 (3) 1 (3) 26 (87) 2 (7)
10 (8) 14 (11) 98 (75) 8 (6) 0.62
0.97
42 (74) 15 (26)
22 (73) 8 (27)
103 (79) 27 (21) 0.36
0.23
38 (93) 3 (7)
18 (82) 4 (18)
87 (84) 16 (16)
0 (0–1) 2 (1–3) 8 (6–9)
1 (0–1) 3 (1–4) 8 (6–10)
0 (0–1) 2 (1–3) 7 (4–9)
0.12 0.64 0.12 0.41
0.11 0.55 0.50 0.83
17 (30) 24 (42) 1 (2) 13 (23) 2 (4)
7 (23) 15 (50) 0 (0) 8 (27) 0 (0)
28 (22) 71 (55) 4 (3) 26 (20) 0 (0)
Note: P-value 1 was testing the overall difference across all 3 groups; p-value 2 was testing the difference between =1 readmitted and N1 readmitted.
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
related to neutropenia (15%) (Table 3). Twenty-nine (64%) of forty-five patients admitted for gastrointestinal complaints had nausea and vomiting; other gastrointestinal causes for admission were constipation (4%), diarrhea (7%), abdominal pain (9%) and other (16%). The number of days from chemotherapy administration to readmission for nausea and vomiting ranged from 1 to 29 days. The median number of days was 6; 55% (16/29) readmissions for nausea and vomiting occurred within 1–6 days of chemotherapy. When comparing all reasons for readmission for patients receiving neoadjuvant chemotherapy to those undergoing adjuvant chemotherapy (all regimens combined), there was a statistically significant difference. Thirty-three patients (15%) received granulocyte colony stimulating factors during their chemotherapy course: 9 in the adjuvant IV cohort, 11 in the adjuvant IV/IP cohort and 13 in the neoadjuvant cohort. There was a significant difference between those with unplanned admissions and those without in body mass index (BMI, p = 0.048) and hypertension (p = 0.001) (Table 1) on univariable analysis. When the entire cohort was separated into those with one readmission, those with more than one readmission and those without any readmissions (Table 2), there was a significant difference between all three groups in race (p = 0.03), hypertension (p = 0.003) and aspirin use (p = 0.02). There was also a significant difference between those with one readmission and those with more than one readmission in race (p = 0.04) and aspirin use (p = 0.01). Although aspirin use was included in our data because of reports suggesting an association between aspirin use and ovarian cancer risk and mortality [21], testing for collinearity revealed aspirin use was highly associated with hypertension (p = 0.0005); therefore, aspirin use was excluded from the logistic regression model. There was no significant difference in the readmission rates when comparing different adjuvant chemotherapy regimens and neoadjuvant chemotherapy. The logistic regression model demonstrated that BMI greater than thirty was associated with increased odds of readmission of 3.32 (95% CI: 1.14–9.65, p = 0.03) (Table 4). There was an increased odds for readmission for patients with hypertension 3.72 (95% CI: 1.58–8.79, p = 0.003). Additional logistic regression modeling grouping all neoadjuvant chemotherapy regimens versus all regimens of primary debulking surgery plus adjuvant therapy showed increased odds of readmission for patients with BMI greater than thirty and hypertension (Appendix 1). Kaplan-Meier estimates for the entire cohort stratified by unplanned hospital admission showed a significant difference in overall survival (p = 0.004) and progression-free survival (p = 0.01) between those with no hospitalizations, those with one hospitalization and those with more than one hospitalization (Fig. 1, Fig. 2). 4. Discussion Unplanned hospital admission rates in our ethnically and racially diverse population of women with epithelial ovarian cancer were higher
5
Table 4 Logistic regression model for readmission. Variable Race White (ref) Black Hispanic Other Age b50 (ref) 50–59 60–69 70–79 ≥80 BMI ≥20 20–25 (ref) 25–30 N30 Hypertension No (ref) Yes Histology Serous (ref) Endometrioid Other Stage 1 (ref) 2 3 4 Tobacco No (ref) Yes Type of chemotherapy Adjuvant IV/IP cisplatin & paclitaxel Adjuvant carboplatin and paclitaxel every 3 weeks (ref) Adjuvant carboplatin and weekly paclitaxel Neoadjuvant chemotherapy Optimally debulked No (ref) Yes
Odds ratio
95% CI
P value
1 1.29 0.95 0.77
0.48–3.47 0.33–2.79 0.27–2.22
0.61 0.93 0.63
1 1.43 1.19 1.18 1.10
0.51–4.00 0.37–3.85 0.34–4.11 0.12–9.93
0.50 0.77 0.80 0.94
1.54 1 2.33 3.32
0.22–10.90 0.66 0.80–6.84 1.14–9.65
0.12 0.03
1 3.72
1.58–8.79
0.003
1 0.34 0.45
0.09–1.29 0.18–1.15
0.11 0.10
1 1.63 3.99 0.46
0.45–5.91 0.46 1.25–12.78 0.02 0.08–2.62 0.38
1 1.09
0.46–2.59
0.85
0.50 1
0.19–1.33
0.16
0.17 0.73
0.01–2.51 0.20 0.02–26.32 0.86
1 1.34
0.34–5.25
0.68
The odds ratio with 95% CI for neoadjuvant vs primary surgery was 1.65 (0.04–60.58), pvalue = 0.79.
PFS: No readmission 1 readmission >1 readmission
Table 3 Reason for unplanned hospital admission for patients receiving chemotherapy for epithelial ovarian cancer
Neutropenic fever Gastrointestinal toxicity Poor nutrition Hematologic toxicity Venous thromboembolism Infection Neurologic toxicity Cardiovascular toxicity Pleural effusion Other
Entire cohort (%)
Neoadjuvant (%)
Adjuvant: IV (%)
Adjuvant IV/IP (%)
25 (18) 45 (32) 11 (11) 23 (15) 5 (4)
5 (13) 14 (35) 0 (0) 5 (13) 0 (0)
11(15) 18 (25) 9 (13) 18 (25) 1 (1)
9 (30 13 (43) 2 (2) 0 (0) 4 (13)
12 (9) 5 (4) 3 (2) 4 (3) 8 (4)
6 (15) 3 (8) 1 (3) 2 (5) 4 (10)
5 (7) 2 (3) 2 (3) 1 (1) 3 (4)
1 (3) 0 (0) 0 (0) 1 (3) 0 (0)
P value comparing all reasons for readmission for neoadjuvant vs combined adjuvant chemotherapy = 0.03.
Fig. 1. Kaplan-Meier curves for PFS vs readmission. Log-rank test p-value = 0.01 (overall comparison).
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
6
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
OS: No readmission 1 readmission >1 readmission
Fig. 2. Kaplan-Meier curves for OS vs readmission. Log-rank test p-value = 0.004 (overall comparison).
(40%) than the previously reported rates in the literature. Hypertension and obesity were predictive of a single unplanned readmission. Race was associated with multiple unplanned readmissions. Poorer diseasefree survival and overall survival estimates in women with unplanned hospital readmissions were seen. Identification of obesity and hypertension as a risk factor may help identify patients who would benefit from care navigators and outpatient resources to identify preventable problems before admission. Patient care navigators could aid patients in coordinating appointments with primary care physicians as well as their oncologists to optimize treatment of comorbidities; patients with other cancer types have significantly fewer problems with care coordination when given support from patient care navigators [22]. We also identified nausea and vomiting as the most common reason for readmission. From the data available in this retrospective review, it is unclear why nausea and vomiting were the most common reasons for readmission. A quality improvement initiative prospectively examining anti-emetic prophylaxis at time of chemotherapy administration and anti-emetic use in the outpatient setting is warranted to further characterize this finding. The addition of a neurokinin-1 receptor antagonist to the pre-medication regimen may lead to reduced admissions for nausea and vomiting in our population with delayed phase chemotherapy induced nausea and vomiting. In a systematic review, dos Santos et al. demonstrated that the addition of a neurokinin-1 receptor antagonist to the standard regimen of a 5-HT3 receptor antagonist and glucocorticoid significantly improved effectiveness in both eliminating or reducing nausea in the acute and delayed phases of chemotherapy induced nausea and vomiting as compared to the standard regimen [23]. Available retrospective studies in the literature looking at hospital readmission rates after initiating treatment for ovarian cancer including surgery and chemotherapy suggest rates of readmission from 10 to 35% [12,13]. These studies largely focused on postoperative readmissions. Clark et al. found a readmission rate of 12% following surgical cytoreduction in their patients in Boston with stages II–IV ovarian cancer. Perioperative complications placed the patients at highest risk while age, medical comorbidities, surgical approach did not affect the likelihood of readmission [15]. In another study, Fauci et al demonstrated a hospital readmission rate of 16% within 30 days of primary surgery in Alabama [16]. Eskander et al looked at readmission rates using the SEER database and found a rate of 19.5% readmission rate. The increased readmission rates may be partially secondary to the limitation of the database, which only includes patients 65 years or older [14].
Duska et al, focused on both surgery and chemotherapy and demonstrated a readmission rate of 12% after primary debulking and adjuvant chemotherapy with body mass index, residual tumor after debulking and surgical stage being predictive of readmission [17, 18]. Similarly, we found that body mass index was a risk factor for readmission. Duska et al also suggested that patients with advanced disease may have lower readmission rates if they had undergone neoadjuvant chemotherapy prior to debulking. The unplanned admission rate in the cohort of patients receiving intraperitoneal therapy reported by Wright et al. most closely approximates the unplanned admission rate identified in our study (35%) [20]. The dose dense and standard intravenous cohorts from that study at unplanned admission rates of 25% and 21% respectively. Patients with breast cancer are often treated with taxane containing chemotherapy regimens; rates of hospital admission related to chemotherapy in breast cancer patients are reported as 8–13%. These rates are lower than the readmission rate that our study found [24,25]. Our study showed a difference in readmission rates as compared with previously reported literature (10–35%). While there has been much focus on hospital readmission after primary debulking and adjuvant chemotherapy there has been less on readmission after neoadjuvant chemotherapy. It has been suggested that patients at higher risk of readmission after primary surgery and chemotherapy may benefit from neoadjuvant chemotherapy instead [18]. Our study demonstrates no significant difference in readmission rates after neoadjuvant chemotherapy and primary debulking surgery followed by adjuvant chemotherapy. We found a statistically significant difference when combining reasons for readmission when comparing neoadjuvant chemotherapy to all regimens of adjuvant chemotherapy; however, the number of patients in each group was too small to further delineate which specific causes for readmission were significantly different. This would suggest that future approaches to reduce readmission would be applicable in neoadjuvant patients as well. Unique to our study, we examined the progression free and overall survival rates from the prospective of unplanned hospital readmissions. Prior studies have focused on treatment regimens (primary debulking versus neoadjuvant, IV versus IV/IP adjuvant chemotherapy) when reporting these outcomes. When combining all treatment regimens, we found a significant difference in both progression free and overall survival rates for those with unplanned hospital readmissions. Patients with more than one hospital readmission have the worst survival outcomes in our study population. Effects of unplanned hospital readmission rates on both progression free and overall survival warrants further research in patients treated for ovarian cancer. In contrast to the previously reported studies on readmission, our study population included an ethnically diverse population in one of the poorest urban counties in America. Previous studies have focused on a more racially homogenous and economically affluent population than our study population with rates of white patients between 75 and 90% while in our study the percentage of white patients was 34%. Reasons for unplanned readmissions (gastrointestinal 32%, neutropenic fever 18%, and hematological 15%) also differed in our study population as compared to the reasons described by Wright et al (infection 15%, gastrointestinal 13%, and electrolyte disorders 12%) [13]. These differences may have been related to variation in use and access to supportive therapies such as colony stimulating factor and anti-emetics [13]. The limitations of this study include the possibility of underestimating the total number of unplanned hospitalizations since patients may have been hospitalized at other hospitals not affiliated with our institution. The monetary consequences and impact on quality of life of over one thousand days of hospital admission were beyond the scope of this study. There is also the possibility of under-capturing toxicity by selecting only one reason for admission. These complications that we report may not have been caused by the chemotherapy administered and instead caused by other factors, such as disease progression or surgery. While planned treatment regimens were included in the analysis, the lack of accommodation for variations between actual doses prescribed
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021
S. Dioun et al. / Gynecologic Oncology xxx (2018) xxx–xxx
and delivered as well as dose reductions is a limitation. In comparison of the unplanned hospitalizations for our patients as compared to previous studies we did not evaluate compulsory use of granulocyte colony stimulating factors or use of these factors in subsequent cycles after unplanned admission. Our data suggests that unplanned hospitalizations in our racially diverse patient population are higher than those reported in the literature. Unplanned hospitalization may be a predictor for overall survival and disease-free survival. Identifying patients that are more likely to be readmitted may play a role in chemotherapy decision-making, and resource allocation including care navigators. Going forward, further studies to stratify patients to optimize adjuvant chemotherapy for women with ovarian cancer are needed. Supplementary data to this article can be found online at https://doi. org/10.1016/j.ygyno.2018.08.021. Conflict of interest statement None of the authors have any conflicts of interest to disclose
Author contribution Shayan Dioun: Study conception and design, data collection, analysis and interpretation, manuscript writing. Jennifer R. Jorgensen: Analysis and interpretation, manuscript writing, critical revision. Eirwen M. Miller: Data collection, critical revision. Joan Tymon-Rosario: Data collection and critical revision. Xianhong Xie: Analysis and interpretation of data, critical revision. Xiaonan Xue: Analysis and interpretation of data, critical revision. Dennis Yi-Shin Kuo: Study conception and design, critical revision. Nicole S. Nevadunsky: Study conception and design, analysis and interpretation, manuscript writing, critical revision. References [1] J.D. Wright, L. Chen, A.I. Tergas, S. Patankar, W.M. Burke, J.Y. Hou, et al., Trends in relative survival for ovarian cancer from 1975 to 2011, Obstet. Gynecol. 125 (2015) 1345–1352. [2] G.A. Omura, B.N. Bundy, J.S. Berek, S. Curry, G. Delgado, R. Mortel, Randomized trial of cyclophosphamide plus cisplatin with or without doxorubicin in ovarian carcinoma: a Gyneco- logic Oncology Group Study, J. Clin. Oncol. 7 (1989) 457–465. [3] W.P. McGuire, W.J. Hoskins, M.F. Brady, P.R. Kucera, E.E. Partridge, K.Y. Look, et al., Cyclophosphamide and cisplatin compared with paclitaxel and cisplatin in patients with stage III and stage IV ovarian cancer, N. Engl. J. Med. 334 (1996) 1–6. [4] M.J. Piccart, K. Bertelsen, K. James, J. Cassidy, C. Mangioni, E. Simonsen, et al., Randomized intergroup trial of cisplatin- paclitaxel versus cisplatin-cyclophosphamide in women with advanced epithelial ovarian cancer: three-year results, J. Natl. Cancer Inst. 92 (2000) 699–708. [5] E.A. McGlynn, S.M. Asch, J. Adams, J. Keesey, J. Hicks, A. DeCristofaro, et al., The quality of health care delivered to adults in the United States, N. Engl. J. Med. 348 (2003) 2635–2645. [6] J.M. Unger, W.E. Barlow, D.P. Martin, S.D. Ramsey, M. Leblanc, R. Etzioni, et al., Comparison of survival outcomes among can- cer patients treated in and out of clinical trials, J. Natl. Cancer Inst. 106 (2014), dju002.
7
[8] J.P. Neijt, S.A. Engelholm, M.K. Tuxen, et al., Exploratory Phase III study of paclitaxel and cisplatin vs paclitaxel and carboplatin in advanced ovarian cancer, J. Clin. Oncol. 18 (2000) 3084–3092. [9] A. du Bois, H. Lück, W. Meier, et al., A randomized clinical trial of cisplatin/paclitaxel vs carboplatin/paclitaxel as first-line treatment of ovarian cancer, J. Natl. Cancer Inst. 95 (2000) 1320–1329. [10] N.K. Aaronson, S. Ahmedzai, B. Bergman, et al., The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology, J. Natl. Cancer Inst. 85 (1993) 365–376. [11] Jeffrey Clough, D. Kamal, H. Arif, Oncology care model: short- and long-term considerations in the context of broader payment reform, J. Oncol. Pract. 11 (4) (2015) 319–321. [12] Blase Polite, N. Miller, D. Harold, Medicare innovation center oncology care model, J. Oncol. Pract. 11 (2) (2015) 117–119. [13] A.A. Wright, A. Cronin, D.E. Milne, M.A. Bookman, R.A. Burger, D.E. Cohn, et al., Use and effectiveness of intraperitoneal chemotherapy for treatment of ovarian cancer, J. Clin. Oncol. 33 (2015) 2841–2847. [14] P. Basch, Quality of health care delivered to adults in the United States, N. Engl. J. Med. 349 (19) (2003) 1866–1868. [15] R.M. Clark, W.B. Growdon, A. Wiechert, D. Boruta, M. Del Carmen, A.K. Goodman, et al., Patient, treatment and discharge factors associated with hospital readmission within 30 days after surgical cytoreduction for epithelial ovarian carcinoma, Gynecol. Oncol. 130 (3) (2013) 407–410. [16] J.M. Fauci, K.E. Schneider, P.J. Frederick, G. Wilding, J. Consiglio, A.L. Sutton, et al., Assessment of risk factors for 30-day hospital readmission after surgical cytoreduction in epithelial ovarian carcinoma, Int. J. Gynecol. Cancer 21 (5) (2011) 806–810. [17] R.N. Skander, J. Chang, A. Ziogas, H. Anton-Culver, R.E. Bristow, Evaluation of 30-day hospital readmission after surgery for advanced-stage ovarian cancer in a Medicare population, J. Clin. Oncol. 32 (36) (2014) 4113–4119. [18] L.R. Duska, J.J. Java, D.E. Cohen, R.A. Burger, Risk factors for readmission in patients with ovarian, fallopian tube, and primary peritoneal carcinoma who are receiving front- line chemotherapy on a clinical trial (GOG 218): an NRG oncology/gynecologic oncology group study (ADS-1236), Gynecol. Oncol. 139 (2) (2015) 221–227. [19] M.E. Charlson, P. Pompei, K.L. Ales, C.R. MacKenzie, A new method of classifying prognostic comorbidity in longitudinal studies: development and validation, J. Chronic Dis. 40 (1987) 373–383. [20] J.D. Wright, J.Y. Hou, W.M. Burke, A. Tergas, L. Chen, J. Hu, C. Ananth, A. Neuget, D. Hershman, Utilization and toxicity of alternative delivery methods of adjuvant chemotherapy for ovarian cancer, Obstet. Gynecol. 127 (6) (2016) 985–991. [21] B. Trabert, R.B. Ness, W.H. Lo-Ciganic, M.A. Murphy, E.L. Goode, E.M. Poole, L.A. Brinton, P.M. Webb, C.M. Nagle, S.J. Jordan, Australian Ovarian Cancer Study Group, Australian Cancer Study (Ovarian Cancer), H.A. Risch, M.A. Rossing, J.A. Doherty, M.T. Goodman, G. Lurie, S.K. Kjaer, E. Hogdall, A. Jensen, D.W. Cramer, K.L. Terry, A. Vitonis, E.V. Bandera, S. Olson, M.G. King, U. Chandran, H. Anton-Culver, A. Ziogas, U. Menon, S.A. Gayther, S.J. Ramus, A. Gentry-Maharaj, A.H. Wu, C.L. Pearce, M.C. Pike, A. Berchuck, J.M. Schildkraut, N. Wentzensen, Ovarian Cancer Association Consortium, Aspirin, nonaspirin nonsteroidal anti-inflammatory drug, and acetaminophen use and risk of invasive epithelial ovarian cancer: a pooled analysis of the Ovarian Cancer Consortium, J. Natl. Cancer Inst. 106 (2) (2014), djt431. [22] E.H. Wagner, E.J. Ludman, E.J. Aiello Bowles, R. Penfold, R.J. Reid, C.M. Rutter, J. Chubak, R. McCorkle, Nurse navigators in early cancer care: a randomized, controlled trial, J. Clin. Oncol. 32 (1) (2014) 12–18. [23] L.V. dos Santos, F.H. Souza, A.T. Brunetto, A.D. Sasse, J.P. da Silveira Nogueira Lima, Neuorkinin-1 receptor antagonists for chemotherapy-induced nausea and vomiting: a systematic review, J. Natl. Cancer Inst. 104 (17) (2012) 1280. [24] N.M. Pittman, W.M. Hopman, M. Mates, Emergency room visits and hospital admission rates after curative chemotherapy for breast cancr, J. Oncol. Pract. 11 (2) (2015) 120–125. [25] J.M. Beana-Canada, S. Estalella-Mendoza, P. Rosado-Varela, I. Exposito-Alvarez, M. Gonzalez-Guerrero, M.C. Diaz-Blanco, C. Cortes-Carmona, P. Ramirez-Daffos, E. Arriola-Arellano, A. Reuda-Ramos, L. Solana-Grimaldi, E. Benitez-Rodriguez, Use of health-care services during chemotherapy for breast cancer, Eur. J. Cancer 48 (18) (2012) 3328–3334.
Please cite this article as: S. Dioun, et al., Unplanned hospitalizations in a racially and ethnically diverse population of women receiving chemotherapy for epithelial ovarian ..., Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.08.021