Surgical care of elderly women with ovarian cancer: A population-based perspective

Gynecologic Oncology 99 (2005) 352 – 357 www.elsevier.com/locate/ygyno

Surgical care of elderly women with ovarian cancer: A population-based perspective Teresa P. Dı´az-Montes a,b,*, Mariana L. Zahurak b,c, Robert L. Giuntoli IIa,b, Ginger J. Gardner a,b, Toby A. Gordon b,d, Deborah K. Armstrong b,e, Robert E. Bristow a,b a

The Kelly Gynecologic Oncology Service, Department of Gynecology and Obstetrics, The Johns Hopkins Medical Institutions, Baltimore, MD 21287, USA b The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Baltimore, MD 21231, USA c Department of Biostatistics, Johns Hopkins Oncology Center, The Johns Hopkins Medical Institutions, Baltimore, MD 21287, USA d Department of Planning and Marketing, The Johns Hopkins Medical Institutions, Baltimore, MD 21287, USA e Department of Medical Oncology, The Johns Hopkins Medical Institutions, Baltimore, MD 21287, USA Received 2 April 2005 Available online 1 August 2005

Abstract Objective. To characterize the primary surgical care and short-term outcomes for ovarian cancer in women aged 80 years and older compared to women younger than 80 years. Methods. A statewide hospital discharge database was used to identify women undergoing primary surgery for ovarian cancer from 1990 to 2000. Logistic regression models were used to evaluate for significant differences in demographic characteristics and short-term outcomes comparing women aged 80 years with those aged <80 years. Results. A total of 2417 women were identified; women aged 80 years comprised 7.0% (n = 168) of cases. Compared to younger women, those aged 80 years were significantly more likely to be admitted under emergent conditions (25.6% vs. 14.9%, P < 0.0003) and less likely to undergo surgery at a university hospital (6.6% vs. 18.6%, P = 0.001). Ovarian cancer patients aged 80 years were significantly more likely to have a longer hospital stay (median 10 days vs. 7 days, P < 0.0001) and a higher adjusted cost of hospital related care (median $76,760 vs. $52,649, P < 0.0001). The 30-day mortality rate was 2.3-fold higher for women aged 80 years (5.4% vs. 2.4%, P = 0.036). For women aged 80 years, there was a trend toward a higher risk of peri-operative death among low-volume hospitals (8.8%) compared to highvolume hospitals (3.0%, P = 0.16). Conclusion. Primary surgical care for ovarian cancer in women aged 80 years is associated with utilization of significant health care resources and worse short-term outcomes compared to younger women. Additional research is needed to identify opportunities for improving the cost-effectiveness of care in this population. D 2005 Elsevier Inc. All rights reserved. Keywords: Ovarian cancer; Surgery; Elderly women; Access to care

Introduction In the United States, ovarian cancer is the fourth most common cause of cancer-related death among women [1]. It is * Corresponding author. The Kelly Gynecologic Oncology Service, Department of Gynecology and Obstetrics, The Johns Hopkins Medical Institutions, 600 North Wolfe Street, Phipps #281, Baltimore, MD 21287, USA. Fax: +1 410 614 8718. E-mail address: [email protected] (T.P. Dı´az-Montes). 0090-8258/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2005.06.005

estimated that 22,220 new cases will be diagnosed and approximately 16,210 women will die of this disease in the United States during 2005 [1]. Ovarian cancer is more frequently diagnosed among elderly women. The age-specific incidence of ovarian cancer peaks in the eighth decade. Seventy percent of women with ovarian cancer are diagnosed with advanced stage disease, and approximately 60% of them will die within 5 years [2]. Specific and reproducible factors associated with the increasing mortality among older women with ovarian cancer remain to be defined.

T.P. Dı´az-Montes et al. / Gynecologic Oncology 99 (2005) 352 – 357

Among all gynecologic malignancies, ovarian cancer is the deadliest, with an overall 5-year survival estimate of 37% [2]. Significant prognostic factors for ovarian cancer include tumor stage, volume of residual disease, and histologic grade [3]. Ovarian cancer treatment typically consists of aggressive surgical debulking combined with adjuvant chemotherapy. Due to the disproportionate number of older women that are affected with ovarian cancer and their often compromised medical condition, concerns arise with respect to the advisability of an aggressive treatment program of initial surgical debulking and chemotherapy. Multiple studies have demonstrated that the management of older patients affected with ovarian cancer is often less aggressive than that received by their younger counterparts [4,5]. A review of the SEER data has revealed that up to 40% of women 85 years of age and older were not offered a definitive cancer treatment program [6]. As described by Berkman et al. [7], some of the myths and biases related to cancer in the elderly encountered by health professionals include: (1) the overlooking of early signs and symptoms that are attributed to other intercurrent illness or to natural physiologic aging changes, (2) decisions regarding treatment are made on the basis of patient age rather than on the basis of the patient’s actual medical condition and availability of treatment, and (3) that chronologic age is used as a proxy for negative characteristics like poor prognosis and limited life expectancy. The study of cancer and aging has emerged as a critical issue during recent years. The incidence of cancer rises steadily with advancing age during adult life. Cancer is the second most common cause of death among persons aged 65 years and older, with approximately half of all cancers occurring in this population [8]. Sixty percent of cancerrelated deaths occurred among people older than 65 years [9]. It is estimated that persons in this age group will comprise 20.1% of the U.S. population by 2030 [9,10]. The group over 75 years will triple by 2030, and the group 85 and older will double in the same period [9,10]. These statistics outline the significance of an increasing older cancer population that will require oncologic management specific for their needs. Despite the projected growth in the number of elderly United States women, there is a paucity of data regarding the primary surgical care and the short-term outcomes of women aged 80 and older diagnosed with ovarian cancer. Therefore, the goals of this study were to characterize the primary surgical care and short-term outcomes for ovarian cancer in women aged 80 and older compared to women younger than 80 years using a statewide population-based data set.

Methods Data source

(HSCRC) was done. The HSCRC database provides information regarding the index hospital admission and is limited to 30 days of follow-up. All adult female patients, 18 years of age and older, who underwent a surgical procedure including oophorectomy for a malignant ovarian neoplasm in Maryland between January 1, 1990 and December 31, 2000 were included in the study. The International Classification of Disease, 9th revision (ICD-9) code 183.0 was used for sorting [11]. All histologic classifications were included. The surgical procedures included in the analysis were limited to those incorporating oophorectomy, with or without hysterectomy, as this were felt to be the most likely procedures that patient underwent for initial ovarian cancer surgery. The surgical procedures included were: abdominal Table 1 Demographic characteristics among women of different age groups undergoing primary surgery for ovarian cancer in Maryland, 1990 – 2000 Variable

<80 years, n (%)

80 years, n (%)

Number of patients Ethnic classification White African-American American-Indian or Eskimo Asian or Pacific Islander Unknown Other Hospital admission type Elective Emergency Urgent Other Unknown Insurance payer Medicaid Medicare Commercial Health maintenance organization Self pay Other Hospital type Community Community teaching University Surgical procedure Hysterectomy Other Hospital volume Low volumea High volumeb Surgeon volume Low volumec High volumed Attending physician or operating surgeon concordance Concordance Discordance

2249 (93.0)

168 (7.0)

1825 313 2 41 12 56

(81.2) (13.2) (0.1) (1.8) (0.5) (2.5)

143 19 0 0 1 5

(85.0) (11.0) (0.0) (0.0) (1.0) (3.0)

1561 335 330 22 1

(69.4) (14.9) (14.7) (1.0) (0.0)

104 43 20 1 0

(61.9) (25.6) (11.9) (0.6) (0.0)

95 707 845 529 60 13

(4.2) (31.4) (37.6) (23.5) (2.7) (0.6)

0 158 3 7 0 0

(0.0) (94.0) (1.8) (4.2) (0.0) (0.0)

a

A cross-sectional analysis of hospital discharge data from nonfederal acute care hospitals in Maryland collected by the Maryland Health Services Cost Review Commission

353

b c d

19 cases/year. 20 cases/year. 9 cases/year. 10 cases/year.

1294 (57.4) 537 (23.9) 418 (18.6)

103 (61.3) 54 (32.1) 11 (6.6)

1641 (73.0) 608 (27.0)

118 (70.2) 50 (29.8)

893 (39.7) 1356 (60.3)

68 (40.5) 100 (59.5)

1007 (44.8) 1242 (55.2)

63 (37.5) 105 (62.5)

1920 (85.4) 329 (14.6)

131 (78.0) 37 (22.0)

354

T.P. Dı´az-Montes et al. / Gynecologic Oncology 99 (2005) 352 – 357

Table 2 Variables influencing age among women treated by primary surgery for ovarian cancer in the state of Maryland during 1990 – 2000 Covariate

<80 years, median (range)

80 years, median (range)

P value

Total procedures billed Total adjusted cost

4 (1 – 15) $52,649 (0 – $2,061,787) 1 (0 – 8) 0 (0 – 64) 7 (0 – 79)

5 (1 – 15) $76,760 ($7654 – $896,148) 2 (0 – 7) 0 (0 – 19) 10 (2 – 64)

0.0020 <0.0001

Comorbidities ICUa days Length of stay (days) a

0.1000 0.0700 <0.0001

Intensive care unit.

hysterectomy (68.3, 68.4, and 68.6), oophorectomy (65.01, 65.09), unilateral oophorectomy (65.31, 65.39), unilateral salpingo-oophorectomy (65.41, 65.49), bilateral oophorectomy (65.51 –65.54), and bilateral salpingo-oophorectomy (65.61 –65.64). The primary independent variable in the study was patient age. Patient age was modeled using a categorical variable (patients younger than 80 years vs. patients aged 80 years and older). Secondary independent variables included were: ethnicity, insurance payer status, hospital admission type, hospital volume, surgeon volume, hospital type, surgical procedure, inpatient death, and concordance between attending and operating physician. Frequency distributions for these variables were tabulated for women who underwent surgery for ovarian cancer aged 80 years and older and women younger than 80 years. Logistic regression was employed to evaluate total number of procedures billed, adjusted cost of hospital-related care, comorbidities, intensive care unit length of stay, and admission length of stay. Logistic regression was done for the subset group of women aged 80 years and older to evaluate the variables of total number of procedures billed, adjusted cost of hospital-related care, comorbidities, length of intensive care unit stay, admission length of stay, and inpatient death among hospital’s volume and surgeon’s volume. For the purpose of this study, surgeons performing 9 cases or less per year were categorized as low-volume surgeons, and those performing 10 or more cases per year were categorized as high-volume surgeons. Similarly, hospitals with 19 or less cases per year were categorized as low-volume hospitals, and those with 20 or more cases per year were categorized as high volume. A community teaching hospital was defined as a nonuniversity hospital with a residency program in Obstetrics and Gynecology. Approval to conduct this study was obtained from the Johns Hopkins Medical Institutions Clinical Research Committee and Joint Committee on Clinical Investigation. Data were extracted from the HSCRC database in order to characterize the primary surgical care and short-term outcomes of women 80 years of age and older compared to women younger than 80 years diagnosed with ovarian cancer. Factors associated with women older than 80 years were explored using cross tabulations and logistic regression analysis models [12]. All statistical computations were

performed using the SAS system [13], and all reported P values are two-sided.

Results Patient demographics and patterns of surgical care A total of 2417 women who met the criteria for a primary surgical procedure for a malignant ovarian neoplasm were identified in the state of Maryland during the period of 1990 to 2000. Women aged more than 80 years comprised 7.0% (n = 168) of the cases identified during the 11-year study period. Most of the 80 years and older women were white (85%), had Medicare insurance coverage (94%), and were treated in community-based hospital (61%) (Table 1). Women 80 years of age and older were significantly less likely to be admitted to a university type hospital when compared to women younger than 80 years (OR = 0.33, 95%CI: 0.18, 0.62; P = 0.0010). Elective surgery comprised the great majority of the admissions (61.9%) for women aged 80 years and older. Women aged 80 years and older were significantly more likely to have surgery under emergency conditions than women younger than 80 years (OR = 1.96, 95%CI: 1.36, 2.80; P = 0.0003). Overall, 118 of the surgical cases (70%) performed on women aged 80 years and older and 1641 of the cases (73%) performed on women younger than 80 years included an abdominal hysterectomy. There was no significant difference between groups in terms of type of procedure performed for ovarian cancer surgery (OR = 1.14, 95%CI: 0.80, 1.62; P = 0.44). Women aged 80 years and older were less likely than women younger than 80 years to have concordance between the operating physician and the attending physician (OR = 0.61, 95%CI: 0.41, 0.89; P = 0.01). In other words, older women were more likely to have an operating surgeon that was a distinctly separate clinician from the attending physician of record. Compared to younger women, those aged 80 years and older were significantly more likely to have a longer length of hospitalization stay (median 10 days vs. 7 days, P <

Fig. 1. Peri-operative (30-day) mortality rate among women that underwent surgical treatment for ovarian cancer in Maryland, 1990 – 2000 according to age and hospital volume. Legend: <80 years, 80 years.

T.P. Dı´az-Montes et al. / Gynecologic Oncology 99 (2005) 352 – 357

355

Table 3 Variables influencing surgeon volume among women aged 80 years in the state of Maryland during 1990 – 2000

Table 4 Variables influencing hospital volume among women aged 80 years in the state of Maryland during 1990 – 2000

Covariate

Low-volume surgeon, median (range)

High-volume surgeon, median (range)

P value

Covariate

Low-volume hospital, median (range)

High-volume hospital, median (range)

P value

Total procedures billed Total adjusted cost Comorbidities ICUa days Length of stay (days)

3 (1 – 12)

6 (2 – 15)

<0.0001

4 (1 – 14)

5 (2 – 15)

0.0500

$42,993 ($7654 – $896,148) 1 (0 – 4) 0 (0 – 15) 10 (2 – 49)

$118,971 ($10,910 – $849,306) 2 (0 – 7) 0 (0 – 19) 10 (3 – 64)

<0.0001

Total procedures billed Total adjusted cost

$74,490 ($10,910 – $849,306) 2 (0 – 7) 0 (0 – 19) 11 (3 – 64)

0.1500

a

0.0010 0.8200 0.8400

$78,471 ($7654 – $896,148) Comorbidities 1 (0 – 5) ICUa days 0 (0 – 15) Length of stay (days) 10 (2 – 49) a

0.0200 0.5700 0.4000

Intensive care unit.

Intensive care unit.

0.0001), a higher adjusted cost of hospital-related care (median $76,760 vs. $52,649, P < 0.0001), and more billable procedures (mean 5 vs. 4, P = 0.0020). There were no significant differences between the number of medical comorbidities and length of intensive care unit stay between the two groups (Table 2). However, the 30-day mortality rate was 2.3 times higher for women aged 80 years and older (5.4%) compared to women younger than 80 years (2.4%) (95%CI: 1.14, 4.84, P = 0.0200) (Fig. 1). In total, 531 different surgeons provided primary ovarian cancer surgical care, although not all surgeons provided continuous care for the entire duration of the study period. Only 18 (3.4%) of the surgeons were categorized as highvolume surgeons (10 cases/year) and 513 (96.6%) of the surgeons were categorized as low-volume surgeons (9 cases/year). Sixty-two percent (n = 105) of women 80 years and older were managed by high-volume surgeons. In the subgroup of women aged 80 years and older, those that were managed by high-volume surgeons were significantly more likely to have more procedures billed (median 6 vs. 3, P < 0.0001), had a higher adjusted cost of hospital-related care (median $118,971 vs. $42,993, P < 0.0001), had more comorbidities (median 2 vs. 1, P = 0.0010). There was no significant difference in terms of length of hospitalization stay and intensive care unit stay between high- and lowvolume surgeons (Table 3). Six percent (n = 6) of the deaths

Fig. 2. Peri-operative (30-day) mortality rate among women that underwent surgical treatment for ovarian cancer in Maryland, 1990 – 2000 according to age and surgeons volume. Legend: <80 years, 80 years.

happened among high-volume surgeons and 5% (n = 3) among low-volume surgeons. No significant difference was noted in the peri-operative death rate between high- and low-volume surgeons (OR = 0.82, 95%CI: 0.20, 3.42; P = 1.0000) (Fig. 2). A total of 49 hospitals provided care for ovarian cancer patients during the study period. Only 9 hospitals (18.4%) were categorized as high volume (20 cases/year) and 40 hospitals (81.6%) were categorized as low volume (19 cases/year). Sixty percent (n = 100) of women 80 years and older were managed at high-volume hospitals. In the subgroup of women aged 80 years and older, those admitted at high-volume hospitals were significantly more likely to have more procedures billed (median 5 vs. 4, P = 0.0500) and have more medical comorbidities (median 2 vs. 1, P = 0.0200). There was no significant difference in adjusted costs of hospital-related care, intensive care unit stay, or length of hospitalization stay between high- and lowvolume hospitals (Table 4). Although there was a trend toward a lower 30-day mortality rate for women age 80 years and older operated on high-volume hospitals (3.0%) compared to low-volume hospitals (8.8%), this trend did not reach statistical significance (OR = 3.12, 95%CI: 0.75, 13.1; P = 0.1602) (Fig. 1).

Discussion Our society is aging. There are now more than 35 million persons age 65 and older in the United States, representing more than 13% of the population [9]. By 2030, about 20% of the United States population will be older than age 65 [10]. These increases reflect dramatic increases in the life expectancy over the last century. With the population aging, cancer in the elderly is becoming an increasingly common problem [14]. As the population ages over the next 50 years, the number of cancer patients is expected to double from the current 1.3 million to 2.6 million, and the majority of these patients will be at least 75 years old [15]. In view of an upsurge in cancer incidence, it is imperative to evaluate the adequacy of health care that is available for geriatric ovarian cancer patients.

356

T.P. Dı´az-Montes et al. / Gynecologic Oncology 99 (2005) 352 – 357

Our analysis of the HSCRC database during the period of 1990 – 2000 demonstrated that even though the majority of women aged 80 years and older underwent elective surgery for ovarian cancer, they were more likely than younger women to undergo surgery under emergency conditions. Also, the older group was less likely to be treated at university hospitals. At present, there is no published report available to compare these results. A healthcare infrastructure with a high level of expertise and capacity to provide intensive postoperative care is crucial for achieving good surgical results, especially in the elderly. Hightower et al. [16] reported that in women older than 80 years, surgery was done more frequently by nonspecialized surgeons or general gynecologists, and often these patients did not receive standard surgery and postoperative chemotherapy. To a certain extent, the differences in the surgical management of elderly ovarian cancer patients reflect the different backgrounds and experience of the physicians. Hightower et al. revealed that most elderly ovarian cancer patients were cared for by nononcologists, such as general surgeons (31%), and obstetrician/gynecologists (29%) [16]. Only 18% of the older women in this report were cared by a gynecologic oncologist. The results of the current study seem to mirror these observations. Specifically, the fact that older patients were more likely to have an operating surgeon different from their attending physician of record suggests that, in many cases, primary surgical care was not being directed by a gynecologic oncologist. Further evaluation of surgical expertise of operating surgeon was limited by the way the HSCRC data set is configured. The current study revealed no significant differences between age groups in terms of the number of the surgical procedure performed. This observation contradicts reports presented by Bruchim et al. that revealed that elderly patients underwent proportionally fewer surgeries and less aggressive surgery for primary treatment of ovarian cancer and were more likely to receive neoadjuvant chemotherapy than their younger counterparts [17]. Hightower et al. also reported that operations most commonly performed in the staging and treatment of ovarian cancer were performed less frequently in the group of 80 years and older [16]. Unfortunately, further evaluation of the adequacy of surgical debulking according to age group is limited by the configuration of the HSCRC data set. Perhaps most importantly, the current data indicate that the 30-day mortality rate was 2.3 times higher for women aged 80 years and older, compared to women younger than 80 years. Although this is not surprising, previous studies have had difficulty in quantifying the absolute risk of death associated with ovarian cancer surgery among elderly women. Furthermore, for women aged 80 years and older, there was a notable trend, although not statistically significant, toward a higher risk of peri-operative death among those cared for at low-volume hospitals (8.8%) compared to patients undergoing surgery at high-volume hospitals (3.0%). For comparison, Cloven at al. have reported that

the overall median survival for the group of women older than 80 years was 6 months and 83% of the women older than 80 years were able to receive chemotherapy after the surgical procedure [18]. There are several limitations of the current study that must be considered in interpreting the data. First, the Maryland HSCRC database provides no follow-up beyond 30 days from the date of discharge and contains no information on stage of disease. Given these restrictions, a meaningful analysis of long-term surgical outcome was not practical from the available data. Second, although the case mix was limited to surgical procedures incorporating oophorectomy, the possibility that some patients may not have been undergoing their initial surgical effort for ovarian cancer cannot be entirely excluded. Despite these limitations, the current data highlight several important aspects of primary surgical care for women age 80 years and older that may have implications for improving the delivery of health services to this population. Specifically, short-term outcome analyses indicate that primary surgical care for ovarian cancer in women aged 80 years is associated with utilization of significant health care resources while providing worse short-term outcomes compared to younger women. Additional research is needed to identify opportunities for improving the costeffectiveness of care in this population.

Acknowledgment This work was supported by the Elizabeth Frost Ovarian Cancer Research Fund.

References [1] Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A, et al. Cancer statistics 2005. CA J Clin 2005;55:10 – 30. [2] L.A.G. Ries, M.P. Eisner, C.L. Kosary, B.K. Hankey, B.A. Miller, L. Clegg, et al. editors. SEER cancer statistics review, 1975 – 2001, National Cancer Institute, Bethesda, MD. Bethesda, MD: http://www.seer.cancer.gov/csr/1975_2001/. [3] Ozols RF, Rubin SC, Thomas GM, Robboy JS. Epithelial ovarian cancer. In: Hoskins WJ, Perez CA, Young RC, Barakat RB, Markman M, Randall ME, editors. Principles and practice of gynecologic oncology. Philadelphia’ Lippincott Williams and Wikins; 2005. p. 895 – 987. [4] Fentiman IS. Denying surgery to elderly people with cancer. Oncol Pract 1998;2:12. [5] Fentiman IS, Tirelli U, Monfardini S, et al. Cancer in the elderly: why so badly treated? Lancet 1990;335:1020 – 2. [6] Ries LAG. Ovarian cancer: survival and treatment differences by age. Cancer 1993;71(2):524 – 9. [7] Berkman B, Rohan B, Sampson S. Myths and biases related to cancer in the elderly. Cancer 1994;74(7):2004 – 8. [8] Balducci L, Extermann M. Cancer and aging. An evolving panorama. Hematol/Oncol Clin North Am 2000;14:1 – 16. [9] Yancik R, Ries LA. Aging and cancer in America. Demographic and epidemiologic perspectives. Hematol/Oncol Clin North Am 2000;14:17 – 23. [10] Murray CJ, Lopez AD. Alternative projections of mortality and

T.P. Dı´az-Montes et al. / Gynecologic Oncology 99 (2005) 352 – 357

[11] [12] [13] [14]

disability by cause 1990 – 2020: global burden of disease study. Lancet 1997;349:1498 – 504. World Health Organization. International classification of diseases. 9th Revision. Geneva’ WHO; 1975. Cox DR, Snell EJ. The analysis of binary data. 2nd ed. London’ Methuen; 1989. SAS Institute Inc. SAS user’s guide: statistics. 5th ed. Cary (NC)’ SAS 15 Institute Inc; 1985. Yancik RM, Ries L. Cancer in older persons. Magnitude of the problem. How do we apply what we know? In: Balducci L, Lyman GH, Ershler WB, editors. Comprehensive geriatric oncology. Armsterdam’ Harwood Academic Publishers; 1998. p. 95 – 104.

357

[15] Edwards B, Howe H, Ries L, et al. Annual report to the nation on the status of cancer, 1973 – 1999, featuring implications of age and aging on US cancer burden. Cancer 2002;94:2766 – 92. [16] Hightower RD, Nguyen HN, Averette HE, Hoskins W, Harrison T, Steven A. National survey of ovarian carcinoma IV: patterns of care and related survival for older patients. Cancer 1994;73:377 – 83. [17] Bruchim I, Altaras M, Fishman A. Age contrasts in clinical characteristics and pattern of care in patients with epithelial ovarian cancer. Gynecol Oncol 2002;86:274 – 8. [18] Cloven NG, Manetta A, Berman ML, Kohler MF, DiSaia PJ. Management of ovarian cancer in patients older than 80 years of age. Gynecol Oncol 1999;73:137 – 9.