Journal of the National Cancer Institute Advance Access originally published online on March 11, 2008
JNCI Journal of the National Cancer Institute 2008 100(6):399-406; doi:10.1093/jnci/djn033
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ARTICLES |
Specialized Care and Survival of Ovarian Cancer Patients in The Netherlands: Nationwide Cohort Study
Affiliations of authors: Departments of Gynecologic Surgery and Oncology (FV, APMH) and Medical Oncology (POW) and Julius Center for Health Sciences and Primary Care (YvdG), University Medical Center Utrecht, Utrecht, The Netherlands; Comprehensive Cancer Center Middle Netherlands, Utrecht, The Netherlands (MvdHvdL); Department of Public Health, Erasmus University Medical Center, Rotterdam, The Netherlands (JWC)
Correspondence to: Flora Vernooij, MD, Department of Gynecologic Surgery and Oncology, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands (e-mail: f.vernooij{at}umcutrecht.nl).
 |
ABSTRACT |
|---|
 Top Abstract Context and CaveatsMaterials and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Background: There is much debate on the necessity of regionalization of ovarian cancer care. We investigated the association between hospital type and survival of patients with ovarian cancer in The Netherlands.
Methods: A retrospective, population-based cohort study was performed on all Dutch patients diagnosed with ovarian cancer from January 1, 1996, through December 31, 2003. We used data from the Netherlands Cancer Registry that were linked to mortality data from the Statistics Netherlands database to obtain the date and cause of death. Five-year relative survival ratios, defined as the ratio of the observed survival in the patient population to the expected survival of women in the general population with the same age, were determined for the total population and for groups stratified by tumor stage and/or hospital type. The association between hospital type and disease-specific survival was analyzed by use of multivariable Cox regression analyses.
Results: We analyzed data from 8621 women with epithelial ovarian cancer, of whom 3482 (40%) were treated in general hospitals, 3510 (41%) were treated in semispecialized hospitals, and 1557 (18%) were treated in specialized hospitals. Five-year relative overall survival ratios of patients treated in general, semispecialized, and specialized hospitals were 38.0% (95% confidence interval [CI] = 36.0% to 39.9%), 39.4% (95% CI = 37.5% to 41.4%), and 40.3% (95% CI = 37.4% to 43.1%), respectively; median survival of patients aged 50–75 years was 36 months (interquartile range [IQR] = 13 to >54 months), 37 months (IQR = 14 to >54 months), and 38 months (IQR = 15 to >55 months), respectively. Age and cancer stage were associated with the relationship between hospital type and ovarian cancer–specific survival but histologic tumor type, grade, year of diagnosis, and socioeconomic status were not. Among patients with early-stage ovarian cancer, treatment in semispecialized and specialized hospitals was associated with lower risks of ovarian cancer–specific mortality than treatment in general hospitals. Among patients with stage I–IIA disease who were aged 50–75 years, risk of ovarian cancer–specific mortality was 30% and 42% lower after treatment in semispecialized and specialized hospitals, respectively, than in general hospitals (for semispecialized hospitals, hazard ratio [HR] = 0.70, 95% CI = 0.53 to 0.93; for specialized hospitals, HR = 0.58, 95% CI = 0.38 to 0.87). Among patients with advanced ovarian cancer, hospital type was not associated with survival.
Conclusion: Hospital type was statistically significantly associated with survival among Dutch ovarian cancer patients with early-stage ovarian cancer: Patients who were treated in specialized and semispecialized hospitals survived longer than patients treated in general hospitals.
Prior knowledge Some countries have developed referral guidelines that recommend that ovarian cancer patients be treated in specialized centers because treatment in such hospitals has been shown to result in better survival. Study design Retrospective, population-based cohort study performed on all Dutch patients diagnosed with ovarian cancer from January 1, 1996, through December 31, 2003, investigating the association between hospital type (general, semispecialized, and specialized) and disease-specific survival. Contribution Hospital type was statistically significantly associated with survival among patients with early-stage ovarian cancer in The Netherlands. Ovarian cancer patients with early-stage cancer who were treated in specialized or semispecialized hospitals survived longer than patients treated in general hospitals. Among patients with advanced ovarian cancer, no association was found between hospital type and survival. Implications The level of collaboration between general hospitals and semispecialized or specialized hospitals during the study period was not sufficient to deliver optimal care to all Dutch patients with ovarian cancer. Regionalized care thus seems necessary. Limitations Adjustment for comorbidity could not be done, so it is possible that more patients without severe comorbid conditions were referred to specialized hospitals than patients with comorbid conditions.
|
Treatment of ovarian cancer patients in a specialized hospital has been shown to result in longer survival (1–5). In some countries, this observation has led to the development of referral guidelines that recommend that patients with ovarian cancer be treated in (regional) specialized centers. However, there is much debate on the necessity of regionalization. Referral may not be needed when the multidisciplinary teams from specialized hospitals collaborate closely with general hospitals. Furthermore, general gynecologists with a special interest in gynecologic oncology, who operate on all patients with ovarian cancer in their clinic and work with gynecologic oncologists from regional specialized centers, might also provide adequate care to ovarian cancer patients.
In The Netherlands, the treatment of ovarian cancer is not regionalized, and a minority of the Dutch patients with ovarian cancer are treated in specialized hospitals. In some parts of The Netherlands, multidisciplinary teams are involved in the treatment of ovarian cancer patients, and, in many hospitals, surgery is performed by general gynecologists with a special interest in gynecologic oncology (ie, semispecialized gynecologists). As in other countries, there is an ongoing debate in The Netherlands about implementation of the results from the literature on regionalized care for ovarian cancer patients. The debate is hampered by the fact that it is not known whether collaboration and semispecialization lead to optimal care because the results for Dutch ovarian cancer care are not known. Therefore, we conducted a population-based study to investigate the magnitude of the association between hospital type and ovarian cancer–specific survival of ovarian cancer patients in The Netherlands.
|
Materials and Methods |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Design
We performed a retrospective cohort study on all patients with ovarian carcinoma who were newly diagnosed from January 1, 1996, through December 31, 2003, in The Netherlands. Patients were categorized according to the hospital in which the initial treatment took place.
The primary outcome measure was disease-specific survival that was defined as the interval from the date of diagnosis to the date of death from ovarian cancer, as registered in the database of Statistics Netherlands. When the patient died from another cause, she was censored at the date of death. If the patient had not died by December 31, 2005, she was censored on that date.
Databases and Linkage
Data on patients with ovarian cancer from the population-based Netherlands Cancer Registry were linked to mortality data from the database of Statistics Netherlands. The data from the Netherlands Cancer Registry are gathered by the Comprehensive Cancer Centers that receive lists of all patients with newly diagnosed ovarian cancer from pathology departments and from the National Registry of Hospital Discharge Diagnosis. After notification, trained registrars from the Comprehensive Cancer Centers extract relevant information from the medical records of the newly diagnosed patients. In this way, data from more than 98% of the eligible cancer patients are recorded (6).
We excluded patients with ovarian tumors of borderline malignancy. We restricted our cohort to patients with epithelial ovarian malignancies with the International Classification of Diseases for Oncology (ICD-O) 2 and 3 topography codes C56, C48.1, and C48.2 and ICD-O morphology codes 8470, 8471, 8480, 8481, 8482, 8380, 8381, 8382, 8570, 8441, 8443, 8460, 8461, 8310, 8313, 8140, 8211, 8260, 8263, 8323, 8440, 8450, 8010, 8012, 8013, 8020, 8021, 8033, 8041, 8045, 8070, 8071, 8076, 8130, 8230, 8255, 8560, 8574, 8934, and 9100.
The following information was registered: hospital in which the patient was diagnosed, hospital in which the patient had surgery, the patient' s date of birth, the patient's postal code, tumor–node–metastasis (TNM) stage of the tumor, histology of the tumor, grade of the tumor, operative treatment (yes or no), treatment with chemotherapy (yes or no), and the date of death (when available). When the operative procedure was limited to taking a biopsy specimen, this procedure was registered as no operation. Two of the nine Comprehensive Cancer Centers also provided data on the specialty of the gynecologist who performed the first operation.
Statistics Netherlands registers vital status, including, if applicable, date of death of a person as recorded by the municipal registries. Statistics Netherlands also documents the cause of death as recorded by the physician who determined it. The ICD-10 codes that we used to determine ovarian cancer–related causes of death were C54, C56, C57, C76.2, C76.3, C80, and R990. In addition, information on the socioeconomic status was provided by Statistics Netherlands and was based on the average household income and the housing prices in the residential area of the patients.
Linkage of Netherlands Cancer Registry and Statistics Netherlands data was performed by use of postal codes and dates of birth. If the patient could not be found in the Statistics Netherlands files by use of postal code and date of birth, the date of death, if available in the Netherlands Cancer Registry, was additionally used to track the patient in the Statistics Netherlands files.
Hospitals
The hospital in which the patient was diagnosed with ovarian cancer (hospital of diagnosis) and the hospital in which the patient was treated (hospital of treatment) were recorded in the cancer registry. The hospital of diagnosis is the place in which the definite diagnosis was made by use of imaging studies and/or pathologic specimens. The hospital of treatment is the place in which surgical or medical therapy was started. Patients were categorized according to the hospital in which treatment was performed. If the patients were not treated surgically or if the hospital of treatment was not known, patients were categorized by the hospital of diagnosis.
Hospitals were classified according to their level of specialization as a general, semispecialized, or specialized hospital. In total, there were 60 general hospitals, 32 semispecialized hospitals, and 13 specialized hospitals. The staff in specialized hospitals includes gynecologic oncologists, and specialized hospitals are regional centers for gynecologic oncologic care. A physician is recognized as a gynecologic oncologist by the Dutch Society of Gynecologic Oncology when he or she has received subspecialty training in gynecologic oncology during a 2-year fellowship. Of the 13 clinics that fulfilled the criteria for a specialized center, eight were university hospitals. Community hospitals with a semispecialized gynecologist were categorized as semispecialized hospitals. Semispecialized gynecologists are not trained in oncology but operate on most ovarian and endometrial cancer patients in large community (usually teaching) hospitals. General hospitals without semispecialized oncologic care were classified as general hospitals. The specialization level of the institution (general, semispecialized, or specialized) was verified by a panel of gynecologists and gynecologic oncologists. The hospitals were not additionally characterized according to teaching status because almost all general hospitals were nonteaching hospitals and almost every semispecialized institution and all specialized institutions were teaching hospitals. In The Netherlands, all chemotherapy is provided by medical oncologists in hospital settings.
Data Analysis
The stages of disease were converted from TNM to International Federation of Gynecology and Obstetrics (FIGO) stages (7). The following histologic groups were distinguished: serous, mucinous, endometrioid, clear cell, adenocarcinoma not otherwise specified, and other. The statistical significances of differences in patient and treatment characteristics among hospitals were tested with 
2 analysis.
Five-year relative survival ratios were determined for the total population and for groups stratified by tumor stage and/or hospital type. The relative survival ratio is the ratio of the observed survival in the patient population to the survival that would have been expected had the patients experienced only the age- and period-specific mortality of the general population from which they were drawn (8). All-cause mortality was the endpoint for the analysis of relative survival ratios, specifically. The general population in this study was all women in The Netherlands. We used a generalized linear model with a Poisson error structure that was based on collapsed data and exact survival times (9). Relative 5
year survival was computed with SAS statistical software (version 8.2, SAS Institute, Cary, NC).
For multivariable survival analyses and the assessment of median survival, ovarian cancer–specific survival was used as the endpoint. Median survival was estimated by use of actuarial life tables and was stratified by age group, stage, and hospital type. Interquartile ranges (IQRs) are shown as the time at which 75% of the patients were alive and the time at which 25% of the patients were alive. To allow for statistical adjustments for prognostic variables (ie, stage, age, histologic tumor type, grade of tumor, year of diagnosis, and socioeconomic status), a Cox proportional hazards model was used, and hazard ratios (HRs) with their corresponding 95% confidence intervals (CIs) were calculated. The proportional hazards assumption was confirmed by visual inspection of the curves plotting survival of patients treated in different hospital types. We also examined whether age and stage modified the association between hospital type and survival by introducing interaction terms into the model. Data were analyzed primarily by the hospital of treatment but were also analyzed by the hospital of diagnosis. The association between hospital of treatment and survival of patients with advanced disease was also analyzed in patients who had received both surgical treatment and chemotherapy. In addition, we separately examined the data of the patients from the two Comprehensive Cancer Center regions (in this study called regions X and Y) that provided information on the specialty of the operating gynecologist. These analyses were conducted with SPSS software (version 11.5, SPSS Inc., Chicago, IL). All statistical tests were two-sided.
|
Results |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Characteristics of the Study Population
Among the 9932 ovarian cancer patients who were newly diagnosed from January 1, 1996, through December 1, 2003, 8915 (90%) had a primary epithelial tumor. Survival data were missing for 294 (3.3%) of the 8915 patients, and the hospital of treatment was not registered for 283 (3.3%) of the remaining 8621 patients. When the hospital of diagnosis was used to categorize these 283 patients, the hospital category remained unknown for 72 patients.
Characteristics of the 8621 patients with epithelial ovarian carcinoma whose vital status was known are shown in Table 1. Of these 8621 patients, 2750 (32%) presented with early-stage disease (FIGO I and II) and 6821 (79%) were younger than 75 years (Table 1). A total of 3482 (40%) patients were treated in general hospitals, 3510 (41%) patients were treated in semispecialized hospitals, and 1557 (18%) patients were treated in specialized hospitals (Table 1). For the whole cohort, stage of disease at diagnosis did not differ between the different hospital types. However, among patients younger than 70 years, 645 (31%) of the 2093 patients in general hospitals had stage I disease vs 280 (25%) of the 1107 patients in specialized hospitals (663 [30%] of 2188 patients in semispecialized hospitals). Also among patients younger than 70 years, 889 (41%) of the 2188 patients in semispecialized hospitals had stage III disease vs 497 (45%) of the 1107 patients treated in specialized hospitals (916 [44%] of the 2093 patients in general hospitals). In the other age and stage groups, no statistically significantly differences in stage among the three hospital types were observed (results not shown). The distributions of the histologic tumor types were also similar among the hospital groups. In specialized hospitals, relatively fewer patients with middle-level socioeconomic status were treated (548 [35%] of 1557 patients) than in general and semispecialized hospitals (1373 [39%] of 3482 and 1300 [37%] of 3510 patients, respectively). Only 348 (4%) of the 8621 patients were diagnosed and treated in different types of hospitals (Table 1). At least 244 (16%) of the 1557 patients treated in specialized hospitals had been diagnosed in general or semispecialized hospitals. Patients older than 75 years, with either early- or advanced-stage ovarian cancer, were almost never referred to specialized hospitals, whereas patients younger than 75 years with advanced disease were the largest group of referred patients (results not shown). Patients with stage II or more advanced disease who were treated in general hospitals were less likely to undergo surgery and receive chemotherapy than similar patients treated in semispecialized and specialized hospitals.
|
Association Between Hospital Type and Survival
The mean follow-up time was 3 years (range = 0–10 years). Survival increased as the level of specialization of the hospital increased: 5-year relative survival ratios of patients treated in general, semispecialized, and specialized hospitals were 38.0% (95% CI = 36.0% to 39.9%), 39.4% (95% CI = 37.5% to 41.4%), and 40.3% (95% CI = 37.4% to 43.1%) (Table 2), respectively; median survival of patients aged 50–75 years was 36 months (IQR = 13 to >54 months), 37 months (IQR = 14 to >54 months), and 38 months (IQR = 15 to >55 months), respectively (Table 3).
|
|
In multivariable analysis, stage and age were statistically significantly associated with hospital type and ovarian cancer–specific survival and modified this relationship (P for interaction = .001 and .002, respectively), whereas histologic tumor type, grade, year of diagnosis, and socioeconomic status were not associated with the relationship between hospital type and survival (data not shown). Patients aged 50–75 years with stage I–IIA ovarian cancer who were treated at semispecialized and specialized hospitals had a 30% and 42% lower risk, respectively, of dying from ovarian cancer (for semispecialized hospitals compared with general hospitals, HR = 0.70, 95% CI = 0.53 to 0.93; for specialized hospitals compared with general hospitals, HR = 0.58, 95% CI = 0.38 to 0.87) (Figure 1). Likewise, among patients older than 75 years with the same stage of disease, treatment in semispecialized and specialized hospitals was associated with better survival (for semispecialized hospitals compared with general hospitals, HR = 0.63, 95% CI = 0.43 to 0.94; for specialized hospitals compared with general hospitals, HR = 0.57, 95% CI = 0.30 to 1.1). The favorable outcome associated with treatment in semispecialized and specialized hospitals as compared with general hospitals was also observed among patients older than 50 years with stage IIB–IIIA ovarian cancer. Among patients with more advanced disease, there was little evidence that treatment in semispecialized and specialized hospitals was associated with improved survival. For example, among patients aged 50–75 years with stage IIIC ovarian cancer, treatment in specialized hospitals was associated with a hazard ratio of ovarian cancer death of 0.89 (95% CI = 0.77 to 1.03) compared with general hospitals, and treatment in semispecialized hospitals was associated with a hazard ratio of 0.91 (95% CI = 0.81 to 1.03) compared with general hospitals (Figure 1). Among a subgroup of these patients who received both surgical treatment and chemotherapy, the association between survival and hospital type was further reduced (for treatment in semispecialized hospitals, HR = 0.99, 95% CI = 0.85 to 1.14; for specialized hospitals, HR = 0.99, 95% CI = 0.84 to 1.18). Categorization of patients according to hospital of diagnosis did not alter the results substantially (data not shown).
|
Association Between Type of Gynecologist and Survival
In regions that provided data on gynecologist specialties, a gynecologic oncologist performed the surgery for 254 (27%) of the 925 patients treated in general hospitals, 56 (8%) of the 686 patients treated in semispecialized hospitals, and 160 (68%) of the 236 patients treated in specialized hospitals. Associations between survival and the type of hospital and gynecologist differed by region (Table 4). Nevertheless, treatment by a consulting gynecologic oncologist was consistently associated with better survival than treatment by a general gynecologist in a general hospital (Table 4), even when the analysis of the effect of type of gynecologist was restricted to the patients receiving surgery (results not shown).
|
|
Discussion |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionFundingReferencesNotes |
|---|
This population-based study showed that treatment of early-stage ovarian cancer patients in specialized and in semispecialized Dutch hospitals was associated with longer survival. These results are in accord with previous reports that patients with ovarian cancer who are treated in specialized hospitals receive adequate treatment more often (1,4,10–13) and survive longer (1–5) than patients treated in general, nonspecialized hospitals. In many countries, including The Netherlands, results of such studies have led to a policy discussion about instituting regionalized care for ovarian cancer patients. In The Netherlands, approximately 1100 patients are diagnosed with ovarian cancer each year and treated in nearly all 105 hospitals throughout the country (6). These 105 hospitals, which are of all three types, collaborate through the framework provided by the nine regional Comprehensive Cancer Centers. Collaboration consists of treatment advice and sometimes surgery by a consultant gynecologic oncologist in general and semispecialized hospitals. The Comprehensive Cancer Center North was one of the first regions to systematically provide assistance in treatment of ovarian cancer patients; this assistance has resulted in 42% of the patients in general hospitals being operated on by a specialized gynecologist (14). Gynecologic oncologists often argue that care for ovarian cancer patients should become more regionalized, but many general gynecologists claim that the present system suffices. We performed this study to provide data for this discussion.
We found that patients with early-stage ovarian cancer who were aged 50–75 years had a 30%–42% lower risk of dying from ovarian cancer after treatment in a semispecialized or specialized hospital than a general hospital. This association likely results from better surgical staging and thus less undertreatment in these hospitals. Indeed, a recent study of a sample of 1077 Dutch ovarian cancer patients showed that only one-fourth of the patients treated in general hospitals received an adequate surgical staging procedure, compared with 60% of the patients treated in semispecialized and specialized hospitals (F. Vernooij, MD, A.P.M. Heintz, MD, PhD, J.W. Coebergh, MD, PhD, L.F.A.G. Massuger, MD, PhD, P.O. Witteveen, MD, PhD, Y. Van der Graaf, MD, PhD, unpublished data, 2008).
According to the literature (4,15), differences in survival between specialized and nonspecialized hospitals appear to be largest for patients with stage IIIC ovarian cancer. Junor et al. (15) found that survival for stage III patients treated by gynecologic oncologists was better than survival for patients treated by general gynecologists (adjusted HR = 0.75, 95% CI = 0.62 to 0.92). In addition, Paulsen et al. (4) reported that survival for stage IIIC patients who were treated in teaching hospitals was better than that for patients treated in nonteaching hospitals (adjusted HR = 0.54, 95% CI = 0.33 to 0.90). In stage IIIC ovarian cancer, the tumor has spread throughout the whole abdomen and often presents as bulky disease, which is hard to debulk adequately. Optimal debulking, which is an important prognostic factor (16), is more often achieved in specialized hospitals than in general, nonspecialized hospitals (1,4,10,12). We found that the risk of mortality was approximately 10% lower after treatment in semispecialized or specialized hospitals than in general hospitals; however, this difference was not statistically significant. We hypothesize that this observation is attributable in part to gynecologic oncologists from specialized hospitals sometimes assisting in surgery in general and semispecialized hospitals, thereby reducing any differences in debulking results among hospital types. It is also possible that, in The Netherlands, the differences in surgical outcomes between general and specialized gynecologists might not be large enough to result in a detectable survival difference. The extent to which surgical therapy was adequate could not be discerned in this dataset, although we did find that gynecologists in general hospitals refrained from surgical therapy or limited the operative procedure to taking a biopsy more often than gynecologists in semispecialized or specialized hospitals. The adequacy of staging and debulking by Dutch gynecologists requires further investigation.
When we restricted the analysis to patients with advanced ovarian cancer who received both surgical treatment and chemotherapy, the differences between hospitals decreased. The high number of patients who did not receive treatment in general hospitals thus could explain the modest improvement in survival that was observed for patients with stage IIIC disease in specialized hospitals. Patients may not have received treatment because they were in poor medical condition and thus had a poor a priori prognosis. Differences in percentages of patients not receiving treatment (and, thus, differences in survival) could result from the referral of patients who are in good condition, but not those in poor condition, to the specialized hospitals, thereby introducing confounding by indication. Alternatively, patients may not have undergone treatment because their physician judged that the tumor was not resectable or that the patient would not benefit from therapy. The selection of patients who would benefit from therapy and who have resectable tumors is subjective and a part of the treatment. A specialized gynecologist may have a different opinion about the effect of therapy on a patient's prognosis or the resectability of her tumor than a general gynecologist.
In addition to investigating the association between the hospital of treatment and survival, we investigated the association between the hospital of diagnosis and survival because adequate referral can also be considered to be a part of the treatment. If no survival differences would be found between the three types of hospitals of diagnosis, referral guidelines would be redundant because patient referral would be adequate. However, semispecialized and specialized hospitals still had lower risks of mortality among patients with early-stage disease, and so more patients should probably be referred to such hospitals.
Our study had several limitations. We could not adjust for comorbidity. It is possible that patients without severe comorbid conditions were referred to specialized hospitals because they could endure aggressive therapy, whereas patients who were in a poor medical condition were not referred. By adjusting for comorbidity, we could have reduced the potential effect of confounding by indication. However, the strongest association between hospital type and survival was among patients with early-stage disease, who are usually in a good medical condition. Furthermore, only approximately 4% of the total population had been referred, and it is thus unlikely that confounding by indication fully explains our results.
Another limitation is that stage migration could be part of the explanation of the differences in survival found in early-stage patients. However, when the results were not split by stage, patients treated in semispecialized and specialized hospitals still survived longer than patients treated in general hospitals.
We did not know exactly which patients in general and semispecialized hospitals had undergone surgery by consulting gynecologic oncologists; consequently, the survival associated with treatment in a specialized hospital might be even higher than we reported. In the two Comprehensive Cancer Center regions that contributed this information, 530 (29%) of the 1847 patients were treated by (consulting) gynecologic oncologists. The risk of mortality was statistically significantly lower among these 530 patients than among patients treated by a general gynecologist; this result confirms previous findings from one of these regions (14). Thus, in our nationwide dataset, gynecologic oncologists from a specialized hospital performed surgery on some patients in general and semispecialized hospitals, thereby reducing the differences in survival between the hospital types.
One of the strengths of this study is its size; to our knowledge, this is the largest population-based study ever published on the association between hospital type and survival of ovarian cancer patients. Using the data of the population-based Dutch Cancer Registry allowed us to include more than 98% of the eligible patients. The size of the population thus provided the necessary statistical power to analyze relevant subgroups. In addition, we could classify the hospitals into three categories that adequately represent the different levels of specialization of Dutch hospitals. Our finding that 5-year and median survival increased as the level of specialization of the hospital increased supports the association between specialization and survival. Furthermore, follow-up was complete for 97% of the patients in the study population. Another strong point of our study is its generalizability. Although we studied the results obtained in Dutch hospitals, the care system in The Netherlands is not unique. In many countries, more than two levels of hospital specialization can be distinguished (2,10,17). Furthermore, also in other countries, collaboration and semispecialization are viewed as viable alternatives for regionalization (18).
Treatment of ovarian cancer patients in Dutch specialized and semispecialized teaching hospitals results in longer survival than treatment in general hospitals in large subgroups of patients. This result indicates that the level of collaboration during the study period did not suffice to deliver optimal care to all Dutch ovarian cancer patients, and regionalization of the care for such patients thus seems necessary.
|
Funding |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Netherlands Organisation for Health Research and Development (ZonMW; project no. 945-06-216); Foundation Reserves Voormalige Vrijwillige Ziekenfondsverzekeringen (RVVZ; project no. 763).
|
NOTES |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionFundingReferencesNotes |
|---|
The authors would like to acknowledge the Dutch Comprehensive Cancer Centers, the Netherlands Cancer Registry and Statistics Netherlands for providing the data, and Maryska Janssen-Heijnen and Ingeborg Deerenberg for their assistance, and the members of the support group of the Dutch Comprehensive Cancer Centers for their critical evaluation of the data.
ZonMW and RVVZ have had no role in the study design; the collection, analysis, and interpretation of the data; or the writing or submission of the manuscript.
|
REFERENCES |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionFundingReferencesNotes |
|---|
1. Tingulstad S, Skjeldestad FE, Hagen B. The effect of centralization of primary surgery on survival in ovarian cancer patients. Obstet Gynecol (2003) 102(3):499–505.
2. Wolfe CD, Tilling K, Raju KS. Management and survival of ovarian cancer patients in south east England. Eur J Cancer (1997) 33(11):1835–1840.[CrossRef][ISI][Medline]
3. Oberaigner W, Stühlinger W. Influence of department volume on cancer survival for gynaecological cancers. A population-based study in Tyrol, Austria. Gynecol Oncol (2006) 103(2):527–534.[CrossRef][ISI][Medline]
4. Paulsen T, Kjaerheim K, Kaern J, Tretli S, Tropé C. Improved short-term survival for advanced ovarian, tubal, and peritoneal cancer patients operated at teaching hospitals. Int J Gynecol Cancer (2006) 16(suppl1):11–17.[CrossRef][ISI][Medline]
5. Schrag D, Earle C, Xu F, et al. Associations between hospital and surgeon procedure volumes and patient outcomes after ovarian cancer resection. J Natl Cancer Inst (2006) 98(3):163–171.
6. Visser O, Siesling S, van Dijck JAAM, eds. Incidence of Cancer in The Netherlands 1999/2000 (2003) Utrecht, The Netherlands: Vereniging van Integrale Kankercentra.
7. Hermanek P, Sobin LH, eds. TNM Classification of Malignant Tumours (1998) Berlin, Germany: International Union Against Cancer, Springer-Verlag. 111–114.
8. Ederer F, Axtell LM, Cutler SJ. The relative survival rate: a statistical methodology. Natl Cancer Inst Monogr (1961) 6:101–121.[Medline]
9. Dickman PW, Sloggett A, Hills M, Hakulinen T. Regression models for relative survival. Stat Med (2004) 23(1):51–64.[CrossRef][ISI][Medline]
10. Kumpulainen S, Kuoppala T, Leminen A, et al. Surgical treatment of ovarian cancer in different hospital categories—a prospective nation-wide study in Finland. Eur J Cancer (2006) 42(3):388–395.[CrossRef][ISI][Medline]
11. Goff BA, Matthews BJ, Wynn M, Muntz HG, Lishner DM, Baldwin LM. Ovarian cancer: patterns of surgical care across the United States. Gynecol Oncol (2006) 103(2):383–390.[CrossRef][ISI][Medline]
12. Olaitan A, Weeks J, Mocroft A, Smith J, Howe K, Murdoch J. The surgical management of women with ovarian cancer in the south west of England. Br J Cancer (2001) 85(12):1824–1830.[CrossRef][ISI][Medline]
13. Munoz KA, Harlan LC, Trimble EL. Patterns of care for women with ovarian cancer in the United States. J Clin Oncol (1997) 15(11):3408–3415.
14. Engelen MJ, Kos HE, Willemse PH, et al. Surgery by consultant gynecologic oncologists improves survival in patients with ovarian carcinoma. Cancer (2005) 106(3):589–598.[CrossRef][ISI]
15. Junor EJ, Hole DJ, Gillis CR. Management of ovarian cancer: referral to a multidisciplinary team matters. Br J Cancer (1994) 70(2):363–370.[ISI][Medline]
16. Bristow RE, Tomacruz RS, Armstrong DK, Trimble EL, Montz FJ. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. J Clin Oncol (2002) 20(5):1248–1259.
17. Elit L, Bondy SJ, Paszat L, Przybysz R, Levine M. Outcomes in surgery for ovarian cancer. Gynecol Oncol (2002) 87(3):260–267.[CrossRef][ISI][Medline]
18. Akeson M, Zetterqvist BM, Holmberg E, Horvath G. Improved survival with clinical guidelines? Evaluation of a quality register linked to clinical guidelines for ovarian cancer in the western health care region in Sweden between 1 September 1993 and 1 June 1998. Acta Obstet Gynecol Scand (2005) 84(11):1113–1118.[CrossRef][ISI][Medline]
Manuscript received July 13, 2007; revised January 3, 2008; accepted January 23, 2008.
Editorial about this Article
Related Article in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2008 100: 378-379.
J Natl Cancer Inst 2008 100: 375.
This article has been cited by other articles:
|
|
 |
|
  Does Specialized Care Improve Ovarian Cancer Outcomes? Journal Watch Women's Health, May 8, 2008; 2008(508): 2 - 2. [Full Text]
|
|
|
|
 |
|
 D. Schrag Enhancing Cancer Registry Data to Promote Rational Health System Design J Natl Cancer Inst, March 19, 2008; 100(6): 378 - 379. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||