Variations in treatment policies and outcome for bladder cancer in the Netherlands.

AIM: To describe the population-based variation in treatment policies and outcome for bladder cancer in the Netherlands. METHODS: All newly diagnosed patients with primary bladder cancers during 2001-2006 were selected from the Netherlands Cancer Registry (n = 29,206). Type of primary treatment was analysed according to Comprehensive Cancer Centre region, hospital type (academic, non-academic teaching or other hospitals) and volume (< or =5, 6-10 or >10 cystectomies yearly). For stage II-III patients undergoing cystectomy we analyzed the proportion of lymph node dissections and 30-days mortality. RESULTS: 44% of patients with stage II-III bladder cancer underwent cystectomy, while 26% were not treated with curative intent. Cystectomy was the preferred option in three of nine regions, radiotherapy in two, and two regions waived curative treatment more often. Between 2001 and 2006 the number of cystectomies increased with 20% (n = 108). Twenty-one percent (n = 663) of these procedures were performed in 44 low-volume hospitals. In 79% of the cystectomies lymph node dissections were performed, more often in high and medium-volume centers (82% and 81% respectively) than in low-volume hospitals (71%, the odds ratio being 1.5). The overall 30-days post-operative mortality rate was 3.4% and increased with older age. It was significantly lower in high-volume centers (1.2%). CONCLUSION: Treatment policies for muscle-invasive bladder cancer in the Netherlands showed regional preferences and a gradual increase of cystectomy. Cystectomy albeit considered as golden standard, was performed in a minority of the muscle-invasive cases. In high-volume institutions, lymph node dissection rates were higher and post-operative mortality rates were lower.

Disparities in quality of care for colon cancer between hospitals in the Netherlands.

BACKGROUND: Aim of this study was to describe treatment patterns and outcome according to region, and according to hospital types and volumes among patients with colon cancer in the Netherlands. METHODS: All patients with invasive colon carcinoma diagnosed in the period 2001-2006 were selected from the Netherlands Cancer Registry. Logistic regression analyses were performed to examine the influence of relevant factors on the odds of having adequate lymph node evaluation, receiving adjuvant chemotherapy and postoperative mortality. Relative survival analysis was used to estimate relative excess risk of dying according to hospital type and volume. RESULTS: In total, 39 907 patients were selected. Patients diagnosed in a university hospital had a higher odds (OR 2.47; 95% CI 2.19-2.78) and patients diagnosed in a hospital with >100 colon carcinoma diagnoses annually had a lower odds (OR 0.70; 95% CI 0.64-0.77) of having >/=10 lymph nodes evaluated. The odds of receiving adjuvant chemotherapy was lower in patients diagnosed in teaching hospitals (OR 0.85; 95% CI 0.73-0.98) and university hospitals (OR 0.56; 95% CI 0.45-0.70) compared to patients diagnosed in non-teaching hospitals. Funnel plots showed large variation in these two outcome measures between individual hospitals. No differences in postoperative mortality were found between hospital types or volumes. Patients diagnosed in university hospitals and patients diagnosed in hospitals with >50 diagnoses of colon carcinoma per year had a better survival. CONCLUSIONS: Variation in treatment and outcome of patients with colon cancer in the Netherlands was revealed, with differences between hospital types and volumes. However, variation seemed mainly based on the level of the individual hospital.

Variation in management of early breast cancer in the Netherlands, 2003-2006.

BACKGROUND: To describe variation in staging and primary treatment by hospital characteristics including type and volume and region in patients with early breast cancer (BC) in the Netherlands, 2003-2006 after completion of national guidelines in 2002. METHODS: All patients newly diagnosed with invasive BC in 2003-2006 and recorded in the Netherlands Cancer Registry were included (n = 51 354). Multivariable logistic regression analyses examined the influence of patient and hospital characteristics, also by region, on type of breast surgery, axillary lymph node dissection (ALND), sentinel node procedure (SNP), and adjuvant irradiation and/or systemic treatment. RESULTS: Patients <40 years more often underwent breast conserving surgery (BCS) in general hospitals (OR 1.4 (95%CI 1.1-1.5)) than in teaching and academic hospitals, whereas patients of 40-69 years less often received BCS in an academic hospital (OR 0.9 (95%CI 0.8-1.0)) than in teaching hospitals. Patients with pT1-2N0 cancer more often underwent primary ALND in a general hospital than in a larger teaching or academic hospital. Type of hospital did not seem to affect utilization of adjuvant systemic therapy, but patient age and tumour size and grade did. Over time, patients more often received SNP, BCS, and adjuvant systemic therapy, primary ALND being on the decline, but with substantial regional variation between geographic regions. CONCLUSION: With detailed evidence-based national guidelines since 2002 the considerable regional and hospital variation in staging procedures and primary treatment among newly diagnosed patients with early breast cancer in the Netherlands decreased markedly, suggesting the presence of late adaptors rather than specific hospital characteristics.

Variation in treatment and outcome of patients with rectal cancer by region, hospital type and volume in the Netherlands.

BACKGROUND: Aim of this study was to describe treatment patterns and outcome according to region and hospital type and volume among patients with rectal cancer in the Netherlands. METHODS: All patients with rectal carcinoma diagnosed in the period 2001-2006 were selected from the Netherlands Cancer Registry. Logistic regression analyses were performed to examine the influence of relevant factors on the odds of receiving preoperative radiotherapy and on the odds of postoperative mortality. Relative survival analysis was used to estimate relative excess risk of dying according to hospital type and volume. RESULTS: In total, 16 039 patients were selected. Patients diagnosed in a teaching or university hospital had a lower odds (OR 0.85; 95% CI 0.73-0.99 and OR 0.70; 95% CI 0.52-0.92) and patients diagnosed in a hospital performing >50 resections per year had a higher odds (OR 1.95; 95% CI 1.09-1.76) of receiving preoperative radiotherapy. A large variation between individual hospitals in rates of preoperative radiotherapy and between Comprehensive Cancer Centre-regions in the administration of preoperative chemoradiation was revealed. Postoperative mortality was not correlated to hospital type or volume. Patients with T1-M0 tumours diagnosed in a hospital with >50 resections per year had a better survival compared to patients diagnosed in a hospital with <25 resections per year (RER 0.11; 95% CI 0.02-0.78). CONCLUSION: This study demonstrated variation in treatment and outcome of patients with rectal cancer in the Netherlands, with differences related to hospital volume and hospitals teaching or academic status. However, variation in treatment patterns between individual hospitals proved to be much larger than could be explained by the investigated characteristics. Future studies should focus on the reasons behind these differences, which could lead to a higher proportion of patients receiving optimal treatment for their stage of the disease.

The Quality of Cancer Care initiative in the Netherlands.

BACKGROUND: In 2007 the Dutch Cancer Society formed a 'Quality of Cancer Care' taskforce comprising medical specialists, from all disciplines involved in the care for cancer patients. This taskforce was charged with the evaluation of quality of cancer care in the Netherlands and the development of strategies for improvement. OBJECTIVE: The experts first focused on the relation between procedural volume and patient outcome and later aimed to identify other factors associated with high and low quality of the care provided in different regions and (types of) hospitals in the Netherlands. The question if cancer care in the Netherlands could be organized differently to assure high quality of care for all patients, was the main subject of investigation. METHODS: An extensive review of the literature on infrastructure, volume and specialization on the one hand and outcome on the other was performed. In addition, a meta-analysis of the volume-outcome relationship for pancreatectomies, bladder, lung, colorectal and breast cancer resections was performed. Finally, variation in quality of cancer care between regions, groups of hospitals and individual hospitals in our country was investigated on data from the Netherlands Cancer Registry. RESULTS: In the Netherlands quality of care varies by hospital and region. These differences are not limited to surgical procedures and postoperative mortality, but are also demonstrated in other parts of the care process. Differences are only partly explained by differences in infrastructure, procedural volume and specialization between hospitals. Essential information on differences in case mix between these hospitals are lacking from the Netherlands Cancer Registry. More detailed clinical data are needed to reveal the mechanisms behind the differences in quality of care between Dutch hospitals. CONCLUSION: On a population level, there is potential for improvement of outcome for cancer patients in the Netherlands by reducing variation in optimal treatment rates between hospitals. Not only treatment of tumours with a low incidence but also other complex or high risk cancer procedures should be provided in a specialized setting, with the right infrastructure, sufficient volume and adequate expertise. In addition, outcomes should be monitored continuously and fed back to individual caregivers.

Improved diagnosis and treatment of soft tissue sarcoma patients after implementation of national guidelines: a population-based study.

AIM: The majority of clinicians, radiologists and pathologists have limited experience with soft tissue sarcomas. In 2004, national guidelines were established in The Netherlands to improve the quality of diagnosis and treatment of these rare tumours. This study evaluates the compliance with the guidelines over time. PATIENTS: Population-based series of 119 operated patients with a soft tissue sarcoma (STS) diagnosed in 1998-1999 (79 before implementation of new guidelines) and in 2006 (40 after implementation). METHODS: Coded information regarding patient and tumour characteristics as well as (the results of) pathology review was collected from the medical patient file by two experienced data-managers. RESULTS: Diagnostic imaging of the tumour was performed according to the guidelines in 75-100% depending on the site of the tumour (abdominal versus non-abdominal) as well as the time of diagnosis. Adherence to the guidelines with respect to invasive diagnostic procedures in patients with non-abdominal STS improved over time. A pre-operative histological diagnosis was obtained in 42% of the patients in 1998-1999 and in 72% of the patients in 2006 (p<0.001). The guidelines for reporting on pathology were increasingly adhered to. In 2006, (nearly) all pathology reports mentioned tumour size, morphology, tumour grade, resection margins and radicality. This represents a major improvement compared to the pathology reports in 1998-1999, where these aspects were not mentioned in 14-40% of the cases. The proportion of prospective pathology reviews by (a member of) the expert panel increased from 60% in 1998-1999 to 90% in 2006 (p=0.001). DISCUSSION: The compliance with the guidelines has been optimised by the increased attention to this group of patients. Most important factors have been the reporting of the results of the first evaluation and (discussions about) the centralisation of treatment. Further improvements could be reached by the prospective web based registry monitoring logistic aspects as well as parameters useful for the evaluation of the quality of care.

Improved survival after resectable non-cardia gastric cancer in The Netherlands: the importance of surgical training and quality control.

BACKGROUND: In The Netherlands, standardised limited D1 and extended D2 lymph node dissections in the treatment of resectable gastric cancer were introduced nationwide within the framework of the Dutch D1-D2 Gastric Cancer Trial between 1989 and 1993. In a population-based study, we evaluated whether the survival of patients with resectable gastric cancer improved over time on a regional level. METHODS: We compared 5-year overall and relative survival of patients with curatively resected non-cardia gastric cancer in the regional cancer registry of the Comprehensive Cancer Centre West in The Netherlands before the Dutch D1-D2 trial (1986 to mid 1989; n = 273), during the trial period (mid 1989 to mid 1993; n = 255), and after the trial (mid 1993 to 1999; n = 219), adjusting for prognostic variables. RESULTS: Unadjusted survival was highest in the post-trial period: 5-year overall and relative survival were 42% and 52%, respectively, compared to 34% and 41% in the pre-trial period, and 39% and 46% in the trial period (p = 0.31 and p = 0.06, respectively). After adjustment for age, gender, tumour site, pT-stage, nodal status and hospital volume, the effect of period on survival was more apparent (p = 0.009). Compared to the pre-trial period, the hazard ratio was 0.83 (95% confidence interval, 0.68-1.02) for the trial period, and 0.72 (0.58-0.89) after the trial. Less than 1% of the patients received adjuvant therapy. CONCLUSION: Survival of patients with curatively resected non-cardia gastric cancer has improved. Standardisation and surgical training in D1 and D2 lymph node dissection are the most likely explanation for this improvement.