Hospital volume and surgical outcomes for elderly patients with colorectal cancer in the United States1

https://doi.org/10.1016/S0022-4804(03)00207-5Get rights and content

Abstract

Background

Recent emphasis has been placed on the quality of surgical care in the United States. As such, patients, providers, and payers are increasingly aware of the outcomes of surgical care as a marker of quality. The objective of this study was to determine the impact of hospital volume on mortality for patients of different age groups to determine whether elderly patients would derive more benefit from selective referral policies.

Methods

Data from the Nationwide Inpatient Sample for all patients undergoing surgery for colorectal cancer during 1997 were obtained (N = 20,862). Differences in mortality associated with increasing age and hospital volume quartiles were determined. Risk-adjusted analyses of mortality were performed using multiple logistic regression.

Results

The overall mortality rate was 3.1% for the 842 hospitals included. Patient age breakdown was the following: age <50, 7%; age 50 to 64, 19%; age 65 to 80, 51%; and age >80, 22%. Increasing age was associated with higher mortality rates: age <50, 0.8%; age 51 to 65, 1.3%; age 66 to 80, 2.9%; and >80, 6.9%. Overall, the highest volume hospitals (HVH) (>150/year) had lower mortality than the lowest volume hospitals (LVH) (<55/year) (2.5% vs. 3.7%; P = 0.006). However, the effect of volume on mortality was primarily due to differences in older patients. For patients greater than 65 years old, the mortality rate was 3.1% at HVH and 4.5% at LVH (P = 0.03). For patients greater than 80 years old, the mortality rate was 4.6% at HVH and 7.3% at LVH (P = 0.04). The results were unchanged after adjustment for patient demographics, comorbid disease, site of cancer, and type of resection.

Conclusions

The majority of deaths after surgery for colorectal cancer occur in older patients. Hospitals that perform higher volumes of colorectal resection have lower mortality rates, especially for older patients. In the absence of other information about the quality of surgical care, provider volumes are a useful marker of postoperative outcomes for older patients in need of surgery for colorectal cancer.

Introduction

The quality of healthcare is not uniform across medical centers in the United States. Recent attention has focused on the large gaps between the high quality of health care some patients receive and the low quality others receive 1, 2. For surgical procedures, these differences in quality can be measured by examining the outcomes of care across medical centers. Payers, providers, and patients are becoming increasingly interested in identifying centers with the highest quality outcomes. Surgery for colon cancer is commonly performed in a variety of settings across the nation’s hospitals. In general, higher volume hospitals have been shown to have superior outcomes when compared to lower volume centers 3, 4, 5, 6. However, the effect of volume on outcome for this procedure is small when compared to other, more complex procedures, such as pancreatic, esophageal, or hepatic resection 6, 7.

The differences in outcomes associated with provider volume have resulted in efforts by various stakeholders to concentrate surgery in high volume centers 8, 9, 10. Such a policy, however, will have several adverse effects on lower volume centers and may threaten the practice of surgeons in these settings [10]. An alternative strategy for colorectal resection, especially given the small effect of volume on outcome for this procedure, would be to target a higher-risk group of patients for selective referral [11]. Since older patients having elective surgical procedures have been shown to have an increased risk of adverse postoperative outcomes, they may provide an easily identifiable population to target for selective referral [12]. The objective of the current study was to determine the differential effect of hospital volume on outcome for older patients using a nationally representative sample of hospitals.

Section snippets

Data source

The Nationwide Inpatient Sample (NIS) is a 20% stratified random sample of all hospital discharges in the United States. It is maintained by the Agency for Health Care Research and Quality (AHRQ) as part of the Healthcare Cost and Utilization Project (HCUP) [13]. Data for the current study were derived from the 1997 version of the NIS. During this time period, 842 hospitals from 22 states performed colorectal resections for colon cancer. Records for all patients who were discharged from these

Hospital characteristics

There were 842 hospitals in 22 states that performed colon resection for cancer in the 1997 version of the NIS. Of these hospitals, 536 (64%) were low volume, 149 (18%) were medium volume, 97 (12%) were high volume, and 60 (7%) were very high volume. Several hospital characteristics differed across the hospitals volume categories (Table 1). Very-high-volume hospitals were more likely to be teaching hospitals (55% versus 6%, P <0.001) and have a large bed size (78% versus 14%, P <0.001)

Discussion

The safety of undergoing certain surgical procedures is not uniform across medical centers. There is a consistent association of provider volume and outcomes for several operations, with a stronger relationship for more complex procedures. In the current study, we demonstrate that, overall, colon resection for cancer has a significant but small effect of volume on outcome. However, given the higher risk of death for older patients, selectively targeting this population would achieve most of the

References (18)

There are more references available in the full text version of this article.

Cited by (104)

  • Operative Mortality Prediction for Primary Rectal Cancer: Age Matters

    2019, Journal of the American College of Surgeons
  • The effect of hospital familiarity with complex procedures on overall healthcare burden

    2018, American Journal of Surgery
    Citation Excerpt :

    Rogers et al. investigated the impact of surgeon and hospital volume on 28,644 colorectal cancer patients who had tumor resection, and found the adjusted 30-day postoperative mortality to be inversely related to hospital volume, and 5-year survival to be significantly lower in low volume hospitals.10 Other studies have similarly reported an association between volume and short-term complications and mortality and long-term oncologic outcomes.14,16,18,34 Surgery at NCI-designated cancer centers and by trained colorectal surgeons has also been shown to be associated with better overall survival.15

View all citing articles on Scopus
1

Presented at the Annual Meeting of the Association for Academic Surgery, Boston, MA, November 7th–9th, 2002

View full text