Annals of Surgical Oncology Cite Track
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
10.1245/s10434-006-9130-9
Annals of Surgical Oncology 14:411-416 (2007)
© 2007 Society of Surgical Oncology
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ahuja, N.
Right arrow Articles by Gearhart, S. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ahuja, N.
Right arrow Articles by Gearhart, S. L.

Original Article

Disparities in Colon Cancer Presentation and In-Hospital Mortality in Maryland: A Ten-Year Review

Nita Ahuja, MD, David Chang, MPH, PhD, MBA and Susan L. Gearhart, MD

Department of Surgery, Johns Hopkins Medical Institutions, Baltimore, Maryland 21287, USA

Correspondence: Address correspondence and reprint requests to: Susan L. Gearhart, MD; E-mail: sdemees1{at}jhmi.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Much attention has focused on in-hospital treatment disparities in colon cancer outcomes. Little is known about the effect of prehospital factors on outcomes. We hypothesized that racial and socioeconomic disparities exist in the presentation of colon cancer and that these disparities affect in-hospital outcomes.

Methods: Ten-year data on colon cancer patients were obtained from the Maryland Hospital Discharge Database. Life-threatening symptoms at presentation served as a proxy for delay in diagnosis. Patients with the primary diagnosis of colon cancer treated with surgical resection were included. Outcomes of interest were obstruction, hemorrhage, perforation, and in-hospital mortality.

Results: A total of 14,291 patients had primary colon cancer, and 13,031 underwent resection. Among this group, 52% were male, 22% were African American (AA), and mean age of AA was 66.0 years versus non-AA mean age of 70.5 years (P < .001). Overall, 27.6% of patients presented with life-threatening symptoms. In-hospital mortality rate was 3.8%. Symptomatic patients had a 2-fold higher rate of in-hospital mortality (odds ratio [OR], 6.06 vs. 2.89, P < .001). Multivariate analysis demonstrated that AA were more likely to have life-threatening symptoms at presentation independent of socioeconomic status (SES) (OR, 1.36). In addition, AA had a higher in-hospital mortality, both overall (OR, 1.39) and in the higher SES (OR, 1.81).

Conclusions: Racial disparities exist in the rate of presentation with life-threatening symptoms that may be a proxy for a delay in diagnosis. These findings were independent of SES, implying that factors beyond health care access may account for poorer outcomes among AA.

Key Words: Colon cancer • Presentation • African American • Surgery


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Colorectal cancer is the third most common cancer diagnosed in the United States. Diagnosis is achieved by means of colonoscopy, and under the current U.S. recommendation, everyone should have a screening colonoscopy by the age of 50.1 If the disease is caught early and there is no evidence of metastasis, the cure rate approaches 90%. Despite this knowledge, only 50% of the U.S. population undergoes screening colonoscopy at the appropriate age, and as a result, only 38% of colorectal cancers are diagnosed at an early stage.2 Nearly 70% of patients present with symptoms from their cancer that are associated with more advanced disease.3–6 Emergent laparotomy for bowel obstruction or perforation resulting from colon cancer is required in 20% of patients, and this is associated with a 17% to 30% in-hospital mortality rate.7,8

Epidemiological data indicate that racial disparities exist within outcomes from colorectal cancer in that African Americans (AA) have a 1.4-fold increase in mortality and a decrease in overall survival when compared with whites.9–11 It has been suggested that this may be due to in-hospital provider or treatment disparities.12,13 However, more recent studies suggest that disparities in outcomes from colorectal cancer may not be due to in-hospital treatment but the result of prehospital factors such socioeconomic status (SES) and participation in preventative health care.14 According to the American Cancer Society, disparities in outcomes from colorectal cancer in Maryland exist.2

In this study, we chose to use the Maryland Hospital Discharge Database to determine whether this disparity was due to prehospital factors that affect in-hospital outcomes. Our primary outcome was presentation of colon cancer with life-threatening symptoms, which served as a proxy for delay in diagnosis in this study. The life-threatening symptoms included were intestinal hemorrhage, obstruction, and perforation. We hypothesized that racial and socioeconomic disparities exist in the presentation of colon cancer, and that these disparities affect in-hospital outcomes.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Data Sources
Discharge data was obtained from the Maryland Hospital Discharge Database. This comprehensive database contains information regarding hospital discharge, length of stay, treating diagnoses, in-patient procedures, in-patient mortality, complications, demographics, and payer information for the state of Maryland. A total of 50 institutions, including academic medical centers and community hospitals, are required to enter discharge information regarding all patients cared for in Maryland. Submission is mandated and regulated through the Maryland Health Services Cost Review Committee for the purpose of hospital reimbursement and quality improvement.

Patient Population
All patients discharged with a diagnosis of colon cancer in Maryland from 1994 through 2003 were initially included. Specifically, the International Classification of Disease (ICD-9) codes of 153.1–4 and 153.6–9 for colon cancer were used. To evaluate those patients who sought medical care for colon cancer as their primary complaint and not as an incidental in-hospital finding, we chose to include only those patients who received a primary discharge diagnosis of colon cancer. Furthermore, we chose only those patients receiving surgical therapy to limit the bias of coexisting medical conditions.

Outcomes
The primary outcomes of interest studied from the database included presentation with life-threatening symptoms, which included intestinal obstruction, hemorrhage, and perforation. We chose these outcomes because we believe that they may be proxy for a delay in diagnosis and treatment, indicating the presence of prehospital factors contributing to poorer outcomes. To validate the use of presentation with life-threatening symptoms as our outcome, we also evaluated in-hospital mortality. These outcomes were compared with the primary demographic variables of race. Only AA and white patients were included in this study population. Secondary demographic variables included age, sex, payer status, and mean household income. To determine payer status, patients listed as self-pay were compared with patients listed as insured. As a further indication of SES, patients who had a lower-than-average income were compared with those who had an income greater than the mean income as determined by the 2000 census information per patient residence. Finally, the Charlson Index was used to adjust for the presence of comorbidity within the patient population. To exclude cancer-related comorbidities, the modified Charlson Index was used.15 This study was approved by the Johns Hopkins Institutional Review Board.

Statistical Analysis
Bivariate analysis of categorical data was performed by the {chi}2 test. Analysis of continuous data was performed by Student’s t-test. Logistic regression was used for multivariate analysis. All statistical analyses were performed by STATA software (College Station, TX). Significance was reported for findings demonstrating P < .05.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Population Statistics
Between the years 1994 and 2003, a total of 24,109 patients were identified from the Maryland Discharge Database to have a diagnosis of colon cancer. The primary discharge diagnosis of colon cancer was recorded in 59.3% (n = 14,291). Of those patients with a primary diagnosis of colon cancer, 91.2% (n = 13,031) underwent surgical resection, and this population served as our study population (Table 1Go). This population consisted of 51.9% men, and 22% of patients were AA. The overall mean age was 69 years, and the median age was 71 years. The mean household income for study participants was $63,762. More than half of the patients had a Charlson Index Score of ≥5.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Characteristics of patients whose data were recorded in the Maryland Hospital Discharge Database
 
On bivariate analysis, the mean age of men versus women with colon cancer did not significantly differ (68.1 vs. 70.6 years, P = 1.0). However, AA with colon cancer were significantly younger than whites (66.0 vs. 70.5, P < .001, Table 2Go). Furthermore, colon cancer was more common among female AA than female whites (55.48% vs. 51.14%, P < .001). The mean household income for AA with colon cancer was $55,284, significantly less than the mean household income for whites ($66,130, P < .001). Furthermore, AA patients reported self-pay status significantly more often than whites (50.6% vs. 35.6%, P < .001).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Bivariate analysis of effect of race on age, symptomatic presentation, socioeconomic status, and in-hospital mortality for patients undergoing resection for primary diagnosis of colon cancer
 
Presentation of Disease
Within this study population, 27.6% of patients had life-threatening symptoms at the time of presentation (Table 1Go). Specifically, 22.9% presented with symptoms of obstruction, 3.9% had intestinal hemorrhage, and 2.3% had developed intestinal perforation. The overall mortality was 3.77%. On bivariate analysis, symptomatic patients had a 2-fold higher rate of in-hospital mortality than asymptomatic patients (6.1% vs. 2.9%; P < .001, Table 3Go). This finding indicates that the risk of in-hospital mortality was directly associated with symptomatic status at presentation and supports the use of symptomatic status as a valid measure that may affect patients’ inhospital outcomes. Furthermore, symptomatic patients were more likely to be older and to have a higher mean household income. AA presented more often with life-threatening symptoms from their colon cancer than whites (26.8% vs. 30%, respectively; P = .001; Table 2Go). Specifically, AA were more likely to present with obstruction or hemorrhage (P = .03 and .009, respectively), but not perforation. However, there was no difference in the in-hospitality mortality rate with regards to race (AA 4.2%, whites 3.7%; P = .292).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Bivariate analysis of the effect of sex, age, payer status, income, and in-hospital mortality on symptomatic status
 
By means of multivariate analysis, race, age, sex, and SES were evaluated as they relate to life-threatening symptoms and in-hospital mortality (Table 4Go). Independent of income (SES), AA were more likely to have life-threatening symptoms from their colon cancer (odds ratio [OR], 1.36). Specifically, these finding were true for symptoms of obstruction and intestinal hemorrhage (OR, 1.17 and 1.45, respectively), but not perforation. The risk of AA presenting with obstruction or hemorrhage increased independent of insurer status (OR, 1.18 and 1.46, respectively). Furthermore, the risk of AA presenting with intestinal obstruction or hemorrhage increased even more in the higher income group (OR, 1.27 and 1.55, respectively). These findings were independent of age and sex (data not shown). Although no difference in in-hospital mortality was seen on bivariate analysis, on multivariate analysis, when controlling for insurer status and symptoms at presentation, the risk of in-hospital mortality was increased among AA (OR, 1.35). In fact, AA with a mean household income exceeding $60,000 per year had the greatest risk of in-hospital mortality (OR, 1.71). Use of the Charlson Index confirmed that disparities in the risk of in-hospital mortality existed despite the presence of unknown comorbidities (Table 4Go). Identical results were also seen when the modified Charlson Index was used to adjust for cancer-related comorbidities (results not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 4. Multivariate analysis demonstrating effect of race on life-threatening symptoms at presentation and socioeconomic status
 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Previous methods of assessment of outcomes in health care have focused on in-hospital factors. Recently, several prehospital factors such as adherence to screening guidelines, social bias, and education have been shown to be important in health-related outcomes.14–17 In this study, we chose to use life-threatening symptoms as a potential proxy for delay in diagnosis, indicating the presence of prehospital factors affecting outcome from colon cancer. Specifically, the risk of presenting with life-threatening symptoms of intestinal hemorrhage, obstruction, and perforation as a result of colon cancer were evaluated among patients of differing race and SES in Maryland. Our study demonstrated that independent of SES, AA present more frequently than whites with life-threatening symptoms as a result of colon cancer and, on multivariate analysis, have a higher in-hospital mortality risk. Furthermore, independent of comorbid conditions, affluent AA with an average household income greater than $60,000 per year have the highest inhospital mortality risk.

Racial disparities in outcomes among different diseases have often been attributed to disparities in treatment. However, Wudel et al.11 demonstrated that AA had a decrease in overall and disease-free survival from colorectal cancer when compared with whites undergoing similar treatment at a single institution. These findings were independent of stage and socioeconomic factors. The authors suggested that because this study was performed at a single institution, these findings would imply that factors outside of in-hospital treatment were responsible for this disparity. Dominitz et al.18 evaluated outcome disparities in colon cancer treatment within the Veterans Medical System, an equal-access system for health care. This study demonstrated that when demographics, comorbidities, and tumor stage are adjusted for, no racial difference in treatment with regards to surgery or adjuvant therapy existed within the Veterans Administration medical system. The conclusion from this study was that in an equal access system such as in the Veterans Administration system, no disparities in the treatment of colorectal cancer occurred.

In this study, we chose to evaluate how prehospital factors such as health care access affect outcomes from colon cancer. In particular, we examined the risk of presenting with life-threatening symptoms and in-hospital mortality from colon cancer among the demographic variables of race and SES. The use of life-threatening symptoms as a proxy for delay in diagnosis indicated the presence for prehospital factors effecting outcome. This was validated by the finding that presentation with life-threatening symptoms was associated with a much higher in-hospital mortality rate. Our findings have been corroborated by the findings of others indicating that the in-hospital mortality rate for emergent surgery for colorectal cancer approaches 30% in some series.6–8 Furthermore, on multivariate analysis, the risk of presenting with life-threatening symptoms was independent of age, sex, and the presence of comorbidities. These findings may suggest that a higher percentage of AA patients in Maryland allow "life-threatening" symptoms to occur before seeking medical care. Other investigators have corroborated our findings. Freeman and Alshafie19 reported that a delay in diagnosis was exceedingly common among AA, and a major portion of patients refused appropriate therapy. Baldwin et al.13 demonstrated that AA and whites are equally offered chemotherapy for stage III colon cancer, but that AA are more likely to refuse adjuvant therapy. These finding were corroborated by Govindarajan et al.20 in their single-institution study. Collectively, these reports imply that educational programs highlighting the early symptoms of colorectal cancer and emphasizing how colorectal cancer can be treated more successfully if caught early may improve these disparities.

This study demonstrated that prehospital barriers may exist among AA in Maryland such that the rate of presenting with life-threatening symptoms from colon cancer was much higher for AA. However, a higher SES had a negative effect on outcome from colon cancer: AA with a higher mean income had a greater risk of delay in treatment and in-hospital mortality. But McGory et al.,15 when evaluating outcomes in California, demonstrated that SES does impact survival from colorectal cancer. He demonstrated that patients with advanced colorectal cancer and an income at poverty level received appropriate adjuvant therapy far less often than those patients of a higher SES. Their findings were independent of race and comorbidities. This study differs from our study in that their primary outcome was related to cancer treatment after diagnosis, whereas our primary outcome evaluated prehospital factors before hospital presentation. It is plausible that different barriers exist when prehospital factors are evaluated rather than factors related to treatment. Other reasons for the differences in the findings of these two studies may be a result of the techniques used in evaluating SES or the result of differences unique to the state of Maryland.

Finally, other possible reasons for the presence of this racial disparity may include genetic factors that predispose this population to colon cancer. Ashktorab et al.21 demonstrated that microsatellite unstable (MSI-H) tumors were 2- to 3-fold more common among AA than whites, although the rate of a defect in the mismatch repair gene (MMR) expression was similar. Furthermore, among AA and whites, differences in methylation patterns for cell cycle genes important in oncogenesis (p16, hMLH1) were also seen. The authors suggest this may be a result of different environmental exposures. Mayberry et al.22 also noted survival among AA with earlier-stage colon cancer (II and III) was worse than white patients with a similar stage of disease. The authors suggested that this may be due to the characteristic of the tumor or the lack of aggressive treatment for disease.

We acknowledge that several limitations exist in interpreting the results from our study. First, the results rely solely on the independent institutional collection of data and insertion of this data into the Maryland Hospital Discharge Database. The information in this database depends on the correct coding and recording of ICD-9 codes, which may not be as precise as chart reviews. However, the error distribution in databases of this type is likely to be random with respect to our study question because the data were recorded by a clerical sta3 who could not possibly have any knowledge of our study question. Given the size of the study population, sporadic errors in the collection of data would be unlikely to greatly alter the outcome of this study. Furthermore, this database does not record data about tumor stage. Given that all patients underwent surgery and the outcome evaluated was based on prehospital factors, we do not believe the lack of tumor stage information effects the findings of this study. Finally, this database has been previously used to evaluate hospital outcomes per case volume for pancreatic and colorectal cancer surgery.23,24 Future studies are necessary to examine other prehospital factors such as education, preventative health initiatives, and newer procedures for screening that may affect outcomes for patients with colorectal cancer.


   FOOTNOTES
 
Presented at the Society of Surgical Oncology, March 23–26, 2006, San Diego, CA.

Received for publication June 11, 2006. Accepted for publication June 14, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Smith RA, Cokkinidis V, Eyre HJ. American Cancer Society guidelines for the early detection of cancer 2006. CA Cancer J Clin 2006; 56:11–25.[Abstract/Free Full Text]
  2. American Cancer Society. Cancer Prevention and Early Detection, Facts and Figures 2006. American Cancer Society, Atlanta, GA, 2006.
  3. Beart RW, Melton IJ, Marute M, et al. Trends in right and left sided colon cancer. Dis Colon Rectum 1983; 26:393–8.[Medline]
  4. Fuchs CS, Giovannucci EL, Coiditz CA, et al. A prospective study of family history and the risk of colorectal cancer. N Engl J Med 1994; 331:1669–74.[Abstract/Free Full Text]
  5. Umpleby HC, Bristol JB, Rainey JB, Williamson RC. Survival of 727 patients with single carcinoma of the large bowel. Dis Colon Rectum 1984; 27:803–10.[Medline]
  6. Kelly WE, Brown PW, Lawrence W, Terz JJ. Penetrating, obstructing, and perforating carcinoma of the colon and rectum. Arch Surg 1981; 116:381–4.[Abstract/Free Full Text]
  7. Phillips RKS, Hittinger R, Fielding LP. Malignant large bowel obstruction. Br J Surg 1985; 72:296–302.[Medline]
  8. Ohman U. Prognosis of patients with obstructing colorectal carcinoma. Am J Surg 1982; 143:742–7.[CrossRef][Medline]
  9. US Dept of Health and Human Services Public Health Services. Cancer. October 16, 2002. Available at: http://www.omhrc.gov/templates/content.aspx?ID=2826.
  10. Cancer facts and figures for African Americans, 2005–2006. American Cancer Society 2005–2006 Report, Atlanta, GA.
  11. Wudel LJ, Chapman WC, Shyr Y, et al. Disparate outcomes of colorectal cancer patients: the influence of race on long-term survival. Arch Surg 2002; 137:550–4.[Abstract/Free Full Text]
  12. Cooper GS, Yuan Z, Landefeld CS, Timm AA. Surgery for colorectal cancer: race-related differences in rates and survival among Medicare beneficiaries. Am J Public Health 1996; 86:582–6.[Abstract/Free Full Text]
  13. Baldwin L, Dobie S, Billingsley K, et al. Explaining black-white difference in receipt of recommended colon cancer treatment. J Natl Cancer Inst 2005; 97:1210–20.
  14. O’Malley A, Forrest C, Feng S, Mandeblatt J. Disparites despite coverage: gaps in colorectal cancer screening among Medicare beneficiaries. Arch Intern Med 2005; 165:2129–35.[Abstract/Free Full Text]
  15. McGory M, Zingmond C, Sekeris E, Bastani R, Ko C. A patient’s race/ethnicity does not explain the underuse of appropriate adjuvant therapy in colorectal cancer. Dis Colon Rectum 2006; 49:319–29.[CrossRef][Medline]
  16. Whynes DK, Frew EJ, Manghan CM, Scholefield JH, Hardcastle JD. Colorectal cancer, screening and survival: the influence of socio-economic deprivation. Public Health 2003; 117:389–95.[CrossRef][Medline]
  17. Spencer B, Babey S, Etzioni D, et al. A population-based survey of prostate-specific antigen testing among California men at higher risk for prostate cancer. Cancer 2006; 106:765–74.[CrossRef][Medline]
  18. Dominitz J, Samsa G, Landsman P, Provenzale D. Race, treatment, and survival among colorectal carcinoma patients in an equal-access system. Cancer 1998; 82:2312–20.[CrossRef][Medline]
  19. Freeman H, Alshafie T. Colorectal carcinoma in poor blacks. Cancer 2002; 94:2327–32.[CrossRef][Medline]
  20. Govindarajan R, Shah R, Erkman L, Hutchins L. Racial differences in the outcome of patients with colorectal carcinoma. Cancer 2003; 97:493–8.[CrossRef][Medline]
  21. Ashktorab H, Smoot D, Farzanmehr H, et al. Clinicopathological features and microsatellite instability (MSI) in colo-rectal cancer from African Americans. Int J Cancer 2005; 116:914–919.[CrossRef][Medline]
  22. Mayberry R, Coates J, Hill H, et al. Determinants of black/white differences in colon cancer survival. J Natl Cancer Inst 1995; 87:1686–93.[Abstract/Free Full Text]
  23. Harmon J, Tang D, Gordon T, et al. Hospital volume can serve as a surrogate for surgeon volume for achieving excellent outcomes in colorectal resection. Ann Surg 1999; 230:404–11.[CrossRef][Medline]
  24. Sosa JA, Bowman HM, Gordon TA, et al. Importance of hospital volume in the overall management of pancreatic cancer. Ann Surg 1998; 228:429–38.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 J. Thorac. Cardiovasc. Surg.Home page
R. A. Meguid, E. S. Weiss, D. C. Chang, M. V. Brock, and S. C. Yang
The effect of volume on esophageal cancer resections: what constitutes acceptable resection volumes for centers of excellence?
J. Thorac. Cardiovasc. Surg., January 1, 2009; 137(1): 23 - 29.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ahuja, N.
Right arrow Articles by Gearhart, S. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ahuja, N.
Right arrow Articles by Gearhart, S. L.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS