Colon Cancer; Study results from University of South Australia in the area of colon cancer

Stage at diagnosis plays an important role in colorectal cancer (CRC) survival. Understanding what factors to an advanced stage at diagnosis is crucial for improving survival. Comorbidity, race, age, and are known to impact the receipt of the cancer treatment and survival, but the relationship of these factors on the stage at diagnosis of CRC is less clear. The aim of this study is to investigate how co-morbidity, race and age affect the stage CRC diagnosis.
Methods

Two different populations of health care in the United States (U.S.) were retrospectively investigated. The Cancer Care Outcomes Research and Surveillance Consortium database, we found CRC patients at 15 Veterans Administration (VA) hospitals from 2003-2007. We assess metastatic CRC patients from 2003-2006 in 10 non-VA fee-for-service (FFS) practices. Stage at diagnosis was dichotomized (non-metastatic, metastases). Race was dichotomized (white, not white). Charlson comorbidity index and age at diagnosis was. Associations between stage, comorbidity, race and age were by logistic regression.
Results

342 VA and 340 FFS patients were. Difference from the population, the proportion of patients with metastatic CRC in the diagnosis (VA 27% and 77% FFS), the differences in the criteria for inclusion. VA patients were mean (standard deviation, SD) age 67 (11), Charlson Index 2.0 (1.0), and 63% white. FFS patients average age 61 (13), Charlson Index 1.6 (1.0), and 73% were white. In the VA cohort, higher co-morbidity was associated with earlier stage at diagnosis after adjusting for age and race (odds ratio (OR) 0.76, 95% confidence interval (CI) 0.58-1.00, p = 0045); no significant relationship was in the FFS cohort (OR 1.09, 95% CI 0.82-1.44, p = 0.57). In both cohorts, no association was found between the stage and either in the diagnosis of age or race.
Conclusion

Higher comorbidity may lead to earlier diagnosis of CRC. Several factors, perhaps even greater interaction with the health system because of comorbidity, in this finding. These interactions are increased in patients seen in a health system like the VA system in the U.S. in comparison to sporadic interactions, you can use FFS health care.
Background

CRC is the second leading cancer in Europe and the fourth most common cancer in the United States (USA) [1,2]. CRC is the second leading cause of cancer death in Europe and the USA [1,2]. Advanced age is the most important risk factor for the diagnosis of CRC, since the vast majority of patients are diagnosed when 65 and older with a peak incidence of 415.9 cases per 100000 in the 85 and older [3]. Despite advances in early detection and treatment, stage at diagnosis remains the most important predictor for mortality. The five-year survival rate for localized disease is 90.4%, but only 39% of CRCs are diagnosed at this early stage [3]. The diagnosis of metastatic disease, less than 10% are still alive after 5 years [3].

Since the stage at diagnosis plays a significant role in the survival of CRC, which factors contribute to a more advanced stage at diagnosis is crucial for improving survival. Possible factors include comorbidity, age and race. These factors are known to influence cancer outcomes, at least in relation to the receipt of standard therapy, which in turn influence the likely survival [4-10]. While the relationship between the delivery of treatment, and these factors, the association between comorbidity, age, race, and stage at diagnosis of CRC is less clear (Figure 1).

Figure 1 thumbnail Relationship between stage at diagnosis and comorbidity, age and race. Comorbidity, age, breed and are known to influence the delivery of phase-appropriate chemotherapy, and thus influence on survival. We hypothesis (dashed line), the co-morbidity and age also influence survival by determining stage at diagnosis.

Access to health care and health insurance was associated with differences in cancer survival and late stage of diagnosis [11.12]. The objective of this analysis is the relationship between stage at diagnosis and comorbidity, age, race, and in patients with CRC, and the influence that health care has led to this association. Two U.S. health care system, fee for service (FFS) and the Veterans Administration (VA), were chosen because they assess the predictors of stage of diagnosis, only for the health care system in comparison to the population. Health care in the United States is supported by several parallel and overlapping systems roughly between commercial, public and non-insured / self-pay programs [13]. FMS provides the context in which patients usually have a form of commercial insurance for most medical costs, access to specialist care is more readily available, and primary care can not be promoted, depending on the type of insurance ( for example, managed care, Preferred Provider Organization) [13]. The VA system provides care to military service veterans and their families but is a closed system with access to primary health care facilities, as gatekeepers to specialist care. The VA system is generally regarded as a U.S. health care system where access to health care is less constrained by the ability to pay [14]. We hypothesis that patients with more comorbidity, age, white race and would rather be diagnosed with early-stage CRC, due to frequent contact with the health system. We suspected that these differences would be best known in the VA system, where access to quality primary care is recommended [15.16].
Methods
Study population

The first patient cohort was derived from patients with CRC in the Cancer Care Outcomes Research and Surveillance Consortium (CanCORS), a cohort study to assess the care of patients with newly diagnosed lung or CRCs recruited in geographically diverse populations and health systems [17]. Eligible patients ≥ 21 years old and was diagnosed with colorectal adenocarcinoma within 3 months after registration. Approximately 4920 patients with CRC were CanCORS. CanCORS With the database, we identified patients with CRC diagnosed and treated at 15 Veterans Administration (VA) hospitals from the year 2003 (cohort initiation) until 2007. Of the 4920 CRC patients in the CanCORS study, 477 treatment in the VA health system. Of those, 342 had complete chart abstraction at the time of this analysis, the inclusion criteria are met (as described above), and were in this analysis. Informed consent was signed by all living patients and deceased HIPAA consent waiver was for deceased patients.

The second patient cohort was supported by the academic community and FFS practices in the southeastern United States, as part of a larger study to examine regional structures for the treatment of metastatic CRC [18]. Eligible patients were adults diagnosed with metastatic CRC (recurrent or primary) from January 2003 to June 2006, and treated in one of the ten participating academic or community websites. Medical data for patients before 2003 for non-metastatic CRC were also responsible for assessing the stage at diagnosis. Charts of 743 initially screened by the local tumor registry lists, 340 met eligibility criteria and were abstracted charts available at the time of analysis. Informed consent was waived by the Institutional Review Board for this patient cohort. The research was presented in accordance with the principles of the Declaration of Helsinki [19].
Data Collection

We have a retrospective review of patients from two different groups of patients, as mentioned above. The data were abstracted based on race, cancer diagnosis, stage, age at diagnosis, and presence of comorbid diseases. Stage at diagnosis was based on review of pathology reports in cooperation with the radiology and the physician notes. The TNM staging system was used. Stage at diagnosis was dichotomized into non-metastatic (defined as ≤ stage III, localized, or regional) and metastatic due to the small sample size limits. Age at diagnosis was made in the years for each patient. Race was dichotomized to white and not white.

The presence of comorbid conditions was based on chronic diseases, the Deyo adaptation of Charlson comorbidity index [20]. The Charlson index was developed as a reproducible measure of the prognostic impact of comorbid disorders [21]. The index was validated in the oncology population and has been designed for use with administrative databases [20]. Patient medical records were abstracted for chronic diseases are already in the CRC diagnosis (Table 1). The terms were weighted and patients were treated with a result in line with the Charlson Index [21]. The co-morbidity outcome was modeled as an integer value of quantitative variables.

Table 1. Charlson comorbidity index and weights
Data Analysis

The primary objective of the analysis was to make connections between the stage at diagnosis and an a priori set of variables (comorbidity, age and race) using logistic regression modeling in cohorts from two different groups of patients. The two cohorts were analyzed together because of differences in the patients from both studies and the overall study design. The results for each group, as unadjusted and adjusted odds ratios associated with 95% confidence interval and p-values. Odds ratios and associated confidence intervals are also presented graphically for better understanding of the similarities and differences in the patterns of association between the cohorts [22]. Linearity assumptions of continuous covariates (age and comorbidity) were using graphical techniques, and no evidence of systematic non-linearity was. Statistical analysis was with SAS for Windows version 9.1 (SAS Institute, Cary, NC). Two-sided p-values in the standard 0.05 level were used to determine statistical significance.
Results
Patients

Six hundred eighty-two patients met the criteria for this analysis (Table 2). The VA cohort was older than the FFS group (mean ± standard deviation age, 67 ± 11 years in the VA compared with 61 ± 14 years in FFS). As expected, the VA cohort was predominantly male (98%) in contrast to the FFS cohort, evenly divided by gender. More patients than VA FMS patients had a Charlson score ≥ 3 (11% versus 7% VA FFS). Both cohorts were predominantly white. The FFS cohort consisted of a higher proportion of patients with stage IV disease at diagnosis on original intent and design of the study from which this patient cohort was [18].

Table 2. Characteristics of patients in both VA and fee-for-service settings.
Logistic regression

The statistical results of the logistic regression are given in Table 3 and discussed.

Table 3. Association of metastatic colorectal cancer stage of diagnosis with co-morbidity, age and race in two patient cohorts
Influence of comorbidity

In the VA cohort, higher co-morbidity was associated with non-metastatic stage at diagnosis. This relationship was after adjusting for age and race (adjusted odds ratio (OR) 0.76, 95% confidence interval (CI) 0.58-1.00). In the FFS cohort, but no correlation was between comorbidity and stage at diagnosis (adjusted or 1.09, 95% CI 0.82-1.44).
Influence of age

In the VA and FFS cohorts, no evidence was found of an association between the stage of diagnosis and age. The CI's for the OP's in the two populations span 1.0 and overlap each other, which means that older people are not connected with the stage at diagnosis. These claims hold, whether the results were for the race and comorbidity.
Influence of race

The influence of race was not statistically significant for the two cohorts. The unadjusted and adjusted confidence intervals for the odds ratios largely spanned 1.0 for both cohorts.
Discussion

This analysis provides a health system-based view of the relationship between stage of CRC diagnosis and comorbidity, race and age. Higher comorbidity was associated with non-metastatic stage at diagnosis in the VA cohort, but not in the FFS group. Neither age nor race was associated with stage at diagnosis. The results may be due to several factors, some of them in this analysis is not measured.

A critical, co-morbidity associated with cancer care refers to the results, such as co-morbidity influenced survival in cancer patients. Comorbidity is already known that a negative impact on the delivery of appropriate treatment phase (Figure 1), no appropriate treatment negatively affect survival results [23-29]. Due to the aging of the population, the increasing burden of comorbid illness is likely to play a greater role in the treatment of cancer and the toxicity results. Higher co-morbidity has shown the difference in survival between blacks and whites with CRC [24]. Both comorbidity and advanced age have been associated with worse outcomes after surgery for CRC [23,25,26,29], decreased referral to a medical oncologist [23.28], and incomplete courses adjuvant chemotherapy [27.30]. With the best understanding, if the cancer treatment continuum comorbidity, age, race, and its greatest influence, we can better determine how to improve the care of patients with comorbid disorders and CRC.

In the VA cohort wherever comorbidity was associated with earlier stage at diagnosis, patients with multiple comorbid diseases could be more frequent health care contact and thus the larger clinical trial, he provides. Evidence that VA patients are more likely to experience higher quality primary care than FFS patients; addition to improved access to primary health care is a routine increase in chronic and preventive medical care to be overlooked in the health systems with less access to basic healthcare. In non-VA population, FFS health care could lead to erratic interaction with the healthcare system, through the ability to pay. For example, VA patients have shown that it works better with diabetes and hyperlipidemia control than patients in health care [15]. VA patients are more likely to report diabetes education than those for private health insurance [31]. Therefore, the medical check-ups increases if a person has multiple co-morbidities in conjunction with a health system that provides easy access to primary health care facilities could be the delivery of age-appropriate screening or could lead to accidental diagnosis of cancer through blood tests or studies in order for other purposes.

Other factors can affect the relationship between comorbidity and stage in the VA cohort and not in the FFS group. Above all, the difference could be due to differences in the two cohorts. The VA cohort were older, predominantly male, and a higher burden of comorbid illnesses than the FFS cohort. This difference between the non-VA and VA cohorts in our analysis are representative of the VA and non-VA population as a whole [32-35]. The association may be influenced by the comorbid conditions themselves, leading to a manifold influence of comorbidity on stage at the population rather than on the health system. For example, symptoms of comorbid diseases could actually from a neglected, is developing malignancy [36]. Second, the pathophysiology of comorbid disease or its treatment could contribute to the development or progression of cancer. For example, a growing number of data has shown an association between chronic insulin therapy and the development of CRC in patients with type II diabetes mellitus [37]. These factors could have decreased (in the VA group) or not (in the FFS group) the impact of comorbidity on the stage at diagnosis.

Another factor was the definition of the FFS cohort. Eligible patients with metastatic disease in the period 2003-2006, although their prior non-metastatic was also taken into account. As a result, the cohort was enriched with patients who have stage IV at diagnosis. This group may be less screening and health involvement at study entry, and the impact of comorbidity status May, diluted. Nevertheless, this group is still a quarter of patients who are non-metastatic at diagnosis.

In our analysis, race and age were not influential factors in the phase of CRC diagnosis. Age is not likely, because influential numerical age is less important, clinically, if the degree of comorbidity is. In other words, in elderly patients without significant comorbidity should be the similar clinical outcome as relatively younger patients without comorbidity. For example, if elderly patients chemotherapy, they are able to tolerate and respond to treatment as well as their younger counterparts [38-41]. The absence of interactions with the race and stage at diagnosis can be done by a similar process. Bach et al found that after the control of population mortality (non-cancer deaths), the difference in cancer mortality between black and white was reduced [24]. The differences of race, in the diagnosis at the stage could be reduced when adjusted for comorbidity. We have not, however, for differences in the socio-economic status, which may also influence the role of race.

In determining the significance of the results, the limitations of this study must be discussed. Firstly, in relation to the categorization of co-morbidity, we opted for the use of the Charlson index, which has been validated in the oncology setting. Although updated, the index was launched in 1984 and to the inclusion in a hospital over a period of one month [20.21], which call into question the generalizability index for diagnosis and treatment of cancer today [42]. The Charlson index is not grade severity of comorbidity, nor does it capture functional disability. It can not be that the measurement comorbid conditions are the most for a cancer population. On the other hand, the user-friendliness, reliability, validity and content of the index, it is a reasonable choice [42].

Second, asymptomatic screening was not as variable in this study as a patient screening information was not available for both patient cohorts. Patients with a high co-morbidity index also less likely to undergo cancer screening by an increased risk for non-cancer mortality. If this is the case, the relationship between comorbidity and stage at diagnosis, less important than the relationship between comorbidity and receipt of screening, such screening could be delayed to later stage at diagnosis. However, studies conducted in both VA and VA populations have shown that patients are at CRC, regardless of comorbidity status, so that there is no relationship between the degree of comorbidity and rates of screening asymptomatic [43-46]. If screening rates of comorbidity are not, then step in diagnosis is the next most important variable for the study. In addition, if younger and healthier patients are more likely to check for CRC, then this analysis should have found that younger and healthier patients with non-metastatic disease, and older, sick patients with metastatic disease. He did not.

This analysis is not explicitly limited to access to screening tests such as colonoscopy. Several factors have combined with the access or control of the colorectal cancer screening, including age, education, insurance status and usual source of care [47]. Our present data can not be applied to these features entirely, if the literature supports the assumption that the VA patients have greater access to many aspects of primary health care than Medicare fee-for-service and privately insured patients [15,16,31]. While the data is not available for our sample, the rate of CRC screening with endoscopy is fairly equivalent between VA and non-VA population [48.49].

This study has strengths, overcome limitations of previous studies. Firstly, we are consistent data from two different health systems. Our study is the first, the relationship between co-morbidity, age and stage at diagnosis in the VA health system. As the receipt of care in this system is less dependent on the ability to pay, it serves as a useful mechanism for the inability to access to appropriate cancer treatment [50]. Secondly, this study did not dichotomize comorbidity, but models such as an integer value quantitative variable (0 - ≥ 3) with Charlson Score ≥ 3 equated with severe comorbidity. A value of ≥ 3 predicted a significantly higher risk of non-cancer mortality over a period of one year [21]. A previous study showed a slightly higher prevalence of comorbidity in patients diagnosed with early stage (Dukes' A) CRC, but this study dichotomized co-morbidity with an average of ≥ 1 [51]. A study by Gonzalez et al found the opposite: patients with a comorbid condition (Charlson Score ≥ 1 vs 0) were more likely to be diagnosed at the end of this phase, although interesting, this result is not in a Charlson Score of ≥ 2 [52] . Based on these findings, we do not have to dichotomize comorbidity. The quantitative use of co-morbidity index, as opposed to simply measuring presence to no comorbidity, a more clinically-useful model in assessing the role of a comorbidity index for the diagnosis, treatment and outcomes of cancer. A third strength of our study lies in our ability to measure comorbidity. Phase of the test studies with larger cohorts of cancer registries such as Surveillance, Epidemiology and End Results (SEER) database, have no access to data comorbidity.

Our results help exploratory opinion on which future efforts to improve CRC outcomes. Since the association between comorbidity and cancer outcomes is clear, future prospective studies should examine where the patient in the treatment of cancer of a certain trajectory comorbid illness could impact results [53]. For example, in patients with breast cancer, studies have shown that only certain comorbid diseases (like diabetes) at a later date at diagnosis, while others (such as cardio-vascular diseases) are not [54]. Our opinion exploratory results should be further investigation into comorbid, such as diseases, and not only co-morbidity indices, impact, diagnosis, treatment and survival. When patients are even more likely to be diagnosed at an early stage in the definition of comorbid illness, it is even more important, the development of CRC treatment strategies in which co-morbidities. Future analysis of similar questions you will find the complete Multi-Health Systems CanCORS sample taken from approximately 10000 lung and colon cancer patients. Socioeconomic characteristics such as income and education, should also be examined in conjunction with co-morbidity. These factors can not be adequately investigated by SEER-Medicare linked data, but can CanCORS.

As cancer treatment improves, more patients live longer. As a result, more attention must be paid to the role of comorbidity and general health of the patient as a co-morbidity could actually play a greater role than in the survival of the cancer itself. Studies have shown that some cancer patients also are less likely to diet, exercise and healthy lifestyle changes after a cancer diagnosis [55]. The future of interventional studies in cancer patients could investigate how to better control of health habits, and comorbid diseases could improve the health related quality of life and cancer associated with survival outcomes. The data from this study suggests that the role of public health and primary health care provider may have an important influence on the time of diagnosis of CRC, similar to access to primary health care and routine is important for maintaining the health of cancer survivors in the period .

Currently, oncologists are older or frailes patients based on data from clinical trials, younger, healthier patients [56-58]. If a better understanding gained of the role of concurrent disease outcomes in cancer patients with comorbidity and advanced age can be evaluated separately and in clinical studies to understand how they really benefit from the advances in cancer therapy. This diversification of the clinical trial population would clinical data that are more representative of the typical cancer patient who may not otherwise be required for a clinical study, but could still make some form of treatment.
Conclusion

In this analysis of patients with CRC treated in two different health systems, we found that higher co-morbidity is associated with earlier age of CRC diagnosis for VA patients. The result could be the result of many factors, some unmeasured in this analysis. One factor could be the supply is within certain U.S. health system where access to health care is less in terms of performance and primary care is promoted. Future studies should be designed to focus on the effects of specific co-morbidities on both stage at diagnosis and survival outcomes.