Dig Dis Sci (2014) 59:504–506 DOI 10.1007/s10620-013-2922-4
EDITORIAL
Radiating Disparity in IBD Mark Flasar • Seema Patil
Published online: 8 December 2013 Ó Springer Science+Business Media New York 2013
In remote corners of the rapidly expanding IBD literature universe, two separate areas of investigation and controversy continue to grow independently. The first is in regard to the risks of diagnostic radiation exposure, whereas the second concerns the disparities in access to care and health care resource utilization in IBD patients. Abdominal imaging is frequently utilized to diagnose disease, detect complications, and ascertain response to treatment in IBD patients. Given a 20-fold rise in the amount of CT imaging between 1980 and 2000, it is not surprising that an estimated three-quarters of the radiation to which IBD patients are exposed is due to CT scanning [1, 2], mostly from abdomino-pelvic examinations, which have largely supplanted conventional contrast enterography in IBD patients. As an example, an 840 % increase in CT enterography use was reported between 2002 and 2007 [3]. Further, as many as 1 in 6 IBD patients are exposed to moderate-high cumulative effective doses (CED) of diagnostic ionizing radiation as measured in millisieverts (mSv) [4]. These figures are important, as frequent exposure to low levels of ionizing radiation have been hypothesized to increase the incidence of malignancy, with as many as 2 % of all cancers worldwide attributable to diagnostic radiation exposure [1, 5]. Although equipoise exists as to the legitimacy of this possible association, there is nonetheless a growing interest in and discussion surrounding the M. Flasar (&) S. Patil Division of Gastroenterology and Hepatology, Department of Medicine, University of Maryland School of Medicine, 100 North Greene Street, Lower Level, Baltimore, MD 21201, USA e-mail: [email protected] M. Flasar Veterans Affairs, Maryland Heath Care System, Baltimore, MD, USA
123
limitation of diagnostic radiation exposures in IBD patients to what is truly necessary for optimal patient management [6]. Parallel to concerns about radiation exposure, several studies have revealed that the IBD universe is not immune to some of the issues that plague the field of medicine as a whole. Racial and socioeconomic disparities in the availability, delivery, and utilization of IBD healthcare resources have been repeatedly demonstrated [7–9]. As these disparities have included differences in use of highly effective medical therapies, emergency department (ED) visits, and gastroenterologist subspecialist care, resultant differences in the degree and frequency of diagnostic radiation exposures are likely. Possible reasons may include barriers to highly effective medical therapy, delays in definitive surgical therapy, and reduced access to gastroenterology specialist and IBD subspecialist care. In this issue of Digestive Diseases and Sciences, Hou et al. [10] report the findings of a retrospective cohort study with the aim of identifying factors associated with highdose radiation exposure in a cohort of IBD patients treated in a county-based ‘‘safety net’’ healthcare system in Houston, TX, USA, from 2000 to 2010. Such healthcare delivery systems are designed to offer universal care access regardless of ability to pay. Since these systems have high proportions of uninsured and under-insured individuals, it is likely that this study population is enriched in IBD patients at risk for increased diagnostic radiation exposure stemming from care disparity issues. The authors identified all possible IBD patients from the 3 inpatient and 14 outpatient facilities in the studied healthcare system using ICD-9 codes. The authors subsequently reviewed electronic medical records and applied accepted IBD diagnostic criteria to reasonably verify a racially and ethnically diverse 278-patient IBD cohort.
Dig Dis Sci (2014) 59:504–506
Patients were approximately 57 % Hispanic, 26 % African American, and 9 % Caucasian. Two-thirds were ‘‘selfpay,’’ whereas \2 % had commercial insurance. The remainder was a 3:1 ratio of Medicaid and Medicare, respectively. The EMR (electronic medical record) of each patient was hand-searched for all diagnostic imaging studies performed following cohort entry; median followup was 6 years. Standardized conversion factors for each study type enabled calculation of CED and annualized radiation exposure for each patient. Given the comprehensive EMR, the authors were able to phenotype patients using the Montreal classification, in addition to obtaining detailed demographic, IBD medication, and surgical histories. Distribution of IBD disease location as defined by Montreal Classification variables was similar to that reported in a large diverse referral cohort, though the reported median age at diagnosis of 48 years was older than expected [11, 12]. Similarly, the cohort proportion of B3 (penetrating) phenotype was relatively higher at the expense of B1 (non-stricturing/non-penetrating), possibly reflecting disease duration and diagnostic delay. Median CED from diagnostic imaging studies was consistent with levels reported in prior tertiary center analyses. Abdominopelvic CT accounted for 58 % of total cohort radiation exposure. Age, gender, and race were not associated with significant differences in cumulative radiation exposure. Compared to UC, CD patients had significantly higher median CED (26.31 vs. 16.18 mSv, p \ 0.01) and annualized radiation dose. Approximately 13.3 % of IBD patients had high CED (defined as the commonly accepted cutoff value of [50 mSV), with CD patients exhibiting the previously reported increased odds of high-dose radiation exposure compared to those with UC [2–4]. Other risk factors for increased odds of high-dose radiation exposure uncovered using univariate analyses included the use of IBD therapeutic medications with the exception of mesalamine, prior bowel resection, increased follow-up duration, and a history of perianal disease in CD patients. Disease distribution was not associated with radiation exposure in either CD or UC patients. Multivariate logistic regression of the entire IBD cohort revealed that only a history of bowel resection and length of follow-up remained significantly associated with increased odds of high-dose radiation exposure, a finding that has been previously reported [4]. In sub-analyses limited to CD and UC, no significant associations were found between any of the covariates and high CED, though trends towards increased odds of high-dose radiation exposure were associated with prior bowel resection in both CD and UC, length of follow-up in UC, and fistulizing disease in CD patients. The authors also performed
505
separate sensitivity analyses in which they limited to patients with CED [75 mSv, analyzing only abdominopelvic imaging studies. Univariate and multivariate results were similar to the primary analyses. Based on these findings, a small but significant proportion of IBD patients in a racially and ethnically diverse urban Texas ‘‘safety-net’’ healthcare network receive high CED from diagnostic radiation, though no factors associated with high CED were identified. The authors should be congratulated for attempting to address two such provocative and controversial issues simultaneously. They utilized a unique dataset which attenuated biases related to differing insurance status, access to care, some referral biases, and limited patient diversity, biases inherent in analyses of radiation exposure and IBD care disparities based on tertiary center populations [2, 4]. However, their analysis highlights both the promise and pitfalls of observational studies in specialized populations. On the surface, what appears initially to be a populationbased analysis should avoid the inherent biases which hinder tertiary center-based studies. However, external validity of such data is problematic, owing to the extremely specific patient cohort analyzed. Further, while the nature of the safety-net system seems to balance issues of insurance and socioeconomic status on some level, the patient characteristics and access to care in the majority ‘‘self pay’’ group are not likely homogenous. As a result, the potential bias attenuation with regard to insurance status and access to care may not be complete, as subtle differences affecting the results may remain. The complexities of socioeconomic status, race, and insurance status are a certain quagmire where even the demonstration of associations can be difficult, let alone causality or determination of underlying mechanism. In the current study, the authors’ reasonable premise is that participants in the safety-net system have generally reduced care access, with resultant less optimal and timely medical and surgical care for their IBD. These factors presumably increase abdominal and overall exposure to unnecessary ionizing radiation, though their data did not support this hypothesis. Nonetheless, the authors’ analysis cannot account for CT-sparing variables such as availability of abdominal MRI and capsule endoscopy. Furthermore, access to subspecialist care and location of imaging (inpatient, ED, etc.), both important mechanistic determinants of any differences in the data, could not be accurately captured. In the end, as is often the case, the strength can become the weakness in observational studies depending on the data source. The same diversity of patient population and insurance status leads to relatively small numbers of outcomes (i.e. CED [50 mSv) in any one subgroup. Therefore, it can be difficult to find significant variables of
123
506
association unless the differences are rather large. For example, an administrative claims analysis aimed at assessing disparities in use of biologics using the Medicaid database reported overall use of anti-TNF to be seemingly low (\7 %), limiting the conclusions about associations between drug exposure and several covariates including race, the variable of most interest [13]. In summary, the authors did not detect overexposure to ionzing radiation in a racially diverse and socioeconimically disadvantaged population. Whether this is due to an absence of disparities or an inability to detect them is not entirely clear. Nonetheless, this study adds to a growing body of literature focused on the possible risks of diagnostic ionizing radiation exposure in IBD patients. Further research should be designed to more clearly assess the relative contributions of factors such as disease severity, access to specialist care, and ED utilization and ordering location of diagnostic imaging (i.e. ED, inpatient, outpatient) on diagnostic ionizing radiation exposure, as well as ways to limit superfluous and redundant testing. In the meantime, IBD providers should remain cognizant of issues surrounding (and thoughtful in the ordering of) IBD diagnostic imaging tests.
Dig Dis Sci (2014) 59:504–506
4.
5.
6.
7.
8.
9.
10.
11.
References 1. Brenner DJ, Hall EJ. Computed tomography—an increasing source of radiation exposure. N Engl J Med. 2007;357: 2277–2284. 2. Kroeker KI, Lam S, Birchall I, Fedorak RN. Patients with IBD are exposed to high levels of ionizing radiation through CT scan diagnostic imaging: a five-year study. J Clin Gastroenterol. 2011;45:34–39. 3. Peloquin JM, Pardi DS, Sandborn WJ, et al. Diagnostic ionizing radiation exposure in a population-based cohort of patients with
123
12.
13.
inflammatory bowel disease. Am J Gastroenterol. 2008;103: 2015–2022. Newnham E, Hawkes E, Surender A, James SL, Gearry R, Gibson PR. Quantifying exposure to diagnostic medical radiation in patients with inflammatory bowel disease: are we contributing to malignancy? Aliment Pharmacol Ther. 2007;26:1019–1024. Berrington de Gonzalez A, Darby S. Risk of cancer from diagnostic X-rays: estimates for the UK and 14 other countries. Lancet. 2004;363:345–351. Israeli E, Ying S, Henderson B, Mottola J, Strome T, Bernstein CN. The impact of abdominal computed tomography in a tertiary referral centre emergency department on the management of patients with inflammatory bowel disease. Aliment Pharmacol Ther. 2013;38:513–521. Sewell JL, Velayos FS. Systematic review: the role of race and socioeconomic factors on IBD healthcare delivery and effectiveness. Inflamm Bowel Dis. 2013;19:627–643. Nguyen GC, LaVeist TA, Harris ML, Wang MH, Datta LW, Brant SR. Racial disparities in utilization of specialist care and medications in inflammatory bowel disease. Am J Gastroenterol. 2010;105:2202–2208. Flasar MH, Johnson T, Roghmann MC, Cross RK. Disparities in the use of immunomodulators and biologics for the treatment of inflammatory bowel disease: a retrospective cohort study. Inflamm Bowel Dis. 2008;14:13–19. Hou JK, Malaty HM, Thirumurthi S. Radiation exposure from diagnostic imaging studies among patients with inflammatory bowel disease in a safety-net health-care system. Dig Dis Sci. (Epub ahead of print). doi:10.1007/s10620-013-2852-1. Nguyen GC, Torres EA, Regueiro M, et al. Inflammatory bowel disease characteristics among African Americans, Hispanics, and non-Hispanic Whites: characterization of a large North American cohort. Am J Gastroenterol. 2006;101:1012–1023. Silverberg MS, Satsangi J, Ahmad T, et al. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: report of a working party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol. 2005;19 Suppl A:5A–36A. Flasar M, Chao J, Yang M, Mulani P, Lu M, Cross R. Racial differences in use of biologics for Crohn’s disease in a medicaid population. Am J Gastroenterol. 2010;105:S443.
EDITORIAL
Radiating Disparity in IBD Mark Flasar • Seema Patil
Published online: 8 December 2013 Ó Springer Science+Business Media New York 2013
In remote corners of the rapidly expanding IBD literature universe, two separate areas of investigation and controversy continue to grow independently. The first is in regard to the risks of diagnostic radiation exposure, whereas the second concerns the disparities in access to care and health care resource utilization in IBD patients. Abdominal imaging is frequently utilized to diagnose disease, detect complications, and ascertain response to treatment in IBD patients. Given a 20-fold rise in the amount of CT imaging between 1980 and 2000, it is not surprising that an estimated three-quarters of the radiation to which IBD patients are exposed is due to CT scanning [1, 2], mostly from abdomino-pelvic examinations, which have largely supplanted conventional contrast enterography in IBD patients. As an example, an 840 % increase in CT enterography use was reported between 2002 and 2007 [3]. Further, as many as 1 in 6 IBD patients are exposed to moderate-high cumulative effective doses (CED) of diagnostic ionizing radiation as measured in millisieverts (mSv) [4]. These figures are important, as frequent exposure to low levels of ionizing radiation have been hypothesized to increase the incidence of malignancy, with as many as 2 % of all cancers worldwide attributable to diagnostic radiation exposure [1, 5]. Although equipoise exists as to the legitimacy of this possible association, there is nonetheless a growing interest in and discussion surrounding the M. Flasar (&) S. Patil Division of Gastroenterology and Hepatology, Department of Medicine, University of Maryland School of Medicine, 100 North Greene Street, Lower Level, Baltimore, MD 21201, USA e-mail: [email protected] M. Flasar Veterans Affairs, Maryland Heath Care System, Baltimore, MD, USA
123
limitation of diagnostic radiation exposures in IBD patients to what is truly necessary for optimal patient management [6]. Parallel to concerns about radiation exposure, several studies have revealed that the IBD universe is not immune to some of the issues that plague the field of medicine as a whole. Racial and socioeconomic disparities in the availability, delivery, and utilization of IBD healthcare resources have been repeatedly demonstrated [7–9]. As these disparities have included differences in use of highly effective medical therapies, emergency department (ED) visits, and gastroenterologist subspecialist care, resultant differences in the degree and frequency of diagnostic radiation exposures are likely. Possible reasons may include barriers to highly effective medical therapy, delays in definitive surgical therapy, and reduced access to gastroenterology specialist and IBD subspecialist care. In this issue of Digestive Diseases and Sciences, Hou et al. [10] report the findings of a retrospective cohort study with the aim of identifying factors associated with highdose radiation exposure in a cohort of IBD patients treated in a county-based ‘‘safety net’’ healthcare system in Houston, TX, USA, from 2000 to 2010. Such healthcare delivery systems are designed to offer universal care access regardless of ability to pay. Since these systems have high proportions of uninsured and under-insured individuals, it is likely that this study population is enriched in IBD patients at risk for increased diagnostic radiation exposure stemming from care disparity issues. The authors identified all possible IBD patients from the 3 inpatient and 14 outpatient facilities in the studied healthcare system using ICD-9 codes. The authors subsequently reviewed electronic medical records and applied accepted IBD diagnostic criteria to reasonably verify a racially and ethnically diverse 278-patient IBD cohort.
Dig Dis Sci (2014) 59:504–506
Patients were approximately 57 % Hispanic, 26 % African American, and 9 % Caucasian. Two-thirds were ‘‘selfpay,’’ whereas \2 % had commercial insurance. The remainder was a 3:1 ratio of Medicaid and Medicare, respectively. The EMR (electronic medical record) of each patient was hand-searched for all diagnostic imaging studies performed following cohort entry; median followup was 6 years. Standardized conversion factors for each study type enabled calculation of CED and annualized radiation exposure for each patient. Given the comprehensive EMR, the authors were able to phenotype patients using the Montreal classification, in addition to obtaining detailed demographic, IBD medication, and surgical histories. Distribution of IBD disease location as defined by Montreal Classification variables was similar to that reported in a large diverse referral cohort, though the reported median age at diagnosis of 48 years was older than expected [11, 12]. Similarly, the cohort proportion of B3 (penetrating) phenotype was relatively higher at the expense of B1 (non-stricturing/non-penetrating), possibly reflecting disease duration and diagnostic delay. Median CED from diagnostic imaging studies was consistent with levels reported in prior tertiary center analyses. Abdominopelvic CT accounted for 58 % of total cohort radiation exposure. Age, gender, and race were not associated with significant differences in cumulative radiation exposure. Compared to UC, CD patients had significantly higher median CED (26.31 vs. 16.18 mSv, p \ 0.01) and annualized radiation dose. Approximately 13.3 % of IBD patients had high CED (defined as the commonly accepted cutoff value of [50 mSV), with CD patients exhibiting the previously reported increased odds of high-dose radiation exposure compared to those with UC [2–4]. Other risk factors for increased odds of high-dose radiation exposure uncovered using univariate analyses included the use of IBD therapeutic medications with the exception of mesalamine, prior bowel resection, increased follow-up duration, and a history of perianal disease in CD patients. Disease distribution was not associated with radiation exposure in either CD or UC patients. Multivariate logistic regression of the entire IBD cohort revealed that only a history of bowel resection and length of follow-up remained significantly associated with increased odds of high-dose radiation exposure, a finding that has been previously reported [4]. In sub-analyses limited to CD and UC, no significant associations were found between any of the covariates and high CED, though trends towards increased odds of high-dose radiation exposure were associated with prior bowel resection in both CD and UC, length of follow-up in UC, and fistulizing disease in CD patients. The authors also performed
505
separate sensitivity analyses in which they limited to patients with CED [75 mSv, analyzing only abdominopelvic imaging studies. Univariate and multivariate results were similar to the primary analyses. Based on these findings, a small but significant proportion of IBD patients in a racially and ethnically diverse urban Texas ‘‘safety-net’’ healthcare network receive high CED from diagnostic radiation, though no factors associated with high CED were identified. The authors should be congratulated for attempting to address two such provocative and controversial issues simultaneously. They utilized a unique dataset which attenuated biases related to differing insurance status, access to care, some referral biases, and limited patient diversity, biases inherent in analyses of radiation exposure and IBD care disparities based on tertiary center populations [2, 4]. However, their analysis highlights both the promise and pitfalls of observational studies in specialized populations. On the surface, what appears initially to be a populationbased analysis should avoid the inherent biases which hinder tertiary center-based studies. However, external validity of such data is problematic, owing to the extremely specific patient cohort analyzed. Further, while the nature of the safety-net system seems to balance issues of insurance and socioeconomic status on some level, the patient characteristics and access to care in the majority ‘‘self pay’’ group are not likely homogenous. As a result, the potential bias attenuation with regard to insurance status and access to care may not be complete, as subtle differences affecting the results may remain. The complexities of socioeconomic status, race, and insurance status are a certain quagmire where even the demonstration of associations can be difficult, let alone causality or determination of underlying mechanism. In the current study, the authors’ reasonable premise is that participants in the safety-net system have generally reduced care access, with resultant less optimal and timely medical and surgical care for their IBD. These factors presumably increase abdominal and overall exposure to unnecessary ionizing radiation, though their data did not support this hypothesis. Nonetheless, the authors’ analysis cannot account for CT-sparing variables such as availability of abdominal MRI and capsule endoscopy. Furthermore, access to subspecialist care and location of imaging (inpatient, ED, etc.), both important mechanistic determinants of any differences in the data, could not be accurately captured. In the end, as is often the case, the strength can become the weakness in observational studies depending on the data source. The same diversity of patient population and insurance status leads to relatively small numbers of outcomes (i.e. CED [50 mSv) in any one subgroup. Therefore, it can be difficult to find significant variables of
123
506
association unless the differences are rather large. For example, an administrative claims analysis aimed at assessing disparities in use of biologics using the Medicaid database reported overall use of anti-TNF to be seemingly low (\7 %), limiting the conclusions about associations between drug exposure and several covariates including race, the variable of most interest [13]. In summary, the authors did not detect overexposure to ionzing radiation in a racially diverse and socioeconimically disadvantaged population. Whether this is due to an absence of disparities or an inability to detect them is not entirely clear. Nonetheless, this study adds to a growing body of literature focused on the possible risks of diagnostic ionizing radiation exposure in IBD patients. Further research should be designed to more clearly assess the relative contributions of factors such as disease severity, access to specialist care, and ED utilization and ordering location of diagnostic imaging (i.e. ED, inpatient, outpatient) on diagnostic ionizing radiation exposure, as well as ways to limit superfluous and redundant testing. In the meantime, IBD providers should remain cognizant of issues surrounding (and thoughtful in the ordering of) IBD diagnostic imaging tests.
Dig Dis Sci (2014) 59:504–506
4.
5.
6.
7.
8.
9.
10.
11.
References 1. Brenner DJ, Hall EJ. Computed tomography—an increasing source of radiation exposure. N Engl J Med. 2007;357: 2277–2284. 2. Kroeker KI, Lam S, Birchall I, Fedorak RN. Patients with IBD are exposed to high levels of ionizing radiation through CT scan diagnostic imaging: a five-year study. J Clin Gastroenterol. 2011;45:34–39. 3. Peloquin JM, Pardi DS, Sandborn WJ, et al. Diagnostic ionizing radiation exposure in a population-based cohort of patients with
123
12.
13.
inflammatory bowel disease. Am J Gastroenterol. 2008;103: 2015–2022. Newnham E, Hawkes E, Surender A, James SL, Gearry R, Gibson PR. Quantifying exposure to diagnostic medical radiation in patients with inflammatory bowel disease: are we contributing to malignancy? Aliment Pharmacol Ther. 2007;26:1019–1024. Berrington de Gonzalez A, Darby S. Risk of cancer from diagnostic X-rays: estimates for the UK and 14 other countries. Lancet. 2004;363:345–351. Israeli E, Ying S, Henderson B, Mottola J, Strome T, Bernstein CN. The impact of abdominal computed tomography in a tertiary referral centre emergency department on the management of patients with inflammatory bowel disease. Aliment Pharmacol Ther. 2013;38:513–521. Sewell JL, Velayos FS. Systematic review: the role of race and socioeconomic factors on IBD healthcare delivery and effectiveness. Inflamm Bowel Dis. 2013;19:627–643. Nguyen GC, LaVeist TA, Harris ML, Wang MH, Datta LW, Brant SR. Racial disparities in utilization of specialist care and medications in inflammatory bowel disease. Am J Gastroenterol. 2010;105:2202–2208. Flasar MH, Johnson T, Roghmann MC, Cross RK. Disparities in the use of immunomodulators and biologics for the treatment of inflammatory bowel disease: a retrospective cohort study. Inflamm Bowel Dis. 2008;14:13–19. Hou JK, Malaty HM, Thirumurthi S. Radiation exposure from diagnostic imaging studies among patients with inflammatory bowel disease in a safety-net health-care system. Dig Dis Sci. (Epub ahead of print). doi:10.1007/s10620-013-2852-1. Nguyen GC, Torres EA, Regueiro M, et al. Inflammatory bowel disease characteristics among African Americans, Hispanics, and non-Hispanic Whites: characterization of a large North American cohort. Am J Gastroenterol. 2006;101:1012–1023. Silverberg MS, Satsangi J, Ahmad T, et al. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: report of a working party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol. 2005;19 Suppl A:5A–36A. Flasar M, Chao J, Yang M, Mulani P, Lu M, Cross R. Racial differences in use of biologics for Crohn’s disease in a medicaid population. Am J Gastroenterol. 2010;105:S443.
