Hemanth Veluswamy; Kunal Suryawala; Ankur Sheth; Shannon Wells; Erik Salvatierra; Walter Cromer; Ganta V Chaitanya; Annette Painter; Mihir Patel; Kenneth Manas; Ellenmarie Zwank; Moheb Boktor; Kondal Baig; Balaji Datti; Michael J Mathis; Alireza Minagar; Paul A Jordan; Jonathan S Alexander
BMC Gastroenterology. 2010;10(104) © 2010 BioMed Central, Ltd.
Abstract and Introduction
Abstract
Background: Inflammatory Bowel Diseases (IBD) remain significant health problems in the US and worldwide. IBD is most often associated with eastern European ancestry, and is less frequently reported in other populations of African origin e.g. African Americans ('AAs'). Whether AAs represent an important population with IBD in the US remains unclear since few studies have investigated IBD in communities with a majority representation of AA patients. The Louisiana State University Health Sciences Center in Shreveport (LSUHSC-S) is a tertiary care medical center, with a patient base composed of 58% AA and 39% Caucasian (W), ideal for evaluating racial (AA vs. W) as well and gender (M vs. F) influences on IBD.
Methods: In this retrospective study, we evaluated 951 visits to LSUHSC-S for IBD (between 2000 to 2008) using non-identified patient information based on ICD-9 medical record coding (Crohn's disease 'CD'-555.0- 555.9 and ulcerative colitis 'UC'-556.0–556.9).
Results: Overall, there were more cases of CD seen than UC. UC and CD affected similar ratios of AA and Caucasian males (M) and females (F) with a rank order of WF > WM > AAF > AAM. Interestingly, in CD, we found that annual visits per person was the highest in AA M (10.7 ± 1.7); significantly higher (* -p < 0.05) than in WM (6.3 ± 1.0). Further, in CD, the female to male (F: M) ratio in AA was significantly higher (*- p < 0.05) (1.9 ± 0.2) than in Caucasians (F:M = 1.3 ± 0.1) suggesting a female dominance in AACD; no differences were seen in UC F: M ratios.
Conclusion: Although Caucasians still represent the greatest fraction of IBD (~64%), AAs with IBD made up >1/3 (36.4%) of annual IBD cases from 2000–2008 at LSUHSC-S. Further studies on genetic and environments risks for IBD risk in AAs are needed to understand differences in presentation and progression in AAs and other 'non-traditional' populations.
Background
Crohn's disease (CD) and ulcerative colitis (UC), the major forms of inflammatory bowel disease (IBD), are characterized clinically by diarrhea, weight loss and fever as well as endoscopic, radiologic, histopathologic findings and biochemical markers (e.g. perinuclear anti-neutrophil cytoplasm (p-ANCA), anti-Saccharomyces cerevisiae (ASCA) and IBD-specific p-ANCA markers).[1] The development of IBD is thought to depend on several factors including genetic background, environmental influences e.g. dietary "hygiene", parasite burden and subclinical infectious diseases.[2–5] These factors lead to a complex overall distribution of IBD, with some described patterns in disease prevalence. IBD is considered a disease of developed nations especially Northern Europe and the United States, usually of colder climates with increased incidence as distance from the equator increases.[6–10] IBD is now increasingly reported in non-classical populations and in developing regions such as Asia, the Mid-East and in Africa.[11,12]
In a study of central-African (Ghana) black patients[13] CD was viewed as a common, but often under-diagnosed condition, while Zaahl et al.[14] have reported that UC is relatively uncommon in South African blacks. In the US, IBD, (especially CD) is still regarded as a syndrome affecting individuals of European descent, but there is increasing evidence for IBD in the Hispanic and African American (AA) populations.[15]
It is possible that clinical features of IBD may differ in AAs, leading to misdiagnosis and under-reporting of 'minority' IBD.[15] Eidelwein et al.[16] reported that pediatric CD was possibly more severe in AAs, often presenting with lower body mass scores, blood sedimentation rates and hemoglobin (Hb). AAs may also require more treatment (e.g. steroids, anti-TNF-α) to manage their IBD. In a study of pediatric/adolescent AA cases, White et al.,[17] found that AAs were older at the time of diagnosis/onset, had a greater incidence of CD (>UC>indeterminate colitis), and lower Hb. Conversely, a meta-analysis of 8 studies with >2000 individuals suggested that no race-based differences exist in IBD susceptibility.[18] However, regional studies may suggest differences in IBD in AA populations exist. Ogunbi et al. report a CD incidence in AAs of 7–12 per 100,000 and 5–7 per 100,000 for UC.[19] In African populations, Shapira and Tamir[20] suggested that African CD incidence equals or exceeds that in European-Americans, consistent with underestimates in both African or AA IBD.
Incongruities about the incidence of IBD in AA populations may thus reflect low sampling, geographical influences, variations in disease presentation, environmental and socioeconomic factors that confound diagnosis and ultimately treatment. Gender-based differences in IBD have also been reported;[21] the F: M ratio in IBD may also be affected by urban/rural status[22–24] and may be changing in recent years.[25]
Sample Population of Shreveport/LSUHSC-S
One important limitation in this type of study is adequate sampling, in this case, ideally selecting populations in which AA patients at least match, or even exceed that of Caucasians. Our study carried out at the LSU Health Sciences Center in Shreveport (in accordance with HIPAA regulations regarding privacy) should approximate local ethnic profiles in Shreveport,/Arkansas-Louisiana-Texas region (''Arklatex''). The racial composition of Shreveport has been >50% AA since 2002;[26] the self-identified ethnic background of patients at LSUHSC-S (1999–2008) was 58.11% AA, and 39.42% Caucasian, (plus 1.43% Hispanic and 1.04% Asian, Native American plus other groups). Our goal was to evaluate the proportion of IBD cases as related to race and gender at LSUHSC-Shreveport, LA, a hospital serving the Northwest LA region with a roughly equal racial (AA : W) ratio.
Methods
This study was reviewed and approved by the LSUHSC-Shreveport Human Research Protection Program/Institutional Review Board (IRB). We used non-identified ICD-9 codes (international statistical classification of diseases) from information medical record codes for CD (555.0 - 555.9) and UC (556.0 - 556.9). For every year in the study, individual non-identified patients were evaluated for self-identified race, gender and ICD-9 coding, (each patient was counted only once annually). Ages of individual patients at the time of annual visits were also recorded. Cumulative cases of CD or UC visits per each group were also counted (independent of the number of patients). We then determined: 1) the annual number of patients for each code, 2) the number of hospital visits per group, 3) the age at visit for individual racial and 4) gender populations at the LSUHSC in Shreveport, LA, (between 2000 and 2008). Hospital visits per individual were calculated as the total number of hospital visits (separated by race and gender) for each specific condition (CD or UC) and dividing this number by the number of cases for each condition. One-way ANOVA analysis was used to determine statistical significance with Bonferroni Multiple Comparisons post-testing. Statistical significance between pairs of data was determined using unpaired student's t-test. Nonlinear regression was used to calculate correlation and determine fit for trend lines in graphs. Because Hispanic, Asian, and Native American populations constitute <1% of total case reports for ICD-9 codes, these groups were omitted from analysis; we compared Caucasian and African-American populations with a further sub-grouping for gender. Co-morbid conditions with CD and UC, including diabetes (ICD-9 code 250.0), multiple sclerosis (ICD-9 code 340), and arthritis, were also investigated.
BMC Gastroenterology. 2010;10(104) © 2010 BioMed Central, Ltd.
Abstract and Introduction
Abstract
Background: Inflammatory Bowel Diseases (IBD) remain significant health problems in the US and worldwide. IBD is most often associated with eastern European ancestry, and is less frequently reported in other populations of African origin e.g. African Americans ('AAs'). Whether AAs represent an important population with IBD in the US remains unclear since few studies have investigated IBD in communities with a majority representation of AA patients. The Louisiana State University Health Sciences Center in Shreveport (LSUHSC-S) is a tertiary care medical center, with a patient base composed of 58% AA and 39% Caucasian (W), ideal for evaluating racial (AA vs. W) as well and gender (M vs. F) influences on IBD.
Methods: In this retrospective study, we evaluated 951 visits to LSUHSC-S for IBD (between 2000 to 2008) using non-identified patient information based on ICD-9 medical record coding (Crohn's disease 'CD'-555.0- 555.9 and ulcerative colitis 'UC'-556.0–556.9).
Results: Overall, there were more cases of CD seen than UC. UC and CD affected similar ratios of AA and Caucasian males (M) and females (F) with a rank order of WF > WM > AAF > AAM. Interestingly, in CD, we found that annual visits per person was the highest in AA M (10.7 ± 1.7); significantly higher (* -p < 0.05) than in WM (6.3 ± 1.0). Further, in CD, the female to male (F: M) ratio in AA was significantly higher (*- p < 0.05) (1.9 ± 0.2) than in Caucasians (F:M = 1.3 ± 0.1) suggesting a female dominance in AACD; no differences were seen in UC F: M ratios.
Conclusion: Although Caucasians still represent the greatest fraction of IBD (~64%), AAs with IBD made up >1/3 (36.4%) of annual IBD cases from 2000–2008 at LSUHSC-S. Further studies on genetic and environments risks for IBD risk in AAs are needed to understand differences in presentation and progression in AAs and other 'non-traditional' populations.
Background
Crohn's disease (CD) and ulcerative colitis (UC), the major forms of inflammatory bowel disease (IBD), are characterized clinically by diarrhea, weight loss and fever as well as endoscopic, radiologic, histopathologic findings and biochemical markers (e.g. perinuclear anti-neutrophil cytoplasm (p-ANCA), anti-Saccharomyces cerevisiae (ASCA) and IBD-specific p-ANCA markers).[1] The development of IBD is thought to depend on several factors including genetic background, environmental influences e.g. dietary "hygiene", parasite burden and subclinical infectious diseases.[2–5] These factors lead to a complex overall distribution of IBD, with some described patterns in disease prevalence. IBD is considered a disease of developed nations especially Northern Europe and the United States, usually of colder climates with increased incidence as distance from the equator increases.[6–10] IBD is now increasingly reported in non-classical populations and in developing regions such as Asia, the Mid-East and in Africa.[11,12]
In a study of central-African (Ghana) black patients[13] CD was viewed as a common, but often under-diagnosed condition, while Zaahl et al.[14] have reported that UC is relatively uncommon in South African blacks. In the US, IBD, (especially CD) is still regarded as a syndrome affecting individuals of European descent, but there is increasing evidence for IBD in the Hispanic and African American (AA) populations.[15]
It is possible that clinical features of IBD may differ in AAs, leading to misdiagnosis and under-reporting of 'minority' IBD.[15] Eidelwein et al.[16] reported that pediatric CD was possibly more severe in AAs, often presenting with lower body mass scores, blood sedimentation rates and hemoglobin (Hb). AAs may also require more treatment (e.g. steroids, anti-TNF-α) to manage their IBD. In a study of pediatric/adolescent AA cases, White et al.,[17] found that AAs were older at the time of diagnosis/onset, had a greater incidence of CD (>UC>indeterminate colitis), and lower Hb. Conversely, a meta-analysis of 8 studies with >2000 individuals suggested that no race-based differences exist in IBD susceptibility.[18] However, regional studies may suggest differences in IBD in AA populations exist. Ogunbi et al. report a CD incidence in AAs of 7–12 per 100,000 and 5–7 per 100,000 for UC.[19] In African populations, Shapira and Tamir[20] suggested that African CD incidence equals or exceeds that in European-Americans, consistent with underestimates in both African or AA IBD.
Incongruities about the incidence of IBD in AA populations may thus reflect low sampling, geographical influences, variations in disease presentation, environmental and socioeconomic factors that confound diagnosis and ultimately treatment. Gender-based differences in IBD have also been reported;[21] the F: M ratio in IBD may also be affected by urban/rural status[22–24] and may be changing in recent years.[25]
Sample Population of Shreveport/LSUHSC-S
One important limitation in this type of study is adequate sampling, in this case, ideally selecting populations in which AA patients at least match, or even exceed that of Caucasians. Our study carried out at the LSU Health Sciences Center in Shreveport (in accordance with HIPAA regulations regarding privacy) should approximate local ethnic profiles in Shreveport,/Arkansas-Louisiana-Texas region (''Arklatex''). The racial composition of Shreveport has been >50% AA since 2002;[26] the self-identified ethnic background of patients at LSUHSC-S (1999–2008) was 58.11% AA, and 39.42% Caucasian, (plus 1.43% Hispanic and 1.04% Asian, Native American plus other groups). Our goal was to evaluate the proportion of IBD cases as related to race and gender at LSUHSC-Shreveport, LA, a hospital serving the Northwest LA region with a roughly equal racial (AA : W) ratio.
Methods
This study was reviewed and approved by the LSUHSC-Shreveport Human Research Protection Program/Institutional Review Board (IRB). We used non-identified ICD-9 codes (international statistical classification of diseases) from information medical record codes for CD (555.0 - 555.9) and UC (556.0 - 556.9). For every year in the study, individual non-identified patients were evaluated for self-identified race, gender and ICD-9 coding, (each patient was counted only once annually). Ages of individual patients at the time of annual visits were also recorded. Cumulative cases of CD or UC visits per each group were also counted (independent of the number of patients). We then determined: 1) the annual number of patients for each code, 2) the number of hospital visits per group, 3) the age at visit for individual racial and 4) gender populations at the LSUHSC in Shreveport, LA, (between 2000 and 2008). Hospital visits per individual were calculated as the total number of hospital visits (separated by race and gender) for each specific condition (CD or UC) and dividing this number by the number of cases for each condition. One-way ANOVA analysis was used to determine statistical significance with Bonferroni Multiple Comparisons post-testing. Statistical significance between pairs of data was determined using unpaired student's t-test. Nonlinear regression was used to calculate correlation and determine fit for trend lines in graphs. Because Hispanic, Asian, and Native American populations constitute <1% of total case reports for ICD-9 codes, these groups were omitted from analysis; we compared Caucasian and African-American populations with a further sub-grouping for gender. Co-morbid conditions with CD and UC, including diabetes (ICD-9 code 250.0), multiple sclerosis (ICD-9 code 340), and arthritis, were also investigated.