Srawani Sarkar1, Marc J Dauer1, Haejin In2,3. 1. Department of Surgery, Montefiore Medical Center/Albert Einstein College of Medicine, Bronx, NY, USA. 2. Department of Surgery, Montefiore Medical Center/Albert Einstein College of Medicine, Bronx, NY, USA. hin@montefiore.org. 3. Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY, USA. hin@montefiore.org.
Abstract
INTRODUCTION: Socioeconomic status (SES) is a known risk factor for gastric cancer (GC). This study seeks to examine education, income, and occupation variables separately to identify the single variable that can be best used to assess SES risk for GC. METHODS: Data from a case-control survey study were used. Logistic regression models were created for education, income, and occupation adjusted for age, sex, and race. Models were compared using AIC, c-statistics, and pseudo-R square to determine the model that had the highest risk predictive ability. RESULTS: GC cases had lower education levels and more commonly held jobs in unskilled labor. Annual household income was lower in cases compared to controls. Age, gender, race, education, and occupation were associated with increased risk of GC. The education model adjusted for age, gender, and race found < high school (HS) education to have an OR of 3.18 (95% CI 1.09-9.25) for GC compared to > HS education. The occupation model demonstrated that employment in unskilled labor had OR of 4.32 (95% CI 1.05-17.76) for GC compared to professional occupation. Model fit was best for the education model (AIC: 113.583, lower AIC is better) compared to income (117.562) or occupation (117.032). Education contributed the most to model variability (% delta pseudo-R square (4.7%)) compared to occupation (4.0%) or income (3.8%). CONCLUSION: Education level was the single most reliable measure of GC risk among 3 SES variables and can be employed as an ideal single indicator of SES-related GC risk when multiple SES factors cannot be obtained.
INTRODUCTION: Socioeconomic status (SES) is a known risk factor for gastric cancer (GC). This study seeks to examine education, income, and occupation variables separately to identify the single variable that can be best used to assess SES risk for GC. METHODS: Data from a case-control survey study were used. Logistic regression models were created for education, income, and occupation adjusted for age, sex, and race. Models were compared using AIC, c-statistics, and pseudo-R square to determine the model that had the highest risk predictive ability. RESULTS: GC cases had lower education levels and more commonly held jobs in unskilled labor. Annual household income was lower in cases compared to controls. Age, gender, race, education, and occupation were associated with increased risk of GC. The education model adjusted for age, gender, and race found < high school (HS) education to have an OR of 3.18 (95% CI 1.09-9.25) for GC compared to > HS education. The occupation model demonstrated that employment in unskilled labor had OR of 4.32 (95% CI 1.05-17.76) for GC compared to professional occupation. Model fit was best for the education model (AIC: 113.583, lower AIC is better) compared to income (117.562) or occupation (117.032). Education contributed the most to model variability (% delta pseudo-R square (4.7%)) compared to occupation (4.0%) or income (3.8%). CONCLUSION: Education level was the single most reliable measure of GC risk among 3 SES variables and can be employed as an ideal single indicator of SES-related GC risk when multiple SES factors cannot be obtained.
Authors: D F Cella; E J Orav; A B Kornblith; J C Holland; P M Silberfarb; K W Lee; R L Comis; M Perry; R Cooper; L H Maurer Journal: J Clin Oncol Date: 1991-08 Impact factor: 44.544
Authors: Paula A Braveman; Catherine Cubbin; Susan Egerter; David R Williams; Elsie Pamuk Journal: Am J Public Health Date: 2010-02-10 Impact factor: 9.308
Authors: Seri Hong; Young-Joo Won; Young Ran Park; Kyu-Won Jung; Hyun-Joo Kong; Eun Sook Lee Journal: Cancer Res Treat Date: 2020-03-16 Impact factor: 4.679
Authors: Leila Cattelan; Feras M Ghazawi; Michelle Le; François Lagacé; Elham Rahme; Andrei Zubarev; Denis Sasseville; Ivan V Litvinov; Kevin A Waschke; Elena Netchiporouk Journal: Curr Oncol Date: 2021-05-28 Impact factor: 3.677