Relation between Demographic Factors and Type of Gastrointestinal Cancer using Probit and Logit Regression


Background: Approximately 50,000 new cases of cancer occur each year in the Iranian population of 70.4million. The organ system involved with more than 38% of all cancers is the gastrointestinal (GI) tract. Theobjective of this study was to investigate the relation between demographic factors and type of gastrointestinalcancer using probit and logit models.
Methods: This study was designed as a cross-sectional survey including allconsecutive GI cancer patients admitted over a one year period in a randomly selected hospital group located inTehran in 2006.
Results: The largest number of cases were colorectal cancers (40.0%), followed by gastriccancers (34.5%) and esophagus cancers (17.1%). There was a significant gender effect in the colorectal, gastricand esophagus cancer also there was a significant association between age and gastrointestinal cancers in bothlogit and probit regression. The factor of duration was not significant in gastric cancer.
Conclusion: Men aremore likelyhave colorectal cancer than women. Older people are more likely to have gastric cancer than youngerpeople. In esophagus cancer all factors were significant. Results from probit and logit models were similar,indicating that probit analysis can be employed as a logit model to analyze relationships between demographicfactors and cancer type.