Analysis of patients with colorectal cancer treated with surgery in 2015 and 2016, using a statistical model

Abstract

The purpose of this study was to analyze the data of the open and laparoscopic colorectal cancer surgeries countrywide in an administrative database of the National Health Insurance Fund of Hungary in 2015-2016. Hungarian surgical data of colorectal cancer from 2015-2016 were analyzed based on the following objectives: diagnosis of care, complications and co-morbidities; age; sex and medical institutions. We used crosstabs and partial least squares path modeling to analyze the relations between data groups. Results of three LVPLS models (group of open, laparoscopic interventions and the whole population of the survey) were compared. From the 4941 analyzed operative cases, 4562 were open and 379 laparoscopic. Based on the three LVPLS models, we found nexus between complication groups and co-morbidity groups and progressivity of medical institutions in laparoscopic operations. As far as complications, in the laparoscopic model bleeding and anaemia, in cases of total population and open operations septic complications played dominant role. Laparoscopic resections in patients with multiple co-morbidities are carried out in medical institutions with high professional progressivity most frequently. In cases of laparoscopic colorectal operations, the higher the progressivity of an institution, the lower the complication rate was. Differences in the results in the models may be due to patient selection, and further research should be carried out in this regard. Anaemia and bleeding are dominant complication factors in the laparoscopic model, which are influenced by patient selection besides the differenced in surgical technique. Orv Hetil. 2019; 160(11): 426-434.