An assessment of colonoscopy recommendations and the provision of genetic counseling for families of colorectoral cancer patients in Newfoundland and Labrador

McNicholas, Patrick J. (2021) An assessment of colonoscopy recommendations and the provision of genetic counseling for families of colorectoral cancer patients in Newfoundland and Labrador. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (2MB)

Abstract

Introduction Family members of patients who develop colorectal cancer (CRC) are at higher than population risk of developing CRC. Screening can prevent CRC by finding and excising precancerous polyps. Family members at risk of CRC require screening, starting at different ages and varying frequency depending on family history. The purpose of this thesis is to investigate clinical, pathological, demographical and family history variables in patients with CRC in an attempt to predict clinical decisions important in their management: referral to genetic counselors and frequency of colonoscopy in family members. Methodology The cohort used was from a familial CRC clinic started in 2010 in urban and rural regions of Newfoundland and Labrador (NL). Patients in the clinic were pathologically confirmed incident CRC patients from the Newfoundland Cancer Registry presenting between 2008-2010. Of the 1091 cases presented to the two health regions, 525 were considered for analysis. Pedigrees were reviewed by genetic counselors and family risk was determined using a variety of criteria. Multivariate models were created to predict the need for genetic counseling and the frequency of screening colonoscopy in family members as recommended by the genetic counselor. Results The multivariate model created to predict the need for provision of genetic counseling (GC) included eight predictive variables, had a sensitivity of 81.3% and specificity of 87.4%. Scores in 25 of 134 (19%) patients recommended for GC differed from the decision of the genetic counselor. Review of clinical information demonstrated that the decision of our model was more clinically apt than that of the genetic counselor. The model to predict screening at 5-10-year intervals included nine predictive variables, had a sensitivity of 90.3% and specificity of 76.3%, and identified 30 (9%) patients that differed from the decision of the genetic counselor. The model to predict screening at 2-3-year intervals included 7 predictive variables, had a sensitivity of 69.1% and specificity of 87.4%, and identified 21 (31%) patients that differed from the decision made by the genetic counselor. The model to predict 3-5-year screening intervals compared to 2-3-year intervals included five predictive variables, had a sensitivity of 75.0% and specificity of 55.6%. This prediction of this decision was poor and was not considered to be clinically important. Conclusion Reasonable prediction models of patients who had families that needed to see a genetic counselor and of those whose family members needed colonoscopy at 5-10-year, or 2-3-year intervals were obtained. Clinical information on patients where the decision made by the multivariate score differed from the decision made by the genetic counselor suggested the score’s decision was more clinically apt. Application of these multivariate scores in a population-based program of incident CRC patients has the potential to reduce mortality rates from CRC in NL.

Actions (login required)

View Item