In the article Predicting 10-year survival of patients with primary cutaneous melanoma: corroboration of a prognostic model
Question:
In the article “Predicting 10-year survival of patients with primary cutaneous melanoma: corroboration of a prognostic model” published in Cancer
(Sahin, 1997), the authors reported a logistic regression model for the 10-year survival probability of melanoma patients. The variables in the model are age coded as 1 when a patients age is ≤ 60 and 0 otherwise, sex coded as 1 for female and 0 for male, lesion thickness entered as three dummy variables for the intervals < 0.76, 0.76–1.69, and 1.70–3.60, and a dummy variable for the location of the lesion coded as 1 for a lesion on an extremity and 0 otherwise. The model was fit using the data from ???? = 780 melanoma patients. Use the summary of the fitted logistic regression model given in Table 13.18 to answer the following:
(a) Determine the odds ratios associated with each of the explanatory variables.
(b) How many times more likely is it for female to survive 10 years than for a male, assuming all of the other explanatory variables are the same?
(c) Estimate the 10-year survival probability for a 65-year-old female with a primary extremity lesion of tumor depth 1 mm.
(d) Estimate the 10-year survival probability for a 65-year-old male with a primary extremity lesion of tumor depth 1 mm.
(e) Estimate the 10-year survival probability for a 55-year-old-female with a primary extremity lesion of tumor depth 4 mm.
(f) Estimate the 10-year survival probability for a 55-year-old male with a primary extremity lesion of tumor depth 4 mm.
Step by Step Answer:
Applied Biostatistics For The Health Sciences
ISBN: 9781119722694
2nd Edition
Authors: Richard J. Rossi