Cancer Survival Analysis

Day 1 (November 6, 2017)
Concept of time to event analysis
Descriptive analysis (Life table, Kaplan Meier, Median survival, Log rank)
Regression: semi-parametric vs parametric

Cancer survival analysis: clinical trial data, hospital-based data, population-based data
Core measures of cancer surveillance (Incidence, Survival, Mortality)
Bias (lead time, length time, overdiagnosis)
Population-based cancer survival: importance and use


Day 2 (November 7, 2017)
Overall survival
Cancer prognosis (remove competing causes of death): relative survival, cause-specific survival
Actual prognosis (include competing causes of death)
Relative survival ratio and Net survival: Ederer I, Ederer II, Hakulinen, Pohar Perme
Age-standardized relative survival (Standard cancer patient population)

Cohort vs Period vs Complete approach
Practice with given a data set
Practicum in relative survival analysis
Practicum in risk models for survival

Workshop materials

Cancer Incidence Trends Analysis

Day 1 (October 30, 2017)
Joinpoint and age-period-cohort model
Over-all trend (APC)
Age, period, cohort effects

Population denominators – past and future
Making a data frame for joinpoint and APC analyses
Merging incidence and population dataframes
(lecture and practicum)


Day 2 (October 31, 2017)
Steps in APC tutorials – trend analysis
(lecture and practicum)

Selection of the best model – AP, AC, AP-C, AC-P, APC
Interpretation of age, period, cohort effects
(lecture and practicum)


Day 3 (November 1, 2017)
Steps in joinpoint and APC projection
Cut trend methods
(lecture and practicum)

Interpretation of the projection
Limitation of use of trend projection
  

Files used in the workshop for 'Cancer Incidence Trends Analysis'. They are arranged in the following folders.

0 data

1 basic concepts of joinpoint analysis

2 basic concepts of age-period-cohort analysis

3 annual percent change and average APC

4 age, period, cohort effects 

5 data management for joinpoint 

6 steps in joinpoint tutorials - trend analysis 

7 data management for age-period-cohort analysis 

8 steps in APC tutorials - trend analysis 

9 data management for nordpred 

10 steps in nordpred tutorial - analysis

11 population denominators 

12 selection of the best model -AP, AC, APC

13 interpretation of age, period, cohort effects 

14 data management for projection 

15 steps in joinpoint APC and nordpred projection

16 cut trend methods

17 interpretation of the projection 

18 limitation of use of trend projection 

19 demonstration of StMoMo 

Cancer Trend Analysis Using Joinpoint Regression Part 1 The Basics.pdf

epaac-wp9-session2-crocetti.pdf

PHT2011 Trend Analysis.ppt 

software

Proposal writing

Files used in the workshop for 'proposal writing'.

Drafting a Proposal
Vancouver Condensed Guide
CI5vol10
International Rules for Multiple Primary Cancers
Good Clinical Practice-Hutcha Sriplung
Curriculum Vitae-Hutcha Sriplung

Journals and Instruction to Authors
Introduction
Methods, Statistical Analysis, and Results

Software for sample size calculation
GPower3Windows.zip
GPowerMac_3.1.9.3.zip

Analysis of Cancer Registry Data

Materials for the workshop on analysis of cancer registry data at Faculty of Medicine, Prince of Songkla University, 28-30 August 2017.

Slide presentations
Breast and cervical cancer
Lung and liver cancer
Additional issues
Age standardization
International resources
Concept of trend analysis
Cancer registry data
CONCORD and SURVCAN data
Maps
Concepts of net/relative survival
Analysis of cancer registry data
Data exploration and data quality

Instructions
คู่มือการใช้โปรแกรม R สำหรับวิเคราะห์ข้อมูลทะเบียนมะเร็ง
Ex.Thailand Songkhla CI5-X

R script files
1-R basics and data import export.R
2.1-Tabulation and Graphics.R
2.2-Tabulation and Graphics.R
2.3-Tabulation and Graphics.R
2.4-Tabulation and Graphics.R
3-Stage distribution.R
4-histology distribution.R
5.1-Population pyramids 2.R
5.2-Population pyramids.R
6.1-General trend plots-male.R
6.2-General trend plots-female.R
7-analysis of cancer data.R
8-data exploration.R
9-number and incidence tables.R

Packages
CRStat_3.4.0.1.tgz
CRStat_3.4.0.1.tar.gz
CRStat_3.4.0.1.zip

Poisson regression, 2016

Hi,

These are materials for Poisson regression course, academic year 2016 (2017). This year I use R version 3.3.2. The results are similar to those obtained from the previous versions of R.

montana.dat
montana.dta
docsmoke.dat
docsmoke.dta
welsh.dat
welsh.dta

The functions for Poisson regression are within ice package.

The module can be downloaded here.
Poisson1602.pdf

You can also download ICE modules from ice and epid. However, they are not necessary for this session.

Finally, you can follow my R script file below. Don't forget to change working directory to yours.

poisson.R

Logistic regression II, Conditional logistic regression, 2016

Hi,

These are materials for logistic regression course II: Conditional logistic regression, academic year 2016 (2017). This year I use R version 3.3.2 and the regression outputs are the same as that produced from the previous version of R.

Conditional logistic regression requires library survival that already exists on your R library folder. Just call library(survival) or require(survival) and you are ready to use clogit function.

agechd.dta
cca-match.dta

The module can be downloaded here.

logistic1602-2.pdf

The required modules for this course are ice and epid which can be installed into R by typing the following two lines on your R console.

install.packages("ice", repos="http://r-ice-project.org")
install.packages("epid", repos="http://r-ice-project.org")

In this session, we are going to use multilevel logistic regression (glmer) in dealing with data stratified by age group as if the stratification is a matching condition. This method is recently used by some authors. Then we need another package from CRAN called "lme4". You can install the package by typing the following line on your R console.

install.packages("lme4")

(Since the package is installed from CRAN or its mirror, a repos argument is not required.)

Finally, you can follow my R script file below. Don't forget to change working directory to yours.

conditional2.R

Additional reading:
The article "Medication risk factors associated with healthcare-associated Clostridium difficile infection: a multilevel model case–control study among 64 US academic medical centres" shows how to use multilevel logistic regression so that the effect of the variables at the contexual level can be identified. This method can be used to handle the matched analysis of a case-control study as well.

J. Antimicrob. Chemother.-2014-Pakyz-1127-31.pdf

Exercise:
Try analyzing the following data. It is the data set of a matched case-control study. The matching variables are sex, age, and alcohol drinking and smoking habits. Explore the case:control ratio of the matched sets and determine how to analyze to see the effects of the following genes on cancer status: gstm1, gstt1, p53,  cyp2E1, mEH3, mEH4, and MPO. This is just an exercise, so don't worry about the action of these genes on cancer, alcohol drinking and cigarette smoking.

oralex.dta
lowbwtm11.dta 

Logistic regression I, 2016

Hi,

These are materials for logistic regression course I, academic year 2016 (2017). This year I use R version 3.3.2. The outputs of R are the same as with the previous versions of R.

For the exercises in the module, download the following files.
anc.dta
agechd.dta
cca.dta
lowbwt.dta

The revised module can be downloaded here.
logistic1602-1.pdf

The required modules for this course are ice and epid which can be installed into R by typing the following two lines on your R console.

install.packages("ice", repos = "http://r-ice-project.org")
install.packages("epid", repos = "http://r-ice-project.org")

Finally, you can try following my R script file. Please change the working directory to yours, otherwise the script will not run correctly.
exercises.R
exercise-additional.R