Chapter 4 SURVIVAL AND LIFE TABLES

Slides:



Advertisements
Similar presentations
SURVIVAL AND LIFE TABLES
Advertisements

CHAPTER 16 Life Tables.
Math Qualification from Cambridge University
Main Points to be Covered
Lecture 3 Survival analysis. Problem Do patients survive longer after treatment A than after treatment B? Possible solutions: –ANOVA on mean survival.
Measure of disease frequency
Samuel Clark Department of Sociology, University of Washington Institute of Behavioral Science, University of Colorado at Boulder Agincourt Health and.
Sampling and Sampling Distributions Simple Random Sampling Point Estimation Sampling Distribution.
Introductory Mathematics & Statistics
Measures of disease frequency (I). MEASURES OF DISEASE FREQUENCY Absolute measures of disease frequency: –Incidence –Prevalence –Odds Measures of association:
Manish Chaudhary MPH (BPKISH)
Analysis of Complex Survey Data
Psy B07 Chapter 1Slide 1 ANALYSIS OF VARIANCE. Psy B07 Chapter 1Slide 2 t-test refresher  In chapter 7 we talked about analyses that could be conducted.
Positional Number Systems
1 1 Slide © 2009 Thomson South-Western. All Rights Reserved Slides by JOHN LOUCKS St. Edward’s University.
1 1 Slide © 2008 Thomson South-Western. All Rights Reserved Slides by JOHN LOUCKS St. Edward’s University.
1 POPULATION PROJECTIONS Session 2 - Background & first steps Ben Jarabi Population Studies & Research Institute University of Nairobi.
1 1 Slide IS 310 – Business Statistics IS 310 Business Statistics CSU Long Beach.
1 1 Slide © 2008 Thomson South-Western. All Rights Reserved Slides by JOHN LOUCKS St. Edward’s University.
1 Chapter 7 Sampling and Sampling Distributions Simple Random Sampling Point Estimation Introduction to Sampling Distributions Sampling Distribution of.
Chapter 7 LIFE TABLES AND POPULATION PROBLEMS
Introduction to Survival Analysis Utah State University January 28, 2008 Bill Welbourn.
Relative Values. Statistical Terms n Mean:  the average of the data  sensitive to outlying data n Median:  the middle of the data  not sensitive to.
01/20151 EPI 5344: Survival Analysis in Epidemiology Actuarial and Kaplan-Meier methods February 24, 2015 Dr. N. Birkett, School of Epidemiology, Public.
Biostatistics Case Studies 2007 Peter D. Christenson Biostatistician Session 2: Aging and Survival.
Life expectancy Stuart Harris Public Health Intelligence Analyst Course – Day 3.
SURVIVAL ANALYSIS PRESENTED BY: DR SANJAYA KUMAR SAHOO PGT,AIIH&PH,KOLKATA.
Methods and Statistical analysis. A brief presentation. Markos Kashiouris, M.D.
Health Indicators.
Mortality: The Life Table Its Construction and Applications
Instructional Objectives:
Chapter 4 Systems of Linear Equations; Matrices
Virtual University of Pakistan
Significant Figures.
April 18 Intro to survival analysis Le 11.1 – 11.2
Chapter 4 Systems of Linear Equations; Matrices
St. Edward’s University
Relative Values.
Z-scores & Shifting Data
Virtual University of Pakistan
Distribution of the Sample Means
Measures of Association
Measures of Disease Frequency
LIFE TABLES The life table is one of the most important devices used in demography. In its classical form, it is a table that displays various pieces of.
SQL for Predicting from Likelihood Ratios
Survival Analysis {Chapter 12}
SQL for Calculating Likelihood Ratios
Gathering and Organizing Data
Overview probability distributions
Measures of Disease Frequency
Summary measures of mortality – Life expectancy and life tables
II. Survivorship.
The Measures of Mortality
Fractions.
Measures of Disease Occurrence
Virtual University of Pakistan
Population Ecology pp ; 972
Sampling Design Basic concept
Demographic Analysis and Evaluation
The Normal Curve Section 7.1 & 7.2.
Gathering and Organizing Data
Epidemiological Measurements of health
Mortality rate = No. of deaths * K
Module appendix - Attributable risk
CHAPTER 3 FERTILITY MEASURES .
Chapter 7 Sampling and Sampling Distributions
Sampling Method.
Where are we?.
Risk Adjusted P-chart Farrokh Alemi, Ph.D.
Approaches and Challenges in Accounting for Baseline and Post-Baseline Characteristics when Comparing Two Treatments in an Observational/Non-Randomized.
Presentation transcript:

Chapter 4 SURVIVAL AND LIFE TABLES Dr. A. PHILIP AROKIADOSS Assistant Professor Department of Statistics St. Joseph’s College (Autonomous) Tiruchirappalli-620 002.

THE FIRST FOUR COLUMNS OF THE LIFE TABLE ARE: 1. AGE (x) 2. AGE-SPECIFIC MORTALITY RATE (qx) 3. NUMBER ALIVE AT BEGINNING OF YEAR (lx) 4. NUMBER DYING IN THE YEAR (dx)

PROCEDURE: We use column 2 multiplied by column 3 to obtain column 4. Then column 4 is subtracted from column 3 to obtain the next row’s entry in column 3.

EXAMPLE: 100,000 births ( row 1, column 3) have an infant mortality rate of 46.99/thousand (row 2, column 2), so there are 4,699 infant deaths (row 3, column 4). This leaves 95,301 left (100,000 – 4,699) to begin the second year of life (row 2 column 3).

If we stopped with the first four columns, we could still find out the probability of surviving to any given age. e.g. in this table, we see that 90.27% of non-white males survived to age 30.

THE NEXT THREE COLUMNS OF THE LIFE TABLE ARE: THE NUMBER OF YEARS LIVED BY THE POPULATION IN YEAR X (Lx) THE NUMBER OF YEARS LIVED BY THE POPULATION IN YEAR X AND IN ALL SUBSEQUENT YEARS (Tx) THE LIFE EXPECTANCY FROM THE BEGINNING OF YEAR X (ex)

WE CALCULATE COLUMN 5 FROM COLUMNS 3 AND 4 IN THE FOLLOWING WAY: The total number of years lived in each year is listed in column 5, Lx. It is based on two sources. One source is persons who survived the year, who are listed in column 3 of the row below. They each contributed one year. Each person who died during the year (column 4 of the same row) contributed a part of year, depending on when they died. For most purposes, we simply assume they contributed ½ a year.

The entry for column 5, Lx in this table for age 8-9 is 94,321 The entry for column 5, Lx in this table for age 8-9 is 94,321. Where does this number come from? 94,291 children survived to age 9 (column 3 of age 9-10), contributing 94,921 years. 60 children died (column 4 of age 8-9) , so they contributed ½ year each, or 30 years. 94,921 + 30 = 94,321.

EXCEPTION TO THE ½ YEAR ESTIMATION RULE Because deaths in year 1 are not evenly distributed during the year (they are closer to birth), infants deaths contribute less than ½ a year. Can you figure out what fraction of a year are contributed by infant deaths (0-1) in this table?

Lx = 96,254 95,301 contributed one year 96,254 - 95,301 = 953 years, which must come from infants who died 0-1 4,699 infants died 0-1 953/4,699 = .202 or 1/5 of a year, or about 2.4 months

HOW DO WE GET COLUMN 6, Tx The top line of Column 6, or Tx=0 , is obtained by summing up all of the rows in column 5. It is the total number of years of life lived by all members of the cohort. This number is the key calculation in life expectancy, because, if we divide it by the number of people in the cohort, we get the average life expectancy at birth, ex=0, which is column 7.

COLUMN 7, LIFE EXPECTANCY, or ex=0 For any year, column 6, Tx, provides the number of years yet to be lived by the entire cohort, and column 7, the number of years lived on average by any individual in the cohort. (Tx/lx) Thus column 7 is the final product of the life table, life expectancy at birth, or life expectancy at any other specified age.

WHAT IS LIFE EXPECTANCY? Life expectancy at birth in the US now is 77.3 years. This means that a baby born now will live 77.3 years if………….. that baby experiences the same age-specific mortality rates as are currently operating in the US.

Life expectancy is a shorthand way of describing the current age-specific mortality rates.

SOME OTHER MEASURES OF SURVIVAL AND THE PROBLEM OF CENSORED DATA

5-year survival. Number of people still alive five years after diagnosis.  Median survival. Duration of time until 50% of the population dies. Relative survival. 5-year survival in the group of interest/5-year survival in all people of the same age. Observed Survival. A life table approach to dealing with censored data from successive cohorts of people. Censoring means that information on some aspect of time or duration of events of interest is missing.

THREE KINDS OF CENSORING COMMONLY ENCOUNTERED Right censoring Left censoring Interval censoring Censoring means that some important information required to make a calculation is not available to us. i.e. censored.

RIGHT CENSORING Right censoring is the most common concern. It means that we are not certain what happened to people after some point in time. This happens when some people cannot be followed the entire time because they died or were lost to follow-up.

LEFT CENSORING Left censoring is when we are not certain what happened to people before some point in time. Commonest example is when people already have the disease of interest when the study starts.

INTERVAL CENSORING Interval censoring is when we know that something happened in an interval (i.e. not before time x and not after time y), but do not know exactly when in the interval it happened. For example, we know that the patient was well at time x and was diagnosed with disease at time y, so when did the disease actually begin? All we know is the interval.

DEALING WITH RIGHT-CENSORED DATA Since right censoring is the commonest problem, lets try to find out what 5-year survival is now for people receiving a certain treatment for a disease.

OBSERVED SURVIVAL IN 375 TREATED PATIENTS Number Number alive in Treated 1999 00 01 02 03   1999 84 44 21 13 10 8 2000 62 31 14 10 6 2001 93 50 20 13 2002 60 29 16 2003 76 43 Total 375

WHAT IS THE PROBLEM IN THESE DATA? We have 5 years of survival data only from the first cohort, those treated in 1999. For each successive year, our data is more right-censored. By 2003, we have only one year of follow-up available.

What is survival in the first year after treatment? It is: (44 + 31 + 50 + 29 + 43 = 197)/375 = 52% Number Number alive in Treated 99 00 01 02 03   1999 84 44 21 13 10 8 2000 62 31 14 10 6 2001 93 50 20 13 2002 60 29 16 2003 76 43 Total 375

What is survival in year two, if the patient survived year one? (21 + 14 + 20 + 16 = 71)/154 = 46% Note that 154 is also 197 (last slide’s numerator) – 43, the number for whom we have only one year of data Number Number alive in Treated 96 97 98 99 00   1995 84 44 21 13 10 8 1996 62 31 14 10 6 1997 93 50 20 13 1998 60 29 16 1999 76 43 Total 375

By the same logic, survival in the third year (for those who survived two years) is: (13 + 10 + 13 = 36)/(71 - 16 = 55) = 65%  Number Number alive in Treated 99 00 01 02 03   1999 84 44 21 13 10 8 2000 62 31 14 10 6 2001 93 50 20 13 2002 60 29 16 2003 76 43 Total 375

In year 4, survival is(10 + 6)/(36-13) = 70% Number Number alive in Treated 99 00 01 02 03   1999 84 44 21 13 10 8 2000 62 31 14 10 6 2001 93 50 20 13 2002 60 29 16 2003 76 43 Total 375

The total OBSERVED SURVIVAL over the five years of the study is the product of survival at each year:   .54 x .46 x .65 x .70 x .80 = .08 or 8.8%

Subsets of survival can also be calculated, as for example:   2 year survival = .54 x .46 = .239 or 23.9%

Five-year survival is averaged over the life of the study, and improved treatment may produce differences in survival during the life of the project. The observed survival is an average over the entire period.

Changes over time can be looked at within the data Changes over time can be looked at within the data. For example, note survival to one year, by year of enrollment:   1999 - 52.3% 2000 - 50.0% 2001 - 53.7% 2002 - 48.3% 2003 - 56.6% Little difference is apparent.

These data also do not include any losses to follow-up, which would make our observed survival estimates less precise. The calculation is only valid if those lost to follow-up are similar in survival rate to those observed.