Analytical epidemiology

analytical epidemiology

1. ANALYTICAl EPIDEMIOLOGY 1 Presented by: re MAN dha KAL CODSH-NMC FIRST BATCH

2. Introduction Types of epidemiology Types of analytical epidemiology Case control study Cohort study Comparison between case control and cohort study 2 Content

3. John M. last: "the study of the distribution and determinants of health-related states in specified populations, and the application of this study to control health problems. 3 EPIDEMIOLOGY

4. 4 EPIDEMIOLOGY

5. 5 Analytical epidemiology Second major type of epidemiology. Focus on individual within population unlike descriptive epidemiology .. Objective not to formulate hypothesis but to test hypothesis . TYPES CASE CONTROL STUDY COHORT STUDY

6. Retrospective or trohoc study Distinct features: Both exposure and outcome have occurred before the start of disease Study proceed backward from effect to cause Uses a control or comparison group to support or refute an inference. 6 A. CASE CONTROL STUDY

7. 7 The basic study design Control (those without condition) eg : those free of oral cancer Cases (those with condition) eg : cases with oral cancer Unexposed (without characteristic or risk factor) Eg . Non chewers E xposed (with characteristic or risk factor) Eg . tobacoo chewers

8. Selection of cases and controls Matching Measurement of exposure , and Analysis and interpretation. 8 BASIC STEPS

9. 9 Selection of cases and controls

10. Define as:”process by which we select controls in such a way that they are similar to cases with regard to certain pertinent selected variables( eg . Age) which are known to influence the outcome of disease and which, if not adequately matched for comparability, could distort or confounded the result”. CONFOUNDING FACTOR 10 2. MATCHING EXPOSURE ( eg . Consumption of alcohol) DISEASE ( eg . Oesophageal cancer) CONFOUNDING FACTOR ( eg . smoking, age )

11. Definition and criteria about exposure are just as important as those used to define cases and controls. This may be obtained by : Interviews Questionnaries Studying past record of cases such as hospital records, employment records etc. Clinical or laboratory examination. Investigator should not know whether a subject is in case or control group. 11 3.MEASUREMENT OF EXPOSURE AND OTHER FACTORS

12. The final step is analysis, to find out: Exposure rates among cases and controls to suspected factors Estimation of disease risk associated with exposure (ODD RATIO) 12 4. ANALYSIS AND INTERPRETATION

13. 13 a) EXPOSURE RATES Total 35 (a +c) 82 (b +d) Example of case control study of tobacco chewing and oral cancer EXPOSURE RATES Cases =a/( a+c ) = 33/35 = 94.2% Controls =b/( b+d ) = 55/82 = 67.0%

14. 1. Relative risk = Incidence among exposed incidence among non exposed = a c ( a+b ) ( c+d ) 2. Odds ratio (OR) = (a/b) (c/d) It measure strength of the association between risk factor and outcome. , b) ESTIMATION OF RISK:

15. 1. Selection bias : special types: a) Prevalence incidence (selective survival) b) Admission rate ( Berkson / Berkesonian ) 2. Information bias: a) Memory or recall bias b) Telescopic bias c) Interviewer’s bias 3. Bias due to confounding 15 BIAS

16. 16 ADVANTAGES: Relatively easy to carry out. Rapid and inexpensive Require fewer subjects. Suitable for investigation of rare diseases. No risk of subject. Allows the study of several different etiological factors. Risk factor can be identify No attrition problem because do not require follow up. Minimal ethical problem. DISADVANTAGES: Problem of bias since it relies on past memory or past records. Difficulty in selection of appropriate control group. Can not measure incidence only RR. Doesn’t distinguish between cause and associated factors. Not suited for the evaluation of therapy or prophylaxis of disease. CASE CONTROL STUDY

17. Prospective ,longitudinal, incidence and forward-looking study Distinguishing features: The cohorts are identified prior to the appearance of the disease The study groups, so defined, are observed over a period of time to determine the frequency of disease among them Study proceeds from cause to effect. 17 B. COHORT STUDY

18. Cohort Design time Study begins here Study population free of disease Factor present Factor absent disease no disease disease no disease present future

19. When there is good association between exposure and disease. When exposure is rare, but the incidence of disease is high among exposed. When attrition of study population can be minimized. 19 INDICATION

20. 20 TYPES OF COHORT STUDIES PROSPECTIVE COHORT STUDY RETROSPECTIVE COHORT STUDY A COMBINATION OF PROSPECTIVE AND RETROSPECTIVE COHORT STUDY.

21. - Outcome has not yet occurred the time of investigation begins . 1.PROSPECTIVE COHORT STUDY Measure exposure and confounder variables Exposed Non-exposed Outcome Outcome Baseline time Study begins here

22. Outcomes have all occurred before the start of the investigation . 22 2. RETROSPECTIVE COHORT STUDY Measure exposure and confounder variables Exposed Non-exposed Outcome Outcome Baseline time Study begins here

23. Selecting of study subject Obtaining data on exposure Selection of comparison group Follow up analysis 23 ELEMENTS OF A COHORT STUDY

24. 24 Selecting of study subject General population When exposure or cause of death is fairly frequent in the population b) Special groups:- Select group – Professional group ( doctors,nurses ) ii. Exposure group- High risk situation ( eg.radiologist exposed to x-ray) 2. Obtaining data on exposure Information about exposure may be obtained directly from:- Cohort member- Review of records- Medical examination- Environmental surveys-

25. 25 3. Selection of comparison group Internal comparisons:- Comparison groups are in built ( eg . Smoking, bp etc.) within same cohort group. b. External comparisons:- Eg . Smoker and non smoker, radiologists with opthalmologists . c. With General population:- If none is available, mortality of exposed group with general population 3. Follow up Main problem Procedures to obtain data for assessing the outcome are: Periodic medical checkup Reviewing hospital records Routine surveillance of death records Mailed questionnaries , telephone calls, periodic home visits.

26. Data are analysed in term of: Incidence rates of outcome among exposed and non-exposed: 26 5. ANALYSIS

27. 27 RISK FACTOR (TOBACCO) DEVELOPED ORAL CANCER DID NOT DEVELOP TOTAL PRESENT (CHEWERS) 45 (a) 9955 (c) 10000 (a + c) ABSENT (NON CHEWERS) 5 (b) 9995 (d) 10000 (b + d) 5. ANALYSIS Data are analysed in term of: a. Incidence rates of outcome among exposed and non-exposed Incidence rates: 1. Among tobacco chewers: = 45/10000 =4.5 per 1000 2.Among non chewers = 5/10000 = 0.5 per 1000

28. b. Estimation of risk Relative risk (RR): = incidence of disease among exposed incidence of disease among non-exposed = 4.5/0.5 = 9 It implies 9 times higher risk of development of oral carcinoma in tobacco chewers compared to non chewers. 28

29. B. Attributable risk (AR) Or “risk difference”: Incidence of disease rate among exposed- incidence among non exposed Incidence rate among exposed = 4.5 – 0.5 4.5 = 88.9% Indicates to what extent the disease under study can be attributed to the exposure. 29

30. 1. Selection bias: Non consent bias Missing data bias 2. information bias: Error in classification of individual Diagnostic bias 3. Confounding bias : Due to confounding factors 4. Post hoc bias: 30 Bias in cohort studies

31. Incidence can be calculated Several possible outcomes related to exposure can be studied simultaneously. Provide a direct estimate of RR. Dose response ratios can be calculated. Since comparison groups are formed before disease develops, certain forms of bias can be minimized like misclassification of individual. 31 Advantages of cohort studies:

32. 1. Unsuitable for investigating uncommon disease. 2. Long time to complete study and obtain results. 3. Administrative problem – Extensive record keeping 4.Expensive 5. Alter people behavior Stop or decrease smoking Loss of interest migration 5. Ethical problem of varying important 32 Disadvantages of cohort studies

33. Start disease :effect cause First approach to test hypothesis. Involve fewer subject. Yield result quickly. Suitable for studying rare disease. Gives RR only. Relatively inexpensive. Does not give information about diseases other than that selected for the disease. 33 Comparison of case control and cohort Start people: cause effect. Reserved for testing precisely formulated hypothesis. Involve larger number of subjects. Results are delayed due to long follow up. Unsuitable for studying for rare diseases. Yield RR and AR. Expensive Can give information more than one disease. 1 2 3 4 5 6 7 8

34. 34