Case-Control Study Design.pptx

The second major type of observational epidemiology
The subject of interest is individual
The object is testing of hypothesis:
Two distinct type
Case-Control Study
Cohort Study
These studies determine the statistical association between RF & diseases and if yes the strength of the association
Also c/d “retrospective studies”
First approach to test causal hypothesis
Also c/d “retrospective studies”
First approach to test causal hypothesis
Use two group cases & control
Has three distinct features
Both exposure & outcome (disease) occurred before the start of the study
The study proceeds backward from effect to cause
It uses a control or comparison group to support or refute an inference

1. CASE-CONTROL STUDY Professor Dr. AB Rajar, MBBS, Dip-Diab, MPH, Ph.D. CPHE Director of Research and Innovative Center [IBN-E-SINA UNIVERSITY AB Rajar /[email protected] 1

2. Learning objectives Theme and Historical Perspective Definition. Classification. Analytical Study Design Case-Control Study. Biases of Case-Control Study Advantages & Disadvantages. MCQs AB Rajar /[email protected] 2

3. Theme AB Rajar /[email protected] 3

4. Example from History Thalidomide Tragedy: Thalidomide was first marketed as a safe, non-barbiturate hypnotic in Britain in 1958. In 1961 at a Gynecological congress it was discussed that a large number of babies with congenital abnormalities were being born (phocomelia) which was associated with thalidomide. Confirmed that thalidomide was Teratogenic. No % who took thalidomide Cases (with congenital defects) 46 41(89.13) Controls 300 AB Rajar /[email protected] 4

5. Example from History OCP and Thromboembolic Disease: By Aug 1965, BRITISH COMMITTEE ON SAFETY DRUGS received 249 reports of adverse reactions and 16 deaths in women taking OCPs. Thus there was a need to conduct an epidemiological study. A case-control study was conducted by Vassey and Doll in 1968 . Controls were matched for age, marital status, and parity. Relative Risk of the users to the non-users was 6.3:1 Confirmation was established. No % who used OCP Cases (venous thrombosis and pulmonary embolism) 84 50 (59.52) Controls 168 14 (8.33) AB Rajar /[email protected] 5

6. Example from History Again in the 1940s , Sir Norman Gregg , an Australian ophthalmologist, observed a number of infants and young children in his ophthalmology practice who presented with an unusual form of cataract . Gregg noted that these children had been in utero during the time of the rubella ( German measles ) outbreak. He suggested that there was an association between prenatal rubella exposure and the development of unusual cataracts . Keep in mind that was a time there was no knowledge that a virus could be teratogenic . Thus, he proposed his hypothesis solely on the bases of observational data, the equivalent of data from ambulatory or bedside practice today. AB Rajar /[email protected] 6

7. Example from History In the early 1939s, Alton Ochsner , a surgeon of New Orleans, observed that virtually all of the patients on whom he was operating for lung cancer gave a history of cigarette smoking. He hypothesized that cigarette smoking was linked to lung cancer. Although this relationship is accepted and well-recognized today, it was relatively new and controversial at the time that Ochsner made his observation. Based only on his observations in cases of lung cancer, was this conclusion valid? AB Rajar /[email protected] 7

8. AB Rajar /[email protected] 8

9. DEFINITION "The study of the distribution and determinants of health-related states or events in specified populations , and the application of this study to the prevention and control of health problems ". AB Rajar /[email protected] 9

10. CLASSIFICATION AB Rajar /[email protected] 10

11. TYPES OF EPIDEMIOLOGICAL STUDIES AB Rajar /[email protected] 11

12. ANALYTICAL STUDIES In analytical studies, the subject of interest is the individual within the population. The object is not to formulate but to test the hypothesis. To evaluate an association between exposure and disease. Analytical studies focuses on the magnitude of the association between the exposure and the health problem under the study. AB Rajar /[email protected] 12

13. CASE-CONTROL STUDY The observational epidemiologist study of persons with the disease (or other outcome variables) of interest and a suitable control (comparison/reference) group of persons without the disease. (Descriptive of Epidemiology: 4 th ed:John M Last.2001) AB Rajar /[email protected] 13

14. CASE-CONTROL STUDY [Retrospective Studies] A case-control study is an observational study in which subjects are sampled based on the presence or absence of disease and then their prior exposure status is determined. A case-control study involves two populations – cases and controls and has three distinct features : Both exposure and outcome occurred before the start of the study. The study proceeds backward from effect to cause . It uses a control or comparison group to support or refute an inference. AB Rajar /[email protected] 14

15. CASE-CONTROL STUDY The investigator selects Cases with the disease And Appropriate Controls without the disease And obtains Data regarding past exposure to possible etiologic factors in both groups. The investigator then compares the frequency of exposure of the two groups AB Rajar /[email protected] 15

16. Dogma of case-control study AB Rajar /[email protected] 16

17. DESIGN OF CASE-CONTROL STUDY CASES CONTROLS Hall Mark of case-control study : Begins with people with the disease (cases) and compares them to people without the disease (controls) and searches for exposure. AB Rajar /[email protected] 17

18. FRAMEWORK OF CASE-CONTROL STUDY FIRST SELECT Cases (With disease) Controls (Without disease) THEN Were exposed a b Measure Exposure Were not exposed c d TOTALS a +c b +d Proportions Exposed FIRST SELECT Cases (With disease) Controls (Without disease) THEN Were exposed a b Measure Exposure Were not exposed c d TOTALS a +c b +d Proportions Exposed AB Rajar /[email protected] 18

19. STEPS OF CASE-CONTROL STUDY. AB Rajar /[email protected] 19

20. AB Rajar /[email protected] 20

21. SELECTION OF CASES CASE : A person in the population or study group identified as having a particular disease, health disorder, or condition under investigation. (Descriptive of Epidemiology: 4 th ed:John M Last.2001) CONTROL: Person or persons in a comparison group that differs, in disease experience (or other health-related outcome) in not having the outcome being studied. (Descriptive of Epidemiology: 4 th ed:John M Last.2001) AB Rajar /[email protected] 21

22. SELECTION OF CASES Definition of the case : it involves two specifications- 1. Diagnostic Criteria: Enunciate clear-out diagnostic criteria for the disease of interest. Must be specified before the study was undertaken . Once established, it should not be altered or changed till the study is over. 2. Eligibility Criteria: It is always advisable to take the incident cases (new cases) since the prevalent cases (old cases in advanced stage) might have changed their exposure status due to medical advice etc. AB Rajar /[email protected] 22

23. SELECTION OF CASES 3-Hospitals : Convenient Can be chosen from one hospital or a network of hospitals. Admitted during a specified period of time Entire case series or random sample is selected 4-General Population: All cases of the study disease occurring within the same geographical area during a specified period of time. Through a survey, disease registry or hospital network Entire case series or random sample should be fairly representative of all cases in the community. AB Rajar /[email protected] 23

24. STEPS OF CASE-CONTROL STUDY. AB Rajar /[email protected] 24

25. AB Rajar /[email protected] 25

26. SELECTION OF CONTROLS Sources of controls Hospitals: control may be patient of another disease Chance of selection bias (one condition may influence another) Smoking & lung cancer Control are patients of MI Relatives Neighborhood Person living in the same locality Person working in the same factory Student of the same school, college, or University General population: in a population-based study, random sample, cases represent the community AB Rajar /[email protected] 26

27. SELECTION OF CONTROLS Control must be free from disease Similar to cases as much as possible except for the presence of disease or some other factor influencing the disease COMPARABLE: Should the controls be similar to the cases in all respects other than having the disease? REPRESENTATIVE : Should the controls be representative of all non-diseased people in the population from which the cases are selected? AB Rajar /[email protected] 27

28. SELECTION OF CONTROLS Compatibility vs Representativeness The control group should be representative of the general population in terms of the probability of exposure to the risk factor. AND they should also have had the same opportunity to be exposed as the cases have. Not that both cases and controls are equally exposed but only that they have had the same opportunity for exposure. AB Rajar /[email protected] 28

29. SELECTION OF CONTROLS SOURCES Source Advantage Disadvantage Hospital Based Easily identified Available for interview More willing to co-operate Tend to give complete and accurate information ( Recall bias) Not typical of general population Possess more risk factors for disease. Some diseases may share risk factors with the disease under study. (whom to exclude) Population-Based (Registry cases) Most representative of the general population. Generally healthy Time, money, and energy. Opportunity of exposure may not be same as that of cases. Neighborhood controls/Telephone exchange random dialing. Controls and cases similar residence. Easier than sampling the population Non co-operation. Security issues Not representative of general population Best friend control/Sibling control Accessible, Cooperative, Similar to cases in most aspects. Over-matching Source Advantage Disadvantage Hospital Based Not typical of general population Possess more risk factors for disease. Some diseases may share risk factors with the disease under study. (whom to exclude) Population-Based (Registry cases) Most representative of the general population. Generally healthy Time, money, and energy. Opportunity of exposure may not be same as that of cases. Neighborhood controls/Telephone exchange random dialing. Controls and cases similar residence. Easier than sampling the population Non co-operation. Security issues Not representative of general population Best friend control/Sibling control Accessible, Cooperative, Similar to cases in most aspects. Over-matching AB Rajar /[email protected] 29

30. SELECTION OF CONTROLS Large study: Cases: Control:: 1:1 Small study: Cases: Control:: 1:2, 1:3, 1:4. Use of multiple controls. Multiple control of different types: Control-1 hospital, 1 neighborhood e.g. Children with a brain tumor, control- children with other cancer, normal children, risk factor-h/o radiation exposure. AB Rajar /[email protected] 30

31. Example of multiple controls. AB Rajar /[email protected] 31

32. STEPS OF CASE-CONTROL STUDY. AB Rajar /[email protected] 32

33. AB Rajar /[email protected] 33

34. Matching AB Rajar /[email protected] 34

35. Matching The process by which we select controls in such a way that they are similar to cases with regard to certain pertinent selected variables . Like age, sex, occupation, social status , etc are all known to influence the outcome of the disease and if not adequately matched for comparability can distort or confound the results. Gordis Epidemiology; 6 th Ed AB Rajar /[email protected] 35

36. Matching Matching Types : Matching may be of two types Group matching Individual matching Group matching/Frequency matching : Consists of selecting the controls in such a manner that the proportion of controls with a certain characteristic is identical to the proportion of cases with the same characteristic. If 25% of the cases are married, the controls will be selected so that 25% of that group is also married. Individual matching/matched pairs : In this approach, for each case selected for the study, a control is selected who is similar to the case in terms of the specific variable or variables of concern. - 50 year old man with a disease compared with a 50-year-old man without a disease -Pair of patients & control of the same age , sex, duration and severity of illness etc AB Rajar /[email protected] 36

37. Matching A way to account for possible effects of confounding AB Rajar /[email protected] 37

38. Age wise comparison AB Rajar /[email protected] 38

39. Matching Problems Individual Matching on too many variables – is time consuming, costly and may be lead to too less control. Cannot explore possible association of disease with any variable on which cases and controls have been matched. Therefore only factors which are known to be associated with the disease are studied. Suppose we know that breast caner rates are higher among single women than in married women; then matching cases for marital status would spuriously NOT detect any relation regarding this factor. AB Rajar /[email protected] 39

40. AB Rajar /[email protected] 40

41. STEPS OF CASE-CONTROL STUDY. AB Rajar /[email protected] 41

42. AB Rajar /[email protected] 42

43. Information about the exposure should be obtained in precisely the same manner for both cases and controls. This may be obtained by: Interview Questionnaire Past records of cases Hospital records Employment records . Measurement of exposure AB Rajar /[email protected] 43

44. On analysis of the case-control study we find out Exposure Rates : The frequency of exposure to suspected risk factors in cases and controls. Estimation of Risk : Relative Risk or Risk Ratio Odds Ratio : Strength of association between risk and outcome. Measurement of exposure AB Rajar /[email protected] 44

45. Exposure Rate: A case-control study provides a direct estimation of the exposure rates (frequency of exposure) to the suspected factor in disease and non-disease groups. Cases= = 33/35=94.2% Controls= = 55/82=67.0%   Cases (Lung cancer) Controls (without lung cancer) Smokers 33 (a) 55 (b) Non-Smokers 2 (c) 27 (d) Total 35 ( a+c ) 82 ( b+d ) Measurement of exposure Outcome of case-control study AB Rajar /[email protected] 45

46. Relative Risk or Risk Ratio (RR): RR= RR=   A typical case Control study does not provide incidence rates from which RR can be calculated directly. There is no appropriate population or denominator at risk. In general RR can be exactly calculated from a cohort study. Measurement of exposure Outcome of case-control study AB Rajar /[email protected] 46

47. Odds Ratio: Odds: Odds of an event are defined as the ratio of the number of ways an event occurs to the number of ways an event cannot occur. If the probability of event X occurring is P, then the odds of it occurring is =P/P-1. Odds ratio : Ratio of the odds that the cases were exposed to the odds that the controls were exposed. Measurement of exposure Outcome of case-control study AB Rajar /[email protected] 47

48. Measurement of exposure Outcome of case-control study Odds ratio: Odds ratio= Odds ratio= =   Cases (Diseased) Controls (Not Diseased) Exposed a b Not exposed c d AB Rajar /[email protected] 48

49. Odds ratio (=cross product ratio) can also be viewed as the ratio of: the product of two cells that support the hypothesis of an association (cells a & d -diseased people who were exposed and non-exposed people who were not exposed) to the product of the two cells which negates the hypothesis of an association (cells b & c -non diseased people who were exposed and diseased people who were not exposed). Measurement of exposure Outcome of case-control study AB Rajar /[email protected] 49

50. Odds Ratio: Odds ratio= Odds ratio= =   Cases (Lung cancer) Controls (without lung cancer) Smokers 33 (a) 55 (b) Non-Smokers 2 (c) 27 (d) Total 35 ( a+c ) 82 ( b+d ) Measurement of exposure Outcome of case-control study AB Rajar /[email protected] 50

51. STEPS OF CASE-CONTROL STUDY. AB Rajar /[email protected] 51

52. AB Rajar /[email protected] 52

53. Odds ratio= = = = 8.1 INTERPRETATION : The odds of smoking more than5 cigarettes per day was 8.1 time more in the lung cancer patient than without lung cancer. OR Smoking (<5/day) was found be associated 8.1 times more in patients with lung cancer than those without lung cancer.   Analysis & Interpretation AB Rajar /[email protected] 53

54. Interpretation of odds ratio OR=1: Exposure is not related to the disease. OR>1: Exposure is a related disease OR<1 :Exposure is protective against disease. AB Rajar /[email protected] 54

55. Bias in Case-Control Study Definition : Any systemic error in the design, conduct, or analysis of a study that results in mistaken estimates of the effect of the exposure on disease. Types of bias in case control studies: Selection bias Information Confounding bias AB Rajar /[email protected] 55

56. Bias in Case-Control Study SELECTION BIAS: Sources- Selective loss to follow-up. Incomplete ascertainment of cases (Detection or Diagnostic bias). Inappropriate control group Differential motivation to participate AB Rajar /[email protected] 56

57. Bias in Case-Control Study Selection Bias: Selective Survival : only surviving subject available to be studied; those surviving from those dying in potentially important ways. AB Rajar /[email protected] 57

58. Bias in Case-Control Study INFORMATION BIAS: Occurs due to- Imperfect definition of study variables OR Flawed data collection procedures. Leads to- Misclassification of disease and exposure Types of information bias- Recall bias Interviewer bias AB Rajar /[email protected] 58

59. Bias in Case-Control Study Recall Bias: [usually in case control studies]. Cases who are aware of their disease status may be more likely to recall exposure than controls. E.g , congenital malformation with prenatal infections Results in misclassification. AB Rajar /[email protected] 59

60. Bias in Case-Control Study Interviewer bias: When the interviewer is not blinded (knows) the case status of subjects there is potential for interviewer bias. AB Rajar /[email protected] 60

61. Bias in Case-Control Study Confounding : When a measure of the effect of an exposure on risk is distorted because of the association of exposure AB Rajar /[email protected] 61

62. Advantages Only realistic study design for uncovering etiology in rare diseases. Efficient for the study of chronic diseases. Require comparatively few subjects. Commonly used in outbreaks investigation. Tend to require a smaller sample size than other designs. Relatively inexpensive AB Rajar /[email protected] 62

63. Limitation/ Disadvantages Risk of disease cannot be estimated directly Not efficient for the study od rare exposure More susceptible to selection bias than alternative designs Information on exposure may not be less accurate than that available alternative designs AB Rajar /[email protected] 63

64. MCQs Q-Which one of the following studies has the highest risk of bias? A- Case report/series B- Cross-sectional study C- Case-control study D- Cohort study E- RCT AB Rajar /[email protected] 64

65. MCQs Q-What is the main reason why it is important to use precise, specific criteria for what constitutes a "case," i.e. in defining the outcome? (Select the best answer.) A. To avoid misclassification with respect to the outcome. B. To limit the number of subjects in the study. C. To avoid selection bias. D. To avoid interviewer bias AB Rajar /[email protected] 65

66. MCQs Q-Which of the following are legitimate source of cases for a case-control study? (Select all that apply). A. Disease registries, e.g. cancer registries. B. Hospitals. C. Members of the general population responding to an advertisement seeking subject with a particular condition. D. Patients at an outpatient clinic AB Rajar /[email protected] 66

67. MCQs Q-Which of the following are advantages to case-control studies? (Select all that apply). They tend to be less expensive and more efficient than prospective cohort studies. They are feasible for rare diseases. They allow you to study multiple outcomes of a single risk factor. They are good for diseases that have a long latency period (i.e., a long time between exposure and manifestation of disease.) AB Rajar /[email protected] 67

68. MCQs Q-Suppose a study looking at the association between smoking and bladder cancer found an odds ratio = 2.4. What would be the best way to interpret this? (Select one answer.) Smokers are 140% times as likely to develop bladder cancer compared to non-smokers. People with bladder cancer are about 40% more likely to e smokers compared to people who don't have bladder cancer. Smokers have a 2.4 % more risk of bladder cancer compared to nonsmokers. Smokers have 2.4 times the risk of d eveloping bladder cancer compared to non-smokers AB Rajar /[email protected] 68

69. MCQs Q-All of the following are true of odds ratio except: It is an estimate of the relative risk It is the only measure of risk that can be obtained directly form a case-control study It tends to be biased towards 1 (neither risk or protection at high rates of disease It is the ratio of incidence in exposed divided by incidence in nonexposed It can be calculated without data on rates (as in a case-control study AB Rajar /[email protected] 69

70. Thank You AB Rajar /[email protected] 70