Research process & study designs

1. Research process & study designs A.Prof. Dr. Maha Al_Nuaimi

2. Learning objectives • Describe the components of research process. • Review basic epidemiological study design aspects. • Understanding of advantages and drawbacks of these design aspect.

4. A research design: is the set of methods and procedures used in collecting and analyzing measures of the variables and steps conducted and implemented to reach specific inference. • The design of a study defines the study type (descriptive, correlation, semi-experimental, experimental, review, meta-analytic)

5. Most problems in studies are due to poor design (not poor analysis) Results from a single study are seldom definitive (or even clear)

6. We select the study design according to: • Nature of the research question • Objective(s) of the study • Our resources

7. The main types are divided as:

9. Descriptive studies • Hypothesis formation studies about the disease and risky groups. • DESCRIBE the occurrence of disease in terms of: person-place-time, or agent-host-environment, to determine type of exposure and susceptible group. TYPES • Case Report • Case Series • Ecological studies • Cross-sectional ?? • Biometry • Hospital-records Study • Meta-analysis

10. Case report • Describes unusual clinical observations, interesting or novel variation of a disease.Advantages: - simple, quick &easy.- Calls attention to a novel observation.Disadvantages: cannot be generalized (single case represent itself).

11. Case Series: Refers to a group of similar cases, to recognize well a clinical pattern and characteristics of specific disease.Advantages:-simple, quick &easy.- Better recognition of any disease, case definition of new one. Disadvantages:- non-generalizable inferences (small sample size).- Estimation of risk cannot be done (no comparison).

12. Ecological studies • e.g. there were cigarette consumption rate of 1500 per capita /year in a city in 1945 (lung cancer rate 10/100,000) which raised to 3500 in 1965 (LCR: 40/100,000). - This type of studies depends on the correlation between factor and disease occurrence, either (+) ve or (-) ve or no correlation. Does this mean that every person in the city was a smoker?

13. Ecological studies • These are studies where exposure data relating to a place (hardness of water) are correlated with health data collected but summarised by place (CHD rates). Disadvantages: • Require additional caution in their interpretation. • Ecological fallacy. • Confounding (uncontrolled) Advantages: • (formulate hypothesis). • simple & quik

14. Cross-sectional study PURPOSES: • To examine the health problem frequency/nature. • To examine the exposure and health problem at the same time. • Who are affected ?, and how they behave ? • KAP of people about disease. • Attempts to assess relationship between exposure and effect (comparative cross-sectional studies).

15. Cross-sectional Study • Data collected at a single point in time • Observation of all of a pop, or a representative subset. • Describes associations A “Snapshot”

16. Cross-sectional study Advantages: • cheaper, easier and faster • Hypothesis formation study • measure burden of disease (prevalence) • identify high risk group. As a baseline for future cohort study • Start with reference population so generalization possible Disadvantages: • Can show association only but NOT CAUSATION – • No temporal sequence, Chicken-egg dilemma; • Survivors problems • Possible biases - selection, memory, recall.

17. Biometry • Hospital-records study • Meta analysis • (Descriptive studies are hypothesis formation studies .. not to test hypothesis)

18. Analytic studies Observational analytic studies: • Case control study • Cohort study • Exposure & disease can be the Dependent or Independent variable according to the study used.

19. Case-control study • Case control studyA type of observational analytic studies, where the subjects are selected on basis of “whether they do have the disease (cases) or they do not have the disease (control) under study”. • The 2 groups then compared for the proportion of having a previous exposure (s) or characteristic of interest.

20. Direction of inquiry • Design of a Case - Control Time risk factor present Start with: Cases risk factor absent Population risk factor present Control risk factor absent

21. Case-control study Advantages: (1) a relatively quick, cheap, small sample size, (2) Used initially to test an association (3) An excellent to study risk factor of rare diseases, long latency? (4) Odds ratio can be estimated Disadvantages: • Cannot measure incidence or RR. • Bias: selection, memory, and recall • Not sutable to study rare exposure.

22. Case-control study Selection of control group: • is a major determinant validity! • Similarity .. Matching .. as similar as the cases except the disease status (account for confounding). • from the same source pop. • The number of control should be (up to 1:4)

25. Cohort study • A study in which two or more groups of people that are free of disease and that differ according to the extent of exposure (exposed and unexposed) are compared with respect to disease incidence. • Synonyms : incidence study, longitudinal study,

26. CohortDesign disease Factor present no disease Study population free of disease disease Factor absent no disease present future time Study begins here

27. Purpose Of A Cohort Study To identify risk, protective factors, prognostic factors When exposure is rare. assessing effectiveness of preventive measures.

28. Type Of Cohort Study Prospective Cohort Study: • the exposure at the start, follow-up, to determine the incidence. Historical Cohort Study: • the exposure in the past (before onset of the study) and are followed through existing records in to the future. Exposure* Start of the studyoutcome Retrospective cohort: • the exposure and disease before the time of the study, (cannot know which precedes which). • It is especially important in diseases with long latency period.Exposure out come * start of the study

30. Cohort Study Advantages: • Test hypotheses of casual relationship & give directly RR. • Able to measure risk • Can directly measure incidence • Temporal sequence between exposure & disease can be more clearly established. Disadvantages • High cost, longer time & relatively more difficult. • Loss to follow-up ( attrition). • Not suitable for rare diseases.

31. A well designed cohort study is a reliable means of showing an association between risk factors and disease . Why ?

32. Experimental or Interventionalstudy A study in which a planned trial of exposure, whose effects are measured by comparing the outcome in the experimental group versus the outcome of another regimen in the control group. • manipulation of the study exposures? • Interference in the form of a new drug, surgical procedure, vaccine, test, or new method of diagnosis.It can give a degree of assurance about the validity, not provided with any observational design option.

33. Types Of Interventional Studies • Randomized Controlled Trail (Therapeutic trial) RCT:- Two different drugs or procedures.- Drug & placebo- Single drug (before & after Rx). • Field Trail;(Preventive trial)- Conducted on healthy peopleat risk of developing a disease - Data collection takes place "in the field" from the general population. - Example: salk vaccine to prevent poliomyelitis on > 1 million child were divided randomly into 2 groups salk vaccine group & placebo group • Community Trail (Preventive trial)- the treatment groups are communities rather than individuals. - Entire pop. (community trial) an in floride fortification of water.

34. Steps in conduct of RCT • The protocol • Selecting reference and experimental populations. • Measure baseline variables. • Randomization • Blinding the Intervention • Follow up • Assessment the outcomes.

35. Detailed design of a RCT Reference Population Experimental Population Non-Participants Participants Treatment Allocation Treatment Group Comparison Group Improved Improved Not-improved Not- improved Selection by defined criteria Invitation to participate randomization

36. EXPERIMENTAL STUDIES • Advantage: • High validity of research • May approve causality … • Unique problems:- Ethical- Feasibility - Cost

37. EXPERIMENTAL STUDIES • Other problems:- Chance: managed by increase sample size & randomization.- Bias(subjectivity): treated by masking & blinding (single, double & triple blinding)- Confounders: by matching of all factors (or as much as we can) except the factor under study

38. Weight Gain on Different Diets deermice.sav (Labs 2 & 3)Explanatory variable = diet group (1=standard, 2=junk, 3=health) Response variable = weight gain (grams) Data are experimental because the investigator assigned the explanatory variable

39. Cigarettes and Lung Cancer Mortality doll-ecol.sav (Chap 12 and 13 labs)Explanatory var = per capita cigarette consumption (cig1930)Response var = lung cancer mortality per 100,000 (mortalit) Data are observational with data on aggregate-level. This is an ecological study

40. How to select study design?