Research Designs-II PDF

Research Designs-II STEPS IN DESCRIPTIVE STUDIES 1. Defining the population 2. Defining the disease under study 3. Describing the disease by Time - Place – Person 4. Measurement of disease 5. Comparing with known indices 6. Formulation of aetiological hypothesis...

Research Designs-II STEPS IN DESCRIPTIVE STUDIES 1. Defining the population 2. Defining the disease under study 3. Describing the disease by Time - Place – Person 4. Measurement of disease 5. Comparing with known indices 6. Formulation of aetiological hypothesis Dr.Fawaz P'Shery BMC 1.Defining the population Define the population in relation to: 1. Number 2. Age 3. Gender 4. Occupation 5. Cultural and other characteristics Dr.Fawaz P'Shery BMC  The defined population can be: 1. Whole population 2. Sample 3. Specially selected groups Defined population should be: Large enough Stable (no migration) Clear on who belongs to the population Community participation This step is very important as it forms the basis for all calculations Dr.Fawaz P'Shery BMC 2. Defining the disease under study Need for an operational definition. Researcher or investigator needs a definition that is precise and valid to obtain an accurate estimation about the disease. An operational definition is more specific and provides a detailed description of how a concept or variable will be measured or observed in a particular study or research context. Dr.Fawaz P'Shery BMC For example: DEFINITION OPERATIONAL DEFINITION 1. Gingivitis “Inflammation of the 1. Gingival bleeding in one or more gingiva” sites after gently probing the gingival sulcus 2. Dental caries -Infectious microbial 2.Dental caries The lesion is disease affecting the calcified tissues clinically visible and obvious of teeth Explorer tip can penetrate deep into soft yielding material. There is discoloration or loss of translucency. Explorer tip resists removal after moderate to firm pressure Dr.Fawaz P'Shery BMC 3. Describing the disease Dr.Fawaz P'Shery BMC  Time distribution Week, month, year Seasonal in occurrence Periodic decrease or increase / consistent trend ◉ Three types of time trends: 1. Short term fluctuations 2. Periodic fluctuations 3. Long term/ secular trends Dr.Fawaz P'Shery BMC Short-term fluctuations Commonly know as “Epidemic” Types I. Common source epidemics 1.Single source or point source epidemic (Food Poisoning) 2.Continuous or multiple exposure epidemic (contaminated water) II. Propagated epidemics 1.Person-to-person (hepatitis A, poliomyelitis) 2.Arthropod vector 3.Animal reservoir III. Slow or modern epidemics (HIV/AIDS) Dr.Fawaz P'Shery BMC Periodic fluctuations 1) Seasonal trends Related to environmental conditions - Measles and Varicella – seen in spring - Upper Respiratory Tract infection- winter 2) Cyclic trends Some conditions occur in cycles spread over short period of time (days, weeks, months , years etc) -traffic accidents – common in weekends Dr.Fawaz P'Shery BMC Long-term or secular trends Changes in occurrence of a disease over long periods of time (years or decades) -EX: Cancer show an upward trend for the last 30-40 years, Whereas diseases like tuberculosis , polio, diphtheria show a downward trend Interpretation of time trends: 1. Disease increasing/decreasing 2. Effectiveness of measures 3.Formulates etiological hypothesis based on characteristics observed Dr.Fawaz P'Shery BMC b) Place distribution Geographical comparisons Differences in disease patterns b/w countries and within countries Importance of genes versus environment; changes with migration and role of diet Cultures and standard of living and external environment vary in different parts of world Helps distinguish role of genetic and environmental factors Range of studies vary with: International variations National variations Rural-urban differences Local distribution Dr.Fawaz P'Shery BMC c) Person distribution Age If equal susceptibility in all ages – no previous immunity Progressive increase with age – persistent exposure to causative agent Bimodality: two separate peaks Two separate sets of causative factors Gender: Biological difference between two sexes Cultural difference between two gender Dr.Fawaz P'Shery BMC ETHINICITY: difference in disease occurrence in different population subgroups OCCUPATION: occupational disorders SOCIAL CLASS: Higher social class – diabetes, coronary heart diseases Lower social class – nutritional disorders BEHAVIOR: cigarette smoking, sedentary life, drug abuse, over-eating STRESS: affects response to variables Susceptibility to diseases Exacerbation of symptoms Compliance with medical regimen MIGRATION: eg. Malaria /ebola from rural to urban areas Dr.Fawaz P'Shery BMC 4. Measurement of disease Knowing “ DISEASE LOAD ” is important Measured based on: Mortality (death) Morbidity (disease) Disability (inability to perform some activity) Mortality – straightforward Morbidity – incidence and prevalence “Incidence” – longitudinal studies ‘Prevalence’ – cross sectional studies Dr.Fawaz P'Shery BMC Incidence Definition: Incidence is defined as the occurrence of new cases of disease that develop in a population over aspecified time period. Dr.Fawaz P'Shery BMC Prevalence Definition: all current cases (old and new) existing at a given point in time, or over a period of time in a given population. Two types: Point prevalence: proportion of population that is diseased at a single point in time. It is a single snapshot of the population Period prevalence: proportion of population that is diseased during a specific period of time Dr.Fawaz P'Shery BMC Dr.Fawaz P'Shery BMC 5. Comparing with known indices Make comparisons Ask questions Helps in: Aetiology Identify groups at increased risk Dr.Fawaz P'Shery BMC 6.Formulation of etiological hypothesis It is a supposition, arrived at from observation or reflection It should specify following: Population Specific cause being considered Expected outcome – the disease Dose-response relationship Time-response relationship Dr.Fawaz P'Shery BMC Uses of descriptive studies 1. Contributes to research by describing variations in disease occurrence by time, place and person 2. Clues to disease epidemiology – aetiological hypothesis 3. Data regarding magnitude of disease load and types of disease problems in community in terms of morbidity and mortality 4. Background data for planning, organizing and evaluating preventive andcurative services Dr.Fawaz P'Shery BMC Analytical study designs Case control study Cohort study Case Control Studies ʻRetrospective studyʼ 3 distinct features Both exposure and outcome [disease] have occurred before the start of study. Proceeds from Effect to cause. Uses control or comparison group - support or refute an inference  Exposed (with risk factor) Cases (those with condition) Unexposed (without risk factor) Exposed (with risk factor) Controls (without the condition) Unexposed (without risk factor)  Tobacco chewers Cases of oral cancer Non Tobacco chewers Tobacco chewers Those free from oral cancer Non Tobacco chewers  Steps in case – control study Selection of cases and controls. Matching Measurement of exposure Analysis and interpretation.  1. Selection of Cases and Controls Cases Selection Sources Diagnostic Hospitals Eligibility General Population Relatives Neighbourhood  b. Selection of Controls - Absence of disease under study. - Similar to cases as far as possible except for the absence of disease.  Sources Hospital General Population Neighbourhood Relatives  II. Matching Selection of control in such a way that they are similar to cases with regards to certain pertinent selected variables [e.g..age] which are known to influence the outcome of the disease.  “Confounding Factor” - Associated both with exposure and disease and is distributed unequally in study and control group. Eg: Age – diabetes and periodontal status Suspected etiological factor - should not be matched.  III. Measurement of Exposure Information - interviews - questionnaires - studying past hospital records, employment records etc. Two characteristic of information is important – its comparability and validity.  IV. Analysis Exposure rate Estimation of disease risk associated with exposure.  Exposure rates Direct estimation of the exposure rates to a suspected factor in diseased and non – diseased groups.  Cases Controls (Oral Cancer) (without oral cancer) Tobacco 80 (a) 40 (b) Chewers Non- Tobacco Chewers 20 (c) 60 (d) ER in Cases = a /(a+c) = 80/100 = 80 % ER in Controls = b / (b+d) = 40/ 100 = 40 % Odd Ratio (Cross Product Ratio) Measure of strength of association between risk factor and outcome. The disease under investigation must be relatively rare The cases must be representative of those with disease. The controls must be representative of those without disease.  Cases Controls (without (Oral Cancer) oral cancer) Tobacco Chewers 80 (a) 40 (b) Non- Tobacco Chewers 20 (c) 60 (d) ad 80 x 60 Odds ratio = = =6 bc 40 x 20  b. Estimation of Risk Incidence among exposed Relative risk = Incidence among non-exposed A typical case control study cannot provide the incidence rates as no appropriate denominator or population at risk is known.  Bias in Case-Control Study “ Bias is any systematic error in the determination of association between exposure and disease.” Confounding Bias Memory / Recall Bias Selection Bias Berkesonian Bias Interviewerʼs Bias  Advantages Relatively easy to carry out. Rapid and less expensive than cohort study. Relatively few subjects required. Allow study of several aetiological factors. Risk factors can be identified. No attrition problem - no follow up required. Ethical problem minimal.  Disadvantages Problem of bias. Selection of proper control may be difficult. Cannot measure incidence, but only relative risk. Do not distinguish between cause and associated factors. Not suited for evaluation of therapy or prophylaxis of disease. Representativeness of cases and control is important. Dr.Fawaz P'Shery BMC Thank You