Research Design (2021) PDF

 8/12/2023 Introduction RESEARCH DESIGN It is vital to design a...

 8/12/2023 Introduction RESEARCH DESIGN It is vital to design any kind of research study before initiation of a research DR. ADYANI MD REDZUAN Poorly design study give misleading results FACULTY OF PHARMACY, UKM The study design is important as it will oversee how data are collected & analyzed Outline Learning outcomes Introduction Classification Observational studies Experimental studies Learning outcomes Classification Describe different types of study designs Observational study I.Case-control studies Describe the advantages & disadvantages of different II.Cross-sectional studies, surveys study designs III.Cohort studies Experimental studies I. Clinical trials 1  8/12/2023 Observational studies EXAMPLE There is a suspicion that zinc oxide, the white non-absorbent sunscreen traditionally worn by lifeguards is more effective at preventing sunburns that Specific characteristics are observed & measured lead to skin cancer than absorbent sunscreen lotions. A case-control study without any attempt to modify the subjects was conducted to investigate if exposure to zinc oxide is a more effective skin cancer prevention measure. The study involved comparing a group of former lifeguards that had developed cancer on their cheeks and noses (cases) to a group of lifeguards without this type of cancer (controls) and assess their prior ‘Nature’ determines who is exposed to the factor of exposure to zinc oxide or absorbent sunscreen lotions. interest and who is not This study would be retrospective in that the former lifeguards would be asked to recall which type of sunscreen they used on their face and approximately how often. This could be either a matched or unmatched study, but efforts would need to be made to ensure that the former lifeguards are of the same average age, and lifeguarded for a similar number of seasons and amount of time per season i. Case-control/Retrospective ii. Cross-sectional A study that compares patients who have a disease or Data are observed, measured & collected at one point of outcome of interest (cases) with patients who do not time have the disease or outcome (controls) Also labelled as survey, epidemiological or prevalence History/previous events of both cases & controls are studies analyzed by looking back in time (retrospectively) through examination of medical records, interviews to identify a characteristic/risk factor to determine the Determine “What is happening?” at that moment relationship between the risk factor and the disease Advantages/Disadvantages EXAMPLE Appropriate for examining condition that develop over a long time A medical study investigates the prevalence of breast cancer in a population. The researcher can look at a Also for investigation of a preliminary hypothesis wide range of ages, ethnicities and social backgrounds. If a significant number of women from a certain social background are found to have the disease, then the Fast & least espensive researcher can investigate further BUT have the greatest number of possible biases/errors Depend on high-quality existing records Challenge in the selection of appropriate control group 2  8/12/2023 Advantages/Disadvantages Best for determining the status quo of a disease/condition Quick & relatively inexpensive BUT only provide a snapshot in time which may result to misleading information Difficult to get large sample & sample may not be representative of the entire population iii. Cohort : Prospective Experimental studies studies/Retrospective A cohort refers to a group of people who shares Experimental studies that involve humans are known as something in common and remains part of a group over clinical trials an extended time Controlled clinical trials are studies whereby In cohort study, a particular group with a certain experimental drug or procedure is compared with characteristic is observed over a certain period of time another drug/procedure Address “What will happen?”(prospective), what had In randomized clinical trials, subjects are randomized to contributed to the problem? (retro) treatment options providing the best insurance that the result was due to intervention or most rigorous evidence of causality Advantages/Disadvantages Advantages/Disadvantages Appropriate for study that investigates the causes of a Gold standard against which other studies are judged condition, the course of a disease or risk factors Provides the greatest justification for concluding Possess correct time sequence to provide strong causality & subject to least biases evidence for possible causes/effects BUT costly and time consuming BUT extended study period results in large cost Evidence from a clinical trial may not 100% applicable to Other events occurring during study period may current practice as certain factors may have been have affected outcomes excluded Susceptible to problems in patient follow up, drop- out 3  8/12/2023 Research Study Designs Name Class Type Protocol Examples and uses Experiment Experimental Longitudinal Select a sample; Measure Before and after baseline for parameter, apply measurements such as in a intervention, measure outcome. clinical trial for a new therapy, Or divide sample into laboratory experiments, field intervention and control groups. trials e.g. of pesticides or Apply intervention to one fertilisers. group, measure outcome for both and compare. Cohort Observational Longitudinal Define a cohort and assess How a disease progresses, current status and influential Whether family income affects factors. Observe outcomes. educational achievement. Case-control Observational Longitudinal Define a cohort and divide into Whether most lung cancer those exhibiting and those not patients have previously exhibiting parameter of interest. smoked. Investigate history to assess Whether parental divorce risk factors. affects future employment choice. Cross- Observational Cross- Define a sample and collect Prevalence estimates, such as sectional sectional information about parameter of how many people have a given interest. disease, or how many people watch a particular television programme. Repeated Observational Cross- Define a sample and collect Trends, such as how many cross- sectional information about parameter of people are giving up smoking, sectional interest. Repeat at defined the effect of publicity on the intervals, but not necessarily number of people who cycle with the same sample. regularly, or the popularity of a political party over time 4