CONCEPTUALIZING A RESEARCH DESIGN Detty Nurdiati Dept of Obstetrics & Gynecology Fac of Medicine, Universitas Gadjah Mada Yogyakarta, Indonesia
Conceptualizing a Research Design The Research Design The Definition of a Research Design The Functions of as Research Design Selecting a Study Design
RESEARCH PROCESS 1 Formulating a research problem 2 Conceptualizing a research design 3 Constructing instruments for data collection 4 Selecting a sample 5 Writing a research proposal 6 Collecting data 7 Processing data 8 Writing a research report
1. Formulating a research problem Reviewing the literature Formulating a research problem Identifying variables Contructing hypothesis
2. Conceptualising a research design Overview of research design Selecting a study design
3. Constructing an instrument for data collection Seleting a method of data selection Collecting data using appropriate scales Establishing the validity and realibility of your research instrument
4. Selecting a sample Sample Size Sampling technique
5. Writing a research proposal Magnitude of the research problem Objectives of the research problem How to find the answer the research problem
6. Collecting data Team and technical management Considering ethical issues
7. Processing data Data management Data analyses Displaying data
8. Writing a research report Report the findings Compare to other studies Limited of the study Conclusion Suggestion remarks
RESEARCH DESIGN
Definition Research is an activity that contributes to the understanding of a phenomenon (Kuhn, 1962; Lakatos, 1978) Research Design is a procedural plan that is adopted by the researcher to answer research questions validly, objectively, accurately and economically (Kumar, 1996) Includes an outline of what investigators will do from writing the hypotheses and their operational implications to the final analysis of data
The Functions of Research Design 1 Relates to the identification and/or development of procedures and logistical arrangement required under the study 2 Emphasizes the importance of quality in these procedures to ensure their validity, objectivity and accuracy
Independent variable Type of Diet Counseling Study population Pregnant Women Extraneous variables Sosioeconomic condition Prepregnancy BMI Dependent variable Total Weight Gain Change in total weight gain: -Change attributable to type of diet counseling ± -Change attributable to extraneous variables ± -Change attributable to chance or random variables
Which Research Design?
Borgman, J (1997). The Cincinnati Enquirer. King Features Syndicate.
Methodology Comparison Quantitative Explanation, prediction Test theories Known variables Large sample Standardized instruments Deductive Qualitative Explanation, description Build theories Unknown variables Small sample Observations, interviews Inductive
Types of Quantitative Study Design Number of contacts Reference period Nature of investigation 1: cross sectional 2: before and after study 3: longitudinal study Retrospective Prospective Retrospectiveprospective Experimental Non-experimental Semiexperimental
Study Pyramid Best Worst
Rating clinical evidence Assessment system of the US Preventive Services Task Force
Source: The Lancet 2002; 359:57-61 (DOI:10.1016/S0140-6736(02)07283-5)
Descriptive Study Do not feature a comparison (control) group. Do not allow assessments of causal association. Case reports/case-series reports Surveilance Studies
Projects Worked On Is the Study Design appropriate for the Research Question? assessed through more rigorous research such as analytical studies or RCT Descriptive Studies the first foray into a new area of medicine. Describe the frequency, natural history, and possible determinants of a condition. Show how many people develop a disease or condition over time, describe the characteristics of the disease and those affected generate hypotheses about the cause of the disease Time Spent
Source: The Lancet 2002; 359:57-61 (DOI:10.1016/S0140-6736(02)07283-5)
Source: The Lancet 2002; 359:57-61 (DOI:10.1016/S0140-6736(02)07283-5)
Cross Sectional Study: A Snapshot in time A frequency survey or a prevalence study To examine the presence or absence of disease and the presence or absence of an exposure at a particular time. Prevalence, not incidence, is the focus. Both outcome and exposure are ascertained at the same time the temporal relation between the two might be unclear.
Cross Sectional Study: Example A cross-sectional study finds obesity to be more common among women with than without arthritis. Did the extra weight load on joints lead to arthritis, or did women with arthritis become involuntarily inactive and then obese?
Cohort Study: looking forward in time Cohort studies proceed in a logical sequence: from exposure to outcome. Investigators identify a group with an exposure of interest and another group or groups without the exposure then follow the exposed and unexposed groups forward in time to determine outcomes. If the exposed group develops a higher incidence of the outcome than the unexposed, then the exposure is associated with an increased risk of the outcome.
Lancet 2002; 359: 341-45 Cohort Study: marching towards outcomes
Cohort Study: looking forward in time The cohort study has important strengths and weaknesses. Because exposure is identified at the outset, one can assume that the exposure preceded the outcome. Recall bias is less of a concern. Enables calculation of true incidence rates, relative risks, and attributable risks. For the study of rare events or events that take years to develop, this type of research design can be slow to yield results and thus prohibitively expensive.
Cohort Study
Case Control: thinking backwards Starting with an outcome, such as disease, this type of study looks backward in time for exposures that might have caused the outcome. Investigators define a group with an outcome (for example, ovarian cancer) and a group without the outcome (controls) through chart reviews, interviews, or other means ascertain the prevalence (or amount) of exposure to a risk factor (eg, oral contraceptives, ovulation-induction drugs) in both groups. If the prevalence of the exposure is higher among cases than among controls, then the exposure is associated with an increased risk of the outcome.
Case-control Study: research in reverse
Case Control: thinking backwards Especially useful for outcomes that are rare or that take a long time to develop, such as cardiovascular disease and cancer. Require less time, effort, and money than would cohort studies. The Achilles heel of case-control studies is choosing an appropriate control group. Controls should be similar to cases in all important respects except for not having the outcome in question. Inappropriate control groups have ruined many case-control studies and caused much harm.
Case Control Study
Case Control: thinking backwards Recall bias (better recollection of exposures among the cases than among the controls) is a persistent difficulty in studies that rely on memory. Because the case-control study lacks denominators, investigators cannot calculate incidence rates, relative risks, or attributable risks. Instead, odds ratios are the measure of association used; when the outcome is uncommon eg, most cancers the odds ratio provides a good proxy for the true relative risk.
Cohort versus Case-Control Study
E(+) E(-) Participants, Patients, Subjects E(+) E(-) Cases Controls Exposure No Expo. Case Control Case Control Observational Designs Retrospective (Case-control) Prospective (Cohort) Today Retrospective Cohort E(+) Cases E(-) Controls E(+) E(-) Cross-sectional Time
Non-Randomised Trial: penultimate design? Some experimental trials do not randomly allocate participants to exposures eg, treatments or prevention strategies. Instead of using truly random techniques, investigators often use methods that fall short of the mark eg, alternate assignment. The US Preventive Services Task Force and Canadian Task Force on the Periodic Health Examination designate this research design as class II-1, indicating less scientific rigour than randomised trials but more than analytical studies.
Non-Randomised Trial: penultimate design? After the investigators have assigned participants to treatment groups, the way a nonrandomised trial is done and analysed resembles that of a cohort study. The exposed and unexposed are followed forward in time to ascertain the frequency of outcomes. Advantages of a non-randomised trial include use of a concurrent control group and uniform ascertainment of outcomes for both groups. However, selection bias can occur.
Randomised Controlled Trial: gold standard The only known way to avoid selection and confounding biases in clinical research. Approximates the controlled experiment of basic science. Resembles the cohort study in several respects, with the important exception of randomisation of participants to exposures. The hallmark of RCT is assignment of participants to exposures purely by the play of chance. Reduce the likelihood of bias in determination of outcomes.
Randomised Controlled Study
Randomised Controlled Trial: gold standard Random allocation precludes selection bias. Feature uniform diagnostic criteria for outcomes and, often, blinding of those involved as to the exposure each participant is receiving reducing information bias. A unique strength is that it eliminates confounding bias, both known and unknown. The trial tends to be statistically efficient. If properly designed and done, RCT is likely to be free of bias and is thus especially useful for examination of small or moderate effects.
The Double Blind Method
Randomised Controlled Trial: gold standard The RCT has internal validity ie, it measures what it sets out to measure it might not have external validity. This term indicates the extent to which results can be generalised to the broader community. The RCT includes only volunteers who pass through a screening process before inclusion the volunteer for trials tend to be different from those who do not; for example, their health might be better. The RCT cannot be used in some instances, since intentional exposure to harmful substances eg, toxins, bacteria, or other noxious exposures would be unethical. As with cohort studies, the RCT can be prohibitively expensive.
Systematic Reviews
Systematic Review Rigorous research methodology Objective review of all the evidence Can be appraised like other research Trustworthy because potential sources of bias can be assessed
Source: The Lancet 2002; 359:57-61 (DOI:10.1016/S0140-6736(02)07283-5)
Selection of Design Because some research questions can be answered by more than one type of research design, the choice of design depends on a variety of considerations, including: speed cost availability of data.
Weight Gain on Different Diets 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
Cigarettes and Lung Cancer Mortality 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
HIV in a Women s Prison Explanatory var = IV drug use (1 = users, 2 = non-user) Response var = HIV serology (1 = positive, 2 = negative) Data are observational on the individual-level. But onset data cannot be unraveled. Thus, data are cross-sectional
Toxicity in Cancer Patients Explanatory variable = generic drug use (generic: 1 = yes, 2 = no) Response variable = cerebellar toxicity (tox: 1 = yes, 2 = no) Data are observational, individual-level, longitudinal, with all individuals followed over time. Thus, data are cohort. Comment: This is a retrospective cohort based on data abstracted data from medical records.
Esophageal Cancer and Alcohol Consumption Explanatory var = alcohol consumption (alc2: 1 = high, 2 = low) Response var = esophageal cancer (case: 1 = case, 2 = control) Data are observational, individual-level, with study of all population cases but only a sample of non-cases. Thus, data are case-control.
Summary Understanding what kind of study has been done is a prerequisite to thoughtful reading of research. Clinical research can be divided into experimental and observational; observational studies are further categorised into those with and without a comparison group. Only studies with comparison groups allow investigators to assess possible causal associations, a fact often forgotten or ignored.
QUESTIONS?
LITERATURES AN OVERVIEW OF CLINICAL RESEARCH: THE LAY OF THE LAND. DAVID A GRIMES, MD AND KENNETH F SCHULZ, PHD. THE LANCET, VOLUME 359, ISSUE 9300, PAGES 57-61 (JANUARY 2002). DOI: 10.1016/S0140-6736(02)07283-5 RESEARCH METHODOLOGY: A STEP-BY-STEP GUIDE FOR BEGINNERS. RANJIT KUMAR. 1999. LONDON: SAGE PUBLICATIONS LTD