Unit 4 - THE REAL Introduction to Research Designs
A good epidemiologic research design should include?
1)Enable comparison of a variable between two or more groups. 2)Allow the comparison to be quantified 3)Permit the investigators to determine temporality (i.e: which occurs first?) 4)Minimize biases, confounding, and other problems that would complicate interpretation of the data.
can you perform case control study prospectively or retrospectively?
Because participants are selected on the basis of disease, exposures for ALL PARTICIPANTS are obtained RETROSPECTIVELY... Example: lung cancer cases and non-cancerous controls recall past exposure to cigarette smoke
What observational/descriptive designs are used as correlational studies?
CORRELATIONAL STUDIES Ecological & Cross-Sectional
Which observational study design are analytic model of study designs?
Case-controlled Cohort
What is definition of Cohort Studies?
Definition: A study in which two or more groups of people that are free of disease and that differ according to the extent of exposure (e.g. exposed and unexposed) are compared with respect to disease incidence
Ecological Fallacy: Examples
Example 1 •Assume that you measured the math scores of a particular classroom and found they had the highest average score in the district. •Later you run into one of the kids from that class and you assume that she's a math genius Example 2 •A study of people with eyeglasses showed that they have above-average intelligence. You wear eyeglasses, so you must be intelligent.
Ecological Study Examples
Example 1 •Geographical distribution of farm animals compared with geographical clustering of asthma cases Example 2 •Study that showed a period with increasing internet cable connections was correlated with a decrease in sexual assaults
•A drawback to ecological studies is that they require an extensive budget and staffing. -True or False
False What are Strengths of Ecologic Studies? •Can be done quickly and inexpensively, often using already available information •Surveillance data can permit comparisons of disease rates in different geographic regions •Analysis and presentation are relatively simple and easy to understand •Ability to achieve a wider range of exposure levels than could be expected from a typical individual-level study
What is Randomized Controlled Trials (RCT)
Generally regarded as the most scientifically rigorous method of hypothesis testing available in epidemiology
•An experimental study was conducted among adults with Type 2 diabetes in order to determine if a new medication was more effective in reducing blood glucose levels than the currently used medication. What type of study is this? A.Case Report B.Case-Control C.Case Series D.Retrospective Cohort E.Cross-Sectional F.Prospective Cohort G.Ecologic H.Randomized Trial
H.Randomized Trial
What are difference between in-depth interview vs focus groups?
Interviews Appropriate for Eliciting individual experiences, feelings Strengths Addressing sensitive topics Elicits in-depth response, with nuances and contradictions Focus Group Appropriate for Identifying group norms Discovering variety within population Strengths Elicits information on a range of norms in a short time Group dynamics stimulates conversation, reactions
What are descriptive study designs include?
Observational - Descriptive +Case reports and series +Ecological studies +Cross sectional surveys
What are Classification of Epidemiologic Study Designs?
Observational - Descriptive +Case reports and series +Ecological studies +Cross sectional surveys - Analytic + case control + Cohort Experimental -Randomized controlled trials -Others (e.g. before - after studies and nonrandomized trails)
What are two types of research designs?
Qualitative vs Quantitative
What are adv. and disadv. of qualitative methods?
Qualitative Methods: Advantages •Flexibility •Useful for describing complex phenomena •Rich and exploratory in nature •Meaningful and culturally salient to the participant Disadvantages •Findings might not be generalizable •Cannot test hypotheses; needs to be confirmed with survey research •Takes more time to collect and analyze •More easily influenced by bias
What is Ecological Fallacy?
The error that arises when information about groups of people is used to make inferences about individuals •"An association observed between variables on an aggregate level does not necessarily represent the association that exists at the individual level." •Implication - group-level data cannot be used for hypothesis-testing unless it is a summary of individual measures
•Case control studies are good for studying rare diseases. -True or False
True Case control •Good for studying rare diseases •Can use smaller sample sizes •Cost/time effective when using previously collected (RETROSPECTIVE) exposures •Tend to support causal hypotheses by establishing associations
More questions
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7150150/ https://www.cdc.gov/csels/dsepd/ss1978/lesson1/section7.html
What is Systematic Review?
}Systematic reviews: A high-level overview of primary research on a particular research question }The Cochrane Collaboration promotes use of systematic reviews }Key points: ◦Scientific investigation in itself, with a pre-planned methods section ◦Comprehensive search of all potentially relevant studies and the use of explicit, reproducible criteria in the selection of studies for review ◦Seeks to collate all evidence that fits pre-specified eligibility criteria in order to address a specific research question
What is Observational Studies?
•"Natural" experiments: investigator lets nature take its course •People expose themselves to noxious or healthy substances through personal habits, choice of occupation, place of residence, etc •Provide information on exposures that occur in natural settings; not limited to preventions and treatments
Quantitative vs. Qualitative
•"While quantitative research can tell us much about the incidence and outcomes of disease, it cannot answer how to get patients to use medication when it does not make them feel better, or why, despite improving medical technology, asthma is getting worse."
What are Limitations of Clinical Trials?
•Are time-consuming and usually costly •Can study only interventions or exposures that are controlled by investigator •May have problems related to therapy changes and dropouts •May be limited in generalizability •Are often unethical to perform at all
What is bias?
•Bias -Any systematic error (not random) in a study that leads to an incorrect estimate of the association between exposure and disease -Can occur in the design, implementation, or analysis stages of a study -Tendency to overestimate or under estimate a parameter Types of Bias •Selection bias - Variation in how participants are selected, i.e. the sample is non-representative •Information bias - inaccuracy in measurement or classification •
What is difference between Experimental vs. Observational Studies?
•Both compare "exposed" and "unexposed" groups •In experimental (intervention) studies, the researcher determines who is exposed •Most epidemiologic studies are observational (natural experiments) - exposed and unexposed groups exist before an epidemiologist arrives on the scene
What are Strengths of Ecologic Studies?
•Can be done quickly and inexpensively, often using already available information •Surveillance data can permit comparisons of disease rates in different geographic regions •Analysis and presentation are relatively simple and easy to understand •Ability to achieve a wider range of exposure levels than could be expected from a typical individual-level study
What are limitation of Case Series?
•Cannot be used to test for a valid statistical association •Hard to interpret because there is not an appropriate comparison group •This lack can either obscure a relationship or suggest an association where none actually exists
What are limitation of Case Reports?
•Cannot be used to test for presence of a valid statistical association •Based on the experience of only one person •The presence of any risk factor, however suggestive, may simply be coincidental
What are advantage of ecological studies?
•Cheap •Easy •Suggestive •Fun •dangerous
What is Case Series?
•Collections of individual case reports, usually in a fairly short time window •Often used as an early means to identify the beginning or presence of an epidemic --- Example: 5 young, previously healthy homosexual men diagnosed with Pneumocystis carinii pneumonia (PCP) at three LA hospitals during a 6-month period in 1980-1981
What are Analytic Studies?
•Comparison is explicit, since investigator assembles groups of individuals for specific purpose of systematic determination of whether or not risk of disease differs for exposed versus unexposed • •Use of an appropriate comparison group allows testing of epidemiologic hypotheses in analytic study designs
What are Strengths of Clinical Trials?
•Considered as providing most reliable evidence from epidemiologic research •Randomization offers ability to control both known and unknown influences (confounders) •When well-designed and conducted can provide most direct evidence on which to judge whether an exposure causes or prevents a disease
What is Cross-sectional Studies?
•Cross-sectional studies are carried out for public health planning and for etiologic research. •Most governmental surveys conducted by the National Center for Health Statistics are cross-sectional in nature -Example: the US Cens •Deals with Prevalence
What statistical measure is commonly used in ecologic studies?
•Data typically presented as summary statistics such as means, median, or mode. •A simple correlation coefficient is used as the measure of association •hypothesis-generating studies focusing on the relationship between exposures and outcomes
What are descriptive study?
•Describe patterns of disease in terms of person, place and time •Useful for healthcare and education planning •Often provide important clues about possible determinants of a disease •Good for building hypotheses
What are Limitations of Cross-sectional Studies?
•Does not separate cause-effect relationships in associations established; temporal relationships unclear •Deals only with survivors •Not good for rare diseases •Does not identify risk or future likelihood of disease from given characteristic •Limited usefulness for explosive epidemics or acute, short duration illnesses such as measles or upper respiratory infections
What is a quasi-experimental design?
•Establish whether the treatment or intervention had a population impact •Attribute causality +++Was the change in outcome uniquely attributed to the treatment or intervention? •The main difference between RCT vs Quasi is - Randomization
What are Objectives of Randomized Trials?
•Evaluate therapeutic and preventive aspects of medical practice •Evaluate new approaches to health care delivery •Evaluate impact of health education on health behavior
What are Strengths of Case-control Studies?
•Good for studying rare diseases •Can use smaller sample sizes •Cost/time effective when using previously collected (RETROSPECTIVE) exposures •Tend to support causal hypotheses by establishing associations
What are Limitations of Ecologic Studies?
•Lack of individual-level information** →"ecological fallacy" •Correlational data represent average exposure levels rather than actual individual values •Investigators' inability to detect subtle or complicated relationships (e.g. j-shaped or other curvilinear relationships) •Lack of information on potential confounding factors
What are Limitations of Cohort Studies?
•Likely to be large and expensive •Inefficient for studying rare diseases •Potentially long duration of follow-up for some outcomes •Loss to follow up of subjects** •Must anticipate secular trends in technology, behaviors, etc... •Difficulties in measuring confounding variables •Exposures can change through study •The quality of the data is not consistent, when compared to other types of studies
What are Strengths of Cohort Studies?
•Maintain temporal sequence between exposure & outcome •Assess the various effects of a particular exposure •Study new or rare exposures •Calculate measures of risk •Avoid bias in the exposure measurement •Provides a complete picture of experience following exposure
What does Blinding & Double Blinding mean in clinical trial?
•Masking knowledge of exposure status to - Study participants -Observers/data collectors -Single Blind/Masked •Study participants blind to exposure status -Double Blind/Masked •Both study participants AND observers/data collectors blind to exposure status
What are the difference between systematic review vs meta-analysis?
•Meta-analysis is a type of systematic review -Not all systematic reviews are meta-analyses • •Meta-analysis is the use of statistical methods to combine results of individual studies. -By statistically combining the results of similar studies we can improve the precision of our estimates of treatment effect, and assess whether treatment effects are similar in similar situations.
What is Case Reports?
•Most basic type of descriptive study of individuals •Careful, detailed profile of single patient by one or more clinicians Case Report: Example In 1961 a 40-year-old premenopausal woman developed a pulmonary embolism 5 weeks after beginning to use an oral contraceptive --- 1901: Alois Alzheimer treated 51-year-old patient at Frankfurt asylum Strange behavioral symptoms including short term memory loss 1906: Mrs. Deter died; Alzheimer took patient records and brain to Munich With two Italian physicians, he used staining techniques to identify amyloid plaques and neurofibrillary tangles
What are some key points about Case-control Studies? •Those who are diseased are called ______ •Those who are not diseased are called ______
•Most frequently used study design •Participants selected on the basis of whether or not they are DISEASED (in a cohort study participants are selected based on exposure status) •Those who are diseased are called CASES •Those who are not diseased are called CONTROLS •Examples: -Smoker and non-smoker who develop lung cancer Smoker and non-smoker who develop CAD
What is Placebo & Placebo Effect?
•Placebo- Pharmacologically inactive substance given as a substitute for an active substance, especially when the person taking/receiving it is not informed whether it is an active or inactive substance
What are two types of cohort studies?
•Prospective (concurrent): -begin with a group of exposed and unexposed and follow them for years •Retrospective (non-concurrent): -look back into the past to find a group of exposed and unexposed and 'follow' them to the present -------------------------- *Cohort follows for exposure. **Case control retrospective follows the disease.
What is Qualitative Research?
•Qualitative research is a field of investigation used by many disciplines to explore a variety of different subject matter •"Qualitative research aims to gather an in-depth understanding of human behavior and the reasons that govern such behavior" •"The qualitative method investigates the why and how of decision making, not just what, where, when" •Concentrates on words and observations to express reality •Describes people in natural situations and settings
Qualitative Study How is it used? •When is it appropriate to be used?
•Qualitative research is used to gain insight into people's attitudes, behaviors, value systems, concerns, motivations, aspirations, culture or lifestyles [again why and how, not when, where, what] •It can also provide information about often contradictory behaviors, beliefs, opinions, emotions, and relationships of individuals •When is it appropriate to be used? -A problem or issues needs to be explored in more-depth
What are qualitative Methods of Data Collection?
•Qualitative researchers typically rely on the following methods for gathering information: -Participant Observation -Passive Observation -Field Notes -Receptively listening -Reflexive Journals -In-Depth Interview -Focus Groups -Analysis of documents (archival data)
What are Strengths of Cross-sectional Studies?
•Quick, requiring only a "one-time" examination and/or interview •Less expensive than other designs •Helpful in program planning and determining types of health services needed •Useful in determining associations between variables of interest, thereby generating hypotheses
What are Correlational Studies: Ecological and Cross-Sectional USED FOR?
•SIMULTANEOUSLY assess exposure and disease •Establish the STRENGTH and DIRECTION of association between exposure and disease •Used to generate hypotheses NOT test causality
What is confounder?
•Similar to bias and often confused •The influence of third variables in a study which lead to an incorrect estimate of the association between the exposure and disease variables •EXAMPLE: People who carry matches are more likely to develop lung cancer. •Smoking is the confounding factor in this relationship (or "Third Variable")- Smokers are more likely to carry matches and they are also more likely to develop lung cancer. •To be a confounder, the third variable must: •Be associated with the exposure •Be an independent risk factor for the disease •Not be in the causal pathway between exposure and disease
What are Experimental Studies?
•Studies preventions and treatments for diseases •Investigator actively manipulates which groups receive the agent under study
What are key concept of basic and clinical biostatistic? part 2
•Studies with interventions are called experiments or clinical trials. → They provide stronger evidence than observational studies. •The single best way to minimize bias is to randomly select subjects in observational studies or randomly assign subjects to different treatment arms in clinical trials. •Bias occurs when the way a study is designed or carried out causes an error in the results and conclusions. Bias can be due to the manner in which subjects are selected or data are collected and analyzed. •Clinical trials without controls (subjects who do not receive the intervention) are difficult to interpret and do not provide strong evidence. •Each study design has specific advantages and disadvantages.
What are key concept of basic and clinical biostatistic? part 1
•Study designs in medicine fall into two categories: studies in which subjects are observed, and studies in which the effect of an intervention is observed. •Observational studies may be forward & backward-looking (cohort), backward-looking (case-control), or looking at simultaneous events (cross-sectional). Cohort studies generally provide stronger evidence than the other two designs. •Studies that examine patient outcomes are increasingly published in the literature; they focus on specific topics, such as resource utilization, functional status, quality of life, patient satisfaction, and cost-effectiveness.
What are Limitations of Case-control Studies?
•Subject to bias (selection and recall) •Can't calculate incidence •Selecting appropriate controls can be challenging •Data from clinical records may be inadequate or incomplete
What is Quantitative Analysis?
•Test hypotheses and examine associations •Describe a population (e.g. diabetes prevalence, socio-economic characteristics) •Uses Statistical power and sample size •Generalizability and representativeness
What is a Ecologic Studies?
•Use data from entire populations (Aggregate data) •Identify groups by place and time •Can compare disease frequencies -between different populations (same time period) -in same population at different time periods •Disease described relative to population-wide factors (e.g. age, calendar time, health service utilization, food consumption, medication) •The unit of analysis is a population rather than an individual •Very crude measures •Often draw associations between occupation, environment and disease