Introduction to Research Methods Flashcards
5 types of quantitative observational research designs
- case reports/case series
- ecological
- cross sectional
- case-control
- cohort
2 Broad types of quantitative research designs
- observational (which has 5 subgroups)
- experimental (which has 2 subgroups)
2
2 subgroups of quantitative experimental research designs
- quasi-experiemtnal (nonrandom allocation)
2. randomized controlled trials (random allocation.).
in qualitative research, its exploratory, seeking understanding and using ____ analysis,
in quantitative research, it’s structures; seeks explanation or causation and uses ___- analysis
in qualitative research, its exploratory, seeking understanding and using INDUCTIVE analysis,
in quantitative research, it’s structures; seeks explanation or causation and uses DEDUCTIVE- analysis
CASE REPORTS VS CASE SERIESc
case report = 1 patient
case series = more than 1 patient
a case report is a ____ study of patients based on some variable of interest
descriptive study
case report/case series is a quantitative observational subset.
example of case report/case series
locating at liver function in a group of patients exposed to a particular toxin. no controls– you just look at the patients that happened to have exposure to a toxin
t/f case reports and case series have a control group
false. there is no control group
3 pros and 2 cons to case reports/series (quantitative observational)
pros
- often quick and inexpensive
- no risk to patient
- ideal for rare diseases, new medical conditions (HYPOTHESIS GENERATING)
cons
- can’t assess causation
- small sample size (can lead to selection bias)
ecological studies is a subset of ___ ____ studies
subset of quantitative observational studies
ecological studies is a snapshot in ___ of the characteristics of a ___ population group. make an example of an ecological stude.
snapshot in TIME of the characteristics (ex/ exposure and disease) of a large population group.
example: looking at rates of breast cancer in relation to dietary fat consumption; mean population exposure and disease.
ecological fallacy
when characteristics of a group are assigned to individuals without considering variability amongst individuals.
example of an ecological fallcy
in a study designed to examine relationships between diet, lifestyle, heart disease, and stroke, researchers found that the mean entry-level blood pressures and stroke mortality rates were inversely correlated for certain cohorts (study groups) of men aged 45 to 59 with 25-year follow-up. The finding was contrary to expectations. Subsequent analyses carried out at the individual level showed that the association between blood pressure and stroke mortality was strongly positive in most of the study groups. The explanation of this paradox is that within each cohort, individuals who had experienced a stroke and who had died from a stroke tended to have high blood pressure. However, when the individual values in each cohort were averaged and used to calculate the correlation, cohorts with higher average blood pressures may have turned out to have smaller mortality rates simply because of the heterogeneity of correlations among the cohorts.
cross sectional studies are aka ___ studies
prevalence studies. cross sectional studies are a subset of observational quantitative studies.
difference between ecological and cross sectional studies in terms of sample sizes.
ecological studies focus on the entire population, whereas XS studies focus a little closer on more of an individual level. cross sectional studies are snapshots of exposure and disease outcome for each subject.
example of cross sectional studies
measuring cholesterol levels and ECG evidence of IHD; exposure and disease outcome OVER DIFFERENT TYPES OF PTS (ex/ high cholesterol vs no cholesterol, IHD vs no IHD) is measured simultaneously for each subject.
3 pros and 4 cons of ecological or cross sectional studies.
pros
- often quick and inexpensive ( no long periods of followup, you just measure what’s going on in the sample
- no risk to patient– it’s observational
- good for generating hypothesis
cons
- can’t assess causation/association can be difficult to interpret
- not suitable for rare diseases or diseases wit short duation
- can’t measure incidence
- ecological fallcy
- prone to recall bias.
cross sectional studies allow for the measurement of ____
prevalence;; you can calculate the prevalence of disease or the prevalence of exposure.
POD: exposed compared to non-expoaesd persons
POE: diseased compared to non-diseased persons.
gotta be wearing of prevalence vs incidence, and temporal relationships that can confound the data.
In ___-___ studies, it involves comparing exposure data in patients who have the disease (cases) and those who don’t (control)
case-control studies.
in the ‘cases’ group– you can have the ‘were exposed’ subset and ‘were not exposed’ subset
in the controls group – you can have the ‘were exposed’ and ‘were not exposed” subset.
recall bias
systematic differences leading to differential recall between cases and controls
limitation of recall:
random limitation in ability to recall information.
in cohort studies, you compare incidence of disease in groups that are exposed vs those without exposure. of these two groups, you can have exposed –? disease or NO disease and you can have NOT-exposed –> disease vs NO disease. outline an example of a cohort study
ex/ looking at number of people who smoke who go on to develop CHD, vs number of people who don’t smoke (NOT EXPOSED) and develop CHD
information bias involves problems with ___
problems with recall
what factors can cause selection bias?
- non participation and non response
- loss to follow up
- people volunteering (they may just have common personality traits)
factors of information bias
- quality of data (especially with retrospective cohorts)
2. subjective classification of disease (observer bias)
case control studies vs cohort studies
case-control studies are always retrospective
cohort studies can be retrospective or prospective.
**this might not be true lol
The cohort study design identifies a people exposed to a particular factor and a comparison group that was not exposed to that factor and measures and compares the incidence of disease in the two groups. A higher incidence of disease in the exposed group suggests an association between that factor and the disease outcome. This study design is generally a good choice when dealing with an outbreak in a relatively small, well-defined source population, particularly if the disease being studied was fairly frequent.
The case-control design uses a different sampling strategy in which the investigators identify a group of individuals who had developed the disease (the cases) and a comparison of individuals who did not have the disease of interest. The cases and controls are then compared with respect to the frequency of one or more past exposures. If the cases have a substantially higher odds of exposure to a particular factor compared to the control subjects, it suggests an association. This strategy is a better choice when the source population is large and ill-defined, and it is particularly useful when the disease outcome was uncommon.
Hawthorne effect
participants alter behaviour if they know they’re being observed.
this might be a design challenge especially in a drug and addiction trial.
advantages and weaknesses of case-control and cohort studies
pro of cohort:-
- can examine multiple effects of single exposure, able to assess incidence, better assessment of CAUSALITY (but not proved, just better assessment of causality compared to case-control
pro of case control;
- good for rare disease/diseases with long latency. NOT GOOD FOR RARE EXPOSURES, AND CAN’T ASSESS INCIDENCE OR TEMPORALITY.
common weakness of both: risk fo bias (selection, loss to follow up, recall, observer bias)
common advantage
- can both examine multiple exposures
- can be retrospective ( more-so in case- control studies)
2 measures of association that are seen when we look at observational studies.
- relative risk: ratio of risk of disease in exposed individuals versus the risk of disease in non-exposed individuals (ie/ incidence of disease in exposed/ incidence of disease in non exposed
- odds ratio = ratio of odds that a case was exposed versus the odds that controls were exposed.
relative risk:
the incidence of disease in exposed vs incidence of disease in non-exposed.
odds ratio in a cohort study
ratio of odds that a case was exposed versus the odds that controls were exposed.
the only study that can measure relative risk is a
cohort study; in these studies, you look at all the people who have been exposed vs non exposed and see if how many of each group develop the disease.
cohort studies can allow you to measure relative risk because it is the only study that can calculate incidence.
odds ratio in a case-control study
odds of exposure among cases/ odds of exposure among controls.
what is the equation for a cohort study odds ratio
OR = (a/b)/(c/d) or (ad)/(bc)
what is the equation for an odds ratio for a case control study
OR = (a/c)/(b/d) = (ad/bc)
why can’t you calculate relative risk in a case-control study?
in a case- control study, your exposed and non-exposed are dependent variables. you pick the number of people who have the disease, the exposed vs non-exposed are measred/dependent variable.
vs in a cohort study, the exposed and non-exposed are THE INDEPENDENT VARIABLe. you pick your total patient population (exposed vs non exposed numbers) and then measure the amount of people with the disease
a study was carried out to assess the association of CHD with smoking. Of the 200 patients with CHD selected, 112 were found to be smokers. 400 controls were selected from the same geographic region, general ethnicity and SES. 176 of those patients were found to be smokers.
calculate the odd ratio for the association between smoking and CHD
in this study, the exposure (smoking) was the dependent variable. they chose people with CHD. therefore, this is a case control study.
odds ratio in case control study = odds of exposure among cases/odds of exposure among controls.
= (112/88)/(176/224) = 0.56/0.44 = 1.62
the odds that a person with CHD is a smoker versus a person without CHD being a smoker is 1.62. THIS IS NOT THE SAME AS THE STATEMENT AS the odds that a person with CHD is a smoker is 1.62 TIMES HIGHER than not having CHD– the odds ratio approximate relative risk when disease prevalence is low.
cohort studies can determine relative risk and an odds ratio, whereas case-control studies can only determine odds ratio– CC studies cannot determine relative risk.
when will an odds ratio be similar to a relative risk?
- when the cases studied is representative of all people with the disease in the population from which the cases were drawn
- controls studied must be representative (in regard to history of exposure) of all people without disease in the population from which the cases were drawn
- disuse studies does not occur frequently (low incidence)
what is a quasi experimental study and give an example
- a term for experimental study where interventions are assessed on non-randomized and non-controlled gorups
ex/ measuring the difference in wellbeing before and after applying an education initiative to a patent group. there is no control ( no group where no education initiative is applied)
an ______ is an experimental study where patients are randomized into treatment and control groups, and effects of intervention of interest are assessed.
a RCT
design characteristics of an RCT (RIDTA)
- randomization (multiple methods; stratified randomization like grouping patients by age and then randomly putting equal numbers of each age into the treatment and control grop– prevents age being a confounding difference between the two groups)
2 inclusion/exclusion criteria
- blinding
- trial DESIGN (run in period, cross over; experimental and control groups are switched halfway through, so each group acts as their own control)
- analysis (pre-determined sample size, intention to treat, pre-determined variables)
sources of error in RCT
- bias: systematic error in design, execution or analysis of a study that affects the relationship between exposure and outcome (ex/ the method used to select subjects leads to incorrect conclusion
- confounding: where another variable affects the relationship between exposure and outcome
- effect modification: where a variable differentially modifies observed association between exposure and outcome; different groups have different risk estimates. (ex/ different relative risk for males vs females or ethnicities)
what is effect modification?
effect modification is a source of error in a randomized controlled trial where a variable differentially modifies theobserved association between exposure and outcome; different groups have different risk estimates. (ex/ different relative risk for males vs females or ethnicities)
examples of selection bias (5)
- exclusion bias
- non-response bias
- prevlance-inciidence bias
- healthy worker bias
- loss to follow up.