Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
36 Cards in this Set
- Front
- Back
|
Types of Study Design
Experimental |
Randomized clinical trials
|
|
|
Types of Study Design
Observational |
Follow-up (cohort) study
Case-control study Cross-sectional Case series Case reports |
|
|
Relative Strength of Causal Relationships Based on Study Design
|
Randomized clinical trials
Follow-up (cohort) study Case-control study Cross-sectional Case series Case reports ** increasing from the bottom up ** |
|
|
Study Design Case Reports and Case Series
|
--Observations related to a drug or technology being applied to single patient or group of patients
--Most basic type of descriptive study of individuals, consisting of a careful, detailed report by one or more clinicians of the profile of a single patient or group of pts. |
|
|
Study Design Case Reports and Case Series
advantages |
Hypothesis screening or hypothesis generating
|
|
|
Study Design Case Reports and Case Series
disadvantages |
Can not prove causality
|
|
|
Study Design
Cross Sectional Studies |
Quantify outcome (i.e., prevalence) in pts exposed to risk factor vs. outcome in pts not exposed to risk factor at one single point in time
|
|
|
Study Design
Cross Sectional Studies Advantages |
Defines prevalence
Little time required for completion |
|
|
Study Design
Cross Sectional Studies Disadvantages |
--Selection bias: Selective differences between comparison groups that impacts on relationship between exposure and outcome
--Weak evidence for causality |
|
|
Cross Sectional Studies
Health Interview Survey is: |
a national cross-sectional survey that periodically collects extensive information by questionnaire from a representative sample of over 100,000 individuals throughout the United States.
|
|
|
Cross Sectional Studies
participants are asked to: |
record their current status, as of the date of the questionnaire, with respect to personal and demographic characteristics, illnesses, health habits, and utilization of health care resources.
|
|
|
Cross Sectional Studies
The frequencies of various diseases, injuries and other health outcomes are calculated and examined in relation to age, sex, race, socioeconomic variables, medication use, cigarette smoking and other risk factors |
true
|
|
|
Study Design
case control (design...) |
see slide 10
|
|
|
Case Control Studies
Advantages: |
Good for rare diseases (prospective trials would require very large sample size), less expensive, can be completed rapidly (no follow up needed)
|
|
|
Case Control Studies
Disadvantages |
Highest potential for biases (recall, selection, others), weak evidence for causality
|
|
|
Case Control Study
example |
To evaluate the possible association between consumption of artificial sweeteners and risk of bladder cancer, investigators examined 592 patients hospitalized in the Boston area with a primary cancer of the bladder and 592 control patients without bladder cancer who were selected at random from the general population.
All participants were interviewed to obtain information on their history of consumption of artificially sweetened beverages and foods, and use of sugar substitutes, and a number of other possible risks factors for bladder cancer, including smoking, medication use, and coffee consumption |
|
|
Study Design
Cohort (design) |
see slide 13
|
|
|
Cohort Study
|
--Usually prospective; occasionally retrospective
--Compares outcomes in group exposed to risk factor with outcomes in group not exposed to risk factor |
|
|
Cohort Study
Advantages |
Defines incidence and decreases biases (sampling, measurement, reporting)
Stronger evidence for causality |
|
|
Cohort Study
Disadvantages |
Expensive, requires follow up time; does not eliminate confounding (other factors besides exposure may alter outcomes)
May not be practical for rare diseases, outcomes Weak evidence for causality |
|
|
Cohort Studies
example |
The Nurses’ Health Study, enrolled over 120,000 married female nurses, aged 30 to 55 years, who were registered in one of 11 US states at the time of the initial mail survey in 1976.
On a baseline questionnaire, participants provided information on a number of demographic, reproductive, medical history and life-style variables. The enrolled nurses then completed, at 2-years intervals, follow-up questionnaires that asked about the development of outcomes during that period, updated the exposure information that had been collected on the baseline questionnaire, and obtained data on new variables of interested. By comparing those classified as exposed or nonexposed to a particular risk factor, such as use of oral contraceptives, postmenopausal hormones, or hair dyes, consumption of dietary fat, age at first birth and menopause, and family history of disease. Provided important data about the relationship of these variables with the development of cancer and cardiovascular disease. |
|
|
Randomized Clinical Trials
(design) |
see slide 17
|
|
|
Clinical Trials
|
“Experiment”
Randomization to treatment or control group Treatment or no treatment (placebo) New treatment vs. current standard treatment (active) Compare rate of outcome between randomized groups |
|
|
Clinical Trials
Advantages |
Gold standard if placebo-controlled and double-blinded
Randomization reduces confounding Best evidence for causality |
|
|
Clinical Trials
Disadvantages |
Expensive and may expose treatment group to unnecessary risk
Longer time to complete and inappropriate for rare diseases or outcomes |
|
|
Types of Blinding
Single Blind |
Either subjects or investigators are unaware of assignment of subjects to active or control groups
|
|
|
Types of Blinding
Double Blind |
Both subjects and investigators are unaware of assignment of subjects to active or control groups
|
|
|
Types of Blinding
Triple Blind |
Both subjects and investigators are unaware of assignment of subjects to active or control groups; another group involved with interpretation of data is also unaware of subject assignment
|
|
|
Methods of Data Analysis
The issue: |
How should investigators analyze study data if one or more pts have not adhered to the allocated treatment strategy?
|
|
|
Methods of Data Analysis
The options |
Per protocol analysis: Excluding the participants who did not adhere to the treatment allocated from the study (“method effectiveness”)
Intention to treat analysis: Including the data of all subjects who were lost in the study (“use effectiveness”) |
|
|
Per Protocol Analysis
Problems with the "per protocol efficacy" analysis |
Estimate of treatment effect likely to be flawed (i.e., overestimated), since reasons for non-adherence to protocol may be related to patients’ prognosis
Empirical evidence suggests that participants who adhere tend to do better than those who do not, even after adjustment for all known prognostic factors, and irrespective of assignment to treatment or placebo Excluding non-adherent participants from the analysis leaves those who may be destined to have a better outcome, and destroys the unbiased comparison afforded by randomization |
|
|
Intention to treat analysis
|
Data of lost subjects censored at time of departure
Imputation of outcomes The subject’s last score or measurement at the time of discontinuation Analysis of best case and worst case scenarios Average score or measurement for the entire group Multivariate analysis of prognostic factors to predict most likely outcome Significant loss to follow up negatively affects intention to treat analysis, as numerous assumptions are made about the data censored (imputed outcomes are guessed) |
|
|
Study Design
Cross Over Study (design) |
see slide 24
|
|
|
Cross Over Study
|
Stronger, more powerful design
Subject serves as own control, eliminating differences in prognostic factors |
|
|
Cross Over Study
Limitations |
Not appropriate when outcomes are permanent (e.g., death) or effect is long-lasting (bone density changes, hair loss or hair growth, etc…)
|
|
|
Important Considerations for Cross-Over Studies
|
Washout period must be of sufficient duration to prevent carry-over effects
Multiple cross-over periods are useful when studying disease with exacerbations and remissions Subjects should be randomized to treatment order Both investigators and subjects should be blinded to time when cross-over occurs Subject drop-outs and deaths should be minimized |
