EBM Quiz, DILE Midterm, DILE -- Chapter 2
Define- Evidence Based Reviews
Critically appraised topics and guidelines: summaries of evidence that may or may not be systematically reviewed but uses higher level resources to inform clinicians of best practices. Examples: Dynamed Up To Date Best Evidence Topics National Guidelines Clearinghouse
A prospective study in which patients are moved from one group to another. - Allocation of patients from one group to the next can be randomized. - Random allocation makes a this type of trial stronger. - Length of washout timeframe needs to be sufficient duration to eliminate effects of first intervention before employing the second intervention.
Cross-Over Trial.
A type of method used with research paradigms in which you start with 2 treatments, randomizing treatments into 2 groups, and after certain period of time, you cross them over.
Cross-Over.
- Both descriptive and analytic. - Assessment of a segment of the population at a single time point. - Conduct prevalence studies. - Limitations: Selection bias, recall bias, nonresponse bias, instrument bias, confounding and co-varying factors, and reciprocal factors.
Cross-Sectional Study Design.
A study in which you are taking a slice of the population and will do a one-time assessment such as a survey in which some questions might require you to think back in time ("In the past 6 months...").
Cross-Sectional Study.
A special type of clinical trial where subjects are moved from one treatment group to another during the treatment period. Subjects are in the control group for a period of time, allowed a wash-out time period and then given the other treatment. Most common subtype
Cross-over clinical trials
Cross-sectional studies
Descriptive study looking at a sample of the population to see how many have a particular outcome/disease & how many have a particular risk factor at a single instant in time. Examples: Prevalence studies (how many have it at a given time) Survey research Quality of life studies
What questions to Clinical trials answer?
Designed to: answer contributory cause. improve treatment of specific diseases.
What is the order of systematic steps?
Determine the question Categorize the question Select a specific resource Review and evaluate the information If necessary, try another resource
When you are analyzing and interpreting the literature, and you are assessing whether the study design is optimal for the intent of the research, which step are you performing?
Determining the Validity of the Research Design
A type of blinding that: - Group Allocation Concealed from Subjects. - Subject Group Allocation Concealed from Health Care Providers.
Double Blind.
A blinding mechanism in which both the subject and investigators do not know who is on what treatment.
Double-Blinded Study.
What type of blinding is most commonly used?
Double-Blinding.
What type of tertiary literature contains FDA approved drugs and investigational and orphan agents; legend and OTC listings; good quick-source; not referenced
Drug Facts and Comparisons
Which type of tertiary literature has an alphabetical listing of drugs by generic name with many charts and algorithms?
Drug Information Handbook
What 2 types of tertiary literature would be best for if you needed information on drug interactions?
Drug Interaction Facts and Hansten & Horn's Drug Interaction Analysis and Management
Which type of tertiary literature has NDC numbers and average wholesale prices of prescription and OTC drugs?
Drug Topics Red Book
Which tertiary literature is the best reference that discusses all aspects of drug dosing in pregnancy and lactation?
Drugs in Pregnancy and Lactation
Cause & effect occur together more frequently than by chance alone. *statistics to show relationship isn't random Cause must always precede the effect. **Cohort, randomized clinical trials Changing the cause (intervention), changes the effect.
Elements of Contributory Cause
In PICO, Outcome addresses
Endpoint of interest
What style is described below: The article by Stilwell and Harman is titled "I didn't pay her to teach me how to fix my back: A focused ethnographic study exploring chiropractor's and chiropractic patient's experiences and beliefs regarding exercise adherence".
Ethnographic Study.
description and interpretation of a cultural or social group or system
Ethnography
A type of naturalistic research that uses direct observation and extended field research to produce a thick, naturalistic description of a people and their culture. - Description and Interpretation of a cultural or social group or system.
Ethnography.
Clinical Research Study Questions want to answer questions in these four areas
Etiology Therapy Prognosis Diagnosis
What is the last component of PPAARE?
Evidence Use the best current available evidence relevant to the PPAAR components. Use higher levels of evidence to make decisions based on the hierarchy established on the quantitative (positivistic) and qualitative (naturalistic) pyramids.
Process of combining the best available research evidence with your knowledge/ skill to make collaborative, patient centered decisions within a given health care situation.
Evidence Based Practice
Summary EBM lecture 1
Evidence Based Practice is vital to improving patient outcomes. Creating structured questions is essential to searching for the best evidence available. EBM skills you develop here are important to your success both as a student and practitioner. There are many different research designs, but all are not created equal or capable of answering the question of interest.
A type of naturalistic research that is a type of qualitative study in which several different studies of the same phenomenon are brought together to develop theories, generate theoretical models, inform the design of further primary research, or evaluate health interventions.
Evidence Synthesis. - Forming analysis.
This type of evidence is Pre-populated, intended to be used at point of care, found in many databases
Evidence-based reviews
Treatment STudies
Examine if a drug or device improves the condition in terms of morbidity or mortality. Consist of: Interventional, prospective trials with the purpose of treating a diagnosed condition. RCT's Observational, prospective studies with the purpose of determining outcomes in patients with a diagnosed condition already receiving treatment. Cohort studies & systematic reviews/meta-analysis
- Randomized-Controlled Trial. - Often times unethical: Take group who isn't smoking, randomize this group into smokers and nonsmokers --> Unethical because smoking is addictive and detrimental.
Experimental Research.
What occurs in phase III of clinical trials?
Experimental treatment is given to large groups of people (1,000 to 3,000) to confirm its effectiveness, monitor side effects, compare it with commonly used treatments, and collect evidence allowing the experimental treatment to be used safely.
A style of quantitative research that involves dividing patients randomly, selecting subjects from the entire population or randomly assigning participants into intervention and control groups, and measuring differences between them or associations between predictor and response variables.
Experimental.
True/False: Larger, the sample, the worse the estimate.
False! *Larger the sample, the better the estimate.*
What would be an example of a prospective cohort study with regard to whether or not smoking causes lung cancer?
Find a group of smokers and nonsmokers, follow them for 20 years, and check what proportions developed cancer. - Having lung cancer (or not having it) is the outcome. - Cigarette smoking is the exposure.
Steps of the Scientific Process
Formulate a hypothesis Perform an experiment Control statistically for secondary factors Reason causation from the collected data
Prospective studies look where?
Forward
What are two examples of cohort studies?
Framingham Heart Study, MA= Primarily all White Study Jackson Heart Study, MS= All Black/African American Study
Why should both Relative risk reduction and Absolute risk reduction be reported to readers?
Relative risk reduction of a therapy overestimates impact of a therapy or treatment in comparison to absolute risk reduction.
The degree to which subjects in the sample match the population they represent. - Sample should match the population in terms of characteristics such as the distribution of people by sex, age, race/ethnicity, severity of disease, and so on. - Characteristics included in analysis are based on the nature of the study itself.
Representativeness.
What occurs in phase I of clinical trials?
Researchers test an experimental treatment in a small group of people (20 to 80) for the first time to evaluate its safety, determine a safe dosage range, and identify side effects.
- A retrospective study design ~ start with outcomes and examine differences in exposure. - Can be categorized as both observational and retrospective. - Clinicians are interested in retrospective research because it helps identify risk factors with disease (or health-related behaviors). - Subject selection could be problematic ~ patient charts, which are common data source in retrospective research, can be fraught with errors and omissions.
Retrospective Research (Case-Control Study). - Cheaper, easier.
A study in which you go back such as using chart reviews; Looking at data that's already been collected.
Retrospective.
Formulate a focused question based off of this PPAARE: 1. Problem a. Increase in mental and physical exhaustion 2. Patient a. Caregivers for geriatric patients 3. Action a. Community intervention program 4. Alternative Action (not a required component) a. None 5. Result a. Decrease in mental and physical exhaustion
What is the highest level of evidence available to determine whether a population with an increase in mental and physical exhaustion exhibiting caregivers of geriatric patients will benefit from a community intervention program to decrease the mental and physical exhaustion?
Similar to publication bias in that it can lead to a study not being published for reasons than the quality and relevance of the research.
Funding Bias.
What type tertiary literature contains general principles of action, absorption, distribution, and metabolism?
Goodman and Gillman's Pharmacological Basis of Therapeutics
A type of naturalistic research that is the study of abstract problems and their process. - Generalizations are inferred from the collected data. - The research product constitutes a theoretical formulation or integrated set of conceptual hypotheses about the substantive area under study.
Grounded Theory.
What type of tertiary literature has an alpha listing of drugs and compatibility with other drug products and fluids?
Guide to Parenteral Admixture
What type of tertiary literature is a well-known pocket guide to geriatrics?
Handbook of Geriatric Drug Therapy
Which type of tertiary literature would be the best source for over the counter products?
Handbook of Nonprescription Drugs
What type of tertiary literature cover parenteral infusions, drug stability, and compatibility?
Handbook on Injectable Drugs
Which resource would you use when inquiring about drug interactions?
Hansten and Horn's Drug Interaction Analysis and Management
A mother wants to give her 8-month old child Tylenol for a fever. Which resource is the most appropiate to use to find the correct dose for the child?
Harriet Lane Handbook
What type of tertiary literature is a well-known pocket guide to pediatrics?
Harriet Lane Handbook
A block sections the entire experimental units into?
Homogeneous Groups.
When is a drug considered ready for the market?
If it passes phase III of clinical trials.
Principles of the Null Hypothesis
If you start with the assertion that no association exists, stats allow estimation of the observations occurring by chance alone. You can either accept or reject the null hypothesis. If you reject the null hypothesis, you accept the alternative hypothesis by default.
Why are prospective studies important/beneficial?
Important because you have more control over it because you're designing it. Thus, you have a say in what you're going to collect, how you'll collect it, and how you'll analyze it.
Randomized Controlled Trials
In RCT's patients are randomly (think coin-flip) assigned into experimental & control groups, then followed up for the outcomes of interest. Randomization attempts to avoid possibility of selection differences or bias. RCT's are the gold standard of primary research study design for questions on therapy!
In case-control studies, do we start with the outcome or exposures?
In case-control studies, we start with the outcome and investigate the exposure.
In cohort studies, do we start with the outcome or exposures?
In cohort studies, we start with exposures and investigate the outcomes.
WRITE A PICO: A 75 year old woman with chronic venous stasis presents to your clinic with a large stasis ulcer on her left shin. After your exam, you recommend Hydrogel dressings. The patient mentions a friend who was treated in the UK with maggots for the same problem.
In elderly patients with chronic venous stasis ulcers, compare larval (maggot) therapy with hydrogel dressing in ulcer healing. (are maggots better than hydrogel dressings) for ulcer healing time.
Summary of Inf stats
In treatment studies, readers should be attentive to the stated primary outcomes, differences b/t groups and alignment. Evaluation of confidence intervals allow reader a fuller understanding of true differences between groups. Absolute risk, attributable risk, relative risk and odds ratios are commonly used terms that readers should become familiar with in evaluating studies of therapy, harm and prevention.
Lower standard deviations will ____ power
Increase Standard deviation: SD ↓'s, power ↑'s.
An increase in effect size ____ power
Increases Effect size: ↑ ES= ↑power.
Increase in level of significance (alpha), ____ power
Increases Level of significance α: ↑ α = ↑ power.
It usually represents a major methodology flaw in the Quantitative study if the ____ and ______ cannot be identified by a casual reading.
Independent, Dependent Variables (cause, effect)
What do databases provide?
Indexing and abstracting services
What is an advantage of tertiary literature?
Information has already been compiled
What are some applications of ethical principles?
Informed consent Assessment of risks and benefits- Investigators must perform a systematic assessment of risks and benefits Selection of subjects- must be fair
Out of the 4 clinical trials, which stages are considered randomized-controlled trials?
Stages 1, 2, 3, of the 4 clinical trials are as such.
SEM=
Standard error of the mean is the standard deviation of those sample means over all possible samples (of a given size) drawn from the population.
What are the steps for conducting critical appraisal?
Step 1: Determine the paradigm and style of research conducted Step 2: Complete the Critical Appraisal form that goes with the type of study.
Scatter diagrams represent what?
Strength and direction of relationship (correlation!)
R= -0.75 indicates what kind of correlation
Strong NEGATIVE association
What are mixed studies?
Studies involving quantitative & qualitative research paradigms.
Qualitative (Naturalistic) Quantitative (Positivistic) Mixed-method
Study Paradigms
What is the group assignment of systematic review?
Study results are weighted with respect to methodologic quality.
Does the Qualitative Research paradigm have a subjective or objective perspective?
Subjective!
Using one outcome to reflect another. Selected based on the association of a physiologic or biologic measure with another known clinical end point.
Surrogate Outcomes
What are the two outcomes of care?
Surrogate and outcomes that matter
What was the surrogate and outcomes that matter in the Accord Trial?
Surrogate- Blood glucose, HbA1c OTM- cardiovascular events and mortality
What was the surrogate and outcomes that matter in CAST?
Surrogate- asymptomatic arrhythmias Outcomes that matter- mortality after myocardial infarction
What is the goal of a systematic review?
Synthesize findings from individual studies to provide a coherent, integrated answer to a focused research question.
A research procedure in which all prior studies on a given topic are brought together and analyzed collectively. - Deemed to be near the top of the evidence pyramid, but should be treated with caution due to possible bias. - Especially valuable in comparative effectiveness research.
Systematic Review.
A review where we pool data from different studies and get the big picture.
Systematic Review.
True/False: SR should follow a strict protocol that is developed and published before the study is conducted.
Systematic Reviews.
What are examples of naturalistic research?
•Enthnography •Biography •Phenomenology •Grounded theory •Case study
What are examples of positivistic research?
•Survey study •Cohort study •Case-control study •RCT •Systematic review •Meta-analysis
An increase in sample size leads to ___ in power
↑in SS = ↑power.
In Medline /Pubmed searches on studies of harm, use theseterms or subheadings:
Interactions, adverse events, side effects, harm, safety or tolerability.
What is the fundamental question for both studies to evaluate them?
Is the information provided in the publication trustworthy?
What is the fundamental question for both qualitative and quantitative types of studies?
Is the information provided in the publication trustworthy? - Note that characteristics of trustworthiness differ b/w the two.
Differences- define
Is the length of stay in the intensive care unit (ICU) shorter for patients with commercial insurance than for patients with public insurance? Analysis- Test to determine length of stay between the two groups.
What is the scientific method?
Is the process of formulating a hypothesis, performing objective experiments, and engaging in sound reasoning supported by the collected data.
Correlation- define
Is there a relationship between a patient's age and the time spent in a hospital after a heart attack (MI)? Analysis- correlation coefficient
Why is qualitative research an essential component of health services research?
It enables us to reach areas not amendable to quantitative research.
What is the defining characteristic of prospective research?
Its forward orientation in time.
A nurse manager calls the pharmacy wanting to know if IV furosemide may be injected into Y-site IV line currently infusing dobutamine? Which of the following references would you most likely consult?
King Guide to Parenteral Admixture
A type of bias in which numerical finding of significance does not indicate that a difference between groups was clinically meaningful.
Lack of clinical value or practical significance.
Disadvantage of Case control/ case studies/ case reports?
Less validity/ generalizability for application to patients
What would be quasi-experimental, cross-sectional studies?
Longitudinal cross-sectional studies.
What is the THEME of advantages of surrogate outcomes?
MONEY!!!
Why is it important to consider both the p-vaue and r or r2 statistic when evaluating correlations?
"It is not uncommon for statistically significant correlations to be found in data that are in fact not strongly associated." "The p-value may indicate a statistically significant result even when the r value is not indicative of a strong association."
Define: Outcomes (of care)
"Measurable or observable results of illness or treatment"
Minimal (clinically) important difference
"The Minimal Important Difference (MID) is "the smallest difference in score in the domain of interest that patients perceive as important, either beneficial or harmful, and which would lead the clinician to consider a change in the patient's management".1 MID quantify important changes in both subjective and objective measures based o n therapies. MCID= MID
Prevalence
% o people in a population with a disease
Temporal Perspectives
(time) Prospective Retrospective Cross-sectional
NIH Phasees of Clinical Trials
- Phase I: small groups (usually healthy volunteers) tested to determine safety, dosing and side effects - Phase II: larger group (usually condition of interest) tested to determine efficacy and further safety - Phase III: larger groups to confirm effectiveness, monitor SE's, compare to control group, and determine use. - Phase IV: Post-marketing studies for additional info on long term safety and efficacy.
What are the three types of blinding?
- Placebo Effect. - Single-Blinded Study. - Double-Blinded Study.
A rule of thumb employed with correlation tests is that an r of at least_______is considered a strong positive association (whereas ______ would be a strong negative association). A so-called weak relationship would be at an r of _____ or less.
- r=at least 0.75= strong - (-0.75)= strong negative - 0.160 = weak
Relative Risk (RR)
- ratio of the risk of an event among the exposed to the risk among the unexposed. - Tells you if the risk causes or prevents the outcome. Reported commonly in: RCT for therapies. Cohort studies on risk factors. Any studies with harm endpoints.
Pearson Product Moment Correlation Coefficient
- symbolized by r - RELATIONSHIP between 2 interval, continuous or ratio scale variables - most commonly used correlation test statistic
An ANOVA test produces:
-F statistic resulting from a comparison of means between two or more groups
In an RCT, how do you tell if randomization has been successful?
-See that different treatment groups have same characteristics at baseline. - eg- same number of men and women, or older or younger people, or degree of disease severity. *At beginning: must have equal likelihood of ending up in either group. Groups should end up equal in all variables. Eg- CAD study, if all diabetics or smokers were in the same group= effect outcome
A Chi-square test produces:
-a χ² statistic resulting from several different types of tests
absolute Risk
Absolute risk: Event rate or incidence rate of a group exposed to a risk/intervention. i.e. Incidence of lung cancer in a middle-aged smoker. (ARI and ARR)
In primary literature, the ______ section of the article provides a brief overview of the study and includes information on the methods used. results found, and conclusions.
Abstract
- Use cumulating evidence from the trials to modify the trial design in the conduct of the trial to increase the efficiency and benefit to trial participants. - Often reemployed with pragmatic clinical research and comparative effectiveness research. - Based on *Bayesian Statistical Inference*, where prior information are considered and incorporated in the analysis of the data derived from the trials.
Adaptive Research.
Hypothesis testing refers to:
After a descriptive review of the data and determining the level of measurement, the researcher again asks the original question: "What is the hypothesis?
Stats can tell us what about the null hypothesis?
All we can know from stats is the probability that the null hypothesis was falsely rejected.
Explain Qualitative Research Paradigm
An inquiry process of understanding based on distinct methodological traditions that explore social or human problems. Qualitative researchers build a rich complex, holistic picture, analyze words, report detailed views of the informants, and conduct the study in a natural setting. Subjective rather than objective perspective.
What is a cross-sectional study design?
Assessment of a segment of the population at a single time point. Example: People who have/had cancer, Do/did they smoke versus People who never had cancer, do/did they smoke
Outcomes Measures (look for these first!)
Authors should explicitly state the primary and secondary outcome measures and they should align across those sections. You should find them from the article or the abstract. Look at: Title of the article. Purpose statement. Background section. Results table. Conclusion/discussion section. - Results section should include an endpoint for each outcome measure.
Number needed to treat
Average number of patients needed to be treated to prevent one more adverse event. 1/absolute risk reduction. (or produce one more outcome of difference) eg- Crohn's gets better
What type of tertiary literature is the primary reference for many pharmacology courses?
Basic and Clinical Pharmacology
Adaptive trials are often referred to as?
Bayesian Adaptive Trials.
A statistical approach used with research paradigms where you can't ignore prior information. Thus, you assume certain things about the person; you use probabilistic methods to help analyze the data.
Bayesian.
What can limitations lead to in the outcomes of a study?
Bias.
study of an individual and his or her experiences as told to the researcher
Biography
A type of naturalistic research that is the study of an individual and his/her experiences as told to the researcher or found in documents and archival materials.
Biography.
Causation
Biomedical research tries to prove relationship between a cause, usually a risk factor or treatment, & the effect, a desired outcome.
- Study design where treatment group allocation is concealed from the participants, investigators/providers and/or assessors. - Aims to minimize observer and participant bias
Blinded Clnical Trials
The technique used to increase the *precision* of an experiment ~ a means to reduce and control experimental error variance to achieve more precision.
Blocking.
What are some examples of resources for clinical practice guidelines?
ClinicalKey Guidelines (Elsevier Inc) CMA Infobase: Clinical Practice Guidelines (Canadian Medical Association) Essential Evidence Plus Guidelines and EBMG evidence summaries National Guideline Clearinghouse (Agency for Healthcare Research & Quality) NICE Guidance (National Institute for Health & Clinical Evidence, UK)
What is an outcome that matters?
Clinically relevant outcomes that provide direct measures of disease, including outcomes that patients and providers care about, such as the patient's ability to function or the cost of care
Which resource would you use when trying to identify a foreign drug?
Martinadale's
Which type of tertiary literature contains foreign drug information?
Martindale: The Complete Drug Reference
What's the problem with too large of a sample?
May find differences that aren't really there
A type of bias where bias is introduced in any procedure used for quantifying information (Counting the number of fall incidents in a hospital ward, weighing patients, and measuring cholesterol levels).
Measurement Bias.
Can include inaccurate measurement tools, calculation errors, mistakes in recording measurements, participant bias, recall bias, and more.
Measurement Bias.
What are the 5 types of secondary literature we talked about?
Medline International Pharmaceutical Abstracts The Medical Letter OVID The Cochrane Library
These types of study represent the BEST available evidence, highest quality, but also smallest amount of information
Meta Analyses and Systematic Reviews (Cochrane rev)
Example of gap between best research and clinical practice
Meta Analysis proved Flomax didn't work.. still prescribed!
Which type of secondary research uses quantitative mathematical methods to summarize the group results of a systematic review. (systematic review + math)
Meta- Analysis!
Similar to systematic review, but we're now trying to find a global estimate. - A lot of studies might not have used big enough sample size and thus can't be trusted. So, we pool studies together and develop an estimate. - It's using *math and statistics* to help us come up with estimates we initially didn't feel comfortable with making.
Meta-Analysis.
An added step to a systematic review in which a statistical analysis is performed to quantify the findings of the review.
Meta-analysis.
Which tertiary literature would be best for looking up side effces associated with drug therapy?
Meyler's Side Effects of Drugs
What type of tertiary literature is an online pharmacology database which has a vast array of information dealing with pharmacology, therapeutics, poison information, etc.
Micromedex
Which resource would you use when inquiring about adverse effects?
Micromedex
Koch's postulates don't work well for what?
Modern diseases with multiple causes
Outcomes that Matter Include:
Morbidity Mortality Quality of Life Pain Cost
- Developed as treatment modality that identities and utilizes the intrinsic motivation of individuals with substance-related problems. - Offers strategy for providers to collaborate with patients to identify their health-related goals and make self-directed changes toward achieving those goals.
Motivational Interviewing.
Is a controlled trial a randomized-controlled trial?
No, it does not have randomization of participants to groups, and it often cannot be concealed.
Levels of Measurement in Inferential Stats
Nominal Ordinal Interval/Ratio
Non Parametric data- define
Nominal or ordinal scale Non-normal shape Small sample size Non-equivalent variability across groups **Data are non-parametric if ANY of these characteristics are present
Can be acceptable for certain types of studies, such as proof-of-concept studies, cohort studies, and others.
Nonrandom Sampling.
Parametric data is _____, while non-paramedic data is _____
Normative, non-normative
What's the problem with too small of a sample?
Not representative
Does the Quantitative Research paradigm take an objective or subjective view?
OBJECTIVE
What does the Cochrane Collaboration explanation of Systematic Reviews state?
Systematic reviews are studies that collate all evidence that fits pre-specified eligibility criteria in order to address a specific research question and minimize bias by using explicit, systematic methods.
Explain Quantitative Research Paradigm
Takes an objective view vs. subjective - Reality is external and measurable - "Reality" is measured using the scientific method: (Formulate a hypothesis Perform an experiment Control statistically for secondary factors Reason causation from the collected data) - Majority of biomedical research follows a quantitative paradigm.
How are differences analyzed?
Test of differences
What is the assumption in classifying clinical questions (especially in the novice?)
That all background information is correct@
Formulate a focused question based off of this PPAARE: 1. Problem= HPV vaccine 2. Patient= 15-year old male 3. Action= Provide the vaccine 4. Alternative =Not get the vaccine 5. Result = Reduce risk of getting HPV
What is the highest level of evidence available to determine whether the Human Papilloma Virus (HPV) vaccine should be administered to a person who is a 15-year old male, benefits from receiving the vaccine compared with not receiving the vaccine to reduce the risk of spreading HPV to others?
What is tertiary literature?
The general literature; has already been compiled and reviewed for you; textbooks, dictionaries
Study Approaches
Observational Quasi-experimental Experimental/Interventional
- A type of method used by case-control and cohort studies. - You're not intervening but instead just observing, trying to see what happened.
Observational.
- A type of quantitative research that quantifies phenomena but does not involve the use of an intervention. - Applies the same rigor as experimental studies.
Observational.
Odds Ratio
Odds ratio (OR): ratio of the odds of an event occurring in one group to the odds of an event occurring in another group. Most commonly seen in epidemiological and harm studies, mainly case-control studies & meta-analyses. Its is a less reliable measure than relative risk. Odds ratio best estimates the RR when a disease is rare.
Odds ratio interpretation
Odds ratio =1 means that exposure is not associated with disease/outcome. Odds ratio > 1 means that exposure is associated with disease/outcome. Odds ratio< 1 exposure means it provides protection from disease/outcome.
Odds ratio calculation
Odds ratio= (a/c)/(b/d) Or by cross multiplying= AD/BC A= Disease +, Risk+ B= Disease -, Risk + C= Disease present, risk absent D= Disease and risk absent
What are studies where there is no blinding?
Open Label Trials (Cancer, chemo).
A type of bias that includes the motivation for a study to be conducted or to be published (Or not published, as the case may be).
Other Source of Bias.
Loss of Toe(s) Due to Peripheral Artery Disease
Outcome that matters
Loss of Vision Due to Diabetic Retinopathy?
Outcome that matters
Clinically relevant outcomes that provide direct measures of disease, including outcomes that patients and providers care about, such as the patient's ability to function or the cost of care.
Outcomes that Matter
How is data collected in a prospective research?
Over a defined period of time, of sufficient duration to draw reliable inferences based on the sample.
What does PPAARE stand for?
P- problem P- patient/ population A- action A- alternative R- result E- evidence
Research Categorization Criteria (4)
Paradigm Approach Temporal Perspectives Study Design
A type of method used with research paradigms in which you start with 2 groups, stay on same treatment, and evaluate which drug does the best job.
Parallel Groups.
Treatment Studies Considerations
When assessing treatment studies look at: Outcome measures Minimal Clinically Important Differences Efficacy including confidence intervals Description of treatment modality Indications and Contraindications Side effects and complications Interactions Administration & Dosage Adherence Cost and Access
PICO format
Patient Intervention Comparison Outcome
What is patient-centered interprofessional practcie?
The interaction among the patient/population individual practitioners collaborative team members
What would be a solution to a sample size issue?
Perform a sample size determination or power analysis prior to conducting study.
What is a way a pharmacist could be useful with regard to washout timeframe?
Pharmacist could look at half-life of drug.
What would be an example of the method of cross-over used with research paradigms?
Pharmacists start a drug trial, wait for some time, and then switch patient's treatment.
Which styles of qualitative research are used more in healthcare?
Phenomenology and ethnography are most commonly used Grounded theory less frequently used Semi—structured interviews and focus groups are more widely used in healthcare-related qualitative research
A type of naturalistic research that is the belief that the object of interest be examined without any preconceived notions or expectations. - The study of lived experiences of several individuals centered on a single __________.
Phenomenology.
Studying phenomenon; Typically qualitative in which we interview people and tie themes, stores together. - Look for themes & domains that explain why phenomena are occurring.
Phenomenology.
Surrogate Outcomes include:
Physiologic Biologic Laboratory
A blinding mechanism that addresses psychological effect of treatment and tests the efficacy of treatment.
Placebo Effect.
Which paradigm's role of researcher is no or minimal engagement with participants and phenomena; Subjects and instruments are the sources of data, not researchers?
Positive/Quantitative.
Holds that reality is both external and objective. - Reality can be measured by systematic observations of nature -- a process referred to as Scientific Method.
Positivism.
What type of study involves the computation of numerical values?
Positivistic Study.
Is Positivistic objective or subjective? What about Naturalistic?
Positivistic research is objective whereas naturalistic is subjective.
Which paradigm's analysis is descriptive and/or inferential?
Positivistic/Quantitative.
Which paradigm's data is numerical, measurable, objective data collected from subjects only?
Positivistic/Quantitative.
Which paradigm's design is fixed, predetermined, utilizing one or more intervention?
Positivistic/Quantitative.
Which paradigm's reasoning is typically deductive, seeking predictors?
Positivistic/Quantitative.
Which paradigm's sampling is large, preferably random samples blinded to the intervention?
Positivistic/Quantitative.
Which paradigm's setting is laboratory or controlled by design of research?
Positivistic/Quantitative.
Which paradigm has focused questions?
Positivistic/Quantitative. - Do you consider yourself being food-insecure? Yes, No? Not sure?
Where would the location and style be in qualitative research?
Possibly in the title or methods section of publication.
What occurs in phase IV of clinical trials?
Post marketing studies delineate additional information, including the treatment's risks, benefits, and optimal use.
Calculate power
Power=1-B β should be < 0.20. Power should be >or equal to 0.80.
An original investigation undertaken in order to gain new knowledge partly by means of practice and the outcomes of that practice.
Practice-Based Research.
Research designed for the primary purpose of informing decision-makers regarding the comparative balance of benefits, burdens, and risks of a biomedical or behavioral health intervention at the individual or population level.
Pragmatic Clinic Trial.
A type of method used with research paradigms that involves people coming in, we assess them, we apply an intervention, and we then assess them again.
Pre-/Post-Interventional. - Randomized-Controlled trial uses this type of method.
CIs measure what?
Precision!
What would be observation, cross-sectional studies?
Prevalence studies surveys.
Prevention Studies
Prevention: actions aimed at eradicating, eliminating or minimizing the impact of disease and disability. Consists of: Observational studies with the purpose of determining whether patients develop a condition. Cohort studies Systematic reviews/meta-analysis of cohort studies Interventional, prospective studies with the purpose of preventing patients from developing a condition. RCT's and cohort studies
Types of Prevention
Primary Prevention Secondary Prevention Tertiary Prevention Vaccines?= primary Colon cancer screening? = secondary Mammograms?= secondary (dysplastic) Cardiac or stroke rehab? = tertiary (ameliorate) Seatbelt laws?= primary Aspirin to prevent further MI/stroke?= primary (never had)- prevent CAD. Secondary= already had one, preventing another Chronic disease management? = tertiary?
Practice-based research networks (PBRNs) funded by the Agency for Healthcare Research and Quality (AHRQ) have focused on?
Primary care practices -- working together to answer community-based health care questions and translate research findings into practice.
A single-group study in which the plausibility of an underlying principle is tested.
Proof-of-concept experiment.
What would be experimental, cross-sectional studies?
Proof-of-concept experiments.
What are the three perspective in research?
Prospective Retrospective Cross-sectional
- Involves the formation of a hypothesis, followed by the collection of data and subsequent analysis of findings.
Prospective Research.
A study in which you start now and go forward with the study; You design a study and collect data going forward.
Prospective Study.
What are some examples of databases?
PubMed Cochrane Library CINAHL (Cumulative Index of Nursing & Allied Health Literature) OVID
Occurs when the determination of whether to publish a study is bade on factors other than the quality of the research and the relevance of its findings.
Publication Bias.
For consumers of research information, what is the key to detecting bias?
The key to detecting bias is consideration of all of the possible causes of an outcome reported in a study.
What is evidence based practice?
The process of combining the best available research evidence with your knowledge and skill to make collaborative, patient- or population-centered decisions within the context of a given healthcare situation.
Blindedness- define
The process used in clinical trials where treatment group allocation is concealed from the participants, investigators/providers and/or assessors.
Coefficient of Determination (r2)
The proportion of variability in the response variable that is predicted by variability in the predictor variable. A correlation coefficient of 0.755 would produce an r2 of 0.570, which means that a correlation coefficient of 0.755 represents a coefficient of determination of 0.570 (57% of the variation in the response variable is explained by variation in the predictor variable). If an r is reported without an r2, look for an explanation about why the r2 was not an appropriate statistic for the data.
Confidence interval
The range in which the true population parameter can most likely be found
What is the principal procedural foundation of quantitative research?
The scientific method.
What are the two research paradigms?
Qualitative and Quantitative
Studying the interaction between patients and providers (Examine differences and similarities), examining barriers to practice change, examining social and cultural factors that influence health care are all applications of which paradigm of research?
Qualitative.
The majority of biomedical research follows a ___ paradigm
Quantitative
This research paradigm says that 'Reality is external, and measurable'
Quantitative
Which research paradigm uses the scientific method to measure reality
Quantitative
- Two famous examples of studies: 1. Framingham Heart Study, MA (Primarily all white study). 2. Jackson Heart Study, MS (All Black study). - Inherently limited in terms of generalizability of findings to general population i.e. looking at specific group. - Can examine multiple exposures (Independent variables) and outcomes (Dependent variables). - Can be retrospective or prospective.
Quasi-Experimental Research (Cohort Study).
A style of quantitative research that involves studying a phenomenon in which researchers cannot randomly select subjects or randomly assign subjects to treatment and control groups; However, researchers are able to control some independent variables.
Quasi-Experimental.
What kind of study are the 'gold standard' of primary research design for questions on THERAPY?
RCTs!
What is a way to achieve representativeness?
Random Sampling.
Blinding can effect what?
Randomization - limits which intervention participants are randomized to
In RCTs, how do we attempt to avoid possibility of selection differences or bias?
Randomization!
- The cornerstone of underlying the use of statistically methods in experimental design ~ draw samples of a known population within limits of a sampling error. - Assists in "averaging out" the effects of extraneous factors that may be present ~ create comparable groups.
Randomization.
What falls under experimental research?
Randomized Controlled Studies (RCTs)
Patients randomly assigned to experimental and control groups, then followed up for outcomes of interest.
Randomized controlled trials
What is a type of experimental style?
Randomized-Controlled Trials.
What is grounded theory?
The study of abstract problems and their process.
What is ethnography?
The use of direct observation and extended field research to produce a thick, naturalistic description of a people and their culture.
The test statistic is
The value resulting from a statistical hypothesis test (t, F, x2)
The period of time in which you wait before switching treatments in the cross-over method is called?
The wash-out period (You have to know the half-life of a drug).
In PICO, Comparison addresses
Therapy, exposure or test (standard) you will compare the intervention (novel) against
In PICO, Intervention addresses
Therapy, exposure, or test you will apply to the patient ***NOVEL therapy here
Why Evidence-based practice?
There is a substantial gap between the best research evidence and clinical practice with potentially serious health consequences for patients. Evidence based practice helps bridge that gap by providing strategies that facilitate finding and appraising research, and resources that pre-filter and synthesize research to make the process more efficient. Need to know best practices in real time. Traditional information often inadequate. Practice changing research is often ignored. Inefficient practice is harmful, costly or ineffective.
Are extraneous variables a type of bias?
Yes, extraneous variables are a type of bias.
When interviewing, how do you know you've done enough interviewing?
You want to reach the point of "information saturation" in which you begin to hear the same thing over and over again.
Cross-sectional studies focus on
a single point in time
The point estimate is
a single value (statistic) measured from a sample (mean, proportion)
Cochrane Reviews are
a systematic review- the best!
R less than or = 0.160 indicates what kind of correlation
WEAK
What would be an example of mixed research paradigms?
Wanting to know the determinants and factors that cause food insecurity. - Give the people a survey i.e. quantitative. - Talk to some of the people i.e. qualitative (Narrative analysis).
Correlations include:
What is a positive association? What is a negative association? How much correlation is sufficient? Correlation is not Causation. What out for anomalous associations.
Formulate a focused question based off of this PPAARE: Problem- Rheumatoid Arthritis Patient- 48 year old male Action- Gain perspective of having this disease Alternative- none Result- Help patient prepare for challenges
What is the highest level of evidence available to determine the experiences or perceptions of an adult male with Rheumatoid Arthritis to gain knowledge to assist and prepare them for the challenges of living with this disease?
Formulate a focused question based off of this PPAARE: 1.Problem a. COVID19 Outbreak 2. Patient a. Louisiana residents 3. Action a. Determining protocol 4. Alternative Action 5. Result a. Stop spread of COVID19
What is the highest level of evidence available to determine whether a population in the COVID-19 outbreak from Louisiana benefits from determining better protocol to stop the spread of COVID19?
Power (define) - Microscope analogy
ability to detect a statistically significant difference when one exists. - Microscopes with GREATER power can detect smaller differences n cells
Number needed to harm
average number of patients needed to be treated in order to produce one more adverse event. 1/absolute risk increase. (eg- develop HepB)
Number needed to screen
average number of persons who must undergo screening or diagnostic testing in order to prevent one case of disease. 1/absolute risk reduction.
Longitudinal Studies
control group 1. Interventional - Clinical trials 2. Observational - Case control studies (retrospective usually) - Cohort studies
Examples of Quantitative study designs
control trial, cohort study, RCT, cross-over trial, systematic review, meta-analysis, clinical trial
Examples of Qualitative Study Design
ethnography, biography, phenomenology, case study, grounded theory.
What is gray literature?
evidence not published through commercial sources (academic papers, conference papers, dissertations and theses, ongoing research, government reports, and committee reports)
What is secondary literature?
interpretation, synthesis, evaluation of the primary literature; systematic reviews and meta-analysis; compiled by indexing and abstracting services
activity intended to alter an outcome, such as a risk reduction strategy
intervention
What is evidence synthesis?
is a type of qualitative study in which several different studies of the same phenomenon are brought together to develop theories, generate theoretical models, inform the design of further primary research, or evaluate health interventions.
What is a case study?
is an exploration of a bounded system (a case or multiple cases) over time through detailed, in-depth data collection involving multiple sources of information rich in context.
A wide CI indicates
lack of certainty
Barriers to practicing EBM
lack of personal time personal & organizational inertia problems with the 'evidence' lack of evidence/unrelated evidence attitudes of colleagues patients' expectations
We generally accept results when the probability that the null hypothesis was falsely rejected is equal to:
less than 1 in 20 times: p< 0.05.
A study of the same group of people at more than one point in time. Case-control studies Cohort studies Clinical trials
longitudinal studies
Retrospective studies
look back in time
Hazard ratio
measures how often a particular adverse event occurs in one group compared to another over time. It is a commonly reported type of relative risk calculation. It describes the outcome, usually in therapeutic trials, where the question is, "To what extent does treatment shorten the duration of the illness?" It is interpreted the same as relative risks So the distinguishing characteristic of a hazard ratio is?
statistics applied to medical research
medical statistics
What do search engines provide?
meta-search engines are useful to practitioners for EBP
combining qualitative and quantitative perspectives
mixed-method research
A narrow CI indicates
more certainty
treatment modality that identifies and utilizes the intrinsic motivation of individuals with substance-related problems
motivational interviewing
Descriptive Studies
no control group- what happened - Case reports - Case series - Cross-sectional studies
Type 2 Error
null hypothesis is not rejected even though it is really false false negative result. Type 2 errors only happen in negative clinical trials.
What are some strategies for when you find relevant evidence?
o Bookmark the evidence or add evidence to a folder to promote easy access. o Create an Alert or Update notification to your email when new information is added. o Determine when the evidence was last updated. o Determine whether the resource identifies the evidence used in the update to ensure the information is current. o Determine whether recommendations or guidelines are graded or rated on: o Strength (e.g., weak or strong) o Quality of evidence (e.g., low, moderate, high)
Name some search strategies.
o Learn how to use the basic and advanced search features. o Use the autofill feature to locate words or phrases relevant to PPAARE components. o Determine the organization of health information in the resource. o Use hyperlinks to access additional relevant information including pictures, diagrams, illustrations, videos, algorithms, calculators, patient education materials, etc.
What are some supplemental clinical databases?
o MicroMedex o Lexi-Comp o National Center for Complementary and Alternative Medicine (NIH) o Natural Medicine Comprehensive Database
What is primary literature?
original research, the method how it was done; most current; found in biomedical journals and peer-reviewed
Self-report of excessive urination
outcome that matters
Stroke
outcome that matters
heart attack
outcome that matters
A significant p value tells us:
p,0.05, tells us that the probability the null hypothesis was falsely rejected is less than 1 in 20
What is subjective data?
patient reports; chief complain; assesses an outcome that matters
the study of the lived experiences of several individuals centered on a single phenomenon
phenomenology
proportion of the population with a given disease at a single point in time
prevalence
The P value is equal to
probability of obtaining the observed result by chance if the null hypothesis is true.
What does a t-test produce?
produces a t statistic resulting from a comparison of means between two groups
What do synthesized summaries provide?
provide practitioners with a high level of evidence while at the time of delivering clinical care.
2 types of study design
qualitative, quantitative
Which study paradigm do we largely work within
quantitative/positivistic
Incidence
rate of new disease cases over time
What is the principal procedural foundation of quantitative research?
scientific method
Indexing and abstracting represent
secondary literature sources
An r = 0.75 or greater indicates what kind of correlation
strong positive association
Body mass index
surrogate
High blood pressure
surrogate
asymptomatic arryrthmia shown on the ECG
surrogate
fasting plasma glucose
surrogate
What are primary outcomes?
the measure(s) that researchers choose as the one(s) to influence with the intervention, as the most important reflection of the condition of interest
What are secondary outcomes?
the measure(s) that researchers identify as also being influenced by the intervention but are either not most important or are not intentionally influenced by the intervention
grouped according to their time orientation
temporal style
Controlled Clinical Trials
test a specific treatment involving 2 or more groups of patients with the same disease. The experimental group receives the treatment that is being tested & the other (comparison or control group) receives alternative treatment, placebo or no treatment. The groups are compared for differences in outcomes to determine effectiveness of the experimental group. *2 groups *Intervention compared to control
What is alpha?
the probability of a Type I error. The risk of finding a statistically significant difference when in fact one does not truly exist. α is set at 0.05 usually.
The p-value is:
the probability that the test statistic represents normal variability
What is Naturalistic or Qualitative evidence?
the researcher's goal is to gain a deep understanding of the lived experiences of individuals or groups and to develop a rich, thick description of these experiences. Naturalistic re-searchers attempt to understand the meaning of experiences according to those who live them.
What is a biography?
the study of an individual and his/her experiences as told to the researcher or found in documents and archival materials.
What is the general rule for quantitative research?
thought as objective as compared to naturalistic research, which is thought to be subjective.
What are the limitations to qualitative research?
time consuming and difficult to share data
When you think about critically appraising the evidence you are concerned with assessing..
validity, clinical importance, and applicability
In PICO, Patient addresses
who/what population is the question about?
A case report can be helpful for studying what?
Rare/emerging diseases
What is the most mathematically useful level of measurement?
Ratio/Interval
A type of measurement bias that is pervasive in case-control studies and, to some degree, in most observational studies.
Recall Bias.
A bias in which a person is not able to remember things vividly.
Recall bias.
general characteristics of descriptive studies
Record events. Look for particular characteristics of the patients studied.
A patient calls your pharmacy and asks if Prozac comes in liquid form, if so, how much would it cost. Which resource is the most appropriate?
Red Book
What are quantitative research methods meant to reduce?
Reduce potential for errors through sufficient sample sizes, representative samples, adequate duration, and objective analysis.
What are two things "control" implies?
- Group selection or assignment. - Research procedure.
What are three different styles of quantitative research?
- Observational. - Quasi-Experimental. - Experimental.
What do you need to look at to evaluate correlations/ association strength?
- P values AND r or r2 statistic
What are limitations for quantitative research?
- Generalizability -- ability to generalize findings from the sample to the target.
Studies of Harm
*Etiology Examine: Natural course of a untreated disease, harm from complications or interactions of treatment, or exposure to a toxin. Consist of: Observational or interventional prospective studies with the purpose of identifying harmful exposures. Cohort studies Systematic reviews/meta-analysis of Cohort studies RCT's Observational, retrospective studies with the purpose of identifying harmful exposures. Case-control studies
Relative Risk Increase
- (RRI) - another way of reporting risk. - Also reported with a CI and p value
How many questions does the NIH divide clinical trials into and how many phases?
- 5 types of questions. - 4 phases.
Commonly Reported Measures of Efficacy in Treatment, Harm and Prevention Studies
- Absolute Risk (ARR, ARI) - Attributable Risk - Relative Risk (RR, RRR, RRI)
What are three important characteristics of a pragmatic clinic trial?
- An intent to inform decision-makers (Patients, clinicians, administrators, policy makers), as opposed to elucidating a biological or social mechanism. - An intent to enroll a population relevant to the decision in practice and representative of the patients or populations and clinical settings for whom the decision is relevant. - Either an intent to (a) streamline procedures and data collection so that the trial can focus on adequate power for informing the clinical and policy decisions targeted by the trial or (b) measure a broad range of outcomes.
What would be experimental, prospective studies?
- Blinded RCTs. - Phase I, II, and III clinical trials.
What are some harmful practices that were supported by expert opinion?
- Blood letting - Thalidomide for morning sickness (teratogen) - Bed rest for low back pain (still happens) - Benzos for mild anxiety and insomnia - Lidocaine for acute MI
Relative Risk Interpretation
- Calculated only from RCT's or cohort studies. - If RR >1, the risk of outcome of interest is increased. - If RR <1, the risk of outcome is decreased. - If RR=1, risk is equal between groups. - They are reported with a confidence interval and p value (chance). - If the CI crosses 1, there is no difference between the groups.
What would be observational, retrospective studies?
- Case Studies (medical records, researches). - Case-Control Studies (Risk factors, dependent variable & controlled).
What would be observational, prospective studies?
- Case studies. - Controlled trials. - Phase IV clinical trials.
What would be quasi-experimental, prospective studies?
- Cohort studies. - Non-blinded RCTs. - Phase I, II, and III clinical trials.
Advantages of Surrogate Outcomes
- Effects seen sooner than with outcomes that matter. - Treatments available sooner potentially reducing harm and suffering from a disease. - Surrogates are less expensive to study. - Surrogates are usually continuous scale variables= smaller sample sizes. - Surrogates are used in every day clinical practice, so clinicians can compare. - Some surrogates are reliably associated with clinically important outcomes.
What are examples of Qualitative (Naturalistic) research paradigms?
- Enthnography. - Biography. - Phenomenology. - Grounded theory. - Case study.
Name some Best- practices of Sampling (considerations)
- Population & disease demographics and geography - Sampling frame in context of population - Methods of identifying, recruiting, selecting subjects - Inclusion/Exclusion criteria - determines how generalizable it will be (who can participate) - Cost to subjects/ incentives - Sample size calculation, w/ effect size and power in terms of alpha and data scale for each outcome measure - Literature on the chosen effect size - extant literature should make reference to effect size, what they used (referencing older tests/literature) to determine what sample size they should use - Sampling fraction that will be attained - Sample's representativeness of the population - Differences between groups at baseline (if groups are compared) - Effects of drop-outs
Applications of quantitative research (Observational, Quasi-Experimental, Experimental studies) to healthcare: - - - - - - - -
- Prevalence or incidence of disease and associated risk factors. - Accuracy, harm, or side effects of screening tools and diagnostic procedures. - Interactions between interventions. - Cost of care. - Survival longevity (With and without treatment). - Effectiveness and complications of treatment modalities. - Quality of life. - Patient satisfaction.
Relative Risk calculation
- RR = Experimental event rate (EER)/control event rate (CER).
Relative Risk Reduction
- RRR- way of reporting risk. - Also reported with a CI and p value - difference in the event rates over the event rate in the untreated (control) group. *RRR=CER-EER/CER
Parametric data- define
- Ratio, continuous, or interval scale - Normal shape of distribution - Sufficient sample size - Equivalent variability across groups
What are the three major groups of bias?
- Sampling Bias. - Measurement Bias. - Other source of bias.
What are the two types of measurement bias?
- Self-reporting bias. - Recall bias.
Traditional RCT: - Setting? - Purpose? - Questions? - Subjects? - Comparison? - Outcomes?
- Setting: Artificially controlled. - Purpose: Identifying causes and effects of interventions. - Questions: How an intervention works under ideal conditions. - Subjects: Cohorts of individuals who meet strict inclusion and exclusion criteria. - Comparison: Comparing a novel intervention with placebo or non-treatment. - Outcomes: Surrogate or process measures.
Pragmatic Clinical Trial: - Setting? - Purpose? - Questions? - Subjects? - Comparison? - Outcomes?
- Setting: Routine patient care (Clinic). - Purpose: Developing generalizable knowledge that improves local care and informs policy-decisions. - Questions: How an intervention works under usual conditions. - Subjects: Broad/diverse populations seen in routine settings. - Comparison: Different interventions are compared with one another. - Outcomes: Patient-centered outcomes that matter.
What are random methods of sampling techniques?
- Simple random sampling. - Stratified proportional sampling. - Stratified constant sampling.
2 kinds of outcomes
- Surrogates - Outcomes that Matter
What are examples of Quantitative (Positivistic) research paradigms?
- Survey study. - Cohort study. - Case-control study. - RCT. - Systematic review. - Meta-analysis.
What are non-random methods of sampling techniques?
- Systematic sampling. - Convenience sampling. - Purposive sampling. - Quota sampling.
What are some limitations of qualitative research?
- Time-consuming. - Difficulty in sharing data (displaying, size, protection of subject identity).
What are limiting factors to all types of research (Quantitative and Qualitative)?
- Time. - Cost.
What are the 5 types of research questions of the clinical trials?
- Treatment. - Prevention. - Diagnosis. - Screening. - Quality of life.
Attributable Risk (AR)
- difference between the exposed (experimental) event rate and the unexposed (control) event rate. - As a measure of impact, the proportion of disease cases linked to an exposure or intervention. - ARR eg risk of cancer FROM smoking- consider nonsmokers who get cancer to assess the attribution of smoking to cancer
Framingham Calculator
- example of regression analyss- Looks for the combination of predictor variables that best explain the response variable: - based on regression analysis in which researchers identified the 7 variables that, when combined, offer the best prediction of an adverse coronary event within 10 years.
What is Effect Size?
- is the magnitude of the difference between groups. - The absolute effect size is the difference between the average, or mean, outcomes in two different intervention groups.1 - doesn't capture the variability within the group... just average of their response. - clinical significance: statistics show that the drug is effective, but comparison of the groups show a very small difference between it and placebo (1%)- ex.drug for IBS- only 1% better than placebo, eventually people died from drug (removed from market!!) ... statistically significant but not clinically significant. Importance of phase 4 long-term effect studies Most meaningful for quantitative data but can be applied to qualitative data and correlation. "On the basis of previous studies, we estimated that 20% of patients would achieve the primary outcome with intensive medical therapy, 83% with gastric bypass and 50% with sleeve gastrectomy."2
What are the two steps for conducting critical appraisal?
1. Determine the paradigm and style of research conducted. 2. Complete the Critical Appraisal form that goes with the type of study.
Sample size calculations require knowledge of THREE ELEMENTS:
1. Effect size 2. Alpha (α) 3, Power and the type of data
What are the layers for the naturalistic evidence pyramid?
1. Evidence syntheses 2. Confirmatory or transferability studies 3. Theory construction research- grounded theory 4. Single group research- phenomenology, ethnography 5. Review articles, analysis of documentary resources
What are the steps in evidence-based practice?
1. Formulate focused questions 2. Identify articles and other evidence-based resources 3. Critical appraisal of evidence 4. Apply the evidence 5. Evaluate application of evidence
What are two advantages of replication?
1. Allows experimenter to obtain an estimate of the experimental error ~ this error becomes a basic unit of measurement for determining whether observed difference in the data are really statistically different. 2. If the sample mean is used to estimate the effect of factor in the experiment, then replication permits the experimenter to obtain a more precise estimate of this effect.
5 Values to be reported in studies
1. Alpha= Risk of Type I Error 2. Point Estimate- A single value (statistic) that is measured from a sample (e.g. a mean or a proportion) 3. Test Statistic- The value that results from the statistical hypothesis test (e.g. t, F, χ²) 4. p-value- The probability that the test statistic represents normal variability 5. Confidence Interval- The range in which the true population parameter can most likely be found.
What are four types of observational style?
1. Case Report. 2. Case Series. 3. Cross-Sectional Studies. 4. Case-Control Studies.
**What are the Essential Steps for EBM? (5)
1. Convert information needs into answerable questions. 2. Efficiently track down the best evidence to answer those questions. 3. Critically appraise the evidence to assess its validity & usefulness*. 4. Implement the results in practice. 5. Evaluate our performance.
Studies can be based on:
1. Grounded Theory. 2. Randomized-Controlled Trial. 3. Cohort Study. 4. Case Control Study. 5. Phenomenology. 6. Systematic Review. 7. Meta-Analysis.
Three main phases of the Clinical Research Study process
1. Identify need 2. Design study 3. Perform study
What are the components of evidence based practice?
1. Looking for the best available evidence. Highest quality and most recent. 2. Knowledge and skills. Clinical Expertise. 3. Collaborative patient centered decisions 4. Using it in the context of a given healthcare situation
What are the elements of patient-centered decisions?
1. Macro level external influences- political systems, education systems, insurance systems. 2. Mesio level external influences- interprofessional collaboration, community 3. Micro level external influences- clinical knowledge and skill, best available evidence 4. Patient or population- needs, ability to afford care, health literacy
Top of the Evidence Heirarchy/Pyramid.
1. Meta Analysis/ Systematic Reviews 2. Evidence Based reviews 3. Randomized control trials
What types of evidence are important?
1. Meta-analyses and systematic reviews: BEST/highest quality, smallest amount of info. Cochrane reviews- best of systematic reviews! can find in PubMed (free) 2. Evidence- based reviews: Many databases. Pre-populated, intended to be used at point of care 3. Randomized clinical trials: 4. Case control/case study/case reports- less validity for application to patients 5. Textbooks: decreased greatly in value/usability since the internet
What are some methods that can be used with regard to research paradigms?
1. Pre-/Post-Interventional. 2. Cross-Over. 3. Observational. 4. Parallel Groups. 5. Bayesian.
What are the three different kinds of temporal studies?
1. Prospective Study. 2. Retrospective Study. 3. Cross-sectional Study.
What are the two kinds of research paradigms?
1. Qualitative (Naturalistic); Example: Focus Group. 2. Quantitative (Positivistic); Example: Survey.
What are four different ways for controlling effects of variables?
1. Randomization. 2. Blinding. 3. Replication. 4. Blocking.
What are the ethical principles of the Belmont report?
1. Respect for persons · Acknowledge independence/self-control · Protect those with diminished independence/self-control 2. Beneficence · Do no harm · Maximize possible benefits and minimize possible harms 3. Justice · Who should receive benefits of research · Who should bear the burdens of research
4 factors effecting Power
1. Sample Size 2. Effect Size 3. Level of significance 4. Standard deviation
What are some applications of qualitative research in health care?
1. Study the interaction between patients and providers - examine differences and similarities 2. Examine barriers to practice change 3.Examine social and cultural factors that influence health care.
What are the levels of the positivistic evidence pyramid?
1. Synthesized summaries or guidelines with ratings. 2. Meta-analyses 3. Systematic reviews 4. RCTs 5. Cohort studies 6. Case-control studies 7. Case series 8. Case report 9. Review, synthesis, opinion, laboratory research
Three test statistics
1. T-test 2. ANOVA test 3. Chi-square test - Each test statistic has an associated p-value (area under the curve, AOC). - The point estimate, usually the mean of the groups is usually reported along with a confidence interval.
Influences on health care practice
1. The Patient/population (ability to afford care, needs, change readiness, culture, values, health literacy) 2. Micro Level External Influences (clinical setting, resources, clinician knowledge and skills, best available evidence, patient support system) 3. Mesio Level External Influences (inter professional collaboration, community's health services, community social support system, community's environment) 4. Macro Level External Influences (Education systems, Policy systems, regulatory legislative political systems, insurance systems)
How do you categorize medical information?
1. Top is the highest level of evidence (most trustworthy) 2. It is less important that you correctly identify the layer into which a given resource should be placed, but more important that you recognize if a resource is near the top, near the middle, or near the bottom of the hierarchy.
How to plan the search strategy?
1.Identify words or phrases identified in the PPAARE components. 2. Identify synonyms or alternative words or phrases. 3. Identify words or phrases to exclude irrelevant evidence. This step might be easier after searching databases and getting irrelevant results.
Most research studies are asking which two questions?
1.What is the difference between 2 or more variables? or 2. What is the relationship or association (AKA correlation)between 2 or more variables?
What makes up the bottom half of the Evidence heirarchy/pyramid?
4. Case-control studies, case studies, case reports 5. Textbooks, review articles, expert opinions and laboratory research
Define- Confidence Intervals
95% Confidence Interval (CI 95%): 95% of the time, mean value of the variable for that sample will be between the upper and lower limits of the interval. If you are measuring independent events in a normal distribution then: 95% CI = μ +/- 1.96 (SEM)
A significant p value is:
<0.05
The conscientious, explicit, & judicious use of the current best evidence in making decisions about the care of individual patients1.
= Evidence based medicine!
B- error refers to
= maximum probability of making a type II error . Type II error occurs when you fail to reject the null hypothesis Ho even though it's false.
What should the value of Power be?
> or equal to 0.80
Confidence Interval
A Confidence Interval calculated for a measure of treatment effect shows the range within which the true treatment effect is likely to be. It is standardly reported as a 95% confidence interval. For absolute differences, 0 is the no difference point. (applies to absolute/relative risk, division/subtraction) CI's are preferable to p values as they report the range of possible effect sizes. More contemporary way for reporting 'meaningfulness'
A study in which there is at least one treatment group and one control group.
A Controlled Trial.
What is patient-centered decision making?
A collaborative process between the practitioner and the patient, where the patient or population is always at the center.
Describe Case Report
A report on a single patient with an outcome of interest. Case reports usually contain detailed information about the individual patients including demographics, clinical features, & information on the diagnosis, treatment, response to treatment & follow-up.
Sample size considerations
A sample that is too small will not accurately represent the population. The sample should include all of the variability in the population. A sample that is too large can waste resources and time and also expose subjects unnecessarily to experimental interventions. A very large sample can make statistical significance appear as a result of statistical power, rather than treatment effect. Watch out for very large sample sizes.
Why EBM for mE?
ARC-PA Standards-4th Edition B1.07: The curriculum must include instruction related to the development of problem-solving and medical decision-making skills. B2.10: The program curriculum must include instruction to prepare students to search, interpret & evaluate the medical literature, including its application to individualized patients. PA Competencies
What is ARC-PA?
ARC-PA is the Accreditation Review Commission on Education for the Physician Assistant is the accrediting agency that protects the interests of the public and physician assistant profession by defining the standards for physician assistant education and evaluating physician assistant educational programs within the territorial United States to ensure their compliance with those standards.
Absolute Risk reduction- calculation
ARR =experimental event rate(EER) minus the control event rate (CER). *ARR=EER-CER
ARI
Absolute Risk Increase - Exposure INCREASES risk compared to control
ARR
Absolute Risk Reduction - Exposure decreases that risk compared to control
Which type of tertiary literature discusses the basic principles of pharmacokinetics?
Basic Clinical Pharmokinetics
Classifying clinical questions
Background: Answered in past Core of medical knowledge Typically found in print books Foreground: Cutting edge of practice May or may be not fully answerable At the heart of practicing EBM
How is an outcome predicted in prospective research?
Based on a predetermined set of characteristics, a specified experience, or a chosen intervention.
*What is the ONLY study design that can clearly answer contributory cause?
CLINICAL TRIALS
A study used to find risk factors (Example: Does smoking cause lung cancer?).
Case Control Study.
What is the hierarchy of Epi Design Strategies?
Case Reports Case Series Cross-Sectional Surveys Case-Control Studies Cohort (Follow-Up) Studies Randomized Controlled Trials The farther you go down, the more complex and confidence you have.
The number of cases that we're combining and putting together is referred to as the?
Case Series.
A type of naturalistic research that is an exploration of a bounded system (A case or multiple cases) over time through detailed, in-depth collection involving multiple sources of information rich in context.
Case Study.
Applications of Qualitative Research paradigm in biomedical research
Case reports/studies, impacts of illness/treatments on patients and studies of motivation or market behavior are common applications
Case Series
Case study, but with a group of patients A group or series of case reports involving patients who were given similar treatment or have a similar illness. Examples: Description of a rare disease Rare presentation of a common disease Novel uncontrolled treatment of disease-Budesonide in pregnancy Novel diagnostic test
Correlation is NOT:
Causation!
What are the two Quantitative variables?
Cause = independent variable. Effect = dependent variable. The effect depends on the cause.
Either directly involve a particular group of people or use materials from humans.
Clinical Trials.
Clinical Trials- what do they do?
Clinical trials test drugs or other interventions to assess their effectiveness & safety.
What are some examples of Point of Care (POC)?
ClinicalKey First Consult (Elsevier Inc) DynaMed Plus Best Practice (British Medical Journal) GIDEON (Global Infectious Disease and Epidemiology Network) UpToDate (Wolters Kluwer)
What falls under quasi-experimental studies?
Cohort
What would be quasi-experimental, retrospective studies?
Cohort Studies.
When you take a group of people and follow them over a period of time (Can be a prospective or retrospective study).
Cohort Study.
What is a general difference between general quasi-experimental design and cohort study?
Cohort study is an observation study.
What is a type of quasi-experimental style?
Cohort.
Systematic Reviews- describe
Combines 2 or more research studies to identify & summarize all the available evidence on the effectiveness of interventions or exposures. *Blended method- not specific type of study
Secondary Research
Combines/synthesizes previously done research for specific aims. Examples: Evidenced Based Reviews Systematic Reviews Meta-analysis
How does knowing risk help patient care?
Communicate quantitative answers to patients on the estimated probability of outcome based on a therapy or risk factor. Demonstrated efficacy of an intervention is important to many people/patients. The most beneficial interventions can targeted.
Type 1 Error
Conclude there is a difference between the groups in the hypothesis when there isn't one false positive result.
A group that serves as a basis for comparison to ascertain the magnitude of the effect of an intervention.
Control Group.
What group in the controlled trials is the placebo, usual/standard care?
Control Group.
Types of Non-probability sampling (3)
Convenience Sample (less generalizable) Purposive Sample (go after specific people) Snowball Sample (more in psychological literature- go after fractionalized population and follow one person to connect with others within group/culture)
How is correlation analyzed?
Correlation coefficient
What isn't EBM?
Decision making by antedote. - experience rules! 2. Just reading papers or decision making by press-clipping. - If it is new, it must be correct? 3. Good old boys sitting around the table. 4. Pure cost-minimization.
Primary Quantitative Research Study Designs
DESCRIPTIVE STUDIES LONGITUDINAL STUDIES
Problem w/ textbooks?
Decreased in value/usability since the internet. may be dated.
What would be experimental, retrospective studies?
N/A.
Can Case studies prove cause-effect relationship?
NO
Are all study designs capable of proving causation?
NO!
Which tertiary literature sources would be best to use when needing information on natural products/medicines?
Natural Medicine database and The Review of Natural Products
Which resource would you use if you were looking for evidence-based information regarding natural products?
Natural Medicines Comprehensive Database
Which paradigm's analysis is thematic, narrative, content analytic procedures?
Naturalistic/Qualitative.
Which paradigm's data is observation, artifact, textual, visual, field notes, audio recordings, includes data from researchers as well as participants?
Naturalistic/Qualitative.
Which paradigm's design is emergent, guided by data as study proceeds, utilizing no intervention?
Naturalistic/Qualitative.
Which paradigm's reasoning is typically inductive, seeking descriptions?
Naturalistic/Qualitative.
Which paradigm's role of researcher is active engagement with participants and phenomenon; Researcher is one of many data?
Naturalistic/Qualitative.
Which paradigm's sampling is small, purposeful, and sometimes emergent selection or serial selection of participants?
Naturalistic/Qualitative.
Which paradigm's setting is natural, uncontrolled and studied as part of the research?
Naturalistic/Qualitative.
Which paradigm has open-ended questions?
Naturalistic/Qualitative. - Would you consider yourself being food-insecure? If so, why?
What are some examples of primary literature?
New England Journal of Medicine Journal of the American Medical Association Archives of Internal Medicine Annals of Pharmacotherapy American Journal of Pharmacist Association Annals of Internal Medicine Lancet British Medical Journal
Case Reports strengths/weaknesses
STRENGTHS Cheap Quick & Easy Alert physicians to emerging diseases All-or-none case series * Milwaukee Rabies protocol. WEAKNESSES No control group- causation cannot be proved. Retrospective- many biases
RCT Strengths/ Weaknesses
STRENGTHS: Can prove contributory cause Biases are limited by design. WEAKNESSES: Expensive. Require cooperation. Open to industry influence. Subject compliance/attrition. Ethical concerns.
Strengths/Weaknesses of Cohort studies
STRENGTHS: Determine incidence of disease. Can show cause is associated with effect more than by chance. Can show cause precedes effect. Ideal design for answering etiology, harm or prognosis ?'s. WEAKNESSES: Expensive. Long duration Poor for determining new risk factors No intervention usually
Case control strengths/weaknesses
STRENGTHS: Screen for potential risk factors. Cheap, easy, & relatively quick. Can study rare diseases. Used as a pilot for a stronger study. WEAKNESSES: Subject to review bias, recall bias, and sampling/referral biases. Cannot prove contributory cause. Cannot alter the cause to change effects
What are common sources of sampling bias?
Sample size, representatives of the sample, selection bias. - Erroneous conclusions can be drawn from studies that are too small or too large.
A type of bias that involves selecting a sample, a subset of a population, where the sample is not representative of the population.
Sampling Bias.
The process of formulating a hypothesis, performing objective experiments, and engaging in sound reasoning supported by the collected data.
Scientific Method.
Occurs when there is a systematic procedural error, or limitation, in the selection of subjects or when a nonrandom (nonprobability) sample is utilized.
Selection Bias.
What modifications (additions) were made to PA Certification maintenance process in 2014?
Self-assessment component Self-reflection & meta-cognition to identify learning needs & address them. Clinical Quality Improvement Project Practice-based learning & improvement using quantitative skills.
Types of probability sampling (3)
Simple Random Sample Stratified Random Sample (ex- different doses w/in control groups) Multi-stage Sample
A type of blinding that: - Group allocation concealed from subjects.
Single Blind.
Blinded Clinical Trials
Single blind study- participants or investigators are blind to the treatment. Double blind study- Both participants & investigators are blind to treatment allocation. *most common (but many therapies can't be done this way) Triple blind study- participants, investigators and assessors are blind to the treatment.
A blinding mechanism in which the subject doesn't know who is on what treatment.
Single-Blinded Study.
"It is much more important to know what sort of a patient has a disease, than what sort of a disease a patient has." This quote describes the centrality of what in EBM?
The Centrality of the Individual
Considered the leading organization in the world for this systematic research, and sets the standards for design and quality of systematic reviews and meta-analyses.
The Cochrane Collaboration.
What is the aim of Blindedness?
The aim is to minimize observer and participant bias.
What is phenomenology?
The belief that the object of interest be examined without any preconceived notions or expectations; studying a single phenomenon
non-concurrent cohort studies
The cohort is identified on the basis of a feature or features that were determined in the past. Exposures and outcomes are collected sometime in the past, but the ascertainment of exposures predates the development of the outcomes. - NCCS are AKA database studies & are often done in f/u to previously designed study. Doctors Health Study, Nurses Health Study, Whitehall Study, Framingham Study, Olmstead County (Mayo)
What is Positivistic or Quantitative evidence?
The computation of numerical values to represent phenomena in order to describe them and/or make predictions about them
What occurs in phase II of clinical trials?
The experimental treatment is given to a larger group of people (100 to 300) to determine if it is effective and to further evaluate its safety.
The ACCORD trial (Action to Control Cardiovascular Risk in Diabetes) studied the effects of intensive glucose lowering in type 2 diabetes. What did it demonstrate about surrogate outcomes/ outcomes that matter?
They can ne different! After 1 year, patients with intensive glucose control has improved HbA1C, but ALSO Increased mortality and failure to decrease cardiovascular events
What are the drawbacks to a retrospective study?
This data has limitations because they did not collect this data with you and your study in mind. Thus, data might've been collected in a way you didn't want collected which can result in differing definitions and explanations.
What skills are needed to keep up to date with the best evidence?
To be able to find evidence more efficiently and to use good quality evidence more sytematically
What is the strategy of systematic review?
Use critical thinking approach to gather, evaluate, and synthesize information from individual research studies to address a question of clinical relevance.
What is a surrogate outcome?
Using one outcome to reflect another. Surrogates are selected based on the association of a physiologic or biologic measure with another known clinical end point. (less expensive)
A type of blinding that: - Group Allocation Concealed from Subjects. - Subject Group Allocation Concealed from Health Care Providers. - Subject Group Allocation Concealed from Researchers and/or Data Analysts.
Triple Blind.
T/F: EBM and evidence based practice are not very evolved (new)
True
True/False: A prospective research can be observational, quasi-experimental, or experimental in design.
True.
True/False: Although not always feasible, when a crossover trial is randomized and blinded, it can be considered to be high on the evidence pyramid.
True.
True/False: Sometimes the qualitative part will give you context for the surveys (i.e. quantitative data) that are going on.
True.
COHORT STUDIES
Two groups or cohorts of subjects. One receives the exposure of interest & one does not. The cohorts are generally followed forward in time for the outcome of interest.
The Alpha statistic tells us the risk of
Type I error
What do we know about risk?
Underestimate common risks. Overestimate rare risks. Respond to risks based on emotion rather than facts. Patients are risk averse when faced with medical interventions vs. voluntary decisions. Patients want info on even very rare adverse events.
Case control studies
begin in present and look back in time. Largely done in rare diseases *The purpose of this study is to compare the odds of exposure to the risk factor among controls and cases.
Statistics applied to research
biostatistics
explains the course of an illness and often the patient's response to treatment
case report (case study)
What falls under observational studies?
case report, case series, cross-sectional studies, and case-control studies
Exploration of a bounded system (a case of multiple cases) over time through detailed, in-depth data collection involving multiple sources
case study
examines an outcome that has already occurred and utilizes numerical data to test for relationships between variables or differences between groups that can be associated with phenomenon
case-control study
Usefulness=
clinically applicable
What is objective data?
clinician observes or measures; heart rate; assess a surrogate outcome
*Validity=
closeness to the truth.
Evidence based practice emphasizes what in the healthcare provider-patient relationship?
collaborative decision-making
Nominal measurements are
descriptive
Ordinal measurements are
descriptive, nonparametric
Ratio/Interval measurements are
descriptive, parametric
What do clinical practice guidelines provide?
developed to assist healthcare professionals with their clinical decision making
External validity:
does it apply to our patient population
the study of abstract problems and their processes
grounded theory
Inclusion/Exclusion criteria determine
how generalizable a study results will be (who can participate)
Alternative hypothesis (H1) states :
states that there is a difference between two groups or there is an association between independent & dependent variables.
Null hypothesis (H0) states:
states that there is no difference exists between groups or there is no association between independent & dependent variables.