Evidence-Based Medicine

• Evidence-based medicine (EBM) appraises literature efficiently to address clinical questions for patients, aiding preparedness
• In EBM, physicians have a duty to find and judiciously apply the best, current information for the benefit of their patients
• The goal in EBM is to provide the best care possible using clinical expertise and addressing patient values and expectations for treatment.

5 Steps to Evidence based Practise

• Ask the right clinical questions, by formulating a searchable question

• Collect the most relevant publications through efficient literature searching and selection of appropriate studies

• Critically appraise and synthesize the evidence

• Integrate the best evidence with personal clinical expertise, patient preferences and values, applying it to clinical practice

• Evaluate practice decisions or changes by assessing the outcomes of applied evidence

• The clinical process requires knowledge about etiology in terms of causation

• The clinical process requires knowlegde about patient diagnosis from presentation

• The clinical proccess requires diagnosis knowledge about prognosis

• Clinical proccess change therapy based of prognosis

• Clinical proccess requires knowledge about therapeutic effectiveness to determine clinical outcome

• Etiognostic Research seeks evidence from literature to search on etiology knowledge about causation

• Diagnostic Research seeks evidence from literature to search on patient presentation knowledge about diagnosis

• Prognostic Research seeks evidence from literature to search on diagnosis knowledge about prognosis

• Therapy Research seeks evidence from literature to search on Therapy with knowledge about therapeutic effectiveness which will lead to a clinical outcome

• For non-life-threatening conditions, start with prevalence, screening, diagnosis, and prognosis.

• Questions to address include prevalence, screening, diagnosis, prognosis and the effects and harms of treatment

• A clinical question is formed by considering: the patient, intervention, comparison and outcome, and can translate into: "PICO"

Constructing A Clinical Question

• P: How would I describe a group of patients similar to this particular patient?
• I: Which treatment, est or other intervention?"
• C: Compared to what other treatment, test, or perhaps compared to doing nothing
• O: What is the patient oriented outcome, better prognosis, higher rate of cure

Types of Questions Answered by EBM

• Diagnosis Question addresses a test's ability to predict likelihood of disease: Example: Is duplex ultrasound as good as magnetic resonance angiography in detecting significant carotid stenosis in geriatric patients.
• Prognosis Question concerns patient outcome with condition: Example: In diabetic patients with foot ulcers, is osteomyelitis diagnosis with MRI as predictive of healing as an audible pulse on Doppler exam?
• Therapy Question concerns treatment or preventative effectiveness: Example: Is Depakote more effective than Inderal for migraine prophylaxis without auras?
• Harm Question addresses the likelihood of intervention causing harm: Example: Does high coffee consumption increase spontaneous abortion rate for pregnant patients, compared to non-coffee drinkers?

Searching For Evidence

• Perform de novo literature search.
• Use critically appraised topics (CATs), which are summaries of a study and its results for retrieval and review
• CATs include title, clinical bottom line, the main question, a results summary, comments, publication date, and relevant citations

CAT Components

• Title: The topic of clinical interest in the reviewed article.

• Clinical Bottom Line: The study's major findings related to the topic of interest.

• Clinical Question: The information sought regarding the topic of interest.

• Study Design: Pertinent information about the study's methodology.

• Evidence Summary: A summary of the article's main findings.

• Comments: Suggestions for how to use the study's findings.

• Publication Date: The year the study was published.

• Citations: References to the cited article and other related studies.

• The primary care clinic with a 58-year-old man with hypertension can consider reseach relevant systematic reviews and meta-analyses

• Access systematic reviews and meta-analyses at cochrane.org and critically appraise them

• Look at relevant individual articals like cohort studies, case control sudies, case series and expert opinion

• When researching relevant individual articles, confirm the validity by assessing if study results and conclusions are accurately deduced and methods are free of error or bias

• Confirm there an an independent and blind comparison to a reference standard and confirm a criterion standard should ideally have 100% sensitivity and specificity

• Confirm those whom preform tests and interperet the results is independent

• Subject need have similar baseline characteristics

• View with supsicion if not independently preformed on every participant

Diagnostic Tests

• Sensitivity refers to the proportion of people with the disease who test positive

• SNOUT (Highly Sensitive test, Negative result rule OUT disease)

• Specificity refers to the proportion of people free of the disease who test negative.

• SPIN (Highly Specific test, Positive result rules IN disease)

• High sensitivity means low number of false negatives.

• High specificity means low number of false positives

• Best accuracy has both factors near 100%

• It can only deal with dichotomous tests

• Positive likelihood ratios define how much the odds of disease increase with a positive test result.

• This a the likelihood of +ve test among individuals with disease relative to likelihood of +ve test among those without disease

• Negative likelihood ratios define how much the odds of disease decrease with a negative test result.

• This a the likelihood of -ve test among individuals with disease relative to liklihood of -ve test among those without disease

Advantages of Likelihood Ratio Approach

• The likelihood ratio form of Bayes Theorem is easy to remember: Posttest Odds = Pretest Odds x LR.

• Likelihood ratios can deal with tests with more than two possible results (not just normal/abnormal).

• The magnitude of the likelihood ratio give intuitive meaning as to how strongly a given test result will raise (rule-in) or lower (rule-out) the likelihood of disease

• Computing posttest odds after a series of diagnostic tests is much easier than using the sensitivity/specificity method. Posttest Odds = Pretest Odds x LR₁ x LR₂ x LR₃ ... x LRn

• Positive predictive value is the proportion with +ve tests who have the disease

• Negative predictive value is the proportion with -ve tests who are free of the disease

• Nomograms can interprets diagnostest tests that include pretest probability, likelihood ratio, and posttest probability

• ROC (Receiver Operating Characteristic)

• Measures the sensitivity and specificity of a diagnostic test

• ROC are used for diagnosis (yes/no) cases

• ROC can compare 2 diagnostic tests

• In a ROC curve, the true positive rate (sensitivity) plot vs false positive rate (100-Specificity) for different parameters cut off points

• A sensitivity/specificity pair is an indicator of a decison threshold

• The AUC (area under the curve) measures the two diagnostic groups (diseased or normal)

Useful Definitions

• Relative risk measures how much the risk is reduced in experimental group compared to the control group.

• Absolute risk reduction is the absolute difference in outcome rates between the control and treatment groups, defined as CER - EER

• Number needed to treat describes how many patients to treat to prevent one bad outcome.

• How to research relevant individual articles

• Ascertain the validity of an individual article, by determining if an individual article

• If the study's results and conclusions were accurately deduced

• The methods used to arrive at the conclusions were free of error and bias

• Can these results benefit the patient? -- If the answer is yes, then apply -- If the answer is no, then start over

• Is the diagnostic test available and affordable in the physician's clinical setting?

• How can the physician determine a specific patient's pretest probability of having the target disorder? Using the study's inherent disease prevalence as well as the physician’s.

• Progonostic Studies involves Cohort studies, that identifies a group of individuals with a previously specified characteristic of interest and follow up with them over a pretermined period of time to compare with normal subject with out the disease

Validity of Prognostic Studies

-- Are the study subjects truly representative of the population of interest? -- Studies analyze a representative sample of a patient population and make generalized conclusions -- Where the study subject chosen at a common point in the disease course --The point of the disease course and its effect, validity, and applicability to the patient

• The study subjects and investigators are blinded from measures of interest - Single blinding with clinicians being blinded

Study outcome of progonostic studies

How likely for outcomes from time depending on the endpoint The principle measure of precision is the 95% confidence interval that quantifies the uncertainty of a measurement by reporting a range of values

Validity of Articles on Therapy

• To assess the validity of a study is to ask if its findings are true and accurate -- Were subjects randomly assigned to treatment groups? -- Where subjects blindly assigned? --- Were the study groups similar at the start of the investigation? --- Where the study groups treated equally during the course of the experiments --- How big was treatment effect --- How precise is the estimate of the treatment effect

To Assess if Causation is Presumed:

• Exposure should precede outcome

• Dose and response should corelate

• When assessing articles to identify harm caused by the article, see objective truth, but exclude competing explanation

• Do the researchers use blind studies?

• Short follow up may allow too little to occur

• Studies on prognosis stratify the group

To assess the validity of a study, is the experiment true and accurate?

Is statistical significance expressed in a CI within 95% value Does the P value prove additional information on the magnitude

Is there an independent association between the subject's demo, history and comorbidities

Practical Use In Studies

• • Did a patient meet inclusion/exclusion criteria?
  • Were the diseases/ patient oriented outcomes considered

Appropriate Design for Clinical Research

-- Diagnosis/dx testing, Cross sectional study, not a randomized trial --- Prognosis, Follow-up studies of patients evaluated at the same early point of their illness ---- Therapy/treatment, Rct, or systematic rewiew

• When is the treatment successful in a fatal condition?

EBM Databases Are:

Cochrane, Clinical Evidence, DARE, Acp journal club, infopoems, usef, National guidelines clearing house.