CLINICAL DECISION USING AN ARTICLE ABOUT

TREATMENTGeorge G. Lim, MD, FPSGS, FPCS, FPSCRSGeorge G. Lim, MD, FPSGS, FPCS, FPSCRS

Professor I, Department of SurgeryProfessor I, Department of SurgeryUST Faculty of Medicine & SurgeryUST Faculty of Medicine & Surgery

Principles of Decision MakingPrinciples of Decision Making

• Knowing the patient’s true state is often unnecessaryKnowing the patient’s true state is often unnecessary• Treatment error is always a possibility when the Treatment error is always a possibility when the

diagnosis is uncertaindiagnosis is uncertain• The need for diagnostic certainty depends on the The need for diagnostic certainty depends on the

penalty for being wrongpenalty for being wrong

Decision Making in MedicineDecision Making in Medicine

• The main issue is making choicesThe main issue is making choices• DiagnosisDiagnosis– Should I request for a laboratory examination or treat right Should I request for a laboratory examination or treat right

away?away?– Between two diagnostic alternatives, which one should I Between two diagnostic alternatives, which one should I

request? request? • TreatmentTreatment– Should I start treatment or observe first?Should I start treatment or observe first?– Between two or more drugs, which one should I prescribe?Between two or more drugs, which one should I prescribe?

Decision Making in the Old ParadigmDecision Making in the Old Paradigm

• Pathophysiologic understandingPathophysiologic understanding• Common senseCommon sense• ExperienceExperience• ExpertsExperts• Medical EvidenceMedical Evidence

Decision Making in the New ParadigmDecision Making in the New Paradigm

• Medical EvidenceMedical Evidence• Pathophysiologic understandingPathophysiologic understanding• Common senseCommon sense• ExperienceExperience• ExpertsExperts

A 72-year old female consulted you for hypertension A 72-year old female consulted you for hypertension and elevated cholesterol. During your discussion on and elevated cholesterol. During your discussion on cardiac risks, she inquires about the need to take an cardiac risks, she inquires about the need to take an anti-cholesterol drug.anti-cholesterol drug.

What advise will you give?What advise will you give?

The PatientThe Patient

Decision Making in the Old Decision Making in the Old ParadigmParadigm

• PathophysiologyPathophysiology– Hypercholesterolemia promotes coronary artery disease thus Hypercholesterolemia promotes coronary artery disease thus

cholesterol is a risk factorcholesterol is a risk factor• Common senseCommon sense

– Lowering cholesterol also lowers the risk of coronary artery Lowering cholesterol also lowers the risk of coronary artery diseasedisease

• ExperienceExperience– If cholesterol is elevated by laboratory examination, it can be If cholesterol is elevated by laboratory examination, it can be

lowered by drugslowered by drugs• ExpertsExperts

– National Cholesterol Education Program: screen all adult National Cholesterol Education Program: screen all adult patients and intervene when necessarypatients and intervene when necessary

Decision Making in the New ParadigmDecision Making in the New Paradigm

• Formulate the problem into an answerable questionFormulate the problem into an answerable question• Conduct a systematic medical literature searchConduct a systematic medical literature search• Critically appraise the medical literatureCritically appraise the medical literature• Apply the resultsApply the results• Evaluate the application Evaluate the application

The EBM PractitionerThe EBM Practitioner

• EBM PractitionerEBM Practitioner– Among elderly patients who have elevated cholesterol, Among elderly patients who have elevated cholesterol,

will anti-cholesterol drugs be beneficial? will anti-cholesterol drugs be beneficial? • The searchThe search– MEDLINEMEDLINE– Randomized trial of cholesterol lowering in 4444 patients Randomized trial of cholesterol lowering in 4444 patients

with coronary heart disease: the Scandinavian with coronary heart disease: the Scandinavian Simvastatin Survival StudySimvastatin Survival Study

Case Control Studies‐Case Control Studies‐• AdvantagesAdvantages– Quick and cheapQuick and cheap– Only feasible method for very rare disorders or those Only feasible method for very rare disorders or those

with long lag between exposure and outcomewith long lag between exposure and outcome– Fewer subjects needed than cross sectional studies‐Fewer subjects needed than cross sectional studies‐

Case Control Studies‐Case Control Studies‐• DisadvantagesDisadvantages– Reliance on recall or records to determine exposure Reliance on recall or records to determine exposure

statusstatus– ConfoundersConfounders– Selection of control groups is difficultSelection of control groups is difficult– Potential bias: recall, selectionPotential bias: recall, selection

Cross Sectional Survey‐Cross Sectional Survey‐• AdvantagesAdvantages– Cheap and simpleCheap and simple– Ethically safeEthically safe

• DisadvantagesDisadvantages– Establishes association at most, not causalityEstablishes association at most, not causality– Recall bias susceptibilityRecall bias susceptibility– Confounders may be unequally distributedConfounders may be unequally distributed– Neyman biasNeyman bias– Group sizes may be unequalGroup sizes may be unequal

Cohort StudyCohort Study

• AdvantagesAdvantages– Ethically safeEthically safe– Subjects can be matchedSubjects can be matched– Can establish timing and directionality of eventsCan establish timing and directionality of events– Eligibility criteria and outcome assessments can be Eligibility criteria and outcome assessments can be

standardizedstandardized– Administratively easier and cheaper than RCTAdministratively easier and cheaper than RCT

Cohort StudyCohort Study

• DisadvantagesDisadvantages– The controls may be difficult to identifyThe controls may be difficult to identify– Exposure may be linked to a hidden confounderExposure may be linked to a hidden confounder– Blinding is difficultBlinding is difficult– Randomization not presentRandomization not present– For rare disease, large sample sizes or long follow up ‐For rare disease, large sample sizes or long follow up ‐

necessarynecessary

Crossover DesignCrossover Design

• AdvantagesAdvantages– All subjects serve as own controls and error variance is All subjects serve as own controls and error variance is

reduced thus reducing sample size neededreduced thus reducing sample size needed– All subjects receive treatment (at least some of the time)All subjects receive treatment (at least some of the time)– Statistical tests assuming randomization can be usedStatistical tests assuming randomization can be used– Blinding can be maintainedBlinding can be maintained

Crossover DesignCrossover Design

• DisadvantagesDisadvantages– All subjects receive placebo or alternative treatment at All subjects receive placebo or alternative treatment at

some pointsome point– Washout period lengthy or unknownWashout period lengthy or unknown– Cannot be used for treatments with permanent effectsCannot be used for treatments with permanent effects

Randomized Controlled TrialRandomized Controlled Trial

• AdvantagesAdvantages– Unbiased distribution of confoundersUnbiased distribution of confounders– Blinding more likelyBlinding more likely– Randomization facilitates statistical analysisRandomization facilitates statistical analysis

• DisadvantagesDisadvantages– Expensive: time and moneyExpensive: time and money– Volunteer biasVolunteer bias– Ethically problematic at timesEthically problematic at times

Randomized Controlled TrialRandomized Controlled Trial

• Provide the BEST evidence of effectivenessProvide the BEST evidence of effectiveness• Individuals are randomly assigned (randomization) Individuals are randomly assigned (randomization)

to either of the groupsto either of the groups• Randomization tries to make the groups SIMILARRandomization tries to make the groups SIMILAR• They are observed FORWARD in time and their They are observed FORWARD in time and their

outcomes comparedoutcomes compared• The outcome can be the cure of a disease, relief of The outcome can be the cure of a disease, relief of

symptoms, or improvement in quality of life symptoms, or improvement in quality of life

RelevanceRelevance

• Is the objective of the article comparing interventions Is the objective of the article comparing interventions similar to your clinical dilemma?similar to your clinical dilemma?– Population of the study (P) – should be similar to the Population of the study (P) – should be similar to the

characteristic of your patientcharacteristic of your patient– Intervention/comparative intervention/exposure (I) – Intervention/comparative intervention/exposure (I) –

should include the therapeutic intervention you want to should include the therapeutic intervention you want to testtest

– Outcome of the study (O) – one of the outcomes measured Outcome of the study (O) – one of the outcomes measured should be the goal you and your patient wish to work forshould be the goal you and your patient wish to work for

RelevanceRelevance

• Is the objective of the article comparing Is the objective of the article comparing interventions similar to your clinical dilemma?interventions similar to your clinical dilemma?– YESYES• The objective of the study is to compare Simvastatin, an anti-The objective of the study is to compare Simvastatin, an anti-

cholesterol drug with placebocholesterol drug with placebo

EBM on TherapyEBM on Therapy

• ValidityValidity– Is the article likely to be true?Is the article likely to be true?

• ImportanceImportance– How large is the size of the effect?How large is the size of the effect?

• ApplicabilityApplicability– Will the findings work in my patient?Will the findings work in my patient?

Are the Results Valid?Are the Results Valid?

• Was the assignment of patients to treatments Was the assignment of patients to treatments randomized?randomized?– YESYES• The title, abstract and the methodology section stated that the The title, abstract and the methodology section stated that the

study was a double-blind randomized controlled trialstudy was a double-blind randomized controlled trial

Are the Results Valid?Are the Results Valid?

• Were all patients who entered the trial properly Were all patients who entered the trial properly accounted for and attributed at its conclusion?accounted for and attributed at its conclusion?– Was follow-up complete?Was follow-up complete?– Were patients analyzed in the groups to which they were Were patients analyzed in the groups to which they were

randomized?randomized?

Drop-outsDrop-outs

Control Drop-out Rate

TreatmentDrop-out Rate

ControlDeath Rate

TreatmentDeath Rate

A.1% 1% 20% 10%

B. 1% 1% 2% 1%

C. 10% 10% 50% 10%

D. 10% 10% 10% 5%

E. 1% 10% 5% 5%

× Drop-out rates are greater than event ratesDrop-out rates are greater than event rates× Gross imbalance in drop-out rates between the groupsGross imbalance in drop-out rates between the groups×Worst assumptions on what happened lead to opposite Worst assumptions on what happened lead to opposite conclusionsconclusions

Are the Results Valid?Are the Results Valid?

• Were all patients who entered the trial properly Were all patients who entered the trial properly accounted for and attributed at its conclusion?accounted for and attributed at its conclusion?– YESYES• 4,444 patients were randomized and analyzed according to 4,444 patients were randomized and analyzed according to

intention-to-treat principle. intention-to-treat principle. • In the Table for the main analysis, the sum of patients in each In the Table for the main analysis, the sum of patients in each

group totaled 4,444, the number who were randomizedgroup totaled 4,444, the number who were randomized

Intention to Treat PrincipleIntention to Treat Principle

Subjects Treatment Failure A

1000 (0)

6-months 100/1000 = 10%

1000 (100)

1-year 45/900 = 5%

Intention to Treat PrincipleIntention to Treat Principle

Subjects Treatment Failure A Failure B

1000 (0)

6-months 100/1000 = 10%

100/1000 =10%

1000 (100)

1-year 45/900 = 5%

135/1000 =13.5%

Are the Results Valid?Are the Results Valid?

• Were patients, health workers and study personnel Were patients, health workers and study personnel blind to treatment?blind to treatment?– YESYES• The study was a double-blind controlled trialThe study was a double-blind controlled trial

Are the Results Valid?Are the Results Valid?

• Were the groups similar at the start of treatment?Were the groups similar at the start of treatment?– YESYES• Table 1 of the results section showed no major difference in the Table 1 of the results section showed no major difference in the

baseline characteristics between the two groupsbaseline characteristics between the two groups

Are the Results Valid?Are the Results Valid?

• Aside from experimental intervention, were the Aside from experimental intervention, were the groups treated equally?groups treated equally?– YESYES• There were no planned co-intervention for the two groupsThere were no planned co-intervention for the two groups

What are the Results?What are the Results?

• If you weighed 80 kg after the Christmas holidays, If you weighed 80 kg after the Christmas holidays, and 60 kg after a summer diet, what would be the and 60 kg after a summer diet, what would be the ways of expressing your weight loss?ways of expressing your weight loss?– I lost 20 kg (absolute weight reduction)I lost 20 kg (absolute weight reduction)– I am now 75% of what I used to weigh (relative weight)I am now 75% of what I used to weigh (relative weight)– I lost 25% of my weight (relative weight reduction)I lost 25% of my weight (relative weight reduction)

What are the Results?What are the Results?

Measures Formula Figures

Risk in Control (Rc) Death Control / N Control 256/2222 = 11.5%

Risk in Treatment (Rt) Death Treatment/ N Treatment 182/2222 = 8.2%

Absolute Risk Reduction (ARR) Rc –Rt 11.5 – 8.2% = 3.3% or0.03

Relative Risk (RR) Rt / Rc 0.082 /0.115 = 0.71

Relative Risk Reduction (RRR) 1 – RR 1 – 0.71 = 0.29 or 29%

Number Needed to Treat (NNT) 1/ARR 1/0.03 = 33.3

What are the Results?What are the Results?

Measures Figures

Risk in Control (Rc) 256/2222 = 11.5%

Risk in Treatment (Rt) 182/2222 = 8.2%

Absolute Risk Reduction (ARR) 11.5 – 8.2% = 3.3% or0.03

Death is prevented in 3% of patients taking the drug

Relative Risk (RR) 0.082 /0.115 = 0.71 The risk of death is now 71% of what it used to be

Relative Risk Reduction (RRR) 1 – 0.71 = 0.2929%reduction in death when taking drug compared to taking placebo

Number Needed to Treat (NNT) 1/0.03 = 33.3 You will have to treat 33

patients to prevent 1 death

What are the Results?What are the Results?

• How precise was the estimate of the treatment How precise was the estimate of the treatment effect?effect?– RR of death in the Simvastatin group = 0.71RR of death in the Simvastatin group = 0.71– 95% CI 0.58 0.85, ‐95% CI 0.58 0.85, ‐ p = 0.0003p = 0.0003

Can the results help me in caring for my Can the results help me in caring for my patient?patient?

• Can the results be applied to my patient care?Can the results be applied to my patient care?– YESYES• Subjects included in this study were patients with angina or MI Subjects included in this study were patients with angina or MI

and elevated cholesterol and elevated cholesterol

Can the results help me in caring for my Can the results help me in caring for my patient?patient?

• Were all clinically important outcomes considered?Were all clinically important outcomes considered?– YESYES• The main outcomes considered were new onset of MI, coronary The main outcomes considered were new onset of MI, coronary

death, and overall cause of deathdeath, and overall cause of death

Can the results help me in caring for my Can the results help me in caring for my patient?patient?

• Are the likely treatment benefits worth the potential Are the likely treatment benefits worth the potential harm and cost?harm and cost?– 33 patients to prevent 1 death (or save 1 life)33 patients to prevent 1 death (or save 1 life)– 40 mg tab at @ P 25.0040 mg tab at @ P 25.00• At 365 days = P 18,250At 365 days = P 18,250• For 33 patients = P 602,250For 33 patients = P 602,250• For 5 years = P 3MFor 5 years = P 3M

– 72-year old patient72-year old patient

Can the results help me in caring for my Can the results help me in caring for my patient?patient?

• Are your patient’s values and preferences satisfied Are your patient’s values and preferences satisfied by the regimen and its consequences?by the regimen and its consequences?– Do your patient and you have a clear assessment of their Do your patient and you have a clear assessment of their

values and preferences?values and preferences?– Are they met by this regimen and its consequences?Are they met by this regimen and its consequences?

A 72-year old female consulted you for hypertension A 72-year old female consulted you for hypertension and elevated cholesterol. During your discussion on and elevated cholesterol. During your discussion on cardiac risks, she inquires about the need to take an cardiac risks, she inquires about the need to take an anti-cholesterol drug. anti-cholesterol drug.

What advise will you give?What advise will you give?

I would rather not give an anti-cholesterol drug. The I would rather not give an anti-cholesterol drug. The benefit is too small for the cost it will take.benefit is too small for the cost it will take.

What advice will you give What advice will you give your 72-year old patient?your 72-year old patient?

A 72-year old female consulted you for hypertension A 72-year old female consulted you for hypertension and elevated cholesterol. During your discussion on and elevated cholesterol. During your discussion on cardiac risks, she inquires about the need to take an cardiac risks, she inquires about the need to take an anti-cholesterol drug. anti-cholesterol drug.

What advise will you give?What advise will you give?

I will prescribe an anti-cholesterol drug. My patient I will prescribe an anti-cholesterol drug. My patient can afford the cost even if the benefit is not large.can afford the cost even if the benefit is not large.

What advice will you give What advice will you give your 72-year old patient?your 72-year old patient?

• Keep ASKINGKeep ASKING

• Keep SEARCHINGKeep SEARCHING

• Keep LEARNINGKeep LEARNING

Thank you.