Principle of Evidence-Based

Practice CPP 416

BY PROF SIREEN SHILBAYEH

WEEK 10 LECTURE
Applying Results to Individual
Patients

BY PROF SIREEN SHILBAYEH

2019_2020
SECOND SEMESTER
 THE APPROACH OF THE USERS’ GUIDES
TO THE MEDICAL LITERATURE

 Our approach to
addressing diagnosis,
treatment, harm, and PART A
prognosis begins when
the clinician faces a Chapter 2
clinical dilemma (Figure 1-
1).
 Having identified the Chapter 3
problem, the clinician then
formulates a structured
clinical question (see Chapter 4
Chapter 3, What Is the
Question?), and continues
with finding the 1stbest Chapter 5
3 steps in evaluating
relevant that
evidence (see
evidence
Chapter 4, Finding the
Evidence) (Figure 1-1).2nd Chapter 6

3rd PART B
 Content of this lecture

 10. 5. Measuring Patients’ Experience: How can I

apply the results to patient care?
 11 Advanced Topics in Applying the Results of
Therapy Trials
 11.1 Applying Results to Individual Patients
10. 5. Measuring Patients’ Experience:

How can I apply the results to patient care?

TYPES OF INSTRUMENTS AND TESTS IN MEDICINE

THIS CHAPTER
IS DESIGNED FOR CLINICIANS ASKING
THE QUESTION ,
WILL THIS TREATMENT MAKE THE
PATIENT FEEL BETTER?
Guidelines for Using Articles About Health-Related
Quality of Life (HRQL)
 Finding the Evidence
 Are the Results Valid?
 Have the Investigators Measured Aspects of Patients’ Lives
That Patients Consider Important?
 Is the Instrument Reliable (When Measuring Severity) or
Responsive (When Measuring Change)?
 Does the Instrument Relate to Other Measurements in the
Way It Should?
 Are There Important Aspects of Health-Related Quality of Life
That Have Been Omitted?
 What Are the Results?
 How Can We Interpret the Scores?

 How Can I Apply the Results to Patient Care?

 Has the Information From the Study Addressed Aspects of
Life That Your Patient Considers Important?
 TYPES OF INSTRUMENTS AND TESTS IN MEDICINE

Why do we offer treatment to patients? There are 3

reasons??.
 We believe that our interventions increase
longevity , decrease symptoms, or prevent future
morbidity.
 Decreasing symptoms or feeling better includes
avoiding discomfort (pain, nausea, breathlessness,
and so forth), distress (emotional suffering), and
disability (loss of function).
 TYPES OF INSTRUMENTS AND TESTS IN MEDICINE (cont”)

Why do we offer treatment to patients? There are 3

reasons??. (cont’)
 At least in part because of the difficulty in measurement, for
many years, clinicians were willing to substitute physiologic
or laboratory tests for the direct measurement of these
endpoints, or tended even to ignore them altogether.
 During the past 20 years, however, the growing prevalence
of chronic diseases has led clinicians to recognize the
importance of direct measurement of how people are
feeling and the extent to which they are functioning in
daily activities.
Or Clinicians rated
domains
USING THE GUIDE
Scenarios
 Clinical Scenario (Cont”)

 After hearing about the treatment options, the

patient comments,
 “Doctor, tell me how much better I will actually feel
while taking these medications and also what side
effects I might get. I care more about these than
being able to stay on one medication longer than the
others.”
 USING THE GUIDE
Scenarios

Patient’s concern: insomnia, fearfulness, & hearing

voices.
(Psychiatric patient): The patient asked you 2
specific questions:
•What is the nature of the AE he might experience, and
•How much better he will feel while taking alternative
medications (Olanzapine vs Risperidone).
•Patient: not concerned about his tremor AEs, but his
family is.
Specific: efficacy, safety, cost
 USING THE GUIDE

First step: Literature Search

1-Best Evidence---------------------•CATIE study:
 Large pragmatic trial comparing 4 newer antipsychotic
drugs and 1 old-generation drug (chlorpromazine
belongs to this latter class) among 1500 patients with
chronic schizophrenia.
2- initial skimming
 The authors concluded that, although the majority of the
patients in each group discontinued their medication,
olanzapine proved the most effective in terms of the rates
of overall discontinuation and symptom reduction.
 Patients did not, however, tolerate olanzapine as well,
and the drug can produce weight gain and elevations in
blood glucose and lipid levels.
 USING THE GUIDE

•Second step: outcomes evaluation

(Are the Results Valid?)

 In the Clinical Antipsychotic Trials of Intervention

Effectiveness (CATIE) trial the investigators
 not only administered the PANSS but also
 monitored adverse events through systematic query,
 administered 3 rating scales of extrapyramidal signs, and
 measured changes in weight, electrocardiogram, and
laboratory analyses.
The assessment appears adequately comprehensive.
 USING THE GUIDE

•Patient’s concern: insomnia, fearfulness, & hearing voices.

(efficacy), AE (tremor or others)

Third step: statistical evaluation of results (WHAT

ARE THE RESULTS?)
 Neurologic side effects of olanzapine & risperidone were
very similar, with approximately 8% experiencing some
extrapyramidal signs in olanzapine.
 •Additional AEs - olanzapine - also weight gain (>7% ) in
30% on olanzapine vs 14% taking risperidone; P <0.001)
 Increase in glycosylated hemoglobin in olanzapine - but
it does not tell - any patient-important consequences of
increased blood glucose level.
 USING THE GUIDE

•Patient’s concern: insomnia, fearfulness, & hearing voices.

Fourth Step: Additional evidence to answer

specific questions
1-Another report
 •Reported: A greater increase in plasma prolactin for
patients taking risperidone > those taking other
medications (P < 0.001), does not tell us if it led to any
patient-important consequences.
 •No report of these changes insomnia, fearfulness, &
hearing voices, but, given changes in PANSS (Positive &
Negative Syndrome Scale ), one would anticipate small
average effects & a low important improvement with
olanzapine vs risperidone.
 USING THE GUIDE

•Patient’s concern: cost

Fourth Step: Additional evidence to answer

specific questions

3- Another meta-analysis
Olanzapine was the cheapest atypical antipsychotic
drug but may be less effective than the others (not
statistically significant).
 Possible CLINICAL RESOLUTION

1-•concern 1 (shift to newer ASs): Based on available information, you inform the
patient that he is less likely to experience intolerable extrapyramidal AEs with
newer antipsychotics, & given his concern about tremor & his family’s concern
about his looking ill - strongly recommend a switch to one of newer agents. The
patient agrees.

2-Patient’s concern 2: insomnia, fearfulness, & hearing voices. •E.g. newer

antipsychotics: Olanzapine - greater reduction in symptoms, but probability -
patient will benefit is small: 1 in 20 to 100 pts experienced a small but important
change in symptoms when taking olanzapine that he/she would not have
experienced if taking risperidone.

3 Patient concern 3: AEs-•Considering tradeoff between a small likelihood of

benefit in terms of decreased symptoms (olanzapine) & probability of increased
weight gain & an increase in blood glucose of uncertain significance, patient
decides to try olanzapine first while being ready to switch to risperidone soon if
significant AEs (such as substantial weight gain or polyuria as a result of
hyperglycemia) occur.
 CONCLUSION

 Clinicians: consider the effect of their treatments on

patients’ HRQL & to look for information regarding
this influence in clinical trials.
 •Responsive, valid, and interpretable instruments
measuring experiences of importance to most
patients should increasingly help guide clinical
decisions.
 Applying Results to Individual Patients:
CLINICAL SCENARIO

For Which Myocardial Infarction (MI) Patients Is

Thrombolytic Therapy Indicated in the Philippines?
 •A 40-year-old history professor presents to the emergency
department of a general hospital in the Philippines.
 •Severe chest pain for 2 h, associated with clammy
perspiration.
 •Pain is now settling, & patient is not feeling dyspneic or
otherwise in distress.
 •Physical examination: A BP of 110/70 mm Hg, a HR of
92/min, a normal first heart sound, & clear lungs.
 •An electrocardiogram discloses 3-mm ST-segment elevation
… suggesting an acute inferior wall MI.
 •IV lines is placed & patient is admitted to the coronary care
unit.
 Your Work

 you consider the possible benefits and risks of

administering thrombolytic therapy.
 Because, to be fully prepared to advise just this sort
of patient, you have recently examined the literature,
you move quickly and confidently to the bedside.
 FINDING THE EVIDENCE

 Streptokinase is the only thrombolytic agent that most of your patients

might afford.
 •Best evidence from an appropriate RT or, if available, a meta-analysis of
many trials.
 •PubMed: “myocardial infarction” AND “streptokinase.” – limit: “meta-
analysis.”:
 –Meta-analysis that deals with effectiveness but not with safety.
 –A single RCT: Second International Study of Infarct Survival (ISIS-2),
 •On the basis of its size (17000 patients), strong design (which includes blinding), &
performed in a wide variety of centers.
 •Meet the validity criteria for systematic reviews and trials.
 •Meta-analysis: Treatment reduced the event rate from 17.4% to 12.8%.
 •Outweighs the potential harm of bleeding requiring transfusion, which
occurred in 0.5% of streptokinase-treated patients compared with 0.2% in
the placebo group in the ISIS-2 trial.
•Asians compose only a minority of the patients in the trial & in the meta-
analysis.
 INTRODUCTION

 Clinicians looking at RCTs to guide medical

decisions must decide how to apply results to
individual patients. Chapter 6, Therapy, suggested 2
criteria for deciding on applicability:
 (1) Can you apply the results of the study to the
patient before you?
 (2) Are the benefits worth the risks and costs?
In this chapter, we discuss these guides in greater
detail.
 INTRODUCTION

 Clinical trialists typically spend a lot of effort ensuring

comparability of treatment and control groups (internal
validity) through strategies such as randomization,
blinding, and intention-to-treat analyses.
 They spend much less effort on ensuring comparability of
trial patients to actual patients (external validity)
through strategies such as population sampling because
the main focus of trials has been to answer the question,
can the drug work at all?, rather than the question, will it
actually work in real life?
 Can the results be applied to the local
population?

Were the Study Patients Similar to the Patient in

My Practice?
 •Inclusion and exclusion criteria- are they similar
 •Are there differences - make the trial irrelevant?

Be pragmatic (practical) !! And better ask:

 –Why the results do not apply to the patient.
 –You usually will not find a convincing reason, and most
often you can generalize the results to your patient with
confidence
 The extent to which we can generalize
findings:

 From a study using a particular drug to another closely

(or not so closely) related agent.
–Class effects and how conservative one should be in
assuming class effects remains controversial.
 Generalizing findings of surgical treatment may be even
riskier.
–RTs of carotid endarterectomy, for instance, demonstrate
much lower perioperative rates of stroke and death than
one might expect in one’s own community
 Were All Patient-Important Outcomes
Considered?

What clinicians and patients require: Evidence that

the treatments improve outcomes (patient-important
outcomes) e.g.
Reducing shortness of breath during the activities
required for daily living,
Avoiding hospitalization for heart failure,
Decreasing the risk of myocardial infarction.
 Were All Patient-Important Outcomes
Considered?..... Use only Substituted
endpoint

Impact of antiarrhythmic drugs after MI illustrate the

danger of using substitute outcomes or endpoint:
 –A reduction in abnormal ventricular depolarization
(VD) (Flecanide Story!!!!)
 •It made sense for their use to reduce the occurrence of
life-threatening arrhythmias.
 –RTs on 3 agents (encainide, flecainide, & moricizine)
effective in suppressing the substituted endpoint of
abnormal VD.
 – Trials - stopped: The mortality was substantially higher
in patients receiving antiarrhythmic treatment than in
those receiving placebo.
 Were All Patient-Important Outcomes
Considered? Use only Substituted endpoint
…..

 Even when favorable effects of treatment on one

patient-important outcome,:
 –Must consider - may be harmful effects on other
outcomes.
 –Cancer chemotherapy may lengthen life but decrease
its quality of life!
 •RTs often fail to adequately document the toxicity
or adverse effects of the experimental intervention.
 Were All Patient-Important Outcomes
Considered? Use Composite endpoints

 Dangerous trend in presenting outcomes.

 Composite endpoints are attractive for reducing sample
size and decreasing length of follow-up.
 –Unfortunately, they can mislead.
 –A trial that reduced a composite outcome of death,
renal failure requiring dialysis, and doubling of serum
creatinine level actually demonstrated:
 –A trend toward increased mortality with the
experimental therapy and showed convincing effects only
on doubling of serum creatinine.
 Are the Likely Treatment Benefits Worth
the Potential Harm and Costs?

 If you can apply the study’s results to a patient, and

its outcomes are important:
 –The treatment benefits are worth the effort
 –A 25% reduction in the RR of death may sound
quite impressive, but its impact on your patient and
practice may nevertheless be minimal.
 – Better to use number needed to treat (NNT), the
number of patients who must receive an intervention
of therapy during a specific period to prevent 1
adverse outcome or produce 1 positive outcome.
 TABLE 11.1-1: Users’ Guides for Applying
Study Results to Individual Patients

 A. Can I apply the results to my patients?

1. Have biologic factors that might modify the treatment response
been excluded?
2. Can the patients comply with treatment requirements?
3. Can the health care providers comply with treatment
requirements?
 B. Are the benefits worth the risks & costs?
 Have Biologic Factors That Might Modify
the Treatment Response Been Excluded?

 Table 11.1-2 lists 5 biologic factors that sometimes

lead us to reject the idea of applying results to a
particular patient. “SCRAP” is a mnemonic to
remember these 5 factors, which include a patient’s
sex, presence of comorbidity, race or ethnicity, age,
and pathology of the disease.