step 3. 嚴格評讀文獻 (Critical appraisal ) 實施實證醫學五大步驟 step 1. 整理出一個可以回答的問題 step 2. 尋找文獻證據 step 3. 嚴格評讀文獻 (Critical appraisal ) step 4. 應用於病人身上 step 5. 對過程進行稽核

What is CAT? Systematic evaluation of study/studies To critically appraise -- not only “read”, but also “assess” Simplify appraisal with VIP (or PICO/RAMbo)

CRITICAL APPRAISAL SKILLS What is VIP? Validity (Reliability) 效度/信度 Can we believe it ? (研究方法的探討) Importance (Impact) 重要性 We believe it ! But does it matter? (研究結果的分析) Practice (Applicability) 臨床適用性 If we believe it - does it apply to our patients? (如何在臨床運用)

Validity (Reliability) What is RAMbo? “ Do we believe it? ” Randomized：Was it randomised? Were the groups similar? Accounted for all patients Measurement: blinded or objective

“該治療性研究結果是可信的嗎?” EVIDENCE-BASED MEDICINE How to Practice and Teach EBM 病人及醫師是否對治療不知情(blind)？實驗組及對照組是否被同等對待？(M)

Importance (Impact) “ We believe it !But-does it matter? ” What were the results? -How large was the treatment effect? -How precise was the estimate of the treatment effect?

In therapy study Number needed to treat (NNT) Relative risk (RR) Relative risk reduction (RRR) Absolute risk reduction (ARR)

將12,562位有不穩定型心絞痛的患者分兩組進行隨機對照試驗，分別接受aspirin或合併使用clopidogrel ，九個月後研究結果：單獨使用aspirin(對照組)其心肌梗塞、中風或心臟血管疾病死亡率為11.47%；合併使用aspirin及clopidogrel (實驗組)其心肌梗塞、中風或心臟血管疾病死亡率為9.28%。對照組發生非致死性出血副作用為2.7% ；實驗組為3.6% 。 CER=11.47%; EER=9.28% RR=9.28%/11.47%=0.81 RRR=(11.47%-9.28%)/11.47%=19% ARR=11.47%-9.28%=2.19% NNT=1/2.19%=46 ARI(絕對危險性增加率)=3.6%-2.7%=0.9% NNH=1/0.9%=111

Practice (applicability) If we believe it, does it apply to our patient? Can it be applied to my patient? 病患差異 Can it be done here? 風土(可運用的資源) How do patient values affect the decision? 民情(病患的偏好)

How to Decide on the Applicability of Clinical Trials Results to Your Patient(1) Biologic Issues If the treatment can work?有效 ~同樣的治療應用在不同的病患族群是否有不同的反應? ~我們的病人與研究中的病人是否非常不同，以致於無法應用這個研究結果？ 

~病患配合度? 醫療提供者(醫事人員、醫療單位、保險制度)的配合度與能力? How to Decide on the Applicability of Clinical Trials Results to Your Patient(2) Social and Economic Issues If the treatment will work?可行 ~這個治療適用於我們的診療環境嗎？ ~病患配合度? 醫療提供者(醫事人員、醫療單位、保險制度)的配合度與能力?

How to Decide on the Applicability of Clinical Trials Results to Your Patient(3) Epidemiologic Issues How efficient the treatment will be? ~我們的病人是否有其他共病狀況可能改變治療的結果? 影響有多大? ~病人可能從治療中得到什麼好處或壞處？經由治療而減少的不良後果是否比不治療有明顯的差別?

如何進行文獻搜尋及評讀 how to find and determine current best evidence 直接搜尋已經評讀的文獻 - ACP journal club, Cochrane Library 自行評讀文獻 - Simplify appraisal with VIP or PICO/RAMbo 必要時使用評讀工具 - 如CATMaker, CASP

我們的臨床狀況 An 80 year-old Chinese women with 20 years of type 2 DM with triopathy, dyslipidemia, and under medications. Recent myocardial infarction 6 months ago. Poor glycemic control, HbA1c: 10, fasting sugar usually above 200.

假設以下為您找到的文獻之一 Am J Med. 2004 Feb 15;116(4):230-5

逐步逐項嚴格評讀該文獻(1) Paul Glasziou, University of Oxford 確認該研究所提出的問題及所回答的問題(PICO)-PURPOSE

逐步逐項嚴格評讀該文獻(2) Paul Glasziou, University of Oxford 確認該研究的內部效度 (V-R1-randomization) -METHOD

逐步逐項嚴格評讀該文獻(3) Paul Glasziou, University of Oxford 確認該研究的內部效度 (V-R2-similarity)- RESULT the result should have a table of “baseline characteristics” (always table 1)

逐步逐項嚴格評讀該文獻(4) Paul Glasziou, University of Oxford 確認該研究的內部效度 (V-A-accounted)-RESULT

逐步逐項嚴格評讀該文獻(5) Paul Glasziou, University of Oxford 確認該研究的內部效度 (V-M-measurement/blind)-METHOD&RESULT

逐步逐項嚴格評讀該文獻(6) Paul Glasziou, University of Oxford 若該研究通過效度審核，審視並解讀其結果。該介入的效果有多大? 相關性如何?(I)—RESULT

逐步逐項嚴格評讀該文獻(7) Paul Glasziou, University of Oxford DO IT YOURSELF!!