COCHRANE ROB 2 · 偏誤風險評讀
IBS 症狀評分(病人自評)療程後
← 回到「腸躁症」的圖表
總體判斷 OVERALL-High(高風險)
Domain 與 overall 判斷由程式依 Cochrane RoB 2 官方演算法計算;AI 只負責回答下方各訊號問題。符號:+ 低風險、? 有些疑慮、- 高風險(不只靠顏色)。
風險偏誤紅綠燈
- +Domain 1 隨機化過程:Low(低風險)
- -Domain 2 偏離預定介入:High(高風險)
- ?Domain 3 缺失結局資料:Some concerns(有些疑慮)
- +Domain 4 結局測量:Low(低風險)
- ?Domain 5 選擇性報告:Some concerns(有些疑慮)
- -總體(Overall):High(高風險)
中文摘要(給一般讀者)
這篇論文的偏誤風險與 limitation
設計為平行三臂、隨機、雙盲、安慰劑對照試驗(個體化中藥/標準中藥/安慰劑,116 人,16 週)。主要結局取病人自評總 Bowel Symptom Scale(BSS)分數於療程結束時、中藥 vs 安慰劑之組間比較(P=.03)。隨機化:以從封閉袋盲抽密封信封分配、基線均衡,隨機與隱匿風險低(未載明亂數產生法故 PY)。盲性(domain 2 之知情、domain 4 之評估)佳:安慰劑外觀氣味不可分辨、統一 30 分鐘等候與封閉配藥、且以可信度量表主動驗證患者/中醫師/消化科醫師設盲維持成功,故即便主要結局為病人自評的主觀量表,評估者(患者)不知分組,測量偏誤風險低。主要疑慮在 domain 2(偏離介入/ITT)與 domain 3(缺失資料):本連續型結局採『符合方案(僅完成者)』分析而非 ITT(2.6 判 N),約 15% 缺失且各組不對稱(個體化 24% vs 安慰劑 9%),退出原因含『治療無效』可能與真值相關,故 ITT 目標下有實質影響之虞(2.7 PY)。domain 5 因無法確認統計分析計畫是否解盲前定案(5.1 NI),且採混合 ITT/per-protocol 策略而略有保留。整體而言,本結局偏誤傾向主要來自 per-protocol 分析與差異性缺失(domain 2/3),盲性與隨機面向風險低。
詳細評讀(給專業讀者)
Domain 1:隨機化過程 — 判斷:Low
偏誤來源:隨機分組的過程——分配序列是否隨機產生、是否隱蔽到分組為止、基線(第一期開始時)是否平衡。
1.1 Was the allocation sequence random?
分配序列是否隨機產生?
- 回答:可能是(Probably yes — 綜合判斷後傾向「是」)
- 原文引述:Randomization was done by selection of a sealed envelope from a closed bag. Seventy sealed envelopes were prepared for each of the standard and individualized groups, and 60 envelopes were prepared for the placebo group.隨機化以從封閉袋中抽取一個密封信封的方式進行。標準組與個體化組各準備 70 個密封信封、安慰劑組準備 60 個。
- 理由:受試者被「隨機分配(randomly allocated)」到三組,分配靠從封閉袋中盲抽密封信封這個物理隨機過程完成(比例 70:70:60,故患者知道分到活性治療機率較高)。抽籤本身屬隨機程序,故判 PY;但論文未描述信封內容是用亂數表/電腦亂數等正式方法產生,因此不到完全確定的 Y。
1.2 Was the allocation sequence concealed until participants were enrolled and assigned to interventions?
分配是否隱蔽到分組為止?
- 回答:可能是(Probably yes — 綜合判斷後傾向「是」)
- 原文引述:The first consultation with the Chinese herbalist occurred at week 2, at which time the patient was randomized (by an assistant) to placebo, standard CHM, or individualized CHM treatment.與中醫師的第一次會診在第 2 週進行,此時患者由一位助理隨機分配到安慰劑、標準中藥或個體化中藥治療。
- 理由:分配在納入後由助理以「從封閉袋抽密封信封」執行,且全程雙盲、安慰劑與真藥外觀氣味無法分辨,配藥在封閉房間由不接觸患者的助理處理,指派前無法預知落點,隱匿大致充分,故 PY。信封法未載明不透光/連續編號等強化措施,未達完全確定的 Y。
1.3 Did baseline differences between intervention groups suggest a problem with the randomization process?
(第一期開始時)基線差異是否暗示隨機化有問題?
- 回答:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:Patient groups were similar in terms of age, weight, and sex distributions. There were no significant differences among patients in the 3 groups in terms of total severity of symptoms... Patients allocated to the placebo group had a higher mean score for constipation, while patients allocated to the standard treatment had a higher mean score for diarrhea.各組在年齡、體重、性別分布相似;三組在總症狀嚴重度上無顯著差異。安慰劑組便秘分數平均較高、標準組腹瀉分數平均較高。
- 理由:基線(年齡、體重、性別、病人與醫師自評總 BSS 分數 P 值 .59~.68)相當均衡,僅個別症狀次量表(便秘/腹瀉)有輕微不均,屬隨機分組的正常波動,未暗示隨機化出問題,故 PN。
Domain 2:偏離預定介入 — 判斷:High
偏誤來源:實際執行時偏離預定的介入——盲性如何、是否依「隨機分到的組別」分析(ITT)。
2.1 Were participants aware of their assigned intervention during the trial?
參與者是否(在每一期)知道自己的分組?
- 回答:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:Patients, gastroenterologists, and herbalists were all blinded to treatment group... the placebo was deemed indistinguishable from raw powdered Chinese herbs. All herbs and the placebo formulation were supplied in the same opaque capsules... The success of blinding patients to treatment was tested... No significant difference was noted between groups at outset and at end of treatment.患者、消化科醫師、中醫師皆對分組設盲;安慰劑被判定與生的中藥粉無法分辨,所有藥物與安慰劑都裝在相同的不透光膠囊;並以可信度量表檢驗設盲成效,各組間在試驗開始與結束時皆無顯著差異。
- 理由:雙盲、安慰劑外觀氣味味道與真藥不可分辨、且以可信度量表主動驗證設盲成功(各組間無顯著差異,作者確信設盲維持),故判受試者不知道自己分組,PN。
2.2 Were carers and people delivering the interventions aware of participants' assigned intervention during the trial?
照護/執行者是否(在每一期)知道分組?
- 回答:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:The authors are convinced that patients, herbalists, and gastroenterologists were all successfully blinded... All medication preparation occurred in a closed room by assistants who were restricted from contact with the patients... A blinded primary research assistant managed all the questionnaires and was responsible for giving the capsules to the patients.作者確信患者、中醫師與消化科醫師都成功設盲;所有藥物在封閉房間由不接觸患者的助理準備;一位設盲的主要研究助理管理問卷並負責發放膠囊給患者。
- 理由:施治的中醫師與評估的消化科醫師均設盲;為避免個體化組(現場配製)與預製膠囊(標準/安慰劑)被辨識,所有人統一等候 30 分鐘、配藥於封閉房間由專責助理完成。故照顧/施治者不知分組,PN。
2.3 Were there deviations from the intended intervention that arose because of the trial context?
是否因試驗情境造成偏離?
- 回答:不需評估(NA)
- 原因:條件題:依 2.1/2.2 的答案,本題依 RoB 2 官方路由不需評估(NA)。
2.4 Were these deviations likely to have affected the outcome?
這些偏離是否可能影響結局?
- 回答:不需評估(NA)
- 原因:條件題:依 2.3 的答案,本題依 RoB 2 官方路由不需評估(NA)。
2.5 Were these deviations from intended intervention balanced between groups?
這些偏離在兩組間是否平衡?
- 回答:不需評估(NA)
- 原因:條件題:依 2.4 的答案,本題依 RoB 2 官方路由不需評估(NA)。
2.6 Was an appropriate analysis used to estimate the effect of assignment to intervention?
是否用了適當分析(ITT)?
- 回答:否(No — 有明確證據支持)
- 原文引述:Data are presented according to an intention-to-treat protocol, in which patients who withdrew from the trial were recorded as having worsened (if appropriate) for categorical items only. Data for all other outcome measures are presented as per protocol analysis... Missing scale and item scores were not replaced.資料依意向治療(ITT)呈現,其中退出試驗者被記為惡化(若適當)僅限於類別型項目。所有其他結局指標(含連續型分數)以符合方案(per protocol)分析呈現;缺失的量表與項目分數未做替補。
- 理由:本主要結局為連續型的病人自評總 BSS 分數,論文明言連續型指標一律採「符合方案(per protocol,僅完成者)」分析,未做替補;Table 3 各組療程結束 n(安慰劑 32、標準 38、個體化 29)小於隨機人數即反映此點。以「指派效果(ITT)」為估計目標時,此非適當分析,故判 N。(僅類別型的總體改善用了 ITT 惡化插補。)
2.7 Was there potential for a substantial impact (on the result) of the failure to analyse participants in the group to which they were randomized?
未依隨機分組分析,是否可能對結果造成重大影響?
- 回答:可能是(Probably yes — 綜合判斷後傾向「是」)
- 原文引述:Fifteen patients withdrew during the 4-month course of the trial, and 2 patients were withdrawn from the trial for commencing a variety of relevant medications... Withdrawn (n=9) [Individualized]... Withdrawn (n=5) [Standard]... Withdrawn (n=3) [Placebo].試驗四個月期間有 15 名患者退出、另 2 名因開始使用相關藥物而被退出;退出人數:個體化組 9 人、標準組 5 人、安慰劑組 3 人。
- 理由:以符合方案分析排除退出者,且退出比例在各組不對稱(個體化 9/38≈24%、標準 5/43≈12%、安慰劑 3/35≈9%),退出原因含「治療無效」等可能與結局相關者;此差異性流失在僅分析完成者下,確有可能實質影響組間比較結果,故判 PY。
Domain 3:缺失結局資料 — 判斷:Some concerns
偏誤來源:結局資料缺失——是否幾乎所有人都有資料、缺失是否可能與病情(結局真實值)相關。
3.1 Were data for this outcome available for all, or nearly all, participants randomized?
是否幾乎所有受試者都有此結局的資料?
- 回答:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:Patient total BSS score ... End of treatment ... 150.0 (81.6) (n = 32) [Placebo] ... 106.1 (73.7) (n = 38) [Standard] ... 103.0 (74.7) (n = 29) [Individualized].病人自評總 BSS 分數在療程結束時:安慰劑 150.0(n=32)、標準 106.1(n=38)、個體化 103.0(n=29)。
- 理由:隨機共 116 人,本結局療程結束時可得資料 32+38+29=99 人(約 85%),缺失約 15%,且個體化組僅 29/38(約 76%)。並非「幾乎所有受試者」都有結局資料,故 PN。
3.2 Is there evidence that the result was not biased by missing outcome data?
是否有證據顯示結果未受缺失資料偏誤?
- 回答:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:Data for all other outcome measures are presented as per protocol analysis... Missing scale and item scores were not replaced.其他結局指標以符合方案分析呈現,缺失的量表分數未做替補。
- 理由:針對此連續型結局,論文僅呈現完成者的符合方案分析、未替補缺失,也未提供敏感度分析或其他證據證明結果未受缺失資料偏誤,故無「未受偏誤」的證據,判 PN。(類別型結局另用 ITT 惡化插補,但不適用於本連續型結局。)
3.3 Could missingness in the outcome depend on its true value?
缺失是否可能取決於結局真實值?
- 回答:可能是(Probably yes — 綜合判斷後傾向「是」)
- 原文引述:Withdrawn (n=9) Intervention Ineffective (n=1) Lost to Follow-up (n=8) [Individualized] ... Withdrawn (n=3) Intervention Ineffective (n=2) ... [Placebo] ... Withdrawn (n=5) Intervention Ineffective (n=1) ... [Standard].退出原因:個體化組 9 人(治療無效 1、失去追蹤 8);安慰劑組 3 人(治療無效 2);標準組 5 人(治療無效 1)。
- 理由:退出/缺失原因中包含「治療無效(intervention ineffective)」,此與症狀真實嚴重程度直接相關(症狀未改善者較可能退出),故缺失有可能取決於結局真值,判 PY。
3.4 Is it likely that missingness in the outcome depended on its true value?
缺失是否很可能取決於結局真實值?
- 回答:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:Lost to Follow-up (n=8) [Individualized] ... Lost to Follow-up (n=3) [Standard] ... Lost to Follow-up (n=0) [Placebo] ... Intervention Ineffective (n=1/2/1).失去追蹤:個體化 8 人、標準 3 人、安慰劑 0 人;因治療無效退出各組僅 1~2 人。
- 理由:明確記錄為「治療無效」而退出者各組僅 1~2 人,多數缺失歸類為「失去追蹤」原因不明、未必與症狀真值連動;缺失「可能」取決於真值(3.3 判 PY),但尚無足夠證據顯示「很可能」如此,故 3.4 判 PN。
Domain 4:結局測量 — 判斷:Low
偏誤來源:結局的測量——測量工具是否恰當、評估者是否盲化、評估是否可能受分組知情影響。
4.1 Was the method of measuring the outcome inappropriate?
結局測量方法是否不適當?
- 回答:否(No — 有明確證據支持)
- 原文引述:The BSS consists of 100-mm visual analog scales related to each symptom of IBS... The reliability of the BSS... was high for total score (r = 0.7; P<.01)... The visual analog scales within the BSS had high face validity... and have high content validity.BSS 由各項 IBS 症狀的 100mm 視覺類比量表構成;總分信度高(r=0.7, P<.01),且具高表面效度與內容效度。
- 理由:以效度、信度均經檢驗的 BSS 症狀量表作為量測工具,方法適當、量表本身不失當,故 4.1 判 N(測量方法並非不當)。
4.2 Could measurement or ascertainment of the outcome have differed between intervention groups?
(每一序列內)結局測量是否可能因介入而異?
- 回答:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:Patients and gastroenterologists completed this scale independently at the beginning and end of the treatment period... All patients were treated in an equivalent fashion.患者與消化科醫師在療程開始與結束時各自獨立完成此量表;所有患者以相同方式對待。
- 理由:三組使用相同的 BSS 量表、相同時間點、相同流程量測,量測與確認方法在各組間一致,故不同組之間量測不會有系統差異,判 PN。
4.3 Were outcome assessors aware of the intervention received by study participants?
結局評估者是否知道受試者接受的介入?
- 回答:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:Patients ... were all blinded to treatment group... The success of blinding patients to treatment was tested... No significant difference was noted between groups at outset and at end of treatment... The authors are convinced that patients ... were all successfully blinded.患者對分組設盲;患者設盲的成效經檢驗,各組間在開始與結束時無顯著差異;作者確信患者成功設盲。
- 理由:本主要結局為病人自評,結局評估者即患者本人;患者設盲並經可信度量表主動驗證維持成功,故評估者(患者)不知道所接受的分組,判 PN。
4.4 Could assessment of the outcome have been influenced by knowledge of intervention received?
結局評估是否可能受介入知情影響?
- 回答:不需評估(NA)
- 原因:條件題:依 4.3 的答案,本題依 RoB 2 官方路由不需評估(NA)。
4.5 Is it likely that assessment of the outcome was influenced by knowledge of intervention received?
結局評估是否很可能受介入知情影響?
- 回答:不需評估(NA)
- 原因:條件題:依 4.4 的答案,本題依 RoB 2 官方路由不需評估(NA)。
Domain 5:選擇性報告 — 判斷:Some concerns
偏誤來源:選擇性報告結果——是否依事前訂定的分析計畫分析、有無從多個結果中挑選有利者報告(交叉另含:是否因偵測到殘留而只報第一期)。
5.1 Were the data that produced this result analysed in accordance with a pre-specified analysis plan that was finalized before unblinded outcome data were available for analysis?
是否依事前訂定(取得未盲化資料前定版)的分析計畫分析?
- 回答:資訊不足(No information — 原文未提供足以判斷的資訊)
- 原文引述:For the BSS, analysis of variance was used to determine the differences among groups at baseline, end of treatment, and follow-up... Data are presented according to an intention-to-treat protocol... for categorical items only. Data for all other outcome measures are presented as per protocol analysis.BSS 以變異數分析比較各組在基線、療程結束與追蹤的差異;資料依 ITT 呈現僅限類別型項目,其他結局指標以符合方案分析呈現。
- 理由:本文為 1998 年試驗,未提及事前註冊或可取得的統計分析計畫,無法確認分析是否在解盲前定案;且採「類別型用 ITT、連續型用 per protocol」的混合分析策略,難以判定是否事前規定。依規格無法確認統計計畫事前定案即判 NI。
5.2 Is the numerical result being assessed likely to have been selected, on the basis of the results, from multiple eligible outcome measurements (e.g. scales, definitions, time points) within the outcome domain?
此數值結果是否可能依結果,從多個量表/定義/時間點中選出?
- 回答:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:For all 5 main outcome measures—total mean BSS scores and global improvement as assessed by patients and gastroenterologists, and interference with life as assessed by patients—patients receiving the standard CHM formulation responded significantly better than patients in receiving placebo.五項主要結局指標(病人與醫師評的總 BSS 分數、總體改善,以及病人評的生活干擾程度)皆有報告,標準中藥組均顯著優於安慰劑。
- 理由:主要結局為事前定義的 BSS 總分,五項主要結局皆完整報告、各時間點(基線/療程結束/追蹤)也都呈現,未見從多個量表或時間點中挑選有利者只報一項的跡象,故 PN。
5.3 Is the numerical result being assessed likely to have been selected, on the basis of the results, from multiple eligible analyses of the data?
此數值結果是否可能依結果,從多種分析中選出?
- 回答:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:Data are presented according to an intention-to-treat protocol... for categorical items only. Data for all other outcome measures are presented as per protocol analysis.資料依 ITT 呈現僅限類別型項目,其他結局指標以符合方案分析呈現。
- 理由:分析策略(連續型 per protocol、類別型 ITT)雖非理想,但為明文、系統性的分工,非事後在多種分析中挑選有利結果只報之;未見以子群/校正與否等多種分析擇優呈現的跡象,故 PN。(惟此策略事前是否定案無法確認,見 5.1。)