COCHRANE ROB 2 · 偏誤風險評讀
Constant-Murley 分數(療程結束)
← 回到「癌因性疼痛」的圖表
總體判斷 OVERALL?Some concerns(有些疑慮)
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:Low(低風險)
- +Domain 3 缺失結局資料:Low(低風險)
- ?Domain 4 結局測量:Some concerns(有些疑慮)
- +Domain 5 選擇性報告:Low(低風險)
- ?總體(Overall):Some concerns(有些疑慮)
中文摘要(給一般讀者)
這篇論文的偏誤風險與 limitation
本研究為前瞻、開放標籤的平行組 RCT(針灸 vs 常規照護 usual care),主要結局為 4 週療程結束時的 Constant-Murley 複合分數(ITT,針灸優於對照 +11.2 分、P=.008)。隨機化健全:分層+隨機長度區組+電腦化資料庫確保完整分配隱匿;雖有性別、基線分數與用藥的偶發不均衡,但已以共變數校正,未暗示隨機化問題。因開放標籤,病人與施治者皆知分組(無法盲化為此療法本質);分析採 ITT,故指派效果的偏離風險低。缺失資料約 17%(70 隨機、58 可評估),但作者以多重插補(含保守假設)做敏感度分析結果幾乎不變,缺失偏誤風險受控。結局測量為本篇關鍵:主要量表由盲化第三方研究人員評估(客觀的活動度/肌力尤佳),但複合分數含病人自陳的疼痛與 ADL 子量表,作者自承易受未設盲者偏誤回報影響,故 4.3/4.4 取保守 PY;惟僅取純客觀成分重分析仍顯著(10.1 分、P=.037),顯示效果非單由知情自陳驅動,4.5 判 PN,Domain 4 傾向『部分疑慮』。選擇性報告方面,主要結局、主分析與樣本數事前明訂且為登錄試驗,報告透明,風險低。整體而言本結局主要偏誤來源為開放標籤下半主觀複合結局的測量疑慮,其餘面向偏誤風險低。
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:是(Yes — 有明確證據支持)
- 原文引述:Random assignment was stratified by neck procedure type (selective, modified, or radical) and baseline Constant-Murley score using blocks of random length. Random assignment was implemented via a secure computerized database.隨機分派依頸部手術類型(選擇性、改良式或根治性)與基線 Constant-Murley 分數分層,使用隨機長度區組,並經由安全的電腦化資料庫執行。
- 理由:明確使用分層+隨機長度區組,並以電腦化資料庫產生序列,屬真正隨機化。
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:是(Yes — 有明確證據支持)
- 原文引述:Random assignment was implemented via a secure computerized database, ensuring full allocation concealment.隨機分派經由安全的電腦化資料庫執行,確保完整的分配隱匿。
- 理由:作者明言以安全電腦化資料庫確保完整分配隱匿,納入前無法預知分派結果。
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:Chance imbalances between treatment arms included a higher proportion of women (46% for acupuncture v 23% for control) and somewhat poorer Constant-Murley scores (41.9 for acupuncture v 48.1 for control) ... the use of any type of pain medication before enrollment was more common among controls (53% v 14%).兩組間的偶發不均衡包括針灸組女性比例較高(46% 對照 23%)、基線 Constant-Murley 分數略差(41.9 對照 48.1),且對照組入組前使用止痛藥較普遍(53% 對 14%)。
- 理由:確有數項基線不均衡(性別、基線分數、用藥),但隨機方法健全(分層+區組+完整隱匿),作者將其歸為偶發不均衡;此類不平衡未暗示隨機化程序本身出問題,故 PN。基線分數與用藥已於主要分析以共變數校正。
Domain 2:偏離預定介入 — 判斷:Low
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:是(Yes — 有明確證據支持)
- 原文引述:The study was a prospective, open-label, randomized controlled trial ... Both the Numerical Rating Scale of Pain and the Xerostomia Inventory were completed by the patient who was not blinded to treatment arm.本研究為前瞻性、開放標籤的隨機對照試驗……疼痛數字評分量表與口乾量表皆由未對分組設盲的病人填寫。
- 理由:開放標籤設計,病人明確知道自己被分到針灸或常規照護。
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:是(Yes — 有明確證據支持)
- 原文引述:The study was a prospective, open-label, randomized controlled trial ... All treatments were performed at the MSKCC Integrative Medicine Center by staff acupuncturists.本研究為前瞻性、開放標籤的隨機對照試驗……所有治療皆由整合醫學中心的專任針灸師執行。
- 理由:開放標籤,施治的針灸師與提供常規照護的照顧者皆知道病人分組。
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:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:No formal limitation was placed on other therapy (eg, rehabilitation) the patient could receive while getting acupuncture. However, any additional treatments were recorded. ... Twelve patients (40%) on the control arm pursued such options, whereas 20% of patients on the acupuncture arm pursued these options.對病人在接受針灸期間可另外接受的治療(如復健)未設正式限制,但任何額外治療皆有記錄……對照組有 12 位(40%)病人尋求此類選項,針灸組則為 20%。
- 理由:此為務實(pragmatic)設計,允許共同治療本即介入定義的一部分;雖兩組共同治療使用比例不同,但無證據顯示存在因『試驗情境』而生、超出常規實務的協議偏離。以指派效果(ITT)評估,此屬設計預期而非偏誤性偏離,故 PN。
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:是(Yes — 有明確證據支持)
- 原文引述:All analyses were conducted using Stata 9.2 and were based on the intent-to-treat principle for all patients who had evaluable outcome data.所有分析皆以 Stata 9.2 進行,並對所有具可評估結局資料的病人採意向治療原則。
- 理由:分析明言採意向治療原則(依隨機分組分析),為估計指派效果的適當分析,故 Y。(可評估資料的限制屬缺失資料議題,於 Domain 3 處理。)
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:不需評估(NA)
- 原因:條件題:依 2.6 的答案,本題依 RoB 2 官方路由不需評估(NA)。
Domain 3:缺失結局資料 — 判斷:Low
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:Patients eligible, consented, and randomly assigned (n = 70) ... Completed study (n = 28) ... Completed study (n = 30) ... A total of 58 randomly assigned patients (28 ... acupuncture and 30 ... usual care) were evaluable for the primary end point.符合資格、簽署同意並隨機分派者 n=70……完成研究者針灸組 n=28、對照組 n=30……共 58 位隨機分派病人(針灸 28、常規照護 30)可評估主要結局。
- 理由:CONSORT 流程圖顯示隨機分派 70 人,最終 58 人(約 83%)具主要結局資料,約 17% 缺失;未達『幾乎全部』,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:是(Yes — 有明確證據支持)
- 原文引述:To evaluate the impact of missing follow-up data on our findings, we conducted multiple imputations using the ice command in Stata ... a second analysis to adjust for missing data ... under the conservative assumption that treatment had no effect in patients with missing data. ... Adjusting for missing data by multiple imputation had little effect on our results.為評估追蹤缺失資料對結果的影響,我們以 Stata 的 ice 指令進行多重插補……並在『缺失病人治療無效』的保守假設下做第二種校正分析……以多重插補校正缺失資料對結果影響甚微。
- 理由:作者進行多重插補(含保守假設模型)之敏感度分析,結果幾乎不變,提供結果未受缺失資料偏誤的證據,故 Y。
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:不需評估(NA)
- 原因:條件題:依 3.2 的答案,本題依 RoB 2 官方路由不需評估(NA)。
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:不需評估(NA)
- 原因:條件題:依 3.3 的答案,本題依 RoB 2 官方路由不需評估(NA)。
Domain 4:結局測量 — 判斷:Some concerns
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 primary outcome measure was a composite score of pain, function, and activities of daily living provided by the Constant-Murley instrument (score of 0 to 100). ... The scale has low levels of observer error, is often used as a gold standard against which other shoulder instruments are measured, shows good responsiveness.主要結局為 Constant-Murley 量表所提供的疼痛、功能與日常生活活動複合分數(0-100)……該量表觀察者誤差低、常被當作評估其他肩部量表的黃金標準、反應性良好。
- 理由:Constant-Murley 為經驗證、觀察者誤差低、被視為黃金標準的肩部量表,量測方法適當,故 N。
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:Constant-Murley scores were assessed by a researcher who was unaware of the patient's treatment allocation.Constant-Murley 分數由不知病人治療分派的研究人員評估。
- 理由:兩組皆以同一量表、由同一位盲化研究人員在相同時點評估,量測方式兩組間無系統差異,故 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:可能是(Probably yes — 綜合判斷後傾向「是」)
- 原文引述:Constant-Murley scores were assessed by a researcher who was unaware of the patient's treatment allocation. ... data on the pain and activities of daily living subscales were excluded because these seem more susceptible to possible biased reporting by unblinded trial participants.Constant-Murley 分數由不知病人治療分派的研究人員評估……疼痛與日常生活活動子量表的資料被排除,因為這些子量表較易受未設盲受試者的偏誤性回報影響。
- 理由:評估量表由盲化第三方研究人員執行(客觀的活動度、肌力子量表尤然),此為降低偏誤的正面設計;但複合分數中疼痛(15 分)與 ADL(20 分)子量表仰賴未設盲病人自陳,作者亦承認其易受偏誤回報影響。就整體主要結局而言,部分評估來源(病人)知情,故取較保守之 PY 而非 N。
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:可能是(Probably yes — 綜合判斷後傾向「是」)
- 原文引述:we reanalyzed the Constant-Murley data limiting our analysis to those components of the scale that lend themselves to more objective assessment (ie, range of motion and power; data on the pain and activities of daily living subscales were excluded because these seem more susceptible to possible biased reporting by unblinded trial participants).我們重新分析 Constant-Murley 資料,僅限於較適合客觀評估的部分(即活動度與肌力;疼痛與 ADL 子量表被排除,因較易受未設盲受試者的偏誤性回報影響)。
- 理由:疼痛與 ADL 子量表屬主觀、由知情病人自陳,其評估可能受分組知情影響,故 PY。
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:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:The difference between groups was of a similar magnitude (10.1 points) and remained significant (P = .037), favoring the acupuncture arm.(僅取客觀成分時)組間差異幅度相近(10.1 分)且仍顯著(P=.037),偏向針灸組。
- 理由:以完全客觀成分(活動度+肌力,且評估者盲化)重新分析仍得相近且顯著的效果,顯示結果並非主要由知情自陳偏誤所驅動,故『很可能受影響』判為 PN。
Domain 5:選擇性報告 — 判斷:Low
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:可能是(Probably yes — 綜合判斷後傾向「是」)
- 原文引述:The Constant-Murley score, a composite measure of pain, function, and activities of daily living, was the primary outcome measure. ... we estimated that 58 evaluable patients would be needed for an 80% power to detect a difference between groups of 12 points on the Constant-Murley scale. ... Clinical Trials repository link available on JCO.org.Constant-Murley 分數(疼痛、功能與日常生活活動的複合量表)為主要結局……我們估算需 58 位可評估病人,才有 80% 檢定力偵測 Constant-Murley 量表 12 分的組間差異……臨床試驗登錄連結見 JCO.org。
- 理由:主要結局、主分析模型(ANCOVA 校正基線與手術類型共變數)與樣本數計算皆於方法中事前明訂,且為登錄試驗,主要結果與事前指定一致,故 PY。(無法取得完整統計計畫原件,未給 Y。)
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:The Constant-Murley score, a composite measure of pain, function, and activities of daily living, was the primary outcome measure.Constant-Murley 分數為主要結局量表。
- 理由:主要結局為單一事前指定的量表與時點(療程結束),修正版 Constant-Murley 與純客觀成分皆明示為敏感度分析、透明呈現,非從多量表中挑選最佳者,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:All analyses were conducted using Stata 9.2 and were based on the intent-to-treat principle ... conducted by analysis of covariance with group, baseline score, and type of surgical procedure ... as covariates.所有分析皆以 Stata 9.2 進行、採意向治療原則……以共變數分析(含分組、基線分數、手術類型為共變數)進行組間比較。
- 理由:主要結局採事前指定的 ANCOVA 模型;多重插補、事後用藥校正等皆額外透明呈現,主要結果非從多種分析中擇優報告,故 PN。