COCHRANE ROB 2 · 偏誤風險評讀
每週 CSBMs 增加≥3 次達標率(追蹤期末,中藥 40% vs 安慰劑 19%)
← 回到「便秘」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:Some concerns(有些疑慮)
- +Domain 4 結局測量:Low(低風險)
- +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
本試驗為前瞻註冊(NCT 03186079)之隨機、雙盲、安慰劑對照試驗,N=200(中藥 100、安慰劑 100),以小積導滯湯治療 4–14 歲兒童功能性便秘。隨機序列以 Random Allocation Software 產生、由未參與招募者執行,安慰劑外觀完全配對且盲性檢測成功,基線兩組均衡,故隨機過程(領域一)與偏離介入(領域二,採適當 ITT 分析)偏誤風險低。結局衡量採每週完全自主排便(CSBMs)日誌、評估者盲化,測量偏誤(領域四)與報告選擇偏誤(領域五)風險亦低。主要疑慮在遺漏資料(領域三):實際完成僅中藏 82%、安慰劑 71%,兩組失訪不對稱(16 vs 9),且退出原因含『無改善』並有家長疑似分到安慰劑而停藥之情形,作者僅以最後觀測值結轉填補、未做敏感度分析,使遺漏可能依結局真值而定,屬 Some concerns。整體而言主要偏誤來源為差別性失訪與結轉填補處理。
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:According to the sequence generated by Random Allocation Software (version 1.0.0), the grouping was randomized in a ratio of 1:1, which was performed by a nonrecruited researcher.依據 Random Allocation Software(1.0.0 版)所產生的序列,以 1:1 比例進行隨機分組,由一位未參與招募的研究者執行。
- 理由:明確使用電腦隨機分配軟體(Random Allocation Software)產生 1:1 序列,屬含隨機成分的序列生成方法,符合『Yes』。
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 grouping was randomized in a ratio of 1:1, which was performed by a nonrecruited researcher. The masking process was double-blind, and the appearance of the placebo was the same as that of the CHM group.隨機分組以 1:1 比例進行,由一位未參與招募的研究者執行;遮盲過程為雙盲,安慰劑外觀與中藥組相同。
- 理由:分組由『未參與招募的研究者』執行,分配與招募人員分離;加上安慰劑與中藥外觀、重量、顏色、味道、氣味完全相同(藥物容器形式的遮盲),可合理判斷分配序列在納入與分派前被適當隱匿。雖未明述中央隨機或信封細節,但綜合雙盲與相同外觀的藥物容器設計,判為『Probably yes』。
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:否(No — 有明確證據支持)
- 原文引述:Age (yr) 6.61 ± 2.79 / 6.24 ± 2.20 ... Bowel movements per wk 2.99 ± 1.47 / 2.74 ± 1.27 ... Male 48 (48%) / 51 (51%)年齡 6.61±2.79 對 6.24±2.20;每週排便次數 2.99±1.47 對 2.74±1.27;男性 48% 對 51%。
- 理由:Table 2 基線人口學與便秘相關變項(性別、年齡、病程、體重、每週排便、便型、疼痛、滯便、遺糞)兩組相近,組別人數皆為 100:100,無顯示隨機過程出問題的失衡,判為『No』。
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:否(No — 有明確證據支持)
- 原文引述:The masking process was double-blind, and the appearance of the placebo was the same as that of the CHM group. ... Patients, researchers, evaluators, and sponsors did not know which patients received which treatments.遮盲過程為雙盲,安慰劑外觀與中藥組相同;病患、研究者、評估者與贊助者皆不知道哪位病患接受何種治療。
- 理由:雙盲安慰劑對照,安慰劑於外觀、重量、顏色、味道、氣味完全配對;盲性檢測顯示約 68.63% 受試者無法判斷分組、正確猜對者多基於療效而非藥物特徵,顯示參與者大致未察覺分組,判為『No』。
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:否(No — 有明確證據支持)
- 原文引述:Patients, researchers, evaluators, and sponsors did not know which patients received which treatments. ... the grouping ... was performed by a nonrecruited researcher.病患、研究者、評估者與贊助者皆不知道哪位病患接受何種治療;分組由未參與招募的研究者執行。
- 理由:藥物外觀完全配對且分組由獨立研究者執行,照護與給藥/評估人員亦被納入雙盲(明列 researchers、evaluators 不知分組),判定提供介入之人員未察覺分組,為『No』。
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:是(Yes — 有明確證據支持)
- 原文引述:Intent-to-treat analysis was performed, and outcomes at the last follow-up were considered the final outcomes in all the participants even if they were dropped. ... Switch to analysis (n = 100) / Switch to analysis (n = 100)進行意向治療(ITT)分析,所有受試者(即使已退出者)皆以最後一次追蹤的結果作為最終結果;分析人數兩組各 100。
- 理由:明確採 ITT,兩組各 100 名(全部隨機者)納入分析、依隨機分派分組,無依實際接受治療重新分組、亦無排除合格隨機者,符合適當的意向治療分析,判為『Yes』。
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: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:Overall, 153 subjects completed the trial, 82 in the CHM group (82%) and 71 in the placebo group (71%). ... outcomes at the last follow-up were considered the final outcomes in all the participants even if they were dropped.整體 153 名完成試驗,中藥組 82 名(82%)、安慰劑組 71 名(71%);退出者亦以最後一次追蹤結果作為最終結果。
- 理由:雖宣稱兩組各 100 納入 ITT,但對退出者係以『最後一次追蹤結果結轉』填補(屬填補資料,依本題規定應視為遺漏)。實際具觀測結果者為中藥 82/100、安慰劑 71/100,遺漏比例約 18% 與 29%,並非全部或近乎全部具可用結局資料,判為『Probably no』。
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:outcomes at the last follow-up were considered the final outcomes in all the participants even if they were dropped. ... Lost to follow-up (n =9) ... Lost to follow-up (n =16)退出者以最後追蹤結果結轉作為最終結果;中藥組失訪 9 人、安慰劑組失訪 16 人。
- 理由:作者僅以『最後觀測值結轉』方式處理退出者,未提供校正遺漏偏誤的分析或敏感度分析;依本工具,結轉式填補不能視為校正了遺漏資料偏誤,故無證據顯示結果未受遺漏資料影響,判為『Probably no』。
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:是(Yes — 有明確證據支持)
- 原文引述:Discontinued intervention (n=9) 6 due to low compliance ... Discontinued intervention (n =13) 5 due to no improvement ... 5 due to low compliance中藥組停藥 9 人(其中 6 人因順從性低);安慰劑組停藥 13 人(5 人因無改善、5 人因順從性低)。
- 理由:退出原因包含『無改善 no improvement』(安慰劑組 5 人),此與結局真值(仍便秘)直接相關;且作者自承『家長若認為被分到安慰劑組可能停藥,導致失訪率升高』。故結局遺漏與否可能取決於其真值,判為『Yes』。
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 =9) ... Lost to follow-up (n =16) ... 5 due to no improvement [placebo] ... because of the bitterness of traditional CHM, children may refuse to take the medicine or some parents may stop using the medicine if they perceive they may be in the placebo group, leading to a higher rate of lost interviews.中藥組失訪 9 人、安慰劑組失訪 16 人;安慰劑組有 5 人因『無改善』退出;作者指出中藥味苦、家長若覺得被分到安慰劑組可能停藥,使失訪率升高。
- 理由:雖兩組遺漏比例與退出原因有差異(安慰劑組較多、且含『無改善』),存在遺漏依真值而定的可能;但本研究主要療效終點為達標『有效/無效』二分類反應率,事件數相對於遺漏數不算極小,且觀察到的療效差距大(40% vs 19%)。在最惡與最佳情境邊界內,差異方向不易翻轉,故傾向判為遺漏很可能依真值而定的程度尚未達『High』門檻,判為『Probably no』(即定位於 Some concerns 而非 High)。
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:Complete spontaneous bowel movements (CSBMs) referred to the frequency of defecation without drugs or other auxiliary methods. ... Increased complete spontaneous bowel movements ≥3 per week from baseline完全自主排便(CSBMs)指未用藥物或其他輔助方式下的排便次數;以自基線增加每週 ≥3 次 CSBMs 為指標。
- 理由:以每週完全自主排便次數變化作為療效衡量,是兒童便秘臨床試驗公認且合適的結局衡量方式,對可能的介入效果具敏感性,判為『No』(衡量方法並非不當)。
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:否(No — 有明確證據支持)
- 原文引述:The symptoms before treatment, 8 weeks after treatment, and 12 weeks after follow-up were recorded ... Four visits were arranged for each participant. ... participants were required to record defecation details ... daily.於治療前、治療後 8 週、追蹤 12 週記錄症狀;每位受試者安排四次回診,並每日記錄排便細節。
- 理由:兩組採相同的排便日誌記錄方式、相同時間點與相同回診安排衡量結局,無證據顯示量測方法或時點在組間不同,判為『No』。
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:否(No — 有明確證據支持)
- 原文引述:Patients, researchers, evaluators, and sponsors did not know which patients received which treatments. ... Among the 153 participants who finished the protocol, 48 (31.37%) correctly guessed the group they were assigned to病患、研究者、評估者與贊助者皆不知分組;完成方案的 153 人中僅 48(31.37%)猜對分組。
- 理由:本主要結局為病患/家長自評(參與者即為結局評估者),且雙盲成功(多數無法判斷分組、猜對者多基於療效);評估者(含參與者與評估人員)對分組保持盲性,判為『No』。
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: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 study protocol was registered prospectively with https://clinicaltrials.gov/(NCT 03186079). ... Main outcome measures were complete spontaneous bowel movements and satisfaction with bowel function.研究計畫前瞻性註冊於 ClinicalTrials.gov(NCT 03186079);主要結局衡量為完全自主排便與排便滿意度。
- 理由:試驗於 ClinicalTrials.gov 前瞻性註冊(NCT 03186079),且全文一致以 CSBMs 與排便滿意度為預先指定之主要結局、分析方法(ITT、χ² 檢定)事先載明。雖未取得完整統計分析計畫核對,但有前瞻註冊且結局與分析方向一致,合理判為『Probably yes』。
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:Main outcome measures were complete spontaneous bowel movements and satisfaction with bowel function. ... At the end of the treatment, there were 46 cases ... and 26 cases in the placebo group ... 40 patients in the CHM group and 19 patients in the placebo group were found at the end of the follow-up.主要結局衡量為完全自主排便與排便滿意度;CSBMs 達標於治療後(46 vs 26)與追蹤期末(40 vs 19)兩時點皆完整報告。
- 理由:兩個並列的主要結局(CSBMs、排便滿意度)皆於治療後與追蹤期末兩時點完整報告,未見只挑選對介入有利的單一量表/時點,判為『Probably no』。
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:Intent-to-treat analysis was performed ... The χ2 test was used for counting data, and the rank-sum test was used for ranking data. ... All tests are reported as double-sided tests.採意向治療分析;計數資料用 χ² 檢定、等級資料用秩和檢定,所有檢定皆為雙側。
- 理由:分析方法單純(ITT + χ² 雙側檢定),所有主要與次要結局於兩個時點皆一致報告,無多重模型擇優報告之跡象,判為『Probably no』。