COCHRANE ROB 2 · 偏誤風險評讀
人工淚液使用
← 回到「乾燥症」的圖表
本試驗為交叉(crossover)設計。RoB 2 設有交叉試驗專屬變體(會額外評估洗脫期是否足夠、殘留 carryover 效應、是否採配對分析等);本評讀以平行組版工具近似,未完整納入上述交叉專屬考量,解讀時請留意。
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:High(高風險)
- +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
這篇研究在「人工淚液使用次數」這個結果上的整體偏誤風險被評為「高」。
與淚液分泌相同,本試驗的隨機化與雙盲都做得好(隨機化風險「低」);人工淚液次數雖是病人自填,但病人有被有效設盲(安慰劑仿製真藥、病人不知分組),所以測量偏誤風險評為「低」。
但這個結果有兩個較高的風險。其一是「分析方式」:只用 57/70 名完成者做 per-protocol 分析、非 ITT(風險「高」)。其二是「缺失資料」:人工淚液次數是反映眼乾感受的自評指標,而部分退出原因(不良事件、嚴重不良事件、藥物順從度)可能與病情、也就是這個結果本身相關,缺失因此可能取決於真實值,研究又沒有做敏感度分析來檢驗(風險「高」)。選擇性報告同樣因缺研究計畫書而「有些疑慮」。
總結:自評結果在高比例退出、又非 ITT 分析下,整體可信度受到影響(另見上方交叉設計警語)。
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:A permuted-block randomization scheme was employed by our team's statistician (JRL) to generate randomization codes, and equal numbers of patients were assigned to each group.本研究團隊的統計師(JRL)採用排列區組(permuted-block)隨機化方案產生隨機分配碼,且每組分配相同人數的病人。
- 理由:報告明確指出由統計師以排列區組隨機化方案產生分配碼,屬含隨機成分的序列產生方法,符合官方判準。
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 randomization table was then sent to the pharmaceutical company that manufactured the SS-1 and placebo formulations... Due to the double-blind trial design, none of the patients, doctors, pharmacists, care providers, or investigators knew which patients were in which group.隨機分配表送交製造 SS-1 與安慰劑的藥廠,再轉送中醫藥局;由於雙盲設計,病人、醫師、藥師、照護者與研究者均不知道分組。
- 理由:分配表由獨立統計師產生並交藥廠/藥局保管,安慰劑外觀、味道、氣味與 SS-1 一致,雙盲下醫師與研究者均不知分組;雖未逐項描述容器封緘,但綜合可合理判斷分配在納入前已隱蔽,故以判斷給「可能是」。
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:None of the baseline characteristics differed significantly between the two groups, and none of the patients received cyclosporine, cevimeline, pilocarpine, rituximab, or other biological agents before enrolling in this trial.兩組之間沒有任何基線特徵達統計顯著差異,且所有病人在納入前均未接受 cyclosporine、cevimeline、pilocarpine、rituximab 或其他生物製劑。
- 理由:Table 1 顯示兩組所有基線變項 p 值皆 >0.05、無顯著差異;組間人數(36 vs 34)與 1:1 分配相符,差異可由機遇解釋,亦無過度相似之疑。
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:否(No — 有明確證據支持)
- 原文引述:the placebo contained 1/100 of the SS-1 dose used for treatment, which was necessary to mimic the color, taste, and smell of SS-1... none of the patients, doctors, pharmacists, care providers, or investigators knew which patients were in which group.安慰劑含 SS-1 治療劑量的 1/100,以模擬其顏色、味道與氣味,維持雙盲;病人不知道自己的分組。
- 理由:安慰劑刻意以玉米澱粉加色素加 1/100 劑量匹配 SS-1 的顏色、味道、氣味,並明文敘述病人不知分組,屬有效的對病人盲化。
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:Due to the double-blind trial design, none of the patients, doctors, pharmacists, care providers, or investigators knew which patients were in which group.由於雙盲設計,病人、醫師、藥師、照護者與研究者皆不知道分組。
- 理由:報告明確指出醫師、藥師與照護者(提供介入者)皆不知分組,故執行者亦盲。
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:2.1/2.2 皆為「否」(受試者與執行者均已盲化),無需評估因試驗情境造成的偏離。
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:上一題 2.3 不需評估,故略過。
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:上一題 2.4 不需評估,故略過。
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:The trial outcomes were analyzed using the per-protocol model, and outcome differences between the SS-1–Placebo and Placebo–SS-1 groups were evaluated through analysis of variance in a 2 × 2 crossover design.試驗結果以 per-protocol(符合方案)模型分析,兩組差異以 2×2 交叉設計的變異數分析評估。
- 理由:作者明確採 per-protocol 分析:70 名隨機者僅以完成者(30+27=57)進入分析,排除隨機後未完成者。此非適切的 ITT 分析。
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:是(Yes — 有明確證據支持)
- 原文引述:a total of 70 patients remained for randomization... the crossover analysis proceeded with the 30 and 27 patients who remained in the Placebo–SS-1 and SS-1–Placebo groups, respectively.共 70 名病人進入隨機分組……交叉分析最終以 30 名與 27 名(共 57 人)進行。
- 理由:70 名隨機者中有 13 名(約 19%)因退出/停藥被排除,且排除原因與安全性/順從度(可能與預後相關)有關,足以對結果造成實質影響。
Domain 3:缺失結局資料 — 判斷:High
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:a total of 70 patients remained for randomization... the crossover analysis proceeded with the 30 and 27 patients... The trial outcomes were analyzed using the per-protocol model共 70 名病人進入隨機分組……交叉分析以剩餘的 30 名與 27 名病人進行……試驗結果以 per-protocol 模型分析。
- 理由:70 名隨機分組僅以 57 名完成者分析,缺失約 18.6%,遠超連續型結局約 95% 的門檻;人工淚液使用為自填連續變項,缺失比例大且未對全部隨機者提供結局。
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:The trial outcomes were analyzed using the per-protocol model試驗結果以 per-protocol 模型分析。
- 理由:僅以 per-protocol 分析,未提供 ITT 或敏感度分析、未插補、未比較完成者與缺失者,無證據顯示結果未受缺失影響。
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:six members of the Placebo–SS-1 group and seven of the SS-1–Placebo group left the trial by their own request or due to AEs, SAEs, and low drug accountabilityPlacebo–SS-1 組 6 人、SS-1–Placebo 組 7 人因自身要求或因 AE、SAE 及藥物服用率過低而退出。
- 理由:人工淚液使用反映眼乾感受;部分退出原因(AE、SAE、藥物耐受/順從)可能與病情、即此結局真值相關,故缺失可能取決於真實值。
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:可能是(Probably yes — 綜合判斷後傾向「是」)
- 原文引述:left the trial by their own request or due to AEs, SAEs, and low drug accountability因自身要求或因 AE、SAE 及藥物服用率過低而退出。
- 理由:各組退出原因與人數不盡對稱,且退出常與治療反應/病情(眼乾程度)相關,自填人工淚液次數的缺失很可能取決於結局真實值。
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:可能否(Probably no — 綜合判斷後傾向「否」)
- 原文引述:The trial's secondary endpoints were 1) self-reported numbers of artificial-tear applications per day, ranging from 0 to 30 or more, with more use indicating more eye dryness次要終點為每日自我回報人工淚液使用次數(0 到 30 次以上),使用越多代表眼睛越乾。
- 理由:以每日自我回報人工淚液次數作為眼乾代理指標,對療效變化具合理敏感度(觀察到 −1.38 次/日的顯著差異),屬常見且可接受的測量方式。
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:Due to the double-blind trial design, none of the patients... knew which patients were in which group.由於雙盲設計,所有病人均不知道分組。
- 理由:兩組皆以相同方式(病人自填、相同時間點)收集,且雙盲、安慰劑相配,就診與測量機會一致,無證據顯示組間測量方式有別。
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:Due to the double-blind trial design, none of the patients... knew which patients were in which group... the placebo... mimic the color, taste, and smell of SS-1雙盲設計下病人不知分組;安慰劑模擬 SS-1 的顏色、味道與氣味。
- 理由:此為病人自評,評估者即病人本人;試驗雙盲、安慰劑外觀/味道/氣味相配,病人被有效盲化,故評估者不知所接受的介入。
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:4.3 為「可能否」(病人已被有效盲化),無需評估知情影響。
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:上一題 4.4 不需評估,故略過。
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:This clinical trial was registered at ClinicalTrials.gov (NCT02110446).本臨床試驗已於 ClinicalTrials.gov 註冊(NCT02110446)。
- 理由:試驗有註冊,但全文未提供解盲前定稿的統計分析計畫或詳細計畫書,無法確認人工淚液此次要結果是否依事前指定、解盲前定稿之計畫分析。
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 trial's secondary endpoints were 1) self-reported numbers of artificial-tear applications per day... We evaluated the above parameters at four time-points (V1–V4).次要終點為自我回報每日人工淚液使用次數……於 V1–V4 四個時間點評估。
- 理由:此結果域只有單一既定測量方式(自填每日次數 0–30+),文中明列為次要終點並於四個固定時間點測量,無證據顯示在此測量域內擇優挑選報告。
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:The trial outcomes were analyzed using the per-protocol model... in a 2 × 2 crossover design... presented as means with 95% confidence intervals試驗結果以 per-protocol、2×2 交叉設計之變異數分析評估,並以平均值及 95% 信賴區間呈現。
- 理由:所有結果統一採單一既定分析方法,未見對同一測量以多種分析計算後僅擇優報告之跡象。