Efficacy and Economic Assessment of Conventional Ventilatory Support versus Extracorporeal Membrane Oxygenation for Severe Adult Respiratory Failure (CESAR)

• p. 8 Table 6 column header

  Probability of survival to 6 months

Judge's note: Table 6 image: conventional 0·47 matches 41/87, the primary outcome given in the abstract. Table 3 shows that 45 (50%) conventional patients did not die. The text describes Table 6 as cost per additional survivor without severe disability, so the column label is incomplete. For ECMO the two quantities coincide (57/90).

Second check: another model agreed

• p. 12 Reference 26

  Health Care Anal 2006; 4: 273–83

Only the authors can say what is right here.

Judge's note: Rests on outside knowledge: that volume 4 of Health Care Analysis appeared in 1996, not 2006. The image of ref 26 on p12 confirms the printed text 'Health Care Anal 2006; 4: 273–83'. Nothing inside the files contradicts it, so this goes to the authors as a query.

Second check: another model agreed

• p. 12

  Bersten AD, Edibam C, Hunt T, Moran J; Australian and New Zealand Intensive Care Society C

Only the authors can say what is right here.

Judge's note: Rests on outside knowledge: the Crossref record for doi:10.1164/ajrccm.165.4.2101124 gives volume 165. The image of ref 2 on p12 confirms the printed volume '15'. A volume of 15 is implausible for this journal in 2002, but this cannot be checked from the files.

Second check: another model agreed, code check

• p. 1 Abstract, Findings

  68 (75%) patients actually received ECMO

• p. 7 Results, para 2

  68 (75%) patients considered for treatment with

• p. 3 Table 2, Treatment by ECMO

  68 (76%)

Judge's note: I checked the abstract (p1), Results paragraph 2 (p7) and Table 2 (p3). The same count and denominator (n=90) are printed with two different percentages. The rounded value is 76%.

Second check: another model agreed

• p. 1 Summary, Methods

  severe (Murray score >3·0 or pH <7·20)

• p. 2 Methods, Participants

  of 3·0 or higher

• p. 1 Summary, Interpretation

  whose Murray score exceeds 3·0 or who have a pH of less than 7·20

• p. 2 Methods, Participants

  of 3·0 or higher, or uncompensated hypercapnoea

Judge's note: I checked the Summary Methods and Interpretation (p1), the Methods Participants section (p2) and the Figure 1 exclusion box (p2). The eligibility threshold is ≥3·0 in the Methods and the figure but strictly >3·0 in the abstract. The two differ for a patient scoring exactly 3·0.

Second check: another model agreed

• p. 9 Table 7, Scenario 3

  US$31 112 (12 317 to 95 507)

• p. 1 Summary

  7622–59 200

Only the authors can say what is right here.

Judge's note: Table 7 and Summary images. The point estimate and lower bound convert at 1·616, and 95 507/1·616 = 59 101, which suggests 59 100 against 59 200. The conversion is not exact everywhere, though: scenario 4 gives 9389×1·616 = 15 173 against the printed 15 183. So it is unclear which figure is off, and only the authors can say.

Second check: another model agreed

• p. 5 Table 4, footnote ‡

  Data for 33 patients receiving extracorporeal membrane oxygenation

Judge's note: Table 3 gives 33 deaths in the ECMO-allocated group (n=90). The Results say 3 died before transport and 2 in transit, 17 had gentle ventilation and 14 of them survived, and 68 received ECMO, of whom 43 survived. So only 25 of the deaths were among patients who received ECMO. The column header itself refers to the allocated group, so the footnote wording names the wrong population.

Second check: another model agreed

• p. 7 Results, para 1

  of whom 180 were enrolled from 68 centres

• p. 11 Participating centres list

  For all centres that recruited patients

Only the authors can say what is right here.

Judge's note: I counted the entries on the page 11 image and in the text layer: 69 hospitals with bracketed counts, totalling 180. Several sites could reasonably count as one centre (for example the three Leicester hospitals including Glenfield, or Walsgrave and Warrington, which list the same staff), but the paper does not say so. The authors should clarify how centres were counted.

Second check: another model agreed

• p. 10 Discussion, right column para 1

  additional average cost per patient of referral for ECMO is more than double the average cost of treatment with conventional manage-

• p. 8 Table 6, Scenario 1

  £73 979

• p. 10 Discussion, right column, para 1

  the additional average cost per patient of referral for ECMO is more than double the average cost of treatment with conventional manage-ment

• p. 8 Results text

  Mean health-care costs per patient were more than twice as high

Judge's note: Discussion image (page 10) checked against Table 6 scenario 1 (£73 979 against £33 435). The Results sentence 'Mean health-care costs per patient were more than twice as high' is correct. The Discussion misdescribes the incremental cost.

Second check: another model agreed

• p. 9 Table 7, scenarios 1–4

  3·66 (··)

• p. 9 Table 7, scenarios 3 and 4

  3·66 (··) £48 533 £19 252

Only the authors can say what is right here.

Judge's note: Table 7 image. The table is labelled 'bootstrap estimates', so a mean of bootstrap ratios could differ from the ratio of means. However, the footnote defines the ICER as the ratio of the differences, and the gaps (45% in scenario 3, 22% in scenario 4) are large. For scenario 2, 26 772/732 818 implies a QALY gain of 0·037, which would round to 0·04 rather than the printed 0·03. Only the authors can say how the columns relate.

Second check: another model agreed

• p. 12 Reference 15

  Peek GJ, Tirouvopaiti R, Firmin RK.

• p. 1 Author list

  Ravindranath Tiruvoipati

• p. 12 Ref 15

  Tirouvopaiti R

Judge's note: Looked at the reference list on p12 and the byline on p1. The byline reads 'Ravindranath Tiruvoipati' and ref 15 reads 'Tirouvopaiti R'. The two spellings are inconsistent inside the paper.

Second check: none (judge only)

• p. 2 Methods, Participants

  for every paricipating centre

• p. 2 Methods, Participants

  every paricipating centre

• p. 8 Results

  consideration for treament

• p. 2

  paricipating centre

Judge's note: Both spellings are visible in the page images, so they are not extraction artefacts.

Second check: none (judge only)

• p. 10 Discussion, left column para 2

  which in uninterpretable

• p. 10 Discussion, left column, para 2

  then becomes death or ECMO, which in uninterpretable.

• p. 2 Methods

  paricipating centre

• p. 8 Results

  consideration for treament

Judge's note: Seen on the page 10 image (crop C042-b). The other misspellings listed in one reviewer note also appear in the text layer ('paricipating' on page 2, 'Carergiver' in Table 5). 'treament' is handled separately under C035.

Second check: none (judge only)

• p. 3 Table 1 footnote §

  acute milliary tuberculosis

Judge's note: The misspelling is visible in the Table 1 footnote on p3.

Second check: none (judge only)

• p. 3 Table 2 footnote ||

  The remaining patients was deemed too sick to move by the transport team.

• p. 3 Table 2 footnote ||

  Three of these patients died

Judge's note: The footnote says four patients were not transferred and three of them died (before transfer, family refusal, too sick to move). The final sentence covers the one remaining survivor, so the content is coherent. Only the number agreement is wrong.

Second check: none (judge only)

• p. 3 Table 1, positive end-expiratory pressure

  13·7 (9.6)

• p. 3 Table 1

  19·68 (6·3)

Judge's note: In the p3 image, the point in 9.6 sits on the baseline, unlike the mid-dot in 13·7. This is cosmetic. The APACHE precision point is handled under C012.

Second check: code check

• p. 3 Table 2, time between randomisation and treatment

  6·1 (4·0–7 ·1)§

Judge's note: The space is visible in the p3 image, so it is not just an extraction artefact.

Second check: none (judge only)

• p. 3 Table 1, APACHE II score

  19·68 (6·3)

Judge's note: The p3 image confirms the inconsistent precision. This is cosmetic.

Second check: none (judge only)

• p. 4 Methods, ECMO management

  Servo 300 ventilators (Solma, Sweden)

• p. 4 Study design, ventilators

  Solma, Sweden

Only the authors can say what is right here.

Judge's note: Rests on outside knowledge: Siemens-Elema/Maquet Servo ventilators are made in Solna, Sweden. The p4 image does read 'Solma, Sweden'.

Second check: none (judge only)

• p. 4 Methods, ECMO management

  Medizintechink

• p. 4 Study design, ECMO protocol

  Medos Medizintechink

Only the authors can say what is right here.

Judge's note: Rests on outside knowledge: the manufacturer's name is Medos Medizintechnik. The page 4 image clearly reads 'Medizintechink', which is not a German word. Both crops show the same single occurrence.

Second check: none (judge only)

• p. 6 Methods, economic

  Organisation of Economic Co-operation and Development

• p. 13 Ref 45

  Organisation for Economic Cooperation and Development

Only the authors can say what is right here.

Judge's note: Rests on outside knowledge: the organisation's official name is 'Organisation for Economic Co-operation and Development'. Both images show the two forms, so the paper is inconsistent with itself.

Second check: none (judge only)

• p. 7 Table 5

  Carergiver strain index (0–12)

• p. 7

  Carergiver strain index

Judge's note: The page 7 image and crop clearly show 'Carergiver'. The Methods use 'carer strain index'.

Second check: none (judge only)

• p. 8 Results, para 1

  consideration for treament

• p. 8

  treament

Judge's note: Seen on the page 8 image (crop C032-a).

Second check: none (judge only)

• p. 8 Results, cost-effectiveness acceptability

  GP pounds (2005)

• p. 8 Results, cost-effectiveness acceptability

  willingness to pay in GP pounds (2005)

• p. 8

  in GP pounds (2005)

Judge's note: Seen on the page 8 image (crop C036-b). The paper uses GP for general practitioner, so 'GP pounds' is a typographical error.

Second check: none (judge only)

• p. 11 CESAR trial collaboration list

  It support: A King, M Bennett.

• p. 11 Other hospitals list

  Goodmayes Hospital, Illford

Only the authors can say what is right here.

Judge's note: Rests on outside knowledge: the town is spelled Ilford. 'It support' for the acronym IT is evident from the page itself. Both are seen on the page 11 image.

Second check: none (judge only)

• p. 12 Ref 13

  severe adults respiratory failure

Judge's note: The image of ref 13 on p12 confirms the wording 'severe adults respiratory failure'. The phrase is ungrammatical, and the paper's own title (p1) uses 'severe adult respiratory failure'.

Second check: none (judge only)