Selective Laser Trabeculoplasty versus Eye Drops for First-Line Treatment of Ocular Hypertension and Glaucoma (LiGHT)
2019 · paper · Clinical trials
A UK randomized trial of 718 newly diagnosed patients comparing initial laser treatment with eye drops for glaucoma or raised eye pressure. Quality of life at three years did not differ significantly; laser was more cost-effective, and the authors conclude it should be offered first-line.
18 findings
Substance Abstract Findings: "Average EQ-5D score was 0·89 (SD 0·18) in the selective laser trabeculoplasty group versus 0·90 (SD 0·16) in the eye drops group". Table 4, the Results text and Figure 2 give SLT 0·90 (0·16) and eye drops 0·89 (0·18).
- p. 1 Summary, Findings
Average EQ-5D score was 0·89 (SD 0·18) in the selective laser trabeculoplasty group versus 0·90 (SD 0·16) in the eye drops group
- p. 6 Results text and Table 4, EQ-5D at 36 months
group had an average EQ-5D score of 0·90 (SD 0·16), compared with 0·89 (SD 0·18) in the eye drops
- p. 6 Results text and Table 4
the selective laser trabeculoplasty group had an average EQ-5D score of 0·90 (SD 0·16), compared with 0·89 (SD 0·18) in the eye drops group
- p. 1 Abstract, Findings
0·89 (SD 0·18) in the selective laser trabeculoplasty group versus 0·90 (SD 0·16) in the eye drops group
Judge's note: I checked the abstract on p1 and Table 4 and the Results paragraph on p6. Table 4 at 36 months gives eye drops 336, 0·89 (0·18) and SLT 337, 0·90 (0·16). The text reads 'selective laser trabeculoplasty group had an average EQ-5D score of 0·90 (SD 0·16), compared with 0·89 (SD 0·18) in the eye drops group'. In Figure 2 (p7), Laser is above Drops at 36 months. The positive difference of 0·01 (SLT minus drops) fits the table, not the abstract. The p1 margin says the online version was corrected on July 4, 2019, but the files contain no correction notice and the swap is still in this version.
Second check: another model agreed
Report a problem with this finding
Substance The protocol-deviation list says "one received eye drops despite randomisation to selective laser trabeculoplasty", but Figure 1 shows the reverse crossover: in the eye drops arm "1 decided to receive SLT after randomisation", while the SLT arm's only non-recipient "withdrew consent before treatment".
- p. 6 Results, protocol violations
one received eye drops despite randomisation to selective laser trabeculoplasty
- p. 4 Figure 1, eye drops arm
1 decided to receive SLT after randomisation
Only the authors can say what is right here.
Judge's note: Checked Figure 1 on p4 and the text on p6. In the SLT arm, 355 of 356 received SLT and 1 withdrew consent. In the drops arm, 361 of 362 received drops and 1 chose SLT. The two double-randomised patients are handled by their second randomisation (Figure 1 footnote), so they do not explain this. No other statement in the article or supplement describes an SLT-allocated patient who received drops. The direction of the crossover contradicts the figure, and only the authors can say which is right.
Second check: another model agreed
Report a problem with this finding
Substance The Results give the GQL-15 36-month difference as "–0·4, 95% CI –0·6 to 1·3", but Table 4 (analysis at 36 months) prints "–0·4 (–1·3 to 0·6)". As printed in the text, the interval does not contain the estimate.
- p. 6 Results, GQL-15 paragraph
difference –0·4, 95% CI –0·6 to 1·3
- p. 6 Table 4, GQL-15 analysis at 36 months
–0·4 (–1·3 to 0·6)
- p. 6 Results text
adjusted mean difference –0·4, 95% CI –0·6 to 1·3
- p. 6
adjusted mean difference –0·4, 95% CI –0·6 to 1·3)
Judge's note: Checked the p6 image. Table 4 GQL-15 row: drops 19·8 (7·8), SLT 19·8 (7·2), –0·4 (–1·3 to 0·6). The repeated-measures 36-month row is –0·4 (–1·2 to 0·4). The text cites 19·8 vs 19·8 and –0·4, so it refers to the same analysis. The text's lower bound –0·6 is above the estimate –0·4, so that interval cannot belong to this estimate. The table's interval appears correct; the text looks transposed or sign-swapped.
Second check: another model agreed
Report a problem with this finding
Substance The text says "the selective laser trabeculoplasty group had a mean score of 83·3 (SD 17·3) at 36 months, compared with 83·1 (SD 17·7) for the eye drops group", but Table 4 gives eye drops 83·3 (17·3) and SLT 83·1 (17·7). The arms are swapped.
- p. 6 Results, GSS sentence
selective laser trabeculoplasty group had a mean score of 83·3 (SD 17·3) at 36 months, compared with 83·1 (SD 17·7) for the eye drops group
- p. 6 Table 4, GSS analysis at 36 months
83·3 (17·3)
- p. 6 Results para 4
mean score of 83·3 (SD 17·3) at 36 months, compared with 83·1 (SD 17·7) for the eye drops group
- p. 6 Results paragraph 4
the selective laser trabeculoplasty group had a mean score of 83·3 (SD 17·3) at 36 months, compared with 83·1 (SD 17·7) for the eye drops group
Judge's note: Checked the p6 image. In Table 4 the columns are Eye drops then SLT. The GSS row at 36 months is 281 | 83·3 (17·3) | 294 | 83·1 (17·7), and the repeated-measures 36-month row gives the same arm assignment. Figure 2 (GSS panel, p7) shows the drops curve slightly above the laser curve at 36 months, which agrees with the table. The text has reversed the arms.
Second check: another model agreed
Report a problem with this finding
Substance The text says "499 (93·1%) of the 526 eyes treated in the eye drops group", but the eye drops group had 536 eyes analysed, and 93·1% corresponds to 499/536.
- p. 7 Results, clinical outcomes paragraph
499 (93·1%) of the 526 eyes
- p. 7 Results
536 eyes (86·2%) of 622 eyes in 312 patients
- p. 7 Results paragraph 6
499 (93·1%) of the 526 eyes treated in the eye drops group
- p. 7 Results paragraph 5
536 eyes (86·2%) of 622 eyes in 312 patients in the eye drops group
Judge's note: Checked the p7 text: 536 eyes of 622 in the drops group were available at 36 months. In Table 5 (p8), the drops-group medication categories sum to 499, and 499 + 37 not at target = 536. No subset of 526 eyes is defined anywhere. The percentage matches 536, not 526.
Second check: another model agreed
Report a problem with this finding
Substance For the multiple-imputation + SUR ophthalmology-cost analysis, the article prints "mean incremental QALY of 0·011, with 95% bootstrap iterations falling between –0·024 and 0·050", while the supplement prints "0.014 additional QALYs with 95% bootstrap replications falling between -0.018 and 0.046".
- p. 8 Results, economic paragraph
mean incremental QALY of 0·011, with 95% bootstrap iterations falling between –0·024 and 0·050
- p. 25 4.6 ICER and CEAC
results in 0.014 additional QALYs with 95% bootstrap replications falling between -0.018 and 0.046
- p. 25 4.6 ICER and CEAC
0.014 additional QALYs with 95% bootstrap replications falling between -0.018 and 0.046
- p. 8 Results, cost-effectiveness paragraph
mean incremental QALY of 0·011
Only the authors can say what is right here.
Judge's note: Checked p8 of the article and supp p25 (section 4.6). Both describe MI-imputed QALYs, SUR and bootstrap with ophthalmology-only costs. Both give an identical cost saving (–£458; –£585 to –£345) and identical CEAC probabilities (97%/93%; 68%). Only the QALY figures differ. Only the authors can say which run is correct.
Second check: another model agreed
Report a problem with this finding
Substance In Table 6 the total number of adverse events is printed as "2096", but the arm totals 1196 + 906 sum to 2102, and so do the Total-column subrows.
- p. 9 Table 6, Adverse events row, Total column
2096
- p. 9 Table 6, Adverse events row
1196 ... 906 ... 2096
Judge's note: Checked on the page 9 image. Each arm's subrows sum to its printed total (1196 and 906), and the Total-column subrows sum to 2102. Every other Total cell in the table, including the patient count 521 = 260+261 and the serious adverse events 202 = 95+107, equals the sum of the arms. The 2096 cell is the only one that does not.
Second check: another model agreed
Report a problem with this finding
Substance In Table 6, Serious adverse events, Other systemic, eye drop group: "50 (15·3%)" is printed, but 50/362 = 13·8%.
- p. 9 Table 6, Serious adverse events, Other systemic, Eye drop group
50 (15·3%)
Judge's note: Checked on the page 9 image. The column denominator is n=362. In the same row the SLT cell (43/356 = 12·1%) and the Total cell (93/718 = 13%) are consistent with 50 patients. The 15·3% figure is the only inconsistent cell.
Second check: another model agreed
Report a problem with this finding
Substance Supplement 4.1 gives the multiple-imputation QALY difference as '0.014 (SE 0.220 95% CI -0.029 to 0.057 p=0.526)'. The CI and p value fit an SE of about 0.022, not 0.220.
- p. 23 4.1 QALYS, multiple imputation
0.014 (SE 0.220 95% CI -0.029 to 0.057 p=0.526)
Judge's note: Checked on the image of supp p23. The CI half-width of 0.043 implies SE ≈ 0.022. With SE 0.022, z ≈ 0.64 and p ≈ 0.52, which matches the printed p=0.526. An SE of 0.220 would match neither the CI nor the p value, so this looks like a decimal slip.
Second check: another model agreed
Report a problem with this finding
Substance Supplement 4.2 says the machine cost 'is £32 per patient if you assume each site sees 300 patients per year', but the stated inputs give about £37 at 300 patients and £32 at 350.
- p. 23 4.2 Cost of SLT
is £32 per patient if you assume each site sees 300 patients per year
Only the authors can say what is right here.
Judge's note: I reworked the annuity: a 10-year factor at 3.5% is about 8.317, so 38995/8.317 + 6395 ≈ £11,084 a year. Dividing by 200 gives £55.4, which matches the printed £55, so the method is confirmed. Dividing by 300 gives £36.9 and by 350 gives £31.7. An annuity paid in advance (factor 8.608) gives £36.4 at 300 patients, and no discounting gives £34.3, so neither reaches £32. Either the patient count should be 350 (the upper bound stated just before) or the figure should be about £37. The £32 figure also feeds the £96 lower total.
Second check: another model agreed
Report a problem with this finding
Substance In Table SM3 the 3-year totals '133 (109)' (Medicine-1st) and '141 (130)' (Laser-1st) are not the sums of their components (125+23=148; 139+19=158), and the difference total of 8 is not 14+(−4)=10. The baseline columns do add up.
- p. 24 Table SM3, 3-years columns
133 (109)
Only the authors can say what is right here.
Judge's note: Checked on the image of supp p24. Only one n is given per column (223, 217), and nothing in sections 3.5, 3.6 or the caption says the total was derived differently from its components. With a common n, the mean of a sum equals the sum of the means. Only the authors can say which figures are right.
Second check: another model agreed
Report a problem with this finding
Substance In Table SM4 the 3-year totals '1776 (2538)' and '1389 (2100)' do not equal the sums of the printed components (1934 and 1566), and the difference total −387 does not equal the component sum −368.
- p. 25 Table SM4, 3-years columns
1776 (2538)
- p. 24 Table SM3
133 (109)
Only the authors can say what is right here.
Judge's note: Checked on the images of supp p25. The baseline totals also fail to add exactly (431 vs 425; 390 vs 386). Each column has a single n, and no stated rule explains why the totals fall short of the components. The caption says 'Inpatient data includes SAE costs', but the table has no inpatient row. Only the authors can say which figures are right.
Second check: another model agreed
Report a problem with this finding
Reference or label Table 1: "Mean age (years, SD) 62·7 (11·6%) / 63·4 (12·0%)". The SD of age in years is printed with a percent sign.
- p. 4 Table 1, Mean age (years, SD)
62·7 (11·6%)
Judge's note: The p4 image shows % signs on both SDs, although the row label states the units are years. This is a wrong unit symbol.
Second check: another model agreed
Report a problem with this finding
Reference or label "Several sensitivity analyses were performed to verify the results of the primary analysis (appendix p 25)". These analyses are on appendix printed p 18 (Supplementary Online Material 3).
- p. 5 Statistical analysis
(appendix p 25)
- p. 19 Supplementary Online Material – 3 (printed page 18)
Sensitivity analyses for LiGHT
- p. 5 Statistical analysis
Several sensitivity analyses were performed to verify the results of the primary analysis (appendix p 25)
Judge's note: Appendix printed page = PDF page - 1. Printed p 18 (PDF p19) is 'Sensitivity analyses for LiGHT', with the seven EQ-5D sensitivity analyses. Printed p 25 (PDF p26) holds Figure SM1 (CEAC) and 4.7, the economic sensitivity analysis. PDF p25 (printed 24) holds cost tables. Neither page 25 matches the primary-analysis sensitivity analyses.
Second check: another model agreed
Report a problem with this finding
Reference or label The Discussion says "In the eye drops group, 36 patients (5·8%) had disease progression compared with 23 (3·8%) of patients", but the Results and Table 5 count 36 and 23 eyes, and the percentages use eye denominators.
- p. 8 Discussion, second paragraph
36 patients (5·8%) had disease progression compared with 23 (3·8%) of patients
- p. 7 Results
36 eyes in the eye drops group showed algorithm-confirmed disease deterioration
- p. 8 Discussion, para 2
In the eye drops group, 36 patients (5·8%) had disease progression compared with 23 (3·8%) of patients in the selective laser
- p. 8 Table 5, Disease progression during the trial
36 (5∙8%)
Judge's note: Checked p7 ('36 eyes in the eye drops group showed algorithm-confirmed disease deterioration ... compared with 23 eyes') and Table 5 on p8. The percentages match 622 and 611 eyes, not 362 and 356 patients. The unit of analysis is mislabelled.
Second check: another model agreed
Report a problem with this finding
Reference or label Supplement section 3.5 cites '(CSRI Beecham et al 19952)', but the only Beecham entry in the supplement references is 'Beecham JK, Knapp MRJ … Gaskell: London; 1992'.
- p. 21 3.5 Other health care costs
CSRI Beecham et al 19952
- p. 27 References
Gaskell: London; 1992
- p. 27 References
Beecham JK, Knapp MRJ Costing psychiatric interventions. In: Thornicroft G, Brewin C, Wing JK (eds) Measuring Mental Health Needs. Gaskell: London; 1992.
- p. 21
(CSRI Beecham et al 19952)
Only the authors can say what is right here.
Judge's note: On the page image, '19952' is printed at full size on the baseline, not as a superscript. Even reading the trailing '2' as a note marker, the in-text year 1995 does not match 1992 in the reference list. It also says 'et al' for a two-author entry. Only the authors can say which year is intended.
Second check: another model agreed
Report a problem with this finding
Wording Discussion: "This difference may be because control of intraocular pressure after laser relies on patient adherence with treatment". As printed, this is the reverse of the argument being made; presumably it should say that control after laser does not rely on adherence.
- p. 9 Discussion, first column
control of intraocular pressure after laser relies on patient adherence with treatment
- p. 9 Discussion, left column
This difference may be because control of intraocular pressure after laser relies on patient adherence with treatment
Only the authors can say what is right here.
Judge's note: Read on the page 9 image. The sentence explains why the laser group had better IOP control. The supporting evidence that follows is about poor adherence to drops (drops available 69% of the time; concordance 76–86%), and the next sentence contrasts SLT's continuous effect with episodic medication. The logic therefore requires that laser control does not depend on adherence. A word is probably missing, but only the authors can confirm the intended wording.
Second check: another model agreed
Report a problem with this finding
Wording Sensitivity analysis item 3 says 'Laser-1st dominates Medicine-1st, with more costs and less QALYs', which reverses the definition given in 4.6 ('more QALYs for a lower cost') and the positive QALY difference (0.032) reported in the same item.
- p. 26 4.7 Sensitivity analysis, item 3
Laser-1st dominates Medicine-1st, with more costs and less QALYs.
- p. 23 4.1
an adjusted difference of 0.025 (95% CI -0.020 to 0.070 p=0.277)
- p. 26 4.7(3)
adjusted difference of 0.032 (95% CI -0.003 to 0.068 p=0.075)
Judge's note: The image of supp p26 confirms the wording. Dominance as defined in 4.6, and the item's own figures (Laser 2.63 vs 2.61 QALYs), imply 'lower costs and more QALYs'. The differing CIs relative to section 4.1 are handled under C060.
Second check: another model agreed
Report a problem with this finding
33 minor (typos and the like; most checked by the judge only)
Reference or label "Selective laser trabeculoplasty has also been proposed to provide better diurnal intraocular pressure stability..." is cited to ref 29, "Greenidge KC... Effect of argon laser trabeculoplasty on the glaucomatous diurnal curve. Ophthalmology 1983", which predates SLT ("introduced in 1995").
- p. 9 Discussion, left column
Selective laser trabeculoplasty has also been proposed to provide better diurnal intraocular pressure stability
- p. 12 Reference 29
Effect of argon laser trabeculoplasty on the glaucomatous diurnal curve. Ophthalmology 1983
- p. 1 Introduction
It was introduced in 1995
Only the authors can say what is right here.
Judge's note: Checked ref 29 on the page 12 image and the 1995 introduction date in the page 1 text. The cited study concerns argon laser trabeculoplasty, not SLT. It may have been cited deliberately as evidence on laser trabeculoplasty in general, so the authors should confirm whether this is the intended reference (ref 30, Nagar 2009, deals with SLT and IOP fluctuation).
Second check: another model agreed
Report a problem with this finding
Reference or label "Health-care resource use was based on cost using published sources (appendix pp 23–24)". The unit costs and their published sources are in Table SM1 on appendix printed p 21. Printed pp 23–24 hold the cost results (Tables SM2–SM4).
- p. 5 Statistical analysis, economic assessment
cost using published sources (appendix pp 23–24)
- p. 22 Table SM1 (printed page 21)
Table SM1: Health care unit costs used in the cost-effectiveness analysis
Only the authors can say what is right here.
Judge's note: I read appendix PDF pp 21–25. Costing methods are in sections 3.4–3.5 (printed p 20), and Table SM1 with its sources is on printed p 21 (PDF 22). Printed pp 23–24 (PDF 24–25) report cost results. The pointer may have been meant for the cost results, so I ask the authors which they intended.
Second check: another model agreed
Report a problem with this finding
Reference or label Reference 20 is dated '2005' but its URL is '…/primary-open-angle-glaucoma-ppp-2015', so the edition year and the link disagree.
- p. 12 Reference 20
Preferred Practice Pattern. 2005. https://www.aao.org/preferred-practice-pattern/primary-open-angle-glaucoma-ppp-2015
Only the authors can say what is right here.
Judge's note: Looked at the crop of the main reference list on p12. The entry reads 'Primary Open-Angle Glaucoma: Preferred Practice Pattern. 2005. https://www.aao.org/preferred-practice-pattern/primary-open-angle-glaucoma-ppp-2015 (accessed Feb 27, 2019)'. The entry contradicts itself about which edition is cited, and only the authors can say which one they used.
Second check: another model agreed
Report a problem with this finding
Reference or label Supplement 4.6 says 'The CEAC is presented in Figure 1.', but the CEAC is labelled 'Figure SM1', and Figure 1 of the main paper is a different figure (the trial profile).
- p. 25 4.6 ICER and CEAC
The CEAC is presented in Figure 1.
- p. 26 Figure caption
Figure SM1: CEAC of Laser-1st compared to Medicine-1st
Judge's note: Checked the text on supp p25 and the caption on p26. The pointer should read Figure SM1.
Second check: another model agreed
Report a problem with this finding
Wording Abstract: "glaucoma surgery to lower intraocular pressure required in none versus 11 patients". The Results say "11 eyes (1·8%) required surgery to lower intraocular pressure (trabeculectomy) in the eye drops group".
- p. 1 Summary, Findings
required in none versus 11 patients
- p. 7 Results
11 eyes (1·8%) required surgery to lower intraocular pressure
Only the authors can say what is right here.
Judge's note: The p7 Results count 11 eyes against a denominator of 622 eyes. The abstract changes the unit to patients. The paper does not say whether the 11 eyes belong to 11 different patients, so only the authors can confirm the abstract's unit.
Second check: another model agreed
Report a problem with this finding
Wording The Table 3 footnote "*n=716" applies to EQ-5D, but Table 4 baseline EQ-5D lists 362 + 355 = 717 patients.
- p. 5 Table 3 footnote
*n=716
- p. 6 Table 4, EQ-5D Baseline
362
- p. 6 Table 4 baseline EQ-5D
362 ... 355
Only the authors can say what is right here.
Judge's note: In the Table 3 image (p5), the asterisk is on both EQ-5D* and Glaucoma Utility Index*†, and the column headers are n=362 and n=355. In Table 4 (p6), baseline EQ-5D n is 362/355 (717) and baseline GUI n is 361/355 (716). The footnote fits the GUI but not the EQ-5D. Only the authors can say which count is right for EQ-5D.
Second check: another model agreed
Report a problem with this finding
Wording Methods: "stratification factors (diagnosis and centre)". The Table 4 footnote reads "adjusted for baseline score, severity, centre", and the appendix (3.6, 3.7) calls the randomisation factors "(severity and centre)".
- p. 5 Statistical analysis
stratification factors (diagnosis and centre)
- p. 6 Table 4 footnote
adjusted for baseline score, severity, centre
Judge's note: The p3 randomisation section stratifies by diagnosis (OHT vs OAG) and centre, and the p5 Methods use the same wording. The Table 4 footnote (p6) and appendix pp 18 and 21 say severity. Severity categories (OHT, mild, moderate or severe OAG) may include diagnosis, so both forms could be legitimate. The covariate is still described inconsistently.
Second check: another model agreed
Report a problem with this finding
Wording The Results say "the results were confirmed in sensitivity analyses (data not shown)", but the Methods cite the appendix for these analyses and supplement 3 tabulates their results.
- p. 6 Results paragraph 3
the results were confirmed in sensitivity analyses (data not shown)
- p. 19 Sensitivity analyses table
The results of the sensitivity analysis are presented in the following table
Judge's note: Checked supp p19 (printed p18): it lists seven sensitivity analyses and a results table, all with Laser–drops differences between 0·002 and 0·019. The Methods on p5 point to 'appendix p 25'. 'Data not shown' therefore misdescribes data that are in the appendix; it should be a pointer to the appendix. Wording issue.
Second check: another model agreed
Report a problem with this finding
Wording The complete-case adjusted discounted QALY difference is given as '0.025 (95% CI -0.020 to 0.070 p=0.277)' in supplement 4.1 but as '0·024 (–0·02 to 0·07)', p=0·286, in Table 4 and the main text; the n values (261/263) and covariates are the same.
- p. 23 4.1 QALYS
adjusted difference of 0.025 (95% CI -0.020 to 0.070 p=0.277)
- p. 6 Table 4, QALY (discounted)
0·024 (–0·02 to 0·07)
- p. 8 Results
not significant p=0·286
Only the authors can say what is right here.
Judge's note: Compared supp p23 section 4.1 with the 'QALY (discounted)' row of Table 4 (main p6) and the main text on p8 ('p=0·286'). Both are described as complete-case, adjusted for the same covariates and discounted, with the same n, so they appear to be one analysis reported two ways. The supplement says its CIs are bias-corrected bootstrap, which could explain a different p value, but neither place says which test produced each P value. Query for the authors: which model and test gives each figure? The 0.024 vs 0.025 gap may be rounding.
Second check: another model agreed
Report a problem with this finding
Wording Sensitivity item 3 repeats the wording of 4.1 ('In the complete case analysis … 2.63 … n=261 … 2.61 … n=263') but gives different CIs (2.60–2.65 vs 2.60–2.66; 2.57–2.62 vs 2.57–2.64) and a different difference ('0.032' vs '0.025'), without saying which QALY measure it uses. It also has the typo 'with and adjusted'.
- p. 26 4.7 Sensitivity analysis, item 3
(n=261 95% CI 2.60 to 2.65), with 2.61 QALYs in the Medicine-1st arm (n=263 95% CI 2.57 to 2.62) with and adjusted difference of 0.032
- p. 23 4.1 QALYS
(n=261 95% CI 2.60 to 2.66), with 2.61 QALYs in the Medicine-1st arm (n=263 95% CI 2.57 to 2.64) with an adjusted difference of 0.025
Only the authors can say what is right here.
Judge's note: Compared supp p26 with p23. Section 3.9(3) plans a secondary analysis using GUI-derived QALYs, so item 3 is probably that analysis, but the text never says so and reuses the primary-analysis description. Only the authors can confirm what the item reports.
Second check: another model agreed
Report a problem with this finding
Typo Abstract Findings: misspelling "within target intracoluar pressure at more visits"; it should read 'intraocular'.
- p. 1 Summary, Findings
within target intracoluar pressure
- p. 1 Summary, Findings
within target intracoluar pressure at more visits
- p. 1
selective laser trabeculoplasty group were within target intracoluar pressure
Judge's note: The misspelling is visible in the p1 page image (abstract crop).
Second check: none (judge only)
Report a problem with this finding
Typo Summary Funding gives "National Institute for Health Research, Health and Technology Assessment Programme", whereas the Acknowledgments give "Health Technology Assessment Panel" and "the Health Technology Assessment programme".
- p. 1 Summary, Funding
Health and Technology Assessment Programme
- p. 11 Acknowledgments
Health Technology Assessment Panel
Judge's note: The p1 Funding line includes 'and'. Both mentions in the p11 Acknowledgments omit it. This is an internal inconsistency in a proper name, and the Acknowledgments form is the one the paper uses itself.
Second check: none (judge only)
Report a problem with this finding
Typo Randomisation and masking: "system (www.sealedenvelope.com) We randomly assigned" has no full stop between the two sentences.
- p. 3 Randomisation and masking
system (www.sealedenvelope.com) We randomly assigned
- p. 3 Randomisation and masking
(www.sealedenvelope.com) We randomly assigned
Judge's note: The p3 page image shows no full stop after the closing parenthesis before 'We randomly assigned'.
Second check: none (judge only)
Report a problem with this finding
Typo Figure 1, SLT discontinued box: "1 was longer contactable" is missing 'no'. The eye-drops box reads "1 no longer contactable".
- p. 4 Figure 1, SLT discontinued box
1 was longer contactable
Judge's note: Checked in the Figure 1 crop on p4. A word is missing, which reverses the literal meaning, but the intent is clear from the parallel box.
Second check: none (judge only)
Report a problem with this finding
Typo Table 1, A-level or equivalent, SLT: "39 (11.0%)" uses a full-stop decimal where the rest of the table uses mid-dots.
- p. 4
39 (11.0%)
Judge's note: The p4 image shows '11.0%' with a full stop, while neighbouring cells use the mid-dot. The value itself is consistent with 39 of 356.
Second check: code check
Report a problem with this finding
Typo "The cost of drops for OAG and ocular hypertension based on prescribed medications, using the British National Formulary." The sentence has no finite verb (it should read 'was based').
- p. 5 Statistical analysis, economic assessment
The cost of drops for OAG and ocular hypertension based on prescribed medications, using the British National Formulary.
Judge's note: The sentence is visible as printed in the p5 page image, economic assessment paragraph.
Second check: none (judge only)
Report a problem with this finding
Typo Methods use 'worse eye' then 'worst eye' in the same sentence: "based on the worse eye, where the worst eye was defined using visual field loss".
- p. 5
where the worst eye was defined using visual field loss
Judge's note: Checked the crop of p5. The sentence says 'worse eye' and then 'worst eye'. With two eyes the comparative 'worse' is the correct form. Cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo The Table 4 footnote defines the difference as "*(SLT eye drops)" with no minus sign or dash. The text defines it as 'selective laser trabeculoplasty–eye drops'.
- p. 6 Table 4 footnote
*(SLT eye drops)
Judge's note: Checked the crop: there is no dash between 'SLT' and 'eye drops'. Cosmetic omission; the meaning can be recovered from the text.
Second check: none (judge only)
Report a problem with this finding
Typo Table 4 uses full-stop decimals in a paper that otherwise uses mid-dots, e.g. "0·89 (0.18)", "0·90 (0.16)", "4·0 (2·0 to 6.0)".
- p. 6
0·89 (0.18)
Judge's note: Checked the p6 crops. These values are printed with baseline full stops, unlike the mid-dot used elsewhere. The QALY rows also use full stops. Cosmetic only.
Second check: code check
Report a problem with this finding
Typo In Table 5, SLT eyes at target on one medication are printed "64 (12∙0%)"; 64/536 rounds to 11·9%. Likewise "4 (0∙8%)" for three medications, where 4/536 rounds to 0·7%.
- p. 8 Table 5, Number of medications per eye – One, SLT group
64 (12∙0%)
Judge's note: Checked the p8 image. The other cells in the SLT column (419 = 78·2%, 21 = 3·9%, 1 = 0·2%, 27 = 5·0%) match a denominator of 536, and the counts sum to 536. Two cells are off by 0·1 in rounding.
Second check: none (judge only)
Report a problem with this finding
Typo In Table 5, eye drops eyes not at target are printed "37 (7∙0%)"; 37/536 rounds to 6·9%, and with 499 (93·1%) at target the two shares sum to 100·1%.
- p. 8 Table 5, Eyes not at target at 36 months, Eye drops
37 (7∙0%)
Judge's note: Checked the p8 image. The other drops-column cells (16 = 3·0%, 346 = 64·6%, 99 = 18·5%, 35 = 6·5%, 3 = 0·6%) match a denominator of 536. This cell is off by 0·1 in rounding.
Second check: none (judge only)
Report a problem with this finding
Typo Table 5 percentages for medications per eye do not all match one denominator of 536 eyes: "64 (12·0%)" (64/536 = 11·9%) and "37 (7·0%)" (37/536 = 6·9%); "4 (0·8%)" (4/536 = 0·7%) is also off.
- p. 8 Table 5, number of medications per eye at target IOP
64 (12·0%)
- p. 8 Table 5, Eyes not at target at 36 months
37 (7·0%)
Judge's note: Checked Table 5 on the page 8 image. Each arm has 536 eyes: drops 16+346+99+35+3 at target plus 37 not at target, and SLT 419+64+21+4+1 plus 27. Most cells round correctly with 536 (346→64·6%, 419→78·2%, 499→93·1%, 509→95·0%). The 64, 37 and 4 cells are each off by 0·1 percentage point, and no single denominator fits every cell in the column. This is a minor rounding inconsistency.
Second check: none (judge only)
Report a problem with this finding
Typo Table 6, SLT column (n=356): Cancer "13 (3·6%)" and Ophthalmic allergic reactions "13 (3·7%)" give two percentages for the same count; 13/356 = 3·65%, which rounds to 3·7%.
- p. 9 Table 6, SLT, Serious adverse events – Cancer
13 (3·6%)
- p. 9 Table 6, SLT, Ophthalmic allergic reactions
13 (3·7%)
Judge's note: Checked on the page 9 image. The two cells share the same count and the same denominator but show different percentages. This is a rounding slip of 0·1 percentage point.
Second check: none (judge only)
Report a problem with this finding
Typo Table 6 shows "2 (0·5%)" for SLT pulmonary problems, SLT cerebrovascular accidents and eye-drop deaths, whereas 2/356 and 2/362 both round to 0·6%, and the SLT uveitis row prints "2 (0·6%)".
- p. 9 Table 6, SLT Pulmonary problems/Cerebrovascular accidents (SAE) and drops Death
2 (0·5%)
- p. 9 Table 6, SLT Uveitis
2 (0·6%)
Judge's note: Checked on the page 9 image. The other small counts in the table round normally (for example 1/362 = 0·3% and 3/362 = 0·8%), so truncation does not explain these cells. This is a rounding inconsistency.
Second check: none (judge only)
Report a problem with this finding
Typo Table 6, Systemic drug-related events, Total: "75 (10·5%)" is printed, but 75/718 = 10·4%.
- p. 9 Table 6, Systemic Drug-related events, Total
75 (10·5%)
Judge's note: Checked on the page 9 image. 75 = 52+23 is consistent with the arms. Since 75/718 = 10·446%, the value is not a rounding tie.
Second check: none (judge only)
Report a problem with this finding
Typo Table 6, Other transient events, Total: "123 (17·2%)" is printed, but 123/718 = 17·1%. The SLT cell "122 (34·4%)" likewise does not match 122/356 = 34·3%.
- p. 9 Table 6, Other transient events, Total
123 (17·2%)
Judge's note: Checked on the page 9 image. Both cells would fit denominators one smaller (717 and 355), but the table headers state n=718 and n=356, and other SLT-related cells (for example 14 (1·9%) of 718) do not fit 717. No alternative denominator is stated. These are small rounding inconsistencies.
Second check: none (judge only)
Report a problem with this finding
Typo Discussion: "...had lower rates of disease deterioration compared with those in the eye drops group) and that 11 eyes required surgery..." has an unmatched closing parenthesis and broken syntax.
- p. 9 Discussion
with those in the eye drops group) and that 11 eyes
- p. 9 Discussion, para 1
compared with those in the eye drops group) and that 11 eyes
Judge's note: The unmatched ')' is visible on the page 9 image at the top of the right column, and no opening parenthesis appears earlier in the sentence. It is not an extraction artefact.
Second check: none (judge only)
Report a problem with this finding
Typo Discussion: "Primary selective laser trabeculoplasty gave drop-free control intraocular pressure" is missing the word 'of' (control of intraocular pressure).
- p. 9 Discussion, right column
gave drop-free control intraocular pressure for at least 36 months
Judge's note: Visible on the page 9 image, right column.
Second check: none (judge only)
Report a problem with this finding
Typo Declaration of interests (KB): "Glaucos, Amakem, Thea" misspells the company spelled "Glaukos" earlier in the same declaration (GG).
- p. 11 Declaration of interests
Glaucos, Amakem, Thea
Judge's note: The page 11 image shows both spellings in the Declaration of interests paragraph. Internal inconsistency in the spelling of a company name.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 21: "SEAGIG. South East Asia Gluacoma Interest Group" misspells 'Glaucoma'.
- p. 12 Reference 21
South East Asia Gluacoma Interest Group
Judge's note: Visible in ref 21 on the page 12 image.
Second check: none (judge only)
Report a problem with this finding
Typo The supplement's CSRI medication table misspells 'CONTINUOUS' as 'CONTINOUS', in both 'Please TICK if CONTINOUS' and 'If NOT CONTINOUS'.
- p. 15 CSRI medication table
CONTINOUS
Judge's note: The page image shows the misspelling twice in the column headers. It is cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo In supplement section 1 (Aim), 'Health and social care costs and QALYs was calculated' should read 'were calculated'.
- p. 20 1. Aim
Health and social care costs and QALYs was calculated
Judge's note: The image of supp p20 confirms the wording. The plural subject takes a singular verb, a grammar slip.
Second check: none (judge only)
Report a problem with this finding
Typo Supplement 4.4 prints '(95% -£757 t0 118)': 't0' should be 'to', and 'CI' is missing after '95%'.
- p. 24 4.4 Other health care resource use
(95% -£757 t0 118)
- p. 24
with a difference of -£319 (95% -£757 t0 118)
Judge's note: The image of supp p24 shows 't0' with a zero, so this is not an extraction artefact.
Second check: none (judge only)