Table A3.
RoB-2—Risk of Bias in Randomised Studies of Interventions.
| Cochrane Tool for RoB-2—Risk of Bias in Randomised Studies of Interventions (RoB 2) | |||
|---|---|---|---|
| (1) Yes[Y]; (2) Probably Yes[PY]; (3) Probably No[PN]; (4) No[No]; and (5) No Information[NI]. | |||
| Domain 1: Risk of bias arising from the randomization process | |||
| Signalling questions | Comments | Response options | Actual responses—Russell et al., 2017 |
| 1.1 Was the allocation sequence random? | The study was randomized; the allocation sequence was performed under direction of an independent biostatistician assigned to either intervention or control and the list of patients was created before enrolment. | Y/PY/PN/N/NI | Y |
| 1.2 Was the allocation sequence concealed until participants were enrolled and assigned to interventions? | Y/PY/PN/N/NI | Y | |
| 1.3 Did baseline differences between intervention groups suggest a problem with the randomization process? | There were unequal groups (Treated n = 21 vs. Untreated n = 10); in addition, the baselines between Treated and Untreated eyes were unequal; “some concerns”. | Y/PY/PN/N/NI | Y |
| Risk-of-bias judgement | There is a risk-of-bias due to the small sample and this concern was identified the issue from the authors themselves (see Russell, 2017, page 858). | Low/High/Some concerns | Low/Some concerns |
| Optional: What is the predicted direction of bias arising from the randomization process? | Favours expirmental; change of LogMAR, ambulatory navigation/mobility and full-field sensitivity improves visual function | NA/Favours experimental/Favours comparator/Towards null/Away from null/Unpredictable | Favours expirmental |
| Domain 2: Risk of bias due to deviations from the intended interventions (effect of assignment to intervention) | |||
| Signalling questions | Comments | Response options | Actual responses—Russell et al., 2017 |
| 2.1. Were participants aware of their assigned intervention during the trial? | The study was “open-label” therefore the patients were unblinded. | Y/PY/PN/N/NI | N |
| 2.2. Were carers and people delivering the interventions aware of participants’ assigned intervention during the trial? | Y/PY/PN/N/NI | N | |
| 2.3. If Y/PY/NI to 2.1 or 2.2: Were there deviations from the intended intervention that arose because of the trial context? | Deviations may arise on the basis of being unblinded due to the “trial context”, specifcally the absence of unequal baselines) (see interpretation in the Cochrane advice, Rob-2 explanation no. 2.3) | NA/Y/PY/PN/N/NI | PY |
| 2.4 If Y/PY to 2.3: Were these deviations likely to have affected the outcome? | There is “NI”; there was no trial protocol avabilable in the Russell paper, or no IND 13804 document or not included in the BLA 125610 (FDA) | NA/Y/PY/PN/N/NI | NI |
| 2.5. If Y/PY/NI to 2.4: Were these deviations from intended intervention balanced between groups? | See question above, 2.4 | NA/Y/PY/PN/N/NI | NI |
| 2.6 Was an appropriate analysis used to estimate the effect of assignment to intervention? | Y/PY/PN/N/NI | Y | |
| 2.7 If N/PN/NI to 2.6: Was there potential for a substantial impact (on the result) of the failure to analyse participants in the group to which they were randomized? | NA/Y/PY/PN/N/NI | N | |
| Risk-of-bias judgement | Low/High/Some concerns | Some concerns | |
| Optional: What is the predicted direction of bias due to deviations from intended interventions? | NA/Favours experimental/Favours comparator/Towards null/Away from null/Unpredictable | Favours expirmental | |
| Domain 3: Missing outcome data | |||
| Signalling questions | Comments | Response options | Actual responses—Russell et al., 2017 |
| 3.1 Were data for this outcome available for all, or nearly all, participants randomized? | Y/PY/PN/N/NI | Y | |
| 3.2 If N/PN/NI to 3.1: Is there evidence that the result was not biased by missing outcome data? | NA/Y/PY/PN/N | N | |
| 3.3 If N/PN to 3.2: Could missingness in the outcome depend on its true value? | NA/Y/PY/PN/N/NI | N | |
| 3.4 If Y/PY/NI to 3.3: Is it likely that missingness in the outcome depended on its true value? | NA/Y/PY/PN/N/NI | N | |
| Risk-of-bias judgement | Low/High/Some concerns | Low | |
| Optional: What is the predicted direction of bias due to missing outcome data? | NA/Favours experimental/Favours comparator/Towards null/Away from null/Unpredictable | Favours expirmental | |
| Domain 4: Risk of bias in measurement of the outcome | |||
| Signalling questions | Comments | Response options | Actual responses—Russell et al., 2017 |
| 4.1 Was the method of measuring the outcome inappropriate? | The outcome methods for the mobility test (MLMT) had several measurements rolled into one final outcome without sufficient data incuding: (i) speed; (ii) time; (iii) accuracy; (iv) obstacles; (v) time penalties; (vi) lux, and; (vii) the scales of ordinal vs. logarithmic interpretation. Consequently, (a) the full direct data and measurements were not avaiable, and; (b) the final outcome may have different interpretations; see comments from FDA reviewers in the BLA 125610. | Y/PY/PN/N/NI | Y |
| 4.2 Could measurement or ascertainment of the outcome have differed between intervention groups? | As above. | Y/PY/PN/N/NI | PY |
| 4.3 If N/PN/NI to 4.1 and 4.2: Were outcome assessors aware of the intervention received by study participants? | Open-label, therefore there was no blinding. | NA/Y/PY/PN/N/NI | Y |
| 4.4 If Y/PY/NI to 4.3: Could assessment of the outcome have been influenced by knowledge of intervention received? | The study is an open-label, unblinded and approved design; while there may be some unconscious influence, there is no evdience for such, therefore was no influence reported; PN or N. | NA/Y/PY/PN/N/NI | PN/N |
| 4.5 If Y/PY/NI to 4.4: Is it likely that assessment of the outcome was influenced by knowledge of intervention received? | NA/Y/PY/PN/N/NI | PN/N | |
| Risk-of-bias judgement | Low/High/Some concerns | Some concerns | |
| Optional: What is the predicted direction of bias in measurement of the outcome? | NA/Favours experimental/Favours comparator/Towards null/Away from null/Unpredictable | Favours expirmental | |
| Domain 5: Risk of bias in selection of the reported result | |||
| Signalling questions | Comments | Response options | Actual responses—Russell et al., 2017 |
| 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? | There was no pre-specified analysis plan (a protocol) so there is: NI. | Y/PY/PN/N/NI | NI |
| Is the numerical result being assessed likely to have been selected, on the basis of the results, from... | The MLMT assay is a novel primary outcome; see question and response above in 4.2 | Y/PY/PN/N/NI | NI |
| 5.2 … multiple eligible outcome measurements (e.g., scales, definitions, time points) within the outcome domain? | Y/PY/PN/N/NI | NI | |
| 5.3 … multiple eligible analyses of the data? | Y/PY/PN/N/NI | ||
| Risk-of-bias judgement | Low/High/Some concerns | Some concerns | |
| Optional: What is the predicted direction of bias due to selection of the reported result? | NA/Favours experimental/Favours comparator/Towards null/Away from null/Unpredictable | Favours experimental | |
| Overall risk of bias | |||
| Comments | Response options | Actual responses—Russell et al., 2017 | |
| Risk-of-bias judgement | Further data would be valauble to conclude an overall risk of bias/judgement. | Low/High/Some concerns | Some concerns |
| Optional: What is the overall predicted direction of bias for this outcome? | NA/Favours experimental/Favours comparator/Towards null/Away from null/Unpredictable | Favours experimental | |