Table A6:
Risk of Biasa Among Nonrandomized Trials, Bilateral Cochlear Implantation in Children (ROBINS-I Tool)
| Pre-intervention | At Intervention | Post-intervention | |||||
|---|---|---|---|---|---|---|---|
| Author, Year | Confounding | Study Participant Selection | Classification of Interventions | Deviations From Intended Intervention | Missing Data | Measurement of Outcomes | Selection of Reported Results |
| Cullington et al, 201783 | Seriousb | Low | Low | Low | Seriousc | Moderated | Low |
| Galvin et al, 201684 | Seriousb | Seriouse,f | Low | Low | Low | Moderateg | Low |
| Godar et al, 201085 | Seriousb | Seriouse,f | Low | Low | Low | Low | Moderateh |
| Peters et al, 200786 | Seriousb | Seriouse,f | Low | Low | Moderatei | Moderatej | Low |
| Reeder et al, 201787 | Seriousb | Moderatef | Low | Low | Moderatei | Low | Low |
| Scherf et al, 200788 | Seriousb | Seriouse,f | Low | Low | Moderatei | Seriousj,k | Low |
| Scherf et al, 200989 | Seriousb | Seriouse,f | Low | Low | Moderatei | Seriousj,k | Low |
| Scherf et al, 200990 | Seriousb | Seriouse,f | Low | Low | Moderatei | Seriousg,l | Low |
| Scherf et al, 200991 | Seriousb | Seriouse,f | Low | Low | Moderatei | Seriousg,l | Low |
| Sparreboom et al, 201192 | Seriousb | Moderatef | Low | Low | Moderatei | Low | Low |
| Sparreboom et al, 201293 | Seriousb | Moderatef | Low | Low | Moderatei | Seriousj,m | Low |
| Sparreboom et al, 201494 | Seriousb | Moderatef | Low | Low | Moderatei | Seriousg,j,n | Low |
| Strom-Roum et al, 201295 | Seriousb | Moderatef | Low | Low | Moderatei | Low | Low |
| Tait et al, 201096 | Seriousb | Moderatee | Low | Low | Low | Low | Low |
Abbreviation: ROBINS-I, Risk of Bias in Non-randomized Studies—of Interventions.
Possible risk of bias levels: low, moderate, serious, critical, and no information.
Potential differences in patient characteristics at baseline (e.g., age at first and second cochlear implantation, degree and duration of hearing loss, duration of unilateral and bilateral auditory deprivation, history of prior hearing aid experience).
Children dropped in and out of the study, suggesting that data were not available for each child at each test interval for patient-related reasons (e.g., attention, fatigue, language) or non-patient-related reasons (e.g., equipment, scheduling).
Children were not included in testing if clinicians considered them unco-operative; results may be skewed toward older children with more advanced development, better hearing performance, better language, and more co-operative behaviour.
There was no description of the selection process (e.g., consecutive or random within a time period).
In studies in which patients acted as their own controls, those who underwent bilateral cochlear implantation were asked to deactivate one implant to assess the difference between unilateral and bilateral hearing. This did not represent true unilateral hearing, because implantation may cause insertion damage to the cochlea, deteriorating residual hearing. Deactivating one cochlear implant for patients with bilateral cochlear implants would not reflect day-to-day listening conditions.
The self-reported nature of the questionnaires for tinnitus, subjective benefits of hearing, and quality of life increased the potential for bias in favour of bilateral cochlear implantation.
Statistical testing on outcomes were not reported.
Not all children returned for all postoperative test intervals. Not all tests could be administered to every child because of young age, limited attention span, etc.
Test materials for speech perception in quiet were presented at above 60 dB (average conversational level in quiet).8
It was unclear which test materials were used at each test interval, although the authors stated that the test materials used were appropriate for the speech development stages of the child.
There were potential ceiling effects, because some children were doing well with their first cochlear implant, making it difficult to demonstrate a bilateral benefit.
Generic questionnaires were not sensitive to changes in hearing status and may have underestimated the gains in quality of life from bilateral cochlear implantation.
Although this study used the same cohort of children with bilateral cochlear implants as Sparreboom et al,92,93 the unilateral cochlear implantation control group had nine children at 2-year follow-up and 26 children at 5- to 6-year follow-up. The results may not be directly comparable within the study series.