Nilsson 1980.
| Methods | CBA study | |
| Participants | Ship builders; n = 231 Highly exposed group with > 94 dB(A); n = 1838 Low exposed group with < 88 dB(A); n = 1354 Sweden 1 shipyard | |
| Interventions | Intervention: workers wearing earmuffs; n = 1883 Control: workers wearing earplugs; n = 1309 | |
| Outcomes | STS > 10 dB any frequency in either ear per 100 person‐years; frequencies tested: 0.25, 0.5, 1, 2, 3, 4, 6, 8 kHz | |
| Notes | Long‐term Comparability ‐ intervention/control: age: ? hearing: both groups < 35 dB all frequencies |
|
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| 14. Blinding (subjects) | High risk | not blinded |
| 15. Blinding (outcome assessors) | High risk | not blinded |
| 16. Retrospective unplanned subgroup analyses | Low risk | no subgroup analyses |
| 17. Follow‐up | Low risk | all about 3.3 years' follow‐up |
| 18. Statistical tests | Unclear risk | one sided Chi2 test |
| 19. Compliance | Unclear risk | not reported |
| 20. Outcome measures | Low risk | audiometry quality reported |
| 21. Selection bias (population) | High risk | same factory but different departments with different noise exposure |
| 22. Selection bias (time) | Unclear risk | same time |
| 23. Randomization | High risk | not randomised |
| 24. Allocation concealment | Unclear risk | not randomised, not applicable |
| 25. Adjustment for confounding | High risk | 5 years' difference in average age |
| 26. Incomplete outcome data | Unclear risk | not reported |