Before training, 63% of 878 Chinese manufacturing workers who went through hearing protector fit testing had personal attenuation ratings (PARs) below the OSHA-derated noise reduction rating (NRR) of their earplugs; after one-on-one training, 99% beat it (Murphy et al., Journal of the Acoustical Society of America 2022). The pattern repeats elsewhere: 42.9% of 3,634 workers in seven Chinese industries failed their baseline fit test (Su et al., Chinese Journal of Industrial Hygiene and Occupational Diseases 2025), and fewer than half of 126 offshore oil inspectors reached a 25 dBA PAR before NIOSH stepped in (Murphy, Themann and Murata, International Journal of Audiology 2016). The gains also fade: among Marine recruits trained with the best-performing method, the share passing a 25 dB PAR fell from 78% to 37% within about a year (Federman et al., International Journal of Audiology 2026). We aggregated data from NIOSH, OSHA, the Cochrane Collaboration, the National Hearing Conservation Association, the Journal of the Acoustical Society of America, the Journal of Occupational and Environmental Hygiene, the International Journal of Audiology, Ear and Hearing, and Military Medicine. For the broader picture of exposure, non-use and the device market, see our hearing protection statistics roundup. This page covers what happens once the earplug is actually in the ear.
TL;DR
- 63% of 878 workers had pre-training PARs below the OSHA-derated NRR; 99% exceeded it after training (Murphy et al., JASA 2022).
- Only 15% of 503 workers reached the attenuation their earplug’s labeled NRR predicted before training (Gong et al., American Journal of Industrial Medicine 2021).
- 1,559 of 3,634 workers (42.9%) failed a baseline fit test across seven industries (Su et al., 2025).
- Versus simple instruction, extensive instruction lifted foam earplug PAR by 8.62 dB and the pass rate from 50.5% to 88.1% (Cochrane Review CD015066, 2024).
- The mean earplug PAR gain of 12.88 dB right after training shrinks to 4.46 dB at 12 months (Gong et al., JOEH 2026).
- Marine recruits passing a 25 dB PAR fell from 78% to 37% about 12 months after training (Federman et al., 2026).
- OSHA subtracts 7 dB and then applies a 50% safety factor to the NRR when weighing engineering controls (OSHA Technical Manual).
- NIOSH replaced its 25% / 50% / 70% derating scheme with a call for individual, quantitative fit testing (NIOSH Science Policy Update 2025-104).
- Service members with a 0 dB PAR had a 17.9% significant threshold shift rate, against a 7.7% sample average (Kulinski et al., 2026).
- Three commercial field fit-test systems landed within about 2 to 3 dB of the laboratory reference (Karch et al., JOEH 2024).
1. The Label Versus the Ear: NRR and Measured PAR
The NRR printed on the box was never meant to describe one person. It is a laboratory figure built from 10 test subjects, set at the mean attenuation minus two standard deviations (roughly the 98th-percentile performance capability of the protector) with a further 3 dB subtracted (see the first row below). Field data show how far real workers fall from that number. In a study of 503 workers at nine manufacturing facilities in eastern China, only 15% reached the attenuation predicted by the labeled NRR before training, and just 26% did after one-on-one training (Gong et al., American Journal of Industrial Medicine 2021).
The device type matters as much as the wearer. Among 306 automotive workers tested with the 3M E-A-Rfit dual-ear system, preformed earplugs produced a baseline median PAR of 0.5 dB, essentially no protection, while foam earplugs reached 19 dB (Ou et al., Chinese Journal of Industrial Hygiene and Occupational Diseases 2026). For a sense of the exposures these workers face, see our manufacturing noise exposure statistics.
| Metric | Value | Source |
|---|---|---|
| Basis of the US NRR | Mean attenuation minus 2 SD (about 98th percentile), minus a further 3 dB; 10 test subjects | Murphy et al., JASA 2022 |
| Workers reaching labeled NRR attenuation before training (503 workers, 9 facilities) | 15% (14% for the SNR) | Gong et al., American Journal of Industrial Medicine 2021 |
| Workers reaching labeled NRR attenuation after one-on-one training | 26% | Gong et al., American Journal of Industrial Medicine 2021 |
| Baseline median PAR, preformed vs foam earplugs (306 automotive workers) | 0.5 dB vs 19 dB | Ou et al., Chinese Journal of Industrial Hygiene and Occupational Diseases 2026 |
| Baseline median PAR, 485 machinery manufacturing workers | 6.0 dB | Wang et al., Chinese Journal of Industrial Hygiene and Occupational Diseases 2024 |
| Median PAR among 714 workers in Southern China | 13.0 dB (8.0 dB after adjusting for actual wear time) | Xu et al., JOEH 2025 |
| Mean PAR among 385 noise-exposed auto parts workers | 7 dB | Gong et al., International Journal of Environmental Research and Public Health 2021 |
| Workers reaching the corrected labeled NRR (94 foam-earplug users) | 11.70% (11 workers) | Liu et al., Chinese Journal of Industrial Hygiene and Occupational Diseases 2024 |
Context: the Xu et al. study shows a second gap beyond fit. Once wear time was factored in, the median effective PAR dropped by 5 dB, because only 31% of the 714 workers wore their earplugs properly throughout noise exposure.
2. How Many Workers Fail Their First Fit Test
In the largest field samples, somewhere between two in five and more than half of workers fail on the first try, which turns fit testing from a nice-to-have into a screening tool. Of 3,634 workers across seven Chinese industries, 1,559 (42.9%) failed their baseline fit test (Su et al., Chinese Journal of Industrial Hygiene and Occupational Diseases 2025). In five machinery plants the share was even higher, at 56.7%.
The NIOSH offshore study is the US reference point: fewer than half of 126 inspectors and engineers reached the 25 dBA PAR target before intervention, and more than 85% did after refitting and retraining (Murphy, Themann and Murata, International Journal of Audiology 2016). Most recent available data for that US field benchmark: 2016.
| Metric | Value | Source |
|---|---|---|
| Offshore oil inspectors with adequate protection before NIOSH training (126 tested, 25 dBA target) | Fewer than half | Murphy, Themann and Murata, International Journal of Audiology 2016 |
| Same group after refitting and retraining | More than 85% | Murphy, Themann and Murata, International Journal of Audiology 2016 |
| Workers failing the baseline fit test, 7 industries | 1,559 of 3,634 (42.9%) | Su et al., 2025 |
| Workers failing the baseline fit test, 5 machinery plants | 275 of 485 (56.7%) | Wang et al., Chinese Journal of Industrial Hygiene and Occupational Diseases 2024 |
| Under-protection rate before vs after training (94 machinery workers) | 45.74% vs 7.45% | Liu et al., Chinese Journal of Industrial Hygiene and Occupational Diseases 2024 |
| Baseline pass rate, foam vs preformed earplugs | 84.8% (178/210) vs 37.5% (36/96) | Ou et al., 2026 |
| Fit-test trials showing ineffective protection, roll-down foam earplugs (Taiwan) | 23 of 105 (20%) | Chiu and Wan, IJERPH 2020 |
| Workers needing a fit intervention after baseline testing (3 enterprises; baseline median PAR 7 dB) | 54.55% (120 of 220) | Wang et al., Chinese Journal of Industrial Hygiene and Occupational Diseases 2025 |
Most recent available data for Taiwan: Chiu and Wan, 2020. Even after education, training and switching earplug type, seven testees in that study still could not reach an effective attenuation, which is why the NIOSH offshore team recommended other hearing protection solutions for workers who could not reach the target PAR.
3. What Fit-Training Actually Buys
The training effect is large, but only when the instruction is hands-on. Extensive instruction raised foam earplug PAR by 8.62 dB (95% CI 6.31 to 10.93) over simple instruction, and lifted the pass rate from 50.5% to 88.1% (RR 1.75, 95% CI 1.44 to 2.11) in a trial of 321 Marine recruits pooled in the 2024 Cochrane review. The review notes that every 3 dB reduction halves the sound energy reaching the ear, so a 9 dB gain is three halvings, or one-eighth of the energy (1/2 x 1/2 x 1/2).
Simple instruction is a different story. In the same review, a short instruction paired with fit testing produced a 1.62 dB difference at one month (95% CI -0.93 to 4.17), which the authors read as probably no improvement. The lesson for program managers: the fit test is the measuring stick, and the coaching is what moves the number.
| Metric | Value | Source |
|---|---|---|
| PAR gain, extensive vs simple instruction (foam earplugs, 321 participants) | 8.62 dB (95% CI 6.31 to 10.93), moderate certainty | Cochrane Review CD015066, 2024 |
| Pass rate, simple vs extensive instruction | 50.5% vs 88.1%; RR 1.75 (95% CI 1.44 to 2.11) | Cochrane Review CD015066, 2024 |
| Mean PAR improvement after one-on-one training (503-worker study) | 15 dB (p < 0.01) | Gong et al., American Journal of Industrial Medicine 2021 |
| Binaural PAR before vs after training (94 foam-earplug users) | 11.1 +/- 7.4 dB(A) vs 19.3 +/- 6.4 dB(A), P < 0.001 | Liu et al., 2024 |
| Median PAR of the intervention group, before vs after (485-worker study) | 0.0 dB vs 15.0 dB | Wang et al., 2024 |
| Pass rate, current vs experiential training (341 Marine recruits) | About 50% vs nearly 92%; delta PAR more than 9 dB higher | Federman, Karch and Duhon, International Journal of Audiology 2021 |
| MIRE pass rate, video training vs no intervention (52 novices, 13 per group) | 31% to 62% vs 24% to 15% | Samelli et al., La Medicina del Lavoro 2025 |
Outlier note: in the Samelli et al. Brazilian trial, the group that received no instruction got worse on retest, while package reading and video both doubled the pass rate. Small groups of 13 limit how far that result can be generalized.
4. How Long the Gains Last
Fitting an earplug well is a perishable skill. A 2026 mixed-methods review of 23 before-and-after studies covering 5,575 workers found the mean earplug PAR gain fell from 12.88 dB immediately after extensive training to 8.84 dB within six months and 4.46 dB at 12 months (Gong, Morata and Samelli, Journal of Occupational and Environmental Hygiene 2026). The authors flag that the confidence intervals at six and 12 months include zero, and they rate all 23 studies at serious risk of bias.
The best randomized evidence points the same way. Among 390 Marine Corps infantry recruits, 57% of the control group and 78% of the experiential-training group passed a PAR of at least 25.0 dBA right after training; about 12 months later, those shares were 19% and 37% (p < 0.05 at both time points), and the authors concluded that annual retraining is justified (Federman et al., International Journal of Audiology 2026).
| Metric | Value | Source |
|---|---|---|
| Mean earplug PAR gain: immediate / within 6 months / at 12 months | 12.88 dB / 8.84 dB / 4.46 dB | Gong, Morata and Samelli, JOEH 2026 |
| PAR change without refresher (within 6 months / 6 months or longer) vs with refresher (within 6 months / annual extensive) | -2.64 dB / -3.09 dB vs +1.53 dB / +3.45 dB | Gong, Morata and Samelli, JOEH 2026 |
| Pass rate (PAR at least 25.0 dBA), control vs experiential: immediately after training | 57% vs 78% | Federman et al., 2026 |
| Same, about 12 months after training | 19% vs 37% | Federman et al., 2026 |
| Average PAR drop over a 2-hour shift, six earplug types (30 brewery workers) | -0.7 to -2.6 dB; 95% of final PARs still met the 15 dB target | Gong et al., Seminars in Hearing 2023 |
| Median PAR of 192 textile workers, initial visit vs follow-up 17 months later | 10 dB vs 13 dB | Fu et al., International Journal of Audiology 2025 |
| Subjects still below adequate PAR after two training sessions (30 workers, 6 months) | 23 to 27% | Assuncao et al., La Medicina del Lavoro 2019 |
Most recent available data for the Brazilian six-month cohort: Assuncao et al., 2019. Earmuff evidence is thin: the same review found a single study, with PAR increases of 5 dB, 2 dB and 3 dB at immediate, short-term and long-term follow-up.
5. Derating Rules: OSHA, NIOSH and the 2025 Policy Shift
Derating was the workaround for a world without field fit-test systems. OSHA’s version is blunt: subtract 7 dB from the NRR for A-weighted measurements, then apply a 50% safety factor when deciding whether engineering controls are needed. In OSHA’s own worked example, a worker exposed to 98 dBA wearing a 25 dB NRR protector is estimated at 89 dBA with the 50% factor, versus 80 dBA with the 7 dB subtraction alone (OSHA Technical Manual, Section III Chapter 5). A 9 dB swing on paper decides whether the employer must engineer the noise down.
The problem is that a single factor cannot describe a population whose PARs run from essentially zero for some preformed earplugs (section 1) to around 19 dB for trained foam-earplug users (section 3). Murphy et al. (JASA 2022) found that for untrained workers the derating may overestimate attenuation for as many as 25% of workers, while 90% to 100% of well-trained workers beat the derated value. NIOSH drew the conclusion in 2025: its Science Policy Update 2025-104 replaced the derating guidance in Chapter 1 (Section 1.5) and Chapter 6 of the 1998 criteria document with a recommendation for individual, quantitative fit testing. On paper, fit is already in the rules: 1910.95(i)(5) requires employers to ensure proper initial fitting, and in construction 29 CFR 1926.101 requires protectors to be individually fitted, or determined to fit, by a competent person, but neither asks for a measured PAR. For how that plays out on job sites, see our construction worker hearing loss statistics.
| Metric | Value | Source |
|---|---|---|
| OSHA A-weighted estimate with safety factor | TWA (dBA) minus [(NRR - 7) x 50%] | OSHA Technical Manual, Section III Chapter 5 |
| OSHA worked example: 98 dBA exposure, 25 dB NRR | 89 dBA with 50% factor; 80 dBA with 7 dB subtraction only | OSHA Technical Manual, Section III Chapter 5 |
| Extra attenuation credited for dual protection (muffs plus plugs) | 5 dB added to the higher NRR | OSHA Technical Manual, Section III Chapter 5 |
| Required attenuated exposure, general industry / after a standard threshold shift | 8-hour TWA of 90 dB / 85 dB or below | OSHA 29 CFR 1910.95(j) |
| NIOSH 1998 derating of labeled NRR | 25% earmuffs, 50% formable earplugs, 70% other earplugs | NIOSH Criteria Document 98-126 |
| Workers whose attenuation derating may overestimate (untrained) vs workers beating the derated value (well-trained) | As many as 25% vs 90% to 100% | Murphy et al., JASA 2022 |
| Year NIOSH replaced derating with individual fit testing | 2025 (Publication No. 2025-104) | NIOSH Science Policy Update 2025-104 |
Context: the OSHA noise standard does not require fit testing. The Technical Manual says a fit test cannot be substituted for the NRR when judging a protector’s adequacy, although it can be used alongside it, and OSHA Appendix B to 1910.95 notes that calculated attenuation is realistic only to the extent that protectors are properly fitted and worn.
6. Fit-Test Systems, ANSI S12.71 and Adoption
The technology question is largely settled; the adoption question is not. NIOSH names ASA/ANSI S12.71-2018 (R2022), the performance criteria for systems that estimate the attenuation of passive hearing protectors, as the benchmark, and it does not favor any single method. When 57 adults were tested on up to seven earplugs each, three commercial field systems came within about 2.1 to 3.1 dB of the laboratory reference and within 1.0 dB of one another (Karch et al., JOEH 2024).
Methods split two ways in the research base, between microphone-in-real-ear (MIRE) testing and real-ear attenuation at threshold (REAT), with MIRE the more common in recent fit-training studies (see the table). In the US military, DoD Instruction 6055.12, updated on November 22, 2023, now requires quantitative hearing protector fit testing, yet implementers reported eight barriers, from choosing a target PAR to moving results into the electronic medical record (Murphy, Taylor and Schulz, Military Medicine 2025). For the outcome side of military noise, see our veterans hearing loss and tinnitus statistics.
| Metric | Value | Source |
|---|---|---|
| Performance standard NIOSH cites for PAR systems | ASA/ANSI S12.71-2018 (R2022) | NIOSH Science Policy Update 2025-104 |
| PAR84 definition under S12.71 calculations | One standard deviation below the median (alpha = 1.0); E-A-Rfit described as the only commercial product applying the S12.71 corrections | Murphy et al., JASA 2022 |
| Field system vs laboratory reference (Benson CCF-200 narrowband, pure tone; FitCheck Solo) | +3.1 dB, +2.1 dB, +2.5 dB | Karch et al., JOEH 2024 |
| Agreement among field systems | Within 1.0 dB | Karch et al., JOEH 2024 |
| Measurement method in 23 fit-training studies | 15 MIRE, 8 REAT | Gong, Morata and Samelli, JOEH 2026 |
| Typical fit-test time in the NIOSH offshore study | 6 to 12 minutes | Murphy, Themann and Murata, 2016 |
| Photo-based neural network classifying fit above or below median attenuation | 73% accuracy (960 photos, 160 people, 4 protector types) | Smalt et al., JASA 2021 |
| DoD hearing conservation instruction requiring quantitative fit testing | DoDI 6055.12, updated November 22, 2023 | Murphy, Taylor and Schulz, Military Medicine 2025 |
Most recent available data for photo-based fit estimation: Smalt et al., 2021. The NIOSH 2025 update also accepts real-time protected exposure level monitoring as meeting its recommendation, which leaves room for in-ear dosimetry systems alongside classic fit testers.
7. PAR, Over-Protection and Hearing Outcomes
A PAR is only useful if it predicts hearing outcomes, and the newest military data says it does. Among 1,829 service members fit-tested during routine hearing conservation visits, those with a 0 dB PAR had a 17.9% significant threshold shift rate (p = 0.001), more than double the 7.7% sample average (Kulinski, Sheffield and Brungart, International Journal of Audiology 2026). The same team found pure-tone thresholds tracked octave-band PAR calculations closely (R² = 0.85), which could cut testing time by about 50%.
The opposite failure exists too. In US metal manufacturing, most measurements (Sayler et al., Ear and Hearing 2019) showed workers blocking so much sound that their effective exposure fell below 70 dBA, a level that can isolate people from alarms and speech. NIOSH’s own guidance aims for a protected exposure of 75 to 85 dBA and notes that most workers need only about 10 dB of reduction (NIOSH, Hearing Protection).
| Metric | Value | Source |
|---|---|---|
| Significant threshold shift rate, 0 dB PAR group vs sample average (1,829 service members) | 17.9% vs 7.7% (p = 0.001) | Kulinski, Sheffield and Brungart, 2026 |
| Agreement of pure-tone vs octave-band PAR | R² = 0.85; about 50% potential test-time reduction | Kulinski, Sheffield and Brungart, 2026 |
| Attenuation of miners with vs without hearing loss (200 workers, 10 sites) | Lower by beta = -5.1 dB (SE = 1.7); protection worn 73.9% of the time | Ullman et al., Occupational and Environmental Medicine 2021 |
| Measurements giving effective exposure below 70 dBA (251 measurements, 91 workers) | 84% (over-protection); under-protection in under 1% of shifts | Sayler et al., Ear and Hearing 2019 |
| Cumulative noise exposure, unadjusted vs PAR-adjusted (714 workers) | 102.3 vs 91.3 dB(A)-year; 22.5% with hearing loss | Xu et al., JOEH 2025 |
| High-frequency hearing loss, auto parts workers with mean PAR of 7 dB vs non-exposed staff | 65% vs 33% | Gong et al., IJERPH 2021 |
| Ear canals classed as extremely large, linked to lower earplug attenuation (121 workers) | 12.7% | Poissenot-Arrigoni et al., International Journal of Audiology 2025 |
| NIOSH target exposure range with protection worn | 75 to 85 dBA | NIOSH, Hearing Protection |
Most recent available data: Sayler et al., 2019, and Ullman et al., 2021. In the auto parts study, earplug use was not significantly associated with hearing loss prevalence (OR 0.964, 95% CI 0.925-1.005, p = 0.085), which fits with a 7 dB mean PAR offering little real protection.
Summary: Hearing Protection Fit Testing by the Numbers
| Metric | Value | Source |
|---|---|---|
| Workers below OSHA-derated NRR before vs after training (878 workers) | 63% below before; 99% above after | Murphy et al., JASA 2022 |
| Workers reaching labeled NRR attenuation before training | 15% of 503 | Gong et al., American Journal of Industrial Medicine 2021 |
| Baseline fit-test failure, seven industries | 42.9% of 3,634 | Su et al., 2025 |
| Baseline fit-test failure, machinery plants | 56.7% of 485 | Wang et al., 2024 |
| Offshore inspectors adequately protected, before vs after NIOSH training | Fewer than half vs more than 85% | Murphy, Themann and Murata, 2016 |
| Baseline median PAR, preformed vs foam earplugs | 0.5 dB vs 19 dB | Ou et al., 2026 |
| PAR gain, extensive vs simple instruction; pass rate | 8.62 dB; 50.5% to 88.1% | Cochrane Review CD015066, 2024 |
| Earplug PAR gain, immediate vs 12 months | 12.88 dB vs 4.46 dB | Gong, Morata and Samelli, JOEH 2026 |
| Marine recruits passing 25 dBA PAR, immediately vs about 12 months later (experiential) | 78% vs 37% | Federman et al., 2026 |
| PAR drop over a 2-hour shift | -0.7 to -2.6 dB | Gong et al., Seminars in Hearing 2023 |
| OSHA worked example, 98 dBA with 25 dB NRR; credit for dual protection | 89 dBA (50% factor) vs 80 dBA; 5 dB | OSHA Technical Manual |
| NIOSH 1998 derating factors | 25% / 50% / 70% | NIOSH Criteria Document 98-126 |
| Field fit-test systems vs laboratory | About 2.1 to 3.1 dB | Karch et al., JOEH 2024 |
| Threshold shift rate, 0 dB PAR vs average | 17.9% vs 7.7% | Kulinski, Sheffield and Brungart, 2026 |
| Measurements showing over-protection below 70 dBA | 84% | Sayler et al., Ear and Hearing 2019 |
| Workers wearing earplugs properly throughout exposure | 31% of 714 | Xu et al., JOEH 2025 |
Methodology and Sources
We read each study’s abstract or full text directly (through PubMed Central, Europe PMC records and CDC Stacks) and cite the original paper, not secondary summaries. Fit-test PARs depend on the earplug, the test system and the target used, so figures from different studies should not be pooled.
- Murphy, Gong, Karch, Federman and Schulz, Personal attenuation ratings versus derated noise reduction ratings for hearing protection devices, Journal of the Acoustical Society of America 2022
- Morata, Gong, Tikka, Samelli and Verbeek, Hearing protection field attenuation estimation systems and associated training, Cochrane Database of Systematic Reviews 2024 (CD015066)
- Gong, Morata and Samelli, Enhancing and sustaining noise attenuation from hearing protection devices, Journal of Occupational and Environmental Hygiene 2026
- NIOSH Science Policy Update: Individual Fit-Testing Recommendation for Hearing Protection Devices, Publication No. 2025-104
- NIOSH Criteria for a Recommended Standard: Occupational Noise Exposure, Revised Criteria 1998 (No. 98-126)
- NIOSH, Noise and Hearing Loss Prevention: Hearing Protection (2024)
- Murphy, Themann and Murata, Hearing protector fit-testing with off-shore oil-rig inspectors in Louisiana and Texas, International Journal of Audiology 2016
- OSHA Technical Manual, Section III Chapter 5: Noise
- OSHA 29 CFR 1910.95, Occupational Noise Exposure and Appendix B
- National Hearing Conservation Association, Intro to Fit Testing for Hearing Protection
- Federman, Karch, Qureshi and Duhon, Comparing long-term effects of different earplug fit-training methods on U.S. Marine Corps training recruits, International Journal of Audiology 2026
- Federman, Karch and Duhon, How hearing conservation training format impacts personal attenuation ratings in U.S. Marine Corps training recruits, International Journal of Audiology 2021
- Kulinski, Sheffield and Brungart, Personal attenuation rating as an indicator of audiometric threshold shifts, International Journal of Audiology 2026
- Karch, Federman, Ginsberg and Qureshi, Measuring earplug noise attenuation: a comparison of laboratory and field methods, JOEH 2024
- Gong, Zhang, Liu, Ding and Zhu, Verifying earplug attenuation and evaluating one-on-one training at nine facilities in China, American Journal of Industrial Medicine 2021
- Xu et al., Dose-response relationship between NIHL and cumulative noise exposure adjusted by hearing protection devices, JOEH 2025
- Su et al., Protective effects and influencing factors of hearing protection devices in different industries, Chinese Journal of Industrial Hygiene and Occupational Diseases 2025
- Murphy, Taylor and Schulz, Using implementation science to improve the adoption of hearing protector fit testing in the Department of Defense, Military Medicine 2025
- Gong, Murphy, Meinke, Feng and Stephenson, Evaluating earplug performance over a 2-hour work period with a fit-test system, Seminars in Hearing 2023
- Samelli et al., Enhancing hearing protection: evaluating innovative training modalities, La Medicina del Lavoro 2025
- Also cited from their published abstracts: Ou et al. 2026, Wang et al. 2024, Wang et al. 2025 and Liu et al. 2024 (Chinese Journal of Industrial Hygiene and Occupational Diseases); Fu et al. 2025 and Poissenot-Arrigoni et al. 2025 (International Journal of Audiology); Gong et al. 2021 (IJERPH); Chiu and Wan 2020 (IJERPH); Ullman et al. 2021 (Occupational and Environmental Medicine); Sayler et al. 2019 (Ear and Hearing); Smalt et al. 2021 (JASA); Assuncao et al. 2019 (La Medicina del Lavoro).
- Data watch: (1) Federman et al. 2021 reports pass rates of about 50% vs nearly 92% for 341 recruits, while the Cochrane review’s reanalysis of the same trial uses 321 participants and reports 50.5% vs 88.1%; we cite each figure to its own source. (2) Gong et al. 2021 reports that 65% of its workers already exceeded the derated NRR before training, while Murphy et al. 2022 found 63% below it in a pooled Chinese cohort; the cohorts, earplugs and derating method differ (Murphy et al. use OSHA’s (NRR - 7)/2), so the two are not contradictory but should not be combined. (3) Two abstracts on the same 3,634-worker Su et al. cohort give 1,559 (42.9%) and 1,536 (43.4%) baseline failures; we use the figure whose percentage matches its count. (4) The 2026 review counts 13 studies for the immediate earplug effect in its abstract and 12 in its results, and its 6- and 12-month confidence intervals include zero. (5) Several outcome studies (Sayler 2019, Assuncao 2019, Chiu 2020, Ullman 2021, Smalt 2021) are older than three years and are flagged as most recent available data. (6) Vendor specifications for commercial fit-test systems were not used as statistics.
- Last updated: October 18, 2026. We update this roundup quarterly, and the next refresh is expected when the next update of Cochrane Review CD015066 or the next NIOSH fit-testing field report is published.