Impact of Occupational Noise Exposure on Physical and Mental Health of Water Pumping Station Operators in Lebanon

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript is clear, relevant to the field, and presented in a well-structured manner. 

Many of the cited references were recently published (within the last 5 years). Some old NIOSH, WHO references? More and a few extra recent references!

The manuscript includes no excessive number of self-citations.

The manuscript is scientifically sound, and the experimental design is appropriate to test the hypothesis.

The manuscript’s results are reproducible based on the details given in the Methods section.

The sound level meter 'type 2230' (brand Bruel & Kjaer ?)

The figures/tables/images/schemes do not properly present the data.

Two figures after the comma is enough for all results.

In Figure 5 delete: 120%. 

The conclusions are consistent with the evidence and arguments presented.

The ethics and data availability statements are adequate.

 

Comments for author File: Comments.pdf

Author Response

Please see attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript addresses an important and under-studied occupational setting (Lebanese public water infrastructure) and combines field noise measurements with hearing assessment and worker-reported psychological effects. The work is timely given the widespread reliance on diesel generators and the plausible exposure burden.

However, major methodological and inferential limitations currently prevent confident interpretation of the findings, particularly regarding: 1) exposure assessment (spot SPL vs. 8-hr dose), 2) outcome measurement validity (smartphone audiometry implementation details; unvalidated binary mental health items), 3) confounding and model specification (age/exposure collinearity; limited covariate control), and 4) causal language exceeding a cross-sectional design.

Major comments

The exposure assessment does not operationalize “dose” consistent with the cited standards. Risk is evaluated against the NIOSH 85 dBA 8-hr TWA benchmark. But the methods describe sound level meter measurements at duty locations during regular hours and an “average sound level” per station, without sufficient detail to interpret as an 8-hr TWA. What is missing here is essential information regarding measurement duration per station (seconds/minutes), number of repeats, time-of-day, and whether readings occurred during generator operation (vs. grid only). Also, the authors need to describe/justify how station “average SPL” approximates personal exposure (for example, do operators move between rooms/inside versus outside such that personal exposure is, to some degree, variable?).

The hearing outcome measurement is missing important details regarding both implementation and validity. The authors rely on a smartphone-based “Hearing Test” app with calibrated headphones, classifying impairment via WHO categories and summarizing impairment as the maximum of left/right ear. “Quiet room” is stated, but the field context is unclear (e.g., onsite vs offsite; ambient levels verified?). Who administered the test and what training did they have? Also, the choice of maximum impairment across ears may inflate occupational signal and could preferentially capture unilateral factors unrelated to workplace noise – i.e., if unilateral issues are common in this sample, using a “worse ear” approach can mistakenly overstate the degree of impairment arising from exposure to occupational noise. Furthermore, given a mean age of ~52, age-related hearing loss is a major competing explanation; you partially model age, but measurement validity and potential misclassification still matter.  It is recommended that the authors provide a methods appendix detailing test environment/ambient noise confirmation, calibration steps, administration protocol, and sensitivity analyses using (a) better-ear PTA, (b) worse-ear PTA, and (c) binaural average. Also, please inlude a more explicit limitations paragraph about smartphone-based hearing testing compared with clinical PTA.

The sample selection and comparison logic are unclear. First, the one-sample t-test evaluates whether the mean station exceeds 85 dBA; it does not test whether all stations exceed this level. Indeed, the manuscript reports stations with average levels well below 85 dBA (e.g., S37). In turn, reframe the research question/conclusion to “mean station noise levels exceed 85 dBA” rather than implying a universal hazard across stations. In addition, the test appears to be conducted across all 52 stations, yet the inference is phrased as “stations operating on diesel generators” (p.8); if diesel operation is the causal/exposure mechanism, this should be supported by stratified analyses (hybrid vs non-hybrid) or explicitly reworded. The authors state that the fraction of stations exceed 85 dB the authors have stated 68% - if the goal is to inform system-wide policy, this proportion should be framed using 95% confidence interval for that proportion (e.g., Wald, Wilson); this will directly support policy language about how widespread the exceedences are.

Also, there is no internal control group for the health outcomes. The authors state that they measured noise at 52 stations, but then conducted health assessments “exclusively” among operators in stations defined as noisy (hybrid system). Please clarify how the 50 workers map onto the the subset of “noisy”/hybrid stations – how these were identified, were some stations unmanned/workers unavailable? Provide a flow chart of eligiblity or a clear description. Furthermore – very importantly – without a comparison group (operators in lower-noise stations, or an external reference group), causal attribution of mental health effects to noise is limited, particularly given concurrent stressors (working conditions, broader contextual stress). If you have any operators from lower-noise stations, include them as a comparator; if not, explicitly position this as a single-group exposure–outcome association study with limited causal inference.

Overall, confounding control in the modeling is insufficient and the regression likely suffers from collinearity and omitted variables. The linear regression uses sound level, age, and exposure years as predictors. In this sample/workforce, age and years of exposure are almost certainly highly correlated, and the authors do not report diagnostics (VIF/tolerance; model validity including nonlinearity – e.g., hearing threshold shift may increase non-linearly with age and/or show threshold effects). Several key omitted confounders for hearing outcomes include: prior exposure to noisy jobs, military service, recreational noise, ototoxic medications/solvents, smoking, diabetes/hypertension, and baseline hearing status. At minimum, report collinearity diagnostics, residual plots; also, the authors might consider alternative models to predict predict impairment category (ordinal logistic) rather than continuous dB HL. Add even a small set of covariates (prior noisy work; comorbidity; smoking; recreational noise), or acknowledge strongly as a limitation.

Regarding the psychological health measurement and statistical testing, this approach is generally substandard. The psychological tool is a binary-response questionnaire created for this study (14 items across domains). You then run one-sample z-tests for each item against p=0.50. The 50% threshold is arbitrary; significance here does not establish “impact,” only that endorsement exceeds 50%. Multiple testing (14 tests) inflates false positives; there is no correction for this, or domain-level aggregation. Items, it seems, are not a validated scale, so interpretability and comparability to prior work are limited. Some interpretive text speculates about reluctance/awareness as an explanation for non-significant findings; this is not empirically tested. I suggest that the authors replace the z-test framing with descriptive prevalence plus confidence intervals, and either aggregate into domain scores (with internal consistency checks), or use validated instruments (e.g., perceived stress, sleep quality, anxiety) in a follow-up study and clearly position the current study as exploratory.

The manuscript overstates causal language and recommendations vis-à-vis the methodology and results. The authors often use causal phrasing (“noise exposure significantly increases…”) based on cross-sectional self-report and non-clinical hearing assessment. To be sure, recommendations (engineering controls/HPDs/monitoring) are sensible, but should be tied to an appropriately cautious interpretation and, ideally, a hearing conservation framework. Revise language to “is associated with,” “workers reported,” “levels exceed benchmarks,” and add a limitations section explicitly addressing causality.

There is a potential bias in data collection context should be acknowledged related to the fact that supervisors accompanied site visits. With face-to-face interviews, this may influence responses (social desirability; fear of reprisal), especially for questions about PPE use/training. The authors are encouraged to state whether supervisors were present during interviews; if so, acknowledge potential bias and consider anonymous data collection in future work.

The methods contain non-standard elements that should be removed/reframed; for example, a “brief consultation” with a physician is used to justify a 10-year exposure cut-point. This is not necessary and reads as anecdotal. The cut-point should be justified via literature only and cite the literature directly for exposure-duration rationale.

More minor comments and editorial corrections:

-Correct typographical/terminology errors (e.g., “NOISH” vs NIOSH in the abstract).

-Duplicate heading: “Physical Physical and Psychological…”

-Statistical reporting: avoid “p ≈ 0.00”; report as p<0.001.

-Replace Excel “printout” style tables with journal-formatted regression tables (β, SE, 95% CI, p), and describe model assumptions checks.

-Data availability statement (“presented in tables/figures”, p.16) is weak for reproducibility; consider providing a de-identified dataset and analysis script in a repository.

-Add detail on voluntariness, confidentiality, and whether participation affected employment.

 

Author Response

Please see attached file

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Reviewer Comment
This study focuses on an important and unique public health issue by addressing the effects of occupational noise exposure on the physical and mental health of operators working in water pumping stations. Its strengths lie in its reliance on field data and its combined assessment of hearing and mental health outcomes. However, the current version of the article has significant structural, methodological, and reporting deficiencies. In particular, the abstract, introduction, methods, statistical analysis, presentation of findings, and discussion sections do not fully adhere to the systematic approach of scientific article writing. Restructuring the study based on an appropriate reporting checklist (e.g., STROBE) is important for improving methodological transparency and reporting quality. The methods section needs to be more systematic, statistical analyses need to be reported in a detailed and consistent manner, regression results need to be presented clearly, and the findings need to be discussed in more depth in relation to the literature in the discussion section. For these reasons, a comprehensive revision is needed to make the study more compliant with scientific article writing systematics and to clarify its scientific contribution. Revision suggestions are as follows:
1. Page 1, line 36; The methods section of the abstract should explain the purpose of the evaluation scales used.
2. Page 1, line 36; The results section of the abstract should explain the results statistically in a clear and specific manner.
3. Page 2, lines 68-76; “Water pumping stations are critical infrastructures in water systems. They generate considerable noise due to heavy machinery operations, such as pumps, generators, boosters, ventilation fans, and fluid turbulence within pipelines. This situation poses health risks for workers. This research focuses on water pumping stations in the South and Nabatieh Governorates. These stations are managed by the South Lebanon Water Establishment (SLWE). SLWE, headquartered in Saida, is a public authority responsible for the production, distribution, and maintenance of water systems in South Lebanon. SLWE operates under the umbrella of the Ministry of Energy and Water in Lebanon. SLWE is responsible for providing safe and potable water to all cities, towns, and villages in South Lebanon.” The information from the literature cited in the introduction should not be used without citing the source. Sources should be added for any necessary information in this text. 4. Page 2, line 58; The introduction, in its current form, primarily contains descriptive information, making it difficult to focus the text. To ensure a more fluid and analytical structure, it is recommended that the introduction begin by briefly addressing the prevalence of occupational noise exposure and the effects of high-decibel noise on general health. Following this, it would be appropriate to summarize the key studies in the literature examining the effects of noise exposure on physical and mental health, highlighting their scope and limitations. Next, it should be clearly stated in what ways the current study differs from these studies in the literature and what knowledge gap it aims to fill. The introduction should be structured by referencing the research results in the literature.
5. Page 3, line 119; The materials and methods section of the study should not be included in the introduction but should be at the end.
6. Page 3, line 119; The Materials and Methods section should be structurally divided into subheadings and presented systematically under the headings “Study Design”, “Participants”, “Assessments/Measurements”, and “Statistical Analysis”.
7. Page 3, line 119; The Study Design section should clearly and explicitly define the study design (e.g., cross-sectional/observational study); it should include a brief methodological summary specifying the time period and geographical/institutional setting in which the study was conducted; it should also explicitly state the ethical committee approval, the date and number of approval, the informed consent process, and a declaration that the study was conducted in accordance with the principles of the Declaration of Helsinki; and it should be indicated in this section, with a source, that a relevant checklist (e.g., STROBE) was used in reporting the study.

 

8. Page 4, line 140; In the Participants section, the target population of the study should be clearly defined; the method of determining the sample size should be explained in detail, and if a power analysis was performed, the method, assumptions, and parameters used should be stated; if not, the reason should be clearly stated; furthermore, inclusion and exclusion criteria should be presented clearly and systematically, and the number of participants invited to the study, individuals who refused to participate or were excluded, and the final number of participants analyzed (n) should be reported numerically. In the exclusion criteria, it should be clarified that chronic diseases with uncontrolled symptoms may affect the results.

9. Page 4, line 166; In the Assessments/Measurements section, all measurement and evaluation tools used in the study should be described separately and in detail; for each assessment, the name of the tool or device used, the application protocol, measurement conditions, measurement duration and repetition, calibration information, who performed the assessments (specifying their training, area of ​​expertise, and role), and the approximate duration of each assessment should be clearly reported; The type of decibel used in noise measurements, measurement points, timing, and how exposure is summarized (e.g., time-weighted average) should be explained in detail; the method used in hearing assessment, the application method, classification system, and reference standards should be presented with references; the structure of the tools used in psychological/mental health assessments, the areas they measure, and the assessment approach should be explained based on the literature, and the scientific basis of each measurement tool must be stated in this section with appropriate references.
10. Page 6, line 201; All statistical analyses used in the Statistical Analysis section should be described in detail and systematically; it should be clearly stated for which variables and for what purpose the analyses were used; how normality tests (e.g., Shapiro–Wilk, Kolmogorov–Smirnov) were applied in evaluating the distribution characteristics of the data and according to which criteria the assumption of normal distribution was accepted should be explained; how the necessary assumptions (linearity, homogeneity of variance, independence, and distribution of residuals) for the use of parametric analyses were tested should be detailed; The regression analysis should clearly report which methods were used to assess multicollinearity, normality of residuals, homoskedasticity, and the assumption of independence; information on the statistical software and version used, the accepted statistical significance level in the analyses (e.g., p < 0.05), whether the reported p-values ​​are two-sided or one-sided, confidence intervals, and any corrections for multiple comparisons should be clearly stated in this section, and all analyses should be presented transparently enough to ensure the reproducibility of the study.
11. Page 12, line 349; The R² (R Square) value appears inconsistent across text, tables, and figures; this value should be reported consistently across all presentation formats.
12. The regression table is not sufficiently explanatory; specific effects for each of the variables included in the model—sound level, age, and exposure to noise—should be clearly presented in the regression table. The direction and magnitude of the effect of each variable should be reported along with the regression coefficients, confidence intervals, and p-values; The table should clearly show which variable caused the observed effect and at what level.
13. All abbreviations in the tables should be explained with a footnote below the table.
14. Page 14, line 411; The discussion section should be restructured in accordance with the systematic approach of scientific article writing. In its current form, the section is limited in interpreting the findings and the results are not addressed within a sufficiently analytical framework. Each key finding obtained in the study should be discussed in separate paragraphs, comparing it with the relevant and current literature; the possible mechanisms, contextual reasons, and relationships with similar or contradictory results in the literature should be clearly presented. These discussions should not only rely on repeating the results, but should be structured with inferences that make the scientific contribution of the study and its original value to the literature visible. In addition, the absence of bullet points or numbered statements in the discussion section will strengthen the academic fluency and integrity of the text.
15. Page 15, line 459; The limitations of the study are not clearly and systematically stated at the end of the discussion section. Measurement methods, study design, and limitations regarding generalizability should be clearly addressed in these sections.

Comments on the Quality of English Language

While the language and style of the article are generally understandable, there are occasional grammatical, sentence structure, and fluency issues throughout the text; therefore, it is recommended that the text undergo professional editing to ensure it meets academic English standards.

Author Response

Please see attached file:

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you to the authors for substantially addressing my comments on the prior draft. Overall, the revised manuscript represents a meaningful improvement and now meets an acceptable standard for a cautious, descriptive contribution to the occupational and environmental health literature. Any remaining concerns would require the collection of new data rather than further revision of the current manuscript.

That said, to ensure full alignment between the study design, analyses, and interpretation—particularly with respect to causal inference—I suggest a small number of minor editorial refinements prior to publication:

Abstract: Please reword the relevant sentence to read:
“A high prevalence of hearing impairment was observed among operators, with 88% exhibiting impairment in the worse ear.”

Regression results: Please revise the phrasing to:
“For every additional dBA in sound level, hearing impairment was higher by approximately 0.772 dB HL.”

Psychological health (Discussion): Please soften causal phrasing and revise the sentence to read something along the lines of:
“In addition to the hearing assessment, the study identified a high prevalence of self-reported psychological symptoms among noise-exposed workers, consistent with prior research on occupational noise.”

Conclusions: Please further temper directive-causal language. For example:
“Results indicate that water pumping stations located in the South and Nabatieh Governorates are characterized by elevated noise levels and a high prevalence of adverse physical and psychological health indicators among workers.”

Policy recommendations: Please similarly soften the language to reflect precautionary inference rather than demonstrated causality. For example:
“Based on benchmark exceedances and observed health indicators, precautionary evidence-based interventions should be implemented…” and explicitly acknowledge the need for additional comparative and/or longitudinal research to establish causal relationships.

With these minor refinements, the manuscript will be well aligned with its methodological scope and positioned appropriately for publication.

Author Response

Comment1: to ensure full alignment between the study design, analyses, and interpretation—particularly with respect to causal inference—I suggest a small number of minor editorial refinements prior to publication.

Response1:

Thank you very much for your careful reading of the revised manuscript and for your encouraging assessment. I sincerely appreciate your constructive suggestions, which helped further improve the clarity and interpretation of the study.

All suggested minor editorial refinements have been addressed in the revised manuscript. In particular:

• Abstract: The sentence has been revised to:
“A high prevalence of hearing impairment was observed among operators, with 88% exhibiting impairment in the worse ear.”

• Regression results: The phrasing has been revised to:
“For every additional dBA in sound level, hearing impairment was higher by approximately 0.772 dB HL.”

• Psychological health (Discussion): The causal wording has been softened to reflect association rather than causation, and the sentence now reads:
“In addition to the hearing assessment, the study identified a high prevalence of self-reported psychological symptoms among noise-exposed workers, consistent with prior research on occupational noise.”

• Conclusions: Directive causal language has been tempered to better reflect the descriptive nature of the findings.

• Policy recommendations: The text now adopts precautionary language and explicitly acknowledges the need for further comparative and longitudinal research to establish causal relationships.

The English language, figures and tables are now reviewed and edited through the MDPI Author Services as shown in the new revised version.

All changes have been incorporated and highlighted in the revised manuscript.

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for the comprehensive revisions you made to the article based on the peer review. The revised version significantly improves the methodological structure, reporting transparency, and the analytical nature of the discussion section. The article has reached a stronger and more consistent scientific level compared to the previous version.

Comments on the Quality of English Language

Language usage has been significantly improved compared to the previous version.

Author Response

Comment:Thank you for the comprehensive revisions you made to the article based on the peer review. The revised version significantly improves the methodological structure, reporting transparency, and the analytical nature of the discussion section. The article has reached a stronger and more consistent scientific level compared to the previous version.

Response: Thank you very much for your careful review and your positive feedback on the revised manuscript. I am pleased to know that the revisions have improved the clarity, methodological structure, and discussion of the paper. The English language, figures and tables are now reviewed and edited through the MDPI Author Services as shown in the new revised version.