Hair Metal Concentrations in Adults Living in Campania, Italy: A Cross-Sectional Biomonitoring Study in a Mixed Geogenic–Anthropogenic Exposure Setting

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript addresses an important topic in environmental biomonitoring and presents data that may contribute to understanding trace element exposure in the study population. The study is generally well organised, and the objectives are clearly stated. However, several issues related to the structure, methodological reporting, and presentation of the results should be addressed before the manuscript can be considered for publication. In particular, improvements are needed in the organisation of the manuscript, the reporting of analytical methods, and the clear separation between the Results and Discussion sections. The following comments are intended to help strengthen the scientific quality and readability of the manuscript.

Recommendation: Major Revision

Abstract

1. The abstract would benefit from an introductory sentence at the very beginning to contextualise the study before presenting specific findings.
2. The results section within the abstract is too long and overly detailed; it should be condensed to highlight only the key findings.

Introduction

• Lines 75–78 are better suited to Section 2.1 (Study Design and Population) rather than the Introduction.

Materials and Methods

• The ICP-MS instrument model, operating conditions, and LOD/LOQ values (among other analytical parameters) are missing and should be reported for methodological transparency and reproducibility.

Results

• Lines 216–218 read as discussion/interpretation rather than results and should be moved to the Discussion section.
• Similarly, lines 223–225 and 255–257 should be revised/relocated, as they also contain discussion-type content within the Results section.

Author Response

Dear Reviewer, you can find our detailed response to all your comments in the attached file.

Best regards

Angelo Riccio, on behalf of all co-authors

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Review Comments

This cross-sectional study conducted multi-element biomonitoring using adult hair samples from the Campania region of Italy. The authors measured 24 elements via ICP-MS to examine associations between hair metal concentrations and gender, lifestyle exposures, and geographical zones. Spatial visualization and cluster analyses were performed. The work aligns with the harmonized research framework of the European Human Biomonitoring Initiative (HBM4EU) and bears regional environmental public health relevance. Nevertheless, substantial methodological shortcomings exist in quality control, statistical approaches and spatial inference. Major Revision is recommended. Three major concerns are outlined below:

Major Comment 1

The authors adopted an alternating acetone–aqueous ultrasonic washing protocol for hair pretreatment. Ample literature has demonstrated that hair washing cannot completely eliminate external surface contamination from dust and cosmetic products. Furthermore, the analytical method was not validated using certified reference materials (CRMs) for hair matrices; only procedural blanks and replicate digestions were reported. The study identified significant negative correlations between sample mass and concentrations of Mg and Zn, which the authors attributed to weighing errors or detection limit issues without supporting validation experiments using serial sample mass gradients.

Supplementary information is required as follows: (1) spike recovery data for the complete ICP-MS analytical workflow; (2) accuracy validation of the washing and digestion protocol using matrix-matched hair CRMs; (3) pilot tests with gradient sample masses to disentangle the origin of negative mass-concentration correlations, discriminating instrumental bias, surface contamination and biological effects.

Major Comment 2

For elements with detection frequencies below 70% (e.g., Cd, Fe, Mn), non-detect values were directly assigned zero prior to skewed log model analysis. The authors themselves acknowledge that censored regression models such as Tobit regression or order-statistic regression are better suited for datasets with low detection rates.

The authors need to justify the use of the simplified zero-substitution strategy instead of widely accepted censored statistical approaches. Sensitivity analyses should be provided to compare association outcomes of key metals (Cd, As, V) under three handling schemes: zero substitution, substitution with half the limit of detection, and Tobit modelling, to verify whether the current statistical strategy alters core conclusions.

Major Comment 3

Figure 2 presents spatial visualizations of elemental medians and detection frequencies across four subareas of Naples, highlighting the Sorrentine Peninsula as the zone with the highest detection rates and concentrations of Pb, Cr, As and Cd. However, no formal spatial statistical tests (e.g., spatial autocorrelation tests, between-group comparisons of regional mean levels) were implemented. Geographic disparities were described merely qualitatively, precluding statistical confirmation of differing regional exposure levels. In addition, only 109 participants were included in spatial analyses, among which merely 9 samples originated from the Sorrentine Peninsula, indicating severely unbalanced sample sizes across zones.

The authors should implement Kruskal–Wallis tests with post-hoc multiple comparisons to quantitatively verify significant spatial disparities in metal concentrations across the four subareas. Stability assessments of median values are also needed to evaluate the impact of the extremely limited sample size (n=9) from the Sorrentine Peninsula.

Minor Suggestions

1. Figures: The legend text in Figure 2 is too small. Indicate zone-specific sample sizes below the figure (Naples: 32; Northern Naples: 38; Southern Naples: 30; Sorrentine Peninsula: 9). Add error bars representing interquartile ranges to bar charts to illustrate data dispersion.
2. References: When comparing findings with Italian population studies, distinguish physiological disparities between adult and child cohorts. Elaborate variations in hair washing and digestion protocols across published works to temper overinterpretation of cross-study comparisons.
3. Manuscript language: Multiple verbose and repetitive sentences require revision. Standardize units for trace metals throughout the text. Define abbreviations of statistical terms (e.g., FDR, HC3 robust standard errors) upon first mention.

Recommendation: Major Revision

All the above core concerns must be addressed point-by-point with supplementary laboratory experiments or statistical analyses. Without adequate revisions, the manuscript cannot satisfy the methodological and evidentiary standards for human biomonitoring research in Toxics. Acceptance will be re-evaluated only after comprehensive revision.

Author Response

Dear Reviewer, you can find our detailed response to all your comments in the attached file.

Best regards

Angelo Riccio, on behalf of all co-authors

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Review Comments

The authors have systematically addressed the three major methodological shortcomings and multiple minor suggestions raised in the initial review. All review points were responded to point‑by‑point, with substantial supplementary analyses including statistical sensitivity tests and spatial diagnostics, as well as optimized visualizations. Constraints of analytical approaches and interpretive boundaries for results have been emphasized in the main text, discussion and limitation sections, which greatly improves the overall rigor of this work. Nevertheless, inherent missing experimental conditions in the original dataset prevent retrospective addition of several key validation experiments, leaving non‑negligible methodological limitations. Minor Revision is recommended. Upon completion of remaining textual, logical and technical refinements, the manuscript will meet the publication criteria for human biomonitoring studies in Toxics.

Comment 1: Within the limitations section of the main text, the absence of matrix‑matched certified reference materials (CRM) and sample‑weight gradient validation experiments should be highlighted in bold as critical inherent weaknesses of the present study. Practicable quality‑control experimental workflows (e.g., gradient sample weighing, parallel digestion of hair certified reference materials, and whole‑process spike‑recovery tests) are suggested to be elaborated in the future‑work outlook, to strengthen reference implications for analogous follow‑up investigations.

Comment 2: Explicitly note in Section 3.2 (Spatial Results) that small sample size for the Sorrentine Peninsula compromises estimation precision. Stratified sampling with enlarged sample size for this region is recommended to mitigate sampling bias.

Final Decision

The authors have provided comprehensive responses to all core concerns from the initial review. Required statistical tests and sensitivity analyses have been supplemented; flaws in figures, text and comparative interpretations have been corrected. Limitations stemming from intrinsically unavailable experimental data of the original dataset have been objectively and sufficiently acknowledged, leading to more prudent and restrained interpretation of study conclusions. The manuscript satisfies methodological standards for human biomonitoring in Toxics.

Decision: Minor Revision

No further external peer‑review will be required once the authors implement the minor textual adjustments listed above. Acceptance for publication may be granted directly by the editorial office.

Author Response

Dear Reviewer, you can find a detailed response in the attached document.

Best Regards

Angelo Riccio

Author Response File: Author Response.pdf