Impact of Occupational Noise on Menstrual Disorders in Female Workers: A Systematic Review and Meta-Analysis

Abstract

Background:

Occupational noise exposure may impair female reproductive health, but evidence on its association with menstrual disorders remains limited.

Objective:

This systematic review and meta-analysis evaluated the association between occupational noise exposure and menstrual disorders among female workers, providing evidence for future research and occupational health policy.

Methods:

Peer-reviewed studies published from January 1, 1995, to December 31, 2024, were searched in PubMed, Web of Science, CNKI, Wanfang, and VIP. Eligible studies examined occupational noise exposure and menstrual disorders in female workers. Data on study characteristics, exposure assessment, diagnostic criteria, menstrual outcomes, and effect estimates were extracted. Study quality was assessed using tools for cross-sectional and cohort studies. Meta-analyses were performed using odds ratios (ORs) or risk ratios (RRs), as appropriate, with 95% confidence intervals (CIs) and subgroup analyses were conducted by study design, noise level, diagnostic criteria, and industry type.

Results:

Twenty-two studies were included in the review, and 18 studies were included in the meta-analysis. All were conducted in China, mostly in textile-related settings. Occupational noise exposure was significantly associated with menstrual disorders among female workers (OR = 3.27, 95% CI: 2.50–4.29, P < 0.01), including dysmenorrhea (RR = 1.76, 95% CI: 1.58–1.95), menstrual cycle irregularities (OR = 3.54, 95% CI: 2.54–4.94), abnormal menstrual duration (OR = 3.29, 95% CI: 1.65–6.59), and abnormal menstrual flow (OR = 2.52, 95% CI: 1.74–3.66). Higher noise levels, especially ≥90 dB(A), longer exposure duration, and nonsteady or impulsive noise may further increase risk. Substantial heterogeneity was observed.

Conclusion:

Occupational noise exposure may increase the risk of menstrual disorders among female workers. However, evidence is limited by observational designs, heterogeneous exposure and diagnostic methods, inadequate control for confounders, and a concentration in Chinese textile-related settings. Prospective multicenter studies with standardized criteria, objective noise assessment, and rigorous adjustment for confounders are needed.

KEY MESSAGES

• (1)Occupational noise exposure may be associated with menstrual disorders in female workers.
• (2)The associations were observed for dysmenorrhea, menstrual cycle irregularities, abnormal menstrual duration, and abnormal menstrual flow.
• (3)Higher noise intensity, longer exposure duration, and nonsteady or impulsive noise may be related to a greater prevalence of menstrual abnormalities.
• (4)Current evidence should be interpreted cautiously because of high heterogeneity, inconsistent diagnostic criteria, limited confounder control, and restricted geographic and industrial representativeness.

INTRODUCTION

Accelerated industrialization has led to a substantial increase in the female workforce across various sectors. By 2021, women comprised 45.3% of the global labor force, with particularly high representation in the manufacturing and healthcare industries.[1] By 2019, women constituted 47.0% of the total US labor force, reflecting near gender parity in workforce participation.[2] China’s female workforce has reached 300 million, constituting 44.0% of the total employed population. Significant representation was reported in the manufacturing sector (58.2%) and the healthcare sector (68.4%).[3] As the participation rate of women in the labor force continues to rise, the likelihood of female workers being exposed to noise is also increasing. Many women were employed in manufacturing industries, such as textiles and food processing, where they were frequently exposed to high levels of noise.[4] Increased noise exposure among female workers is a concern for the future.

Long-term exposure to occupational noise has been associated with multiple hearing impairments and significantly reduced workers’ quality of life.[5,6] In recent years, nonauditory health effects (e.g., cardiovascular diseases, hypertension, endocrine disorders, and reproductive system impairments) associated with noise exposure have been extensively investigated.[7,8,9] Evidence indicated that occupational noise exposure was associated with multiple reproductive health risks in female workers, including adverse pregnancy outcomes, reproductive endocrine disorders, and menstrual abnormalities.[4] Menstrual disorders were among the most prevalent reproductive health issues in women, with nearly half of female workers reporting menstrual symptoms.[10]

Several potential mechanisms by which noise exposure may lead to menstrual disorders in female workers have been reported. Occupational noise activated crosstalk between the hypothalamic–pituitary–adrenal (HPA) axis and hypothalamic–pituitary–ovary (HPO) axis, potentially contributing to reproductive endocrine disorders such as menstrual irregularities, anovulatory cycles, amenorrhea, and infertility.[11,12] Increased activation of the corticotropin-releasing hormone (CRH) system further disrupted the regularity of menstrual cycles.[13] Additionally, the combined effect of noise and psychological stress in specific occupational environments elevated the risk of menstrual disorders among female workers, negatively impacting their physiological functioning.[14] Furthermore, occupational noise exposure may also affect menstrual regularity indirectly by disrupting sleep and increasing stress levels.[15]

To date, there have been few literature reviews addressing the impact of occupational noise exposure on the reproductive health of female workers, particularly regarding menstrual disorders. In 1995, Nurminen et al.[16] analyzed the effects of occupational noise and shift work on women’s reproductive health. They found that high noise exposure [>85 dB(A)] increased the risk of reproductive health issues, notably a higher likelihood of low birth weight. Harmse et al.[17] identified a potential association between occupational noise exposure and menstrual cycle disorders through a comprehensive literature review. However, these reviews were limited to qualitative assessments. They lacked a quantitative analysis of the impact of noise exposure characteristics (e.g., noise levels, exposure duration, and noise type) on menstrual disorders. Therefore, there is a need for a literature review that examines both qualitative and quantitative evidence regarding the effects of occupational noise exposure on menstrual disorders in female workers.

This study employed a systematic review to map existing evidence on occupational noise exposure and menstrual disorders in female workers and to investigate the impact of exposure characteristics (e.g., noise intensity, exposure duration, and noise type). We subsequently conducted a meta-analysis to calculate pooled odds ratios (ORs) for overall and subtype-specific menstrual disorders, to assess the association between noise exposure and these outcomes. This integrated approach will provide a basis for protecting the reproductive health of female workers.

MATERIALS AND METHODS

Literature Retrieval

Systematic searches were executed across five databases: PubMed, Web of Science, CNKI, Wanfang, and VIP. A comprehensive review of published studies examining the effects of occupational noise exposure on menstrual abnormalities in female workers was conducted. Search parameters restricted publications to the English and Chinese languages, covering the period from January 1, 1995, to December 31, 2024. To ensure comprehensive search coverage, the term “occupational noise” and its synonyms were searched as keywords (connected with the OR is a Boolean operator used to combine synonyms or related terms), and “menstrual abnormalities” was searched as a subject heading. Both search histories were then combined using the AND operator. Additionally, eligible studies that were not retrieved through database searches were supplemented by reviewing the reference lists of included articles and other manual search methods. The Chinese search terms were defined as “occupational noise,” “noise,” “noise induced,” “noise exposure,” and “menstruation disturbance.” In contrast, the English search terms were specified as “occupational noise,” “noise,” “noise induced,” “noise exposure,” and “menstruation disturbance.” Additionally, the “literature review method” was used to screen relevant reviews on occupational noise exposure and menstrual abnormalities in female workers to identify studies that met the inclusion criteria.

Eligibility Criteria

A literature search was conducted using the population/participants/problem, intervention, comparison/comparator, outcome, and Study design (PICOS) and sample, phenomenon of interest, design, evaluation, and research type (SPIDER) tools for both quantitative and qualitative studies.[18] Studies were included if they met the following criteria: (1) they explicitly explored the direct relationship between occupational noise exposure and abnormal menstruation in female workers, including cycle disorders, irregular menstrual flow, dysmenorrhea, and similar issues; (2) the study participants were clearly defined as female workers, and the noise exposure was described either objectively or subjectively; (3) for cross-sectional, cohort, and case-control studies that met the inclusion criteria, a quality evaluation was performed. The literature needed to provide extractable effect size indicators, such as ORs, risk ratios, or hazard ratios, along with their 95% confidence intervals (CIs), or raw data for calculation. Only studies that scored 5 points or more on the quality assessment were included in the meta-analysis.

Exclusion criteria in this study were as follows: (1) studies where noise was not the primary exposure of interest, or where concurrent exposures (e.g., vibration, chemical toxins, extreme temperatures) precluded isolation of noise-specific effects; (2) the sample size was insufficient relative to the study design and outcome frequency; major uncontrolled recall bias was present; or exposure assessment was based on unvalidated tools; (3) duplicate publications identified via author overlap, identical datasets, or >80% text similarity.

Based on the inclusion and exclusion criteria, a total of 22 studies were included in the analysis, as shown in Figure 1.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of screening and selection process.

Literature Quality Assessment

This systematic review and meta-analysis were reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.[19,20,21] The methodological quality of the cross-sectional studies was assessed using the appraisal tool recommended by the Joanna Briggs Institute Centre for Evidence-Based Healthcare.[22] Study quality was assessed across eight domains, including study population, disease measurement, exposure assessment, confounding control, analytical methods, and others. Each item was rated as “Yes,” “No,” “Unclear,” or “Not Applicable.” Responses of “Yes” were scored 1 point, whereas “No,” “Unclear,” and “Not Applicable” received 0 points. Studies with a total score of ≥5 points were classified as high quality, whereas those with <5 points as low quality. According to the predefined inclusion and exclusion criteria, 18 cross-sectional studies were deemed eligible for inclusion in the systematic review and were therefore subjected to a methodological quality assessment. Among these studies, 12 achieved quality scores of 7 or higher, and the detailed results of the quality assessment are presented in Table 1. For the quantitative synthesis, studies were further screened based on the availability and suitability of data for meta-analysis. Consequently, 16 studies were ultimately included in the meta-analysis.

Table 1.

Quality Assessment Results of Included Studies.

No.	Reference (Publication Year)	Item 1	Item2	Item3	Item4	Item5	Item6	Item7	Item8	Total Score
1	Zhang et al.[23] (1995)	1	1	1	1	1	1	1	1	8
2	Zhou et al.[24] (1999)	1	1	1	1	1	1	1	1	8
3	Wang and Ren[25] (2001)	1	1	1	1	0	0	1	1	6
4	Du and Zhai[26] (2002)	1	0	1	0	1	0	1	0	5
5	Huo[27] (2004)	1	1	1	1	0	0	1	1	6
6	Zhang and Zheng[28] (2004)	0	1	1	1	1	1	1	1	7
7	Bai and Tang[29] (2005)	0	1	1	1	1	1	1	1	7
8	Zhao and Dong[30] (2005)	1	1	1	1	1	0	1	1	7
9	Song[31] (2007)	1	0	1	0	1	1	1	1	6
10	Hu et al.[32] (2007)	1	1	1	1	0	0	1	0	5
11	Liang et al.[33] (2011)	1	0	1	1	1	1	1	1	7
12	Ge[34] (2016)	1	1	1	1	0	0	1	1	6
13	Ou et al.[35] (2017)	1	1	1	1	1	1	1	1	8
14	Zhou and Liu[36] (2017)	1	1	1	1	1	1	1	1	8
15	Jiang et al.[11] (2019)	1	1	1	1	1	1	1	1	8
16	Song and Liu[37] (2020)	1	1	1	1	1	1	1	1	8
17	Lan et al.[38] (2020)	1	1	1	1	1	1	1	1	8
18	Wang et al.[39] (2023)	1	0	1	1	1	1	1	1	7

Item 1: Were the sample inclusion criteria clearly defined? Item 2: Were study subjects and setting described in detail? Item 3: Were exposure measures valid and reliable? Item 4: Were health outcomes defined using objective, consistent criteria? Item 5: Were potential confounders identified? Item 6: Were strategies applied to control for confounders? Item 7: Were outcome measures valid and reliable? Item 8: Were statistical methods appropriate? Scoring: 1 = Yes, 0 = No. Responses of “Yes” were scored 1 point, while “No,” “Unclear,” and “Not Applicable” received 0 points. Studies scoring ≥5 points were classified as high quality, and those with <5 points as low quality.

The methodological quality of the included cohort studies was independently evaluated using the Newcastle–Ottawa Scale (NOS).[40] The NOS assesses the quality of nonrandomized studies across three domains: selection (4 points), comparability (2 points), and outcome (3 points). Specifically, selection was assessed based on the representativeness of the exposed cohort, selection of the nonexposed cohort, ascertainment of exposure, and confirmation that the outcome of interest was absent at the start of follow-up. Comparability was evaluated according to whether the studies controlled for the most important confounder and additional potential confounding factors. Outcome assessment included the method of outcome ascertainment, adequacy of the follow-up period, and completeness of follow-up. The total NOS score ranges from 0 to 9, with higher scores indicating higher methodological quality. The complete results of the quality assessment are presented in Table 2.

Table 2.

Cohort Study Quality Assessment (NOS Scores).

No.	Reference (Publication Year)	Selection (★★★★)	Comparability (★★)	Outcome (★★★ or ★★)	Score
1	Xiao et al.[41] (2001)	★★★	★★	★★★	8
2	Wu et al.[42] (2004)	★★★	★	★	5
3	Li et al.[43] (2004)	★★★	★★	★	6
4	Sun et al.[44] (2007)	★	★	★	3

The Newcastle–Ottawa Scale (NOS) for cohort studies evaluates study quality across three domains: Selection, Comparability, and Outcome, with a maximum score of 9 stars (Selection, 4 stars; Comparability, 2 stars; Outcome, 3 stars). One star is awarded for each item if the criterion is met, except for the Comparability domain, in which up to two stars may be assigned.

Data Items

Data were organized and synthesized using Microsoft Excel 2019 (Microsoft Corporation, Redmond, WA, USA), with a comprehensive master spreadsheet used for final compilation and analysis. The extracted and synthesized information comprised: (1) study characteristics (e.g., authors, country, publication year, and study design); (2) key study components (outcomes, study population, and noise exposure metrics); and (3) investigative parameters: sample size, data sources. The primary study findings were summarized, and meta-analyses were conducted to quantify the magnitude of occupational noise exposure effects on both menstrual cycle abnormalities and dysmenorrhea among female workers.

Synthesis of Results

Based on the characteristics of the included studies, both quantitative and qualitative analytical approaches were employed. The main findings were systematically synthesized and presented in tables and figures, followed by a narrative summary. A total of 18 studies that met the inclusion criteria for meta-analysis were selected and analyzed using Review Manager 5 (RevMan 5.4: The Cochrane Collaboration, Copenhagen, Denmark). In accordance with Cochrane guidelines,[45] we implemented a rigorous methodology to enhance precision and strengthen the evidence base for addressing our research questions. Heterogeneity was interpreted using the I² statistic, with thresholds defined as follows: 0% to 40% may not be important, 30% to 60% may represent moderate heterogeneity, 50% to 90% may indicate substantial heterogeneity, and 75% to 100% may indicate considerable heterogeneity. Forest plots were generated to visualize effect sizes with 95% CIs and study weights for all included studies. All effect measures were reported with 95% CIs, and statistical significance was set at P < 0.05. For the meta-analysis examining the effect of occupational noise exposure on dysmenorrhea, a fixed-effect model was used when heterogeneity tests were nonsignificant (P > 0.05) and I² ≤ 50%. Otherwise, a random-effects model was employed. Publication bias was assessed using funnel plots. Additionally, subgroup and sensitivity analyses were performed to explore potential sources of heterogeneity.

For studies that did not directly report ORs and 95% CIs but provided the total number of participants and the number of menstrual abnormalities in both the exposed and control groups, we calculated crude ORs and 95% CIs from the four-fold table data and included them in the corresponding subgroup meta-analysis.

RESULTS

General Characteristics of Included Studies

A total of 22 studies were included, comprising 18 cross-sectional studies and 4 cohort studies. Among these, 18 studies were included in the meta-analysis. As summarized in Table 3, the studies encompassed 32,794 female workers exposed to noise and corresponding controls, with ages ranging from 18 to 55 years and noise exposure durations ranging from 1 to 23 years.

Table 3.

General Characteristics of the Included Studies.

No./Study Type	Authors and Year	Country	Research Subjects				Noise Level	Diagnostic Criteria*	Menstrual Abnormalities
			Subjects (n)	Industry	Age (Years)	Exposure Duration (Years)
Cross-sectional
1	Zhang et al.[23] (1995)	China	973	Textile	18–49	3–36	Exposed >107 dB(A)	Gynecology	Cycle, duration, volume abnormalities; dysmenorrhea
2	Zhou et al.[24] (1999)	China	289	Textile	24–40	4–20	Exposed 95–98 dB(A)	Gynecology	Cycle abnormalities, dysmenorrhea, prolonged duration, reduced volume
3	Wang and Ren[25] (2001)	China	322	Textile	24–39	4–18	Exposed >85 dB(A)	Occupational health	Cycle, duration, volume abnormalities; dysmenorrhea
4	Du and Zhai[26] (2002)	China	202	Textile	20–35	3–13	Exposed >85 dB(A)	Occupational health	Prolonged cycle, increased volume, dysmenorrhea
5	Huo[27] (2004)	China	973	Textile	25–40	1–26	Exposed >90 dB(A)	Occupational health	Cycle, duration, volume abnormalities; dysmenorrhea
6	Zhang and Zheng[28] (2004)	China	360	Textile	27.8 ± 6.7	9.1 ± 4.7	Exposed >97dB(A)	Occupational health	Cycle, duration, volume abnormalities; dysmenorrhea
7	Bai and Tang[29] (2005)	China	433	Textile	30.5 ± 4.8	4–20	Exposed >85 dB(A)Control no noise	Gynecology	Menstrual abnormalities
8	Zhao and Dong[30] (2005)	China	196	Textile	25–35	3–14	Exposed >90 dB(A)	Occupational health	Menstrual changes
9	Song[31] (2007)	China	2221	Machinery	20–45	—	Subjective assessment	Occupational health	Cycle, volume abnormalities; dysmenorrhea
10	Hu et al.[32] (2007)	China	963	Textile	24–44	—	Exposed >90 dB(A)Control <75 dB(A)	Gynecology	Irregular duration, timing, volume; dysmenorrhea
11	Liang et al.[33] (2011)	China	330	Steel	20–50	1–23	Exposed >85 dB (A)	Gynecology	Menstrual disorders
12	Ge[34] (2016)	China	380	Textile	18–45	—	Exposed >85 dB(A)Control <85 dB(A)	Occupational health	Volume, cycle abnormalities
13	Ou et al.[35] (2017)	China	2020	Railway	18–55	≥1	Subjective assessment	Gynecology	Menstrual irregularities
14	Zhou and Liu[36] (2017)	China	594	Steel	38.9 ± 8.2	14.8 ± 8.9	Exposed >85 dB(A)Control <85 dB(A)	Gynecology	Menstrual abnormalities
15	Jiang et al.[11] (2019)	China	12,881	Healthcare	—	—	Subjective assessment	American Clinical Gynecologic Endocrinology	Menstrual disorders
16	Song and Liu[37] (2020)	China	2358	Textile	—	—	Exposed >85 dB(A)Control <85 dB(A)	Occupational health	Cycle disturbances
17	Lan et al.[38] (2020)	China	298	Pharmaceutical	38.0 ± 5.1	≥1	Exposed >70 dB(A)	Gynecology	Menstrual abnormalities
18	Wang et al.[39] (2023)	China	1202	Telecommunication	36.81 ± 5.44	≥1	Subjective assessment	Gynecology	Menstrual abnormalities
Cohort study
1	Xiao et al.[41] (2001)	China	1640	Textile	18–40	≥1	Exposed >85 dB(A)Control <80 dB(A)	Gynecology	Menstrual disorders, dysmenorrhea
2	Wu et al.[42] (2004)	China	214	Textile	24–44	—	Exposed >85 dB(A)Control <85 dB(A)	Gynecology	Changes in cycle length
3	Li et al.[43] (2004)	China	175	Textile	20–34	—	Exposed >85 dB(A)Control <85 dB(A)	Gynecology	Cycle length
4	Sun et al.[44] (2007)	China	3770	Textile	18–45	≥1	Exposed 89–100 dB(A)Control 60–76dB(A)	Gynecology	Menstrual disorders, dysmenorrhea, amenorrhea, premenstrual tension

^*Gynecology: the diagnostic criteria from the 7th Edition of Obstetrics and Gynecology^[46]; Occupational health: analytical indicators for reproductive occupational epidemiological surveys proposed in 1989^[47]; American Clinical Gynecologic Endocrinology: the diagnostic criteria from Clinical Gynecologic Endocrinology.^[48,49]

As illustrated in [Figure 2], the number of studies reported per year was small, with peaks in 2004 and 2007 (four and three publications, respectively). All included studies (100%, n = 22) were conducted in China. In terms of study design, cross-sectional studies accounted for the highest proportion (81.8%, n = 18), followed by cohort studies (18.2%, n = 4). The included studies encompassed seven industrial sectors, with textiles representing the highest proportion (68.2%, n = 15), followed by steel manufacturing (9.1%, n = 2) and railway systems (4.5%, n = 1).

Figure 2.

Distribution of literature publication quantity regarding menstrual disorders associated with noise exposure. The number of studies reported per year was low, with publications from China accounting for 100%. Cross-sectional studies constituted the majority (81.8%), and the textile industry was the most frequently studied sector (68.2%). (a) Literature publication quantity grouped by countries in different years. (b) Literature publication quantity grouped by research type and industry.

Types and Definitions of Menstrual Disorders in Female Workers

The main types of menstrual disorders associated with noise exposure included menstrual cycle irregularities and dysmenorrhea. Among the included studies, 82.1% used menstrual cycle irregularities as outcome measures, while 42.9% used dysmenorrhea. The definitions of menstrual disorders varied across studies. They were primarily based on the 7th Edition of Obstetrics and Gynecology (China),[46] the analytical indicators for reproductive occupational epidemiological surveys proposed in 1989,[47] and Clinical Gynecologic Endocrinology (US).[48,49] A summary of the various diagnostic definitions indicated that standardized indicators for menstrual disorders could be categorized into four primary domains: menstrual frequency, duration of menstruation, volume of bleeding, and dysmenorrhea. This definition did not exclude other symptoms, such as premenstrual syndrome. The similarities and differences among these diagnostic definitions are presented in Table 4.

Table 4.

Diagnostic Criteria for Menstrual Disorders.

Clinical Manifestation	Diagnostic Criteria from China	Occupational Health Standards from China	American Clinical Gynecologic Endocrinology
Menstrual cycle	Frequent: <21 days; Infrequent: >35 days		>6 Days variation between consecutive cycles
Menstrual duration	Prolonged: >7 days; shortened: <3 days
Bleeding volume	Light: <20 mL; Heavy: >80 mL;with clots	Heavy: visible clots or >3 packs (30 pads)	—
Dysmenorrhea	Severe pain requiring medication/rest

Overall, the diagnostic heterogeneity across studies was substantial. Although most outcomes could be categorized into the four broad domains of menstrual cycle, duration, flow, and dysmenorrhea, the exact thresholds and operational definitions differed across diagnostic systems.

Subgroup Analysis Results

To further explore the possible sources of heterogeneity among studies, subgroup analyses were performed by study type, noise exposure level, diagnostic criteria, and industry type in this study, with results shown in Table 5.

Table 5.

Subgroup Analysis (Random-Effects Model) of the Association between Occupational Noise Exposure and Menstrual Abnormalities in Female Workers, Stratified by Different Factors.

Stratification Factor	Subgroup	K	OR (95% CI)	P	I2 (%)	P Value for Subgroup Difference
Study type	Cross-sectional	15	3.05 (2.38–3.90)	<0.001	89.7	0.060
	Cohort	4	1.39 (0.64–3.03)	0.402	94.2
Noise exposure level	85–90 dB(A) vs. control	2	1.17 (0.14–9.80)	0.887	97.0	<0.001
	≥90 dB(A) vs. control	14	3.24 (1.96–5.35)	<0.001	95.3
Diagnostic criteria	Occupational health	7	2.64 (2.05–3.40)	<0.001	70.1	<0.001
	Gynecology	10	2.72 (1.74–4.27)	<0.001	93.4
	American Clinical Gynecologic Endocrinology	1	1.75 (1.58–1.94)	<0.001	–
Industry type	Textile	12	3.13 (2.11–4.66)	<0.001	91.9	<0.001
	Other industries	6	1.82 (1.69–1.96)	<0.001	0.0

Subgroup Analysis by Study Type

Subgroup analysis by study type showed that a total of 15 effect sizes were included in the cross-sectional study subgroup, with a pooled OR of 3.05 (95% CI: 2.38–3.90) using a random-effects model, indicating that occupational noise exposure was associated with an increased risk of menstrual abnormalities in female workers, with a statistically significant difference and high heterogeneity (I² = 89.7%). A total of four effect sizes were included in the cohort study subgroup. Because some studies did not directly report OR values, we calculated crude ORs from the provided total numbers of participants and cases in the exposed and control groups and then included in the pooled analysis. The results showed that the pooled OR for the cohort study subgroup was 1.39 (95% CI: 0.64–3.03), with no statistically significant difference and high heterogeneity (I² = 94.2%). The test for subgroup differences showed that the difference between cross-sectional and cohort studies was not statistically significant (χ² = 3.54, P = 0.06).

Subgroup Analysis by Noise Exposure Level

Subgroup analysis according to noise exposure levels revealed the following findings: When using the control group as a reference, the subgroup exposed to noise levels of 85 to 90 dB(A) included two effect sizes. The pooled OR, calculated using a random-effects model, was 1.17 (95% CI: 0.14–9.80). This result was not statistically significant, and there was high heterogeneity (I² = 97.0%). In contrast, the subgroup exposed to noise levels greater than 90 dB(A) included 14 effect sizes. The pooled OR for this group was 3.24 (95% CI: 1.96–5.35), indicating a statistically significant difference, although there was still high heterogeneity (I² = 95.3%). The test for subgroup differences revealed a statistically significant variation in effect sizes across different noise exposure levels (χ²  = 21.61, P < 0.01). This suggests that the risk of menstrual abnormalities among female workers may increase with higher levels of noise exposure.

Subgroup Analysis by Diagnostic Criteria

Subgroup analysis according to diagnostic criteria revealed that the occupational health criteria subgroup included seven effect sizes. The pooled OR, using a random-effects model, was 2.64 (95% CI: 2.05–3.40), indicating a statistically significant difference with moderately high heterogeneity (I² = 70.1%). The obstetrics and gynecology criteria subgroup consisted of 10 effect sizes, with a pooled OR of 2.72 (95% CI: 1.74–4.27). This difference was also statistically significant, and the heterogeneity was high (I² = 93.4%). The American Clinical Gynecologic Endocrinology criteria subgroup included only one effect size, resulting in an OR of 1.75 (95% CI: 1.58–1.94). This difference was statistically significant, but due to the limited number of studies, we could not assess within-group heterogeneity. Furthermore, the test for subgroup differences indicated a statistically significant variation in effect sizes across the diagnostic criteria (χ² = 30.97, P < 0.01). This suggests that the differences in diagnostic criteria may be a contributing factor to the heterogeneity observed between studies.

Subgroup Analysis by Industry Type

Subgroup analysis by industry type showed that the textile industry subgroup included 12 effect sizes. The pooled OR using a random-effects model was 3.13 (95% CI: 2.11–4.66), with a statistically significant difference and high heterogeneity (I² = 91.9%). The other industries subgroup included six effect sizes, with a pooled OR of 1.82 (95% CI: 1.69–1.96). The difference was statistically significant, and no substantial heterogeneity was observed (I² = 0.0%). The test for subgroup differences showed a statistically significant difference in effect sizes across industry types (χ² = 47.21, P < 0.01), suggesting that industry type may be one source of between-study heterogeneity.

Effect of Noise Exposure Characteristics on Menstrual Disorders

The effect of exposure duration on menstrual disorders was examined in six included studies. Workers were stratified by exposure duration into three groups: exposure group 1 (0–5 years), exposure group 2 (5–10 years), and exposure group 3 (≥10 years). Multivariable logistic regression analysis of the five cross-sectional studies in Table 6 revealed that cumulative exposure duration had a significant effect on the occurrence of menstrual disorders (P < 0.05). The risk of menstrual disorders increased with longer exposure duration (P < 0.05), indicating exposure duration was a risk factor.

Table 6.

Logistic Regression Analysis of Menstrual Disorders by Exposure Duration in Cross-sectional Studies.

No.	Authors	Industry	Prevalence (%, Number of Menstrual Disorders/Total Subjects)
			Exposure 1*	Exposure 2*	Exposure 3*
1	Du and Zhai[26]	Textile	43.6 (17/39)	46.43 (13/28)	46.15 (18/39)
2	Zhang and Zheng[28]	Textile	20 (9/45)	27.78 (30/108)	27.66 (13/47)
3	Zhao and Dong[30]	Textile	42.5 (17/40)	48.28 (14/29)	48.72 (19/39)
4	Zhou and Liu[36]	Steel	21.88 (7/32)	19.05 (20/105)	53.19 (125/235)
5	Wang et al.[39]	Telecommunication	20.49 (101/493)	19.68 (73/371)	18.34 (62/338)
Summary			23.27 (151/649)	23.40 (150/641)	33.95 (237/698)†,‡

^*Exposure group 1: 0–5 years; Exposure group 2: 5–10 years; Exposure group 3: ≥10 years; ^†P < 0.05, compared with exposure group 1, OR = 1.459, 95% CI: 1.159–1.937; ^‡ P < 0.05, compared with exposure group 2, OR = 0.689, 95% CI: 0.547–0.868.

Only one study showed that nonsteady noise (OR = 2.5) had a more pronounced effect on menstrual disorders than steady noise (OR = 1.8). This is likely due to its high-intensity bursts disrupting endocrine function.[30] The prevalence of menstrual disorders varied by industry. Textile workers exposed to steady noise showed a prevalence of 32.1%, while machinery manufacturing workers exposed to nonsteady noise had a rate of 41.5%. All menstrual outcomes were more prevalent among workers exposed to impulsive noise,[31] confirming its greater detrimental impact.

Meta-analysis of Occupational Noise Exposure and Menstrual Disorders

Overall Effect of Occupational Noise Exposure on Menstrual Disorders

[Figure 3] shows the meta-analysis results and scatter plots for the 14 included studies. Substantial heterogeneity was observed among the studies (I² = 88%, P < 0.01), and a random-effects model was employed. The analysis included 5990 noise-exposed female workers and 5423 controls. A significant association was found between noise exposure and menstrual disorders (OR = 3.27, 95% CI:2.50–4.29, P < 0.01).

Figure 3.

Forest and funnel plots for the association between occupational noise exposure and menstrual disorders. (a) Forest plot showing a statistically significant association between occupational noise exposure and menstrual disorders [odds ratio (OR) = 3.27, 95% confidence interval (CI): 2.50–4.29, P < 0.01]. (b) Funnel plot evaluating publication bias for the association between occupational noise exposure and menstrual disorders.

Effect of Occupational Noise Exposure on Dysmenorrhea

Thirteen studies reported the association between occupational noise exposure and the prevalence of dysmenorrhea. Substantial heterogeneity was observed (I² = 56%, P < 0.01), and a random-effects model was employed. The results demonstrated that compared with the control group, occupational noise exposure was closely associated with dysmenorrhea in female workers (RR = 1.76, 95% CI: 1.58–1.95; P < 0.01), as shown in [Figure 4]a. The funnel plot is shown in [Figure 4]b.

Figure 4.

Forest and funnel plots for the associations of occupational noise exposure with dysmenorrhea and menstrual cycle irregularities. (a) Forest plot of occupational noise exposure and dysmenorrhea risk [relative risk (RR) = 1.76, 95% confidence interval (CI): 1.58–1.95, P < 0.01]. (b) Funnel plot assessing publication bias for dysmenorrhea. (c) Forest plot of occupational noise exposure and menstrual cycle irregularities [odds ratio (OR) = 3.54, 95% CI: 2.54–4.94, P < 0.01]. (d) Funnel plot assessing publication bias for menstrual cycle irregularities.

Effect of Occupational Noise Exposure on Menstrual Cycle

Twelve studies reported the association between occupational noise exposure and an irregular menstrual cycle. Substantial heterogeneity was observed (I² = 67%, P < 0.01), and a random-effects model was applied. Compared with the control group, occupational noise exposure was closely associated with menstrual cycle irregularities in female workers (OR = 3.54, 95% CI:2.54–4.94; P < 0.01), as shown in [Figure 4]c. The funnel plot is shown in [Figure 4]d.

Effect of Occupational Noise Exposure on Menstrual Duration

Twelve studies reported the association between occupational noise exposure and an abnormal menstrual duration. Substantial heterogeneity was observed (I² = 91%, P < 0.01), and a random-effects model was employed. The results showed that, compared with the control group, occupational noise exposure was closely associated with an abnormal menstrual duration in female workers (OR = 3.29, 95% CI: 1.65–6.59; P < 0.01), as shown in [Figure 5]a. The funnel plot is shown in [Figure 5]b.

Figure 5.

Forest and funnel plots for the associations of occupational noise exposure with abnormal menstrual duration and abnormal menstrual flow. (a) Forest plot of occupational noise exposure and abnormal menstrual duration [odds ratio (OR) = 3.29, 95% confidence interval (CI): 1.65–6.59, P < 0.01]. (b) Funnel plot assessing publication bias for abnormal menstrual duration. (c) Forest plot of occupational noise exposure and abnormal menstrual flow (OR = 2.52, 95% CI: 1.74–3.66, P < 0.01). (d) Funnel plot assessing publication bias for abnormal menstrual flow.

Effect of Occupational Noise Exposure on Menstrual Flow

Twelve studies reported the association between occupational noise exposure and an abnormal menstrual flow in female workers. Substantial heterogeneity was observed among the studies (I² = 83%, P < 0.01), and a random-effects model was used. The results showed that, compared with the control group, occupational noise exposure was closely associated with an abnormal menstrual flow in female workers (OR = 2.52, 95% CI: 1.74–3.66; P < 0.01), as shown in [Figure 5]c. The funnel plot is shown in [Figure 5]d.

Egger’s regression test showed significant funnel plot asymmetry for overall menstrual disorders, suggesting potential publication bias or small-study effects (k = 14, P = 0.016). The result for dysmenorrhea was also statistically significant but close to the significance threshold, indicating possible publication bias (k = 13, P = 0.048). A significant result was observed for menstrual cycle irregularities (k = 12, P < 0.001), suggesting a higher likelihood of publication bias for this outcome. In contrast, no significant evidence of publication bias was detected for abnormal menstrual duration (k = 12, P = 0.330) or abnormal menstrual flow (k = 12, P = 0.218), as shown in [Figure 6].

Figure 6.

Egger’s regression test for publication bias across different menstrual outcomes. Points indicate Egger’s intercepts, horizontal lines indicate approximate 95% confidence intervals (CIs), and the dashed line represents an intercept of zero. The corresponding results for each data point were as follows: overall menstrual disorders (intercept = 4.09, t = 2.79, P = 0.016), dysmenorrhea (intercept = 1.59, t = 2.23, P = 0.048), menstrual cycle irregularities (intercept = 3.20, t = 4.82, P < 0.001), abnormal menstrual duration (intercept = 1.79, t = 1.03, P = 0.330), and abnormal menstrual flow (intercept = 1.75, t = 1.31, P = 0.218).

DISCUSSION

Noise is an environmental stressor that reduces quality of life and overall well-being.[50] For workers, occupational noise is a common hazard in many industries, including textiles, electronics manufacturing, and food processing. In China’s manufacturing sector, approximately 30.06 million workers were exposed to occupational noise in 2020.[51] According to the 2020 National Survey on Occupational Disease Hazards in China, approximately 71.95% of workers in the sampled enterprises were exposed to noise hazards.[52] A European Union review of women’s work and health conditions found that noise levels in female work environments were often overlooked and inadequately monitored.[53] Additionally, one study reported that the utilization rate of hearing protection devices was three times lower among female workers than among male workers.[54] Furthermore, the prevalence of occupational hearing loss was high among female manufacturing workers.[55] The World Health Organization has emphasized the need for further investigation into occupational noise exposure and its potential gender-specific health effects.[56] Menstruation serves as an important indicator of female health. Evidence suggests that irregular and prolonged menstrual cycles are associated with an elevated risk of premature mortality.[57]

In this systematic review and meta-analysis, occupational noise exposure was associated with an increased prevalence of menstrual disorders among female workers. The pooled analysis showed that noise-exposed workers had a higher risk of overall menstrual disorders than unexposed or lower-exposed controls. Outcome-specific analyses further suggested significant associations between occupational noise exposure and dysmenorrhea, menstrual cycle irregularities, abnormal menstrual duration, and abnormal menstrual flow. Subgroup analyses indicated that higher noise intensity, particularly exposure levels ≥90 dB(A), may be associated with a greater risk. In addition, longer exposure duration and non-steady or impulsive noise appeared to be related to a higher prevalence of menstrual disorders. These findings suggest that both the intensity and characteristics of occupational noise exposure may be relevant to female reproductive health.

However, the findings should be interpreted cautiously. Although the overall pooled estimates indicated positive associations, substantial heterogeneity was observed across several analyses. This heterogeneity may be partly explained by differences in study design, diagnostic criteria, exposure assessment methods, industry type, noise intensity, exposure duration, and the degree of confounding control. In particular, the cross-sectional subgroup showed a stronger association than the cohort subgroup, whereas the pooled estimate from cohort studies was not statistically significant. This difference suggests that temporality and causal inference remain uncertain. Therefore, the current evidence supports a potential association between occupational noise exposure and menstrual disorders, but it does not establish a definitive causal relationship.

Diagnostic heterogeneity is an important methodological issue in the current evidence base. The included studies used different criteria to define menstrual disorders, including Chinese gynecological diagnostic criteria, occupational health standards, and criteria derived from clinical gynecologic endocrinology. These definitions differed in the operational thresholds for menstrual cycle length, cycle regularity, menstrual duration, menstrual flow, and dysmenorrhea severity. Such inconsistency may have caused outcome misclassification and reduced comparability across studies. Future studies should adopt standardized and internationally comparable diagnostic criteria for menstrual disorders, including clear definitions of cycle frequency, regularity, duration, bleeding volume, and dysmenorrhea severity. The international federation of gynecology and obstetrics (FIGO) systems for abnormal uterine bleeding provide a useful framework for standardizing terminology and improving comparability in reproductive-age populations.[58]

From a biological perspective, the observed association is plausible. Occupational noise may act as a chronic physical stressor, activating the sympathetic nervous system and the HPA axis. Noise-induced stress responses may alter the secretion of cortisol, catecholamines, and other stress-related hormones.[59] The HPA axis is closely connected with the HPO axis, which regulates gonadotropin-releasing hormone (GnRH) pulsatility, luteinizing hormone (LH), follicle-stimulating hormone (FSH), ovarian steroidogenesis, ovulation, and menstrual cyclicity. Stress mediators, including corticotropin-releasing hormone and glucocorticoids, may suppress GnRH secretion, disrupt LH pulsatility, and interfere with ovarian hormone production.[60] Through this HPA–HPO axis crosstalk, chronic occupational noise exposure may contribute to menstrual cycle irregularities, anovulatory cycles, luteal phase dysfunction, dysmenorrhea, or changes in menstrual bleeding patterns.

Another important issue is the insufficient control of confounding factors. Menstrual disorders are multifactorial and may be influenced by age, body mass index (BMI), age at menarche, reproductive history, parity, hormonal contraceptive use, smoking, alcohol consumption, physical activity, sleep quality, psychological stress, gynecological disease, and endocrine conditions. Occupational factors may also play an important role. Shift work, especially night-shift work, can disrupt circadian rhythms and has been associated with menstrual irregularities and dysmenorrhea.[61] Psychological stress is also associated with menstrual dysfunction, particularly irregular menstruation and abnormal menstrual flow.[62] In addition, female workers exposed to occupational noise may also be exposed to vibration, heat, chemical substances, organic solvents, heavy physical workloads, or other workplace hazards. These co-exposures may confound or modify the association between noise exposure and menstrual disorders.

In the included studies, adjustment for these confounding factors was inconsistent and often insufficient. Some studies identified potential confounders but did not adequately control for them in the analysis, while others relied on crude comparisons between noise-exposed and control groups. As a result, residual confounding cannot be excluded. Therefore, the pooled estimates should be interpreted as associations that may be affected by residual confounding rather than as direct causal effects of occupational noise alone.

Exposure assessment was another limitation of the current evidence. Several included studies used broad exposure categories, historical workplace measurements, or subjective exposure assessment, and relatively few studies performed individual-level quantitative noise monitoring. This may have introduced exposure misclassification and limited the ability to evaluate dose–response relationships. Future studies should use objective and standardized exposure assessment methods, including personal noise dosimetry, task-based measurements, 8-hour time-weighted average exposure levels, cumulative exposure indices, and indicators for impulsive or nonsteady noise. The national institute for occupational safety and health (NIOSH) recommends 85 dBA as the recommended occupational noise exposure limit over an 8-hour shift, and personal dosimeters are recommended for assessing individual worker exposure.[63] In addition, the international organization for standardization (ISO) 9612:2025 provides methodological guidance for measuring workers’ noise exposure and calculating occupational noise exposure levels, which may be useful for future epidemiological studies.[64]

The geographical and industrial distribution of the current evidence also limits generalizability. All included studies were conducted in China, and most were from the textile industry. Therefore, the pooled findings mainly reflect specific Chinese occupational settings, industrial structures, regulatory environments, work organization patterns, and workforce characteristics. These results should not be generalized uncritically to other countries, industries, or occupational populations. Noise exposure patterns, the use of hearing protection, shift schedules, co-exposures, occupational health regulations, socioeconomic conditions, and access to reproductive health services may differ substantially across countries and industries. Therefore, although the current review suggests that occupational noise may be an important reproductive health concern among female workers, the magnitude of risk estimated in this study should not be directly extrapolated to other national or industrial contexts.

This study has several limitations that should be considered when interpreting the findings. (1) Most included studies were observational, particularly cross-sectional studies, which limits causal inference. (2) Diagnostic criteria for menstrual disorders and noise exposure assessment methods varied across studies, contributing to between-study heterogeneity. (3) Although some studies considered potential confounders, adjustment was generally incomplete. Important factors such as age, BMI, reproductive history, shift work, psychological stress, sleep disturbance, and co-exposure to other occupational hazards were not consistently controlled, and residual confounding may therefore have influenced the observed associations. (4) The evidence base was geographically and industrially concentrated, as all included studies were conducted in China and most were from textile-related settings. Therefore, the present findings should not be generalized uncritically to other countries, industries, or workforce populations.

Future research should address these limitations through multicenter prospective cohort studies conducted across multiple countries and diverse industries. Such studies should enroll reproductive-aged female workers before outcome onset, collect repeated menstrual cycle data prospectively, and apply standardized diagnostic definitions. They should also conduct objective individual-level noise exposure assessment and collect detailed information on exposure duration, cumulative dose, shift schedules, night work, psychological stress, sleep quality, reproductive history, BMI, and co-exposure to other occupational hazards. Where feasible, future studies should incorporate biomarkers of endocrine function, including cortisol, LH, FSH, estradiol, progesterone, and ovulation indicators, to clarify whether occupational noise affects menstrual function through HPA–HPO axis dysregulation. These improvements would help strengthen causal inference, clarify dose-response relationships, and determine whether the observed associations are consistent across different occupational and national contexts.

CONCLUSIONS

Based on the available evidence, occupational noise exposure appears to be associated with an increased risk of menstrual disorders among female workers, including dysmenorrhea, menstrual cycle irregularities, abnormal menstrual duration, and abnormal menstrual flow. However, this conclusion should be interpreted cautiously because the current evidence base is dominated by studies from China, is concentrated in specific industries such as textiles, applies heterogeneous diagnostic criteria, and shows limited control for important confounding factors. Therefore, the present findings should not be generalized uncritically to other countries or industries. More well-designed, multicenter prospective studies with standardized diagnostic definitions and objective quantitative noise assessment are needed to confirm these associations and clarify their mechanisms.

Availability of Data and Materials

Data and materials may be requested from the corresponding author upon reasonable request.

Author Contributions

X.H. Li wrote the manuscript, designed the methodology, performed analysis, and curated data. Y.W. Huang designed the methodology, performed analysis, and curated data. Z.H. Wang contributed to methodology development and data curation. A.K. Zeng curated data and assisted in methodology design. L.J. Wu conceptualized the study and developed the methodology. J.R. Xin conceptualized the study and curated data. C.Y. Yu conceptualized the study, designed the methodology, supervised the work, and wrote and revised the manuscript. M.B. Zhang conceptualized the study, designed the methodology, supervised the work, and wrote and revised the manuscript. All authors read and approved the final manuscript.

Ethics Approval and Consent to Participate

Not applicable.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding Statement

This research was supported and funded by the Research on the Association Between Occupational Noise Exposure Characteristics and Reproductive Aging in Female Workers, Chinese Preventive Medicine Association; the Research on the Mode of Action of Kurtosis-Based Temporal Structure in Noise-Induced Hearing Loss, Capital Special Scientific Research Program for Health Development (Project No. 2026-2G-4271); and the Research on Comprehensive Prevention and Control Technologies for Occupational Diseases Caused by Physical Factors.