Innovative Questionnaire for Assessing Occupational Exposure to Airborne Chemical Pollutants: Development and Validation

Abstract

Background:

Industrial workers are often exposed to various chemical agents, making awareness of their exposure crucial for disease prevention and health improvement. Consequently, a low-cost and efficient method in this area would be highly beneficial.

Methods and Materials:

In the first stage, factors affecting exposure were extracted from scientific sources and be completed by interviewing experts (initial questionnaire). Then, the questionnaire standardization process (validity and reliability evaluation), face validity, content and structure were checked step by step and in necessary cases. In all these steps, some questions were deleted, some changed, and finally 105 questions remained which entered the next stage.

Results:

In the phase of face validity and quantitative content, respectively, questions with content validity ratio (CVR) and content validity index (CVI), less than 0.51 and 0.71 (32 questions) were deleted and some questions (19 questions) were corrected. The 115-question questionnaire entered the construct validity phase, and finally 105 questions remained. The result of the Kaiser–Meyer–Olkin test for the designed questionnaire was equal to 0.805. For Bartlett’s sphericity test, the value of 21547 = 6 was obtained with P > 0.001. For exploratory factor analysis, seven main characteristics dimensions (process, pollutant, workplace, pollutant source, operator, technical engineering, and organizational-management) were extracted. A cutoff point of 0.4 was considered for varimax rotation on 115 items included in the exploratory factor analysis. The total interclass correlation coefficient (ICC) was 0.777 for reliability.

Conclusion:

In the present study, a questionnaire was prepared and validated, which collects all information and objective and nonobjective factors affecting the airborne chemical exposure.

INTRODUCTION

Work activities and industrial processes lead to human exposure to various risks, including exposure to chemicals.[1,2] To reduce these risks and their effects, it is crucial to implement control measures.[3] Proposing necessary and effective controls relies on the prediction, identification, measurement, and evaluation of health risks by occupational health experts.[4,5,6,7] According to European Union regulations (Council Directive 89/391/EEC) and occupational health laws in countries like Estonia, employers must conduct systematic and written risk assessments.[8] Accurate identification of factors affecting exposure is the first stage of risk assessment and control,[9] and its results should be applicable in different areas of risk management.[10,11,12,13] Various methods and instruments have been developed by organizations and experts,[9,14,15] but many countries still face shortcomings in identifying, measuring, and evaluating workers’ exposure to chemicals.[16]

Determining exposure through sampling, analysis, and evaluation of workplace air pollutants typically occurs annually, requiring professional health experts, complex guidelines, and high costs, often leading to systematic errors. During pollutant measurement, the main causes of exposure and defects leading to pollutant release are not easily identified.[11,17,18,19,20] Moreover, pollution levels in many factories do not change throughout the year without control measures. Developing countries, including Iran, often prioritize spending on controlling workplace pollutants rather than measuring existing risks. This raises the question of whether factories should incur the costs of annual measurements. The complex process can be expensive and entails administrative challenges. According to Franxman et al.,[21] reducing red tape in organizations is essential. The costly procedures for annual air pollutant measurement contradict the theory of red tape cutting programs, aimed at reducing the burden of health and safety issues for increased productivity. Reputable organizations like the British Standard and European Norm, along with the American Industrial Hygiene Association (AIHA) and the National Institute for Occupational Safety and Health, suggest monitoring schedules based on measured exposure relative to permissible limits.[20,22] Changes in work style can also alter exposure. AIHA strategies indicate that if the environment remains unchanged, annual measurement may not be necessary. However, the challenge remains in verifying whether exposure has changed during this period.

Chronic occupational diseases, such as respiratory diseases or cancer, often result from long-term exposure to harmful agents. Evaluating the dose-response relationship requires measuring exposure values throughout the working period, a method that seems impractical. Collecting quantitative exposure data for the entire working period is complex and not readily available.[23]

Burzoni et al. noted that while quantitative assessment has advantages, it also has limitations; qualitative risk assessment, or initial risk assessment, is essential for any preventive approach.[24] Awareness of different effective factors in occupational exposure is crucial for recognizing these factors and predicting possible health effects.[25,26]

Therefore, a simple preventive solution and checklist[27] that outlines relationships, conditions, and evidence of exposure to pollutants is needed.

The purpose of this study was to design, standardize, and validate a practical questionnaire for the identification, monitoring, and screening of all objective and subjective factors affecting occupational exposure to airborne chemical pollutants in industrial settings. The tool is intended to be used systematically and consistently by occupational health professionals, providing feedback to improve the quality of control measure implementation and to prioritize high-risk jobs.

METHODS

Study design

This research was conducted in three stages:

Stage 1: Literature Review and Factor Identification

A comprehensive literature review was performed on the relevant topics to identify factors affecting occupational exposure. These factors were further refined and supplemented through interviews with subject-matter experts.

Stage 2: Questionnaire Design and Validation

Based on the identified factors, the initial version of the questionnaire was developed. Its face validity, content validity, and structural aspects were then assessed to ensure clarity, relevance, and comprehensiveness.

Stage 3: Reliability Testing

Finally, the reliability of the questionnaire was evaluated, and the tool was finalized, standardized, and prepared for practical application in industrial settings [Figure 1].

Figure 1.

Flow diagram of sample attendance and questionnaire validation procedures

Procedures

The development and validation of the questionnaire were conducted following the DeVellis (2003) scale development framework.[28] Phase I involved defining the constructs and generating the initial item pool through literature review and expert interviews. Phase II included face validity, content validity (CVR/CVI), and construct validity (exploratory factor analysis). Phase III assessed reliability through internal consistency (Cronbach’s alpha) and test–retest (interclass correlation coefficient (ICC)). Items were revised or removed based on expert feedback and statistical criteria, resulting in a standardized and validated tool suitable for occupational exposure assessment.

This questionnaire was designed and developed by professional health experts and other trained experts for the purpose of initial evaluation and identification of factors affecting the exposure to chemical pollutants.

Phase I: Literature review (item determination/development)

To systematically identify the key determinants influencing occupational exposure to airborne pollutants among workers, a targeted literature search was conducted using predefined keywords in Scholar, Embase, Scopus, PubMed, Science Direct, Web of Science, Wiley, Taylor and Francis, Springer, SID, and IranMedex databases, covering publications from 1996 to 2018. The screening process yielded 89 original research articles, 12 review articles, and 8 Work Procedures and Reports for in-depth analysis. A subset of the reviewed studies is shown in Table 1 for comparative evaluation.

Table 1.

Some studies with the factors affecting the exposure and concentration of chemicals mentioned in them

Study	Factors influencing exposure/concentration
Hopkins & et al.[29]	Worker behaviors, such as the use of ventilation systems, other good work practices, and cleaning and sanitizing measures (5s), are altered by training and subsequently cause changes in the potential to reduce exposure levels.
Cherrie & et al.[30]	Work space dimensions Ventilation rate/number of air changes Duration of work
Herber & et al.[31]	Physical processes, chemical processes, the workplace, workers’ behavior
Lee & et al.[32]	Position, direction, and activity of the worker pollutant distribution pattern (depending on: air flow speed in the room, effects of source factors such as pollutant composition, density, and pollutant production rate) Some other factors affecting the speed and concentrations of the workshop: type of air supply diffusers, temperature difference between the walls and the room air, room configuration
Basinas & et al.[33]	Ventilation rates and specifications Work frequency, work conditions, work duties, health conditions of the working environment and the person, the type and material of the structure and the type of activity or process, equipment repair, cross section of the working environment Frequency and method of cleaning the working environment and equipment, temperature, humidity, and wind speed
Jones & et al.[34]	Activities and assigned tasks
Mater & et al.[35]	Individual protection, type of work environment, general ventilation, group protection, estimated production activity, temperature range of the work environment, type of process that produces the chemical pollutant (causing exposure), frequency of exposure, product or process causing exposure, temperature of use exposure product

Searched Keywords: Determinants of Exposure, Design of Control Measures, Measures Evaluation of Control, Exposure Modifying Factors, Exposure Control Banding, Environmental Interventions, Behavioral Interventions, Occupational Exposure Interventions, Occupational Engineering Techniques, Occupational Technical Controls, Chemical Agent Engineering Controls, Workplace Environment Control, Work Practice Management, Administrative Controls, Workplace Exposure Control Strategies, Industrial Hygiene Prevention, Engineering Control Strategies, Prevention Strategies, Administrative Strategies, Hierarchy Controls of Workplace Exposure, Chemical Agent exposure Determinants, New Strategies Control of Workplace, Workplace Modifications, Reducing Worker Exposure, Methods for Reduction of Exposures, Design of Control Measures.

Finally, the factors affecting the exposure were completed through interviews with experts and in the form of a qualitative study, the results of which are given in other studies [Table 1].

Phase II: Content, face, and construct validity

For the initial design of the questionnaire, the findings of the literature review and interviews were used. The similar items in the findings of the two sections were merged together with the independent items in both sections in the initial version. In this phase, face validity, content validity, and construct validity were used to standardize the tool.

A) Face validity

The qualitative method was used to determine the face validity. This work was reviewed by 15 experts from industries and health centers (female, male, and bachelor’s level) who were familiar with the subject and were selected in an easy (accessible) way. Regarding ambiguous and controversial questions and options, they were interviewed and revised.

B) Content validity

In this study, two quantitative and qualitative methods were used to determine the content validity. In the qualitative validity section, the opinions of 10 people were used, and in the quantitative validity section, the opinions of 14 experts and professional health professors were used who reviewed the content of the questions of the proposed tool in a comprehensive and regular manner using relevant methods in the qualitative section.

In the qualitative content validity review, the above people were asked to give their opinions about each of the questions designed in the questionnaire based on the criteria of following the grammar, using the right words, necessity, importance and proper placement of phrases.

To quantitatively check the content validity, two content validity checking methods were used including: content validity ratio (CVR) and content validity index (CVI). Table 2 shows the questionnaire used to collect expert opinions for determining the CVR and CVI in this section.

Table 2.

Questionnaire used to collect expert opinions to determine CVR and CVI

Question	Suggested Likert scale for questions	CVI
		Simplicity				Related
		Quite simple	Simple	Somewhat complicated	Complicated	Totally relevant	Related	Somewhat related	Irrelevant

Question	CVI Clarity				CVR
	Quite clear	Clear	Somewhat clear	Unclear	Necessary	Useful but unnecessary	Unnecessary

CVI=Content validity index; CVR=Content validity ratio

First, to determine the CVR, experts were asked to review each question based on a three-part spectrum (it is necessary, it is useful but not necessary, it is not necessary).

Then, the CVR was calculated based on equation 1.[36] Finally, according to the obtained value and the Lawshe table [Supplementary 2], the CVR of that item was judged.[36]

SUPPLEMENT 2.

Lawshe’s table for CVR

Number of experts	Minimum Value
5	0.99
6	0.99
7	0.99
8	0.78
9	0.75
10	0.62
11	0.59
12	0.56
13	0.54
14	0.51
15	0.49
20	0.42
25	0.37
30	0.33
35	0.31
40	0.40

Questionnaire standardization scores and factor loading of remaining questions after varimax rotation

Number of question	CVR	CVI				Result	factor loading
		Simplicity	related	Clarity	Total
1	1	0.86	1	0.93	0.93	*	0.792
2	0.71	0.79	0.86	0.79	0.81	*	0.664
3	1	0.93	1	0.93	0.95	*	0.693
4	1	0.79	1	0.86	0.88	*	0.664
5	0.43	0.79	0.71	0.71	0.74	eliminated
6	0.86	0.71	0.93	0.79	0.81	*	0.736
7	0.29	0.71	0.64	0.79	0.71	eliminated
8	0.57	0.79	0.86	0.64	0.71	*	eliminated
9	0.71	0.71	0.86	0.64	0.76	*	0.672
10	0.71	0.93	0.86	0.86	0.88	*	0.575
11	0.57	0.79	0.79	0.79	0.79	*	0.481
12	0.57	0.86	0.86	0.93	0.88	*	0.517
13	1	0.86	1	0.86	0.9	*	0.597
14	1	0.79	1	0.79	0.86	*	-0.674
15	1	0.71	1	0.79	0.83	*	0.547
16	0.43	0.57	0.93	0.5	0.67	eliminated
17	1	0.79	1	0.79	0.86	*	0.437
18	0.86	0.71	1	0.71	0.81	*	0.646
19	1	0.71	1	0.71	0.81	*	0.623
20	0.57	0.71	0.86	0.71	0.76	*	eliminated
21	1	0.86	1	0.64	0.83	*	0.825
22	1	0.71	1	0.64	0.79	*	0.796
23	1	0.64	1	0.64	0.76	*	0.721
24	0.29	0.71	0.57	0.79	0.69	eliminated
25	0.86	0.43	0.93	0.5	0.62	eliminated
26	1	0.71	0.93	0.86	0.83	*	0.583
27	1	0.86	1	0.86	0.9	*	0.772
28	1	0.71	0.93	0.71	0.79	*	0.547
29	0.57	0.71	0.86	0.71	0.76	*	0.472
30	0.57	0.64	0.86	0.64	0.71	*	0.604
31	0.57	0.64	0.93	0.71	0.76	*	0.643
32	0.57	0.57	0.86	0.71	0.71	*	0.441
33	0.29	0.93	0.57	0.93	0.81	eliminated
34	0.71	0.79	0.93	0.86	0.86	*	-0.670
35	0.57	0.71	1	0.64	0.79	*	0.481
36	0.57	0.79	0.93	0.86	0.86	*	0.634
37	0.71	0.79	0.93	0.86	0.86	*	0.593
38	0.71	0.79	0.93	0.86	0.86	*	eliminated
39	0.29	0.71	0.64	0.71	0.69	eliminated
40	0.14	0.64	0.71	0.71	0.62	eliminated
41	1	0.93	1	0.93	0.89	*	-0.763
42	0.29	0.93	0.71	0.93	0.83	eliminated
43	1	0.86	1	0.79	0.88	*	0.533
44	0.71	0.57	1	0.43	0.67	eliminated
45	0.14	0.64	0.57	0.79	0.67	eliminated
46	0.71	0.57	0.93	0.79	0.76	*	0.656
47	0.71	0.57	0.93	0.79	0.76	*	0.646
48	0.43	0.57	0.79	0.79	0.71	eliminated
49	0.57	0.86	0.71	0.86	0.81	*	0.538
50	1	0.71	1	0.86	0.86	*	0.778
51	1	0.93	1	0.93	0.95	*	0.796
52	0.57	0.64	0.93	0.64	0.74	*	0.433
53	0.29	0.57	0.79	0.5	0.62	eliminated
54	0.57	0.79	0.86	0.71	0.79	*	eliminated
55	0.43	0.36	0.71	0.36	0.48	eliminated
56	0.14	0.79	0.57	0.79	0.71	eliminated
57	0.14	0.79	0.57	0.79	0.71	eliminated
58	0.57	1	0.79	1	0.93	*	0.663
59	0.57	1	0.79	1	0.93	*	0.655
60	0.57	1	0.79	1	0.93	*	0.644
61	0.57	0.79	0.93	0.86	0.86	*	0.544
62	0.57	0.79	0.86	0.79	0.83	*	0.494
63	0.57	1	0.79	1	0.93	*	-0.723
64	0.86	0.79	0.93	0.86	0.86	*	0.586
65	1	0.86	1	0.86	0.9	*	0.785
66	0.86	0.71	1	0.79	0.83	*	0.716
67	0.57	0.86	0.93	0.86	0.88	*	0.481
68	0.29	0.86	0.71	0.86	0.81	eliminated
69	0.57	0.71	1	0.64	0.79	*	eliminated
70	0.29	0.71	0.64	0.71	0.67	eliminated
71	0.71	0.71	0.86	0.86	0.81	*	0.535
72	0.71	0.71	0.86	0.86	0.81	*	0.577
73	0.57	0.71	0.86	0.71	0.76	*	0.471
74	0.86	0.71	0.86	0.86	0.81	*	0.583
75	0.86	0.71	0.86	0.86	0.81	*	0.631
76	0.57	0.71	0.79	0.71	0.73	*	0.465
77	0.86	0.71	0.93	0.86	0.83	*	0.615
78	0.71	0.71	0.86	0.86	0.81	*	0.497
79	0.86	0.71	0.93	0.86	0.83	*	0.566
80	0.71	0.71	0.86	0.86	0.81	*	0.503
81	0.57	0.86	0.86	0.86	0.86	*	0.529
82	0.29	0.64	0.79	0.86	0.76	eliminated
83	0.14	0.64	0.64	0.64	0.64	eliminated
84	0.43	0.71	0.71	0.86	0.76	eliminated
85	0.43	0.71	0.71	0.86	0.76	eliminated
86	1	0.79	1	1	0.93	*	eliminated
87	1	0.71	1	0.86	0.86	*	eliminated
88	0.86	0.79	1	0.79	0.86	*	eliminated
89	0.43	0.86	0.79	0.79	0.81	eliminated
90	1	0.79	1	0.79	0.86	*	-0.545
91	1	0.86	1	0.86	0.9	*	-0.617
92	1	0.64	1	0.71	0.79	*	-0.519
93	1	0.71	1	0.71	0.81	*	-0.553
94	1	0.79	1	0.79	0.86	*	-0.685
95	1	0.71	1	0.79	0.83	*	-0.821
96	0.57	0.79	0.79	0.86	0.81	*	0.573
97	0.57	0.86	0.86	0.79	0.83	*	0.512
98	0.57	0.71	0.93	0.86	0.83	*	0.449
99	0.29	0.71	0.79	0.79	0.76	eliminated
100	1	0.79	1	0.79	0.86	*	-0.669
101	1	0.71	1	0.79	0.83	*	-0.627
102	0.86	0.64	0.93	0.79	0.79	*	0.461
103	0.57	0.93	0.86	0.93	0.9	*	0.437
104	0.14	0.86	0.5	0.86	0.74	eliminated
105	0.71	0.79	1	0.79	0.86	*	0.535
106	0.43	0.71	0.79	0.79	0.76	eliminated
107	0.57	0.86	0.86	0.79	0.83	*	0.478
108	1	0.71	1	0.79	0.83	*	0.625
109	1	0.71	1	0.79	0.83	*	-0.543
110	1	0.79	1	0.79	0.86	*	-0.605
111	0.57	0.79	1	0.79	0.86	*	0.547
112	0.57	0.79	1	0.79	0.86	*	0.573
113	0.43	0.79	1	0.79	0.86	eliminated
114	0.57	0.79	0.86	0.79	0.81	*	0.538
115	0.57	0.79	0.86	0.79	0.81	*	-0.479
116	0.57	0.79	0.86	0.79	0.81	*	0.436
117	0.57	0.79	0.86	0.79	0.81	*	eliminated
118	1	0.79	1	0.79	0.86	*	-0.583
119	1	0.79	1	79/0	0.86		-0.492
120	0.86	0.64	1	71/0	0.79	*	-0.568
121	0.86	0.64	1	71/0	0.79	*	0.658
122	0.57	1	0.79	0.93	0.9	*	0.749
123	0.71	0.71	0.86	0.86	0.81	*	0.763
124	0.57	0.79	0.79	0.93	0.83	*	0.726
125	0.57	0.93	1	0.86	0.93	*	eliminated
126	0.29	0.71	1	0.5	0.74	eliminated
127	0.29	0.93	0.86	0.93	0.9	eliminated
128	0.57	0.71	0.86	0.71	0.76	*	0.424
129	1	0.93	1	0.93	0.95	*	0.535
130	0.86	0.71	0.86	0.64	0.74	*	-0.684
131	0.43	0.86	0.79	0.79	0.81	eliminated
132	0.43	0.86	0.79	0.79	0.81	eliminated
133	1	0.79	1	0.93	0.9	*	0.483
134	1	0.79	1	0.93	0.9	*	0.454
135	1	0.71	0.86	0.71	0.76	*	-0.777
136	0.43	0.86	0.79	0.79	0.81	eliminated
137	0.43	0.86	0.79	0.79	0.81	eliminated
138	1	0.71	1	0.79	0.83	*	0.665
139	0.86	0.71	0.93	0.71	0.81	*	0.565
140	0.86	0.64	0.93	0.71	0.79	*	0.578
141	0.71	0.71	1	0.79	0.83	*	0.518
142	1	0.71	1	0.79	0.83	*	0.546
143	0.86	0.79	0.93	0.71	0.81	*	0.564
144	1	0.79	1	0.71	0.83	*	0.804
145	0.71	0.79	1	0.79	0.86	*	0.582
146	1	0.93	1	0.93	0.95	*	0.526
147	1	0.79	1	0.79	0.86	*	0.488

where

ne: Number of experts who select the “Nessesary” option.

N: Total experts

According to the Lawshe table to determine the minimum value of the CVR, expressions whose CVR value was higher than 0.51 (based on the evaluation of 14 experts) were retained.

The evaluation of CVI was conducted with the help of Waltz and Bausell method.[37] Regarding ensuring whether the items are designed to measure the indicators in the best way or not, regarding the three criteria of “simplicity and fluency,” “relevance or specificity,” and “clarity or transparency” of each of the questions in the questionnaire according to Table 2, the panel of experts commented on a four-point Likert scale. Then, the CVI was determined based on the experts’ opinions. Separately, in the three areas of “relevance,” “clearness,” and “simplicity,” the number of experts who chose the two higher options (clear-completely clear, simple-completely simple, and relevant-completely relevant) was divided by the total number of all experts. Then, the average of all three results was obtained.[21,38]

To determine the CVI for each phrase, the number of experts agreeing for the phrase with ranks 3 and 4 was divided by the total number of experts and thus the desired index was calculated.[39] Then, the average CVI was calculated based on the average scores of the CVI of all questionnaire statements. In this way, the content validity of the questionnaire of determinants of exposure to airborne chemical pollutants was prepared. The judgment and acceptance of the items were based on the following criteria:

CVI score higher than 0.79 was recognized as a suitable value. The score between 0.7 and 0.79 was questionable and needed correction and revision. The CVI score less than 0.7 was unacceptable and was removed.[36]

C) Construct validity—exploratory factor analysis and determination of factor loadings

To check the construct validity (exploratory factor analysis) of the questionnaire, working experts who will probably use this tool in the future were used. To do so, after providing explanations and training about the questionnaire, the experts were told that this questionnaire is better to be used for a pollutant and a specific task in the industry. Therefore, they were asked to complete the questionnaire by performing a basic job analysis for each job (for the considered activities and pollutants they should do in the form of easy (available) sampling). Then, in each job, the main tasks that take the most time are considered and the questionnaire for its specific pollutants is completed.

A sample of 523 people was collected for different pollutants (people working in the same job in the same industry and similar industries, which were collected by 33 experts). It was evaluated by exploratory factor analysis using the “principal component analysis” method with Varimax rotation to determine the main dimensions of the tool for assessing exposure to airborne chemical pollutants.[40]

To judge the results of exploratory factor analysis, Kaiser–Meyer–Olkin (KMO) index and Bartlett’s sphericity test were used.[41,42] For analysis, after calculating the correlation matrix between variables, factors were extracted. The decision on the number of factors was made using the eigenvalue and sand graph to determine the number of factors that make up the tool. The turning point of 0.4 was considered as the minimum factor loading needed to maintain the item in the factors extracted from the factor analysis, and the items with a coefficient lower than this value were not included in the instrument.

Initial analysis was performed with an eigenvalue greater than one. Because the number of factors was large and could not be interpreted with the results of the research, therefore, the analysis with an eigenvalue of 2.3 and above was performed from the sand graph to determine the extracted factors and make them interpretable. Twenty-four factors were selected using the sand chart. The factors that were on the steep slope of the graph were separated from the factors with low eigenvalue on the slow slope of the graph. Since the content analysis process of the qualitative part had reached seven main indicators (seven main themes), the process of exploratory factor analysis based on seven main factors was repeated again. In principal components analysis, Varimax rotation, which is an orthogonal rotation, was used to examine classes of variables that have the most correlation with each other. In this case, the factors are mathematically rotated to make the extracted factor structure simple and interpretable. The steps of construct validity (dimension) using exploratory factor analysis are shown in Figure 2.

Figure 2.

Stages of construct validity using exploratory factor analysis

Phase III: Reliability

Instrument reliability

To measure the reliability of the designed tool, two methods were used: internal consistency and retesting.

To measure the internal homogeneity of the instrument, Cronbach’s alpha coefficient was calculated for each dimension as well as the entire questionnaire for 708 questionnaires on which confirmatory factor analysis was performed. It should be noted that the instrument designed in this study had Likert scale questions, but the number of items varied from four to nine items. However, because the instrument is completed by experts and has a checklist mode (most of its questions are visible in the environment) and only the questions that have a question mode needed a reliability test, but it was conducted for the entire questionnaire.

In this study, to check the stability of the questionnaire, the retest method was also used. To determine the retest coefficient of the instrument, the questionnaire was completed for the second time in two parts, one in the content validity evaluation stage and in the data collection stage with an interval of two weeks. In the first stage, content validity was evaluated by 10 out of 14 professors and experts, and in the second stage, data collection was completed by 15 experts from industry and health centers. Then, the retest coefficient of the scores obtained in each was calculated separately using the ICC and values of 0.4 or higher were accepted [Table 3].[43]

Table 3.

The questionnaire development and validation steps to the DeVellis scale framework

DeVellis stage	Activities conducted in this study	Tools/methods used	Output/result
1. Define what to measure	Identification of factors affecting airborne chemical exposure	Systematic literature review, expert interviews	Clear definition of the scope and indicators of exposure
2. Generate item pool	Integration of literature review findings and expert interviews	Content analysis, item extraction	Initial list of questionnaire items
3. Determine response format	Selection of Likert scale and multiple-choice options	Review by the research team	Defined response format for accurate measurement
4. Expert review	Face validity and content validity assessment	Interviews with 15 experts, CVR and CVI calculation	Revision of ambiguous items, removal or modification of inappropriate items
5. Consider validation items	Construct validity assessment	Exploratory factor analysis (EFA) with PCA and Varimax rotation	Identification of main dimensions and factor loadings of each item
6. Administer to development sample	Data collection from a sample of 523 participants	Questionnaire implementation across different industries	Prepared dataset for factor analysis
7. Evaluate the items	Removal of weak items, reliability assessment	Cronbach’s alpha, ICC, factor loading ≥0.4	Ensured stability and internal consistency of the instrument
8. Optimize scale length	Removal or modification of invalid items	Combination of statistical analysis and expert feedback	Final standardized questionnaire suitable for industrial application

ICC=Interclass correlation coefficient; PCA=Principal component analysis

RESULTS

Phase I: Literature review (item determination/development)

In the conducted search, approximately 89 documents were examined, both partially and fully. Relevant sources were thoroughly analyzed, and determining factors affecting exposure were extracted. After five revisions, these factors were categorized into appropriate classes and subclasses. Factors that were directly mentioned in the sources were compiled into a table according to their titles. Duplicate items were removed, and the remaining factors were classified based on the main and subheadings identified in the sources. Table 1 lists several of the sources along with the factors influencing the exposure and concentration of chemical substances.

To enhance the factors affecting exposure, interviews with experts were conducted. This qualitative study provided additional insights, and the results will be presented in a separate publication.

Phase II: Content, face, and construct validity

Content and face validity

The initial questionnaire comprised 225 questions. However, 44 questions were removed due to overlap and redundancy. From the remaining 181 questions, four questions that estimated the qualitative amount of exposure were separated from the factors affecting exposure. This left a total of 177 questions, which were categorized into seven dimensions.

To determine qualitative face validity and address ambiguities, the opinions of 15 experts from various industries and health centers were solicited through a convenient sampling method. Based on their feedback, 19 questions were deleted, and 34 questions that exhibited ambiguity or difficulty were modified. Further necessary corrections were made to achieve the desired qualitative validity of the content, involving 10 experts. This process resulted in the deletion of 13 questions, the addition of two new questions, and the modification of 24 others.

Additionally, to measure the CVR, the opinions of 14 experts were utilized. As a result, 29 expressions were removed for scoring below 0.51 according to the Lawshe table. Following this, the CVI was assessed based on the methodology of Waltz and Bausell. The average scores indicated that three questions with a CVI score below 0.71 were removed, while 19 questions with scores between 0.71 and 0.79 were modified. Ultimately, this process yielded a questionnaire consisting of 147 questions, with the structure of the questions detailed in Supplementary 1.

SUPPLEMENT 1.

Final Questinnaire

Area	Number	Question
Workplace Environmental characteristics (13question)	1	In what environment is the workstation located?
	2	What is the work station like in terms of geometric shape and presence of dead spaces?
	3	What is the size of the workplace?(m3)
	4	How is the relative location of the workshop compared to other potential sources of pollutant emission in the neighborhood outside the workshop?
	5	What is the condition of the building in terms of structure and the existence of disturbing washing and cleaning facilities?
	6	What is the condition of the design and structure of the workshop building for the implementation of control systems?
	7	What is the type of climate in which the workshop is located?
	8	What is the type and nature of the industry (index) based on the investigated pollutant?
	9	What is the type and nature of neighboring and surrounding industries affecting the exposure (according to the investigated pollutant and based on the index (amount) of industrial pollution)?
	10	What is the ambient temperature of the workplace?
	11	What is the environmental humidity of the workplace?
	12	What is the regional barometric pressure of the workplace?
	13	What is the speed of the ambient wind flow according to the Beaufort scale?
Contaminant Source (7question)	14	What is the location of the pollutant emission source in terms of the proximity of the worker to the emission source?
	15	What is the frequency, temporal and spatial pattern (dominant) of pollutant emission from the main pollutant sources?
	16	Based on external evidence (smell, turbidity and darkness, etc.), what is the rate of pollutant emission?
	17	What is the direction and relative position of the operator in relation to the pollutant emission source and the potential flows of the working environment?
	18	What is the shape of the main pollutant sources and how they are distributed?
	19	What is the number of sources, workstations or similar processes producing the pollutant under investigation?
	20	What is the height of the source of pollutant leakage in relation to the worker’s breathing zone, considering the density and the potential of the pollutant to float?
Process (18question)	21	What is the main production process and pollutant generator?
	22	According to the answer to the previous question, what is the type of production process and the source of pollution?
	23	According to the previous question, what are the specific conditions of the process (affecting pollutant production)?
	24	What is the working phase or life cycle of the process (facilities)?
	25	What is the origin, the reason or the main deviations that cause pollutant emission and exposure?
	26	What is the working method of the process or the assigned production task?
	27	What is the level of automation (control, feeding or exploitation technique) in the process and assigned task?
	28	What is the average working age of machines, equipment or facilities involved in pollutant emission?
	29	What is the state of changes in the settings of the operational parameters of the process, including temperature, pressure, humidity, duration of the process, flow and speed of the equipment?
	30	How is the complexity of the process and equipment affecting pollutant release and exposure?
	31	What is the technology used based on output efficiency?
	32	What is the status and position of the technology used based on the life cycle?
	33	What is the level of energy required for work or the level of physical activity?
	34	What is the amount of care, attention and involvement required in the activity under consideration?
	35	What is the situation of overload in the operation of the process (over the capacity of the process and control systems)?
	36	The quality of changing, replacing and replacing process components and equipment during repairs in such a way as to increase pollutant emissions?
	37	What is the state of error, defect and interruption in the performance of the department or the entire equipment and process that leads to an increase in exposure?
	38	What is the condition of the evacuation equipment, the exit routes of the evacuation and industrial air conditioning, and the pressure breaker and control systems of the industrial processes?
Contaminant (19question)	39	What is the type of chemical being investigated?
	40	What is the known functional base of the material?
	41	What is the physical state of the substance or pollutant?
	42	What is the role of matter in production activity?
	43	What is the average amount of material used in the process per hour?
	44	What is the estimated amount of pollutant emitted per hour based on evidence, previous measurements, or similar processes?
	45	In which class is the toxicity of the pollutant?
	46	What is the weight fraction or purity percentage of the desired compound or pollutant?
	47	What is the nature of the substance used or the spread pollutant in terms of the presence of impurities and how it is combined in the composition?
	48	What is the compatibility of the type of permitted limits provided for the composition with the pattern of encountering it?
	49	What are the protective features of the pollutant (including the sense of smell, taste, burning in the nose, throat and lungs, and tearing, dirtiness and deposition of the pollutant on the person, and the Tyndall phenomenon causes the worker to avoid exposure to the pollutant)?
	50	What is the ability to vaporize (evaporation) or the amount of vapor pressure at ambient temperature (open process)?
	51	What is the ability of particulate pollutants to be airborne?
	52	What is the amount of moisture content of the material (particulate pollutant), their adhesion and resuspension?
	53	What is the reactivity of the substance with water, air, and other chemicals for the pollutant under investigation?
	54	What is the dust generation potential of the material under operation?
	55	Are diseases related to pollution present and recorded in the medical records of the individual or colleagues with similar activities?
	56	What is the viscosity of the material?
	57	What is the density of material?
Operator (28question)	58	In which age group is the operator?
	59	What is the operator’s experience and work record, including similar jobs?
	60	What is the operator’s level of education?
	61	How is the operator’s financial situation?
	62	What is the operator’s physical condition?
	63	What is the type of employment and employment of the operator (employment relationship)?
	64	What is the operator’s level of skill, knowledge and understanding of methods, techniques and work steps?
	65	What is the status of in-service training for workers regarding exposure to airborne chemical pollutants in the workplace?
	66	What is the person’s knowledge and information about exposure to airborne chemical pollutants in their environment?
	67	What is a person’s job class or job group/work department?
	68	What is the operator’s metabolism or physical activity level?
	69	What is the diversity of individual tasks and the diversity of the operator’s workplace?
	70	At what level is the operator’s ability to work continuously and with only moderate fatigue?
	71	What is the opinion of the supervisors or managers regarding the abilities, skills, personal characteristics and personal competence of the operator to avoid exposure?
	72	How much effort does the operator have to do to minimize exposure while doing the job following the rules, instructions and doing the job safely (by doing the job right)?
	73	How is a person’s mental ability, alertness and cleverness to understand safe work instructions, to face different (new) conditions and solve challenging conditions or difficult situations?
	74	Can you trust an operator who does not expose himself in any way and requires minimal supervision?
	75	What is the quantity and quality of work that the operator does in a working day?
	76	What is the talent, creativity and imagination of the operator to have new ideas and initiative in doing things in a way that is associated with less exposure and emissions?
	77	What are the personality, characteristics and behavioral characteristics of a person to stop working if he feels worried about his exposure, and continue working by making the situation safe?
	78	How is the operator’s ability to focus on work, understand working conditions, and make decisions and take action in a timely, appropriate manner?
	79	Is the operator enthusiastic about his work and does the organization value him, does he pay a fair salary for his work, and does he actively participate in reducing his exposure due to his loyalty to the organization?
	80	How is the operator’s adaptability to organizational changes and planning, organizing and managing assigned tasks to have less exposure?
	81	Does the operator have an active practice in workplace discipline, personal hygiene and health maintenance?
	82	How is the operator’s ability to withstand pressure and the right to relax in the work environment?
	83	How is the operator’s ability to manage time and act independently within the limits defined for him?
	84	How is the operator’s loyalty to come and attend the daily work and adapt to the working hours?
	85	What is the effectiveness and ability of a person to control assigned tasks and work with minimal exposure?
Technical-engineering (33question)	86	What quality has the replacement of the pollutant been done to reduce the exposure?
	87	What quality has been done to reduce exposure through changing or changing the process, equipment and facilities?
	88	What quality has been done to reduce the exposure by replacing or changing the type and amount of energy consumption of the pollutant emission agent?
	89	What is the type of energy used in the production process (pollutant)?
	90	What is the condition of the physical isolation of the equipment or the pollutant emission source?
	91	What is the status of operator confinement?
	92	For activities where the operator and pollutant source (process) are forced into an enclosed area, which of the enclosure ventilation methods is used to control exposure?
	93	How is the separation of the pollutant emission source?
	94	What is the type of local suction ventilation system used?
	95	What is the overall quality of the local suction ventilation system? (It is determined according to the following 3 questions)
	96	How is the configuration and general structure of the channeling, the presence of specific and appropriate hoods for each process and efficient and refined design compared to the standards and recommended principles of design?
	97	Considering and controlling the consequences of changes in the ventilation system and performing the process of monitoring, maintenance and management of the system compared to the manufacturer’s recommendations, with what quality?
	98	How is the quality of preventing the pollutant from returning to the work environment according to the chimney and its outlet?
	99	Is any temporary activity with pollutant emission carried out in different places under the portable local suction ventilation system?
	100	What is the type of dilution or general mechanical ventilation used?
	101	How is the general quality of dilution ventilation? (according to the following questions)
	102	What is the ratio of the compound limit to the proposed limit for a pollutant with low toxicity (preferably >100 ppm)?
	103	For what pollutant group (proportion of dilution ventilation with pollutant) has it been used?
	104	Are pollutants required to be recycled?
	105	What is the position of the air entering and leaving the working environment in order to avoid dead space and the airflow not passing through the worker’s breathing area due to the arrangement of the equipment?
	106	Is it possible for polluted air to enter from nearby buildings or re-enter the pollutant from windows and doors?
	107	The consequences of changes in the general ventilation system, monitoring and control of the efficiency and effectiveness of general ventilation has been done with what quality?
	108	What is the number of room air changes per hour? 1- (room volume divided by the total flow rate of fans m3/min) and 2- (60 minutes divided by the time obtained from equation 1)
	109	What type of natural ventilation is used in the place of the process and workshop?
	110	How is the general quality of natural ventilation, including the appropriateness of the type of ventilation used according to the environmental and structural conditions of the building, and compliance with standards? (according to the following 6 questions)
	111	What is the appropriateness of environmental factors (humidity, ambient temperature, temperature difference, wind flow and geographic location and upstream industries and workshops) for the implementation of natural ventilation?
	112	What is the appropriateness of the nature and main structure of the structure (design and map of the workplace hall) for the implementation of natural ventilation (including the height and installation location of entrances and exits and natural ventilation equipment)?
	113	How much are the pollutants in the work environment (according to Golbad) affected by the pollutants of neighboring workshops or neighboring workshops or the pollutants of the outside environment?
	114	What is the ratio of the width of the workshop to the lateral height of the workshop and the area of open windows to the area of the workshop in one-way windows?
	115	What is the ratio of the width of the workshop to the lateral height of the workshop and the ratio of the total area of the smallest open window on one side of the workshop to the area of the workshop in the opposite windows (two sides)?
	116	What is the height difference between the inlet and outlet and the smallest area of the inlet or outlet compared to the floor to create air blowing with the help of the flue process (vents) naturally?
	117	What quality of natural solar induction ventilation is used?
	118	Which of the techniques to reduce, prevent the emission and suppression of the pollutant is applied?
	119	What is the quality of pollutant reduction, emission prevention and suppression techniques?
	120	Which of the protective layers of control and management of the process, equipment, pollutant emission producing lines have been used? (In exploratory factor analysis transferred to the process)
	121	What is the quality of the protective layers of the control and management of the process, equipment, lines used to generate pollutant emissions? (In exploratory factor analysis transferred to the process)
Organizational management (26question)	122	What is the type of ownership of the company or workshop?
	123	How is the company’s financial situation?
	124	How many employees does the company have?
	125	Which of the following methods is used to clean the work environment?
	126	What is the condition of the workshop and the process in terms of creating pollution and dirt?
	127	What is the number of repetitions and the cleaning period of the environment?
	128	What is the general state of cleanliness and cleanliness of the workplace, to prevent re-dispersion of the pollutant?
	129	How is the quality of work methods and job content to reduce exposure?
	130	How is the quality of the implementation of preventive maintenance measures?
	131	What is the condition of the workshop and the process in terms of sensitivity and the need to implement preventive maintenance standards?
	132	What is the method of implementation and schedule of implementation of preventive maintenance criteria?
	133	What is the frequency of the number of exposure days per working day per year?
	134	What is the duration of exposure to the duration of shift work (standard)?
	135	How is the quality of implementation of operational maintenance measures?
	136	What is the status of the process in terms of sensitivity and the need to implement operational maintenance repair criteria?
	137	What is the method of implementation and schedule of implementation of preventive maintenance criteria?
	138	How is the quality of the implementation of the set of measures related to cleaning, discipline and workshop in (implementing the S5 system) to reduce exposure in the workplace?
	139	Which of the management controls to reduce exposure time has been used to reduce exposure?
	140	How is the quality of management controls to reduce exposure time in order to reduce exposure?
	141	How is the quality of the implementation of the measures of the notification system and notification of the pollution status and reduction of exposure in the workplace?
	142	What is the state of government requirements?
	143	What is the status of documentation and quality assurance?
	144	What is the status of macro-management strategies to reduce exposure in the workplace?
	145	What is the status of specific occupational health management measures to reduce exposure?
	146	Which respiratory protective equipment does the operator use?
	147	What is the status of necessary measures regarding personal protective equipment?

During the CVR/CVI measurement stages, a total of 32 questions were eliminated from the 147-question questionnaire, reducing it to 115 questions. In the next step, the average CVI for all questionnaire items was calculated. Supplementary 3 displays the scores for CVR and CVI for the 147 questions, along with the acceptance or rejection results for each question. The average validity index for the questionnaire assessing factors influencing exposure to airborne chemical pollutants was 0.85. The final instrument included 115 questions, distributed across various dimensions: 11 in the workshop environment dimension, 15 in the process dimension, 6 in the pollutant source dimension, 10 in the contaminant dimension, 22 in the operator dimension, 31 in the engineering technical measures dimension, and 20 in the managerial and organizational action dimension. Some questions were removed at the beginning of the questionnaire as they pertained to contextual and demographic information.

SUPPLEMENT 3.

Questionnaire standardization scores and factor loading of remaining questions after varimax rotation

Number of question	CVR	CVI				Result	factor loading
		Simplicity	related	Clarity	Total
1	1	0.86	1	0.93	0.93	*	0.792
2	0.71	0.79	0.86	0.79	0.81	*	0.664
3	1	0.93	1	0.93	0.95	*	0.693
4	1	0.79	1	0.86	0.88	*	0.664
5	0.43	0.79	0.71	0.71	0.74	eliminated
6	0.86	0.71	0.93	0.79	0.81	*	0.736
7	0.29	0.71	0.64	0.79	0.71	eliminated
8	0.57	0.79	0.86	0.64	0.71	*	eliminated
9	0.71	0.71	0.86	0.64	0.76	*	0.672
10	0.71	0.93	0.86	0.86	0.88	*	0.575
11	0.57	0.79	0.79	0.79	0.79	*	0.481
12	0.57	0.86	0.86	0.93	0.88	*	0.517
13	1	0.86	1	0.86	0.9	*	0.597
14	1	0.79	1	0.79	0.86	*	-0.674
15	1	0.71	1	0.79	0.83	*	0.547
16	0.43	0.57	0.93	0.5	0.67	eliminated
17	1	0.79	1	0.79	0.86	*	0.437
18	0.86	0.71	1	0.71	0.81	*	0.646
19	1	0.71	1	0.71	0.81	*	0.623
20	0.57	0.71	0.86	0.71	0.76	*	eliminated
21	1	0.86	1	0.64	0.83	*	0.825
22	1	0.71	1	0.64	0.79	*	0.796
23	1	0.64	1	0.64	0.76	*	0.721
24	0.29	0.71	0.57	0.79	0.69	eliminated
25	0.86	0.43	0.93	0.5	0.62	eliminated
26	1	0.71	0.93	0.86	0.83	*	0.583
27	1	0.86	1	0.86	0.9	*	0.772
28	1	0.71	0.93	0.71	0.79	*	0.547
29	0.57	0.71	0.86	0.71	0.76	*	0.472
30	0.57	0.64	0.86	0.64	0.71	*	0.604
31	0.57	0.64	0.93	0.71	0.76	*	0.643
32	0.57	0.57	0.86	0.71	0.71	*	0.441
33	0.29	0.93	0.57	0.93	0.81	eliminated
34	0.71	0.79	0.93	0.86	0.86	*	-0.670
35	0.57	0.71	1	0.64	0.79	*	0.481
36	0.57	0.79	0.93	0.86	0.86	*	0.634
37	0.71	0.79	0.93	0.86	0.86	*	0.593
38	0.71	0.79	0.93	0.86	0.86	*	eliminated
39	0.29	0.71	0.64	0.71	0.69	eliminated
40	0.14	0.64	0.71	0.71	0.62	eliminated
41	1	0.93	1	0.93	0.89	*	-0.763
42	0.29	0.93	0.71	0.93	0.83	eliminated
43	1	0.86	1	0.79	0.88	*	0.533
44	0.71	0.57	1	0.43	0.67	eliminated
45	0.14	0.64	0.57	0.79	0.67	eliminated
46	0.71	0.57	0.93	0.79	0.76	*	0.656
47	0.71	0.57	0.93	0.79	0.76	*	0.646
48	0.43	0.57	0.79	0.79	0.71	eliminated
49	0.57	0.86	0.71	0.86	0.81	*	0.538
50	1	0.71	1	0.86	0.86	*	0.778
51	1	0.93	1	0.93	0.95	*	0.796
52	0.57	0.64	0.93	0.64	0.74	*	0.433
53	0.29	0.57	0.79	0.5	0.62	eliminated
54	0.57	0.79	0.86	0.71	0.79	*	eliminated
55	0.43	0.36	0.71	0.36	0.48	eliminated
56	0.14	0.79	0.57	0.79	0.71	eliminated
57	0.14	0.79	0.57	0.79	0.71	eliminated
58	0.57	1	0.79	1	0.93	*	0.663
59	0.57	1	0.79	1	0.93	*	0.655
60	0.57	1	0.79	1	0.93	*	0.644
61	0.57	0.79	0.93	0.86	0.86	*	0.544
62	0.57	0.79	0.86	0.79	0.83	*	0.494
63	0.57	1	0.79	1	0.93	*	-0.723
64	0.86	0.79	0.93	0.86	0.86	*	0.586
65	1	0.86	1	0.86	0.9	*	0.785
66	0.86	0.71	1	0.79	0.83	*	0.716
67	0.57	0.86	0.93	0.86	0.88	*	0.481
68	0.29	0.86	0.71	0.86	0.81	eliminated
69	0.57	0.71	1	0.64	0.79	*	eliminated
70	0.29	0.71	0.64	0.71	0.67	eliminated
71	0.71	0.71	0.86	0.86	0.81	*	0.535
72	0.71	0.71	0.86	0.86	0.81	*	0.577
73	0.57	0.71	0.86	0.71	0.76	*	0.471
74	0.86	0.71	0.86	0.86	0.81	*	0.583
75	0.86	0.71	0.86	0.86	0.81	*	0.631
76	0.57	0.71	0.79	0.71	0.73	*	0.465
77	0.86	0.71	0.93	0.86	0.83	*	0.615
78	0.71	0.71	0.86	0.86	0.81	*	0.497
79	0.86	0.71	0.93	0.86	0.83	*	0.566
80	0.71	0.71	0.86	0.86	0.81	*	0.503
81	0.57	0.86	0.86	0.86	0.86	*	0.529
82	0.29	0.64	0.79	0.86	0.76	eliminated
83	0.14	0.64	0.64	0.64	0.64	eliminated
84	0.43	0.71	0.71	0.86	0.76	eliminated
85	0.43	0.71	0.71	0.86	0.76	eliminated
86	1	0.79	1	1	0.93	*	eliminated
87	1	0.71	1	0.86	0.86	*	eliminated
88	0.86	0.79	1	0.79	0.86	*	eliminated
89	0.43	0.86	0.79	0.79	0.81	eliminated
90	1	0.79	1	0.79	0.86	*	-0.545
91	1	0.86	1	0.86	0.9	*	-0.617
92	1	0.64	1	0.71	0.79	*	-0.519
93	1	0.71	1	0.71	0.81	*	-0.553
94	1	0.79	1	0.79	0.86	*	-0.685
95	1	0.71	1	0.79	0.83	*	-0.821
96	0.57	0.79	0.79	0.86	0.81	*	0.573
97	0.57	0.86	0.86	0.79	0.83	*	0.512
98	0.57	0.71	0.93	0.86	0.83	*	0.449
99	0.29	0.71	0.79	0.79	0.76	eliminated
100	1	0.79	1	0.79	0.86	*	-0.669
101	1	0.71	1	0.79	0.83	*	-0.627
102	0.86	0.64	0.93	0.79	0.79	*	0.461
103	0.57	0.93	0.86	0.93	0.9	*	0.437
104	0.14	0.86	0.5	0.86	0.74	eliminated
105	0.71	0.79	1	0.79	0.86	*	0.535
106	0.43	0.71	0.79	0.79	0.76	eliminated
107	0.57	0.86	0.86	0.79	0.83	*	0.478
108	1	0.71	1	0.79	0.83	*	0.625
109	1	0.71	1	0.79	0.83	*	-0.543
110	1	0.79	1	0.79	0.86	*	-0.605
111	0.57	0.79	1	0.79	0.86	*	0.547
112	0.57	0.79	1	0.79	0.86	*	0.573
113	0.43	0.79	1	0.79	0.86	eliminated
114	0.57	0.79	0.86	0.79	0.81	*	0.538
115	0.57	0.79	0.86	0.79	0.81	*	-0.479
116	0.57	0.79	0.86	0.79	0.81	*	0.436
117	0.57	0.79	0.86	0.79	0.81	*	eliminated
118	1	0.79	1	0.79	0.86	*	-0.583
119	1	0.79	1	79/0	0.86		-0.492
120	0.86	0.64	1	71/0	0.79	*	-0.568
121	0.86	0.64	1	71/0	0.79	*	0.658
122	0.57	1	0.79	0.93	0.9	*	0.749
123	0.71	0.71	0.86	0.86	0.81	*	0.763
124	0.57	0.79	0.79	0.93	0.83	*	0.726
125	0.57	0.93	1	0.86	0.93	*	eliminated
126	0.29	0.71	1	0.5	0.74	eliminated
127	0.29	0.93	0.86	0.93	0.9	eliminated
128	0.57	0.71	0.86	0.71	0.76	*	0.424
129	1	0.93	1	0.93	0.95	*	0.535
130	0.86	0.71	0.86	0.64	0.74	*	-0.684
131	0.43	0.86	0.79	0.79	0.81	eliminated
132	0.43	0.86	0.79	0.79	0.81	eliminated
133	1	0.79	1	0.93	0.9	*	0.483
134	1	0.79	1	0.93	0.9	*	0.454
135	1	0.71	0.86	0.71	0.76	*	-0.777
136	0.43	0.86	0.79	0.79	0.81	eliminated
137	0.43	0.86	0.79	0.79	0.81	eliminated
138	1	0.71	1	0.79	0.83	*	0.665
139	0.86	0.71	0.93	0.71	0.81	*	0.565
140	0.86	0.64	0.93	0.71	0.79	*	0.578
141	0.71	0.71	1	0.79	0.83	*	0.518
142	1	0.71	1	0.79	0.83	*	0.546
143	0.86	0.79	0.93	0.71	0.81	*	0.564
144	1	0.79	1	0.71	0.83	*	0.804
145	0.71	0.79	1	0.79	0.86	*	0.582
146	1	0.93	1	0.93	0.95	*	0.526
147	1	0.79	1	0.79	0.86	*	0.488

Construct validity—exploratory factor analysis and determination of factor loadings

To further standardize the instrument, exploratory factor analysis was performed to investigate its construct validity. This analysis utilized questionnaires collected from 523 polluting activities. The exploratory factor analysis, which was conducted by determining the main components, yielded four key outputs.

The first output consisted of the KMO test report, and the second was Bartlett’s test of sphericity, both shown in Table 4. The KMO test result for the designed questionnaire was 0.805, indicating good sampling adequacy. Bartlett’s test showed significant results (P < 0.001), confirming that the observed correlation matrix was appropriate for exploratory factor analysis.

Table 4.

KMO’s test results and Bartlett’s test of sphericity

Test	Indicator	Value
KMO Test	KMO sampling adequacy index	0.805
Bartlett’s test of sphericity	Chi-square estimation	21547/6
	Degrees of freedom	3128
	P	>0.001

KMO=Kaiser–Meyer–Olkin test

Table 5 shows the third output, which includes the eigenvalues of the extracted factors after varimax rotation. According to this output, seven main dimensions with eigenvalues greater than 2.3 were identified (Engineering controls, Process characteristics, Management/Organization, Worker characteristics, Contaminant characteristics, Workstation characteristics and Source characteristics). The eigenvalue of the first dimension was 9.548, explaining 12.961% of the variance, while the cumulative variance explained by all seven dimensions was 58.004%.

Table 5.

Total variance and questions dimensions

Dimensions	1RSL
	Total	Variance %	Cumulative %	Number of questions	Factor loading range
Engineering controls	9.548	12.961	12.961	25	0.436–0.821
Process characteristics	7.168	10.387	23.348	16	0.441–0.825
Management/organization	6.314	9.479	23.845	19	0.424–0.804
Worker characteristics	5.053	8.517	41.362	21	0.465–0.785
Contaminant characteristics	4.873	6.986	48.348	9	0.433–0.796
Workstation characteristics	4271	5.771	54.119	10	0.481–0.792
Source characteristics	2.428	3.885	58.004	5	0.437–0.674

¹Rotation sums of loadings

After calculating the correlation matrix between the variables, factors were extracted using principal component analysis with varimax rotation on the 115 items, applying a cutoff point of 0.4. The results revealed that the items were grouped into seven factors with eigenvalues exceeding 2.3. Items with a factor loading below 0.4 were removed, resulting in a reduction of the questionnaire from 115 to 105 questions. The fourth output of the exploratory factor analysis included the factor loadings of each remaining question after rotation.

Phase III: Reliability

To assess the stability of the instrument, the test–retest method was utilized with a two-week interval and a sample size of 15 participants. The ICC was calculated to analyze the results. Table 6 presents the results of internal consistency reliability (Cronbach’s alpha) and test–retest reliability analyses. According to the results, the ICC was 0.777, with the contaminant dimension showing the highest reliability at 0.86 and the source dimension recording the lowest value at 0.72.

Table 6.

Internal consistency reliability (Cronbach’s alpha) and test–retest reliability results

Dimensions	Number of questions	Cronbach’s alpha (n=708)	ICC (n=15)
Engineering	25	0.81	0.81
Management/organization	17	0.82	0.78
Worker	21	0.79	0.78
Process	15	0.66	0.73
Contaminant	8	0.84	0.86
Workstation	9	0.73	0.76
Source	5	0.58	0.72
Total	100	0.747	0.77

ICC=Interclass correlation coefficient

Table 7 provides a summary of all activities carried out in this study. The research was designed in three main phases, and the actions undertaken in each phase, along with the number of deleted, added, and revised items, are shown in Table 7.

Table 7.

A summary of all questions removed, added, and corrected at each stage

Stages (phases)		Total input questions to stage	Removed questions	Added questions	Corrected questions	Total questions left at end	Output results
Phase (I)	Literature review	255	48	0	0	177	Initial questionnaire
	Industry and health centers experts	177	19	0	34	158	Correcting ambiguous and difficult expressions
	Occupational health Experts	158	13	2	24	147
Phase (II)	Qualitative	147	32	0	19	115	Grammatical correction
	Quantitative (CVR)						Eliminate question (below 0.51)
	Quantitative (CVI)						Eliminate question (below 0.79)
	Construct validity	115	10	0	0	105	0.4 for varimax rotation
Phase (III)	Internal consistency (Cronbach’s alpha)	105	0	0	0	105	Total alpha coefficient 0.74
	Test retest						ICC=0.77

CVI=Content validity index; CVR=Content validity ratio; ICC=Interclass correlation coefficient

DISCUSSION

This study introduces a questionnaire designed to screen for airborne chemical pollutant exposure in industrial settings. The questionnaire comprehensively identifies and tracks objective and subjective factors influencing exposure, with a particular emphasis on the effectiveness of implemented control measures. Its applications include assessing overall workplace pollutant exposure levels, evaluating the quality of influential factors, analyzing the impact of changes on individual exposure over time, and determining the necessity of costly pollutant measurements to optimize resource allocation. This cost–benefit analysis enables year-on-year comparisons of employee exposure levels, facilitating informed decisions regarding the prioritization of workplaces and tasks for detailed pollutant measurement.

The questionnaire’s development leveraged both a thorough literature review and extensive interviews with subject matter experts. This represents a pioneering effort in designing and standardizing a questionnaire that assesses airborne chemical pollutant exposure by examining its contributing determinants.

The process of classifying contributing factors, integrating findings from literature reviews, expert interviews, and validity/reliability assessments within the qualitative study, revealed some discrepancies.

Specifically, while expert interviews did not identify items classifiable as dimensions of career records and associated criteria, the literature review, along with Hibbs and Moore’s research[44] on colleague illness rates and the monitoring, evaluation, and periodic examination of exposure and individual physical condition, highlighted these factors. These considerations were incorporated into the questionnaire as questions focusing on pollutant health effects.

Numerous studies have identified workplace atmospheric conditions—barometric pressure, humidity, temperature, and airflow—as significant dimensions.[33,45,46] These factors, along with geographical location, were categorized as elements of geographical weather within the broader dimension of workshop characteristics.

The initial questionnaire, entering the qualitative formal validity phase, comprised 177 questions. This extensive question set incorporated all data extracted from library research, expert interviews, and workplace observations. Unlike conventional questionnaires, this instrument employs a nonstandard Likert scale. Instead of a simple Likert scale, it incorporates a combination of Likert scales and checklists, adapting to the diverse nature of influencing factors and observable environmental characteristics. The questionnaire’s intended users are experts and evaluators with expertise in airborne chemical pollutant exposure. Further distinguishing it from typical questionnaires, the number of answer choices on the Likert scale varies according to the specific question.

The questionnaire designed to investigate factors influencing airborne chemical pollutant exposure achieved an average validity index of 0.85, exceeding the 0.79 threshold for acceptability. Following CVR/CVI measurement, 32 questions were removed, reducing the total number of items from 147 to 115. The majority of deleted questions pertained to the pollutant profile dimension, while the fewest deletions were within the workplace profile dimension.

The KMO value calculated was 0.805. This value indicates adequate sample size for factor analysis, with values closer to one signifying better sampling adequacy.[41] Thus, the data were deemed suitable for factor analysis.

Bartlett’s test of sphericity yielded a P value < 0.001. This result confirms the existence of correlations between the variables, suggesting the presence of discernible relationships within the data.

The observed significant differences between the correlation coefficients of the questionnaire items indicate their suitability for integration into a structured framework.[42] Failure to reject the null hypothesis of no correlation between variables would have undermined the validity of applying factor analysis.

During construct validity testing (exploratory factor analysis), certain questions were excluded. Questions pertaining to natural solar induction ventilation were removed due to its limited prevalence and nonusage in the studied Iranian industries, leading to uniform responses across all participants and hindering analysis.

Similarly, three questions focusing on the quality of control measures—namely, those related to replacing or modifying polluting substances, processes, equipment, and facilities, and those targeting exposure reduction through changes in energy consumption of pollutant emission agents—were omitted because they impeded the analytical process.

Furthermore, a question concerning the condition of air evacuation equipment, industrial air conditioner exit routes, pressure breakers, and industrial process control systems was removed due to insufficient applicability across the studied job types and processes. However, it was retained within the main questionnaire for use in relevant process industries.

The engineering/technical dimension, encompassing 25 questions (including 13 subquestions evaluating the quality of local, dilution, and natural ventilation), constituted the largest section of the questionnaire. The operator dimension followed, with 21 questions (10 subquestions focusing on operator abilities and characteristics). The design allows for flexibility; the 23 subquestions related to ventilation quality and operator details can be omitted if detailed assessments of these aspects are deemed unnecessary.

Several researchers, including Shin et al.,[47] Gorce et al.,[48] and Zhang et al.,[49] have developed exposure assessment screening instruments. However, these instruments exhibit high uncertainty, requiring specialized equipment and detailed toxicological data while overlooking practical environmental conditions. These methods often examine exposure but fall short in identifying the root causes and sources of pollutant release. Consequently, their findings provide limited assistance in pinpointing influential factors and developing effective control strategies.[35,50,51,52,53]

Existing risk assessment methods, comprising hazard identification, exposure assessment, toxicity assessment, and risk characterization,[54] rely on data and measurements of harmful factors to quantify exposure. Parameters, such as volatility/vapor pressure, consumption amount (mass/volume), exposed substance mass/volume, absorption rate, and body weight, are utilized. Some methods incorporate limited information on control measures in risk evaluation.[35,50,51,52,53] The reliance solely on consumption amounts, pollutant physical properties, and brief toxicity data elevates uncertainty. In contrast, comprehensive exposure assessment necessitates consideration of exposure levels and all objective and subjective factors contributing to uncertainty.[15,55,56,57]

Mik[25] and Pan[26] emphasize the importance of understanding effective factors in occupational exposure and their interrelationships for a given work setting. This knowledge aids in identifying influential factors and predicting potential health consequences. A simple preventive solution, like a checklist,[27] is valuable for visualizing relationships, conditions, and evidence of pollutant exposure. Therefore, this checklist questionnaire was designed and validated by systematically compiling factors impacting exposure.

Friesen et al.[58] highlighted that exposure assessment can be conducted through various methods, including questionnaires which regularly and systematically collect information on workplaces and assigned tasks, combined with expert judgment to reduce misclassifications. In line with this, the questionnaire developed in this study facilitates structured data collection related to workplace processes, technical engineering, and management measures.

According to Nieuwenhuijsen,[59] questionnaires are indispensable instruments in exposure assessment for occupational and environmental epidemiological studies. Developing questionnaires is a complex process, requiring significant effort for validation. The present study’s inclusion of all information on influential factors, along with question alternatives on the Likert scale, contributes to reconstruction of exposure history in epidemiological studies, providing comprehensive exposure details.

Rice et al.[60] underscored the importance of questionnaire design, stating that properly formulated questionnaires can identify problems and enhance data quality for improved job evaluations. Their research demonstrates the value of questionnaires in surveys for accurate estimation, yet further research is needed to develop instruments for assessing workplace damage levels. This study aligns with Rice et al.’s work, undergoing rigorous validation procedures.

Anderson and Gerbing[61] note that while structural equation modeling allows for causal inferences, caution is warranted. Neither the study’s design nor the use of structural equation modeling establishes causality.

CONCLUSION

This study developed and validated a comprehensive questionnaire designed to assess both objective and nonobjective factors influencing airborne chemical exposure. Utilizing a Likert scale, the tool offers a systematic and optional method for detailed data collection. It facilitates the classification, screening, and monitoring of exposure types and changes, while also suggesting the need for annual measurements, thereby aiding in exposure history reconstruction for epidemiological research. However, the questionnaire’s length resulted in partial completion by some experts during various study phases, indicating a need for revalidation of the final version.

Ethics approval

Ethical approval was obtained from the Medical Ethics Committee of Shahid Beheshti University of Medical Sciences in Tehran, Iran (ethics code IR.SBMU.PHNS.REC.1396.115).

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.