Influence of occupational exposure on hyperuricemia in steelworkers: a nested case–control study

The participants in this study were workers of a steel company who underwent occupational health screening at Hongci Hospital, Tangshan City, Hebei Province, China, from March 2017 to September 2017. Cohort inclusion criteria: formally employed workers, at least one year of service. Cohort exclusion criteria: hyperuricemia and refusal to participate in this cohort study. A total of 4,247 workers from the steel company were included in the cohort. A follow-up survey of a total of 3,706 steelworkers between March 2019 and September 2019 was carried out, with a 12.74% missing rate. Workers in the cohort with new-onset hyperuricemia in steel companies are regarded as the case group. From this cohort population, each new-onset patient was matched with a worker in the steel company without hyperuricemia as a control according to the principle of matched design (the same sex and age) to form a control group. A total of 641 case and control pairs were eventually included as samples after those with incomplete information were excluded.

All participants have read and signed the informed consent form. The subject was approved by the Medical Ethics Committee of the North China University of Technology (No.15006).

A one-on-one survey was conducted by uniformly trained enumerators with a questionnaire designed by the subject group. The contents of the questionnaire included sex, age, household size, monthly household income, education level, marital status, smoking status, drinking status, tea status, physical exercise, type of work, change in type of work, shift work, type of shift, change in shift work, change in working hours, daily working hours, monthly rest time, and personal disease history. The participants were also asked about the frequency of food intake (0 day per week, 1–2 days per week, 3–6 days per week, and 7 days per week). Food was divided into different groups: vegetables, fruits, meat, eggs, dairy products, soy products, and seafood.

The participants took off their shoes when their height and weight were measured through the ultrasound height and weight measuring device. Measurements were taken three times and averaged. The body mass index (BMI) was calculated based on the measurement. The participant was instructed not to drink tea, coffee, alcohol or other beverages that may affect the results of blood pressure. The participant was asked to have a five-minute break before the blood pressure measurement and to take the measurement three times with an interval of not less than three minutes.

Subjects were required to fast for 12 h. Fasting blood and morning urine were collected by the Laboratory Department of Tangshan Hongci Hospital before 9 a.m. the next day. Blood, urine, and blood biochemistry were examined by specialist physicians.

The on-site hygienic survey involved heat, dust, and noise.

The measurement tool for temperature is Wet Bulb Black Globe Temperature Gauge (WBGT). According to relevant standard [20], the temperature should be measured during the hottest season of the year. Temperatures were measured at different workplaces in consideration of the specific conditions of the steel production unit. Three to six measurement points were selected for each workplace, and the test was repeated three times at each measurement point, with the average taken as the final result.

Dust was measured with a dust sampler. The sampling points were chosen according to relevant standard [21] and specific conditions of the workshop. The sample collection time for each sampling point was 45 min, and the flow rate of the dust sampler was set at 40 L/min. The calculation formula is as follows:

where: C- dust concentration, mg/m.³

m2—the mass of the filter membrane after sampling, mg.

m1—the mass of the filter membrane before sampling, mg.

Q—flow rate, L/min.

t—sampling time, min.

Noise testing was carried out according to relevant standards [22, 23] and specific circumstances of the workplace. When the noise distribution in the workshop was uneven, the noise was divided into different sound zones according to the sound level and two test points were set up in each zone. When the noise distribution in the workshop was relatively even (the sound level difference is less than or equal to 3 dB (A)), three measurement points were set up. The average value was taken as the final result after measurement. The calculation formula for sound level measurement is as follows:

where: \({L}_{Aeq, {T}_{i}}\)- equivalent sound level during the time period T_i.

\({L}_{Aeq, T}\)—equivalent sound level for a full day.

n—the total number of periods.

T—the duration of each period.

T_i- i period of time.

Cumulative exposure measurement (CEM) for steelworkers is calculated based on the results of the on-site hygienic survey, combined with the change in work status and duration of occupational exposure. The formula is as follows:

where: L_n is the average exposure to the target harmful factor over a period of time T_n.

Those having a blood uric acid value greater than or equal to 7.0 mg/dL in men and 6.0 mg/dL in women, as well as previous or ongoing gout treatment, were diagnosed with hyperuricemia [24]. Never smokers were defined as those who had never smoked from birth to the time of the survey. Former smokers were defined as those who had previously smoked but had quit smoking for 6 months or longer as of this survey. People who had smoked at least 1 cigarette per day for six months or longer as of the survey were defined as current smokers. People who drank alcohol more than twice a week, regardless of the type of alcohol, and who had done so for more than a year were considered to be drinking. The frequency of food consumption was divided into four categories: never (0 day per week), occasionally (1–2 days per week), frequently (3–6 days per week) and daily (7 days per week). In this study, the International Physical Activity Questionnaire (IPAQ) was used to analyze the physical activities of steelworkers [25]. Physical activities were classified into light, moderate, and heavy activities based on intensity, frequency, and overall weekly physical activity level. Shift work is a system of irregular working hours in which one or more teams perform tasks continuously for 24 h by working in shifts without stopping [26]. The cumulative number of days of night work represents the total number of days of night work done by workers in the steel plant as of the date of the survey. According to relevant standard [20], work with a productive heat source and WBGT ≥ 25 °C was defined as heat-exposed work. Dust exposure was defined based on the type of work, the work environment, and the findings of the site hygiene [21]. Exposure to noise was defined as workers who were exposed to a noisy environment where the 8 h/d or 40 h/week equivalent A-weighted sound pressure level is ≥ 80 dB, which may be harmful to health and hearing [27].

Continuous variables were described by means and standard deviations, and the differences between groups were obtained through Student’s t-test. The categorical variables were expressed through the number of individuals (%), and χ² tests were used for comparisons between groups. Multifactorial analyses were performed and multiplicative interactions between occupational hazard factors were explored with the help of conditional logistic regression models. Additive interactions were assessed using the attributable proportion of interaction (AP), the relative excess risk of interaction (RERI), and the synergy index (SI), calculated using the Excel spreadsheet by Andersson et al. [28]. The AP is the proportion of the risk due to the interaction in the doubly exposed group. When RERI is positive, it indicates increased risk due to the additive interaction. SI can be interpreted as the ratio of an increased risk due to both exposures to the sum of individual increased risks.

A total of 3,706 workers (3,352 males and 354 females) were followed up in the steel enterprise. The follow-up period was (25.88 ± 1.68) months. 641 new cases of hyperuricemia (587 males and 54 females) were found among the 3,706 workers in the steel company, with a mean age of (44.67 ± 7.58) years. The case and control groups have the same sex and age composition because the same sex and age were required and 1:1 matching was used according to the matching principle. The two groups were comparable with the same basic data. The incidence rate was 17.30% (17.51% for males and 15.25% for females). The incidence density was 81.32/1,000 person-years (82.35/1,000 person-years for males and 71.56/1,000 person-years for females).

The results of the analysis are shown in Table 1 (at the end of the manuscript). The analysis revealed that the proportions of junior, senior, and secondary were the highest in both the case and control groups, at 78.5 and 71.1%, respectively. The proportion of people with a junior college or above in the case group was 20.4%, lower than 27.6% in the control group, with a statistically significant difference (P < 0.05). The proportions of workers in the steel enterprise with a per capita monthly income of RMB1,500—and RMB ≥ 2,500 were 46.8 and 32.6%, respectively, higher than those of the control group at 41.0% and 25.6%, respectively, with statistically significant differences (P < 0.001). Regarding the health status of steelworkers, the proportions of overweight, obesity, hypertension, diabetes, dyslipidemia, abnormal kidney function, and abnormal liver function in the case group were all higher than those in the control group, with statistically significant differences (P < 0.05 for all diseases). In terms of behavioral lifestyle, the proportion of workers in the steel company who consumed fruit daily was the highest in both the case (51.6%) and control (59.3%) groups, and was lower in the case group than in the control group, showing a statistically significant difference (P < 0.05). The proportion of workers in the steel company who consumed meat daily was 33.4% in the case group, higher than that in the control group at 24.2%, with a statistically significant difference (P < 0.001). The proportion of workers in the steel company who consumed seafood daily was 11.4% in the case group, compared with 6.4% in the control group, hence the difference was statistically significant (P < 0.05). The proportions of heavy physical activities and physical exercises 5–7 times per week were 75.2 and 14.8% in the case group, respectively, compared with 82.2 and 22.9% in the control group, presenting statistically significant differences (P < 0.001 for both). The proportion of alcohol consumption in the case group was 53.8%, higher than that of 40.1% in the control group, showing a statistically significant difference (P < 0.001). The differences in the distribution of marital status, frequency of vegetable consumption, frequency of egg consumption, frequency of soy product consumption, smoking, and tea-drinking between the case and control groups were not statistically significant (P > 0.05 for all parameters).

A comparison of the distribution of exposure to occupational hazards between the case group and the control group revealed that ever shifts, current shifts, heat exposure, and dust exposure differed significantly between the two groups, but the differences were not statistically significant in the two noise exposure groups. Please refer to Table 2.

In the cumulative exposure, those not exposed to occupational hazards with a cumulative exposure of 0 formed a group. The median among exposures to occupational hazards was obtained. With 0 and the median as bounds, each cumulative exposure was divided into three groups (0,0-median, ≥ median). The median cumulative exposure to night shifts in the shift work population was 1,972.80. The median cumulative exposure to heat in the heat-exposed population was 567.83. The median cumulative exposure to dust in the dust-exposed population was 30.02. The median cumulative exposure to noise in the noise-exposed population was 1,660.68.

After comparing the distribution of cumulative exposure to occupational hazards between the case and control groups, we found that the proportion of the cumulative number of days of night work in the case group was 46.5% for ≥ 1,972.80 day, which was higher than that of the control group (41.2%), presenting a statistically significant difference (P < 0.05). The proportions of cumulative exposure to heat in the case group were 26.1% and 30.7% for 0–567.83(°C/year) and ≥ 567.83(°C/year), respectively, which were higher than those of the control group (24.6% and 20.0%), which shows statistically significant differences (P < 0.001). The difference in the cumulative dust exposure between the two groups was also statistically significant (P < 0.05). The difference in the cumulative noise exposure between the two groups was not statistically significant (P > 0.05). Please see Table 3.

Table 4 shows the analysis results of the effect of occupational hazard exposure on HUA by dint of the Conditional Logistic Regression model. The dependent variable was the presence or absence of hyperuricemia, and the independent variable was the occupational exposure to harmful factors. After the adjustment of possible confounders, the OR (95% CI) for the risk of developing hyperuricemia was 2.18 (1.28–3.69) times and 1.81 (1.11–2.95) times for ever shifts and current shifts, respectively, compared to the reference group. Heat exposure and dust exposure were associated with an elevated risk of developing hyperuricemia with ORs (95% CI) of 1.58 (1.17–2.14) and 1.34(1.01–1.81), respectively. In addition, the effect of noise exposure on the risk of hyperuricemia development in steelworkers was not statistically significant (P > 0.05).

Cumulative exposure to occupational hazards was segmented as above. A Conditional Logistic Regression model was employed with the presence or absence of hyperuricemia as the dependent variable, and the segmented cumulative exposure to occupational hazards as the independent variable. After the adjustment of possible confounders, the results showed an increased risk of developing hyperuricemia in the cumulative days of night work in the 0–1,972.80 and ≥ 1,972.80 (days) compared to the reference group, with ORs (95% CI) of 1.87(1.12–3.13) and 2.02(1.21–3.37), respectively. The ORs (95% CI) for the risk of developing hyperuricemia in steelworkers in the cumulative exposure to heat groups of 0–567.83 and ≥ 567.83 (°C/year) were 1.50(1.02–2.22) and 1.64(1.11–2.43), respectively. Compared to the reference group, the OR (95% CI) for the risk of hyperuricemia in steelworkers was 1.56 (1.05–2.32) for cumulative dust exposure ≥ 30.02 (mg/m³/year). The OR (95% CI) for the risk of developing hyperuricemia in steelworkers at dust exposure of ≥ 30.02 (mg/m³/year) was 1.16(0.78–1.73), which was not statistically significant (P > 0.05) compared to the reference group. The association between cumulative noise exposure and hyperuricemia development in steelworkers was not statistically significant (P > 0.05). Please refer to Table 5.

The multiplicative interactions between shift work, heat exposure, and dust exposure were analyzed in the case of hyperuricemia. The main effects of the two occupational hazards and the product terms of the two multiplicative interactions were jointly introduced into the Conditional Logistic Regression model for analysis. The model analysis was adjusted for service years, marital status, education level, monthly per capita household income, physical activity level, BMI, hypertension, diabetes, dyslipidemia, abnormal kidney function, abnormal liver function, frequency of vegetable consumption, frequency of fruit consumption, frequency of egg consumption, frequency of dairy consumption, frequency of soya product consumption, frequency of seafood consumption, physical exercise, smoking, drinking, and tea drinking. The results showed a multiplicative interaction between dust exposure and heat exposure (P _interaction < 0.05). Dust exposure significantly increased the effect of heat exposure on hyperuricemia development (OR = 2.45, 95% CI: 1.34–4.46). No multiplicative interactions were found between shift work and heat exposure or shift work and dust exposure (P _interactions > 0.05 for all variables). Please see Tables 6, 7 and 8.

Since the additive interaction model can only be employed to explore interactions between dichotomous variables, the ever shifts and current shifts were combined into one group. The shift states were divided into never-shift and shift for the analysis of additive interactions. The results revealed no additive interactions between shift work and heat exposure, shift work and dust exposure, or heat exposure and dust exposure. Please refer to Table 9.