Impact of cold exposure on shift working seafood handlers in Northern Norway: a comparative analysis across work shifts

Abstract

Objective: This study aimed to investigate the impact of occupational thermal exposure on shift workers, specifically whether cold exposure elicits distinct physiological responses and thermoregulatory recovery across different tasks and shift types.

Methods: Observational study at two factories processing prawns in Northern Norway in which 32 shift-working seafood handlers with different task responsibilities were followed for a single shift (morning, evening, night). The participants answered questionnaires regarding thermal exposures at work and related symptoms; these were compared to answers from 12 administration workers. Personal thermal loggers measured the range of temperature exposures associated with four different seafood handler work tasks. Pre- and post-shift plasma levels of FGF21, GDF15 and cytokines were analysed using immunoassays. As a proxy for thermoregulatory response across different shift types, hand temperature was measured repeatedly before and after breaks using a thermal imaging camera.

Results: Most seafood handlers reported subjective impact from cold exposure. Cold working conditions of ≤ 10 ℃ were measured across all shifts and three different seafood handling tasks. The morning shift-seafood handlers displayed lower plasma FGF21 post-shift vs. pre-shift; the evening and night shifts showed no difference. GDF15 levels remained unchanged regardless of shift types but were positively correlated with age. Night shift was associated with increased plasma IL6 post-shift vs. pre-shift. Thermoregulatory responses showed a positive linear relationship with break duration but did not differ between shifts.

Conclusions: The findings suggest that exposure levels are closely linked to specific tasks and shifts, with thermoregulatory responses varying by task type and time of day.

Keywords: Breaks; Cold exposure; Occupational health; Shift work; Thermoregulation.

Fig. 1

The participants'characteristics and their questionnaire responses. A The study was conducted in two seafood factories in Northern Norway with a total of 32 shift-working seafood handlers participating. For the questionnaire, an additional 12 non-shift day-working personnel participated from the same factories. B Cold exposure-related symptoms were assessed using a questionnaire. C Each participant was asked to respond to questions about their perception of the ambient working environment, specifically whether they perceive it as cold or warm. If the participants responded affirmatively to these questions, they were presented with the follow-up questions. Statistical significance was tested by Chi-square test; *P < 0.05. Data is shown as the percentage

Fig. 2

Thermal exposure associated with specific work tasks and shifts among seafood handlers. A A personal logger, carried in a backpack, recorded work environment temperature at a supraclavicular position, while an iButton, taped to the skin, measured brachial skin temperature. B Image shows personal thermal loggers that were used. C Examples of ambient temperature of the working environment (blue line) and brachial skin temperatures (red line) recorded using loggers carried by individual seafood handlers working as thawer, operator, controller, and packer. Shaded time intervals indicate breaks outside the production area. The dotted line indicates the limit of cold working conditions defined as ≤ 10 ℃. D Pooled ambient and skin temperature data of all seafood handlers. E Work environment temperature recorded during morning, evening, and night shifts, skin temperatures of the corresponding shift workers. Mean values are indicated. Statistical significance was tested by one-way ANOVA; ***P < 0.001, ****P < 0.0001

Fig. 3

The effects of shift types on the plasma levels of putative cold exposure biomarkers. A FGF21 and GDF15 plasma levels in seafood handlers were analysed pre-shift (left) and post-shift (right) for different shift types (morning, evening, night). B FGF21 and GDF15 data stratified based on Body Mass Index (BMI) categories: normal weight (< 25 kg/m²), overweight (≥ 25 kg/m² and < 30 kg/m²) and obesity (≥ 30 kg/m²); and C) Age quartiles: Q1 (21–27 yrs), Q2 (29–36 yrs), Q3 (39–52 yrs), Q4 (53–64 yrs). Statistical tests by two-way ANOVA; pre/post sampling (W), age (A) as independent factors; statistical significance is indicated as *P < 0.05; **P < 0.01.; ***P < 0.001. Plots show individual values with Mean ± SEM

Fig. 4

The effects of shift types on the plasma levels of inflammatory biomarkers. Panels show the effects of shift types (morning, evening, and night) on plasma levels pre- and post-shift of the cytokines IL1α, IL6, IL10, TNF, IL12p70. Two-way ANOVA was used with shift types (S) and pre/post sampling (W) as independent factors. Statistical significance is indicated as *P < 0.05. Data is presented as Mean ± SEM

Fig. 5

Change in hand temperature of participants between work bouts, during breaks. The hand temperature of participants was measured using infrared imaging before and after each break during one shift. A Representative images show the hand temperature of an individual after a work bout in the Packing area (upper image) and immediately after a 28-min break, going back to work (lower image). B Hand temperature changes associated with different work tasks; C) shift types; and D) break durations. The median and 5^th percentile temperature pixel values were identified before and after each break. Each dot represents one break-associated temperature change for participants’ left and right hands, as indicated. Data was collected for all breaks those participants had during one shift. Black bars show Mean ± SEM of overall break- related temperature change. Statistical significance was tested by two-way ANOVA with temperature change as the dependent factor; and left/right hand (H), break duration (D) as independent factors (HxD indicates the interaction). Statistical significance is indicated as *P < 0.05

Fig. 6

Hand temperature changes pre-post break in relation to break duration. Plots present regression lines to demonstrate the relationship between break duration and hand 5th percentile temperature change associated with A) different shifts and C) different work tasks. Corresponding B) shift data and D) task data for median temperature changes are shown. Each dot represents the mean temperature changes of both right and left hands for each individual break. R² values indicate the variability of the data. Statistical significance is indicated as *P < 0.05