This study was designed as a randomized crossover trial with a total of four independent experiments, which differed in ambient temperature and the use of PPE: 1) normal temperature (22 °C) and no PPE (NN), 2) normal temperature (22 °C) and PPE (NP), 3) warm temperature (27 °C) and no PPE (WN) and 4) warm temperature (27 °C) and PPE (WP) (Fig. 1). The selection of the high temperature was based on the guidelines of the German Federal Ministry of Labour and Social Affairs, where the ambient temperature in workspaces should not exceed 26 °C [17]. However, we chose 27 °C in order to have a sufficient temperature difference to the reference temperature of 22 °C, which was the lower technical limit of the climate chamber. The composition of the PPE was adapted from our hospital regulations for infectious diseases and consisted of protective gloves, FFP2 masks, a disposable plastic gown and a face shield.

Summary of the experimental procedure

Participants were recruited using convenience sampling through both offline and online postings at the LMU Munich University Hospital. Inclusion criteria comprised individuals aged 18 to 60 years with a background or experience in healthcare-related activities. Exclusion criteria included sensitivity to heat (e.g., dizziness, redness on the skin), obesity (BMI > 30) and severe chronic diseases such as COPD or acute coronary heart disease. The study team conducted a preliminary examination to confirm eligibility for participation. All participants provided their informed consent through signed consent forms and were granted a 300 € allowance after completion of the entire study. Data collection was performed between October 2021 and March 2022.

The climate chamber at the University Hospital Munich is a 5 m by 3 m by 2.2 m (L/W/H) room (33 m3) connected to an external air-conditioning unit. The chamber was accessed via a windowed safety door that could be opened from both sides. Communication with participants was conducted by either eye contact, hand signs or via an intercom system. In this room, a patient care situation was simulated. In detail, the interior included a table with a chair, a treadmill, a patient bed and a patient dummy (CLA1®, 21 kg, Coburger Lehrmittelanstalt, Coburg, Germany). A picture of the experimental setting can be found in the supplementary information (Figure S1). The temperature in the chamber was controlled and set to either 22 °C or at 27 °C depending on the experiment. The relative humidity was set to 40% for all experiments to limit the number of variables. Climatic conditions in the chamber were monitored using a QUESTemp 34 Heat Stress Monitor® (TSI, Shoreview, Minnesota, USA), which was placed on a table at a height of 1 m.

The procedure of all trials was identical with the exception of wearing and changing PPE. Participant arrived at the chamber and changed into a standard hospital gown. Thereafter, they entered the chamber to follow a standardized 3.5-h protocol, which can be found in the supplementary information. The protocol was developed under supervision of an experienced HCW at the State Vocational School for Nursing at the University Hospital of LMU Munich. It simulated a typical healthcare setting during a regular shift and included taking care of a patient (mobilization, washing, changing clothes), walking and sitting at a table (administrative tasks). In experiments involving PPE (NP, WP), participants were required to put on and take off PPE at multiples times during the experiment. Participants were allowed to drink water (room temperature) throughout the experiment. The amount of total consumed water was noted. Every participant conducted each experiment at the consistent time of the day (mornings or afternoons). The interval between two experiments was at least one week, but not longer than ten days. The order of experiments was randomized for each participant.

To assess participants’ subjective well-being during the experiments, a questionnaire in German language consisting of several survey instruments was used. An English translation of the questionnaire as well as the original document can be found in the supplementary information. Prior to each trial, the participants assessed their perceived personal state of health using a visual analog scale ranging from 0 (worst health status) to 10 (best health status). After each trial, participants were once again asked to assess their perceived personal state of health. Moreover, thermal, physiological and psychological stress were evaluated using the same scale ranging from 0 to 10. Questions about the effect of heat and PPE on personal restrictions and perceived health issues were adapted from a previous HCW survey [8]. Possible responses on the 4-point Likert scale were “no “, “rather no”, “rather yes” and “yes”. Furthermore, the NASA task load index (NASA-TLX, scale ranging from 0 to 10) was included to evaluate the influence the effect of heat and PPE on the perceived workload [19]. Finally, the participants were asked to provide free-text answers if they experienced anything unusual not covered by the questionnaire.

Excel (Version 16.0, Microsoft Corporation®, Redmond, USA) was used for initial data processing. The results were stratified by experimental condition as well as by experimental order. Due to the non-parametric distribution of the data, Friedman-Tests with Bonferroni correction (α = 0.05, SPSS®, Version 29.0, IBM, Armonk, USA) were used for outcomes with continuous results. For outcomes using a 4-point Likert scale (“no”, “rather no”, “rather yes” and “yes”), a descriptive analysis was performed based on the share of each answer. The visualization of the results was performed using R Statistical Software® (version 4.0.0). For box-whisker plots, individual differences between the reference experiment NN and the other experiments (NP, WN and WP) were calculated.