Portal de Periódicos da CAPES

Connected devices to evaluate sleep, physical activity and stress pattern of anaesthesiology and intensive care residents

2020; Lippincott Williams & Wilkins; Volume: 37; Issue: 7 Linguagem: Inglês

10.1097/eja.0000000000001207

1365-2346

Franck Ehooman, Lucille Wildenberg, Elsa Manquat, Sarah Makoudi, Suela Demiri, Hélène Carbonne, Jean Bardon,

Workplace Health and Well-being

Editor, Anaesthesiology and intensive care residency is challenging with long hours, night calls and a high level of stress and its consequences on physical and mental health has become a major concern.1 Connected devices are increasingly involved in everyday practice. The amount of data collected, so-called big data, permits us to uncover unknown correlations or trends in large data sets of populations. The iFAR study (Impact de la Formation en Anesthésie-Réanimation), is a prospective longitudinal study describing sleep patterns, physical activity and stress using connected wristbands in real-life conditions in a cohort of 20 French residents working in an anaesthesiology, surgical ICU or medical ICU. They wore a connected wristband during 2 consecutive months allowing real-time monitoring of sleep, physical activity and heart rate (HR) 24 h a day, during working days (including on-call days) and holidays. Residents also responded to the Perceived Stress Scale 10 (PSS-10) questionnaire and self-evaluated their sleep duration needs. We studied sleep duration, sleep debt by comparing actual sleep duration and sleep needs, physical activity and HR during working days, on-call days and holidays, each participant being its own control. The connected wristbands used were CHARGE HR manufactured by Fitbit (Fitbit Inc., San Francisco, California, USA) due to their ability to monitor HR, physical activity and sleep duration continuously with a real time synchronisation between the wristband and the manufacturer's web platform. The devices were synchronised with residents’ smartphones and an account was created on the manufacturer's website using an anonymised login. The study was approved by the national CERAR (comité d’éthique de la recherche en Anesthésie-Réanimation) ethics committee 74 rue Raynouard, Paris, France, under the reference IRB 000102542015012, on 19 February 2015, and consent forms were signed by all participating residents. Quantitative variables were expressed as median [interquartile range], while qualitative variables were expressed as a percentage. Nonparametric tests were analysis of variance (ANOVA) Friedman test and parametric tests were one way ANOVA and Student's t test. The alpha risk was set to 0.05. Analyses were performed using Prism Software (GraphPad Software, La Jolla, California, USA). We collected 1027 data items per resident during the study period. Participants’ characteristics are described in Table 1. In total, 1320 days were analysed, including 189 on-call days, representing a median of 9 [8.75 to 11] on-call days per resident during the 2 months study period, 302 days of holidays and 829 working days excluding on-call days.Table 1: Participants’ characteristicsMedian sleep time was 6.5 h [5.3 to 7.6] on working days compared to 3.8 h [2.7 to 5.3] during on-call days (P < 0.0001) and 7.5 h [6.3 to 8.6] on holidays (P < 0.0001) (Fig. 1). Ninety percent of residents had a nap on the day after a night call with a median duration of 59 min [0 to 150]. By subtracting actual sleep duration from estimated sleep needs, the median sleep debt was 49.8 min [34.8 to 79.3] for each night. Cumulated sleep debt was 64.7 h [28.7 to 82.5] per resident during the 2 months period. Forty percent of residents presented a cumulated sleep debt of more than 72 h, corresponding to 10 nights.Fig. 1: Hours of sleep on regular, on-call days and holidays.Daily walking distance per resident was 8.3 km [6.3 to 10.8] corresponding to 11 888 steps [9109 to 15 755]. Physical activity more than three metabolic equivalents of tasks was significantly longer during on-call days compared with regular days, 6.4 h [5.5 to 7.3] versus 5.4 h [4.2 to 6.5], respectively (P < 0.0001) (Fig. 2) and longer during working days than during holidays, 5.6 h [4.4 to 6.8] versus 5.0 h [3.9 to 6.0], respectively (P < 0.0001).Fig. 2: Physical activity more than three metabolic equivalents of tasks in hours on regular, on-call days and holidays.HR did not vary significantly between regular (65 bpm [62 to 69]), on-call days (65 bpm [62 to 69]) and holidays (66 bpm [58 to 70]). Median PSS-10 score was 21 points [20 to 26] corresponding to the threshold above which respondents are emotionally troubled by their professional environment. Five residents (25%) presented a PSS-10 score superior to 27 points corresponding to a strong feeling of powerlessness facing professional situations, at risk of burnout according to PSS-10 scale interpretation.2 PSS-10 was not different between ICU and anaesthesiology residents, with a median of 21 points [20 to 34] and 23.5 points [19 to 22.5] (P = 0.15), respectively. While it is not surprising to observe that residents sleep less during night calls, the tremendous sleep debt accumulated during the 2 months period is alarming, since it is demonstrated that the effects of sleep deprivation are cumulative.3 In addition, despite a debate on individual sleep needs, current guidelines recommend 7 to 9 h of sleep per night to promote optimal health,4 and sleeping more than 9 h might be appropriate for young adults and individuals recovering from sleep debt.5 On top of that, connected wristbands use actigraphy to monitor sleep thus giving no information on sleep quality. Indeed, if sleep is possible during night calls, it is often limited and fragmented,6 so that the sleep debt might have been underestimated. The results of this study performed in France can hardly be generalised to other countries where duty regulations, schedules and activities may be significantly different. Among the solutions to deal with professional stress, combining these new measure technologies with high fidelity medical simulation seems a possibly attractive approach. Furthermore, measures could be taken to withstand stressful situations (simulation for the most stressful situations, regular evaluation, anonymous help call numbers, psychological support in and outside medical wards). To conclude, using connected wristbands, an innovative method to quantify sleep and physical activity, this prospective preliminary study (iFAR) observed that anaesthesiology and intensive care residents sleep less during work periods than during holidays, accumulate sleep debt (64.7 h in 2 months), are highly active and present high levels of stress when compared with the general population.7