Are Work-related Accidents Associated with Specific Times of the Day and the Elapsed Time after the Start of the Work Shift?

Introduction: This study investigates the temporal distribution of work accidents (WA) in Brazil using data from the Notifiable Diseases Information System (SINAN) for the year 2022. SINAN is the official reporting system for public health events in Brazil, including notifications of work accidents among both formal and informal workers.

Methods: An analytical, cross-sectional study was conducted using data accessed from SINAN on October 20, 2024. WA cases were selected based on the completeness of information regarding the time of the accident and hours elapsed since the start of the work shift. Accidents were categorized into three groups: typical accidents, commuting accidents, and all accidents. A chi-square (χ²) goodness-of-fit test was used to assess the distribution of accidents throughout the day. Analyses were conducted with Jamovi software version 2.2.5, supplemented by histograms and moving averages.

Results: In 2022, 65,535 WAs were recorded in Sinan. Of these, 47,367 included information on the time of occurrence, and 32,530 specified the time elapsed since the start of the shift. For all accidents, the highest frequency was observed at 10 am, with 4,897 records (10.34% of the total), followed by 9 am (4,636 accidents, 9.79%) and 8 am (4,286 accidents, 9.04%). For typical accidents, the peak also occurred at 10a.m. (4,433 accidents, 11.47%), followed by 9 am (4,180 accidents, 10.82%) and 8 am (3,529, 9.13%). Commuting accidents had the highest number of records at 7 am (973 accidents, 13.91%) and 6 am (745 accidents, 10.65%). Chi-square tests confirmed a non-random distribution of accidents, with significant values for all types analyzed: χ² = 28,962 (all accidents), χ² = 27,929 (typical accidents), and χ² = 4,109 (commuting accidents), all with p < 0.001. The analysis regarding hours after the start of the shift also revealed occurrence patterns. For all accidents, the highest concentration was observed between 0 and 4 hours, with peaks after 2 hours of work (4,455 accidents, 13.24%) and after 1 hour (4,128 accidents, 12.26%). For typical accidents, peaks were also observed after 2 hours (4,118 accidents, 14.18%) and 1 hour (3,823 accidents, 13.16%). For commuting accidents, 33.37% of records occurred before the first hour of work, with 1,162 accidents. Goodness-of-fit tests reinforced the unequal distribution of events after the start of the shift for all types of accidents (p < 0.05).

Conclusion: The results indicate a non-random and statistically significant distribution of WAs, influenced by specific times of the day and hours after the start of the shift, highlighting the importance of considering temporal factors in accident prevention strategies. Support: CNPq- Productivity grant to F.M. Fischer n^o. 306963/2021–3.