Key result
Only ~24% of hospital clocks display the correct time, highlighting widespread timepiece unreliability in clinical areas.
Why the study?
The accuracy and availability of clocks and monitors in clinical areas are uncertain following guidelines discouraging wrist watch use by clinicians.
Observational (n=78)
No
Hospital clocks and monitors are frequently inaccurate, which may have detrimental effects on time-critical clinical care and medicolegal documentation.
Clinicians should not rely on hospital clocks or monitors for precise timing; leaves open effects on care and need for synchronization standards.
Recent Department of Health guidelines encourage the adoption of a ‘bare below the elbows’ dress code in hospitals that includes the absence of a wrist watch [1]. This brings into question the availability and the accuracy of clocks and other timepieces in clinical areas. On the day of the audit the ‘Talking Clock’ was accessed via the hospital switchboard. A control watch was then synchronised with Greenwich Mean Time using ‘The Talking Clock’. All available clocks and monitors in clinical areas pertinent to anaesthetists were checked against the control watch. The accuracy of timepieces was recorded as minutes fast or slow with respect to the control watch, to the nearest minute. The proposed standard was for 100% agreement between all available timepieces and the control watch. The results of this audit were presented at the Department of Anaesthetics monthly audit meeting. The directors of the various clinical areas audited were then contacted by email to inform them of the results, and also to inform them of our intention to re-audit the timepieces. This re-audit was performed 5 months after the original audit using the same methods described above. The initial audit included 38 clocks and 40 monitors spread throughout the following areas: main theatres; daycase unit theatres; elective surgical unit theatres; main and paediatric recovery wards; HDU; ITU; labour ward; CT scanner; resuscitation area of the emergency department; and the anaesthetic office. Only 9 of the 38 clocks (24%) displayed the same time as the control watch (range −12 to +6 min), while only 4 of the 40 monitors (10%) agreed with the control watch (range −11 to +6 min). The re-audit included a total of 36 clocks and 29 monitors. The difference in the number of timepieces from the initial audit was due to ongoing work to upgrade some of the operating theatres. Of the 36 clocks 10 (28%) were in agreement with the control watch (range −3 to +6 min), and of the 29 monitors 4 (14%) were in agreement with the control watch (range −60 to +9 min). The importance of accurate timekeeping in all disciplines of medicine is extremely important. As anaesthesia is often a time-critical specialty, accurate timekeeping becomes all the more important. This applies particularly to the practice of obstetric anaesthesia, where recent national enquiries and guidelines have highlighted aspects of timekeeping in the management of maternal emergencies [2] and grade-1 caesarean sections [3]. We have shown with this simple audit that the clocks and monitors within the clinical areas in which anaesthetists practise cannot be relied upon to keep time accurately. If, as per Department of Health guidelines, clinicians are not permitted to wear watches, there may be a detrimental effect on clinical care and audits, along with potential medicolegal ramifications.
No takes yet. Share an insight, caveat, or question.
Topping et al. (2009) conducted an observational in Accuracy of timepieces in clinical areas (n=78). Clocks and monitors in clinical areas vs. Control watch synchronized with Greenwich Mean Time was evaluated on Agreement with control watch to the nearest minute. Only 24% of clocks and 10% of monitors in clinical areas displayed the correct time compared to a control watch, highlighting the unreliability of hospital timepieces for accurate timekeeping.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: