CL IN ICAL ENGIN E E R ING
system. Incorrect loose alarm settings can cause failures to detect patient deterioration leading to patient harm or even death. Conversely, very tight settings can cause frequent nuisance alarms, resulting in alarm fatigue for clinical teams, and disturbance of patients.
A frequent response to adverse events is to ascribe the failure as a device malfunction or a user error, though the causes (and they are often more than one) are often not as simple. Models such as that of figure 1 can guide investigators to be more open-minded. Keeping with the alarms, a poorly designed user interface for them can hinder adjusting them to match the patient’s condition, or perhaps even predispose the carer to unintentionally switch an alarm off. If investigation of an adverse event points to alarm settings as a possible cause, investigators should check whether alarm setting is straightforward.
The human usability of medical devices is of such importance that an international standard4
has been developed to specify
processes that will help developers improve the usability of their products. This suggests that adverse events resulting from poor ergonomic design should be judged as caused by device rather than user error. Design that encourages correct operation and/or prevents incorrect operation, known as “mistake proofing”, is encouraged.5 Mistake proofing is sometimes referred to as poke-yoke, a Japanese term meaning “to avoid inadvertent errors”. An every-day example is the UK’s 13Amp mains plug: visually its appearance encourages correct placement in the mains outlet socket; mechanically its design prevents incorrect
Patient’s
deterioration detected by medical device which raises alarm
Carer sets alarms
appropriate for patient
Carer observes patient, taking corrective action
Patient harm:
Deterioration not detected
Failure to monitor alarm
Patient condition deteriorating
Alarms not configured
Monitoring fails
Alarm setting inappropriate - fails to detect
Figure 2: Protective barriers that should detect a patient’s deteriorating condition shown in the form of Reason’s 1
Swiss Cheese Model
insertion. Where possible medical device design should adopt this approach. Human usability and mistake proofing also apply to the design of the healthcare environment and to policies and procedures. A severe adverse event has been described whose origins were in the poor layout of medical devices in the care environment.6 The nurse caring for the patient had responded to an alarm by pressing what was thought to be the alarm silence button, but the layout prevented clear sight of the alarm controls. What had been pressed was the control to increase the infusion rate significantly.
A simplistic assessment might judge this as entirely due to user error, the nurse not making sure to clearly see which alarm was being pressed. Such a judgement could leave the layout unchanged, perhaps causing a
repeat incident. A holistic appraisal of the care system can uncover the actual causes: device layout; operator pressing control without checking; device human usability design. When medical devices are chosen their selection should include assessing their usability. Ideally this should be carried out by the carers who will use the devices, and in the clinical environment where they will be used. An additional pre-procurement check could be to try and use the device incorrectly, including attempting to assemble the device and its accessories incorrectly. Human usability and mistake proofing should be given a strong weighting when combining all evaluation factors in deciding what to procure.
National recording of adverse events About twenty years ago, two seminal reports called for reducing the many repeated adverse events by learning from them.7,8
Patient
Health systems and independent organisations responded by recording and analysing incidents. NHS England’s National Reporting and Learning System9
now records Care Team
over half a million incidents every quarter (compared to 153 when started in 2003). Most incidents do not involve medical devices. Summary data for 2017 lists about fifty thousand device incidents, less than 3% of all incidents.10
Of these 86% involved no Medical Devices
Care Plans Procedures
Infrastructure and support system including clinical engineers who support the devices and organisation management who fund them
Figure 1 Healthcare system showing the elements that support the patient 64 l
WWW.CLINICALSERVICESJOURNAL.COM
patient harm, but moderate or severe harm or death occurred in about 1%. The data only includes minimum information on the causes but that which is available suggests device failure as the dominant cause (39%), with 31% caused by device unavailability and slightly over 10% due to ‘user’ error. The analysis reliability depends crucially on the comprehensiveness of the underlying incident reports. Global percentages of failure mechanisms overlooks the different causes between device types.6
Targeted reporting of
medical devices incidents is carried out by the UK’s regulatory authority, the Medicines
FEBRUARY 2021
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72
