Nurses’ clinical alarm-related behaviors and influencing factors in China^†

Alarm learning behaviors in the study mean seeking ways of acquiring alarm-related knowledge. Alarm learning scored highest for all NCAB items; 41.0% (n = 971) of nurses responded that they often actively seek ways of acquiring alarm-related knowledge, and 32.7% (n = 775) responded that they always do. These data reflect a lack of clinical alarm knowledge in nurses. In this study, 49.1% of surveyed nurses indicated that they had not received alarm-related training during the last year. Other studies report that more than 60% of nurses indicated that they needed more training on the use of bedside and central monitors,⁵ and it reported that nurses lacked knowledge of equipment alarms and equipment with alarm functions.⁷ More than 90% of nurses reported that they have learned how to operate the equipment, the purpose of using equipment, and the function of equipment.¹³ Our results, taken together with those from other studies, suggest there is an urgent need to enhance clinical alarm-related training and education for nurses and other hospital personnel.

Alarm response behaviors refer to how to handle an alarm once the nurse is informed of it, in the study. As for frequent false or interfering alarms, 33.1% (n = 784) and 18.5% (n = 437) of nurses responded that they often or always, respectively, turn off/mute alarm. Nearly half (49.2%; n = 1165) of the nurses reported that they always respond immediately upon receiving an alarm signal, which was in agreement with previous studies. Voepel-Lewis et al. reported that 26% of oxygen saturation alarms in general surgery wards were not responded to by nurses,¹⁴ and another study reported that the response rate to ECG alarms was only 46.8% in non-critical care settings.¹⁵ Another study found that 14/34 (41%) red alarms never responded immediately and 70% of patient monitor alarms never responded promptly.¹⁶ Funk et al. ⁷ reported that nurses sometimes do not respond to alarms or lower sound volume, heighten alarm thresholds, inappropriately silence alarms, or even disconnect them.

In this study, 41.4% (n = 981) nurses responded that “if they can’t handle the alarm, they will always actively seek the assistance of other colleagues.” Similarly, another related study found that ICU nurses relied on the collective efforts of the team to be aware of the alarms.¹⁷ In response to “You will not respond to alarms that you believe are false or interfere with.” A total of 36.6% (n = 867) of nurses responded “often” and 13.5% (n = 320) responded “always”. Insufficient staffing, high severity of illness on the unit, and unbalanced nursing skills are likely to contribute to inadequate alarm response as described above. These findings are in agreement with others who reported that nurses’ responses to ECG monitors were affected by limited reflection on changes of illness, an inability to distinguish different emergency levels, insufficient recognition of the risk from the alarms, and lack of energy to respond.¹⁸

Alarm setting refers to adjusting the alarm thresholds, the priority of the alarm according to the patient's situation when using the alarm device, and adjusting the alarm volume according to the ward environment and time. In this study, when using alarming devices, 34.7% (n = 821) of nurses always adjusted alarm-related settings, and 40.2% (n = 952) often adjusted them. Only 15.4% (n = 364) of nurses responded that they generally did not adjust alarm volume according to the environment and time on the ward, and 30.9% (n = 732) of nurses always stated that personalized alarm thresholds were followed according to the patient condition. In a recent study, only 35% of nurses agreed with a requirement for the documentation of alarm settings at their institution.¹⁹ This may be due to the disproportionate increase in alarm devices, or nurses may be too busy or not pay attention to alarm settings. As a diagnostic test, monitoring was set up to be highly sensitive, but poor with specificity. Predictably, the resulting high false-positive rate led to the emergence of alarm fatigue, which ultimately reduced the sensitivity of monitoring.²⁰ The challenge, therefore, is to reduce the frequency of false-positive, clinically meaningless alarms with appropriate, personalized alarm settings.

Alarm recognition behaviors refer to identifying accurately the alarm trigger device and alarm priority according to the alarm signal once the nurse is informed of it, and the recipient of the alarm understands the meaning of the alarm and can predict what changes in the patient's condition may occur.in the study. When asked if they could accurately identify the alarm-triggering device according to the alarm signal, 48.5% (n = 1148) of nurses responded “often,” and only 25.5% (n = 604)responded “always.” Only 22.2% (n = 526) of nurses responded that they can always identify different levels of alarm according to different alarm signals, and 42.3% (n = 1001) responded with “often.” One study reported that 99% (n = 164) of nurses stated that alarms should indicate alarm priority, and only 36% (n = 59) of nurses agreed that alarms frequently could not be heard or were missed and 62% (n = 102) of nurses felt that environmental background noise had interfered with alarm recognition.²¹ With currently used clinical alarms, it is not easy to distinguish the alarm parameters and priority, which may make the nurses handle one alarm and miss other, more important alarms²²; therefore, better training on alarm recognition is required.

Alarm notification means that once an alarm is generated, the nurse can be notified in time, as mentioned in the study. Only 13.0% (n = 309) of nurses responded that they “always” missed an alarm because they could not hear it, and 32.5% (n = 769)responded “often”. When asked if they were alerted to an alarm by the patient or escort, 22.7% (n = 537) and 8.5% (n = 202) responded “often” and “always,” respectively. This agrees with a study that reported that family presence at the bedside helps nurses prioritize alarm responses in hectic work environments.²³ A descriptive qualitative study conducted using focus groups has shown that nurses mentioned patient education and explanations about alarms as a way to minimize patient concerns and reduce anxiety.¹⁷ Ruppell et al. ²⁴ reported that although initial alarm notifications declined by 68% after their intervention, these notifications accounted for only about half of all alarm notifications. At present, there are two ways to be alerted of the ECG monitor: sound alarm and visual alarm. If the noisy environment of the hospital affects the audibility of the alarm, when nurses are busy with other work and far away from the monitor, the alarm will not be received. To overcome this, some institutions have made massive investments in personnel, such as monitor “watchers” to help nurses identify actionable alarms. As alarm notification behaviors were the lowest scores of all NCAB items, it is important to take measures to improve timely alarm notifications in the alarm management process.

Nurses 46 years of age and older scored highest for alarm behavior, which may be explained by the eagerness for professional knowledge learning and rich experience with proper handling of alarm-related events. The low scores for alarm-related behavior in the 20–30 years age group were probably due to insufficient professional knowledge, inadequate equipment-related know-how, shorter working experiences, and less safety awareness of patients.

The longer the working years of nurses, the higher the alarm behavior scores. Nurses-in-charge scored the highest for alarm behavior, likely related to their long working life, rich working experiences, and higher levels of self-confidence. Senior nurses scored high for alarm-related knowledge and attitude yet low for behavior. Senior nurses generally have abundant clinical experiences, meticulous handling of safety issues, and comprehensive, specialized knowledge and skills. However, with little practical experience in alarm management, the low scores for behavior may be due to weaker risk awareness, and a reduced ability to predict risks.

ICU nurses scored highest for behavior, followed by anesthesia and surgical nurses; these findings are in agreement with a previous study in which nurses who worked in ICUs performed better in relation to clinical alarms than those working in other wards. As physiologic monitoring is most common in ICUs, alarm fatigue in the ICU setting has been well studied.²⁵ We hypothesize that there are better clinical alarm management initiatives in the ICU than in other departments.

Compared to second-level hospitals, the alarm behavior scores of nurses in the third-level hospitals were higher. The behavior scores of nurses at hospitals affiliated with colleges or universities were higher than for nurses at non-affiliated hospitals. This may be explained in part by the fact that critical patients at third-level hospitals are relatively complex and more patients may require monitoring equipment, thus providing these nurses with richer clinical alarm experiences.

Higher scores for alarm attitude or alarm knowledge were positively correlated with higher scores for alarm-related behavior. The theoretical model of knowledge, belief, and action²⁶ divide change in human behavior into three interrelated processes: acquiring knowledge, generating beliefs, and forming behaviors. Knowledge is the basis of change; faith indicates behavioral tendency, and the premise of changing behavior is to change beliefs and attitudes. To change behaviors, we must have knowledge and information as the basis and correct beliefs and positive attitudes as the driving forces. In this study, nurses with high alarm knowledge scores had high attitude and behavior scores, which corroborates the view of Lorenz,⁸ who suggests reducing alarm fatigue through training and using personalized patient-related alarm restrictions.

For nurses who have received alarm-related education and training, the alarm knowledge and behavior scores were positively correlated, (On the other side, those who have not received the clinical alarmrelated education and training (control = yes) (P < 0.001) have a negative correlation with the clinical alarm behavior score). Additionally, nurses that were more willing to participate in alarm-related training had higher scores for alarm-related attitude and behavior. Education and training have been considered effective measures for improving nurses’ alarm behavior: these factors may be mutually beneficial.²⁷

There were statistical differences in alarm-related behavior scores depending upon the use of auxiliary technology such as central monitoring stations for improving alarm management or formulation of alarm management-related systems or norms. Nurses who had used auxiliary technology to improve the management of alarms scored higher for alarm-related behaviors than those who had not used the technology. Similarly, nurses who worked at departments or hospitals with alarm management-related systems or norms scored higher than those who did not have such systems in place. Although critically ill patients are concentrated in ICUs at hospitals, there are still no authoritative medical equipment alarm management processes or parameter setting guides for ICUs.²⁸ In our study, if no improvement project for alarm management had been implemented over the last 3 years (control = yes) (P < 0.001) and no relevant system or norm of alarm management had been formulated (control = yes) (P < 0.001), there was a negative correlation with scores for alarm behavior. Quality control circle activities or quality management project improvements have recently been implemented in some hospitals here for improving alarm proficiency.²⁹ Optimal alarm management systems should be created through multi-channel collaborations at the levels of clinical staff, hospital administrators, and equipment manufacturers.²⁸ However, now, there is lack of national clinical management norms and guidelines in China. Clinical norms are usually based on departmental operating rules, such as checking alarms and requiring timely responses to alarms during shifts. Few hospitals use such mobile devices as wristband watches, pagers, or mobile bracelets to improve alarm notification.

There were no statistically significant differences in the scores for alarm behavior of nurses whether or not they had experienced adverse events related to clinical alarms in the past 2 years. Although 7% (n = 165) of nurses in this study experienced alarm-related adverse events, which was less than those in foreign studies (18%),⁷ did not improve their scores for alarm-related behavior. One possible explanation is that clinical nurses do not consistently report adverse events related to medical equipment. Also, administrative departments may not pay enough attention to or draw useful lessons from these adverse events. Interestingly, in China, unlike other countries, there is no systematic alarm management or alarm adverse event warning training and no protocols for analysis or improvement of clinical alarm adverse events. For example, six hospitals in Ireland reported adverse events related to clinical alarms in 2018, and more than half of nurses (54%, n = 88) knew about adverse patient events related to clinical alarms.²¹

In the present study, there were no statistically significant differences in the scores for alarm behavior among nurses regardless of gender, educational background, marital status, administrative position, organizational attribute, or whether they have children. In our study, highly educated nurses scored higher for alarm-related knowledge and attitude, but lower for alarm behavior. However, some researchers have shown no differences in alarm-related knowledge among nurses with different educational backgrounds.³⁰ The lack of statistically significant differences in scores for alarm-related behavior for nurses with varying educational backgrounds may indicate a dearth of teaching content about clinical alarm management in nursing schools in China and warrants further exploration.