Abstract
BACKGROUND:
Desensitisation to alarms, or alarm fatigue, is a concern for healthcare staff. Little is known about how physiotherapists relate to, or are affected by clinical alarms. This pilot study aimed to explore physiotherapists’ attitudes and practices towards physiologic monitor alarms (PMA) in critical care.
METHODS:
An online survey of physiotherapists with critical care experience working at a Model 4 Irish Hospital. A sample of convenience was used with all eligible physiotherapists invited to complete the online survey via email (n = 33). Demographic information was captured, as well as information on experiences, practices, and barriers and facilitators to managing PMA.
RESULTS:
The response rate was 76% (25/33). All respondents worked on-call and weekends, with one respondent managing a current day-to-day critical care caseload. The majority of respondents (20/25, 80%) perceived all PMA as clinically important, but a workplace distraction (19/25, 76%). Negative emotions were commonly experienced by respondents on hearing PMA. All respondents (25/25, 100%) reported to notice their patient’s PMA, feeling they had a responsibility to respond. Respondents indicated varying levels of self-confidence in responding to PMA but commonly assessed the cause of the alarm (24/25, 96%) and checked the patient’s condition (24/25, 96%). Education and training was identified as a key barrier and facilitator for physiotherapists in terms of managing alarms in critical care.
CONCLUSION:
This study provides preliminary data on physiotherapists’ attitudes and practices towards PMA in critical care. Additional studies are necessary in order to verify the findings of this pilot study and further explore alarm fatigue amongst critical care physiotherapists.
Introduction
Physiologic monitors in critical care can generate hundreds of alarms per bed per day, with 80% to 90% of these alarms estimated as false [1]. False physiologic monitor alarms (PMA) have no clinical or therapeutic consequence for the patient. However excessive numbers of clinically insignificant PMA can result in alarm fatigue in healthcare staff [2]. When staff become desensitised to PMA they may ignore or miss PMA completely [3, 4]. Alarm fatigue can also lead staff to lower the volume of PMA, modify alarm limits unsafely, and silence or disable PMA entirely [3]. As a result, response to actionable PMA and urgent clinical situations may be delayed, ultimately impacting patient safety [4–6].
Recent descriptive studies exploring alarm fatigue, and perceptions of healthcare workers towards PMA, have focused primarily on nurses. Research has shown that critical care nurses view frequent false PMA as distracting and disruptive to patient care [7, 8]. In a study of an Australian critical care unit (n = 48) 93% of nurses perceived exposure to frequent nuisance PMA can lead to inappropriate silencing and disabling of PMA, as well as slower reaction times to PMA [9]. In the United States, the vast majority of the 1241 respondents to the Healthcare Technology Foundation (HTF) clinical alarm survey (2016) also perceived false PMA occur frequently in clinical practice (87%) and reduce staff trust in PMA, which leads to inappropriate disabling of PMA (83%) [10]. Research, however, investigating how other healthcare professionals perceive PMA in critical care is lacking.
The critical care multidisciplinary team (MDT) underpins the care of the critically unwell patient. It is recognised that in order to address the complex needs of the patient with a critical illness a collaborative approach is required [11]. Physiotherapists are considered an important part of the critical care MDT and their role in the respiratory management and early mobilisation of patients in critical care is well documented [11–13]. Physiotherapy is also the only allied health profession to provide cover to critical care 24 hours per day, 7 days per week. Consequently physiotherapists are required to maintain a core level of competency in the management of critically ill patients irrespective of the clinical speciality in which they are involved in during normal working hours [13–15]. However, in providing a service to critical care physiotherapists are inadvertently exposed to PMA and may also be susceptible to alarm fatigue.
Currently, little is known as to what extent physiotherapists are affected by PMA in clinical practice or how they react to them. Given the lack of available evidence relating to alarm fatigue and physiotherapists, the aim of this pilot study was to generate preliminary data on physiotherapists’ attitudes and practices towards PMA in critical care, through an online survey. The main objectives were to: 1) Identify physiotherapists’ attitudes and perceptions towards PMA in critical care; 2) Identify how physiotherapists respond to PMA in critical care; 3) Identify physiotherapists’ barriers and facilitators to managing PMA in critical care.
Methods
Design
An anonymous online survey was conducted with physiotherapists who provided input to critical care patients at a Model 4 Irish hospital. The critical care unit at the hospital site, a supra-regional centre, consisted of an adult intensive care unit (ICU; 11 beds), high dependency unit (HDU; 6 beds), cardiothoracic ICU (CTICU; up to 5 beds), and post anaesthetic care unit (PACU; 4 beds). Ulster University Institute of Nursing and Health Research Filter Committee provided ethical approval for the study. Ethical approval was also gained from the Clinical Research Ethics Committee at the hospital site where recruitment took place (Ref: C.A. 2143).
Survey sample and recruitment
A convenient sample of physiotherapists who were currently working in critical care and/or providing out of normal working hours (on-call and weekend) physiotherapy services to patients at the study site were invited to participate. Physiotherapy assistants, physiotherapy students, and physiotherapists not trained in critical care or currently providing on-call and weekend services were not considered eligible for inclusion. Due to the exploratory nature of the study the sample size was guided by the resources available to the research team i.e. all eligible physiotherapists, as determined by the inclusion and exclusion criteria, were invited to participate (n = 33).
Agreement was obtained from the physiotherapy manager at the hospital site to distribute an invitation email via the physiotherapy department’s electronic mailing list. The email included a link to the survey, which was available for a period of three weeks, and a participant information sheet. The physiotherapy manager also forwarded reminder emails to participants one and two weeks after the initial invite in order to increase participation [16]. All responses were collected anonymously. Participants indicated their informed consent via an online consent form prior to taking part in the survey.
Survey development
The survey was developed using Survey Monkey, an online survey software package [17], and was purpose-designed by the investigators to explore physiotherapists’ attitudes and practices towards PMA in critical care. The term ‘physiologic monitor alarm’, for the purpose of this pilot study, referred to an audible heart rate, blood pressure, oxygen saturation, and/or respiratory rate alarm. Survey questions were informed by the aims of the study, current literature, and by survey instruments used in studies of critical care nurses but considered unsuitable to adapt for use in this study [9, 18].
The survey was pre-tested by three senior respiratory physiotherapists precluded from participation in the finalised survey. Informal face-to-face feedback was sought on survey content and format, relevance of questions, and completion time. Following feedback minor revisions were made to the wording, flow and order of some of the questions. The time to complete the survey was determined to be no more than 15 minutes. As minimal revisions were required the amended survey was finalised and administered to participants without additional pilot testing.
The final version of the survey consisted of 29 questions presented over six sections: demographics, experiences and opinions of PMA, responses to PMA, barriers to alarm management, facilitators to alarm management, and general comments. The majority of questions were closed questions with open-ended questions used to allow participants to elaborate on answers as necessary. Survey questions were not made mandatory to minimise measurement error [19]. A copy of the survey is available as an online supplement.
Statistical analysis
Survey responses were downloaded from Survey Monkey onto a Microsoft Excel spread sheet. Data were transferred to and analysed using Statistical Package for Social Sciences (SPSS) for Mac version 25.0. Calculations of frequencies and percentages were made for nominal and ordinal data generated by single answer, multiple response, and Likert scale questions. Numerical data (age, years experience working as a physiotherapist, years experience working on-call and weekends) were assessed for normality using the Shapiro-Wilk test of normality. Data were found to be normally distributed and were reported as mean and standard deviation (SD).
Responses to open-ended questions were manually coded by the primary researcher, using an open coding method, and independently crosschecked by the second author. Open-text responses relating to who should respond to PMA, actions taken in response to PMA, competency training needs, and barriers and facilitators to alarm management were analysed using frequencies and percentages.
Results
Survey response
Twenty-five surveys were returned, a response rate of 76% (25/33). All were included for final data analysis. As respondents were not required to answer every question results are provided as the number and percentage of total respondents to each question where appropriate.
Respondent characteristics
Respondent characteristics are presented in Table 1. All respondents worked on-call and weekends and had done so for between 1 and 25 years (mean 10 years). As part of their on-call and weekend duties respondents worked in critical care between two and six times a year, with 96% (24/25) last in critical care within the previous six months. Only one respondent had a current day-to-day clinical caseload in critical care.
Demographic characteristics of respondents
Demographic characteristics of respondents
Note. SD, standard deviation.
Q 20. Actions taken by physiotherapists in response to physiologic monitor alarms generated by a patient during physiotherapy assessment and/or treatment (n = 25)
The majority of respondents (20/25, 80%) perceived all PMA as clinically important, but 44% (11/25) felt that when PMA continually sounded they could not trust them. When asked if they were distracted by PMA 76% (19/25) of respondents said yes, with 68% (17/25) overwhelmed by the number of PMA that occurred in critical care. Feelings experienced by respondents in response to PMA are summarised in Fig. 1. The most frequent response was that PMA made them ‘anxious’ (13/24, 54%).
Q 11. Respondents’ emotional responses to PMA. Proportion of respondents who reported experiencing the suggested feelings on hearing physiologic monitor alarms in critical care (n = 24).
All respondents (25/25, 100%) indicated that they take notice of PMA generated by their patients. When asked who should respond to these alarms over half of respondents (13/24, 54%) believed the healthcare professional at the bedside was responsible, and that if they were with the patient then they should respond (25/25, 100%). Less than half (10/25, 40%) felt the same level of responsibility for patients they were not treating, with 52% (13/25) reporting to not paying active attention to other patients’ PMA.
The majority of respondents (24/25, 96%) reported that they would assess the cause of PMA, and check the patient’s condition, with only one respondent reporting that they would ignore PMA (Table 2). Most respondents also stated that they would never disable PMA to permanently prevent them from sounding or alter limits of PMA (Table 2). Although respondents did report silencing and pausing PMA to temporarily stop them from sounding they commonly alerted nursing staff and reassured the patient and/or family (Table 2).
When asked to describe any other measures taken in response to PMA, of those that commented, most respondents said they would modify their treatment (2/6, 33%) and clinically reason the suitability of continuing treatment (2/6, 33%). One respondent explained “I consider my handling and treatment choice with that patient, I consider what level of sedation that patient has received and I decide whether to continue treatment and discuss with the nurse in charge” (Physiotherapist #5). Another stated, “If the alarm is in relation to low O2 [oxygen] saturations, I may increase O2 or provide O2 breaths to see if it improves” (Physiotherapist #16). Respondents also commented that they would seek help (1/6, 17%) or stop physiotherapy treatment (1/6, 17%).
Respondents’ perceived level of confidence in responding to PMA are detailed in Table 3. Respondents reported being confident in performing most suggested alarm responses but did not feel confident in terms of disabling an alarm (18/25, 72%) and changing alarm limits for physiologic measurements (16/25, 64%). Of the respondents, 16% (4/25) reported not knowing how to disable PMA, with one respondent indicating a lack of knowledge in terms of how to identify alarm limits for physiologic measurements (1/25, 4%), change alarm limits (1/25, 4%), and differentiate technical and patient alarms (1/24, 4%).
Q 24. Physiotherapists’ perceived level of confidence in performing actions related to managing physiologic monitor alarms (n = 25)
Q 24. Physiotherapists’ perceived level of confidence in performing actions related to managing physiologic monitor alarms (n = 25)
Note. * n = 24.
Respondents’ level of agreement to suggested barriers to managing alarms is presented in Fig. 2. The main responses reported were a lack of knowledge/training on the bedside monitoring equipment (20/25, 80%) and on how to customise alarm limits (20/25, 80%). When asked about other potential barriers seven respondents reported fear of causing harm (2/7, 29%), habit of deferring to experienced staff (3/7, 43%), lack of confidence (3/7, 43%), and limited opportunities to develop alarm management skills (2/7, 28%). Sixty-eight percent (17/25) felt that a lack of time was not a significant barrier to managing alarms.
Q 25. Perceived barriers to alarm management. Proportion of respondents who agreed, were unsure, and disagreed with suggested barriers to managing physiologic monitor alarms in critical care (n = 25).
Respondents were also asked to indicate what would best enable them to manage PMA in clinical practice. With most respondents (24/25, 96%) indicating that physiotherapy competency training for critical care should include the physiologic monitor and PMA, education and training was identified as a key facilitator (Fig. 3). All respondents agreed that practical training on both the bedside monitor (25/25, 100%) and how to respond to PMA appropriately (24/24, 100%), and education on factors contributing to high alarm rates in critical care (25/25, 100%) would facilitate their management of PMA (Fig. 3). Only one respondent reported that the development of a physiotherapy specific alarm training package would not be helpful in facilitating alarm management (Fig. 3). When asked to consider their own personal learning needs 18 respondents provided additional suggestions for competency training, with topics summarised in Fig. 4.
Q 27. Perceived facilitators to alarm management. Proportion of respondents who agreed, were unsure, and disagreed with suggested facilitators to managing physiologic monitor alarms in critical care (n = 25). Key: *n = 24.
Q 18. Topics for competency training. Proportion of respondents who made suggestions of topics for physiotherapy critical care competency training on the physiologic monitor and monitor alarms (n = 18).
Respondents were also asked to give details on any other facilitators that may help physiotherapists in managing PMA in clinical practice in critical care. Two respondents felt that simulation training would be very helpful for physiotherapists to learn how to manage PMA. In addition one respondent encouraged “common sense, being alert and reactive” (Physiotherapist #15), while another highlighted the importance of being kept up to date on adverse events associated with alarms “… hearing cases where the implication of ignoring alarms has resulted in a bad outcome for the patient to aid avoiding alarm fatigue” (Physiotherapist #5).
Alarm fatigue occurs in critical care staff who, when exposed to excessive numbers of PMA, become desensitised, nonreactive and mistrusting of PMA [5, 20]. This study explored if physiotherapists working in critical care experience similar reactions towards PMA. The findings from this study showed that physiotherapists notice PMA, and respond appropriately. Unlike critical care nurses [8–10, 21] physiotherapists reported to not ignore or disable PMA when they sound, but instead assess the cause and check the patient as recommended by the European Respiratory Society/European Society of Intensive Care Medicine Guidelines Statement on Physiotherapy for Critically Ill Patients [22]. It has been reported that critical care nurses are aware that a significant proportion of PMA in critical care are false or nuisance alarms [21, 23–25] and acknowledge this influences their alarm behaviours [7–10, 18]. Interestingly, this study found that physiotherapists perceive all PMA as clinically relevant which may account for the differences in reported alarm responses.
However, in terms of the impact PMA have on physiotherapists the results from this study are comparable to studies investigating alarm fatigue in critical care nurses [7–10]. Physiotherapists reported being overwhelmed by the number of PMA generated by critically ill patients, and this provoked feelings of anxiousness and nervousness. PMA tend to distract physiotherapists, interfering with their work and, when they continually sound, lead to alarm mistrust. However despite these findings physiotherapists do not appear to be apathetic towards PMA. This is in contrast to critical care nurses [26] but may reflect the limited exposure this physiotherapy cohort had to PMA. Typically critical care nurses work a 12-hour shift and can be exposed to hundreds of alarms per day [23, 27]. In this study most physiotherapists only worked in critical care when on-call or at weekends, a minimum of twice a year. In a larger multi-centered study it would be essential to recruit physiotherapists who regularly work in critical care in order to determine if, with more exposure to PMA on a day-to-day basis, they are impacted by alarm fatigue.
Previous research has shown that threshold alarms caused by parameters falling outside a set range [28], and technical alarms resulting from patient and/or equipment movement [28] contribute significantly to false alarm rates in critical care [20, 27]. Physiotherapy interventions, often aimed at eliciting an acute physiologic response in the critically ill patient [22, 29], may inadvertently generate such false alarms. Recent evidence has shown that false or nuisance alarms can be effectively reduced by appropriately altering and/or customising alarm limits [30–32] and by temporarily pausing alarms prior to patient intervention [21], and is reflected in critical care nursing practice [8, 24]. Results from this study indicate, although physiotherapists respond to PMA, they do not proactively manage PMA as part of their clinical practice in critical care. This may be attributed to an acknowledged lack of confidence in terms of changing alarm limits. However, it is important to consider that through ineffective use of physiologic monitoring equipment, physiotherapists may be impacting the alarm load in critical care and contributing to alarm fatigue in other members of the critical care MDT.
A lack of training on PMA has been identified as a contributing factor to alarm fatigue [33–35], with emerging evidence to support alarm management education and training programs for critical care staff [30, 35]. From this study it appears that physiotherapists also perceive training and education as a key barrier and facilitator to managing alarms in critical care, which has implications for competency training programs for physiotherapists working in critical care. Previous studies have deemed the ability to accurately interpret readings from the physiologic monitor as a core skill for physiotherapists working in critical care [12–14], however other essential skills for physiotherapists in terms of working with the physiologic monitor and PMA have yet to be identified. In order to address the specific learning needs of physiotherapists relating to alarm management further studies are needed to establish physiotherapists’ baseline level of knowledge and skill relating to the physiologic monitor and PMA in critical care.
Implications
Research on alarm related issues in critical care has, to date, focused primarily on critical care nurses. This is the first study, to the authors’ knowledge, to explore how physiotherapists are affected by PMA in clinical practice and how they react to them. In this pilot study physiotherapists appeared to value and pay attention to PMA generated by their critically ill patient. In order to promote a culture of alarm safety on critical care units and avoid potentially adverse situations in the management of the critically ill patient it is important that we encourage physiotherapists to consider PMA in their clinical practice. In order to further advance and support the role of physiotherapy in alarm management in critical care future large-scale studies, incorporating all Irish physiotherapists who provide services to critical care, are recommended.
It is worth noting, however, that to date there are no specific national guidelines relating to the management of clinical alarms in Irish critical care units. We recommend the findings of this pilot study are used to help inform and guide local alarm management strategies and polices. We would also encourage that appropriate roles and responsibilities for physiotherapists are identified to ensure safe working practices with PMA. In order to address and overcome barriers to working with PMA in clinical practice we also recommend physiotherapists receive training and education relating to the physiologic monitor and monitor alarms. Physiotherapists in this study reported an unmet need in terms of physiologic monitor and monitor alarm training and education. Based on our preliminary findings we suggest inservice training on the causes and impact of alarms on staff and patients, and practical skill sessions in the critical care environment.
This study highlights the need to consider how PMA affect other members of the critical care MDT, other than nurses. It is important to acknowledge that effective alarm management requires a collaborative multi-disciplinary team approach. We would therefore encourage that physiotherapists are considered key stakeholders in developing appropriate alarm management solutions. The scope of future research should also be broadened to include other allied health professionals who work in critical care.
Limitations
Some limitations to this study should be noted. This was a pilot study conducted at a single hospital site. Consequently the ability to generalise the findings to physiotherapists who provide services to critical care units of differing capacities, specialties, and complexities is limited. The sample size was also small. Although a response rate of 76% was achieved only one respondent was working in critical care full-time. As a result the generalisabilty of results to physiotherapists who primarily work in critical care is limited. Also, due to the under representation of critical care physiotherapists in the sample, it was not possible to determine if, statistically, there were differences in attitudes and practices towards PMA between physiotherapists working full-time in critical care and those who do so only in providing out of normal working hours on-call and weekend services.
Conclusion
PMA are integral to the care and safety of the critically ill patient, and are a necessary part of the critical care environment. In this pilot study physiotherapists perceive all PMA as clinically important. Although the preliminary findings suggest physiotherapists are overwhelmed and distracted by the number of PMA that occur in critical care they do not appear to be apathetic towards alarms that occur during physiotherapy. However, in terms of managing PMA in critical care physiotherapists in this study identified a lack of training and competency in bedside monitoring devices as a key barrier. Additional studies are necessary in order to verify the findings of this pilot study and to further explore alarm fatigue amongst critical care physiotherapists.
Footnotes
Acknowledgments
The authors would like to thank those who gave valuable feedback on the survey and all those who gave their time to complete the survey.
Conflict of interest
The primary author (DO’D) is a practising physiotherapist at the hospital stite included in the research.
Ethical considerations
Ethical approval for this study was provided by Ulster University Institute of Nursing and Health Research Filter Committee and Galway University Hospitals Clinical Research Ethics Committee (Ref: C.A. 2143).
Funding
This research did not receive any specific grant from funding agencies in the public, commerical, or not-for-profit sectors.
