The prevention and reduction of serious medical incidents are important challenges worldwide.1 In the United States, an estimated 400,000 premature deaths per year are associated with preventable harm,2 and in Japan, the associated mortality numbers are estimated to be 1326 to 1433 deaths per year.3 Therefore, medical errors are a substantial threat to patients and their families.

A ventilator is a device that helps patients breathe when they have a respiratory disease by enabling the exchange of oxygen and carbon dioxide and reducing the burden on the respiratory system. However, ventilator use is also associated with complications, such as ventilator-associated pneumonia. The treatment of ventilator-associated pneumonia has been extensively studied.4,5

Human failure and machine malfunction can lead to medical practice errors.6 Several studies on ventilator-related adverse events have focused on human factors.7–9 Ventilator-related errors are often related to specific usability issues, exacerbated by staff often working with unfamiliar ventilators.10 Incident reporting systems are essential for true learning, sustainable risk reduction, and patient safety improvements.11 The Pharmaceuticals and Medical Devices Agency in Japan has collected data on all adverse effects and defects reported by companies or medical personnel since April 2004.12 Although these systems can help solve problems with medical devices, errors related to health care devices are still not well understood.13 Incident reporting for medical devices is currently performed in Canada, and in Germany and Austria, use of incident reporting devices is enforced by law. The Canadian/European Union regulations on incident reporting are intended to improve monitoring and reduce the recurrence of incidents related to medical devices.14,15

In Japan, the number of medical expense claims for artificial respiration guidance at home, which covers the fee associated with home teaching and monitoring of equipment by medial staff, was 228,862 for the 2019 fiscal year16 and 216,300 for the 2016 fiscal year.17 The proportion of individuals 65 years or older was 29.1% in 2021.18 Therefore, Japan requires serious policy challenges with respect to its aging society. Moreover, the average length of hospital stay has been decreasing,19 and patients receiving medical care, such as mechanical ventilation, are discharged to their homes directly. Thus, the number of patients using mechanical ventilation outside a hospital setting is also increasing.

As the number of home-care patients on mechanical ventilators increases, the number of cases in which ventilators used at home are brought back to hospitals will also increase. Hospitals generally use 2 to 5 types of ventilators; some even use more than 5 different types of ventilators and such hospitals account for approximately 83% of the respondents.20

Ventilator-related accidents can be life-threatening for patients. The present study aimed to assess factors related to medical device incidents, especially those related to ventilators. To the best of our knowledge, this is the first study to examine the causes of accidents related to ventilators based on incident reports from nationwide medical institutions. The results of this study may provide patient safety administrators or medical device development engineers insight into the causes of ventilator accidents to develop recurrence prevention measures.

METHODS Data Sources

Since 2004, the Japan Council for Quality Health Care (JCQHC)21 has collected adverse event information and evaluated medical services provided at hospitals. As a neutral third-party organization, the JCQHC publishes near-miss/adverse event information on the JCQHC Web site. In the years 2019 and 2020, 1512 and 1531 medical institutions, respectively, participated in this project on either a voluntary or mandatory basis, accounting for 0.9% of all medical institutions in Japan in 2020 (n = 178,724). Among those hospitals, 660 and 661, respectively, provided case reports of incidents to the JCQHC, accounting for 0.4% of the total. From those reports, we collected data related to near-misses and adverse events.

Sample Data

Data reported from January 2019 to December 2020 were collected from the JCQHC database. A total of 232 reports on adverse events, assigned as patient harm event cases by the participant hospitals, due to medical device incidents were downloaded from the JCQHC Web site; 34 of the 232 reports (14.7%) were ventilator-associated incidents. Data related to the person, situation, and incident were collected in detail.

Analytical Approach

Incident reports include 2 types of data—qualitative and quantitative. First, we analyzed ventilator-associated errors quantitatively. Next, the textual data of the qualitative data in the reports were analyzed using content analysis, which can be used inductively and deductively.22 Two research nurses were involved: one with experience as a patient safety manager and the other, a qualified clinical engineering technologist, with surgical nurse experience. Both had experience in mixed-methods research and content analysis.23 The researchers independently read the incident reports several times and the reasons for errors coded according to the Mattox report13 and the SHELL model.24 The SHELL model consists of 4 domains: software (procedure, protocol, and training), hardware (machines and medical instruments), environment (operating theater, wards, and consultation room), liveware (human factors: doctors, nurses, and other health care professionals, or patients), liveware-liveware (human factors: people surrounding the person, people who were with the person). Mattox reported that factors contributing to medical device errors could be divided into 4 large categories and 60 small categories: device (12 small categories), organizational factors (17 small categories), environment (12 small categories), and device users (19 small categories).13 After the 2 researchers discussed and agreed on the division of codes, a third researcher, a nationally qualified clinical engineer, checked the divisions. The codes were not forcefully assigned, and when data could not be coded per the Mattox report, new codes were created. The discussion was continued until researchers reached a consensus.

Ethical Considerations

This study was conducted using secondary data from the JCQHC. The data are open access, and sensitive information was deleted by the JCQHC staff. All information was anonymized to prevent the possible identification of individuals and facilities.

RESULTS Characteristics of the Accident “Conditions”

Table 1 shows the characteristics of ventilator-associated accidents. Ventilator-associated accidents occurred more frequently on weekdays (n = 26 [76.5%]). The situation period was divided into daytime and night/early morning, which accounted for 58.8% (n = 20) and 41.2% (n = 14) of cases, respectively. The highest frequency of ventilator-associated accidents occurred in the hospital room (n = 22 [64.7%]), and the most common number of related persons was one (n = 25 [73.5%]).

TABLE 1 - Characteristics of Ventilator-Associated Accidents n (%) Day  Weekday 26 (76.5)  Weekend/holiday 8 (23.5) Period  Night/early morning 14 (41.2)  Daytime 20 (58.8) Situation  Hospital room 22 (64.7)  ICU 4 (11.8)  Other 8 (23.5) No. related persons  1 25 (73.5)  2 6 (17.6)  >3 3 (8.8) Patient complications  No impairment/no probability of impairment 19 (55.9)  Reversible patient harm 9 (26.5)  Permanent harm 4 (11.8)  Death 2 (5.9)

Although not shown in the table, the total number of related persons, excluding 1 case with missing data, was 46, and incidents most commonly involved nurses (n = 35 [76.1%]), followed by physicians (n = 6 [13.0%]) and clinical engineers (n = 5 [10.9%]). The most common outcome was no impairment/no probability of impairment (n = 19 [55.9%]), although patient death occurred in 2 ventilator-associated accidents.

Summary of Ventilator-Associated Incident Reports

Ventilator corruption was the only ventilator-related problem that occurred (n = 4 [11.8%]); postaccident investigations noted the probability of failure in the ventilator switch and power supply unit as well as continued use of a medical device beyond the manufacturer-defined, planned maintenance interval. The remaining 30 cases were caused by mistakes committed by medical professionals with respect to ventilator use. A summary of the ventilator-associated incidents is shown in Table 2. The most common error was misuse/misapplication (n = 17 [50.0%]), which included cases of home ventilators, rental equipment, or power supply errors. This was followed by disconnection errors (n = 15 [44.1%]). Although not shown in the table, there were 4 cases (11.8%) related to issues with the standby mode of the ventilator, 4 cases of problems with home ventilators (11.8%), and 5 cases (14.7%) where the error occurred during the transfer of patients.

TABLE 2 - Summary of Ventilator-Associated Incident Reports n (%) Problem with the ventilator itself 4 (11.8)  Corruption of ventilator 4 (11.8) Problem with the usage or management of ventilators by medical professionals 30 (88.2)  Misuse/misapplication (misuse of unfamiliar equipment, such as home ventilators, or rental equipment, misapplication of battery, standby mode) 17 (50.0)  Disconnection (e.g., tracheal cannula, artificial nose) 15 (44.1)  Setting error (e.g., forgot to set the alarm volume, forgot to change the respiratory setting) 4 (11.8)  Connection error (e.g., misconnection of ventilator exhalation and inspiratory circuits, misconnection between the artificial nose and the humidifier) 4 (11.8)

Patient complications and ventilator issues are shown in Appendix 1. Two cases of death and 3 of 4 cases of permanent harm were assigned to disconnection. The number of ventilatory-associated incidents per ventilator type is shown in Appendix 2. By type, 16 cases (47.1%) occurred with general hospital-use ventilators, 7 (20.6%) were with general/transport ventilators, 6 (17.6%) were with home/hospital ventilators, 1 (2.9%) occurred with a mask ventilator.

Factors Related to Ventilator-Associated Accidents

Table 3 shows the factors related to ventilator-associated accidents. The people involved in ventilator-associated accidents were physicians, nurses, clinical engineers, patients, and patients’ families.

TABLE 3 - Factors Related to Ventilator-Associated Accidents Hardware  Power (OI) 7 (20.6) Nos. 7, 9, 10, 15, 22, 30, 33  Degree of intuitiveness of the design 4 (11.8) Nos. 4, 19, 21, 30  Transparency of operations (i.e., can users easily determine what the device is doing?) 2 (5.9) Nos. 16, 30  Improper maintenance, testing, or repair 2 (5.9) Nos. 8, 22  Default mode (poor feedback to users about default mode, inadvisable, unsafe, or unexpected default mode) 2 (5.9) Nos.13, 16 Software  There are no rules for managing before, during, or after device-use (OI) 25 (73.5) Nos. 1, 2, 3, 4, 5, 6, 7, 8, 9, 13, 14, 15, 16, 19, 22, 24, 25, 26, 28, 29, 30, 31, 32, 33, 34  Policy or protocols are not consistent with manufacturing guidelines; failure to heed warnings or adhere to guidelines related to the safe use of a device 9 (26.5) Nos. 9, 12, 14, 17, 22, 25, 26, 30, 32  There are rules, but there is a tendency to ignore the rules (OI) 5 (14.7) Nos. 2, 9, 12, 32, 34  Introduction of devices without adequate education before implementation 2 (5.9) Nos. 1, 24  Improper storage of devices 2 (5.9) Nos. 9, 14  Existence of local rules (OI) 2 (5.9) Nos. 9, 14  Introduction of devices without adequate assessment before implementation 1 (2.9) No. 2  Organizational responsiveness to poorly designed or suboptimal devices and work-arounds 1 (2.9) No. 2 Environment  Overlapping schedules, multiple tasks 11 (32.4) Nos. 6, 7, 12, 13, 20, 23, 24, 25, 27, 29, 30  Use during patient transport or transfer 8 (23.5) Nos. 2, 4, 13, 17, 30, 32, 33, 34  Staffing levels 7 (20.6) Nos. 4, 6, 7, 9, 22, 24, 34  Environment changing (i.e., private room, administration from home) (OI) 7 (20.6) Nos. 1, 6, 11, 12, 14, 18, 28  Not familiar devices (OI) 5 (14.7) Nos. 14, 17, 18, 33, 34  Physical layout of care setting 4 (11.8) Nos. 9, 21, 30, 32  Multiple types of devices adopted in the same department/hospital (OI) 3 (8.8) Nos. 19, 22, 31 Liveware  Lack of awareness of risk or falsely low perception of risk 26 (76.5) Nos. 1, 2, 5, 6, 7, 9, 11, 12, 13, 14, 15, 16, 17, 18, 20, 21, 22, 23, 24, 25, 26, 27, 29, 30, 32, 33  Low attentiveness 18 (52.9) Nos. 1, 2, 5, 7, 9, 10, 12, 13, 16, 19, 20, 22, 24, 25, 26, 30, 31, 34  Problem-solving skills 12 (35.3) Nos. 3, 4, 5, 9, 10, 15, 16, 19, 21, 22, 27, 31  Poor ability to hear and interpret sounds, including alarms and other tones 11 (32.4) Nos. 3, 4, 5, 9, 12, 22, 25, 26, 30, 31, 34  Lack of information-sharing among the staff members (OI) 8 (23.5) Nos. 1, 7, 10, 11, 13, 24, 30, 34  Cognitive level of task (automatic or skill-based, rule-based, or knowledge-based tasks) 6 (17.6) Nos. 4, 5, 17, 19, 31, 33  Degree of formal and informal training on the device 6 (17.6) Nos. 3, 5, 7, 25, 27, 34  Personal interpretation of the meaning of actions and commands such as return or restart or enter 5 (14.7) Nos. 10, 15, 27, 30, 34  Use in environments or care settings for which a particular device was not intended, but no awareness of potential risks 5 (14.7) Nos. 8, 15, 16, 17, 21  Failure to detect a modified or malfunctioning device 4 (11.8) Nos. 4, 13, 22, 26  Alert fatigue 4 (11.8) Nos. 9, 12, 30, 32  Low degree of reporting among front-line staff experiencing difficulty with devices 4 (11.8) Nos. 5, 15, 21, 22  Emotional state (anxiety, fear, stress level) 2 (5.9) Nos. 9, 33  Failure to adhere to safety mechanisms 2 (5.9) Nos. 2, 33 Liveware-liveware  Lack of awareness of risk or falsely low perception of risk 18 (52.9) Nos. 6, 7, 9, 12, 13, 14, 16, 17, 18, 22, 23, 24, 25, 26, 27, 29, 33, 34  Problem-solving skills 11 (32.4) Nos. 3, 4, 5, 9, 10, 15, 16, 22, 25, 27, 28  Lack of information sharing among staffs (OI) 6 (17.6) Nos. 1, 7, 10, 13, 24, 34  Unexpected behavior of the patient (OI) 5 (14.7) Nos. 6, 11, 23, 25, 26  Alert fatigue 3 (8.8) Nos. 9, 12, 32  Lack of education and instructor for families (OI) 2 (5.9) Nos. 6, 17  Lack of information-sharing among families (OI) 1 (2.9) No. 6  Personal interpretation of the meaning of actions and commands such as return or restart or enter 1 (2.9) No. 25

Numbers show the assigned cases. These are items where multiple answers were allowed.

OI, original items that are not included in Mattox’s results.

The factors related to ventilator-associated accidents were divided into 43 items: hardware (5 items), software (8 items), environment (7 items), liveware (14 items), and liveware-liveware interaction (8 items). Mattox’s report13 has 60 categories, but these categories were assigned to 29 items (48.3%); there were 4, 5, 4, 13, and 4 Mattox categories in hardware, software, environment, liveware, and liveware-liveware interaction, respectively.

Hardware had one original item, “power,” which was the most frequent error. Power-related cases included examples such as “the ventilator was kept in the standby mode or was battery-powered, but medical staff did not notice until the patient’s oxygen gas monitor rang an alarm.” A design-related code included in hardware was the degree of intuitiveness of the design (n = 4 [11.8%]).

The software contained 3 original codes: “there are no rules for managing the device before, during, or after using the device,” and “there are rules, but the atmosphere is that of ignoring the rules,” and “existence of local rules.” These original codes were related to hospitals/wards. The code, “there are no rules for managing the device before, during, or after using the device” occurred most frequently (n = 25 [73.5%]). This code included other cases, such as “the hospital did not have rules for using a ventilator that had already been used for a patient, so another medical staff considered the ventilator to be disinfected and reused it for another patient.”

The environment had 3 original items, namely, “environment change,” “not familiar devices,” and “multiple types of devices adopted in the same department/hospital.” The most assigned code was “the schedules overlap, multiple tasks” (n = 11 [32.4%]), followed by “use during patient transport or transfer” (n = 8 [23.5%]).

Liveware had one original item: “lack of information-sharing among staff” (n = 8 [23.5%]). Liveware-liveware had 4 original items: “lack of information sharing among staff,” “unexpected behavior of the patient,” “lack of education and instruction for families,” and “lack of information sharing with families.” The most assigned item was “lack of awareness of risk or falsely low perception of risk” (liveware: n = 26 [76.5%]; liveware-liveware interaction: n = 18 [52.9%]).

Complicated Specifications of Ventilators

There were 14 cases in which the user wanted the equipment to be improved. Table 4 shows the list of the improvements, equipment-related points, and reason for the equipment being difficult to use.

TABLE 4 - Complicated Specifications of Ventilators Issues Cause for Specifications Not Being User-Friendly Intuitiveness of the design when users encounter some trouble The device has no display, or display message not easy to understand when users encounter some trouble The device does not start working just by pressing the start button (concluded as standby mode) Two accessories can be connected for different purposes Robustness of design: alarms and accessories cannot be easily removed Monitoring equipment (e.g., SPO2 monitor) or the ventilator’s accessories were easily removed with external force The alarm continues to sound when a device has an issue The trouble was not resolved, but the alarm bell stopped ringing, or the monitor screen notified the trouble Automatic calculation system when the ventilator is connected to mobile oxygen gas The user needs to calculate how much oxygen can be kept from the remaining gas amount and the gas flow rate of the oxygen cylinder

The following 14 cases were included in these parts. Nine of the 14 cases were assigned to “intuitiveness of design” when a user encountered trouble. These included the problems of display message, connection parts. The code of the “robustness of devices and device accessories” was assigned to 3 cases. Moreover, the “improvement of alarm function” was assigned to 2 cases, and “automatic calculation of the system” was assigned to 1 case.

DISCUSSION

Our study revealed 3 major findings with respect to factors associated with ventilator-associated accidents, which in turn relate to patient safety at the hospital. First, ventilator-associated accidents were caused by an entanglement of complex factors, such as hardware, software, environment, liveware, and liveware-liveware interaction. Second, inadequate communication among caregivers and families was related to ventilator-associated accidents. Third, alarms were overlooked owing to inattentiveness. Nurses are the sharp end of patient care,25 but when faced with ventilator-related issues, they were unable to troubleshoot.

Detangling Complex Factors by A Systems Approach

Medical care circumstances are changing with the use of various kinds of medical devices. Consequently, patients with chronic illnesses requiring ventilation are treated in general wards. However, the types of ventilation equipment involved are complex. The types of medical devices used should be uniform to improve patient safety. When patients with home-based ventilators are admitted to a hospital, hospital staff are faced with a complex situation. On one hand, substituting the patient’s home ventilator with a hospital-ventilator imposes a heavy physical burden on the patient. On the other hand, our findings revealed that medical device accidents were not caused by problems with the ventilator itself, but by the usage or management of ventilators by medical professionals. Most ventilator-related incidents are caused by human factors, education, and training systems, rather than the failure of the device itself.7,26

In this study, a lack of risk awareness, inadequate problem-solving skills, and inattentiveness were frequent causes of ventilator-associated accidents. Mistakes were generally caused by inadequate experience, insufficient training, or outright negligence. Reducing the risk of slip-ups requires attention to the designs of protocols, devices, and work environments using checklists so that key steps are not omitted, thus implementing functions forcefully to minimize workarounds.27 Power failure and intra-hospital transfer have been recognized as important risks to patients,7,28 a trend that was also recognized in our study. Our results focused on the administrative rules for ventilator usage, with respect to their software-related aspects. Canadian registrations suggested potential abnormal uses, such as continued use of a medical device beyond the manufacturer-defined, planned maintenance interval because of the users’ failure to arrange for maintenance.12 We found only one case of ventilator corruption because medical device beyond the manufacture-defined maintenance interval. Moreover, using new devices creates a burden for the hospital staff with respect to education, training, and financial resources.

In Japan, enforcement regulations in the Medical Care Act are set to ensure the safety management of medical devices. In this system, one appointed person oversees the safe use of medical devices, and he/she is responsible for providing employee training for the safe use of these devices.29 When introducing a new medical device, training should be provided to those who plan to use the medical device, and the content of the training should be recorded. Whereas advanced treatment hospitals regularly conduct training (approximately twice a year) on medical equipment that requires technical proficiency, including ventilators, other hospitals do not have such an obligation. Although a previous study13 did not list factors related to hospital regulations or policies, our findings revealed original findings related to policies in different institutions. In Japan, there are no regulations for hospitals regarding the management of medical devices. To prevent incidents related to medical devices, hospital administrators need to make corresponding policies or protocols and set rules for managing devices before they are used. The education and training on the use of medical devices are often regarded as the responsibilities of professional medical staff, but we believe that education and training are responsibilities of the hospital administration.

Improve Communication Among Staff, Patients, and Families

Inadequate communication among caregivers and families is caused by complex factors. The Joint Commission, a U.S. task force that aims to continuously improve health care for the public, proposed 7 goals for hospital national patient safety in 2022.30 The goals focus on problems in health care safety, and one of them is “improvement in staff communication.” Communication breakdown is an important cause of adverse events in clinical practice, particularly during handovers.31 Communication errors are frequently reported in the intensive care unit (ICU) and operating room.32,33

Our findings revealed that lack of information or education sharing among staff and families was related to ventilator-associated accidents. Hesitancy to speak up can be an important cause of communication errors that affect the safety culture, the perceived risks for patients, and the ambiguity or clarity of the clinical situation.34 “Lack of awareness of the risk or false low perception of risk” was also the most common human-related contributing factor for ventilator-associated accidents in our study. The people involved in ventilator-associated accidents were physicians, nurses, clinical engineers, patients, and their families. Lack of perception of risk by health care workers can lead to severe patient impairments. Encouraging patients to advocate for themselves is an important patient safety policy.

Improvement in the Use of Alarms for Patient Safety

The 2022 Joint Commission Hospital National Patient Safety goals included improvements to ensure that alarms on medical equipment were heard and promptly addressed. Our results suggested that liveware and liveware-liveware interactions were associated with alarm-related errors: the ability to hear and interpret sound and alert fatigue. Frequent false alarms reduce attention or suppress the response to alarms.35 In our study, the attentiveness-related aspects of liveware included reduced attention to the alarm, lack of awareness of risk, or false low perception of risk related to alert fatigue. In some cases, long-sounding alarms meant that medical staff did not have adequate problem-solving skills and failed to detect a modified or malfunctioning device. Ventilators have numerous functions, such that even when a skilled medical worker, such as a nurse, encounters ventilator-related trouble, they cannot pinpoint the exact source of trouble. This leads to delayed reporting of device-related troubles among front-line staff.

In Japan, the mean number of clinical engineers in the hospital is 20.6 (range, 2–70),36 and the mean number of clinical engineer enrollments per 100 beds is 3.1 (range, 0.6–8.5).36 The percentage of full-time clinical engineers with daytime and nighttime shifts is 39.7%, and that for only daytime shift is 21.7% in acute hospital sections, such as ICU and the emergency department.37 Our study revealed that 41.2% of ventilator-associated accidents occurred during the night/early morning. A shortage of clinical engineers among the nighttime staff might worsen ventilator trouble, as the staff might hesitate to contact the clinical engineers on call or there may be a delay in service because of the clinical engineer’s travel time from home to hospital. Although our study comprised only few cases, our findings indicate that the number of clinical engineers staffed during the daytime and night/early morning might be insufficient to maintain and provide a high level of intensive care.

Improvement in Intuitiveness and Design for Patient Safety

Human factors and systems engineering to improve patient safety have been demonstrated in several areas, such as medication safety, medical device design and usability, and reliability design.38 However, our findings suggest that medical devices are complicated and unintuitive for users. Medical device incidents are reportedly related to poor medical device design.13 Usability problems have the potential to create a high cognitive burden on nurses and increase the likelihood of mistakes.38 Thus, improvement in the design and usability of medical devices may reduce medical device incidents.

Limitations

Our study has some limitations. First, the number of cases was small because we only used 2 years of data. Opportunities to use medical devices are expanding; the functions of medical devices are becoming more diverse and complex, and circumstances surrounding medical care are dynamically changing. Thus, we studied data over a short term to collect analogous cases. Two previous studies indicated that organizational and personal barriers hindered reporting behavior.39–41 We do not know if the number of ventilator-associated accidents is underreported by hospital staff because we used secondary data from the JCQHC, which is partly based on hospital records. A study by Tomas et al26 reported that the proportion of ventilator-related incidents was 164 (16.1%) among 1021 incident reports. This percentage was similar to our study result, wherein the proportion of incident reporting data for ventilators was 11.8%. Second, although our study results show many incidents reported from general wards, there are no incident reports from the operation room. We collected medical device incident reports, but the JCQHC codes medical device incidents that occur during surgery as surgery incidents. Moreover, the JCQHC does not record information about whether patients are on home or hospital ventilators, or on how many patients were on ventilators during the survey period. Therefore, we cannot argue whether our results for the probability of ventilator-related accidents are reasonable relative to the number of ventilators used. Third, the staff and hospital administrators experience complex feelings and barriers to reporting incidents, the risk of litigation issues, which makes it difficult to collect incident reports and hospital information. Incident reporting research has focused on nurses, physicians, and pharmacologists, but there are few studies focusing on other hospital staff, including hospital administrative staff,42 nursery teachers, and care-certified staff.23 Fourth, this is a cross-sectional study; thus, we cannot investigate the causal relationship between the improvement of complicated specifications of ventilators and the actual prevention of errors. Because this study was based on the data collected by the JCQHC, it cannot be said whether these codes improved or if the accidents were preventable. Nevertheless, intuitive and robust designs could undoubtedly reduce device-related accidents.

CONCLUSIONS

Our study demonstrated that ventilator-associated accidents were related to an entanglement of complex factors, including not only human factors but also hardware, software, and the environment surrounding humans. Medical professionals had problems with the usage or management of the ventilators in 30 of the 34 cases; only 4 cases involved a problem with the ventilator itself. To ensure patient safety, it is necessary to educate and train medical staff and improve the design of medical devices.

Inadequate communication and responsiveness to alarms were important factors causing accidents. It is the hospital administration’s responsibility to set adequate rules and educate staff to prevent medical device accidents. In addition, when using a medical device that functions in a complicated manner and an error occurs, the user cannot easily understand the meaning of the error and cannot immediately address the error. Because system approaches are important to prevent errors caused by human factors, hospital administrators need to consider securing personnel to handle nighttime errors and adapting the system to friendly and intuitive designs for users.

REFERENCES 1. Jha AK, Prasopa-Plaizier N, Larizgoitia I, et al. Patient safety research: an overview of the global evidence. Qual Saf Health Care. 2010;19:42–47. 2. James JT. A new, evidence-based estimate of patient harms associated with hospital care. J Patient Saf. 2013;9:122–128. 3. Fujita S, Iida S, Nagai Y, et al. Estimation of the number of patient deaths recognized by a medical practitioner as caused by adverse events in hospitals in Japan: a cross-sectional stydy. Medicine. 2017;96:e8128. 4. Hunter JD. Ventilator associated pneumonia. BMJ. 2012;344:e3325. 5. Papazian L, Klompas M, Luyt CE. Ventilator-associated pneumonia in adults: a narrative review. Intensive Care Med. 2020;46:888–906. 6. Krishna Kumar B, Ravi M, Dinesh K, et al. Ventilator malfunction. J Anaesthesiol Clin Pharmacol. 2011;27:576. 7. Kamio T, Masamune K. Mechanical ventilation-related safety incidents in general care wards and ICU settings. Respir Care. 2018;63:1246–1252. 8. Pham JC, Williams TL, Sparnon EM, et al. Ventilator-related adverse events: a taxonomy and findings from 3 incident reporting systems. Respir Care. 2016;61:621–631. 9. Drews FA, Musters A, Samore MH. Error producing conditions in the intensive care unit. In: Henriksen K, Battles JB, Keyes MA, et al., eds. Advances in Patient Safety.  New Directions and Alternative Approaches. (Vol. 3: Performance and Tools). Rockville MD: Agency for Healthcare Research and Quality (US); 2008. 10. Coldewey B, Diruf A, Röhrig R, et al. Causes of use errors in ventilation devices—systematic review. Appl Ergon. 2022;98:103544. 11. World Health Organization. Patient safety incident reporting and learning systems: technical report and guidance. 2020. Available at: https://www.who.int/publications/i/item/9789240010338. Accessed April 26, 2022. 12. Pharmaceuticals and Medical Devices Agency. Safety measures work. Available at: https://www.pmda.go.jp/safety/info-services/0001.html. Accessed April 26, 2022. 13. Mattox E. Medical devices and patient safety. Crit Care Nurse. 2012;3:60–68. 15. EUR-Lex. Access to European Union law. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, amending Directive 2001/83/EC, Regulation (EC) No 178/2002 and Regulation (EC) No 1223/2009 and repealing Council Directives 90/385/EEC and 93/42/EEC. Available at: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32017R0745. Accessed August 26, 2022. 16. Ministry of Health, Labour and Welfare. 6th NDB Open Data Japan. Available at: https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/0000177221_00010.html. 2019. Accessed August 26, 2022. 17. Ministry of Health, Labour and Welfare. 3th NDB Open Data Japan. Available at: https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/0000177221_00002.html. 2016. Accessed August 26, 2022. 18. The Japan Times. Japan’s older population hits record high. 2021. Available at: https://www.japantimes.co.jp/news/2021/09/20/national/japans-older-population-hits-record-high/. Accessed April 26, 2022. 19. Ministry of Health, Labour and Welfare. Summary of Static/Dynamic Surveys of Medical Institutions and Hospital Report. 2017. Available at: https://www.mhlw.go.jp/english/database/db-hss/mih_report_2017.html. Accessed April 26, 2022. 20. Osaka Prefectural Pharmaceutical Affairs Council Medical Device Safety Measures Promotion Subcommittee. Ventilator safety measures learned from the case of near-miss. 2020. Available at: https://www.pref.osaka.lg.jp/attach/5743/00347914/kokyuuki_torimatomehonbun.pdf. Accessed April 26, 2022. 21. Japan Council for Quality HEALTH CARE. Outline. Available at: https://www.med-safe.jp/contents/english/index.html. Accessed April 26, 2022. 22. Elo S, Kyngäs H. The qualitative conten