Maintenance Status and Availability of Public Automated External De brillators: Results of On-Site Inspection

Tae Youn Kim Dongguk University College of Medicine, Dongguk University Ilsan Hospital Yun-kyoung Jung Korean Association for Safe Communities Sun Hwa Yoon Korean Association for Safe Communities Sun Ju Kim Yonsei University Wonju College of Medicine Kyoung-chul Cha Yonsei University Wonju College of Medicine Woo Jin Jung Yonsei University Wonju College of Medicine Young Il Roh Yonsei University Wonju College of Medicine Soyeong Kim Yonsei University Wonju College of Medicine Sung Hwa Kim Yonsei University Wonju College of Medicine Dae Ryong Kang Yonsei University Wonju College of Medicine Sung Oh Hwang (  shwang@yonsei.ac.kr ) Yonsei University Wonju College of Medicine


Background
Immediate cardiopulmonary resuscitation (CPR) and rapid de brillation are the most important measures to save patients with cardiac arrest from ventricular brillation. 1 Installation of automated external de brillators (AEDs) in public places is an important way to provide rapid de brillation to patients with out-of-hospital cardiac arrest (OHCA).
By enabling the general public to perform de brillation, the survival rate of such patients can be increased. [2][3][4][5] It has been reported that the survival rate of patients with OHCA had improved in a community with public access de brillation (PAD) programs where AEDs were installed in public places as a way to increase the chance of survival. [6][7][8] Accordingly, the number of AEDs being installed in public places is increasing with the spread of the PAD programs.
However, AED devices might be prone to failure or malfunction. 9,10 Importance Good maintenance of AEDs is important to ensure reliable operation and reduce the risk of failure or malfunctioning of AEDs. The de brillators in medical facilities or ambulances are used frequently and regularly undergo maintenance checks made by experts. On the other hand, public AEDs may not be maintained e ciently since they are not used frequently or checked by experts. With the increase in the number of publicly installed AEDs increasing, communities need to pay attention to the maintenance of AEDs installed in their jurisdictions to ensure that they are always available and ready for use.

Goals of this investigation
In this context, it is necessary to assess the maintenance status and availability of AEDs in public places and determine strategies for good maintenance in the process of spreading (disseminating) PAD programs. However, there is little information on the maintenance and availability of AEDs installed in public places. Thus, the objective of this study was to investigate the management status and availability of AEDs installed in public places and provide basic data for establishing future management plans for AEDs.

Study design and setting
This was a longitudinal, descriptive, observational study using questionnaires and checklist for AED maintenance. Ethical approval was obtained from the Institutional Review Board of Wonju Christian Hospital, Yonsei University Wonju College of Medicine (Approval number: CR320033).
The government of the Republic of Korea has mandated the installation of AEDs for public housing with more than 500 households, subway stations, passenger terminals, casinos, racecourses, prisons, athletic elds with more than 5,000 seats, and government buildings. These AEDs are required to be registered with the government, and the equipment maintenance status is to be reported annually to the government. These AEDs have managers assigned to maintain the equipment and report on equipment status regularly. AEDs installed in places other than the mandatory installation sites must be registered with the government, but there is no obligation to report their maintenance status or appoint a manager for maintenance.

AEDs
AEDs installed in Seoul City from 2013 to 2017 were included in this study. Seoul City is the largest metropolitan city in South Korea with an area of 605 km 2 and a population of about 10 million. Only the AEDs installed in non-mandatory locations with nancial support from the Seoul government were included in this study. AEDs installed with self-funded resources were excluded from this study.

AED inspection
The Korean Association for Safe Communities (Seoul, Republic of Korea), a non-pro t organization registered with the Ministry of Public Administration and Security, South Korea, was commissioned by the Seoul Metropolitan Government to conduct the inspection and surveys of AEDs installed in Seoul. The AED inspection is conducted by inspectors annually at the sites where the AEDs are installed.
The inspection items for each AED were selected from the checklist recommended by the guidelines for AED placement and management published by the Korean Ministry of Health and Welfare. These selected inspection items were reviewed by an advisory committee composed of two emergency physicians and one specialist each from the AED manufacturing company, AED management company, and consumer protection group. To con rm the suitability of the developed checklist, a pilot inspection of 10 AEDs in different places was carried out. The nal version of the inspection checklist included 11 items consisting of ve items on the AED manager and accessibility (designation of AED manager, CPR and AED training of the manager, regular internal check, obstacles near AED, AED guide sign), four items on the maintenance status (operation status, battery status, accessibility and validity of electrodes), and two items on the availability and actual use of AED (web-only appendices). The de nition of a ready-to-use de brillator is a de brillator that has valid electrodes, has a good battery status, and is functioning normally. A 24-h available de brillator was de ned as a de brillator that has valid electrodes, has a good battery status, operates normally, and can be used for 24 h.
Persons with valid basic life support (BLS) certi cation were selected as inspectors. A manual that included the concept of PAD, related laws, CPR and AED usage, AED management, survey procedure, survey items and their de nitions, and how to input survey results was developed for training inspectors. The inspectors were trained with a 1-day course (7 hours of lectures and 1 hour of practice on how to inspect an AED).
Every year in November during the study period, the inspector visited the AED installation sites for inspection of the AEDs. The inspector checked each AED using a structured inspection checklist and surveyed the associated manager. After inspecting the AED, the inspector introduced the AED maintenance manual to the manager and allowed the manager to practice how to maintain the AED. Upon completion of the AED inspection, the inspector entered inspection and survey results into a spread sheet (Excel, Microsoft 2010) and sent it to the data management center which kept input data and performed quality control.

Statistical analysis
A retrospective descriptive time series analysis was conducted in this study. Nominal data were calculated as the percentage of the frequency of occurrence and compared using Chi-square or Fisher's exact test, as appropriate. The Cochran-Armitage trend test was used to assess temporal trends of the proportion of "yes" responses to the parameters. Ordinary least squares regression analysis was used to estimate linear temporal trends among variables for modeling and analyzing variables. All statistical analyses were conducted using R Statistical Software (version 3.6.3; R Foundation for Statistical Computing, Vienna, Austria).

Results
Through recruitment notices distributed via webforms (www.sa a.org) and emails, 100 persons with a valid certi cation for BLS were selected and trained to inspect PADs every year from 2013 to 2017. Among the 36,313 AEDs installed during the study period, 11,783 AEDs installed with self-funded resources were excluded. A total of 911 AEDs that could not be inspected were also excluded (737 for refusal of inspection; 174 for duplicate data, lost data, demolition, or in 2017, were inspected and included in the nal analysis ( Figure 1). The places where the de brillator was installed were residential settings (49%), multi-use facilities (12%), and schools (10.2%) ( Table 1).   In terms of availability, non-ready-to-use AEDs accounted for 15.4% of the total AEDs. The causes of non-ready-ready-touse status were invalid electrode (83.7%), AED malfunction (22.1%), and bad battery status (17.2%). AEDs with limited 24-h availability accounted for 44.1% of the total AEDs, and the causes were limited 24-h accessibility (75.3%), invalid electrode (29.3%), AED malfunction (7.8%), and bad battery status (6.5%) ( Table 4) ( Figure 2). The proportion of AEDs with limited 24-h availability according to installation location was the highest in residential settings (27.2%), followed by in multi-use facilities (18.4%), schools (16.3%), public buildings (16.3), and transportation facilities (Table 5).

Discussion
On observing public AEDs for a period of 5 years, we found that most AEDs had a relatively good maintenance status, with more than 97% of the AEDs operating normally. However, 15% of the AEDs were not ready for use, and invalid electrode was the most common cause for this. Further, 44% of the AEDs had limited 24-h use, and limited 24-h accessibility was the most common cause for this. Factors related to management and maintenance of AED and accessibility to AED were found to have improved over time. The proportion of valid electrodes decreased over time.
Only around 1% of the AEDs were actually used. This rate did not change during the study period.
For successful implementation of PADs, four essential elements are required, namely planned and practiced response, training of anticipated rescuers in CPR and use of an AED, link to the local emergency medical system (EMS), and a process for continuous quality improvement. 11 In the process of developing and implementing the PAD program, the government or community has paid attention mainly to the installation of AEDs and link to the EMS system through legislation or guidelines. 12,13 In addition, the AED must be maintained in a state of being ready-to-use for 24 h a day. AEDs need to be maintained and tested regularly in accordance with the applicable rules and regulations established by governmental authorities. However, the maintenance and management of AEDs may be the responsibility of the locations holding the AEDs, considering that the community or government may not be able to directly manage the maintenance of the AEDs. Although each country or community has legal provisions for AED registration and management, many public AEDs are not registered in the national registry system or their management status is often unknown. In the Swedish experience, a large proportion (43%) of AEDs was not registered in their registry because of the unawareness of the AED registry or di culty in registering although those AEDs had high functionality. 14 In a report assessing Canadian public AED registries, governance and administrative processes across registries were found to be irregular. Some registries do not use a standardized validation or quality surveillance process, which might result in the loss of important information on AED usability, including battery and electrode validity. 15 In the present study, a trained inspector checked the management and maintenance status of each AED with regard to manager, accessibility, and equipment and electrode status by annually visiting the installation site. To the best of our knowledge, this is the rst report on the management and maintenance status of AEDs by on-site inspection. In this study, the maintenance of the de brillator itself gradually improved during the observation period. However, a signi cant proportion (44%) of AEDs had limited 24-h availability and this proportion increased over time. In particular, 3 years after the inspection began, the percentage of electrodes that had passed their expiration date was found to have increased. The de brillator itself does not need a separate function check as it reports data by performing self-tests on its internal circuitry to ensure readiness. 16 Two important accessories, i.e., batteries and electrodes, are subjected to inspection during de brillator maintenance checks along with the de brillator equipment itself. Since the battery is installed in the de brillator, the charging status can be checked using the indicator. It is checked along with the de brillator operation status. On the other hand, since the electrode is separate from the de brillator, the validity of the electrode must be checked separately by its expiration date. Therefore, to ensure the working status of the electrode, the manager needs to be aware of the periodic replacement plan. In addition, in cases where the AED is installed with nongovernmental external nancial support, there is often no nancial plan for replacing de brillator accessories. In such cases, even if the AED manager or inspector nds a problem with the electrodes, the problem cannot be solved. In this respect, when purchasing a de brillator and installing it in a public space, it is necessary to establish a supply or nancial plan for maintaining its accessories along with an inspection plan.
The use of AEDs in public places is related to the number of cardiac arrests in the installation area, the willingness of witnesses to use AEDs, and 24-h availability of the AEDs. [17][18][19] AEDs are highly accessible during weekdays, but their accessibility declines in the evenings, including nighttime, and on weekends. This limitation in accessibility is associated with reduced use of AEDs. 20 As observed in our study, the 24-h availability was limited for AEDs installed in places that were not open for 24 h, such as multi-use facilities, schools, public buildings, and welfare facilities. In addition, we found that the proportion of limited 24-h availability was the highest in residential settings. The limited use of AEDs in a residential setting can be a major hindrance to the PAD program. Only approximately 1% of the AEDs were actually used. This low utilization rate might be associated with the low 24-h availability. Thus, when planning AED installation, it is necessary to consider whether the installation site is open for 24 h.

Limitations
This study has several limitations. Since the AEDs included in this study were not installed according to a systematic plan considering the incidence of cardiac arrest nor were they linked to the EMS, the results of this study cannot be applied to all public AEDs. Most de brillators were inspected repeatedly but some were replaced during the study period. Since the inspectors were recruited annually, the same inspectors did not check de brillators during the study period. In order to reduce the bias caused by the inspectors, the recruited inspectors were trained on the inspection method. The results of this study cannot be generalized to other countries because the AED implementation and maintenance regulations are based on the relevant laws of each country.
In conclusion, although the AEDs had a relatively good maintenance status, a signi cant proportion of public AEDs were not available for 24-h use. Invalid electrodes and limited 24-h accessibility were the main reasons that limited the 24-h availability of public AEDs. Community attention and initiatives are needed to increase the 24-h availability of public AEDs.

Declarations
Author contribution: SOH, SHY, and YJ conceived and designed the study, and obtained research funding. KC supervised the conduct of the study. SJK, WJC, YIR, and SK conducted the data collection and managed the data, including quality control. YJ undertook recruitment of AED installation agency. SHK and DRK provided statistical advice on study design and analyzed the data. TYK drafted the manuscript, and all authors contributed substantially to its revision. SOH takes responsibility for the paper as a whole. Figure 1 Flow chart showing the enrollment process for AEDs AED, automated external de brillator Figure 2 The proportion of non-available AED and their causes A. The proportion of non-available AED B. Causes of nonavailable AED AED, automated external de brillator

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. PADappendices.docx