Analysis of Hospital’s Emergency and Disaster Preparedness Using Hospital Safety Index in Indonesia

Abstract

Indonesia is country with abundant high-risk areas for various disasters that can affect both the structural and non-structural safety of various vital establishments, particularly hospitals. This present study aims to examine the level of the hospital safety index in nine hospitals in four provinces based on the guidelines from the WHO/PAHO (World Health Organization/Pan American Health Organization). The Hospital Safety Index (HSI) guidelines consist of four parameters that include the types of hazards, structural safety, non-structural safety, disasters and emergency managements. This study was a cross-sectional study on data obtained through interviews, focus group discussions (FGDs), observations, and document reviews to assess the parameters of the HSI. Data were calculated for the HSI score, and descriptive statistics and multiple correspondence analysis (MCA) were carried out. The SPSS software version 25.0 was used for the statistical analysis. Results show that the overall safety index was 0.673 (Level A), meaning that it is likely the hospital will maintain functionality in emergencies and disasters. By province, the level A index was identified in DKI Jakarta (0.76), Yogyakarta (0.709), and West Java (0.673), showing that hospitals in these provinces will maintain functionality in emergency and disaster situations; however, in North Sumatera, the index was categorized in B category (0.507), demonstrating that the hospital’s ability to function during and after emergencies and disasters is potentially at risk. The multiple correspondence analysis shows that the hospitals in the provinces of Yogyakarta and West Java tend to achieve similar categories in almost all assessment modules; therefore, control measures of preparedness should be considered, such as improvements in equipment and facilities; hospital emergency and disaster response and recovery planning; communication and information management; training; and relevant stakeholders awareness.

1. Introduction

Indonesia, as a country that has thousands of islands, is a disaster-prone country. The fact that the country has dense population of around 270,20 million people [1], and is located on a Ring of Fire, with more than 160 active volcanoes, has made this country prone to a high risk of disaster impacts [2]. Indonesia is also prone to seismic risks from the interactions between Pacific, Eurasian, and Australian faults [3]. Furthermore, the tropical climate, with its rainy and dry seasons, has also caused Indonesia to be vulnerable to various hydro-meteorological hazards that trigger floods, landslides, mud floods, extreme weather, abrasions, and drought-triggered forest fires [4]. These disasters could endanger people, infrastructures, and constructions. During 2021, it was reported that 2009 disasters occurred in Indonesia. The top disasters were flooding, typhoons, landslides, and forest fires. These disasters have accumulatively caused 709 people to die, 73 to go missing, and 13,088 injured, with a further 5,084,979 people being negatively affected and displaced. Furthermore, the disasters have also damaged 1542 public facilities, including 356 health facilities [5]. Additionally, Indonesia has also been experiencing the COVID-19 pandemic since early 2020. During the pandemic, hospitals are expected to remain operational during peak cases of COVID-19 and to increase the surge capacity [6]. Until 31 January 2022, there have been 4,353,370 cases of COVID-19, with 194,776 active cases, 4,140,454 recovered cases, and 144,320 deaths [7].

Hospitals play important roles when facing all kinds of disasters, especially in terms of providing health services for the affected population. Hospitals should be able to function during and after the occurrence of disaster [8]. When hospitals fail to function in disaster and emergency situations, whether for structural or functional reasons, they will not be available to treat the victims in a timely fashion when they are most needed. Thus, over the course of many years, hospital preparedness in dealing with disasters has become a concern, which is evident from its inclusion into the Hyogo Framework for Action 2005–2015 and Sendai Framework for Disaster Risk Reduction 2015–2030, which highlight the importance of safe and effective operations during and after disaster events [9,10]. In addition, the Sendai Framework for Disaster Risk Reduction 2015–2030 has recognized the importance of making hospitals safe from disasters by ensuring that all new hospitals are built with a level of resilience that strengthens their capacity to remain functional in disaster situations, and of implementing mitigation measures to reinforce existing health facilities, particularly those providing primary health care [10]. Furthermore, hospitals are an important part of healthcare infrastructure that should always remain safe and operational. Their continuous operations are particularly important in disaster management. In this case, a safe hospital is a facility whose services remain accessible and functioning at maximum capacity, and with the same infrastructure, before, during, and immediately after disasters. The continuing functionality of the hospital depends on a range of factors, including the safety of its buildings, critical systems, equipment, availability of supplies, and emergency and disaster management capacities of the hospital, especially in terms of its ability to respond to and recover from possible hazards or events [8]. When a hospital has the robustness, strength, and ability to recover from any disruption or disaster, it will contribute directly to the achievement of the Sustainable Development Goals.

Sunindijo et al. has conducted a study on 10 hospitals in West Java and 5 hospitals in Yogyakarta, and reported that, on average, the preparedness level of these hospitals to be well-functioning during and after disasters is still in the at-risk category. This definitely requires immediate intervention [11]. Another study conducted by Firdausi et al. in two public health centers in DKI Jakarta also reported that the facilities are still at risk in terms of facing disasters, as they are still at risk in terms of their structural and non-structural safety [12]. Public health centers provide primary level in-patient and out-patient services to the community, and they perform public health prevention and promotion measures for infectious diseases and non-infectious diseases [12]. Public health center facilities face various disaster risks in DKI Jakarta, including floods, blackouts, and earthquakes. In 2019, electricity blackouts lasted for 3 days when the electricity backup was only adequate for 1–2 days [12]. Moreover, in the Mamuju District, West Sulawesi, the 6.2 SR earthquake in 2021 caused the Mitra Manakarra hospital to collapse. As a result, the Local Disaster Management Agency of the district was forced to build emergency tents and run a field hospital on the hospital yard to substitute the hospital functions in providing services to patients [13]. The COVID-19 pandemic has also become a new challenge for Indonesia’s healthcare facilities, including hospitals, due to the rapid increase of confirmed cases day by day [14,15], resulting in difficulties in referring patients to hospitals because the hospitals are full of COVID-19 patients [16]. When a hospital, which is considered as an important asset for disaster management, experiences faults or structural damages, the necessary health service provision will be disrupted and will directly affect the safety and security of the affected population [8].

The Hospital Safety Index is published by the Pan American Health Organization (PAHO) and the World Health Organization (WHO) for evaluating Hospital and Healthcare facility readiness in facing emergencies and disasters. This index has been used substantially in many countries [13,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34]. There are three aspects measured in the index: structural, non-structural, and emergency and disaster management. Not only does the index predict the operational capacity of the hospital during and after disaster, but the result of HSI can also be used as a reference for hospital to strengthen planning and improve implementation in the future [8]. The hospital safety index is presented as a function of several parameters that characterize the different types of disaster hazards and risks: hazard, vulnerability, and exposition [19,20]. An evaluation of hospital emergency and disaster preparedness is deemed significant in order to identify the gaps within the hospital emergency and disaster plan, as well as to ensure the effectiveness of hospital functional capacity when disasters occur [20,22,23,24]. Many researchers and government authorities have created hospital emergency and disaster evaluation methods and checklists [20,21,22,23,24,25,26,27,28,29,30]; however, there is no definitive solution in terms of a reliable and valid instrument that can evaluate hospital emergency and disaster preparedness [20,21,22,23,24,25,26,27,28,29,30]. Indeed, many nations prefer to use their own specific instruments in evaluating their hospital emergency and disaster readiness [20,21,22,23,24,25,26,27,28,29,30]. Mojtahedi et al. (2021) [31] have developed a new way for assessing hospital disaster preparedness using the TOP-SIS method, where human resources; finance; logistics and evacuation; emergency and disaster management coordination; patient care and support services; response and disaster recovery planning; decontamination; security; and communication and information management are used as the parameters to be evaluated. On the other hand, Yazdani et al. (2021) [35] focus more on assessing patient evacuation and disaster readiness planning. Furthermore, Aghapour (2019) [36] emphasizes the significance of the increasing hospital surge capacity for disaster management, which is also relevant to the COVID-19 situation [6]. Recently, a catastrophic fire occurred in a COVID-19 ward of the al-Hussain hospital in Southern Iraq, resulting in 92 fatalities and injuring more than 100 persons. The accident investigation findings reported that the hospital was not ready to handle a fire accident as there were inadequate quantity of hospital fire alarms and sprinklers, as well as a limited hospital emergency response plan and team [37,38].

It can be concluded that as a country that has been hit by various past disasters, Indonesia has experienced the destruction of healthcare facilities, including hospitals, that render them as being non-functional. These hospitals experience structural and non-structural damages, which causes them to collapse, thus rendering them non-functional. Based on these experiences, it is recognized that other hospitals in Indonesia may also be vulnerable to disasters; therefore, the implementation of a mitigation program is crucial to reduce the possibility of hospital damages during disasters in the future. To achieve this, an evaluation of hospital readiness when facing disasters or emergencies is necessary. Nevertheless, studies that comprehensively assess hospital preparedness in dealing with emergencies and disasters are scarce in Indonesia, despite the fact that Indonesia has selected the Hospital Safety Index (HSI) published by the WHO/PAHO to evaluate hospital emergency and disaster preparedness as a part of hospital national accreditation, known as the Indonesia National Hospital Accreditation Standard (SNARS) [39]. The objective of this study was to assess the level of hospital preparedness in Indonesia using the WHO/PAHO Hospital Safety Index. It is expected that the implementation of a reliable and validated international instrument such as HSI [13,34] will result in valuable findings that enable comparison of the readiness of Indonesian hospitals when facing emergencies and disasters.

2. Literature Review

2.1. Overview of Study Area

The Java and Sumatera Islands have been known as the areas with the highest number of extreme events and disasters [40]. The human health consequences derived from disasters has been identified in the highly disaster-prone areas on the two islands. Based on the frequency of extreme events and disasters, these areas include four provinces: the Special Capital Area of Jakarta (DKI Jakarta), West Java, Yogyakarta, and North Sumatera [40]. The role of hospital readiness to reduce the human health consequences from different kinds of extreme events and disasters is believed to be significant, particularly the readiness of the hospital disaster committee and its capacity to conduct evacuation in the hospital [35,36]. In this study, these four provinces in the Java and Sumatera Islands (DKI Jakarta, West Java, Yogyakarta, and North Sumatera) were selected as the study sites.

2.1.1. DKI Jakarta Province

The DKI Jakarta Province is prone to flooding due to its location, which is on the northwest coastal area of Java Island and includes the estuary of Ciliwung river in the Jakarta Bay. The average height of areas in Jakarta is 8 m above the sea level. Climate change, and rapid development of city areas in Jakarta, have contributed significantly to flooding. A study by the UNDRR highlighted that Jakarta will sink in the next 20 years, making it the fastest sinking city in the world [41]. This also means that the DKI Jakarta province is also prone to earthquakes. When the classification of risks is applied to this province (<13 categorized as low, 13–144 categorized as medium, >144 categorized as high), DKI Jakarta has a risk index score of 66.74, in other words, it is in a medium risk category, and is depicted with the amber color code on the disaster risk map, as seen in Figure 1 [8]. Other potential disasters that could occur in this province are extreme weather, extreme tide, or abrasion. Additionally, due to its status as the capital city of Indonesia, Jakarta is one of the provinces with the highest population density in the country, making it also prone to social conflicts caused by political, social, religious, and economic issues [42,43,44]. DKI Jakarta is also currently the province with the highest number of confirmed COVID-19 cases in the country, accounting for 48,264 cases as of 27 June 2021 [45].

2.1.2. West Java Province

The West Java province is known as the ‘disaster’s supermarket’ as this province has experienced all kinds of disaster. The West Java province is prone to earthquakes due to its position across several active faults such as the Cimandiri fault, Lembang fault, Baribis Kendeng fault, Garsela fault, and Ciremai fault. In the southern area of this province is the Java Megathrust zone, which triggered the Pangandaran Earthquake in 2006 [46]. Furthermore, West Java also has many active volcanoes, such as Mount Salak, Mount Gede, Mount Tangkuban Perahu, Mount Papandayan, Mount Guntur, Mount Galunggung, and Mount Ciremai, that may erupt at any time. Mount Tangkuban Perahu has erupted several times since 1829, with the most recent one being in August 2019 [46]. The West Java province has a risk index of 150.46, is categorized as having a high risk, and is represented by the red color code on the disaster risk map (Figure 2). Other disasters that may occur in West Java are tsunamis, flooding, landslides, droughts, extreme weather, forest fires, and extreme tides and abrasion [41]. The West Java province also has the second highest number of confirmed COVID-19 cases, accounting for 350,719 cases (as of 27 June 2021) in Indonesia [45].

2.1.3. Yogyakarta Special Regions Province

The Yogyakarta Special Regions Province, or DI Yogyakarta province, is prone to earthquakes and volcanic eruptions. The Opak fault that runs through this province triggered a significant earthquake in 2006 [41]. Another significant disaster that occurred in this area is the Mount Merapi eruption in 2010, which severely affected the surrounding area and community [47,48]. This province has a risk index score of 140.92, in other words, it is in a medium risk category, with an amber color code (Figure 3). Other disasters that could potentially occur in DI Yogyakarta are flooding, landslides, droughts, extreme tides/abrasion, forest fires, extreme weather, and tsunamis [41]. Currently, Yogyakarta is the 8th province with the highest number of confirmed COVID-19 cases in Indonesia, accounting for 53,978 (as of 27 June 2021) of the total cases [45].

2.1.4. North Sumatera Province

The North Sumatera province is prone to various natural disasters, especially earthquakes and volcanic eruptions. This area is located across four faults that can trigger earthquakes: Renun fault, Toru fault, Angkola fault, and Barumun fault [41]. Some parts of the North Sumatera area were included in the areas affected by the devastating Aceh earthquake and tsunami in 2004 [49]. This province is also the home of Mount Sinabung, which is located in the Karo District, that has been intermittently erupted since 2013 and causes many people to be displaced from their home and have not been able to return [41,50]. Based on the Indonesian Disaster Risk Index 2019, the North Sumatera Province receives a score of 145.18, and thus it is placed in the High-Risk category, with the red color code (Figure 4) [41]. Other disasters that frequently occur in North Sumatera are tsunamis, flooding, landslides, droughts, extreme weather, extreme tides and abrasion, and forest fires. In addition, as of 27 June 2021, North Sumatera is ranked 13th in terms of being one of the areas with the highest confirmed COVID-19 cases, accounting for 34,869 cases in the country [45].

2.2. Hospital Safety Index for Evaluating Hospital Preparedness

The Hospital preparedness was assessed using the Hospital Safety Index: Guide for Evaluators 2nd Edition in this study. This tool includes four modules: disaster hazard and risk; structural safety (18 items); non-structural safety (93 items); and emergency and disaster management (40 items). The index comprises various elements, including understanding the multi-hazards encountered by hospital facilities (module 1); structural safety that describes how the hospital’s structure ensures the safety of workers, patients, and visitors (module 2); non-structural safety that describes how the hospital’s facilities and utilities such as water, electricity, communication, and medical gases, remains operational during or after disasters (module 3); and disaster management that describes how the hospital manages different types of hazards, including the presence of a disaster management committee and hospital, and its relevant procedures for addressing all hazards encountered by the hospital (module 4). The Hospital Safety Index combines all information from all modules and indicators to present the whole picture of hospital disaster preparedness and resilience. Each module addresses different aspects as described below [8]:

2.2.1. Module 1 (Potential Hazard and Disaster)

This module is used to determine the hazards and geotechnical properties of soils at the hospital that may affect the safety of the hospital. Module 1 also identifies hazards that could lead to disaster or emergency situations, in which the hospital should be prepared for dealing with those events. Some potential hazards classified in this module include geological, hydro-meteorological, biological, and human-made hazards (technological and societal hazards). Data required for this module include the geographical condition of the site, population information, hazard map, and history of disasters at the hospital. Module 1 is not included in the Hospital Safety Index (HSI) calculation as per guidance from the WHO; however, this module describes various potential hazards in the hospital (as we describe in Section 2.1 and Section 4.1 of this article) by providing detailed explanations of the hazards, including the natural hazards, non-natural hazards, and man-made hazards faced by the hospital. This information is important initial information for assessing whether the hospital addresses all hazards identified in this module. Population information includes the population size and density in the hospital area, and the disaster vulnerability of the population. This information is not provided in this article but is normally recorded during the hospital safety index evaluation process for hospital accreditation and for the development of the district’s disaster contingency plan.

2.2.2. Module 2 (Structural Safety)

The structural elements addressed in this module include columns, beams, walls, foundations and other elements that form parts of the building’s load-bearing system. The structural safety can be determined by assessing the building’s condition, structure plan, and prior events affecting the safety of the building.

2.2.3. Module 3 (Non-Structural Safety)

The non-structural elements assessed in this module consist of architectural elements; infrastructure protection; emergency access; critical systems including electrical systems, telecommunication systems, water supply systems, fire protection systems, waste management systems, fuel storage systems, medical gas, and HVAC system; office and laboratory equipment; and medical equipment and supplies.

2.2.4. Module 4 (Emergency and Disaster Management)

This module assesses the preparedness level of the hospital’s organization and personnel in response to emergencies or disasters. The elements included in module 4 consist of disaster responses and recovery plans; human resources; communication and information management; and resource availability including medicine, supplies, and equipment.

3. Materials and Methods

3.1. Data Collection and Sampling

The sampling method used in this study was purposive sampling. The hospitals were selected based on the purpose of the study. The inclusion criteria for the participating hospitals are: to have not been assessed for the HSI level; to be a Type B Hospital (provincial hospital) or Type C Hospital (district/city hospital); and for the hospital to be willing to participate in the study. There was a total of nine sample hospitals in four provinces: two hospitals in DKI Jakarta, three hospitals in West java, three hospitals in DI Yogyakarta, and two hospitals in North Sumatera. DKI Jakarta, which is the capital city of Indonesia with a disaster risk index score of 64.02 (moderate), is prone to floods, landslides, and other hydro-meteorological disasters, whereas West Java with a disaster risk index score of 145.81 (high) is prone to earthquakes, tsunamis, volcanic eruptions, and floods. DI Yogyakarta, with a disaster risk index score of 140.92 (moderate), is prone to earthquakes, volcanic eruptions, floods, and landslides. North Sumatera, with a disaster risk index score of 145.18 (high), is prone to earthquakes, floods, and tornados [4].

The ‘mixed method’ of using qualitative and quantitative methods was applied in this study. Qualitative data were collected from key informants, who were the people in charge of their respective units, through in-depth interviews according to the HIS tools, as well as through direct observations of the hospitals, and reviews of related documents to assess the parameters of the types of hazards, structural safety, non-structural safety, and disaster and emergency management. The quantitative data were obtained by determining the hospital preparedness level based on the Hospital Safety Index: Guide for Evaluators 2nd Edition. According to the WHO, the HSI is a rapid, reliable, and low-cost diagnostic tool that addresses the structural safety, non-structural safety, and functional capacity of a hospital. This is a standardized tool to measure hospital preparedness through an assessment [51].

Although the HSI does not replace detailed vulnerability studies, it provides the decision makers with an overall idea of the hospitals’ ability to respond to major emergencies and disasters. In other words, an assessment using the HSI is the first step toward prioritizing a country’s investments in hospital safety. This helps the decision makers to prioritize resource allocation [52].

Accreditation status may reflect the hospital preparedness, because since January 2018, self-assessment for hospital disaster preparedness using the WHO’s HSI has become one of the components of the hospital accreditation in Indonesia [32].

This study was conducted during the period of April to October 2021. Selected key informants included 3 groups of participants (i.e., hospital disaster team members, building facility service personnel, and public health officers) working at the selected hospital with minimum length of work in the hospital of 1 year. Data were collected by 8 evaluators whose backgrounds were in public health (2 evaluators), nursing (2 evaluators), disaster management (2 evaluators), and civil engineering (2 evaluators).

3.2. Data Analysis, Weighting, and Score Calculation

There were two data analysis methods applied in this study. The first method was based on the HSI Guideline to determine the hospital disaster preparedness level. Each item in the HSI was given 1, 0.5, or 0 score, which indicated high, average, or low, respectively. The score from each module, namely, structural safety, non-structural safety, and emergency and disaster management modules were obtained by dividing the total score with the number of items.

The weighting for each module was determined as per the HSI Guideline. When there is a higher risk of earthquake, it is recommended by the guidelines that the structural safety should have a weighted value of 50%, whereas the non-structural safety and emergency and disaster management are weighted 30% and 20%, respectively. After multiplying the weighted value with scores of each module, the final scores of each module were then summed and the safety index was obtained.

The safety index obtained was then analyzed using the univariate analysis to describe the disaster preparedness level of the hospital. According to the HSI guidelines, the safety index score can be categorized into three classifications, which is given as follows:

• Category A (hospital safety index of 0.66–1.00)

The hospital is likely to function during emergency and disaster situations; however, it is recommended to continue measures to improve the emergency and disaster management capacity, and to carry out medium- and long-term measures to improve the hospital safety level for potential emergencies and disasters.

• Category B (hospital safety index is 0.36–0.65)

The hospital’s current level of safety and emergency and disaster management is such that the safety of patients and hospital staff, as well as the hospital’s ability to function during and after emergencies and disasters, are potentially at risk; therefore, short-term intervention measures are needed.

• Category C (hospital safety index is 0.00–0.35)

The hospital is unlikely to function during and after emergencies and disasters, and the current levels of hospital safety and emergency and disaster management are inadequate to protect the lives of patients and hospital staff during and after emergencies or disasters; therefore, urgent interventions are needed.

A Multiple Correspondence Analysis (MCA) was also performed in the study to map the correlations between the hospital safety index (dimension 1) and case provinces (dimension 2). The analysis was used to identify and cluster provinces in Indonesia that had the tendency to have the same problems regarding the disaster management variables in the Hospital Safety Index. The MCA was performed using the SPSS Software version 25.0.

4. Results

4.1. Potential Hazards

A common potential hazard for natural disasters and emergencies within the study locations are geological and hydro-meteorological hazards. Earthquakes, volcanic eruptions, and floods are major threats in all study locations. The DI Yogyakarta and West Java provinces had experienced earthquakes that brought significant damages to the infrastructure of the provinces, including hospitals [53]. Additionally, West Java and DI Yogyakarta provinces also experienced volcanic eruptions and severe landslides [54,55]. A similar situation was seen in the North Sumatera Province, with potential floods, volcanic eruptions, and landslides. The DKI Jakarta province, as the capital of Indonesia, also faced similar potential hazards. This is concerning as this province has a very dense population and many high-rise buildings. Several hospitals in the DKI Jakarta province have identified and assessed possible hazards in the hospital environment using the Hazard Vulnerability Assessment (HVA) method. The results demonstrated that the potential incidents in these hospitals are fire as the major technological incident, infant abduction as the most prominent human-caused incident, dengue hemorrhagic fever or DHF as the major disease event, and flooding as the most frequent natural disaster. Another threat is the high intensity of rainfall which can cause flooding, which explains the history of annual flooding in DKI Jakarta. Major electrical blackouts are also a threat in Java Island. During the major electrical blackout in 2019, the DKI Jakarta, West Java, and DI Yogyakarta provinces experienced a prolonged blackout that took 2–3 days to resolve [56]. The lessons learned from this electrical blackout emphasize the need for the hospitals to be prepared for future prolonged electrical blackouts. Hospitals need to be able to operate for up to 3 days without electricity by having electrical back-ups, such as by storing enough fuel for diesel-fueled electricity generators and installing electrical solar panels so the hospital can remain operational during a blackout.

4.2. Structural Safety Assessment

The assessment of the Hospital building structural safety was based on the prior events affecting the hospital safety and building integrity. Below,

Table 1. Structural Safety Assessment.

No	Item	Max Score	Average Score	Ratio
1	Prior events affecting hospital safety	3	2.125	0.708
2	Building integrity	15	9.438	0.629
Total Score		18	11.563	0.642

presents the dimensions of module 2 on structural safety assessment. The maximum score indicates the number of items within that dimension, since the maximum score for each item is 1. The average score is the average score of the 9 hospitals studied. Finally, the ratio was derived from dividing the average score with the maximum score.

Table 1 shows that the total score of the Hospital Structural Safety element in this study was 0.642, or in the “B” category. This means that the structural safety of the buildings in all hospitals was at risk when dealing with disasters. According to guidelines from the WHO, the “B” category means that all the buildings in these hospitals need to receive short-term interventions to improve the structural safety of the hospitals.

4.3. Non-Structural Safety Assessment

The non-structural safety module assesses the architectural safety, infrastructure protection, access and physical security, critical system, and equipment and supplies.

The results of the non-structural safety assessment of the sampled hospitals are listed below.

From

Table 2. Non-Structural Safety Assessment.

No	Item	Max Score	Average Score	Ratio
1	Architectural safety	15	10.688	0.713
2	Infrastructure protection, access, andphysical security	4	3.000	0.750
	Critical systems
	a. Electrical system	10	6.813	0.681
	b. Telecommunication system	8	5.750	0.719
	c. Water supply system	6	3.875	0.646
	d. Fire protection system	5	3.000	0.600
	e. Waste management system	5	4.500	0.900
	f. Fuel storage system	5	3.125	0.625
	g. Medical gas system	6	4.438	0.740
	h. Heating, ventilation, and air conditioning	8	6.000	0.750
3	Equipment and supplies
	a. Office and storeroom furnishings and equipment	2	0.938	0.469
	b. Medical and laboratory equipment and supplies     used for diagnosis and treatment	19	13.750	0.724
Total Score		93	65.875	0.732

, it is apparent that the total ratio of all hospitals in this study was 0.732, or in the “A” category. This means that in terms of the non-structural safety, these hospitals will be able to remain functional in emergency situations and disasters.

4.4. Disaster and Emergency Management

The disaster and emergency management of the hospitals were assessed based on the coordination of emergency and disaster management activities; hospital emergency and disaster response and recovery planning; communication and information management; human resources; logistics and finance; patient care and support services; evacuation; decontamination; and security. Below,

Table 3. Disaster and Emergency Management.

No	Item	Max Score	Average Score	Ratio
1.	Coordination of emergency and disaster management activities	8	4938	0.617
2.	Hospital emergency and disaster response and recovery planning	5	2688	0.538
3.	Communication and information management	4	2375	0.594
4.	Human resources	5	2938	0.588
5.	Logistics and finance	4	2938	0.734
6.	Patient care and support services	9	6375	0.708
7.	Evacuation, decontamination, and security	5	3563	0.713
	Total	40	25,813	0.645

presents the results of the disaster and emergency management assessment in sampled provinces.

The overall score for the disaster and emergency management of the hospitals in all provinces was 0.645 or in the “B” category. Thus, all sampled hospitals still carry risks when dealing with disasters in this variable.

4.5. Overall Hospital Safety Index (HSI)

Table 4. Overall Hospital Safety Index.

No	Module	Score	Index
1	Structural Safety	11.563	0.642
2	Non-structural Safety	65.875	0.732
3	Disaster and Emergency Management	25.813	0.645
Overall Safety Index			0.673

presents the overall HSI score obtained from averaging all scores from the structural safety, non-structural safety, and emergency and disaster management aspects.

Based on Table 4 (shown above), the overall safety index for all samples in all provinces was 0.673, or in the “A” category. This indicates that the hospitals are likely to be functional in emergencies and disasters. It is recommended, however, to continue measures to improve the emergency and disaster management capacity and to carry out medium- and long-term measures to improve hospital safety levels when dealing with emergencies and disasters.

4.6. Comparison of HSI Scores in All Provinces

To gain an overview on the differences and similarities between sampled hospitals in this study, a comparison was made of the average score and ratio of each study area. The description and comparison of each province are presented in Supplementary S1. Results showed that almost all hospitals were in either the A or B category. There were only few aspects of the assessment that prompted some hospitals to be assigned to the C category, such as office and storeroom furnishing and equipment in hospitals in the DI Yogyakarta province, and some aspects on disaster and emergency management in hospitals in the North Sumatera province. The overall results of the hospital safety index assessment showed that 3 provinces, DKI Jakarta, DI Yogyakarta, and West Java, were in the A category, and only 1 province, North Sumatera, was in the B category.

5. Discussion

5.1. Potential Hazards

The potential for natural disasters and emergencies in the study areas tends to be similar, with geological and hydro-meteorological hazards being the most prominent hazards in these areas. Earthquakes, volcanic eruptions, and floods are the main threats for the entire areas of the four provinces. Historically, the DI Yogyakarta and West Java provinces have experienced earthquakes that cause damage to buildings, including hospitals [53]. In addition to earthquakes, the DI Yogyakarta and West Java provinces have also experienced volcanic eruptions and deadly landslides [54,55]. Similar to DIY Yogyakarta and West Java, the North Sumatera province also has the potential for volcanic eruptions, floods, and landslides. As a result of an earthquake on Nias Island, North Sumatera, on 28 March 2005, several hospitals experienced structural and physical damage which directly affected health care provisions [56]. The volcanic eruption of Mount Sinabung of the Karo District, North Sumatera also resulted disruptions of health service system in the district [57]. Geologically, the potential risks triggering emergencies and disasters in the DKI Jakarta area are no different than those associated with other provinces; however, as the capital city of Indonesia, the population density brings other threats brought about by social challenges. Hospitals in DKI Jakarta have identified and assessed existing hazards in the hospital environment using the Hazard Vulnerability Assessment (HVA) method. Based on this assessment, the potential incidents in this hospital are fire as the most prominent technological incident, infant abduction as the most prominent human-caused incident, dengue hemorrhagic fever or DHF as the most prominent disease incident, flood as the most prominent natural disaster, and man-made actions that change natural environment as the most prominent socio-natural disaster. High rainfall has also impacted the operations of several hospitals in Jakarta. In 2020, one of the hospitals located in DKI Jakarta, which is also the national referral hospital, was affected by flooding, causing the hospital equipment in the radiology unit to be submerged in water [58]. Thus, hazard identification and mapping are greatly required. A study by Mulyasari et al. in several hospitals in Japan demonstrated that the hazard mapping of areas around hospitals is necessary, especially for disaster-prone areas [59].

5.2. Structural Safety

Structural safety involves the assessment of items, such as prior events affecting hospital safety and building integrity. The structural safety is a crucial factor in Indonesia as the country is located in a hazardous zone where approximately 90% of the global earthquakes take place [60]. Furthermore, Indonesia is also home to 15% of the world’s most active volcanoes, making volcanic eruptions, earthquakes, and tsunamis more frequent in this country [30,60]. Indeed, this high-risk condition affects the structural safety element of vital buildings such as hospitals, and the HSI has become the most vital parameter for ensuring hospital preparedness in Indonesia when facing various types of emergencies and disasters. There are many hospital facilities that were damaged during the earthquakes in Lombok and Palu, which was attributed to the noncompliance of building structural safety [61,62]. A study conducted in two European countries, Serbia and Croatia, showed that the most challenging and common drawbacks encountered by hospitals are related to building structural safety. Based on this study, it is known that many hospital buildings are composed of old structures [63].

From the results of the assessment carried out in hospitals in the DKI Jakarta and DI Yogyakarta provinces, the hospitals in these areas are in category A (the hospital safety index is 0.66–1.00). Moreover, hospitals in the West Java and North Sumatera provinces are in category B (the hospital safety index is 0.36–0.65) in term of structural safety.

The classification of hospitals in the DI Yogyakarta and DKI Jakarta provinces into category A indicates that these hospitals are built with proper design and construction, including the use of the universal design principles and the standardization of building materials; retrofitting and rebuilding; nurturing a culture of maintenance; and taking into account the results of the economic, social, structural, technological, and environmental impact assessments. Structural safety is very crucial for hospital resilience. When a disaster strikes and the hospital building remain safe, they can provide healthcare for the victims. Hospitals that remain safe, resilient, and sustainable will contribute to the sustainability of health care. Hence, safe hospitals contribute to the achievement of sustainable cities and communities (SDG 11). The structural safety issues found in this study mainly occurred because the hospital buildings were old, given that they were built in the 1990s. In addition, when the DI Yogyakarta province experienced a major disaster in 2006, several hospitals experienced severe structural damages; therefore, the structural safety aspect is required, especially for measuring the building’s resistance when dealing with earthquakes. This aligns with a study that took place in a hospital in Japan which reported that hospitals located in earthquake-resistant buildings can deliver better medical services to their beneficiaries [59]. Furthermore, the age of the hospital buildings also affects aspects of its structural safety. Research conducted in two European countries, Serbia and Croatia, revealed that the problem frequently faced by health services in both countries is that infrastructure issues influence structural safety. This is attributed, in the study, to the old buildings of the hospitals [63].

Results showed that for several aspects within the structural safety module, a “C” category is obtained, particularly in North Sumatera’s hospitals for the sub-item of prior major structural damage or failure of the hospital building(s). This means that the buildings have experienced prior major structural damage with no repairs. It is essential that these hospitals are improved immediately in terms of its structural safety, since the current condition may endanger the safety of workers, patients, and the public. Furthermore, results suggested that in North Sumatera’s hospitals, several aspects of building integrity are at risk with the presence of flaking rust, cracks larger than 3 mm (concrete), and excessive deformations (steel and wood), leading to these hospitals being classified in the “C” category. In order to protect the hospital from poor building integrity, the hospital should urgently carry out repairs. Hospitals in DI Yogyakarta were also categorized in the “C” category for building integrity due to the proximity of buildings, which could create a wind tunnel effect and rapid spread of fire.

5.3. Non-Structural Safety

In terms of non-structural safety, there are four elements that have been assessed according to the Hospital Safety Index tools. These elements are architectural safety, infrastructure protection, access and physical security, critical systems, equipment, and supplies. Hospitals in three provinces, DKI Jakarta, DI Yogyakarta and West Java, are in category “A”, whereas hospitals in the North Sumatera province are in category B. The lower scores for non-structural safety in all provinces are obtained for the electrical system, architectural safety, water supply system, fire protection system, fuel storage system and office storeroom furnishing and equipment. This is similar to the results of a study conducted in a hospital in Sweden, which reported that the critical system has a lower score compared with other aspects concerning non-structural safety [64]. A cross-sectional study conducted on healthcare facilities in Malaysia on the non-structural safety assessment revealed that non-structural safety is considered to be a very important parameter and required urgent corrections, particularly for medical equipment, as well as laboratory and medical supplies for clinical diagnostic and medical treatment [65].

Non-structural safety also includes fire safety management within the hospital. It is essential to control fire hazards in locations with very dense populations through the establishment of a fire safety regulatory system at the city level [66]. Due to drawbacks regarding compliance with the fire safety system, a tragic fire accident occurred in a COVID-19 ward in the al-Hussain hospital in Southern Iraq [37,38]. The fire incident investigation report revealed that the hospital was not prepared for a fire accident, as indicated by the lack of fire alarms and sprinklers, an inadequate hospital emergency and disaster response plan, and an inadequate hospital emergency response team [37,38]. This tragic fire incident suggests the significance of an HSI evaluation and the identification of any gaps to be improved in order to protect the safety of patients, workers, and visitors. The results of this study support the result of a study of hospitals in Malaysia. A cross-sectional survey conducted on health services in Malaysia demonstrated that in the non-structural assessment of preparedness, the non-structural safety aspect is found to be in the very critical category and immediate improvements are needed, especially for medical laboratory equipment and supplies for diagnosis and treatment [65]. Regarding medical laboratory equipment, a similar study in Japan also suggests that medicine management implementation is the crucial non-structural aspect of healthcare services [59].

For aspects categorized in the “C” category, hospitals in the DI Yogyakarta province should pay special attention to aspects of external electrical systems installed for hospital use, such as the alternate water supply being connected to the regular water supply. In DKI Jakarta, regular electrical system tests of alternate sources of electricity in critical areas are not conducted; hence it is categorized in the “C” category. Hospitals in North Sumatera are categorized in the “C” category for emergency maintenance and the restoration of standard and alternate communications systems; the fire protection system, including fire/smoke detection systems; fire suppression systems (automatic and manual); the water supply for fire suppression; emergency maintenance and restoration of the fire protection system; the fuel storage system; the condition and safety of above-ground fuel tanks and/or cylinders; emergency maintenance and restoration of fuel reserves; the medical system; the availability of alternative sources of medical gases; emergency maintenance and restoration of medical gas systems; the condition and safety of medical equipment in emergency care service units; and the condition and safety of medical equipment for obstetric emergencies and neonatal care. Hospitals in DI Yogyakarta are categorized as “C” for office and storeroom, in terms of the safety of shelving and shelf contents; the safety of computers; the printers’ office, storeroom furnishings, and equipment (fixed and movable); and the condition and safety of medical equipment and supplies for emergency care for burns.

5.4. Disaster and Emergency Management

The result of the disaster and emergency management assessment is known to obtain the lowest score when compared with the other two aspects. The result showed that hospitals in the DI Yogyakarta and North Sumatera provinces are categorized into category B, whereas the other two provinces, DKI Jakarta and West Java, are in category A. Major disasters that hit the DI Yogyakarta province in the last several years have increased the awareness of health care facilities in the province in terms of the need for improvements in preparing for disaster management in their facilities; however, further improvement is still very much required. Overall, there are similar issues experienced by hospitals in all study areas. These issues concern human resources, logistics, finance, hospital emergency and disaster responses, and recovery planning. With regard to the human resource issue, it is apparent that many hospitals have not had an emergency response plan and emergency response training scheme for hospital personnel. Personnel training is often an issue when implementing emergency and disaster management as demonstrated by a study conducted by Yari et al. on primary health care facilities in Iran. It was shown that the HSI index for training and training personnel aspect is considerably low in Iran [67]. Furthermore, the financial and logistical aspect is also inadequate during the implementation of the emergency and disaster preparedness program. A study by Djalali et al. reported that the hospitals in Sweden and Iran experience financial issues, which require special attention as it is one of the more important aspects for implementing hospital emergency and disaster management in both countries [68]. Additionally, communication and information, in hospitals in the North Sumatera and DI Yogyakarta provinces, are categorized as C and D, respectively. This is similar to the findings of a study of HSI in Serbia. In Serbia, it is reported that safety is poor due to inadequate telecommunication systems, a lack of alternative communications systems, and the inadequate condition and protection of the external and internal communications system [65]. With regard to functionality, the significance of hospital emergency and disaster committee organizations has been shown to be important in a previous study [69], because the hospital emergency and disaster committees enable the proper implementation of emergency and disaster management when an emergency or disaster occurs [34,67,70].

The elements that were classified as category C in hospitals in the North Sumatera province were the coordination of emergency and disaster management activities, such as the hospital emergency/disaster committee; committee member responsibilities and training; designated emergency and disaster management coordination; the preparedness program for strengthening emergency and disaster response and recovery; the Emergency Operations Center (EOC); and coordination mechanisms and cooperative arrangements with local emergency/disaster management agencies. Other elements that also need to be improved in terms of hospital emergency and disaster response and recovery planning are the hospital emergency or disaster response plan; hospital hazard-specific sub plans; procedures to activate and deactivate plans; hospital emergency and disaster response plan exercises; evaluation and corrective actions; and hospital recovery plans. Communication and information management is also categorized as C in North Sumatera, which includes aspects of emergency internal and external communication, the external stakeholder directory, and procedures for communicating with the public and media. Patient care and support services are classified as category C in North Sumatera, including the continuity of emergency and critical care services. In Yogyakarta, financial resources for emergencies and disasters are classified in category C.

5.5. Multiple Correspondence Analysis

The MCA was used in this study to further analyze the data and find the relationship among categorical variables between participating hospitals that includes two hospitals in the DKI Jakarta province, two hospitals in the DI Yogyakarta province, two hospitals in the West Java province, and two hospitals in the North Sumatera province. The model summary for the MCA analysis is presented in

Table 5. SPSS Outputs of MCA.

Dimension	Cronbach’s Alpha	Total Eigenvalue	Inertia	% of Variance
1	0.816	2.578	0.644	64.442
2	0.378	1.396	0.349	34.889
Total		3.973	0.993
Mean	0.662	1.987	0.497	49.665

, whereas the joint plot of category points is presented in Figure 5. The model summary shows that the two-dimensional space captures a representative 99.3% of the total variance in the data set, and each dimension yields an eigenvalue that is higher than one. Dimension 1 represents the level of HSI in the hospital, whereas dimension 2 represents the study provinces as the case.

The results of the multiple correspondence analysis showed that the hospitals in the DI Yogyakarta and West Java provinces tend to achieve similar categories in almost all assessment modules. In 2019, a study conducted on the HSI assessment of hospitals in the West Java and Yogyakarta provinces demonstrated that class C hospitals in Indonesia have a relatively low HSI index and require urgent improvement [11]. Based on the latest data in our study, the hospitals in the DI Yogyakarta and West Java provinces have already taken action to overcome the hospital safety issues. Different from the hospitals in other provinces, the hospitals in the North Sumatera province were classified as category B for non-structural safety. The DKI Jakarta province achieves the highest score, almost achieving a classification in the “A” category for the structural safety and disaster management modules. Overall, there are no significant differences in the results of the hospital HSI assessment in all provinces, except for the hospitals in North Sumatera that are categorized as B.

6. Conclusions

Hospitals play an essential role during and after emergency and disaster events, which thus emphasizes the need for hospitals to remain operational during these situations to provide health services to the affected community. Given the current COVID-19 situation, hospitals are also crucial for referring COVID-19 patients with severe symptoms and need intensive health care. In some areas, disasters triggered by natural hazards have occurred at the same time as the pandemic, meaning that maintaining the functionality of health care system, including in hospitals, to be even more critical. Hence, ensuring that hospitals remain operational during disasters is very important, meaning that it is crucial to conduct routine HSI level assessments. This study shows that the HSI score for hospitals in the Jakarta (0.766), West Java (0.673), and DI Yogyakarta (0.709) provinces are in category A (score between 0.66 and 1.00). This means that the hospital will likely to be able to provide services during and after disasters; however, the hospitals in the DI Yogyakarta province should improve their office and storeroom furnishing and equipment, which is currently still in category C. In contrast, the hospitals in North Sumatera are categorized in category B (score between 0.36 and 0.65) with a score of 0.507. Major and immediate improvements should be made on some items in terms of emergency and disaster management, which are still in category C, especially in terms of the coordination of emergency and disaster management activities, hospital emergency and disaster response and recovery planning, and communication and information management. Hospitals in category A will remain functional in emergencies and disasters. It is recommended, however, to continue measures to improve the emergency and disaster management capacity and to carry out medium- and long-term measures to improve the safety level in case of emergencies and disasters. Hospitals in category B still carry risks when dealing with disasters. It is recommended that all elements (structural, non-structure and emergency disaster management) in the hospital should be improved.

Findings from previous similar studies in Primary Health Care show that PHCs will be able to recover during disasters, but several services will be exposed to danger. The results suggested that there are several gaps that need urgent interventions to improve the structural safety of buildings; water supply systems; fuel storage; disaster committee organization; furniture and fittings; offices and storage equipment; and capacity of workers through a structured and systematic training framework for disaster readiness. The results from this study can be used for prioritizing budgets and resource allocation, cost planning, providing specific solutions for local and national government, and efforts to achieve disaster risk reduction [70].

The results of this paper are significant from a practical perspective because damages to hospitals during disasters will complicate relief measures due to a delay in treatment of trauma and other diseases. Damaged hospitals also cause secondary disasters by compromising the achievements of national and global health goals, and by creating political risks for the government [71].

The results are also significant from an economic perspective as the economic and social costs associated with the damages to hospitals during disasters will influence the economic impacts of a disaster or emergency in a country. Hospitals represent more than 70% of public spending on health in countries [8]. Studies on post major disasters found that damages to hospitals and health care facilities are estimated to be within a range of 61 to 214 days of government health spending on the entire population [72]. Population increase, climate change, and urbanization are bound to worsen natural disasters in the future; thus, resilient hospitals are paramount for Indonesia.

The Hospital Safety Index is an innovative way of offering a new perspective on the disaster readiness of hospitals in Indonesia. It is critical that hospitals continue to work during emergencies and disasters since people immediately go to the nearest hospital for medical assistance when emergencies occur, without considering whether the facilities might not be functional. Consequently, it is vital to identify the level of safety and functionality that a hospital will have if an emergency or disaster occurs. Hospital evaluations aim to identify elements that need improvement in a specific hospital or network of hospitals, and they also aim to prioritize interventions in hospitals, which, because of their type or location, are essential for reducing the mortality, morbidity, disability, and other social and economic costs associated with emergencies and disasters [8]. This is a crucial step to build recommendations to strengthen the resiliency of health care facilities in Indonesia, as it is a disaster-prone country.

There are several recommendations for improving hospital preparedness in Indonesia. First, immediate improvement should be made to the poor quality aspects in the C category, including office and storeroom furnishing and equipment; coordination of emergency and disaster management activities; hospital emergency and disaster response and recovery planning; and communication and information management. Second, more in-depth assessments and studies need to be done for every hospital in Indonesia, especially those located in disaster prone areas with a higher risk of disasters. Third, readiness training and raising awareness should be performed in every hospital in Indonesia.

This study has some limitations. Firstly, the study was conducted during the COVID-19 pandemic, and thus hospital approval was quite difficult to obtain; therefore, it was required to collaborate with the Ministry of Health and Hospital Organization when obtaining the permits for this study, not only in public hospitals, but also private hospitals. In addition, all data collection activities were done through online or virtual methods. Secondly, the sample size is small; hence, a bigger sample size by involving more hospitals from various provinces in the high-risk category for disasters is needed to representatively analyze the differences in the disaster preparedness level among hospitals in Indonesia by comparing the type of hospitals and their preparedness. Moreover, long-term research can be conducted by comparing the hospital preparedness and their real readiness when disaster occurs to assess the reliability of the hospital safety index tool. Thirdly, the study only focused on the elements of the Hospital Safety Index as published by the WHO; thus, other determinants contributing to the occurrence of disasters that affect hospitals were not examined. Further studies are required to investigate these determinant factors of hospital preparedness in Indonesia.