Hospital Preparedness Assessment against COVID-19 Pandemic: A Case Study in Turkish Tertiary Healthcare Services

Hospitals play a critical role in providing essential health services to people in the healthcare system. Healthcare systems around the world have faced some issues in responding to patients with various disease severity levels. Nowadays, the world as a whole is combating a pandemic called COVID-19. )is pandemic causes an increase in the disease spread with fluctuated patient demand that may affect the hospitals’ capacity and overall functioning and risks rising based on hospital site, medical staff, patient, and healthcare process. To deal with the challenges of the COVID-19 pandemic, hospitals must have completed their preparations before these events occur. )erefore, this study proposes an integrated approach based on the decision-making concept with interval-valued spherical fuzzy sets (IVSFSs) to the hospital preparedness assessment problem. A technique for order preference by similarity to ideal solution (TOPSIS) extended with IVSFSs is used to rank hospitals from Turkish tertiary healthcare services. A checklist reported by the World Health Organization (WHO) is adapted to conform to Turkey’s COVID-19 pandemic management. Ninety-nine subcomponents of the adapted checklist under ten components are weighted by interval-valued spherical weighted arithmetic mean (IVSWAM) operator. )e hospitals in the problem are then ranked concerning these weighted subcomponents. With the proposed approach, a COVID-19 pandemic preparedness index is determined for the observed hospitals. In addition, a separate index based on each main component (component-based ranking) is determined. )ese indexes are vital indicators in determining in which aspects hospitals are ready and in what aspects hospitals are not prepared for pandemics.)e proposed approach can be adaptable and applied by national policymakers in assessing all hospitals of the country against the COVID-19 pandemic.


Introduction
Hospitals play a critical role in providing primary health care to people, especially in natural or human-made disasters. Pandemics can lead to an increasing spread of disease, with irregular and suddenly increasing patient demands that can affect hospitals' capacity and the overall functioning of the health system. To cope with the difficulty of such an epidemic disaster, hospitals must have completed their preparations before these events occur. Nowadays, an overwhelming majority of the world is fighting against an epidemic called COVID-19. e world faces the demand for infected patients who arrive at the hospitals heavily and irregularly.
As of April 22, 2020, similar to countries' fights against this pandemic throughout the world, Turkey continues its struggle. e mortality rate, the number of intensive care patients, and the number of intubated patients are better than the most developed countries in the world. e mortality rate was 2.4% as of April 22 [4]. Also, the total number of infected cases to the total number of cases examined is around 13%. Turkey is among the top six countries in the world in terms of the total number of tests and has reached this level quickly. All these statistical indicators show that Turkey is at a better level than the world average because of Turkey's early isolation decisions and the case of being prepared for such an event. erefore, hospitals are required to be ready for such pandemics.
For Europe, the World Health Organization Regional Office has published a checklist entitled "Hospital Readiness Checklist for COVID-19" to take in the context of a continuous hospital emergency preparedness process [5]. In this document, they have advised hospitals to manage the preparatory process under ten critical components as follows: surge capacity, infection prevention and control, case management, human resources, continuity of essential health services and patient care, surveillance: early warning and monitoring, communication, logistics and supply chain management including pharmaceuticals, laboratory services, and vital support services. Hospitals are facilities with complex processes, mostly connected to external support and supply lines [6]. Even in regular times, many hospitals operate at full capacity or close to maximum capacity. In epidemic conditions, with surge demand, hospitals may find it challenging to carry out their necessary functional activities, and capacity may no longer meet this demand [7]. Even a well-prepared hospital for disasters will have a hard time coping with the consequences of a COVID-19 pandemic. It is an effective hospital management policy that will reduce these difficulties to some extent. In this checklist report, it is highlighted that this effective hospital management will help (1) continuity of essential services, (2) well-coordinated implementation of priority action, (3) clear and accurate internal and external communication, (4) swift adaptation to increased demands, (5) effective use of scarce resources, and (6) safe environment for health workers [5].
Based on the World Health Organization's checklist report, this study aims to develop a ranking model for hospitals regarding their preparation for the COVID-19 pandemic. In this context, components and subcomponents in the checklist are adapted to conform to Turkey's pandemic management. e model consists of many components and has a multicriteria decision-making structure in the final decision expressed as "goal-alternative-criterion and subcriterion-performance rating of each alternative with respect to the criteria." erefore, in this study, a "multicriteria decision-making (MCDM)" based model was used to evaluate hospitals' level of preparedness against the COVID-19 pandemic. e model is based on the technique for order preference by similarity to ideal solution (TOPSIS) concept. It was first developed by Yoon and Hwang [8]. It represents the MCDM problem, aiming to find the best alternative considering the distances from the positive and the negative ideal solutions. In the scientific literature, several TOPSIS versions are suggested for selection, ranking, and sorting problems [9][10][11][12][13][14][15].
Since we do not have precise data for the evaluation of these components and their related subcomponents, linguistic expressions and corresponding fuzzy numbers were used in the evaluation. Fuzzy logic is a concept proposed by Zadeh to eliminate the disadvantages of classical 0-1 logic. Several extensions have been developed in the first version. In this study, spherical fuzzy sets, which are proposed by Kutlu Gündogdu and Kahraman [16] for the first time in the literature, are used together with the TOPSIS MCDM method. Kutlu Gündogdu and Kahraman [16] proposed spherical fuzzy sets, which reflect uncertainty and ambiguity in real-world decision problems better than classical fuzzy set theory. ey are mathematically based on a membership function on a spherical surface. ey independently describe the degree of membership, nonmembership, and hesitancy in a larger domain (the sum of these three values must be between 0 and 1). ey are considered as the integration of Pythagorean fuzzy sets and neutrosophic sets. ey eliminate some aspects of neutrosophic sets and Pythagorean fuzzy sets as follows [16][17][18]: they cannot permit the sum of membership, nonmembership, and hesitancy degrees to be larger than one, and they do not disregard an independent hesitancy, unlike Pythagorean fuzzy sets.
In the existing study, a particular branch of spherical sets, "interval-valued spherical fuzzy sets (IVSFSs)," is used with TOPSIS. Interval-valued fuzzy sets provide decision makers to model their decision-making problem via a fuzzy interval framework instead of a single point. By this aspect, it helps decision makers to clarify their judgments more accurately. Initially, TOPSIS is merged with IVSFSs to evaluate hospital preparedness levels against the COVID-19 pandemic. e experienced hospital directors perform linguistic evaluations of hospitals with respect to the adapted components. en, a final score is determined for each assessed hospital.
e proposed model is tested on three Turkish hospitals that provide health care under the tertiary step. With the obtained results from the application of this proposed approach, a pandemic preparedness index is determined for the observed hospitals. In addition, it is considered to make a ranking and calculate the index based on each main component (component-based ranking). ese indexes are vital indicators in determining in which aspects hospitals are ready and in what aspects hospitals are not prepared for pandemics.

Literature Review
e topic of hospital readiness evaluation against disasters is studied very limitedly in the literature. Scholars deal with this topic either as an MCDM structure or a conceptual/questionnaire-based structure. Nekoie-Moghadam et al. [19] presented a systematic literature review of tools/checklists used for the evaluation of hospital disaster evaluation. ey mentioned some relevant themes such as logistics, planning, human resources, triage, communication, command and control, structural and nonstructural preparedness, training, evacuation, recovery after a disaster, coordination, transportation, surge capacity, and safety [19][20][21]. Fifteen studies are included in the scope of this review. However, none of them concern with the MCDM concept. Some of the reviewed ones are reports suggested by the WHO as we adapted in this study and guidelines [19]. Most of them are based on a questionnaire-based assessment tool [22][23][24][25][26][27]. Apart from this type of disaster preparedness evaluation studies, there are MCDM-based papers in the literature. For example, Hosseini et al. [28] developed a hospital ranking model based on disaster preparedness using the TOPSIS multicriteria method. ey used four dimensions and twentyone indicators under these dimensions, as in the study of Mulyasari et al. [25], to assess hospitals. Eight hospitals in Iran were evaluated in terms of four dimensions. e results of their research showed that the structural and functional preparedness dimensions had the highest and lowest weights, respectively. However, in their study, no MCDM method is used to assign the importance of these dimensions. Instead, they were just determined via subjective judgment. MCDM was used in the ranking of hospitals. A second significant contribution is performed by Ortiz-Barrios et al. [29]. In that study, the disaster readiness of emergency departments is assessed with an analytical MCDM approach. e approach consists of the analytic hierarchy process (AHP), decisionmaking trial and evaluation laboratory (DEMATEL), and TOPSIS. Ortiz-Barrios et al.'s study [29] is different from Hosseini et al.'s study [28] in the following aspects: it applies a pairwise comparison between the dimensions and their subdimensions via AHP. ey also determined interrelations between criteria via DEMATEL. Finally, the ranking of Colombian emergency departments in terms of disaster readiness is carried out via TOPSIS. ere are also cross-sectional studies using statistical methodologies, Delphi, and similar tools [30,31]. For a broad literature review in disaster preparedness of hospitals, scholars can refer to the papers of Fallah-Aliabadi et al. [20], Verheul and Dückers [21], Alruwaili et al. [32], and Nekoie-Moghadam et al. [19]. Tabatabaei and Abbasi [33] performed a cross-sectional study in some Iranian hospitals to assess risks during disasters based on the hospital safety index. ey designed questionnaires for hospitals' disaster ability with 145 metrics in structural, functional, and nonstructural factors. is study is not a fully numerical or MCDM-based study. Otherwise, it benefits from a semiquantitative disaster preparedness assessment method. Similarly, Naser et al. [34] made a cross-sectional study to assess hospital disaster preparedness in South Yemen. e results of the study showed that hospitals had not reached an unacceptable level of readiness. Samsuddin et al. [35] performed a cross-sectional study for disaster preparedness attributes and the hospital's resilience in Malaysia. e results showed that human resources and training and the ability to adapt in a timely manner are ranked as the most critical attribute. Marzaleh et al. [36] proposed an approach using Delphi for hospital emergency room preparedness against radiation and nuclear incidents in Iran. ey mentioned 31 criteria under three main classes: staff, stuff, and structure (system). Results indicated that staff and stuff preparedness had the highest and lowest priority levels, respectively. Shabanikiya et al. [37] designed a Delphi-based tool for hospital preparedness for surge capacity during disasters and assessed 64 components in five categories and 13 subcategories.
In light of the studies mentioned above, many contributions are performed for hospital disaster preparedness. Most of the papers contribute to the literature by proposing the cross-sectional questionnaire-based frameworks that we discussed. On the other hand, conceptual-based models that suggest new attributes regarding hospital disaster readiness and review papers that provide a comprehensive overview of the topic are also widespread. Our brief reviews in this section show the importance of hospital disaster preparedness assessment from the viewpoint of MCDM through the literature. It can be observed that there exists a wide range of MCDM methods used in the literature applied to various areas. So far, however, there have been limited papers regarding the applications of such methods in hospital disaster preparedness. Moving from this point, we aim to develop an MCDM model supported with interval-valued spherical fuzzy set concept for hospital preparedness assessment against the COVID-19 pandemic. We developed our model for a newly occurred pandemic that spreads over the world in a short time. We followed and adapted the checklist of the WHO for the COVID-19 pandemic. Our proposed model specifically deals with hospital disaster readiness assessment and hospital ranking. Our study is different from similar contributions in the literature in the following aspects: (1) e suggested approach adapted the components and subcomponents of the WHO's checklist [5]. As this checklist is created considering all countries of the world's status, our model is based on a strong background. Our model is also adaptable for all countries and the other disasters except the COVID-19 pandemic. (2) We developed an MCDM-fuzzy integrated approach that includes TOPSIS and interval-valued spherical fuzzy sets. In determining hospital disaster preparedness's evaluation components/subcomponents, an interval-valued spherical weighted arithmetic mean (IVSWAM) operator is used. en, in the hospital ranking phase, interval-valued spherical fuzzy TOPSIS is applied. Spherical fuzzy sets eliminate some missing aspects of neutrosophic sets and Pythagorean fuzzy sets by not permitting the sum of membership, nonmembership, and hesitancy degrees to be larger than one and not disregarding an independent hesitancy, unlike Pythagorean fuzzy sets. In view of the TOPSIS MCDM method's characteristics and spherical fuzzy set concept either individually or an integrated style, our approach can handle the problem systematically and analytically. (3) Our approach is implemented in Turkish tertiary healthcare services. For implementation, three tertiary hospitals placed in the Eastern Black Sea Region of Turkey are selected. As an additional analysis, a secondary ranking of hospitals by major components Mathematical Problems in Engineering was studied (10 components, as stated in the WHO's adapted checklist) and an overall ranking was obtained. is analysis can help hospital decision makers and national policymakers determine in which aspects hospitals are ready and in what aspects hospitals are not prepared for pandemics.

Overview of Spherical and Interval-Valued Spherical
Fuzzy Sets. Spherical fuzzy sets are the integration of Pythagorean fuzzy sets and neutrosophic sets. ere are three membership degrees named membership, nonmembership, and hesitancy in intuitionistic and Pythagorean fuzzy sets. Membership functions in neutrosophic sets are defined under three pillars: truthiness, falsity, and indeterminacy membership. e sum of these three membership values can be between 0 and 3. In spherical fuzzy sets, while the squared sum of three parameters can be between 0 and 1, each of them can be defined between 0 and 1 independently. For detailed information on this type of fuzzy sets, readers can refer to Kutlu Gundogdu and Kahraman [16] who are the authors and first-time developers of this type of fuzzy set.
Unlike the previous TOPSIS studies extended with spherical fuzzy sets (mostly related to numerical examples to demonstrate the methodology), this study focuses on hospital readiness assessment against COVID-19. Although it has some application of hospital location analysis [46], 3D printer selection [16], and supplier selection [18], it has no application domain in public health, especially in the emerging COVID-19 pandemic. us, this work is the first of its kind in addressing (1) the use of IVSF-TOPSIS in the public health domain and (2) the application of compromise ranking MCDM under uncertainty in the management of COVID-19 pandemic as a global public health emergency. It also advances the evolving literature of COVID-19 by effectively adopting the comprehensive critical criteria for hospital readiness assessment, which may be set as guidelines for relevant policymakers and decision makers.
IVSFS is a special sub-branch of spherical fuzzy set, and an IVSFS S i of the universe of discourse U is defined as follows: (2) Some mathematical operations with IVSFSs are defined in the following formulas benefiting from Kutlu Gundogdu and Kahraman [16].
Let S 1 and S 2 be two different interval-valued spherical fuzzy numbers of the universe of discourse U.
To avoid any complexity in the calculation, a leaner formulation is designed as follows: let IVSWAM operator is defined as follows: Here, w i ϵ[0, 1]; n i�1 w i � 1. Score and accuracy functions in ranking spherical fuzzy numbers are defined as in equations (9) and (10):

Mathematical Problems in Engineering
Here, it should be noted that S 1 < S 2 if and only if Score(S 1 ) < Score(S 2 ) or Score(S 1 ) � Score(S 2 ) and Accuracy(S 1 ) < Accuracy(S 2 ).

Proposed Approach by Interval-Valued Spherical Fuzzy TOPSIS (IVSF-TOPSIS).
A payoff matrix is mandatory to construct the decision-making process for all MCDM problems. Since the problem that this study has dealt with is related to the assessment of hospitals' preparedness against the COVID-19 pandemic, we have designed this matrix whose elements include the values of all alternatives with respect to each criterion under IVSFSs. Let H � h 1 , h 2 , . . . , h m m ≥ 2 be a set of alternatives (for this study "hospitals"), D � d 1 , d 2 , . . . , d n be a set of criteria set (for this study "components or dimensions of hospital preparedness assessment against COVID-19 pandemic"), and w � w 1 , w 2 , . . . , w n be a set of criteria weights for this study "components' weights") that satisfy the conditions of 0 ≤ w j ≤ 1 and n j�1 w j � 1. Steps of the proposed approach by IVSF-TOPSIS are described as follows: Step 1: decision matrices are created. e construction of the importance weight vector carried out by each expert is also carried out in this step. In determining the ratings of experts (for this study, we refer to "hospital decision maker") regarding alternative hospitals with respect to the readiness components, the IVSFSs-based linguistic scale (adapted from [16]) given in Table 1 denote the rating of a hospital with respect to a component. X � D j (H i ) mxn refers to the decision matrix, which is defined as follows: Step 2: the decision matrices of each hospital decision maker, which are constructed under IVSFSs, are aggregated. Also, the aggregation regarding evaluations of hospital decision makers on component weight determination is performed in this step. We follow the aggregation procedure given in equation (8).
Step 3: the aggregated interval-valued spherical fuzzy decision matrix is converted into a weighted intervalvalued spherical fuzzy decision matrix using a multiplication operator which is given in equation (5).
Step 4: the weighted interval-valued spherical fuzzy decision matrix is defuzzified using an adaptable version of equation (9). It should be noted that the interval values of three membership functions are weighted interval-valued spherical fuzzy numbers, which are obtained in Step 3.
Step 5: the interval-valued spherical fuzzy positive ideal solution (IVSFPIS) and the interval-valued spherical fuzzy negative ideal solution (IVSFNIS) are calculated based on the score values obtained in Step 4. e formulas are given in equations (12) and (13) as follows: Step 6: the distances from IVSFPIS and IVSFNIS are calculated using equations (14) and (15) as follows: 6 Mathematical Problems in Engineering Step 7: the closeness ratio (CR) of IVSF-TOPSIS is computed using the following equation: Step 8: the last step prioritizes the ranking of each hospital to the descending order.
e demonstration flow of the proposed approach is shown in Figure 1. In Figure 1, a step-by-step flow is presented.

Preparation.
In the preparation step of the implementation, the study's main aims, the adapted checklist, components and subcomponents, healthcare decision makers, and characteristics of assessed hospitals are explained. As stated in the previous sections, the aims of the study are as follows: (1) to assess hospitals in terms of readiness against COVID-19 pandemic concerning ninetyeight subcomponents in total under ten components and (2) to determine an index value for each assessed hospital considering all components and according to each component individually.
us, an overall component-based index and an individual component-based index can be computed using the IVSF-TOPSIS procedure.
Regarding the adapted checklist, Figure 2 represents the components that the World Health Organization Regional Office for Europe published with the name of Hospital Readiness Checklist for COVID-19 in Copenhagen in 2020.
Under these ten components, there are several subcomponents. Each of them is explained in the WHO's checklist [5]. In the following, each subcomponent is described by highlighting its effects on hospital readiness against the COVID-19 disaster situation. For the first component of "surge capacity," eight subcomponents are determined as follows: D1.1: availability of maximum patient admission capacity (this component refers not only to the total number of beds but also to the availability of human resources, areas that can be converted to intensive care, mechanical ventilators, and other resources) D1.2: ability to use existing planning power and tools to predict the demand for hospital services in the time of the COVID-19 pandemic D1.3: ability to expand inpatient capacity of the hospital (in terms of physical area, staff, equipment, and process) D1.4: ability to identify potential gaps in providing health care by giving importance to intensive care (in cooperation with senior managers and neighboring hospitals) D1.5: possibility to create additional capacity by determining an alternative location for the treatment of noncritical patients (e.g., sending less urgent patients home) D1.6: availability of extra locations for conversion to healthcare units in coordination with the local authorities (e.g., hotels, schools, community centers, and gyms) D1.7: cancellation flexibility of noncritical medical services (e.g., elective surgery) if necessary D1.8: flexibility in adapting admissions and discharge criteria and possibility in prioritization of patients and clinical interventions according to available treatment capacity and demand Regarding the second component of "infection prevention and control," fifteen subcomponents are determined as suitable to assess the readiness of hospitals. ey are depicted as follows: D2.1: availability of verbal instructions, informational posters, cards, hand hygiene stations (water, soap, paper towel, and alcohol hand rub), and waste bins at strategic locations across the hospital to provide healthcare workers, patients, and visitors' awareness Table 1: e linguistic scale used in the assessment.

Linguistic term
Interval-valued spherical fuzzy number    4: ability to designate a special waiting and examination area for individuals applying with COVID-19 symptoms depending on the algorithm of the Republic of Turkey Ministry of Health D3.5: ability to create additional areas for triage of patients in tents set up outside the hospital D3.6: appointment of a triage supervisor authorized for the whole triage process D3.7: ability to create a triage algorithm for the detection of acute respiratory infection cases D3.8: applicability of standard and droplet measures at all times D3.9: applicability of the hospital strategy for admission, internal transfer, referral, and discharge of patients with severe acute respiratory infections (ARIs) D3.10: consideration of home care for mild cases of COVID-19 ARI in patients with no comorbidities D3.11: consideration of hospital admission for cases of COVID-19 ARI in patients with comorbidities D3.12: availability of staffed beds for the admission of severe COVID-19 ARI cases requiring supportive care D3.13: ability to continuously monitor vital signs and oxygen saturation D3.14: availability of oxygen and sufficient sedation for intubated patients by means of respiratory support D3.15: patient care by following national and international guidelines (status of all staff aware of national and international guidelines for case management) D3. 16: communication status to hospital staff responsible for admission criteria and processes related to triage logistics (e.g., location and entry/exit routes) D3.17: availability of awareness of healthcare workers regarding the protocols for off-license use of medicines, which should be done against observational trial protocol, and outcomes recorded against standardized variables Mathematical Problems in Engineering D4.5: prioritization of personnel needs by the unit or service and distribution of personnel accordingly D4.6: ability to collect and train additional staff based on expected needs D4.7: ability of ward staff to work in high demand areas (e.g., infectious disease department, emergency department, and intensive care unit) D4.8: ability in providing training and exercises to ensure staff competency and safety D4.9: availability of domestic support measures that could enhance staff flexibility for shift work and longer working hours and define off work time for recuperation D4.10: formation of psychosocial support teams for staff and patient families D4.11: consideration of liability, insurance, and temporary license issues for personnel working outside of their specialties D4.12: availability of volunteer workers' policies (vetting, accepting, rejecting, liability issues, etc.) D4.13: reassignment of staff at high risk for complications of COVID-19 acute respiratory infection e fifth component named as "continuity of essential health services and patient care" has only four subcomponents, as explained below. D8.1: developing/maintaining an updated inventory of all equipment, supplies, and pharmaceuticals; availability of a shortage alert and reordering mechanism D8.2: estimation of the consumption of required equipment, materials, and medicines according to the most likely pandemic scenario D8.3: ability to consult with the authority to ensure uninterrupted supply of essential medicine and supplies D8.4: quality assessment and requesting a quality certificate before purchasing D8.5: agreements with vendors to ensure resource availability and immediate delivery in times of shortage D8.6: storage of additional materials and physical availability in the hospital for this D8.7: stocking ability of fundamental supplies and medicine based on recommended guidelines D8.8: defining the role of the hospital pharmacy in providing medicines for the treatment D8.9: finding a mechanism for rapid maintenance and repair of essential equipment for basic services D8.10: coordination of an emergency transport strategy to be created to ensure uninterrupted patient transfers D8.11: availability of a policy in place for managing donations of medical supplies, food for staff, etc. e ninth and tenth components, "laboratory services" and "essential support services," have seven components per each, as investigated in detail below.
D9.1: continuous availability of necessary laboratory testing D9.2: identifying essential laboratory supplies and resources and their continuous availability D9.3: identifying backup laboratory personnel and/or alternative laboratory services D9.4: prioritizing testing for respiratory viruses (e.g., COVID-19) D9.5: availability of a laboratory referral pathway for the identification, confirmation, and monitoring of COVID-19 D9.6: establishing and training staff on packaging and transportation procedures for specimen referrals D9.7: availability of mechanisms for the prompt provision of laboratory data to the physicians and health authorities D10.1: estimating the additional supplies required by the support services and introducing a mechanism to ensure the continuous availability of these supplies D10.2: ability of support services to cope with demand boom D10.3: anticipating the impact of COVID-19 on hospital food supplies; taking proactive measures D10.4: availability of backup arrangements for water, power, and oxygen D10.5: availability of hospital security in managing security and safety of hospital D10.6: availability of an area for a temporary morgue and the sufficient body bags and shroud packs D10.7: formulation of a postmortem care contingency plan with proper stakeholders After adapting the components and related subcomponents in this regard, healthcare decision makers are determined. ey are then contacted to fill the questionnaire, which is designed for the decision-making problem. e team consists of hospital managers who know the hospital's operations well and follow all of the activities in most of his/ her time during the COVID-19 pandemic. e team's evaluation forms were sent on April 20, 2020, and forms were received on May 10, 2020.
is study is conducted in three hospitals serving approximately 700,000 people per year. e hospitals are coded as H1, H2, and H3. H1 is one of the oldest hospitals in the region. is hospital has been serving COVID-19 patients, approximately 10% of its total admission capacity. H2 is a hospital that serves as a private hospital outside the city center. As a result of the decision taken by the Republic of Turkey Ministry of Health, the number of patients increased considerably with the diagnosis and treatment of pandemic cases free of charge during the pandemic. H3 is a relatively newly established state hospital in the region. It has served as the main hospital in the fight against COVID-19 in the region. During the peak time of the epidemic, it has operated close to its full capacity. In addition, this hospital is superior to other hospitals in the region in terms of bed capacity, number of intensive care beds, and number of respirators.

Assessment.
In the first step of IVSF-TOPSIS, a decision matrix is constructed. ree hospital decision makers have rated the hospitals with respect to the ninety-nine subcomponents using the linguistic scale in Table 1. At the same time, they have rated the subcomponents using the same scale to obtain importance weights. It is indeed the second step of the approach. Also, the aggregation is performed in this step using an IVSWAM operator. Table 2 provides the obtained final weights of these subcomponents in IVSFSs.
As an example, Figure 3 shows the calculation process to obtain the weight of the D1.1 subcomponent from the expert evaluations in IVSFS. e aggregation of the decision matrix, which is constructed under IVSFSs, is then made. As an example of this, Figure 4 shows the process to obtain weighted IVSF values of H1, H2, and H3 hospitals against the D1.1 subcomponent from the expert evaluations.
In the third step, the aggregated decision matrix is converted into a weighted interval-valued spherical fuzzy decision matrix by considering the weight values obtained in Table 2. As an example, Figure 5 demonstrates the process to get a weighted aggregated decision matrix in IVSFSs using equation (5).
Following this step, the weighted interval-valued spherical fuzzy decision matrix is defuzzified in the fourth step. In the fifth and sixth steps, the IVSFPIS and IVSFNIS values based on the score functions and separation measures from these two values are calculated. Finally, the CR for each hospital is obtained, and the prioritization is performed. Figure 6 presents the CR values for the hospitals.
e CR values of IVSF-TOPSIS indicate that the readiest hospital among COVID-19 is H1. It has an IVSF-TOPSIS CR value of 0.79. is hospital is an educational and research hospital and serves many admissions during the pandemic time. On the other side, H2 has a value of 0.44. H3 has the lowest CR value of 0.24. e least ready hospital is H3.

Analysis.
is section provides the additional analysis results, which show a secondary ranking of hospitals with respect to main components (10 components, as mentioned in the WHO's adapted checklist). is follow-up analysis may help hospital executives on a regional and national scale determine in which aspects hospitals are ready and in what aspects hospitals are not prepared for pandemics. Figures 7  and 8 show detailed analysis results and ranking orders of hospitals against the COVID-19 fight, respectively. e results are based on the D1 dimension, and hospital rankings are obtained as H1, H3, and H2, respectively. e relatively new hospital "H3," although superior in terms of technical possibilities, is ranked in the second position. It can be concluded that for the surge competence, not only the infrastructure is sufficient, but also a strong health team is needed. Although hospital "H2" has a relatively excellent physician staff because it is a private hospital, it has been ranked in the second position due to lack of infrastructure  and equipment. As a result of the evaluation for the D2 dimension, the hospitals are listed as H3, H2, and H1, respectively. H3 is the hospital that treats the most COVID-19 patients in the region. It can be said that the extra training given to the hospital staff of H3 during the epidemic is beneficial. It can also be noted that as the number of patients increases, the awareness of protection from infection increases, too. e ranking order of "H2, H1, H3" is obtained based on the D4 dimension. is ranking evokes the rate of healthcare workers per patient. Increasing the number of staff may enable the D4 dimension to be improved. e evaluation results of the dimensions "D3, D5, D6, D7, D8, and D9" yield the same ranking order (H1, H2, H3).
is ranking has the same order as the time the hospital has been operating. ere is a close relationship between case management, continuity of health services, early warning and monitoring, communication capability, supply chain management, laboratory services, and the institutional structure of the hospital and duration of the operation. In addition, these parameters are not related to the novelty and adequacy of hospital equipment but are closely related to the number and transformation of health personnel. It should be made attractive for the qualified health personnel to operate in the hospital to create a pandemic-prepared hospital within the scope of the dimensions of D5, D6, D7, D8, and D9. Job satisfaction of the staff should also be increased. e ranking in the evaluation made for the D10 dimension has been obtained as H2, H1, and H3. Transportation facilities at H2 and H1, which are in the top two in essential support   services, are high.
e D10 dimension has a tight relationship with transport and logistics capabilities. e condition of meeting the D10 dimension shows a relationship with the location of the hospital.
A comparative study has also been studied to demonstrate the solidity of the proposed approach. For this aim, the results of the proposed approach by IVSF-TOPSIS and traditional F-TOPSIS by Chen [54] are compared. Aggregated experts' evaluations are transformed into triangular fuzzy numbers, and computational steps in Chen [54] are followed. e variations in final CR values and hospital rankings are then observed. e results are shown in Table 3. Table 3 shows that hospital rankings are the same by both approaches according to the nine of ten analysis bases. e only different ranking was obtained in the analysis based on the D10 dimension. When we compare the results obtained with both approaches according to the CR values, we observe close results. Since we do not observe significant differences     between the current study and the benchmark model, we can conclude that this proposed approach is appropriate and applicable for this area.

Conclusion
Fighting against the pandemic is carried out with the proper management of human resources, equipment, materials, and information. One of the essential parameters in combating pandemics is planning resources in environments where resources and time are limited. Hospital preparedness is a vital component of an emergency plan that can significantly reduce the impact of large-scale epidemics. erefore, evaluating organizational readiness is an essential step in this planning process.
is study proposes a hospital preparedness assessment against the COVID-19 pandemic using the TOPSIS decisionmaking method alongside its IVSFS extension. e proposed approach is applied to rank three Turkish hospitals serving as tertiary healthcare services. e paper has adapted and used ninety-nine subcomponents under ten components in the model initially reported by the WHO in 2020. ese components are concerned with the patient surge, infection monitoring and control, human resource, case management, communication, supply chain management, laboratory services, surveillance, and essential services. Under the weights of these subcomponents, a preparedness index for each hospital is obtained. e index is specifically computed individually for each hospital under each main dimension. is helps decision makers and policymakers observing the readiness level of the hospital from ten different viewpoints. e results of the study indicate that H1 is the readiest hospital with the highest IVSF-TOPSIS CR value of 0.79, considering all ten components. H3 is vice versa. It has the lowest CR value (0.24), which means that it is the least ready hospital among them. H2 is placed at the middle point with a CR value of 0.44. Assuming that 0.5 is a threshold, we can say that only H1 is ready for the COVID-19 pandemic. Although the other two hospitals have acceptable CR values for the ten different analyses, they are not fully prepared. At this point, they should formulate a preparedness improvement plan considering their deficiencies and negative aspects.
Regarding the methodological side, the study follows a multicriteria decision-based approach since the nature of the problem fits well with this concept. Additionally, a relatively complex argument "IVSFSs" is merged with the TOPSIS multicriteria decision model. However, the approach benefits from the linguistic evaluation of hospital decision makers and does not require any precise hospital readiness assessment data in this regard. is is an essential advantage of the approach. It can be easily adapted to other hospitals. A national-scale assessment may be possible to improve mitigation ability against a pandemic. is assessment will be an essential guide for reducing the effects of outbreaks and improving personnel disaster preparedness training, organizing material management activities during the epidemic period, accelerating the response to outbreaks, and managing human resources appropriately.

Data Availability
No data were used to support this study.