• 沒有找到結果。

The disaster response performance of hospitals in Taiwan: evaluation and classification

N/A
N/A
Protected

Academic year: 2021

Share "The disaster response performance of hospitals in Taiwan: evaluation and classification"

Copied!
17
0
0

加載中.... (立即查看全文)

全文

(1)

DOI 10.1007/s11135-010-9319-7

The disaster response performance of hospitals

in Taiwan: evaluation and classification

Fuh-Hwa Franklin Liu· Ying-Hock Teng · Ching-Hsiang Lai

Published online: 7 April 2010

© Springer Science+Business Media B.V. 2010

Abstract Hospitals need disaster response plannings and temporary shelters to rescue victims in disasters. In Taiwan, there are 82 emergency medical service (EMS) hospitals are requested to possess guildlines of preparedness and responses to mitigation the damage and recovery the pre-event status in a disaster. A measurement chart including three major factors—facility, manpower, and disaster plan and drill mode—was created to evaluate the performance of disaster management of these EMS hospitals. Based on the expert opinions, some hospitals were selected as a reference set prior to the accreditation. Data envelopment analysis (DEA) is applied to classify each of the remaining hospitals into qualified or unqual-ified classes. Then, each unqualunqual-ified hospital was recommended to improve its practice of disaster preparedness and responses into the qualified level. We also find that, on average, private hospitals perform better than public hospitals and medical centers perform better than the regional hospitals. But, the differences are not statistically significant.

Keywords Disaster response planning· Data envelopment analysis (DEA) ·

Performance· Classification · Emergency medical service (EMS)

F.-H. F. Liu

Department of Industrial Engineering and Management, National Chiao Tung University, 1001, University Road, Hsinchu 300, Taiwan

Y.-H. Teng

Department of Emergency Medicine, Chung Shan Medical University Hospital, 110, Sec. 1, Jianguo North Road, Taichung 402, Taiwan

C.-H. Lai (

B

)

School of Applied Information Sciences, Chung Shan Medical University, 110, Sec. 1, Jianguo North Road, Taichung, 402, Taiwan

(2)

1 Background of motivation

Man-made and natural disasters often occur around the world. Taiwan suffered a major earth-quake in the early morning on 21 September 1999. More than 2,400 people died and 8,000 people sustained injuries in the fault areas. Hospitals located near the fault received mass casualties after the quake. Most of the victims arrived at an emergency department (ED)

in the initial hours (Chen et al. 2001). Although many health care providers and medical

resources were sent into the disaster area, the hospitals were severely damaged, and some of them even had function loss. Based on the emergency medical service (EMS) level, when hospitals were damaged beyond operating facility, patients needed to be transferred because of the lack of temporary shelters or field hospitals.

The American Red Cross often coordinates and staffs mass shelters to care for people who have physical or mental limitations that restrict their mobility or ability to function indepen-dently. These individuals may require assistance with daily life activities, or medication to sustain them. Long road journeys may be harmful to their health. Because these individuals are often at the highest risk, human and economic resources should be allocated to enhance

their survival and promote their well being during a catastrophe.Pretto et al.(1992,1994)

demonstrated that there were many victims whose deaths might have been prevented if they

had received medical attention in the first 6 h after the tremor.Safar(1986) showed that about

25–50% of the patients could have been saved if emergency care had been rendered at once. Therefore, it is imperative for a medical response system to be maximally efficient so that medical care can be administered as soon as possible to reduce mortality rates.

The Joint Commission on Accreditation of Healthcare Organizations (JCAHO) requires that hospitals have emergency plans “designed to manage the consequences of natural disas-ters or other emergencies that disrupt the hospital’s ability to provide care and treatment” (JCAHO 1994). The plans must be specific in the hospital’s role in community-wide emer-gency preparedness, the procedures of implementation, the management of key materials and activities, staff deployment and roles, the management of patient care services, staff preparation, disaster drills, and monitoring and evaluation of hospital performance.

A successful, comprehensive, and risk-based emergency management program of disaster preparedness, response, and recovery will reduce the loss of life and resources. A standard requires detailed written plans for all potential emergencies and disasters in the facility or community, training for all employees as soon as they begin work at the facility, periodic reviews of procedures with staff, and unannounced drills.

There are several such instruments of national or regional importance in the world. For instance, in Europe, the Sphere Project, a consortium of non-governmental organizations, has

formulated a “Humanitarian Charter and Minimum Standards in Disaster Response” (Sphere

Project 2004). It is the first general standard for humanitarian operations, and a distillation of great experience and worldwide consultation. In the USA, the National Fire Protection Asso-ciation (NFPA) has prepared a “Standard on Disaster/Emergency Management and Business

Continuity Programs” (NFPA 2007). In Australia, the Australian Standards Organization has

an “Australia/New Zealand Risk Management Standard” (The Joint Australian/New Zealand

Standard 2004). This instrument uses a standardized and widely accepted risk management methodology, and is thus an important and authoritative source of support for risk manage-ment policies. In Taiwan, the law regarding emergency medical care was legislated in 1995

and revised in 2000, 2002, 2005, and 2007 (Department of Health 2007), and the rule

per-taining to disaster protection was implemented in 2000 (Department of Health 2000). There

(3)

Disaster scholarAghababian et al.(1994) indicated that a hospital should plan to take care of inner and outer disasters in advance. The hospital is required to consider manpower and equipment deficiency in disaster. The problem of manpower refers to the lack of the medical staff required to take care of a massive number of patients. The equipment includes the supply of electricity (power), water, oxygen, and so on. Besides these, the EMS hospitals should plan for a temporary shelter to confront major disasters. The book, “Community Medical Planning

and Evaluation Guide” (Heide 1995), published by America College Emergency Physician

(ACEP), provides a detailed form to evaluate the building, telephone, radio, records, and computer information, and is a very good reference for disaster preparation and recovery.

In Taiwan, hospital accreditation was initiated in 1986. The purpose of accreditation was to promote the quality of medical care. In 2000, 16 public (government) and non-profit hos-pitals were accredited as teaching medical centers, and 88 moderate-sized hoshos-pitals were accredited as regional teaching hospitals. During a field visit, the evaluation of EDs was performed in isolation by authorized emergency physicians. Also, evaluation was used to increase the quality of emergency services. However, it did not include emergency evacua-tion and temporary shelters as a part of the evaluaevacua-tion issue. To improve the performance and quality of disaster emergency care of Taiwan’s EMS hospitals, we established an evaluation program that focuses on disaster preparedness planning and temporary shelters. This research is a pioneer study in the field of EMS and disaster preparedness in Taiwan.

2 Accreditation of disaster preparedness and temporary shelters

In Taiwan, hospital accreditation is usually based on a measurement chart which is created by an evaluation committee convened by the Department of Health (DOH). We performed the accreditation of disaster preparedness planning and temporary shelters in EMS hospitals in 2001. A measurement chart and evaluation guide were created by the disaster planning evaluation committee. The chart was based on the “Community Medical Planning and

Eval-uation Guide” (Heide 1995), consulted the “Hospital Accreditation and Evaluation Table”

of DOH (Taiwan Joint Commission on Hospital Accreditation 1999), and modified through

discussion among the members of the committee and some experts in the fields of health, fire and construction in a conference. The final version of the measurement chart and evaluation guide is listed in the appendix.

The chart includes three major factors: facility, manpower, and disaster plan and drill mode. Each factor consists of some related items, and each item was assigned a different point in accordance with its different consequences. The total points for each factor is the sum of the points of items included in it. For instance, the facility factor includes the following 18 items: space ownership, space type, space area, space site, space capacity, space design, power supply, ventilation facilities, water supply, communication facilities, oxygen supply, transportation, emergency equipment, eating space, dressing and sleeping supply, toilet, bath-room, and kitchen. The points assigned to space design is 7, while for transportation it is 3. The total points of facility, 65, is the sum of the points of the above 18 items. Similarly, the total points of manpower, and disaster plan and drill mode are 20 and 15, respectively.

For each item, the accredited ranks are divided into five levels: excellent, good, adequate, mild deficiency, and deficiency, each with corresponding weights, 1, 0.8, 0.6, 0.4, and 0.2, respectively. If a hospital was accredited as at a good level in the space design item, the point score gained in this item was 5.6, which equals the item point, 7, multiplied by the rank weight, 0.8. In this chart, the accredited points of all items were designed higher as the hospital performed better in the accreditation.

(4)

The accreditation process was performed from December 2000 to December 2001. After deleting non-teaching hospitals and teaching hospitals without EMS, we interviewed 82 hos-pitals, among which 16 (19.5%) were medical centers and 66 (80.5%) were regional hospitals.

The three accredited points of the hospitals are listed in Table1, in the increasing order of

their monthly emergency patients. Our study is based on the following two assumptions: (i) In the disaster area, if a hospital ordinarily receives more emergency visits, then it will receive relatively more patients if a disaster occurs, and (ii) A good quality of preparedness plan will rescue more victims in a disaster.

Thus, the 82 hospitals were partitioned into two groups according to the monthly emer-gency patients: the hospitals with minor EMS as a reference set, and the hospitals with major EMS to be accredited, denoted as R and A, respectively. The committee should choose a

specific percentage of hospitals on the top of Table1as R and the rest as A. To illustrate our

methodology, we set the specific percentage at about 10%, and groups R and A have 9 and 73 hospitals, respectively. It is reasonable to expect that the hospitals in A will perform better than those in R in terms of disaster preparedness. However, a hospital in A is accredited as “qualified” if it outperformed all members of group R, otherwise it is “unqualified”.

We employed data envelopment analysis (DEA) model (Seiford and Zhu 1998) to assess

the performance of each hospital with respects to R. Its optimal value is used to classify hos-pitals in A into the qualified or unqualified class. It also provides the direction and extent of improvement for the unqualified hospitals. The DEA models for the performance assessment

are introduced in the Sect.3. The results are presented in Sect.4. Conclusions and discussions

are presented in Sect.5.

3 Methodology

Many researchers have used DEA methods to examine differences in health care system

performance.Burgess and Wilson(1996) used it to analyze the differences of four types of

ownership structure in the hospital industry in the USA—private non-profit, private for-profit, federal, and state and local government. They also used DEA efficiency scores as regressive variable for exogenous variables to gain insight into various issues impacting the debate over

health-care reform in USA (Burgess and Wilson 1998).Hofmarcher et al.(2002) investigated

the evolution of efficiency and productivity in the hospital sector of an Austrian province

from 1994 to 1996.Grosskopf et al.(2001) used a frontiers approach to compare the teaching

and non-teaching hospitals under their best practice performance.

Data envelopment analysis (DEA), originally proposed by Charnes et al.(1978), is a

multiple input-output efficient technique that measures the relative performances of deci-sion-making units (DMUs) using linear programming based models. It is non-parametric because it requires no assumption on the weights of the underlying production function. In

this paper, let Hj denote hospital j and the symbols y1 j, y2 j, and y3 j denote the points of

facility, manpower, and disaster plan and drill mode corresponding to Hj, respectively. The

relative performance of a specific hospital k is obtained by the following DEA model:

tk∗= max 3  r=1 uryr k s.t. 3  r=1 uryr j ≤ 1, j ∈ R ur ≥ 0, r = 1, 2, 3. (1)

(5)

Ta b le 1 Accredited data and ef fi cienc y scores of T aiw an’ s EMS hospitals Hj Number of ER patients per m onth y1 j y2 j y3 j Score t ∗ j Hj Number of ER patients per m onth y1 j y2 j y3 j Score t ∗ j R : R eference set A : A ccredited hospitals 1 TP 600 26 .21 0. 28 .61 .000 42 4000 40 .41 5. 21 0. 01 .490 2 650 26 .67 .07 .20 .766 43 4000 41 .01 2. 01 1. 61 .243 3 900 34 .88 .87 .20 .863 44 U 4000 39 .89 .07 .00 .956 4 1050 31 .09 .49 .00 .970 45 4000 43 .41 2. 81 0. 01 .255 5 TP 1100 41 .41 0. 28 .41 .000 46 4000 40 .01 3. 61 0. 21 .333 6 TP 1200 42 .08 .29 .20 .824 47 4250 43 .01 1. 61 0. 21 .161 7 TP 1250 36 .49 .69 .41 .000 48 4300 38 .61 1. 48 .61 .118 8 1350 29 .88 .46 .81 .000 49 U 4500 29 .29 .07 .80 .894 9 1400 31 .88 .87 .60 .875 50 4500 35 .21 3. 29 .21 .294 A : A ccredited hospitals 10 U 1500 33 .66 .27 .60 .822 51 4500 31 .41 1. 07 .61 .078 11 1500 44 .81 2. 41 1. 41 .261 52 4500 32 .41 1. 08 .21 .078 12 1650 45 .21 2. 01 0. 41 .195 53 4500 43 .61 3. 08 .21 .275 13 U 1800 34 .81 0. 28 .21 .000 54 4500 35 .61 1. 48 .21 .118 14 2100 48 .21 3. 01 0. 81 .276 55 4500 48 .61 1. 81 0. 81 .219 15 2200 43 .01 1. 88 .81 .157 56 4650 37 .49 .69 .41 .008 16 2200 35 .21 1. 48 .61 .118 57 4700 37 .01 1. 81 0. 21 .171 17 2300 41 .01 2. 08 .81 .176 58 4750 38 .61 2. 69 .01 .235 18 2500 33 .41 2. 66 .81 .235 59 4750 53 .61 4. 41 1. 81 .412 19 U 2500 33 .09 .87 .80 .961 60 U 4800 24 .49 .09 .00 .957

(6)

Ta b le 1 continued Hj Number of ER patients per m onth y1 j y2 j y3 j Score t ∗ j Hj Number of ER patients per m onth y1 j y2 j y3 j Score t ∗ j 20 2700 34 .01 2. 89 .61 .255 61 4800 33 .21 3. 49 .81 .314 21 2800 39 .61 2. 41 0. 61 .226 62 5000 48 .01 3. 41 0. 01 .314 22 2850 46 .41 4. 49 .21 .412 63 5250 49 .61 2. 61 0. 41 .235 23 3000 33 .21 1. 67 .81 .137 64 5400 40 .21 0. 09 .01 .018 24 3000 48 .81 5. 69 .01 .529 65 6000 50 .61 2. 41 0. 61 .238 25 U 3000 34 .07 .46 .20 .815 66 U 6000 40 .68 .26 .60 .969 26 3000 40 .81 2. 01 1. 21 .225 67 6000 38 .81 0. 47 .41 .020 27 3000 42 .61 2. 88 .21 .255 68 6300 38 .21 1. 21 0. 01 .125 28 U 3050 40 .81 0. 28 .21 .000 69 6500 44 .01 4. 81 1. 21 .451 29 U 3350 35 .48 .67 .60 .874 70 U 6500 37 .66 .28 .00 .895 30 U 3400 41 .69 .28 .60 .998 71 6500 32 .61 0. 21 0. 01 .064 31 3450 39 .21 2. 09 .41 .176 72 6750 39 .81 2. 28 .61 .196 32 3500 41 .21 4. 61 0. 01 .431 73 7000 48 .01 4. 41 0. 81 .412 33 3500 33 .41 1. 48 .21 .118 74 7000 39 .41 3. 48 .61 .314 34 3500 36 .21 2. 09 .01 .176 75 7000 28 .01 2. 01 0. 01 .176 35 U 3600 39 .64 .67 .60 .943 76 7500 49 .01 3. 41 1. 41 .326 36 U 3750 41 .09 .88 .60 .998 77 8000 40 .21 3. 49 .41 .314 37 3750 44 .41 3. 21 0. 41 .294 78 10000 42 .21 0. 69 .61 .079 38 3750 33 .81 1. 48 .61 .118 79 11000 39 .61 3. 41 0. 01 .314 39 3750 40 .09 .81 0. 21 .087 80 12000 35 .61 3. 49 .41 .314 40 3800 28 .89 .69 .81 .043 81 13000 37 .81 4. 49 .01 .412 41 U 3850 33 .06 .26 .60 .786 82 14000 46 .61 2. 89 .41 .255 TP hospitals with top p erformance in reference set, U unqualified hospitals

(7)

The decision variable ur is the weight assigned to factor r. The relative performance, tk∗, is evaluated without predetermining weights of these factors. The weights are settled by

maxi-mizing the performance of Hkand restricting all hospitals in R with performance not greater

than one.

In the event of a hospital k belonging to R, tk= 1 indicates that Hkhas the top performance

in the whole group. We collect a set of hospitals that possess the top performance in R and

denote it as TP. Hence, the index set R in model (1) can be substituted by TP (Seiford and

Zhu 1998). For hospital k in A is evaluated by (1), tk> 1 indicates that Hkhas the possibility

to perform better than R (so as TP) if weights are set by its optimal solutions; tk∗≤ 1 indicates

that Hkcould not perform better than all members of R for any given weights. The dual form

of model (1) is written as follows:

tk= min tk s.t.  j∈R λjyr jyr k tk , r = 1, 2, 3; (2)  j∈R λj = 1; λj≥ 0, j ∈ R; tk≥ 0.

The model could be used to measure the extent of qualification of an accredited hospital Hk with respect to group R. The classification is based on the following rule:

Rule 1. If tk> 1, then Hkis qualified. Otherwise tk≤ 1, Hkis unqualified.

In the case where Hk is unqualified, the ratio(1/tk− 1) is the extent of

under-qualifi-cation, and this hospital is required to improve the performance of facility, manpower, and

disaster plan and drill mode until they reach beyond the levels of y1k/tk, y2k/tk, and y3k/tk∗,

respectively. In the case where Hkis qualified,(1 − 1/tk) is the ratio of outperformance,

and is the maximum extent of deterioration for Hkto remain qualified. The deterioration rate

indicates how Hkis superior to TP. The larger value of tkis, the better the Hkdoes.

However, the optimal value tk∗to model (2) reveals the performance of Hkwith respect

to TP. For a set of hospitals that is under accredited, one can employ model (2) to assess

their relative performances and rank them accordingly. Figure1illustrates the relationship

between the reference surfaces construct by TP and accredited group A graphically, where the qualified and unqualified classes are separated by the TP surface. It also shows that an

unqualified hospital Hkrequires improvement until it is better than the point Pkon the TP

surface, where Pkis the intersection of the TP surface and the line passing through the origin

and Hk.

4 Results

To determine whether a given hospital, Hk, is qualified or not under accreditation, we employ

model (2) to calculate tkand follow Rule 1 to obtain the classification of Hk. We used the tool

“Solver” embedded in Microsoft Excel to solve model (2). Table1presents the following

(8)

(qualified class) (unqualified class) Hk Pk y1 y2 unqualified hospitals qualified hospitals reference hospitals reference hospitals with top performance The TP surface

0

Fig. 1 The reference surface generated by TP and the relationship between qualified and unqualified hospitals

in the accredited set

1. The relative performances of the reference hospitals, H1 ∼ H9, are the optimal value

to model (2). The performances of H1, H5, H6, and H7all equal to one. That is, they

performed the best in the whole set R. Hence, set TP consists of these four hospitals.

2. Each hospital in A under accreditation was set as Hk in model (2) to assess the

per-formance, tk, with respect to R. The performances of the 73 accredited hospitals are

distributed as: 58 (79%) of them are greater than one, 2 (3%) are equal to one, and 13 (18%) are less than one. It means that 58, 2, and 13 of them are superior, even, and inferior to TP, respectively. Hence, there are 58 qualified hospitals while the other 15 hospitals are unqualified.

3. The accredited hospitals are ranked appropriately according to their relative

perfor-mances. For instance, H24with top performance 1.529 is the first rank and H42with

the secondary performance 1.49 is the second rank, and so on.

However, an unqualified hospital is required to improve its performance until it becomes

qualified. Table2depicts the boundary values of the three factors in which the 15

unquali-fied hospitals should be improved to go beyond. For example, the unqualiunquali-fied H10needs to

increase its facility, manpower, and the disaster plan and drill mode factor points beyond the values 40.85, 7.54, and 9.24, respectively.

We also looked at the performance differences by ownership type and by hospital

accred-ited levels in A. These results are presented in Table3and4. When they are classified by

ownership type, there are 25 public hospitals and 48 private hospitals; when they are classified by accredited levels, there are 16 medical centers and 57 regional hospitals. Our findings indi-cate that on average the private hospitals performed better than the public hospitals, although the differences are not statistically significant by using t, Kruskal–Wallis, and median tests. On average, the medical centers are better than the regional hospitals, although this is not statistically significant.

(9)

Table 2 The target value that the unqualified hospitals should be improved

Hj Score (tj) Facility (y1 j) Manpower (y2 j) Plan & drill (y3 j)

10 0.822 >40.85 >7.54 >9.24 13 1.000 >34.80 >10.20 >8.20 19 0.961 >34.35 >10.20 >8.12 25 0.815 >41.74 >9.08 >7.61 28 1.000 >40.80 >10.20 >8.20 29 0.874 >40.49 >9.84 >8.69 30 0.998 >41.69 >9.22 >8.62 35 0.943 >42.00 >4.88 >8.06 36 0.998 >41.08 >9.82 >8.62 41 0.786 >42.00 >7.89 >8.40 44 0.956 >41.64 >9.42 >7.32 49 0.894 >32.66 >10.07 >8.72 60 0.957 >25.48 >9.40 >9.40 66 0.969 >41.92 >8.47 >6.81 70 0.895 >42.00 >6.93 >8.94

Table 3 Efficiency comparison

by ownership types Ownership type No. Mean SD Min. Max.

Public 25 1.148 0.184 0.815 1.529

Private 48 1.176 0.161 0.786 1.490

Test Level P-value

t-test 0.690 0.494

Kruskal–Wallis 0.546 0.460

Median 1.082 0.298

Table 4 Efficiency comparison

by accredited levels Accredited level No. Mean SD Min. Max.

Regional 57 1.159 0.172 0.786 1.529

Medical center 16 1.193 0.157 0.894 1.412

Test Level P-value

t-test 0.710 0.478

Kruskal–Wallis 0.729 0.393

(10)

5 Conclusions and discussions

In this research, the performance of disaster preparedness planning and temporary shelters in Taiwan’s hospitals were measured using three factors. The DEA technique, dealing with non-parametric data, was used in the two-group discriminant problem. The reference set was predetermined by the accreditation committee and their top-practice was regarded as an objective rule of classification. The hospitals under accreditation were expected to be better than the objective. The TP played top in R and constructed a piece-wise linear ref-erence surface to separate the accredited hospitals into qualified and unqualified classes. If a hospital was unqualified, it was recommended to improve its practice until it moved beyond TP.

This research did not focus on discriminating efficient hospitals from inefficient ones. We proposed the reference boundary of passing the evaluation and suggested that all hospitals needed to reach beyond the boundary. The DEA classification technique is important for decision-making problems where decision makers are interested in knowing the changes in attributes for moving an unqualified unit into the qualified class. In this application, the classification method promotes the quality of disaster preparedness planning and temporary shelters, indicates which hospital does not satisfy our requirements, and provides what the hospital needs to improve in order to achieve the qualification level.

In this research, the reference boundary is not assigned prior to the accreditation. We just selected some hospitals based on minor emergency patients as the least needed for this accreditation. Then, a detached classification boundary was relatively determined by the top practice of these hospitals. This analysis is not consistent with other discriminant analysis. It used only the data with one group to develop a boundary, instead of data with two or more groups. This non-parametric DEA classification can be widely applied in future accreditation or risk management problems.

The limitation of this study is the selection of the reference set. The setting of the threshold value as the reference set R may affect the accreditation results. What is the suitable per-centage of hospitals that should be set to R? or, what would happen if the reference hospitals had performed better? Setting a higher value, or including a better hospital in the reference, would result in more hospitals needing to improve their disaster preparedness capability. Setting a lower value would result in less hospitals needing to be improved. In this study, the committee subjectively selected nine reference hospitals based on the assumptions and their experience. However, a region under higher threat from disasters may set a larger value to have a higher preparedness capability, or one may set a value in a gradually increasing manner if the accreditation process is held periodically.

In the long run, some regional hospitals could be selected as a reference set. It is expected that a medical center under accreditation should operate superior to the reference set, oth-erwise it should improve its performance beyond the boundary. In the risk management problem, organizations that do not operate well or are even bankrupt could be identified in advance, then the top performance of these pre-selected organizations could be assessed as the reference. It requires that all of the current organizations perform better than the bound-ary value as risk prevented units. While the organizations are not better than the boundbound-ary, the organizations have a high risk of going bankrupt, and must improve they go beyond the boundary.

Acknowledgements This research was partly supported by the Department of Health, Executive Yuan of Taiwan, DOH90-TD-1064.

(11)

Appendix

Measurement chart and measurement guides

Part 1. Fixed structure

Proportion of scores: 65 points s t n i o p _ _ _ _ l a t o t b u S

Ranks Excellent Good Adequate Mild

deficiency

Deficiency

Weights 1.0 0.8 0.6 0.4 0.2

Space ownership (3 points) 1. private land, oral appointment 2. rental

3. hospital property 4. others

Excellent: with characteristic of feasibility, stability, flexibility, and safety.

Good: with 3 of above characteristics. Adequate: with 2 of above characteristics. Mild deficiency: with 1 of above characteristic. Deficiency: none.

Space type (5 points) 1. building

2. underground building 3. open-ended space 4. others

Same as above.

Space area (4 points) 1. <200 m2 2. 200-500 m2 3. 500-1000 m2 4. >1000 m2 5. others Same as above.

Space site (2 points)

1. private land, oral appointment 2. rental

3. hospital property 4. others

Same as above.

Space capacity (4 points) 1. <50 people 2. 50-100 people 3. 100-200 people 4. >200 people 5. others Same as above.

Space design (7 points) 1. clear sign for day and night 2. control gate

3. easy entrance and exit 4. more than 2 exits 5. good ventilation

6. examination/operation room 7. space for drug & supplies storage 8. isolated room and space 9. others

Excellent: with >5 of left items. Good: with 5 of left items. Adequate: with 4 of left items. Mild deficiency: with 3 of left items. Deficiency: with <3 of left items.

(12)

Part 1. Fixed structure (continue 1)

Ranks Excellent Good Adequate Mild

deficiency

Deficiency

Weights 1.0 0.8 0.6 0.4 0.2

Emergence power supply (4 points) 1. small generator

2. big generator 3. un-disconnect system 4. others

Excellent: with characteristic of feasibility, stability, flexibility, and safety.

Good: with 3 of above characteristics. Adequate: with 2 of above characteristics. Mild deficiency: with 1 of above characteristic. Deficiency: none.

Ventilation facility (3 points) 1. fan 2. exhaust fan 3. air-conditioner 4. central air-conditioner 5. others Same as above.

Water supply (3 points) 1. warm drinking water 2. drinking fountain 3. central warm water 4. U-V antiseptic 5. retrograde system 6. others

Same as above.

Communication facilities (4 points) 1. telephone 2. cell phone 3. fax machine 4. walky-talky 5. broadcast system 6. satellite phone 7. computer line 8. others Same as above.

Oxygen supply (3 points) 1. small container 2. big container 3. pre-set central tube 4. pre-set single tube 5. others Same as above. Transportation (2 points) 1. small bus 2. big bus 3. ambulance 4. others Same as above.

(13)

Part 1. Fixed structure (continue 2)

Ranks Excellent Good Adequate Mild

deficiency

Deficiency

Weights 1.0 0.8 0.6 0.4 0.2

Emergence equipments (6 points) 1. defibrillator

2. vital sign monitor 3. rescue package 4. ventilator 5. EKG monitor 6. X-ray machine 7. blood lab 8. suction 9. operation room

10. stretcher/wheel chair/ER bed 11. others

Excellent: with >8 of left items. Good: with 7-8 of left items. Adequate: with 5-6 of left items. Mild deficiency: with 3-4 of left items. Deficiency: with <3 of left items.

Eating space (3 points) 1. indoor

2. outdoor 3. whole day supply 4. fire safety 5. others

Excellent: with characteristic of feasibility, stability, flexibility, and safety.

Good: with 3 of above characteristics. Adequate: with 2 of above characteristics. Mild deficiency: with 1 of above characteristic. Deficiency: none.

Dressing & sleeping supply (4 points) 1. clothes 2. blanket 3. mosquito net 4. sleeping bag 5. tent 6. others Same as above. Toilet (3 points) 1. fix toilet 2. movable toilet 3. privacy 4. hand wash 5. tissue paper 6. others Same as above. Bathroom (2 points) 1. privacy 2. hot water 3. shower 4. others Same as above. Kitchen (3 points) 1. indoor 2. outdoor 3. water supply

4. residual food management 5. fire safety

6. others

(14)

Part 2. Manpower

Proportion of scores: 20 points s t n i o p _ _ _ _ l a t o t b u S

Ranks Excellent Good Adequate Mild

deficiency

Deficiency

Weights 1.0 0.8 0.6 0.4 0.2

ER doctor (4 points) 1. specific arranged 2. at least internist & surgeon 3. at least leaded by senior resident 4. standard process protocol 5. at least 2 group for shift 6. others

Excellent: with >4 of left items. Good: with 4 of left items. Adequate: with 3 of left items. Mild deficiency: with 2 of left items. Deficiency: with 1 or none of left items.

ER nurse (3 points) 1. specific arranged

2. ER cares training (eg. ACLS) 3. standard process protocol 4. at least 2 group for shift 5. others

Excellent: with >3 of left items. Good: with 3 of left items. Adequate: with 2 of left items. Mild deficiency: with 1 of left items. Deficiency: none.

ER volunteer (1 point) 1. specific arranged 2. first aid training

3. know the standard procedure 4. good communication ability 5. at least 2 group for shift 6. others

Excellent: with >4 of left items. Good: with 4 of left items. Adequate: with 3 of left items. Mild deficiency: with 2 of left items. Deficiency: with 1 or none of left items.

Food provider (2 points) 1. specific arranged 2. with license 3. first aid training

4. know of standard procedure 5. at least 2 group for shift 6. others

Same as above.

Transportation staff (2 points) 1. specific arranged 2. with license 3. EMT-1 training

4. know of standard procedure 5. at least 2 group for shift 6. others

Same as above.

Maintenance staff (2 points) 1. specific arranged 2. with license 3. first aid training

4. know of standard procedure 5. at least 2 group for shift 6. others

(15)

Part 2. Manpower (continue)

Ranks Excellent Good Adequate Mild

deficiency

Deficiency

Weights 1.0 0.8 0.6 0.4 0.2

Social worker (2 points) 1. specific arranged 2. first aid training

3. good communication ability 4. have standard procedure 5. at least 2 group for shift 6. others

Excellent: with >4 of left items. Good: with 4 of left items. Adequate: with 3 of left items. Mild deficiency: with 2 of left items. Deficiency: with 1 or none of left items.

Housekeeping staff (2 points) 1. specific arranged 2. first aid training

3. good communication ability 4. know of standard procedure 5. at least 2 group for shift 6. others

Same as above.

Administrative staff (2 points) 1. specific arranged 2. first aid training

3. good communication ability 4. know of standard procedure 5. at least 2 group for shift 6. others

Same as above.

Part 3. Disaster planning and dril mode

Proportion of scores: 15 points s t n i o p _ _ _ _ l a t o t b u S

Ranks Excellent Good Adequate Mild

deficiency

Deficiency

Weights 1.0 0.8 0.6 0.4 0.2

Disaster written plan (5 points) 1. have written plan 2. have improvement process 3. scheduled meeting 4. have activation guide 5. others

Excellent: with characteristic of feasibility, stability, flexibility, and safety.

Good: with 3 of above characteristics. Adequate: with 2 of above characteristics. Mild deficiency: with 1 of above characteristic. Deficiency: none.

(16)

Part 3. Disaster planning and dril mode (continue)

Ranks Excellent Good Adequate Mild

deficiency

Deficiency

Weights 1.0 0.8 0.6 0.4 0.2

Disaster drill plan (2 points) 1. carry out biannually 2. carry out annually 3. carry out irregularly 4. biannually, not carry out 5. annually, not carry out 6. others

Excellent: with item 1 of left. Good: with item 2 of left. Adequate: with item 3 of left.

Mild deficiency: with item 4 or 5 of left. Deficiency: none.

Drill mode (2 points) 1. one hospital only

2. with other nearby organizations 3. both

4. others

Excellent: with characteristic of feasibility, stability, flexibility, and safety.

Good: with 3 of above characteristics. Adequate: with 2 of above characteristics. Mild deficiency: with 1 of above characteristic. Deficiency: none.

Improvement meeting plan (2 points) 1. carry out biannually

2. carry out annually 3. carry out irregularly 4. biannually, not carry out 5. annually, not carry out 6. others

Excellent: with item 1 of left. Good: with item 2 of left. Adequate: with item 3 of left.

Mild deficiency: with item 4 or 5 of left. Deficiency: none.

Response time (2 points) 1. within one hour 2. within 1-3 hours 3. within 3-6 hours 4. within 6-12 hours 5. more than 12 hours 6. others

Excellent: with item 1 of left. Good: with item 2 of left. Adequate: with item 3 of left. Mild deficiency: with item 4 of left. Deficiency: with none or item 5 of left.

Capability in 2 hours (2 points) 1. <25 patients 2. 25-50 patients 3. 51-100 patients 4. 101-200 patients 5. >200 patients 6. others

Excellent: with item 5 of left. Good: with item 4 of left. Adequate: with item 3 of left. Mild deficiency: with item 2 of left. Deficiency: with none or item 1 of left.

References

Aghababian, R.V., Lewis, C.P., Gans, L.: Disaster within hospitals. Ann. Emerg. Med. 23(4), 771–777 (1994) Burgess, J.F., Wilson, P.W.: Hospital ownership and technical inefficiency. Manag. Sci. 42(1), 110–123 (1996) Burgess, J.F., Wilson, P.W.: Variation in inefficiency among US hospitals. INFOR 36(3), 84–102 (1998) Charnes, A., Cooper, W.W., Rhodes, E.: Measuring the efficiency of decision making units. Eur. J. Oper.

Res. 2, 429–444 (1978)

Chen, W.K., Cheng, Y.C., Ng, K.C., Hung, J.J., Chuang, C.M.: Were there enough physicians in an emergency department in the affected area after a major earthquake? An analysis of the Taiwan Chi-Chi earthquake in 1999. Ann. Emerg. Med. 38(5), 556–561 (2001)

(17)

Department of Health. The Rules of Massive Patient Emergency Care in Emergency Medical Service. Department of Health, Executive Yuan, Taiwan.http://dohlaw.doh.gov.tw/Chi/FLAW/FLAWDAT0202. asp(2000). Accessed 15 May 2008 (in Chinese)

Department of Health. Emergency Medical Service Act. Department of Health, Executive Yuan, Taiwan.http:// www.doh.gov.tw/ufile/doc/Emergency%20Medical%20Services%20Act.pdf(2007). Accessed 15 May 2008

Grosskopf, S., Margaritis, D., Valdmanis, V.: Comparing teaching and non-teaching hospitals: a frontier approach (teaching vs. non-teaching hospitals). Health Care Manag. Sci. 4(2), 83–90 (2001)

Heide, E.A.: Community Medical Planning and Evaluation Guide. America College Emergency Physician, Texas (1995)

Hofmarcher, M.M., Paterson, I., Riedel, M.: Measuring hospital efficiency in Austria—a DEA approach. Health Care Manag. Sci. 5(1), 7–14 (2002)

Joint Commission on Accreditation of Healthcare Organizations (JCAHO). 1994 Accreditation Manual for Hospital, II, pp. 14–18. JCAHO, Oakbrook Terrance (1994)

NFPA. NFPA 1600: Standard on Disaster/Emergency Management and Business Continuity Programs.http:// www.nfpa.org/assets/files/PDF/NFPA1600.pdf(2007). Accessed 15 May 2008

Pretto, E.A., Ricci, E., Klain, M.: Disaster reanimatology potentials: a structured interview study in America. III. Results, conclusions and recommendations. Prehosp. Disaster Med. 7, 327–337 (1992)

Pretto, E.A., Angus, D.C., Abrams, J.I.: An analysis of prehospital mortality in an earthquake. Disaster Reanimatology Study Group. Prehosp. Disaster Med. 9(2), 107–117 (1994)

Safar, P.: Rescuscitation potentials in mass disasters. Prehosp. Disaster Med. 2, 34–47 (1986)

Seiford, L.M., Zhu, J.: An acceptance system decision rule with data envelopment analysis. Comput. Oper. Res. 25(4), 329–332 (1998)

Sphere Project. Sphere Handbook: Humanitarian Charter and Minimum Standards in Disaster Response, 2nd edn. Oxfam Publishing, Oxford (2004)

Taiwan Joint Commission on Hospital Accreditation. Hospital Accreditation and Evaluation Table.http:// www.tjcha.org.tw/NewsDetail.asp?NewsId=317(1999). Accessed 15 May 2008 (in Chinese) The Joint Australian/New Zealand Standard. AS/NZS 4360: 2004, Risk Management. SAI Global, Sydney

數據

Fig. 1 The reference surface generated by TP and the relationship between qualified and unqualified hospitals
Table 2 The target value that the unqualified hospitals should be improved

參考文獻

相關文件

 Evaluated deadline and cost perfor mance of various scheduling polici es under a large range of SLA cost function and

The Performance Evaluation for Horizontal, Vertical and Hybrid Schema in Database Systems.. -A Case Study of Wireless Broadband

Because of path planning and refuge activity for a community shelter will cause disaster increasing and result second disaster, hence it has a great relationship between refuge

The main objective of this article is to investigate market concentration ratio and performance influencing factors analysis of Taiwan international tourism hotel industry.. We use

T., “The qualitative habitat evaluation index (QHEI), rationale, methods, and application,” Ohio EPA, Division of Water Quality Planning and Assessment, Ecological Assessment

Keywords: Standard Hotels, Service Quality, Kano’ s Model, Decision Making Trial and Evaluation Laboratory (DEMATEL), Importance-Performance Analysis

C., 2004b, The Market Microstructure and Relative Performance of Taiwan Stock Index Futures a Comparison of the Singapore Exchange and The Taiwan Futures Exchange, Journal of

The government, under pressure from the public, gave the central task of disaster relief, at this time and in the future, to the military and in July 2010