Formación acreditada

SPECIAL ARTICLE

Balanced scorecard management of a hospital emergency department FRANCISCO JAVIER MONTERO-PÉREZ, JOSÉ MANUEL CALDERÓN DE LA BARCA GÁZQUEZ, LUIS JIMÉNEZ MURILLO, FRANCISCO DE BORJA QUERO ESPINOSA, FRANCISCO GRACIA GARCÍA, JUAN JOSÉ ROIG GARCÍA Unidad de Gestión Clínica de Urgencias, Hospital Universitario Reina Sofía de Córdoba, Córdoba, Spain.

CORRESPONDENCE: Francisco Javier Montero-Pérez C/ Gutiérrez de los Ríos, 26 14002 Córdoba, Spain E-mail: [email protected]

RECEIVED: 26-10-2011

ACCEPTED: 5-12-2012

CONFLICT OF INTEREST: None. Research Project, finalist in the 4th edition of the Awards of the Health Quality Agency of Andalusia -2010, in the category Organization of Health Services Innovation and Creativity.

We report the design of a balanced scorecard for managing an emergency department in a tertiary care university teaching hospital; data derived by implementing the scorecard system are also presented. The project was carried out in the following phases: 1) selection of indicators of activity and quality of processes and outcomes for the scorecard, 2) validation of the indicators, 3) analysis of indicators from 2007 through 2009, and 4) conclusions regarding clinical performance in relation to the indicators that make up the scorecard. For 2009, we analyzed 124 720 emergencies. The mean wait time before triage was 5.2 minutes; 31.7% of the patients waited longer. Triage took a mean of 1.5 minutes; triage took longer for 1.0% of the emergencies. Emergencies with a wait time of longer than 5 minutes before triage were distributed bimodally, with peaks at 11 A.M. to 12 noon and from 4 P.M. to 5 P.M. The rate of staff reassignments was 4.8%, and 2.8% of patients were lost. Among noncritical cases, 41.2% exceeded the maximum wait time before physician contact. The overall mortality rate was 0.24%. We conclude that a customized balanced scorecard approach allows an emergency department to manage relevant time intervals and adapt them to care standards, among other advantages of the system. The scorecard sheds light on how a department actually works and encourages the adoption of corrective measures based on analysis of results. [Emergencias 2012;24:476-484] Keywords: Balanced scorecard. Emergency health services. Triage.

Introduction Current clinical management increasingly requires useful, relevant and reliable data for accurate, timely and proactive decision-making. A balanced scorecard (BSC) is a set of indicators that provide information on the activity and quality of production of a company; it is a strategic planning tool designed to achieve objectives 1. The concept of a balanced scorecard was developed by Kaplan and Norton in 19922,3 in the financial business sector, and includes four key areas: knowledge and growth (employee competencies, motivation, working hours, etc.), internal operational processes (company-specific), economic aspects and customer perspective. In 2003, Zelman4 established the importance of a BSC for health care, although with modifications to adapt to the specific characteristics of the field of healthcare. The BSC for healthcare only incorporates the sec476

ond area of the model developed by Kaplan and Norton, i.e. measurement of all healthcare activities and intermediate procedures occurring from first medical contact to the end of attendance. Hospital emergency departments (ED) are the great unknown of the Spanish public health system, largely because their care activity is scarcely analyzed and what is actually known is not disseminated, probably because its results are highly susceptible to improvement and ED managers responsible consider such information as appropriate for internal use only. The objective of this article is to present the design and results of a BSC performed in a hospital ED (BSC-ED) to offer a real picture of what happens in an ED from the time a patient requests assistance until he/she leaves, either discharged home or admitted to hospital. Unlike other studies on this topic5-7, our work is not limited to the first phases of design, but also Emergencias 2012; 24: 476-484

BALANCED SCORECARD MANAGEMENT OF A HOSPITAL EMERGENCY DEPARTMENT

shows some activity results obtained using the indicators that make up the BSC-ED.

Scope and methodology of the BSC-ED design The study was carried out in the ED of a tertiary university hospital with 1,319 beds serving a population of 788,287 inhabitants. The ED attends approximately 125,000 emergencies per year, all adults (> 14 years). It does not attend pediatric, obstetric or gynecological patients. The staff consist of 46 physicians, 104 nurses, 74 nursing assistants and 53 orderlies, all full-time workers. In addition, 6 first-year and 2 second-year resident physicians join the permanent staff at 3 pm daily to perform 24-hour shift work on weekdays and at 9 am on Sundays and holidays. At the time the study was performed, the ED was structured as follows: a triage area where a nurse classified patients on a 4-level scale of priority (PI - PIV) using an unstructured protocol based on expert consensus since 1988, a consulting/attending area, divided into one for the critically ill (with 3 attendance positions), 11 for medical-surgical pathologies, one for trauma patients and 2 for lowest priority patients. There was and is also an observation area, comprising 31 beds and a 14 seats, with a maximum stay time of 24 and 12 hours respectively, exclusively for patients admitted from the consulting/attending area. The design involved several stages: 1) selection of appropriate indicators for the design of the BSC, 2) validation of these indicators; 3) analysis of the data in the DIRAYA-URGENCIAS computer application data-base (Version 2.4.01) for the hospital where the study was conducted during the years 2007 to 2009 inclusive, and 4) drawing conclusions and use of the indicators comprising the BSC-ED. The scorecard was designed using the application "Modulo de Tratamiento de la Información” (MTI) Version 4.0", included in the DIRAYA-URGENCIAS application, which is the management and information system for healthcare of the government of Andalucía. The design followed that of Pink et al.8, which includes the choice of indicators, the definition of hospital parameters, identification of the data source and the determination of their relative performance. Data loading in the MTI application is performed with a 1-week delay. The selection of indicators (activity, process quality and outcome) for the BSC-ED was performed in accordance with others previously deEmergencias 2012; 24: 476-484

signed5,9-11 and the possibilities for measurement offered by MTI. In addition, we designed new indicators based on expert consensus. The BSC-ED was structured in the MTI in 6 folders that included reports and these, together with the indicators, are shown in Table 1. Figure 1 graphically depicts the indicators of attendance times selected. Indicator validation was performed by random case sampling and checking to confirm if the result of its analysis was consistent with manual measurement. Design errors detected were corrected in the computer application, and a correction for marginal cases was also performed. Finally, the computer application issued the results in exportable format (Microsoft Excel®, 2007® Office package), expressed as absolute frequencies, percentages and arithmetic means. The results presented here refer to year 2009. The relevance of the various indicators comprising the BSA-ED and the conclusions were the result of expert consensus. The literature review was conducted through the Virtual Library of Andalucia, using the GERION search engine and the health data contained in the databases CINAHL, EMBASE, ERIC (USDE), IME-Biomedicine, MEDLINE, PubMed and SciELO (Scientific Electronic Library Online). For this, the keywords used in Spanish were “cuadro de mando, urgencias, triaje” and in English “balanced scorecard, emergency department, triage”, without date limitations. We excluded articles not written in English or Spanish. In addition, a manual search of the contents of the journal EMERGENCIAS was performed with the same keywords.

Results of the main indicators 1. Indicators for the triage area. Figures 2 and 3 respectively show the distribution of attended episodes according to triage priority and month, and globally for the whole attendance circuit. Mean waiting time for triage (TWT) was 5.21 minutes, with 31.7% exceeding 5 minutes, with a bimodal distribution showing a peak at 11-12 hours and another at 16-17 hours (Figure 4). Mean triage duration (TD) was 1.5 minutes, with 1% exceeding 5 minutes. The attendance circuit relocation rate was 4.8%.

2. Global attendance indicators During 2009, 125,720 urgent episodes were seen in the ED, with a mean daily rate of 344 cas477

F. J. Montero-Pérez et al.

Table 1. List of folders, reports and indicators included in the balanced scorecard 1. Folder containing triage area indicators. Report 1.1. Classification: – Patients admitted and classified by priority level and care circuit. – Mean waiting time for triage (mins). – Triage waiting time > 5 mins. – Mean duration of triage process (mins). – Triage process duration > 5 mins. Report 1.2. Time-slot distribution of episodes with triage process time > 5 min. Report 1.3. Weekly distribution of episodes with triage process time > 5 min. Report 1.4. Relocations after triage, by priority and location. 2. Folder containing global care indicators. Report 2.1. Patients by age group and sex. Report 2.2. Mean visits by time and day of the week. Report 2.3. Mean ED visits by shift, day of the week and priority. Report 2.4. Mean ED visits by day of the week and priority. Report 2.5. Mean ED visits by week of the month, day of the week and priority. Report 2.6. Workload and mean time spent at each location (ED area). Report 2.7. Distribution of emergency episodes by discharge destination, priority and location. Report 2.8. Emergencies according to district of origin, by care circuit and day of the week. Report 2.9. Nº of patients attended according to district of origin, by hour, day and priority. Report 2.10. Nº of patients attended according to district of origin, by priority and mode of ED arrival. 3. Folder containing consulting/attending area indicators. Report 3.1-3.2. Care times – Mean time waiting for first physician attendance (mins). (by priority and final location after evaluation): – Mean time being attended by a physician (mins). – Mean time spent in consulting/attending area (mins). Report 3.3. Compliance with the standard on waiting time for first physician attendance (PII> 15 '; PIII> 30; PIV> 120')*. Report 3.4. Waiting time for first physician attendance, by final consulting area location. Report 3.5. Compliance with the standard on waiting time for first physician attendance by last consulting area location, shift and day of the week (PII> 15'; PIII>30'; PIV> 120'). Report 3.6. Time spent in the ED, mean attendance time and time to leaving the consulting area (by priority and last consulting area location). Report 3.7. Time spent in the consultation area and time to leaving the ED (by priority and final consulting area location). Report 3.8. Distribution of emergency episodes by destination on discharge, priority and last location Report 3.9. Destination departments of patients admitted from the consulting area (by last location). 4. Folder containing care indicators from the observation area. Report 4.1. Care times by priority and location – Mean time waiting for first physician attendance (mins) by priority and location. (beds / seats / total): – Emergencies waiting >30 mins. – Mean time spent in observation area (mins). – Mean time before leaving the observation area (episodes without immediate discharge) (mins). – Mean time spent in the observation area (mins). Report 4.2. Emergencies exceeding maximum stay time in the observation area. Report 4.3. Distribution of admissions to observation area, according to discharge destination, priority and location (beds/seats/total). Report 4.4. Destination departments of patients admitted from the observation area. 5. Folder containing professional activity: diagnostic coding, revisits and mortality. Report 5.1. Care activity of ED physicians (I): – Discharges ordered by EP. Report 5.2. Care activity of ED physicians (II): – Admissions to destination departments by each EP. Report 5.3. Care activity of ED physicians (III): – Type of discharge destination. Report 5.4. Overall diagnostic coding. Report 5.5. Diagnostic coding by location. Report 5.6. Diagnostic coding by EP. Report 5.7. Number of revisits to ED within 72 hours. Report 5.8. Mortality in the ED. 6. Folder containing population reports. Report 6.1. Reference population general TIS by area: Nº of TIS by health area. Report 6.2. Reference population general TIS by area, Nº of TIS by district and primary care center of origin. district and primary care center of origin: Report 6.3. Reference population general TIS by area, age and sex: Nº of TIS by area, age and sex. *Priority 1 patients (critical emergencies) were excluded from the analysis. ED: emergency department; EP: emergency physician; TIS: Individual Health Card.

es. Figures 5 and 6 respectively show the distribution of episodes by sex and age group and time and weekday attendance. The overall attendance rate at our ED for the districts served by the hospital (Districts 1 and 3) was 434.86/1,000 population. Finally, Table 2 shows the distribution of episodes attended as well as attendance rate by 478

day of the week, mode of ED arrival, triage priority, shift and attendance circuit. For space reasons we have omitted the distribution of patients discharged home, but would highlight the fact that 2.8% of all patients passing the triage stage were not further attended (i.e. apparently left the ED before being seen by a physician). Emergencias 2012; 24: 476-484

BALANCED SCORECARD MANAGEMENT OF A HOSPITAL EMERGENCY DEPARTMENT

Figure 1. Definition of indicators of ED response times. *Time of leaving the Consultation and Observation areas. ED: emergency department.

3. Indicators of the consulting/attending area By way of example, we present the mean waiting time before physician attendance (TE1 ª F in Spanish) during 2007-2009, and how this indicator met the standard according to triage level of priority for the year 2009 (Figure 7).

4. Indicators of the observation area Figure 8 shows the evolution of mean time spent in the observation area, as an example of the indicators for this area of care.

5. Indicators of professional activity, diagnostic coding, revisits and mortality The overall diagnostic coding rate was 8.8%. By areas and considering only the figures for emergency physicians (excluding medical residents) this was 14.1% (0.1 to 63.2%) in the consulting/attending area and 45.5% (3.4%-80%) for the observation area. Other indicators, such as the annual rate of revisiting the ED within 72 hours was 5.3%, while the overall mortality rate was 0.24%, all with respect to the year 2009.

Discussion In the literature of health sciences there is extensive material on the theoretical framework of balanced scorecards3,4,7,8,12-14, but not so much on

Figure 2. Distribution of urgent episodes by priority level and month (in 2009). Emergencias 2012; 24: 476-484

Figure 3. Distribution of urgent episodes according to care circuit (year 2009).

479

F. J. Montero-Pérez et al.

Figure 4. Distribution of cases with triage waiting time (TWT) > 5 minutes.

their implementation and, even less containing analysis of their component indicators in order to draw conclusions for continued improvement in the quality of health services. According to Felisart et al.5, monitoring a series of indicators that cover various aspects of the urgent care process provides an overview of the quality of service, and failures or deviations from established standards. The analysis of these deviations always brings to light new opportunities for improvement in the ED. The determination of triage times, their compliance with standards15 and the time-slots when deviations occur is important for human resource adjustment16 of this nurse activity and allows testing the effectiveness of such adjustments. In our case, this indicator allowed us to detect the need for a second triage nurse for a certain time band and to adapt the physical structure and staffing thereof after a recent refurbishment. The interest in measuring the rate of internal relocations of patients (from one internal ED care circuit to another) is that they generate delays and represent inefficient human resource use

Figure 5. Population pyramid and gender distribution of cases attended. *The negative sign in the column "Men" is due to technical reasons on constructing the population pyramid in Excel.

480

Figure 6. ED visits according to daily time slots and day of the week.

since they require attendance twice; in addition this rate indirectly measures the quality of triage assignment. The proportion of relocation should be as low as possible. The reasons for relocating patients are usually: a change in priority level on

Table 2. Distribution of cases attended according to health district: rate of visits by day of the week, ED arrival mode, triage priority level, work shift and ED care circuit District 1

District 3

Nº attendances/ Attendance rate

Nº attendances/ Attendance rate

Day of the week 82,465/257.30 21,403/177.56 Monday 12,972/40.47 3,496/29.00 Tuesday 11,876/37.05 3,372/27.97 Wednesday 11,798/36.81 3,223/26.74 Thursday 11,784/36.77 3,099/25.71 Friday 11,902/37.14 3,117/25.86 Saturday 10,767/33.59 2,611/21.66 Sunday 11,366/35.46 2,485/20.62 Mode of ED arrival 82,465/257.30 21,397/177.51 Health center 65,370/203.96 13,932/115.58 Centro de salud 4,242/13.24 4,242/35.19 DDCCUU 490/1.53 71/0.59 Emergency team 11,327/35.34 2,938/24.37 Public body 170/0.53 26/0.22 Hospital 835/2.61 185/1.53 Other 31/0.10 3/0.02 Triage priority level 82,465/257.30 21,397/177.51 I 703/2.19 279/2.31 II 10,499/32.76 3,187/26.44 III 43,429/135.50 11,483/95.26 IV 27,834/86.84 6,448/53.49 Work shift 82,465/257.30 21,403/177.56 Morning 08-15 h 34,628/108.04 9,729/80.71 Afternoon 15-22 h 30,950/96.57 8,132/67.46 Night 22-08 h 16,887/52.69 3,542/29.38 Care circuit 82,020/255.91 21,262/176.39 Critical care 1,659/5.18 646/5.36 First level (mild) 17,136/53.47 4,357/36.15 Medical-Surgical 43,499/135.72 11,747/97.45 Trauma 19,726/61.55 4,512/37.43 DDCCUU = Peripheral emergency units. *The rate of attendance is per 1000 inhabitants. Emergencias 2012; 24: 476-484

BALANCED SCORECARD MANAGEMENT OF A HOSPITAL EMERGENCY DEPARTMENT

Figure 8. Evolution of mean time spent in the observation area (2007-2009).

Figure 7. Evolution of the mean waiting time for first attendance by a physician (2007-2009 inclusive) and episodes exceeding the standard22, by month and assigned triage priority shown in the table (2009). * Priority 1 patients (critical emergencies) were excluded from the analysis. Percentages are calculated on the total number of patients seen by priority.

re-evaluation of the patient by a physician and ignorance of circuit assignment on the part of novice professionals, primarily. Logically, triage indicators included in the BSCED should be considered general indicators of this activity, but they do not replace those of computer applications of structured triage systems which allow more detailed analysis. Our triage system is still of the unstructured 4-level type, typical of many Spanish EDs. The data on annual attendance provide information on ED visits made by the population 17. Knowing global demand is important, but so is knowing how and when it occurs, the distribution of patients in different circuits, mode of ED arrival, where patients come from and degree of severity etc. Knowing all these data is possible today thanks to the electronic medical record and powerful information systems, and allows better clinical management in our setting18. This study analyzes the main indicators of ED activity, some referred to in literature and others which are newly proposed. Detailed knowledge of shift work, level of priority and care circuits allows optimizing resources in the ED. Moreover, if we know the workload of each care circuit within the ED and mean time spent there for each patient, we can optimize the use of human and material resources. The dropout rate of patients (those who are not seen by a physician after triage) was Emergencias 2012; 24: 476-484

2.8%, slightly higher than the 2% maximum recommended by the Spanish Society of Accident and Emergency Medicine (SEMES)back in 200111. Other studies have reported a median rate of 3.1% (range 1.1 to 15%) 11,19,20, and our rate is therefore consistent with those described in the literature. Finally, in relation to ED use, our BSC-ED provides population data for epidemiological rate calculation (Table 2). The rate of overall attendance at our ED of the population served by our hospital (Districts 1 and 3) was 434.9/1,000 population, within the range reported in Spain (292.8 585.3/1,000) 17,21, although above the national average for 2001 (394.6/1,000 population)17, due to generally increased annual ED attendance. The differences in the attendance rates between the two districts is explained by the geographical proximity of District 1 with respect to District 3. The problem of delays in ED attendance is another facet that requires a determined approach. Detecting when they occur and what factors contribute to them requires detailed knowledge of the care process over time. Most authors22-27 consider TE1 ªF (time to first attendance by a physician) as the time lapse between admission or registration on ED arrival (the latter term being rather imprecise in our view) and actually being seen by a physician. In the present study, we considered that this time interval begins with triage classification by nurses entered in the computer application. And in reality, until a patient is classified by the triage nurse, he/she cannot be attended by a physician, and the time spent performing triage classification is not known, so we believe that this indicator must be calculated as proposed in this study. This study examined indicators of activity, effectiveness and safety. In the model of clinical management through participatory management 481

F. J. Montero-Pérez et al.

by objectives28, as important as measuring the activity and indicators of quality of the service as a whole is to do the same thing for each of its professionals, in an objective manner as in an information system. This allows knowing the degree to which each ED professional participates in achieving the objectives, which acts as a great personal incentive. It might be argued that these indicators do not truly and completely reflect this activity. We agree, but it is a start and every journey begins with a first step. And this first step is not a bad start. Other indicators, such as individualized pharmaceutical prescription (also used in our ED but not included in the proposed BSC-ED), the degree of adherence to current clinical guidelines, the ordering of laboratory and imaging tests, individualized mortality and 72h-revisit rates, number of complaints etc. could be used to complete individual profiles on performance, effectiveness and efficiency of each professional. ED diagnostic coding is a good indicator of quality that facilitates management by those responsible for the service, teaching and scientific work of all its members and allows feedback to professionals11. The standard is set at 100%5,11. Diagnostic coding is considered difficult to perform during ED working hours by many professionals, and a hard habit to acquire and perform systematically on the part of the physician him/herself. Furthermore, in an ED with resident physicians from other specialties, achieving a rate of 100% is a real challenge. Our very poor overall coding rate (8.8%), obtained during the first year of coding in our ED, reflects this. Analysis of diagnostic coding by care areas within the ED showed great variability, suggesting different motivation to perform the task. In fact, in the year following the study period, after the management included coding as an objective with incentives, the overall diagnostic coding rate exceeded 50% in the consulting/attending area. Regarding the observation area, the rate was considerably higher; the explanation is that coding was actively encouraged in this area and the nature of the work there allowed more time for the task of coding. Considering the rate reported by Gomes et al.19 (99.9%), although in a small hospital, there is obviously much room for improvement in this regard, Cleary, the successful implementation of diagnostic coding depends on professionals receiving some kind of feedback so that it becomes seen as a tool, e.g. for the purposes of teaching and research. We are currently initiating the application of ICD-9 diagnoses for these purposes, and one of our objectives is to in482

clude this in our BSC-ED when this practice is consolidated in our unit. The rate of ED revisits within 72 hours, or revisit index, allows us to gauge the effectiveness of treatment received on first attention. Revisits may indicate either poor effectiveness or the development of unforeseen complications11. Some authors consider the standard for this indicator should be 2.5-5% of patients19,29, but there is no consensus as to whether this should only include revisits for the same reason or not11,30. In the present study the revisit rate was slightly above the higher standard and it remained stable throughout the months of study, suggesting that it is not dependent on ED visits, prevalent diseases or other factors such as the incorporation of medical residents in the month of June. Interestingly, the monthly rate was slightly higher in June. In our setting, Miró et al.30 reported a revisit rate of 1.4% and Gomes et al.19 reported 1.6%. The Estudi on Activitat l’i l‘Organització dels serveis d'urgències hospitalaris31 which included all hospitals in the Network of Public Hospital of Catalonia reported an ED readmission rate within 72 hours of 6%. There is therefore wide variability for this indicator in the areas of our country where it has been analyzed. Mortality rate indirectly measures ED capacity and effectiveness at rapidly dealing with critical cases, either in the ED itself or referring the patient to a reference center11. There is however no clearly established standard. Gomes et al.19 reported a mortality rate of 0.03% (with a proposed standard of less than 0.2%) and Rodriguez Maroto et al.32 reported 0.1%. In both cases, the hospitals involved offered reduced care options. Mortality rates from other studies33-37 range between 0.1 to 0.4%, which includes our overall mortality rate. Finally we would highlight the importance of tools such as the BSC-ED described here, to provide more detailed information for better ED management, considering workload and how it fluctuates according to times and the various phases of the urgent care process. Its use will undoubtedly lead to improvements in ED organization and resource assignment, and hopefully help banish the word chaos from the description of many Spanish EDs.

References 1 Villalbí J, Guix J, Casas C, Borrell C, Durán J, Artazcoz L, et al. El cuadro de mando integral como instrumento de dirección en una organización de salud pública. Gac Sanit. 2009;23:14-5. 2 Kaplan R, Norton DP. The balanced scorecard-measures that drive performance. Harv Bus Review. 1992;70:71-9.

Emergencias 2012; 24: 476-484

BALANCED SCORECARD MANAGEMENT OF A HOSPITAL EMERGENCY DEPARTMENT

3 Kaplan R, Norton DP. Putting the balanced scorecard to work. Harv Bus Rev. 1993;71:134-9. 4 Zelman WN, Pink GH, Matthias CB. Use of the balanced scorecard in health care. J Health Care Finance. 2003;29:1-16. 5 Felisart J, Requena J, Roqueta F, Saura RM, Suñol R, Tomás S. Servicios de urgencias: indicadores para medir los criterios de calidad de la atención sanitaria. Barcelona: Agéncia d`Avaluació de Tecnología i Recerca Mèdiques. Barcelona: Servei català de la Salut. Departament de Sanitat i Seguretat Social. Generalitat de Catalunya; 2001. 6 Huang SH, Chen PL, Yang MC, Chang WY, Lee HJ. Using a balanced scorecard to improve the performance of an emergency department. Nurs Econ. 2004;22:140-6. 7 Groene O, Brandt E, Schmidt W, Moeller J. The Balanced Scorecard of acute settings: development process, definition of 20 strategic objectives and implementation. Int J Qual Health Care. 2009;21:25971. 8 Pink GH, McKillop I, Schraa EG, Preyra C, Montgomer, Baker GR. Creating a balance scorecard for a hospital system. J Health Care Fin. 2001;27:1-20. 9 Grupo de Trabajo SEMES-Insalud. Calidad en los Servicios de Urgencias. Indicadores de calidad. Emergencias. 2001;13:60-5. 10 Tejedor Fernández M, Burillo Putze G, Roqueta Egea F, Jiménez Murillo L, Leciñena Esteban MA, Carretero Alcántara L. Gestión clínica en un Servicio de Urgencias. Todo Hospital. 2003;462:70. 11 Sociedad Española de Medicina de Urgencias y Emergencias (SEMES). Manual de indicadores de calidad para los Servicios de Urgencias de hospitales. 2ª ed. Madrid: SANED; 2009. 12 Walker KB. Improving hospital performance and productivity with the balanced scorecard. Acad Health Care Manag J 2006. (Consultado 1 Julio 2011). Disponible en: http://findarticles.com/p/articles/ mi_m1TOQ/is_2/ai_n25009491/. 13 Voelker K, Rakich R, French G. The balanced scorecard in healthcare organizations: a performance measurement and strategic planning methodology. Hosp Top. 2001;79:13-24. 14 Wells R, Weiner B. Using the balanced scorecard to characterize benefits in the safety net. Health Services Management Research. 2005;18:109-23. 15 Sociedad Española de Medicina de Urgencias y Emergencias (SEMES). Sistema Español de Triaje (SET). Madrid: Edicomplet; 2004. 16 Sociedad Española de Medicina de Urgencias y Emergencias (SEMES). Estándares de acreditación para Servicios de Urgencias de hospitales, 2ª ed. Madrid: Grupo Saned; 2008. 17 Sociedad Española de Medicina de Urgencias y Emergencias. Urgencias sanitarias en España: situación actual y propuestas de mejora. Granada: Escuela Andaluza de Salud Pública; 2003. 18 Busca P, Marrón R. La informatización en urgencias y emergencias. An Sist Sanit Navar. 2010;33:(Supl. 1):69-76. 19 Gómes Jiménez J, Faura J, Burgues L, Pámies S. Gestión clínica de un servicio de urgencias hospitalario: indicadores de calidad, benchmarking y análisis de la casuística (cas-mix). Gestión Hospitalaria. 2004;15:3-12. 20 Guttmann A, Schull MJ, Vermeulen MJ, Stuket TA. Association between waiting times and short term mortality and hospital admission after departure from emergency department: population based cohort study from Ontario Canada. BMJ. 2011;342:2983. 21 Estadística de Establecimientos Sanitarios con Régimen de Internado. Indicadores hospitalarios. Año 2007. Instituto de Información Sanitaria. AC-SNS. Madrid: Ministerio de Sanidad y Política Social; 2007. 22 Dirección General de Asistencia Sanitaria. Subdirección de Asistencia Especializada–Gestión Sanitaria Plan Andaluz de Urgencias y Emer-

Emergencias 2012; 24: 476-484

gencias. Plan funcional de la sección de urgencias del servicio de cuidados críticos y urgencias. (Consultado 20 Julio 2011). Disponible en: http://www.sas.junta-andalucia.es/contenidos/gestioncalidad/ PlanAndUrgenciasEmergencias/Plan_Func_Urgencias_SCCU/Plan_Fun c_Urgencias_SCCU.pdf. 23 Horwitz LI, Bradley EH. Percentage of US Emergency Department Patients Seen Within the Recommended Triage Time. 1997 to 2006. Arch Intern Med. 2009;169:1857-65. 24 Álvarez Álvarez B, Gorostidi Pérez J, Rodríguez Maroto O, Antuña Egocheaga A, Alonso Alonso P. Estudio del triaje y tiempos de espera en un servicio de urgencias hospitalario. Emergencias. 1998;10:100-4. 25 Carbonell Torregrosa MA, Girbés Borrás J, Calduch Broseta JV. Determinantes del tiempo de espera en urgencias hospitalarias y su relación con la satisfacción del usuario. Emergencias. 2006;18:30-5. 26 Wilper AP, Woolhandler S, Lasser KE, McCormick D, Cutrona SL, Bor DH, et al. Waits To See An Emergency Department Physician: U.S. Trends And Predictors, 1997-2004. Health Affairs 2008;27:84-95. 27 US Government Accountability Office. Hospital Emergency departments: crowding continues to occur, and some patients wait longer than recommended time frames. (Consultado 20 Julio 2011). Disponible en: http://www.gao.gov/highlights/d09347high.pdf. 28 Díaz Garrido E, Pinillos Costa MJ, Soriano Pinar I. Cátedra Madrid Excelente. Dirección por objetivos: la dirección participativa. Madrid: Cátedra Madrid Excelente. (Consultado 9 Agosto 2011). Disponible en: http://www.madridexcelente.com/files/5ce0f7e0c060.pdf. 29 Plan Andaluz de Urgencias y Emergencias. Manual de indicadores de actividad y calidad para urgencias y emergencias sanitarias. Sevilla: Junta de Andalucía. Consejería de Salud. Servicio Andaluz de Salud, 2000. 30 Miró O, Jiménez S, Alsina C, Tovillas-Morán FJ, Sánchez M, Borrás A, Millá J. Revisitas no programadas en un servicio de urgencias de medicina hospitalario: incidencia y factores implicados. Med Clin (Barc). 1999;112:610-5. 31 Estudi sobre l’activitat i L’organització dels serveis D’urgències hospitalàris. Informe final. Generalitat de Catalunya. Departament de Salut, 2005. (Consultado 5 Agosto 2011). Disponible en: http://www.gencat.cat/salut/depsalut/html/ca/dir2385/estudi_urgencies.pdf. 32 Rodríguez Maroto O, Llorente Alvarez S, Casanueva Gutiérrez M, Alvarez Alvarez B, Menéndez Somoano P, De la Riva Miranda G. Mortalidad en un Servicio de Urgencias Hospitalarias. Características clínico epidemiológicas. Emergencias. 2004;16:17-22. 33 Miró O, De Dios A, Antonio MT, Sánchez M, Borrás A, Millá J. Estudio de la mortalidad en un servicio de urgencias hospitalario: incidencias, causas y consecuencias. Med Clin (Barc.). 1999;112:690-2. 34 Requena López A, Lando Tesan JF, Gamollón Rubio LG, Parrilla Herranz P, Franco Sorolla JM, Moreno Vernis M. Análisis de la mortalidad en las áreas de observación de urgencias. Emergencias. 1998;10(extra):208. 35 Córdoba Victoria A, Delgado Lozano LC, Cabrera Vélez R, Kessler P, Perpiña C, Castro C, et al. Estudio de la mortalidad en el Servicio de Urgencias del Hospital 12 de Octubre durante 1989. An Med Interna. 1991;8:487-90. 36 Sahuquillo Llamas JC, Tudela Hita P, Segura Egea A, Estrada Cuxarto O. Análisis de la mortalidad en el Servicio de Urgencias de un Hospital General. Emergencias. 2000;12(extra):377. 37 Nieto Sánchez A, Arranz Gómez F, Lana Soto R, Torres Villaredo P, Rodríguez Carrillo M, Jiménez de Diego L. Análisis descriptivo de la mortalidad en un servicio de urgencias terciario. Emergencias. 2000; 12(extra):291.

483

F. J. Montero-Pérez et al.

Gestión clínica de un servicio de urgencias hospitalario mediante un cuadro de mando asistencial específico Montero-Pérez FJ, Calderón de la Barca Gázquez JM, Jiménez Murillo L, Quero Espinosa FB, Gracia García F, Roig García JJ Este estudio da a conocer el diseño y, a modo de ejemplo, algunos de los resultados de la implantación de un cuadro de mandos asistencial en un servicio de urgencias hospitalario (CMA-SUH) como herramienta para la gestión clínica en un hospital universitario de tercer nivel. El estudio constó de varias fases: 1) selección de indicadores para diseño de un CMA; 2) validación de dichos indicadores; 3) análisis de los indicadores durante los años 2007 a 2009; y 4) elaboración de conclusiones asistenciales y pertinencia de la explotación de los diversos indicadores que constituían el CMASUH. Se analizaron 125.720 episodios urgentes correspondientes al año 2009. El tiempo medio de espera para el triaje (TET) fue de 5,2 minutos, con una proporción de casos con TET mayor de 5 minutos de 31,7%. El tiempo de duración del triaje (TDT) medio fue de 1,5 minutos, con una proporción de casos con TDT mayor de 5 minutos de 1,0%. La proporción de episodios asistenciales con TET mayor de 5 minutos mostró una distribución bimodal con un pico a las 11-12 horas y otro a las 16-17 horas. La tasa de reubicación de circuitos asistenciales fue de 4,8% y la de fugas del 2,8%. Un 41,2% de los episodios no críticos superaron el estándar de demora máxima para la primera consulta facultativa. La tasa global de mortalidad fue de 0,24%. Se concluye que un CMA SUH permite gestionar, entre otros aspectos, los tiempos asistenciales y su adecuación a los estándares, propicia la adopción de medidas correctoras una vez analizados los resultados, arroja luz al conocimiento de cómo funcionan realmente los SUH y constituye, pues, una potente herramienta de gestión. [Emergencias 2012;24:476-484] Palabras clave: Cuadro de mando. Urgencias. Triaje.

484

Emergencias 2012; 24: 476-484