Dynamic network data envelopment analysis for university hospitals evaluation

Lobo, Maria Stella de Castro; Rodrigues, Henrique de Castro; André, Edgard Caires Gazzola; Azeredo, Jônatas Almeida de; Lins, Marcos Pereira Estellita

doi:10.1590/S1518-8787.2016050006022

Acessibilidade / Reportar erro

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

Public Health Practice • Rev. Saúde Pública 50 (0) • 2016 • https://doi.org/10.1590/S1518-8787.2016050006022 copy

Dynamic network data envelopment analysis for university hospitals evaluation

Authorship SCIMAGO INSTITUTIONS RANKINGS

ABSTRACT

OBJECTIVE

To develop an assessment tool to evaluate the efficiency of federal university general hospitals.

METHODS

Data envelopment analysis, a linear programming technique, creates a best practice frontier by comparing observed production given the amount of resources used. The model is output-oriented and considers variable returns to scale. Network data envelopment analysis considers link variables belonging to more than one dimension (in the model, medical residents, adjusted admissions, and research projects). Dynamic network data envelopment analysis uses carry-over variables (in the model, financing budget) to analyze frontier shift in subsequent years. Data were gathered from the information system of the Brazilian Ministry of Education (MEC), 2010-2013.

RESULTS

The mean scores for health care, teaching and research over the period were 58.0%, 86.0%, and 61.0%, respectively. In 2012, the best performance year, for all units to reach the frontier it would be necessary to have a mean increase of 65.0% in outpatient visits; 34.0% in admissions; 12.0% in undergraduate students; 13.0% in multi-professional residents; 48.0% in graduate students; 7.0% in research projects; besides a decrease of 9.0% in medical residents. In the same year, an increase of 0.9% in financing budget would be necessary to improve the care output frontier. In the dynamic evaluation, there was progress in teaching efficiency, oscillation in medical care and no variation in research.

CONCLUSIONS

The proposed model generates public health planning and programming parameters by estimating efficiency scores and making projections to reach the best practice frontier.

Hospitals, University, organization & administration; Efficiency, Organizational, economics; Programming, Linear, utilization; Health Services Evaluation

Dimension	Variable	Observation	Type
Care	Number of beds	Total number of active beds per year.	Input
	Technology sum index, or Service mix	Sum of equipment and technology reported to the National Registry of Health Institutions. The first two variables, beds and service mix, can be jointly treated as capital proxyl¹².	Input
	Human Resources (total, except doctors)	Number of staff, except doctors. Considers total workforce, excluding doctors. This variable highly correlates with the number of doctors, which will be considered in another dimension of the model.	Input
	Outpatient visits adjusted by complexity	Number of adjusted annual outpatient visits. A Complexity Index was created for adjustment, based on the percentage of high complexity procedures performed at outpatient level, then multiplied by the volume of visits.	Output
	Output budget (goals agreement with manager)	The annual output budget (in millions of reais) is the input defrayal for one year and the proxy output measure for the following year. Since negotiation of budget values is based on the historical output series of previous years, it is useful to program a more efficient funding system.	Carry-over
Teaching	Doctors	Number of doctors involved in service training activities in wards and outpatient departments.	Input
	Number of multi-professional residents	Total number of multi-professional residents only.	Output
	Sum of undergraduate students in all health areas	Total number of undergraduate students in all health professions in activity at the hospital. This output aims to emphasize the view of multidisciplinary activities.	Output
	Professor without doctoral degree	Number of professors with a masters, specialization or bachelor degree, mainly dedicated to teaching. Fixed variable, given the low variability from one year to another and the low control of the hospital management over the number of places.	Carry-over
Research	Number of students in graduate courses	Total number of enrolled graduate students.	Output
Research	Professor with doctoral degree	Number of professor with a doctoral degree. Fixed variable, given the low variability from one year to another and the low control of the hospital management over the number of places. Following the logic of not repeating a variable in more than one dimension, PhD professor is considered to be mostly dedicated to research.	Carry-over
Care - Teaching	Number of medical residents	Care input and teaching output. These doctors play an important role in the care workforce, and, at the same time, are new graduates undergoing service training.	Link
Care - Research	Admissions adjusted by complexity	Care output and research input. Annual admissions are the main care output and, at the same time, the basis for the development of clinical research and care technologies in the hospital. Complexity adjustment considers the percentage of high complexity procedures performed in the admissions unit.	Link
Research - Teaching	Research projects	Teaching input and research output. Annual number of projects approved by CEP (Research Ethics Commission). The results are the main research output and, at the same time, they generate knowledge to be transmitted in the classroom.	Link

Year	Measure	R/C Link	Care dimension (C)				T/C Link	Teaching dimension (T)				T/R Link	Research dimension (R)

		Adjusted admissions (per 100)	(I) Technological structure	(I) Beds	(I) Human resources	(O) Adjusted outpatient visits (per 1,000)	Medical residents	(I) Doctors	(C) Professor without PhD	(O) Multi-professional residents	(O) Undergraduate students	Research projects	(C) Professor with PhD	(O) Graduate students
2010	Mean	132	822	295	1,708	308	159	292	83	21	1,127	184	112	256
	Maximum	597	2,355	745	4,217	2,373	792	1,084	151	68	2,327	1,868	1,002	3,362
	Minimum	3.5	31	119	501	3	8	71	18	7	197	5	5	4
	SD	157	712	162	1,051	499	151	225	38	15	484	328	175	610
2011	Mean	137	838	299	1,724	295	167	299	102	38	1,230	179	111	300
	Maximum	585	2.395	737	4,291	1.973	816	1.159	635	90	2,781	1,722	568	3,413
	Minimum	1	21	93	484	2	16	82	26	17	408	13	10	3
	SD	145	666	163	1,083	405	155	235	105	19	652	312	111	626
2012	Mean	131	916	2,856	1,662	310	177.6	287	92	50	1,213	188	111	206
	Maximum	498	2,478	734	4,557	1.842	867	1.230	487	123	3,063	1,609	575	2,246
	Minimum	0	35	96	303	0	19	19	21	7	235	6	17	1
	SD	133	748	155	1,072	442	162	243	81	31	644	298	111	405
2013	Mean	143	954	284	1,677	333	194	299	98	61	1,283	65	130	260
	Maximum	537	2,483	732	4,750	1,740	1,058	1.202	740	284	3,198	588	553	2,416
	Minimum	1	41	97	598	1	27	69	11	12	57	2	7	3
	SD	161	759	151	1,073	463	191	241	126	61	704	103	121	484

DMU	Overall score	Care score	2010			2011			2012			2013

			Value*	Projection	Dif.	Value*	Projection	Dif.	Value*	Projection	Dif.	Value*	Projection	Dif.

	%	%			%			%			%			%
UFGD	100	100	0.6	0.6	0	2.6	2.6	0	21.2	21.2	0	12.3	12.3	0
UFMA	100	100	64.8	64.8	0	80.5	80.5	0	181.9	181.9	0	140.3	140.	0
UFS	100	100	8.8	8.8	0	13.8	13.8	0	18.1	18.1	0	21.2	21.2	0
UFU	100	100	85.0	85.0	0	88.3	88.3	0	103.4	103.4	0	98.4	98.4	0
UNIFESP	100	100	135.9	135.9	0	119.5	119.5	0	176.2	176.2	0	181.9	181.9	0
HCPA	88.8	100	123.6	123.6	0	167.5	167.5	0	186.8	186.8	0	155.0	155.0	0
UFRJ	87.2	100	60.0	60.0	0	63.9	63.9	0	57.2	57.2	0	85.5	85.5	0
FURG	64.4	89.1	15.4	16.7	8.3	18.7	18.7	0	29.0	29.0	0	30.2	30.2	0
UFC	50.5	48.7	43.4	38.4	-11.5	55.6	48.0	-13.5	45.2	38.0	-15.8	62.1	52.9	-14.8
UFPR	49.6	56.0	81.9	74.3	-9.3	149.2	78.8	-47.2	104.7	104.5	-0.1	106.4	106.9	0.5
UFPE	47.2	41.1	68.6	44.7	-34.9	56.2	51.0	-9.3	48.6	61.6	26.8	96.9	66.4	-31.5
UFMT	45.2	76.6	11.7	11.7	0.0	18.8	18.8	0.0	16.8	16.8	0	13.5	13.5	0.0
UnB	44.0	35.8	49.4	29.2	-40.9	76.7	35.6	-53.6	58.6	42.8	-27.0	45.2	54.0	19.5
UFTM	39.2	20.4	46.3	42.0	-9.3	58.2	53.5	-8.2	16.2	55.9	245.8	63.5	53.1	-16.4
UFAL	39.2	87.8	21.2	21.2	0	21.2	21.2	0	22.6	22.6	0	16.1	16.1	0
UFG	38.2	54.8	38.5	44.7	16.0	63.1	44.3	-29.8	56.4	46.0	-18.3	58.8	46.0	-21.6
UFES	34.1	17.9	42.8	31.5	-26.4	48.9	40.4	-17.5	46.4	46.1	-0.6	33.3	31.3	-6.1
UFBA	31.3	47.6	40.5	38.9	-3.9	69.3	49.6	-28.5	62.7	52.1	-16.8	70.2	40.9	-41.8
UFAM	28.8	77.8	14.1	14.1	0	33.4	33.4	0	23.2	23.2	0	19.6	19.6	0
UFF	18.3	12.3	27.7	37.9	37.0	47.3	38.1	-19.6	39.6	44.2	11.7	31.1	55.0	76.6
UFCG	17.6	63.1	10.4	15.0	44.0	16.8	20.9	24.1	11.6	27.2	133.6	24.7	28.0	13.4
UFSC	17.0	35.8	30.8	40.1	30.4	59.3	45.9	-22.7	45.3	36.8	-18.9	49.4	39.0	-20.9
UFMS	11.0	33.5	28.2	30.3	7.4	52.7	31.1	-41.0	22.1	25.8	16.6	52.7	10.2	-80.7
UFSM	10.9	33.6	36.8	42.3	15.0	61.1	51.1	-16.5	74.5	66.9	-10.2	73.2	53.8	-26.5
UFPel	10.6	100	25.2	25.2	0	26.9	26.9	0	32.1	32.1	0	46.8	46.8	0
UFJF	8.9	100	11.9	11.9	0	23.2	23.2	0	27.4	27.4	0	28.2	28.2	0
UFRN	5.2	26.1	13.9	24.6	77.0	13.9	26.3	89.2	19.6	31.8	62.1	9.3	32.1	246.8
UFPA	1.8	3.2	26.6	29.0	9.0	41.4	32.5	-21.5	38.6	37.6	-2.6	30.0	31.0	3.4
UFMG	1.3	22.4	79.3	77.4	-2.4	99.7	81.1	-18.7	135.8	111.8	-17.7	128.6	120.2	-6.5
UNIRIO	1.0	1.7	32.0	28.8	-10.2	2.5	27.0	989.4	32.0	30.7	-4.3	2.5	39.3	1.478.9
UFPB	0.5	26.1	29.0	31.7	9.3	27.2	36.6	34.5	17.7	32.4	82.9	10.3	9.7	-5.7
Total	-	-	1,304.2	1,280.1	-1.9	1,677.4	1,469.6	-14.1	1,771.5	1,788.1	0.9	1,797.1	1,718.8	-4.6

Faculdade de Saúde Pública da Universidade de São Paulo Avenida Dr. Arnaldo, 715, 01246-904 São Paulo SP Brazil, Tel./Fax: +55 11 3061-7985 - São Paulo - SP - Brazil
E-mail: revsp@usp.br

Acompanhe os números deste periódico no seu leitor de RSS

[1] Correspondence
: Maria Stella de Castro Lobo. Hospital Universitário Clementino Fraga Filho – UFRJ/SEAV. R. Professor Rodolfo Rocco, 255 5º andar Sala 5 a 26 Cidade Universitária – Ilha do Fundão 21941-913 Rio de Janeiro, RJ, Brasil. E-mail: clobo@hucff.ufrj.br