Performance Analysis of Hospitals in Kerala Using Data Envelopment Analysis Model

Abstract

Performance analysis in any industry plays a vital role in understanding the current scenario and thereby improving the overall efficiency. Using a sample of 20 hospitals randomly selected in Kerala, performance measures of quality were examined as they related to technical efficiency. Efficiency scores for the study hospitals were computed using data envelopment analysis (DEA). The study found that the technically efficient hospitals were performing well as far as quality measures were concerned. DEA can be used to benchmark both dimensions of hospital performance, that is, technical efficiency and quality. The variables selected for the study were divided under input and output measures. Using the DEA model, the factors considered were weighed and analysis was done. The input variables under study are bed number, trained medical staff and services offered. The output variables considered were outpatient rate, mortality rate and number of surgical operations in a month. Through the study, performance of each hospital is measured, and it aims to find out a relation between the input and output variables.

Keywords

Hospital performance input measure output measure data envelopment analysis Kerala

Introduction

Kerala is one state in India where health indicators remain on the high end. Even before India became an independent country, the maharajas of the kingdoms of Travancore and Cochin had made huge contributions in the domain of healthcare through specific policies and targeted efforts. Even among the other developing nations, Kerala stands out to be one of the few regions that has achieved substantial progress in the realm of health (Patrick, 2017). The state has substantial demographic morbidity, mortality, epidemiological and health transitions, which follow a pattern that is similar to many advanced countries. Kerala has become a model Indian state viewed in terms of low birth and death rates, low infant and maternal mortality rates, high life expectancy at birth and favourable sex ratio among other achievements (Kumar, 1993). These have been made possible through a robust public healthcare system coupled with charitable medical institutions in the private sector.

Kerala has witnessed a significant increase in the health indicators in the past two decades, which owe a great deal to the private sector. Private sector hospitals have grown immensely, surpassing the Indian standards and competing with the international standards (Kutty, 2000). The private health sector consists of various health services provided by Non-governmental Organizations (NGOs), charitable institutions, missions, trusts and various types of practitioners and institutions. The corporative medical institutions are also added to this sector. The licensed practitioners ranging from general practitioners (GPs) to the super specialists, various types of consultants, nurses and paramedics, licentiates and rural medical practitioners (RMPs) are all included in this healthcare system (Nandraj & Duggal, 1996). Expansion of private medical sector is due to the interruption between the intrinsic worth of service the public and private medical sectors can contribute. Most of the patients preferred to utilize the private medical sector because the care provided at the public sector did not fulfil them (Lekshmi, 2014). This inefficiency of the public facilities paved the way for the expansion of the private medical care set-up in the state, which has brought about the commercialization and the commoditization of healthcare (Basu et al., 2012). Through this article, we aim to focus on the performance of 20 private sector hospitals in Kerala in areas such as service provided, the facilities offered, the care offered by the medical practitioners, the number and capability of surgical operations and also the satisfaction level of the customers, in this case, the patients, to understand the actual performance achieved by them and the part played by them in the healthcare industry in Kerala.

In healthcare sector, the efficiency can be determined by using the four E formula, that is, Economy, Efficiency, Effectiveness and Equity (Peacock et al., 2001). For the purpose of our study, we are considering only the efficiency aspect, which depends upon the input and output under consideration. In this scenario, to provide better healthcare facilities, it is important to understand the performance of the hospitals. To satisfy this cause, defining performance becomes primarily important. To define ‘Performance’, first, one must understand the criteria for measuring performance. Only then can the performance of individual hospitals be identified and measured. In this article the criteria identified for measuring performance are facilities and amenities provided by the hospital, the care and efficiencies offered by the medical practitioners and the output generated as a result like the outpatient rate and number of surgical operations successfully performed (Clement et al.,2008). This article also focuses on the opinion of the customers, in this case, the patients, to define and understand the performance of hospitals and the level of satisfaction obtained as a consequence which directly measures performance.

To measure performance, methods and models are available in plenty. But the efficiency and accuracy in measuring these bring about the need of using a simple yet powerful tool like data envelopment analysis (DEA) (Hollingsworth, 2008; Hollingsworth, 2016; Kohl et al., 2018). Conventional performance measurement system brings about a very unbalanced picture of performance that can lead to misinterpretation and loosing of important opportunities for improvement. Until recently, the most common methods of comparison or performance evaluation were regression analysis and stochastic frontier analysis. These measures have now been classified inadequate due to the multiple inputs and outputs related to different resources, activities and environmental factors and their lack of dealing with multiple items simultaneously (Chirikos & Sear, 2000). In DEA study, efficiency of an organization is calculated relative to the group’s observed best practice. In this study, utilizing DEA method, the performance of various hospitals is measured and analysed. Technical efficiency is measured using DEA, which indicates the maximum production with the technology currently present, and technical efficiency can be bettered using three methods—producing more output with the same input, maintaining the same output using less input or improving production by using advanced technology. The first section gives an introduction to the study conducted. The second section describes the literature of study undertaken. The third section details about the methodology used, describing in detail on DEA model. The fourth section discusses the inferences and the fifth section concludes.

Literature Review

Kerala has attained extensive growth in the field of education and health and is also one of the fastest developing states with a comparatively low level of per capita income and the state domestic product. Kerala also has attained a remarkable place and is regarded as a model state due to the low birth and death rates, infant and maternal mortality rates, high life expectancy at birth, favourable sex ratio and triumphs in the public health surveillance. The change in nature of the system from the traditional illness structure to the modern neoplastic diseases was in check with the sickness pattern (Kannan et al., 1991; Ramachandran, 1996.

The two main medical systems, namely Unani and Ayurveda, are a lot unique in several aspects, and they have greatly contributed and enriched the material, medical and pharmacopoeia towards the improvement and betterment in diagnostic as well as therapeutic skills.

The infrastructure and other additional facilities provided in the healthcare during the British rule in the country have been significantly lower than the counterparts. In 1925, The local maharaja’s fared better with respect to the number of institutions and number of beds per 100,000 population in comparison to the British rule, namely in Mysore and Travancore (Sadanandan, 2001).

Lack of vision and development of strategy were clearly visible in the healthcare infrastructure in British-ruled Malabar. The improper utilization of the public healthcare services was due to the institutional and cultural factors that were passed down from tradition in Kerala. The policy of small-pox vaccination was not very effective in colonial Malabar compared to Travancore. Female education and their restriction to participation were also causes as the colonial administrators were hesitant to appoint women vaccinators (Kabir & Krishnan, 1992).

Macro, Meso and Micro are three levels in the healthcare sector through which efficiency is measured (Hakkinen & Joumard, 2007; Joumard et al., 2010). Macro level mainly focuses in the national or the regional healthcare systems. The efficiencies across all boundaries over time are measured and compared to infer (Afonso & St. Aubyn, 2006; Berger & Messer, 2002; Retzlaff-Roberts et al., 2004; Spinks & Hollingsworth, 2009).

Meso-level mainly focuses on the amenities and organizations which aid the healthcare services. For example: acute care hospitals and the ones at rural and district level (Athanassopoulos & Gounaris, 2001; Butler & Li, 2005; Hofmarcher, Paterson, & Riedel, 2002; Hollingsworth, 2008; Kirigia et al., 2004; Osei et al., 2005; Rebba & Rizzi, 2007). Micro-level in the end on the line mainly focuses on the precise disease to measure it for efficiency and the quality improvement programme at the healthcare (Chilingerian, 1995; Siciliani et al., 2013).

The origins of efficiency or benchmarking trace back to Farrell’s (1957) study, theoretical development of the DEA approach was by Charnes et al. (1978), who produced a measure of efficiency for decision-making units (DMU). One of the main approaches is based upon the calculation of individual partial indicators that can be developed as ratios like the responsiveness of each healthcare system, expended resources. This ratio provides measurements for different sub-dimensions of efficiency at the macro- or the system level (Davis et al. 2004; Storto, 2017). The first person to use DEA for calculating the overall hospital efficiency was Sherman (1984).

DEA is a technique which is based on non-parametric linear programming that builds and develops an efficiency frontier by optimizing the weighted output to input ratio of each provider. The efficiency measures of a DMU can be obtained as the maximum of a ratio of weighted output to weighted input subject to the condition that similar ratios for every DMU be less than or equal to unity (Charnes et al., 1981).

The key advantage of DEA is that it does not require the strong assumptions and is even able to replicate the results with the small samples, which is why it is popularly used for the benchmarking analysis (Storto, 2017; Jafarov & Gunnarson, 2008; Mirmirani et al., 2008; Verhoeven et al., 2007).

The major spending in Kerala was for health from the very early years. We can even spot hospitals that are 50 years older in Kerala. The two main aspects that have played a key role in the upbringing of the health status of the state is the ease of accessibility and coverage (Gangadharan, 2005).

‘Good Health at Low Cost’ is an apt title suitable for the state of Kerala that was achieved through collective handiness and act of government healthcare delivery system to the bottom of the poverty-stricken sections of society. And also the opposition from the government amenities assists in the form of an important element in defining the treatment cost in privatized hospitals (Kunhikannan & Aravindan, 2000).

The spreading of awareness and education in Kerala among the women with regard to health realization has raised the bar high and the focus on attaining good health is growing. The blend of healthcare with tourism in order to meet the international standards has encouraged private hospitals to offer quality services (Soman, 2007).

We can always contradict that there are numerous factors that promote as well as degrade the growth of the private healthcare institutions in India. The growing interest for the benefit is always accounted in the open. Public and private sectors should join hands and work together to treat patients in the near future at least for the betterment of society as the diseases are spreading and many are affected (Chattarjee, 2002). Dilip (2008) made an attempt to apprehend the traits of the private hospitals and their fairness in assessing the service with the aid from the data for the period from 1986 to 2004 using the secondary sources. The derived result was that there was not much expansion in terms of number, but, surely, there was a meaty alliance between large hospitals.

Nabae (1997) has looked into the past triumphs and evaluated in terms of the new challenges that popped up and suggested some measures for the public sector to overcome the challenges, which did not bother the private sector as much in the healthcare system in Kerala. First, the tax revenue must be increased, followed by streamlining of the system via decentralization, and a way needs to be found for the public and private sectors to match each other and work towards a common goal, that is, to meet the needs of the people. The state must invest in the public sector to regenerate the system.

Figure 1.

Source: Nutley and Smith (1998).

Methodology

While healthcare is a very sensitive and primary subject in society, specific methodology has to be followed to measure the aspects of performance. Here, we have followed a four-step process (Nutley & Smith, 1998) which is detailed below (see Figure 1).

Stage 1: Measurement

The purpose of study was identified in order to begin the process. The purpose includes identifying the current scenario of healthcare and the part played by private sector in its growth. Also, the aim was to identify the efficiency of randomly selected hospitals and the improvements to be made in order to better them. Next step is to identify the elements to be measured. Commonly used dimensions include health outcomes, cost-effectiveness, quality, safety, finance, patient satisfaction, responsiveness, etc (Atun, 2004). Out of these, for the purpose of this study, we have chosen the various services offered by each hospital, the number of available beds and number of trained medical staff to attend to the patients as our inputs. These were selected to have a commonness in measurement and ease of data availability for the purpose of measurement. Each of the above-mentioned points could also act as inputs but the data availability and time constraint for the study limit us here. For the selection of the dimensions of study, a simple analysis was done listing out the advantages and disadvantages of the indicators to come up with the most suitable for the study (see Figure 2).

The above inferences were made through the visits at the hospitals and having a discussion with the medical practitioners. For the purpose of measurement, after identifying the dimensions, we selected DEA technique.

Data Envelopment Analysis

DEA is a leading model based on linear programming technique for measuring the relative performance of organizational units where the presence of multiple inputs and outputs makes comparisons difficult. It helps out the decision-makers in such a way that, with little or no background in economics and operational research, efficiency can be measured and analysed easily (Adnan & Abdelkhader, 2013). Using this technique is far more advantageous as mathematics is kept to a minimum, which makes it less complicated. The major benefits of using DEA are:

Figure 2.

Advantages and Disadvantages of Dimensions Used

efficiency score helps to determine if a firm is efficient or has capacity for improvement;

helps to understand by how much input or output considered must be increased or decreased to attain efficiency; and

the return to scale helps to indicate if a firm has to decrease or increase its scale (or size) to minimize the average cost.

The usual measure of efficiency, that is, efficiency = output/input has often proved to be inadequate because of the inefficiency in handling multiple inputs and outputs related to different resources, activities and environmental factors (Al-Shammari, 1999). This gave rise to the development and popularity of DEA model, which considers the above-mentioned deficiencies.

By optimizing calculations, DEA models obtain input and output weights. On the basis of the results generated, units can be divided as efficient and inefficient. In case of inefficient units, the target values of inputs and outputs which would steer to efficiency are described. The efficiency of an organization or DMU can be evaluated by means of either an input or an output orientation. Input-oriented measures keep output constant and evaluate the reduction in input usage, while output-oriented measures keep input levels constant and find out the proportional increase in output quantities (Coelli et al. 2005; Vincova, 2005; Andersen & Petersen, 1993).

Efficiency Measures

Measuring efficiency is the ratio of output over input. For better efficiency, two techniques can be used—either increase the output or decrease the input. Another method to achieve higher efficiency is to bring about technological changes or to reengineer service processes (Ozcan, 2008). The classic difference between DEA and other methods is that it finds out the optimal ways of performance rather than the averages.

Efficiency measurement by DEA model is of different types, such as Technical efficiency, where both output and input are calculated in physical term. Cost-efficiency which is similar to technical efficiency, but, here, cost (or price) information about input is added to the model. Revenue efficiency is again similar to technical efficiency, except that price information about outputs is incorporated into the model. Scale efficiency is obtained when its size of operations is optimal that any further modifications in its size makes it less efficient. Scale efficiency is calculated by dividing the aggregate efficiency by the technical efficiency. Technical efficiency ignores the size of the unit measured as it compares a DMU only with units of similar size.

For this study, to analyse performance of the hospitals, we are making use of multiple inputs and outputs, which is one added advantage of using DEA model. While using multiple items for analysis, one has to make sure the total number of outputs and inputs is not limitless. If the number of firms is less than three times the sum of the total number of inputs and outputs, there are great chances of obtaining 100 per cent efficiency score, making DEA value incorrect. The other feature of DEA is that the weights assigned to outputs and inputs are not decided by the user. Also, it does not make use of common set of weight to all the firms under consideration but calculated and assigned by means of linear optimization procedure. In addition, in order to apply DEA, the number of inputs and outputs used must be equal in number, which brings us to the selection of three inputs and three outputs, and the number of hospitals selected for the study is taken as 20 in total number.

For this study, we are making use of the input-oriented DEA model, taking into account the Variable Return to Scale (VRS). There are two scales available in the DEA model—the constant returns to scale (CRS) and VRS. The difference between the two scales is that, in CRS, the output dimensions are considered to change proportionally with the input which is not applicable in VRS Paco and Perez (2013). Here, we make use of input-oriented DEA, which helps to determine how much the input use could be adjusted by the firm in question to make it optimum.

Data Employment Analysis Software

The software used for the purpose of study was DEAP version 2.1.

Patient Satisfaction Survey

Further, to validate the study, patient satisfaction survey was done in addition to a few interviews with patients to understand the requirements and perception of efficiency from their viewpoint, to arrive at suggestions for improvement. The patient satisfaction survey questionnaire is attached as Appendix 1 and the compilation of the results gathered are detailed in the analysis section.

Stage 2: Analysis

Analysis was done via a two-stage process in which one stage was understanding the efficiency of the selected hospitals through DEA model and the other stage comprised collecting reviews from patients regarding the efficiency of hospitals based on carefully identified questions or the patient satisfaction survey.

Data Employment Analysis Analysis

For the process of DEA analysis, as mentioned earlier, the output and input variables were identified and is detailed in the Table 1.

After the selection of input and output variable data were collected from the randomly selected 20 hospitals which is shown in Table 2.

Table 1.

Selected List of Input and Output Variables

Input Variable	Output Variable
No. of hospital beds	Outpatient rate
No. of trained medical staffs	Mortality rate
No. of services offered	No. of surgical operations performed per month

Source: Author’s compilation.

Table 2.

Input and Output Variable Data from Randomly Selected 20 Hospitals

DMU No.	DMU Name	No. of Hospital Beds	No. of Trained Medical Staffs	No. of Services Offered	Outpatient Rate	Mortality Rate	No. of Surgical Operations per Month
1	A	200	115	6	440	10	75
2	B	460	713	22	980	45	90
3	C	300	227	20	860	55	80
4	D	635	780	25	1200	75	480
5	E	540	590	27	1150	56	420
6	F	175	120	15	490	45	70
7	G	900	1400	27	1800	150	540
8	H	230	123	10	420	55	75
9	I	280	140	20	520	34	60
10	J	800	920	28	400	120	450
11	K	1000	1200	33	1500	100	580
12	L	1100	1100	33	1432	100	440
13	M	350	980	30	920	55	320
14	N	300	750	17	480	77	380
15	O	350	820	25	510	35	350
16	P	300	720	20	420	55	420
17	Q	250	250	10	360	60	230
18	R	250	255	18	520	70	350
19	S	200	175	5	340	35	125
20	T	375	125	21	625	55	185

Source: Author’s compilation.

The efficiency score in DEA is measured on a scale ranging from 0 to 1, where 1 indicates the system to be efficient and any value less than 1 to be relatively inefficient (8). From the data obtained through DEA analysis, most of the selected hospitals have been identified to be efficient, having obtained a score of 1. This indicates that the input provided by the hospitals suffice the needs to ensure required efficiency. DEA versions of DEAP measures the weighted output to weighted input to calculate efficiency. Here, both input and output are assigned with a weight of 1 for each.

For the purpose of study as mentioned previously, we have selected three inputs and three outputs each of which is assigned with a weight of 1. The entities used and the notations are mentioned below:

N DMUs (index, i, j)

K inputs

M outputs

The parameters under consideration are:

x_i = K × 1 input vector for DMU i

y_i = M × 1 output vector for DMU i

Variables used are:

v = K × 1 input weight vector

u = M × 1 output weight vector

Here, K = 3 inputs, M = 3 outputs.

Therefore, x_i = K × 1 = 3 × 1 input vector for DMU i

y_i = M × 1 = 3 × 1 output vector for DMU i

We make use of CRS DEA model, which indicates a linear relation between input and output. For each DMU I, the objective is to obtain a vector v of input weights and a vector u of output weights, such that the weighted ratio of both under consideration is maximal.

max u^Ty_i ≤ 1

v^T x_i

where u, v are variable vectors and x_i, y_i are input and output parameter vectors.

The collected data were then applied to the DEA software, which generated the following results on efficiency parameter, shown in Table 3.

Table 3.

Efficiency Parameter Result

DMU No.	DMU Name	Input–Output CRS Efficiency
1	A	1.00000
2	B	0.84461
3	C	1.00000
4	D	0.95827
5	E	0.97191
6	F	1.00000
7	G	1.00000
8	H	1.00000
9	I	0.84312
10	J	0.68702
11	K	0.80740
12	L	0.67447
13	M	1.00000
14	N	1.00000
15	O	0.71563
16	P	1.00000
17	Q	1.00000
18	R	1.00000
19	S	1.00000
20	T	1.00000

Source: Author’s compilation.

Table 3 represents the results obtained from DEA model. Through this method, to say that a particular hospital is efficient it must obtain a value of 1. Any value less than 1 shows a discrepancy in the efficiency of the workings of the hospital and clearly indicates a need for improvement. A total of 12 hospitals has achieved an overall efficiency score of 1 which indicates that these hospitals are in par with the utilization of available resources to utmost value. That is, the inputs considered such as number of beds, number of staff and services offered are sufficed here. From the results obtained for the 20 hospitals surveyed, 8 have to improve upon their resources and activities to better attain efficiency. The results clearly indicate a relative inefficiency in the hospitals such as “B”, “D”, “E”, “I”, “J”, “K”, “L” and “O”. Of the identified hospitals, hospital “L” ranks the least efficient with 67 per cent efficiency, closely followed by hospital “J” with 68 per cent. This can broadly be interpreted as hospital “L” having been able to support its activity levels with only 67 per cent of its resources. In addition, of the identified hospitals, seven have an efficiency below 0.9, indicating a fair degree of discrimination. In solving each linear programme, the technique adopted tries to make the efficiency of the target unit as large as possible. How this functions is that the search procedure ends when either the efficiency of the target unit or the efficiency of one or more of other units hit the upper limit of 1.

Using the DEA model, both input and output slack values were calculated, which indicate the additional improvement, that is, the increase in output and/or reduction in input was required for a unit to become efficient (Tone, 2001). Slacks are helpful in recognizing the problems faced by hospitals in terms of performance with an appropriate direction for improvement in order to achieve fully efficient performances. Hence slacks are seen only on DMUs that perform inefficiently as they indicate the remaining portion of inefficiency. When a DMU is incapable in reaching frontier efficiency, then slack is required to push the DMU into efficient performance. Slack values on both input and output variables have zero value when the efficiency score is equal to one on CRS model or VRS model, or both of them. Table 4 indicates the input slack value, and Table 5 indicates the output slack value.

The results in Table 4 indicate that hospitals need to improve in terms of the number of trained medical staff availability. Here, it shows that hospital “L” should either increase or decrease the medical staff by 363.26172 to become efficient. Slack value shows the variable discrepancy between constant output and input. That is, the value needs to be increased or decreased. Similar is the case with output slack, as shown in Table 5. The slack value for hospital “K” in terms of number of trained staff is 468.17600, which indicates that the hospital needs to reduce the number by the same. Furthermore, from output slack value, it can be seen that the same hospital needs to improve by reducing the mortality rate by 20.40800 and outpatient rate by 509.24800 to attain efficiency. Similar is the case with the rest of the hospitals in terms of attaining efficiency.

Table 4.

Efficiency in Terms of no. Beds, Trained Staff and Services

DMU No.	DMU Name	No. of Hospital Beds	No. of Trained Staffs	No. of Services Offered
1	A	0.00000	0.00000	0.00000
2	B	0.00000	361.61905	0.00000
3	C	0.00000	0.00000	0.00000
4	D	0.00000	260.11600	0.00000
5	E	0.00000	165.07143	0.00000
6	F	0.00000	0.00000	0.00000
7	G	0.00000	0.00000	0.00000
8	H	0.00000	0.00000	0.00000
9	I	0.00000	0.00000	0.00000
10	J	0.00000	283.04688	0.00000
11	K	0.00000	468.17600	0.00000
12	L	0.00000	363.26172	0.00000
13	M	0.00000	0.00000	0.00000
14	N	0.00000	0.00000	0.00000
15	O	0.00000	465.11905	0.00000
16	P	0.00000	0.00000	0.00000
17	Q	0.00000	0.00000	0.00000
18	R	0.00000	0.00000	0.00000
19	S	0.00000	0.00000	0.00000
20	T	0.00000	0.00000	0.00000

Source: Author’s compilation.

Table 5.

Efficiency in Terms of Outpatient Rate, Mortality Rate and Surgical Operations Per Month

DMU No.	DMU Name	Outpatient Rate	Mortality Rate	No. of Surgical Operations Per Month
1	A	0.00000	0.00000	0.00000
2	B	8,57143	22.90476	282.61905
3	C	0.00000	0.00000	0.00000
4	D	46.36800	28.92800	0.00000
5	E	7.14286	30.14286	46.07143
6	F	0.00000	0.00000	0.00000
7	G	0.00000	0.00000	0.00000
8	H	0.00000	0.00000	0.00000
9	I	96.75676	21.69730	65.77297
10	J	1152.50000	0.00000	89.45313
11	K	509.24800	20.40800	0.00000
12	L	836.12500	0.00000	82.36328
13	M	0.00000	0.00000	0.00000
14	N	0.00000	0.00000	0.00000
15	O	218.57143	61.90476	135.11905
16	P	0.00000	0.00000	0.00000
17	Q	0.00000	0.00000	0.00000
18	R	0.00000	0.00000	0.00000
19	S	0.00000	0.00000	0.00000
20	T	0.00000	0.00000	0.00000

Source: Author’s compilation.

The results in Table 5 indicate that hospitals such as “O” in order to achieve efficiency need to increase or reduce the outpatient rate by 218.57143 and mortality rate by 61.90476.

Patient Satisfaction Survey Analysis

The second stage of analysing the efficiency of the selected hospitals is through the patient satisfaction survey. The questionnaire was distributed randomly across 10 patients in each of the selected hospitals. Through this survey, the perception of the efficiency of the hospital by the general public was to be determined and the necessary improvements on the same to be identified. The questionnaire has been attached as Appendix 1 for reference.

From the survey, the patients’ level of ease and satisfaction were determined, which directly links to the efficiency of the hospitals. Table 6 indicates the data collected through the survey.

Table 6.

Level of Efficiency as Perceived by Patients (Survey Data Collection)

DMU No.	DMU Name	Q1	Q2	Q3	Q4	Q5	Q6	Q7	Q8	Q9	Q10	Q11	Q12	Q13	Q14
1	A	4	4	5	4	3	3	4	2	3	2	4	4	3	2
2	B	2	2	2	2	3	3	2	4	2	3	2	2	3	3
3	C	3	4	4	3	3	2	2	4	3	3	3	4	2	4
4	D	2	2	1	3	3	2	2	3	2	3	3	2	2	3
5	E	3	2	2	1	3	3	2	2	3	2	3	3	2	2
6	F	3	3	3	4	4	2	3	4	3	3	4	4	2	4
7	G	4	3	4	4	4	3	3	2	2	4	4	3	3	4
8	H	3	3	3	3	4	2	2	2	3	3	4	3	2	3
9	I	2	2	2	1	3	3	2	2	3	2	3	3	2	2
10	J	2	3	1	3	3	2	2	3	2	3	3	2	2	3
11	K	2	2	1	3	3	2	2	3	2	3	3	2	2	3
12	L	2	1	2	3	2	1	2	1	3	3	2	2	3	2
13	M	4	3	3	4	4	4	3	3	2	2	2	3	2	3
14	N	4	3	4	4	3	3	3	4	3	3	2	2	3	2
15	O	2	2	1	2	3	2	1	2	1	3	3	2	2	3
16	P	3	3	3	3	4	4	2	3	4	3	3	4	4	2
17	Q	3	4	3	4	4	4	3	3	2	2	4	4	3	3
18	R	4	3	3	3	3	4	2	2	2	3	3	4	3	2
19	S	3	3	4	3	4	4	4	3	3	2	2	4	4	3
20	T	3	3	4	3	4	4	4	3	3	2	2	4	4	3

Source: Author’s compilation.

Table 6 indicates the data collected from the survey. The data show the mean average indicated for each question by 10 patients for each hospital. From the results, it indicates the level of efficiency as perceived by the patients. Hospital “A” has the maximum value of 4, that is, out of 14 questions asked, 6 of them have a maximum value of 4. This indicates that out of the maximum satisfaction level of 5, hospital “A” is perceived by patients as efficient. Hospital “B” has a maximum value of 2, hospital “C” an efficiency score of 3, hospital “J” an efficiency score of 3, and hospital “D” with an efficiency score of 2. In order to measure the perception of the patients regarding each hospital, the questions were formulated. The questions mainly focus on the care provided by the hospital and the amenities they have in order to achieve the mandated efficiency. Similar is the case with other hospitals under study. The results indicate issues such as the ease in appointment scheduling, professionalism of staff, cleanliness and amenities they provide and clarity on directions. The questionnaire also focuses on the doctor’s efficiency such as listening skills, time spent and interactions. These questions were examined, and the above-mentioned results were obtained.

By attaining an average score of 4 out of 5 by hospital “A” indicates that as perceived by the patients, the staff at the hospital is very helpful in providing the right care and offer the right directions with ease. The cleanliness and amnesties provided are also well rated by most of the patients surveyed. The doctors are well qualified and experienced who seem to be capable of handling patients’ issues. The efficiency score as marked by the patients is least for hospital “B”, which, on further analysis, revealed the less than average performance by the staff. They lacked in terms of providing care and assistance to patients. On further analysis, this seemed to have occurred due to overall dissatisfaction by the staff on account of remuneration. Similarly, each hospital was surveyed. Hospital “J” seemed to have achieved the low-efficiency score because of the lack of time spent by the doctors with the patients. Most of the patients reported to have spent lesser time with the doctor as there were large queues of patients for consultation.

Stage 3: Action

Based on the inferences gathered from DEA model and patient satisfaction survey, the efficiency level of each of the randomly selected 20 hospitals was received. Of the hospitals under consideration, “J” and “L” rank the lowest in terms of efficiency by a total compilation of both the DEA and patient satisfaction survey. This could be due to the inefficiency in the supply of input. In case of Aswini Hospital, which is one of the largest in Thrissur district, to the hospital lacks efficiency in handling large number of incoming patients suitably. Hence, in order to bring about improvements, the dimension to be considered is handling multiple patients for which the staff number needs to be increased. In case of BMH Hospital, similar is the case. The patients have mentioned the amount of time spent with them to be inadequate, which can be attributed to fewer number of doctors available with respect to the incoming patient rate. By identifying the areas in which the improvements are to be made, the efficiency rate of the low-ranking hospitals can be increased sufficiently.

Conclusion

With this article, we tried to understand the impact of private sector hospitals in the healthcare sector in Kerala, for which the efficiency of the randomly selected 20 hospitals were measured using DEA model and patient satisfaction survey. Considering both the aspects together, the efficiency was calculated, indicating the highest to hospital “G” and lowest to “J” and “L” hospitals. DEA is a novel approach in relative efficiency measurement, which can easily handle multiple inputs and outputs. Additionally, the method identifies peer units and targets for inefficient units. Through this article, we aimed at measuring the efficiency of the selected hospitals using this novel approach, providing significant analysis on the same. From this study, it has become evident that the operational efficiencies of sample hospitals measured had overall trends towards efficiency; however, a few among the selected have yet to achieve this benchmark, which can be attributed to the ineffective utilization of the available resources.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Appendix 1: Patient Satisfaction Survey

Hospital Name:

Please indicate your satisfaction level for each of the following questions

(1-not satisfied 5-very satisfied) 1 2 3 4 5

References

Adnan

Abdelkhader

(2013). An empirical comparison of DEA and SFA method to measure hospital unitsâ™ efficiency . Proceedings of the 2nd International Conference on Operations Research and Enterprise Systems. doi:10.5220/000427450242025.

Afonso

Aubyn

M. S.

(2006). Cross-country efficiency of secondary education provision: A semi-parametric analysis with non-discretionary inputs. Economic modelling , 23(3), 476–491.

Al-Shammari

(1999). A multi-criteria data envelopment analysis model for measuring the productive efficiency of hospitals. International Journal of Operations & Production Management , 19(9), 879–891.

Andersen

Petersen

N. C.

(1993). A procedure for ranking efficient units in data envelopment analysis. Management Science , 39(10), 1261–1264. doi:10.1287/mnsc.39.10.126.

Athanassopoulos

Gounaris

(2001). Assessing the technical and allocative efficiency of hospital operations in Greece and its resource allocation implications. European Journal of Operational Research , 133(2), 416–431. doi:10.1016/s0377-2217(00)00180-.

Berger

M. C.

Messer

(2002). Public financing of health expenditures, insurance, and health outcomes. Applied Economics , 34(17), 2105–2113.

Butler

T. W.

(2005). The utility of returns to scale in DEA programming: An analysis of Michigan rural hospitals. European Journal of Operational Research , 161(2), 469–477.

Chilingerian

J. A.

(1995). Evaluating physician efficiency in hospitals: A multivariate analysis of best practices. European Journal of Operational Research , 80(3), 548–574.

Clement

J. P.

Valdmanis

V. G.

Bazzoli

G. J.

Zhao

Chukmaitov

(2008). Is more better? An analysis of hospital outcomes and efficiency with a DEA model of output congestion. Health Care Management Science ,11(1), 67–77.

10.

Charnes

Cooper

W. W.

Rhodes

(1978). Measuring the efficiency of decision making units. European Journal of Operational Research , 2(6), 429–444.

11.

Coelli

T. J.

Rao

D. S. P.

O’Donnell

C. J.

Battese

G. E.

(2005). An introduction to efficiency and productivity analysis . Springer Science & Business Media.

12.

Davis

Schoen

Schoenbaum

Audet

Doty

Tenney.

(2004). Mirror, mirror on the wall: Looking at the quality of American health care through the patient’s lens. Boston: The Commonwealth Fund.

13.

Dilip

T. R.

(2008). Role of private hospitals in Kerala: An exploration (Working paper). Retrieved from https://opendocs.ids.ac.uk/opendocs/bitstream/handle/20.500.12413/3105/wp400.pdf?sequence=1&isAllowed=y

14.

DeRosario

J. M.

(1999). Healthcare system performance indicators: a new beginning for a reformed Canadian healthcare system. Journal for Healthcare Quality , 21(1), 37–41.

15.

Farrell

M. J.

(1957). The measurement of productive efficiency. Journal of the Royal Statistical Society: Series A (General) , 120(3), 253–281.

16.

Häkkinen

Joumard.

(2007). Cross-Country Analysis of Efficiency in OECD Health Care Sectors: Options for Research. OECD Economics Department Working Papers , No. 554, Paris: OECD Publishing. Retrieved from https://doi.org/10.1787/178861806081

17.

Joumard

André

Nicq

Chatal

(2010). Health status determinants: lifestyle, environment, health care resources and efficiency. Environment, Health Care Resources and Efficiency (May 27, 2010) . OECD Economics Department Working Paper, (627).

18.

Kabir

Krishnan

T. N.

(1996). Social Intermediation and Health Change: Lessons from Kerala. In Gupta

Das

Chen

L.G.

Krishnan

T.N.

(Eds.), Health, Poverty and Development in India . New Delhi: Oxford University Press.

19.

Kannan

K. P.

Thankappan

K. R.

Ramankutty

Aravindan

K. P.

(1991). Health and Development in Rural Kerala: A study of the linkages between socioeconomic status and health status . Palakkad: Integrated Rural Technology Centre.

20.

Gangadharan

(2005). Utilization of Health Services in Urban Kerela: A Socio Economic Study . Serials Publications.

21.

Jacobs

Smith

P. C.

Street

(2006). Measuring efficiency in health care: Analytic techniques and health policy . Cambridge University Press.

22.

Jafarov

Gunnarsson

(2008). Government spending on health care and education in Croatia: Efficiency and reform options. IMF Working Paper (WP 136). Washington: International Monetary Fund.

23.

Hollingsworth

(2008). The measurement of efficiency and productivity of health care delivery. Health economics , 17(10), 1107–1128.

24.

Kirigia

J. M.

Emrouznejad

Sambo

L. G.

Munguti

Liambila

(2004). Using data envelopment analysis to measure the technical efficiency of public health centers in Kenya. Journal of Medical Systems , 28(2), 155–166.

25.

Kohl

Schoenfelder

Fügener

Brunner

J. O.

(2018). The use of data envelopment analysis (DEA) in healthcare with a focus on hospitals. Health Care Management Science . doi:10.1007/s10729-018-9436-.

26.

Kutty

V. R.

(2000). Historical analysis of the development of health care facilities in Kerala state, India. Health Policy and Planning , 15(1), 103–109. doi:10.1093/heapol/15.1.10.

27.

Mirmirani

H. C.

Ilacqua

J. A.

(2008). Health care efficiency in transition economies: An application of data envelopment analysis. International Business & Economics Research Journal , 7(2), 47–56.

28.

Nabae

(2003). The health care system in Kerala: its past accomplishments and new challenges. Journal of the National Institute of Public Health , 52(2), 140–145.

29.

Ozcan

Y. A.

(2008) Health care benchmarking and performance evaluation: An assessment using data envelopment analysis (DEA) . New York, NY: Springer Science+Business Media.

30.

Osei

d’Almeida

George

M. O.

Kirigia

J. M.

Mensah

A. O.

Kainyu

L. H.

(2005). Technical efficiency of public district hospitals and health centres in Ghana: a pilot study. Cost Effectiveness and Resource Allocation , 3(1), 9.

31.

Ramachandran

V. K.

(1996). In ‘On Kerala’s development achievements’, Jean

Dreze

Amartya

Sen

(eds), Indian Development: Selected Regional Perspectives. New Delhi, India: Oxford University Press.

32.

Sadanandan

(2001). Government Health Services in Kerala: Who Benefits? Economic and Political Weekly , 3071-3077.

33.

Rebba

Rizzi

(2007). Measuring Hospital Efficiency through Data Envelopment Analysis when Policy-makers’ Preferences Matter. An application to a sample of Italian NHS hospitals. Politica economica , 23(3), 233-258.

34.

Retzlaff-Roberts

Chang

C. F.

Rubin

R. M.

(2004). Technical efficiency in the use of health care resources: A comparison of OECD countries. Health policy , 69(1), 55-72.

35.

Sherman

H. D.

(1984). Hospital efficiency measurement and evaluation: empirical test of a new technique. Medical care , 922-938.

36.

Spinks

Hollingsworth

(2009). Cross-country comparisons of technical efficiency of health production: a demonstration of pitfalls. Applied Economics , 41(4), 417-427.

37.

Tone

(2001). A slacks-based measure of efficiency in data envelopment analysis. European journal of operational research , 130(3), 498-509.

38.

Siciliani

Sivey

Street

(2013). Differences in length of stay for hip replacement between public hospitals, specialised treatment centres and private providers: selection or efficiency?. Health economics , 22(2), 234-242.

39.

Soman

C. R.

(2007). Fifty years of primary health care: the Kerala experience. Bulletin of the Nutrition Foundation of India , 28(4), 91-99. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.576.9295&rep=rep1&type=pdf

40.

Verhoeven

Gunnarsson

Lugaresi

, 2007. The health sector in the Slovak Republic: Efficiency and reform. IMF Working Paper, WP 226. Washington: International Monetary Fund.