Background

Hospitals are an essential component of health systems, while also being the most costly. They account for 50–80% of total health expenditures [1]. Hospital costs continue to rise due to the development of new technologies. New diagnostic and therapeutic methods are implemented to combat the rising proportion of chronic diseases, the increasing demand for health services, and the subsequent medical errors [2]. This has become a primary challenge and concern for governments [3].

Hospitals in the Eastern Mediterranean Region (EMR) differ in size, proprietorship, assignment, and performance. The total number of hospital beds is estimated to be 740,000 and, except for Lebanon, the majority of hospital beds are in the public sector (80%), with the remaining in private for-profit (18%) and private not-for-profit (2%) hospitals. The range of hospital beds per 10,000 population vary from 3.9 to 32 in 22 countries in the EMR. Hospitals also vary widely in size, location (rural and urban), resources, specialization (general versus specialty hospitals) and organization, as well as their position in the health system (first-level hospitals, secondary care hospitals and large teaching institutions) [4]. A large proportion of hospitals are financed by the government, but out-of-pocket payments are rising due to limited public sector resources [5]. This leads to limited access to health services for vulnerable communities. Private hospitals in the EMR are usually small to medium size and located in capitals and other large cities. These hospitals are not the result of comprehensive health system planning, as such, they can also lead to inequity in access to healthcare. Most countries in the EMR have addressed inequalities by implementing reforms to increase productivity, transparency, and cost flexibility [5,6,7]. To facilitate this process and increase hospital efficiency, it is necessary to provide the healthcare sector with additional resources and management tools.

According to Farrell (1957), efficiency is defined as “the firm’s success to produce the maximum feasible amount of output from a given amount of input or producing a given amount of output using the minimum level of inputs where both the inputs and the outputs are correctly measured” [8]. Three different types of efficiency were defined by Farrell: technical efficiency, allocative efficiency, and economic efficiency. Technical efficiency is the ability of a business to gain a maximum output from the specific input. In contrast, allocative efficiency refers to the directing of resources toward products or services with the highest demand. Economic efficiency is allocative efficiency and technical efficiency from a joint unit of cost efficiency. An organization has an economic efficiency Which be efficient in terms of both technical and allocational [8]. In general, different methods have been used to measure hospital efficiency: Data Envelopment Analysis (DEA), Stochastic Frontier Analysis (SFA), and measures of performance, such as Pabon Lasso’s model. DEA is a non-parametric linear programming method used to evaluate the efficiency of decision-making units [8, 9]. SFA is parametric and calculates the difference between the organization’s predicted and expected outputs [10]. Pabon Lasso’s model (1986) assesses hospital performance using three performance indicators: bed occupancy rate (BOR), bed turnover rate (BTR), and average length of stay (ALS) [11].

A decline in hospital efficiency has been observed worldwide. In a global report by the World Health Organization (WHO) published in 2010, 10 sources of hospital inefficiency were identified: (1) underuse or overpricing of generic drugs; (2) use of substandard or counterfeit drugs; (3) inappropriate and ineffective drug use; (4) overuse or oversupply of equipment, investigations and procedures; (5) inappropriate or costly workforce mix, unmotivated worker; (6) inappropriate hospital admissions or length of stay; (7) inappropriate hospital size (low use of infrastructure); (8) medical errors and suboptimal quality of care; (9) waste, corruption and fraud; and (10) inefficient mix or inappropriate level of strategies [12]. However, thus far there has not been a comprehensive review to assess the source of hospital inefficiency in the EMR. This study aims to comprehensively identify the sources of hospital inefficiency in the EMR, and compare these to previously identified sources of hospital inefficiency. This will provide insight into the current condition of healthcare in this region.

According to the aforementioned WHO report, hospital efficiency in the EMR is low, particularly in low and middle-income countries (LMICs) [5]. To increase hospital efficiency in a context of rising costs and limited resources, it is necessary to identify sources of inefficiency and to suggest improvement strategies. Identifying these sources and identifying improvements are the objectives of this study.

Methods

This is a systematic review of existing evidence on hospital inefficiency in the EMR. This study recruited English peer-reviewed articles published between January 1999 and May 2018. To identify relevant articles, a database search was conducted in MEDLINE (via PubMed) (Additional file 1), Web of Knowledge, Embase, Google and Google Scholar. Keywords used included “efficiency”, “productivity”, “inefficiency”, “hospital”, “data envelopment analysis”, “Pabon Lasso”, and “stochastic frontier analysis”. Moreover, the reference lists of selected articles were searched for relevant papers. Economic journals in the field of health economy and efficiency such as the Journal of the Knowledge Economy, the American Journal of Economics and Business Administration, Cost Effectiveness and Resource Allocation, and the International Journal of Economics and Financial Issues were searched individually. An initial review was conducted to determine the scope of the study, and no study published before 1999 was found. Therefore, the review included studies between 1999 and May 2018.

Following the screening of 1087 identified articles, 80 full texts were assessed for eligibility. After assessing these articles, 56 were included in the review. The screening process and search results are shown in the PRISMA Flow Diagram [13] of Fig. 1.

Fig. 1
figure 1

PRISMA Flow Diagram: Database search and article selection process

A data extraction form with entries for the first author, year of publication, country of study, data collection method, number of hospitals studied, inputs and outputs for efficiency, sources of hospital inefficiency, and factors affecting efficiency, was used to collect data from the selected studies. For higher reliability, two researchers independently extracted data from a randomly selected sample of the chosen articles. Any disagreements were solved by discussion and consensus and, if necessary, by a third reviewer.

Mitton et al.’s fifteen-point scale [14] was used for quality appraisal. The criteria used to assess quality included: literature review and identification of research gaps; research question and design, validity and reliability; data collection; population and sampling; and analysis and reporting of results. These criteria were rated 0 (not present or reported), 1 (present but low quality), 2 (present and mid-range quality), or 3 (present and high quality). Articles were rated independently by two researchers using the article quality rating sheet. Given that the review was qualitative, articles were not removed at this stage, but more weight was given to articles with a quality rating of 10 or above in the data analysis and interpretation of results.

The data were analyzed using qualitative content analysis. Data were coded and managed using MAXQDA 10 for Windows (VERBI GmbH, Berlin, Germany), and themes and subthemes were extracted to identify patterns and relationships between themes.

Results

A total of 56 articles on hospital efficiency in the EMR, published between January 1999 and May 2018, were reviewed. A large number of studies (91%) were published after 2010. The reviewed studies were only conducted in 10 out of 22 EMR countries included in the search. Iran (n = 35) was most represented in the included studies, followed by Saudi Arabia (n = 5) and Tunisia (n = 5), Jordan (n = 4), Pakistan (n = 2), and finally UAE, Palestine, Iraq, Oman, and Afghanistan (n = 1 each).

Overall, 1995 hospitals were examined in these studies; most of them located in Iran (n = 858), Saudi Arabia (n = 573), Tunisia (n = 266), UAE (n = 96), Jordan (n = 72) and Afghanistan (n = 68). Out of 56 reviewed studies, 21 used DEA (37%), 12 used Bayesian SFA (21%), 10 used Pabon Lasso’s model (18%), and four studies used the Malmquist index (7.5%). Moreover, four studies (7.5%) used a hybrid approach by comparing DEA and Pabon Lasso’s model. Finally, five studies (9%) used other methods (the Cobb-Douglas Model, the Lean model, and efficiency and performance indicators).

Calculating efficiency requires input and output variables. In data analysis, the number of workforce, active beds, total costs, hospital size, medical equipment, technological capacity, and budget have been used as input variables (Fig. 2). Total outpatient visits, inpatient admissions and days, number of inpatients, emergency visits, number of surgeries, ratio of major surgeries to total surgeries, total number of medical interventions, BOR, BTR, average length of stay (ALS), number of ambulances, ratio of active beds to fixed beds, hoteling expense (bed-day costs) and employee expense total survival rate, number of discharged patients, number of imaging service users, and number of laboratory test users, were used as output variables (Fig. 3). The input and output selection depends on the objective of the study and efficiency measurement. It is reasonable to consider total costs on the input side; however, few studies have employed hospital hoteling and workforce expenses as output in their evaluation. For example, Hatam [15] used hoteling and workforce expenses and found that most cases had more workforce and hoteling expenses than the similar ones showing significant inefficiency.

Fig. 2
figure 2

Frequency of input variables used to measure hospital efficiency in EMR countries

Fig. 3
figure 3

Frequency of output variables used to measure hospital efficiency in EMR countries

Operational definitions for acronyms and terms of input and output measures are given below:

  • Number of active beds: alternative term for ‘available beds’ [16].

  • Number of beds or hospital size: “Hospital beds include all beds that are regularly maintained and staffed and are immediately available for use. They include beds in general hospitals, mental health, and substance abuse hospitals, and other specialty hospitals. Beds in nursing and residential care facilities are excluded” [17].

  • Number of inpatient admissions: Mean number of hospital admissions in a certain hospital per year [16].

  • Number of bed-days: “number of days during which a person is confined to a bed and in which the patient stays overnight in a hospital” [18].

  • Bed occupancy rate (BOR): “The occupancy rate for curative (acute) care beds is calculated as the number of hospital bed-days related to curative care divided by the number of available curative care beds, multiplied by 365”.

  • Bed turnover rate (BTR): the number of times there is change of occupant for a bed during a given time period [17].

  • Average length of stay (ALS): “Average length of stay refers to the average number of days that patients spend in hospital. It is generally measured by dividing the total number of days stayed by all inpatients during a year by the number of admissions or discharges. Day cases are excluded” [17].

  • Day surgery: Day surgery is defined as the release of a patient who was admitted to a hospital for a planned surgical procedure and was discharged the same day [16].

Table 1 provides a summary of the studies reviewed, presenting the type and total number of hospitals examined, the methods used to calculate efficiency, inputs and outputs, and the source of inefficiency.

Table 1 Summary of reviewed studies

Various sources of hospital inefficiency were identified and divided into four themes, each with a set of subthemes: hospital products and services, hospital workforce, hospital services delivery, hospital system leakage (Table 2).

Table 2 Source of inefficiency in Eastern Mediterranean hospitals and strategies for improvement

The most frequent sources of inefficiency in EMR hospitals are excess workforce, excess beds, and inappropriate hospital sizes. Helal et al. [66] investigated the effect of health reforms (privatization) on the efficiency of 270 hospitals in Saudi Arabia and reported a 0.90 average efficiency in 2006 and a 0.92 average efficiency in 2014. The average efficiency of one is considered the best level of performance. Despite a reduction in inputs, outputs increased by 2%. Moreover, there was a 10.1% increase in the number of inpatients from 2006 to 2014. Therefore, reducing excess inputs such as excess workforce, excess beds or/and increasing outputs can be beneficial to hospitals. A 2013 analysis in Saudi Arabia showed that there was a reduction in the number of beds, doctors, nurses, and allied health workforce as inputs. Moreover, there was an increase in the number of inpatients, outpatients, the number of daily laboratory tests and the number daily of radiography services as outputs [39]. The most common strategies proposed in the included studies are: developing health policies for accurate recruitment planning, calculating the required number of beds for each community, and making proper use of hospital beds based on community needs.

Discussion

The purpose of this research was to examine the sources of hospital inefficiency and strategies available to increase hospital efficiency in the EMR. In recent years, there has been an increasing focus on hospital efficiency for health policymakers in developing countries. A total of 56 studies have been conducted on hospital efficiency in the EMR from January 1999 to May 2018. These studies have shown that hospital care is an economic activity requiring adequate funding and budgeting. As such, reducing inputs can improve performance and efficiency [56, 74].

The WHO Regional Office for the EMR classifies countries to there groups: high income countries (six countries), middle income countries (ten countries), and low income countries (six countries). The present research identified 56 articles on hospital efficiency in three high-income countries, five middle-income countries, and two low-income countries. General government expenditure allocated to health in the EMR countries remains between 2 and 16%, a low figure. Regarding hospital service utilization, the overall average bed occupancy rate and length of stays were 60.7% and 4.12 days, respectively, in the Region in 2013. Only a few countries have well-defined and functioning referral networks between hospitals and primary health care facilities, or between hospitals at different levels. Hospitals do not serve geographically defined catchment areas based on national policy mandates. Most countries are entrenched in the historical model of public provision and financing, and there is a mix of funding patterns, including public sector funds (through central government budgets and national insurance funds) and out-of-pocket payments made directly by users. In most countries, there is misalignment between the distribution of hospital beds and high-technology equipment and population health needs [4]. Contextual challenges exist, such as security issues, internal conflict and political volatility in EMR countries, leading to economic problems influencing health policies, health system budgets, and health system efficiency as a result [75, 76].

Some health system challenges are common to all EMR countries: “limited capacity in MoHs for evidence-based policy analysis and formulation and strategic planning through better use of information in adequate capacity to legislate, regulate and enforce rules and regulations” or “most countries lack national medicines policy” [75]. Both this study and the WHO have reported similar findings.

The most common input variables used in these studies were workforces numbers and the number of beds, while the most common output variables were the total number of outpatient visits, admissions and inpatient days. A systematic review of new approaches to measure hospital performance in LMICs in 2015 [77] identified seven key performance indicators. These included total inpatient days; recurrent expenditure per inpatient day; ALS; infection prevention rate; BOR; inpatient days per technical workforce; and unit cost of outpatient care. Seven performance indicators were also identified for high-income countries (HICs): mortality rate from emergency heart attack admissions after 28 days; mortality rate from emergency surgery after 30 days; number of patients on waiting lists; infection rate of methicillin-resistant Staphylococcus aureus per 10,000 bed-days; net profit; probability of workforce leaving within 12 months; and average healthcare commission rating [77].

On average, out-of-pocket payments differ between HICs and LMICs. In HICs, patients rarely pay directly for their care compared to LMICs where direct payment by patients is necessary due to lower insurance coverage. Furthermore, the mortality rate for non-elective admission is not the optimal output indicator for LMICs, as access to healthcare is a significant problem. These explain the differences in outputs between LMICs and HICs [77, 78].

The themes related to inefficiency extracted in this review, and the sources of inefficiency identified in the WHO report 2010 [11], highlight that studies have failed to address the issue of medical drugs. Using drug-related inputs and outputs can provide useful insights into drug-related sources of inefficiency in the EMR. For example, a study in Ethiopia used the cost of drug supply as input [79]. This can provide further insights into how to improve hospital efficiency.

In addition to excess workforce, excess beds and inappropriate hospital sizes, the inefficiency of hospitals in the EMR is also due to inappropriate workforce composition, lack of workforce motivation and inefficient use of health system inputs. According to a WHO report about National Health Accounts published in 2009, 15 to 25% of hospital inefficiency is related to workforce [80]. The workforce is at the core of the health system and accounts for almost half of the total health budget, in the form of wages and other payments [81]. The shortage of human resources is a major obstacle in implementing national healthcare plans, causing ineffective recruitment, inappropriate training, poor supervision, and suboptimal workforce distribution, which can further reduce efficiency [82]. Strategies to increase workforce efficiency focus on assessment and training based on needs, reviews of incentive policies, flexible contracts and performance-based payments [83].

Hospitals can result in lower efficiency if healthcare products and services are not optimal. Hospitals will face higher inputs against the specific output or lower outputs against the specific input. Excessive lengths of hospital stays, unnecessary admissions, and unnecessary referrals to specialists are examples of overuse of healthcare services. Reduced demand for hospital services and low BORs indicate underuse of available services [25,26,27,28,29,30,31,32]. A WHO report showed that suboptimal use of hospital resources, such as doctors, nurses, and beds, reduce demand for services and thus reduce hospital efficiency [82]. Optimal hospital management plays a vital role in optimizing healthcare services, improving hospital outcomes, and reducing costs [84,85,86]. Hospital managers and health policymakers can increase hospital efficiency and productivity through economies of scale. Strategies include optimizing hospital size, providing more products and services, and reducing ALS [38, 84,85,86].

Two of the principal sources of inefficiency in the EMR are inappropriate hospital sizes and excess numbers of active beds. These have been analyzed in studies conducted in countries outside the EMR, including in HICs [14, 21, 24,25,26, 33,34,35, 62]. These studies revealed the significant impact of hospital size and bed numbers on efficiency [87, 88]. The optimal number of active hospital beds typically lies between 200 and 300 beds. Generally, hospitals with less than 200 beds or more than 600 beds have higher costs [89]. According to international standards, a threshold BOR range between 84 and 85% indicates that use of hospital facilities and hospital resources are optimally efficient [90]. Therefore, optimizing hospital sizes and bed numbers can ensure that hospitals respond to population needs thus increasing efficiency. Indeed, it may be necessary for governments to build hospitals of a specific size, to take into account geographical considerations and difficulties accessing healthcare facilities.

The payment system has a vital role in improving hospital efficiency and productivity. In the EMR, payment systems are typically fee-for-service systems. In developed countries payments are often based on performance at clinical and organizational levels, increasing efficiency through performance incentives [91, 92]. Strategies to increase hospital efficiency include developing healthcare policies to implement appropriate payment systems, fair tariffs, and meticulous workforce recruitment plans, calculating required bed numbers for each community, making optimal use of hospital beds based on demand, and developing two-way electronic referral systems.

Conclusion

The results of this study have elucidated numerous sources of hospital inefficiency in the EMR. These sources should be addressed with targeted strategies, to improve hospital performance. Severe resource scarcity and increased costs of healthcare services, particularly in developing countries, require policymakers to ensure maximum use of available resources. Hospitals are highly complex, multidisciplinary social entities, whose performance can be improved through accurate, effective, and timely planning, organization, leadership, and management. Efficiency depends on multiple factors. As such, using various methods to measure hospital efficiency can be an effective strategy for managers and policymakers. Needs-based assessments and training, reviews of incentive policies, flexible contracts, performance-based payments, optimal hospital sizes based on community needs, increased resource availability and preservation of hospital social functions are crucial to increasing hospital efficiency.