Evidence-Based Experiences of Using Time-Driven Activity-Based Costing in Telemedicine-Based Health Care Delivery Protocols

Abstract

Background:

In the era of value-based health care, maximizing health outcomes and minimizing costs require different value optimization strategies. To maximize value and ensure control of expenditure, time-driven activity-based costing (TDABC) is widely used in health care organizations. In our study, we examined the impact of telehealth technologies on value creation by using the TDABC approach.

Methods

: We mapped four pairs of (traditional and telemedicine supported) health care delivery processes in terms of time, resource use, and information flow. Data were collected from four sources: approved protocol descriptions, in-depth interviews with senior clinicians, a financial controlling database of unit costs, and additional comments from controlling experts.

Results

: We found that technological improvements do not necessarily increase the value of protocols. Of the protocols studied, two telemedicine protocols proved to be more cost-effective (80.37% and 45.29% compared with the originals). However, significant cost overruns were detected in two other telemedicine protocols (902.90% and 161.01%, respectively). An increased value could be detected only when the use of telemedicine technology resulted in greater savings in net human capacity compared with the additional expenditure related to telemedicine technology.

Conclusions

: We concluded that the use of telemedicine technology leads to modifications in protocols at numerous points, which have a significant impact on cost levels. It is not sufficient to examine only the costs of modified steps, as proposed in the TDABC methodology. Our study also suggests that a refined TDABC method is a potential tool for assessing the complex effects of technological change.

Introduction

Technology change is radically reshaping market behaviors, how organizations operate, and how actors interact. This is the case with health care organizations, in which the pandemic caused by the SARS-CoV-2 virus has accelerated the adaptation of technology. The demand from patients, health care providers, and policymakers alike to adopt telemedicine and telehealth applications has increased.^1,2 This also means that value creation in health care organizations is being transformed.

In health care services, value can be described as the outcome achieved regarding a patient’s health status relative to the inputs utilized to accomplish the latter.³ From this perspective, value can be maximized by improving the former outcome using a given set of resources or minimizing expenditure across the whole health care delivery cycle without reducing the health outcome.⁴ To maximize value and ensure control of expenditure, time-driven activity-based costing (TDABC) is widely used in health care organizations.⁵ TDABC directly allocates resources to activities required to achieve a given health outcome by calculating the expenditure of the health care delivery cycle based on the time required to use the resources for a given activity.⁶ TDABC method is well suited for examining the economic value of new or novel technology-supported procedures in the field of health care.

TDABC METHODOLOGY

The study of cost management effectiveness of telemedicine as compared with traditional methods is an increasingly studied issue.^1,7
–9 A more accurate and sophisticated version of the traditional activity-based costing method is TDABC. TDABC is most commonly applied by health care organizations to health care systems or health care facilities.^9

–12 The growing health care application of TDABC is rooted in the value-based health care (VBHC) framework. VBHC encourages health care organizations to optimize the value equation by maximizing health outcomes while reducing costs.¹³

TDABC allows providers to accurately measure the cost of patient treatment under specific health conditions throughout the delivery cycle.^6,14 It provides information that is of use to health care management when making investment decisions or organizing activities.¹⁰ Moreover, TDABC provides information about which form of health care represents greater value from the perspective of a health care protocol.^15,16 Institutions most often use the method to explore opportunities for the development of the delivery cycle, improve information about costs, and compare traditional and time-based cost information.^17,18 Frequent operational applications include redeployment, workflow development, and facility management.^19,20 When applying the TDABC method, the costs of resources are not allocated to activities and then products; instead, managers first estimate the resources required for each transaction, product, or customer. Then, two parameters must be estimated for each resource group. On the one hand, the unit cost of providing a resource, and on the other, the unit time of resource use per product (resource use is accounted for at the practical capacity level).⁵

To implement TDABC costing and exploit the benefits of the method, Kaplan and Anderson⁵ emphasize the following. (1) No estimations of the cost of resource capacity per unit of time are required; it is sufficient for managers to estimate (based on experience) the rate of utilization of the resource in relation to total capacity. The cost of resource capacity per unit of time is the capacity cost rate, which can be obtained by multiplying the cost of the provided capacity by actual capacity.²¹ (2) When estimating the unit times of activities, managers should determine the time required to complete each sub-activity associated with each activity. These periods may be defined based on interviews with employees or direct observation. (3) Based on the product of the previous two estimated factors, it is possible to calculate capacity per unit of time (so-called deriving cost-driver rates).^5,7

APPLYING TDABC IN TELEMEDICINE/TELEHEALTH

TDABC has been widely used in telemedicine/telehealth projects and experiments over the last 10 years. The literature shows documented results in surgery, radiology, internal medicine, pediatrics as well as emergency care. In its application, the initiators take the VBHC as their starting point, emphasizing that instead of the widely used system of accounting prices in health care, the procedure involves an analysis of the actual costs incurred by a given provider using micro-costing. The TDABC aims to follow two patterns in each application: comparing the economic attributes of telemedicine/telehealth applications with the attributes of in-person health care protocols^{1,9,12,22
–24} or identifying the cost attributes of telemedicine/telehealth solutions.¹⁴ The use of TDABC involves either a multidimensional analysis of specific outcome characteristics (e.g., vital parameters, safety of care, patient satisfaction factors, sociometric factors)^9,24,25 or an analysis of cost characteristics only.¹²

In most cases, the research process is modified: either the individual steps are modified by identifying time drivers,¹⁴ excluding the observation of fixed costs,¹² or simplifying and sometimes merging the steps of the analysis.^1,22,23 In all cases, the implementation of process mapping, the estimation of time requirements of protocol steps, the calculation of unit costs of each resource, and the summation of the total cost estimate are observed as identifiable steps. In most cases, as originally recommended, the information required for the analysis is collected through interviews with key actors^1,23 or supplemented by observation,^12,14,25 with one case where the analysts made use of information extracted from telemedicine databases.²² In applying TDABC in health care, analysts include in their observations broadly defined personnel, equipment, material, and infrastructure costs^1,25; in many cases only personnel and infrastructure costs¹²; and sometimes only personnel costs.^22,23 Where only labor costs are used as the unit cost, effective capacity is typically not estimated. Effective capacity is estimated using expert judgment or based on local or regional standards.

The TDABC-based assessments of telemedicine/telehealth solutions show a wide range of results. A number of studies point to a significant reduction in infrastructure and equipment overheads during virtual visits, while the length of visits is shortened, reducing personnel overheads.^19,22,24,25 In these studies, where outcomes were measured, it was clear that interactions supported by telemedicine/telehealth applications did not reduce the effectiveness, safety, or perceived satisfaction of care. However, other studies point to the complexity of the phenomenon. Portney et al.¹² found no cost-saving potential when evaluating specialist visits using different technologies, but process analysis allowed for the substitution of resources, which has already clearly led to cost reductions. Goh et al.²³ also could not clearly demonstrate the predominance of telemedicine or face-to-face care as an economic option. Their results suggest that the ideal combination of the two technologies is the optimal solution.

Methods

The aim of our research was to identify the added value of telemedicine compared with traditional medicine in a pilot program. The analysis did not allow sufficient time to observe telemedicine outcomes in detail, so we focused our attention on the characteristics of the costs that have changed as a result of the changes in technology. The TDABC methodology^3,4,6,10,26 was found to be an ideal tool to study this area of focus.

Our study involved a systematic analysis of four health care protocols conducted in parallel with face-to-face medicine and telemedicine support as part of the above-mentioned pilot program: Case #1—Diagnosis and therapy assessment of ear, nose, and throat diseases—71 patients involved initially; Case #2—Applying peritoneal dialysis—6 patients involved initially; Case#3—Risk reduction for patients with metabolic syndrome—61 patients involved initially; and Case#4—Monitoring patients with heart failure—18 patients involved initially. The study was conducted at the University of Szeged, Albert Szent-Györgyi Clinical Center between October 1, 2020, and June 30, 2022. The protocols included in our study were approved by the Human Institutional and Regional Biomedical Research Ethics Committee, University of Szeged.

The research was carried out in three phases. In the first phase, on the basis of document analysis (1), a process map of the (face-to-face and telemedicine) protocols was compiled (2). During this process, in accordance with the TDABC methodology, the sequence and dependencies of activities were recorded along a given health care protocol. We then examined the resources involved by activity (3), distinguishing between human, tangible, and intangible resources for each of the two technologies used. As for human resources, we distinguished between the patients involved in the process and the medical, nursing, auxiliary, and managerial staff.^27

–31

When examining additional resources directly used for carrying out activities, we distinguished three types of resources: raw materials, equipment, and facilities. In assessing raw materials, we distinguished between raw materials used for the patient’s benefit in care, medicine, ancillary materials used in care, and other materials used in care (depending on the circumstances of the treatment).^32,33 When taking into account equipment and facilities, we examined the assets and equipment used directly for treatment, diagnostic tools and equipment, assets and equipment used for patient care during treatment, assets and property of the background infrastructure for treatment, and assets and property of the infrastructure for patient care.³⁴

In the second phase, the previously collected data were confirmed and refined through in-depth interviews (4) with the specialists responsible for each of the examined protocols. The interviews lasted 1–2 h per protocol, during which we first clarified (for both face-to-face and telemedicine procedures) the scope of activities, their interdependence and possible parallelism, the range and roles of actors involved in each activity, and the range and usage time of additional resources required for each activity, according to the categories previously recorded. The interviews were conducted between January and July 2021 and resulted in (5) the activity times for each protocol (6), the effective capacities, and capacity cost rates for each human and equipment capacity. In addition, the controlling expert responsible for reporting and the program manager responsible for the pilot program were involved in the assessment and refinement of the information obtained during each interview, in accordance with the TDABC methodology.

In the third phase of the study, the results of the interviews were coded (8) for clarity and comparability, and the costs of each (face-to-face and telemedicine) protocol were calculated (9) using information from the clinic’s staffing and procurement system.

Main Findings

We present the process costs of four pairs of health care protocols included in the study in relation to each other. Table 1 displays the cost type groups accounted in the TDABC methodology, where we considered 100% of the total process costs we had counted in the traditional protocols. The table also shows the percentage of process costs for each of the telemedicine protocols examined, both by cost type and in aggregate. For staff expenses, the basic salary per person and social contribution tax (13%) were taken into account for each job role related to the implementation of the protocol. For the inputs used, the unit net purchase price of the input and the related VAT were considered. In the case of assets, in addition to the net purchase price and VAT, we calculated different depreciation periods (0, 3, and 7 years). The calculations were made according to the order of magnitude of the purchase price of the asset (there are three main categories: <100,000 Hungarian forints [HUF] [approximately 255 USD]; 100,000–500,000 HUF [approximately 255–1,280 USD], >500,000 HUF [approximately 1,280 USD]) and the Hungarian public procurement procedures. Finally, in the case of real estate, the maintenance costs per square meter of real estate and the annual depreciation value per square meter were factored in, as well as the square meters required for each related activity. Then, for each resource, we calculated the expenditure per unit of time (minutes) and multiplied it by the unit of time of resource use for each process step to obtain the total expenditure for that resource.

Table 1.

Comparison of the Relative Process Costs of Protocols by Cost Type

TYPES OF COSTS	FACE-TO-FACE PROTOCOLS	TELEMEDICINE PROTOCOLS
TYPES OF COSTS	FACE-TO-FACE PROTOCOLS	CASE #1	CASE #2	CASE #3	CASE #4
Staff expenses	100.00%	75.77%	146.50%	44.73%	223.39%
Inputs	100.00%	93.14%	101.63%	27.12%	104.67%
Assets	100.00%	42.63%	13,711.50%	139.85%	400.16%
Properties	100.00%	101.30%	216.53%	28.43%	181.75%
Total	100.00%	80.37%	902.90%	45.29%	161.01%

Source: Own compilation.

CASE #1—EAR, NOSE, AND THROAT

The diagnosis of ear, nose, and throat diseases and the corresponding therapeutic recommendations traditionally are made in the outpatient clinic in the presence of a specialist. With otoscopic imaging and telemedicine software to support the transmission and analysis of clinical data, the telemedicine protocol allowed the specialist to view the results sent remotely on their own smart device and make a diagnosis based on them off-site. In the telemedicine protocol, staff expenses have been reduced (75.77%), as the presence of a resident has become sufficient to replace the specialist and the assistant involved in the traditional protocol. In the case of inputs used, the role of disposable tools (e.g., mask, paper, pen), which are directly necessary for the examination, has been eliminated, resulting in fewer input-related costs for the telemedicine protocol (93.14%). Although new tools have been introduced in the telemedicine protocol (e.g., otoscopic camera, computer for data analysis, smart devices), the basic equipment for face-to-face examination have been downsized, reducing the associated costs by around 40%. The location of care has also changed, with activities being carried out by the resident in a medical room rather than in the outpatient room where the specialist normally resides, although this has led to a minimal increase in the costs associated with properties (101.30%).

CASE #2—DIALYSIS

While the traditional protocol is to perform peritoneal dialysis at the clinic using the infrastructure there, the telemedicine protocol involves the use of telemedicine software and smart devices to self-monitor the at-home dialysis. A moderate increase in personnel costs can be detected (146.50%), as the telemedicine support for home dialysis requires further training in the use of smart devices and telemedicine software by a specialist nurse, and a nephrology nurse regularly checks the data sent during home dialysis, and, if necessary, an unscheduled visit is organized. In the traditional protocol, minimal input materials were used, and in the telemedicine protocol, the additional use of these inputs (pens, paper, masks, disposable treatment sheets) was not significant (101.63%) during an incidental unscheduled visit. In contrast, the purchase of home dialysis equipment (tablet, smart weight scale, smart blood pressure monitor, smart kitchen scale) resulted in a very high-cost increase (13,711.50%), which led to a significant additional cost increase during the telemedicine protocol. Additionally, the use of the nurses’ room for analyzing the data automatically recorded and sent by the telemedicine devices and the cost of using the examination room for unscheduled visits led to an additional increase in costs (216.53%).

CASE #3—METABOLIC SYNDROME

With the help of a dietician and a physiotherapist, the face-to-face protocol involves 3 weeks of hospital rehabilitation for patients with metabolic syndrome, while the telemedicine protocol involves patients to keep a food diary and doing exercises at home, thus receiving therapy at home with a few hospital check-ups. The telemedicine data on dietary and physical activity are sent through smart devices and are regularly monitored by the dietician and physiotherapist at the clinic, overall causing a cost decrease in personnel costs (44.73%). There is also a significant reduction in the cost of inputs used, as there is no need for disposable materials used and patient meals during hospital stays, resulting in a reduction of input costs by approximately a third (27.12%).

Traditional rehabilitation is more human rather than equipment intensive, but in the telemedicine protocol, the purchase of smart devices and software for monitoring patients at home has significantly increased the equipment type costs (139.85%). The 3-week hospital therapy requires a considerable amount of property, but home rehabilitation in the telemedicine protocol frees up real estate capacity and thus a considerable amount of costs (28.43%), which reduces the costs of this type of expenditure by about a third.

CASE #4—CARDIOLOGY

As in previous cases, the telemedicine protocol for monitoring patients with heart failure involved the use of smart devices (a smart sphygmomanometer, a smart body weight scale, and telemedicine software for data transmission). Patients need to be educated on using the software and these devices, which requires extra work on the part of a specialist assistant. An additional cardiology nurse is involved to analyze the transmitted data, resulting in an increase of personnel costs (223.39%). As for inputs, home monitoring does not require clinical inputs; however, the cardiology nurse has to use pens, paper, and hospital uniforms at the clinic for telemedicine consultations, resulting in a small increase in input costs (104.67%). As in the previous cases, the equipment requirements differ considerably (400.16%), as the telemedicine protocol involves months of home monitoring, which requires the purchase of clinical equipment allowing remote monitoring (see above). Finally, a cardiology nurse on-site must conduct the analysis of the data sent during remote monitoring. Data analysis is performed in an examination room during the course of the telemedicine protocol; thus, the use of this property results in an overall cost increase of approximately 182%.

Summary

As a result of our analysis, we can conclude that it is not sufficient to examine only the costs of the modified steps and the change in outcomes when introducing a new technology to examine the change in value. The new technology used leads to modifications in the protocols at numerous points, sometimes at points far from the intervention, which have a significant impact on the change in costs. Besides the recording of outcomes, TDABC is a useful tool for monitoring the change in value because it requires a systematic analysis of protocols as well as an evaluation of resources used for the protocols and their intensity of use.

Our results are promising; however, a few limitations need to be considered. Four cases have been studied; thus, the number of cases does not allow us to generalize our experiences. Moreover, some cases contain specialty-specific information; thus, it would be worth comparing several protocols per medical specialty case separately. Finally, all four cases were studied in one institution. Therefore, specific features of this organizational culture may have an impact on the results, which we were not able to filter out. Overall, our studies have also provided evidence that technology development does not necessarily increase the value of health protocols. In the cases we studied, if technology development resulted in higher (human) net capacity utilization savings than the additional expenditure of technology use, the development created positive value. Otherwise, no additional value creation could be detected. Our experience suggests that TDABC is a potential tool for assessing the complex effects of technological change on a given protocol, capturing changes in the value created in health care.

Footnotes

Authors’ Contributions

M.V.: Conceptualization (lead), formal analysis (lead), methodology (lead), writing—original draft (equal), writing—review and editing (supporting), project administration (lead), and supervision (lead). A.R.M.: Conceptualization (supporting), formal analysis (supporting), visualization (lead), writing—original draft (equal), and writing—review and editing (lead).

Disclosure Statement

The authors have no conflicts of interest to disclose.

Funding Information

No funding was received for this article.

References

Parikh

, Chang

, Kishan

, et al. Time-driven activity-based costing analysis of telemedicine services in radiation oncology. Int J Radiat Oncol Biol Phys, 2020; 108(2):430–434; doi: 10.1016/j.ijrobp.2020.06.053

Middleton

, Simpson

, Bettger

, et al. COVID-19 pandemic and beyond: Considerations and costs of telehealth exercise programs for older adults with functional impairments living at home-lessons learned from a pilot case study. Phys Ther, 2020; 100(8):1278–1288; doi: 10.1093/ptj/pzaa089

Porter

. What is value in health care? N Engl J Med, 2010; 363(26):2477–2481; doi: 10.1056/NEJMp1011024

Kaplan

, Porter

. How to solve the cost crisis in health care. Harv Bus Rev, 2011; 89(9):46–52, 54, 56-61 passim.

Kaplan

, Anderson

. Time-driven activity-based costing. Harv Bus Rev, 2004; 82(11):131–138, 150; doi: 10.2139/ssrn.485443

Kaplan

, Witkowski

, Abbott

, et al. Using time-driven activity-based costing to identify value improvement opportunities in healthcare. J Healthc Manag, 2014; 59(6):399–412; doi: 10.1097/00115514-201411000-00005

Garoon

, Lin

, Young

, et al. Cost savings analysis for a diabetic retinopathy teleretinal screening program using an activity-based costing approach. Ophthalmol Retina, 2018; 2(9):906–913; doi: 10.1016/j.oret.2018.01.020

Bauer

, Nguyen-Phan

, Ong

, et al. Cardiology electronic consultations: Efficient and safe, but consultant satisfaction is equivocal. J Telemed Telecare, 2020; 26(6):341–348; doi: 10.1177/1357633X19828130

Rodrigues

DLG

, Belber

, Padilha

FVDQ

, et al. Impact of teleconsultation on patients with type 2 diabetes in the Brazilian public health system: Protocol for a randomized controlled trial (TELEconsulta Diabetes Trial). JMIR Res Protoc, 2021; 10(1):e23679; doi: 10.2196/23679

10.

Etges

, Cruz

, Notti

, et al. An 8-step framework for implementing time-driven activity-based costing in healthcare studies. Eur J Health Econ, 2019; 20(8):1133–1145; doi: 10.1007/s10198-019-01085-8

11.

Zanotto

, Etges

, Siqueira

, et al. Avaliação Econômica de um Serviço de Telemedicina para ampliação da Atenção Primária à Saúde no Rio Grande do Sul: O microcusteio do Projeto TeleOftalmo. Ciênc Saúde Coletiva, 2020; 25(4):1349–1360; doi: 10.1590/1413-81232020254.28992019

12.

Portney

, Ved

, Nikolian

, et al. Understanding the cost savings of video visits in outpatient surgical clinics. Mhealth, 2020; 6:32–32; doi: 10.21037/mhealth-20-33

13.

Keel

, Muhammad

, Savage

, et al. Time-driven activity-based costing for patients with multiple chronic conditions: A mixed-method study to cost care in a multidisciplinary and integrated care delivery centre at a university-affiliated tertiary teaching hospital in Stockholm, Sweden. BMJ Open, 2020; 10(6):e032573; doi: 10.1136/bmjopen-2019-032573

14.

Cidav

, Marcus

, Mandell

, et al. Programmatic costs of the telehealth ostomy self-management training: An application of time-driven activity-based costing. Value Health, 2021; 24(9):1245–1253; doi: 10.1016/j.jval.2021.03.018

15.

Sadri

, Sinigallia

, Shah

, et al. Time-driven activity-based costing for cataract surgery in Canada: The case of the Kensington Eye Institute. Healthc Policy, 2021; 16(4):97–108; doi: 10.12927/hcpol.2021.26496

16.

Oderanti

, Li

, Cubric

, et al. Business models for sustainable commercialisation of digital healthcare (eHealth) innovations for an increasingly ageing population. Technolog Forecast Soc Change, 2021; 171:120969; doi: 10.1016/j.techfore.2021.120969

17.

Etges

, Polanczyk

, Urman

. A standardized framework to evaluate the quality of studies using TDABC in healthcare: The TDABC in Healthcare Consortium Consensus Statement. BMC Health Serv Res, 2020; 20(1):1107; doi: 10.1186/s12913-020-05869-0

18.

Etges

, Ruschel

, Polanczyk

, et al. Advances in value-based healthcare by the application of time-driven activity-based costing for inpatient management: A systematic review. Value Health, 2020; 23(6):812–823; doi: 10.1016/j.jval.2020.02.004

19.

Popat

, Gracia

, Guzman

, et al. Using time-driven activity-based costing to model the costs of various process-improvement strategies in acute pain management. J Healthc Manag, 2018; 63(4):e76–e85; doi: 10.1097/JHM-D-16-00040

20.

Etges

, Zanotto

, Ruschel

, et al. Telemedicine versus face-to-face care in ophthalmology: Costs and utility measures in a real-world setting. Value Health Reg Issues, 2022; 28:46–53; doi: 10.1016/j.vhri.2021.06.011

21.

Kaplan

, Anderson

. Time-Driven Activity-Based Costing: A Simpler and More Powerful Path to Higher Profits. Harv Bus Rev. Press: Boston; 2007.

22.

Dooley

, Simpson

, et al. A modification of time-driven activity-based costing for comparing cost of telehealth and in-person visits. Telemed J E Health, 2022; 28(10):1525–1533; doi: 10.1089/tmj.2021.0338

23.

Goh

OQM

, Xin

, Lim

, et al. Economic evaluation of novel models of care for patients with acute medical problems. JAMA Netw Open, 2023; 6(9):e2334936; doi: 10.1001/jamanetworkopen.2023.34936

24.

Kuo

, Anjum

, Nguyen

, et al. Utilization of remote diabetic retinal screening in a suburban healthcare system. Clin Ophthalmol, 2021; 15:3865–3875; doi: 10.2147/OPTH.S330913

25.

O’Donnell

, Haberli

, Martinez

, et al. Telehealth visits after shoulder surgery: Higher patient satisfaction and lower costs. J Am Acad Orthop Surg Glob Res Rev, 2022; 6(7); doi: 10.5435/JAAOSGlobal-D-22-00119

26.

Keel

, Savage

, Rafiq

, et al. Time-driven activity-based costing in health care: A systematic review of the literature. Health Policy, 2017; 121(7):755–763; doi: 10.1016/j.healthpol.2017.04.013

27.

Ley

. Eficiencia productiva: Un estudio aplicado al sector hospitalario. Investigaciones Economicas, 1991; 15(1):71–88.

28.

Byrnes

, Valdmanis

. Analyzing technical and allocative efficiency of hospitals. In: Data Envelopment Analysis: Theory, Methodology, and Applications. ( Charnes

, Cooper

, Lewin

, et al. eds) Springer Netherlands: Dordrecht; 1994; pp. 129–144; doi: 10.1007/978-94-011-0637-5_7

29.

Lopez-Valcarcel

, Perez

. Changes in the efficiency of Spanish public hospitals after the introduction of program-contracts. Investigaciones Economicas, 1996; 20(3):377–402.

30.

Maniadakis

, Thanassoulis

. Assessing productivity changes in UK hospitals reflecting technology and input prices. Appl Econ, 2000; 32(12):1575–1589; doi: 10.1080/000368400418970

31.

Farmer

, Stevens

, Hirst

, et al. Optimal strategies for identifying kidney disease in diabetes: Properties of screening tests, progression of renal dysfunction and impact of treatment—systematic review and modelling of progression and cost-effectiveness. Health Technol Assess, 2014; 18(14):1–128; doi: 10.3310/hta18140

32.

Chattopadhyay

, Ray

. Technical, scale, and size efficiency in nursing home care: A nonparametric analysis of Connecticut homes. Health Econ, 1996; 5(4):363–373; doi: 10.1002/(SICI)1099-1050(199607)5:4<363::AID-HEC217>3.0.CO;2-1

33.

Parkin

, Hollingsworth

. Measuring production efficiency of acute hospitals in Scotland, 1991–94: Validity issues in data envelopment analysis. Appl Econ, 1997; 29(11):1425–1433; doi: 10.1080/000368497326255

34.

Lavy

, Shohet

. Integrated maintenance management of hospital buildings: A case study. Construct Manag Econ, 2004; 22(1):25–34; doi: 10.1080/0144619042000186031