Determining the cost of implementing and operating a remote patient monitoring programme for the elderly with chronic conditions: A systematic review of economic evaluations

Abstract

Introduction

Remote patient monitoring (RPM) in conjunction with home nursing visits is becoming increasingly popular for the follow-up of patients with chronic conditions and evidence exists that it improves patients’ health outcomes. Current cost data is reported inconsistently and often gathered from studies of poor methodological quality, making it difficult for decision-makers who consider implementing this service in their organizations. This study reviewed the cost of RPM programmes targeting elderly patients with chronic conditions.

Methods

After evaluation against the inclusion and exclusion criteria and appraisal against two criteria which are important for economic evaluations, data from selected studies were extracted and grouped into meaningful cost categories, then adjusted to reflect November 2015 US dollars.

Results

In the 13 selected studies, the newly-created cost category ‘Combined intervention cost’ (reflecting equipment purchasing, servicing and monitoring cost) for the various RPM programmes ranged from US$275–US$7963 per patient per year. The three main findings are: (a) RPM programme costs have decreased since 2004 due to cheaper technology; (b) monitoring a single vital sign is likely to be less costly than monitoring multiple vital signs; and (c) programmes targeting hypertension or congestive heart failure are less costly than those targeting respiratory diseases or multiple conditions.

Conclusions

This review recommends that future studies present their cost data with more granularity, that grouping of costs should be minimized and that any assumptions, such as amortization, should be made explicit. In addition, studies should compare programmes with similar characteristics in terms of type of conditions, number of vital signs monitored, etc. for more generalizable results.

Keywords

Health economics home telecare telecare telehealth telemedicine

Introduction

Today, digital technology for the purpose of providing healthcare services is increasingly making its way into patients’ homes. One of its forms, remote patient monitoring (RPM), consists of the transmission of healthcare data from a patient to a healthcare provider in different locations for assessment and intervention purposes and is used primarily to support the care and self-management of patients with chronic conditions. Incorporating RPM in disease management is beneficial in improving clinical outcomes and medication compliance and in reducing cost related to patient travel, number of emergency room visits and hospital admissions.^1–4 However, considering the cost of RPM is essential in justifying these services within the current context of budgetary constraints and austerity measures. Often RPM is integrated into other home healthcare services, making it challenging to isolate and allocate its cost appropriately.⁵ The assessment of cost effectiveness when comparing RPM programmes is further complicated by the use of different RPM systems, the focus of programmes on one or multiple patient diagnoses, and the monitoring of one or multiple vital signs.^5–8 Hence, these factors, in addition to the poor methodological quality of the available literature, makes it difficult for healthcare organizations to rely on these findings as a basis for estimating costs when planning RPM services. Organizations are interested in RPM programme cost in dollars in addition to cost-effectiveness metrics, however most financial systematic reviews limit themselves to the latter, expressing results as incremental cost effectiveness ratios often without presenting the underlying cost data. Therefore this economic review has determined the actual cost of RPM programmes, limited to those targeting seniors with chronic conditions. Data from primary studies were grouped according to cost categories and diagnoses allowing for a meaningful comparison.

Methods

A search of the literature was conducted on various databases (CINAHL, Cochrane Library, EMBASE, Medline, PsycINFO and PubMed) until November 2015, using the search strategy and key terms as elaborated in Table 1. All primary studies, obtained from the review and cross-referencing, that were written in English and disclosed the cost for RPM services solely or in combination with other home healthcare services that target seniors with an average age greater than or equal to 65 years and having at least one chronic condition were included. Any studies that used Markov or other cost simulation models were excluded, as were any studies that did not separately report the costs of RPM equipment. After removing duplicates and screening abstracts, a full review was conducted on the remaining articles as well as appraisal using two criteria from the Drummond et al.⁹ checklist, which were deemed of critical importance to this economic review. These criteria were: (a) the economic importance of the research question is stated; and (b) the quantities of resource use are reported separately from their unit costs. The first criterion ensures that financial conclusions are of primary importance thus minimizing the risk of including papers presenting low quality cost data and the second criterion ensures that cost figures can be adjusted to scale up or down in terms of quantity of resource use. After this appraisal, the final studies were used for data extraction and analysis. Due to the fact that the included studies reported costs at varying degrees of granularity, six cost categories related to monitoring, equipment servicing, equipment purchasing, home visits, emergency room visits and hospitalization were created to classify all reported costs. What follows is a description of each cost category. The cost category ‘monitoring’ includes all costs related to the monitoring of patients’ health information remotely including the nurses’ time providing telephone intervention. The equipment servicing category (‘servicing’) includes the cost of the technician, modem and other recurring monthly expenses related to operating the RPM equipment. The third category, equipment purchasing (‘equipment’), includes costs related to the installation of the equipment, the training of employees and patients, home RPM terminals or mobile devices, providers’ computers and RPM software, and any other equipment required for the provision of the RPM services. This category also included equipment maintenance costs when these were reported grouped together with equipment costs. The fourth cost category included all costs, intervention and travel time, related to making home visits by various health professionals. Emergency room visits and hospitalization cost, the final two cost categories, included standard fees associated with treatment.

Table 1.

Search strategy.

1.	Telemedicine/OR telemedicine OR telecare OR telehealth OR remote care OR telenursing
2.	Remote monitoring OR telemonitoring OR telemetric OR telehomecare OR telemanagement
3.	eHealth OR e-health OR mhealth OR m-health OR telerehabilitation OR telesurveillance
4.	1 OR 2 OR 3
5.	Home Care Services/OR home care OR homecare OR home OR telehomecare
6.	Home telecare OR home telehealth
7.	5 OR 6
8.	‘Costs and cost analysis'/OR cost-benefit analysis/OR cost analysis OR cost-effectiveness
9.	Cost-utility OR cost-benefit OR economic evaluation OR economics
10.	8 OR 9
11.	Aged/or Health Services for the Aged/OR ageing OR aging OR elderly OR seniors
12.	Geriatrics OR veterans OR older people
13.	11 OR 12
14.	4 AND 7 AND 10 AND 13

All data were adjusted and presented in a standardized way, on a per patient per year basis, in order to make the data comparable. Local currencies were converted to US dollars using exchange rates published in individual studies, where available, or otherwise using the World Bank purchasing power parity conversion factor¹⁰ for the study period. Subsequently, costs were adjusted to November 2015 values using the Consumer Price Index for medical care¹¹ and presented as ranges and lists rather than aggregates or averages. The Wilcoxon rank-sum test was used to identify significant cost differences between subgroups with different characteristics associated to various RPM programmes, such as varying geographic locations, the monitoring of varying number of medical conditions and the monitoring of varying numbers of vital statistics. A p-value of less than 0.05 was considered statistically significant.

Results

As indicated in Figure 1, a total of 687 studies were obtained from the search strategy (n = 666) and cross-referencing (n = 21). After removing duplicates (n = 101) and screening of the abstracts (n = 494), the remaining 92 articles underwent a full review, of which 71 were eliminated when re-evaluated against the inclusion and exclusion criteria. A total of eight articles were excluded following an appraisal using the Drummond et al.⁹ checklist, leaving 13 studies for data extraction and analysis.

Figure 1.

Study selection results.

Table 2 presents characteristics of the selected studies.^12–24 With regards to the geographical origin, the first identified characteristic, all studies originated from North America, Europe and New Zealand and were published between 2004 and 2015. The duration of RPM service provision varied greatly from three months to 2.5 years, and the number of participants ranged from four to a sample of 538 seniors suffering from various chronic conditions including chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), diabetes mellitus (DM) and hypertension (HTN). As for the number of chronic conditions being monitored, the second identified characteristic, six^{13,15,17,18,20,22} of the 13 studies targeted an elderly population focusing on a single chronic condition, while the remaining seven studies^{12,14,16,19,21,23,24} monitored multiple conditions. In terms of the third and final identified characteristic, three studies^17,18,20 monitored a single vital sign, while all others targeted multiple.

Table 2.

Overview of selected studies.

Author	Year	Country	Patient count	Intervention period (months)	Targeted diagnosis	Vitals monitored
Kenealy et al.¹²	2015	New Zealand	98	4.5	COPD, CHF or DM	Multiple
Boyne et al.¹³	2013	Netherlands	197	12.0	CHF	Multiple^a
Henderson et al.¹⁴	2013	UK	538	3.0	COPD, CHF or DM	Multiple
Pare et al.¹⁵	2013	Canada	60	5.9	COPD	Multiple
Pare et al.¹⁶	2013	Canada	95	4.0	CHF, DM, COPD or HTN	Multiple
Rifkin et al.¹⁷	2013	USA	28	6.0	HTN with CKD	Blood pressure
Stoddart et al.¹⁸	2013	UK	200	6.0	HTN	Blood pressure
Hicks et al.¹⁹	2009	USA	47	6.0	Post-surgery, CVD, lung disease, cancer, DM, infections or other	Multiple
Vitacca et al.²⁰	2009	Italy	118	12.0	Chronic respiratory failure^b	Pulse oximetry
Finkelstein et al.²¹	2006	USA	20	30.0	COPD, CHF or chronic wound care	Multiple
Pare et al.²²	2006	Canada	19	6.0	COPD	Multiple
Walsh et al.²³	2005	USA	4	3.9	Heart disease or DM	Multiple
Noel et al.²⁴	2004	USA	47	6.0	COPD, CHF or DM	Multiple

CHF: congestive heart failure; CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; CVD: cardiovascular disease; DM: diabetes mellitus; HTN: hypertension.

Refers to multiple symptoms being monitored, not vitals.

Targets patients with chronic respiratory failure who receive home mechanical ventilation and/or long term oxygen therapy.

The costs associated with health services utilization, which is a primary concern to healthcare organizations, can fluctuate across geographies and between hospitals within close proximity due to differing cost and reimbursement structures and various models of care. In this instance, it is more meaningful to present such findings in terms of utilization per person per year (see Table 3), as opposed to costs so that individual healthcare organizations can apply these figures to their own cost structures in order to elicit the associated costs of care.

Table 3.

Healthcare utilization rates (per patient per year).

Author	Year	Targeted diagnosis	Home visits (nurse or other)	Home visit minutes (nurse or other)	Hospitalizations	Hospitalization days	ED visits
Kenealy et al.¹²	2015	COPD, CHF or DM	16.0	na	na	na	na
Boyne et al.¹³	2013	CHF	0.4	na	na	11.4	0.6
Henderson et al.¹⁴	2013	COPD, CHF or DM	0.0	na	na	na	na
Pare et al.¹⁵	2013	COPD	11.5	548	0.9	13.3	1.2
Pare et al.¹⁶	2013	CHF, DM, COPD or HTN	6.1	203	0.6	5.8	1.2
Rifkin et al.¹⁷	2013	HTN with CKD	0.0	na	na	na	na
Stoddart et al.¹⁸	2013	HTN	0.2	na	na	na	0.1
Hicks et al.¹⁹	2009	Post-surgery, CVD, lung disease, cancer, DM, infections or other	35.6	2089	0.8	4.7	na
Vitacca et al.²⁰	2009	Chronic respiratory failure^a	7.4	na	1.3	na	0.6
Finkelstein et al.²¹	2006	COPD, CHF or chronic wound care	159.5	7339	na	na	na
Pare et al.²²	2006	COPD	8.4	483	0.2	2.8	na
Walsh et al.²³	2005	Heart disease or DM	82.1	na	na	na	na
Noel et al.²⁴	2004	COPD, CHF or DM	4.0	na	na	na	2.1

CHF: congestive heart failure; CKD: chronic kidney disease; COPD: chronic obstructive pulmonary disease; CVD: cardiovascular disease; DM: diabetes mellitus; ED: emergency department; HTN: hypertension; na: data not available in the underlying reports.

Targets patients with chronic respiratory failure who receive home mechanical ventilation and/or long term oxygen therapy.

Tables 4 and 5 show standardized numbers by reporting total cost in US dollars for each of the six cost categories, adjusted for inflation to reflect figures in November 2015. Across the selected studies, the costs associated with purchasing equipment are presented inconsistently; some amortise equipment costs over a number of years, while other do not. In this review, we assume equipment costs are proportional to the number of participants enrolled in a given study as most equipment and related peripherals for remote monitoring purposes can be found in the patients’ home environment. Taking into account the number of patients, we then calculated the total equipment cost per individual patient and amortised this amount over one, three and five ‘useful life years’ for the equipment assuming a discount rate of 5.24% representing the average cost of capital for hospitals;²⁵ these figures are presented as equivalent annual costs. As the maximum duration of the RPM programmes in the studies was 2.5 years, this was inadequate to suggest a figure to represent the ‘useful life years’ of the equipment before the need for replacement. In any event, such a figure would need be closely related to the specific equipment purchased and patterns of use and therefore it is most appropriate to present these figures over a range. When considering a three-year amortization period as a reference, the resulting annual equipment cost per patient ranged from US$74–US$5809. In addition, when incorporating monitoring and equipment servicing cost to the equipment purchasing cost amortised over three years, referred to as ‘combined intervention cost’, we arrive at a range of US$275–US$7963 per patient per year. It is not helpful to include home and emergency room visits and hospitalization costs as these metrics are highly variable and dependent on how RPM programmes and care protocols were designed, which in our case, were inconsistent across studies. Therefore we suggest that healthcare organizations considering the implementation of an RPM programme use ‘combined intervention cost’ plus their local unit costs for home and emergency room visits and hospitalizations multiplied by the utilization rates (see Table 3) in order to have a reliable estimate of the cost of an RPM programme for a particular institution.

Table 4.

Cost summary table (per patient per year; presented in US$ adjusted to November 2015).

				Equipment purchasing^a			Combined intervention cost^a
Author	Year	Monitoring	Servicing	1	3	5	1	3	5
Kenealy et al.¹²	2015	$566	$872	$754	$278	$175	$2192	$1716	$1613
Boyne et al.¹³	2013	$0	$0	$746	$275	$173	$746	$275	$173
Henderson et al.¹⁴	2013	$2070	$0	$1508	$556	$351	$3578	$2626	$2420
Pare et al.¹⁵	2013	$791	$0	$2492	$919	$579	$3283	$1710	$1371
Pare et al.¹⁶	2013	$1166	$0	$4226	$1559	$983	$5392	$2725	$2149
Rifkin et al.¹⁷	2013	$0	$328	$251	$92	$58	$579	$421	$387
Stoddart et al.¹⁸	2013	$133	$106	$201	$74	$47	$440	$313	$286
Hicks et al.¹⁹	2009	$0	$0	$3032	$1118	$705	$3032	$1118	$705
Vitacca et al.²⁰	2009	$0	$1163	$255	$94	$59	$1417	$1257	$1222
Finkelstein et al.²¹	2006	$2707	$998	$4596	$1695	$1069	$8301	$5400	$4774
Pare et al.²²	2006	$219	$0	$15,061	$5555	$3502	$15,280	$5775	$3721
Walsh et al.²³	2005	$2170	$0	$15,704	$5793	$3651	$17,875	$7963	$5822
Noel et al.²⁴	2004	$0	$0	$15,749	$5809	$3662	$15,749	$5809	$3662

Equipment purchasing prices are presented amortized over one, three and five years and presented as equivalent annual costs.

Table 5.

Extended cost summary table (per patient per year; presented in US$ adjusted to November 2015).

			Intervention + home visit^a					Total cost^a
Author	Year	Home Visits	1	3	5	ED	Hospitalization	1	3	5
Kenealy et al.¹²	2015	$183	$2374	$1899	$1796	na	na	$2374	$1899	$1796
Boyne et al.¹³	2013	$46	$791	$321	$219	$128	$10,676	$11,595	$11,124	$11,023
Henderson et al.¹⁴	2013	$0	$3578	$2626	$2420	na	na	$3578	$2626	$2420
Pare et al.¹⁵	2013	$434	$3717	$2144	$1804	$1097	$4629	$9443	$7870	$7531
Pare et al.¹⁶	2013	$1841	$7233	$4566	$3989	$181	$2148	$9562	$6895	$6318
Rifkin et al.¹⁷	2013	$0	$579	$421	$387	na	na	$579	$421	$387
Stoddart et al.¹⁸	2013	$29	$468	$342	$315	$24	na	$492	$366	$338
Hicks et al.¹⁹	2009	$1448	$4479	$2566	$2153	na	$12,715	$17,194	$15,281	$14,868
Vitacca et al.²⁰	2009	$240	$1658	$1497	$1462	$62	$13,976	$15,696	$15,535	$15,500
Finkelstein et al.²¹	2006	$11,039	$19,340	$16,439	$15,812	na	na	$19,340	$16,439	$15,812
Pare et al.²²	2006	$504	$15,784	$6279	$4225	na	$1993	$17,777	$8272	$6218
Walsh et al.²³	2005	$14,991	$32,866	$22,954	$20,813	na	na	$32,866	$22,954	$20,813
Noel et al.²⁴	2004	$586	$16,335	$6395	$4248	$794	$3103	$20,232	$10,292	$8145

na: data not available in the underlying reports.

Equipment purchasing prices are presented amortized over one, three and five years and presented as equivalent annual costs.

Table 6 presents the ‘combined intervention cost’ per patient per year as an average cost calculated from the 13 studies and further subdivides these costs according to geography, the number of (single or multiple) vital signs monitored and the number of (single or multiple) chronic conditions included. There were three main findings of the study: however only one identified characteristic of the RPM programmes across studies achieved statistical significance. There was no statistically significant difference found between RPM programmes implemented in North America and those in Europe and New Zealand; however, the reported costs of RPM programmes from the North American studies seemed to be higher. RPM programmes including patients with CHF or HTN are significantly less costly in terms of ‘combined intervention cost’ than those focusing on patients with respiratory diseases or where multiple conditions were monitored (p < 0.05). Finally, RPM programmes monitoring a single vital sign seemed to be less costly when compared to those monitoring multiple vital signs, although not statistically significant.

Table 6.

Average costs (per patient per year; presented in US$ adjusted to November 2015).

	Combined intervention cost^a
Category	1	3	5
Number of vitals monitored
Single^17,18,20	$812	$663	$631
Multiple^{12–16,19,21–24}	$7543	$3512	$2641
Conditions targeted
Congestive heart failure¹³	$746	$275	$173
Respiratory diseases^15,20,22	$6660	$2914	$2104
Hypertension^17,18	$509	$367	$336
Multiple conditions^{12,14,16,19,21,23,24}	$8017	$3908	$3021
Geography
North America^{15–17,19,21–24}	$8686	$3865	$2824
Europe^13,14,18,20	$1545	$1118	$1025
New Zealand¹²	$2192	$1716	$1613

Note: These figures represent simple averages of the studies that belong to each category.

Equipment purchasing prices are presented amortized over one, three and five years.

With regards to the relationship between the year of publication and the cost of the RPM programmes, Figure 2 shows that the average ‘combined intervention cost’ per patient per year has decreased steadily from 2004 to December 2015. This reduction in cost can be explained by the fact that RPM equipment has not changed substantially over time and technologies tend to reduce in cost over time. Generally, healthcare organizations have not felt the need to transition to more costly, modern equipment because RPM programmes can be run effectively on cheaper, older equipment thus benefiting from lower costs.

Figure 2.

Cost (per patient per year; US$, November 2015) vs year of publish.

Lastly, in Figure 3, we present this same cost data plotted against study sample sizes. The average cost per patient seems to drop with increases in sample size. This finding suggests that even though equipment costs are assumed to be proportional to the number of patients enrolled, in fact, a large proportion of the cost is likely to be fixed. This suggests that programmes enrolling large number of patients could be significantly less costly on a per patient basis than their smaller counterparts. It also provides a rationale to suggest that future studies should explicitly present fixed equipment costs separately from those that are variable and therefore proportional to the number of patients enrolled in the study.

Figure 3.

Cost (per patient per year; US$, November 2015) vs sample size.

Discussion

In this discussion, the following questions will be addressed: (a) what conclusions can be drawn from these cost findings; (b) what might explain the fact that certain identified characteristics are less costly to monitor than others; and (c) what recommendations can emerge from this analysis to improve the quality of future cost studies?

It is evident that any healthcare organization considering the implementation of an RPM programme would find it difficult to base their decision on a cost estimate ranging broadly from US$275–US$7963 per patient per year. One could suppose that these wide ranging results are likely reflective of the inherent variability of RPM programmes. Therefore, it is imperative for organizations searching for cost estimates to identify RPM programmes most similar to those being considered for implementation. However the difficulty lies in determining the characteristics which should be used to decide if a given RPM programme is similar enough to be used as a comparable.

Our results indicated that RPM programmes monitoring a single vital sign were likely to be less costly than those covering multiple. Programmes monitoring one condition (CHF or HTN) seemed less costly than those covering respiratory diseases or multiple conditions and that programmes implemented in North America seemed more costly than those implemented in Europe or New Zealand. Though only one of these findings is statistically significant, one can hypothesize that each of these individual characteristics could potentially contribute to the heterogeneous results obtained, making it all the more important to understand how these factors contribute to cost differences across studies. The first claim is intuitive in that monitoring a single vital sign is understandably less costly than monitoring multiple. Less equipment and related attachments are required, resulting in less complex monitoring; the logical consequence to this is lower cost. In addition, the statistically significant finding that indicates that programmes which cover CHF or HTN are less costly than those which cover respiratory diseases or multiple conditions can be explained by two factors: patient complexity and number of vital signs being monitored. Patients with COPD or multiple chronic conditions are usually more complex to care for than those with HTN; as an indication of complexity, close to 50% of patients with COPD in the USA have five or more chronic conditions compared to just over 20% of those with HTN.²⁶ The second factor, in addition to the complexity of cases, is the effect of the number of vital signs monitored on cost. To further consider this point, it is important to note that two of three studies including CHF or HTN patients only monitored one vital sign. This contributes to the difficulty in identifying which factor contributes to a lower cost. Lastly, geography seems to have an influence on cost, with North American RPM programmes being more expensive when compared to other regions, which does not seem logical and is likely due to other confounding variables. As mentioned in the results, older studies reported higher costs than more recent ones. The four studies published before 2009 (each reporting costs greater than US$5000 per patient per year) are North American studies. Thus, the cost difference could be explained by the age of the studies rather than geographical differences.

There are two major recommendations that emerge from this systematic review. The first is that future research should focus on isolating confounding variables in order to better determine significant relationships between the identified characteristics and the cost of RPM programmes. Furthermore, studies should compare the cost of programmes that are serving a similar patient population and are monitoring a similar number of vital signs in order to better inform decision-makers. Secondly, considering that there is little to no consistency across studies with regards to the manner in which costs are grouped and presented and that descriptions of interventions are often inadequate, another recommendation that can contribute to the quality of financial information reported in future studies is that cost data should be separated into the cost categories suggested above. Some studies^{13,17,19,20,24} neglected to report monitoring costs as these are typically associated to nursing salaries, however it is essential to present these numbers as incremental costs. When monitoring costs are reported, they are often merged with the costs of home visits as the cost of nursing care is typically regrouped, however they should be presented separately as they represent different interventions. In addition, equipment servicing costs are often grouped with equipment purchasing costs; however, they should also be presented separately as one represents an ongoing expense while the latter is an upfront, one-time, expense.

It is fair to use amortization assumptions; however, the number of years utilized needs to be disclosed to the reader as they can have a significant impact on the reported cost. An amortization period should reflect the estimated ‘useful life of equipment’. For studies lasting less than a year, amortization cost should be prorated. Future studies should disclose total equipment costs, including both fixed and variable components; fixed cost includes the cost of equipment dedicated for the providers’ use, while the variable cost includes the terminals or mobile devices provided to patients. This would allow for an accurate comparison across studies of different sample sizes.

Despite the fact that RPM programmes usually entail a combination of in-home visits and remote monitoring, the cost of both nursing interventions should be presented separately. For the sake of clarity, ideally, all costs should be broken down into monitoring cost, scheduled visits cost which are built into the programme structure, the cost of urgent need home visits and all cost related to hospitalization or ED visit in terms of utilization volumes.

In conclusion, no reliable cost estimates for the implementation of RPM programmes are available in the literature. On the basis of cost, this review suggests that RPM programmes are more suitable for certain chronic conditions and grouping similar characteristics of an RPM programme, such as the type of condition, the number of vital signs monitored, etc. would yield more useful, comparable data that is generalizable and could serve to inform policy and decision makers.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

References

Poole L. Transforming chronic disease prevention and management through telehomecare. Report. OTN, Canada: Toronto, Ontario Telemedicine Network, 2012.

Mierdel

Owen

. Telehomecare reduces ER use and hospitalizations at William Osler health system. Stud Health Technol Inform 2015; 209: 102–108.

Bowles

Hanlon

Glick

et al.

Clinical effectiveness, access to, and satisfaction with care using a telehomecare substitution intervention: A randomized controlled trial. Int J Telemed Appl 2011; 2011: 1–13.

Hæsum

Ehlers

Hejlesen

. Telehomecare technologies enhance self-management and empowerment among patients with chronic obstructive pulmonary disease (COPD) – Where does health literacy fit into this equation? Stud Health Technol Inform 2013; 192: 1182–1182.

Mistry

. Systematic review of studies of the cost-effectiveness of telemedicine and telecare. Changes in the economic evidence over twenty years. J Telemed Telecare 2011; 18: 1–6.

Grustam

Severens

van Nijnatten

et al.

Cost-effectiveness of telehealth interventions for chronic heart failure patients: A literature review. Int J Technol Assess Health Care 2014; 30: 59–68.

Elbert

van Os-Medendorp

van Renselaar

et al.

Effectiveness and cost-effectiveness of ehealth interventions in somatic diseases: A systematic review of systematic reviews and meta-analyses. J Med Internet Res 2014; 16: e110–e110.

Polisena

Coyle

et al.

Home telehealth for chronic disease management: A systematic review and an analysis of economic evaluations. Int J Technol Assess Health Care 2009; 25: 339–349.

Drummond

O’Brien

Stoddart

et al.

Methods for the economic evaluation of health care programmes, 2nd ed. Oxford: Oxford Medical Publications, 1997.

10.

The World Bank. PPP conversion factor, GDP (LCU per international $), http://data.worldbank.org/indicator/PA.NUS.PPP (2015, accessed 23 December 2015).

11.

United States Department of Labor Bureau of Labor Statistics. Consumer price index – all urban consumers, http://data.bls.gov/timeseries/CUUR0000SAM?output_view=pct_12mths (2015, accessed 23 December 2015).

12.

Kenealy

Parsons

Rouse

et al.

Telecare for diabetes, CHF or COPD: Effect on quality of life, hospital use and costs. a randomised controlled trial and qualitative evaluation. PLOS One 2015; 10: e0116188–e0116188.

13.

Boyne

Van Asselt

Gorgels

et al.

Cost-effectiveness analysis of telemonitoring versus usual care in patients with heart failure (the TEHAF-study). Eur Heart J 2013; 34: P3346–P3346.

14.

Henderson

Knapp

Fernandez

et al.

Cost effectiveness of telehealth for patients with long term conditions (Whole Systems Demonstrator telehealth questionnaire study): Nested economic evaluation in a pragmatic, cluster randomised controlled trial. Br Med J 2013; 346: f1035–f1035.

15.

Paré

Poba-Nzaou

Sicotte

et al.

Comparing the costs of home telemonitoring and usual care of chronic obstructive pulmonary disease patients: A randomized controlled trial. European Research in Telemedicine/La Recherche Européenne en Télémédecine 2013; 2: 35–47.

16.

Paré

Poba-Nzaou

Sicotte

. Home telemonitoring for chronic disease management: An economic assessment. Int J Technol Assess Health Care 2013; 29: 155–161.

17.

Rifkin

Abdelmalek

Miracle

et al.

Linking clinic and home: A randomized, controlled clinical effectiveness trial of real-time, wireless blood pressure monitoring for older patients with kidney disease and hypertension. Blood Press Monit 2013; 18: 8–15.

18.

Stoddart

Hanley

Wild

et al.

Telemonitoring-based service redesign for the management of uncontrolled hypertension (HITS): Cost and cost-effectiveness analysis of a randomised controlled trial. BMJ Open 2013; 3: e002681–e002681.

19.

Hicks

Fleming

Desaulnier

. The application of remote monitoring to improve health outcomes to a rural area. Telemed J E Health 2009; 15: 664–671.

20.

Vitacca

Bianchi

Guerra

et al.

Tele-assistance in chronic respiratory failure patients: A randomised clinical trial. Eur Respir J 2008; 33: 411–418.

21.

Finkelstein

Speedie

Potthoff

. Home telehealth improves clinical outcomes at lower cost for home healthcare. Telemed J E Health 2006; 12: 128–136.

22.

Paré

Sicotte

St.-Jules

et al.

Cost-minimization analysis of a telehomecare program for patients with chronic obstructive pulmonary disease. Telemed J E Health 2006; 12: 114–121.

23.

Walsh

Coleman

. Developing a pilot telehealth program. Home Healthc Nurse 2005; 23: 188–191.

24.

Noel

Vogel

Erdos

et al.

Home telehealth reduces healthcare costs. Telemed J E Health 2004; 10: 170–183.

25.

Damodaran A. New York University Stern School of Business. Cost of capital by sector (US), http://pages.stern.nyu.edu/∼adamodar/ (2015, accessed 28 July 2016).

26.

Centers for Medicare and Medicaid Services. Chronic Conditions among Medicare Beneficiaries, Chartbook, Baltimore, MD, 2012.