Abstract
We carried out a cost-benefit analysis of the STAY telecare service (Sandwell Telecare Assisting You) for unpaid carers of people with dementia in the West Midlands. This was done from the perspective of the carers. We employed the willingness to pay (WTP) approach, an established economic method to quantify the benefits of healthcare. A questionnaire was given to 87 unpaid carers of people with dementia, and 34 completed questionnaires were returned (39%). Two WTP question designs were used (open-ended and bounded style). The annual mean WTP for telecare was £242-310 per person. Carers offered higher values if the cared-for person was suffering from moderate dementia. Carers offered WTP values that were similar to the resource costs of providing telecare services. The pilot study suggests that the provision of telecare for unpaid carers of people with dementia is beneficial to society.
Introduction
Dementia is a progressive condition which affects approximately 800,000 people in the UK and costs £1.4 billion per year. 1 Many people suffering from dementia are cared for by family or friends, who provide unpaid care worth about £30,000 per year. 2 The burden of providing unpaid care is a predictor of poorer quality of life, depression and anxiety in carers. 3 People with dementia (PWD) and their carers indicate a preference for care to be provided at home. However, poor physical health and psychological functioning in carers can necessitate the institutionalisation of PWD to costly residential care. 4
Sandwell Metropolitan Borough Council in the West Midlands has a telecare support service called Sandwell Telecare Assisting You (STAY), which is available to unpaid carers of PWD in the borough who meet certain assessment criteria. The service involves the installation of a range of telecare products, including bed leaving sensors, smoke alarms and heat sensors. The service is designed to help people live more independently by monitoring changes and warning the PWD, or alerting carers at home or at a control centre. 5
Telecare can potentially help to reduce the burden on carers by allowing the PWD to remain at home for longer than would otherwise have been possible. Evaluations of the impact of telecare on carers of PWD have been reported in the literature;6–13 but evidence of the effectiveness of telecare on carer wellbeing is mixed, and does not include consideration of cost.
Cost-benefit analysis (CBA) is a tool to aid public sector decision-makers when choosing between alternatives. The use of CBA to value goods and services under non-market conditions is well established. 14 In the UK health care system, patients receive health care free at the point of use and thus prices and resultant demand cannot be observed in the conventional manner used to establish the value of goods. Hypothetical markets therefore need to be generated, using survey approaches, to elicit hypothetical values from consumers (or patients in the context of health care) to reveal the value of goods and services. For example, if the objective of the analysis is to determine the value of a new drug or treatment, patients will be asked, using a survey, to imagine that a market does exist for the drug or treatment, and contingent upon this hypothetical market, to state their valuation of the associated benefits. The most common method of asking about value is to elicit a person’s maximum willingness to pay (WTP) to receive a benefit from the service. The WTP values are then averaged over the survey sample and compared with cost to assess the social net benefit of the service. Economic theory predicts that if the benefits (measured by the average maximum WTP) are greater than the cost then the service is operating efficiently and should be recommended.
We have carried out a cost-benefit analysis of the STAY telecare service based in the West Midlands. This was done from the perspective of carers of people with dementia.
Methods
The data were collected using a questionnaire, designed for self-completion without supervision. The study was approved by the appropriate ethics committee. To distribute the questionnaire we applied two approaches. In the first approach we identified PWD in receipt of telecare through the STAY database and cross-referenced this information with a local database held by Sandwell Council to identify any unpaid carers. Inclusion criteria for carers were as follows: unpaid carer of PWD, PWD had to be living at home and carer had to be able to give informed consent. The exclusion criteria were any carer who had a cared-for PWD living in either a residential or nursing home or who had passed away. We contacted all possible carers who met the inclusion criteria. The unpaid carer was contacted by the local Sandwell Council carer support team, and scripts were provided for the council staff to use when explaining the study. The carers were told that researchers based at the University of Birmingham were interested in finding out about their experiences of telecare, and asked if they would be willing to participate in the study. The questionnaire was then posted to all who agreed to participate.
In the second approach we contacted unpaid carers through a local support charity Better Understanding of Dementia for Sandwell (BUDS), which hosts a support group for carers of PWD. The BUDS charity invited them to participate in the study and questionnaires were posted to those who agreed to participate.
Carers were able to return completed questionnaires by post in pre-paid reply envelopes. With the second recruitment method it was probable that some carers would have had no direct experience of using telecare services, so the questionnaire was modified to accommodate this.
In the questionnaire the carer was asked to provide details about what telecare service they had previously received using tick-box options for each potential product. For the carers identified through the BUDS charity, the questionnaire contained an extra question to ask if they had previously received any form of telecare and if so, to specify the products. For the carers who had no experience of telecare, they were then asked if they would be willing to have telecare installed in the cared-for person's home. Carers who replied no to this question were asked to simply proceed to the sociodemographic section of the questionnaire. The other responses to this question (Yes and Don’t know) were prompted to move onto the WTP section. The nature of the WTP exercise was then explained and the carer was invited to supply a valuation for the telecare service. The question was: “Suppose Sandwell Borough Council and Sandwell Primary Care Trust do not pay for the cost of your telecare services. What is the maximum monthly amount you would be willing to pay for the service out of your own pocket? Bear in mind that the extra cost will reduce what you have left to spend on yourself on a monthly basis”.
The remainder of the questionnaire, for both formats, asked routine socio-demographic and economic data, including gender, age, employment status and household income. We also requested information about the relationship with the cared-for person, whether they lived with that person, how much time they spent caring in an average week and a subjective judgement as to the severity of the dementia of the cared-for person. We asked about quality of life using the validated EQ-5D 3L instrument 15 and the Carer Strain Index (CSI). 16 The EQ-5D 3L instrument is a standard generic utility-based quality of life instrument that measures quality of life on a 0-1 scale (0 = death, 1 = full health). The CSI is a 13-item tool that measures strain related to care provision and can identify carers with potential caregiving concerns. A positive screen (7 or more positive items) indicates a need for more detailed assessment to facilitate appropriate intervention. When carers supplied a WTP value we asked them to state their reasons why they had chosen that value. These reasons were then analysed by thematic analysis to aid interpretation of the WTP values.
Bootstrapped confidence intervals for mean WTP values were calculated using the STATA bootstrap command. We used the response to the type of telecare service received (in terms of combinations of products) to calculate the average cost of providing such a service. Unit cost information was obtained from relevant sources and expressed in 2011/12 prices. For each application of telecare, there is a preferred product used by the council which was used to estimate the cost. These costs were then offset against the value elicited from the respondents to estimate the net benefit of the telecare service. Further analysis exploring factors influencing WTP values was undertaken using the Kruskal-Wallis test when testing categorical independent variables, and correlation coefficients when testing relationships with ordinal/ratio independent variables. We applied a series of univariate statistical tests to determine if any individual sociodemographic or health-related characteristics were significant predictors of WTP. The evaluation was done using a formal CBA framework. The analysis was performed using a standard package (STATA IC Version 11).
Results
Demographic characteristics and caring responsibilities.
percentages do not add up to 100 as there were small amounts of missing data
All carers were related to the person they cared for, with the majority of respondents either the partner, parent or child of the dementia sufferer. The results suggest that there is a considerable burden of care placed on the unpaid carers with many respondents spending more than seven hours every day of the week caring. The majority of dementia sufferers were reported as having moderate or severe dementia. Approximately one-third of respondents (10/34) were classified as ‘at risk’ using the CSI score (CSI+ score of 7 or greater).
Costs, WTP and welfare gains/losses resulting from the telecare service.
Explanations given for the WTP responses.
Tests for differences in WTP values.
P < 0.05
Discussion
The study indicates that under hypothetical market conditions, the provision of telecare for carers of PWD produces a positive net benefit, i.e. the value revealed by carers exceeds the cost of providing the service. The study used a WTP approach which is a well-established economic method of measuring benefit in monetary terms for goods that are not provided under conventional market conditions where consumers actively pay for services. The study showed that the average annual WTP for telecare is within the range £240-300 depending on the style of question. We believe that this is the first such study to perform a comprehensive cost-benefit analysis where the benefits (as expressed by the WTP values) are offset against the costs to estimate the net benefit.
Although our study showed that carers value telecare, the distribution around the WTP values was large, so the results were not conclusive. The study had certain limitations. For example, the sample was small, although this is not uncharacteristic of mail-based WTP studies.18,19 This meant that methods such as regression analysis were not possible. There was an absence of local data to compare the characteristics of the sample with. Compared to national figures for unpaid carers, the study population was characteristic in terms of age (mean age over 60 years), and in terms of the quantities of care 20 provided (with many carers nationally known to provide over 100 hours of care per week). 20 There are a number of stark differences however, particularly in terms of the gender and income distribution of the sample. The majority of respondents in the study were male, whereas the majority of care in the UK is provided by women. 20 Furthermore, the sample included no participants in the higher household income bracket of >£20,000 per annum, indicating that wealth is affected by providing unpaid care which is contrary to what the literature states. 21 This however, is likely to be due to the Sandwell area from which participants were recruited, which in 2010 was ranked as the 12th (of 253) most income-deprived local authority in England. 22
Conventionally, economic evaluations use health-related outcomes as a measure of benefit, and these outcomes fail to capture aspects of the service which are seen as valuable by the carers, e.g. ability to go to work knowing that the PWD is safe and secure at home. Previous reviews have emphasized that very few evaluations of telecare seek to measure economic outcomes. In a recent review 23 of evaluations of telehealth for COPD, only 2 out of 23 studies considered the costs of the service.
Our results provide an early indication that the service is of value but this was based on a relatively small sample, and the work needs to be replicated in a larger, more generalizable sample. The cost data used to estimate the net benefit of telecare was based on the UK prices and not the full operational cost which would be borne by telecare providers. These wider operational costs are not available. Operational costs such as administrative and maintenance costs could be expected to have a negative effect on the net benefit.
The WTP method asks respondents to reveal values under hypothetical market conditions. The nature of the method therefore means that translating hypothetical values to real values is susceptible to methodological uncertainties. However in the case of telecare for carers of PWD there is additional information which puts this into a slightly different context. In the UK, depending on individual circumstances, some carers are entitled to a Carer’s Allowance which amounts to £59.75 a week (April 2014). This is provided to help carers who are spending at least 35 hours a week caring for someone. It is at the carer’s discretion how they spend that money, but one option could be to invest in telecare to help alleviate some of the stresses of caring. By asking the bounded WTP question in this study, albeit using an arbitrary monthly value of £500, we hoped to put the study more into a real-life context.
Quantitative analysis of the factors influencing WTP values yielded one significant result, that WTP varied with the severity of the cared-for person’s dementia. Further investigation indicated that WTP values were higher in respondents who reported their cared-for person as having moderate dementia, rather than mild or severe dementia. This suggests that telecare is most useful to carers of those with moderate dementia. We hypothesise that this relationship between dementia severity and WTP for telecare seems plausible, as it is possible to view telecare as being of limited use in mild cases (as occasions when it is of benefit to carers may occur less frequently), and conversely in severe cases of dementia, the relief provided by telecare alone may be insufficient to guarantee the cared-for person’s safety, and thus have little effect on the burden of care.
Despite the absence of conclusive quantitative evidence to suggest other factors affecting WTP values, the qualitative results obtained from the content analysis provide an insight into what else may be affecting WTP. The most commonly reported reason for a participant’s WTP values was affordability. The idea that ability to pay plays a significant role in determining WTP has been well documented in both theoretical discussions and practical applications of the WTP method24,25 and is thus likely to be a good predictor of both WTP values and also the likelihood of non-response to WTP questions.
Few previous studies have used WTP to value telecare. It has however been used to value telemedicine. Bradford et al. investigated patients with chronic heart failure and found that more than half of the sample were willing to pay $20 per month to have access to telemedicine instead of travelling to a physician appointment. 26 Qureshi et al. asked patients with a history of psoriasis or melanoma about telemedicine versus clinic visits and found that patients were willing to pay $25 for telemedicine. 27 Tsuji et al. interviewed users users of a home telemonitoring system in Japan and found them willing to pay yen 4519 per month. 28 Although, these published WTP values for telemedicine are of limited value for comparison with the results from the present study, they show that WTP is a method that has been successfully applied in the past as a means of placing a monetary value on telemedicine.
In conclusion, our study suggests that the provision of telecare for unpaid carers of PWD is beneficial to society. Furthermore, it appears that telecare is of greatest benefit to those caring for moderately severe dementia patients. However, additional research is required in order to test whether these assertions hold in other population groups, and to test whether other factors influence the value placed on telecare.
Footnotes
Acknowledgements
The work was funded by the National Institute for Health Research through the Collaborations for Leadership in Applied Health Research and Care for Birmingham and Black Country programme. We are grateful to Sarah Cottingham, Ross Bailey, Julie Grant, Martyn Filsak, Alasdair Morrison, Mick Dowd, Sam Hay, Jamie Crew, Julia Sear, Pam Price, Carl Griffin and Billingsley Kaambwa.