The Costs of Blindness
An Analysis of the Costs of
Visual Impairment and Blindness
in the United Kingdom
John Grainger and Robin Hutchinson
The Guide Dogs for the Blind Association
？ Ethical Strategies Ltd, July 2003
Acknowledgements ___________________________________________________ 3 Section 1: Background ________________________________________________ 7 1.1 Purpose of an Economic Evaluation _________________________________ 7 1.2 UK Treatment Environment _______________________________________ 7 1.3 Cost of Illness Studies: Methodological Issues _________________________ 9 Section 2: Literature Review and Analysis _______________________________ 16 2.1 Prevalence and Incidence of Major Causes of Blindness and Vision
Impairment_____________________________________________________ 16 2.2 Number of Blind / Partially Sighted People in the UK __________________ 17 2.3 Cost of Illness Studies ____________________________________________ 19 2.4 Cost-Effectiveness Studies_________________________________________ 24 Section 3: A Cost of Illness Analysis for Major Causes of Blindness __________ 31 3.1 Generating Lifetime Cost Estimates ________________________________ 31 3.2 Treatment Patterns ______________________________________________ 31 3.3 Resource Utilisation Data _________________________________________ 32 3.4 Cost Data ______________________________________________________ 36 3.5 Lifetime Cost Estimates __________________________________________ 46 3.6 Sensitivity Analysis ______________________________________________ 50 3.7 Budgetary Implications ___________________________________________ 52 Section 4: Conclusions _______________________________________________ 64 Key Messages: Lifetime Costs (Case Scenarios) _____________________________ 64 Key Messages: Annual Costs _____________________________________________ 64 Conclusions ___________________________________________________________ 65
Appendices A - L
? ESL July 2003 2
Ethical Strategies Ltd would like to thank Dr. Ian Murdoch, Mandy O‟Keefe and Paddy Martin at Moorfields Eye Trust, London, for their assistance in preparing this
Any correspondence pertaining to this report should be addressed to:
Clare Langley Hawthorne
Ethical Strategies Ltd
55 Eden Street
Kingston upon Thames
Surrey KT1 1BW
Tel: 020 8481 1050
Fax: 020 8547 0990
? ESL July 2003 3
The goals of this report are to present an analysis of current economic data relating to the major causes of blindness in the United Kingdom (UK), to highlight the issues involved in quantifying the cost impact of blindness, and to present some initial estimates of the lifetime and annual costs imposed by blindness and visual impairment from a societal perspective.
It is important to note that this report is not intended as a detailed economic evaluation of the costs imposed by blindness. Our initial estimates are limited by the available data and by the scope of the project. Although there are a number of data gaps, we have not undertaken any primary data collection, and are thus restricted in the form of analysis that may be undertaken. All estimates provided are based on a review of the current literature as well as statistics publicly available in the UK.
The report is divided into three sections. The first section provides an overview of the purpose of an economic analysis and the key issues that need to be addressed in a costing study. Some of the key methodological issues surrounding the development of a cost model and analysis of the impact of early detection or screening programmes, are outlined.
The second section provides a summary of the current literature on treatment patterns, costs and patient outcomes associated with the main causes of blindness and visual impairment in the UK. This section also provides an assessment of the evidence to date and its limitations for use in a costing analysis.
The third section outlines the framework of the costing model developed and presents some initial estimates of the costs associated with blindness and visual impairment in the UK. As the principal aim of this project was to generate lifetime cost estimates, preliminary lifetime cost estimates for a series of case scenarios are presented in this section. In addition, a prevalence-based costing was undertaken to supplement this analysis and to demonstrate the potential burden imposed by blindness and visual impairment on healthcare and non-healthcare budgets in the UK.
The UK government has employed increasingly sophisticated economic analyses of disease to determine the cost-effectiveness of alternative treatments or interventions. The technology appraisal process undertaken by the National Institute for Clinical Excellence (NICE) continues to have ramifications regarding the level of care and services provided by the National Health Service (NHS) for conditions that result in visual impairment. In December 2002 for example, NICE issued its final appraisal determination on the use of photodynamic therapy (PDT) for age related macular degeneration (AMD). As part of the appraisal process, a cost-effectiveness study was commissioned. The guidance issued by NICE recommended use of PDT only for the treatment of wet AMD in those patients with classic disease, no occult subfoveal choroidal neovascularization, and best corrected visual acuity of at least 6/60. ? ESL July 2003 4
On 14 May 2003, the results of an appeal against this appraisal were made public. While the majority of the appeal points were rejected, the appeal was upheld in part. NICE is now required to provide additional information including details of the cost-effectiveness estimates used. This case illustrates not only the impact of the NICE appraisal process on provision of treatment for conditions causing vision impairment, but also the impact of health economic data on the decision-making process. Accordingly, it is important to make the link between early intervention and reduction in cost-burden over time when presenting cost data to the UK government. This report outlines a framework for analysing incidence and prevalence-based estimates, although both approaches are limited by the data available. As the UK government is increasingly concerned about cost containment, it may not be sufficient to demonstrate that blindness imposes a significant cost burden – but
necessary to demonstrate the impact of appropriate early intervention in reducing costs and improving patient outcomes. The principal limitation in this regard is, again, data availability.
Some of the key results of the costing model developed in this report are summarised below:
Results: Lifetime Costs (Case Scenarios)
； Lifetime costs to the UK government for a person with diabetic retinopathy
can be up to ?237,591 per person. Almost 50% of these costs represent
productivity losses due to blindness or vision impairment. An additional 5
years of possible working life (productivity losses to age 65) increases this
estimate to ?314,512. Early detection or screening could potentially avoid
these costs. If 1,000 cases of diabetic retinopathy could be avoided the
potential savings to the government could be as high as ?237 million.
； Lifetime costs of a woman in residential care suffering from AMD are
estimated to be ?196,876. If appropriate treatment could reduce the need for
residential care these costs could be reduced to ?41,652.
； Lifetime costs for the elderly (>65 years) with cataracts are estimated to be
?19,120 while the lifetime costs for those with glaucoma are estimated to be
； The lifetime costs associated with congenital vision loss in adolescence is up
to ?257,184 (?973, 782 undiscounted). Sixty-one percent of these costs are
attributable to productivity losses as a result of their condition.
Results: Annual Costs
； Annual Costs for those registered as blind or partially sighted in England
alone range from ?1.4 to ?2.9 billion (2002 costs). This represents ?7,561 per
person. If we take the RNIB evidence of underreporting of blindness and
visual impairment the cost estimates increase to ?4.1 to ?8.8 billion. These
costs include social benefits and productivity losses and exclude condition ? ESL July 2003 5
specific treatment costs. They therefore understate the total costs imposed by
eye conditions causing blindness and visual impairment.
； Using prevalence estimates from the North London Eye Study and the same
resource assumptions applied in the previous analysis, annual cost estimates
associated with vision impairment in the elderly population (> 65 years) in the
UK were also generated. These annual costs estimates range from ?13 and
?35 billion with a reference (base case) cost estimate of ?25.1 billion. This
represents an annual cost of ?4,980 per person.
； Total costs of glaucoma (in the working age and elderly population) range
from ?16 to ?38 billion annually with average annual costs per patient ranging
from ?7,239 to ?17,246.
； If only 10% of the glaucoma population received earlier treatment that
arrested the development of visual impairment this kind of programme could
save the government between ?555 million and ?1 billion.
； Total costs of cataracts in the elderly population range from ?7 billion to ?15
billion or between ?2,521 and ?5,579 per patient.
； If an early screening programme could reduce the prevalence of cataract by
10% (to a prevalence of 20% in the elderly), the UK government could save
?3.1 billion annually.
The lifetime cost estimates generated in this report provide a mechanism for assessing the magnitude of the cost imposed by blindness and visual impairment on society over the longer term. The annual (prevalence) based estimates provide an assessment of the impact of blindness and visual impairment on UK government expenditures in a given year (2002). The major limitation of the incidence-based approach is that it does not easily translate into population-based budget impact estimates. However, the strength of this approach is that it reveals the extent to which the burden of care rests with government health and social services, as well as the cost burden imposed on patients and their families over a patient‟s lifetime.
The estimates provided and the sensitivity analysis undertaken reveal that any costing of the major conditions causing blindness and visual impairment are dependent on the assumptions made and the data used. Unfortunately we have insufficient data available to make accurate assumptions regarding probable treatment experience, but we can identify, using the estimates generated the key cost drivers for any future costing analysis. Further research is needed, however, to fully assess treatment patterns, resource utilisation and costs from both a patient and societal perspective to develop a more comprehensive framework for assessing the benefits of early intervention and treatment to prevent blindness and visual impairment.
? ESL July 2003 6
Section 1: Background
1.1 Purpose of an Economic Evaluation
An economic evaluation should be viewed as a comparative analysis of alternative 1courses of action in terms of both their costs and consequences. Although cost of
illness studies are used to quantify the magnitude of the burden of disease or disability, they are often of only limited use in aiding decision makers regarding the 2allocation of health care resources. These forms of analyses provide only partial
data on whether a particular intervention strategy will be an efficient use of resources or whether it will improve patient outcomes more efficiently than an alternative intervention or treatment option.
In order to persuade the UK government to fund early intervention programmes, it will be important to generate robust estimates of the costs associated with visual impairment and blindness, and to provide guidance on the potential costs and benefits of early intervention strategies, particularly with respect to lifetime costs. This report discusses a methodology to identify the costs of blindness to the government and society and provides a framework within which investment in early intervention programmes should be assessed.
1.2 UK Treatment Environment
When evaluating cost of illness issues, consideration must be given to the treatment and reimbursement environment within which health care services are provided. In the UK there have been a number of recent reforms that have emphasised the need to establish principles of evidence-based medicine, consistency of care and a more efficient allocation of health care resources.
The National Institute for Clinical Excellence (NICE) was established as a Special Health Authority in 1999 and is responsible for issuing guidance regarding best practices for the National Health Service (NHS). One of its roles is to undertake technology appraisals of new and existing health care interventions and products including pharmaceuticals, diagnostic techniques, medical devices and procedures. In July 2001, NICE issued a revised „Technical Guidance for Manufacturers and
1 See Drummond MF, O‟Brien B, Stoddart G, Torrance GW. Methods for the Economic Evaluation of ndHealth care Programmes. Oxford Medical Publications. 2 ed. Oxford University Press. Oxford. 1997. 2 Medical Research Council and Department of Health. Current and Future research on Diabetes. A Review for the Department of Health and Medical Research Council. October 2002. Pg 41.
? ESL July 2003 7
3 This guidance Sponsors on Making a Submission to a Technology Appraisal‟.
provides valuable information regarding “best practice” (as perceived by the
government) with respect to the development of an economic evaluation within a UK context. It is prudent, therefore, to discuss development of an economic analysis of the impact of blindness (and the potential benefits of early intervention programmes) in light of the evidentiary standards set out in the revised guidance, which are based on four key principles:
； The best available data should be presented
； Final outcome measures are preferred to intermediate (surrogate) outcomes ； Data from controlled prospective studies carried out in a routine care situation are
； Presentation of data and analysis should be transparent
The perspective to be adopted in any technology appraisal is that of the NHS and the personal social services (PSS) decision-maker. This means that a societal perspective should be adopted in the model and costs should include all costs to the NHS and PSS budgets. The guidance specifies that the analysis undertaken should take the form of either a cost-effectiveness or cost-utility analysis (see section 1.3 for a definition of these analytical techniques).
In addition, NICE has issued guidance to patient and carer groups with respect to the 4information these groups could supply as part of the technical appraisal process.
This guidance is useful to consider, especially as it highlights two key areas of concern for NICE: input on meaningful patient outcomes and the provision of information on the costs and benefits of alternative interventions.
In addition to its technology appraisal function, NICE also issues clinical guidelines to provide healthcare professionals with information regarding the appropriate treatment pathways for different conditions. NICE has issued clinical guidelines with respect to treatment of type 2 diabetes including diabetic retinopathy and its early 5management.
The UK National Screening Committee (NSC) assesses proposed new screening programmes against criteria including evidence that the opportunity cost of the screening programme should be economically balanced in relation to medical expenditures as a whole. In April 1999, the NSC commissioned a working group to develop proposals for a national screening programme for diabetic retinopathy. They estimated the initial roll-out cost of a screening programme would be ?67 million,
3 National Institute for Clinical Excellence. Technical Guidance for Manufacturers and Sponsors on Making a Submission to a Technology Appraisal. July 2001. 4 National Institute for Clinical Excellence. Guidance for Patient and Carer Groups. March 2001 5 The Royal College of General Practitioners Effective Clinical Practice Unit, University of Sheffield. Clinical Guidelines for Type 2 Diabetes. Diabetic Retinopathy: Early Management and Screening.
? ESL July 2003 8
and also identified that screening could result in substantial savings as people screened would be less likely to become blind.
Screening targets have also been established as part of the Diabetes National Service Framework launched in January 2003. This sets a target for all people with diabetes to have access to diabetic retinopathy screening services by 2007. The Department of Health announced in April 2003, that ?27 million would be spent on providing digital 6 retinopathy cameras to enhance local screening facilities.
While diabetic retinopathy and age related macular degeneration (AMD) have gained widespread attention within the UK, there remain a number of areas of unmet needs both in relation to these conditions and other causes of visual impairment and blindness.
1.3 Cost of Illness Studies: Methodological Issues
When undertaking a cost of illness study there area a number of key issues to consider. These include:
； The costing approach adopted
； The scope of the study
； Costs measured and costs excluded from measurement, and
； The data utilised, it appropriateness and limitations.
These key issues in turn raise a number of areas for consideration such as approaches to costing, discounting and time horizon for analysis.
Perspective – Identification of Relevant Costs and Consequences
The resources consumed and costs associated with a disease or health care 7intervention can be divided into three main components:
； Healthcare sector costs
； Costs to patients and their families
； Other sector costs
When assessing the overall costs to society associated with blindness and visual impairment, all of these costs need to be considered. Costs should be viewed as resources utilised or forgone as a consequence of disease. Costs arising from the use of resources within the health sector, resources used by patients and their families as well as resource use in other sectors should also be included.
Healthcare sector costs include hospitalisation, surgery, diagnostic procedures, medications, physician visits, nursing time etc. In addition to these costs, however,
6 www.diabetes.org.uk/news/apr03/eye.htm 7 Drummond. ibid.
? ESL July 2003 9
patients and their families incur out-of-pocket expenses such as prescription co-payments, costs to adapt their homes and purchase of visual aids. In addition there is the time spent by families and friends caring for a blind or partially sighted person.
Other sector costs include social services or other benefits provided (such as a disability allowance or tax credit) as well as nursing home or other residential care that may be required. People who are blind or partially sighted are also less likely to be in employment so there are also productivity losses associated with their condition which represent a loss to society as a whole and which should be included in any societal cost evaluation. Even if all of these costs are identified and capable of being quantified, there remains the intangible consequences such as pain and suffering which ideally would also be considered (although placing a monetary value on these can be difficult).
Given the societal perspective of our study, we will attempt to identify and include as wide a range of costs as possible, to fully assess the extent of the burden imposed by blindness and visual impairment on society.
Selecting the Appropriate Form of Analysis
As already noted, a cost of illness analysis is only a partial economic evaluation if the impact of alternatives is not considered. When seeking to compare alternative interventions there are four forms of analysis that have traditionally been used in 8) below: health economics. These are listed (adapted from Drummond et al. 1997
； Cost-minimisation analysis: Compares the costs of competing interventions or
products where the consequences of them are assumed to be equivalent.
； Cost-effectiveness analysis: Compares the costs of competing interventions or
products against the consequences measured in natural outcome units (such as
years of life saved).
； Cost-benefit analysis: Compares costs of competing interventions or products
with benefits measured in monetary values (such as changes in productivity
using wages as a measure).
； Cost-utility analysis: Compares costs of competing interventions or products
with benefits measured in units such as quality-adjusted life years (QALYs)
gained. These benefits take into account patient preferences (sometimes
revealed using patient‟s willingness to pay for a benefit) and the effectiveness
of the intervention or product.
The Technical Guidance issued by NICE indicates a preference for cost-effectiveness or cost-utility analyses. Clearly the quality of life impacts associated with vision loss are significant and many evaluations of screening and treatment options have taken the form of a cost-utility analysis. The unit of outcome of these studies are typically costs per QALY gained. As discussed in Section 2, one of the critical issues when
8 Drummond. ibid.
? ESL July 2003 10