Short-term Revision Risk of Patellofemoral Arthroplasty Is High: An Analysis from Eight Large Arthroplasty Registries.

BACKGROUND: Patellofemoral arthroplasty (PFA) is one option for the treatment of isolated patellofemoral osteoarthritis, but there are limited data regarding the procedure and results. Because isolated patellofemoral arthritis is relatively uncommon, available case series generally are small, and even within national registries, sample sizes are limited. Combining data from multiple registries may aid in assessing worldwide PFA usage and survivorship. QUESTIONS/PURPOSES: We combined and compared data from multiple large arthroplasty registries worldwide to ask: (1) What proportion of patients undergoing primary knee arthroplasty have PFA? (2) What are the patient and prosthesis characteristics associated with PFA in common practice, as reflected in registries? (3) What is the survivorship free from revision of PFA and what are the reasons for and types of revisions? METHODS: Data were provided by eight registries that are members of the International Society of Arthroplasty Registries (ISAR) who agreed to share aggregate data: Australia, New Zealand, Canada, Sweden, Finland, Norway, the Netherlands, and the United States. De-identified data were obtained for PFA performed from either the beginning of year 2000, or the earliest recorded implantation date after that in each individual registry when PFA data collection commenced, up to December 31, 2016. This included patient demographics, implant use, all-cause revision rate (determined by cumulative percent revision [CPR]), and reasons for and type of revision. RESULTS: During the data collection period, 6784 PFAs were performed in the eight countries. PFAs comprised less than 1% of primary knee replacements in all registries. Patient demographics were comparable in all countries. Patients were generally more likely to be women than men, and the mean age ranged from 50 years to 60 years. All registries showed a high rate of revision for PFA. The 5-year CPR for any reason ranged from 8.0% (95% CI 4.5 to 11.5) in Norway to 18.1% (95% CI 15.5 to 20.7) in the Netherlands. The most common reason for revision across all countries was disease progression (42%, 434 of 1034). Most PFAs (83%, 810 of 980) were revised to a TKA. CONCLUSIONS: The revision risk of PFA in all registries surveyed was more than three times higher than the reported revision risk of TKA at the same times. The survivorship of PFA is similar to that of the no-longer-used procedure of metal-on-metal conventional hip replacement. Although there may be potential functional benefits from PFA, these findings of consistent and alarmingly high rates of revision should create concern, particularly as this procedure is often used in younger patients. LEVEL OF EVIDENCE: Level III, therapeutic study.

An instrument for measuring the social willingness to pay for health state improvement.

This paper describes an instrument for measuring the social value of changes in health status, the Relative Social Willingness to Pay. It is a unique combination of measurement attributes designed to minimise cognitive complexity and provide an additional option for measuring 'social value'. Similar to the person trade-off (PTO), it adopts a social perspective and asks respondents to evaluate programmes on behalf of society. Unlike the PTO, trade-offs between the options use dollars, not numbers of patients. Respondents are not, however, asked for their personal willingness to pay. Rather, the opportunity cost of funds spent on one service is as an offsetting reduction in funds for a second service. The amount spent on each service therefore indicates relative, not absolute, value. However, the two services combine to produce one Quality adjusted life year which allows the calculation of a Quality adjusted life year-like unit of social value on a 0-1 scale. A three-stage survey was used to test the instrument's reliability, validity and sensitivity to the framing of the main question. Results indicate that the Relative Social Willingness to Pay produces values similar to but less than the PTO and time trade-off techniques.

Priority setting and patient adaptation to disability and illness: outcomes of a qualitative study.

The study examined the question of who should make decisions for a National Health Scheme about the allocation of health resources when the health states of beneficiaries could change because of adaptation. Eight semi-structured small group discussions were conducted. Following focus group theory, interviews commenced with general questions followed by transition questions and ended with a 'focus' or 'key' question. Participants were presented with several scenarios in which patients adapted to their health states. They were then asked their views about the appropriate role of the public, patients and health professionals in making social judgements of quality of life. After discussion and debate, all groups were asked the key question: 'In light of adaptation, who should evaluate quality of life for the purpose of setting priorities in the allocation of health care?' In all groups participants presented strong arguments for and against decision making by patients, the public and health professionals. However, most groups thought a representative body which included a range of perspectives should make the relevant judgements. This is at odds with the recommendations in most national pharmaceutical guidelines. The main conclusion of the paper is that health economists and other researchers should explore the possibility of adopting a deliberative, consensus-based approach to evaluating health-related quality of life when such judgements are to be used to inform priority setting in a public system.

The initial health-system response to the earthquake in Christchurch, New Zealand, in February, 2011.

At 1251 h on Feb 22, 2011, an earthquake struck Christchurch, New Zealand, causing widespread destruction. The only regional acute hospital was compromised but was able to continue to provide care, supported by other hospitals and primary care facilities in the city. 6659 people were injured and 182 died in the initial 24 h. The massive peak ground accelerations, the time of the day, and the collapse of major buildings contributed to injuries, but the proximity of the hospital to the central business district, which was the most affected, and the provision of good medical care based on careful preparation helped reduce mortality and the burden of injury. Lessons learned from the health response to this earthquake include the need for emergency departments to prepare for: patients arriving by unusual means without prehospital care, manual registration and tracking of patients, patient reluctance to come into hospital buildings, complete loss of electrical power, management of the many willing helpers, alternative communication methods, control of the media, and teamwork with clear leadership. Additionally, atypical providers of acute injury care need to be integrated into response plans.

The monetary value of a life year: evidence from a qualitative study of treatment costs.

A small number of studies have provided suggestive evidence that the general public rejects the idea of giving higher priority to low-cost patients, in the context of a limited budget, in order to maximise health benefits. The study reported here used semi-structured group discussions to investigate the normative bases of such views among the Australian public. Discussion groups help participants reflect critically upon their own reasoning processes and go some way towards revealing underlying values rather than unreflective preferences. As a part of the exercise, participants were asked to allocate a hospital budget. After discussion and deliberation only three out of 41 chose to allocate all of the money to the low-cost patients. Reasons were not based on conceptual confusion or lack of insight into the implications of the different strategies, but rather on views about fairness, including the importance of giving all groups a 'chance' of being treated and of not removing 'hope' from high-cost patients. The results suggest that as costs rise people are willing to pay more than the minimum cost of a quality-adjusted life year for equity reasons, indicating that caution must be exercised in estimating a single monetary value for a QALY.

Completeness and capture rate of publicly funded arthroplasty procedures in the New Zealand Joint Registry.

BACKGROUND: Registry-based studies have become more common due to the availability of a large study cohort. However, the validity of findings is dependent on the completeness of the registry. This study aimed to validate the capture rate of the New Zealand Joint Registry (NZJR) by matching procedures that have been recorded separately via clinical coding by the New Zealand Government's National Surgical Site Infection Improvement Programme (SSIIP). METHODS: The National Health Index, a unique identification code for all patients, was combined with the arthroplasty procedure performed (primary total knee arthroplasty (TKA), primary total hip arthroplasty (THA), revision TKA or revision THA) and operation side. Publicly funded procedures recorded in the NZJR were matched with procedures recorded by the SSIIP on a record-by-record basis. This identified the total number of arthroplasty procedures performed in New Zealand, which was used as the denominator value to calculate the procedure capture rate of the NZJR. RESULTS: Between 2013 and 2018, 24 556 primary TKA, 28 970 primary THA, 2107 revision TKA and 4263 revision THA procedures were recorded by both datasets. The NZJR recorded 95.5% of primary TKA procedures, 96.3% of primary THA procedures, 97.1% of revision TKA procedures and 95.2% of revision THA procedures. CONCLUSION: The NZJR recorded >95% of publicly funded arthroplasty procedures. In contrast, there were inaccuracies in clinical coding by hospitals, particularly with revision procedures, demonstrating the benefits of an arthroplasty registry. However, data recorded by an infection surveillance programme may supplement arthroplasty registry data to strengthen the quality of research.

Treatment costs and priority setting in health care: A qualitative study.

BACKGROUND: The aim of this study is to investigate whether the public believes high cost patients should be a lower priority for public health care than low cost patients, other things being equal, in order to maximise health gains from the health budget. Semi-structured group discussions were used to help participants reflect critically upon their own views and gain exposure to alternative views, and in this way elicit underlying values rather than unreflective preferences. Participants were given two main tasks: first, to select from among three general principles for setting health care priorities the one that comes closest to their own views; second, to allocate a limited hospital budget between two groups of imaginary patients. Forty-one people, varying in age, occupation, income and education level, participated in a total of six group discussions with each group comprising between six and eight people. RESULTS: After discussion and deliberation, 30 participants rejected the most cost-effective principle for setting priorities, citing reasons such as 'moral values' and 'a personal belief that we shouldn't discriminate'. Only three participants chose to allocate the entire hospital budget to the low cost patients. Reasons for allocating some money to inefficient (high cost) patients included 'fairness' and the desire to give all patients a 'chance'. CONCLUSION: Participants rejected a single-minded focus on efficiency - maximising health gains - when setting priorities in health care. There was a concern to avoid strategies that deny patients all hope of treatment, and a willingness to sacrifice health gains for a 'fair' public health system.

Social preferences for prioritising the treatment of disabled and chronically ill patients: beyond the order effect.

Previous evidence suggests that members of the public value life saving services differently when they are for patients with a pre-existing permanent disability and when they are for patients who become disabled at the onset of treatment - for example, as a result of treatment that is not entirely effective. However, the valuation of services in these two cases has also been found to differ with the order in which they are presented in a population survey. This casts doubt upon the validity of the results and leaves unresolved the nature of the public's true preferences. The study reported here had three main objectives: (i) to determine the considered, underlying preferences of a sample of the Australian public with respect to the treatment of the permanently disabled and chronically ill, (ii) to gain insight into the reasons for respondent's distributive preferences and (iii) to eliminate or significantly reduce the order effect. Eight semi-structured, small-group discussions were held with 66 members of the public in Victoria, Australia. Order effects were effectively eliminated. The study found substantial support among participants for the equal treatment of the permanently disabled and chronically ill regardless of when the problem commenced.

Increasing the options for reducing adverse events: Results from a modified Delphi technique.

BACKGROUND: The aim of this paper is to illustrate a simple method for increasing the range of possible options for reducing adverse events in Australian hospitals, which could have been, but was not, adopted in the wake of the landmark 1995 'Quality in Australian Health Care' study, and to report the suggestions and the estimated lapse time before they would impact upon mortality and morbidity. METHOD: The study used a modified Delphi technique that first elicited options for reducing adverse events from an invited panel selected on the basis of their knowledge of the area of adverse events and quality assurance. Initial suggestions were collated and returned to them for re-consideration and comment. RESULTS: Completed responses from both stages were obtained from 20 of those initially approached. Forty-one options for reducing AEs were identified with an average lapse time of 3.5 years. Hospital regulation had the least delay (2.4 years) and out of hospital information the greatest (6.4 years). CONCLUSION: Following identification of the magnitude of the problem of adverse events in the 'Quality in Australian Health Care' study a more rapid and broad ranging response was possible than occurred. Apparently viable options for reducing adverse events and associated mortality and morbidity remain unexploited.

Economic evaluation of services for a National Health scheme: the case for a fairness-based framework.

In this paper we argue that the usual framework for evaluating health services may need modification in the context of a National Health Scheme (NHS). Some costs and benefits may need to be ignored or discounted, others included at face value, and some transfer payments included in the decision algorithm. In contrast with the standard framework, we argue that economic evaluation in the context of an NHS should focus on 'social transfers' between taxpayers and beneficiaries, and that the nature and scope of these transfers is determined by the level of social generosity. Some of the implications of a modified framework are illustrated with a re-examination of (i) costs and transfer payments, (ii) unrelated future costs, (iii) moral hazard, and (iv) the rule that marginal costs should equal marginal benefits. We argue that an explicitly 'fairness-based' framework is needed for the evaluation of services in an NHS. In contrast, the usual welfare economic theoretic framework facilitates the sidelining of issues of fairness.

Social preferences for prioritizing the treatment of severely ill patients: The relevance of severity, expected benefit, past health and lifetime health.

The study examined the preferences of a sample of the Australian public and health professionals regarding the relative importance of four different criteria for prioritizing between patients: the severity of the condition, the size of the benefit from the intervention, past health losses and expected lifetime health. A discussion-group methodology was adopted to elicit social preferences. This allowed participants time to consider all of the alternatives fully, to seek clarification of the task, and to engage in open debate about the issues raised. Participants traded-off cost-effectiveness for priority to the more severely ill. They placed less importance on past health and the lifetime allocation of health in deciding priority for treatment, and more importance on improving the condition of those who will be left more severely ill or disabled in the absence of treatment. The results pose a challenge to studies reporting support for the "fair innings argument". They also support the Norwegian government's decision not to pursue a life-time health loss criterion as recommended by the Norheim Commission. The study question is important given current debate both in the health economics literature and at the policy level in several jurisdictions.

Age Weights for Health Services Derived from the Relative Social Willingness-to-Pay Instrument.

The effect of a patient's age on the social valuation of health services remains controversial, with empirical results varying in magnitude and implying a different age-value profile. This article employs a new methodology to re-examine these questions. Data were obtained from 2 independent Web-based surveys that administered the Relative Social Willingness to Pay instrument. In the first survey, the age of the patient receiving a life-saving service was varied. Patients were left with either poor mental or physical health. In the second survey, patient age was varied for a service that fully cured the patient's poor mental or physical health. In total, therefore, 4 sets of age weights were obtained: weights for life-extending services with poor physical or mental health outcomes and weights for quality-of-life improvement for patients in poor mental or physical health. Results were consistent. Increasing age was associated in each case with a monotonic decrease in the social valuation of the services. The decrease in value was quantitatively small until age 60 years. By age 80 years, the social value of services had declined by about 50%. The decline commenced at an earlier age in the context of physical health, although the magnitude of the decrement by age 80 years was unrelated to the type of service. With 1 exception, there was little difference in the valuation of services by the age of the survey respondent. Respondents aged >60 years placed a lower, not higher, value on quality-of-life improvement for elderly individuals than other respondents. There was no difference in the valuation of life-extending services.

Enhanced hip fracture management: use of statistical methods and dataset to evaluate a fractured neck of femur fast track pathway-pilot study.

The increasing elderly population and subsequent rise in total hip fracture(s) in this group means more effective management strategies are necessary to improve efficiency. We have changed our patient care strategy from the emergency department (ED), acute orthopaedic wards, operating theatre, post-operation and rehabilitation, and called it Fracture Neck of Femur Fast Track Pathway. All clinical data and actions were captured, integrated and displayed on a weekly basis using 'signalfromnoise' (SFN) software. The initial four months analysis of this project showed significant improvement in patient flow within the hospitals. The overall length of stay was reduced by four days. Time in ED was reduced by 30 minutes, and the wait for rehabilitation reduced by three days. Overall time in rehabilitation reduced by 3-7 days depending on facility. On average, fast track patients spent 95 less hours in hospital, resulting in 631 bed days saved in this period, with projected savings of NZD700,000. No adverse effects were seen in mortality, readmission and functional improvement status. Fractured neck of femur has increasing clinical demand in a busy tertiary hospital. Length of stay, co-morbidities and waiting time for theatres are seen as major barriers to treatment for these conditions. Wait for rehabilitation can significantly lengthen hospital stay; also poor communication between the individual hospital management facets of this condition has been an ongoing issue. Lack of instant and available electronic information on this patient group has also been seen as a major barrier to improvement. This paper demonstrates how integration of service components that are involved in fractured neck of femur can be achieved. It also shows how the use of electronic data capture and analysis can give a very quick and easily interpretable data trend that will enable change in practice. This paper indicates that cooperation between health professionals and practitioners can significantly improve the length of stay and the time in which patients can be returned home. Full interdisciplinary involvement was the key to this approach. The use of electronic data capture and analysis can be used in many other health pathways within the health system.

Empiricism, ethics and orthodox economic theory: what is the appropriate basis for decision-making in the health sector?

Economics is commonly defined in terms of the relationship between people's unlimited wants and society's scarce resources. The definition implies a central role for an understanding of what people want, i.e. their objectives. This, in turn, suggests an important role for both empirical research into people's objectives and debate about the acceptability of the objectives. In contrast with this expectation, economics has avoided these issues by the adoption of an orthodoxy that imposes objectives. However evidence suggests, at least in the health sector, that people do not have the simple objectives assumed by economic theory. Amartya Sen has advocated a shift from a focus on "utility" to a focus on "capabilities" and "functionings" as a way of overcoming the shortcomings of welfarism. However, the practicality of Sen's account is threatened by the range of possible "functionings", by the lack of guidance about how they should be weighted, and by suspicions that they do not capture the full range of objectives people appear to value. We argue that "empirical ethics", an emerging approach in the health sector, provides important lessons on overcoming these problems. Moreover, it is an ethically defensible methodology, and yields practical results that can assist policy makers in the allocation of resources.

Deriving health utilities from the MacNew Heart Disease Quality of Life Questionnaire.

INTRODUCTION: Quality of life is included in the economic evaluation of health services by measuring the preference for health states, i.e. health state utilities. However, most intervention studies include a disease-specific, not a utility, instrument. Consequently, there has been increasing use of statistical mapping algorithms which permit utilities to be estimated from a disease-specific instrument. The present paper provides such algorithms between the MacNew Heart Disease Quality of Life Questionnaire (MacNew) instrument and six multi-attribute utility (MAU) instruments, the Euroqol (EQ-5D), the Short Form 6D (SF-6D), the Health Utilities Index (HUI) 3, the Quality of Wellbeing (QWB), the 15D (15 Dimension) and the Assessment of Quality of Life (AQoL-8D). METHODS: Heart disease patients and members of the healthy public were recruited from six countries. Non-parametric rank tests were used to compare subgroup utilities and MacNew scores. Mapping algorithms were estimated using three separate statistical techniques. RESULTS: Mapping algorithms achieved a high degree of precision. Based on the mean absolute error and the intra class correlation the preferred mapping is MacNew into SF-6D or 15D. Using the R squared statistic the preferred mapping is MacNew into AQoL-8D. IMPLICATIONS FOR RESEARCH: The algorithms reported in this paper enable MacNew data to be mapped into utilities predicted from any of six instruments. This permits studies which have included the MacNew to be used in cost utility analyses which, in turn, allows the comparison of services with interventions across the health system.

Diabetes and quality of life: Comparing results from utility instruments and Diabetes-39.

OBJECTIVE: To compare the Diabetes-39 (D-39) with six multi-attribute utility (MAU) instruments (15D, AQoL-8D, EQ-5D, HUI3, QWB, and SF-6D), and to develop mapping algorithms which could be used to transform the D-39 scores into the MAU scores. RESEARCH DESIGN AND METHODS: Self-reported diabetes sufferers (N=924) and members of the healthy public (N=1760), aged 18 years and over, were recruited from 6 countries (Australia 18%, USA 18%, UK 17%, Canada 16%, Norway 16%, and Germany 15%). Apart from the QWB which was distributed normally, non-parametric rank tests were used to compare subgroup utilities and D-39 scores. Mapping algorithms were estimated using ordinary least squares (OLS) and generalised linear models (GLM). RESULTS: MAU instruments discriminated between diabetes patients and the healthy public; however, utilities varied between instruments. The 15D, SF-6D, AQoL-8D had the strongest correlations with the D-39. Except for the HUI3, there were significant differences by gender. Mapping algorithms based on the OLS estimator consistently gave better goodness-of-fit results. The mean absolute error (MAE) values ranged from 0.061 to 0.147, the root mean square error (RMSE) values 0.083 to 0.198, and the R-square statistics 0.428 and 0.610. Based on MAE and RMSE values the preferred mapping is D-39 into 15D. R-square statistics and the range of predicted utilities indicate the preferred mapping is D-39 into AQoL-8D. CONCLUSIONS: Utilities estimated from different MAU instruments differ significantly and the outcome of a study could depend upon the instrument used. The algorithms reported in this paper enable D-39 data to be mapped into utilities predicted from any of six instruments. This provides choice for those conducting cost-utility analyses.

The rule of rescue.

Jonsen coined the term "Rule of Rescue"(RR) to describe the imperative people feel to rescue identifiable individuals facing avoidable death. In this paper we attempt to draw a more detailed picture of the RR, identifying its conflict with cost-effectiveness analysis, the preference it entails for identifiable over statistical lives, the shock-horror response it elicits, the preference it entails for lifesaving over non-lifesaving measures, its extension to non-life-threatening conditions, and whether it is motivated by duty or sympathy. We also consider the measurement problems it raises, and argue that quantifying the RR would probably require a two-stage procedure. In the first stage the size of the individual utility gain from a health intervention would be assessed using a technique such as the Standard Gamble or the Time Trade-Off, and in the second the social benefits arising from the RR would be quantified employing the Person Trade-Off. We also consider the normative status of the RR. We argue that it can be defended from a utilitarian point of view, on the ground that rescues increase well-being by reinforcing people's belief that they live in a community that places great value upon life. However, utilitarianism has long been criticised for failing to take sufficient account of fairness, and the case is no different here: fairness requires that we do not discriminate between individuals on morally irrelevant grounds, whereas being "identifiable" does not seem to be a morally relevant ground for discrimination.

Successful antimicrobial therapy and implant retention for streptococcal infection of prosthetic joints.

BACKGROUND: Streptococci cause up to 20% of prosthetic joint infections but this has received little attention in the published literature. METHODS: We reviewed retrospectively our experience with treatment of streptococcal prosthetic joint infections. Patients were followed up for up to 15 years after discontinuation of antimicrobial therapy and up to 8.5 years while on continuous antimicrobial therapy. RESULTS: Eighteen cases were diagnosed between 1984 and 1995. These included one group A, seven group B, one group D, seven group G and one viridans-group streptococcal infection as well as one group B and D streptococcal co-infection. All were late-onset infections and most (11 of 18, 61%) were acute. Hip and knee joints were equally affected. Six of seven group G streptococcal infections were associated with skin or soft tissue infections. Sixteen patients were treated primarily with antimicrobial agents including 5 days to 6 weeks given intravenously and 2 weeks to 8.5 years given orally. At latest follow up, 10 patients had been off antimicrobial therapy for at least 18 months without relapse, one patient had been off antimicrobial therapy for 7 months without relapse and four infections were successfully controlled with long-term suppressive antimicrobial therapy. One infection was unable to be controlled with antimicrobial therapy. CONCLUSIONS: Our results, and those of others, show that prosthetic joint infections caused by streptococci have a relatively good outcome with primary antimicrobial therapy and, when necessary, drainage, lavage or debridement. Provided the prosthesis is stable and the patient can tolerate long-term antimicrobial therapy, this may be an effective alternative to excision arthroplasty.

Atypical subtrochanteric fractures, bisphosphonates, blinded radiological review.

BACKGROUND: Recent reports have suggested that a certain type of subtrochanteric and diaphyseal femoral fractures maybe associated with bisphosphonate (BP) therapy. We assessed the association between BP use in atypical and typical femoral fractures in a retrospective study and also looked at the rate of coding errors. METHODS: All cases between July 2003 and June 2008 with International Classification of Disease, 10th revision discharge codes for femoral fractures (S72.2 subtrochanteric and S72.3 fracture of shaft of femur) were reviewed. Cases were excluded if there was significant trauma, underlying bone disease or coding error. The remaining cases' films were assessed by an independent, blinded, single radiologist to assess for atypical features (thickened cortices, transverse fractures, medial cortical spike) with additional exclusion criteria of periprosthetic fractures and bone pathology. Odds ratios were calculated comparing BP use in atypical and typical fractures. RESULTS: Six atypical fractures were found in the study period. Compared with the 65 typical fractures, there was an association between BP use and atypical fractures (odds ratio 5.5) but it did not reach statistical significance (0.97-31). Atypical femoral fractures accounted for <0.1% of total fracture admissions during this period. There was a 20% rate of miscoding. CONCLUSION: This study shows a nonsignificant trend towards alendronate/BP use and atypical femoral fractures compared with typical femoral fractures. These fractures were rare <0.1% and the benefit and treatment of osteoporosis with BPs currently seems likely to outweigh the perceived risks. Individual case and radiology review is important as coding errors were frequent.

Social preferences for the inclusion of indirect benefits in the evaluation of publicly funded health services: results from an Australian survey.

The inclusion of both monetary and non-monetary indirect benefits in economic evaluations of public health programmes and services can have significant distributive effects between patient groups. As a result, some patients may be advantaged and others disadvantaged for reasons not directly related to health outcomes or (direct) treatment costs. In pluralistic democracies, there is a case for consulting the community on the fairness of policies that have such distributive implications. This paper reports the results of two pilot studies aimed at uncovering the preferences of the Australian public for the inclusion of indirect benefits in the evaluation of services for its national health scheme, Medicare. The initial survey found some support for taking account of non-monetary indirect benefits - for example, the social contribution made by parents of young children and carers of elderly relatives. By contrast, there was little support for giving high taxpayers priority access to general Medicare services, to life-saving organ transplants, or to very costly drugs, despite the indirect social benefits of doing so. However, such support increased significantly in the follow-up study when the outcomes were characterised as certain, identifiable and health related, and the opportunity costs of failing to take account of indirect benefits were made very clear. The follow-up survey provided evidence of public scepticism about the willingness or ability of government to use additional tax receipts for socially beneficial purposes, and/or a preference for programmes and services that focus on health rather than welfare more generally.