Opt-out organ donation: on evidence and public policy

Introduction

Deceased organ donation in the United Kingdom (UK) is governed by the Human Tissue Act 2004 and the Human Tissue Act (Scotland) 2006 and operates under an opt-in system of donation. Despite its rejection by the Organ Donation Taskforce in 2008, ¹ there have been continued calls to move to an opt-out system (sometimes, perhaps incorrectly, termed presumed consent). For example, the British Medical Association has long supported such a move ² and there have been sustained attempts from some academic commentators who argue for legislative change.^3–6 Recently, the National Assembly for Wales voted in favour of a draft Human Transplantation (Wales) Bill, which will enable the creation of an opt-out system. ⁷ This follows a referendum held in early 2011 which resulted in the Assembly gaining further law-making powers, including in relation to health and health services. ⁸ The new law is slated to be on the statute books by 2015. In addition, earlier this year, the Northern Irish Executive announced plans to consult on organ donation including a possible move to an opt-out system. ⁹

It has been widely assumed that the evidence available makes credible the suggestion that opt-out systems yield significantly higher rates of organ donation, something that a recent review commissioned by the Welsh Assembly supposedly reaffirms. In this essay piece, we take issue with the way in which the evidence has been used to make inferences and reach conclusions not necessarily (strongly) supported by the evidence. The available evidence is weaker than sometimes assumed, yet it is being used to support ideological, ethical and political commitments. In the absence of strong evidence, time and effort spent on legislative change misses the opportunity to focus on non-legislative action, which could have greater impact.

Measuring the potential impact of legislative change

There is a difference between the legal permissibility of opt-out and its implementation in practice. If the mere implementation of such legislation was enough to improve deceased organ donation rates, then those countries with legal opt-out ought to all display high rates of donation. However, while opt-out countries such as Spain, Austria and Belgium have high donation rates, countries such as Luxemburg and Bulgaria, which are also legally opt-out, have some of the lowest rates of deceased donation. ¹⁰ In order to see whether the law in itself makes a difference, we need an analysis which isolates this as a factor.

There are a number of studies available which attempt to look at the effect of the default legislation on organ donation rates and, to date, two reviews examining these have been commissioned in the UK: a systematic review in 2009 ¹¹ by the Organ Donation Taskforce and the other, a single author (Palmer) review, ¹² commissioned in 2012 by the Welsh Assembly. Although the reviews themselves report similar findings regarding the included studies, the resulting policy conclusions in terms of the impetus for legislative change are different. In the first of these Rithalia et al. ¹¹ concluded that, while the introduction of opt-out legislation in different countries was associated with increased donation rates, ‘it cannot be inferred from this that the introduction of presumed consent legislation per se will lead to an increase in organ donation rates’. Subsequently, the Organ Donation Taskforce rejected a move to an opt-out system. By contrast, after updating the 2009 review, and including the results of public surveys and psychology literature, Palmer ¹² concludes that these ‘three strands of evidence provide a convincing basis for the introduction of an opt-out system in Wales’. A key question that arises from this, therefore, is whether the findings from this latter review are enough to support such a strong conclusion.

Best available evidence

Both the surveys and psychology studies looked at by Palmer essentially measure the subjects’ intent given a hypothetical scenario. Thus, they may inform about individual attitudes and apparent willingness to donate organs. However, these represent a weak form of evidence since they are not studies of actual practice. Indeed, intention or willingness does not always translate into practice. Compare, for example, the proportion of those who say they would be willing to be an organ donor (62%) ¹³ with the proportion of those on the UK register (31%). ¹⁴ To study the actual effects of legislation on donation rates, investigators are limited to two options: (1) before-after studies, which measure the donation rates before and after a change in legislation within a country, or (2) between-country comparisons, which compare donation rates in those countries that have an opt-out system with those that have not.

Despite the fact that it is difficult to control for potential confounding factors, before-after studies probably represent the best available design as heterogeneity between countries is not a consideration. Although five such studies identified by Rithalia et al. ¹¹ showed increased donation rates after a change to opt-out legislation, concurrent non-legislative changes were not accounted for in any of them. No new before-after studies were identified by Palmer. ¹² We are, therefore, currently limited to what we can learn from between-country comparisons: these attempt to use data from a sample of different countries and model the effect of legislation on donation rates while attempting to control for potential confounding factors.

In such models, opt-out legislation consistently appears as a significant explanatory variable despite other covariates being influential.^15,16 For example, Bilgel ¹⁶ reported that opt-out countries exhibit 18% higher donation rates on average compared with opt-in countries. Unlike previous modelling studies, this study included a number of institutional covariates such as family consent override, a civil liberties index and the type of donation registry being used. In fact, Bilgel’s ¹⁶ study was the one additional high-quality study (which modelled actual donation rates) identified by Palmer in her updated review. However, while this looks promising, as we note in the next section, there are still grounds to be sceptical about the role of opt-out legislation, particularly when model predictions are compared with observation.

Modelling infrastructure and organisational change

For any organ donation system to be effective, it requires an effective procurement system which is underpinned by a well-organised infrastructure. This is necessary irrespective of the legislation. As a result, focusing on legislative change is to perhaps miss the target if greater gains could arise through organisational and structural change. For example, the Spanish system of organ donation has long been considered to be the ‘gold standard’ of deceased organ donation. In 1979, Spain introduced their current ‘opt-out’ legislation ¹⁷ and between 1979 and 1986 the rate of kidneys transplanted (a surrogate marker for organ donation rates) did increase across the country. ¹⁸ Although it may be tempting to attribute this increase to legislative change, two things should be noted. First, there are no reliable data on organ donation rates in the immediate years pre and post this change. ¹⁹ Second, prior to 1979, very few transplant operations were being conducted outside of Barcelona and Madrid. ¹⁹ After 1979, there was an expansion in transplant programmes across Spain, with different Spanish cities adopting programmes for the first time. ¹⁹ Thus, it is difficult to measure the real effect of legislative change when large portions of the country had no existing organ procurement programmes prior to legislation.

What is known is that transplantation rates in Spain seem to have plateaued between the mid- to late-1980s ¹⁸ and it was not until the 1990s that significant improvements in rates were seen. These changes seemed to coincide with Spain abandoning its separate provincial organ donation programmes and nationalising control of its organ donation programme. It amalgamated the programmes into a single, national approach to the management of organ donation and transplantation through the formation of the Spanish National Transplant Organisation (Organización Nacional de Trasplantes – ONT). ¹⁷ From this point onwards, organ donation rates increased climbing from 14.3 per million population (pmp) in 1989 to 33.6 pmp in 1999. ²⁰ Since then, the Spanish average has held steady at around 34 deceased donors pmp. ¹⁰ There are also several organisational elements within the Spanish system which have been attributed by some to its continued success; these include increased availability of formally trained transplant co-ordinators at the point of care, increased numbers of intensive care beds, increased training and positive communication links between the national transplant authority, press and public.^17,20 The positive effect of non-legislative measures seems to be supported by evidence from countries such as Italy, Australia, Argentina and Uruguay, all of which have tried to implement various aspects of the Spanish Model. ²⁰ Italy, for example, adopted opt-out legislation in April 1999 and this was applied throughout Italy. However, the region which experienced the greatest increase in donation rates in one year, Tuscany (they doubled from 13.1 pmp in 1998 to 27.1 pmp in 1999) achieved this before the change in law had been fully implemented. ²¹ As a result, commentators ascribe the increase to the significant organisational changes which were introduced in the three years prior to law change.^21,22

It is worth noting that in the context of modelling data, none of the models to date has adequately incorporated these non-legislative factors and this may provide some explanation for the disparity between their predictions and what is observed. For example, if Bilgel’s ¹⁶ model accurately depicted the behaviour of countries following legislative change, Spain should have experienced a subsequent increase in donation rates of around 8% when it changed to opt-out legislation. As already noted, the expansion of transplant programmes across Spain means any such effect cannot be determined. However, it is clear that the model does not predict the subsequent doubling of donation rates between 1989 and 1999 after the establishment of the ONT. This is because the model does not accommodate time-dependent variables, such as incremental changes in organisation and infrastructure.

When looking at cause and effect in relation to organ donation rates, determining the relative importance of the different factors which may potentially affect donation rates involves complexity which is yet to be surmounted by investigators. To date, factors such as health expenditures per capita, gross domestic product (GDP) per capita, religion, type of legal system, existence of registries, mortality rates from motor vehicle accidents or cerebrovascular accidents and role of the family have all been added as covariates to such models.^11,12,15,16 A number of these may be coded as binary variables, but important factors such as training, availability of transplant co-ordinators and communication links between the transplant authorities, public and media (all proposed explanations for the Spanish success) are not so easily coded. Thus, any new model is unlikely to adequately reflect all the subtleties and influences associated with infrastructure. Even if these factors could be coded, there is still the problem of obtaining high-quality data on such factors in sufficient quantity. Eliciting country by country data on the sort of factors listed is likely to prove extremely difficult.

Concluding remarks

A review of the research evidence indicates that opt-out legislation is associated with higher organ donation rates^11,12,15,16. However, the question of whether this association is real has not been answered by existing research. The challenge investigators face when studying the drivers of organ donation is that they are restricted to using observational studies and these represent an uncontrolled research environment. Thus, when associations are observed, it is imperative that all the factors which may potentially confound the association are eliminated before inferences are drawn. Such a process is constrained by the quality of data and the difficulty in incorporating covariates in models which adequately characterise changes in infrastructure. Although the modelling of data has become increasingly sophisticated, a number of the factors which have been attributed to the success of the Spanish system, for instance, have yet to be satisfactorily incorporated as covariates.

What is clear is that not all countries that have switched to an opt-out system have benefited and that a number of countries without opt-out legislation have seen increasing donation rates after implementing changes in infrastructure. For instance, in the years prior to the implementation of the Taskforce’s recommendations, the average donation rate in the UK fluctuated around 13 deceased donors pmp (see Figure 1). ²³ This number has steadily risen in recent years to 19.1 pmp in 2012/2013, the highest it has ever been. ¹⁴ Another example of the benefits of organisational change was seen in Tuscany where in one year alone there was an increase in donation rates of over 14 pmp before a change in legislation could be fully implemented. ²¹ Furthermore, the greatest gains in Spain were seen after what was essentially a ‘root and branch’ overhaul of the organ procurement system, 10 years after it had changed its legislation. ²⁰ OPEN IN VIEWER

While we do not yet know exactly which non-legislative measures will produce the greatest gains, this analysis suggests that this is where the largest increases in organ donation rates are likely to come from. It may yet be demonstrated that opt-out legislation has a real beneficial effect, but until then it seems more reasonable to direct efforts and resources towards identifying the key non-legislative factors. For the time being, law and policy-makers in the rest of the UK should be more circumspect than the Welsh Assembly in their treatment of the available evidence regarding the influence of the law.