Article Text

Download PDFPDF

Is safety safe in the market’s hands?
  1. Cam Donaldson
  1. Professor Cam Donaldson, Institute of Health and Society, 21 Claremont Place, Newcastle University, Newcastle upon Tyne NE2 4AA, UK; cam.donaldson{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

There is no doubt that the proposal to establish market-based control mechanisms for patient safety is both interesting and intriguing. However, I am not optimistic that Coiera and Braithwaite’s (C&B’s) proposal (see page 99) will see the light of day1. My objections are based on three fundamental flaws that I see in the foundation for this proposal: first, as an economist, I question whether “safety” can be considered a public good in the same way that the environment might; second, I am not convinced that emissions trading has worked sufficiently well to justify basing a safety strategy on such an approach; and, finally, I propose that our healthcare systems have not been particularly effective in managing resources or priority setting, and I make an economist’s attempt at addressing opportunities that still exist in this domain.

I should add that I am not wholly comfortable with the premise of a lack of widespread uptake of quality and safety measures in healthcare. Intuition would lead one to agree. But what is optimal in provision of safety measures? Indeed, it has become almost an anathema to question the health gains from what might often be disproportionately large amounts of resources invested in measures put in place in response to an unsafe incident. I raise this doubt early because it reflects where I will head with my own proposed next steps. However, it also reflects a related issue, to which I will also return, encapsulated in Woolf’s categorisation of safety as a “subset” of broader medical errors which are in turn a subset of “lapses in quality.”2 How can a market in safety be separated from all of these wider concerns? Nevertheless, accepting the premise for now, the question is whether some form of trading mechanism is the solution.


Twice in the early stages of their article, C&B refer to the environment and safety as “public goods.” I am not convinced of this parallel. Thus, the common proposal of trading in rights to emit carbon or to be unsafe may be flawed from the outset. The term “public good” has a very specific meaning in economics, its key features being non-excludability (people cannot be excluded from consuming it, like the environment) and non-rivalry (one person’s consumption does not diminish the opportunity of another to consume, again like the environment). This, in part, is why we have hit problems with the environment; no one has an incentive to do anything about it; hence, emissions trading as an idea aimed at resolving this.

People can be excluded from consuming healthcare, and one person’s consumption rivals that of another. In most advanced economies, healthcare is publicly financed, but it is not a public good. The solutions to healthcare resource allocation challenges may be very different, therefore. First, it is publicly financed because of market failures, but failures of a different type to those that emanate from the public goods argument.3 Indeed, there are many who would argue that attempts to reintroduce (quasi) markets into healthcare around the world over the past 20 years have failed.34 This does not bode well for C&B’s proposal.


A general answer to the first part of this question could be that the environment continues to be a major challenge. Of course, this cannot all be blamed on a failure in emissions trading, as it is likely that it has not been taken up on a sufficiently widespread basis. Nevertheless, one would have to question the level of price that has been set. Is it perhaps too low, so allowing companies to continue polluting at relatively little penalty? This observation is supported by the actions of organisations such as CoolEarth who are campaigning to persuade private individuals to buy up and “retire” emissions vouchers in attempts to try to force the issue (ie, with fewer vouchers available to buy, there will be a double effect of a reduction in absolute rights to pollute along with forcing companies to work harder on this issue because the price of vouchers will go up as long as campaigners do not resell).

This unintended consequence of emissions trading would indicate that the market has not been functioning as originally intended, which is often the case with markets of course. Price setting in the safety market will be key, and I am not sure that C&B have this right. They suggest prices are based on the cost of averted events, but this is only part of the story. Such prices would surely also have to reflect the risk of such events occurring in the first place and the intrinsic value of less tangible benefits such as morbidity and mortality avoided. In the UK, for example, we already have a value of a prevented death (of £1.4 m) that is used in cost–benefit analyses of road safety policies. As yet, such approaches have never been employed in valuing “safety” in healthcare. Without reflecting the value of such “intangibles,” this market would, at best, be largely ineffectual. Prices would be too low, allowing the safety laggards to continue their bad practices.

On the other hand, if prices are too high, even good performers may struggle, and who knows what will happen to bad performers? Despite repeated macho posturing by governments around the world, they have never let a hospital close due to (internal) market forces. To avoid such volatilities, substantial investment would be required in information to make the market function appropriately. C&B are contradictory on this matter, saying at one point that all that is required is information on price and quantity, in classic first-year-economics-textbook style, but later admitting that information and regulation will be required to inform target setting, generating meaningful outcome measures, monitoring, reporting and avoiding of gaming. These would incur substantial costs.

Finally, the issue of exit and entry to the market would have to be carefully addressed. Entry of CoolEarth and such like into the carbon emissions market may be a good thing, but it is not clear in advance who might want to enter the safety market, why and what the consequences would be. Also, will price be allowed to fluctuate in response to changes in demand engendered by entry and exit, which would normally be the case in a well-functioning market, or will price be “set” by some regulator? The latter would contain all the challenges set out above (eg, what would the regulator take into account in setting price?) as well as the regulator having to provide timely responses to market fluctuations.


The basic problem I have is that our publicly funded health service organisations have not been good at managing scarcity of resources. Market-based reforms have done little to rectify this. Whatever, the latest reform context in which they are asked to operate, health service entities have to manage scarce resources, but have had little training in how to go about this challenging task. What we need are more explicit rules on how to set priorities in healthcare and the leadership to apply the results of priority-setting exercises. For example, a rigorous method of evaluating multiple objectives and the trade-offs across these objectives involves quantitative multiattribute utility theory. Attributes are descriptors of the factors that may influence priority decisions (for example harm caused, risk of harm, effectiveness of interventions), and utility is a measure of “value” or preference associated with one choice over another. Returning to Woolf,2 what is required is a process of multiattribute priority setting to place safety initiatives within the wider healthcare context. Multiattribute decision-making has a long history and strong theoretical basis.56 Prior to building such a process, in-depth qualitative studies with health organisations and their managers are required to understand better the barriers to more rational decision-making and also the criteria which form the attributes of importance when assessing the benefits of safety (and other competing healthcare) initiatives. Leadership will then be required to put such priorities into practice. Only then can we start to use our limited resources to better serve patients and the public who are needlessly suffering.



  • Competing interests: None.

  • Funding: Supported by the Health Foundation in Health Economics.

Linked Articles