Tag Archives: technology

Capturing race

Is HTA like GO? (Photo credit: Wikipedia)

Increasingly widespread amongst the world’s healthcare systems is the assessment of medicines and devices using various types of cost-benefit or cost-utility analysis; this is called health technology assessment or HTA. HTA seeks to determine, using evidence of one sort or another, whether something is broadly speaking affordable, taking account of the cost of the medicine/device taken against the benefit to a particular constellation of diagnostic attributes in patients. This is usually quantified in a measure called a QALY: a quality-adjusted life year, which is a way to assess the value for money of a particular health technology. In short, it is a way of valuing lives.

HTA is a utilitarian approach to assessment. To some extent, this is not surprising as HTA is in the main a method developed by health economists, who, like economists in general, hypothesise that we make daily decisions based on the utilty of this or that, in terms of trade-offs (Pareto optimisation, for instance) and rational decision making (that people seek to maximise value, or utility in what they do). This approach is increasingly in dispute in light of the findings from neurosciences and behaviour economics: by posting that people do not always make decisions that are in their own best interests, a key assumption of traditional economics, that of the rational actor, always calculating trade-offs and maximising benefits, and so on, is questioned.

The problem with utilitarianism, though, is it doesn’t pay attention to the freedom of the individual; it positions the justification of its results on the net benefit to society, regardless of the impact on rights of individuals. Obviously, health economists don’t watch Star Trek or they would know that the needs of the one outweigh the needs of the many. But then, that, too, is a moral position.

Indeed, it is perhaps the sense that utilitarian conclusions don’t seem to correlate with many people’s moral sentiments that may explain why decisions of HTA agencies, for instance NICE in the UK (England) lead to moral outrage and a sense of, if not injustice, at least unfairness. While the results of an HTA process may lead to a quantitatively defensible conclusion, people sense that this conclusion is not morally defensible.

How are we to judge? Few would use utilitarian arguments in this way in other spheres: would we calculate who needs welfare in terms of the net benefit to society in terms of quality of life years, though perhaps we do allocate welfare on moral assumptions that some people deserve welfare while others don’t.

Do we allocate support to communities ravaged by floods based on their overall contribution, or utility, to society.  If you could donate £10 million to a university, would you pick Oxford University or Thames Valley University; which one is more worthy? But would you want to treat people this way?

HTA doesn’t even let us value lives in quite this way, since it neatly avoids deciding about the worth of any particular type of person, who just happens through misfortune to find themselves needing some medicine that fails the HTA tests. HTA keeps us from confronting the fact that HTA is a way of drawing a conclusion, without actually having to decide any allocations for any one person in particular. Bentham would approve.

There is, though, a technical problem with HTA and it has to do with whether at one level of assessment outcome, a utilitarian models can be used when the decision to be made does not have life threatening consequences for some people.

If the QALY threshold is, say £35,000, as it apparently is in the case of NICE, are the decisions below that threshold, which tend toward ‘yes’ or ‘approval’ morally different from decisions above that threshold?  I suggest that different moral criteria come into play above the threshold and this is where I think out moral outrage should be directed and where HTA fails.  Regretfully, HTA models see the results as broadly continuous, that is, decisions above and below this threshold are seen as essentially of the same type.  But I have argued elsewhere that above the threshold, HTA models fail but for reasons other their analytical soundness, because above this threshold, the conclusions may lead to a lessened quality of life, in other words, they actually crystallise the health outcome rather than avoid it.

Therefore, in valuing lives, those above the threshold experience greater injustice than those below; they are treated differently, unfairly, unjustly, perhaps less worthy, but certainly differently.  Indeed, above the threshold, we feel we are more in the realm of our moral sentiments about the value of human life, and less our moral sentiments about the allocation of scarce resources.

If this were not so, then we would be living in a society that believes that the determinant of all important moral and political decisions is affordability, and if that were so, they we could not even afford the costs of inefficiency brought on by democracy, the inconvenience of not being able to exploit people, the costs of equal rights.

Perhaps, though, on our financially contaminated world, all we can think about today is money and that is further contaminating our perception of what sort of society we are actually trying to foster.  Certainly, protests on Wall Street and elsewhere point to the view that there seems to be some unjust allocation of the benefits of government bail-outs that just doesn’t benefit those ‘at the bottom’.

John Rawls wrote that the we should distribute opportunity in a society in such a way as to ensure that the least well off benefit the most. In the context of HTA, medicines and technologies that benefit only a few, but at great cost, represent a cost worth having as the least well off, namely those who would need it most ( have the condition it treats, and in some societies can afford it least), would benefit, even if a little, as that is the price we pay for justice.

This, I suggest, is the root of our moral outrage at HTA, that is unjustly fails to serve those who need it most.

I am left with wondering about the underlying morality of HTA as a government scheme. Governments, as we know, are the last resort, when things are tough and one would hope, ensure that the least well-off in society are not penalised simply in virtue of being least well-off.  In healthcare, someone has to be the carer of last resort; using HTA as a way of avoiding this responsibility is not morally defensible.

Enhanced by Zemanta
Fibre optic strands

a bundle of bright ideas

Twenty greatest engineering achievements of the 20th century, from the National Academy of Engineering.

1. Electrification

2. Automobile

3. Airplane

4. Water supply and distribution

5. Electronics

6. Radio and television

7. Agricultural mechanization

8. Computers

9. Telephone

10. Air conditioning and refrigeration

11. Highways

12. Spacecraft

13. Internet

14. Imaging

15. Household appliances

16. Health technologies

17. Petroleum and petrochemical technologies

18. Laser and fibre optics

19. Nuclear technologies

20. High-performance materials

Enhanced by Zemanta
Data Source

GDP Expenditure

With the new coalition government in the UK, we are seeing early signs of a serious assault on public spending on the state run NHS. Similar challenges await other European countries with bloated public debt. Part of the debt run up by Greece, for instance, arose from efforts to off-shore hospital debt.

In the end the question remains, as it always has, how much money should a country spend on healthcare. The answer, as has always been the case, is as much as you can afford. Research shows that levels of spending (in terms of percentage of GDP, for example) do not correlate well with health status, outcomes and other key indicators of the performance of a health system. Indeed, it can be said with some degree of confidence that GDP spending is NOT an indicator of health system performance.

What does appear to be a factor though is HOW that money is spent and HOW the system is organised to deliver health services.  Recent OECD work has clarified various characteristics of health systems. What is striking are a couple of already familiar features:

  • Not all countries pay 100% of the health bill from the public purse; many, such as France, use co-payments. Countries with socially unacceptable waiting lists have tended to be those with the highest levels of pure public expenditure (such as the UK, Norway and Canada). What this suggests is that there may be important features in how health systems organise themselves to deliver care that is adversely affected when the system is funded from general taxation. Efforts to introduce purchaser/provider separation, for instance, is an effort to create distance between the two quite different objectives, which in tax funded systems have been merged and caused considerable policy confusion, as well as operational difficulties. (I can mention the situation in the Canadian province of Alberta, where the response to funding constraints has been essentially to ‘nationalise’ the system, thus removing key drivers for reform. I can also refer to the Nuffield, UK, study that showed poorer health outcomes in the centralised health system in Scotland compared to now quite devolved purchaser/provider based system in England; and this despite having higher per capita expenditure in Scotland.)
  • Most countries have mixed economies of provision and relatively easier ways for new types of providers to emerge. Lower performing health systems seem to discourage new providers of care to enter the health market; this is an element of overall system design, perhaps regulatory over-reach and dated statist thinking.  But perhaps we are becoming smart enough to know how to design more responsive health systems, which in the end are almost chaotic given the nature of human beings and illness (random?) and so need to be understood as complex adaptive systems rather than tightly managed and controlled (think of the tightly coupled banking system which lacked the ability to realign itself quickly and effectively in response to a financial shock; Homer-Dixon’s remarkably prescient work here is worth looking up).  Managed designs usually end in tears, as they fail to deliver the responsiveness and flexibility that is critical for healthcare to respond to changing demand and fluctuations caused by shocks to the system.

There is no right number of doctors or nurses or hospitals or beds. What there is, though, is the right number of these for the design and structures necessary to deliver effective care.  And these can be designed and developed to use human talent differently, and more effectively.

In the UK, we will hear a lot about ‘front line services’ and protecting them from cuts. I have no problem with protecting front line services, but that does not mean that they will not be delivered in different and novel ways, that may be a better use of the expertise available.  The health professions will undoubtedly circle the wagons and predict dire consequences to the public, so called shroud waving. But what is better is a recognition that healthcare systems are highly inefficient; they are weak adopters of revolutionary change, and they are protective of established working practices — part of the reason for this protectiveness arises from the health professions having become co-dependents to the addiction to public money on the one hand and protected ways of working on the other. In a nutshell, they have become resistant to innovation and reform, and in some respects lost control of the their profession and the profession has ceased to evolve to meet the care needs of people — an emergent adaptive response characteristic of complex systems.

Hospitals are artefacts of industrial era organisational design principles — they embody craft mentalities in the organisation of care, and build on public support to protect their infrastructure (from closure, for example), rather than the public demanding better services, which may not require a hospital in the first place. The difficulty people have in unbundling a hospital (it can be done and I can share the algorithm with you in another post if you like) simply reinforces the protected nature of healthcare work. In part, the emergence of e-health (more precisely, the use of digital information and communication technologies, artificial intelligence/neural networks, predictive algorithms, smart devices, etc) offers a serious challenge to established patterns of working, as these various components have the collective effect of redistributing knowledge, embedding knowledge and skill in devices, and altering the use of bricks and mortar infrastructure — a high-tech/low touch outcome is not the necessary outcome if we are clear on our outcomes.

It is also not just a matter of a cost-effectiveness study of whether an e-consultation is better than a face-to-face consultation.  The evidence for this is actually quite easy — when the telephone was invented, businesses might have one, on a stand, which people would queue up to use. Now, a modern business would hardly do a business case to put a telephone on everyone’s desk — indeed, it hardly needs a business case to ensure everyone has a smart phone — yet in healthcare, smart phones are still rare, yet have the potential to radically alter information flows and hence work flows — 25% of US doctors now have one and ePocrates is one of the most downloaded clinical apps from Apple store, so it is coming. You don’t do a business case when the underlying business logic itself is what will fundamentally change and that is really what e-health is all about.

They say, in capitalism, that it works partly through a process of creative destruction. Otherwise, we’d still be riding around in horse-drawn buggies, and you wouldn’t be reading this note on a computer linked to the internet. There is, however, a general reluctance to apply that process to publicly funded institutions, and by extension to publicly funded ways of working.  The words government and entrepreneur are an oxymoron for many people. But that does not have to mean that public funding cannot be used to incentivise new ways of working and new forms of healthcare delivery. The challenges, in the end, lie in our heart and willingness to change, to create and innovate.

And so to austerity. There is little to fear, except our ability to resist change, protect legacy ways of working, and failing to grasp the real prize, that of doing things better and more effectively.  We will, no doubt, hear the opposite.

Reblog this post [with Zemanta]
Arizona, poisonous snake warning sign.

Beware digital errors as they can bite

We all know accidents (unusual occurances in healthcare) can happen. Where systems are involved, errors can arise from how a system works, the way the various bits mesh, the knowledge and training of everyone involved working together.  It is no real surprise that some errors arise from the technologies that we use. In particular, health information technology systems can cause new types of errors and mistakes, beyond just not working properly.

In the US, the Health IT Policy Committee has proposed establishing a database to track potential safety risks related to IT systems.  These risks include:

  • hardware and software failure and bugs
  • workflow interactions between staff and users
  • interoperability problems
  • implementation and training deficits.

Since healthcare work is complex, the workflow risks are particularly complex and can arise from, for instance, inaccurately understanding how a manual system achieves its results, and thereby designing a software-based system that fails to do just that. There is a funny little thing that happens when a patient sees a doctor; the doctor often will use writing a prescription to terminate the patient encounter — tearing the piece of paper off the tab, a swirl of signature and handing the slip to the patient leads to the patient leaving, a neat way to end the consultation.

In an automated system (electronic prescribing, for instance), the consultation is not terminated in this behavioural manner, but involves essentially hitting the return key on the keyboard to enter the required prescription data in the system, and perhaps handing (or not) the patient a copy — but the Rx is off on electronic wings to the pharmacy for dispensing. There is an error that can occur if the doctor does not hit the return key between patients — the Rx list builds up, from patient to patient, until the return key gets hit (unless some sort of failsafe has been built in); this error actually happened and it was an alert pharmacist commenting to the patient that the doctor had added a lot of new drugs that the alarm was raised. Perhaps the patient should have been more distrustful, too.

We must be mindful of risk and error in any kind of technology, but particularly in systems where it is very hard to look inside the black box of software code.

I wrote a paper on digital risk some years ago, which can be found here: Patient Safety and Digital Risk. I have also raised the issue of risk in the even blacker box of predictive algorithms used to data mine record systems and profile risk of patients and this can be found here: Predictive Health. This second paper suggested that software may need to be subjected to comparable regulatory review like a medical device.

Just because you can’t drop it on your foot, doesn’t mean something can’t be dangerous.

Mike spoke at Advancia in Paris on 22 February on challenges facing entrepreneurs, with a particular emphasis on healthcare. Healthcare faces many challenges, but perhaps the greatest is how to deal with the future: the impact on health from climate change, demography, food production, and technology. Email Mike if you want a copy of the slide presentation.

Sample patient record view from VistA Imaging
Example of EHR (VISTA)

There is trouble in e-health land, at least in Ontario’s funny notion of what they might mean.  EHealth Ontario has been subject to an emergency audit of its procurement or not of an electronic health record [EHR] by the Auditor General of the province.  Apparently, somewhere approaching C$1 billion has been spent with virtually nothing to show for it.  The problems lie in a bad ehealth strategy, and inappropriate use of consultants.

There are lessons here for other jurisdictions, as they seek to embrace the benefits of EHRs, and ehealth more widely, in particular. Of course, what is an EHR for, is the core question.

One of my alma maters, McMaster University, has sprung into the fray saying it has an EHR called OSCAR that could be implemented for perhaps 2% of the estimated cost of a provincial EHR.  Their argument being that a lot of doctors are using it.

EHRs are not a tool for doctors, though.

EHRs are an integrated information repository to facilitate better healthcare.  Doctors are not the only oranges, and nurses, physios, social workers, pharmacists, OTs, oh, yes patients and parents, informal carers, too, need access to health records. In my view, patients should own and hold their own health record, to ensure high audit standards (would you let an error remain on your health record if you knew about it?).

Servicing the specific needs of doctors alone is not an EHR strategy worth having, and doctors themselves should be the first to say this. It is time they showed leadership within the wider healthcare system, and rejected self-serving models, such as McMaster’s, which automate obsolete information models. McMaster, too, should have known better.

The Ontario Ministry of Health has wisely rejected OSCAR’s offer, but for the wrong reasons.  Citing the need for doctors to choose their own systems, just shows their continuing logic of catering to the needs of a particular health profession, rather than addressing the systematic provision of patient information within an integrated decision-support system.

All this is being driven by beleagured officials who really need to think again about their priorities and why they really need an EHR.  Perhaps they are afraid to admit to having made a mistake.  Such hubris.

Clearly more work is needed to define the purpose of the EHR and the goals for an ehealth strategy in Ontario (and other jurisdictions of course), before more taxpayers’ money is spent on ehealth.

Oh yes, apparently Ontario are going for a tender on a diabetes registry. NYC has one. I fear the worst.

Reblog this post [with Zemanta]

The elephant in the room in healthcare is the hospital, about which I have suggested that we will build the last one in 2025.  What will “smart hospitals” look like, and why should we care?

Hospital Universitario Marqués de Valdecilla, ...
Hospital Complex, Spain

Why should we care?

Hospitals are expensive and complex labour intensive organisations originating in industrial era thinking.  Little has been done to modernise the institution itself, although much has been done of course to improve what hospitals do. We also know that hospitals account for a considerable carbon burden and consume a huge amount of energy since they operate 24 hours a day. We know that as labour intensive institutions they suffer from the challenges all such organisations face as they try to improve operating practices and reduce running costs. Healthcare delivery is characterised by regulated cartels, which serve both to protect the public, and protect professional practice from incursion by other health professionals.  A bit like an early 20th century factory with craft guilds.

We should care because these institutions need to become smarter in the use of modern technologies and practices, but this process is slow and cumbersome, and while they evolve, the taxpayer is faced with paying the costs of institutions which in many cases should be replaced. This is not to say that those who lead hospitals are not focused on these issues, but only to say that their job is not easy and with the many vested interests around, challenged.

What would be refreshing would be leadership for clinical workflow change to come from the professions themselves, due recognition of their need to evolve and reform rather than simply protect the status quo.  We need these groups to drive change in healthcare, rather than waiting for politicians or Ministries of Health to set the agenda. Of course, informed and empowered patients will eventually not put up with much of the nonsense that confronts them when they seek healthcare, but that is another story.

What will they look like?

We are left with wondering how to improve how they do what they do.  Enter ‘smarts’. This brings together a constellation of forces currently abroad in the world, ranging from automated building management systems, smart grids, energy recovery systems, to wireless technologies in hospitals to remove the wires.

Coupling smart systems together creates networks that can link patients in their home to monitoring facilities and first-responder capabilities. With the added advantage of wireless, we have untethered remote monitoring.  In the end, we have real-time healthcare.

Smart hospitals will not need to define themselves in terms of their geography or location, that is in terms of buildings. They will define themselves in terms of two factors:

  1. their capabilities and
  2. how they deliver these capabilities.

Indeed, the organising logic of the modern hospital will be replaced with one akin to a dating agency — it will link people with needs to capabilities to meet those needs — built on a sea of clinical, and patient information, and connectivity to various organisations that can deliver the services (healthcare) that is needed.  This breaks the current approach to vertical integration (based on the industrial conglomerate model) and replaces it with the virtual hospital, a network of focused and tasked organisations.

I had scoped such an approach to a redesign effort for a teaching hospital, which would have replaced a campus model (mainly an old building and some attached add-ons) with a distributed and electronically-linked (ehealth stuff here)  network of perhaps 24 centres scattered across a city of a million or so.  But industrial era logic prevailed and they went with the single building.

I guess we won’t get smart hospitals until we have smart planning.