Thursday, March 31, 2011

How to wash your hands properly in hospital

Great video from NEJM, free to view. Obligatory for all healthcare staff, (and also patients, families, kids so they can be sure their carers are doing it the right way.)


I was going to do a piece on immunization. These guys do a far better job than I ever could. Check out

Wednesday, March 30, 2011

How not to deal with the media

St Stephens Green, Dublin


A fantastic piece in the Archives of Internal Medicine, about incidental findings (incidentalomas) picked up while testing for something else. This opens up a whole Pandoras box. A great line about a New York Times columnist who believed he had survived a brush with death following the finding of one such incidentaloma. No, the authors say, he survived a brush with overdiagnosis.
This is very important. The fundamental issue is that each patient will feel they have been saved and each becomes an advocate for more and more testing. However, there is no evidence that many of these findings, even if malignant, will ever pose a threat to health. The authors provide some practical suggestions as to how best to approach these matters.

Clinically Integrated Systems

Following on from my thoughts about disruptive innovation, what structures in the UK and Irish setting might prove disruptive? Starting from the assumption, which may be mistaken, that some disruptors will be players in the current system, what could one possible future look like?
I believe the likeliest candidates are current general practice and primary care structures. They have the benefit of decent infrastructure, a critical mass of clinical, (not just medical) staff, and a population base that is large and varied enough to allow sub-specialistation. What are the missing ingredients?

  1. Seamless integration between priamry care and hospitals; to deliver this, I believe there needs to be consultant presence in the primary care teams. Consultants would work between hospital and primary care, with benefits to staff and patients, (improved access to specialist opinion, greater communication between professionals, enhanced learning opportunities for all)
  2. A payment system that rewards health, and is not just based on following processes, (e.g. check blood pressure every visit). What would this look like? Look at it from the patient perspective. What he/she wants is health, as few visits as possible, as little interruption to daily life, mot missing school or work. Can this be done. I believe it can, but it will require a fundamental shift in the way health, (as opposed to healthcare) is delivered.
  3. A recognition that an unplanned hospital admission represents a failure of care for the team providing care
  4. Quality and outcome metrics allied with an IT system sensitive enough to provide real time data
There is an interesting paper just published on the BMJ site by Chris Ham discussing this. They report the experience of Torbay, summarised below:
  • Reduced use of hospital beds (daily average number of occupied beds fell from 750 in 1998-9 to 502 in 2009-10)
  • Low use of emergency bed days among people aged ≥65 (1920/1000 population compared with regional average of 2698/1000 population in 2009-10)
  • Minimal delayed transfers of care
"hospital providers could also take the initiative in moving in this direction, especially in areas where general practitioners are relatively weak and specialists strong. London is a case in point, not least because it has several academic health sciences centres that present the potential to extend high quality care from hospitals into the community. In this context, integration might build on the strengths of academic health sciences centres by allocating them a capitated budget in conjunction with general practitioners and community health service providers. Patients would be able to choose between integrated systems based on academic health sciences centres and could also access care outside these systems in order to create an incentive for providers to deliver responsive care of high quality."

Are guidelines tainted?

One of the truisms of healthcare is that we should adhere to guidelines in so far as possible, to reduce variation, to improve outcomes and to minimize waste. Guidelines are developed by eminent professionals in a field; it is assumed that they are guided only by the evidence. However, many of these guidelines are sponsored by pharma companies.
A study just published in the Archives of Internal Medicine examined conflicts of interest amongst panel members charged with developing almost 20 cardiology guidelines. They find that 56% of panel members have a declared conflict of interest, e.g. owning stock in a company whose product is recommended, and 81% of panel leaders have a COI. That is not to say that these were unduly influenced by these conflicts, but some wonder whether like Caesars' wife, all should be above suspicion.
One suggestion is to prohibit all guideline developers to have a COI; my own view on this is that it may just exclude those who have the most to offer.

How the iPhone is going to save the world. Seriously!

An artist with an interest in healthcare recently asked me for my thoughts about the future direction of art (broadly defined) in children's healthcare. It's not an area to which I had given much thought but it caused me to consider the benefits, if any, of art and media in healthcare. Simply put, how best to use that which is the favorite activity of children, playing, to assist them in the treatment of their illness.
Just published in JAMA yesterday is a commentary which outlines some of the potential uses and benefits of video games. Before anyone rubbishes the concept, there are good precedents here. My younger patients undergoing a lung function test in which they must breathe into a machine as fast as possible are encouraged to do so by a screen with a birthday cake and candles; the harder they blow, the more candles are extinguished.
The paper in JAMA describes various tools. Mindless Eating Challenge, a weight loss program, is a game where users support one another by sharing photographs of portion sizes. Lit to Quit allows iPhone users to puff into the phone to simulate the experience of smoking.
While the skeptics may rightly laugh off these efforts, bear in mind these are very early days. Analogous to measuring urine glucose in diabetics in the 1920's. This took 5 minutes using some very sophisticated reagents. Underestimate the power of technology at your peril.
While I am not an expert in this area, it seems to me that the real focus of opportunity lies in the nexus of massive multiplayer on line games (MMOG) and social networking. Christakis has published a number of very significant papers showing the effects of social networks on depression, happiness, weight gain and smoking. Imagine a game, aimed at children with diabetes, where children can interact and earn from one another, in which the game incentivizes the appropriate behaviors, such as diet, exercise, compliance with screening and so on. Crazy? Yes, without a doubt. Crazier than our current system where we believe that seeing a child a few times a year in a busy environment with numerous distractions is going to deliver the best outcomes? You decide.

Tuesday, March 29, 2011

Art in Children's Hospitals

Fascinating website and project, thanks to Helene Hugel.

Healthcare; time to be disrupted.

A disruptive innovation (the term was first coined by Clay Christensen) is a technology or strategy, which disrupts an existing market by radically lowering price, increasing productivity or by gaining an alternative set of customers. Examples include the transition from sail to steamships, the adoption of personal computers in place of mainframes and so on. Henry Ford put it well; “to build a better motorcar one could either develop a faster and stronger horse or do something completely different.”

The question for healthcare is whether disruptive innovations are required to radically improve value and outcomes? Alternatively can tinkering with the system, (QIPP, ACO, Medical Homes, pay for performance, increase measurement and accountability) drive sufficient performance and improvement. To truly understand the significance of this question, one has to understand both the power of disruptive innovation and the failure of those agencies being disrupted to challenge the disruption.

Disruptive innovations essentially create a market for a product or service that the incumbents fail to recognize. They do so, often by offering a cheaper, slimmed down product; think Ryanair. One may not like them but they not only offer a cheaper reliable service, but developed a market for services that no-one could have imagined, (London Lodz anyone?).

Existing airlines had no incentive to develop such innovations; they were happy with their quasi-monopolies. Ryanair had to develop such markets in order to grow. if existing companies were to chase these market opportunities, they would cut into their existing markets and reduce their profit margins. In a sense they are helpless to do anything, apart from resisting and fighting back, (think lobbyists, negative advertising campaigns)

Jump to healthcare, the structures are identical. Seemingly strong vested interests, institutional, professional, academic, who will do their best to fight any interloper promising a better cheaper service. And they are helped immensely by the emotional arguments that if the service they offer is undercut, people will die.

Are there examples of such disruption in health? They are likely to appear when the current model no longer makes financial sense, or where Western models are not applicable. It is likely that if this economic downturn persists that we will begin to see disruptors emerge in both the West, and emerge from the developing world into the West. For example, “Aravind the world’s biggest eye-hospital chain, performs some 200,000 eye operations a year. It takes the assembly-line principle literally: four operating tables are laid side by side and two doctors operate on adjacent tables. When the first operation is done, the second patient is already in place.” (Economist April 2010)

A paper by Robert Brook of the RAND corporation outlined one vision of potential disruptors. These include:
  • Considering the entire cost of care, to include not just the cost to the individual but the energy cost, (carbon cost) 
  • Unifying the two great silos of medicine and education so that educators play a central role in improving children’s health and healthcare providers are judged partly on their success at improving educational attainment in their patients 
  • A shift to non-physican care, e.g. nurse practitioners 
  • Reduce the training required for repetitive technical task; for example why go to medical school, residency and fellowship to train how to remove cataracts. Pretty radical. 
  • Routine healthcare delivered 24/7. So no more 9-5 operating rooms or clinics. 
  • Outsourcing or off-shoring of many tasks, including diagnostics, elective surgery which is already happening. 
It is likely that the most marked changes are ones we cannot even begin to envisage yet. One thing is certain, we will have to change. Or perish. As Deming said, “We do not have to change, survival is not mandatory”


For some reason, posting a lot of stuff from the WSJ recently, but it's all good. Check out this piece about the need to innovate in healthcare.

"Unlike many other industries, health care has remained highly fragmented, with a hierarchical culture resistant to change, and a payment system that rewards providers for quantity rather than quality of care. It has been slow to adopt technologies like electronic medical records that have the potential to make care more efficient and safer for patients. Even when new practices have been shown to improve care and reduce error, hospitals and doctors have been slow to adopt them."


Great piece in todays Wall Street Journal, an interview with Pronovost, well worth reading.

Monday, March 28, 2011

Europe's Failing Health (Systems)

Great piece in todays Wall Street Journal. Discusses the challenges facing healthcare systems, from the funding perspective, throughout Europe. My favourite line,

"In 1995 the cost of a hip replacement was the equivalent of buying a flat-screen TV in Germany," he says. "In 2008 you could get 10 flat-screen TVs for the amount of money you paid for a hip replacement."

The conclusion? Healthcare funding and design must change, and change radically. 

Does improving quality save money?

The instinctive answer is, "yes, of course." If only it were that easy. I am more and more coming around to the view that to change a complex system like healthcare will require radical restructuring, which will only happen in the Western World when we realise we can no longer afford to deliver our current service. It wont be easy, it wont be pleasant.
Levette has sent me a paper which suggests that improved quality and lower costs may not go hand in hand. Bottom line;

"The evidence suggests that there are ways of saving money at the same time as improving quality. But doing so on a large scale requires changes in how services are financed, and increasing the capability of ordinary services to make changes. Approaches that look effective in theory rarely seem to have the same impact when implemented in practice and on a large scale. In this respect, the challenges of achieving real finan- cial savings by improving quality and reducing waste are no different from any other attempt to change a complex system."

Training for the future: from 20th century medicine to 21st century healthcare delivery

Check out this great piece from Peter Lachman. Nicely aligns with what I wrote about yesterday.

Disruptive Innovation

Hope to write a bit today about disruptive innovation in healthcare. Just to get into the mood. Thanks Lils.

From Apple Computers Think Different (R) campaign.

"Here’s to the crazy ones. The misfits. The rebels. The troublemakers. The round pegs in the square holes. The ones who see things differently. They’re not fond of rules. And they have no respect for the status quo. You can quote them, disagree with them, glorify or vilify them. About the only thing you can’t do is ignore them. Because they change things. They push the human race forward. And while some may see them as the crazy ones, we see genius. Because the people who are crazy enough to think they can change the world, are the ones who do."

Sunday, March 27, 2011

Medical Education; Fit for the 21st Century?

My personal view is that the current model of medical education is no longer fit for purpose; it is too heavily oriented to rote learning, is excessively hospital focused, pays little attention to the overwhelming needs of disease prevention, is excessively focused on providing healthcare as opposed to health, does not have a word to say about systems thinking and pays lip service to the concept of teamwork. There is no emphasis on the need to provide value, it does not provide the tools that allow professionals to continuously improve their service and has nothing to say about the economics of health and importantly our obligations as professionals to ensure money is spent to ensure that society receives the maximal benefit.
A very clever intern spoke to me recently, and I was discussing the concept of safety and quality. She told me she had never heard anyone in 6 years of medical school discuss these concepts. A shocking state of affairs.
Which brings me to the purpose of this post. A paper was recently published in the Lancet, and I will let the authors speak for themselves. Fundamentally, it is a call to arms, laying out the clear need to radically reform the model of medical education, in similar fashion to the Flexner report 100 years.

"Professional education has not kept pace with these challenges, largely because of fragmented, outdated, and static curricula that produce ill-equipped graduates. The problems are systemic: mismatch of competencies to patient and population needs; poor teamwork; persistent gender stratification of professional status; narrow technical focus without broader contextual understanding; episodic encounters rather than continuous care; predominant hospital orientation at the expense of primary care; quantitative and qualitative imbalances in the professional labour market; and weak leadership to improve health-system performance. Laudable efforts to address these deficiencies have mostly floundered, partly because of the so-called tribalism of the professions—ie, the tendency of the various professions to act in isolation from or even in competition with each other.

Redesign of professional health education is necessary and timely, in view of the opportunities for mutual learning and joint solutions offered by global interdependence due to acceleration of flows of knowledge, technologies, and financing across borders, and the migration of both professionals and patients. What is clearly needed is a thorough and authoritative re-examination of health professional education, matching the ambitious work of a century ago."

The Great Crack Up

"The test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function. One should, for example, be able to see that things are hopeless and yet be determined to make them otherwise."

F.Scott Fitzgerald, The Great Crack-up, 1936

I think the quote above will resonate with many of those who faced with the daily challenge of improving a health service still believe it is possible.

Is informed consent bad for your health?

A letter in this weeks Lancet is worth considering. Having been involved in a number of clinical studies in paediatric intensive care units, this resonated with me. The purpose of seeking informed consent from an individual enrolled in a clinical study is to ensure that they are protected in so far as possible and are fully aware of any possible risks entering this study may entail. The benefits of enrolling in a study, apart from the societal good of increasing knowledge is that there is some evidence that patients in a clinical study, regardless of whether they receive the experimental treatment or conventional treatment receive  better quality of care. There is also the possibility that they may benefit from the experimental intervention.
Using data from the CRASH trial, researchers estimated that in hospitals which sought consent from relatives, there was a 1.2 hour delay in initiating treatment compared to those hospitals in which no consent was sought. Applying these data to an analysis of the CRASH-2 trial, the researchers estimated that a one hour delay in commencing treatment resulted in a reduction in the percentage of patients responding to treatment from 63% to 49% with a likely increase in death and disability being the result.
There are two consequences; first the delay results in an increased risk of poor outcomes, second, the delay may obscure benefits that accrue from administering a time sensitive treatment.
 This has important ethical and scientific implications.

Saturday, March 26, 2011

Using Analytics to Create Your Own Upturn

Check out this SlideShare Presentation:

Sunset, Northern Ireland, Thanks Levette

Are you better off flying with a trainee pilot, or being operated on by a trainee surgeon?

One of the concerns that some people have relates to the way we train our doctors, essentially an apprenticeship system, with a trainee learning from a senior doctor. While I have some concerns about other aspects of this system, one of the unavoidable consequences is that trainees have to practice upon a patient. Is this safe? Surely being operated on by someone with less expertise increases the risk of something going wrong?
A study has just been published seeking to answer that question. The American College of Surgeons examined patient outcomes in more than 600,000 operations. They found that having a resident, (junior surgeon in training) while associated with a slightly higher risk of developing (usually minor) complications, was also more likely to be associated with a slightly reduced risk of death. The reasons obviously are unclear, but this important paper does provide some reassurance.

Wow, Oh Wow

That was Muir Grays Twitter post on seeing this weeks BMJ. Apart from the papers referenced below (Reducing Variation March 24th), there are a number of other papers worth checking out. A report from the Netherlands again finds a marked variation in utilization of medical interventions. An editorial on variation in the NHS; a look at how to redevelop care for long term conditions, and a lot more.

Why such emphasis on variation? Surely the areas with low utilization will balance out the high users for a zero sum game. This is unlikely to be the case, there is mounting evidence, mainly from the US, but also from other countries, that outcomes are not affected by the amount of care given above a certain amount. Therefore there are huge opportunities for reducing waste, saving money, and a lot more good stuff.

Thursday, March 24, 2011

Sunrise on the equinox

Eliminate Variation, Part 2

A piece in the BMJ by John Wennberg summarizes much of the knowledge around variation in medical practice. He breaks unwanted variation into three categories, effective care, preference sensitive care and supply sensitive care.

  • Effective care. This is care that is the right treatment for the vast majority of patients, and variation in this element relates to underuse, e.g. vaccination
  • Preference Sensitive Care. This is care provided when there are more than one effective treatment, and choice of therapy should be dictated by patient preference; in fact, physician choice often determines this. Mastectomy rather than lumpectomy plus radiation for breast cancer for example. 
  • Supply sensitive care. This is care that is dependent on the supply of services in a region. Roehmers law is one example of this; this states that in an insured popualtion, a hosoital bed is a filled bed. Dartmouth have found no corrleation between the amount of care received and outcomes, suggesting that there are areas with massive overuse of resources. Gawande in a brilliant article two years ago examined this discrepancy.
Also check out the NHS health Atlas. There are huge variations in care evident. Likely that this occurs everywhere.

Eliminate Variation

One of the godfathers of determining the extent of variation in medical practice and founder of the Dartmouth Health Atlas, Jack Wennberg, has recently published a book, "Tracking Medicine", available here. I have just downloaded it, will review it when finished, but it gets a great review here.


Wednesday, March 23, 2011


A view looking east from the Delaire Graff wine estate, Stellenbosch, South Africa.

The Future of Nursing

Medicine is excessively hierarchical. That is pretty obvious. While there may have been some benefit to this in the past, (though I am at a loss to think what that might be), clearly healthcare must be seen as a team effort. The role and contribution of nurses in delivering more and more effective healthcare has not been utilized to anything like its full potential. From the Future of Nursing, comes the following recommendations;
  1. Nurses should practice to the full extent of their education and training. 
  2. Nurses should achieve higher levels of education and training through an improved education system that promotes seamless academic progression. 
  3. Nurses should be full partners, with physicians and other health professionals, in redesigning health care in the United States. 
  4. Effective workforce planning and policy making require better data collection and an improved information infrastructure.

Reform of Public Service

This is the subject of a special survey in this weeks Economist. While looking at public services generally, criticizing (rightly in my opinion) the growth and low productivity of services worldwide, it has a specific mention of health. Well worth reading.
Amongst the figures given; "McKinsey points out that American spending on this has grown at an annual lick of 4.9% over the past 40 years, whereas GDP per person has grown by just 2.1%. Pessimists are convinced nothing can be done to restrain it. A refreshingly different perspective is provided by Sir John Oldham, a British doctor who is clinical lead for productivity in the National Health Service."
"America, which currently spends 16% of its GDP on health care is theoretically on track to spend 100% of its GDP on health care by 2065."

Tuesday, March 22, 2011


A view from St Johns bridge, Kilkenny, Ireland looking north at the River Nore on a glorious Spring day. Check out

What is medical harm?

My favorite definition, from a good friend, (Thanks Peter) is that harm is anything that you would not want happen to you or your family.

Some Thought on Value in Healthcare

Value is a dirty word in healthcare, conjuring up visions of ruthless bean counters, scrabbling for pennies while patients suffer. How different though is the reality? Regardless of location, healthcare costs are rapidly climbing, disease burdens, therapies, diagnostics and expectations are rising faster. Like it or not, hard choices must be made, if not by clinicians, then by the payer, ultimately the taxpayer. How does one align this imperative with the need to ensure quality of care is optimized?
A recent paper by former colleagues of mine brought home the reality of this dilemma. Cruz et al sought to improve compliance with goal directed therapy for children with sepsis. They demonstrated a marked reduction in time to first fluid and antibiotic administration, consistent with guideline recommendations. However, part of the solution was to deploy extra nursing, pharmacy, medical and EMT staff. While this may well have been a beneficial use of this scarce resource, it is unclear if the value increased.
Value can be defined as health outcomes (quality) divided by the cost of achieving those outcomes. So while quality almost certainly increased, it is likely that the cost of achieving this level of quality also increased. The question that must be asked, could better or more health outcomes be achieved for the same expenditure. Unfortunately our measurement systems are as yet unable to answer this question. For a more detailed discussion around this topic, please see Michael Porters website.

Boeing to the Rescue?

A recent piece in JAMA by Pronovost is well worth reading. In it he contrasts the way we design (or rather fail to design) healthcare, especially in relation to equipment, (haphazard, no systems thinking, individuals insisting on their preferred piece of technology etc) versus the way airlines buy planes. They do not buy planes, each of which has different toilets seats, lifebelts etc. They buy a standard plane, for economic as well as safety (reduce variation ) reasons. He suggests in an American context that what is needed is a systems integrator, similar to Boeing. It is interesting that national healthcare systems, despite being in a better position to act in this way, singularly fail to do so.
I have worked in intensive care units where the number of different types of ventilator exceeded the daily census of ventilated children; where the choice of a specific ventilator that a child was placed on depended on which physician was on call. This type of variability is hugely damaging, expensive, a safety risk and results in poor training. Compounding this is a perceived need for hospitals to get the latest new thing, often resulting in a situation where there are insufficient patient numbers to allow all team members to develop the required expertise and experience which are necessary to deliver the best outcomes.
While I agree with his arguments, I think he fails to develop the logical conclusion, in that the entire system, inside and outside hospitals needs to be standardized as much as possible.


Snow on the Charles River, Cambridge, Massachusetts.

Monday, March 21, 2011

Psst, want to buy a cheap electronic medical record?

Beginning in the 70's, the Veterans Administration developed an in-house electronic medical record, (EMR), now termed VISTA, which has a number of attractive features as outlined. Notably it is free, which given the financial challenges facing this country is not to be sneezed at.


VISTA includes computerized order entering, barcode administration, electronic prescribing and clinical guidelines. UPTODATE can be accessed from within the system. It also allows (at least for the Veteran population) patients to access their own medical records, similar to the KP system. Vista Imaging allows integration of PACS, pathology, EKG etc. the VISTA system has been used in conjunction with telemedicine to provide surgical care to rural Western States.


VA claims a pharmacy prescription accuracy rate of 99.997%, and the VA outperforms the vast majority of public hospitals, supposedly based on its EMR implementation. There are only three hospital systems in the US, which have achieved HIMSS stage 7, (the highest level of IT implementation), one of which is the VA. A non-VA hospital using VISTA is one of only 42 US hospitals to achieve HIMS level 6.


Approximately 50% of US hospitals with a full EMR implementation are VA hospitals using Vista. It has also been implemented worldwide, including Finland, Germany, and Denmark as well as developing nations.


As it is open source, the license is free. Obviously there are large costs relating to localization, hardware, training, and implementation. I have found one reference to implementation of VISTA over 8 hospitals in one state costing $9 million, as compared to a commercial installation over the same number of hospitals costing $90 million.

See here and here for more information.

Sunset, Discovery Park, Seattle

Sunset looking west over the Olympic Peninsula from Discovery Park, Seattle.

Sunday, March 20, 2011


A very moving post by Anna Roth on her blog about the suicide of her brother in law. She has previously written about suicide, but not this powerfully. The statistics are terrifying, one of the most common causes of death in the 25-44 year old age group, worldwide! Incidence has increased worldwide by 60% over the last few decades. Most common methods are hanging, ingestion, (usually herbicides) and gun shots.
It is estimated that there are 30,000 suicides annually in the US. The most recent data I have seen for Ireland suggest that there are almost 600 reported deaths by suicide annually, a rate substantially higher than the US. The WHO report that the rate in Ireland (11.6 deaths/ 100,000 inhabitants) was 10-15% greater than the average rate in the EU-27.
Can anything be done, an especially pertinent question given our economic collapse and the likely toll that this will cause? While the HSE has a suicide office and a prevention strategy, I have no expertise with which to judge its efficacy. However there are healthcare delivery systems that have shown what can be achieved. The Henry Ford system in Detroit shows what can be achieved. In the first 4 years of the groups suicide prevention program, the rate decreased from 89 to 22 deaths/ 100,000 population. In the most recent analysis the rate over the previous two years had dropped further to zero deaths!
For details, click here. This in a city that has been battered economically over the last decade.
There is another very emotional piece in this weeks JAMA, again an account of suicide from the perspective of a family member. In a future piece, I hope to write about the risks of suicide amongst physicians and ways to reduce this risk.

Friday, March 18, 2011

Optimizing Patient Flow to enhance productivity and safety, Part 2

Coincidentally, following on from my recent post about patient flow, (March 16th) comes  a paper which again demonstrates the critical need to optimize patient flow, not just to improve productivity, but more importantly, to reduce mortality.
Just published this week is a very important paper in the NEJM, here. The authors looked at the effect of nurse staffing numbers on in-hospital mortality in a large academic hospital. It shows that peaks in patient flow (turnovers) are an even greater cause of mortality than patient per nurse staffing ratio. The authors state,
"We also found that the risk of death among patients increased with increasing exposure to shifts with high turnover of patients. Staffing projection models rarely account for the effect on workload of admissions, discharges, and transfers. Our results suggest that both target and actual staffing should be adjusted to account for the effect of turnover. In light of the potential importance of turnover on patient outcomes, research is needed to improve the management of turnover and institute workflows that mitigate the effect of this fluctuation."
The basis of this is simple. Elective admissions are hugely variable, and dependent almost entirely on doctor choice. Because these admissions occur without any reference to the other needs of the hospital, they cause huge peaks and troughs in patient numbers, e.g. not many elective patient will be admitted Friday.
One of the worlds leading experts in patient safety, Peter Pronovost, has also made clear his view that optimizing patient flow is essential for reducing in hospital mortality, see this recent paper.
There are huge opportunities to be had.

St Patricks Day Parade in Dublin

A great parade, weather perfect, crowd friendly, Dublin at its best. The building in the background is the Rotunda Maternity Hospital, the setting for the birth of the eponymous baby in the book by Roddy Doyle, "The Snapper"

Think outside the box

An interesting blog post from Moss Kanter about innovation. Basically very few organization can shift their way of thinking from business as usual. This is a cause of great frustration for the visionaries and innovators within that organization. Great relevance to healthcare.

Thursday, March 17, 2011

Culture eats strategy for lunch

The title of this post refers to an oft quoted piece in business and change literature; to paraphrase, you can have a great plan but if it doesn't take into account the culture of the organization, you will fail. This I think is the crux around which healthcare reform and quality improvement specifically will succeed or fail. This is the great intangible that must be isolated and measured, so that we can truly begin to determine how healthcare providers differ from one another in terms of the quality of the service they deliver. As doctors we believe that the keys to a great service are world class doctors, state of the art facilities, the latest in IT; basically the best that money can buy.

However a study just published in the Annals of Internal Medicine, link here, (subscription required), suggests that this credo is incorrect. Curry et al interviewed 158 staff members from 11 hospitals, which were either in the top or bottom 5% nationally in the US for mortality post heart attack. The factor most highly correlated with outcomes was a cohesive organizational vision that focused on communication and support of all efforts to improve care. In other words, it was the culture, the communication ethos, mutual respect, leadership and desire to improve that should determine where you want to be treated, not all the high tech stuff and big names. This is consistent with my own observations nationally and internationally. All hospitals have problems, the high performers are the ones that seek out the problems and respectfully engage all their staff and leadership to solve them. It is my strongly held belief that one can distinguish between the high and low performers within an hour of visiting them.

Previous studies have suggested that the traditional factors that underpin success include being an academic medical centre, having more beds, and being located in a large city. Curry found that these factors accounted for only 20% of the difference. This is a very significant study, and provides ammunition for those of us who believe the system can be dramatically improved without massive expenditure; the down side is that changing culture can be extremely difficult.

Wednesday, March 16, 2011

Optimizing Patient Flow to enhance productivity and safety

 What if the healthcare system could introduce a method of improving productivity by 15-20% within months, at no extra cost, but by doing so improve safety dramatically, reduce stress, burnout, absenteeism, patient experience, and all the other good things? There is such a system, with very good evidence behind it. See link here.

1.    Safety:
a.    Arguments around safety in healthcare worldwide usually centre around the need for extra resources, rather than in the first instance optimizing the system in place. By reducing variability in the system, cases can be more closely matched to the available infrastructure and staffing resources, thereby reducing the dramatic peaks and troughs that are seen in every hospital, especially with respect to elective activity. These artificial peaks (due almost entirely to elective activity) may then give rise to the impression that the facility is under-resourced and under-staffed; by reducing this artificial variability the current infrastructure and staffing levels can be utilized to deliver a service that is safer (right patient in the right bed at the right time), and avoid a mismatch between staffing (esp. medical and nursing) and patient numbers.

2.    Efficiency/ productivity:
a.    By utilizing resources as outlined in the previous section, expensive plant such as diagnostics, operating rooms, and intensive care beds can be utilized more effectively and productively. The traditional upper estimate of hospital occupancy has been estimated at 80%; the only rationale for this figure is to provide a surge capacity, a buffer for times of increased demand on the system because no control was placed on how the elective component of demand was managed. By properly managing this demand, it is likely that overall occupancy can reach 90-95% on average safely. One hospital has reported a continuous occupancy rate of 91% since implementing this system. Rather than reducing the number of cases performed, this efficient allocation of a scarce resource allows each individual surgeon to perform more procedures.

3.    Financial benefits:
a.    Obviously, under the current Irish hospital reimbursement scheme, financial benefits are harder to estimate and realize. However, it is clear from point 2 that the return on investment is greater by optimizing hospital flow. Whether this equates to greater revenue as in the US system, or more patients treated for a given budget as in the Irish setting, the end result is greater productivity.

4.    Patient satisfaction:
a.    Hospitals that have (in the US) implemented this system have eliminated wait lists, ER waits, cancellation of elective procedures etc.

Evidence base.

1.    The American Hospital Association has identified six priority areas in its efforts to accelerate hospital performance improvement; these include health care acquired infections, health information technology, medication management, patient safety, and of relevance to this document, optimizing patient flow and throughput. The AHA state that “patient safety is negatively impacted when patients do not move through hospitals in a timely and efficient way. The AHA endorses the methodology of the Institute for Healthcare Optimization as a means of improving patient flow.

2.    The Leapfrog Group for Patient Safety is an umbrella group representing some of the largest employers in the US, (IBM, Intel, Toyota, Boeing, Motorola, FedEx). Collectively they represent employers and agencies that purchase care for more than 35 million people in the US. Leapfrog ranks hospitals in the US publicly on a number of patient safety measures. A study by the Commonwealth Fund (June 2008) reported that hospitals publicly reporting their Leapfrog Quality data had a lower mortality rate for Acute Myocardial Infarction and Pneumonia. Participating hospitals are asked if they adhere to the following quality and safety practices:

                                               i.     Computerized Physician Order Entry
                                             ii.     ICU Staffing with intensivists
                                            iii.     Compliance with performance standards for certain high risk treatments
                                            iv.     Leapfrog Safe Practices Score

3.    From 2011, the Leapfrog Group will measure a hospitals policy around flow optimization and safer scheduling, (personal communication). The group states that patient safety can be negatively impacted by a hospitals scheduling policy. The Leapfrog group supports the methodology of the Institute for Healthcare Optimization as a means of improving patient flow and reducing patient risk.

4.    In 2006 the Institute of Medicine (IOM) released “The Future of Emergency Care in the United States Health System”, a series of reports assessing the severe problems facing the nation’s emergency care system and offering recommendations to improve it. The report states "hospitals should reduce crowding by improving hospital efficiency and patient flow and using operations management methods and information technologies."
5.    The Joint Commission, the body responsible for accreditation of US hospitals, endorses a strategy of providing scientifically grounded methods to optimally manage patient flow and reduce variability as part of a policy to improve patient safety. It endorses the methodology of the Institute for Healthcare Optimization.
6.    Don Berwick, the head the Center for Medicare and Medicaid Services (CMS), one of the most powerful positions in American Medicine, has stated that there are two sleeping giants in trying to achieve quality healthcare, patient safety and patient flow.
7.    A number of the largest and most renowned US academic health centers have recently begun work on redesigning their patient flow and throughput, (personal communication). One 212 bed hospital has seen a revenue boost of $50 million annually within one year of implementing this system. Cincinnati Children’s has seen its productivity increase by the equivalent of a 100 million dollar capital rebuild, (100 beds) and its annual revenue increase by $137 million.
8.    A number of Childrens Hospitals including Cincinnati, Lucille Packard, and Great Ormond Street have adopted this approach.
9.    There is emerging evidence in the medical literature that failure to incorporate these principles is associated with serious adverse outcomes. In one study by Pronovost (2009), the odds of being readmitted to an ICU were five times greater when the number of patients admitted to the ICU were above a certain volume. 

Monday, March 14, 2011

Reporting Poor Performance by Doctors

A very concerning piece, but not that surprising to those in practice. A study of >1000 UK doctors found that almost 30% would not report concerns about a colleagues poor performance. The main reason for this was concerns about retribution. It brings to mind a conference I attended 2 years ago in which a keynote speech was by three well known, (non-nmedical) whistleblowers. They made the point that whistleblowers are often motivated by anger or revenge, and that if they were faced with the situation again, they would not reveal their misgivings. This was very sobering; one assumes that whistleblowers (an awful term) are motivated by altruism.

Progress of the Safety Movement

Some (relatively) disappointing news recently. An analysis of the Safer Patients Initiative was published in the BMJ (here and here). This was a large scale intervention comprising a large number of components of care. Unfortunately, there was no evidence of any difference in outcomes between control and intervention hospitals. What could the possible reasons be? There are a number of hypotheses,

  1. hospitals will get safer regardless of interventions, and do not need this type of large scale change, (unlikely)
  2. These interventions were too complex, encompassing 43 different interventions, (probable)
  3. Management and clinician buy in, expertise, knowledge and support were insufficient to show a difference, (highly likely)
  4. The interventions sought has insufficient evidence to underpin their use, (highly likely in some cases)
  5. Many hospitals, (both intervention and control) already had high levels of quality in some domains, hence the big effects were less likely to be seen, (likely)
So what lessons can be drawn? I think there are a number.

  1. Large scale change is difficult, messy, a long term commitment and often fails
  2. Leadership, at management and clinical level is critical
  3. Improvement and quality must be seen to be the only way to do business, not an optional extra
  4. There must be a better system of measurement; even today measuring mortality is contentious. The ideal measurement system should be one that measures patient outcomes from the perspective of the patient, and reimburses the system, not a provider for optimal outcomes. See Micheal Porters work from Harvard for more on this