This past July I spent a week at Imperial College in London, most of it working in the Innovation Group where I have an appointment as adjunct professor. During one of the meetings, we had a fascinating discussion on why it is so difficult to make changes in the UK health system. My colleagues pointed out that if you look at the organization of the healthcare system in the UK, you can identify two main levels of management. At the operational level, you have all the hospitals, doctors, nurses and others delivering health services to their clients. At the strategic level, you have the publicly funded National Health Services (NHS), which is charged with overseeing public health care across the UK, reporting to the Department of Health.
In a typical company, they pointed out, there is a layer of middle management whose primary responsibility is to act as a kind of gear box between senior management and the organization’s operational activities. Generally, middle management is responsible for monitoring the activities of their subordinates, reporting back to upper management, and carrying out the strategies formulated at the senior management level.
But when an organization is so complex, distributed and fragmented, like the NHS, you have a large gap between the senior managers at the top of the organization and everyone else performing their daily operations of delivering health services. There is virtually no middle management. So the NHS can formulate strategy after wonderful strategy to improve the quality and productivity of healthcare in the UK, write great papers and give compelling speeches. But then the strategies fall into the organizational abyss. There is no one to carry out the strategy and oversee the needed changes. Doctors, nurses and administrators will continue to do what they always have done, given the semi-isolation from the top caused by the middle management gap.
I found this hypothesis really intriguing, and have been wondering ever since to what extent it applies to complex organizations in general, as well as what to do about it. I wonder if the hypothesis applies not just to incredibly complex systems like healthcare, but to all forms of complex enterprises, especially those that are doing business around the world and are attempting to improve their global integration.
For example, in MIT's Engineering Systems Division, we have started to investigate how to bring to healthcare engineering practices such as Lean Production and Six Sigma that have been so successfully applied in a variety of industries, such as automotive. As part of our efforts we have reached out to experts in healthcare.
We met Dr. Karen Feinstein, CEO of the Pittsburgh Regional Health Initiative (PRHI) and one of its co-founders, along with former Treasury Secretary Paul O'Neill. PRHI’s mission is to act as "an independent catalyst for improving healthcare safety and quality in Southwestern Pennsylvania, [operating] on the premise that dramatic quality improvement is the best cost-containment strategy for health care.”
In her talks, articles and interviews, Dr. Feinstein eloquently lays out her case for what needs to be done: "We weren’t looking for national policy solutions. We were looking at what happens at the point of care – when care is delivered to patients. We said that if that goes wrong, everything else goes awry. If it’s not delivered with best practice care, efficiently, and safely – if these things aren’t in place, you’re going to be paying for all the wrong things and actually harming people. Our idea was to bring the revolution down to the point of care where worker touches patient and to get out of the policy realm. We asked people to set their sights high."
Their vision of what needs to be done is simple to articulate: "Frontline clinical teams applying daily problem-solving methods and work-process improvement techniques to deliver perfect care to patients." If this approach reminds you of the Toyota Way that revolutionized manufacturing over the last thirty years, it is no coincidence. Dr. Feinstein frequently mentions how much her ideas have been influenced by the work of the Toyota Way, Six Sigma and similar well accepted engineering practices to reduce errors in complex operations systematically.
The problem, she told us, is that you can make progress one hospital at a time by carefully improving their operations and training methods, but there are no practical mechanisms to scale what needs to be done to all hospitals and all physicians across a region, let alone across the whole nation, as we so badly need to do to reduce the runaway costs and the high numbers (100,000 or so) of US lives lost every year due to medical errors.
We know from experience that trying to set up layers of middle management or a relatively large bureaucracy to fill the gap between strategy and operations in a hierarchic organization not only does not work, but inevitably becomes too costly and counter-productive. In fact, over the last twenty years most companies have dismantled such extra layers of management and staffs to reduce costs and improve organizational agility. So, what should we do? How can we best fill the gear box role between strategy and operations in a highly distributed and complex organization where a traditional hierarchy does not work?
Increasingly, I have seen community-based efforts within companies, leveraging Internet-based collaborative platforms acting as a very effective balancing function and gear box between strategy and operations, and contributing innovations to improve both. Working with each other, they come up with new ideas, which they then proceed to refine and filter through their extensive interactions. Only the best ideas survive.
Properly focused to encourage action rather than just talk and analysis, communities can act as very effective agents of innovation and change. Their wisdom of the masses approach - enabled by the right tools and facilitators or coaches - usually leads to a very effective meritocracy, where the people with the best track record in the eyes of their peers gain the respect of the members of the community, thus improving their ability to lead.
These collaborative, social networks nicely complement the strengths and weaknesses of a company's classic hierarchy. Communities are most useful when dealing with complex areas, where the problem to be solved has not yet been well framed, information is incomplete and innovation and extensive change are required. Hierarchic management is not as effective under such circumstances. On the other hand, where the actions are clear, all the needed information is available and it is primarily a matter of execution, an organizational hierarchy performs best.
One nice aspect of Internet-based communities is their scalability in multiple dimensions. First, it is relatively easy to include large numbers of people from all over the world, as we have learned in IBM with collaborative platforms like InnovationJam and Thinkplace. We have also learned the value of reaching out beyond the IBM walls to include business and research partners, industry associations and clients. Diversity of opinions is an important asset when dealing with collaborative knowledge, as long as you have the right technical tools, well trained facilitators and good processes. Internet applications based on open standards make it relatively easy to include a large external, as well as internal, community.
OK, you might ask, but how can the experiences of IBM and Toyota of complementing hierarchic power with community power translate to organizations as amorphous as healthcare systems? Does community power have a role in filling the middle management void in healthcare? Can communities act as effective counterbalances between powerful politicians and powerful physicians?
In fact, the more I think about it, the more I have become convinced that the only counterbalance to physicians, politicians, payers and other powerful institutions are properly organized communities of patients, their families and care givers. A physician can easily blow off a hospital administrator. Perhaps he or she can ignore NHS’s and PRHI’s advice on improving quality and productivity. The doctor might even be able to dismiss what individual patients tell him or her, one at a time.
But I cannot imagine how physicians or anyone involved in healthcare can ignore well organized communities of patients -- especially patients, their families and caregivers who have a critical common interest in their ailments, be it breast cancer or bipolar disorders, and who will not be shy about expressing their opinions as a group, about the care they are getting from specific physicians, nurses or hospitals. To me, this feels like a mighty, free-market weapon to be reckoned with.
Who should organize such patients and offer them platforms and facilitators? Whoever is interested in improving the quality of the healthcare system, from local and regional government and non-profit groups; to enlightened physicians, payers and hospital administrators; to the patients themselves. Lots of other things we have tried have not worked, so perhaps this is an idea whose time has finally arrived.
There have been two significant advances in 'healthcare' over the last couple of centuries, that have lengthened lives.
Firstly, 'public sanitation', which stopped people (in the Western world) from dying aged 30 of waterborne disease.
Second, 'antibiotics', which stopped people (again in the Western world) from dying of infectious disease; and so now we live to 70-80 years typically.
The next step ... getting people to live to 150 ... will require addressing the 'wear-out problems'; the cancers, heart diseases, and other hereditary reasons why some people live longer than others.
Could we but understand the basics of that, the 'what makes you special', maybe we could individualise (or mass-customise) the care giving too ?
But I don't think incremental improvements to the current approaches will even get close. It needs a new 'thing like an antibiotic', a new breakthrough.
Posted by: Chris Ward | October 13, 2007 at 03:03 PM
There's a particularly good example of patients supporting each other : the Yahoo group 'surfacehippy' , for people who wish ot investigate the (recent to the US) technique for replacing hip joints using a hip joint sized metal ball and socket which fits over (rather than substitutes for) the remaining healthy femur.
Collective wisdom on how to work out if you are a candidate, what common FUD to expect from inexperienced doctors, which questions to ask to determine the experience of the surgeon, mutual support on pre operative and post operative recovery optimization. Upwards of 6000 members, more than 5 years of cumulative experiences, several hundred posts some days.
Posted by: anne | October 15, 2007 at 10:28 PM
I think you're up to something, and so I took this as starting point for a post on social software and organizational change management - is this an undervalued benefit of social software in the enterprise?
http://www.martin-koser.de/BMID/index.php/archive/social-networks-and-organizational-pathologies/
ps. Trackback didn't get through?
Posted by: Martin Koser | October 16, 2007 at 10:24 AM
Sermo is a start, Irving (www.sermo.com). As for patients, it already exists, in facebook, ning, etc.
As for enterprise collaboration, a comment I started repeating lately: I've made more connections with my fellow IBMers through facebook and twitter than I do in all of w3, SameTime, BluePages, Notes, Wiki/Blog Central, etc. We need to better harness the power of our 350K people networking.
Posted by: joe | October 28, 2007 at 11:34 AM