Monday, April 22, 2013

The Person, Not the Patient

"Who the person is with the illness is usually more important than what illness the person has." Sir William Osler.

A recent article entitled "Patient -Centered Decision Making and Health Care Outcomes, An Observational Study" in the Annals of Internal Medicine by Saul Weiner and his associates at the University of Illinois shows, to quote the article, "Attention to patient needs and circumstances when planning care is associated with improved health care outcomes". Not only does it lead to improved health outcomes but also to lower costs according to another recent article by that same group published in British Medical Journal Quality and Safety. While in Weiner's work there are many implications related to our health care challenges of high costs, difficult access, and variable quality, I also found within the findings of the research a return to the wisdom of William Osler quoted above.

Osler also said, "The good physician treats the disease; the great physician treats the patient who has the disease". Weiner , in his work, has found a way to measure the great physician. He has developed a research tool to determine if the physician (or nurse or other health professional) is trying to find out "who the person is with the illness" rather than only "what illness the person has" in the words of Osler.

Before describing that research tool an old joke bears repeating. A man is walking down the street late at night and sees another man on his hands and knees under a street light. The first man asks the man on his knees, "Are you looking for something?". The man on his knees replies, "I lost my keys" to which the man walking along responds, "Lucky you lost them near this light". The man on his knees says, "I didn't. I lost them about half a block from here". The upright man then asks, "Why are you looking over here?" The man on the ground answers, "Because this is where the light is".

Weiner and his group are not just looking where the light is, which is what many do when they limit research to claims databases and even medical records. Weiner audio records visits between doctors and patients to discover the true interaction and to discover how often the doctor is investigating the context of the person's medical problem. In his research, Weiner has defined ten elements of context and he and his staff listen to these office visits to determine if the doctor has asked about the person's life and values and whether the doctor has tried to find ways to remove the barriers to good care that the life context may have created. The ten elements, or domaines of a patient's context are access to care, social support, competing responsibilities, relationship with health care providers, skills and abilities, emotional state, financial situation, cultural beliefs, spiritual beliefs, and attitude towards illness. The researchers then use a standardized survey instrument, with multiple listeners independently rating the visits, to detemine the number of errors made in eliciting the "context" from a patient and in addressing that life context. Using this technique and survey tool they have shown that understanding the person who has the disease leads to good, and efficient care while just paying attention to the disease itself is hazardous to the patient's health.

One of the more recent studies, published in 2010 found that if a patient had no significant contextual issues, doctors provided appropriate care 73% of the time. If there were contextual issues that were missed then appropriate care was given only 22% of the time. The costs were significant as well, as described in the BMJ Quality and Safety journal article. They found that these contextual errors on average increased the cost per visit by $234. Medical errors in comparison increased the cost an average of $164 per visit. In total, the visits they recorded and assessed for errors by audio recording wasted $174,000 due to the errors while the errors found by chart review alone accounted for $8,700 in waste. Yet the bulk of the work in health policy and in managed care is related to the medical and not the contextual and is based almost entirely on claims review and chart review rather than actual recording of visits.

This leads to another quote by Osler: "Variability is the law of life, and as no two faces are the same, no two bodies are alike, and no two individuals react alike and behave alike under the abnormal condition, which we know as disease." Yet we keep believing that standard medically focused algorithms that hold doctors accountable to do the same thing with each patient are the answer to our health care dilemma of high costs and inconsistent quality. Don't misunderstand. The algorithms and the technology greatly improve the science and are welcome and necessary. However they may have an unintended consequence of worsening the art of treating the individual by focusing too strongly on steps in medical therapy even when the context makes those steps difficult or impossible. Both knoweldge of standard algorithms and of a person's life context are critical to good patient care. Osler understood this and now Weiner has demonstrated Osler's wisdom using experimental methods and scientific techniques.

As Osler also said, "Medicine is a science of uncertainty and an art of probability". Weiner has shown that the individual's values, family, finances, faith, emotions and everything else that makes that person a complex being must enter into both the science and the art for good, efficient care to take place. I doubt that can be accomplished by technology and evidence based algorithms alone. It also requires caring professionals helping people in need. We must either give our doctors and nurses the time, training and tools to "diagnose and treat" the context or develop new professionals, as we at Accolade are doing, to address these individual and population based clinical, financial and human needs.

Sunday, April 14, 2013

Constraints and Healing: The "Triple Aim" Meets the Reality of Illness

The “triple aim” of health care has been present for a long time but that particular term was first voiced by Don Berwick, ThomasNolan and John Whittington in 2008, in an article in Health Affairs.  That triple aim is care, health and cost.  This month’s Health Affairs is dedicated to the “triple aim” going global as more countries around the world struggle to find solutions to the care, health and cost dilemma. 

In the original article in 2008, the authors wrote about the need for three constraints to be put into place to drive us closer to the triple aim.  They were “(1) recognition of a population as the unit of concern, (2) externally supplied policy constraints (such as total budget limit or the requirements that all subgroups be treated equitably), and (3) existence of an “integrator” able to focus and coordinate services to help the population on all three dimensions at once.”  These constraints have societal and population concerns as the focus.  The “integrator” is there to “help the population” rather than to help the person.  I worry about a certain professional arrogance that is inherent when one defines constraints for people rather than asking people as individuals, what they need.  While populations are made up of individuals, individual’s values and views may be minimized when the population is the focus.  When the first step is the development of constraints, the individual is likely to see those constraints as being against their own best interest. 

This focus on the population, as it is currently interpreted, runs the risk of minimizing the need to understand the many human elements in dealing with illness.  It can downplay the fact that the ability of any one person to heal is often as dependent or even more dependent on the social, economic, psychological and spiritual parts of their life as it is on medical science and health system operations.  People need to have some help balancing their life with their illness when they are sick.  People, as individuals need to have a trusted resource to help them through a maze of difficult decisions and difficult choices that poor health demands.  People need help finding their own voice in a system that can appear to be more concerned with population needs than with individual patient care. 

Good health care is labor intensive and the labor is expensive.  Much of that labor is involved in the technical aspects of care but patients and their families also value expert assistance in decision making requiring knowledge and skill that is totally focused on the individual and their own understanding of how they want to live their life and how they value the options before them.   Traditionally physicians and nurses have taken central roles in that communication and decision making.  However our attempts to create efficiencies, as our population grows more diverse and our health system grows even more complex, have caused doctors and nurses to spend less and less time understanding the person as a person, understanding their values and respecting their autonomy.  Doctors and nurses are considered too expensive to be used in such a way.  But that human need is still there and the emphasis on doctor and nurse productivity has led to a void as the time to understand the patients before them as people is not seen as productive.  That void, has led to higher costs as people give in to their fears and isolation and flail around a health care system as they try to find their way to healing. 

An article in the “triple aim” issue of Health Affairs by Michael Macdonnell and Ara Darzi entitled “A Key to Slower Health Spending Growth Worldwide Will Be Unlocking Innovation to Reduce the Labor Intensity of Care” addresses this issue of the cost of expert health labor.  In the United States, 56% of health care costs are labor costs. They point to labor saving technologies, telemedicine services, and the high productivity centers in India at which “the hospital uses expensive assets such as cardiac catheterization laboratories at a rate five times that of US hospitals” to achieve a resultant lowering of the labor costs for each procedure with an efficient assembly line approach.  They speak of the need for more “patient self-service” especially in the management of chronic disease.  They do not address the need to treat the whole person; the need to address the fear, isolation, loss of autonomy, possible loss of job, and possible loss of feeling of personal worth that can easily be a part of any illness. 

We continue to try to find solutions to the triple aim but always appear to focus on the costs and to focus on disease as divorced from life.  Until we truly understand and address the issue of illness from the point of view of the person who is sick and from the family who cares about that person, we will miss the boat. 

At Accolade, the company I have had a part in building, we have addressed the labor economics by addressing individual’s issues of life and illness directly.  We have created a new profession of Health Assistant.  The Health Assistant offers expert decision support and expert knowledge of the whole person and their family to help with all the implications of the illness.  Each Health Assistant maintains a focus on the specific needs of that person and family and helps them from their perspective.  They are serving that person and that family and purposely not focusing on the triple aim goals of the health policy experts.  The irony of this focus is that by maintaining the focus on the person as part of a family and community unit, savings are achieved that are greater than in programs that constrain and greater than in programs that attempt to tell people what is best for them. 

Our new profession is in the business of helping the person find his or her own path to health and healing but not in diagnosing and treating disease.  Our innovation is focused on building a skilled work force to focus on the needs of the individual and the family from their reality and their prospective as they access health care while living their lives.  

Sunday, April 7, 2013

Genomics, Clinical Risk and Financial Risk

We now have the entire human genome defined and “sequenced” and the costs of this clinical sequencing for individuals have decreased dramatically.  In medicine today, that knowledge is being used to develop new tests so that health professionals can use the genetic tests to assist in making a diagnosis and deciding on treatment.  However as the costs of whole exome and genome testing go down even more, a price point is nearing at which it will be less expensive to do a complete clinical sequencing than to do specific genetic tests.  This promise of more genetic information at lower costs creates the possibility that we will have more information that can potentially help the patient.  There is also the promise of finding abnormalities that we did not anticipate.  This is what the American College of Medical Genetics calls “incidental findings” and is no different than that seen in many areas of medicine today.  When a doctor sends a person for a chest x-ray, or a chest CT for a cough, unanticipated and incidental findings may be seen in the bones that are visible on the imaging study as an example.  With complete clinical sequencing however the number of secondary and incidental findings may be significantly larger than we see today with other diagnostics. 

In recognition of these emerging facts, the Institute of Medicine hosted a 2 day meeting in July 2012 to help understand the economics of adopting whole genome sequence information into health care.    That led to a report of the meeting and also to a “viewpoint” published in the Journal of the American Medical Association entitled "The Economics of Genomic Medicine" that described the main agreed upon insights by the participants.  In my words, their points were:

  1. Genomic researchers, health care practitioners, payors, and economists are all speaking different languages and have no comfort with the languages of the other disciplines making any discussion of value and economics difficult.
  2. There is little to no evidence at this time demonstrating that genomic data favorably affect health outcomes.
  3. We don’t know how to use and explain this clinically sequenced data to people in any sort of logical, understandable and cost effective manner to produce value.  We just don’t know what to do with these genetic facts so health professionals have a hard time educating and counseling their patients on the implications of the clinical sequencing.  
  4. We don’t understand “personal utility” and its role in assessing the value of sequencing. Personal utility describes the meaning and worth any test or intervention brings to an individual from that individual’s perspective rather than from any external metric such as morbidity or mortality or from an expert’s perspective.

To summarize, we don’t know the value that this clinical sequence has for an individual, we don’t know how to talk about it, we don’t know how to measure its value, and the experts don’t completely understand one another.  Admittedly that makes discussing the economics difficult at best.  It also creates a dilemma for the payors, both private and government, about what should be paid. 

In this country, we have a combined pre-paid health care model and a health insurance model to pay for care.  Insurance is a way of sharing the financial risk of relatively unusual high cost events while pre-paid health addresses the costs of preventive care and screening which are not at all unusual.  Preventive care and screening are encouraged and paid as the value from both a public health and a moral point of view are believed to be worth the cost.  The theory is that finding clinical risks early leads to lower costs of treating disease and thus would also be advantageous for financial risk.  The economic reality has proved to be more complex however and is dependent on the specific clinical issues and the specific use of the information that screening uncovers.  That has led to tremendous debates in the public arena concerning specific tests, such as debates around the use of PSA testing for the risk of prostate cancer and the appropriate age to start screening with mammography for breast cancer.  These debates occur as we attempt to find definite answers to questions that can really only provide us with statistical estimates of both clinical and financial risk.    

Risk is nothing more than a probability based on data.  This is true for both clinical risk and financial risk.  We now routinely treat disease risk, as much or even more than we treat disease.  We screen for silent illnesses such as hypertension that may not be symptomatic to prevent them from causing problems in the future and try to find illnesses at earlier points in their course to affect cures such as screening for breast cancer and colon cancer.   But we don't really know what would have happened in each case had we not screened.  We can only estimate the probabilities of both clinical outcomes and financial outcomes.   

With clinical sequencing, we have entered the potential for a new world of health risk.  Right now, the issue is one of deciding how to report the incidental findings from a clinical sequence.  The American College of Medical Genetics has spent the last year developing a policy statement on just this issue with the draft recommendations now being circulated among members of the college.    Part of their effort is an attempt to identify clinically relevant incidental test results and then recommending whether or not such findings should even be reported to the patient.  But this is just the beginning.  As our understanding improves we will quantify new risk probabilities and thus new test "results" based on the clinical sequencing and the number of incidental findings may multiply. 

I wonder if some of their efforts to control what is reported are in vain and are an outgrowth of the different language they, as experts in genetic diseases, speak when compared to the patients and the public who see the open dissemination of all available information as a human rights issue and a patient autonomy issue, rather than a test reporting issue.  The ability of any health professional or any expert organization such as the ACMG to define what results a patient should see and should not see is likely to be strongly challenged and will probably ultimately be refuted. 

This can lead us to a situation in which the multiplied “incidental findings” which are reported can lead to patient demands to find reasons for these findings and to attempt to modify any risks inherent in these results.  That can only lead to higher total costs with little potential benefit at this point of knowledge.   

In this scenario, the economics become challenging.  Who should pay for this type of attempt at risk mitigation when the likelihood of clinical impact will likely be low for some time to come?  Will it be decided by the American College of Medical Genetics which will look at clinical utility or by the yardstick of “personal utility”?  I am as unsure as the members of the esteemed IOM panel.  While the potential for good is exciting, the potential for increased financial and clinical risk from attempting to predict and cure every possible illness is a bit overwhelming.  We will need to be very careful as we try to address the excellent points made by the Institute of Medicine Committee.