A Critical Issue in Health Care Reform

By Gail Garfield Schwartz

From Advocacy News & Views

Health care conferences and health care websites are abuzz these days with conversations about Comparative Effectiveness and Comparative Effectiveness Research (CER). 

Simply put, Comparative Effectiveness means an assessment of the benefits and harms of different treatments or diagnostic processes for the same medical problem. The goal is to find out, in the real world, what works best. Then standard of care can be based on actual results, accurately counted and added up across many hospitals and clinics--not just a physician's favorite plan of attack, or a patient's uninformed preference. Then physicians will be more apt to offer best and most valuable care. 

And if the results of Comparative Effectiveness studies are published and broadcast in plain language, patients will be helped to make wise decisions about their own care.

Yet this good idea is not nearly as simple as one would like. First, Comparative Effectiveness studies can cover a whole host of specifics that make up the clinical value of services:  diagnosis; selection of surgery; radiation; selection of drugs or biologicals; other non-surgical interventions; sequencing of combination and multi-staged treatments.  Even care givers' behavior can be a topic of study.    

Second, outcomes don't necessarily reveal themselves quickly.  Diseases like cancer, especially breast cancer and ovarian cancer, present many complexities arising from the biological diversity of the disease. Each patient's whole health history and genetic characteristics may have to be taken into account so as not to bias outcomes for these diseases.  Moreover, many different combinations of treatments may be required over time, and it may not be easy to pinpoint which is the "best," although it may be easy to show which are the worst.

Nonetheless, a great many studies of Comparative Effectiveness have already been undertaken in the United States. Their recognized value in hospitals and clinics has led public policy makers to launch a huge effort to improve and increase Comparative Effectiveness studies. The 2009 American Recovery and Reinvestment Act plan slotted $1.1 billion for CER. This year's health care reform legislation (Patient Protection and Affordable Care Act) set up a new organization to synthesize results of existing and new studies of Comparative Effectiveness into a more useful knowledge base. This independent organization, called the Patient-Centered Outcomes Research Institute (PCORI, pronounced Picori) will have half a billion dollars to spend.   

The main focus of PCORI will be on Comparative Clinical Effectiveness (CCE). CCE is measured by usual end points of clinical trials: Additional survival time, and/or quality of life after treatment. These end points can be compared among treatments using a standard score for outcomes. To quantify CCE, clinical researchers have married their methods to economists' techniques, and come up with an objective measurement: Quality-Adjusted Life Year (QALY). 

QALY is a measure that can translate objective and subjective descriptions of health status into a single number. It takes harmful side effects into account.  "LY" represents the extra months of life a person might gain as a result of a treatment. "Q" represents the quality of that life. Q is defined as "reasonable comfort," and factors included in assessing Q might be pain, mobility, heart condition, condition of other organs, and emotional state.  The Q rating can range from 1 signifying best possible health, to -1, signifying worst possible health, with 0 indicating no change in quality of life. 

The QALY score is LY multiplied by Q. Using QALY, the longest expected survival time does not necessarily match with the best quality of life. To oversimplify by way of explanation, take chemotherapy. Person X can take drug "cocktails" A or B.  Each cocktail has side effects, some of which are very debilitating, even potentially life threatening.   

Patient X taking A can be expected to live for at least three years compared to no chemotherapy.  But A has serious side effects, sending Q down to .5.  X's expected QALY with A is 3x.5= 1.5

With B, X can be expected to live for at least 2 years and there are no seriously harmful side effects so her Q will be .9.  With B, X's expected QALY is 2x.9=1.8.

 X would be better off with B.  This shows how Comparative Clinical Effectiveness measured by QALY can take a lot of confusion out of decision making.

But such Comparative Clinical Effectiveness alone can not give a complete answer to the question:  For a given medical condition, what's the best thing to do from an individual's viewpoint and from society's viewpoint?  Sure, if a given clinical practice is found to be useless or deadly (as for example stem cell transplants for breast cancer were), it should be scrapped as soon as the information is available.  But if two treatments have almost the same positive outcomes, how to choose one of them? 

The answer is, by knowing which is the most Cost Effective. If two clinically effective treatments have the same outcomes, but greatly different costs, why would the more expensive treatment be justified?   

To evaluate cost effectiveness, numerous formulas exist. The basic approach is to divide cost by QALY to get cost per QALY. The result is Comparative Cost Effectiveness.  (Researchers generally use incremental QALY and incremental cost, but that's a nicety we need not bother with here). 

When Comparative Cost Effectiveness and Comparative Clinical Effectiveness are combined, the result might most conveniently be termed Comparative Effectiveness Overall.

Comparative Cost Effectiveness and Comparative Effectiveness Overall come into play powerfully where one treatment (or diagnosis) is shown by research to be "best" in terms of clinical effectiveness, but worst in terms of cost effectiveness. Then this treatment's cost clearly becomes a prime target for cost reduction measures. This situation is given sharp focus in a study at Columbia University by Victor Grann, a doctor and public health expert with a long track record in comparative effectiveness research. 

Dr. Grann and colleagues reviewed 25000 reported outcomes of treatments for patients who had the BRCA gene. The study showed that prophylactic (preventive) mastectomy is the most cost-effective treatment to prevent breast cancer in those women. But many BRCA carriers don't want surgery; they would rather "watch and wait." The research showed that MRI screening with mammography had the highest QALY survival, but cost about 100 times the cost of surgery.  The researchers reached this practical conclusion: If the cost of MRI could drop by more than 50 % it would be the most cost effective and clinically effective option--that is, the most comparatively effective overall.  (See CancerNetwork.com August 10, 2010).   

Many such conclusions no doubt lurk in the depths of data. Advocates should support digging them out, because such objective knowledge is good for individuals and for society.  But many demands will be made for research dollars.  As representatives of so many women at risk, SHARE advocates can help ensure there's a payoff for people diagnosed with breast cancer or ovarian cancer.  

Posted September 16, 2010.