We all know that American healthcare technology is expensive. But is it worth it?
The October 2013 issue of Contemporary OB/GYN features a fascinating technology update. I would be remiss if I did not comment on the value added to our healthcare system by new technologies.
In a recent New York Times blog, Princeton University health care economist Professor Uwe Reinhardt noted a paradox of modern US health care: It is both high-value and alarmingly wasteful.1 (http://economix.blogs.nytimes.com/2013/09/13/waste-vs-value-in-american-health-care/) How can both statements be true?
There can be no argument that medical advances from penicillin to percutaneous coronary angioplasty have dramatically improved both longevity and quality of life-providing great value to society. However, in an era when every week seems to bring the announcement of another hugely expensive biological therapy or more sophisticated imaging technique, one must ask whether we are seeing the same incremental improvement in quality of life for each incremental dollar spent today that we did 50 years ago.
In other words, are we getting the same “bang” for our healthcare “buck?”
Cutler and colleagues examined this very issue by studying medical spending from 1960 through 2000 and comparing the cost of care to the resultant gains in life expectancy in that period.2 During those 4 decades, overall lifetime healthcare spending climbed nearly 6-fold, from $14,000 to $83,000 per person. But among Americans 65 years and older, who account for most healthcare costs, spending increased more than 13-fold. So what did we get for our money? The answer seems to be diminishing returns!
Assuming that 50% of the observed improvement in life expectancy that the authors documented accrued to medical advances, the average cost per year of life gained at age 65 rose from $75,100 between 1960 and 1970 to $145,000 between 1990 and 2000. Thus, the real “value” (outcome/cost) of medical advances appears to be falling. And my guess is that this unfavorable trend has accelerated greatly over the past 13 years.
Reinhardt notes that plotting increases in quality-adjusted life years (QALY) gained versus per capita health care spending generates a parabolic curve. Thus, the rate by which QALY increase for a given incremental medical cost rapidly accelerates (efficient care), then slows and levels off (inefficient or potentially wasteful care) and finally begins decreasing (unambiguously wasteful care). Reinhardt contends that compared to the past, today substantial additional costs are incurred for very modest QALY gains. He questions whether this incremental cost is worth it or is wasteful.
Reinhardt also points out that at its extreme, where there is a decline in QALY with increasing costs (unambiguous waste), patient harm results from improper care (eg, unnecessary surgery or imaging). According to the Institute of Medicine, waste attributable to unnecessary services and inefficiently delivered care accounts for close to 14% of total healthcare costs, or about $340 billion per year.3 Thus, it is incumbent on us to assess new technology from the perspective of its true value to society: whether it significantly improves QALY for a reasonable cost.
Of course, when it’s you who may marginally benefit from a new diagnostic test or treatment and if you are not directly paying for its high cost, maintaining this intellectual detachment can be difficult.
To give a concrete example of the “Reinhardt curve” in action and the related moral hazard and economic complexity of moving along the curve from left to right, let’s look at 2 examples. When the polio vaccine was introduced 50 years ago, it was quite inexpensive per dose and had an extraordinary public health impact. In fact, even now it costs only $56 per dose.4 By contrast, today the most expensive prescription drug in the world is Idursulfase, an enzyme replacement therapy for patients with Hunter syndrome, that in 2008 was estimated to cost an average of $491,999 per year.5
The first time I saw a robotic hysterectomy, I thought, if this had existed when I was a resident, I never would have become a perinatologist! The images were so clear, hand movements so precise, and ergonomics so improved, that it seemed like a different world. Many people felt the same way. Now, according to the manufacturer of the da Vinci system, more than 2000 hospitals have the machine worldwide and 450,000 procedures were performed in 2012.6 Indeed, it has the potential to become the great leveler of surgical skills, because one no longer needs to be an expert laparoscopist to perform moderately sophisticated minimally invasive surgery. But does it add value to women’s health care?
While most reports comparing traditional laparoscopic to robotic-assisted gynecological surgeries are by subspecialists (eg, gynecologic oncologists), largescale direct comparisons of the 2 modalities employed for benign gynecological surgery show little advantage to the robot. Pasic and colleagues mined the Premier hospital database to review the records of 36,188 patients who underwent minimally invasive hysterectomy in 358 hospitals.7
While the vast majority (95%) of cases were traditional laparoscopic hysterectomies, when compared with those performed with robotic assistance, use of the robot was consistently associated with statistically significant higher per-patient average hospital costs ($9640 [95% CI, $9621 to $9659] vs. $6973 [95% CI, $6959 to $6987]). Furthermore, both inpatient and outpatient surgery times were significantly longer for robot-assisted procedures.
A rather limited Cochrane database analysis concluded that “limited evidence showed that robotic surgery did not benefit women with benign gynaecological disease in effectiveness or in safety. Further well-designed RCTs with complete reported data are required to confirm or refute this conclusion.”8
Of course, such studies do not measure the potential societal benefits of having many gyn surgeons move directly from laparotomy-based to robotic-assisted hysterectomies, thus avoiding large abdominal incisions, wound breakdowns, thromboembolism, lost work time, etc. Moreover, the value of robotic-assisted hysterectomy in morbidly obese patients may also be understated.
On the other hand, the art of vaginal hysterectomy, which confers many of the same advantages as the robot, is slowly being lost. Evidence also exists that the incidence of complications surrounding robotic-assisted hysterectomies is higher than commonly appreciated.
Cooper and associates reviewed various governmental and public device-related complication databases and court records to identify robotic surgery-related complications during a 12-year period and noted that 245 events were reported to the FDA, including 71 deaths and 174 nonfatal injuries, with 5 additional cases identified from other databases.9
Robotic-assisted hysterectomy is likely equipoised on the “Reinhardt curve” between a clearly efficient, value-adding medical advance and an inefficient and potentially wasteful procedure. Further studies of unintended potential societal benefits accruing to its use (“positive externalities,” in economics-speak) are clearly needed to better define the robot’s precise location on the cost-benefit curve.
For example, does it reduce global complication rates by eliminating laparotomies? Does it dramatically reduce adverse outcomes in morbidly obese patients? What are the national economic benefits of the resultant shorter recovery time?
Regardless of the findings of such comparative effectiveness studies, in my opinion, reduced robotic surgery operating costs-due to the introduction of competitors and alternative disruptive innovations- would move it to the left on the Reinhardt curve.
However, the case of robotic-assisted hysterectomy is just one example of the kind of analyses needed to assess the value of new technologies. This type of scrutiny will be increasingly employed by large healthcare systems and insurers as well as by you and your practice partners as healthcare reimbursement moves to global payments and capitation.
Do you really need to buy an office ultrasound with 3-D capabilities? Is remote access to a fetal heart rate tracing needed when your hospital employs “laborists”? Is it worth using the latest anti-ovarian-cancer agent that costs $10,000 per month when, on average, it extends life by only 45 days? And at a more prosaic level: Do you really need to prescribe azithromycin for your patient’s head cold?
Dr Lockwood, Editor in Chief, is Dean of the College of Medicine and Vice President for Health Sciences at The Ohio State University, Columbus, Ohio.
1. Reinhardt UE. Waste vs. value in American health care. http://economix.blogs.nytimes.com/2013/09/13/waste-vs-value-in-american-health-care/. Accessed September 16, 2013.
2. Cutler DM, Rosen AB, Vijan S. The value of medical spending in the United States, 1960-2000. N Engl J Med. 2006;355(9):920-927.
3. Smith M, Saunders R, Stuckhardt L, McGinnis JM, eds. Best care at lower cost: the path to continuously learning health care in America. Washington, DC: The National Academies Press; 2013: Table S-1, page 13.
4. Healthcare Blue Book. Polio vaccine. http://healthcarebluebook.com/ page_Results.aspx?id=13&dataset=CC. Accessed September 14, 2013.
5. Burrow TA , Leslie ND. Review of the use of idursulfase in the treatment of mucopolysaccharidosis II . Biologics. 2008;2(2):311–320. http://www. ncbi.nlm.nih.gov/pmc/articles/PMC2721351/Accessed September 17, 2013.
6. Intuitive Surgical. Company profile. http://www.intuitivesurgical.com/ company/profile.html. Accessed September 16, 2013.
7. Pasic RP, Rizzo JA, Fang H, Ross S, Moore M, Gunnarsson C. Comparing robot-assisted with conventional laparoscopic hysterectomy: impact on cost and clinical outcomes. J Minim Invasive Gynecol. 2010;17(6):730-738.
8. Liu H, Lu D, Wang L, Shi G, Song H, Clarke J. Robotic surgery for benign gynaecological disease. Cochrane Database Syst Rev. 2012 Feb 15;2:CD008978. doi: 10.1002/14651858.CD008978.pub2.
9. Cooper MA, Ibrahim A, Lyu H, Makary MA. Underreporting of robotic surgery complications. J Healthc Qual. 2013 Aug 27. doi: 10.1111/ jhq.12036. [Epub ahead of print].