The Success Factors for Hip Replacement Surgery
Dr. Khaled Saleh, Kamalini Ramdas, and Steven Stern
Posted 01/14/09
Dr. Khaled Saleh
Photo by Tom Cogill
Everyone who watches a hip replacement operation says the same thing: it looks a lot like carpentry. Certainly, there is a superficial resemblance. Both carpenters and surgeons now use cordless drills and saws, and they both exhibit a certain workmanlike pride in their results. The point of the comparison, however, is really to underscore their difference. Carpentry is much more forgiving. Put a wall in the wrong place and you can pry it up and move it. In hip surgery, the stakes are much higher. An error of judgment, a lapse of skill, a moment of inattention, and the result can be a long and painful recovery for the patient, who may never walk properly again.
An appreciation for the exacting nature of his profession is one reason that Dr. Khaled Saleh, an orthopaedic surgeon at the U.Va.
Saleh’s drive to understand the factors that lead to successful replacement surgery is the reason he has joined Darden School of Business professor Kamalini Ramdas and economics professor Steven Stern in their efforts to apply well-established statistical modeling tools to surgical practice.
Ramdas researches how companies can best manage product variety, addressing such questions as the effect of standardization on quality. For instance, a car manufacturer considering adopting a single braking system across its entire fleet must balance a number of issues. Using a single system would increase reliability because it would be easier to ensure that the brakes are manufactured and installed to specification. In addition, producing a single system in volume would entail considerable cost savings. On the other hand, a standardized system might not be the best match for a specific model. Ramdas helps manufacturers determine when variety trumps standardization.
Ramdas’ original impulse was to ask similar questions about the medical devices used in hip operations. What were the considerations that might lead doctors to use a standardized set of devices for all procedures? How did familiarity with devices from the same manufacturer affect outcomes? She recruited Stern for the project because of his expertise in modeling and estimating learning effects.
The team soon had to alter its goals. “We were looking for very subtle differences,” notes Stern. “To measure them with precision would have required us to have a lot more observations than we can gain from looking at hip surgeries at U.Va. alone or a much higher incidence of bad outcomes, like longer surgeries or the need for repeat surgery.” Instead, they are trying to identify the characteristics of surgery or medical hardware that affect the likelihood of a poor outcome. “Rather than determining if a component is the best one for a patient in the long run, we are looking at questions such as how the choice of a particular device affects the duration of a surgery,” says Ramdas. Duration is important because longer surgeries are more likely to result in infection.
To apply their models, Ramdas and Stern have had to find ways to make the data easier to manage. For instance, each of the major manufacturers of medical equipment used in hip replacement operations offers several distinct product lines, with many size variations. With Saleh’s guidance, they have determined which devices have enough in common that they can be considered as a single type. They have also had to integrate data from different sources that include patient charts and electronic records at the U.Va.
The question of surgical outcomes for hip replacements is hardly academic. As Dr. Saleh notes, they are costly procedures that are increasing in number rapidly as the population ages, but they also have outsized benefits. “From the point of view of quality of life, joint replacement is a high-impact procedure,” he says. “It restores patient integrity and reduces dependence on others. It’s important that we understand all the factors that contribute to the best possible outcomes.”