Are patients really “worse off” with older docs?

In what appears to be a largely counter intuitive result, research by the American Journal of Medicine has unearthed that patients end up faring worse when treated by older doctors or more experienced doctors. This Reuters article delves into the issue and discusses several aspects of the study in great lengths. This study has also wracked up a heated discussion on the Evidence Based Healthcare email list on JISCmail, where the members do not seem to be overly shocked by the findings.

Here are my two cents:

The article compares two aspects: length of stay and in hospital mortality as markers for quality of care dispensed. While this seems logical on the first view, there are several caveats:

• Length of stay is a culturally determined variant. The focus of training now is to maximize the resources, minimize the patients’ length of hospital stay and hence has led to a system of aggressive discharging by the younger doctors.

• The study not being a longitudinal one, does not look at the variation of LOS over time. I believe it is an important factor in determining if there were any other confounding factors at the time of the study that may have skewed the results. Since the physician sample size is rather small (59) there is a chance that might have happened. If findings showing that the average LOS for every group of physician was constant over time, it would help in giving more credence to the findings.
• The fact that mortality is higher under the older doctors is also not a very good marker. Having worked in a high mortality unit (intensive care in a developing world set up always is a heart breaking experience) I know that death is not always a by product of a physician’s ineptitude. The study does not really talk about the quality of death and the patient/family satisfaction around this issue.
• What about out of hospital mortality? Since the younger doctors are discharging patients aggressively, are they causing more out of hospital mortalities? Sadly this study does not do that follow up. It would indeed be an interesting point to investigate.

The study found in post hoc analysis that there was no statistically significant association between more years in clinical practice and mortality in less medically complicated cases, and the significance was observed only in case of more complex cases. This indicates that the older doctors are more liable to have poorer outcomes than their younger counterparts. A possible explanation for that maybe attributed to the fact that the younger doctors today are trained to treat more aggressively (at least that is the case in the Indian scenario, as far as my experience goes) than the older doctors. While this is merely speculative on my behalf, the study authors have forwarded some more reasons for this:

The authors content that the older physicians have worsening skills and eroding knowledge. While that may well seem to be a natural thing, one must keep in mind the fact that the study was conducted in a teaching medical hospital. That means even for the patients admitted under the senior doctors, the junior staff had significant role to play in dispension of medical care. There are different kinds of attending physicians – ranging the whole spectrum from the completely absent to the extremely nosy ones. No quantitative study can control for the styles and it seems it would be illogical to class all the different styles of patient care under the same head merely on the basis of their number of years in clinical practice.

Most of the members of the EBHC mailing list were of the opinion that the younger doctors were more up to date, more able to synthesize and implement the best evidence and hence dispense more efficient medical care for their patients. This seems to be a rational explanation, especially since the oldest of the lot are exempt from recertification.

However, if there were some solid evidence, which followed doctors up over time and thus helped to measure their eroding clinical skills and knowledge, it would make more impact than just headline hitting research like this.

Take a look at the study design:

EBMers cry themselves hoarse at the quality of data originating from studies which are not Randomized Controlled Trials (which this is not, hence, the quality of the results are obviously questionable). However, what I also want to point out is that each team consisted of one senior attending and the rest were just the brigade of the younger doctors. And as anyone who has been within a mile of a hospital knows, not all the decisions in the management of the patient are taken by the attending. The thing is, most of them actually are not. And definitely, when it comes to doing the procedures (which are an important source of in hospital LOS), they are generally hands off.

While it would be unreasonable to expect that this kind of micro-data-analysis would be possible in a quantitative study as this one, it still leaves some room for confounders… hence some room to debate the findings.

While the study definitely throws up some really interesting information and opens up newer questions for investigation, there remains enough room for doubt in order to not go gaga over the results and take stern steps. The fact that it was done on just 59 physicians taken from a single institution also does not help in ascertaining the generalizability of the study.

However, this does make a case for the fact that though in the first year of medical school we are handed an important bit of advice by our grizzled professors – that we shall remain students life long – not all physicians seem to have attended that class!

References:

1. Schulz E. Patients worse off with more more-experienced docs? Reuters. July 15, 2011. http://www.reuters.com/article/2011/07/15/us-patients-worse-off-idUSTRE76E62T20110715

2. Southern, W., Bellin, E., & Arnsten, J. (2011). Longer Lengths of Stay and Higher Risk of Mortality among Inpatients of Physicians with More Years in Practice The American Journal of Medicine DOI: 10.1016/j.amjmed.2011.04.011