Surgical Site Infection as a Surrogate Marker of Physician Impairment
Our report details an implant‐associated outbreak of surgical site infections related to the adverse effects of treatment for hepatitis C virus infection administered to surgeon X. During the 12‐month period of this outbreak, 14 (9.5%) of 148 of surgeon X's patients developed a surgical site infection, a rate of SSI that was 8‐fold higher than the rate during the 14‐month baseline period or the 14‐month follow‐up period (
), and higher than the rate among peer surgeons (
).
Received April 14, 2009; accepted June 8, 2009; electronically published September 25, 2009.
Physician impairment is a long‐recognized problem that has come under increasing scrutiny.1‐6 The American Medical Association defines physician impairment as any physical, mental, or behavioral disorder that interferes with the ability to engage safely in professional activities.7 It has been estimated that 5%–15% of physicians in the United States have some sort of impairment.4 Patient complaints, a possible marker of physician impairment, are definitely associated with a greater frequency of malpractice claims and adverse outcomes.1,2,5 The most common reported causes of physician impairment are substance abuse or psychiatric illness.4 The importance of a physical illness sufficient to restrict practice, which affects at least 10% of physicians,4 is very poorly understood.
In our report, we detail the impact of the disabling adverse effects of 1 year of interferon (IFN)–α therapy received by a surgeon who performed implant procedures. It is likely that the excessive number of nosocomial infections represented a surrogate marker of physician impairment.
Methods
Our infection control department previously investigated an outbreak of implant‐associated surgical site infection (SSI) at hospital A. This investigation included a 14‐month baseline period, a 12‐month outbreak period (ie, the period of treatment for hepatitis C virus infection for surgeon X), and a 14‐month follow‐up period. The cases of SSI were prospectively assessed during the entire 40‐month period using the Centers for Disease Control and Prevention criteria. An assessment of surgeon X’s general health was done by conducting interviews with operating room personnel. The rates of SSI were determined, for surgeon X and for the other surgeons who performed the same implant procedures, on the basis of the date of onset of infection and the date of surgery, to eliminate the potential effect of a long incubation period for implant‐associated infections. Rates were compared by use of the 2‐tailed Fisher exact test.
Results
During a 40‐month period, the rates of SSI associated with surgeon X were compared with the rates of SSI associated with the other surgeons who performed the same implant procedures at the same healthcare institution (hospital A) as surgeon X (Figure). During the 14‐month baseline period, surgeon X performed implant surgery on 294 patients, 3 (1%) of whom developed an SSI; this rate of SSI was not significantly different from the rate of SSI associated with the other surgeons at hospital A (Table). During the 12‐month outbreak period, implant surgery was performed on a total of 275 patients, 14 (5.1%) of whom developed an SSI as assessed by date of onset of infection and 17 (6.2%) of whom developed an SSI as assessed by date of surgery. Compared with the 14‐month baseline period, the SSI rate associated with surgeon X increased 6‐fold to 7.4% (relative risk, 3.6; 
) when assessed by date of onset of infection or 8‐fold to 9.5% (relative risk, 4.5; 
) when assessed by date of surgery. During the 12‐month outbreak period, the SSI rate associated with surgeon X and assessed by date of surgery was significantly greater than that associated with the other surgeons and assessed by date of surgery (9.5% vs 2.4%; 
). Compared with the 12‐month outbreak period, the SSI rate associated with surgeon X decreased during the 14‐month follow‐up period to 3.0% when assessed by date of onset of infection (95% confidence interval, 0.9–7.2; 
) and to 1.2% when assessed by date of surgery (95% confidence interval, 1.8–34; 
). During the 14‐month follow‐up period, the SSI rate associated with the other surgeons was 1.2%, identical to the SSI rate associated with surgeon X.
Figure Rates of surgical site infection (SSI) associated with surgeon X were compared with the rates of SSI associated with the other surgeons who performed the same implant procedures at the same healthcare institution (hospital A) as surgeon X. This investigation included a 14‐month baseline period, a 12‐month outbreak period (ie, the period of treatment for hepatitis C virus infection for surgeon X), and a 14‐month follow‐up period.
During the 12‐month outbreak period, there were no significant differences in the microbiology results between cases of SSI associated with surgeon X (5 cases of SSI due to methicillin‐resistant Staphylococcus aureus, 3 due to methicillin‐susceptible S. aureus, 3 due to coagulase‐negative staphylococci, 2 due to gram‐negative organisms, and 1 with a culture‐negative test result) and cases of SSI associated with the other surgeons (1 case of SSI due S. aureus, 1 due to a diphtheroid organism, and 1 with a culture‐negative test result). Prophylactic antibiotics were administered correctly 75% of the time by all surgeons. The duration of surgical procedure did not change for surgeon X. The incubation period for the development of SSI associated with surgeon X had a mean duration (± standard deviation) of 69 ± 85 days during the outbreak period and a mean duration of 28 days during the other 2 periods. The incubation period for the development of SSI associated with the other surgeons had a mean duration of 23 days during the outbreak period and 31 days during the other 2 periods.
Initial detection of the outbreak was obscured for several reasons. First, SSI data were routinely reported only for groups of surgeons. This effectively doubled the denominator and made the increase in cases of SSI much less evident. Second, the long incubation period cited above shifted the detection of cases up to 8 months in time. Because we routinely report our cases of infection by date of onset of infection, this also obscured the detection of the problem until the data were reanalyzed by date of surgery.
Interviews with operating room personnel revealed that surgeon X frequently appeared ill during the 12‐month outbreak period. It was reported that surgeon X was sufficiently ill to require lying down on the operating room floor on more than one occasion and was known to be anemic. Surgeon X denied being ill, and his superiors denied his illness as well, until the analysis was complete. By this time, his IFN‐α therapy was complete, and his visible signs of illness resolved. To our knowledge, no transmission of hepatitis C virus occurred, and no other adverse events were identified.
Discussion
In our report, a surgeon receiving treatment for hepatitis C virus infection was associated with an outbreak of SSI. Reports of outbreaks of SSI have been described elsewhere and usually involve a single organism being transmitted by a surgeon or other operating room personnel, which was clearly not the case for this outbreak, as shown by the microbiology results. This outbreak was associated with a surgeon who was symptomatically ill while receiving IFN‐α therapy; this outbreak stopped immediately after the therapy was stopped and symptoms resolved. This strongly suggests that the symptoms were secondary to the IFN‐α therapy. This type of therapy, which is used to treat hepatitis C virus infection, can cause flulike symptoms severe enough that some patients are bedridden, have difficulty concentrating, are confused, and/or become depressed.8 Any of these adverse effects could have been sufficient to adversely affect the surgeon’s performance. Thus, the increase in the rate of SSI associated with IFN‐α therapy was likely a surrogate marker for physician impairment. To our knowledge, there have been no previous reports in peer‐reviewed medical journals of physician impairment leading to an increased frequency of any kind of hospital complication, outside of lawsuits.
An initial question is whether the lack of reports of patient adverse outcomes secondary to physician impairment means that such events rarely occur. This seems unlikely for several reasons. First, physician impairment is common, with a prevalence of 5%–15% and a lifetime risk of up to 33%.4 Second, there is a clear linkage between adverse outcomes, malpractice lawsuits, and physician impairment.1,2,5 If patient adverse outcomes secondary to physician impairment occur with a significant frequency, then it must be concluded that they are underreported. It is well known that healthcare workers are reticent to report other healthcare professionals that have problems with substance abuse.9 With the additional concerns of professional censure, public humiliation, and lawsuits, there are currently huge disincentives for healthcare workers and healthcare institutions to report an impaired physician. It is our belief that there has been systematic nonreporting of such adverse events for such a long time that it has become one of the sacred cows of medicine.
This underreporting suggests that the administrative system that oversees physician activities should have the responsibility of dealing with impaired physicians.4,10 For such a system to work, it would need to have certain key elements, as has been nicely detailed by Leape and Fromson.4 These elements include having national standards of physician performance, systematic methods for detecting failing physicians, and programs for remediation of such physicians.4 We need to find ways to detect physician impairment and deal with it in a manner that minimizes the adverse impact on both the physician and the patient. Much work remains to be done before we can accomplish this in a satisfactory way.
Acknowledgements
We thank Gerald Smith, Joe Tobin, and Ray Roy for their thoughtful suggestions about this manuscript.
Potential conflicts of interest. Both authors report no conflicts of interest relevant to this article.
References
- 1. Hickson GB, Federspiel CF, Pichert JW, Miller CS, Gauld‐Jaeger J, Bost P. Patient complaints and malpractice risk. JAMA 2002; 287:2951–2957.
- 2. Murff HJ, France DJ, Blackford J, et al. Relationship between patient complaints and surgical complications. Qual Saf Health Care 2006; 15:13–16.
- 3. Papadakis MA, Teherani A, Banach MA, et al. Disciplinary action by medical boards and prior behavior in medical school. N Engl J Med 2005; 353:2673–2682.
- 4. Leape LL, Fromson JA. Problem doctors: is there a system‐level solution? Ann Intern Med 2006; 144:107–115.
- 5. Morrison J, Wickersham P. Physicians disciplined by a state medical board. JAMA 1998; 279:1889.
- 6. Peisah C, Wilhelm K. Physician don’t heal thyself: a descriptive study of impaired older doctors. Int Psychogeriatr 2007; 19:974–984.
- 7. American Medical Association (AMA). Code of medical ethics. Opinion 9.031—reporting impaired, incompetent, or unethical colleagues. Available at: http://www.ama‐assn.org/ama/pub/physician‐resources/medical‐ethics/code‐medical‐ethics/opinion9031.shtml. Accessed September 22, 2009.
- 8. Dieperink E, Willenbring M, Ho SB. Neuropsychiatric symptoms associated with hepatitis C and interferon alpha: a review. Am J Psychiatry 2000; 157:867–876.
- 9. Baldisseri MR. Impaired healthcare professional. Crit Care Med 2007; 35(2 Suppl):S106–S116.
- 10. Brennan TA. Physicians’ professional responsibility to improve the quality of care. Acad Med 2002; 77:973–980.