Although natural disasters regularly disrupt US medical education, the influence of such events on student career choice is unknown. In August of 2005, Hurricane Katrina forced both New Orleans medical schools—Tulane and Louisiana State University Health Sciences Center (LSUHSC)—to relocate from their downtown campuses for nearly a year (1). New Orleans’ subsequent, persistently increased rate of mental illness has been well described (2).

Anticipating the effect of natural disaster on medical education is an important public health concern, given the effect of career choice on future physician supply. Previous work has suggested that students are motivated to enter psychiatric training in part because of perceived demand and a desire for patient advocacy (3). Thus, we hypothesized an increased student interest in psychiatry after Katrina.

At LSUHSC, we compared medical student choice of residency type among Psychiatry and three other specialties: Emergency Medicine (EM), Internal Medicine, and General Surgery, as well as student willingness to remain in Louisiana for residency training.

We surveyed student career choice in the 5 years before and after Katrina (2001 through 2010), using information provided by the LSUHSC School of Medicine Dean’s Office. We also reviewed data included in the annual Association of American Medical Colleges Medical Student Graduation Questionnaire (GQ) from the same period. Students complete the GQ online, voluntarily and confidentially, and must submit their responses no later than mid-June of their graduating year. Our data were examined with chi square and linear-correlation analyses in SPSS Version 15 (SPSS, Inc.; Chicago, IL).

Post-Katrina, alone among the four specialties, Emergency Medicine recruitment increased with statistical significance; 49 students chose Emergency Medicine in 2001–2005, whereas 80 did so in 2006–2010 (χ²[1]=8.86; p <0.005). By contrast, pre- and post-Katrina numbers in other specialties were similar: Psychiatry, 29 versus 32; Internal Medicine, 150 versus 142; Surgery 126 versus 117.

Comparing the two pre- and post-Katrina years, significant differences were observed between recruitment into Emergency Medicine (r=0.845) and Psychiatry (r = −0.899; z=1.91; p <0.05), but not in Internal Medicine (r=0.039) or Surgery (r = −0.028), either directly compared or among all four (ρ[3] = −0.028; NS; see Figure 1). Furthermore, for the entire 10-year study period, LSUHSC students chose Emergency Medicine more frequently than at other United States schools (z=3.14; p <0.001), whereas no difference was observed for other specialties.

More students remained in Louisiana for residency training in Emergency Medicine, versus Psychiatry, in the 4 years pre- and post-Katrina (Figure 2): 13 and 6 in Psychiatry; 17 and 35 in Emergency Medicine, respectively (F[3]=6.7; p=0.01). We observed no significant trends regarding location choice in General Surgery or Internal Medicine training.

Using AAMC Senior Survey data, we compared the 2 years most removed from Katrina: 2001–03 and 2009–2010. Psychiatry and Internal Medicine had significantly more students report the quality of their clerkships as “excellent” (pre-/post-: Psychiatry: 21.0% versus 39.7%; t[2] = −12.58; p <0.005; Internal Medicine: 50.1% versus 77.6%; t[2] = −9.74; p <0.05), but this change was not significant for Emergency Medicine or Surgery.

Contrary to our expectations, fewer students chose psychiatry residency training after Katrina, and those who did were more likely to train out of state. This occurred despite positive regard for the psychiatry clerkship itself. Rather, Emergency Medicine—alone among the specialties studied—became not only more popular, but also more successful at retaining students in Louisiana.

Both the increased popularity of emergency medicine and diminished interest in psychiatry may have been a direct result of Katrina’s effect on healthcare infrastructure. Although LSUHSC’s inpatient psychiatry beds remained closed for more than 2 years, their emergency department reopened more quickly (4). With many primary-care clinics flooded and closed, some still so, a diversity of patients were forced into the emergency department. Some students may have chosen emergency medicine simply because they could see first-hand that its services were in high demand. Furthermore, the fact that it was available at all suggested it was a more valuable specialty than psychiatry. Absent the ability to survey these students, however, we can only speculate about what motivated their career choice.

These data may be difficult to generalize. Natural disasters are similar only in that they disrupt, to varying degrees, the normal delivery of health care and the medical education that accompanies it. Prospective data, which were not available to us here, would be more useful for predicting how disaster might affect a “typical” medical student. Moreover, we were not able to compare our observations with those of others—for example, University of Texas Medical Branch at Galveston’s more recent experience with Hurricane Ike (5)—because specific data about medical student career choice are not available. In order to be able to anticipate post-disaster conditions, at-risk schools should have easy access to relevant information from schools that survived catastrophic events. Unfortunately, the medical literature is scanty.

Nevertheless, the disparate fortunes of psychiatry and emergency medicine in New Orleans are striking. Nationally, the increasing popularity of emergency medicine has been attributed to the “controllable lifestyle” that it and other fields—radiology, dermatology, ophthalmology, pathology—provide (6). Students assume these specialties will allow them to better choose the circumstances under which they practice. If so, in those first, difficult years after Katrina, Louisiana students may have determined that psychiatry could not be left entirely “at the office,” and, instead, trained out of state in other fields. Regardless of their motivation, our study did not show that an increase in medical student experience with trauma-related illness predicts an increase the number of students choosing psychiatry.