We thank Vallabhaneni and colleagues for sharing their experience of a 14-h shift schedule of interns in a medical ICU. The shift model we described in our article (December 2005)1has the potential to adversely affect the continuity of patient care. The model described in the letter by Vallabhaneni and colleagues provides a partial solution to the discontinuity of care. Although the letter does not include any objective data, the interns’ reporting of satisfaction and feeling of familiarity with the patients’ conditions are promising. We may not be able to develop one model that will fit all teaching programs. The optimal resident staffing of an ICU depends on patient mix and availability of attending physicians and critical care fellows. Ideally, attending physicians should be available 24 h/d and 7 d/wk in the ICU.2 The availability of attending physicians provides an opportunity to provide both good patient care and house-staff education, with more than once-a-day teaching rounds when needed. However, lack of resources and shortage of intensivists may not allow all teaching programs to have qualified intensivists in the ICU 24 h/d to guide both teaching and patient care. The model described by Vallabhaneni and colleagues provides an alternative approach. Since their model requires each intern to be on night duty only once every fourth day, its impact on sleep hygiene may be better tolerated than the model described in our study. However, when we introduce a new house-staff/patient care model in an ICU, we need to measure its impact on patient outcome, clinician satisfaction, sleep hygiene, and education. Such an approach may help educators and clinicians to identify the best model.