In reply

Abstract

The letter by Dr. McCormick addresses several issues. The most important of these relate to the sentence in our paper's introduction about simplification of management's strategies among patients with dementia of the Alzheimer type (DAT). This sentence concerns our initial hypothesis, not our conclusion. In our work,1 about 40% of the patients with DAT had four or more associated diseases. Even if the prevalence of associated diseases is slightly lower in patients with DAT than in patients without DAT, it is still high. We stress that point in our conclusion by underlining the high mortality rate of patients with DAT and the questionable external validity of clinical trials where only relatively healthy patients are included. The validity of our study is extensively discussed in our paper. The data recently published by Dr. McCormick add another source of bias to the numerous ones that can be observed in cross-sectional or retrospective studies: underreporting of symptoms by patients having DAT. In our work, this bias could be less than suggested by McCormick, because most diseases are diagnosed after a complete examination of the patients by geriatricians, not only after spontaneous complaints of patients. Concerning the so-called selection bias, the question is not whether our patients are representative of the general population of patients having DAT. They are obviously not because they have been referred in long-term care units, and they are probably more severely ill than patients not admitted in such units. Even in this circumstance, patients with DAT had a slightly lower prevalence of associated diseases than nondemented patients. However, our results could be biased if the selection process leading to hospitalization was different for nondemented patients and demented patients. We cannot rule out this possibility. Finally, a prospective cohort study would be the only appropriate design to assess definitely if DAT is a condition protecting from other diseases. In the meantime, we share the views of Dr. McCormick. Even if slightly lower than among demented patients, the prevalence of associated diseases is quite high. Both DAT itself and comorbid conditions justify the presence of full medical and paramedical staff in geriatric units.