Ouslander and colleagues [1] address a common problem among the elderly—chronic incontinence. They also raise broader ethical issues of research involving nursing home patients. Residents of institutional care settings have a high prevalence of mental and physical disabilities and are vulnerable to neglect and coercion. Similarly, potential protocol participants with dementia may be excluded because of their inability to consent or may be coerced into participating.

This dilemma has been addressed previously. Sachs and colleagues [2] posit that exclusion of patients because of physical or cognitive deficits constitutes discrimination. An inherent right to participate is assumed if the patient is not otherwise excluded. Also, such exclusions skew the study sample, thus limiting the generalizability of the findings. These authors support assent, evidence of the patient's cooperation, as an alternative to informed consent, with proxy consent when possible. Ouslander and colleagues [1] obtained assent when patients could not give informed consent; proxy consent was obtained after procedures.

Although this approach may benefit recruitment and data, ethical issues remain. The concept of assent assumes that a patient's failure to withdraw indicates willingness to participate. However, failure to resist may reflect only a lack of fear or discomfort. The distinction is a fine one: active participation versus passive submission. Assent, then, starkly contrasts with the notion of informed consent, which demands competence and voluntariness. Furthermore, procurement of proxy consent after the fact seems equally contrary to contemporary ethical guidelines of clinical investigation.

If assent is the only option available, certain criteria should be met. Risks to the participant must be minimal; in this case, urinary catheterization is fairly benign. Second, good reason must exist to involve patients who cannot consent. Some investigations by definition require enrollment of patients with dementia (for example, research on Alzheimer disease) [3]. The study of bacteruria and incontinence, however, might actually have been compromised by including cognitively impaired patients if, for example, impairment were independently associated with incontinence.