Clinical factors associated with the quality of interactions between staff and hospitalized older patients with dementia.

The study used baseline data from two medical units of a hospital, and most participants were White and non-Hispanic, so the results may not apply to other hospital populations. The design was cross-sectional, so it cannot show whether pain worsened interactions, better interactions reduced pain, or non-pharmacological interventions caused higher interaction quality. The sample had low average pain, low delirium severity, and lower depressive symptoms and functional dependency, so associations with those factors may have been hard to detect. Staff-reported and observed data may have been affected by social desirability or memory problems, because staff could report interventions or behave differently when being observed. The study did not examine staff education, training, experience, rushed task-oriented care, noisy environment, medications, or medical acuity, all of which could influence interactions.

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Do not conclude that giving more non-pharmacological interventions caused better staff-patient interactions or that pain caused worse interactions. The study observed associations at one time point and did not separate the effects of the many interventions counted.