In Reply

Drs. Venkata, Upadhyay, and Talari expressed concern about our assessment of pain in elderly patients, assuming their limited ability to self-report. We agree that lack of precise evaluation of cognitive function of participants at baseline is a limitation of the present study. However, we excluded persons who already had been certified as disabled, by the Long-term Care Insurance information, at the start of the follow-up; therefore, all analytic subjects were considered to be capable of fully understanding the questionnaire and making valid and reliable responses.¹ 

We evaluated the severity of pain using a verbal rating scale (VRS). Studies comparing various pain intensity scales showed that VRS and simple numeric rating scale (NRS) have the highest validity and reliability in rating pain intensity in older adults, even in those with mild cognitive impairment.^2,3  In 2007, VRS and NRS also were recommended as the best scales for guidance on the assessment of pain in elderly people by the British Pain Society and British Geriatrics Society.⁴ 

Although we discussed that our finding of a negative association between the severity of pain and functional disability due to dementia may be influenced by the impact that treatment with nonsteroidal antiinflammatory drugs has on dementia risk, our study did not assess medication use. Whether pain predicts future cognitive function is still an important question that needs to be answered. We also believe that further studies should be performed using detailed information, such as data on medication, the chronological change in pain, and the appropriate assessment of pain.