Correspondence (reply): In Reply

We thank our correspondents and many other colleagues for their feedback about the PRISCUS list of potentially inadequate medications (PIM) for elderly patients.

The proportion of patients with PIM as determined by Schäfer and Wuttke cannot be compared to the 20% reported for the outpatient setting elsewhere (1). A recently published study from the US shows that implementation of the Beers list into the electronic prescribing system reduced prescriptions of PIM significantly (2).

The note about the increased fracture risk for pethidine that was included in the detailed PRISCUS list reflects expert opinion as per McLeod et al. (3). The study cited by Gehling presents recent results from a case-control study.

Adjuvant pain therapy using antidepressants and anticonvulsants is an important aspect in elderly patients. Experts' assessment of these medications remained primarily focused on the indication depression or seizures. On the other hand, antidepressants and anticonvulsants were deliberately named as alternatives to NSAIDs, which were classified as PIM.

On the basis of the risk assessment for diclofenac and etoricoxib we wish to point out again that we used methodically consensus opinion from experts, whom we furnished with publications that were as comprehensive as possible. The cited study by Fosbøl et al. (4) (registry study, 5 years before index date no admission to hospital, de facto medication “naive” [for example, no beta blockers, diuretics, ACE inhibitors, and many others]) studied a different population than our defined target group of mainly elderly patients with multiple comorbidities.

The dilemma mentioned by Fiß and Hoffman, of a lacking definition for “elderly” and “age” is indeed unsatisfactory.

As the colleagues correctly pointed out, the recommendation of risperidone as an alternative to the potentially inadequately rated neuroleptics does not apply for dementia patients. This important comment provides an opportunity to point out a limitation: The alternatives mentioned in the PRISCUS list are suggestions for treatment, which should be evaluated for their benefit-risk profile in the individual case before they are administered. We actually explicitly mentioned the need to pay attention to the product information in the PRISCUS list.

We welcome the idea that the PRISCUS list should be used by as many groups of researchers as possible, and that it should be monitored with regard to the prevalence of PIM in the context of adverse effects.

We wish to point out that the complete PRISCUS recommendations for medication are meant to support doctors and pharmacists. The list does not claim to be complete nor does it replace the individually tailored consideration of risks and benefits for the individual patient—rather, it is intended to raise awareness of the particular problems associated with drug treatment in elderly patients.