Attitudes Towards Deprescribing: The Perspectives of Geriatric Patients and Nursing Home Residents

Carina Lundby MScPharm, PhD; Peter Glans MScPharm; Trine Simonsen Pharmaconomist; Jens Søndergaard MD, PhD; Jesper Ryg MD, PhD; Henrik Hein Lauridsen DC, MSc, PhD; Anton Pottegård MScPharm, PhD


J Am Geriatr Soc. 2021;69(6):1508-1518. 

In This Article



A total of 358 individuals were screened for eligibility of which 32 were found ineligible due to cognitive impairment, 21 refused to provide written consent, and five were excluded for other reasons. Thus, 300 participants were included in the study, including 44 geriatric inpatients, 94 geriatric outpatients, and 162 nursing home residents.

Participants had a median age of 82 years (interquartile range [IQR] 76–89) and 66% were women (Table 1). Geriatric outpatients were slightly younger than geriatric inpatients and nursing home residents (median age 80 vs 83 and 84 years) and showed less cognitive impairment (median OMC score 26 vs 23 and 18). Participants used a median number of 8 regular medications (IQR 5–10), most commonly paracetamol (81%), laxatives (40%), platelet inhibitors (40%), and proton pump inhibitors (40%) (Table 2 and Tables S1–S3).

Attitudes Towards Deprescribing

Overall, 84% of the participants had a good understanding of their medications and 71% liked being involved in decision-making about their medications (Table 3). Although 51% believed they took a large number of medications, 64% considered their medications necessary and 84% were satisfied with their current medications. Nonetheless, 33% would like to try stopping one of their medications on their own, while 87% would be willing to stop one of their regular medications if their physician said it was possible.

Attitudes were generally similar between the individual participant groups, although considerable differences were seen for two items ("knowledge about current medications" and "feeling being given up on by physician"; Tables S4–S6).

Overall Factor Scores

The "appropriateness" and "concerns about stopping" scores were almost identical between the three participant groups (median 65 and 35–40, respectively) (Table 4). The "burden" score, however, was slightly higher for the geriatric inpatients compared to the geriatric outpatients and nursing home residents (median 50 vs 42, p = 0.11), indicating a greater perceived burden of taking medication among the geriatric inpatients. Further, the "involvement" score was higher for the geriatric outpatients compared to the nursing home residents (median 80 vs 75, p < 0.05) and geriatric inpatients (median 80 vs 70, p < 0.01), indicating more involvement in medication use among the geriatric outpatients.

Overall, the entire score range was generally used for each rPATD factor (Figure 1). The differences observed for the "burden" and "involvement" scores between the participants groups were also reflected in the score distributions for the three individual participant groups (Figures S1–S3).

Figure 1.

Revised Patients' Attitudes Towards Deprescribing (rPATD) factor score distributions for geriatric inpatients, geriatric outpatients, and nursing home residents (n = 300). Higher scores indicate more involvement, greater perceived burden, greater belief in appropriateness, and more concerns about stopping.7 See main text for more detail about score interpretation

Factor Scores Across Participant Characteristics

No pronounced differences in rPATD factor scores were found across sex and age groups, although the "burden" score was slightly higher for men compared to women (median 50 vs 42, p = 0.05) and for participants aged <80 years compared to those aged ≥80 years (median 50 vs 40, p = 0.05) (Table 4). A decreasing OMC score was associated with an increasing "appropriateness" score (median 60 vs 65 vs 70, p trend < 0.05). Further, an increasing number of regular medications was associated with an increasing "burden" score (p trend < 0.001): Those using 1–4 medications had a median "burden" score of 25 (IQR 17–33), while the median score of those using ≥10 medications was 58 (IQR 42–75). Likewise, the "concerns about stopping" score increased with an increasing number of regular medications (p trend < 0.001), from a median of 30 (IQR 20–45) for those using 1–4 medications to a median of 45 (IQR 35–60) for those using ≥10 medications. Conversely, the "appropriateness" score decreased with an increasing number of regular medications (p trend < 0.001), from a median of 75 (IQR 65–75) for those using 1–4 medications to a median of 60 (IQR 40–75) for those using ≥10 medications. Finally, lower physician trust was associated with a higher "concerns about stopping" score (median 45 vs 35, p < 0.01), while more concerns about prescribed medication was associated with a higher "burden" score (median 50 vs 25, p < 0.001) and a higher "concerns about stopping" score (median 45 vs 35, p < 0.001), respectively. Finally, less concerns about prescribed medication was associated with a higher "appropriateness" score (median 75 vs 57, p < 0.001).

When including items 9 and 10 (corresponding to the original rPATD questionnaire),[7] we observed similar trends in factor scores and score distribution for the "burden" factor, although scores generally showed minor increases (Table S7 and Figure S4).