Discussion
This is the first study to investigate the impact of PIMs and PPOs at discharge from geriatric rehabilitation on unplanned readmissions and mortality among geriatric rehabilitation inpatients in Australia. While the prevalence of PIM and PPO use at different time points during the transition into geriatric rehabilitation care was high, the proportion of patients taking any PIMs/PPOs significantly reduced from acute admission to geriatric rehabilitation discharge. The use of at least 1 PIMs, PPOs or PIPs at geriatric rehabilitation discharge were not associated with an increased risk of unplanned readmission and mortality, be it in the short or long-term. Specific groups of PIMs (CNS/psychotropics and fall risk increasing medications) were significantly associated with the 30-day hospital readmission and cardiovascular PPOs with the 12-month mortality. Furthermore, an increased number of discharge medications was significantly associated with 30-day and 90-day hospital readmissions. The use and number of PPOs that also included vaccine omissions were negatively associated with gains in IADL scores at 90-days following discharge from geriatric rehabilitation.
Temporal changes in the prevalence of PIMs and PPOs from acute admission to geriatric rehabilitation discharge were investigated in this study. It showed a significant reduction in PIMs and PPOs as patients moved from acute admission to discharge. While these findings were consistent with other studies,31,32 there were also studies that did not show a significant difference10,33 or have shown a rather increased PIM prevalence during the transition from admission to discharge.13 The current study has shown that the longer that patients stayed in geriatric rehabilitation ward, the fewer patients were likely to be exposed to PIMs. Therefore, it was not surprising to see a decline in the prevalence of PIM use from acute admission to discharge in this study.
In the present study, neither PIMs, PPOs nor PIPs were shown to have an association with readmission and mortality, but a higher number of discharge medications was a major predictor for unplanned hospital readmission at 30-days following discharge from geriatric rehabilitation. This finding is not consistent with previous reports conducted in acute care,7,34 in relation to hospital readmission and mortality. A recent meta-analysis highlighted the burden of medication-related harm due to potentially inappropriate prescribing.7 For example, patients using at least 1 PIM use in the inpatient setting (compared with non-users) were associated with a 91% increased odds of medication-related hospital admission (AOR 1.91, 95% CI 1.21-3.01), and 60% increased odds of dependency in at least 1 ADL (AOR 1.60; 95% CI 1.28–2.01). Informed by this review, a prospective telephone follow-up study with clinical audit of electronic medical record in general medical wards was also subsequently conducted to assess the links between inappropriate medication and clinical and functional outcomes (e.g., unplanned hospital readmission, dependency in activities of daily living).34 It was found that exposure to at least 1 PIM use was associated with an increased occurrence of unplanned hospital readmission (AOR 5.09; 95% CI 2.38─10.85), and there was also an increased in the likelihood of patients experiencing dependency in at least 1 ADL after hospital discharge (AOR 2.31; 95% CI 1.08─4.20).34 The reason for this difference may be that related to the healthcare setting itself, in that patients admitted to geriatric rehabilitation wards stayed longer (median 19.8 days vs 5 days)34 and were likely exposed to fewer PIMs than patients discharged from general medical wards. In addition, the previous study34 was conducted during the intense lock down due to COVID-19 and this may have impacted healthcare provision because of understaffing. The morbidity and mortality pattern were also different – for example, 90-days readmission was 36% in a previous study34 but 21.7% in the current study, it may be due to COVID-related restrictions with limited eligibility for hospital care. The lack of association between hyper polypharmacy (defined as prescription of 10 or more medications) and clinical outcomes in a previous study34 may be because only regular medications were counted unlike the current study. Medications prescribed pro re nata , including benzodiazepines for sleep aid may have been missed in the calculation, which was the main limitation for the previous study.34 It should be noted that the prescription of psychotropic and fall risk increasing PIMs were associated with 30-day readmission in the present study.
The present results are consistent with previous studies,35,36 showing that as the number of medications increase, so does the prevalence of 30-day and 90-day hospital readmissions. Basnet et al.35 found that the number of discharge medications was significantly associated with increased hospital readmission within 30-days, with the risk of rehospitalization increasing by 4% for every added medication (OR 1.04; 95% CI 1.03─1.05). Another study36 of 647 patients discharged from geriatric medicine acute care wards showed that taking eight or more medications was significantly associated with 90-day hospital readmission.