Elderly women more likely than men to have inappropriately prescribed medicines

A recent Swiss study found a high prevalence of polypharmacy (66%) and inappropriate prescribing (67%) among people living with HIV aged 75 and older. Forty per cent of these prescribing issues posed a risk for harmful consequences. Inappropriate prescribing was more frequent in women than in men, which can be partially explained by sex differences in non-HIV co-morbidities.

Polypharmacy (defined in this study as taking five or more non-HIV medications) is common among people with HIV, especially so as a person ages. Managing multiple health conditions poses a challenge because of drug-drug interactions (DDIs), drug-disease interactions, and age-related physiological changes that can affect how the body processes drugs. Lack of co-ordination among multiple providers can also increase the risk for inappropriate prescribing.

Inappropriate prescribing among elderly people with HIV has not been thoroughly evaluated. Existing research mostly recruited people aged 65 years and younger, and did not address co-morbidities, drug dosage, treatment duration, and/or management of DDIs.

This study assessed prevalence of and risk factors for inappropriate prescribing in people living with HIV aged 75 or older in the Swiss HIV Cohort Study. This ongoing, nationally representative study includes nearly 80% of all people living with HIV in Switzerland who are receiving HIV treatment.

A total of 175 people were included in the study, 98% of whom were White. Seventy-one per cent of participants were male, with a median age of 78. Female participants had a median age of 78.5. Almost all (91%) participants were virally suppressed.

Non-HIV co-morbidities were common, the top five being hypertension (61%), renal impairment (56%), dyslipidaemia (44%), neurocognitive problems (39%) and osteoporosis (30%). Women were more likely to have central nervous system (CNS) disorders (62% vs 45%) such as cognitive impairment and depression; renal impairment (62% vs 54%); and musculoskeletal disorders (72% vs 61%) such as osteoporosis and neuropathy.

Almost all participants (98%) received non-HIV-drugs, with a median of five. Sixty-seven per cent of patients had at least one prescribing issue, with 41% having multiple. As expected, the number of prescribing issues rose with longer duration of known HIV infection (median = 18 years), greater number of co-morbid health conditions, and the more non-HIV drugs that a person took.

Many participants (61%) were receiving antiretrovirals that may perpetuate DDIs, but most prescribing issues mostly occurred with non-HIV drugs. Incorrect dosage (26% of errors) included things like failure to adjust dosage to renal function, which can decline with age. Twenty-one per cent of issues were lack of clinical indication or demonstrated benefit.

Prescription omission (e.g. lack of aspirin for people with cardiovascular risk), were seventeen per cent of errors. Drugs that were inappropriate for elderly patients (such as long-term use of benzodiazepines) were 18% of errors. Finally, harmful DDIs (e.g. red flag DDIs or improperly managed amber flag DDIs) were 17% of all issues.

All prescribing issues were more common in female participants, who had more co-morbidities and took more non-HIV drugs than men in this cohort. Dosing errors were particularly elevated in women compared to men (44% vs 28%), as were prescriptions without indication (34% vs 17%), and inappropriate prescriptions for elderly patients (36% vs 15%). Dosing errors were also much more common in people with renal impairment (defined as eGFR < 60 mL/min/1.73 m²) than in those with no impairment, at 47% and 14%, respectively.

While sex differences in type and frequency of co-morbidities accounts for some of the sex differences, female sex still presented an independent, eightfold risk for one or more prescribing issues after adjusting for these and other factors (aOR 8.28, 95% CI, 2.44-28.08). Other significant factors included polypharmacy (aOR 2.5, 95% CI, 1.34-4.65), renal impairment (aOR 2.68, 95% CI, 1.42-5.05) and treatment with CNS-active drugs (aOR 2.09, 95% CI, 1.14-3.82).

CNS-active drugs were mostly benzodiazepines and drugs with anticholinergic properties. These are inappropriate among elderly people, as are other sedative/hypnotic drugs. Their risks include memory loss, falls, accidents, hallucinations and confusion. The authors note that some of the inappropriate prescribing of these may have been driven by patient request.

This was a small study, which lacked a control group; however, the findings on inappropriate prescribing are consistent with existing research on HIV-negative elderly people. More research is needed to replicate these findings and understand why sex posed such an elevated risk for prescribing issues among people with HIV. Some possibilities include sex differences in healthcare utilisation and sex and/or gender bias among healthcare providers, who are more likely to diagnosis and treat certain conditions (such as depression and anxiety) in females.

While this retrospective study did not assess the clinical outcomes of inappropriate prescribing, the authors recommend routine review of prescriptions as a way to reduce potential harms. They also recommend clinicians take care to avoid bias, especially with female patients.

References