A systematic review and meta-analysis of the prevalence and impact of potentially inappropriate prescribing in middle-aged adults

Problem

Potentially inappropriate prescribing (PIP) is known to be common in the elderly and to be associated with adverse clinical outcomes, such as increased healthcare utilisation and adverse drug reactions (ADEs), as well as increased economic cost. Work in Ireland and London have shown that PIP may be common in middle-aged adults (MAA). However, the effects of the of this prescribing in MAA on health, quality of life (QoL) and economic outcomes is unclear.

Approach

A systematic review and meta-analysis were conducted according to the PRISMA reporting guidelines (PROSPERO CRD42020206617- registered 07/10/2020). Searches were performed in Ovid Medline, Embase, CINAHL, Cochrane Library, Web of Science, ProQuest Dissertation and thesis database, OpenGrey, Clinicaltials.gov and WHO ICTRP. All papers were eligible through to September 2020. Inclusion criteria were, studies that: applied an explicit PIP criteria to adults in 45-64 year age group, reported the prevalence of PIP or outcomes associated with PIP. Studies where no English language translation could be obtained, letters, opinion pieces, editorials, case series, case studies and studies using implicit PIP criteria were not included. Risk of bias and overall certainty of findings were assessed using the QUIPS and GRADE tools.

Findings

8,183 records were screened following deduplication. Of these, 87 met our criteria and underwent full text review. Twenty-three studies were included in a narrative synthesis. Fifteen unique PIP criteria were used to define PIP in the middle-aged in these studies. Only one PIP criteria (PROMPT) was designed for use in the middle-aged population. Most studies were focussed on the elderly with some MAA included. Four studies which had disaggregated prevalence data for the middle-aged were included in a meta-analysis (PIP prevalence 38%, 95% CI 25-52%, n=753,030; I2= 99.9%, p<0.01, High certainty (GRADE criteria). Individual studies showed female sex, number of medications, number of long-term conditions and level of education were determinants of PIP in MAA. Two small cohort studies reported the association of PIP with clinical outcomes, healthcare utilisation and a quality-of-life score, however in these studies no statistically significant association between PIP and these outcomes were shown. Risk of Bias assessment of the papers found: 3 were high risk; 11 were moderate risk; 8 were low risk.

Consequences

Potentially inappropriate prescribing in middle aged adults is common. Some determinants of PIP have been described, however more detailed analysis of the effects of deprivation are lacking. Overall, there was insufficient data on whether PIP has any negative health, quality of life or economic outcomes in middle aged adults, therefore no robust evidence was available.