Prevalence of potentially inappropriate prescribing and prescribing omissions in older Irish adults: findings from The Irish LongituDinal Study on Ageing study (TILDA)

Rose Galvin 0 1 Frank Moriarty 0 1 Grinne Cousins 0 1 Caitriona Cahir 0 1 Nicola Motterlini 0 1 Marie Bradley 0 1 Carmel M. Hughes 0 1 Kathleen Bennett 0 1 Susan M. Smith 0 1 Tom Fahey 0 1 Rose-Anne Kenny 0 1 0 G. Cousins School of Pharmacy, Royal College of Surgeons in Ireland , Dublin, Ireland 1 R.<A. Kenny The Irish Longitudinal Study of Ageing, Chemistry Extension, Trinity College Dublin , Dublin 2, Ireland Purpose We sought to estimate the prevalence of potentially inappropriate prescriptions (PIP) and potential prescribing omissions (PPOs) using a subset of the STOPP/START criteria in a population based sample of Irish adults aged 65 years using data from The Irish LongituDinal Study on Ageing (TILDA). Methods A subset of 26 PIP indicators and 10 PPO indicators from the STOPP/START criteria were applied to the TILDA dataset. PIP/PPO prevalence according to individual STOPP/ START criteria and the overall prevalence of PIP/PPO were estimated. The relationship between PIP and PPOs and polypharmacy, age, gender and multimorbidity was examined using logistic regression. Results The overall prevalence of PIP in the study population (n=3,454) was 14.6 %. The most common examples of PIP identified were NSAID with moderate-severe hypertension (200 participants; 5.8 %) and aspirin with no history of coronary, cerebral, or peripheral vascular symptoms or occlusive event (112 participants; 3.2 %). The overall prevalence of PPOs was 30 % (n = 1,035). The most frequent PPO was antihypertensive therapy where systolic blood pressure consistently >160 mmHg (n=341, 9.9 %), There was a significant association between PIP and PPO and polypharmacy when adjusting for age, sex and multimorbidity (adjusted OR 2.62, 95 % CI 2.05-3.33 for PIP and adjusted OR 1.46, 95 % CI 1.23-1.75 for prescribing omissions). Conclusion Our findings indicate prescribing omissions are twice as prevalent as PIP in the elderly using a subset of the STOPP/START criteria as an explicit process measure of potentially inappropriate prescribing and prescribing omissions. Polypharmacy was independently associated with both PPO and PIP. Application of such screening tools to prescribing decisions may reduce unnecessary medication, related adverse events, healthcare utilisation and cost. - Older adults are a heterogeneous group often presenting with multimorbidity and polypharmacy. Prescribing in older adults is challenging for a number of reasons as any new medication must be considered in the context of the physiological effects of the drug, the absorption and excretion of the drug and age-related changes in body composition and physiology [1]. Potentially inappropriate prescribing (PIP) in older people refers to the use of medications that should be generally avoided in such populations and doses or frequencies of administration that should not be exceeded [2, 3]. Medicines that are considered potentially inappropriate lack evidence based indications, are not cost-effective and may pose a higher risk of adverse events including increases in morbidity, adverse drug events, hospitalisation and mortality [4, 5]. In terms of assessing the appropriateness of prescribing in older people, both implicit and explicit measures of process (providers actions) and outcome (e.g., adverse drug events) are used [6]. Implicit process measures involve a clinicians judgment of appropriateness for the individual patient based on patient characteristics and published work [7]. Explicit process measures are criterion-based and are derived from published evidence based reviews, expert opinion and/or consensus [8, 9]. A comprehensive and explicit process measure of PIP has been developed and validated for use in European countries. The STOPP criteria (Screening Tool for Older Persons Prescriptions) are a physiological system based screening tool and comprises 65 clinically significant criteria [10]. The criteria include commonly encountered instances of PIP in older people such as drugdrug and drugdisease interactions, drugs which adversely affect older patients at risk of falls and duplicate drug class prescriptions. In addition, under-prescribing of clinically indicated medications has been identified by an accompanying screening tool known as the START criteria (Screening Tool to Alert doctors to Right Treatment), comprising 22 criteria for potentially appropriate drugs. There have been few national studies of PIP and PPOs in the general population of older people. In 2007, a subset of the STOPP criteria was applied in a national population study carried out in the Republic of Ireland among people aged 70 years, using the Health Services Executive Primary Care Reimbursement Services (HSE-PCRS) pharmacy claims database [6]. This study reported PIP prevalence rates of 36 % in the primary care setting. However, this study was limited in its assessment of PIP as the HSEPCRS database contains no information on diagnosis and so only a subset of the STOPP criteria could be applied. PIP prevalence rates reported may not have adequately reflected all instances of PIP among older people such as drug-disease interactions. Therefore, the overall aim of the current study is to estimate the prevalence of PIP/PPOs, particularly relating to drug-disease interactions, in a nationally representative sample of individuals aged 65 years, using a subset of the STOPP/START criteria. The Irish LongituDinal Study on Ageing (TILDA) is a nationally representative cohort study of people aged 50 years and over and resident in Ireland, charting their health, social and economic circumstances over a 10-year period. The first wave of the TILDA study has been completed, and we have examined the data from this cross sectional wave in our current study. The STROBE standardised reporting guidelines for cross-sectional studies have been followed to ensure the uniform conduct and reporting of the research. All participants aged 65 years are included [11]. Overview of TILDA data collection Data for the current study was captured from three methods used to collect data in the TILDA dataset. Participants first completed a computer-assisted personal interview (CAPI) in their own homes. Detailed information on all aspects of the respondents lives was collected in the CAPI, including economic status, health aspects (physical, mental, service needs and usage) and social participation. Participants were then invited to one of two health centres for a comprehensive health assessment. Health assessments were carried out by qualified and trained research nurses. Cognitive, cardiovascular, gait and balance and vision measurements and tests were taken in the health assessment. Finally, each participant was also requested to complete and return a questionnaire designed to explore certain areas that were considered particularly sensitive for respondents to answer directly to an interviewer (e.g., relationship quality, depression and loneliness). The names of medications that (...truncated)