Statins in Older Danes: Factors Associated With Discontinuation Over the First 4 Years of Use

Abstract

BACKGROUND AND OBJECTIVE Use of statins is considerable among older persons. We investigated factors associated with statin discontinuation in new statin users aged 70 years or older within the first 4 years of use. DESIGN Register‐based descriptive drug utilization study using data from 2008 to 2016. POPULATION/SETTING All Danish persons, aged 70 years or older, initiating statin treatment. MEASUREMENTS Rates and predictors of statin discontinuation after 1 year (early), 2 years, and 4 years. Predictors of discontinuation were estimated using logistic regression. RESULTS We included 83 788 statin initiators. At 1 year, 13% had discontinued their treatment, while another 12% and 13% discontinued after 2 and 4 years, respectively. The overall discontinuation rate over 4 years was 32%. Increasing age was associated with discontinuation at all time points (adjusted odds ratio [OR] = 2.06 [95% confidence interval {CI} = 1.35‐3.16] at 1 year, adjusted OR = 3.94 [95% CI, 1.83‐8.49] at 4 years, comparing those aged >95 years to those aged 70‐74 years). Further, higher comorbidity scores and use of more than 10 medications were modestly associated with discontinuation. Use of statins for secondary prevention was associated with decreased odds of discontinuation compared to primary prevention at 1 year (adjusted OR = 0.74; 95% CI, 0.65‐0.83) and at 4 years (adjusted OR = 0.83; 95% CI, 0.72‐0.95), along with concomitant use of cardiovascular (CV) therapies. The annual proportion of early discontinuers ranged from 14% to 17% for primary prevention and from 9% to 12% for secondary prevention between 2008 and 2015. DISCUSSION Statin discontinuation within the first 4 years after initiation appeared to be influenced most strongly by age, and may also be influenced by comorbidity, polypharmacy, use for secondary prevention, and concomitant CV medication use. Future research should clarify reasons for, and discussions about, statin discontinuation and initiation among older persons, to provide additional insight on this topic. J Am Geriatr Soc 67:2050–2057, 2019

Statin use and discontinuation in Danes age 70 and older: a nationwide drug utilisation study

“…The discontinuation rate we observed in long-term users was slightly lower than what has been reported over the first years following statin initiation. 20,21 The factors we identified to be associated with discontinuation in long-term statin users generally reflect what has been reported in early statin users, with increasing age being the strongest predictor of statin discontinuation. 19,[21][22][23][24] Similar to a previous study 21 of early discontinuation, we found higher co-morbidity scores increased likelihood of discontinuation.…”

“…20,21 The factors we identified to be associated with discontinuation in long-term statin users generally reflect what has been reported in early statin users, with increasing age being the strongest predictor of statin discontinuation. 19,[21][22][23][24] Similar to a previous study 21 of early discontinuation, we found higher co-morbidity scores increased likelihood of discontinuation. It is possible that with increasing co-morbidity, the potential benefit of statins becomes less compelling, prompting discontinuation.…”

Cardiovascular risk management in the elderly with type 2 diabetes prior to death: A Danish nationwide register study of discontinuation patterns

“…38 Evidence on this field proposes a low risk for serious side effects and rejects that milder symptoms such as muscle pains are related to statins. [39][40][41] The decline in the proportion on ASA with increasing calendar year of death could partly be attributed to the uncertainty of the absolute net value of ASA as primary prevention and the subsequent shift in recommendation, from T2D being a singular risk-factor to inclusion of high CVD risk for primary prevention. 25,42 Another factor might be the increased attention on the risk of gastrointestinal bleeding, which is particularly pronounced in the elderly population.…”

Considerations for pharmacoepidemiological analyses in the SARS‐CoV‐2 pandemic

“…This might lead to a paradoxically increased risk of ICU admission in users of preventive medication, for example, statins, which may not be prescribed to the same extent in the oldest and most frail populations. 34 Again, both geographical and temporal variation should be considered when comparing risk and prognosis of ICU admission, as there is considerable geographical variation in the number of ICU beds per inhabitants 35 and as the COVID-19 pandemic may lead to shortage of ICU beds and mechanical ventilators. 36 Finally, in defining and handling outcomes in COVID-19, the lag from transmission to onset of symptoms and to the potential need for hospitalisation needs to be considered.…”