Adherence to antihypertensive therapy : analysis of initiation, implementation, discontinuation and possible risk factors in portuguese primary care units

ABSTRACT: RATIONALE AND BACKGROUND: Cardiovascular disease is a major cause of morbidity Worldwide, being responsible for up to 32% of all deaths in Portugal. Hypertension is highly prevalent and one of the major risk factors for cardiovascular disease. Over 42% of the Portuguese adults (18-90 years) would have hypertension. Even though the benefits of antihypertensive drugs in reducing the risk of major cardiovascular events have been extensively demonstrated, the control of hypertension continues to be inadequate. The precise reasons for patients not achieving target blood pressure despite being treated are not completely clear yet, but one major (and modifiable) reason is the fact that patients often do not only fail to take their medication as has been prescribed, but also fail to use it for a long uninterrupted period of time. A substantially poorer medication adherence rate is observed when analysing newly diagnosed patients, and accounting for those who fail to initiate treatment, fail to ever refill, and time after discontinue, rather than the more commonplace approach of only observing ongoing users. Conventional adherence measures therefore systematically underestimate the public health burden of poor medication adherence of newly prescribed medications. OBJECTIVE: The main objective of this thesis is to determine adherence to antihypertensive therapy in newly treated hypertensive patients in primary health care units from Lisbon and Tagus Valley Region. METHODS: This thesis reports data from a large, population-based, retrospective, cohort study that assessed adherence to antihypertensive therapy, in all its components, i.e., initiation, implementation and discontinuation, in newly diagnosed and treated hypertensive patients in primary health care units of Lisbon and Tagus Valley from January 1st to March 31st 2011 who used no antihypertensive drugs prior to January 1st 2011. We’ve also determined primary adherence rate to antihypertensive drugs, expressed as the number of claims records divided by the total number of prescriptions records. Data were collected from SIARS for each patient during a two-year period after the date of the first acquisition. Initiation was determined by the acquisition of a first prescription in a pharmacy within a six-month period. Implementation was quantified by estimation of Medication Possession Ratio and persistence was determined by the proportion of patients remaining on any antihypertensive drug regardless of switching or the use of multiple drugs during follow-up. Persistence was analysed considering a maximum allowed treatment gap of 90 days. Reinitiation was also analysed. Initiation and persistence were analysed by Kaplan-Meier survival analysis and Cox proportional hazard regression was used to estimate hazard ratios for initiation and discontinuation. Logistic regression was used to estimate the odds ratio for poor implementation of the prescribed AHT therapy. RESULTS: Overall primary adherence rate was 58.5%, increasing with age. Primary adherence rates were higher for men, patients living in the Lisbon Metropolitan Area and diagnosed with ICPC-2 code k87. Drugs acting on the Renin-angiotensin system had the highest primary adherence rates, increasing for fixed-dose combinations and diminishing with the increase of out-of-pocket cost for patients. Of the 10,204 cohort members, 493 (4.8%) never acquired any antihypertensive drug and 855 more (8.4%) initiated hypertension treatment with a considerable delay (six-months or longer) after the first prescription, being classified as ‘non-users’. After adjustment for all the potential predictors of initiation, women, and patients aged 45-64 years, who received an initial prescription of with two or more drugs had higher initiation rates. Among patients with a first dispensing (n=8,856), 638 (7.2%) patients discontinued antihypertensive therapy after acquiring just the first prescription and 519 more (5.9%) completely discontinued treatment during the first year, making a total of 1,157 (13.1%) patients who were no longer on treatment at the end of the first year. During the second year, 904 (10.2%) more discontinued antihypertensive therapy. However, in spite of 6,157 patients being still on treatment two years after the initiation of hypertension treatment, only 539 (8.8%) of them were classified as continuous users, i.e. had no treatment gap or grace period of 90 days or longer, meaning that the remaining 5,618 (91.2%) were using antihypertensive therapy in an ‘on and off’ basis, discontinuing and reinitiating it over time. The risk of complete discontinuation was higher for younger patients, treated with monotherapy and followed by a single physician. Analysing the implementation of hypertension treatment in the two-year observation period, among patients with a first dispensing only 456 (5.1%) had in their possession antihypertensive drugs for 80% or more days, regardless the occurrence of lapses in implementation, i.e. treatment gap or grace periods of 90 days or longer, which occurred in 233 (51.1%) of this patients with a high level of implementation. Younger patients and with a higher buying power had a higher risk for poor implementation. However, and in spite of increasing persistence, the use of more drugs during the follow-up decreased the medication possession ratio for this patients. CONCLUSION: The results of this thesis confirm previous observations that in clinical practice hypertension treatment is frequently abandoned and poorly implemented over time. Our results demonstrated that almost one out of five (19.5%) patients either never initiated treatment, did it with a considerable delay (after six months or more after their first prescription) or completely discontinued it just after acquiring their first antihypertensive drug, being the risk for discontinuation most pronounced during the first year. Low adherence rates to antihypertensive therapy, in all its components, are an especially alarming finding, since this condition contributes greatly to the burden of mortality and morbidity from cardiovascular disease in Portugal. Until this thesis, little was known in Portugal about adherence to antihypertensive therapy, especially at a population level. With this thesis we demonstrated not just the patterns of adherence to antihypertensive therapy but also some possible risk factors for non-adherence