Characterization of long-term oral anticoagulant adherence trajectories among patients with atrial fibrillation: a retrospective observational study

Background

Medication taking is a dynamic behaviour that changes over time. Conventional adherence summary measures (e.g. proportion days covered) used in the OAC adherence studies conducted so far, however, are insensitive to the fluid nature of adherence. For example, identical PDC values can be calculated for patients with initial good adherence followed by poor adherence, and for those with periodic non-adherence throughout the course of therapy.

Purpose

The objective of this study was to characterize atrial fibrillation (AF) patients' long-term unique oral anticoagulant (OAC) adherence trajectories.

Methods

Using linked, population-based administrative data containing physician billings, hospitalization and prescription records of 4.8 million British Columbians (1996–2019), incident adult cases of AF were identified. Only patients who had prescription refill data available for five years were included in the analysis. The primary measure of OAC adherence was the proportion of days covered (PDC) over consecutive 90-day rolling windows. We modelled continuous 90-day PDC values over time. The time variable was number of years since OAC initiation. Group-Based Trajectory Modelling (GBTM) was used to identify patients' unique longitudinal adherence trajectories. To determine the best model, a relative comparison was done between models using Bayesian information criteria (BIC), and the Akaike information criterion (AIC).

Results

The study cohort was 19,749 AF patients [mean age 70.6y (SD 10.64), 56% male, mean CHA2DS2-VASc score 2.77 (SD 1.39]. The model that best fit our data identified four distinct OAC adherence trajectories (Figure). These were “consistent good adherence” (n=14,631 patients, 74.1% of the cohort), “rapid decline and discontinuation” (n=2327, 11.8%), “rapid decline with recovery” (n=1973, 9.99%), and “slow decline and discontinuation” (n=819, 4.2%).

Our results show that there is heterogeneity among non-adherers. PDC dropped significantly in the first year after therapy initiation for those with “rapid decline and discontinuation” trajectory. Patients exhibiting “rapid decline with recovery” also displayed a rapid decline in adherence in the first year but showed improvements around the third year. Those in the “slow decline and discontinuation” trajectory displayed slow decline in adherence over first three years which eventually led to permanent discontinuation of therapy.

Conclusion

In this retrospective study we distinguished between the different kinds of non-adherence in terms of timing and rate. While a majority of our cohort adhered to their medications, we identified three unique trajectories displaying declining adherence over time at varying rates. Our results emphasize the importance of early intervention and have direct implications for improving the design of adherence interventions.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Canadian Institutes of Health Research grant

Comparative adherence between DOAC and VKA in patients with atrial fibrillation: a 23-year retrospective observational study in Canada

Background

A recent systematic review highlighted significant gaps in the evidence on atrial fibrillation (AF) patients' adherence to oral anticoagulants (OAC). Current evidence suffers from short follow-up times, focuses on the first OAC and does not take switching into account. There is also lack of observational data on adherence to warfarin due to its varying dose that complicates the calculations. As such there is lack of evidence on comparative adherence between VKAs and DOACs and whether the convenience of DOACs translates into better adherence in AF patients.

Purpose

Our objective was to measure AF patients' long-term OAC adherence and compare the impact of taking direct oral anticoagulants (DOAC) versus vitamin K antagonists (VKA) on adherence, while accounting for switching.

Methods

Using linked, population-based administrative data containing physician billings, hospitalization and prescription records of 4.8 million British Columbians (1996–2019), incident adult cases of AF were identified. The primary measure of adherence was proportion of days covered (PDC). Consecutive rolling 90-day windows were created for each patient starting from their first OAC prescription fill date until the end of their follow-up. The PDC for each 90-day rolling window was calculated and averaged to yield mean adherence over the follow-up period for each patient. Permanent medication discontinuation resulted in a PDC of 0 for all subsequent rolling windows after their supply ran out. As such, both poor execution and non-persistence were measured simultaneously. The association between drug class and adherence was assessed using generalized mixed effect linear regression models with drug class treated as time-varying covariate to account for switching.

Results

The study cohort was 30,264 AF patients [mean age 72.2 years (SD11.0), 44.6% female, mean CHA2DS2-VASc 2.94 (SD1.4)] with mean follow-up of 7.7 (SD 4.8) years. The mean PDC was 0.71 (SD 0.27) with 51% of the cohort having mean PDC values below the conventional threshold of adherence (PDC<0.8). Adherence dropped over time with the greatest decline in the first two years after therapy initiation. After controlling for all other confounders and accounting for switching, taking VKA compared to DOAC was, on average, associated with a 1-day decrease in number of days of medication-taking per year.

Conclusion

AF patients' OAC adherence was below the conventional threshold of 0.8, and dropped over time, particularly in the first two years. Drug class had no clinically meaningful impact on medication adherence. Our study highlights the need for effective adherence interventions particularly early in OAC therapy. Our findings also emphasizes that prescribers should not assume inherently better adherence for DOACs and should instead choose OAC in conversation with the patient and in accordance with their values and preferences.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): Canadian Institutes of Health Research grant

Implementation of proven interventions in general medical inpatients: development and evaluation of a new quality indicator for drug therapy

BACKGROUND

Among adult general medical inpatients, there are numerous interventions whose benefits outweigh their risks. However, there are no published reports describing the overall use of such proven interventions in this population.

OBJECTIVES

To determine implementation rates of a broad range of interventions while accounting for valid reasons for non-use, predictors of implementation and feasibility of generating new indices to describe quality of care.

METHODS

Based on a review of current practice guidelines and clinical trials related to five common conditions, implementation rates of 17 interventions were assessed retrospectively. Subjects were a complete sample of 150 adults with target medical conditions discharged from general medical units at an urban community hospital.

RESULTS

The Ideal Intervention Index (3I), which described the proportion of ideal intervention opportunities that were implemented, was 0.74 (95% CI 0.70 to 0.78). The Justified Non-Use Index (JNUI), which described the proportion of all the interventions not implemented that were justified by a valid reason for non-use, was 0.49 (95% CI 0.41 to 0.55). Smoking cessation therapy in high-risk patients had the lowest indices (3I 0.30, 95% CI 0.00 to 0.60; JNUI 0.00), and aspirin for secondary stroke prevention had the highest (3I 1.0; JNUI 1.0).

CONCLUSIONS

Overall, proven interventions are underused among the patients studied, and the reasons for non-use are frequently not readily discernible. There is potential for improvement, but research is required to further investigate reasons for non-use. It is feasible to measure implementation rates of proven interventions as an indicator of quality of care using the indices developed.

Upper extremity performance and self-care skill changes in children with spastic cerebral palsy following selective posterior rhizotomy

Changes in upper extremity and self-care performance following selective posterior rhizotomy (SPR) are reported frequently, but rarely quantified. In this study, 36 children with spastic cerebral palsy were assessed preoperatively and 1 year following SPR using the Quality of Upper Extremity Skills Test (QUEST). Twenty-six children were assessed at similar intervals using the Functional Independence Measure for Children (WeeFIM) as a measure of self-care performance. Wilcoxon matched-pair signed-rank tests were used to compare the QUEST total scores and the WeeFIM motor, cognitive, and total scores before and after SPR. One year after SPR, the total QUEST scores were significantly better (median improvement = 3.2%, p < 0.0001), as were the WeeFIM motor (median improvement = 9.5, p < 0. 0001), cognitive (median improvement = 1.0, p < 0.008), and total (median improvement = 11.0, p < 0.0001) scores. The results indicate quantifiable improvements in upper extremity function, and motor and cognitive self-care skills in children 1 year after SPR.

E. coli HPII catalase interaction with high spin ligands: formate and fluoride as active site probes

E. coli catalase (HPII) wild type and mutant enzymes (heme dcis-containing) were examined (i) to study the role of a distal haem cavity residue, asparagine-201, in high spin ligand binding and (ii) to compare the differences in this binding between heme d and protoheme enzymes such as that from beef liver (BLC). High spin fluoride complexes were formed by all three HPII catalases examined, wild type (201 asn) and 201gln and 201asp mutants, but with a lower fluoride affinity than that of BLC. The binding of fluoride was pH-dependent, indicating that a proton is bound as well as a fluoride anion. HPII 201glu and 201 asp mutants showed lower affinities for fluoride than did wild type, unlike their reactions with cyanide which are essentially independent of the nature of residue 201. The equilibria and rates of fluoride and formate binding to BLC were reexamined. The rates of reaction with formate were similar to those reported previously. Dissociation rates for fluoride-catalase are higher than for formate suggesting that the latter may be bound differently. High spin complexes between formate and all three HPII forms showed a substantially higher affinity than that of BLC for HPII wild type and progressively lower affinities for the two mutants. As with fluoride the reactions were pH-dependent, indicating that a proton is bound together with the formate anion (or that undissociated formic acid is the ligand). The known structures of the heme groups and heme pockets involved are discussed. Formate may be bound by secondary H-bounds within the heme pocket in both heme dcis and protoheme enzymes. The nature of the heme pocket and the heme access channel may be more important than the chemical nature of the prosthetic group in controlling both high spin ligand interactions and reactions with the substrate hydrogen peroxide.

The efficacy and safety of combination warfarin and ASA therapy: a systematic review of the literature and update of guidelines

The English-language literature was systematically reviewed to clarify the role of combination antithrombotic therapy with warfarin and acetylsalicylic acid (ASA) versus monotherapy with either agent, including data from several recently published trials. Sixteen published studies with evaluable efficacy and/or safety data were identified. For patients with prosthetic heart valves at high risk of thromboembolism, combined warfarin and ASA therapy may be beneficial compared with warfarin alone. Combination therapy may also be used for primary prevention in patients at high risk for ischemic heart disease, although the expected benefits are small. Evidence does not support the use of combined antithrombotic therapy in patients with established ischemic heart disease, ischemic stroke, coronary artery bypass grafts or atrial fibrillation. Combination therapy is associated with an increased risk of minor and major bleeding. The highest dose of ASA that can be recommended in combination with warfarin is 100 mg daily.

Activity, peroxide compound formation, and heme d synthesis in Escherichia coli HPII catalase

Wild-type Escherichia coli HPII catalase (heme d containing) has 15% the activity of beef liver enzyme per heme. The rate constant for compound I formation with H2O2 is 1.3 x 10(6) M(-1) s(-1). HPII compound I reacts with H2O2 to form O2 with a rate constant of 1.8 x 10(6) M(-1) s(-1). Forty percent of HPII hemes are in the compound I state during turnover. Compound I is reduced by ethanol and formate at rates of 5 and 13 M(-1) s(-1) (pH 7.0), respectively. Incubation of HPII compound I with ferrocyanide and ascorbate does not form a compound II species. Mutation of His128 to alanine or asparagine gives inactive protoheme proteins. Mutation of Asn201 gives partially active heme d forms. Asn201Ala has 24%, Asn201Asp 10%, and Asn201Gln 0.4% of wild-type activity. Asn201His contains protoheme when isolated and converts this via protoheme compound I to a heme d species. Both distal heme cavity residues His128 and Asn201 are implicated in catalytic activity, compound I formation, and in situ heme d biosynthesis. HPII Asn201, like the corresponding residue in protoheme catalases, may promote H+ transfer to His128 imidazole, facilitating (i) peroxide anion binding to heme and (ii) stabilization of a transition state for heterolytic cleavage of the O-O bond.

Probing the structure of catalase HPII of Escherichia coli--a review

Escherichia coli produces two catalases or hydroperoxidases, HPI and HPII. HPI is a bifunctional catalase-peroxidase active as a tetramer of identical 80049-Da subunits encoded by katG. The expression of katG is controlled at the basal level by sigma s (KatF), and its induction by H2O2 is regulated by OxyR. HPII is a monofunctional catalase active as a tetramer of identical 84118-Da subunits encoded by katE. The induction of katE expression in the stationary phase is controlled by sigma s. The core of HPII is similar in sequence to other catalases including the conservation of several residues that have been implicated as playing a catalytic role, His128, Asn201, Ser167 and Tyr415. These residues have served as targets for site-directed mutagenesis in a study that has demonstrated their role in the catalytic mechanism of HPII. In addition, the two Cys residues in HPII have been targeted in a similar study revealing that they do not have a catalytic role, but that Cys438 is blocked by a novel modification. Despite many structural similarities to bovine liver catalase, the heme component of HPII has proved to be quite different. The presence of a cis heme d was determined spectrally and chromatographically, and the inability of certain mutants to generate the modified heme revealed that it was HPII itself that was catalysing the oxidation of heme b to heme d. The recent solution of the crystal structure of HPII and mass spectrometry have revealed that the heme d bound to HPII is a spirolactone structure with a cis orientation of the oxygens on the proximal side of the heme. This has created the problem of explaining how the oxidation of the heme can occur on the opposite side of the heme ring, remote from the catalytic residues.