A Framework for Assessing Adherence and Persistence to Long-Term - PowerPoint PPT Presentation

A Framework for Assessing Adherence and Persistence to Long-Term Medication Thusitha Mabotuwana With: Prof. Jim Warren 31 August 2009 1

A bit of background  Chronic illness is the leading cause of death worldwide - affected 35m (60%) of the world population in 2005, projected to affect 41m (64%) by 2015.  Patients with chronic illness are usually on long-term medication.  Drugs are effective only when taken as directed.  Adherence rates are usually around 50% - poor adherence associated with poor clinical outcomes.  Important to identify patients with adherence issues whose clinical outcomes can be improved. 2

The opportunity  New Zealand in top tier on use of computing in General Practice medicine (near 100%)  Prescribing and results of tests ordered systemically present in Practice Management System (PMS) – Anecdotally, the quality of the record continues to improve on more ‘voluntary’ fields (e.g., Blood Pressures [BPs], diagnoses)  So the PMS data should tell us a lot about how well a patient is being managed Schoen C et al. On the front lines of care: primary care doctors' office systems, experiences, and views 3 in seven countries.. Health Aff (Millwood). 2006 Nov–Dec;25(6):w555-71.

Adherence and persistence  Adherence refers to coverage of medication during a particular time period  Persistence is an indication of the time of continuous therapy.  We all know that taking our medication on time is important to achieve the full benefit of them – patients need to follow prescribed treatment regimens reasonably closely. 4

How should we measure adherence?  No existing gold standard for measuring adherence  Medication Possession Ratio (MPR) is widely used as a measure of adherence to long- term medication, such as AHT medication Number of days supply held during evaluation period X 100 Number of days in evaluation period Andrade SE, Kahler KH, Frech F, Chan KA: Methods for evaluation of medication adherence and 5 persistence using automated databases . Pharmacoepidemiology and drug safety 2006, 15 (8): 565-574; discussion 575-567

MPR calculation considerations  We consider only that portion of coverage which overlaps with evaluation period AHT Pr4 AHT Pr3 AHT Pr2 AHT Pr1 Run-in Period Evaluation Period (EP) (6 months) (12 months) 6

MPR calculation considerations Traditional MPR will be 90 x 5 / 365 = 123% ‘Our’ MPR is [365 – (30+35)]/365 = 82% 7

MPR considerations Only patients on monotherapy are considered “to reduce the complexity in measuring medication adherence” Patients on concurrent therapy are excluded due to the difficulty “to define adherence for more than 1 medication concurrently” 8

Identifying lapses  A lapse should be running-into, during or at the end (on-going) of the evaluation period Lapse1 Lapse2 Lapse3 AHT Pr4 AHT Pr3 AHT Pr2 AHT Pr1 Run-in Period Evaluation Period (EP) (6 months) (12 months) 9

Framework architecture 10

Drug and classification knowledge bases Mabotuwana, T. and Warren, J., An Ontology Based Approach to Enhance Querying Capabilities of General Practice Medicine for Better Management of Hypertension. To appear in: Artificial Intelligence in 11 Medicine , 2009 (epub available online)

Specifying criteria details in XML Lapse constraints MPR constraints Drugs and diagnoses 12

Framework testing and verification  Boundary value analysis (BVA) – eg: 15, 29, 30, 31, 45 day lapses  Equivalence class testing – eg: 45-day nominal lapses prior to, during and after EP  All-pairs testing – eg: a lapse prior to EP together with a lapse during EP  Random testing – for 100,000 patients, 41 different criteria representing the four criteria classes  169 cases covering the first three testing techniques 13

Practice adherence and persistence rates - Antihypertensive adherence - EP = 1-May-08 to 30-April-09 - 6-month run-in N practice-1 = 881 N practice-2 = 940 MPR MPR Maximum Maximum lapse lapse duration duration <80% ≥ 80% <80% ≥ 80% (days) (days) ≥ 30 235 (27%) 70 (8%) ≥ 30 398 (42%) 109 (12%) <30 3 (<1%) 573 (65%) <30 5 (1%) 428 (45%) 14

Individual patient adherence and persistence rates - ACEi/ARB adherence - EP = 1-April-06 to 31-March-07 - HT and DM diagnoses 15

An interactive visualisation tool – non-adherent patient 16

An interactive visualisation tool – adherent patient Combination drugs 17

Prescribing-dispensing matching  Prescription drugs will work only if you take them  Some patients collect their prescriptions, but fail to fill the scripts at the pharmacy  Prescription based adherence calculations are useful – PPV 81%, NPV is 76% Mabotuwana, T., Warren, J., Harrison, J. and Kenealy, T., What Can Primary Care Prescribing Data Tell Us about Individual Adherence to Long-Term Medication? – Comparison to Pharmacy Dispensing Data. Pharmacoepidemiology and Drug 18 Safety , 2009 (Pubmed ref #19609958)

Key messages - There’s lots of good information in routinely collected EMR data that can be used to identify chronic patients whose clinical outcomes can be improved - The framework can be used to identify cohorts of patients with poor adherence and persistence rates - Currently looking at a feasibility study to identify issues behind poor adherence and persistence 19

Contact, Further Reading  Thusitha Mabotuwana thusitha@cs.auckland.ac.nz Methods/results of two recent studies:  – Mabotuwana, T. and Warren, J., ChronoMedIt – A Computational Quality Audit Framework for Better Management of Patients with Chronic Conditions. Journal of Biomedical Informatics , 2009 (epub available online) – Mabotuwana, T., Warren, J. and Kennelly, J., A Computational Framework to Identify Patients with Poor Adherence to Blood Pressure Lowering Medication. International Journal of Medical Informatics , 2009 (epub available online) Opinion/review piece:  - Warren J, ‘General Practice EMRs: What they can tell us, and how,’ Health Care and Informatics Review Online , December 2007 20

Comparison with Quality and Outcomes Framework (QOF)  Our criteria include identifying patients who need a follow-up (eg: “A lapse in AHT >30 days” criterion) which is required for sound adherence  QOF DM15 indicator is “…patients with diabetes… who are treated with ACE inhibitors (or A2 antagonists)” but what is treated with without an EP?  DM 12. The percentage of patients with diabetes in whom the last blood pressure is 145/85 or less  BP 5. The percentage of patients with hypertension in whom the last blood pressure (measured in the previous 9 months) is 150/90 or less 21