Understanding Medication Nonadherence from Social Media: A - PowerPoint PPT Presentation

Understanding Medication Nonadherence from Social Media: A Sentiment-Enriched Deep Learning Approach Presented by: Xiao Liu Department of Operations and Information Systems University of Utah Joint Work with Jiaheng Xie 1 , Xiao Fang 2 , Daniel Zeng 1 1 Department of Management Information Systems, University of Arizona 2 Department of Accounting and Management Information Systems, University of Delaware

What is Medication Nonadherence • Definition: patients not take medications as recommended (Marcum et al. 2013, JAMA ) • Types of nonadherence (Servick 2014, Science ) • Non-fulfillment: not initiate • Non-conforming: take at the wrong time • Non-persistence: discontinue after starting 2

Implications of Medication Nonadherence – Patients’ Health • Detrimental to patients’ health • Drug withdrawal symptoms (Fox et al. 2014, NEJM ) • Treatment failure, disease worsening (Marcum et al. 2013, JAMA ) • Emergency visit, hospitalization (Traverso & Langer 2015, Nature ) Examples of withdrawal symptoms 3

Implications of Medication Nonadherence – Disease Management • Critical to disease management • ≥ 50% patients not follow therapy (Traverso & Langer 2015, Nature ) • Impact on health: increasing adherence rate ≫ improvements in treatment (Shank 2011, NEJM ) • Costs to healthcare systems Account for 19% drug-related visits to emergency rooms (Bailey et al. 2014) • $290 billion preventable costs annually in US (Traverso & Langer 2015, Nature ) • Costs of medication nonadherence 4

Medication Nonadherence Research Medical and healthcare Perspectives from patients system’s perspective • Vital for disease 1. Harmful impact of mitigation & Physicians & medication management (Adler & healthcare Caregivers nonadherence (Gaebel Stead 2015, NEJM) systems et al. 2016; Serper et al. • Why discontinuing 2014) medications? Patients Pharmaceutical 2. Health policy to address companies the issue (Viehman et Prior studies al. 2016; Billimek et al. • Focused patient group & 2015) drug class • Survey: cross-sectional, expensive 5

Research Motivation Methodological Research objective 1. Define reason mining problem Design a deep 2. Sentiment-enriched word representation learning-based – Capture characteristics of patient-generated text framework for and represent noisy patients’ vocabulary real-time – Adjustable based on research context medication 3. Deep learning-based framework for reason extraction nonadherence reason detection in health social Healthcare IT media Investigated medication nonadherence reasons comprehensively by leveraging online health IT platforms 6

Literature Review • Medication nonadherence • Related work in the medical domain • Aspect mining • Approach to understanding factors that lead to certain opinion and behaviors 7

Medication Nonadherence: Three Research Streams Table 1. Selected recent studies on medication nonadherence Sampl Category Author Year Method Drug/Drug class Key findings e size Nafcillin, Viehman et al. 2016 Survey 224 18% patients discontinued therapy Oxacillin Prevalence of Williams et al. 2014 Survey Diabetes 1,264 16% nonadherence rate medication Marcum et al. 2013 Survey Cardiovascular 897 40.7% nonadherence rate nonadherence Rheumatoid Du Pan et al. 2012 Cohort 1,485 Higher retention rate for patients on non-aTNF-Bio arthritis Gaebel et al. 2016 Survey Schizophrenia 19 Re-exacerbation after drug discontinuation Harmful Serper et al. 2014 Interview Liver 105 Adverse clinical outcome if discontinue outcomes of medication Miller 2013 Survey Anticholinergic NA Withdrawal symptoms: Irritability, dysphoria, nausea nonadherence Offord et al. 2013 Cohort Schizophrenia 873 Hospitalization, increased cost if discontinue Mak et al. 2016 Survey Schizophrenia 71 Lack of insight into illness à nonadherence Limitations: Bipolar, Derya et al. 2015 Survey 203 Unwilling to use meds, side effect à nonadherence Reasons for Schizophrenia medication Weid et al. 2015 Survey Diabetes 1,026 Negative expectation, econ burden à nonadherence nonadherence Weight gain, perceived cognitive impairment, tremors, Mago et al. 2014 Review Bipolar 207 sedation à nonadherence 8

Medication Nonadherence Table 1. Selected recent studies on medication nonadherence Sampl Category Author Year Method Drug/Drug class Key findings e size Viehman et al. 2016 Survey Nafcillin, Oxacillin 224 18% patients discontinued therapy Williams et al. 2014 Survey Diabetes 1,264 16% nonadherence rate Prevalence of medication Marcum et al. 2013 Survey Cardiovascular 897 40.7% nonadherence rate nonadherence Rheumatoid Du Pan et al. 2012 Cohort 1,485 Higher retention rate for patients on non-aTNF-Bio arthritis Gaebel et al. 2016 Survey Schizophrenia 19 Re-exacerbation after drug discontinuation Harmful Serper et al. 2014 Interview Liver 105 Adverse clinical outcome if discontinue outcomes of medication Miller 2013 Survey Anticholinergic NA Withdrawal symptoms: Irritability, dysphoria, nausea nonadherence Offord et al. 2013 Cohort Schizophrenia 873 Hospitalization, increased cost if discontinue Prevalent + serious heath outcomes à need to improve adherence rate Take preventive actions and improve drug adherence worldwide à need to understand reasons for medication nonadherence comprehensively 9

Medication Nonadherence Table 1. Selected recent studies on medication nonadherence Drug/Drug Sample Category Author Year Method Key findings class size Mak et al. 2016 Survey Schizophrenia 71 Lack of insight into illness à nonadherence Bipolar, Unwilling to use medication + side effect à Derya et al. 2015 Survey 203 Schizophrenia nonadherence Reasons for medication Negative expectation + economic burden à Weid et al. 2015 Survey Diabetes 1,026 nonadherence nonadherence Weight gain, perceived cognitive impairment, tremors, Mago et al. 2014 Review Bipolar 207 sedation à nonadherence Cross-sectional Focused drug class & patient group Few nonadherence Emerging treatment options Different regimens/effects across drug • • reasons detected à Changing reasons over time classes/patient groups à Varied reasons across drug classes/patient groups 10

Addressing the Limitations Detect nonadherence Patient and drug-specific reasons comprehensively nonadherence reasons & continuously Need comprehensive & real-time dataset + automated approach • Learn nonadherence reasons from heterogeneous patient groups continuously • Predict customized nonadherence reasons for each group timely • Discover nonadherence reasons unnoted by previous studies 11

Comprehensive and Real-time Dataset • Health social media: time-sensitive patients’ feedback about medications • 61% US adults participating in online health community (Xu et al. 2016) • Valuable for understanding medication nonadherence • Social media in healthcare applications • Peer/social/emotional support (Chiaramello et al. 2016) • Drug safety surveillance (Xie et al. 2017; Nikfarjam et al. 2015) à Rarely used by prior medication nonadherence studies 12

Technique to Approach the Problem • Medication nonadherence reasons (Gellad et al. 2011) • Attributes of drug: price, effectiveness, etc. Aspect mining • Attributes of patient: emotion, health literacy • Attributes of patient-provider relationship • Other • Aspect mining • Extract attributes of objects • E.g., aspect of an iPhone (price, screen size) à purchase behavior 13

Aspect Mining Table 2. Selected recent aspect mining studies Task Author Year Method Data Object Aspects Performance Wang et al. 2016 CRF 3,646 reviews Restaurant Food, price, service F-score: 63% Liu et al. 2016 CRF 5.8m reviews Camera Battery life, picture quality F-score: 73% Perplexity: Fang et al. 2015 LDA 5,632 reviews Tourist spot Architecture, park, food 3,375 Jeyapriya & Selvi 2015 Rule-based 100 reviews Camera Pictures, resolution, memory F-score: 80% Aspect extraction Li et al. 2015 CRF 6,847 reviews Mouse Durability, compatibility Precision: 65% Brisson & Torrel 2015 CRF 40,160 reviews Smartphone Battery, camera F-score: 68% Marrese et al. 2014 CRF 200 reviews Hotel Price, food, service, pool F-score: 73% Hai et al. 2014 LDA 10,073 reviews Cellphone Screen, battery, fans, money F-score: 56% Patra et al. 2014 CRF 3,045 sentences Restaurant Service, price, food, ambience F-score: 72% Chen et al. 2014 K-means 36k reviews Camera Battery, service, memory Coherence: -1k Xiong et al. 2016 K-means 1,389 sentences Camera Photo, battery RI: 59% Aspect clustering Xiong et al. 2016 K-means 1,389 sentences Camera Photo, battery Entropy: 1.74 Wang et al. 2016 LDA 1,000 reviews Laptop Screen, battery, usefulness Precision: 79% 14

Applications of Aspect Mining Aspects Food, price, service Aspect of products (Liu et al. 2016; Battery life, picture quality Architecture, park, food, Li et al. 2015; Hai et al. 2014) museum à Lead to purchase behavior Pictures, resolution, memory Durability, compatibility, sensitivity Aspect of Battery, camera interest Price, food, service, pool Screen, battery, fans, money Aspect of tourist destinations (Fang Service, price, food, ambience et al. 2015; Marrese et al. 2014) Battery, service, memory à Lead to choice of destinations Photo, battery Photo, battery Screen, battery, usefulness 15