Comorbidities, Medication Use, and Overdose: A Snapshot of Medicare - PowerPoint PPT Presentation

Comorbidities, Medication Use, and Overdose: A Snapshot of Medicare Beneficiaries with Substance Use Disorders Aida Kuzucan PharmD, Bilal Khokhar, PhD, Linda Simoni-Wastila BSPharm, MSPH, PhD AcademyHealth (Behavioral Health Services Interest Group) Characterizing and Combatting the Opioid Epidemic Session June 24, 2017 1

Acknowledgments • Coauthors − Dr. Khokar, Research Scientist at General Dynamics − Dr. Simoni-Wastila, Professor at University of Maryland School of Pharmacy and Director of Research for the Peter Lamy Center on Drug Therapy and Aging • The Pharmaceutical Research Computing (PRC) center in the Department of Pharmaceutical Health Services Research at the University of Maryland • No funding source • No COI to report 2

Outline • Introduction • Objective and Methods • Results – Characteristics and Comorbidities – Pain and Mental Illness – Opioid Use and Opioid Overdose Hospitalizations – Subset Analysis • Discussion and Future Research 3

The Opioid Epidemic in Medicare (I) • Despite attention paid to opioid overdose in US, Medicare population largely ignored • Growth in opioid use fastest among Medicare beneficiaries – From 1995-2010: 1 • 14-fold increase in total Medicare population and • 9-fold increase in over 65 year olds – From 2007-2012: Use of ≥90 days of C -II and C-III opioids nearly doubled 2 • Concomitant use also a concern – From 2004-2013 : 53.2% of polypharmacy encounters among older adults involved an opioid and a benzodiazepine 3 – From 2015: 27.9% of Part D enrollees of opioid users were concurrently using sedative-hypnotics (SDH) 4 • Mental illness and pain disorders are associated with increased risk of opioid- related overdose 5,6 4

The Opioid Epidemic in Medicare (II) • Remaining questions concerning Medicare beneficiaries – Characteristics associated with SUD? – Outcomes other than fatal overdose (OD)? • Reasons for scant research: – Aged and disabled historically neglected – Stigma of SUD and mental illness – Limitations in administrative claims data • SAMHSA redaction of SUD diagnoses – Bias studies – Prohibit studies of SUD population – This study derives from one examining COPD and depression in Medicare beneficiaries using 2006-2009 unredacted CCW data • Observed high prevalence of SUD and opioid use 5

Outline • Objective and Methods • Results – Characteristics and Comorbidities – Pain and Mental Illness – Opioid Use and Opioid Overdose Hospitalizations – Subset Analysis • Discussion and Future Research 6

Objectives of this Exploratory Study • Better understand of Medicare beneficiaries with SUD by looking at patterns of – Mental illness – Pain syndromes – Opioid and SDH use • To see if this information matters when looking at opioid overdose hospitalizations 7

Methods: Study Design and Population • Cross Sectional Design – 2008-2009 CCW data • Substance abuse diagnoses were not redacted – Population • Inclusion Criteria – Continuous Medicare parts A, B, and D coverage from January 1, 2008 - December 31, 2009 or until death in 2009. • Exclusion Criteria – Individuals with a hospice start date in or before 2009 8

Methods: Defining SUD, Mental Illness, and Pain Pain Disorders 6 SUD (Inpatient and Outpatient ICD-9-CM Codes) • Joint pain “Dependent” and “Non - dependent Abuse” • Back pain • General Abuse • Head pain • Opioids • Neck pain • Sedative-hypnotics • Chronic pain • Stimulants (excluding cocaine) • Alcohol Mental Illness 8 • Illicit Substances • ADHD • Anxiety Overdose (Inpatient ICD-9-CM codes) • Bipolar • “Poisoning by opiates” • Depression and other depressive disorders • “Accidental poisoning by opiates” • Personality • “Adverse effects in therapeutic use with • PTSD opiates” • Schizophrenia and other psychotic disorders 9

Methods: Additional Variables of Interest • Opioid Exposures * – Opioid • Opioid full and partial agonist (e.g., oxycodone, buprenorphine) – Anxiolytics, Sedatives and Hypnotics (SDH) • Barbiturates (i.e., amobarbital, butabarbital) • Benzodiazepines (i.e., Alprazolam, clonazepam) – Benzodiazepines excluded from Part D Coverage from 2006-2009 • Hypnotics (i.e., Propofol, zaleplon) • Comorbidities 8 – Alzheimers and related dementias, Cancer, Cardiovascular diseases (CVD), Chronic Kidney Disease (CKD), Chronic Obstructive Pulmonary Disorder (COPD), Diabetes, Rheumatoid/Osteoarthritis, Osteoporosis, and Human immunodeficiency virus (HIV) • Sociodemographic Characteristics • Age, gender, race, state of residence, original reason for Medicare entitlement * Drug classifications from American Hospital formulary Service (AHFS) Drug Information 2017 Edition 10

Outline • Objective and Methods • Results – Characteristics and Comorbidities – Pain and Mental Illness – Opioid Use and Opioid Overdose Hospitalizations – Subset Analysis • Discussion and Future Research 11

Results: Characteristics of Medicare Beneficiaries With and Without SUD (I) X 2 p-value Characteristic With Any SUD, N(%) Without Any SUD, N(%) N=24,058 N=1,214,336 Male Gender 1,4581 (60.6) 437,132 (36.0) <0.0001 Race <0.0001 White 17,467 (72.6) 993,619 (81.8) Black 5,064 (21.1) 130,180 (10.7) Other/Missing 1,527 (6.4) 90,537 (7.5) Original Reason for Entitlement <0.0001 Age 5,451 (22.7) 904,754 (74.5) Disability and/or ESRD 18,607 (77.3) 309,582 (25.5) Age <0.0001 >=64 17,157 (71.3) 218,066 (29.9) 65-74 4,659 (19.4) 499,827 (41.2) 75-84 1,865 (7.8) 363,043 (29.9) 85+ 377 (1.6) 133,398 (11.0) 12

Results: Characteristics of Medicare Beneficiaries With and Without SUD (II) With Any SUD, N(%) Without Any SUD, N(%) X 2 p-value Characteristic N=24,058 N=1,214,336 CVD* 15,518 (64.5) 435,108 (35.8) <0.0001 COPD 6,307 (26.2) 78,380 (6.5) <0.0001 Diabetes 5,278 (21.9) 153,227 (12.6) <0.0001 CKD 3,743 (15.6) 81,847 (6.7) <0.0001 RA or OA † 2,837 (11.8) 70,889 (5.8) <0.0001 Liver Disorders 2,800 (11.5) 10,150 (0.8) <0.0001 Alzheimer's and 1,727 (7.2) 58,415 (4.8) <0.0001 Related Dementias Cancer 1,157 (4.8) 37,716 (3.1) <0.0001 Osteoporosis 1,068 (4.4) 43,356 (3.7) <0.0001 HIV 756 (3.1) 2,642 (0.2) <0.0001 *Includes acute myocardial infarction, heart failure, hyperlipidemia, hypertension, heart disease, and stroke † Rheumatoid arthritis and osteoarthritis 13

Outline • Objective and Methods • Results – Characteristics and Comorbidities – Pain and Mental Illness – Opioid Use and Opioid Overdose Hospitalizations – Subset Analysis • Discussion and Future Research 14

Results: SUD, Pain, and Mental Illness Diagnoses (Rates per 1,000 Beneficiaries: 2008-2009 18.9 Pain Mental Illness 13.4 12.3 Pain and Mental Illness 11.3 10.0 Prevalence 8.6 8.3 7.7 7.5 7.3 6.7 5.9 3.4 3.5 2.5 2.4 2.1 2.6 2.8 3.1 Any SUD Alcohol SUD Drug SUD Drug and Alcohol Opioid SUD SUD 15

Results: Mental Illness and Non-Cancer Chronic Pain Among Medicare Beneficiaries With and Without SUD X 2 p-value * Characteristic With Any SUD N(%) Without Any SUD N(%) Any Mental Illnesses 14,411 (59.9) 115,455(9.5) <0.0001 Depressive disorders 11,084 (46.1) 84,060 (6.9) Bipolar 5,302 (22.0) 12,509 (1.0) Anxiety 5,470 (22.7) 31,390 (2.6) Psychotic Disorders 4,376 (18.2) 19,167 (1.6) Personality Disorders 2,018 (8.4) 2,024 (0.2) ADHD 683 (2.8) 2,168 (0.2) PTSD 1,111 (4.62) 1,436 (0.1) Pain Conditions 17,115 (71.1) 412,797 (34.0) <0.0001 Joint pain 14,461 (60.1) 372,926 (30.7) Back pain 6,312 (38.7) 144,770 (11.9) Chronic pain 5,350 (22.2) 23.865 (2.0) Neck pain 3,863 (16.1) 48,842 (4.0) Head pain 1,617 (6.7) 10,313 (0.9) * Chi-square tests for differences in proportions were done for all pain and mental diagnoses. All p<0.0001 . 16

Outline • Objective and Methods • Results – Characteristics and Comorbidities – Pain and Mental Illness – Opioid Use and Opioid Overdose Hospitalizations – Subset Analysis • Discussion and Future Research 17

Results: Opioid Use and Opioid OD Hospitalization Among Medicare Beneficiaries with and without SUD 100 9 91.2 90 8 80.5 74.9 80 73.1 7 7.9 67.6 70 6 60 5 48.9 50 3.1 4 40 2.3 3 3.9 30 1.4 2 20 1 0.1 10 0 0 No SUD Any SUD Alcohol SUD Drug SUD Opioid SUD Drug and Alcohol SUD Opioids Opioid Overdose 18

Outline • Objective and Methods • Results – Characteristics and Comorbidities – Pain and Mental Illness – Opioid Use and Opioid Overdose Hospitalizations – Subset Analysis • Discussion and Further Research 19

Subset Analysis: Characteristics prior to Opioid Overdose Hospitalization – We conducted a subset analysis among individuals with a 2009 opioid OD hospitalization • Total: 1,225 individuals • Retrospective Assessment: 12 months prior to the overdose to examine SUD/mental illness/pain diagnoses and opioid use • Overdose includes fatal and non-fatal events • Same criteria from primary analysis 20