Papers by Hsien-yen Chang
BMC Health Services Research, 2019
Background: Overuse is a leading contributor to the high cost of health care in the United States... more Background: Overuse is a leading contributor to the high cost of health care in the United States. Overuse harms patients and is a definitive waste of resources. The Johns Hopkins Overuse Index (JHOI) is a normalized measure of systemic health care services overuse, generated from claims data, that has been used to describe overuse in Medicare beneficiaries and to understand drivers of overuse. We aimed to adapt the JHOI for application to a commercially insured US population, to examine geographic variation in systemic overuse in this population, and to analyze trends over time to inform whether systemic overuse is an enduring problem. Methods: We analyzed commercial insurance claims from 18 to 64 year old beneficiaries. We calculated a semiannual JHOI for each of the 375 Metropolitan Statistical Areas and 47 rural regions of the US. We generated maps to examine geographic variation and then analyzed each region's change in their JHOI quintile from January 2011 to June 2015. The JHOI varied markedly across the US. Across the country, rural regions tended to have less systemic overuse than their MSA counterparts (p < 0.01). Regional systemic overuse is positively correlated from one time period to the next (p < 0.001). Between 2011 and 2015, 53.7% (N = 226) of regions remained in the same quintile of the JHOI. Eighty of these regions had a persistently high or persistently low JHOI throughout study duration. The systemic overuse of health care resources is an enduring, regional problem. Areas identified as having a persistently high rate of systemic overuse merit further investigation to understand drivers and potential points of intervention.
Drug and Alcohol Dependence, 2019
Background: Predicting which individuals who are prescribed buprenorphine for opioid use disorder... more Background: Predicting which individuals who are prescribed buprenorphine for opioid use disorder are most likely to experience an overdose can help target interventions to prevent relapse and subsequent consequences. Methods: We used Maryland prescription drug monitoring data from 2015 to identify risk factors for nonfatal opioid overdoses that were identified in hospital discharge records in 2016. We developed a predictive risk model for prospective nonfatal opioid overdoses among buprenorphine patients (N = 25,487). We estimated a series of models that included demographics plus opioid, buprenorphine and benzodiazepine prescription variables. We applied logistic regression to generate performance measures.
JAMA Network Open, 2018
IMPORTANCE Expanding Medicaid eligibility could affect prescriptions of buprenorphine with naloxo... more IMPORTANCE Expanding Medicaid eligibility could affect prescriptions of buprenorphine with naloxone, an established treatment for opioid use disorder, and opioid pain relievers (OPRs). OBJECTIVE To examine changes in prescriptions of buprenorphine with naloxone and OPRs after the US Affordable Care Act Medicaid expansion. DESIGN, SETTING, AND PARTICIPANTS In this cohort study, longitudinal, patient-level, retail pharmacy claims were extracted from IQVIA real-world data from an anonymized, longitudinal, prescription database. The sample included 11.9 million individuals who filled 2 or more prescriptions for a prescription opioid during at least 1 year between January 1, 2010, and December 31, 2015, from California, Maryland, and Washington (expansion states) and Florida and Georgia (nonexpansion states). Data analysis was conducted from August 1, 2017, to May 31, 2018. Data were aggregated to county-year observations (N = 2082) and linked to county-level covariates. For each outcome, a difference-indifferences regression model was estimated comparing changes before and after expansion in expansion vs nonexpansion counties. Models were adjusted for county demographics, uninsured rate, and overdose mortality in the baseline year (2010). EXPOSURES Presence of Medicaid expansion in the year. MAIN OUTCOMES AND MEASURES For buprenorphine with naloxone and OPRs, rates per 100 000 county residents were calculated separately for any prescriptions overall and by different payment sources. Mean days of medication per county among people filling prescriptions for these agents were also determined. RESULTS The study sample included 11.9 million individuals (expansion states: 40.9% men; mean [SD] age, 44.1 [13.8] years; nonexpansion states: 41.0% men; mean [SD] age, 43.7 [13.7] years). In expansion counties, 68.8 individuals per 100 000 county residents filled buprenorphine with naloxone and 5298.3 filled OPR prescriptions in 2010. After expansion, buprenorphine with naloxone fills per 100 000 county residents increased significantly in expansion relative to nonexpansion counties (8.7; 95% CI, 1.7 to 15.7). Opioid pain reliever fills per 100 000 county residents did not significantly change in expansion counties relative to nonexpansion counties (327.4; 95% CI −202.5 to 857.4). The rate of OPRs per 100 000 county residents paid for by Medicaid significantly increased (374.0; 95% CI, 258.3 to 489.7). There were no significant changes in days per 100 000 county residents of either medication after expansion. CONCLUSIONS AND RELEVANCE Medicaid expansion significantly increased buprenorphine with naloxone prescriptions per 100 000 county residents in expansion counties, suggesting that expansion improved access to opioid use disorder treatment. Expansion did not significantly increase (continued) Key Points Question Did Medicaid expansion under the US Affordable Care Act change prescription fills for buprenorphine with naloxone, a treatment for opioid use disorder, and opioid pain relievers? Findings In this cohort study using difference-indifferences analysis of all-payer prescription fill data from 5 states, Medicaid expansion was associated with a significant overall increase in people filling prescriptions for buprenorphine with naloxone. Expansion was not associated with changes in fills per 100 000 county residents of opioid pain relievers overall, but significantly more people filled prescriptions for opioid pain relievers paid for specifically by Medicaid. Meaning Medicaid expansion may increase the role of states in providing opioid use disorder treatment and in paying for opioid pain relievers for pain management.
Journal of Patient-Centered Research and Reviews, 2017
PLOS ONE, 2016
Background Improving continuity between primary care and cancer care is critical for improving ca... more Background Improving continuity between primary care and cancer care is critical for improving cancer outcomes and curbing cancer costs. A dimension of continuity, we investigated how regularly patients receive their primary care and surgical care for colon cancer from the same hospital and whether this affects mortality and costs. Methods Using Surveillance, Epidemiology, and End Results Program Registry (SEER)-Medicare data, we performed a retrospective cohort study of stage I-III colon cancer patients diagnosed between 2000 and 2009. There were 23,305 stage I-III colon cancer patients who received primary care in the year prior to diagnosis and underwent operative care for colon cancer. Patients were assigned to the hospital where they had their surgery and to their primary care provider's main hospital, and then classified according to whether these two hospitals were same or different. Outcomes examined were hazards for all-cause mortality, subhazard for colon cancer specific mortality, and generalized linear estimate for costs at 12 months, from propensity score matched models. Results Fifty-two percent of stage I-III colon patients received primary care and surgical care from the same hospital. Primary care and surgical care from the same hospital was not associated with reduced all-cause or colon cancer specific mortality, but was associated with lower inpatient, outpatient, and total costs of care.
International journal for quality in health care : journal of the International Society for Quality in Health Care / ISQua, Jan 17, 2016
Process quality measure performance has improved significantly with public reporting, requiring r... more Process quality measure performance has improved significantly with public reporting, requiring reevaluation of process-outcome relationships and the emerging role of patient perspectives on care. To evaluate associations between heart failure patient perspectives of care and publicly reported processes and outcomes. Cross-sectional study, July 2008-June 2011. US hospitals in the Press Ganey database. Heart failure inpatients. Outcomes were Hospital Compare hospital-level risk-adjusted 30-day heart failure mortality and readmissions. Predictors included Hospital Compare heart failure processes of care, a weighted process composite and Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) domains for heart failure. Hospital characteristics included volume of heart failure patients and race, health status and education. Among 895 included hospitals, performance on process measures was high (median by hospital for composite, 95.6%); the median HCAHPS overall rating ...
Journal of oncology practice / American Society of Clinical Oncology, May 1, 2016
Readmissions to a different hospital may place patients at increased risk for poor outcomes and m... more Readmissions to a different hospital may place patients at increased risk for poor outcomes and may increase their overall costs of care. We evaluated whether mortality and costs differ for patients with colon cancer on the basis of whether patients are readmitted to the index hospital or to a different hospital within 30 days of discharge. We conducted a retrospective analysis using SEER-Medicare linked claims data for patients with stage I to III colon cancer diagnosed between 2000 and2009 who were readmitted within 30 days (N = 3,399). Our primary outcome was all-cause mortality, which was modeled by using Cox proportional hazards. Secondary outcomes included colon cancer-specific mortality, 90-day mortality, and costs of care. We used subhazard ratios for colon cancer- specific mortality and generalized linear models for costs. For each model, we used a propensity score-weighted doubly robust approach to adjust for patient, physician, and hospital characteristics. Approximately ...
Drug and Alcohol Dependence, 2016
Background-States have attempted to reduce prescription opioid abuse through strengthening the re... more Background-States have attempted to reduce prescription opioid abuse through strengthening the regulation of pain management clinics; however, the effect of such measures remains unclear. We quantified the impact of Texas's September 2010 "pill mill" law on opioid prescribing and utilization. Methods-We used the IMS Health LRx LifeLink database to examine anonymized, patientlevel pharmacy claims for a closed cohort of individuals filling prescription opioids in Texas * Corresponding author. [email protected] (T. Lyapustina). Contributors Tatyana Lyapustina was responsible for design, analysis, and manuscript preparation. Hsien-Yen Chang and Alim F. Ramji were responsible for statistical support. Matthew Daubresse was responsible for data acquisition. Lainie Rutkow, Mark Faul, and Elizabeth A. Stuart were responsible for critical revision. G. Caleb Alexander was responsible for project oversight and critical revision. All authors have approved the final manuscript and have approved of its submission. Conflict of interest Dr. Alexander is Chair of the FDA's Peripheral and Central Nervous System Advisory Committee; serves as a paid consultant to PainNavigator, a mobile startup to improve patient self-management of pain; serves as a paid consultant to IMS Health; and serves on an IMS Health scientific advisory board. This arrangement has been reviewed and approved by Johns Hopkins University in accordance with its conflict of interest policies.
Journal of managed care & specialty pharmacy, 2015
Dabigatran is a direct thrombin inhibitor approved by the FDA in October 2010 for the treatment o... more Dabigatran is a direct thrombin inhibitor approved by the FDA in October 2010 for the treatment of nonvalvular atrial fibrillation. Little is known regarding patient adherence to this therapy. To examine adherence and persistence to dabigatran among adults with atrial fibrillation. We used IMS Health's LifeLink Health Plan Claims Database from 2010 to 2012 to identify patients with atrial fibrillation who were new users of dabigatran. We derived adherence and persistence for continuously enrolled patients at 6 months, 9 months, and 12 months of follow-up. We measured adherence using the medication possession ratio (MPR), defined as individuals with MPRs of 0.80 or greater as adherent, and examined persistence by identifying individuals with gaps in drug possession of 60 days or greater. Of 5,951 adults with atrial fibrillation who were new users of dabigatran, 49% had prevalent atrial fibrillation and at least 6 months of continuous follow-up. Of these, 89% used dabigatran as ...
The Journal of ambulatory care management
Approximately 7 of 10 (and 95% of the elderly) people in US health plans see one or more speciali... more Approximately 7 of 10 (and 95% of the elderly) people in US health plans see one or more specialists in a year. Controlling for extent of morbidity, discontinuity of primary care physician visits is associated with seeing more different specialists. Having a general internist as the primary care physician is associated with more different specialists seen. Controlling for differences in the degree of morbidity, receiving care from multiple specialists is associated with higher costs, more procedures, and more medications, independent of the number of visits and age of the patient.
Pharmacoepidemiology and drug safety, 2015
We quantified the degree to which the August 2010 reformulation of abuse-deterrent OxyContin affe... more We quantified the degree to which the August 2010 reformulation of abuse-deterrent OxyContin affected its use, as well as the use of alternative extended-release and immediate-release opioids. We used the IMS Health National Prescription Audit, a nationally representative source of prescription activity in the USA, to conduct a segmented time-series analysis of the use of OxyContin and other prescription opioids. Our primary time period of interest was 12 months prior to and following August 2010. We performed model checks and sensitivity analyses, such as adjusting for marketing and promotion, using alternative lag periods, and adding extra observation points. OxyContin sales were similar before and after the August 2010 reformulation, with approximately 550 000 monthly prescriptions. After adjusting for declines in the generic extended-release oxycodone market, the formulation change was associated with a reduction of approximately 18 000 OxyContin prescription sales per month (p ...
Medical Care, 2015
Overuse can be defined as use of a service when the risk of harm exceeds its likely benefit. Yet,... more Overuse can be defined as use of a service when the risk of harm exceeds its likely benefit. Yet, there has been little work with composite measures of overuse. Our goal was to create a composite measure of overuse with claims data. Observational study using 5% of Medicare claims from 2008. All inpatient and outpatient settings of care, excluding nursing homes. Older Americans receiving health care services in hospitals or outpatient settings. We applied algorithms to identify specific cases of overuse across 20 previously identified procedures and used multilevel modeling techniques to examine variation in overuse across all procedures. Included in the model were patient-level factors and both procedure and regional fixed effects for the 306 hospital referral regions (HRR). These estimated regional fixed effects, representing the systematic, region variation in overuse across all measures, was then normalized compared with the overall average to generate a Z score for each HRR. The resulting &amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;quot;Overuse Index&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;quot; was then compared with total costs, 30-day postdischarge mortality, and total mortality at the HRR level, graphically, and associations were tested using Spearman ρ. The Overuse Index varied markedly across regions, but 23 were higher than the average (P&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;0.05). The Index was positively associated with total costs (ρ=0.28, P&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;0.0001). It was positively correlated with 30-day postdischarge mortality (ρ=0.18 P≤0.005), and neither positively or negatively correlated with total mortality. This study confirms previous research hypothesizing that systematic regional variation in overuse exists and is measurable. Addition research is needed to validate index and to test its predictive and concurrent validity in panel data.
The American journal of managed care, 2013
Because laboratory test results are less available to researchers than claims data, a claims-base... more Because laboratory test results are less available to researchers than claims data, a claims-based indicator of diabetes improvement would be valuable. To determine whether a decrease in medication use for diabetes parallels clinical improvement in glycemic control. This was a retrospective cohort study using up to 3.5 years of pharmacy and laboratory data from 1 private insurer. Data included 104 patients with diabetes who underwent bariatric surgery and had at least 1 glycated hemoglobin (A1C) test before and after surgery. We assigned each A1C test to a 90-day interval before or after surgery. Medication availability was noted for the midpoint of the interval (on insulin, on oral medications, count of medications). Each subject could contribute 1 presurgery and up to 3 postsurgery observations. We recorded the changes in A1C test results and medication use from the presurgery to the postsurgery period. Using the A1C test as the reference standard, positive and negative predictive...
JAMA Surgery, 2013
Bariatric surgery is a well-documented treatment for obesity, but there are uncertainties about t... more Bariatric surgery is a well-documented treatment for obesity, but there are uncertainties about the degree to which such surgery is associated with health care cost reductions that are sustained over time. Objective: To provide a comprehensive, multiyear analysis of health care costs by type of procedure within a large cohort of privately insured persons who underwent bariatric surgery compared with a matched nonsurgical cohort. Design: Longitudinal analysis of 2002-2008 claims data comparing a bariatric surgery cohort with a matched nonsurgical cohort. Setting: Seven BlueCross BlueShield health insurance plans with a total enrollment of more than 18 million persons.
Surgery for Obesity and Related Diseases, 2012
Background: Type 2 diabetes mellitus is highly prevalent in obese individuals. Bariatric surgery,... more Background: Type 2 diabetes mellitus is highly prevalent in obese individuals. Bariatric surgery, promoted for reducing the medical problems of morbid obesity, has been increasingly recognized for its particular efficacy in treating diabetes. However, before bariatric surgery can be recommended for the treatment of diabetes, its safety in the diabetic population must be known. We assessed the odds of complications after bariatric surgery in patients with and without diabetes. Methods: This was a retrospective cohort study. Using an administrative database from 7 Blue Cross/Blue Shield plans, we identified 22,288 subjects who had undergone bariatric surgery from 2002 to 2008. From this cohort, we selected 6754 pairs of surgical patients (1 with and 1 without diabetes) matched by age, gender, health plan, and year of surgery. With conditional logistic regression analysis, we determined the relative odds of postoperative complications for Յ12 months after surgery in the 2 groups. Results: The mean age of the surgical patients was 46 years, and 79% were women. Postoperative complications were rare and comparable in those with and without diabetes. The most common complications were nausea, vomiting, and abdominal pain (8.8%), the need for a gastric revision procedure (5.0%), and upper endoscopy (2.3%). Select cardiac, infectious, and renal complications occurred more frequently in the diabetic group. The incidence of cardiac complications was greater in the 2-3-month and 4-6-month postoperative periods (odds ratio [OR] 1.7, P Ͻ .001), the incidence of infectious complications was greater in the 0-1-month (OR 1.3, P Ͻ .02) and 4-6-month (OR 1.8, P Ͻ .001) periods, and the incidence of renal complications was greater in the 2-3-month postoperative period (OR 4.6, P ϭ .01). Conclusions: Our findings support the safety of bariatric surgery in obese individuals with diabetes, although management strategies to avert postoperative cardiac, infectious, and renal complications in this population might be warranted.
The Laryngoscope, 2014
Objectives/Hypothesis: To examine associations between quality of care, survival, and costs in el... more Objectives/Hypothesis: To examine associations between quality of care, survival, and costs in elderly patients treated for laryngeal squamous cell cancer (SCCA). Study Design: Retrospective analysis of Surveillance, Epidemiology, and End Results Medicare data. Methods: We evaluated 2,370 patients diagnosed with laryngeal SCCA from 2004 to 2007 using multivariate regression and survival analysis. Using quality indicators derived from guidelines for recommended care, summary measures of quality were calculated for diagnosis, initial treatment, surveillance, treatment of recurrence, end-of-life care, performance, and an overall summary measure of quality. Results: High-quality care was associated with significant differences in survival for diagnosis [HR 5 0.80, 95% CI (0.66-0.97)], initial treatment [HR 5 0.75 (0.63-0.88)], surveillance [HR 5 0.54 (0.44-0.66)], treatment of recurrence [HR 5 1.54 (1.26-1.89)], end-of-life care [HR 5 0.69 (0.52-0.92)], performance [HR 5 0.41 (0.33-0.52)], and an overall summary measure of quality [HR 5 0.66 (0.54-0.80
Obesity Surgery, 2012
Additional Contributions We thank the Blue Cross and Blue Shield plans and the many staff members... more Additional Contributions We thank the Blue Cross and Blue Shield plans and the many staff members at these sites who actively contributed to this study by providing data and expert advice regarding use of these data.
Obesity Surgery, 2013
Introduction-Micronutrient deficiencies are key concerns after bariatric surgery. We describe the... more Introduction-Micronutrient deficiencies are key concerns after bariatric surgery. We describe the prevalence of perioperative testing and diagnosis of micronutrient deficiencies among a cohort of insured bariatric surgery patients. Methods-We used claims data from seven health insurers to identify bariatric surgery patients from 2002-2008. Our outcomes were perioperative claims for vitamin D, B12, folate, and iron testing and diagnosed deficiencies. We analyzed results by bariatric surgery type: Roux-en-Y gastric bypass (RYGB), restrictive, and malabsorptive. We calculated the prevalence of testing and deficiency diagnosis, and performed multivariate logistic regression to determine the association with surgery type. Results-Of 21,345 eligible patients, 84% underwent RYGB. The pre-surgical testing prevalence for all micronutrients was <25%. The testing prevalence during the first 12 months after surgery varied: vitamin D (12%), vitamin B12 (60%), folate (47%) and iron (49%), and
Medical Care Research and Review, 2013
Health care overuse contributes to unnecessary expenditures and patient exposure to harm. Underst... more Health care overuse contributes to unnecessary expenditures and patient exposure to harm. Understanding and addressing this problem requires a comprehensive set of valid metrics. This article describes and critiques the current state of overuse measurement through a review of the published and gray literature, measures clearinghouses and ongoing work by major measure developers. Our review identified 37 fully specified measures and 123 measurement development opportunities. Many services were considered overuse due to the extension of diagnostic or screening services to low-risk populations. There were more diagnostic or therapeutic overuse measures than for screening or monitoring/surveillance. Imaging services is a major focus of current measures, but opportunities exist to expand overuse measurement in medication, laboratory services. Future development of overuse measures would benefit from new empirical research and clinical guidelines focused on identifying indications or popu...
Uploads
Papers by Hsien-yen Chang