Optum's De-identified Clinformatics Research Articles

Increased Health Care Resource Utilization and Costs Associated with Herpes Zoster Among Patients Aged ≥50 Years with Chronic Obstructive Pulmonary Disease in the United States.

Patients with chronic obstructive pulmonary disease (COPD) are potentially at increased risk of herpes zoster (HZ). Little is known about the impact of an HZ episode on health care resource utilization (HRU) and costs among patients with COPD. This retrospective cohort study of individuals aged ≥50 years in the United States (US) used administrative claims data from Optum's de-identified Clinformatics Data Mart Database for commercially insured and Medicare Advantage members (2013-2018). Two cohorts of patients with COPD, with HZ (COPD+/HZ+) and without HZ (COPD+/HZ-), were identified. All-cause and COPD-related HRU rates and costs (2018 US dollars) were compared between cohorts for up to 12 months of follow-up. Comparisons were controlled for baseline differences through propensity score adjustment. A total of 3415 COPD+/HZ+ and 35,360 COPD+/HZ- patients (mean ages 73.2 ± 9.0 and 72.4 ± 9.4 years, respectively) were identified. Patients in the COPD+/HZ+ versus COPD+/HZ- cohort had increased use of all-cause (adjusted incidence rate ratio [aIRR] 1.17; 95% confidence interval [CI] 1.14, 1.21) and COPD-related (aIRR 1.27; 95% CI 1.21, 1.34) medical services (both P<0.001) and higher mean total all-cause ($4140 versus $3749 per person per month [PPPM]; adjusted cost difference +$313 PPPM) and COPD-related ($1541 versus $1231 PPPM; +$152 PPPM) costs (both P<0.004) in the year after HZ. HRU and cost burden is higher in patients with COPD with vs without HZ. These results could help to estimate the potential cost benefits of HZ vaccination among patients with COPD.

1388. DTaP-containing combination vaccines use and adherence to the recommended infant-toddler vaccination series among privately insured children in the US

BackgroundDespite universal recommendation of the 3 + 1 diphtheria, tetanus, and pertussis (DTaP) vaccine series in infants and toddlers, adherence (i.e. coverage and timeliness) remain suboptimal in the US. DTaP-containing combination vaccines are presumed to improve vaccine coverage rates and timeliness, but research on this topic is limited. The purpose of this study was to compare adherence to the recommended infant-toddler vaccination series between recipients of DTaP-containing combination vaccines (i.e. quadrivalent/pentavalent) and stand-alone vaccines (i.e. trivalent).MethodsWe used the Optum de-identified Clinformatics Data Mart database to create a cohort of children born between 2009 and 2016 with > 24 months of continuous enrollment from birth, and records of > 1 DTaP vaccine receipt. Patients were classified by DTaP-containing vaccine receipt: combination vaccines only, stand-alone vaccines only, or a mixture of both. The primary adherence outcome was completion of the 4-dose series within 20 months of life. We adjusted outcomes for gender, birth year, race, and socioeconomic status via a logistic regression model.ResultsThe cohort contained 200,568 female (48.6%) and 211,882 male (51.4%) children. Of these children, 167,091 received combination vaccines only (40.5%), 61,342 received stand-alone vaccines only (14.9%), and 184,017 received a mixture of both (44.6%). Completion of the 4-dose series was highest among children who received combination vaccines only (75.5%), followed by those who received a mixture of vaccines (72.7%) and those who received stand-alone vaccines only (54.5%). Relative to those who received stand-alone vaccines only, adjusted odds of completion were approximately 2.9 times higher among combination vaccine recipients (odds ratio, OR = 2.93 [95% CI: 2.87, 2.98]) and 2.5 times higher among those who received a mixture of vaccines (OR = 2.54 [2.49, 2.59]).ConclusionDTaP-containing combination vaccine use was associated with significantly greater adherence. Although these results warrant further investigation to better understand the determinants of infant vaccination adherence, such evidence may further support preferential recommendations for combination vaccine use.DisclosuresMatthew M. Loiacono, MSc, Sanofi Pasteur (Employee) Vitali Pool, MD, Sanofi Pasteur (Employee) Robertus Van Aalst, MSc, Sanofi Pasteur (Employee)

A Machine Learning Approach to Identify Predictors of Potentially Inappropriate Non-Steroidal Anti-Inflammatory Drugs (NSAIDs) Use in Older Adults with Osteoarthritis.

Evidence from some studies suggest that osteoarthritis (OA) patients are often prescribed non-steroidal anti-inflammatory drugs (NSAIDs) that are not in accordance with their cardiovascular (CV) or gastrointestinal (GI) risk profiles. However, no such study has been carried out in the United States. Therefore, we sought to examine the prevalence and predictors of potentially inappropriate NSAIDs use in older adults (age > 65) with OA using machine learning with real-world data from Optum De-identified Clinformatics® Data Mart. We identified a retrospective cohort of eligible individuals using data from 2015 (baseline) and 2016 (follow-up). Potentially inappropriate NSAIDs use was identified using the type (COX-2 selective vs. non-selective) and length of NSAIDs use and an individual’s CV and GI risk. Predictors of potentially inappropriate NSAIDs use were identified using eXtreme Gradient Boosting. Our study cohort comprised of 44,990 individuals (mean age 75.9 years). We found that 12.8% individuals had potentially inappropriate NSAIDs use, but the rate was disproportionately higher (44.5%) in individuals at low CV/high GI risk. Longer duration of NSAIDs use during baseline (AOR 1.02; 95% CI:1.02–1.02 for both non-selective and selective NSAIDs) was associated with a higher risk of potentially inappropriate NSAIDs use. Additionally, individuals with low CV/high GI (AOR 1.34; 95% CI:1.20–1.50) and high CV/low GI risk (AOR 1.61; 95% CI:1.34–1.93) were also more likely to have potentially inappropriate NSAIDs use. Heightened surveillance of older adults with OA requiring NSAIDs is warranted.

Long-term safety of drug-coated devices for peripheral revascularisation.

Meta-analyses of randomised trials of paclitaxel-coated peripheral devices found an association with worse long-term survival. We aimed to assess long-term mortality in patients treated with drug-coated versus non-drug-coated devices who are insured by Medicare Advantage (MA), an alternative to traditional Medicare that represents >30% of the Medicare eligible population. We analysed data from an MA administrative claims data source that includes both inpatient and outpatient femoropopliteal artery revascularisation procedures. Patients treated with or without drug-coated devices for femoropopliteal artery revascularisation from 4/2015-12/2017 were studied using Optum's De-identified Clinformatics Datamart Database. Mortality was assessed up to December 2019 using Kaplan-Meier cumulative mortality curves and Cox proportional hazard models. Inverse probability of treatment weighting was used to adjust for differences between groups. Of 16,796 patients revascularised, 4,427 (26.4%) were treated with drug-coated devices: 3,600 (81.3%) balloons and 827 (18.7%) stents. The median follow-up was 2.66 years (IQR 2.02-3.52). Treatment with drug-coated devices was associated with similar long-term mortality to non-drug-coated devices (adjusted HR 1.03, 95% CI: 0.96-1.10; p=0.39). Results were comparable for patients treated with balloons alone (adjusted HR 1.00, 95% CI: 0.92-1.08; p=0.96) or stents (adjusted HR 1.02, 95% CI: 0.88-1.18; p=0.78). These findings did not differ based on treatment setting, disease severity, age, sex or comorbidity burden (interaction p>0.05 for all). In this large cohort, there was no evidence of increased long-term mortality following treatment with drug-coated devices.

Trends in Urine Drug Testing Among Long-term Opioid Users, 2012–2018

Long-term opioid therapy increases the risk of opioid overdose death. Government agencies and medical societies, including the Center for Disease Control and Prevention and the American Society for Clinical Oncology, emphasized risk mitigation strategies, including urine drug testing, in published guidelines. Urine drug testing rates, time trends, and covariates among long-term opioid therapy users were examined to gauge guideline adherence. Using Optum's De-identified Clinformatics DataMart, an incidence cohort (n=28,790) and prevalence cohort (n=621,449) were created to measure baseline and annual urine drug testing, respectively, from 2012 to 2018. Urine drug testing time trends were evaluated by demographics, pain conditions, and Elixhauser comorbidity index. A multivariable generalized estimating model was developed in 2020 to examine the factors associated with urine drug testing. Annual urine drug testing rates doubled from 25.6% in 2012 to 52.2% in 2018, whereas baseline urine drug testing also increased from 3.75% to 11.1%. Annual urine drug testing increased within each age group over time; however, older patients (OR=0.21, 95% CI=0.21, 0.22, aged >79 years) and patients with cancer (OR=0.82, 95% CI=0.80, 0.84) were less likely to receive urine drug testing. Patients residing in the South (OR=1.99, 95% CI=1.96, 2.01) and those with back pain (OR=2.04, 95% CI=2.02, 2.06) or with other chronic pain (OR=1.64, 95% CI=1.62, 1.66) were significantly more likely to be tested. Independent predictors of baseline urine drug testing were similar to predictors of annual urine drug testing. Despite increasing urine drug testing trends from 2012 to 2018, annual and baseline urine drug testing remained low in 2018, relative to numerous guideline recommendations. Findings suggest a need for research on better guideline implementation strategies and the effectiveness of urine drug testing on patient outcomes.

Use of Anti-Infective Prophylaxis in Newly Diagnosed and Relapsed/Refractory Multiple Myeloma Patients Initiating Treatment with Daratumumab

Introduction: Infections continue to be one of the main causes of morbidity and mortality in multiple myeloma (MM) patients. Compared to the general population, MM patients have a 7-fold increased risk of bacterial infections and 10-fold risk of viral infections. The recent National Comprehensive Cancer Network (NCCN) guideline has provided guidance for prevention of cancer-related infections; however, utilization of anti-infective prophylaxis among MM patients in real-world practice has not been fully established. Daratumumab, a human IgGκ monoclonal antibody targeting CD38, is approved as monotherapy and in combination with standard-of-care regimens for newly diagnosed MM (NDMM) and relapsed/refractory MM (RRMM). The primary objective of the study was to describe the proportion of MM patients who received anti-infective prophylaxis when initiating a daratumumab-based treatment regimen. Methods: Medical and pharmacy claims from the Optum's de-identified Clinformatics® Data Mart Database were obtained for a cohort of US patients with NDMM or RRMM who initiated treatment with daratumumab (in combination with other antimyeloma agents or monotherapy) between 1/1/2017 and 12/24/2019. The index date was defined as the first daratumumab treatment date in the study period. Two steps were applied to identify anti-infective prophylaxis. First, use of anti-infective prophylaxis was assessed during the period between the week prior to and the week after the date of daratumumab initiation, including anti-virals for preventing herpesviruses (i.e., herpes zoster, herpes simplex, and cytomegalovirus), systemic anti-bacterials, systemic anti-fungals, Pneumocystis jiroveci pneumonia (PJP) prophylaxis, and intravenous/subcutaneous immune globulin (IVIG/SCIG). Second, use of an anti-infective drug was defined as prophylactic if there was no diagnosis for an infection treated by the anti-infective drug (i.e., patients with infection events one week prior to and one day after the start of drug prescription were excluded). Receipt of vaccines for seasonal influenza (in the year prior to daratumumab initiation) and herpes zoster and pneumococcal infections (at any time prior in the patient's claims history) were also assessed. Results: 929 newly daratumumab-treated patients were eligible for study inclusion; 92 had NDMM (median age 74 years; 48.9% female) and 837 had RRMM (median age 72 years; 46.0% female). At the time of daratumumab initiation, antibacterial prophylaxis was administered to 12.0% of NDMM and 17.0% of RRMM patients; antifungal prophylaxis to 1.1% of NDMM and 1.3% of RRMM patients; and PJP prophylaxis to 4.4% of NDMM and 10.5% of RRMM patients (Table 1). IVIG/SCIG prophylaxis was given to 1.1% of NDMM patients and 2.5% of RRMM patients. Herpesvirus prophylaxis was administered to 57.6% of NDMM and 69.4% of RRMM patients, and 17.4% of NDMM and 11.5% of RRMM patients received vaccination for herpes zoster. Seasonal influenza and pneumococcal vaccinations were given to 56.5% and 47.8% of NDMM patients and 62.5% and 51.3% of RRMM patients, respectively. Conclusion: Use of herpesvirus prophylaxis was approximately 70% in RRMM and 60% in NDMM patients at time of daratumumab initiation, with limited use of herpes zoster vaccination, suggesting that a meaningful minority of patients may not be protected against herpesvirus as recommended by NCCN guidelines for patients treated for MM. While rates of influenza and pneumococcal vaccination were higher than for herpes zoster, nearly half of patients did not have evidence of vaccination for these two pathogens in their claims data. These results indicate that a significant percentage of MM patients are not receiving care to optimally prevent potential viral infections. Further studies are needed to understand the potential benefit of infectious prophylaxis in the clinical management of MM patients. Disclosures Tai: Janssen Scientific Affairs: Current Employment. Ammann:Janssen Scientific Affairs: Current Employment, Current equity holder in publicly-traded company. Kaila:Janssen Scientific Affairs: Current Employment. Pericone:Janssen Scientific Affairs: Current Employment, Current equity holder in publicly-traded company. Singh:Mu Sigma: Current Employment, Other: Mu Sigma was contracted by Janssen Scientific Affairs to conduct the analyses for the present study. Lin:Janssen Scientific Affairs: Current Employment, Current equity holder in publicly-traded company. Davies:Sanofi: Honoraria, Membership on an entity's Board of Directors or advisory committees; Adaptive Biotech: Honoraria; Takeda: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Roche: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Oncopeptides: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene/BMS: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding.

Real-World Economic Burden Among Patients With And Without Heart Failure Worsening After Cardiac Resynchronization Therapy.

Although cardiac resynchronization therapy (CRT) has the potential to improve cardiac function in patients with heart failure (HF), a considerable portion of patients do not respond to therapy. This study assessed the economic burden among patients with and without HF worsening after receiving CRT in real-world practice. In this retrospective claims-based study using Optum's de-identified Clinformatics® Data Mart Database (January 2007-December 2018), adults who received CRT were stratified into two cohorts based on whether they showed evidence of HF worsening within 180days post-CRT implantation. Inverse probability of treatment weighting (IPTW) was used to adjust for confounding, accounting for demographics (e.g., age, sex), the Quan-Charlson Comorbidity Index, other clinical characteristics, healthcare resource utilization (HRU), and healthcare costs during the 180days pre-CRT (baseline period). Annualized all-cause and congestive HF-related HRU and healthcare costs from payer and patient perspectives were assessed from day 181 post-CRT (follow-up period), and compared between cohorts using incidence rate ratios (IRRs) and cost ratios (CRs). This study included 12,753 patients (n = 4785 with HF worsening; n = 7968 without). Mean age was 72years and roughly two-thirds were male. Baseline characteristics were balanced between cohorts post-IPTW. During follow-up, patients with HF worsening had significantly greater annual all-cause inpatient [adjusted IRR (95% confidence interval) = 1.55 (1.44, 1.66), p < 0.001], outpatient [adjusted IRR = 1.46 (1.32, 1.61), p < 0.001], and emergency department [adjusted IRR = 1.31 (1.22, 1.41), p < 0.001] visits. Mean annual total per patient payer-paid amounts were significantly higher for patients with HF worsening versus without HF worsening [adjusted CR = 1.68 (1.56, 1.80), p < 0.001]. Annual patient-paid medical costs were also higher for patients with HF worsening [adjusted CR = 1.31 (1.25, 1.38), p < 0.001]. Results were similar for congestive HF-related HRU and costs. The incremental economic burden among patients with HF worsening following CRT is substantial. Efforts aimed at CRT optimization may help reduce this burden.

Oral anticoagulant use in patients with atrial fibrillation and mitral valve repair

Patients with atrial fibrillation (AF) who have undergone mitral valve repair are at risk for thromboembolic strokes. Prior to 2019, only vitamin K antagonists were recommended for patients with AF who had undergone mitral valve repair despite the introduction of direct oral anticoagulants (DOAC) in 2010. To characterize the use of anticoagulants in patients with AF who underwent surgical mitral valve repair (sMVR) or transcatheter mitral valve repair (tMVR). We performed a retrospective cohort analysis of patients with AF undergoing sMVR or tMVR between 04/2014 and 12/2018 using Optum's de-identified Clinformatics® Data Mart Database. We identified anticoagulants prescribed within 90 days of discharge from hospitalization. Overall, 1997 patients with AF underwent valve repair: 1560 underwent sMVR, and 437 underwent tMVR. The mean CHA2DS2-VASc score among all patients was 4.1 (SD 1.9). The overall use of anticoagulation was unchanged between 2014 (72.2%) and 2018 (70.0%) (P = .49). Among patients who underwent sMVR or tMVR between April 2014 and December 2018, the use of VKA therapy decreased from 62.9% to 32.1% (P < .01 for trend) and the use of DOACs increased from 12.4% to 37.3% (P < .01 for trend). Among patients with AF who underwent sMVR or tMVR between 2014 and 2018, roughly 30% of patients were not treated with any anticoagulant within 90 days of discharge, despite an elevated stroke risk in the cohort. The rate of DOAC use increased steadily over the study period but did not significantly increase the rate of overall anticoagulant use in this high-risk cohort.

Delayed Ejaculation After Lumbar Spine Surgery: A Claims Database Analysis.

Study Design:Retrospective cohort.Objectives:Delayed ejaculation (DE) is a distressing condition characterized by a notable delay in ejaculation or complete inability to achieve ejaculation, and there are no existing reports of DE following lumbar spine surgery. Inspired by our institutional experience, we sought to assess national rates of DE following surgery of the lumbar spine.Methods:We queried the Optum De-identified Clinformatics Database for adult men undergoing surgery of the lumbar spine between 2003 and 2017. The primary outcome was the development of DE within 2 years of surgery. Multivariable logistic regression was performed to identify factors associated with the development of DE.Results:We identified 117 918 men who underwent 162 646 lumbar spine surgeries, including anterior lumbar interbody fusion (ALIF), posterior lumbar fusion (PLF), and more. The overall incidence of DE was 0.09%, with the highest rate among ALIF surgeries at 0.13%. In multivariable analysis, the odds of developing DE did not vary between anterior/lateral lumbar interbody fusion, PLF, and other spine surgeries. A history of tobacco smoking (OR = 1.47, 95% CI 1.00-2.16, P = .05) and obesity (OR = 1.56, 95% CI 1.00-2.44, P = .05) were associated with development of DE.Conclusions:DE is a rare but distressing complication of thoracolumbar spine surgery, and patients should be queried for relevant symptoms at postoperative visits when indicated.

Increased Mortality Among Men Diagnosed With Impaired Fertility: Analysis of US Claims Data

To determine whether male infertility or impaired spermatogenesis is associated with mortality. The Optum de-identified Clinformatics Data Mart database was queried from 2003 to 2017. Infertile men were compared to subjects undergoing semen analysis (ie, infertility testing). Infertile men with oligozoospermia or azoospermia were included. Mortality was determined by data linkage to the Social Security Administration Death Master File. Results were adjusted for age, smoking, obesity, year of evaluation, and health care visits as well as for most prevalent comorbidities. We separately examined men with prevalent or incident cardiovascular disease and cancer diagnoses to determine associations with mortality. A total of 134,796 infertile men and 242,282 controls were followed for a mean of 3.6 and 3.1 years respectively. Overall, infertile men had a higher risk of death (Hazard Ratio [HR]= 1.42, 95% CI: 1.27-1.60) The diagnosis of azoospermia was associated with a significantly increased risk of death (HR= 2.01, 95% CI: 1.60-2.53) with a higher trend among men with oligospermia (HR: 1.17, 95% CI: 0.92-1.49) compared to controls. Subanalysis was done excluding prevalent cardiovascular and malignant disease (alone and combined) showing similar hazard ratios. Male infertility is associated with a higher risk of mortality especially among azoospermic men. Prevalent disease (which is known to be higher among infertile men) did not explain the higher risk of death among infertile men. The implications for treatment and surveillance of infertile men require further study.

Outpatient Cost-effectiveness Study of Hyperkalemia Management

Introduction Patiromer (PAT) is a sodium free, non-absorbed potassium (K+) binder approved for the treatment of hyperkalemia (HK). There is limited real-world evidence on the outpatient cost implications associated with PAT treatment of HK. Hypothesis To assess the cost-effectiveness of treating HK with PAT vs. no K+ binder in a Medicare Advantage population. Methods This retrospective, matched cohort study was conducted using Optum's de-identified Clinformatics® Data Mart Database from 1/1/16-12/31/18. Two HK cohorts were identified: PAT exposed/unexposed (NoPAT). Inclusion criteria included pre-index serum K+ ≥5.0 mEq/L, HK diagnosis (ICD-10 code) and ≥6 months insurance enrollment post-index. Propensity score matching and coarsened exact matching with baseline variables were used to identify the complete set of matching unexposed and exposed HK episodes. Follow-up began on the index date and ended at the first censoring event (insurance disenrollment, death, 12/31/18, sodium polystyrene sulfonate [SPS] or sodium zirconium cyclosilicate [SZC] initiation, PAT discontinuation [exposed only], PAT initiation [unexposed only]). A zero-inflated negative binomial regression model was performed on the outcomes. Outpatient outcomes were defined as any costs not identified as inpatient, ED or pharmacy. Outpatient cost outcomes were measured at 6 months post-index (mean US$ [CI 95%]). Results The study population included 2004 patients (1002 matched pairs), the mean age was 74 years and 60% were male. Patients had a mean of 5 comorbidities, most commonly diabetes mellitus (73%), congestive HF (35%), and end-stage renal disease (10%). At 6 months post-index, 300 (150 matched pairs) PAT and NoPAT patients remained uncensored and, of those, 299 had non-zero outpatient costs evaluated; total PAT outpatient mean cost difference (savings) was $4828 ($2205, $5833; P ≤0.004) for PAT vs. NoPAT patients. ConclusionS In a matched-cohort design of individuals with hyperkalemia covered with Medicare Advantage, patiromer usage was associated with a 35% reduction in outpatient costs over a 6-month period compared to those that were not treated with patiromer. Further study is warranted to replicate these findings in a large cohort.

Cost sharing, postpartum contraceptive use, and short interpregnancy interval rates among commercially insured women

Increasing access to effective birth control after childbirth may meet many women's preferences and reduce short interpregnancy interval rates. Eliminating out-of-pocket costs for contraception has been reported to increase the use of the most effective methods among women with employer-based insurance, but the prevalence and effects of patient cost sharing for contraception have not been studied during the postpartum period. This study aimed to examine the association between cost sharing for long-acting reversible contraception and postpartum contraception use patterns and pregnancies in the 12 months after delivery. We conducted a retrospective cohort analysis of commercially insured women undergoing childbirth from 2014 to 2018 using Optum's (Eden Prairie, MN) de-identified Clinformatics Data Mart database. This large national database includes nonretired employees and their dependents who are enrolled in health insurance plans sponsored by large- or medium-sized US-based employers. Women with 12 months of continuous enrollment postpartum were included. Childbirth, pregnancy, and contraceptive method (female sterilization, long-acting reversible contraceptives, other hormonal methods, and no prescription method observed) were identified using claims data. Contraceptive use patterns were observed at 3, 6, and 12 months postpartum and adjusted for individual and plan characteristics. Median out-of-pocket costs were $0 for sterilization and other hormonal methods but nonzero for long-acting reversible contraception. We therefore used simple and multivariable logistic regressions to examine the association between plan-level cost sharing (no cost sharing, $0; low cost sharing, >$0-<$200; and high cost sharing, ≥$200 out-of-pocket cost) for any long-acting reversible contraceptive insertion and contraceptive use patterns and short interpregnancy interval rates, controlling for age, household income, race and ethnicity, region, and insurance plan type. Among 25,298 plans with cost sharing data, we identified 172,941 women with continuous enrollment for 12 months postpartum, including 82,500 (47.7%) in no cost sharing, 22,595 (13.1%) in low cost sharing, and 67,846 (39.2%) in high cost sharing plans. The percentage of postpartum women in the study sample using any prescription contraceptive method was 39.5% by 3 months, 43.8% by 6 months, and 46.0% by 12 months. At all time points, postpartum women in no cost sharing plans had a higher predicted probability of long-acting reversible contraceptive use (eg, at 12 months: no cost sharing, 22.0%; low cost-sharing, 17.5%; high cost sharing, 18.3%; P<.001) and a lower predicted probability of no prescription method use (eg, at 12 months: no cost sharing, 51.8%; low cost sharing, 55.0%; high cost sharing, 54.9%; P<.001) than those in low or high cost sharing plans. Predicted probabilities of female sterilization and other hormonal method use did not differ substantively by plan cost sharing for long-acting reversible contraception at any time point. The proportion of women experiencing a short interpregnancy interval was low (1.9% by 3 months, 1.9% by 6 months, 2.0% by 12 months) and did not differ by plan cost sharing for long-acting reversible contraception at any time point. Out-of-pocket costs for long-acting reversible contraception influence the method of contraception used by postpartum women with employer-based insurance. Eliminating financial barriers to long-acting reversible contraception access after childbirth may help women initiate their preferred method and increase the use of long-acting reversible contraceptives among interested women who otherwise might utilize less effective methods.

Resource Utilization in the First 2 Years Following Operative Correction for Tetralogy of Fallot: Study Using Data From the Optum's De-Identified Clinformatics Data Mart Insurance Claims Database.

BackgroundDespite excellent operative survival, correction of tetralogy of Fallot frequently is accompanied by residual lesions that may affect health beyond the incident hospitalization. Measuring resource utilization, specifically cost and length of stay, provides an integrated measure of morbidity not appreciable in traditional outcomes.Methods and ResultsWe conducted a retrospective cohort study, using de‐identified commercial insurance claims data, of 269 children who underwent operative correction of tetralogy of Fallot from January 2004 to September 2015 with ≥2 years of continuous follow‐up (1) to describe resource utilization for the incident hospitalization and subsequent 2 years, (2) to determine whether prolonged length of stay (>7 days) in the incident hospitalization was associated with increased subsequent resource utilization, and (3) to explore whether there was regional variation in resource utilization with both direct comparisons and multivariable models adjusting for known covariates. Subjects with prolonged incident hospitalization length of stay demonstrated greater resource utilization (total cost as well as counts of outpatient visits, hospitalizations, and catheterizations) after hospital discharge (P<0.0001 for each), though the number of subsequent operative and transcatheter interventions were not significantly different. Regional differences were observed in the cost of incident hospitalization as well as subsequent hospitalizations, outpatient visits, and the costs associated with each.ConclusionsThis study is the first to report short‐ and medium‐term resource utilization following tetralogy of Fallot operative correction. It also demonstrates that prolonged length of stay in the initial hospitalization is associated with increased subsequent resource utilization. This should motivate research to determine whether these differences are because of modifiable factors.

Undertreatment of primary hyperparathyroidism in a privately insured US population: Decreasing utilization of parathyroidectomy despite expanding surgical guidelines

BackgroundPrimary hyperparathyroidism is associated with substantial morbidity, including osteoporosis, nephrolithiasis, and chronic kidney disease. Parathyroidectomy can prevent these sequelae but is poorly utilized in many practice settings. MethodsWe performed a retrospective cohort study using the national Optum de-identified Clinformatics Data Mart Database. We identified patients aged ≥35 with a first observed primary hyperparathyroidism diagnosis from 2004 to 2016. Multivariable logistic regression was used to determine patient/provider characteristics associated with parathyroidectomy. ResultsOf 26,522 patients with primary hyperparathyroidism, 10,101 (38.1%) underwent parathyroidectomy. Of the 14,896 patients with any operative indication, 5,791 (38.9%) underwent parathyroidectomy. Over time, there was a decreasing trend in the rate of parathyroidectomy overall (2004: 54.4% to 2016: 32.4%, P < .001) and among groups with and without an operative indication. On multivariable analysis, increasing age and comorbidities were strongly, inversely associated with parathyroidectomy (age 75–84, odds ratio 0.50 [95% confidence interval 0.45–0.55]; age ≥85, odds ratio 0.21 [95% confidence interval 0.17–0.26] vs age 35–49; Charlson Comorbidity Index ≥2 vs 0 odds ratio 0.62 [95% confidence interval 0.58–0.66]). ConclusionThe majority of US privately insured patients with primary hyperparathyroidism are not treated with parathyroidectomy. Having an operative indication only modestly increases the likelihood of parathyroidectomy. Further research is needed to address barriers to treatment and the gap between guidelines and clinical care in primary hyperparathyroidism.

Risk of Nonunion with Nonselective NSAIDs, COX-2 Inhibitors, and Opioids.

Cyclooxygenase-2 (COX-2) has been found to be important for fracture-healing in animal models, raising concerns about use of nonsteroidal anti-inflammatory drugs (NSAIDs) and selective COX-2 inhibitors after fractures. We evaluated associations of NSAIDs, COX-2 inhibitors, and opioids with nonunion after long-bone fracture. Using private health insurance claims data from Optum's de-identified Clinformatics Data Mart database from January 1, 2000, to September 30, 2015, we identified adults with a single long-bone fracture or commonly paired long-bone fractures who had 1 year of available follow-up data. Using multivariable logistic regression models, we examined associations between NSAID, COX-2-inhibitor, or opioid prescription fills after the fracture and the risk of nonunion within 1 year, defined as a nonunion diagnosis with a procedure to treat the nonunion. A nonunion diagnosis with a procedure to treat the nonunion was identified after 2,996 (0.9%) of the 339,864 fracture episodes, with rates varying by fracture site. The risk of that outcome was greater in patients who had filled COX-2-inhibitor prescriptions (adjusted odds ratio = 1.84 [95% confidence interval = 1.38 to 2.46]) or opioid prescriptions (1.69 [1.53 to 1.86]), but not in patients who had filled nonselective-NSAID prescriptions (1.07 [0.93 to 1.23]) after the fracture. Results were similar when the outcome definition was changed to just a nonunion diagnosis. COX-2 inhibitors, but not nonselective NSAIDs, were associated with a greater risk of nonunion after fracture. Opioids were also associated with nonunion risk, although patients filling prescriptions for opioids may have had more severe fractures. Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

DIAGNOSTIC DELAY AND CHARACTERIZATION OF THE CLINICAL PRODROME IN AL AMYLOIDOSIS RELATED CARDIOMYOPATHY AMONG 1,173 PATIENTS DIAGNOSED BETWEEN 2001-2018

Prompt recognition and therapy can reverse light chain amyloidosis cardiomyopathy (AL-CM), but diagnosis is commonly delayed. We performed the first population-level study describing the diagnostic delay and clinical prodrome of AL-CM. Using the Optum De-Identified Clinformatics® Datamart

The Total Direct Cost of Healthcare in the United States in Patients with Sickle Cell Disease: A Propensity Score Matched Analysis

Introduction: Sickle cell disease (SCD) is an inherited blood disorder with significant healthcare resource utilization in the United States (US). The clinical management throughout the course of the disease is variable. Children born with SCD in the US receive routine comprehensive care that significantly reduces their morbidity and mortality. Upon transition to adulthood, there are marked increases in disease-related complications and morbidity. The extent to which disease management costs vary by age and overall lifetime cost of treatment are unknown. The aim of this study is to estimate the total direct healthcare cost among patients with SCD in the US across consecutive age ranges. Methods: We estimated direct costs of chronic management for SCD through a retrospective database analysis of Optum's de-identified Clinformatics® Data Mart Database (CDM) commercial insurance medical and pharmacy claims from January 1, 2007 to December 31, 2017. CDM provides information on 15 to 18 million annual insured lives and roughly 57 million unique lives over our study period (2007-2017). The population spans 50 states and is similar to the US distribution with respect to sex, age, and geography. We identified patients using ICD-9-CM (282.4x or 282.6x) and ICD-10-CM (D57.x sans D57.3x) diagnosis codes. Patients were required to have at least one inpatient (IP) claim or three or more outpatient (OP) claims (spaced at least 30 days apart) with a diagnosis of SCD and minimum of one year of continuous enrollment in both medical and prescription drug coverage. Each patient's index date was designated the date of their first SCD-related claim for patients with SCD, or their first date of eligibility for patients without SCD (controls). We matched patients with SCD to controls using 1:1 propensity score matching without replacement with a greedy nearest neighbor algorithm. Propensity scores were generated using probit regression on covariates gender, state, plan type, year of birth, and index year. Total direct costs of healthcare including IP, OP, professional (PF), ancillary (AN), and pharmacy (RX) services were calculated first year post-index (Y1), second year post-index (Y2), third year post-index (Y3), and overall (annualized) post-index. We calculated annualized costs by aggregating all costs after index and dividing by the remaining number of years the patient was eligible for coverage. Average incremental costs (AIC) are defined as the mean difference in costs between patients with SCD and respective controls. We report AIC by decade of age on the patient's index date, inflated to 2018 US dollars. All analyses were performed in SAS 9.4 (SAS Institute, Raleigh, NC) and Stata 16.0 (StataCorp, College Station, TX). This study was approved by the University of Southern California's University Park Institutional Review Board (UPIRB). Results: We identified 10,447 patients with SCD along with 10,447 unique matched controls from 2007 to 2017. Average total costs (sum of IP, OP, PF, AN, and RX costs) for patients with SCD were greater at all age categories, relative to controls. Annualized AIC, i.e. costs attributable to management of SCD, ranged from $18,859 (SD = 44,716) for ages 0-9 to $43,586 (SD = 109,735) for ages 20-29 (Table 1). First-year AIC were consistently greater than second and third-year AIC across all age groups (Figure 1). For patients with SCD, AIC were greatest for patients in the 20-29 years band across all specifications. Details on AIC of SCD management across all consecutive age ranges are available in Table 1. Conclusions: The cost of disease management for patients with SCD in the US is substantial. We observe a rise in SCD-specific costs from childhood to early adulthood. This may reflect the higher risk of disease-associated complications and mortality seen during this period. Additionally, AIC first year after initial SCD-related claim are consistently higher compared to AIC in years two and three. This finding may be attributable to an increased utilization of healthcare services from previously underinsured patients with SCD. Future work should involve simulating estimated lifetime total direct cost of healthcare attributable to management of SCD for patients in the US. Disclosures No relevant conflicts of interest to declare.

Diagnostic Delay and Characterization of the Clinical Prodrome in AL Amyloidosis: Data from 1,313 US Commercially Insured Patients between 2006-2018

Introduction: Light chain (AL) amyloidosis is a rare clonal plasma cell disorder characterized by the organ deposition of amyloid derived from misfolded immunoglobulin light chains. The involvement of vital organs, especially the heart, results in severe morbidity and high early mortality. Diagnostic delays of greater than 1 year from clinical evidence of AL have been reported in surveys and retrospective single-center studies. Earlier diagnosis prior to developing advanced organ dysfunction is a key unmet need in AL. To detect patients earlier, a better understanding of the patterns of AL-related signs/symptoms (AL-S/Sx) and healthcare utilization preceding AL amyloidosis diagnosis is necessary. Methods: Using the Optum De-Identified Clinformatics® Data Mart healthcare claims database (Optum), we identified US commercially-insured patients who had one inpatient code or two outpatient codes of ICD-9-CM 277.30 or ICD-10-CM E85.81, E85.89, or E85.9 and treatment with one multiple myeloma drug within 90 days of diagnosis (01Jan06-31Dec18). Twenty-three AL-S/Sx were defined with sets of diagnosis codes, flagged in prior condition lists, and grouped by system affected (cardiac, nervous, renal/ureter, gastrointestinal, or other). The time between the first occurrence of each symptom and AL diagnosis was summarized and healthcare utilization patterns were described to better characterize the patient journey to diagnosis. Results: Among 1313 AL patients, the median time from the first potential AL-S/Sx to AL diagnosis was 2.5 years. The most common AL-S/Sx were malaise/fatigue (53%), edema and swelling (57%), dyspnea (53%), and abdominal pain (52%), which occurred a median of 1-1.8 years prior to diagnosis. Arrhythmia (51%), cardiomegaly (27%), and cardiomyopathy (21%) were identifiable between a median of 1-1.5 years before diagnosis. Chronic kidney disease (45%) and edema (57%) were seen a median of 2.3 years prior to diagnosis. Patients with a first symptom of nephrotic syndrome presented less than 1 year before diagnosis. Peripheral nerve disease (46%), neuralgias (14%), and paresthesia/anesthesia (20%) appeared a median of 2.8 years prior to diagnosis. The cumulative incidence of each AL-S/Sx by organ system in the 5 years prior to diagnosis is shown in Figure 1. Sequencing of symptoms varied greatly among patients. The most common first symptoms in the sequences were abdominal pain, malaise/fatigue/weakness, peripheral nerve disease, and arrhythmia. Between initial AL-S/Sx and diagnosis, 20% of patients had 3 or more ER visits and 16% had 3 or more hospitalizations. More than 90% of patients visited 5 or more provider types between the first AL-S/Sx and AL diagnosis, with the most frequent specialties visited being cardiology (67%), hematology/oncology (43%), and gastroenterology (40%). About one quarter of patients had a monoclonal gammopathy detected at a median of 1 year prior to diagnosis; detailed analyses of this cohort will be presented at the meeting. Conclusion: AL patients have a prolonged prodrome, including evidence of cardiac and renal dysfunction, with 50% of patients with a first AL-S/Sx observed more than 2.5 years before AL diagnosis. Only one-quarter of patients had a monoclonal protein detected before diagnosis, indicating a low level of suspicion for AL. Limitations of this study include underestimation of AL S/Sx if diagnosis codes are underutilized in claims. Future studies will use controls to understand background rates of symptoms. Overall, these findings suggest that novel approaches to early diagnosis have significant potential to improve outcomes in AL. Figure 1. Cumulative incidence of the first occurrence of potential symptoms prior to the first light chain (AL) amyloidosis diagnosis Figure 1 Disclosures Hester: Janssen R&amp;D, LLC: Employment, Equity Ownership. Gifkins:Janssen R&amp;D, LLC: Employment. Bellew:Janssen R&amp;D, LLC: Employment. Vermeulen:Janssen R&amp;D, LLC: Employment, Equity Ownership. Schecter:Janssen R&amp;D, LLC: Employment, Equity Ownership. DeFalco:Janssen R&amp;D, LLC: Employment, Equity Ownership. Weiss:Janssen R&amp;D, LLC: Employment.

Association of Race and Ethnicity with Vascular Access Type Selection and Outcomes

Black patients with end-stage renal disease (ESRD) represent 30.5% of the prevalent ESRD population in the United States, despite only accounting for 18% of the total population. Black patients are less likely to have pre-ESRD care compared with their white counterparts and are 3-4 times more likely to progress from chronic kidney disease to ESRD than whites, suggesting that black patients are particularly vulnerable to disparities in outcomes related to hemodialysis and ESRD. The objective of this study is to examine the association of race with outcomes of hemodialysis access and selection of arteriovenous fistula (AVF) versus arteriovenous graft (AVG). Patients with chronic kidney disease who initiated dialysis through a tunneled hemodialysis catheter (THC) were identified in the Optum's De-identified Clinformatics® Data Mart (OptumInsight, Eden Prairie, MN) claims database (2011-2017). The odds of AVF versus AVG creation and the odds of repeat vascular access creation were analyzed using logistic regression. Time from initial AVF/AVG to THC removal and time to repeat AVF/AVG were analyzed using Cox proportional hazards. About 7,584 vascular access patients met the inclusion criteria: 5,852 (77%) AVF and 1,732 (23%) AVG. Median follow-up was 583days overall (range, 1-2,543), 589days among AVF patients (range, 1-2,543), and 260days among AVG patients (range, 1-2,529). Between races, there was no clinically significant variation in characteristics or comorbidities, with the exception of a much lower rate of obesity among Asians. Black patients had 36% lower odds of AVF index versus AVG index (P<0.001). Patients 70 years or older and patients with diabetes had lower odds of AVF index, whereas men and obese patients had greater odds of receiving AVF. Overall, graft patients were 73% more likely to have a shorter time to THC removal than fistula patients, but Hispanic graft patients were 25% more likely to have a shorter time to THC removal than whites. Patients with diabetes, patients with cardiac arrhythmia, and obesity were more likely to have a longer time to THC removal. About 1,589 (21%) patients underwent a repeat vascular access creation during the follow-up period: 19% of whites (n=802), 26% of blacks (n=483), 19% of Hispanics (n=250), and 19% of Asians (n=54) (P<0.001). Multivariate analysis demonstrated that black patients had 58% greater odds of requiring a second access than white patients (P<0.001). Graft patients, patients 70 years or older, and men had lower odds of repeat access. Black patients were 45% more likely to have a shorter time until second access creation. Graft patients, patients aged 70 years or older, and men were more likely to have a longer time until second access. Patients with obesity were more likely to have a shorter time until second access. This study's findings suggest that after initial vascular access, compared with whites, blacks have no difference in time to index access success, but their access fails earlier and more frequently, independent of access type, age, and comorbidities. Given blacks constitute 30.5% of the hemodialysis population in the United States, it is imperative that future research investigate the root causes of these disparities.

Trends in EMR for nonmalignant colorectal polyps in the United States

Although most large nonpedunculated colorectal lesions can be safely and efficaciously removed using EMR, the use of colectomy for benign colorectal lesions appears to be increasing. The reason(s) is unclear. We aimed to determine the use and adverse events of EMR in the United States. We used Optum's de-identified Clinformatics Data Mart Database (2003-2016), a database from a large national insurance provider, to identify all colonoscopies performed with either EMR or simple polypectomy on adult patients from January 1, 2011 to December 31, 2015. We measured time trends, regional variation, and adverse event rates. We assessed risk factors for adverse events using multivariate logistic regression. The rate of EMR use in the US increased from 1.62% of all colonoscopies in 2011 to 2.48% of colonoscopies in 2015 (P< .001). There were, however, significant regional differences in the use of EMRs, from 2.4% of colonoscopies in the western United States to 2.0% of colonoscopies in the southern United States. Between 2011 and 2015, we found stable rates of perforation, GI bleeding (GIB), infections, and cardiac adverse events and decreasing rates of admissions after EMR. In our multivariate model, EMR was an independent risk factor for adverse events, albeit the rates of adverse events were low (1.35% GIB, .22% perforation). Use of EMR is rising in the United States, although there is significant regional variation. The rates of adverse events after EMR and polypectomies were low and stable, confirming the continued safety of EMR procedures. A better understanding of the regional barriers and facilitators may improve the use of EMR as the standard management for benign colorectal lesions throughout the United States.