MEDICATION ADHERENCE: DOES ADHERENCE TO SOME THERAPEUTIC CLASSES IMPROVE ADHERENCE TO OTHERS?

MEDICATION ADHERENCE: DOES ADHERENCE TO SOME THERAPEUTIC CLASSES IMPROVE ADHERENCE TO OTHERS?

A274 VA L U E I N H E A LT H 1 9 ( 2 0 1 6 ) A 1 - A 3 1 8 day with the key words of drug, price and pricing. Collected data were verified by two ...

76KB Sizes 0 Downloads 20 Views

A274

VA L U E I N H E A LT H 1 9 ( 2 0 1 6 ) A 1 - A 3 1 8

day with the key words of drug, price and pricing. Collected data were verified by two researchers to ensure timeliness and usefulness. The monitoring lasted for a year from October 2012 to September 2013.  Results: A total of 560 pieces of valid information were extracted focusing on maximum retail price adjustment, zero markup, drug price regulation, essential drugs, different prices for the same drug, traditional Chinese medicine, centralized tender for drug purchase and corruption in pharmaceutical industry. Large differences existed between provinces on the amount of information. The amount of public opinion about any issues was closely linked with government policy of that time.  Conclusions: In the field of health service research, “Internet Public Opinion Monitoring” is an emerging concept, but it has been proved to be an effective tool for Chinese government to know the public reflection to policy. Future work should focus on connection of monitoring results and policy decision-making. PHP93 FACTORS AFFECTING ADOPTION OF COMPUTERIZED PHYSICIAN ORDER ENTRY SYSTEM Malhani M A , Maneno M K , Ettienne E B , Wingate L Howard University, Washington, DC, USA .

.

.

.

.

.

.

Objectives: To determine whether physician specialty and other factors affect adoption of computerized physician order entry (CPOE) system. Methods: A cross-sectional study using the 2006–2010 National Ambulatory Medical Care Survey (NAMCS). The primary outcome, CPOE adoption, was defined as a practice which had a fully functional CPOE. The data was analyzed at the physician level using weighted analyses given the survey’s multistage sampling design. All statistical analysis was conducted using SAS 9.3 at alpha of 0.05. Descriptive statistics for all study variables were estimated. Unadjusted and adjusted logistic regression analyses were used to evaluate predictive factors for CPOE adoption, controlling for geography, physician characteristics, practice characteristics, revenue characteristics, and physician degree.  Results: The weighted proportion of physicians who had adopted CPOE was 37%. Physician specialty was significantly associated with CPOE adoption (p=  0.005), after adjusting for covariates. Surgical and Medical specialties were less likely to adopt CPOE, compared to primary care specialty (p< 0.05). CPOE adoption was less likely among solo practices vs. non-solo practices (p < .0001), practices that submit claims as all paper vs. electronic (p < .0001), practices where physicians were owners vs. employees (p=  0.0008), practices which had less than or equal to 50% of revenue from private insurance vs. more than 50% of revenue from private insurance (p=  0.0367), and physicians who did not conduct internet or e-mail consults with patients (p < .0001). CPOE adoption did not significantly vary by metropolitan status, physician degree, geographic region, and practice ownership.  Conclusions: This study found that physician specialty, solo-practice status, ownership status, use of email/internet consultations and lower private insurance revenue stream were significant predictors of CPOE adoption. PHP94 BRAND NAME DRUG UTILIZATION AMONG MEDICAID PATIENTS IN 340B CLINICS Lee C 1, Chang J 2, Chou J 2, McCombs J 1 of Southern California, Los Angeles, CA, USA, 2PharMedQuest Pharmacy Services, Brea, CA, USA .

.

.

.

1University

Objectives: To determine if primary care physician’s (PCP) prescribing behavior varies across brand name and generic drugs depending on each patient’s eligibility for the Federal 340B program.  Methods: Prescription claims from 10 in-clinic pharmacies in Orange and Los Angeles counties were used to identify 63,384 non-HIV patients who filled 640,480 new prescriptions between 2013 and 2015. Patients were identified as ineligible for the clinic’s 340B program if their claims were processed by fee-for-service state Medicaid (FFS Medicaid). We use Logistic models to analyze the likelihood that a brand name drug was utilized controlling for 340B eligibility, patient characteristics and drug therapeutic class.  Results: PCPs were less likely to prescribe brand name drugs to 340B-eligible patients (OR= 0.537, 95% CI =  0.523-0.551) than FFS Medicaid patients who were ineligible for 340B pricing. The likelihood of brand name prescribing increased monotonically over age 35: 35~49 years (OR= 1.253, 95% CI= 1.227-1.281), 50~64 years (OR= 1.351, 95% CI= 1.324-1.378), 65+ years (OR= 1.709, 95% CI= 1.671-1.748). Males were more likely to receive brand name drugs (OR= 1.185, 95% CI= 1.169-1.200) than females. Patients lived in Orange County were more likely to receive brand name drugs than Los Angeles County (OR= 1.093, 95% CI= 1.071-1.114). Meanwhile, 340B eligible patients and other clinic patients exhibited a lower likelihood of receiving brand name drugs relative to FFS Medicaid patients in all drug classes.  Conclusions: 340B eligible patients are less likely to receive brand name drugs than FFS Medicaid patients. The brand-name drug utilization associated with geographic difference might be due to the different county-based pharmacy formularies in the respective health plans. The differences in generic options among drug classes also affect the brand-name drug utilization. PHP95 IMPACT OF ENHANCED RECOVERY AFTER SURGERY INTERVENTIONS ON PERIOPERATIVE OUTCOMES IN PATIENTS UNDERGOING TOTAL KNEE ARTHROPLASTY Aggarwal J 1, Park S 1, Dagenais S 2, Kang A 2, Menzin J 1 Health Economics, Waltham, MA, USA, 2Pacira Pharmaceuticals, Inc., Parsippany, NJ, USA .

.

.

.

.

1Boston

Objectives: Changes in payment systems for joint replacements have highlighted the need for hospitals to improve the quality of care while also reducing costs. In this analysis we assessed the relationship between various interventions associated with ERAS and perioperative outcomes such as hospital length of stay (LOS).  Methods: Patients undergoing TKA between July 1, 2013 and March 31, 2015 were identified in the Premier Perspective Hospital Database. The use of ERAS interventions, including early mobilization, non-opioid analgesia, and prophylactic treatment of infections and thromboembolism, was identified from hospital chargemaster codes. Multivariate logistic regression was conducted to examine the

impact of these interventions and patient factors (age, comorbidities, race, etc.) on hospital LOS ≤ 2 days.  Results: 105,169 patients undergoing TKA were identified; 61.9% were female and the mean (SD) age was 65.7 (9.6) years. Nearly all patients received antibiotics (98.6%) and thromboprophylaxis (94.8%) preoperatively, as well as short-acting local anesthetics (96.6%) on the day of surgery. Factors associated with hospital discharge ≤ 2 days included male gender (OR 1.61, 95% CI 1.57-1.67), Caucasian race (OR 1.53, 95% CI 1.44-1.62), use of bupivacaine liposomal injectable suspension (OR 2.46, 95% CI 2.38-2.54), use of oral non-opioid analgesia (OR 1.76, 95% CI 1.61-1.91), and early mobilization (OR 1.51, 95% CI 1.44-1.57). Increased age, Charlson comorbidity index, postoperative opioid consumption, use of epidural anesthetics, use of antiemetic and gut motility medications after the day of surgery, and opioid-related adverse events were all associated with a longer hospital LOS.  Conclusions: Important predictors of early discharge after TKA include the use of non-opioid analgesia to manage postsurgical pain and early mobilization. Implementing these elements of ERAS into current care pathways for TKA can result in more efficient care. PHP96 PATIENT CHARACTERISTICS, HEALTH OUTCOMES, AND PREDICTORS OF ORGAN FAILURE OR REJECTION OF ORGAN TRANSPLANT RECIPIENTS IN THE UNITED STATES Zacherle E , Shah S , Blanchette C M , Noone J M University of North Carolina at Charlotte, Charlotte, NC, USA .

.

.

.

.

.

Objectives: Over the past several decades, the organ transplant waiting list has steadily increased and only 10% of eligible patients receive organ transplants worldwide. Innovative methods to end organ shortage are currently under development, and in order to understand the greatest needs in transplantation research, it is essential that transplant recipients are extensively characterized. The purpose of this study is to evaluate the baseline characteristics, health outcomes, and predictors of transplant rejection/failure of transplant recipients in the US.  Methods: Using data from the 2012 Nationwide Inpatient Sample (NIS), we identified patients by diagnosis and procedures codes that received heart (V42.1, 996.83, 37.5, 37.51), liver (V42.7, 996.82, 50.5, 50.59), lung (V42.6, 996.84, 33.5, 33.50, 33.51, 33.52), kidney (V42.0, 996.81, 55.6, 55.69), and pancreas (V4282, 996.86, 528, 5280, 5283) transplantations. Transplant recipient demographics and heath outcomes were assessed and logistic regression was used to determine predictors of organ transplant failure/rejection.  Results: Of the 253,830 patients identified as receiving transplants, the most common organs transplanted were the kidney (59%, n= 148,650), liver (18%, n= 46,340), and heart (8%, n= 21,495). Mean length of stay, inpatient mortality and complications were the greatest in lung transplant recipients (9.97 days, 4%, and 48%, respectively). Additionally, Black ethnicity (OR: 1.42 (1.35-1.49)), liver, lung or multi-organ transplant [(OR: 1.87 (1.72-2.03)), (OR: 1.51 (1.39-1.64)), (OR: 1.49 (1.38-1.61))], age [18-35 years, (OR: 1.45 (1.401.55))], fluid/electrolyte disorders (OR: 1.36 (1.31-1.42)), blood loss anemia (OR: 1.36 (1.31-1.42)), pulmonary circulation disorders (OR: 1.33 (1.19-1.48)) and weight loss (OR: 1.54 (1.42-1.66)) were most predictive of organ transplant rejection.  Conclusions: Key findings of this study included the predictors of organ transplantation failure/ rejection (ethnicity, type of transplant, age and comorbidities) and poor outcomes associated with lung transplantations. These patient characteristics, outcomes, and predictors of transplant rejection/failure highlight where attention is greatly needed as improved methods for organ transplantation are developed. PHP97 PHARMACOVIGILANCE TREND ANALYSIS OF MEDICATION OFF-LABEL USE Ali A K Eli Lilly and Company, Indianapolis, IN, USA .

.

Objectives: This analysis describes the extent and trend of reporting medication off-label use in the FDA Adverse Event Reporting System (FAERS).  Methods: Adverse medication events submitted to FAERS through June 2015 were analyzed. Off-label use was defined by MedDRA PT: Off label use; off label use of device; and contraindicated drug administered. Medication classes were classified by the pharmacological subgroup classification of the Anatomical Therapeutic Chemical (ATC Level 3). Multi-item Gamma Poisson Shrinker disproportionality analysis was applied to estimate signals of off-label use, which are defined as 2.0≥ EB05.  Results: A total 85,026 off-label use reports were submitted for medications, with an average annual increase in reporting of 72% (min= 21%, max= 180%); corresponding 5-year reporting trend was: 2010 (4%); 2011 (5%); 2012 (14%); 2013 (24%); 2014 (30%); and 2015 Q1-Q2 (19%). Signals of reporting of off-label use were prominent in: skin cicatrizants, e.g. hyaluronic acid (EB05= 8.7); topical decongestants, e.g. phenylephrine (EB05= 8.1); hematological agents, e.g. icatibant for hereditary angioedema (EB05= 6.1); laxatives, e.g. bisacodyl (EB05= 5.4); muscle relaxants, e.g. botulinum toxin (EB05= 5.0); psychostimulants, e.g. methylamphetamine (EB05= 4.3); adrenergics, e.g. epinephrine (EB05= 4.0); antacids, e.g. calcium carbonate (EB05= 3.4); other ulcer-healing drugs, omeprazole and ranitidine (EB05= 3.1); anti-parathyroid agents, e.g. calcitonin (EB05= 3.0); intestinal anti-infectives, e.g. vancomycin (EB05= 3.0); inhaled agents for chronic obstructive airway disease, e.g. ipratropium and budesonide (EB05= 3.0); systemic anti-infectives, e.g. ciprofloxacin (EB05= 2.8); systemic corticosteroids, e.g. prednisolone (EB05= 2.7); non-steroidal anti-inflammatories, e.g. ibuprofen (EB05= 2.5); inhaled bronchodilators, e.g. albuterol (EB05= 2.4); vitamin B12 and folic acid (EB05= 2.2); antipsychotics, e.g. clozapine (EB05= 2.0); cytotoxic antibiotics, e.g. doxorubicin (EB05= 2.0); and topical anti-infectives, e.g. neomycin (EB05= 2.0).  Conclusions: Reporting of medication off-label use is increasing in the FAERS, and off-label use is a potential risk with some pharmacological classes. Drug utilization studies are suggested to better evaluate the extent of off-label use in selected medications, e.g. topical and gastrointestinal agents. PHP98 MEDICATION ADHERENCE: DOES ADHERENCE TO SOME THERAPEUTIC CLASSES IMPROVE ADHERENCE TO OTHERS? Peasah S 1, Liu C 2, Mezgebe M 2 .

.

.



A275

VA L U E I N H E A LT H 1 9 ( 2 0 1 6 ) A 1 - A 3 1 8

1Mercer

University College of Pharmacy, Atlanta, GA, USA, 2Mercer University, Atlanta, GA, USA

Objectives: Pharmacotherapy has been an essential part of chronic disease management because of evidence that medications improve health outcomes including hospital admission, emergency visits, and avoidable healthcare costs. Medication non-adherence, however, continues to reduce the impact of pharmacotherapy on health outcomes. Disparities in adherence rates exist among the therapeutic classes, suggesting selective adherence by patients. Currently medication synchronization is one of the interventions utilized to improve adherence. In this study, we addressed whether adherence to a normally high-adherence therapeutic class improves adherence to a normally low-adherence therapeutic class?  Methods: Using prescription claims data from Georgia Medicaid for 2012, we calculated the Proportion of Days Covered (PDC) for the top four prescribed antidiabetic (metformin, glyburide, glimepiride, and glipizide), antihypertensive (lisinopril, amlodipine, losartan, and metoprolol tartrate), antihyperlipidemia (all statins), and some cardiovascular (digoxin, amiodarone, and isosorbide mono/dinitrate) medications for patients with continuous eligibility and at least one refill. We estimated the association between the medication adherence rates of pairs of the four therapeutic classes using student’s t-test and multivariate regression controlling for age, gender, and race for patients on both medications.  Results: The cardiovascular agents had the highest PDC (78%) followed by antihypertensives (75%), antidiabetics (73%), and statins (69%). Patients concurrently on cardiovascular agents (PDC 80%) and statins or antihypertensives or antidiabetics, had a higher PDC for statins (77%) and hypertensives (80%) but not for antidiabetics (69%). However patients concurrently on statins, metformin, and lisinopril had PDC of 73%, 69%, and 80% respectively, an increase from those concurrently on only statins (71%) and metformin (64%) or only lisinopril (77%) and metformin (65%). The multivariate regression results show significant positive association between them.  Conclusions: These results suggest that concurrent medications might have a positive effect on medications with lower adherence rates; suggesting that implementing medication synchronization is likely to improve adherence. PHP99 COMPARISON OF PARENT AND NON-PARENT PREFERENCES IN THE VALUATION OF CHILD HEALTH

1RTI

Health Solutions, Research Triangle Park, NC, USA, 2RTI Health Solution, Research Triangle Park, NC, USA

Objectives: To assess the extent of use of endpoints based on survival, biomarker and clinical outcome assessments (COAs) consisting of Clinician-reported outcomes (ClinROs), Observer-reported outcomes (ObsROs), Patient-reported outcomes (PROs) and Performance outcomes (PerfOs) as primary endpoints in confirmatory studies of new drugs approve from 2011 through 2015.  Methods: New drugs approved between January 2011 and December 2015 were identified using [email protected] FDA database. Labeling and medical review sections from FDA DAPs were reviewed to identify indication and the primary endpoint of confirmatory studies. ICD-10 codes were used to classify disease and the primary endpoints were classified based on the type of outcome assessment. Descriptive data were recorded in Microsoft Excel; frequency of measured characteristics was analyzed.  Results: Of the 182 new drugs approved the majority of the approvals were for drugs related to cancer (27.5%), anti-infective (15.9%) and endocrine, nutritional and metabolic diseases (15.4%). ClinROs and biomarkers constituted 47.8% and 40.7% of the primary endpoints respectively. PROs, survival and PerfO constituted 13.2%, 14.3% and 1.1% of the primary endpoints respectively. PROs were commonly found to be the primary endpoint in diseases related to genitourinary (80%), musculoskeletal (66.7%) and digestive systems (50%). Approvals related to cancer drugs relied on ClinROs (76%), biomarkers (14%) and survival (36%) as primary endpoints. Only six approvals (3.2%) were based on primary endpoints that were composite of PRO and another type of outcome.  Conclusions: ClinROs and biomarkers are used as primary endpoints in the majority of confirmatory studies of recent FDA approvals. PROs were key to assess treatment benefit in diseases such as those related to genitourinary, digestive and musculoskeletal systems as per respective regulatory guidance. PHP102 SUMMARIZING THE BARRIERS AND SUPPORTING FACTORS TOWARDS ORGAN DONATION: A REVIEW OF THE LITERATURE Jordan M R , Pinto S University of Toledo, Toledo, OH, USA .

.

.

Objectives: The goal of this project was to determine whether or not there is any difference in preferences between parents and non-parents for childhood health. While there exists a wide variety of literature focusing on child health, only a few papers have ever attempted to value childhood health problems on a QALY scale.  Methods: 1976 Parents and 2179 non-parents from a nationally representative panel were provided with a series of discrete choice experiments asking them to choose between two different losses in child health using the EQ-5D-Y. An example question may ask if an individual would be willing to trade a child’s lifespan in order for a child not have to experience any pain or discomfort. Using an unnamed 7 or 10 year child with health problems lasting for 1 or 2 years, our statistical analysis estimated the value of childhood health problems on a quality adjusted life year (QALY) scale where a value of 1 is equivalent to a year with no health problems  Results: We find a statistically significant difference in the evaluation of health problems between parents and non-parents. The ordering of the QALY weights by health problem is consistent between both groups indicating a difference of scale between parents and non-parents. By placing lower statistical weights on childhood health problems, parents state that they would prefer longer lifespan rather than more time in a healthier state as compared with non-parents.  Conclusions: This is one of the first surveys to use adult preference to place childhood health outcomes on a QALY scale. It is important for policy makers to take these findings into consideration when allocating resources.

Objectives: Organ transplantation became a new hope for those living with endstage organ disease. The number of patients needing transplants greatly exceeds the number of available donors and continues to rise. Without an increase in the number of donors, many patients will not live to reach an organ transplant. This research seeks to review the literature for barriers and supporting factors that lead patients to becoming organ donors.  Methods: The PubMed database was reviewed using the mesh terms “organ donation” and “barriers.” Studies included were required to: 1) Pertain to solid organ transplantation, 2) Examine a factor that shows a resistance to organ donation 3) Provide patient-reported outcomes. Results were reviewed and summarized into a comprehensive model. A secondary reviewer aided in approving the validity of the findings. Barriers included were significant at the p < 0.05 level as reported by the respective authors.  Results: The mesh terms generated 295 results. Out of these, 67 studies were found to meet all inclusion criteria and examined in full. Some of the most encountered barriers include distrust of the healthcare system, lack of communication between family members, racial/ethnic barriers, lack of knowledge, and preconceived beliefs experienced from the media. Preexisting knowledge about organ donation and religious beliefs, in addition to who speaks to the patient regarding donation can both improve the likelihood of donation and create a barrier.  Conclusions: There are significant barriers to organ donation. Each barrier can be placed into one of four distinct categories (communication, preconceived notions of healthcare system, cultural beliefs, and financial/ethical considerations). Potential cues to action for someone to consider being an organ donor and communicating one’s end of life wishes need to be further examined. Summarizing these findings will assist for future research in developing methods for exploring cues to action for organ donation.

PHP100 AN ASSESSMENT OF RATIONAL DRUG USE AND EVALUATION OF MEDICATION ERRORS IN PUBLIC SECTOR HOSPITALS

PHP103 USE AND IMPACT OF MULTIMODAL ANALGESIA IN TOTAL JOINT ARTHROPLASTY

Jan S U 1, Gul R 1, Ali S H 1, Ullah K 1, Shah A 1, Akhtar M 2 of Balochistan, Quetta, Pakistan, 2The Islamia University of Bahawalpur, Bahawalpur, Pakistan

1Boston

Hartman J D 1, Craig B M 2 1Moffitt Cancer Center, Tampa, FL, USA, 2Department of Economics, University of South Florida, Tampa, FL, USA .

.

.

.

.

.

.

.

.

.

.

.

1University

Objectives: The objective of this study was to assess the prescription pattern by Medical practitioners and to analyze the types of medication errors observed in public sector Hospitals.  Methods: Trainee Pharmacists during internship in my supervision were assigned to examine the prescriptions for drug prescription patterns as well as the same prescriptions and patients were examined for medication errors in three public sector hospitals in the city using a check list and patients’ interviews. The results were collected from 2400 prescriptions/patients. The study period was from July, 1st 2014 to September, 10th 2014.  Results: The results showed that, the average consultation and dispensing time was very short (4 Minutes). The mean number of drugs per prescription was 5. The percentages of prescriptions with antibiotics, analgesics, steroids and injections were 88%, 64%, 13% and 12% respectively. Further, we have observed 1012 medication errors out of 2400 only in drug administration. In each category the medication errors were omission: 926(38.6%), time: 960(40%), unauthorized drug: 200(8.3%), wrong rate: 168(7%), wrong route: 98(4.1%) and wrong dosage form: 48(2%).  Conclusions: We recommend that efforts should be made to improve rational drug use practices in Hospitals. Regular training programs should be organized for health care professionals to promote the concept and practice of rational drug use and to overcome these types of medication errors which lead to about 40% accidents and incidents in drug administration. PHP101 OUTCOME ASSESSMENTS OF PRIMARY ENDPOINTS OF NEW DRUGS APPROVED BY THE FDA (2011-2015) Gnanasakthy A 1, DeMuro C 2 .

.

Aggarwal J 1, Park S 1, Dagenais S 2, Kang A 2, Menzin J 1 Health Economics, Waltham, MA, USA, 2Pacira Pharmaceuticals, Inc., Parsippany, NJ, USA .

.

.

.

.

Objectives: Multimodal analgesia is thought to improve postsurgical recovery. In this analysis we assess the use of multimodal analgesia in total knee arthroplasty (TKA) and total hip arthroplasty (THA) and its impact on perioperative outcomes.  Methods: Patients undergoing TKA and THA were identified in the Premier Perspective Hospital Database from July 1, 2013 to March 31, 2015. Those receiving multimodal analgesia (combination of bupivacaine liposomal injectable suspension (BLIS), intravenous acetaminophen, and ketorolac) were compared to those who did not. Propensity scores were used to match patients on age, gender, race, Charlson comorbidity index, payer, region, hospital teaching status, and length of surgery. Outcomes included hospital length of stay (LOS), proportion discharged in ≤ 2 days, home discharge, and readmission to the same hospital.  Results: A total of 145,288 patients (95,296 TKA and 49,992 THA) were identified during this period; only 7.6% of TKA patients (7,242) and 4.1% of THA patients (2,050) received multimodal analgesia, while 22.3% of TKA patients (21,251) and 29.5% of THA patients (14,748) did not receive any of the three medications of interest. The final analysis included 6,600 matched pairs for TKA and 2,037 matched pairs for THA. Compared to controls, patients who received multimodal analgesia had a shorter LOS for both TKA (mean 2.52 vs 3.08 days, p< 0.001) and THA (mean 2.45 vs 3.04 days, p< 0.001), were more likely discharged in ≤ 2 days for both TKA (56.5% vs 25.4%, p< 0.001) and THA (62.7% vs. 38.8%, p< 0.001), and a higher proportion were discharged home for both TKA (75.9% vs. 65.5%, p< 0.001) and THA (80.3% vs. 69.2%, p< 0.001).  Conclusions: The use of multimodal analgesia combining a long-acting local analgesic, non-opioid analgesic, and non-steroidal anti-inflammatory may be underutilized in patients undergoing TKA and THA, despite the potential benefits of this approach on important perioperative outcomes including LOS and discharge home.