Samantha Alvarez-Madrazo

Multiple policies to enhance prescribing efficiency for established medicines in Europe with a particular focus on demand-side measures : findings and future implications

Introduction: The appreciable growth in pharmaceutical expenditure has resulted in multiple initiatives across Europe to lower generic prices and enhance their utilization. However, considerable variation in their use and prices. Objective: Assess the influence of multiple supply and demand-side initiatives across Europe for established medicines to enhance prescribing efficiency before a decision to prescribe a particular medicine. Subsequently utilize the findings to suggest potential future initiatives that countries could consider. Method: An analysis of different methodologies involving cross national and single country retrospective observational studies on reimbursed use and expenditure of PPIs, statins, and renin-angiotensin inhibitor drugs among European countries. Results: Nature and intensity of the various initiatives appreciably influenced prescribing behavior and expenditure, e.g., multiple measures resulted in reimbursed expenditure for PPIs in Scotland in 2010 56% below 2001 levels despite a 3-fold increase in utilization and in the Netherlands, PPI expenditure fell by 58% in 2010 vs. 2000 despite a 3-fold increase in utilization. A similar picture was seen with prescribing restrictions, i.e., (i) more aggressive follow-up of prescribing restrictions for patented statins and ARBs resulted in a greater reduction in the utilization of patented statins in Austria vs. Norway and lower utilization of patented ARBs vs. generic ACEIs in Croatia than Austria. However, limited impact of restrictions on esomeprazole in Norway with the first prescription or recommendation in hospital where restrictions do not apply. Similar findings when generic losartan became available in Western Europe. Conclusions: Multiple demand-side measures are needed to influence prescribing patterns. When combined with supply-side measures, activities can realize appreciable savings. Health authorities cannot rely on a “spill over” effect between classes to affect changes in prescribing.

Different initiatives across Europe to enhance losartan utilization post generics: impact and implications

Introduction: There is an urgent need for health authorities across Europe to fully realize potential savings from increased use of generics to sustain their healthcare systems. A variety of strategies were used across Europe following the availability of generic losartan, the first angiotensin receptor blocker (ARB) to be approved and marketed, to enhance its prescribing vs. single-sourced drugs in the class. Demand-side strategies ranged from 100% co-payment for single-sourced ARBs in Denmark to no specific measures. We hypothesized this heterogeneity of approaches would provide opportunities to explore prescribing in a class following patent expiry. Objective: Contrast the impact of the different approaches among European countries and regions to the availability of generic losartan to provide future guidance. Methodology: Retrospective segmented regression analyses applying linear random coefficient models with country specific intercepts and slopes were used to assess the impact of the various initiatives across Europe following the availability of generic losartan. Utilization measured in defined daily doses (DDDs). Price reductions for generic losartan were also measured. Results: Utilization of losartan was over 90% of all ARBs in Denmark by the study end. Multiple measures in Sweden and one English primary care group also appreciably enhanced losartan utilization. Losartan utilization actually fell in some countries with no specific demand-side measures. Considerable differences were seen in the prices of generic losartan. Conclusion: Delisting single-sourced ARBs produced the greatest increase in losartan utilization. Overall, multiple demand-side measures are needed to change physician prescribing habits to fully realize savings from generics. There is no apparent “spill over” effect from one class to another to influence future prescribing patterns even if these are closely related.

Data Resource Profile: The Scottish National Prescribing Information System (PIS)

Data Resource Basics: The Prescribing Information System (PIS) covers the prescribed,dispensed and reimbursed prescriptions in community pharmacies from the 5.3 million residents in Scotland. Summary information is available from 1993 and at an individual level from 2009 to the present. Data Collected: The raw data are generated by three data sources: ePrescribed -generated by GPs messages, eDispensed –generated by messages from community pharmacies and Reimbursed messages from scanned paper prescriptions dispensed in the community pharmacies. The four main categories of data collected are: (1) Patient-specific, (2) Prescriber, (3) Dispenser and (4) Drug-specific. PIS data can be linked via a unique identifier to other national databases, including hospital records, maternal and neonatal, the Scottish Cancer Registry and mortality records. The catalogue of databases is available in www.ndc.scot.nhs.uk . Subject to approval of the data controllers other external datasets can also be linked. Data Resource Use: PIS has been used to describe the utilisation of several groups of drugs;factors influencing prescribing and evaluation of interventions to improve it; generation of polypharmacy guidelines; risk of side effects; monitoring of antibiotic use and generation of policy recommendations; associations between community prescription of antimicrobials and deprivation or infection; evaluation of prescription fee abolition; clinical effectiveness, safety and health technology assessment of drugs approved in the last decade. Reasons to be cautious: PIS does not capture information about diagnosis or indication for treatment, over the counter medicines, medicines administered during inpatient hospital stays, upon discharge for short term use, outpatient supplies or some specialist drugs for chronic use. Drug data is currently coded according to the British National Formulary. For longitudinal studies, patient level data is available from 2009 and the frequency of data collection from the three sources is different. Collaboration and data access: PIS data are available upon request to the electronic Data Research and Innovation Service (NSS.eDRIS@nhs.net) and project approval by the Public Benefit and Privacy Panel. Funding and competing interests: This dataset is funded from the public monies available to the NHS. Current work to develop an improved PIS research ready analysis platform and this study is supported by the Farr Institute @ Scotland and its 10-funder consortium. The authors declare no conflict of interest.

Dabigatran - a continuing exemplar case history demonstrating the need for comprehensive models to optimize the utilization of new drugs.

Background: There are potential conflicts between authorities and companies to fund new premium priced drugs especially where there are effectiveness, safety and/or budget concerns. Dabigatran, a new oral anticoagulant for the prevention of stroke in patients with non-valvular atrial fibrillation (AF), exemplifies this issue. Whilst new effective treatments are needed, there are issues in the elderly with dabigatran due to variable drug concentrations, no known antidote and dependence on renal elimination. Published studies showed dabigatran to be cost-effective but there are budget concerns given the prevalence of AF. These concerns resulted in extensive activities pre- to post-launch to manage its introduction. Objective: To (i) review authority activities across countries, (ii) use the findings to develop new models to better manage the entry of new drugs, and (iii) review the implications based on post-launch activities. Methodology: (i) Descriptive review and appraisal of activities regarding dabigatran, (ii) development of guidance for key stakeholder groups through an iterative process, (iii) refining guidance following post launch studies. Results: Plethora of activities to manage dabigatran including extensive pre-launch activities, risk sharing arrangements, prescribing restrictions and monitoring of prescribing post launch. Reimbursement has been denied in some countries due to concerns with its budget impact and/or excessive bleeding. Development of a new model and future guidance is proposed to better manage the entry of new drugs, centering on three pillars of pre-, peri-, and post-launch activities. Post-launch activities include increasing use of patient registries to monitor the safety and effectiveness of new drugs in clinical practice. Conclusion: Models for introducing new drugs are essential to optimize their prescribing especially where concerns. Without such models, new drugs may be withdrawn prematurely and/or struggle for funding.

Use of direct oral anticoagulants in patients with atrial fibrillation in Scotland: applying a coherent framework to drug utilisation studies

To report the use of direct oral anticoagulants (DOACs) for stroke prevention in patients with atrial fibrillation in Scotland and advocate the standardisation of drug utilisation research methods.

Use of text-mining methods to improve efficiency in the calculation of drug exposure to support pharmacoepidemiology studies

Abstract Background Efficient generation of structured dose instructions that enable researchers to calculate drug exposure is central to pharmacoepidemiology studies. Our aim was to design and test an algorithm to codify dose instructions, applied to the NHS Scotland Prescribing Information System (PIS) that records about 100 million prescriptions per annum. Methods A natural language processing (NLP) algorithm was developed that enabled free-text dose instructions to be represented by three attributes – quantity, frequency and qualifier – specified by three, three and two variables, respectively. A sample of 15 593 distinct dose instructions was used to test, validate and refine the algorithm. The final algorithm used a zero-assumption approach and was then applied to the full dataset. Results The initial algorithm generated structured output for 13 152 (84.34%) of the 15 593 sample dose instructions, and reviewers identified 767 (5.83%) incorrect translations, giving an accuracy of 94.17%. Following subsequent refinement of the algorithm rules, application to the full dataset of 458 227 687 prescriptions (99.67% had dose instructions represented by 4 964 083 distinct instructions) generated a structured output for 92.3% of dose instruction texts. This varied by therapeutic area (from 86.7% for the central nervous system to 96.8% for the cardiovascular system). Conclusions We created an NLP algorithm, operational at scale, to produce structured output that gives data users maximum flexibility to formulate, test and apply their own assumptions according to the medicines under investigation. Text mining approaches can provide a solution to the safe and efficient management and provisioning of large volumes of data generated through our health systems.

The use of new oral anticoagulants in patients with atrial fibrillation in Scotland : a population-based drug utilisation study

Background: Type 2 diabetes mellitus (T2DM) has been suggested as a risk factor for liver, pancreatic, and colorectal cancer. T2DM patients show higher incidences of these cancers compared to the non-diabetic (non-DM) population. Current evidence, however, is inconsistent with respect to the incidences of other gastrointestinal (GI) malignancies. Objectives: To determine incidence rates (IRs) of all GI cancers in patients with and without T2DM. Methods: A retrospective cohort study was conducted using the UK Clinical Practice Research Datalink (CPRD) during 1988-2012. A T2DM cohort of antidiabetic drug users was matched to a non-DM reference cohort, by age, sex, and practice. Crude incidence rates (IRs) per 100,000 person-years (105 py) and 95% confidence intervals (CI) were calculated, stratified by age, sex, and calendar period. IRs were compared using the normal theory test. Results: 333,438 T2DM subjects and 333,438 non- DM subjects were analyzed, with a total duration of follow-up of >3.6 million py and 10,977 observed GI cancer cases. Overall, IRs of any GI cancer (IR 330 vs. 276 per 105 py), liver cancer (IR 26 vs. 8.9 per 105 py), pancreatic cancer (IR 65 vs. 31 per 105 py), and colon cancer (IR 119 vs. 109 per 105 py) were significantly higher in the T2DM cohort compared to the non-DM cohort, whereas the IR of esophageal cancer was significantly lower (IR 41 vs. 47 per 105 py, pBackground: Progressive multifocal leukencephalopathy (PML) is a rare, often fatal viral disease, which affects the white matter of the brain. It is caused by John Cunningham (JC) polyomavirus, whi ...The article investigates the special features of state control over international transfer of special-purpose and dual-use goods. It was established what international organizations was created in the international community to determine the principles of control over international transfer of special-purpose and dual-use goods, as well as the question of Ukraines joining the circle of member-states of such organizations. The structure of the system of export control bodies in Ukraine was defined, as well as the main powers of the State Service of Export Control of Ukraine in the sphere of control over international transfer of goods. The essence and the concept of goods over which international transfer state export control is carried out in accordance with the Ukrainian legislation were revealed, as well as special aspects of the procedure of state control over their international transfer.Background: Different antiplatelet regimens are used for secondary prevention after ischemic stroke (IS)/transient ischemic attack (TIA), but studies on the relative effectiveness and safety of each regimen in daily practice are lacking. Objectives: To assess the relative effectiveness and safety of several antiplatelet regimens as secondary prevention in patients after an IS/TIA in clinical practice. Methods: A cohort study was conducted using the Clinical Practice Research Datalink. Patients aged ≥ 18 years with a first diagnosis of IS/TIA in 1998- 2013 were identified. Antiplatelet exposure was categorized into aspirin-dipyridamole, aspirin-only, clopidogrel-only, aspirin-clopidogrel, other regimens, and no-antiplatelet exposure. The primary effectiveness outcome was a composite endpoint of nonfatal IS, nonfatal myocardial infarction (MI), or cardiovascular (CV) death; and the safety outcome was major bleeding. Time-dependent Cox regression analysis was used to assess the association between antiplatelet regimens and CV effectiveness and major bleeding outcomes. Results: We followed 20,552 IS/TIA patients for a median duration of 2.3 years. There were 5,714 composite events during follow-up. All regimens were effective in reducing the primary effectiveness outcome compared to no-antiplatelet exposure. Aspirin-only, clopidogrel-only, aspirin-clopidogrel and other regimens were significantly (p <0.05) less effective compared to aspirin-dipyridamole (HR: 1.35, 1.12, 1.40, and 1.27, respectively), adjusted for age, sex, lifestyle factors, disease history and CV comedications. All other regimens were also significantly (p <0.05) associated with a higher relative risk of major bleeding compared to aspirin-dipyridamole (HR: 1.21, 1.32, 1.78, and 1.37, respectively), adjusted for age, sex, alcohol use, liver and renal disease, major bleeding history and comedications. Conclusions: Compared to aspirin-dipyridamole, all other antiplatelet regimens are less effective in reducing the risk of nonfatal IS, nonfatal MI or CV death, and associated with a higher risk of major bleeding in patients with IS/TIA.Characteristics of Patients at Initiation of Treatment for Primary Chronic Immune ThrombocytopeniaBackground: Guidelines for cardiovascular secondary prevention are based on evidence from relatively old clinical trials and need to be evaluated in daily clinical practice. Objectives: To evaluate effectiveness of the recommended drug classes after an acute coronary syndrome (ACS) for secondary prevention of cardiovascular diseases and all-cause mortality. Methods: This cohort study used data from a representative sample of the French national healthcare insurance system database (EGB). Patients hospitalised for an incident ACS between 2006 and 2011, and aged ≥ 20 years at time of ACS were included in the study. Patients non-exposed to any of the four recommended drug classes (beta-blockers, antiplatelet agents, statins, and angiotensin-converting-enzyme inhibitors, ACEI, or angiotensin II receptor blockers, ARB) in the first 3 months following ACS or who died during this period were not included in the cohort. Exposure status was determined daily during follow-up. Effectiveness of the four therapeutic classes in preventing the composite outcome ACS, transient ischemic attack, ischemic stroke, or all-cause-death was estimated using a time-dependent Cox proportional hazards model, which was adjusted for time-fixed confounders measured at baseline (general characteristics and characteristics of the initial ACS) and time-dependent confounders during follow-up (co-morbidities and co-medications). Results: Of the 2874 patients included in the study, 33.9% were women and the median age was 67 years (interquartile range, IQR: 56-77). The median time of follow-up was 3.6 years (IQR: 2.2-5.3). The risk of the composite outcome decreased with use of antiplatelet agents (adjusted hazard ratio (aHR) 0.76, 95% confidence interval (CI) 0.63; 0.91), use of statins (aHR 0.71, 95%CI 0.57; 0.87), and use of ACEI/ARB (aHR 0.67, 95%CI 0.57; 0.80). Use of beta-blockers was not associated with a lower risk of the composite outcome (aHR, 0.90, 95%CI 0.74; 1.09]). Conclusions: Use of antiplatelet agents, statins, and ACEI/ARB after an ACS, but not beta-blockers, was associated with a lower risk of cardiovascular morbidity and all-cause mortality.Abstracts of the 32nd International Conference on Pharmacoepidemiology & Therapeutic Risk Management, The Convention Centre Dublin, Dublin, Ireland August 25–28, 2016Background: Cough and angioedema are adverse events associated with especially angiotensinconverting enzyme (ACE) inhibitors but also reported with angiotensin receptor blockers (ARBs) and aliskiren, a direct renin inhibitor (DRI). Susceptibility of developing cough/angioedema with ACE inhibitors depends on ethnicity, which is not documented in spontaneous reporting systems of drug safety. Objectives: To assess the impact of ethnicity on the occurrence of cough/angioedema with renin angiotensin system (RAS) inhibitors using information reported to the the World Health Organization database (VigiBase). Methods: A case/non-case study was performed in VigiBase. Cases were defined as reports of cough/angioedema and non-cases were all reports of other adverse events. The reporting countries were divided into three categories: black African countries, East Asian countries and other countries. Logistic regression analysis was used to assess the association between reporting of cough/angioedema with each class of RAS inhibitors stratified by country group and to control for confounding. Results: The reporting of cough with ACE inhibitors was significantly higher in East Asian countries than black African countries and other countries (adjusted reporting odds ratios (RORs): 256, 95%CI (236-278), 48.9, 95%CI (42.7-56.1) and 35.4, 95%CI (34.8- 35.9), respectively. The reporting of angioedema with ACE inhibitors was significantly higher in black African countries than East Asian countries and other countries (adjusted RORs: 55.3, 95%CI (45.5-67.2), 5.29, 95%CI(3.89-7.21) and 16.5, 95%CI (16.1- 16.8), respectively. There was no difference in reporting of cough/angioedema with ARBs and DRI between black African countries, East Asian countries and other countries. Conclusions: Our results by grouping countries according to ethnicity in VigiBase are consistent with previous results in the literature suggesting that the occurrence of cough with ACE inhibitors is higher in East Asian patients and the occurrence of angioedema with ACE inhibitors is higher in black patients. These findings indicate that ethnicity should be included as scientific parameter in pharmacovigilance.An Automatized Model for Sequential Monitoring of Effectiveness of New Drugs using Dronedarone as ExampleGeneral Pharmacological Treatments Preceding A Primary Chronic Immune Thrombocytopenia DiagnosisBackground: Several studies showed a bidirectional association between type 2 diabetes and psychiatric disorders in adults. There is limited information available about the association of type 1 diabetes (T1D) and psychiatric disorders in children and adolescents. Objectives: To assess the extent of psychiatric medication use before and after the onset of T1D in children and adolescents compared with a reference cohort without T1D. Methods: A population-based cohort study was conducted in the Dutch PHARMO Record Linkage System. All children and adolescents <19 years) with at least two insulin dispensings between 1999 and 2009 were identified as a T1D cohort (N=925) and matched with an up to four times larger diabetes-free reference cohort (N=3591) by age and sex. The period prevalences of psychiatric medication use (psycholeptics (ATC N05) and psychoanaleptics (ATC N06)) were calculated by dividing the number of patients with at least one dispensing by the number of patients available in the cohort during that time. Prevalences were calculated from 5 years before until 5 years after the onset of T1D (the index date in both cohorts) and stratified by age, sex, medication subgroup, and before/after the onset of T1D. Results: The mean age of the study participants was 10.1 years and 51% were boys. The 5-year prevalence of psychiatric medication use before the index date was significantly higher in the T1D cohort than in the reference cohort (7.2 vs. 4.7%, respectively, p=0.002). The same pattern was observed for the period after developing T1D (10.4 vs. 7.9% in the T1D and reference cohort respectively, p=0.015). In both cohorts adolescents (15-19 years) and boys had higher prevalences of psychiatric medication use. This increased prevalence of psychiatric medication use both before and after the index date in T1D cohort was mainly driven by an increased use of psycholeptics (mainly anxiolytics). Conclusions: Children with T1D were more likely to use psychiatric medication in the years before and after the onset of type 1 diabetes. This increased use was mainly driven by psycholeptics both before and after onset of T1D.