A. S. Kolbin

Decision modelling for the evaluation of diabetes outcomes

The process of decision modelling in diabetes mellitus (DM) is often complicated by comorbidity among diabetic patients, complexity of endpoint selection, and unclear time horizons. Aim . To review the available recommendations, relevant methods and mathematical approaches to decision modelling in DM. Materials and Methods . We searched through the PubMed database using the ResearchGate and Mendeley networks; we also collected data from the websites of the key opinion leaders in the field of pharmacoeconomics and decision modelling. Results . This review contains up-to-date information on the validity of the most common DM decision models and on the validity of extrapolating the type 2 DM models to patients with type 1 DM. We also provide some clinically relevant comments on the American Diabetes Association’s requirements concerning the decision models in DM. The review incorporates data on the current mathematical approaches to modelling the changes in glycated hemoglobin levels, the body mass index and the quality-adjusted life expectancy – for both type 1 and type 2 DM. Conclusion . Despite recent successes in DM decision modelling, the existing approaches are not always relevant to some groups of DM patients or to some aspects of the disease. Thus, the use of the novel anti-diabetic drugs (liraglutide, semaglutide, empagliflozin) capable of significantly reducing cardiovascular risks in DM patients, require new approaches to decision modelling in diabetes mellitus.

PHARMACOECONOMIC ANALYSIS OF BRENTUXIMAB VEDOTIN THERAPY IN RELAPSED OR REFRACTORY CD30+ SYSTEMIC ANAPLASTIC LARGE CELL LYMPHOMA IN PATIENTS OVER 18 YEARS OF AGE: HEALTH BUDGET IMPACT ANALYSIS

Objectives . To provide a pharmacoeconomic analysis of brentuximab vetodin therapy in relapsed or refractory CD30+ systemic anaplastic large-cell lymphoma in patients over 18 years of age. Methods . Recommendations to carry out a budget impact analysis in theRussian Federation (2016) were used. Results . The analysis covered a period of 3 years. The target population of patients with relapsed or refractory CD30+ systemic anaplastic large cell lymphoma numbered 345 individuals. The results indicate that the introduction of brentuximab vetodin in the List of vital and essential drugs and the Program of state-guaranteed free medical care will not increase the costs of the treatment in patients with relapsed or refractory CD30+ systemic anaplastic large cell lymphoma. Conclusion . The brentuximab vetodin therapy of relapsed or refractory CD30+ systemic anaplastic large cell lymphoma is an economically reasonable and resource-saving medical modality.

ECONOMIC ASSESSMENT OF THE EFFECTIVENESS AND SAFETY OF THE COMMONLY PRACTICED PERIOPERATIVE ANTIBIOTIC PROPHYLAXIS (BASED ON AN EARLIER EPIDEMIOLOGICAL SURVEY OF MULTIDISCIPLINARY HOSPITALS)

A survey conducted in four hospitals located in the city of St. Petersburg revealed that the commonly used perioperative antibiotic prophylaxis (PAP) did not follow (in 88% of cases) the guidelines approved for national clinical practice. Aim . To perform a cost-effectiveness analysis of the commonly practiced PAP among patients with clean, clean-contaminated and contaminated surgical wounds in a multidisciplinary hospital. Materials and methods . The PAP cost-effectiveness analysis was performed using the data from a multicenter epidemiological survey and previously conducted studies. The Markov model was used to compare the effectiveness and safety of the commonly used PAP with that recommended by the clinical practice guidelines. The rate of surgical site infection (SSI) and antibiotic-associated diarrhea (AAD) were chosen for the endpoints. Results . The costs associated with a single case of PAP according to the clinical practice guidelines was 3.5 times less than that associated with the PAP used in the common practice (RUB 4913,67 and 17837,71 respectively). The present analysis demonstrates that the PAP recommended by the clinical practice guidelines was more cost-effective as compared with the commonly practiced PAP. Conclusion . Regular epidemiological monitoring is required to improve effectiveness and safety of the existing PAP practice.

MATHEMATICAL MODEL OF REIMBURSEMENT DECISION MAKING IN RUSSIA. RESULTS OF VIM LIST FOR 2016

Background: In 2014 we firstly analyzed the formalized system (points and expert opinions) of drug inclusion and exclusion into the reimbursement lists in Russian Federation. The liner mathematical model of decision making was developed and adopted. Aim. Update the existing model using the results of reimbursement procedures acting from 2106. Material and methods. The linear models developed and adopted in 2014 were used. In 2015 we included data on 141 drug dossiers. We analyzed the decision of the expert body, chief Ministry of Health expert and the final committee decision. Results. 43 new drugs were included into the reimbursement lists acting from 2016. The model of expert body decision had an error 7,09% (12,4% in 2014). The model of chief Ministry of Health expert decision had an error – 7% (10% in 2014). The above mentioned experts became more experienced in the formalized procedure of decision making. The model of final decision had an error about 42% (35% in 2014). Conclusion. Linear models are working tools for modelling reimbursement system decisions. At the mean time the existing system of decision making needs more formalization.

AN ANALYSIS OF COMPILATION OF THE RESTRICTIVE INVENTORIES IN THE RUSSIAN FEDERATION AS EXEMPLIFIED WITH «VITALLY ESSENTIAL AND INDISPENSABLE DRUGS» (VEID): THE ROLE OF PHARMACOECONOMICS

Background. In 2014, the first experiment of drug inclusion into or exclusion from restrictive inventories based on a score system and independent expert assessments has been carried out. Based on data obtained in 2014, the present authors have developed and tested linear mathematical models for making decisions concerning drug inclusion into restrictive inventories. The objective of the present work was to develop a model for the year 2016 with account of novel and supplementary data derived from the analysis of such inventories. Materials and methods. The linear models that have been developed and adopted earlier were used. The analysis included 141 records of medicinal drugs. Analyzed were verdicts provided by expert institutions and by chief non-stuff experts and the final decisions made by Interdisciplinary Panel. Results. In 2016, the Restrictive Inventory of VEID included 43 drugs. It was shown that the model for making decisions by expert organizations was associated with a 7,09% error (12,4% in 2014). The model for chief on-stuff experts was associated with a 7% error (10% in 2014). The above positive changes suggest that persons involved are able to become trained in using the formal approaches. The model for making the final decision by interdisciplinary panel was associated with a 42% error vs. 35% found earlier. Conclusion. Linear models are effective instruments for making prognoses concerning the inclusion of medicinal drugs into restrictive inventories. However, the currently adopted system should be formalized further.