Budget impact analysis of medicines: estimated values versus real-world evidence and the implications.

OBJECTIVES: Budget Impact Analyses (BIA) of medicines helps managers in promoting health systems' sustainability when assessing the role and value of new medicines. However, it is not clear whether BIAs typically underestimate or overestimate the impact on real-world budgets. This retroactive analysis seeks to compare estimated values obtained by a BIA and Real-World Evidence (RWE) to guide discussions. METHODS: The estimated values were obtained through a BIA concerning the incorporation of adalimumab for the treatment of Rheumatoid Arthritis into the Brazilian Unified Health System (SUS) carried out retroactively and per international guidelines. RWE data was extracted from SUS computerized systems. We subsequently compared the number of treatments, costs, and Incremental Budget Impact (IBI). RESULTS: - The total number of treatments was underestimated by 10% (6,243) and the total expenditure was overestimated by 463% (US$ 4.7 billion). In five years, the total difference between the estimated values and real IBI reached US$ 1.1 billion. A current expenditure of US$ 1.0 was observed for every US$ 5.60 of estimated expenditure. CONCLUSION: - The higher estimates from the BIA might cause decision makers to be more cautious with the introduction of a new medicine to reduce the opportunity costs for other interventions.

Cost Analysis of Cancer in Brazil: A Population-Based Study of Patients Treated by Public Health System From 2001-2015.

OBJECTIVES: The aim of this study was to evaluate the federal government expenditures with oncological care, for the most incident cancer types among the Brazilian population, using registries of all patients treated by the Brazilian National Health Service (SUS) between 2001 and 2015. We adopted the formal healthcare sector perspective in this study, with the costs per patient estimated by the reimbursement price paid by the Ministry of Health to service providers. METHODS: The costs were adjusted by the follow-up time for each patient. We performed multivariate regression analysis using ordinary least squares. We analyzed 952 960 patients aged ≥19 years who underwent cancer treatment, between 2001 and 2015, for breast, prostate, colorectal, cervix, lung, and stomach cancers. RESULTS: The annual mean costs per patient (in USD purchasing power parity) was $9572.30, varying from $5782.10 for breast cancer to $16 656 for cervical cancer. Several variables predicted higher costs of cancer treatment, namely: to be male (+14%), with younger age ranges at treatment initiation, resident in the Northeast region (+26%), treated for colorectal cancer (+482%), with treatment initiation from 2010 to 2014, tumor stages III and IV (III: +182%; IV: +165%), hospitalization for other reasons besides the cancer treatment, and suffering from some a comorbidity. CONCLUSIONS: Given the forthcoming Brazilian demographic changes, which strongly suggest that the economic burden of cancer is about to increase in the near future, our estimates provide relevant information to produce useful projections about future cancer-related costs.

The long-term costs for treating multiple sclerosis in a 16-year retrospective cohort study in Brazil.

BACKGROUND: Multiple Sclerosis (MS) is a disease that appreciably impacts on the quality of life of patients and is associated with high expenditure. MS is a chronic multifactorial disease, characterized by inflammation, demyelination and axonal loss. The Brazilian public health system provides pharmacological treatment as well as hospital and outpatient care for patients with relapsing-remitting and secondary progressive multiple sclerosis. However, we are not aware of any previous publications assessing total direct medical costs in patients with a long follow-up within the Brazilian healthcare system. Consequently, the objective is to analyze public spending on patients with MS to guide stakeholders in future investment and disinvestment decisions. METHODS AND FINDINGS: We retrospectively analyzed public Brazilian spending on patients with MS between 2000 and 2015 using the patient-centered registry of all patients in the public health system (SUS) obtained through deterministic-probabilistic record linkage of the Outpatient Information System, Hospital Information System and Mortality Information Systems in Brazil. Descriptive data analysis and a multiple linear regression model was performed to evaluate the associations between the mean annual cost per patient and the clinical and demographic variables. The suitability of the model was verified from a residue analysis and the level of significance adopted was 5%. RESULTS: 28,401 patients were identified and subsequently 23,082 patients were analyzed. The majority of the patients were female (73.3%), lived in the southeast region (58.9%), had a mean age of 36.8 (± 12.2) years and started treatment using one of the interferons beta (78.9%). The total direct medical cost spending in the sixteen years of the follow-up was US $ 2,308,393,465.60, and the mean annual expenditure per patient was US $ 13,544.40 (± 4,607.05). In the best fit model (p <0.001), approximately 40% of the variability of the mean annual cost per patient was explained by the region of residence; medication used (intention to treat); if the patient was a non-exclusive user of medicines, i.e., used SUS for other procedures other than high-cost medicines; year of treatment start; and presence of events (death; Relapse; change of treatment and/or comorbidity). CONCLUSIONS: In the public health system of Brazil, disease modifying therapies currently represent almost all of the total direct costs of multiple sclerosis treatment. Around the world, new and emerging health technologies to treat of MS impose a challenge to health budgets, highlighting the need for cost-effectiveness studies comparing these technologies to those already available. Our regression model may help in this process, and calls attention to the need to access the real world performance of new therapies available in SUS, with the potential for disinvestment and/ or price reductions if needed.