ABC | Volume 114, Nº3, March 2020

Original Article Silva et al. Warfarin therapy in NVAF patients in Brazil Arq Bras Cardiol. 2020; 114(3):457-466 Table 2 – PMPY costs with and without major bleeds (R$) Values. Costs are expressed as (mean/±SD) Q1 Q2 Q3 Q4 Total All patients Number of patients 303 306 305 306 1220 Cost per patient (total) 39,171 (± 59,728) 33,996 (± 48,637) 31,797 (± 42,030) 24,236 (± 35,158) 32,284 (± 47,480) Cost per outpatient 14,417 (± 31,295) 11,425 (± 18,544) 11,760 (± 17,866) 8,719 (± 12,084) 11,573 (± 21,218) Cost per inpatient 24,754 (± 45,652) 22,570 (± 43,267) 20,037 (± 40,199) 15,517 (± 35,849) 20,710 (± 41,725) INR cost per patient 296 (± 187) 405 (± 223) 417 (± 214) 329 (± 194) 362 (± 211) Without major bleedings Number of patients 285 292 291 301 1169 Cost per patient (total) 36,704 (± 58,663) 33,217 (± 49,138) 30,244 (± 40,852) 24,106 (± 35,376) 30,981 (± 46,858) Cost per outpatient 13,957 (± 31,658) 11,381 (± 18,955) 11,771 (± 18,261) 8,672 (± 12,181) 11,409 (± 21,419) Cost per inpatient 22,747 (± 44,912) 21,835 (± 44,213) 18,473 (± 38,417) 15,434 (± 36,298) 19,572 (± 41,328) INR cost per patient 291 (± 182) 400 (± 220) 416 (± 216) 329 (± 195) 359 (± 210) With major bleedings Number of patients 18 14 14 5 51 Cost per patient (total) 78,236 (± 64,550) 50,248 (± 33,950) 64,092 (± 53,895) 32,072 (± 17,703) 62,145 (± 52,163) Cost per outpatient 21,698 (± 25,175) 12,343 (± 6,755) 11,540 (± 6,290) 11,565 (± 3,997) 15,348 (± 16,170) Cost per inpatient 56,538 (± 49,698) 37,905 (± 30,108) 52,552 (± 50,731) 20,507 (± 15,797) 46,796 (± 44,386) INR cost per patient 382 (± 247) 523 (± 259) 432 (± 190) 357 (± 164) 432 (± 231) Conversion factor: 0.33 USD/BRL. and costs in the Brazilian routine practice. High quality of anticoagulation control was associated with a lower incidence of major and minor bleeds and substantial direct medical cost savings from both reduced inpatient and outpatient costs. Poorly‑controlled patients had 3.3 times more major bleeds and 40% higher PMPY costs than well‑controlled patients. Despite anticoagulation treatment, strokes will still occur, as observed in this study, both ischemic and hemorrhagic ones. Of note, out of 10 confirmed hemorrhagic strokes that were identified in this study, the preceding INR value for 7 of the 10 was within the therapeutic range of 2 to 3, with the other 3 being 3.66, 3.87, and 5.13. This is consistent with the findings from a sub-analysis of the ARISTOTLE trial which showed that for about 80% of the intracranial hemorrhages that occurred in warfarin-treated patients, the preceding INR was between 2 and 3. 34 Past research explored predictors of poor TTR 6,26,32 suggesting that patients with lower TTR were more often females, had less schooling andmore comorbidities, specifically diabetes, chronic kidney disease, heart failure and prior stroke. Quite consistently, female patients and patients with more comorbidities such as chronic kidney disease and ischemic heart disease tended to have lower TTR values in this study, too. Moreover, patients with lower TTR had fewer INR tests and a shorter overall monitoring period. The results suggest that there is a need to identify patients with labile INRs and further assess opportunities to improve their TTR, such as education or closer follow-up. Failing that, other forms of anticoagulation such as the more recently approved non‑vitamin K anticoagulant class should be considered. This class does not require routine monitoring, has fewer drug-drug and drug-food interactions than warfarin, and has been shown to be at least as safe and efficacious as well-controlled warfarin, and to have a lower rate of intracranial haemorrhage. 35 Limitations Our study has several strengths and limitations. The patient cohort of the study was one of the largest thus far among real-world studies in Brazil. The combined use of claims and the care program added significant value to the study, especially by allowing the analysis of INR values, commonly not available in claims. However, given its retrospective observational nature, only associations could be concluded. This study observed TTR variations over time and as such was vulnerable to the effects of repeated measurements as an intervention. No advanced statistical techniques were used to balance characteristics of the TTR patient subgroups and therefore no inferential conclusions about cofactors could be drawn. We could not calculate the mean HAS-BLED risk score, as not all the data points of the score components were captured in the dataset (i.e. alcohol use). The incidence of other outcomes such as stroke, mortality, discontinuation and adherence was not analyzed. Sensitivity analyses at other specific TTR thresholds (e.g. 60% or 70%) were not conducted. The stability of INR over time was not assessed. Only direct medical costs were available; these referred to 463