건강보험심사평가원 자료 기반 고혈압 환자의 고혈압치료제 사용 현황과 처방 지속성 분석

Abstract

본 연구에서는 합병증을 동반하지 않은 치료를 시작하는 고혈압 환자에서 치료 시작 후 1년 동안의 순응도 및 지속성을 평가하고자 하였다. 더불어, 초기 치료 약제의 계열 및 성분의 선택이 치료 순응도 및 지속성에 미치는 영향을 확인하고자 하였다. 건강보험 심사 평가원의 자료를 이용하여, 2012년 고혈압 치료를 처음 시작하는 합병증을 동반하지 않은 전체 환자를 대상으로 하였다. 이후 크게 5가지 코호트로 구분하여 분석을 진행하였다. 순응도는 의약품소지율(medication possession ratio, MPR)을 사용하였으며, 지속은 고혈압 약제 처방의 중단 없이 지속된 경우로 정의되며, 처방간 간격이 60일 미만인 경우 지속으로 간주하였다. 초기 치료 약제 계열 선택에 따른 영향을 살피기 위하여, 단독 요법으로 치료를 시작하는 고혈압 환자의 20%를 단순 무작위 추출하여(45,787명) 분석을 진행하였다. 분석 결과 angiotensin receptor blockers와 비교시, thiazide diuretics (tz-D, adjusted hazard ratio [aHR] 3.16, 95% confidence interval [CI] 2.96-3.74) 와 selective beta-blockers (s-BB, aHR 1.86, 95% CI 1.77-1.95) 로 치료를 시작한 경우, 치료 시작 후 1년 시점에 고혈압 치료 중단의 위험이 높았으며, calcium channel blockers (CCB)에서도 차이는 작으나 다소 높은 양상을 보였다(aHR, 1.12, 95% CI 1.08-1.15). 동일한 계열 내에서 초기 약제 성분의 선택 차이에 따른, 순응도 및 지속성 평가를 위하여, 프로펜서티 스코어 매칭 방법을 이용해 ARB 또는 CCB 단독 요법으로 치료를 시작한 합병증을 동반하지 않은 환자 코호트를 구축하였다. Losartan 코호트와 non-losartan 코호트 사이에 치료 후 12개월 시점의 치료지속율 (각각 66.82% 및 68.25%) 및 순응률(각각 67.48% 및 69.01%)은 차이를 보이지 않았으며, 마찬가지로 amlodipine 코호트와 non-amlodipine 코호트간에도 치료지속율(각각 64.59% 및 64.25%) 및 순응률(각각 65.97% 및 67.07%)에서 차이를 나타내지 않았다. ARB 기반 병용요법으로 치료를 시작하는 환자에서, 복합성분단일제 (N=20,175) 와 단일성분병용(N=20,175) 차이에 따른 순응도 및 지속성을 확인한 결과, 복합성분단일제로 치료를 시작한 경우, 초기 1년 동안 치료 중단(aHR 0.75, 95% CI 0.72-0.78) 및 초기 약제 중단(aHR 0.62, 95% CI 0.61-0.64)의 위험이 낮고, 치료 순응할 가능성이 높은 것으로 확인되었다(adjusted odds ratio [aOR] 1.31, 95% CI 1.25-1.37). 복합성분단일제로 치료를 시작한 환자가운데, ARB/CCB코호트와 ARB/tz-D코호트간에 치료 지속율(각각 66.23% 및 65.99%) 및 순응율(각각 67.92% 및 67.53%)에는 유의한 차이를 보이지 않았다. 본 연구를 통해 합병증을 동반하지 않은, 고혈압 환자에서 초기 치료에 선택된 약제 계열에 따른 지속성 및 순응도의 차이를 확인할 수 있었으며, ARB 및 CCB 계열 내 성분에 따른 치료 지속 및 순응도의 차이는 없었다. ARB 기반 병용요법으로 치료를 시작하는 경우, 복합성분단일제로 시작하는 것이 단일성분병용보다 치료 및 초기 약물 지속, 치료 순응율이 높은 것을 확인하였다. 그러나, ARB/CCB 복합제와 ARB/tz-D 복합제간 치료 지속 및 순응도는 차이가 없었다. 본 연구 결과는 합병증이 없는 고혈압 신환자의 약물 선택에 있어서, 환자 약제 복용 행태를 고려할 수 있는 근거를 제공하는데 그 의의가 있다.|Objectives : We aimed to assess one-year persistence and adherence with antihypertensive therapy (AHT) among newly treated, uncomplicated hypertensive patients in Korea and to evaluate the effect of initial therapeutic classes and drugs on persistence using National Health Insurance claims database. Methods : We identified uncomplicated, treatment naïve hypertensive patients who started AHT in 2012 from national claims data. We then separated the cohort of five. Adherence to AHT was assessed with the medication possession ratio and persistence was defined when there was less than 60 days between prescriptions. Results : To determine the effect of initially chosen therapeutic class, we retrospectively analyzed a random sample of 20% of newly treated uncomplicated hypertensive patients (n=45,787) in 2012. Overall, 62.1% and 42.0% were persistent with any AHT and initial class at one year, respectively, and 64.2% were adherent to antihypertensive treatment. Compared with angiotensin receptor blockers, the risk of treatment non-persistence was significantly increased with initial use of thiazide diuretics (tz-D, adjusted hazard ratio [aHR] 3.16, 95% confidence interval [CI] 2.96-3.74) or selective beta blockers (aHR 1.86, 95% CI 1.77-1.95) and was minimally increased with calcium channel blockers (CCB, aHR 1.12, 95% CI 1.08-1.15). For the analysis of the influence of initial drug, we identified 55,504 patients who started ARB and 9,992 randomly selected patients who started CCB in 2012 among uncomplicated, treatment naïve hypertensive patients after propensity score matching. The proportion of patients remaining on any hypertension treatment at 12 months and the adherence rate were similar between the losartan cohort (66.82% and 68.25%) and the non-losartan ARB cohort (67.48% and 69.01%). Also, there were no differences in the ratio of the patients remaining on any hypertension treatment at 12 months (64.59% and 64.25%) or adherent to antihypertensive treatment (65.97% and 67.07%) between amlodipine and non-amlodipine cohort was similar. To analyze the influence of single pill combination (SPC) initiation, 20,175 patients were included in each SPC and free equivalent combination (FEC) cohort after propensity score matching. After adjusting clinical factors, patients who started SPC were less likely to be non-persistent to AHT (aHR 0.75, 95% CI 0.72-0.78) and initial drug combination (aHR 0.62, 95% CI 0.61-0.64) and more adherent to AHT (aOR 1.31, 95% CI 1.25-1.37). Among patients who started ARB based SPC therapy, treatment persistence (66.23% and 65.99%) and adherence rate (67.92% and 67.53%) between ARB/CCB SPC cohort and ARB/tz-D SPC cohort were not different. Conclusion: We observed meaningful differences in both persistence and adherence among initial AHT classes. Persistence and adherence to AHT were not influenced by the initially prescribed antihypertensive medication within the ARB and CCB class. In addition, treatment initiation with SPC is associated with increased first year treatment persistence and adherence in uncomplicated hypertensive patients compared with two pill combination. Also, there were no differences in both persistence and adherence to AHT between ARB/CCB and ARB/tz-D cohort. These study findings could have significant impacts for providing the comprehensive understandings of the medication taking behavior in uncomplicated, treatment naive hypertensive patients, which would be helpful in choosing initial medications.; Objectives : We aimed to assess one-year persistence and adherence with antihypertensive therapy (AHT) among newly treated, uncomplicated hypertensive patients in Korea and to evaluate the effect of initial therapeutic classes and drugs on persistence using National Health Insurance claims database. Methods : We identified uncomplicated, treatment naïve hypertensive patients who started AHT in 2012 from national claims data. We then separated the cohort of five. Adherence to AHT was assessed with the medication possession ratio and persistence was defined when there was less than 60 days between prescriptions. Results : To determine the effect of initially chosen therapeutic class, we retrospectively analyzed a random sample of 20% of newly treated uncomplicated hypertensive patients (n=45,787) in 2012. Overall, 62.1% and 42.0% were persistent with any AHT and initial class at one year, respectively, and 64.2% were adherent to antihypertensive treatment. Compared with angiotensin receptor blockers, the risk of treatment non-persistence was significantly increased with initial use of thiazide diuretics (tz-D, adjusted hazard ratio [aHR] 3.16, 95% confidence interval [CI] 2.96-3.74) or selective beta blockers (aHR 1.86, 95% CI 1.77-1.95) and was minimally increased with calcium channel blockers (CCB, aHR 1.12, 95% CI 1.08-1.15). For the analysis of the influence of initial drug, we identified 55,504 patients who started ARB and 9,992 randomly selected patients who started CCB in 2012 among uncomplicated, treatment naïve hypertensive patients after propensity score matching. The proportion of patients remaining on any hypertension treatment at 12 months and the adherence rate were similar between the losartan cohort (66.82% and 68.25%) and the non-losartan ARB cohort (67.48% and 69.01%). Also, there were no differences in the ratio of the patients remaining on any hypertension treatment at 12 months (64.59% and 64.25%) or adherent to antihypertensive treatment (65.97% and 67.07%) between amlodipine and non-amlodipine cohort was similar. To analyze the influence of single pill combination (SPC) initiation, 20,175 patients were included in each SPC and free equivalent combination (FEC) cohort after propensity score matching. After adjusting clinical factors, patients who started SPC were less likely to be non-persistent to AHT (aHR 0.75, 95% CI 0.72-0.78) and initial drug combination (aHR 0.62, 95% CI 0.61-0.64) and more adherent to AHT (aOR 1.31, 95% CI 1.25-1.37). Among patients who started ARB based SPC therapy, treatment persistence (66.23% and 65.99%) and adherence rate (67.92% and 67.53%) between ARB/CCB SPC cohort and ARB/tz-D SPC cohort were not different. Conclusion: We observed meaningful differences in both persistence and adherence among initial AHT classes. Persistence and adherence to AHT were not influenced by the initially prescribed antihypertensive medication within the ARB and CCB class. In addition, treatment initiation with SPC is associated with increased first year treatment persistence and adherence in uncomplicated hypertensive patients compared with two pill combination. Also, there were no differences in both persistence and adherence to AHT between ARB/CCB and ARB/tz-D cohort. These study findings could have significant impacts for providing the comprehensive understandings of the medication taking behavior in uncomplicated, treatment naive hypertensive patients, which would be helpful in choosing initial medications.