Department of Medicine, Clinical Epidemiology Unit, Karolinska Institutet, Stockholm.
Scand J Rheumatol. 2011 Jan;40(1):8-15. doi: 10.3109/03009742.2010.493895. Epub 2010 Oct 18.
To measure small-area variations in sales per capita of tumour necrosis factor (TNF) inhibitors.
For 2000-2009, sales data on etanercept, infliximab, and adalimumab were retrieved from the Swedish National Corporation of Pharmacies, which keeps data on drugs dispensed in ambulatory care and hospitals. As points of reference, data were retrieved on all drugs, non-biologic treatments for chronic inflammatory disorders (sulfasalazine, methotrexate, azathioprine), and for a biologic used in a different therapeutic area (trastuzumab). As a corollary measure to sales per capita, penetration of biologics in the rheumatoid arthritis (RA) population was calculated using nationwide registers. Small areas were defined as the 21 counties of Sweden.
From 2000 to 2009, annual TNF inhibitor sales increased 9-fold from 195 to 1779 million SEK (0.7-5.0% of total drug expenditure). The county variation in sales per capita, initially 6.2-fold (coefficient of variation 42%), decreased to 2.3-fold in 2009 (24%). During the same period, total drug expenditure per capita remained at a 1.2-fold county variation (4-6%). Sales per capita variations of non-biologic treatments against chronic inflammatory diseases ranged from 1.5 to 1.8 (12-16%). For trastuzumab, a 3.2-fold variation (30%) was observed in 2009. At the patient level, there was a 2-fold county variation (from 10% to 21%) in biologic penetration in RA. County-specific sales per capita were associated with mean RA duration (r = -0.52, p = 0.015) and C-reactive protein at treatment initiation (r = -0.49, p = 0.025), while pain was borderline significant (r = -0.43, p = 0.055).
Despite universal access to treatment, substantial but decreasing small-area variations were observed. Although geographic variations are anticipated initially, their persistence calls for investigation of patient equity and treatment appropriateness as counties seem to have different initiation thresholds.
测量肿瘤坏死因子（TNF）抑制剂人均销售额的小面积差异。
2000-2009 年，从瑞典国家药房公司检索到依那西普、英夫利昔单抗和阿达木单抗的销售数据，该公司保存了在门诊和医院使用的药物数据。作为参考点，检索了所有药物、慢性炎症性疾病的非生物治疗（柳氮磺胺吡啶、甲氨蝶呤、硫唑嘌呤）以及用于不同治疗领域的生物制剂（曲妥珠单抗）的数据。作为人均销售额的推论衡量标准，使用全国登记册计算生物制剂在类风湿关节炎（RA）人群中的渗透率。小区域被定义为瑞典的 21 个县。
2000 年至 2009 年，TNF 抑制剂年销售额从 1.95 亿瑞典克朗增加到 17.79 亿瑞典克朗（占总药物支出的 0.7-5.0%）。人均销售额的县际差异最初为 6.2 倍（变异系数 42%），到 2009 年降至 2.3 倍（24%）。在此期间，人均总药物支出的县际差异保持在 1.2 倍（4-6%）。针对慢性炎症性疾病的非生物治疗的人均销售额差异在 1.5 到 1.8 之间（12-16%）。2009 年，曲妥珠单抗观察到 3.2 倍的差异（30%）。在患者水平上，RA 中生物制剂的渗透率存在 2 倍的县际差异（从 10%到 21%）。县内特定的人均销售额与 RA 持续时间的平均值（r = -0.52，p = 0.015）和治疗开始时 C 反应蛋白（r = -0.49，p = 0.025）相关，而疼痛呈边缘显著（r = -0.43，p = 0.055）。
尽管有普遍的治疗机会，但仍观察到相当大但逐渐减少的小面积差异。尽管最初预计会有地理差异，但由于各县似乎有不同的启动阈值，因此其持续存在需要调查患者公平性和治疗适宜性。
