von Wyl Viktor, Telser Harry, Weber Andreas, Fischer Barbara, Beck Konstantin
Institute for Epidemiology, Biostatistics & Prevention, University of Zürich, Zürich, Switzerland.
CSS-Institute for Empirical Health Economics, Luzern, Switzerland.
BMJ Support Palliat Care. 2018 Sep;8(3):325-334. doi: 10.1136/bmjspcare-2014-000784. Epub 2015 Oct 15.
Exploration of healthcare utilisation patterns in the final life year to assess palliative care potential.
Retrospective cluster analyses (k-means) of anonymised healthcare expenditure (HCE) trajectories, derived from health insurance claims of a representative sample of Swiss decedents who died between 2008 and 2010 (2 age classes: 4818 <66 years, 22 691 elderly).
3 (<66 years) and 5 (elderly) trajectory groups were identified, whose shapes were dominated by HCE from inpatient care in hospitals and at nursing homes. In each age class, the most expensive group (average cumulative HCE for <66 years: SFr 84 295; elderly: SFr 84 941) also had the largest abundance of cancers (<66 years: 55%; elderly: 32%) and showed signs of continued treatment intensification until shortly before death. Although sizes of these high-cost groups were comparatively small (26% in younger; 6% in elderly), they contributed substantially to the end-of-life HCE in each age class (62% and 18%, respectively).As age increased, these potential target groups for palliative care gained in share among <66-year olds (from 9% in children to 28% in 60-65-year olds), but decreased from 17% (66-70-year olds) to 1% (>90-year olds) among elderly.
Cost trajectory clustering is well suited for first-pass population screenings of groups that warrant closer inspection to improve end-of-life healthcare allocation. The Swiss data suggest that many decedents undergo intensive medical treatment until shortly before death. Investigations into the clinical circumstances and motives of patients and physicians may help to guide palliative care.
探究临终前一年的医疗服务利用模式，以评估姑息治疗潜力。
对2008年至2010年间死亡的瑞士死者代表性样本的医疗保险索赔数据得出的匿名医疗保健支出（HCE）轨迹进行回顾性聚类分析（k均值法）（分为2个年龄组：4818名年龄小于66岁者，22691名老年人）。
确定了3个（年龄小于66岁）和5个（老年人）轨迹组，其形状主要由医院和养老院住院护理的HCE决定。在每个年龄组中，费用最高的组（年龄小于66岁者平均累计HCE：84295瑞士法郎；老年人：84941瑞士法郎）患癌症的比例也最高（年龄小于66岁者：55%；老年人：32%），并且直到死亡前不久都有持续强化治疗的迹象。尽管这些高费用组的规模相对较小（年轻人中占26%；老年人中占6%），但它们在每个年龄组的临终HCE中所占比例很大（分别为62%和18%）。随着年龄增长，这些姑息治疗的潜在目标人群在年龄小于66岁者中的比例有所增加（从儿童中的9%增至60 - 65岁者中的28%），但在老年人中从17%（66 - 70岁者）降至1%（90岁以上者）。
费用轨迹聚类非常适合对那些需要更密切检查以改善临终医疗资源分配的人群进行初步筛查。瑞士的数据表明，许多死者在死亡前不久仍在接受强化医疗治疗。对患者和医生的临床情况及动机进行调查可能有助于指导姑息治疗。
