Beaston-Blaakman Aaron, Shepard Donald, Horgan Constance, Ritter Grant
Family Health International, PO Box 13950, Research Triangle Park, NC 27709, USA,
J Ment Health Policy Econ. 2007 Mar;10(1):3-13.
Understanding variation in the cost of outpatient substance abuse treatment is important for improving the delivery and financing of care. Studies that examine how the cost of treatment relates to treatment program and client characteristics can provide important data about variables that affect unit costs of treatment. Such analyses can inform those who are responsible for setting appropriate reimbursement rates and can give important cost data to program directors responsible for delivering cost-effective treatment.
The aim of this study is to describe the results from cost function analyses of outpatient substance abuse treatment programs sampled in the Alcohol and Drug Services Study (ADSS). The ADSS is a national study conducted in the late 1990s to collect organizational, client, and cost data of the specialty sector.
The authors examined how organizational and client characteristics affect the cost per episode and the cost per enrollment day of outpatient care. The analysis incorporates organizational variables such ownership, average length of stay, and visits per enrollment day, as well as client characteristics such as gender, age, and primary drug of choice. For further applicability for current treatment policy, the ADSS cost data were inflated from 1997 to 2005 dollars. Mixed model regressions using log-log and log-linear relationships were developed.
Several organizational characteristics have statistically significant coefficients in the model estimating cost per episode, including log of point prevalence (-0.53, p<.01), log of average length of stay (0.73, p<.01), log of visits per enrollment day (0.45, p<.01), log of labor index (0.50, p<.01), proportion of counselor time spent in direct counseling (-0.52, p<.01), and location outside a metropolitan area (-0.19. p<.05). None of the client variables are statistically significant in this model. The analysis of cost per enrollment day indicates diseconomies of scope for programs that provide a broader array of ancillary services.
Findings suggest there exist increasing returns to scale in outpatient substance abuse treatment. Mergers of substance abuse treatment programs may be economically beneficial. Other major determinants of cost include the average length of stay, wage rates, visits per enrollment day, and direct client contact time. Increased efficiency may enable programs to control costs in these areas. In addition, many of the patterns identified in the model represent the way in which outpatient substance abuse treatment facilities are reimbursed for services. As these patterns become more specified for client conditions, client factors may become statistically significant in determining costs. The potential problem of endogeneity is addressed. Limitations of the study include possible inaccuracies in non-personnel cost data, changes in the treatment system unaccounted for in the model, and limited market area information with regard to input prices.
If further research indicates economies of scale, policymakers might consider supporting the merging of treatment programs. Also, further research into the optimal-mix of ancillary and treatment services would provide useful data for treatment programs seeking to balance resource constraints while providing important clinical and support activities. Lastly, research is needed to understand the relationship between treatment costs and service reimbursement.
了解门诊药物滥用治疗成本的差异对于改善医疗服务的提供和融资至关重要。研究治疗成本与治疗项目及客户特征之间的关系，可以提供有关影响治疗单位成本的变量的重要数据。此类分析可为负责设定适当报销率的人员提供参考，并能为负责提供具有成本效益治疗的项目主管提供重要的成本数据。
本研究旨在描述在酒精和药物服务研究（ADSS）中抽样的门诊药物滥用治疗项目成本函数分析的结果。ADSS是20世纪90年代末进行的一项全国性研究，旨在收集专业部门的组织、客户和成本数据。
作者研究了组织和客户特征如何影响门诊护理每次发作的成本和每个登记日的成本。分析纳入了组织变量，如所有权、平均住院时间和每个登记日的就诊次数，以及客户特征，如性别、年龄和首选主要药物。为了使ADSS成本数据更适用于当前治疗政策，将其从1997年的美元价值调整到了2005年的美元价值。使用对数-对数和对数-线性关系建立了混合模型回归。
在估计每次发作成本的模型中，几个组织特征具有统计学上显著的系数，包括点患病率的对数（-0.53，p<0.01）、平均住院时间的对数（0.73，p<0.01）、每个登记日就诊次数的对数（0.45，p<0.01）、劳动指数的对数（0.50，p<0.01）、咨询师直接咨询时间的比例（-0.52，p<0.01）以及大都市地区以外的地点（-0.19，p<0.05）。在该模型中，没有一个客户变量具有统计学显著性。对每个登记日成本的分析表明，提供更广泛辅助服务的项目存在范围不经济。
研究结果表明门诊药物滥用治疗存在规模收益递增。药物滥用治疗项目的合并在经济上可能是有益的。成本的其他主要决定因素包括平均住院时间、工资率、每个登记日的就诊次数以及与客户的直接接触时间。提高效率可能使项目能够控制这些领域的成本。此外，模型中确定的许多模式代表了门诊药物滥用治疗设施获得服务报销费用的方式。随着这些模式根据客户情况变得更加具体，客户因素在确定成本时可能会具有统计学显著性。研究解决了内生性的潜在问题，但研究的局限性包括非人员成本数据可能不准确、模型未考虑治疗系统的变化以及关于投入价格的市场区域信息有限。
如果进一步的研究表明存在规模经济，政策制定者可能会考虑支持治疗项目的合并。此外，对辅助服务和治疗服务的最佳组合进行进一步研究，将为寻求在平衡资源限制的同时提供重要临床和支持活动的治疗项目提供有用数据。最后，需要进行研究以了解治疗成本与服务报销之间的关系。
