As predicted by theory: choice and competition in a publicly funded and regulated regional health system yield improved access and cost control.

BACKGROUND: New Public Management (NPM) has been widely used to introduce competition into public healthcare. Results have been mixed, and there has been much controversy about the appropriateness of a private sector-mimicking governance model in a public service. One voice in the debate suggested that rather than discussing whether competition is "good" or "bad" the emphasis should be on exploring the conditions for a successful implementation. METHODS: We report a longitudinal case study of the introduction of patient choice and allowing private providers to enter a publicly funded market. Patients in need of hip or knee replacement surgery are allowed to choose provider, and those are paid a fixed reimbursement for the full care episode (bundled payment). Providers are financially accountable for complications. Data on number of patients, waiting lists and times, costs to the public purchaser, and complications were collected from public registries. Providers were interviewed at three points in time during a nine-year follow-up period. Time-series of the quantitative data were exhibited and the views of actors involved were explored in a thematic analysis of the interviews. RESULTS: The policy goals of improving access to care and care quality while controlling total costs were achieved in a sustained way. Six themes were identified among actors interviewed and those were consistent over time. The design of the patient choice model was accepted, although all providers were discontent with the level of reimbursement. Providers felt that quality, timeliness of service and staff satisfaction had improved. Public and private providers differed in terms of patient-mix and developed different strategies to adjust to the reimbursement system. Private providers were more active in marketing and improving operation room efficiency. All providers intensified cooperation with referring physicians. Close attention was paid to following the rules set by the purchaser. DISCUSSION AND CONCLUSIONS: The sustained cost control was an effect of bundled payment. What this study shows is that both public and private providers adhere long-term to regulations by a public purchaser that also controls entrance to the market. The compensation was fixed and led to competition on quality, as predicted by theory.

Predicting clinical outcome and length of sick leave after surgery for lumbar spinal stenosis in Sweden: a multi-register evaluation.

PURPOSE: Lumbar spinal stenosis (LSS) can be surgically treated, with variable outcome. Studies have linked socioeconomic factors to outcome, but no nation-wide studies have been performed. This register-based study, including all patients surgically treated for LSS during 2008-2012 in Sweden, aimed to determine predictive factors for the outcome of surgery. METHODS: Clinical and socioeconomic factors with impact on outcome in LSS surgery were identified in several high-coverage registers, e.g., the national quality registry for spine surgery (Swespine, FU-rate 70-90%). Multivariate regression analyses were conducted to assess their effect on outcome. Two patient-reported outcome measures, Global Assessment of leg pain (GA) and the Oswestry Disability Index (ODI), as well as length of sick leave after surgery were analyzed. RESULTS: Clinical and socioeconomic factors significantly affected health outcome (both GA and ODI). Some predictors of a good outcome (ODI) were: being born in the EU, reporting no back pain at baseline, a high disposable income and a high educational level. Some factors predicting a worse outcome were previous surgery, having had back pain more than 2 years, having comorbidities, being a smoker, being on social welfare and being unemployed. CONCLUSIONS: The study highlights the relevance of adding socioeconomic factors to clinical factors for analysis of patient-reported outcomes, although the causal pathway of most predictors' impact is unknown. These findings should be further investigated in the perspective of treatment selection for individual LSS patients. The study also presents a foundation of case mix algorithms for predicting outcome of surgery for LSS. These slides can be retrieved under Electronic Supplementary Material.