The new final Clinical Skills examination in human medicine in Switzerland: Essential steps of exam development, implementation and evaluation, and central insights from the perspective of the national Working Group.

OBJECTIVE: Since 2011, the new national final examination in human medicine has been implemented in Switzerland, with a structured clinical-practical part in the OSCE format. From the perspective of the national Working Group, the current article describes the essential steps in the development, implementation and evaluation of the Federal Licensing Examination Clinical Skills (FLE CS) as well as the applied quality assurance measures. Finally, central insights gained from the last years are presented. METHODS: Based on the principles of action research, the FLE CS is in a constant state of further development. On the foundation of systematically documented experiences from previous years, in the Working Group, unresolved questions are discussed and resulting solution approaches are substantiated (planning), implemented in the examination (implementation) and subsequently evaluated (reflection). The presented results are the product of this iterative procedure. RESULTS: The FLE CS is created by experts from all faculties and subject areas in a multistage process. The examination is administered in German and French on a decentralised basis and consists of twelve interdisciplinary stations per candidate. As important quality assurance measures, the national Review Board (content validation) and the meetings of the standardised patient trainers (standardisation) have proven worthwhile. The statistical analyses show good measurement reliability and support the construct validity of the examination. Among the central insights of the past years, it has been established that the consistent implementation of the principles of action research contributes to the successful further development of the examination. CONCLUSION: The centrally coordinated, collaborative-iterative process, incorporating experts from all faculties, makes a fundamental contribution to the quality of the FLE CS. The processes and insights presented here can be useful for others planning a similar undertaking.

Governance mode vs. governance fit: Performance implications of make-or-ally choices for product innovation in the worldwide aircraft industry, 1942-2000

We examine the impact of governance mode and governance fit on performance in make-or-ally decisions. We argue that while horizontal collaboration and autonomous governance have direct and countervailing performance implications, the alignment of make-or-ally choices with the focal firm's resource endowment and the activity's resource requirements leads to better performance. Data on the aircraft industry show that relative to aircraft developed autonomously, collaborative aircraft exhibit greater sales but require longer time-to-market. However, governance fit increases unit sales and reduces time-to-market. We contribute to the alliance and economic organization literatures. (Copyright © 2013 John Wiley & Sons, Ltd.)

Intervals between response choices on a single-item measure of quality of life.

BACKGROUND: A single overall rating of quality of life (QoL) is a sensitive method that is often used in population surveys. However, the exact meaning of response choices is unclear. In particular, uneven spacing may affect the way QoL ratings should be analyzed and interpreted. This study aimed to determine the intervals between response choices to a single-item QoL assessment. METHODS: A secondary analysis was conducted on data from the Lc65+ cohort study and two additional, population-based, stratified random samples of older people (N = 5,300). Overall QoL was rated as excellent, very good, good, fair or poor. A QoL score (range 0-100) was derived from participants' answers to a 28-item QoL assessment tool. A transformed QoL score ranging from 1 (poor) to 5 (excellent) was calculated. The same procedure was repeated to compute seven domain-specific QoL subscores (Feeling of safety; Health and mobility; Autonomy; Close entourage; Material resources; Esteem and recognition; Social and cultural life). RESULTS: Mean (95 % confidence intervals) QoL scores were 96.23 (95.81-96.65) for excellent, 93.09 (92.74-93.45) for very good, 81.45 (80.63-82.27) for good, 65.44 (62.67-68.20) for fair and 54.52 (45.31-63.73) for poor overall QoL, corresponding to transformed QoL scores of respectively 5.00, 4.70, 3.58, 2.05, and 1.00. Ordinality of the categories excellent to poor was preserved in all seven QoL subscores. CONCLUSIONS: The excellent-to-poor rating scale provides an ordinal measure of overall QoL. The intervals between response choices are unequal, but an interval scale can be obtained after adequate recoding of excellent, very good, good, fair and poor.