Adapting the IDEAL Framework and Recommendations for medical device evaluation: A modified Delphi survey.

INTRODUCTION: Current regulatory systems for medical device marketing approval lack adequate requirements for evidence of safety and efficacy. The Total Product Life Cycle (TPLC) concept, with clinical use and marketing expanding as evidence develops, has won support, but lacks a template to define evidence requirements at different stages. The IDEAL Framework & Recommendations, originally developed for new surgical procedures, might provide such a template, but may require modification. METHODS: We conducted a Delphi expert consensus exercise to determine how IDEAL might be modified to accommodate the needs of device regulation. 34 experts were invited to participate in 3 rounds of questioning, with feedback of the results of each round to participants before the next. RESULTS: 27 of 34 experts responded in at least one survey round. Experts agreed that, after appropriate modifications, IDEAL could form an evidence template for a TPLC-based regulatory system. Necessary modifications include a new Stage 0 should guide reporting of pre-clinical studies, expansion of registries to all stages, and omission of IDEAL stages 2 and 3 for "successor" devices under certain conditions. DISCUSSION: A standard approach to TPLC evaluation of medical devices does not currently exist. The IDEAL Framework, if modified appropriately, could fill such a void and improve the safety of new medical devices.

The Safer Delivery of Surgical Services Program (S3): Explaining Its Differential Effectiveness and Exploring Implications for Improving Quality in Complex Systems.

OBJECTIVE: To analyze the challenges encountered during surgical quality improvement interventions, and explain the relative success of different intervention strategies. SUMMARY BACKGROUND DATA: Understanding why and how interventions work is vital for developing improvement science. The S3 Program of studies tested whether combining interventions addressing culture and system was more likely to result in improvement than either approach alone. Quantitative results supported this theory. This qualitative study investigates why this happened, what aspects of the interventions and their implementation most affected improvement, and the implications for similar programs. METHODS: Semistructured interviews were conducted with hospital staff (23) and research team members (11) involved in S3 studies. Analysis was based on the constant comparative method, with coding conducted concurrently with data collection. Themes were identified and developed in relation to the program theory behind S3. RESULTS: The superior performance of combined intervention over single intervention arms appeared related to greater awareness and ability to act, supporting the S3 hypothesis. However, we also noted unforeseen differences in implementation that seemed to amplify this difference. The greater ambition and more sophisticated approach in combined intervention arms resulted in requests for more intensive expert support, which seemed crucial in their success. The contextual challenges encountered have potential implications for the replicability and sustainability of the approach. CONCLUSIONS: Our findings support the S3 hypothesis, triangulating with quantitative results and providing an explanatory account of the causal relationship between interventions and outcomes. They also highlight the importance of implementation strategies, and of factors outside the control of program designers.

Quality and safety on an acute surgical ward: an exploratory cohort study of process and outcome.

OBJECTIVE: To evaluate patient safety in an emergency surgical unit using process and outcome measures in parallel. BACKGROUND: Patient harm from errors in care is common in modern surgical practice. Measurement of the problem is essential to any solution, but current methods of evaluating patient harm are either impractical or inadequate. We have therefore analyzed compliance with safety-relevant care processes, with the aim of developing a process-based system for evaluating ward safety. METHODS: Adverse events (AE), potential adverse events (PAE), and 7 safety-relevant processes were measured on a 38-bed surgical emergency unit over a 16-week period. AE, PAE, and process measures were studied by prospective direct observation in large convenience samples, using objective measures. Possible influences on AE and PAE risk were analyzed. RESULTS: Compliance with the 7 processes studied ranged from 23% to 89%. The AE and PAE rates were 11.9% and 13.8% in a 63% sample of admissions (n = 607). Length of stay was significantly associated with both AE (P < 0.001) and PAE (P < 0.001). Having an operation was also associated with AE (P = 0.001) but not with PAE. No other factors appeared to influence AE/PAE rates. Delays were the commonest causes of both AE and PAE. CONCLUSIONS: Compliance with individual care processes on a ward with average levels of patient harm is poor. Length of hospital stay increases the risk of both AE and PAE, suggesting a system defect. A bundle of care processes may be useful for monitoring safety improvement.