oped (both state-mandate and business coalition models in Washington State, Des Moines, and Memphis) and three where they are not (Rochester, Albany, and Cleveland). Questions raised frequently by hosts were as follows:
- Who would run and administer such a database, and who would own the data?
- What would this effort cost and who would finance it, in terms of both fixed costs (for example, for computer equipment) and variable costs (over the short and long run, such as staff/personnel costs for data entry)?
- What is in it for me or what will this do to me (particularly from physicians in private practice and from employers with national interests whose health policies were set at a corporate level somewhere else)?
- Who would have access to patient-identified and provider-specific data? Interestingly, not everyone was worried about privacy of patient data, believing either that such information could be protected (so the question was moot) or that not much harm would come from judicious release to, for example, employers). Consumers understood the potential value of such databases, but they were also worried about access to patient-level data, especially concerning insurability.
- Could analyses about quality of care realistically be done (in contrast to analyses of cost or charges data or analyses of utilization patterns)?
The visiting committee members, in turn, tried to learn as much as possible about ways—in practice or in theory—groups in these various areas meant to realize the benefits of such databases, to minimize or prevent the exacerbation of current risks to the confidentiality of patient-level data, to improve database security, and to promote appropriate data collection, valid analyses, and useful dissemination of data.