Public Health Laboratories and Health System Change . Appendix C: Poll of Public Health Laboratory Directors

10/06/1997

Purpose

This informal poll recorded and analyzed the responses of state public health lab directors in 48 states and the District of Columbia to three basic questions assessing how, if at all, managed care and other health system changes have affected the function of their laboratory. In addition to helping the project team identify states which were good candidates for case study or further investigation, the responses from this poll summarize the effects managed care and other forms of health system change have on state public health laboratories across the nation.

Study Methods

The informal poll began on March 5, 1997. We e-mailed or faxed the following three questions to all state PHL directors:1

Has managed care's presence in the area of laboratory services changed or affected the practices and/or functions of public health laboratories in your state? If so, how?

Have other market changes (e.g., commercial laboratories, hospital consolidation) had major impacts on the functions of public health laboratories? Please describe.

What (if any) contractual or collaborative arrangements exist between public health labs (state or other) and MCOs or commercial laboratories? Please describe.

Follow-up faxes were sent to public health laboratory directors who did not respond to the initial e-mail, and attempts were made to contact all of the public health laboratory directors who did not respond to e-mail or fax by phone. The study was concluded in early May, with 49 responses out of the 51 state and territorial public health laboratory directors polled.

Once gathered, the responses for each question were categorized by actor (the agent or organization having the effect on the PHL) and impact (the nature of the effect being described). Grouping responses into these categories allowed the detection of general trends across states for each of the three main question topics. The categorized responses were then coded, and statistical analysis software was used to perform tabulations, cross-tabulations, and other simple summary statistics on the data to flesh out and quantify the observable trends. Some simple multivariate analyses (logistic regressions) were used to control for important factors (census region and degree of managed care penetration) that may underlie some of the trends observed. Given that we had data from the census of states (except for Utah and Alabama), all differences reported can be considered to be statistically significant; however, we avoided highly specific quantification of multivariate analyses in favor of more generalizable results.

Study Limitations

In considering the results of this analysis (reported in the main text of this report) it is important to keep in mind the limitations of the instrument. The use of broad, open-ended questions has the advantage of eliciting unprompted opinions from PHL directors, but requires further analysis involving the synthesis of responses that are disparate in focus and level of detail. The anecdotal and vague nature of some of the responses complicated categorization efforts. As a result, a few responses between PHLs could not be compared directly, and the presence of a certain response from one PHL could not be contrasted with its absence from the responses of other PHLs. Despite these limitations, this analysis has revealed some important early impacts and actors in the relationship between PHLs and health system changes.


1 List of PHL directors in 50 states and the District of Columbia was obtained through the Association of State and Territorial Public Health Laboratory Directors (ASTPHLD)