Three possible prioritization schemes have been developed. All of them assume level funding at $500K per year, spread over seven years. Each prioritization scheme uses slightly different priorities and differs from the others in the exact studies funded.
Each prioritization scheme is illustrated with a Gantt chart, which presents a timeline and funding levels for the projects over the 7-year span. Arrows show which projects depend on the results of others. Numbers under each timeline show funding per year for each project.
The Strict Vote-based Prioritization strictly followed the votes given by the Expert Panel votes were cast on the basis of safety urgency, practicality and feasibility). Projects which received the most Expert Panel votes were assigned the highest priority. The Strict Vote-based Prioritization, therefore, is driven entirely by the votes cast by the Expert Panel.
The Diversification Strategy gave greater emphasis to the cost of individual projects. In order to diversify the investment in different research areas, lower-cost projects, from a wider range of topic areas, received higher priority. Overlapping topic areas were limited. If two projects addressed the same topic area, the less-expensive project was recommended. In this way, investment was spread out among as many investigations as possible.
The Topic Coverage Prioritization examined high-priority project in terms of topic coverage. This scheme maximizes coverage of high-priority topic areas, with a special emphasis on feasibility. Overlap of topics is minimized, but with attention paid to the topic coverage rather than price tag. That is, when two projects might have overlapped in the topic, the one providing greater coverage (either in a wider range or in terms of greater depth) was recommended.
Prioritization strictly followed the votes given by the Expert Panel (votes were cast on the basis of safety urgency, practicality and feasibility). The top-voted projects were assigned highest priority. As shown in Table 3, the projects are spaced out through the 7-year time frame, re-divided into subpackages and individual studies. The Table is in Gantt chart format and shows the funding level and timeline of each study.
The cost of individual projects was examined and was given weight in this prioritization scheme. If prioritization were to be assigned predominantly to very large projects, given budget limitations, a single expensive project might tie up all available resources in a calendar year. This limits the type of outcomes produced. In order to diversify the investment in research areas, lower-cost projects, from a wider range of topic areas, were assigned a higher priority in Table 4.
In order to spread out the investment, overlapping topic areas were limited. If two projects covered the same topic area, the less-expensive project was recommended.
For example, the Maryland Model Follow-up addresses many of the same issues as the Screening/Assessment projects - plus it already has a good deal of the infrastructure in place for investigation and follow-up. It may be limited in its ability to generate new instruments; but because the project is testing GRIMPS (a screening instrument developed for NHTSA), it may be wise in any case to let those results arrive before investing heavily in a different instrument. Whether results uphold the current model or point to necessary modifications, the next effort should attempt to maximize the return on those results.
It is clear that a greater number of projects is funded in this current approach: giving up any very large project enables the funding of several smaller programs.
A small project was needed to round out the funding in this scheme. A single Medium-priority project was recommended for funding in this prioritization scheme. “Crash Rates of Older Females”, an investigation of the Gender X Age crash rate interaction was recommended for funding at $125K.
In this prioritization scheme, high-priority projects have been examined with an eye to topic coverage. This scheme is intended to maximize coverage of high-priority topic areas, with a special emphasis on feasibility. Overlap of topics is minimized, but with attention paid to topic coverage rather than price tag (see Table 5).
Again, the Maryland Model Follow-up and the Screening/Assessment projects overlap in their emphasis. Again, the Maryland Model Follow-up is recommended for funding rather than the entire Screening/Assessment package, but a subset of the Screening/Assessment packages have been recommended where there is less overlap with the Maryland Model.
In general, less emphasis has been placed on retaining multi-study packages in their entirety. Where subpackages or individual studies assist in topic coverage, they are “folded back out” and assigned to the timeline.
The study on the “Rate of Impaired Status” is not recommended here for funding. Although the identification of the impaired population proportion is a laudable goal, there are serious challenges in terms of feasibility. Without a valid instrument to define “fitness” to drive, the measurement of this issue is impossible.
Other projects address the development of instruments to measure “fitness” to drive (screening tools and assessments). During the course of reliability and validation testing, the proportion of the population scoring as “unfit” should be revealed. It is recommended that this phase be made explicit in the funding of screening/assessment (specifically, in the Maryland Model Follow-up), and that a separate project not be funded to address this goal.
The “ITS and Implications” study has not been recommended for funding in this prioritization. Although it is important to determine the effects of ITS technology use, a portion of the ITS literature already contains older and younger samples. It is far more efficient for studies to include an older sample than it is to fund an entirely separate project exclusively devoted to age effects.