PUBH 8033 Week 7 Effect Size Argument Example

Real but Too Small to Matter? An Effect Size Argument Translating an Illustrative Crash Rate Ratio for an Older Driver Class Into Crashes Avoided per Hundred Completers, the Number Who Must Finish for One Fewer Crash and the Cost of Each

Student Name

Walden University

PUBH 8033: Interpretation and Application of Public Health Data

Instructor Name

Month Day, Year

Real but Too Small to Matter? An Effect Size Argument Translating an Illustrative Crash Rate Ratio for an Older Driver Class Into Crashes Avoided per Hundred Completers, the Number Who Must Finish for One Fewer Crash and the Cost of Each

From a Ratio to Counts People Can Use

A rate ratio of 0.81 for crash involvement after Larkin County's driving class for people over 70 sounds like a 19% reduction. Decision makers, however, cannot fund a percentage. They fund seats, and they need to know how many crashes those seats might prevent and at what cost. This brief translates the illustrative unadjusted estimate, and the adjusted one, into counts. All figures are composite.

Crashes Avoided per 100 Completers

Among waitlisted applicants, the crash rate was 5.7 per 100 in the following year. Among completers it was 4.6 per 100. Crediting the class with the entire gap, each 100 people who finish it would be spared roughly one crash per year. Using the adjusted rate ratio of 0.92, the difference shrinks to about 0.5 crashes avoided per 100 completers. The adjusted figure is the more credible of the two, since drivers who enroll in such courses early can differ from later applicants before any teaching happens, a self-selection pattern documented for another mature driver course (Nasvadi & Vavrik, 2007).

Number Needed to Complete

The number who must complete the class for one fewer crash is the inverse of the risk difference. With the unadjusted estimate, about 95 people must complete the class to avoid one police-reported crash in a year. With the adjusted estimate, the number rises to about 220. Because both confidence intervals include no effect, the number needed could be infinite, meaning no crashes prevented at all.

Table 1

Effect Size Under Two Estimates

Note. Composite figures; seat cost includes instructor time, room and materials.

Weighing Cost

A seat costs the county about $85. At the unadjusted estimate, each crash avoided would cost about $8,100; at the adjusted estimate, about $18,700. Most police-reported crashes are not fatal, and many cause only property damage, so the value of preventing one varies widely. For a program whose benefit is uncertain and possibly zero, those costs are hard to justify on crash prevention alone.

Other Effects That Might Matter

The class might have benefits not captured by crashes. In an Alabama randomized trial with visually impaired older drivers, those taught safe-driving strategies said they regulated their own driving more, avoiding challenging driving situations such as night driving, even though crash rates did not differ significantly (Owsley et al., 2004). Such changes might matter to families even without fewer crashes. But they could also mean participants drove less, shifting risk to other forms of travel. A systematic review found stronger evidence for programs that combine education with on-road training (Korner-Bitensky et al., 2009), which suggests a different use for the same money.

The Argument

The effect, if real, is small: about one crash avoided per hundred completers at best, and perhaps half that or less after accounting for who got seats. That is too small, and too uncertain, to justify expanding the classroom-only program. It is large enough, however, to justify testing a stronger version with a fair comparison.

References

Korner-Bitensky, N., Kua, A., von Zweck, C., & Van Benthem, K. (2009). Older driver retraining: An updated systematic review of evidence of effectiveness. Journal of Safety Research, 40(2), 105-111. https://doi.org/10.1016/j.jsr.2009.02.002

Nasvadi, G. E., & Vavrik, J. (2007). Crash risk of older drivers after attending a mature driver education program. Accident Analysis & Prevention, 39(6), 1073-1079. https://doi.org/10.1016/j.aap.2007.02.005

Owsley, C., McGwin, G., Phillips, J. M., McNeal, S. F., & Stalvey, B. T. (2004). Impact of an educational program on the safety of high-risk, visually impaired, older drivers. American Journal of Preventive Medicine, 26(3), 222-229. https://doi.org/10.1016/j.amepre.2003.12.005