Evaluating Road Safety Literature and Information

Evaluating Road Safety Literature and Information

Match the source to the decision

Evidence-based road safety begins with a clear question. Are you estimating the crash effect of a treatment, determining a traffic-control requirement, explaining an injury pattern, or selecting a communication strategy? Each question needs different evidence. A standard can establish a requirement without establishing a crash reduction. A controlled evaluation can estimate an effect without granting legal authority to implement the treatment. A resident can identify a missed problem without providing a reliable numerical CMF.

Start with the current TPCB knowledge and skills outline for exam scope. For practice decisions, use the responsible government agency for rules and programs, the original research for effect estimates, and relevant professional guidance for application. Peers are valuable for identifying operational issues and sources, but a colleague's successful project is not proof that the same treatment will have the same effect elsewhere.

Keep an evidence trail: source title, publication or revision date, facility and user population, outcome definition, method, and the claim it supports. If an older source is still authoritative, explain why. If a newer rule supersedes a handbook paragraph, use the rule for the current requirement and the handbook only for the remaining applicable concepts.

Distinguish five kinds of information

Standards and laws define obligations, permissions, and constraints within their jurisdiction. Check whether wording is a requirement, recommendation, or option. The MUTCD applies to traffic-control devices under its governing framework; it is not an all-purpose source for injury costs or public health surveillance.

Government data describe a defined population. FARS concerns qualifying fatal motor-vehicle traffic crashes; it does not count every injury or all transport-related deaths. State crash files have reporting thresholds and coding practices. A graph without the denominator, date, and scope is insufficient for a comparison.

Research studies test a relationship or estimate an intervention effect. Examine design, sample, uncertainty, exposure adjustment, selection bias, and applicability. A statistically significant average effect does not guarantee the same effect at a single site. A non-significant result may be imprecise rather than evidence that an effect is exactly zero.

Evidence summaries and guidance help practitioners find and interpret studies. The FHWA CMF Clearinghouse and NHTSA Countermeasures That Work serve different functions. A Clearinghouse star rating concerns the quality of the CMF evidence. NHTSA's behavioral countermeasure stars communicate its effectiveness assessment. Read the definitions rather than assuming the stars are interchangeable.

Qualitative information includes interviews, field observations, narratives, and community experience. It can reveal mechanisms or barriers omitted from coded files. Collect it systematically and seek corroboration. Several complaints may concern one repeated event; few complaints may reflect inaccessible reporting rather than a safe facility.

Critique a treatment claim

Suppose a vendor says a new device reduces crashes by 60%. Ask: compared with what, for which crashes, over what period, and using which method? An observed before-after decline is not necessarily the treatment effect. The site may have been selected following an unusual spike; traffic may have changed; construction may have diverted vehicles; reporting thresholds may have changed.

Find the underlying study and the untreated comparison. Check whether the estimate concerns all crashes, injuries, pedestrians, or a specific maneuver. Review confidence limits and whether adverse effects were measured. A device tested at rural curves cannot automatically supply a CMF for urban pedestrian crossings. Identify the baseline risk and the mechanism before transferring the result.

Now compare a second study reporting only a 20% reduction but using a more appropriate design and facility population. The larger percentage is not automatically the stronger evidence. Prefer applicable, credible evidence, explain uncertainty, and plan a local evaluation rather than selecting the most attractive headline.

Use research across disciplines and borders

FHWA and NHTSA provide infrastructure and behavioral resources. Transportation Research Board and NCHRP work can address methods and implementation. State research programs can provide locally relevant findings. For Canadian practice, the Canadian Council of Motor Transport Administrators (CCMTA), Transportation Association of Canada, and provincial or municipal agencies provide policy and technical context. IIHS, TIRF, AAA, and CAA also contribute research and synthesis; assess each publication's methods and scope rather than assigning credibility solely by organization name.

Canada's current national framework is Road Safety Strategy 2035 and Beyond, approved in 2026. It guides jurisdictions rather than making every U.S. program or rule applicable in Canada. A U.S. HSIP funding requirement is not a Canadian funding requirement. Conversely, human vulnerability and research methods can inform practice across borders when the population and operational context are considered. CCMTA strategy resources

Evidence checks before transfer

• Identify the outcome and untreated comparison.
• Check design, sample, uncertainty, and selection effects.
• Match facility, users, operation, and legal context.

Translate evidence into a decision

Summarize what the evidence supports, what it cannot resolve, and what local information is still needed. For a pedestrian crossing, an applicable CMF may support a candidate treatment, but traffic conditions, accessibility, legal authority, utilities, maintenance, and community needs still affect implementation. Record the trade-offs and decision owner.

A useful evidence statement is specific: “The selected study supports a reduction in the targeted crash type under comparable conditions; local volume and speed data are needed to assess applicability.” An unhelpful statement is “FHWA says this is safe.” The first can be checked and improved. The second obscures the source, outcome, and uncertainty.

For preparation, practice explaining one source's strengths and limitations aloud. Use the current TPCB reference list to select readings, then connect each to a blueprint task. Reading titles alone does not demonstrate the ability to critique or apply the evidence.