Exploring the Predictive Validity of Drug Evaluation and Classification Program Evaluations [Traffic Tech]
archive: archived pipeline: cataloged verified
Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)
Summary
This study investigates the predictive validity of the Drug Evaluation and Classification (DEC) program, which trains Drug Recognition Experts (DREs) to identify drug-impaired drivers. The DEC protocol utilizes a 12-step procedure to assess clinical and behavioral signs associated with seven drug categories. The primary objective was to determine which specific combinations of signs and symptoms within the DEC protocol most efficiently predict the drug category or combination responsible for impairment. A secondary objective involved analyzing cases ruled out by DREs due to medical conditions or lack of impairment to identify common characteristics. The study focused on identifying the most effective evaluative elements rather than assessing the overall accuracy of DREs in determining impairment. The researchers conducted a statistical analysis on a sample of 2,253 DEC evaluations from 11 U.S. states. Inclusion criteria required that the DRE’s opinion be confirmed by toxicological blood analysis. The dataset included single drug categories and common two-drug combinations, such as CNS depressants, stimulants, narcotic analgesics, cannabis, and alcohol. Additionally, a subset of "rule-out" cases was collected for qualitative review. Data from Drug Influence Evaluation face sheets, narrative reports, and toxicology reports were coded to create a database for statistical modeling. The results identified 22 drug-related signs and symptoms that significantly predicted the correct drug category, achieving an overall correct classification rate of 86.5%. Indicators related to eye appearance and physiological response were the strongest predictors, followed by clinical indicators and psychophysical test performance. Subject observations and statements were the least significant predictors. For drug combinations, 12 key indicators significantly predicted the combination, yielding a 75.6% classification rate. Specific indicators, such as pupil size, reaction to light, and horizontal gaze nystagmus, were critical for both single-category and combination predictions. Qualitative analysis of rule-out cases revealed that these subjects were typically older, more likely to have been involved in crashes, and frequently reported medical conditions like diabetes that could mimic drug effects. The findings confirm that specific DEC elements, particularly ocular indicators, provide high predictive validity for identifying drug categories. However, because prediction is not perfect, the study emphasizes the need for DREs to consider the totality of symptomatology and observational skills. The identification of medical conditions that mimic drug effects highlights the importance of distinguishing between impairment and medical issues. Focusing training on these key indicators may enhance the efficiency and validity of the DEC program, potentially improving enforcement of drug-impaired driving laws and increasing judicial acceptance of DRE evaluations.
Key finding
A set of 22 drug-related signs and symptoms from the DEC protocol achieved an 86.5% classification rate for predicting single drug categories and a 75.6% rate for predicting drug combinations.
Methodology
dataset
Sample size: 2253
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | rosap | — | — | 2 | 2026-05-23 |
| archive | success | — | — | — | 1 | 2026-05-23 |
| extract | success | cached | — | — | 2 | 2026-06-10 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-23 |
| promote | success | — | — | — | 1 | 2026-05-23 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 3 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 19 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Methodological Resource: validation psychometrics