Lost in Translation: Piloting a Novel Framework to Assess the Challenges in Translating Scientific Uncertainty From Empirical Findings to WHO Policy Statements
Bibliographic record
Abstract
BACKGROUND: Calls for evidence-informed public health policy, with implicit promises of greater program effectiveness, have intensified recently. The methods to produce such policies are not self-evident, requiring a conciliation of values and norms between policy-makers and evidence producers. In particular, the translation of uncertainty from empirical research findings, particularly issues of statistical variability and generalizability, is a persistent challenge because of the incremental nature of research and the iterative cycle of advancing knowledge and implementation. This paper aims to assess how the concept of uncertainty is considered and acknowledged in World Health Organization (WHO) policy recommendations and guidelines. METHODS: We selected four WHO policy statements published between 2008-2013 regarding maternal and child nutrient supplementation, infant feeding, heat action plans, and malaria control to represent topics with a spectrum of available evidence bases. Each of these four statements was analyzed using a novel framework to assess the treatment of statistical variability and generalizability. RESULTS: WHO currently provides substantial guidance on addressing statistical variability through GRADE (Grading of Recommendations Assessment, Development, and Evaluation) ratings for precision and consistency in their guideline documents. Accordingly, our analysis showed that policy-informing questions were addressed by systematic reviews and representations of statistical variability (eg, with numeric confidence intervals). In contrast, the presentation of contextual or "background" evidence regarding etiology or disease burden showed little consideration for this variability. Moreover, generalizability or "indirectness" was uniformly neglected, with little explicit consideration of study settings or subgroups. CONCLUSION: In this paper, we found that non-uniform treatment of statistical variability and generalizability factors that may contribute to uncertainty regarding recommendations were neglected, including the state of evidence informing background questions (prevalence, mechanisms, or burden or distributions of health problems) and little assessment of generalizability, alternate interventions, and additional outcomes not captured by systematic review. These other factors often form a basis for providing policy recommendations, particularly in the absence of a strong evidence base for intervention effects. Consequently, they should also be subject to stringent and systematic evaluation criteria. We suggest that more effort is needed to systematically acknowledge (1) when evidence is missing, conflicting, or equivocal, (2) what normative considerations were also employed, and (3) how additional evidence may be accrued.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.490 | 0.578 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.004 | 0.008 |
| Bibliometrics | 0.025 | 0.019 |
| Science and technology studies | 0.011 | 0.023 |
| Scholarly communication | 0.026 | 0.027 |
| Open science | 0.007 | 0.025 |
| Research integrity | 0.008 | 0.011 |
| Insufficient payload (model declined to judge) | 0.010 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".