A research cutoff decides who receives support: read that rule first

Ask what the threshold changes

Cochrane’s nonrandomized-design glossary describes a threshold on a baseline measure that determines intervention assignment. It also distinguishes situations where the probability of receiving the intervention changes at a cutoff. Those descriptions identify study designs, not proof that a particular threshold comparison is sound.

Locate whether the cutoff changed access or assignment before the outcome occurred, or whether researchers applied it afterward to classify outcomes. The same word cutoff can serve either purpose. Write both the measured variable and what crossing its boundary did. Do not infer the assignment rule from a graph alone.

Cochrane: Defining and determining which quantitative study designs to include in your systematic review of effects of a healthcare intervention (April2024)

Use a fictional administrative rule

Imagine an invented support project with an administrative priority index and a stated boundary for offering one arrangement. The index is not a health scale and no number here is a clinical recommendation. Draw the index line and the offer boundary, then place actual receipt on a separate line: an offer and acceptance remain different facts.

Now imagine a separate report that calls someone a responder after a specified change in a routine score. That second cutoff labels an outcome; it does not create the first project’s assignment design. The contrast is a reading exercise only. Neither imaginary project establishes a quitting effect or authorizes a reader to imitate the boundary.

Read how the analysis uses the boundary

Find which observations entered the comparison and how the authors describe the result’s target. A report might focus on people near a boundary rather than everyone in the service population. Do not turn that scope into a universal effect without checking the paper’s own explanation and applicability discussion.

Also read whether the stated rule was followed and what the authors say about other changes associated with the cutoff. If the assignment variable or procedure is unclear, you have not yet established the design. Assessing the assumptions and the model requires methodological expertise; recognizing a threshold is only the start of the appraisal.

Keep the design claim proportionate

A bounded note identifies the assignment variable, boundary, and whether the paper describes assignment or probability of receipt changing there. Then name the population and outcome the reported estimate covers. If all you located is a responder definition, do not call the study regression discontinuity because its results contain a cutoff.

This guide neither analyzes a policy nor gives legal eligibility rules or treatment advice. An individual health decision needs appropriate professional interpretation of actual evidence. A personal puff goal is a planning rule, not a research assignment mechanism. The useful output is a precise description of what the threshold did and the questions the full methods still need to answer.