A vaping transition model can reach a state without a direct arrow
Separate a model’s direct-transition arrows from interval destination probabilities without inventing an unobserved personal route.
No Vape · Updated

Identify the two objects before comparing them
Look for a diagram of allowed transitions and a table or figure of probabilities over time. Record which one you are reading. A transition rate is not already a percentage for the next year. The original msm documentation separately describes a transition intensity matrix and a probability matrix for a specified interval, starting state and covariate values.
For a probability table, preserve the row’s starting state, the column’s destination and the time horizon. The same destination can mean different things depending on where the person started. Do not combine entries from different starting rows into one overall quit percentage or silently replace the report’s horizon with your own date.
Read why a direct edge was excluded
A primary PATH analysis of US adults used five product-use states and estimated continuous-time transitions from repeated surveys. Its model reduction omitted some rare direct transitions, while explicitly allowing an observed change between those states across a wave through an unobserved intermediate state. That is a modeling arrangement, not a claim that people can never change between the endpoint categories.
In a fictional three-state model, direct A-to-C movement is excluded, but A-to-B and B-to-C are allowed. A person can be observed in A at one assessment and C at another without contradicting that model. The example names no real product sequence and does not establish when or whether a particular person passed through B.
Do not turn an endpoint into an uninterrupted history
The PATH authors distinguish states observed at survey times from transitions between them whose timing and number were not directly known. Accordingly, being assigned the same category at two observations does not by itself supply a complete day-by-day history. A fitted model can represent intervening movement without turning those movements into observed diary entries.
Keep category definitions equally visible. In that analysis, non-current use referred to at least 30-day abstinence, and the authors used stopping rather than longer-term quitting language. Another report may define its states differently. Preserve its actual definition instead of substituting continuously vape-free, never used or nicotine-free.
Finish with a bounded model-reading statement
An original reading note can say: “This diagram shows allowed direct movements; this table estimates destinations over the stated interval; intervening routes are not all directly observed.” If a caption appears contradictory, check whether it changed objects before deciding the research contains an error. Leave an unexplained difference open for qualified interpretation.
This worksheet does not select a model, convert rates into probabilities or predict your next behavior. The historical PATH data do not supply a current personal forecast. Your practical quitting plan can use an actual situation and available support without assuming that a research state diagram dictates a sequence you must follow.
Sources and scope
Sources inform the guidance. The worksheets and examples are No Vape editorial suggestions, not tested treatments.
- Christopher Jackson / msm original documentation — Transition intensity and transition probability matrices
- Brouwer and colleagues / Tobacco Control — Transitions between cigarette, ENDS, and dual use in adults in the PATH study (Waves 1–4): multistate transition modeling accounting for complex survey design
- No Vape — No Vape — app features and store links
Written with AI assistance and checked against the cited sources. No clinical review is claimed. This guide covers practical planning, not individual medical treatment. How we research and write