Description
Score each known risk factor a person is exposed to as one, sum to a cumulative risk score, optionally balance it against a protective index, and target the highest-risk groups.
Exposure to several risks predicts poor outcomes better than any single risk, so the index counts them. Each empirically supported risk factor, spanning individual, family, neighbourhood and wider levels, is dichotomised and scored 1 if present; the sum is the cumulative risk score. A parallel cumulative protective index can be built from validated protective factors, and the balance between them examined. The index represents simultaneous risks on a common metric, is easy for decision-makers to read, works with existing data and improves power by reducing many variables to one. Interventions can aim to reduce the number of risks, starting with the factor that covaries most with others. The drawback is that the index cannot say which specific factor matters for which outcome, and dichotomising discards detail.
Attributes
- Discipline
- Statistics & Modelling
- Loss type
- Unmigrated
- Stage
- Strategy, Post-launch
- Data
- Quantitative
- Effort
- Days
Steps
- List empirically supported risk factors for the outcome across levels (individual, context, product exposure).
- Dichotomise each factor with a justified threshold and score exposure as 1.
- Sum to a cumulative risk score per person or group.
- Optionally build a cumulative protective index from validated protective factors and examine the balance.
- Check that outcomes rise with the score, then identify high-risk groups and target the factor that covaries most with the others.
Sources
- Handbook of Child Psychology Vol 4 (Lerner et al., 2015)
Also known as
- Cumulative risk index and cumulative protective index
- CRI
- CPI
- Cumulative risk score
- Risk-protection balance
Categories
- Analytics & Behaviour
- Statistics & Modelling