Innovation Adoption Scale

Typical S-curve: Early Adopters (10–15% of users) adopt within first 1–2 months; Early Majority (next 30–40%) adopt by month 3–4; Late Majority (next 20–30%) by month 6–8; Laggards (last 10%) adopt reluctantly or not at all. High-complexity innovations (e.g., new EMR system) show slower adoption; simple innovations (e.g., new assessment form) may be adopted faster. Benchmark: 70% adoption by month 6 indicates healthy implementation trajectory; <50% by month 6 indicates problems requiring intervention.

Depends on clinical necessity and evidence base. For evidence-based practices shown to improve quality/safety (e.g., surgical checklists), adoption should be organizational policy (approaching 100% for eligible users). For practices where individual provider judgment is acceptable (e.g., choice of sedation protocols with multiple evidence-based options), adoption can be voluntary. Mixed approach: core components are mandatory; flexibility in implementation details is optional. Mandatory adoption typically reaches 90%+; voluntary adoption typically plateaus at 60–80%.

Define eligibility based on frequency/intensity of exposure. Example: For shared decision-making training, a clinician eligible if they conduct ≥2 patient encounters per week; those with <2 per week are not counted as eligible. For each eligible clinician, track whether they ever use shared decision-making (yes/no adoption); then track dosage (percentage of encounters using it). Report separately: 'Adoption (% of eligible users who ever use): 75%; Dosage (average % of encounters using practice): 60%.' This separates adoption (yes/no) from intensity (how much).

70% adoption is neither inherent success nor failure; interpretation depends on context. If 70% of essential clinical staff have adopted a high-impact practice, that may be sufficient to move organizational outcomes (outcomes improve even if not all staff adopt). If 70% includes only willing early adopters but core clinical leaders are among the 30% non-adopters, sustainability is at risk. Examine who is in the 70% vs. 30%; if non-adopters are opinion leaders or high-volume providers, targeted re-engagement is needed.

No—training completion is a process measure (staff were exposed to training), not an outcome measure (staff actually use the practice). Staff may complete training but not adopt (low adoption despite high training completion; indicates training-practice gap). Always measure actual use when possible. If direct measurement is infeasible, use the strongest proxy available (EHR usage logs, billing codes, patient self-report of receiving the practice) rather than training completion or survey intent.