Translating Evidence-Based Lighting into Usable Scenes
Bridging the gap between lighting science and clinical practice — designing lighting scenes that encode evidence-based health benefits into configurations nurses can actually understand and use.
- Role
- Lead Product Experience Engineer
- Year
- 2019–2026
- Organisation
- Chromaviso A/S
Outcome
Lighting scene frameworks deployed across clinical settings, with measurably improved staff adoption rates compared to earlier, purely functional approaches.
Chromaviso’s products are grounded in evidence. The science behind health-promoting lighting — circadian rhythm support, alertness management, patient comfort, surgical precision — is well-established. The design challenge is a translation problem: how do you take that science and put it in the hands of a nurse who has thirty seconds and no particular interest in photobiology?
The problem with “evidence-based”
Evidence-based product features often fail at the UX layer. The research says one thing; the user sees something else entirely. In lighting, this tension is acute.
A lighting system might have scientifically optimal scenes for wound healing, post-operative recovery, and staff alertness — but if those scenes are labelled with parameter values (“4000K, 500 lux, CRI 90”) rather than human-readable intent (“Examination”, “Rest”, “Night round”), clinicians will ignore them and use the manual override instead.
I spent significant time working out how to name, structure, and present scenes in a way that encoded the evidence without exposing the science.
Designing the scene model
The scene design work involved several layers:
Naming and categorisation — working with clinical staff and Chromaviso’s research team to develop a vocabulary for lighting scenes that matched how nurses and doctors actually think about their lighting needs, not how lighting engineers do.
Transition behaviour — when a patient moves from a procedure to recovery, the lighting should change. Designing the timing and character of those transitions (how fast, how obvious, who triggers them) required understanding clinical handover moments.
Contextual defaults — different ward types have different lighting baselines. An ICU default is not a surgical theatre default is not a psychiatric ward default. Designing a scene framework flexible enough to accommodate this while remaining consistent enough to be learnable took multiple iterations across real clinical environments.
Override logic — clinical staff will always override automated systems when they conflict with immediate patient needs. Designing graceful override behaviour — and understanding when to restore defaults — is as important as designing the scenes themselves.
What I learned about designing with science
The most useful framing I found: evidence-based features should be invisible infrastructure, not visible selling points. When lighting supports circadian rhythms, the nurse doesn’t need to know that’s what it’s doing. They need to know that “Night Round” mode won’t wake the patient. The science lives underneath; the interface speaks the user’s language.