For directors of nursing: knowledge that survives the schedule

Picture the night shift on your memory care unit. An agency aide, her second time in your building, is trying to get a resident back to bed at 2:50 a.m. He is pacing near the exit door, upset, saying he has to get to work. She has the care plan. It says "redirect." It does not say that he drove a bread delivery truck for thirty years and started his route at 3 every morning, and that the thing that settles him is being told the truck is loaded and he can go back to bed.

The day aide knew that. The day aide is asleep.

That night is illustrative, but every DON reading this has lived some version of it. This piece is for you, the director of nursing, wellness director or resident care director: the person who owns the care plan and the incident report, and who knows better than anyone how far apart those two documents can drift.

The care plan doesn't work the night shift

The math is against you. In nursing homes, average annual turnover among nursing staff runs around 128 percent, with a median near 94 percent, according to an analysis of payroll records from more than 15,000 facilities.¹ Agency workers covered about 11 percent of direct-care nursing hours in 2022, up from 3 percent in 2018.² RN hours per resident have fallen since 2015.³ And you, by many accounts, are not likely to stay forever either.

CMS has put this problem in writing. In its guidance on physical restraints, the agency lists as a concern for surveyors staff stating that "new staff and/or temporary staff do not know the resident, how to approach, and/or how to address behavioral symptoms."⁴ That is the regulator naming the gap between the plan and the person on shift.

The care plan says "redirect." Only the story tells you where to redirect him to.

Behaviors are often unmet needs, and needs are personal

You know the unmet-needs model. Jiska Cohen-Mansfield's group found that 89 agitated nursing home residents each had about three unmet needs on average, most commonly boredom or sensory deprivation, loneliness, and the need for meaningful activity. They also found that pain and discomfort were under-detected by staff.⁵ In a controlled trial in 12 nursing home buildings, interventions tailored to each resident's unmet need and matched to "their lifelong habits and roles" reduced agitation compared with control and increased pleasure and interest.⁶ One caveat worth remembering: research assistants delivered those interventions, not facility staff.

The larger person-centered care trials point the same way, with modest effects. In CADRES, person-centered care lowered agitation at follow-up across 15 sites (though falls were higher in that arm).⁷ In WHELD, across 69 UK nursing homes, a person-centered care and social interaction program improved quality of life and agitation by small amounts and increased positive care interactions by about 20 percent.⁸ A 10-month staff training program in 12 specialist homes cut neuroleptic use from 42 to 23 percent without changing agitation.⁹

I want to be precise about what that evidence means. Those were training and process programs, delivered with champions and support. None of them was a document, and nobody has tested a software briefing for those outcomes. What they show is that care built on knowing the person works modestly when the knowledge is actually used. Using it is the hard part.

Your regulations already ask for the story

If you are in a certified nursing home, the requirement is explicit. 42 CFR 483.20(b)(1) requires an assessment of each resident's "needs, strengths, goals, life history and preferences," including customary routine. The person-centered care plan under 483.21(b) has to be developed by an interdisciplinary team that includes "a nurse aide with responsibility for the resident," and services must be culturally competent and trauma-informed. F744 guidance on dementia care calls for "individualized, non-pharmacological approaches," and describes meaningful activities as those that address "customary routines, interests, preferences, and choices."¹⁰

The same goes for documentation you already worry about. MDS Section F asks how important daily routines and activities are to each resident, and when a PRN psychotropic or a dose reduction comes up, a documented attempt at a non-drug approach drawn from the person's history is the kind of documentation surveyors expect to see. A life story is raw material for both.

Assisted living is governed state by state, and many states ask for preferences and routines in the service plan. See our state rules library for your state, and life history in senior care regulations for the federal detail.

So collecting life history is not optional. What is missing in most buildings is a way for it to reach the CNA on the IDT, the night aide and the agency nurse, in seconds, every time.

Why the binder fails, and what works instead

The best study of life story documents in care, Gridley and colleagues' 2016 work for the UK's NIHR, found they mostly go unused. "Doing LSW is one thing," they wrote, "using it to inform and improve care is clearly another." Only 24 percent of services said all staff could see the life story, and none of the generalist care homes did.¹¹ The same study warned about stale stories and about freezing people in the past: current preferences must always win over old ones.

From that research, the design rules for a DON are fairly clear:

• Short and at the point of care. If it takes more than a minute to read, an aide with nine residents will not read it.
• Visible to everyone on shift, including agency staff. Not locked in the chart.
• Written as actions. "Tell him the truck is loaded" beats "enjoyed his career in delivery."
• Current. Dated, updated, with room for staff to add what they learn.
• Consented and bounded. Intimate or clinical details don't belong on a door card.
• No new charting. You know better than anyone that a tool called "more documentation" is dead on arrival.

Where Porchlight fits, and where it doesn't

Porchlight was designed around those rules. The resident does the recording: they tap one big button on a tablet and answer questions read aloud, in a fixed order that starts gentle, skipping anything they want. The AI transcribes and lightly tidies what they said, and a screen checks the tidy version against the original. Nobody on your staff types a word.

What your staff get is short: a Know-Your-Resident briefing with conversation starters, a printable door card, and a first-shift card written for someone who has never met the resident. Topics to avoid sit on the profile. There is a per-resident stage setting for dementia: later-stage residents get a slower, partner-paced session with either/or questions, and it stops after two unanswered questions with a warm close and a note to staff. Family and staff can add proxy stories for residents who can no longer tell their own, marked as proxy. Consent runs on a paper form with an opt-out, and there is a page that lays out who sees what.

Here is what it does not do, and I would rather you hear it from me. It does not detect distress. It does not diagnose, assess pain, or flag clinical changes. It is not therapy, and I will not claim it reduces agitation or psychotropic use on its own. It is a way to get a resident's own account of who they are to every person on your schedule. What your team does with it is still nursing.

Something to try on your unit this month

Choose the three residents who generate the most incident reports or PRN calls. For each, write one line on a card by the door: what they like to be called, one thing from their life that settles them, one thing to avoid. Get it from the family, the activity director, or the resident. Brief the night shift on it. Watch the next month of notes.

If that changes anything, you have found the mechanism. Then the question is only how to do it for everyone, without it depending on who happens to be working.