> ## Content Index > Fetch the complete content index at: https://inside-systems.ghost.io/llms.txt > Use this file to discover other available public pages before exploring further. # Thirst stopped telling you the truth - URL: https://inside-systems.ghost.io/thirst-stopped-telling-you-the-truth/ - Published: 2026-08-19T09:05:58.000Z - Updated: 2026-08-19T09:05:58.000Z - Author: Wendell Hollifield ### Same plate, same salt. What changed is the gauge you've been reading it with. ### ![An oil painting of a workshop bench at the end of the day: a brass pressure gauge, reading glasses, and an open paper logbook under a caged lamp](https://storage.ghost.io/c/d6/f3/d6f3e6cb-1577-4ad4-879e-a14dacf1b87a/content/images/size/w2000/2026/08/imgonline-com-ua-Resize-gOA3xtKwAmyWtvO--1-.jpg) Hi, it's Lily. Everything written about falling is written for after. Grab bars. Loose rugs. Night lights. All of it sensible, all of it about the house, and all of it aimed at the moment itself. The moment costs, on average, $18,658 if it puts you in a hospital bed, and about $1,100 if it only gets you as far as the emergency room. That's the bill people picture. It's the small one. What sits behind it isn't treatment — it's help. Dressing, bathing, getting to the bathroom, getting up. That's custodial care, and [Medicare's own page](https://www.medicare.gov/coverage/long-term-care?ref=inside-systems.ghost.io) says it in one line: you pay 100% for most long-term care. Not a share of it. All of it. [Genworth](https://investor.genworth.com/news-events/press-releases/detail/1054/carescout-releases-2025-cost-of-care-survey-results?ref=inside-systems.ghost.io), the insurer that has priced this market every year for two decades, put the national median for a shared room in a nursing home at $315 a day. That is $114,975 for twelve months, and it is the median, which means half the country pays more. The federal analysts who study how any of this gets financed [put the odds](https://aspe.hhs.gov/reports/long-term-services-supports-older-americans-risks-financing-research-brief-0?ref=inside-systems.ghost.io) that someone turning sixty-five will need that kind of care at roughly seven in ten. In the same work they asked Americans aged forty to seventy which government program pays for most of it. One in four got it right. So the sum people insure against is the emergency room visit, and the sum that actually empties a household is the one most of them have never been given. That's what makes the part before the fall worth more attention than it gets. Last week's correspondent had something to say about that, so I gave him the floor. A letter from a reader At the plant, if a man tripped over a hose and caught himself on the rail, that got written up. Nobody was hurt. Nothing was damaged. It went in the book anyway, and it went in the same book as the real ones. We called those near misses, and my department's early warning ran on them for one reason: they're free. The injury costs everything and teaches the identical lesson. Now look at how it's handled for a person. The question you'll be asked, if you're asked at all, is whether you've fallen in the past year. That's a question about the failure, asked after the failure. In my line it would have gotten somebody a short conversation with me. Nobody asks how many times you caught yourself on the counter. I'd count those. Not the falls — the catches. The grab at the doorframe, the hand that went to the wall on the stairs, the step that landed wrong in the yard. A mark on the calendar, no detail, nothing dramatic. What you're after isn't the number. It's whether this month has more marks than last month. One mark is nothing. Four in March and eleven in June is a trend, and a trend is something you can act on while it's still free. I ran a department where we spent money to avoid a shutdown, and the argument I lost most often was that nothing was wrong yet. Nothing being wrong yet is the only time anything is cheap. Wendell Hollifield That's his instinct, and I went to see whether anyone had tested it. Somebody had, and the result is better than his argument. [Two hundred and sixty-six people](https://www.sciencedirect.com/science/article/abs/pii/S0003999309000860?ref=inside-systems.ghost.io), aged seventy to ninety, healthy and living at home, kept a daily log for a year. They wrote down two things: falls, and what the researchers called missteps — a trip, a slip, a loss of balance where you caught yourself and didn't go down. Several missteps logged 3.89× more likely to fall in the months that followed Among repeat fallers 3 to 1 near misses outnumbered actual falls And here's the part that decided this letter. Missteps didn't line up with the other things that get measured — not walking tests, not cognitive tests. They were carrying information those measures weren't picking up. Which means Wendell isn't describing a nice habit. He's describing the one signal that arrives early, arrives often, and isn't on the form. What actually moves the count If you wear bifocals or progressives, the bottom of the lens is ground for reading — about arm's length. When you glance at the ground two paces ahead, you're seeing it through the wrong prescription. The raised paving slab you'd otherwise catch is soft-edged. The lens is doing exactly what it was built for, in exactly the wrong place. [A trial in Australia](https://bmcgeriatr.biomedcentral.com/articles/10.1186/1471-2318-9-10?ref=inside-systems.ghost.io) gave 606 regular multifocal wearers, average age eighty, a second pair of plain distance glasses for walking outdoors. Across the whole group the effect was small and could have been chance. Among the people who actually went outside regularly, falls dropped by 40%. And among those who mostly stayed indoors, outside falls went up. I want to be careful, because that last line matters as much as the first. This isn't advice to abandon your bifocals. It's narrower and better: a second pair, distance only, for walking outdoors — and only if walking outdoors is something you do. For someone who rarely leaves the house, switching between two pairs appears to create the problem it was meant to solve. It's the kind of finding that doesn't survive being summarised, which is probably why it isn't in the pamphlet. Yours, Lily **P.S.** Both studies got their numbers the same way: people marked them down themselves. Two hundred and sixty-six kept a daily log for a year; six hundred kept a monthly calendar for thirteen months. Not records, not hospital data. A pencil and a calendar on the kitchen wall — and that turned out to be enough to see what the instruments missed. Sources [Srygley, Herman, Giladi & Hausdorff — Self-report of missteps in older adults: a valid proxy of fall risk?](https://www.sciencedirect.com/science/article/abs/pii/S0003999309000860?ref=inside-systems.ghost.io), *Archives of Physical Medicine and Rehabilitation*, 2009 · Elsevier [Reider et al. — Cost of U.S. emergency department and inpatient visits for fall injuries in older adults](https://pmc.ncbi.nlm.nih.gov/articles/PMC11829734/?ref=inside-systems.ghost.io) · PubMed Central Haran et al. — the VISIBLE trial · BMC Geriatrics [CareScout Cost of Care Survey, 2025](https://investor.genworth.com/news-events/press-releases/detail/1054/carescout-releases-2025-cost-of-care-survey-results?ref=inside-systems.ghost.io) · Genworth Financial Long-Term Services and Supports for Older Americans · ASPE, U.S. Department of Health and Human Services ###