Weight Loss and Malnutrition in Care Homes

Why weight loss in a care home is usually found late

Unintentional weight loss is weight a resident has lost without trying to. It is not the thinness that comes with age, and it is not a diet that worked. In a care home the food arrives on a tray three times a day, so it is easy to assume that eating is handled. That assumption is where the problem starts.

Intake depends on things that are not on the menu. It depends on whether the resident can reach the plate, whether the food is a texture they can chew, whether their mouth hurts, whether a medicine has dried their mouth or flattened their appetite, and whether anyone has time to sit down beside them. The Department of Health's Elderly Health Service tells carers in residential care homes to screen and monitor nutritional status routinely, so that a resident with an eating problem is found before the condition deteriorates.

The scale is the last place the change appears. A 5% loss in a 50 kg resident is 2.5 kg, roughly 80 g a day, which no photograph and no handover sentence will show. By the time a colleague says the resident looks thinner, the loss is usually well past the first threshold.

This article works forward from that point: the thresholds, the observation at the table, the food that can carry more nutrition, the records, and who is told. Swallowing difficulty has its own assessment and its own article, mealtime safety and dysphagia, and nothing here replaces it.

The screening thresholds, and where they come from

Hong Kong has two written thresholds and they are not identical. The Elderly Health Service nutrition screening guidance asks homes to weigh residents monthly and to suspect malnutrition when weight falls by more than 5% in one month, 7.5% in three months or 10% in six months. It also gives the body mass index bands used locally, where below 18.5 is underweight and 18.5 to 22.9 is normal.

The licensing standard sets the enforceable version. The Code of Practice for Residential Care Homes (Elderly Persons), at paragraph 14.3.2, requires regular weighing, records of it, and a response: where weight loss is more than 5% within a month or more than 10% within six months, the home should seek advice from healthcare professionals and arrange medical treatment for the resident.

The two do not conflict, because one is a subset of the other. A home that screens to the Department of Health indicators will always meet the Code's trigger. A home that screens only to the Code will miss the resident who is sliding slowly, since 7.5% over three months can pass without either the one-month or the six-month figure being crossed on the day of measurement.

Two further indicators catch the resident whose weight has not yet moved. The same guidance treats a resident as at risk when a quarter or more of the food is left at all three main meals for seven consecutive days, and when daily fluid intake stays below six cups, counting water, tea, soup, juice and milk.

• Weigh monthly, on the same scale, at the same time of day, in similar clothing, and write the result down the same shift.
• Score the result rather than filing it: body mass index, unplanned loss, and whether an acute illness has cut intake, which is the three-part shape used by scored tools such as MUST.
• Repeat the screen at admission, every month, and after any hospital stay or illness, because those are the weeks when intake falls and nobody is watching the plate.

What observation at the table adds

The Code requires the home to arrange for staff to monitor eating conditions at meal times and snack times, to give sufficient time so that nobody eats in a hurry, and to provide appropriate utensils and prompts. That duty is written in paragraph 14.6.1, and it is the part of nutrition care that no monthly number can carry.

Medicines belong on the observation list. Some reduce appetite, some dry the mouth, some cause nausea, and the interval between a sedative and a meal can decide whether the meal is finished. Medication management covers the review that should sit behind that observation, including who may change a dose and who may not. Low mood, pain, constipation and a recent bereavement reduce intake too, and none of them appears on a menu.

A weight is one figure a month. A plate is one figure a meal. The practical test is whether the note written at the table names a possible cause: a note reading 'half the congee, stopped after ten minutes, held the left side of the mouth' gives the nurse something to act on, while a note reading 'poor appetite' gives the next shift nothing. Nothing in the room measures what was eaten either. Movement sensing of the kind Steadicore builds covers where a person is and how they move, not what is left on the plate, which is why the intake chart is still written by a person standing at the table.

What to watch, in the order it usually appears:

• How much came back: record quarters or spoons, not the phrase poor appetite, because the phrase cannot be compared with last Tuesday.
• Pace and fatigue: a resident who eats well for five minutes and then stops is reporting stamina, not preference.
• Coughing, a wet voice or food held in the cheek: stop the meal and refer for a swallowing assessment, and the dysphagia guide sets out what to do at the table in the meantime.
• Mouth comfort: a loose denture, a sore gum, oral thrush or a broken tooth ends a meal faster than any other cause, and daily oral care is where it is found.
• Hands and utensils: a resident who has stopped feeding themselves may start again with a different spoon or a plate guard, which is an occupational therapy question rather than a behaviour problem.
• Food brought in by family: the Code asks relatives to check with the nurse on duty first, because a favourite dish from home can be the wrong texture for the resident receiving it.

Fortification: more nutrition in food the resident already eats

Fortification means raising the energy and protein in food a resident already accepts, instead of asking them to eat more of it. It works because the volume stays the same. A bowl of congee with milk powder stirred through it is still a bowl of congee, and it can carry more energy and protein than the plain version without a second bowl being refused.

The Department of Health's leaflet on common eating problems makes the same move from the other direction. It recommends small frequent meals, nutritious snacks between meals, natural seasonings such as ginger and spring onion instead of more salt, and protein-rich additions such as fish with congee.

Two cautions belong in writing. Adding liquid or fat can change the texture of a texture-modified diet, so a change for a resident on minced or pureed food is agreed with the speech therapist who set the level, not improvised in the kitchen. And a supplement is a clinical decision: the Department of Health asks carers to consult a doctor or dietitian before any nutrition supplement is considered, because an unprescribed one gives an unbalanced intake. No product, including ours, prevents malnutrition or guarantees weight gain.

The four records an inspection will ask for

When a resident loses weight and a family member asks what the home did, the answer is four records or it is an opinion. Each one answers a different question, and a home that keeps three of them cannot reconstruct the fourth.

The menu is a record too. The Code requires a menu covering at least two weeks, displayed in a prominent place, with special diet menus for staff and immediate updates when a resident's dietary needs change. A texture level that exists in the care plan but not on the kitchen sheet is a plan that will be broken at the next meal.

The percentage is the finding, and a bare weight is not. A figure with no date, no scale and no note of clothing or time of day cannot be compared with the last one, and the comparison is the whole point. The inspection preparation guide sets out the order in which those records are usually asked for.

Escalation: who is told, and in what order

The chain in a care home is short and should be written down: the care worker who notices, the nurse in charge who checks the record and weighs again, the home manager who owns the response, and then the clinician. Which clinician depends on the question. The cause is a doctor's question, the texture is a speech therapist's, the feeding equipment is an occupational therapist's, and the food itself is a dietitian's.

At the threshold the Code is specific. The home should seek advice from healthcare professionals and arrange medical treatment, so the referral note has to carry the date, the weight, the percentage change over one, three and six months, what was already tried at the table, and what advice came back. A referral that says the resident is not eating well asks the clinician to start from the beginning.

The last step is the one most often skipped. A home should fix its own re-weigh date when it makes the referral, and hold to it. If the trend continues at that date, the case goes back up the chain rather than being weighed again indefinitely, because repeated measurement is not a plan.

Families carry half of this work, and they need the numbers rather than a reassurance. Easikin is the SI（Superintelligence）· AI（artificial intelligence）caregiving assistant for adult children coordinating a parent's care, and the record a home keeps is what makes that conversation specific. What no product does is prevent malnutrition: the screening, the fortified meal and the written referral do that, and they are all done by people.