Two routine blood numbers taken on arrival tracked who was still alive three years after a broken hip

When an older person arrives at hospital with a broken hip, the emergency team draws blood almost straight away. Two of the numbers that come back are routine: albumin, a protein made by the liver that tends to fall with poor nutrition and inflammation, and haemoglobin, which falls with anaemia. A team at Beijing Jishuitan Hospital asked whether those two numbers, multiplied together, could say something about the years after surgery rather than just the days.

They went back to a prospective cohort of people aged 65 and over who had hip-fracture surgery between November 2018 and November 2019. Of 1,092 patients screened, 947 had both admission results and were included. The albumin–haemoglobin index (AHI) is simply albumin in grams per litre times haemoglobin in grams per litre, and the patients were split into four equal groups, from 3,925 or below at the bottom to above 5,357 at the top. Patients were followed at 30 days, 120 days, one year and three years.

The gradient was steady. After adjusting for age, sex, body mass index, diabetes, heart disease, cognitive impairment, mobility before the fall, fracture type, anaesthetic risk grade, time to surgery and more, each 1,000-point rise in the index went with a 43% lower risk of dying within three years (hazard ratio 0.57). Compared with the lowest quarter, the hazard ratios were 0.69, 0.45 and 0.28 for the next three. The curve showed no particular threshold, so the authors call the median value of 4,673 an exploratory cut-off, not a clinical one. Quality-of-life scores were similar across groups at admission, but the lowest group recovered least well at 30 days, 120 days, one year and three years.

The index also travelled with frailty. People in the lowest quarter were older on average (83 against 76 years in the top quarter), lighter, more often needed help walking before the fracture, more often had heart disease and cognitive impairment, waited longer for surgery and stayed in hospital longer. That is the main caution: a low score may mostly be a marker of someone who was already more fragile. This was a post hoc analysis from one hospital, the authors did not test whether the index beats albumin or haemoglobin alone, and later quality-of-life scores could only come from survivors.

Nothing here shows that raising either number improves survival. What it does suggest is practical: a cheap result that is already on the chart can point a team toward the patients who most need a nutrition assessment, a look at anaemia and closer follow-up. For families, it is a reasonable question to ask after a hip fracture — has anyone checked nutrition and blood count, and is there a plan for eating well during recovery?