HumAn-1 randomised 400 children and young adults aged 7 to 25 with type 1 diabetes, at one site in Bangladesh and two in Tanzania, to insulin glargine or to continue usual care — isophane or premixed 70/30. Outcomes were measured on blinded continuous glucose monitors at six months, which removes the reporting bias that makes open-label insulin trials hard to read.
Neither co-primary outcome separated. Time in very low range (below 3.0 mmol/L) was 3.6% on glargine and 3.4% on usual care, an adjusted difference of 0.22% (97.5% CI -0.83 to 1.27, p=0.63). Time in target range was 40.5% versus 38.1%, an adjusted difference of 0.55% (-2.78 to 3.89, p=0.71). Serious adverse events were uncommon in both arms: five participants on glargine, thirteen on usual care.
Why it matters: the assumption that an analogue is simply better has driven a great deal of prescribing and procurement. In this population, over six months, on the two measures that matter most for safety and control, it was not. That is a finding about value rather than about pharmacology.
What to do about it: where cost constrains what a service can offer — which describes much of Indian practice outside the metros — this supports spending on monitoring, education and reliable supply before spending on analogue basal insulin for this age group. Note the limits honestly: six months, three sites, one analogue, and children and young adults rather than adults.
- Do not assume an analogue basal is required for adequate control in this age group
- Where funds are finite, weigh glargine against monitoring and education, not against nothing
- Reliable supply of any insulin beats intermittent supply of a better one
- Applies to children and young adults aged 7 to 25; do not extrapolate to older adults
- Six months is long enough for time in range, not long enough for complications
The statistics, in plain English
Both differences are close to zero with intervals that comfortably span it: 0.22% of time in very low range, plausibly anywhere from 0.8% better to 1.3% worse. A p value of 0.63 says a difference this small would arise by chance most of the time even if the drugs were identical. Note this is a trial that found no difference, which is not the same as a trial that proves the drugs equivalent — but the intervals are narrow enough that a clinically meaningful advantage for glargine would have been visible. The 97.5% interval rather than 95% is because two primary outcomes were tested, and the threshold was tightened to keep the overall error rate honest.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for diabetes & endocrinology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free