Nine Short Children: Percentiles, Parents and a Doctor's Label
A student assists a health camp in a Sundarban block and measures 300 students aged 10–16. Height is taken against a wall tape with shoes on for the first 80 children; weight uses a bathroom scale on a concrete floor. Two volunteers measure on different days. Nine children fall below the 3rd percentile of height-for-age. One is a 13-year-old boy, 128 cm tall, whose mother is 148 cm and father 158 cm; he has no wasting and reports no illness. The visiting doctor wants all nine referred to a district hospital and asks the student to write that the community shows 'genetic adaptation to chronic undernutrition'. The student has age in whole years, no birth records, and no dietary data.
1. What measurement and design problems must be corrected before the nine cases are interpreted?
Read the worked answer
The screening has several avoidable errors, and each affects the percentile count.
- Shoes: height with shoes adds roughly 1–2 cm, enough to move a child across a cut-off; the first 80 are not comparable with the rest.
- Wall tape: a flexible tape can sag or be fixed at an angle; use a rigid rule or stadiometer checked against a known length.
- Scale: a bathroom scale on a concrete floor needs calibration with a known weight and should be zeroed each session.
- Two measurers: inter-observer error is likely; one child should be re-measured by both and the difference recorded.
- Age: whole years make z-scores imprecise near the cut-off; a child recorded as 13 may be 12.5 or 13.9. Use date of birth, or at least months.
- Cut-off logic: 3% of 300 is 9. Nine below the 3rd percentile is exactly what the definition predicts in a healthy reference population; it is not by itself evidence of a problem.
Re-measure a subsample barefoot, calibrate, compute WHO 2007 height-for-age and BMI-for-age z-scores, and check implausible values.
2. For the 13-year-old boy at 128 cm whose parents are also short, what further information should be collected, and what can be concluded now?
Read the worked answer
At 128 cm at age 13 he is far below the reference median and needs clinical assessment, but the cause is not settled by one measurement.
Collect:
- measured heights of both parents (not reported heights), and siblings' heights;
- birth weight and early growth history, and any long illness or steroid use;
- diet recall, puberty stage, and symptoms such as fatigue or chronic diarrhoea;
- growth velocity over 6–12 months, and bone age if available;
- whether his age is accurate, since a boy recorded as 13 may be younger.
A mid-parental height calculation for a boy - (148 + 158 + 13)/2 = 159.5 cm - shows that short parents predict a shorter child, but not usually as short as 128 cm at 13. Familial short stature is therefore plausible but not proven. He has no wasting, which makes severe acute undernutrition less likely, but chronic undernutrition cannot be excluded without diet and growth data.
Conclusion now: refer for clinical assessment, and write 'short stature, cause not determined' rather than a diagnosis.
3. How should the student respond to the phrase 'genetic adaptation to chronic undernutrition'?
Read the worked answer
The phrase is scientifically and ethically questionable in this setting.
- Adaptation in biology means a heritable trait shaped by selection that improves survival or reproduction in a given environment. Demonstrating it requires evidence of heritability and fitness effects, not a cross-sectional mean difference.
- Short stature in a population can be phenotypic plasticity: a response to infection, poor diet and repeated illness. It may be partly reversible when conditions improve, which is the opposite of a fixed genetic label.
- The label can do harm: calling deprivation 'adaptation' naturalises it and can reduce the urgency of nutrition and health interventions.
- The data cannot support it: there are no genetic data, no dietary data and no accurate ages.
A defensible statement is: 'Mean height-for-age in this sample is low; this is consistent with chronic undernutrition and/or normal familial variation, and the present data cannot separate the two.' Recommend referral and a proper growth study, not a genetic claim.