Predicting a Child’s Adult Height: Genetics, Growth Charts, and What Online Tools Cannot Tell You

Why This Question Attracts Bad Tools

"How tall will my child be?" is a question parents ask early and often. Free websites are happy to answer it. Most plug a small number of inputs into a single 1980s-era formula, return a confident-looking figure to one decimal place, and present the result as if it were a measurement. The number is not a measurement. It is an estimate around which the genuine 95 percent confidence interval is roughly seventeen centimetres wide. That is the difference between a short adult and a tall one.

This guide explains what the underlying mid-parental-height formula actually is, what it is and is not capable of, how paediatricians evaluate growth in real life, when concerns about a child's height genuinely warrant medical input, and which lifestyle claims about "increasing height" do not survive contact with the evidence. The goal is to leave you better equipped to interpret an online estimate and to know when the right next step is a conversation with a paediatrician, not another web form.

The Mid-Parental-Height Formula

Most child-height predictors are wrappers around a calculation usually attributed to British paediatricians J. M. Tanner and colleagues in the 1970s and refined in subsequent paediatric growth literature. The standard form is:

Boys: ((father's height + mother's height) / 2) + 6.5 cm
Girls: ((father's height + mother's height) / 2) - 6.5 cm

Some references use roughly +13 cm to the father's height plus the mother's height divided by 2 for boys, and minus 13 cm for girls; algebraically these are the same calculation, just rearranged. The number 13 (or 6.5) is the average sex-related difference in adult height in many studied populations.

The formula is grounded in the heritability of stature: large twin and family studies, summarized in reviews such as Visscher and colleagues (Nature Reviews Genetics, 2008), estimate that 60 to 80 percent of variation in adult height in well-nourished populations is heritable. Population-level genetic studies have since identified hundreds of common variants individually contributing tiny fractions of that variance, with environmental and nutritional factors accounting for most of the remainder.

Heritability of 60 to 80 percent at the population level does not mean a child's adult height can be predicted to the centimetre from two parental heights. It means that if you measured everyone in a population, knowing parental heights would explain a meaningful fraction of the variance — not that any individual prediction will be accurate.

Why a Single Number Misleads: the Prediction Interval Problem

The polite footnote on every honest version of the mid-parental-height formula is that the 95 percent confidence interval around the central estimate is approximately plus or minus 8.5 centimetres. That is a range of about 17 centimetres total — roughly seven inches.

To make that concrete: if a tool tells you your son's predicted adult height is 178 cm, the formula's own honest output is "somewhere between roughly 170 and 187 cm with about 95 percent confidence". The range covers a man who is just below average for many populations and a man who is well above average. Knowing he is "probably somewhere in that range" does not tell you anything you would not have guessed by looking at the parents.

Worse, free tools rarely show the interval. They display the central estimate to one decimal point, often surrounded by satisfying graphics, and let parents read it as a forecast rather than a centred guess. Padmadas and colleagues, in a 2010 review of paediatric growth prediction methods in the European Journal of Pediatrics, were explicit that the formula is appropriate as a rough population reference point and inappropriate as an individual forecast.

An additional complication is that the formula assumes the parental heights themselves are reliable. Self-reported height is biased upward in adults (people round up), particularly in men. A predictor calculated from two slightly inflated parental heights inflates the child's prediction too.

How Paediatricians Actually Assess Growth

A paediatrician assessing growth does not consult a one-shot formula. The standard tools are growth charts, percentile tracking over time, and selective use of bone-age imaging when concerns arise.

Growth Charts

In the United States, the Centers for Disease Control and Prevention publishes growth charts for children aged 2 to 20 (height-for-age, weight-for-age, weight-for-height, BMI-for-age) at cdc.gov/growthcharts. For children under two, the CDC recommends using the World Health Organization international growth standards, available at who.int/tools/child-growth-standards. The WHO charts are based on a multinational study of healthy, breastfed infants raised in optimal conditions and represent how children "should" grow rather than how they "do" grow on average in any one country.

A growth chart does not predict adult height directly. It plots a child's measurement against the percentile distribution for their age and sex. A child at the 25th percentile for height-for-age is taller than 25 percent of same-aged peers and shorter than 75 percent.

Percentile Consistency Over Time

The clinically meaningful pattern is not the percentile itself but its consistency over time. Children typically settle onto a growth percentile by around age two and tend to stay near it through the rest of childhood. A child consistently at the 10th percentile is short for age but is growing predictably; a child who was at the 50th percentile and has dropped to the 10th percentile over two years is showing a change that warrants investigation regardless of where the absolute number sits.

Conversely, a child consistently at the 95th percentile may simply be tall; a sudden jump from the 50th to the 95th could reflect early puberty or, rarely, a hormonal disorder. The pattern over time, not the snapshot, is what a paediatrician reads.

Bone-Age X-Ray

When a paediatrician has a specific concern (a child whose growth has fallen off their established percentile, a child substantially shorter than predicted by the mid-parental height with no obvious explanation, a child with signs of premature or delayed puberty), they may order a bone-age radiograph, typically of the left hand and wrist. The radiograph is read against a reference atlas (the Greulich-Pyle atlas is still widely used; the Tanner-Whitehouse method is an alternative).

Bone age tells the clinician how much growth potential remains. A child whose bone age is younger than their chronological age has more growth ahead and may catch up; a child whose bone age is advanced has less growth potential left. Combined with current height and growth velocity, bone age supports more accurate (though still imperfect) predictions of adult height through methods such as the Bayley-Pinneau or Roche-Wainer-Thissen calculations. These methods are still imperfect and are deployed in the context of a clinical evaluation, not from a web form.

Conditions That Affect Height

Most children who are short for their age are simply on the short end of the normal distribution, often with one or two short parents. A minority have an underlying cause that is worth recognising. The conditions below are described to illustrate why a paediatrician's evaluation matters when a growth pattern is unusual; the right person to assess any of them is a clinician.

• Growth hormone deficiency. Insufficient pituitary production of growth hormone produces predictable growth failure that can be confirmed with stimulation testing. Treatment with recombinant growth hormone, when indicated, is well-studied. Source: NICHD and paediatric endocrinology guidelines.
• Turner syndrome. A chromosomal condition (45,X) affecting roughly 1 in 2,500 girls. Short stature is one of the most consistent features and may be the only obvious sign in some cases; karyotyping is diagnostic. Growth hormone is part of standard management. Reference: the Pediatric Endocrine Society and NIH resources.
• Constitutional delay of growth and puberty. A pattern in which a child grows slowly relative to peers and enters puberty later than average, but ultimately reaches a normal adult height. Often runs in families ("late bloomers"). Diagnosis is clinical and supported by a delayed bone age relative to chronological age.
• Familial short stature. Children whose adult height tracks their parents' relatively short stature, with normal growth velocity and bone age that matches chronological age. Not a disease — a normal variant.
• Idiopathic short stature. A diagnosis of exclusion when growth is below the third percentile and other causes have been ruled out. Sometimes treated with growth hormone in selected cases, with modest height gain that may not justify the cost or burden for many families.
• Chronic illness, malnutrition, malabsorption (coeliac disease, inflammatory bowel disease), and chronic kidney disease. All can present as growth failure and are diagnosed and treated through paediatric care.

The pattern that distinguishes a normal short child from one warranting evaluation is usually a change in growth velocity — falling off the percentile they had previously held. A paediatrician asks for prior measurements, plots them, looks at the parental heights, considers the child's general health and pubertal stage, and decides whether referral to a paediatric endocrinologist is warranted. None of that fits in a web form.

Lifestyle Myths About Height

The internet is rich with claims that specific activities or supplements can "make your child taller". The peer-reviewed evidence base for most of these claims is thin to non-existent.

Sport (basketball, swimming, jumping exercises) does not change adult height. Tall children are over-represented in basketball because basketball selects for height, not the reverse. There is no robust longitudinal evidence that any non-medical activity increases the genetic height ceiling. Adequate physical activity supports normal growth and bone health; it does not make a child taller than they would otherwise be.

"Growth-promoting" supplements (height-boost gummies, herbal blends, vitamin combinations sold for children) are not regulated to clinical standards in the United States and are essentially never supported by adequately controlled trials. The US Food and Drug Administration's authority over dietary supplements is structurally weaker than its authority over drugs (see fda.gov/food/dietary-supplements): supplements do not need to demonstrate efficacy before sale. A meta-analysis would find no consistent effect from these products on adult height.

Adequate nutrition, sleep, and the absence of chronic disease genuinely matter — not as height-boosters but as conditions that allow the genetic potential to be reached. A child with severe untreated coeliac disease, sleep apnoea, or chronic undernutrition will not reach their predicted height. The intervention there is treating the underlying condition, not buying a supplement.

Growth hormone treatment is a real medical intervention with real indications. It is also expensive, given by daily injection over years, has potential side effects, and produces clinically modest height gains in idiopathic short stature. It is appropriate when prescribed by a paediatric endocrinologist for diagnosed deficiency, Turner syndrome, and a small set of other conditions. It is not appropriate as a height-boost for healthy short children, and it is not something a web tool can recommend.

How to Use an Online Height Predictor Honestly

If you do use the mid-parental-height formula, treat it as an order-of-magnitude orientation, not a forecast. A useful framing:

• Compute the central estimate from the formula above.
• Add and subtract 8.5 cm to obtain the rough 95 percent prediction interval.
• Read the result as "my child will probably be somewhere between X and Y cm as an adult, more often near the middle than the edges".
• Recognise that this prediction, even read honestly, contains less information than two paediatric measurements taken a year apart, plotted on a CDC or WHO chart.

If you are concerned about your child's height — substantially shorter or taller than peers, a noticeable change in growth pattern, signs of early or delayed puberty — the next step is an appointment with their paediatrician, not another website. The paediatrician has access to the prior measurements, the family history, and the physical exam findings that determine whether a referral is warranted.

Reliable Sources

• CDC growth charts: cdc.gov/growthcharts — official US paediatric growth charts and the methodology behind them.
• WHO child growth standards: who.int/tools/child-growth-standards — international standards for children under five.
• HealthyChildren.org: healthychildren.org — American Academy of Pediatrics parent-facing resources, including pages on growth concerns and when to ask the paediatrician.
• NHS — Childhood growth and physical development: nhs.uk — UK guidance on growth concerns.
• Pediatric Endocrine Society: pedsendo.org — professional society resources on growth disorders, including patient-facing materials.

The Bottom Line

The mid-parental-height formula is real arithmetic with a real prediction interval. The interval is too wide to be actionable for an individual child. Web tools that hide the interval and present a single number are presenting a confidence the formula does not have.

What actually predicts a child's adult height with reasonable accuracy is a series of paediatric measurements over time, plotted against age- and sex-specific percentiles, interpreted by a clinician who can decide when bone-age imaging or referral to a specialist is warranted. For broader body-composition orientation see our BMI calculator; for the question this article addresses, the right place to look is your child's paediatrician and the CDC or WHO growth charts they use.