Scans: Accuracy and Limitations

Scans in pregnancy: accurate, useful… and sometimes a rod that can be used to beat us into a decision we weren't ready to make.

Ultrasound is often considered one of the great “modern miracles” of pregnancy care: a window into the womb, a due date that feels official, and that first fuzzy flicker that makes it real. But it’s still a test. And every test has limits, margins of error, and knock-on effects.

This post breaks down what pregnancy scans can (and can’t) tell you, how accurate they are, what they’re good for, and why “more scans” isn’t automatically “better care”.

What scans are usually offered (and why)

In many maternity systems (including the UK and Ireland), routine scans commonly include:

A first-trimester scan (often called the dating scan)
Typically around 11–14 weeks, used to:

• confirm the pregnancy is in the uterus

• estimate gestational age (and therefore due date)

• check for multiples

• sometimes, as part of combined screening for chromosomal conditions

A mid-pregnancy anatomy/anomaly scan
Typically around 18–21 weeks, used to:

• look for specific structural anomalies

• check placental location

Some people are offered extra scans (for example: previous growth concerns, hypertension, diabetes, reduced movements, placenta concerns, twins, bleeding, IVF pregnancy, etc.). That’s targeted scanning, and it’s where ultrasound often shines.

Accuracy: what ultrasound is good at, and where it gets wobbly

Dating the pregnancy: ultrasound is strongest early on
First-trimester ultrasound using crown–rump length is widely considered the most accurate way to date a pregnancy

But “most accurate” doesn’t mean “exact”. There’s normal biological variation and measurement variation. Even a small measuring difference can shift dates and downstream decisions.

Why this matters:

• due date changes can affect when you’re labelled “overdue”

• it can influence timing of induction conversations

• it can affect interpretation of growth later (because growth charts depend on accurate gestational age).

Anatomy scans: very useful, but detection varies by condition and context
Mid-pregnancy anomaly scanning is designed to identify a defined list of conditions, not “everything that could ever happen”

Detection depends on:

• the specific condition (some are easier to see than others)

• the equipment and scan quality

• fetal position

• maternal body tissue and abdominal scarring

• gestational age at scan

• sonographer experience and local protocols.

So you can get:

• a reassuring scan and still have an issue that wasn’t detectable at that stage

• a “possible finding” that turns out to be nothing.

3. Growth scans and estimated fetal weight: helpful, but not a crystal ball
   Estimated fetal weight (EFW) is an estimate. A good one, often. But still an estimate.

Studies consistently show meaningful error ranges, commonly in the ballpark of roughly 6–15% (and sometimes more at the extremes, like very small or very large babies)

Why this matters:

• an overestimated weight can fuel unnecessary worry and intervention

• an underestimated weight can miss genuine growth restriction risk

• “big baby” predictions are particularly prone to false alarms, and can change how labour is managed even when the baby is perfectly average at birth.

Efficacy: do routine scans improve outcomes?

This is where it gets interesting: “useful information” and “improved clinical outcomes” are not always the same thing.

Early routine ultrasound (before 24 weeks)
Systematic reviews show early ultrasound can:

• improve gestational age assessment

• reduce missed diagnosis of twins

• reduce induction for “post-term” pregnancy (because dating is clearer)
  But it does not clearly reduce perinatal death in low-risk populations (Whitworth et al., 2015).

In plain language: early ultrasound is good at sorting dates and spotting twins, but it’s not a magic charm that guarantees better outcomes across the board.

Routine ultrasound after 24 weeks in low-risk pregnancies
Evidence reviews conclude that routinely scanning everyone later in pregnancy (without a specific clinical reason) does not show clear benefit for key outcomes like perinatal mortality, preterm birth, caesarean section, or induction rates in low-risk populations (Bricker et al., 2015).

The World Health Organization explicitly does not recommend routine ultrasound after 24 weeks for women who have already had an early scan, unless there’s a clinical indication (WHO, 2018).

So yes: extra scans can be very appropriate. But “routine for everyone, just because” is not strongly supported by evidence.

Why scans might not always be a good thing

Ultrasound itself is generally considered safe when used appropriately, but “safe” doesn’t mean “do unlimited scans for vibes”.

A) False positives and “soft markers” can create real harm
A scan can raise a concern that later resolves or is proven harmless. That’s a false positive, and it can be psychologically expensive.

Research shows that false-positive findings (including soft markers) can increase maternal anxiety and depressive symptoms, and can affect early parent–infant interaction.

Even when everything turns out fine, you still lived through the stress. And stress in pregnancy is not a cute hobby.

B) The cascade effect: one scan can trigger a chain of interventions
A borderline measurement can lead to:

• repeat scans

• consultant appointments

• extra monitoring

• earlier induction discussions

• a lower threshold for operative birth.

Sometimes that chain is life-saving. Sometimes it’s overmedicalisation dressed up as “just being safe”. The tricky part is that, in the moment, it can be hard to tell which story you’re in.

C) Dating and measurement errors can change labels, timelines, and decisions
Small differences in early measurement can alter gestational age assignment. That can cascade into:

• growth centiles later

• “overdue” definitions

• decisions about induction timing.

D) Keepsake/entertainment scans: the benefit is mostly emotional, the risk is unnecessary exposure
Medical bodies emphasise prudent use and the ALARA principle (as low as reasonably achievable), particularly limiting scan time and energy exposure.

The FDA discourages ultrasound for non-medical purposes, and notes keepsake images are reasonable only when they happen during a medically indicated exam and don’t require extra exposure.

In other words: you don’t need a 45-minute ultrasound photoshoot. Your baby is not a celebrity, and your uterus is not a film studio.

So how do you make good decisions about scans?

Try these three questions:

What clinical question are we trying to answer?
Examples:

• confirm dates

• check placenta location

• investigate bleeding

• assess growth because measurements or risk factors suggest an issue

• follow up reduced movements

• monitor a condition like hypertension or diabetes.

If the scan shows X, what will we do differently?

If the answer is “nothing”, it might be reassurance-seeking rather than clinical care. That doesn’t make it “wrong”, but it’s worth being honest about the purpose.

What are the downsides for me, personally?
Not just “medical risk”, but:

• anxiety

• time and cost

• potential for a false-positive spiral

• the emotional toll of uncertainty.

A final feminist note, because it matters

Pregnancy care should be something done with you, not to you. Scans are tools, not commandments. You’re allowed to ask what the scan is for, what it can and can’t tell you, and what choices you have if something uncertain pops up.

You don’t owe anyone passive compliance in exchange for care.

References

American College of Obstetricians and Gynecologists (ACOG) (2016) ‘Practice Bulletin No. 175: Ultrasound in Pregnancy’, Obstetrics & Gynecology.

American College of Obstetricians and Gynecologists (ACOG) (2017) ‘Methods for Estimating the Due Date (Committee Opinion)’.

American Institute of Ultrasound in Medicine (AIUM) (2020) ‘Prudent Use and Safety of Diagnostic Ultrasound in Pregnancy (Official Statement)’.

British Medical Ultrasound Society (BMUS) (2009) Guidelines for the Safe Use of Diagnostic Ultrasound Equipment.

Bricker, L. et al. (2015) ‘Routine ultrasound in late pregnancy (after 24 weeks’ gestation)’, Cochrane Database of Systematic Reviews, CD001451.

Food and Drug Administration (FDA) (2024) ‘Ultrasound Imaging’.

Gadsbøll, K. et al. (2021) ‘Crown-rump length measurement error: impact on pregnancy dating and growth assessment’, Ultrasound in Obstetrics & Gynecology.

Milner, J. and Arezina, J. (2018) ‘The accuracy of ultrasound estimation of fetal weight in comparison to birth weight: A systematic review’, Ultrasound, 26(1), pp. 32–41.

National Health Service (NHS) (2020) ‘Ultrasound scans in pregnancy’.

National Institute for Health and Care Excellence (NICE) (2021) Antenatal care (NG201): Recommendations.

UK Government (2021) ‘Fetal anomaly screening programme handbook: 20-week screening scan’.

Vaughn-Valencia, M.M. et al. (2025) ‘Fetal weight extrapolation following a third-trimester ultrasound’, American Journal of Perinatology.

Viaux-Savelon, S. et al. (2012) ‘Prenatal ultrasound screening: false positive soft markers may alter maternal representations and increase anxiety’, PLOS ONE, 7(1), e30935.

Whitworth, M. et al. (2015) ‘Ultrasound for fetal assessment in early pregnancy’, Cochrane Database of Systematic Reviews, CD007058.

World Health Organization (WHO) (2018) WHO Recommendations on Antenatal Care for a Positive Pregnancy Experience. Geneva: WHO.