Drs Trevor Brown (pictured) and Carina Venter draw together recommendations for managing cow’s milk allergy in children from NICE CG116 and the Clinical Knowledge Summary, and the MAP Milk Allergy Guideline

brown trevor

Read this article to learn more about:

  • taking an allergy-focused clinical history
  • who should be referred for early specialist care
  • how to manage the remaining children with cow's milk allergy in primary care.

Key points

GP commissioning messages


This article has been superseded by a more recent version. Please read:

Cow's milk allergy: MAP publishes international update

The UK has recently seen a significant rise in the number of children suspected of having a food allergy. This is defined as an adverse health effect arising from a specific immune response that occurs reproducibly on exposure to a given food.1 The immune response may then be clinically subdivided into the following:

  • immediate-onset IgE antibody-mediated response, where the adverse effects usually appear within minutes (can be up to 2 hours) after ingestion
  • delayed-onset non-IgE antibody-mediated response, where the effects usually develop more than 2 hours, or even days, after ingestion.2

Milk allergy in primary care: current UK guidance2,3,4

A UK birth cohort study has shown that up to 3% of 1–3-year-olds can be confirmed as having cow's milk allergy (CMA), making it the most common food allergy in the early years of life;5 milk allergy most commonly presents in the early weeks or months of life.2 In a study (published in 2010) of 1000 infants with a diagnosis of milk allergy, randomly chosen from a UK healthcare database, 86% were first diagnosed in primary care and most were then managed in primary care.6 This study highlighted evidence of under-recognition, incorrect diagnosis, significant delay in diagnosis, and subsequent suboptimal management of the infants, especially in choosing the most appropriate initial hypoallergenic formula to prescribe.

As a response, NICE published the 2011 Clinical Guideline 116 (CG116) on the Diagnosis and assessment of food allergy in children and young people in primary care and community settings.3 The NICE Clinical Knowledge Summary (CKS) on Cow's milk protein allergy in children2 (aged 5 years and younger) was subsequently published in 2014, drawing largely on NICE CG116 with regard to the initial diagnosis and assessment of any food allergy in primary care and indications for referral; the CKS also references other guidelines, particularly the UK MAP Milk Allergy Guideline published in 20134,7 to address the ongoing care of those children with milk allergy who should be managed in primary care.

NICE CG116 recommends that the initial recognition and assessment in primary care of a suspected allergy to any food in a child should always be carried out in a certain order (see Box 1, below).3

Box 1: Initial recognition and assessment of suspected food allergy in children (adapted from NICE CG116)3

Initial recognition and assessment of suspected food allergy in children (adapted from NICE CG116)
  • Adapted from NICE Clinical Guideline 116

The stated goals of the CKS are to support primary care healthcare professionals to:2

  1. Recognise the possible features of CMA in children
  2. Assess those children with suspected CMA
  3. Refer appropriate children for early specialist diagnosis and management
  4. Manage those remaining children with suspected CMA in primary care.

Discussion around these four goals forms the basis of the remaining article, which summarises the key evidence-based recommendations relevant to primary care.

1) Recognising the possible features of CMA in children

The primary care healthcare professional is advised to:3

  • recognise that the three main organ systems most commonly affected by food allergy are the skin, gut, and respiratory system
  • pay particular attention to persistent signs or symptoms that involve more than one of these systems and/or where the initial treatment of any of the signs or symptoms has not proved to be helpful
  • consider the early clinical distinction between suspected IgE-mediated and non-IgE-mediated food allergy—see Table 1, below, which lists the typical IgE-mediated and non-IgE-mediated signs and symptoms related to CMA in infancy. (NB Some signs and symptoms appear in both groups; the list is not exhaustive)
  • realise that any family history of atopic disease (e.g. eczema, allergic rhinitis, asthma, or food allergy) in parents or siblings makes either IgEmediated or non-IgE-mediated food allergy more likely to be confirmed— although there will still be children presenting with CMA who will not have a family history of atopic disease.
Table 1: IgE-mediated and non-IgE-mediated signs and symptoms (adapted from NICE CG116 to relate to CMA in infancy)3
 Signs and symptoms
Response rateUsually minutes after food ingestion (may be up to 2 hours)Usually more than 2 hours or even days after food ingestion
  • Pruritus
  • Erythema
  • Acute urticaria (localised or generalised)
  • Acute angioedema (most commonly of the lips, face, or around the eyes)
  • Acute flaring of atopic eczema.
  • Pruritus
  • Erythema
  • Significant atopic eczema.
  • Angioedema of the lips, tongue, and palate
  • Colicky abdominal pain
  • Vomiting
  • Diarrhoea.
  • Infantile colic
  • Vomiting—‘reflux’
  • Gastro-oesophageal reflux disease (GORD)
  • Food refusal or aversion
  • Loose or frequent stools
  • Perianal redness
  • Constipation—especially soft stools with excessive straining
  • Abdominal discomfort, painful flatus
  • Blood and/or mucus in stools
  • Faltering growth plus one or more of the above gastrointestinal presentations (with or without significant atopic eczema).
Respiratory (usually in combination with one or more of the above signs or symptoms)
  • Upper respiratory—e.g. sneezing, rhinorrhoea, nasal congestion (with or without conjunctivitis)
  • More serious signs—Anaphylaxis e.g. stridor, cough, wheezing, shortness of breath, with or without listlessness, pallor.
  • Upper respiratory—e.g. sneezing, rhinorrhoea, nasal congestion (with or without conjunctivitis)
  • Lower respiratory—e.g. cough, wheezing, or shortness of breath.
  • Adapted from NICE Clinical Guideline 116

2) Assessing those children with suspected CMA

When a pattern of signs and symptoms is recognised that could be due to CMA (and this suspicion may or may not be further strengthened by a positive family history of atopy), then an allergy-focused clinical history is the cornerstone of progressing the diagnosis. The required questions that the first-contact clinician should ask, shown in Box 2, below, have been adapted from NICE CG116 to specifically relate to an infant with suspected CMA.

Box 2: Allergy-focused clinical history—'The cornerstone of the diagnosis' (adapted from NICE CG116 to relate to CMA in infancy)3

Ask about:

  • family history of atopic disease (e.g. eczema, asthma, allergic rhinitis, or food allergy) in parents or siblings:
    • a positive history along with signs or symptoms of suspected CMA makes the diagnosis more likely; this applies to both IgE-mediated and non-IgE-mediated milk allergy8
  • the source of an infant's ingested CMP and how much is being or was ingested:
    • exclusive breastfeeding—when some CMP from maternal diet comes through in the breast milk (low risk of allergy—prevalence 0.5%)2
    • mixed feeding—when CMP is given to the breast-fed infant (e.g. top-up formulas, in solids)
    • formula fed—the commonest presentation
  • presenting signs and symptoms (see Table 1,below), to include:
    • if more than one sign or symptom, the sequence of clinical presentation of each one
    • age of first onset
    • speed of onset following ingestion:
      • IgE—usually within minutes, but can be up to 2 hours
      • non-IgE—usually after ≥2 hours or even days:
        • » signs or symptoms often persisting and increasing with continuing ingestion
    • duration, severity, and frequency
    • reproducibility on repeated exposure
    • details of any changes in feeding or diet and any apparent response to such changes
    • details of any other previous management, including medication, for the presenting signs or symptoms and any apparent response to this.
  • CMA=cow's milk allergy; CMP=cow's milk protein
  • Adapted from NICE Clinical Guideline 116

Presuming there are no acute signs needing immediate attention, a physical examination should take place:

  • look for persisting signs such as faltering growth and malnutrition, and signs indicating allergy related comorbidities (e.g. atopic eczema).

Further investigations

Suspected IgE-mediated CMA

Initial investigation

Offer skin-prick tests or blood tests for IgE antibodies specific to cow's milk protein (CMP) and, if indicated, also to likely co-existing allergenic proteins (e.g. soya).

Stated competencies
  • Tests should only be undertaken by a healthcare professional with the appropriate competence to perform them; also, skin-prick tests should only be performed in facilities equipped to deal with an anaphylactic reaction3
  • Test results must be able to be interpreted in the context of the clinical history (e.g. a positive test shows sensitisation, but does not necessarily confirm clinical allergy; the size of the positive test is not related to the clinical severity of the allergy).1

Do not carry out a cow's milk challenge if there is a need to confirm the diagnosis of suspected immediate onset IgE CMA in primary care or community settings.3

Suspected non-IgE-mediated CMA

Initial investigation2

Try eliminating all sources of CMP for a 2- to 6-week trial from the infant's diet or mother's diet if the infant is exclusively breast fed. Then reintroduce CMP at home to either confirm or exclude diagnosis of milk allergy.

Dietary advice2
  • Exclusively breast-fed infants—start the mother on a strict CMP-free diet and prescribe for her daily supplements of 1000 mg of calcium and 10 μg of vitamin D
  • Formula-fed or mixed-fed infants (i.e. breast fed and top-up formula)—prescribe an appropriate replacement hypoallergenic formula: either an extensively hydrolysed formula (eHF) or an amino acid-based formula (AAF)
  • Weaned infants—start infant on an overall CMP-free diet, including an appropriate hypoallergenic formula.
Support of a dietitian2
  • Exclusively breast fed—seek early dietitian support for advising on a nutritionally complete maternal elimination diet
  • Formula fed, mixed fed, and weaned—seek early ongoing dietitian support on confirmation of diagnosis.

3) Referral of appropriate children for early specialist diagnosis and management

On the presentation of suspected food allergy, NICE emphasises the importance of differentiating between immediateonset IgE-mediated and delayed-onset non-IgE-mediated food allergy.2,3 A European-wide CMA guideline8 usefully proposes further dividing CMA into 'mild to moderate' and 'severe'. Relating this to milk allergy in infants and using the limited clinical wording employed by NICE, the referral recommendations are set out in Table 2, below. Whether referral is made or not at this stage, if a change of formula is indicated, it is important to know which category of hypoallergenic formula should be prescribed,2 and this is also highlighted in Table 2. The different categories are as follows:

eHFs—extensively hydrolysed formulas: these formulas are tolerated by 90% of infants and children with milk allergy.2 While they are based on cow's milk protein (CMP; casein or whey fractions), they are extensively broken down into smaller peptides that are less well recognised by the immune system.

AAFs—amino acid-based formulas: based on amino acids, the building blocks that make up proteins, these contain no CMP as such and are indicated for more clinically severe cases. NB Examples of eHFs and AAFs are listed in the NICE CKS on CMA in children.2

With regard to choosing one preferred individual eHF or AAF formula over another, there are very few 'head-to-head' randomised controlled trials comparing different formulas within either the eHF or the AAF categories, and the clinical profiles of patients who improved or did not improve are often very poorly described.4

Table 2: NICE referral and initial formula recommendations (adapted from NICE CG116 and CKS CMA)2,3
Clinical category of CMANICE referral recommendationsInitial formula
Non-IgEmediated CMA*Mild to moderate—commonest presentation of milk allergy in infancyCan be managed in primary care with ongoing input of dietitianeHF
  • faltering growth with one or more gut symptoms or
  • one or more severe delayed reactions or
  • significant atopic eczema with suspected food allergy.
Prompt specialist referral of allAAF
IgE-mediated CMA*Mild to moderate:Guidance on specialist referral is less clear (see below)eHF
  • one or more acute systemic reactions.
Emergency care, if indicated, then referralAAF
Other scenariosClinical suspicion of multiple food allergiesPrompt specialist referralAAF
Persisting parental suspicion of food allergy (especially in children where symptoms are difficult or perplexing) despite a lack of supporting historyFurther opinion at specialist level could be helpful to both theN/A
  • CKS=clinical knowledge summary; CMA=cow's milk allergy; eHF=extensively hydrolysed formula; AAF=amino acid-based formula Adapted from NICE CG116 and CKS.
  • * See Table 1, above for list of possible signs or symptoms
  • Specialist refers to a paediatrician with the skills, experience, and competencies necessary to deal with the particular concern that has been identified by the referring healthcare professional. In practice this may be a general paediatrician, a general paediatrician with an interest in allergy, or possibly a consultant paediatric allergist.
  • NICE CG116 recommended specialist referral when tests are negative but there is strong suspicion of food allergy; however the NICE-stated competency requirements imply that referral was likely to be necessary in most cases of suspected IgE-mediated allergy until competency levels in UK primary care are reached. NICE CKS on CMA in children has brought more up-to-date direction: 'CKS recognises that the expertise to choose, perform, and interpret tests for suspected IgE-mediated CMA may not be readily available in primary care; therefore the diagnosis and management is more.'2

4) Management of those remaining children with suspected CMA in primary care

Mild to moderate non-IgE-mediated CMA (formerly called cow's milk protein intolerance)

NICE CG116 only addressed the initial assessment and recognition of food allergy in children in primary care and community settings. Its commissioned scope did not include the ongoing management of any child with confirmed food allergy.

In 2013, a subgroup of the clinicians on the NICE CG116 guideline development group (namely T Brown, C Venter, AT Fox, and J Walsh), with the co-opted help of a paediatric gastroenterologist (N Shah), decided that a further guideline was needed to focus specifically on recognising, confirming, and then managing infants with suspected mild to moderate non- IgE CMA.4 The guideline would also provide guidance on the referral for more specialist care of those infants presenting with more severe signs and symptoms.4 This guideline has been piloted and audited in a UK NHS Region (Northern Ireland).10 It has evolved as a consequence, and is now called the MAP milk allergy guideline for primary and secondary care use.4 Subsequently, NICE CKS has been published, giving detailed advice over 28 pages of text on the management of these young children with milk allergy in primary care; it frequently references MAP, which focuses on two key algorithms as discussed below.4

The MAP algorithms

The NICE CKS outlines three key aspects related to the management of mild to moderate non-IgE-mediated CMA in primary care, namely:2

  1. Clinical history
  2. Initiation of a trial of a CMP-free diet:
    • with an early home challenge to confirm or exclude diagnosis of CMA
  3. On confirmation of CMA, a CMP-free diet should be continued with later testing for naturally acquired tolerance.

Clinical history—the cornerstone of the diagnosis

The first MAP algorithm (see Figure 1, below)4 outlines all the possible clinical presentations and the need for an allergy-focused clinical history, and advises on the first-choice category of hypoallergenic formula for each of the presentations. It also highlights when specialist referral should be considered.

Initial trial of CMP-free diet with early home challenge

The management of mild to moderate non-IgE-mediated CMA is the focus of the second MAP algorithm (see Figure 2, below). Following an initial trial of a strict CMP-free diet, the infant's signs and symptoms will usually settle comfortably within 2 to 4 weeks if they are due to mild to moderate non-IgE CMA. A planned home challenge should then be carried out to confirm the diagnosis. The MAP protocol for this challenge can be accessed at www.ctajournal.com/content/3/1/23 (see Additional File 1, Part 1).

Continuation of CMP-free diet with later testing for naturally acquired tolerance

Once CMA is confirmed, strict avoidance of all CMP with dietetic support should be continued until at least 9–12 months of age and for at least 6 months in all. A test for tolerance can then be carried out, using a milk ladder—such as the MAP milk ladder4— for home reintroduction (see Figure 2, below). The ladder and additional resources can be accessed at: www.ctajournal.com/content/3/1/23 (see Additional File 1, Part 2).

Most infants with mild to moderate non-IgE milk allergy can be expected at this stage to show a degree of naturally acquired tolerance. The NICE CKS guideline2 confirms that parents or carers should be advised to start the reintroduction with baked milk products, as thermal heating reduces the allergenicity of some CMPs.4

Should there be any delay in dietetic support for service reasons, the following websites may be useful for both the families and primary healthcare professionals:


Cow's milk allergy is the commonest and clinically most complex form of food allergy among children in the UK.4 The first signs or symptoms typically appear in the early weeks of life and usually present in primary care.2,6 There is evidence of apparent missed diagnosis, incorrect diagnosis, delayed diagnosis, and subsequent suboptimal management in a significant number of these infants.6 The challenge is that resulting from this, we now have three UK guidelines that are related to the early recognition, diagnosis, and management of milk allergy in primary care with the very real potential for 'guideline overload'. This article has drawn together the recommendations for primary care from these three guidelines. The MAP milk allergy algorithms on p.48 and p.50, designed by a subgroup of the clinicians on the NICE CG116 food allergy guideline development group, have been piloted in the Northern Ireland region of the NHS and demonstrated a significant positive impact on patient outcomes and prescribing, with potential cost savings.

Figure 1: Suspected CMA in the first year of life4
Suspected cow's milk allergy (CMA) in the first year of life
Figure 2: Diagnosis and management of mild to moderate non-IgE-mediated CMA in UK primary care4
Diagnosis and management of mild to moderate non-IgE-mediated cow's milk allergy (CMA) in UK primary care

GP commissioning messages

written by Dr David Jenner, NHS Alliance GMS contract/PBC Lead

  • The guidelines summarised in this article recognise the difficulty for practitioners in establishing an accurate diagnosis of CMA in primary care
  • Commissioners should consider developing specific community paediatric dietary advice services that GPs and health visitors can easily access for advice and guidance on those infants with CMA who should appropriately stay in primary care, without the need for formal consultant referral and PbR tariff charges
  • Education programmes for GPs on the awareness and early recognition of CMA, based on algorithms from the guidance, would be worthwhile
  • Formulary choices for extensively hydrolysed and amino acid-based formulas should be published in CCG formularies, including suggested quantities to be prescribed and duration of therapy
  • Tariff costs for paediatric outpatients: £212 (new), £123 (follow up).a

CMA=cow's milk allergy; PbR=Payment by results

Back to top

Key points

  • 2% to 3% of 1–3-year-olds in the UK have confirmed CMA
  • 86% of CMA in infancy is diagnosed in primary care
  • There is evidence of:
  • over diagnosis with the perceived parental prevalence of CMA being significantly higher than that confirmed by appropriate tests
  • under-recognition, significant delay in diagnosis, and subsequent suboptimal management, especially in choosing the most appropriate initial hypoallergenic formula to prescribe
  • When CMA is suspected in an infant, the signs and symptoms will fall within the mild to moderate non-IgE-mediated category in 90% of cases
  • There are three UK guidelines aimed at the better recognition, diagnosis, and management of CMA in primary care
  • All the guidelines emphasise the crucial role played by the dietitian in supporting the family and GP
  • This article draws these three guidelines together into a single and practical tool for primary care use

Back to top


  1. Boyce J, Assa'ad A, Burks A et al. Guidelines for the Diagnosis and Management of Food Allergy in the United States: Report of the NIAID-Sponsored Expert Panel. J Allergy Clin Immunol 2010, 126 (6): S1–S58.
  2. NICE. Cow's milk protein allergy in children. Clinical Knowledge Summaries. NICE, 2014. Available at: www.cks.nice.org.uk/cows-milk-protein-allergy-in-children
  3. NICE. Diagnosis and assessment of food allergy in children and young people in primary care and community settings. Clinical Guideline 116. NICE, 2011. Available at: www.nice.org.uk/ guidance/cg116
  4. Venter C, Brown T, Shah N et al. Diagnosis and management of non-IgE-mediated cow's milk allergy in infancy—a UK primary care practical guide. Clin Transl Allergy 2013; 3 (1): 23. Available at: www.ctajournal.com/ content/3/1/23
  5. Venter C, Pereira B, Voigt K et al. Prevalence and cumulative incidence of food hypersensitivity in the first 3 years of life.Allergy 2008; 63 (3): 354–359.
  6. Sladkevicius E, Nagy E, Lack G, Guest J. Resource implications and budget impact of managing cow's milk allergy in the UK. J Med Econ 2010; 13 (1): 119–128.
  7. Venter C, Brown T. Infants with cow's milk allergy have reduced quality of life. Guidelines in Practice 2014; 17 (4): 36–49.
  8. Vandenplas Y, Koletzko S, Isolauri E et al. Guidelines for the diagnosis and management of cow's milk protein allergy in infants. Arch Dis Child 2007; 92 (10): 902–908.
  9. NICE. Anaphylaxis: assessment to confirm an anaphylactic episode and the decision to refer after emergency treatment for a suspected anaphylactic episode. Clinical Guideline 134. NICE, 2011. Available at: www.nice.org.uk/guidance/CG134
  10. Abstract PAAM, Paediatric Asn Allergy Meeting of the European Academy of Allergy and Clinical Immunology (EAACI), Dublin 2014. Full manuscript submitted for publication.