First do no harm: A guide to choosing wisely in General Practice

First do no harm: A guide to choosing wisely in general practice is an evidence-based resource with a primary focus on reducing patient harms and avoiding low-value care. This online, ‘living’ resource will provide both general practitioners (GPs) and patients with information on:

overused tests
overdiagnosed and overtreated conditions
interventions with insufficient evidence.

Importantly, First do no harm is designed to enable positive and constructive conversations between GPs and patients, providing information and tools to assist in shared decision making. First do no harm will help empower patients to make informed and balanced decisions about their health, including information they should consider asking their GP.

Topics

Coronary Artery Calcium Scoring in assymptomatic people
Excluding allergenic foods in maternal and infant diets
Imaging in adults with acute low back pain
Management of subclinical hypothyroidism
MTHFR gene testing
Vitamin C infusions
Vitamin D testing

1. Coronary Artery Calcium Scoring in assymptomatic people

The use of coronary artery calcium scoring for screening and risk reclassification in asymptomatic patients.

RACGP position

The efficacy of coronary artery calcium (CAC) scoring in reducing the risk of cardiovascular disease in asymptomatic people has yet to be determined.
Evidence for reclassification of risk in the Australian population has yet to be determined. CAC scoring may have important clinical benefits in some settings, but this has yet to be well demonstrated in clinical trials. In addition, it may have little overall impact on patient health outcomes.¹

Traffic lights

RED – Do not take this action

Do not use CAC scoring in patients who are already at high risk of cardiovascular disease (CVD) or have established CVD (such as a history of myocardial infarction or revascularisation, or known coronary heart disease).²
Do not use CAC scoring for generalised population screening for CVD risk.²
Do not use CAC scoring to investigate patients with chest pain or other symptoms suggestive of coronary artery disease. Other cardiac investigations (eg stress testing or angiography) are more appropriate in symptomatic patients.

ORANGE – Under specified circumstances, take this action

Consider CAC for a person at moderate risk of CVD for whom the risk may be underestimated, and where the findings are likely to influence intensity of management.²

GREEN – Take this action

Use a shared decision-making approach to discuss the potential benefits and risks of the test. For example, before ordering the test, discuss with the patient whether a high or low result on this test will change your clinical management.²
For patients at high risk of CVD or with established CVD, begin treatment to reduce risk, regardless of CAC result.²

2. Excluding allergenic foods in maternal and infant diets

The practice of attempting to prevent the development of allergies by excluding common childhood food allergens from an infant’s first 12 months of life and/or from the mother’s diet during pregnancy and breastfeeding. RACGP position

Do not exclude allergenic foods from maternal and infant diets in an attempt to prevent allergies.
- Allergenic foods include: egg, peanut, cow’s milk (and dairy products), tree nuts (such as cashew or almond paste), soy, sesame, wheat, fish and other seafood
Avoiding allergenic foods in order to prevent allergies is an outdated practice and not supported by current evidence, which indicates that:
- pregnant and breastfeeding mothers should include commonly allergenic foods in their diet^1,2
- including these foods in the diet of a breastfeeding mother may support immune tolerance of allergenic foods^3,4
- these foods should be introduced into an infant’s diet within the first year of life.^1,2

Traffic lights

RED – Do not take this action

Maternal diet during pregnancy and breastfeeding

Do not recommend that pregnant or breastfeeding women omit specific foods from their diet (including foods that are commonly allergenic) as a way of preventing infant allergies^1,2,5

Infant feeding

Do not recommend partially or extensively hydrolysed infant formula, soy milk formula or goat’s milk formula as a means of allergy prevention.
Do not use acid-suppressive therapy, including histamine-2 receptor antagonists (H2RAs) and proton pump inhibitors (PPIs), to treat simple infant reflux. Reasons for not doing this include poor efficacy and an increased risk of food allergy.^6–8

ORANGE – Under specified circumstances, take this action

If an infant has severe eczema or an egg allergy, recommend regular peanut consumption in their first 12 months of life.⁹
If there is a family history of egg allergy, recommend introducing cooked egg into the infant’s diet before they are aged eight months.¹

GREEN – Take this action

Maternal diet during pregnancy and breastfeeding

With due consideration of any allergies and intolerances the mother has, encourage women who are pregnant or breastfeeding to:
- follow healthy eating guidelines¹⁰
- include foods that are common allergens, even if they are breastfeeding an infant with a high risk of allergy or atopy.^1,2

Infant feeding

If breastfeeding is not possible, recommend that parents start with a standard cow’s milk formula.^1,11
Recommend that parents:
- introduce commonly allergenic foods (particularly peanut butter, cooked egg, dairy and wheat products) into the infant’s diet in the first year of life, even if the infant has a high risk of allergy;^1,2 studies show that this may reduce the chance of food allergies developing in babies with severe eczema or egg allergy¹²
- introduce only one commonly allergenic food at each meal so that if there is a reaction, the probable cause of the reaction can be easily identified
- offer all tolerated foods regularly (eg at least twice weekly), as this may help to reduce the risk of the infant developing an allergy.

3. Imaging in adults with acute low back pain

The routine imaging of adult patients experiencing acute low back pain.RACGP position

Non-specific low back pain is a clinical diagnosis and no tests are required.^1,2
Unnecessary diagnostic imaging causes more harm than benefit because it can result in increased costs, delays in appropriate treatment, exposure to ionising radiation and increased absence from work. It may also lead to unnecessary referrals, procedures and surgery,¹ and is associated with higher rates of prolonged disability.³
Diagnostic imaging for acute low back pain in adults is only recommended after careful clinical assessment that results in a high suspicion that there is a serious cause.^1,4,5
Non-specific low back pain has a good prognosis and usually improves within four weeks if the patient uses simple pain strategies, avoids bed rest and maintains their usual activities.^4,6

Traffic lights

RED – Do not take this action

Do not request imaging for acute low back pain unless you suspect a serious cause.^1,2,5

ORANGE – Under specified circumstances, take this action

If, after clinical assessment, you suspect cauda equina syndrome, spinal infection, an acute high-impact fracture or severe neurological deficit, you must immediately refer the patient to an emergency department for review. Do not delay this by arranging imaging.¹
If, after clinical assessment, you suspect a serious cause, choose the imaging appropriate to the suspected pathology^1,5 and discuss the costs of imaging. There is no Medicare rebate for a lumbar spine magnetic resonance imaging (MRI).

GREEN – Take this action

Take a careful history and perform a clinical examination.^1,2,4
Educate the patient by explaining why you are not recommending imaging and reassure the patient by explaining the natural history of non-specific low back pain recovery.
Encourage the patient to maintain their usual activities.⁴

4. Management of Subclinical Hypothyroidism

The management of subclinical primary hypothyroidism in non-pregnant adults. RACGP position

Avoid routine thyroid function testing in non-pregnant women and well adults.
Avoid routine treatment of subclinical hypothyroidism where the thyroid stimulating hormone (TSH) is raised (4–10 mIU/L) and the T4 is normal.¹
Where there is acute illness, delay the testing.

Traffic lights

RED – Do not take this action

Do not screen for thyroid dysfunction in asymptomatic non-pregnant adults²
Do not order an ultrasound for assessment of hypothyroidism in the absence of a palpable neck mass³
Do not order repeat thyroid peroxidase antibodies (TPOAbs) testing when there has been a previous positive result⁴
Do not routinely treat patients with asymptomatic subclinical hypothyroidism¹
Do not treat subclinical hypothyroidism with triiodothyronine (T3).⁵

ORANGE – Under specified circumstances, take this action

Consider a trial of treatment for subclinical hypothyroidism if the patient’s TSH is between 4 and 10 mIU/L AND the patient has one or more of the following:
- symptoms of hypothyroidism
- positive TPOAbs
- elevated cardiovascular risk, or
- specific risk factors (including previous iodine treatment and surgery).⁴

GREEN -Take this action

Measure free thyroxine (FT4) if the patient’s TSH is above the reference range.⁴
If the patient’s TSH is mildly elevated (4–10 mIU/L) with normal FT4, re-test in two to three months to ascertain if the hypothyroidism is persistent and test for TPOAbs.⁴
Monitor patients with a positive TPOAbs, and repeat TSH and FT4 testing at three months, six months and then annually thereafter,⁴ and start treatment if the patient has a progressive rise in TSH or if symptoms develop.
If symptoms worsen or new symptoms develop, retest for thyroid dysfunction, but only at least six weeks after previous testing.⁴
If non-pregnant adults with subclinical hypothyroidism have a TSH >10 mIU/L on two separate tests three months apart AND symptoms of hypothyroidism,⁴ consider treating with levothyroxine (T4)

5. MTHFR Gene Testing

GPs’ use of methylenetetrahydrofolate reductase (MTHFR) gene testing.RACGP position

Avoid requesting a MTHFR gene test as there is no causal link between MTHFR gene variations and particular diseases,¹ and no substantial evidence to support the use of MTHFR gene testing in routine clinical practice.²
There is no evidence-based treatment that will improve health outcomes for a patient who has one or both MTHFR gene variants,³ so knowledge of a patient’s MTHFR gene status is unlikely to change how you manage and treat the patient.²

Traffic lights

RED – Do not take this action

Do not routinely order MTHFR gene testing in general practice.⁴
Do not order MTHFR gene testing as part of clinical evaluation for thrombophilia or early pregnancy loss.⁴

ORANGE – Under specified circumstances, take this action

Appropriate paediatric referral or a repeat screening may be required if you suspect homocystinuria in a newborn.¹

GREEN – Take this action

Women of childbearing age should take the standard dose of folic acid supplementation to prevent neural tube defects during pregnancy, regardless of their MTHFR status.⁴
Explain to patients that the results of this test would not change their treatment.
For patients who are carriers of the variants of the MTHFR gene, continue to provide normal care.²

6. Vitamin C Infusions

The use of vitamin C infusions as a treatment. RACGP position

There is no evidence to support the use of vitamin C infusions.

While high-dose vitamin C infusions have been proposed as part of treatment for various conditions (for example, cancer, COVID-19, sepsis and herpes zoster infection and cardio protection before percutaneous coronary intervention), these should only be conducted under the governance of a well-conducted clinical trial with ethics approval. Traffic lights

RED – Do not take this action

Do not administer vitamin C infusions in general practice settings, unless part of a properly administered clinical trial with human ethics committee approval.

ORANGE – Under specified circumstances, take this action

If a patient has questions about vitamin C infusions as a treatment:
- discuss the patient’s diet, explain the importance of good nutrition and encourage them to have an adequate dietary intake of vitamin C
- discuss their health beliefs and explain the lack of high-quality evidence for infusion therapy.

GREEN – Take this action

If a patient is particularly interested in having vitamin C infusions, explain to them that vitamin C infusions can interact with some treatment regimens, and encourage them to talk about this with their non-GP specialists, including their oncologist if they are a cancer patient.

7. Vitamin D Testing

The testing for vitamin D deficiency in patients. RACGP position

Testing for vitamin D deficiency (by measuring serum 25(OH)D) is not routinely recommended for adults (including those who are pregnant), children or healthy infants.
It may be appropriate for people at risk of vitamin D deficiency (for example, those in high-risk groups).¹
Population screening for vitamin D deficiency in older adults by measuring serum 25(OH)D is not recommended, but testing high-risk groups is appropriate.²
Vitamin D supplementation should not be routinely given to non-institutionalised elderly people.
Vitamin D testing is appropriate, and is only funded by the Medicare Benefit Schedule (MBS), if the patient:³
- has symptoms of or established osteoporosis/osteomalacia
- has increased alkaline phosphatase in otherwise normal liver function tests
- has hyperparathyroidism, hypocalcaemia or hypercalcaemia, hypophosphataemia
- has malabsorptions (for example, cystic fibrosis, untreated coeliac disease, short bowel syndrome or bariatric surgery)
- has deeply pigmented skin or chronic lack of sun exposure for cultural, medical, occupational or residential reasons
- has medications known to decrease 25(OH)D absorption (for example, anticonvulsants)
- has chronic renal failure or renal transplant recipients
- is younger than 16 years of age and has signs or symptoms of rickets
- is an infant whose mother has established vitamin D deficiency
- is an exclusively breastfed baby and has at least one of the other risk factors
- has a sibling who is younger than 16 years of age and has vitamin D deficiency.

Traffic lights

RED – Do not take this action

Do not routinely measure vitamin D in the general population.
Do not retest vitamin D levels within three months of the patient beginning to take vitamin D replacement.¹
Do not routinely use high-dose vitamin D replacement.

ORANGE – Under specified circumstances, take this action

When a patient’s vitamin D levels have returned to normal, do not retest (do not arrange annual testing), especially if the patient has not changed their lifestyle or is still taking a supplement.

GREEN – Take this action

Test for vitamin D in patients who are at high risk of vitamin D deficiency (refer to the MBS item criteria above).
If retesting vitamin D levels (after three months of the patient taking replacement), use the same laboratory.¹