The following boxes 1 and 2 contain all recommendations developed and externally reviewed. Expansion on the rationale, particularly where a ‘strong’ recommendation strength was issued in the absence supporting high quality evidence, will be provided for in the discussion. Note that recommendations herein only apply to non-diabetic populations. In those who have a diagnosis of T2DM, and have experienced distressing AIWG for which pharmacological management may be suitable, optimisation of metformin prescribing may be considered, or further pharmacological treatment with a glucagon-like peptide-1 agonist, for example, semaglutide, where the evidence base in T2DM is now substantial,23 and in AIWG treatment emerging.24

Box 1 Recommendations developed addressing the point at which metformin should be considered for antipsychotic-induced weight gain (AIWG) management and associated baseline screeningArea 1 (KHQ 1–4): appropriateness of metformin

Recommendation 1: The use of metformin in the treatment of AIWG can be applied in two ways; as part of an early intervention strategy or in the treatment of established weight gain. We recommend that preference should be given to early intervention strategies, where possible.

Strength of recommendation: Strong. Quality of evidence:Low

Recommendation 2: For the purposes of this guideline, early intervention in the management of AIWG is defined as the implementation of an intervention following a ≥7% increase in baseline body weight, within 1 month of antipsychotic treatment. (good practice point)

Recommendation 3: In the case of either early intervention or treatment, where non-pharmacological interventions are deemed appropriate and acceptable to the patient we recommend that these be offered before metformin.

Strength of recommendation: Strong Quality of evidence: Moderate

Recommendation 4: Where lifestyle interventions available to patients are unacceptable to them or are inappropriate, for example in the case of physical disability, we recommend the use of metformin as an alternative first-line intervention.

Strength of recommendation: Strong Quality of evidence: Moderate

Recommendation 5: Where non-pharmacological interventions are appropriate but seemingly ineffective, we recommend metformin be offered as an alternative. It should be noted that evidence supports improved efficacy of metformin in attenuating AIWG when initiated at earlier time points in antipsychotic treatment. Therefore, what constitutes an appropriate trial length of non-pharmacological interventions must consider this.

Strength of recommendation: Strong Quality of evidence: Moderate

Recommendation 6: We recommend the use of metformin to attenuate weight gain induced by any antipsychotic.

Strength of recommendation: Strong Quality of evidence: Moderate

Area 2 (KHQ 7): Initiating metformin (baseline screening)

Recommendation 1: Baseline renal function must be assessed before treatment is started. Where the Estimated Glomerular Filtration Rate (eGFR) is <60 mL/min, dosing of metformin should be adjusted. Metformin is contraindicated in those with an eGFR of <30 mL/min. (good practice point)

Box 2 Recommendations developed addressing optimal metformin dosing, proosed treatment goals, ongoing monitoring and management of side effects, alongside deprescribingArea 3 (KHQ 5): metformin dosing

Recommendation 1: We recommend metformin be started at 500 mg twice daily with meals. Metformin dosing should be increased in increments of 500 mg every 1–2 weeks.

Strength of recommendation: Strong Quality of evidence: Low

Recommendation 2: We recommend a target dose of metformin of 2000 mg/day. The target dose, however, should consider individual tolerability and evidence of efficacy.

Strength of recommendation: Strong Quality of evidence: Moderate

Area 4 (KHQ 6): Assessing response to treatment

Recommendation 1: If metformin is being used as part of an early intervention strategy, we recommend that plateau of weight gain should be the goal of treatment. Reversal of weight gained to date due to antipsychotic treatment may also be feasible.

Strength of recommendation: Strong Quality of evidence: Moderate

Recommendation 2: Where metformin is being used to induce weight loss in those with established antipsychotic-induced weight gain, we suggest the goal of metformin treatment be to induce a weight loss of at least 5% of baseline body weight within 6 months of treatment.

Strength of recommendation: Conditional Quality of evidence: Low

Recommendation 3: Goals of treatment should be individualised and agreed collaboratively with the patient. (good practice point)

Area 5 (KHQ 7): Ongoing monitoring

Recommendation 1: Renal function should be monitored annually. In those who are at increased risk of renal impairment for example, those with chronic kidney disease or the elderly, renal function should be measured every 3–6 months. (good practice point)

Recommendation 2: Intermittent monitoring of vitamin B12 levels is recommended, especially where evidence of megaloblastic anaemia is present. (good practice point)

Recommendation 3: Clinicians should also monitor the patients’ adherence and tolerability to both the antipsychotic and the metformin regularly. (good practice point)

Area 6 (KHQ 7): Management of side effects

Recommendation 1: Gastrointestinal side effects are dose related and can be managed through dose reduction and/or a slower dose titration. (good practice point)

Recommendation 2: The estimated incidence of lactic acidosis is 4.3 per 100 000 person-years in metformin users. Adjustment of dose to account for low levels of renal function will help to mitigate risk. Additionally, avoidance of metformin in certain groups—including those with a history of alcohol misuse or in those who are prescribed interacting medicines will also reduce risk of lactic acidosis occurring. (good practice point)

Area 7 (KHQ 6): Deprescribing

Recommendation 1: Where treatment goals have been reached at 6 months, we recommend metformin be continued. However, lack of evidence to support the continuation of metformin beyond 6 months must be considered as part of the risk–benefit assessment.

Strength of recommendation: Strong. Quality of evidence: Low

Recommendation 2: Where agreed treatment goals have not been reached at 6 months, we recommend that treatment be reviewed and the following undertaken:

The dose of metformin should be increased to 2000 mg/day, where possible.

If treatment has been optimised as much as possible, treatment should be stopped

Clinicians should check adherence, and stop if not mostly adherent.

Strength of recommendation: Strong. Quality of evidence: Low