Overview for healthcare professionals

Regulatory status of omega-3 fatty acids

Purified omega-3 fatty acids are available as approved medicinal products or as food supplements. Food supplements are not regulated in the same way as medicines, and therefore their constituents may vary more, potentially affecting efficacy and safety.

There are numerous products on the UK market that contain omega-3 fatty acids. Amongst them there are 14 that have a current licence as a medicinal product. The majority of the oral formulations are indicated for use as adjuvant treatment in secondary prevention of myocardial infarction and for treatment of hypertriglyceridaemia when dietary measures are not sufficient.

Three other products with UK marketing authorisation contain named unsaturated fatty acids that belong to the group of omega-3 fatty acids (eicosapentaenoic acid and docosahexaenoic acid).

Omega-3 fatty acid medicines are not licensed for treating schizophrenia in the UK, therefore the use of approved medicinal products for this indication is off-label.

The place in therapy of omega-3 fatty acid medicines for their licensed indications is covered in several NICE clinical guidelines. These are summarised in the key therapeutic topic publication on omega-3 fatty acids, which supports the Quality, Innovation, Productivity and Prevention (QIPP) medicines use and procurement work stream.

In line with the guidance from the General Medical Council (GMC), it is the responsibility of the prescriber to determine the clinical need of the patient and the suitability of using omega-3 fatty acid medicines outside their authorised indications.

Evidence statements

Summary of the evidence

This section gives a brief summary of the main evidence. A more thorough analysis is given in the Evidence review section.

Eight relevant RCTs were identified that investigated using omega-3 fatty acids to treat schizophrenia; 7 of these were included in a Cochrane systematic review. The quality of the reported data was often poor, and the studies were small and short term.


The Cochrane review reported on 3 RCTs that measured overall mental health. Two RCTs (Fenton et al. 2001 and Berger et al. 2007) showed that omega-3 fatty acids (ethyl-eicosapentaenoic acid [E-EPA] 2 or 3 g per day) did not lead to any statistically significant improvements in Clinical Global Impression scores or proportion of people achieving symptomatic response compared with placebo at 12 or 16 weeks. One RCT (Peet et al. 2001b) reported outcomes that favoured omega-3 fatty acids (EPA 2 g per day) compared with placebo at 12 weeks in people who were not receiving antipsychotics before randomisation (fewer days on antipsychotic medication). However, this study was small (n=30) and the effects of treatments were unclear because of the concurrent, additional use of antipsychotic medication part way through the trial.

The Cochrane review reported on 4 trials that measured mental state using PANSS. In Peet et al. (2001b) (where people were not receiving antipsychotics before randomisation), a statistically significantly higher proportion of people achieved a 25% improvement on PANSS using omega-3 fatty acids (EPA 2 g per day) compared with placebo (number needed to treat 3, 95% confidence interval 2 to 8). However, this change may represent 'minimally improved' on the Clinical Global Impression (CGI) scale (Levine et al. 2008), and remains a subject of debate (Irving et al. 2006). In people already using antipsychotic drugs (Peet et al. 2001a), this outcome was not statistically significant.

The same 2 RCTs (Peet et al. 2001a and Peet et al. 2001b) found that the average PANSS total score at the end of each 12-week study favoured omega-3 fatty acids compared with placebo, as did the RCT by Emsley et al. (2002). However, no statistically significant difference was found in the larger 16-week study by Fenton et al. (2001).

Table 1 Summary of the 8 randomised controlled trials


Study duration (n=total randomised)


(n=total analysed)

Omega-3 fatty acids (n=total analysed)

Main outcome: omega-3 fatty acids compared with placebo

Fenton et al. (2001)

16 weeks (n=90)



E-EPA 3 g/day

No significant difference for total PANSS score or CGI scorea

Peet et al. (2001a) b

12 weeks (n=55)



EPA 2 g/day

Favoured EPA: % change in total PANSS score baseline to end of study (20.1% reduction with EPA compared with 10.7% reduction with placebo, p=0.05); borderline significance

Peet et al. (2001b) c

12 weeks (n=30)



EPA 2 g/day

Favoured EPA: reduction in mean total PANSS score from baseline (25.8 with EPA compared with 22.2 with placebo, p<0.02)

Favoured EPA: days needing antipsychotics (35.1 days with EPA compared with 65.3 days with placebo, p<0.02)

Peet and Horrobin (2002)

12 weeks (n=122)


E-EPA 1 g/day (n=29), 2 g/day (n=28) or 4 g/day (n=27)

No significant difference for total PANSS score for any EPA dose compared with placebo in people taking typical or atypical antipsychotics, or for EPA 1 g/day or 4 g/day compared with placebo in people taking clozapine

Total PANSS score favoured EPA 2 g/day over placebo in people taking clozapine (26% reduction with E-EPA compared with 6% reduction with placebo, p=0.04)

Emsley et al. (2002)

12 weeks (n=40)



E-EPA 3 g/day

Favoured E-EPA: mean % reduction in total PANSS score from baseline (12.6% with E-EPA compared with 3.1% with placebo, p=0.03)

Emsley et al. (2006)

12 weeks (n=84)



E-EPA 2 g/day

No significant difference for extrapyramidal symptom rating scale scores

Berger et al. (2007)

12 weeks (n=80)



E-EPA 2 g/day

No significant difference for any primary outcome measure of symptom score (assessed by BPRS, SANS, CGI-S, GAF and SOFAS)

Manteghiy et al. (2008)

6 weeks (n=106)



omega-3 fatty acids (EPA/DHA) 3 g/day

No significant difference for PANSS subscales scores

Abbreviations: BPRS, Brief Psychiatric Rating Scale; CGI, Clinical Global Impressions Scale; CGI-S, Clinical Global Impressions-Severity of Illness scale; DHA, docosahexaenoic acid; E-EPA, ethyl-eicosapentaenoic acid; EPA, eicosapentaenoic acid; GAF, Global Assessment of Functioning scale; n, number of participants; PANSS, Positive and Negative Syndrome Scale; SANS, Scale for the Assessment of Negative Symptoms; SOFAS, Social and Occupational Functioning Assessment Scale.

a Unclear whether severity or improvement CGI scale was used.

b This study also had a DHA arm (n=16), but only the EPA compared with placebo results are described here.

c This study has significant limitations relating to confounding.


In the 8 RCTs, the safety of using omega-3 fatty acids in people with schizophrenia was assessed up to a maximum of 16 weeks. An open-label extension to 1 study (Emsley et al. 2008) assessed safety up to 40 weeks. This limited evidence suggests that omega-3 fatty acids are generally well tolerated in the short term.

The summaries of product characteristics for some omega-3 fatty acid medicines (for example, Omacor, Teromeg and Maxepa) report cautions around the possibility of a moderate increase in bleeding time and use in people with non-insulin-dependent diabetes with aspirin-sensitive asthma. The study by Emsley et al. (2008) suggested that clinicians need to be aware of possible increases in bleeding time, as well as potential changes in weight and lipid metabolism, when considering omega-3 fatty acids for people with schizophrenia.

Commonly reported side effects listed in the summaries of product characteristics include dyspepsia and nausea.

Cost effectiveness and cost

No cost-effectiveness studies of omega-3 fatty acids for use in schizophrenia were identified. The costs of omega-3 fatty acid medicines are given in the table below.

Table 2 Costs of omega-3 fatty acid medicines

Omega-3 fatty acid medicine


Omacor 1000 mg capsules

(comprising 460 mg E-EPA and 380 mg DHA)

£14.24 for 28 capsulesa

£50.84 for 100capsulesb

Teromeg 1000 mg capsules

(comprising 460 mg E-EPA and 380 mg DHA)

£11.39 for 28 capsulesb

£40.67 for 100 capsulesb

Maxepa 1000 mg capsules

(comprising 170 mg EPA and 115 mg DHA)

£29.28 for 200 capsulesa

Abbreviations: DHA, docosahexaenoic acid; E-EPA, ethyl-eicosapentaenoic acid; EPA, eicosapentaenoic acid.

a NHS electronic drug tariff, August 2013 (excluding VAT).

b MIMS, August 2013 (excluding VAT).

The dosing regimen used most frequently in the RCTs in this evidence summary was 2 g or 3 g of EPA or E-EPA per day. A dose of 2 g of E-EPA per day is approximately 4 capsules of Omacor per day. The Maudsley Prescribing Guidelines in Psychiatry (11th edition) recommends a dose of 5 capsules daily for Omacor, or 10 capsules daily for Maxepa, when treating schizophrenia. The cost of these dosing regimens for 28 days would be £71.18 for Omacor, £56.94 for Teromeg and £40.99 for Maxepa.