The
 use of alternative medicines for the management of diabetes is 
widespread in the Indian subcontinent, mostly in the form of so-called 
nutraceuticals (eg, capsules made of fenugreek or bitter melon extract).
 In 2014, the Indian Government created an independent Ministry of AYUSH
 (Ayurveda, Yoga and Naturopathy, Unani, Siddha and Homoeopathy), 
emphasising the prominence of alternative medical practices in the 
country. Furthermore, Ayurveda and other forms of alternative medicine 
are increasingly visible in the public domain, with some officials 
describing them as effective and emphasising a supposed absence of 
side-effects.
We searched the scientific literature to 
gauge whether firm claims of efficacy and safety regarding Ayurvedic 
treatments could be made, as is increasingly being done in the 
mainstream media in India. We identified three major systematic reviews 
of studies investigating the use of Ayurvedic treatments for diabetes. 
In 2001, Hardy and colleagues
1
 discussed the major methodological limitations in the evidence base: 
few randomised trials, underpowered studies, and the use of 
inappropriate statistical methods. That human trials with Ayurvedic 
medicines fall short of methodological standards was again emphasised in
 a systematic review in 2004.
2 About 7 years later, Sridharan and colleagues
3
 reviewed the existing literature and noted that, because of 
methodological deficiencies and small sample sizes, no definite 
conclusions regarding the efficacy of Ayurvedic treatments could be 
drawn, affirming that there was insufficient evidence to recommend their
 use in routine clinical practice.
 
There are other 
important issues pertaining to alternative medicines in India. For 
example, there is indistinct compartmentalisation between herbs, 
Ayurvedic medicines, and functional foods. As such, there is the 
possibility of interactions between the potential bioactive ingredients 
of the herbs used in alternative medicines (which are often given as 
mixed formulations), the toxicity of which should be properly studied in
 keeping with national guidelines (
appendix).
 Notably, Indian guidelines have waived or relaxed the rules for 
rigorous pharmacological and toxicology studies for Ayurvedic products (
appendix), provided that medicines are “prepared in same way as mentioned in ancient Ayurvedic treatises” such as 
Charaka Samhita or similar texts, purported to have been written and rewritten between the 6th century BCE and the 1st century CE.
 
The
 safety of Ayurvedic preparations is also a major concern. In addition 
to several reports of heavy metal toxicity due to these drugs, in one 
study, a fifth of Ayurvedic medicines manufactured in the USA and India 
contained detectable lead, mercury, or arsenic compounds, which exceeded
 one or more standards for acceptable daily intake of toxic metals.
4
 Among the metal-containing products, 95% were sold by US websites and 
75% were claimed to have been manufactured in accordance with good 
manufacturing practices.
 
If Ayurvedic medicines are to be
 recommended for use in diabetes care, their active components should be
 defined, their precise pharmacodynamic actions identified, and 
bioassays for their standardisation developed; essentially, rigorous 
methods of modern pharmaceutical drug testing should be applied. 
Importantly, toxicity of these compounds remains an area of major 
concern. Based on current scientific data, there is scant scientific 
basis for the use of these products in the management of diabetes.
We declare no competing interests.