Medicinal cannabis policies in Australia, the US and Canada are being compared in a University of Newcastle study aimed at improving public health.
The study will investigate how products and doses are recommended.
A key issue in the medicinal cannabis sector is finding the right product to treat the right condition.
The research will involve a survey of medical providers and pharmacists, asking where they get information about cannabis products and what they recommend to patients.
The survey will have input from US state cannabis regulators, Health Canada and Australia's Therapeutic Goods Administration.
Myfanwy Graham, who has dual degrees in plant medicines and pharmacy, is leading the project through a Fulbright scholarship.
She is based in Los Angeles for the 12-month project, which involves analysing the effects of cannabis policy on patient health in the three countries.
Researchers will work together to "address important evidence gaps".
These gaps include "product and dose selection for certain physical and mental health conditions".
Medicinal Cannabis Industry Australia chair Peter Crock said doctors need access to "the right information so they can share that with patients and prescribe properly".
Mr Crock said prohibition caused "major delays in mainstream medical research" into cannabis.
"We need more clinical work to be done. Support for that will be critical," he said.
"In Australia, the special access scheme means doctors have been able to prescribe unregistered medicine, which was forward thinking."
Almost 350,000 applications to prescribe medicinal cannabis have been approved in Australia since 2016.
Most have been issued in the past three years, as prescribing was made easier and online medical services boomed.
Chronic pain, anxiety, sleep disorders and cancer pain are the top health conditions cited in approved applications.
Applications to prescribe cannabis have been approved for more than 140 conditions, including post-traumatic stress disorder, depression, epilepsy, arthritis, fibromyalgia, Crohn's and Parkinson's.
Australian Medical Association president Steve Robson said "at the moment only two medicinal cannabis products are registered".
"They are nabiximols, for patients with moderate to severe spasticity due to multiple sclerosis; and epidyolex, for seizures associated with Lennox-Gastaut syndrome or Dravet syndrome," Professor Robson said.
He added that the AMA was confident the TGA will "continue to consider the evidence base for the use of medicinal cannabis".
Mr Crock said "we're at the early stage of all this".
"We don't want doctors to say 'there's no science behind this so I won't touch it' because we know there are people benefiting massively from the use of medicinal cannabis.
"But we've got to get the supporting evidence to catch up, so there is a platform we can use to talk about it.
"At the moment, there's still a lot of smoke and mirrors or myth and mystery around how cannabis can be produced in a way to be reliably prescribed by doctors."
Ms Graham's research involves a University of Newcastle team from the Australian Centre for Cannabinoid Clinical and Research Excellence.
The Australian National University and University of Southern California are also involved.
Their study will also examine whether government policy influences the products and formulations selected, and advice given, for various health conditions.
The cannabis plant's sativa and indica varieties have long been thought to have different effects on the body.
Indica products were considered more like sedatives and sativa products more energising.
Mr Crock said US research showed "there is no difference between the two".
He said there had been a "lot of marketing and myths around sativa versus indica".
Cross-breeding between the two plant varieties meant it was a "fallacy to think they produce different cannabinoids".
"Just because a plant looks different, doesn't mean that you understand what the plant is producing," he said.
"You've got to do the fundamental research to look at what cannabinoids are being produced by the plant."
More scientific work was needed to examine "what cannabinoids are present and effective in certain situations".
And clear labelling is needed of the expected therapeutic effects of particular products.
Ms Graham is also examining the legal recreational cannabis markets in the US.
"In some US states, the legalisation of recreational use has increased the diversity of cannabis products available," she said.
She said these aren't necessarily the type of products that medicinal patients should use.
Medicinal cannabis isn't suitable for everyone.
THC-containing products may harm "vulnerable patient groups", such as those with "certain psychiatric or cardiovascular conditions".
"Well-designed public health policy frameworks protect vulnerable patient groups, while ensuring access to patients with treatment-resistant conditions," she said.
Mr Crock added that "the whole idea of home grown and people self-medicating is not the medical environment we base things on".
