The five categories, defined by their specific symptoms and areas of brain activation, are: tension, anxious arousal, general anxiety, anhedonia - the inability to feel pleasure - and melancholia.
"We are trying to disentangle the symptom overlap in our current diagnoses which can ultimately guide tailored treatment choices," the researchers wrote in the journal JAMA Psychiatry.
Currently, depression and anxiety are the leading cause of disability and lost productivity worldwide with only one- third of patients recovering from treatment, according to the study led by researchers at Stanford University in the US.
"Currently, the treatments would be the same for anyone in these broad categories," said Leanne Williams, professor at Stanford.
More From This Section
"By refining the diagnosis, better treatment options could be prescribed, specifically for that type of anxiety or depression," said Williams.
The researchers collected and processed data from 420 participants both with healthy diagnoses and with multiple anxiety and depression diagnoses.
Researchers measured how well participants functioned in everyday life, their capacity for building social relationships and general outlook on life.
The same tests were conducted with a second independent sample of 381 people.
Using a data-driven approach that involved machine learning algorithms, researchers processed the data and were able to identify the same five new categories across both groups.
Results showed that 13 per cent of participants were characterised by anxious arousal, 9 per cent by general anxiety, 7 per cent by anhedonia, 9 per cent by melancholia and 19 per cent by tension.