The research is published in the British Medical Journal (BMJ) on Saturday, less than three weeks after the US authorities issued a warning to doctors in the light of evidence that some youngsters prescribed with these drugs became suicidal and hostile.
The Australian authors assess six published studies of trials in which children were given antidepressants for treating anxiety or depression.
They found the studies exaggerated the benefits of newer kinds of the drug or downplayed the drugs' adverse effects.
They also say the results were skewed by large numbers (up to 25 percent) of volunteers who dropped out of the programme.
And, they contend, the studies' ethical quality can be doubted because of financial links with pharmaceutical giants, which paid for the trials and remunerated the authors in at least three studies.
"Biased reporting and overconfident recommendations in treatment guidelines may mislead doctors, patients and families," the Australian team say.
Because of it, "many will undervalue non-drug treatments that are probably both safer and more effective. Accurate trial reports are a foundation of good medical care."
They conclude: "Antidepressant drugs cannot confidently be recommended as a treatment option for childhood depression."
Of the six studies, the BMJ report says, four offered no statistically significant advantage for antidepressants over placebos -- harmless lookalike pills taken by one group of volunteers as a control.
And whenever a drug was rated as significantly effective, this judgement was found to have been given by doctors. If a patient, or his parent, was asked to rate its effectiveness, the antidepressant showed no great advantage.
As for potential adverse effects, none of the studies said anything about attempts at suicide, school attendance or whether the child had needed emergency care or sought help from mental health experts, the authors contend.
The six studies had 477 patients who were treated with paroxetine, fluoxetine, sertraline or venlafaxine, and 464 who were given a placebo.
The BMJ article was lead-authored by John Jureidini of the Department of Psychological Medicine at the Women's and Children's Hospital in Adelaide, South Australia.
It comes against a growing backdrop of concern about an apparent link between antidepressants and hostility, self-harm and infatuation with suicide among children who are prescribed these drugs.
In June last year, the GlaxoSmithKline pharmaceutical company sent a letter to all doctors in Britain to tone down prescription of paroxetine (marketed as Paxil) to patients under 18, and this was followed in August by a similar warning about venlafaxine (Effexor) by the US maker Wyeth.
In December 2003, Britain banned all remaining antidepressants in this class (a category of drugs called selective serotonin uptake inhibitors) for patients under 18, with the exception of fluoxetine (Prozac).
On March 22, the United States' Food and Drug Administration (news - web sites) (FDA) issued a public health advisory about 10 antidepressants.
It warned doctors, families and other caregivers to monitor all patients for any signs of worsening depression and obsession with suicide and asked manufacturers to change the labels on drug bottles to that effect.
The 10 drugs are bupropion, citalopram, fluvoxamine, mirtazapine, nefazodone, paroxetine, sertraline, escitalopram and venlafaxine, as well as fluoxetine, which is the only FDA-approved drug for children with a major depressive disorder.