This is just wrong. Depression
is a descriptive condition,
not a prescrtive one.
Many psychiatrists--the quack ones--will use the language of "chemical imbalance" as a sort of "appeal to authority" fallacy. It should be obvious that this is a fallacy because:
A) at no point have they actually measured the patient's brain chemistry (and it's not obvious what it would even mean to do this? an imbalance where? compared to what?)
B) the doctor diagnosed the depression doing exactly what TFA did: matched a set of symptoms to a list.
One of the harshest criticisms of the DSM-V -- the latest version of this descriptive manual--is that it grossly expands the symptoms that qualify as depression (to include things like bereavement, for example).
The "chemical imbalance" theory was based on the success of SSRIs to treat depression. These are chemicals that inhibit the reuptake of serotonin, which results in overproduction of that neurotransmitter. "Ah, so depression is a lack of serotonin"
Counterfactual evidence is conveniently ignored. For example, the fact that antipsychotics, which block the reception of serotonin (and so are basically the opposite of SSRIs), are also powerful antidepressants!
It's also a nonsense theory to anyone with an undergraduate knowledge of neuroscience. There is no "happiness" neurotransmitter. Serotonin is simply an inhibitory neurotransmitter, which reduces the odds of a neuron 'firing'. Which neuron? Part of what network? Wait, what? In the immortal words of Paul Dirac "it's not even wrong."
In fact, most of the most promising new therapies for depression have nothing to do with serotonin: psylocibin (many orders of magnitude more effective than SSRIs), ketamine and transcranial magnetic stimulation are examples.
Of course, the mind is a side effect of physiology: there must be some biological explanation to depression. But that doesn't mean that it's the same explanation for everyone, or that it has anything to do with a "chemical imbalance" of one neurotransmitter vs another.