I think think that one needs to be clear about what one means by effectiveness with these sorts of illnesses. The common disorders are episodic, so that symptom levels fluctuate over time. For major depression the recovery rates from an episode are extremely high in the early weeks and then decline with each passing week as chronicity increases. This dynamic accounts for a substantial component of the (so called) high "placebo response rate" seen in this condition - as with any episodic condition where regression to the mean is a mathematical reality. If one defines "effectiveness" as the difference in the proportion improving by some specified amount in some specified time period (e.g. 6-8 weeks) or a difference in mean scores between a treated and control group in that time period then one may expect antidepressants will be "ineffective" in mildly depressed samples where most will recover within a few weeks with or without treatment and also in chronic samples where even if the medications double or triple the chance of recovery the actual proportions recovering will be low.

Hello

To be honest with you as an AI engineer of knowledge base (I think the first computer expert system with fuzzy logic) and a psychiatrist I am not happy and I was not happy with DSM V and the direction chosen in DSM IV too. Therefore I built my system on the Research Diagnostic Criteria for ICD-10 than DSM.

Please see this: (DSMV)

Five (or more) of the following symptoms have been present during the same 2-week period and represent a change from previous functioning.

Functioning is related to many things not necessary connected with mental health at all.

You can see reduction from ICD-10 (Research Diagnostic Criteria with somatic syndrome 16 item typical for depression) to 9 item.

Kielholz and Poldinger in 1987 described a few dimensions such as: endogenous (genetic), reactive, physical. Nowadays we need to add anxiety, problem solving, constant learning and markers recorded in frontal lobes and prediction of the future based on previous experiences.

Depression and anxiety seem to be normally distributed therefore it is a function in time and technically symptoms are fuzzy not black and white with huge errors at the beginning (during assessment) , therefore extremum of the function only is needed for assessment and for comparison. Of course later you will have a chance for having ranking of effective methods, which you cannot have with classic approach.

It is difficult to judge effectiveness against two criteria. There is little else available at the acute stage, apart from psychopharmaceutical intervention that can control (suppress) florid symptoms. Emergency intervention is a completely different matter from long term treatment of chronic psychiatric illness. The growing professional objection to often unduly high doses of anti-psychotic or anti-anxiolytic medication is based on the risk benefit equation of old. All psychiatric medications are detrimental to physical and mental well being and the longer the patient is exposed to them and the higher the dose the more detriment they suffer. However the emergency intervention can save them from serious injury, death or trouble with the criminal justice system and at the acute stage the risk-benefit is almost certainly skewed to justify psychopharmaceutical intervention. It is interesting to note the lack of emergency psychiatric provision by the 'talking cure' supporters and the lack of 'talking cures' from those who prefer the Rx approach.

I really agree with what Dr. Turner has said. We need better research to discern these risk-benefit distinctions and to refine these distinctions for different patients groups. Long-term treatment (psychological and pharmacological), for example, is likely to be of value to a subset with highly recurrent disorders and fail to be of net benefit for those at lower risk of recurrence. Consumer preference and choice for different therapeutic options will always be a deciding factor but we as scientists need to provide solid information on risks and benefits of alternative decisions.

I agree, especially with the lack of EBM. But we have to remember the efficacy of maintenance treatment and anti-relapse (for MDD, OCD, Anxiety disorder such as GAD and PD, less for BPD). There is lack of Evidence, surely, but common clinical practice would show it, in adults AS WELL AS IN CHULDREN and ADOLESCENTS.

I think also that the subject is different, the effectiveness of placebo was assessed as 33% of responders and you need to remember that reduction of 50% of symptoms is "good enough", the majority of chronic depressions are severe cases which by the definition didn't change after antidepressants' treatment

It depends, for example for OCD, meds are more effective in the long term as compared to the placebo.

Plenty of evidence for that but tantalisingly insufficient to be definitive. Some patients treated in the acute phases of psychosis recover and do not require long term medication. Other that have recovered and shown few symptoms for many years relapse. The old argument that 'the medications work' is demonstrable but misses a point. Since we have no idea what causes mental illness we have no idea why they work. We do know that psychiatric medications can stop the more florid and life disrupting stages of psychosis but we all know that they have adverse effects that can be life disrupting. Risk/benefit is always the gold standard test and far more insight is need by some physicians into what this means. One thing that is never acceptable is for medication to be given because it is more convenient to do so. This is a fundamental breach of medical ethics and apart from the damage it might unnecessarily cause the patient is corrosive to correct medical practice.

Yes indeed. The attached slides include a study with some startling results on acute/chronic variations in drug therapy in psychosis.

My clinical experience fully agrees with the Harrow study. Thanks for the contribution you sent