We have seen this question in the past and some good responses were posted-- and, Alto Stratus on SurvivingAntidepressants made an excellent post on this a couple of years ago, but unfortunately I could find neither.
On August 3rd Sleepyhead posted the question under the heading Why Does Weaning Down The Last 10-15 mg Seem Harder? It'a a worthwhile topic, so I did some research which yielded precious little except that it's not too common with those who withdraw, and in tapering there is a simple math-trap that is easy to overlook-- here's a hypothetical example to illustrate: Let's say your 100 mg capsule has 100 beads, each bead is 1 mg, and you plan to taper to 0 over 100 days.
Seems simple enough.
But here's the catch-- it's very important to consider the percentage of drop from the foregoing dosage-- The very first bead you remove from the above example is just 1%. But near the end, the second-to-last bead is a 50% drop! Therefore, you can see the need to alter the percentage of tapering toward the end-- stretch out the timeline and stretch out the remaining dosages. If you have a week to go-- consider stretching that out to a month or more.
Dr Stuart Shipko of Pasadena CA is a good friend to those suffering psychiatric medications. He has written much on the subject, and you'll see his name alongside other outspoken professionals like Robert Whitaker, Peter Breggin, and David Healy. Dr Shipko posts on madinamerica.com and his own website is http://www.stuartshipko.com
I asked Dr Shipko this question about the last few milligrams, and here is his response:
First of all, this is not the case for everyone. I don't know why this should be the case. I can speculate that towards the lower doses the percentage reduction is greater. For example, a cut from 2 mg to 1 mg is a 50% reduction, while a cut from 5 to 4 mg is a 20% reduction. There may be other factors. It seems that the SSRIs both cause and mask drug toxicity. Perhaps when the dose gets low enough there is a secondary problem related not just to receptor changes but to loss of the covering effect. My answer is that I don't know why this is the case, and that I see this in my practice, but it is not as common as you might think.