So will trying it for longer confirm things if you're saying a few weeks won't?

A few weeks of self-experimentation cannot confirm an ADHD diagnosis, and extending the trial does not fix the core problem: you still lack a baseline, a blinded comparison, and a proper evaluation.

The placebo response is not a few days of wishful thinking. In a randomized, double-blind, cross-over n-of-1 trial of 133 drug-naïve children with ADHD, 18.8% showed a 30% or greater improvement on placebo alone, and 61.6% did as well or better on placebo as on methylphenidate [1]. That is not a brief honeymoon effect. It is a substantial, sustained response that persists across weeks of blinded treatment. If you try the drug for a month and feel better, you have no way to know whether you are in that 18.8% or the 61.6%, because you are not blinded and you have no baseline measure.

The longer you take it, the more you risk anchoring on a false positive. A positive response to a stimulant is not diagnostic. In the same trial, placebo response correlated with methylphenidate response, meaning the people who improved on sugar pills also improved on the drug [1]. That makes the self-experiment even more misleading: you cannot distinguish a genuine drug effect from your own expectation, and the overlap is so large that a "positive" result proves almost nothing.

The cardiovascular risks do not shrink with time. In a separate adult cohort, 10.9% reported adverse events, most commonly reduced appetite and headache [4]. Stimulants raise heart rate and blood pressure [2, 3]. Atomoxetine caused a mean heart rate increase of 9 beats per minute and a diastolic blood pressure increase of nearly 11 mmHg in one pediatric study [3]. A longer trial means more cumulative exposure to those effects, not less, and you still do not know if you have an underlying condition that amplifies them.

The only thing a longer trial confirms is whether the drug does something for you subjectively. It cannot tell you whether you have ADHD, whether the benefit is real or placebo, or whether the long-term risks are worth it. That requires a proper diagnostic evaluation, including baseline cardiovascular assessment, before you start.

My call: no, a longer trial will not confirm anything. It will only deepen the false certainty. Get the diagnosis first, or do not take the drug. Confidence: high.