According to a 2011 study, Plan B, the levonorgestrel-based "morning-after pill", is probably less effective in the half of American women whose mass is ≥ 70kg.1 No studies have ever been conducted on the effectiveness of higher dose for higher body weight (as of 2011), even though dose proportional to body size is correct for some medications.2 (Maybe a double dose is safe and effective, but science hasn't checked.) Ulipristal acetate (UPA) works for a few more people but also becomes less effective at common body masses.
The study stating this is so-so. The study contradictorily implies that body mass and BMI are each better predictors than the other.3 It has a small sample size of the subjects it makes conclusions about.4 It considers this pill ineffective if its small sample could not prove the pill effective at p=95%. Please correct me and add info if you are better at epidemiological research than amateur myself.
Glasier, Anna, et al. "Can we identify women at risk of pregnancy despite using emergency contraception? Data from randomized trials of ulipristal acetate and levonorgestrel." Contraception 84.4 (2011): 363-367. http://wcsh.org.uk/sites/wcsh.org.uk/files/Glasier%202011.pdf↩
"It might be tempting to suggest doubling the dose of [emergency contraception] for women over 70 kg (as is advised for women on enzyme-inducing anticonvulsant drugs), but data are necessary to support such practice."↩
"Moreover, a woman who weighs over 70 kg may have a low BMI if she is tall, but she would, nonetheless, be advised that she may be at greater risk of [emergency contraception] failure." but "Three of the six covariates tested were found to have a statistically significant effect on the risk of pregnancy: BMI, conception probability and further intercourse." (The other three are "age", "weight" and "time (hours) from unprotected intercourse to treatment with [emergency contraception]", unless I parsed the confusingly punctuated list wrong. The latter being insignificant also makes me dubious, because taking Plan B sooner is well known to be more effective; the study doesn't even comment on this.)↩
"Moreover, the number of women in the studies with a BMI of 35 kg/m² or greater was small, and the number of pregnancies among this group was extremely small." The tables bear this out.↩