The manuscript, published in the Archives of Internal Medicine can be accessed here.
1017 subjects responded to the question "How many times a week do you consume chocolate?" Body mass index was determined for 972 subjects (95.6%), who had both weight and height recorded at the screening visit.
The mean age of the subjects was 57years and 68% were male. The average BMI was 28. Subjects ate chocolate a mean 2.0 times/wk and exercised 3.6 times/wk. Greater chocolate consumption frequency was linked to lower BMI, even after adjusting for confounders.
The findings of this study are interesting but by no means conclusive. Diet composition and calorie count influence BMI. The underlying hypothesis is that modest-frequent chocolate consumption might result in reduced fat deposition, likely offsetting the added calories. It is known that chocolate is rich in antioxidants which can modify metabolism and BMI in laboratory animals. Perhaps these mechanisms may exist in humans too.
I would not recommend consuming chocolate with reckless abandon. A prospective, randomized trial of chocolate consumption for metabolic benefits is needed in humans. If such a study is funded by the chocolate industry, and, if the results are favorable, be skeptical.