A quick glance at the overview shows that the study only lasted for 2 weeks (and 6 training sessions). I don't attach much weight that suggests one method of training is better than (or just as good as) another based on such a short period.
From what I've read, he took two sets of people with unspecified fitness levels; some trained using established recommendations from health agencies (x hours of low intensity cycling per week, y hours of medium intensity cycling per week, etc), and others trained using interval training techniques (albeit not at all-out pace, but slightly lower than all-out, and for slightly longer1). The subjects showed the same "benefits", despite the second lot exercising for a shorter duration overall.
People who may not have exercised enough in the past show benefit of exercising more than they used to (despite using different regimes). Shock horror.
Study too small, too short, not adequately controlled, and outcome too subjective. If I were a cynic I'd smell free advertising for an upcoming book promoting a new training regime.
1. Which seems to be the actual claim of this paper. Not that interval training works just as a well as hours of low intensity work, but that interval training at a more comfortable level just below all-out works just as a well as shorter durations of interval training at all-out pace. Extending this claim to say that as intensity decreases, duration should increase to compensate, and that you can pick where on this trade-off you'd like to train is certainly bold and something that this research doesn't back up at all.