Results
A systematic search process retrieved a sole randomised study, and so we broadened inclusion criteria to encompass quasi-experimental designs. Fifteen studies formed our dataset including seven prospective trials and eight before-and-after studies. Nine studies evaluated cycling versus a control group and produced conflicting results whilst three studies compared cycling with antibiotic mixing, with none of the strategies appearing superior. The rest evaluated resistance dynamics of each of the on-cycle antibiotics with contradictory findings. Research protocols differed in parameters such as the cycle length, the choice of antibiotics, the opportunity to de-escalate to narrow-spectrum agents and the measurement of indicators of collateral damage. This limited our ability to evaluate the replicability of findings and the overall policy effects.