The UK Quality and Outcomes Framework rewards general practices for achieving quality indicators for chronic disease management. Some indicators are multi-rewarded. For example, there are indicators for controlling blood pressure for patients with diabetes and for patients with chronic heart disease. Thus if a patient has diabetes and heart disease the practice is rewarded twice for controlling her blood pressure. Other indicators are singly rewarded: the incentivised activity is only for patients with single specific condition. We compare general practice performance on single and multi-reward indicators. We use a 2005/6-2012/13 panel of over 800 Scottish general practices, control for practice characteristics, practice fixed effects, indicator characteristics (whether the indicator was for measurement, treatment, or intermediate outcome, maximum payment, upper thresholds), condition, and year and cluster on indicators. We find that the proportion of patients with a given condition for whom a quality indicator was achieved was higher, and the proportion who were exception reported was lower, for multi-reward indicators than for single reward indicators. We also exploit the replacement of multi-reward smoking indicators by single reward indicators in 2006/7. Compared to indicators which were always single or always multi-reward, the proportion of the relevant patients for whom the smoking indicators were achieved fell when the smoking indicators were no longer multi-reward. Fine details of pay for performance schemes matter: they affect physician behaviour and patient outcomes.