Does pay-for-performance influence outcomes for people with serious mental illness?

Talk Code: 
Rachael Williams
Panos Kasteridis, Christoph Kronenberg, Nils Gutacker, Tim Doran, Anne Mason, Nigel Rice, Hugh Gravelle, Maria Goddard, Tony Kendrick, Najma Siddiqi, Simon Gilbody, Ceri Owen, Lauren Aylott, Rowena Jacobs
Author institutions: 
Clinical Practice Research Datalink, University of York, University of Southampton


The UK’s Quality and Outcomes Framework (QOF), the largest pay-for-performance scheme in primary care worldwide, offers GP practices payments to ensure comprehensive care plans (CPs) are in place for patients with serious mental illness (SMI). Well-coordinated primary care may reduce the need for hospital admissions; however, previous studies have relied on practice-level data and couldn’t ascertain which patients received CPs or when, prohibiting analysis of the temporal effect of financial incentives. This study aimed to determine whether better quality primary care, reflected through the documentation of CPs, is associated with SMI hospitalisations or death.


Patients with newly diagnosed SMI and at least one year of prior registration were identified from Clinical Practice Research Datalink UK primary care data between 2006/07 and 2013/14. Patients were included in the study cohort if eligible for linkage to admitted patient care Hospital Episode Statistics and death certificates. Patients were followed from diagnosis until the outcome or censoring. Cox survival models were estimated, with the hazards of the first admission with a primary diagnosis of SMI and death specified as functions of time-varying CPs (none, current, expired), demographics, local area characteristics and illness severity proxies. Subgroup analyses were conducted by age and diagnosis (schizophrenia vs. bipolar disorder).


The study included 5,231 newly diagnosed patients with SMI (3,000 schizophrenia, 2,231 bipolar), of which 3,590 (69%) received a CP, 568 (11%) had an SMI admission and 533 (10%) died. Overall, current and expired CPs weren’t associated with time to SMI admission (hazard ratio (HR) 1.11, p-value (p) 0.388, and 1.02, p=0.914 respectively), but current CPs were associated with decreased time to admission for patients aged ≥66 or with bipolar disorder (HR 1.87, p=0.039, and 1.45, p=0.017 respectively). Overall, current and expired CPs were associated with increased time to death (HR 0.57, p<0.001 and 0.72, p=0.046 respectively).


This study shows current CPs are associated with decreased time to SMI admission in older patients and those with bipolar disorder, and increased time to death overall.

Submitted by: 
Funding acknowledgement: 
This research is funded by the National Institute for Health Services & Delivery Research programme (project number 135440). The views and opinions expressed are those of the authors and do not necessarily reflect those of the HS&DR programme, NIHR, NHS or the Department of Health.