Better GP care for mental illness linked to more admissions
Patients receiving the best quality care for severe mental illness may actually be the most likely to be admitted to hospital, according to preliminary research findings.
The study of over 8,000 practices found both unplanned and elective admissions went up with higher achievement of QOF points on four key indicators for severe mental illness.
Researchers said the surprising positive association between quality of care and unplanned admissions could reflect increased follow-up in the better performing practices after patients are admitted for an acute illness. The increase in planned admissions was expected because of better monitoring of patients’ physical health, they added.
The results, presented at the annual conference of the Society of Academic Primary Care in Nottingham, highlight the complexity of demonstrating clear benefits on long-term outcomes as a result of quality improvement efforts.
A team led by Professor Tony Kendrick, professor of primary medical care at Southampton University, looked at performance data on four key indicators on severe mental illness for over 8,400 practices, as well as hospital admissions in the same regions, over the period from 2006/07 to 2010/11.
Unexpectedly this revealed that better performance on these indicators was associated with higher unplanned admission rates for physical and mental health problems.
There was also an increase in elective admissions for physical problems as performance increased, which the authors had predicted would be the case.
The team speculated that ‘reverse causality’ could explain the increase in unplanned admissions, with patients discharged after an acute admission tending to receive more assessments.
Another possible explanation is that higher achieving practices tend to attract and retain more of those patients who are high-risk, they said.
Professor Kendrick told Pulse that the team is now aiming to use patient-level data to clarify the relationship, linking patients’ unique NHS numbers to hospital data, which will allow them to look at how the timing of assessments relate to hospital admissions.