Thirteen RCTs were included in the review with a total of 387 physicians (not stated in one study) and 7518 patients.
The majority of the studies were conducted in the US (n=12) and were based on the general population (n=3) or people with unrecognised psychological problems such as anxiety or depression (n=4). Most studies employed educational methods (n=8), either group (n=5) or individual (n=3), though a significant number did not report providing clinicians with any introduction to the collection and feedback of patient-based health information. In all of the studies, the information was fed back to the medical profession rather than other health professionals. Most studies used generic instruments, most commonly SF-36 (n=3), however almost half of the studies used a condition specific instrument and one study used both.
Detection rates (n=7):
Four studies found statistically significant higher detection rates for psychological problems (only for one subgroup of patients in one study) and one for functional problems (but only one of eight functional problems), in the intervention group versus the control group.
Change to treatment (n=10): Two studies found a statistically significant increase in changes to treatment in the intervention group compared with the control group.
Test ordering (n=10):
One study showed a statistically significant increase in test ordering in the treatment as compared to the control group.
Referral rates (n=7):
Two studies showed significantly higher referral rates to other professionals in the intervention groups compared with the control.
Patient adherence (n=1):
No significant differences between the control and intervention groups were reported.
Doctor-patient communication (n=4):
One study reported that the intervention group had improved doctor-patient communication (patient self-report) versus the control group.
Patient functioning (n=7):
One study reported significant improvements in mental functioning in the intervention group compared with the control. Another study showed had significantly better self-reported functional status and fewer anxiety symptoms in the intervention group versus the control group, despite similar anxiety scores on the SCL-90.
Patient satisfaction (n=3):
No significant differences between the control and intervention groups were reported.