Introduction To assess the effectiveness of self-completed triage forms in eliciting symptomatic status and predicting subsequent diagnoses.

Methods Consecutive patients attending a GUM clinic 3/10/16–7/10/16. Data from self-completed patient triage forms were extracted and correlated with clinician findings documented in electronic patient records at the visit. Fisher’s Exact was used to calculate association.

Results 339 patients were included of whom 56.6% were female. Median age was 29 years (14–84) and 86.4% identified as heterosexual (n=293). 54.6% of patients (n=185) indicated symptoms on the triage forms c.f. 58.7% (n=199) documented as symptomatic by clinicians. Clinicians and patients agreed on symptomatic status in 85.3% (289/339) of cases. 57.7% (n=71) of symptomatic women reported lower abdominal pain (LAP), inter-menstrual/post-coital bleeding (IMB/PCB) or dyspareunia on triage forms which were subsequently documented by clinicians on 66.2% (41/71) of occasions. These symptoms were not significantly associated with a diagnosis of PID, or other infections, when documented by clinicians or patients (p<0.05). Patient and clinician documented ‘change in vaginal discharge’, ‘lumps on genitals’ and ‘genital blisters or sores’ were significantly associated with candidiasis and bacterial vaginosis (p<0.05), genital warts (p<0.05), and genital herpes (p<0.05) respectively. Patient and clinician reported dysuria was significantly associated with NSU in men and UTI in women (p<0.05).

Discussion There was a high level of concordance between patients and clinicians regarding symptomatic status. Specific symptoms, when included in triage, are effective predictors of associated diagnoses with the exception of LAP, IMB/PCB and dyspareunia which appear to be non-specific.