Table 3.

Accuracy of patient generated pre-visit differential diagnosis compared with clinician differential, by (A) Proportion of matched diagnoses and (B) Odds ratios (95% CI) for matched diagnoses between groups

A
Diagnostic accuracy (%)	Overall (n = 300)	Google search (n = 100)	HFD (n = 99)	No search (n = 101)	p-value
Primary outcome (≥2/3 matched)	79 (26.3)	27 (27.0)	28 (28.3)	24 (23.8)	0.76
Sensitivity analysis
3/3 matched	6 (2.0)	3 (3.0)	2 (2.0)	1 (1.0)	0.62
≥1 matched	223 (74.3)	80 (80.0)	71 (71.7)	72 (71.3)	0.28

B
Diagnostic accuracy (Odds ratio (95% CI)	Search groupa (n = 199)	p-value	Google searchb (n = 100)	p-value
Primary outcome (≥2/3 matched)	1.23 (0.70–2.13)	0.47	0.94 (0.50–1.75)	0.84
Sensitivity analysis
3/3 matched	2.58 (0.30–22.36)	0.39	1.50 (0.25–9.18)	0.66
≥1 matched	1.27 (0.74–2.17)	0.39	1.58 (0.82–3.04)	0.17

Search group = Google + HFD

^acompared to No Search group (ref. ¹)

^bcompared to HFD group (ref. ¹)