Table 1

Details of the OIAM dataset used in this work, with patient information for the complete dataset

ICPC-2 code	No of transcripts	%
A: General	14	5.9
B: Blood, blood forming	8	3.3
D: Digestive	44	18.4
F: Eye	5	2.1
H: Ear	11	4.6
K: Circulatory	32	13.4
L: Musculoskeletal	65	27.2
N: Neurological	20	8.4
P: Psychological	50	20.9
R: Respiratory	37	15.5
S: Skin	32	13.4
T: Metabolic, endocrine, nutritional	24	10.0
U: Urinary	18	7.5
W: Pregnancy, family planning	11	4.6
X: Female genital	14	5.9
Y: Male genital	7	2.9
Total ICPC-2 code labels	392	164
Total unique consultations	239	100
No of ICPC-2 codes assigned to a consultation (see figure 1)
0	2	1
1	128	53
2	62	26
3	40	17
4+	8	3
Duration (minutes)
<5	13	5.4
5–10	79	33.1
10–15	82	34.3
15–20	52	21.8
20–35	13	5.4
Dataset statistics below are for the original patient sample of N=334.16 This information was not available to compute for the N=239 subset in our experiments	No of patients	%
Sex
Female	212	63.5
Male	122	36.5
Age
18–34	91	27.2
35–54	94	28.1
55–74	99	29.6
≥75	36	10.8
Not reported	14	4.2
Ethnic group
White	291	87.1
Other	43	12.9
IMD (Indices of Multiple Deprivation) quintile
1st (least deprived)	106	31.7
2nd	54	16.2
3rd	35	10.5
4th	53	15.9
5th (most deprived)	84	25.1
Data unavailable	2	0.6