Table 3.

Clinical actions: use of appropriate investigations (%)

	Hospital A (n = 56)				Hospital B (n = 26)				Hospital C (n = 20)				Hospital D (n = 13)
	NA	No exam	Omitted	Done	NA	No exam	Omitted	Done	NA	No exam	Omitted	Done	NA	No exam	Omitted	Done
Chest radiograph	0	-	38	63	0	-	35	65	0	-	35	65	0	-	0	100
Sputum Smear	0	-	52	48	4	-	42	54	10	-	85	5	15	-	0	85
Sputum Culture	0	-	75	25	4	-	54	42	10	-	90	0	15	-	0	85
LN aspiration	11	80	9	0	23	77	0	0	45	35	15	5	69	8	15	8
Pleural tap	66	21	2	11	0	96	0	4	10	70	0	20	69	8	8	15
Lumbar puncture	54	25	2	20	19	62	0	19	30	40	5	25	46	15	0	39
Other	25	-	68	7	8	-	89	4	10	-	45	45	54	-	0	46

The TB-PBPR tool was used to assess whether appropriate investigations were performed. We recorded 'NA' where the investigation was not applicable (for example, where 'Sputum Smear' had identified acid fast bacilli), we recorded 'no exam' where the relevant clinical examination was not documented, we recorded 'omitted' where an investigation was applicable but not performed, and we recorded 'done' where the investigation was performed. For example, in hospital C, lymph node (LN) aspiration was not applicable in 45% of patients and was not documented in 35% of patients. LN aspiration was omitted in 15% and done in 5% of the patients.