Exploring the potential cost-effectiveness of a new computerised decision support tool for identifying fetal compromise during monitored term labours: an early health economic model

Table 1 Current practice labour alert and baby compromised classifications and numbers of alerts^{a, b}

	(Alert & SC) (A)	(No Alert & SC) (B)	Total SC (A + B)	Sensitivity (A/(A + B))	(Alert & NSC) (C)	(No Alert & NSC) (D)	Total NSC (C + D)	Specificity (D/(C + D))	Total Alerts (A + C)
Current practice	38	63	101	0.38	2,984	19,748	22,732	0.87	3,022
Stage 1 analysis ^c
Base-case	56	45	101	0.55	2,273	20,459	22,732	0.90	2,329
Stage 2 analysis ^d
25% reduction	52	45	97		2,278	20,459	22,737		2,329
50% reduction	47	45	92		2,282	20,459	22,741		2,329
75% reduction	43	45	88		2,287	20,459	22,746		2,329
100% reduction	38	45	83		2,291	20,459	22,750		2,329

^aFor the John Radcliffe Hospital cohort
^bAlso shown are the implications of potential improvements in alert sensitivity and specificity with attentive CTG (Stage 1), and with subsequent reductions in the risk of severe compromise for additionally identified babies (Stage 2)
^cThe improved sensitivity with attentive CTG enables the identification of more compromised babies during labour, thus leading to altered delivery types and delivery outcomes (see Fig. 1 and supplementary Tables S5 and S6) but does not alter the overall number of severely compromised babies. Movement is between column B ‘current practice’ to column A ‘base-case’
^dThe improved sensitivity with attentive CTG and altered delivery management does lead to reductions in the risk of severe compromise for the additionally identified babies. Risk reductions of varying magnitudes are modelled. Movement is between column B ’current practice’ to column A ‘% reduction’’ and column C ‘% reduction’
SC severely compromised, NSC non-severely compromised

ISSN: 1478-7547