Cost-effectiveness of screening strategies to detect heart failure in patients with type 2 diabetes

Table 1 Sensitivity, specificity, and costs of heart failure screening strategies in patients with type 2 diabetes of 60 years or older

Parameters	0 No screening	1 EMR symptoms	2 EMR symptoms Physical exam	3 EMR symptoms Physical exam NTproBNP	4 EMR symptoms Physical exam NTproBNP ECG	5 Echo cardio- graphy	Distribution	Source
Sensitivity^a
NYHA 1	0.000	0.250	0.250	0.250	0.500	1.000	Beta	Cohort [5]
NYHA 2	0.000	0.853	0.853	0.879	0.862	1.000	Beta	Cohort [5]
NYHA 3	0.000	0.923	0.949	0.897	0.897	1.000	Beta	Cohort [5]
NYHA 4	0.000	1.000	1.000	1.000	1.000	1.000	Beta	Cohort [5]
Specificity	1.000	0.610	0.617	0.652	0.676	1.000	Beta	Cohort [5]
Screening
GP	€0.00^b	€6.39	€15.17	€36.67	€61.77	€0.00^c	Gamma	Dutch tariff [24–26]
Echocardiography	€169.38	€169.38	€169.38	€169.38	€169.38	€169.38	Gamma	Dutch tariff [24–26]
ECG stress test	€94.75	€94.75	€94.75	€94.75	€94.75	€94.75	Gamma	Dutch tariff [24–26]

For each screening strategy NYHA-specific sensitivities were calculated and for each of these sensitivities a beta-distribution was specified with the true positives and total positives as parameters. Similarly, beta distributions were assigned to the specificities of each of the screening strategies. Gamma distributions with parameters using a variance equal to the mean were used for the costs
EMR Electronic Medical Record, GP general physician
^aIn general, more extensive screening strategies yielded higher sensitivity and specificity at higher costs, except for NYHA 2 when adding ECG and for NYHA 3 when adding NTproBNP and/or ECG
^bStrategy costs in case of no screening are kept fixed at €0 in the sensitivity analyses
^cThere are no GP costs here as everyone is, after their regular diabetes checkup, directly sent for echocardiography

ISSN: 1475-2840