Cost-effectiveness of the diagnostic work-up for stillbirth
Stillbirth occurs in 5 per 1000 pregnancies, resulting in around 1100 cases yearly in the Netherlands. Finding a cause of stillbirth is essential for the bereavement process of parents and for an estimation of recurrence risks, as well as audit of the specific care provided and evaluation of health care in general. To identify the cause of stillbirth, a battery of tests is available, including pathological examination, testing for infection and haemostatic disorders as well as cytogenetic evaluation. Evidence based guidelines on a generally accepted diagnostic workup for stillbirth are lacking. The UMCG has completed the ZonMw funded ZOBAS study, involving 1025 cases of stillbirth (‘What tests are needed for finding the cause of
antepartum fetal death? Development of a National protocol.’ project: 2100.0082). Although this work showed the clinical accuracy (sensitivity, specificity) of each of these tests in identifying the cause of death, neither their mutual dependence nor their cost-effectiveness have been studied. Classification of stillbirths into a system of causes is essential for further evaluation. Several systems are currently in use and although we developed the TULIP
classification specifically for the ZOBAS study it has been internationally acknowledged. The Tulip classification consists of six main causes: congenital anomaly; placental pathology;
prematurity/immaturity; infection; other and unknown. The cause was defined as the initial, demonstrable pathophysiological entity initiating the chain of events that had irreversibly led to death. The mechanism of death was defined as organ failure not compatible with life, initiated by the cause. Origin of mechanism was defined as explanation of the mechanism. Contributing
factors such as smoking, obesity and small for gestational age were also identified. There is at present a national protocolised work-up for stillbirth based on pathophysiological reasoning. Each test is effective in itself, but the mutual dependence between the tests as well as the cost-effectiveness of the work-up as well the value of the information provided to the patient are not known.
To assess the cost-effectiveness of different diagnostic strategies for finding the cause of stillbirth. This project will answer the following questions:
Data are already available of the diagnostic work-up for 1,025 patients who experienced stillbirth (ZOBAS study). Apart from medical history, diagnostic tests included maternal serum tests, maternal and paternal hemostatic status, maternal and fetal viral and bacterial testing, macroscopic fetal examination, autopsy, MRI, umbilical cord blood testing, karyotyping and
placental examination. The present study will model the costs and effects of these tests and various test-combinations. We will calculate efficiency for detecting causes of stillbirth for each of these diagnostic strategies, by comparing the results of the different stepwise strategies
to reach the final diagnosis. The gold standard will be the diagnosis according to the TULIP classification as already established through a panel, based on full diagnostic work-up.
In the economic evaluation we will estimate costs associated with each test or test battery in the workup. As the aim is to identify an optimal diagnostic strategy that - from a clinical point of view - is non-inferior to the presently used extensive diagnostic workup, the economic evaluation is setup as a cost-minimization analysis. The economic evaluation will be
performed from a societal perspective, and will include direct medical, non-medical and indirect costs.
The study builds on a previous cohort study, there is no formal sample size calculation. We expect, however, that the sample size is large enough to answer the study questions. As the prevalence of abnormal test results varies from 0.2-89%, and in terms of final diagnosis, the prevalence of the main causes of death varies from 0.1 to 31%, this will allow analyses with
18 months. Twelve months for the cost-effectiveness analysis and six months for the report.
dr. J.J.H.M. Erwich, gynecologist, UMCG Groningen
dr. H. Groen, epidemiologist, UMCG Groningen
Prof. dr. P.M. Bossuyt, epidemiologist, AMC Amsterdam
Prof. dr. B.W. Mol, gynecologist and clinical epidemiologist, AMC Amsterdam
drs. J.W. Nijkamp, researcher, UMCG Groningen
ZonMW (www.zonmw.nl) File number: 80-82310-97-12019
For more information about the study, please contact Ellen Nijkamp: firstname.lastname@example.org,