Page 184 - 16Neonatal Jaundice_compressed

Appendix C: Economic evaluation of testing strategies for hyperbilirubinaemia

3. routine serum bilirubin testing prior to discharge with follow-up within 2 days of discharge if
the bilirubin measurement is greater than the 40th percentile value on the nomogram
4. routine transcutaneous bilirubin testing prior to discharge with the percentile value on the
34
nomogram developed by Bhuthani et al. guiding decisions about the need for serum
bilirubin testing before discharge and subsequent follow-up.
In strategies 2–4 the threshold for laboratory testing is lowered in recognition of the unreliability
of visual estimation of bilirubin.

The authors estimated that with current practice 2.3% of infants would receive phototherapy
compared with 8.1%, 5.6%, and 7.8% for strategies 2–4, respectively. The savings from an
averted kernicterus case were assumed to be $921,000 (at 2003 prices). The authors assumed
that strategies 2–4 would be equally effective at preventing kernicterus cases and their primary
outcome was to estimate the cost per kernicterus case averted in each of those strategies, with a
relative risk reduction of 0.7, an assumption made in the absence of data.
The results suggested that routine serum bilirubin was the cheapest strategy at $5.75 million per
case averted. Using transcutaneous bilirubin meters prior to discharge gave a cost per averted
kernicterus case of $9.19 million. Universal screening was the most expensive strategy, with a
cost per case averted of $10.32 million. One-way sensitivity analysis suggested that the
magnitude of these costs were sensitive to incidence of kernicterus although this did not alter
the ranking of the strategies in terms of costs. The authors concluded that their data suggested
that it would be premature to implement large-scale routine bilirubin screening.

However, there are often difficulties in generalising the result of an economic evaluation from
one setting to another. In particular, the US hospital charges seem unlikely to accurately reflect
NHS costs and the costs of a kernicterus case may have been understated. Furthermore, the
GDGs remit did not cover screening. Therefore, a de novo economic model was developed to
reflect the UK context and to enable the GDG to consider cost-effectiveness issues in making
their recommendations.

C.3 Background to the economic evaluation

Kernicterus is a largely preventable disease if severe hyperbilirubinaemia is identified early and
promptly treated (using phototherapy or, for more acute cases, exchange transfusion). Therefore,
early identification of raised (or rapidly rising) bilirubin levels is the key to reducing severe
morbidity.
There are studies which demonstrate that more intensive monitoring reduces the need for
exchange transfusions. Evidence from the USA reports that during the 1970s kernicterus was
practically eradicated, which was probably due to the liberal use of phototherapy. 235 The
disease re-emerged in the 1990s, largely among babies cared for in the home environment in
the neonatal period, often with limited medical supervision during the first week after birth. 235
Kernicterus has fallen again in the USA since the adoption of the 1994 American Academy of
Pediatrics (AAP) guidelines; 123 estimates are that the rate has fallen from 5.1 per 100 000 in
1988 to 1.5 per 100 000. 124

In the UK, babies are discharged earlier and are monitored less often than in previous
decades. 236 Reduced contact with experienced midwives and reliance on intermittent visual
examination to assess bilirubin levels may be one of the reasons for the failure to detect babies
with significantly elevated serum bilirubin levels. A newborn baby might only be visited once
by a midwife in the postnatal period if there are no risk factors, although the norm is currently
around two or three visits in the first week. Visual examination by a midwife to assess for
jaundice during these postnatal visits is currently the standard of care, with a small proportion of
these jaundiced children being subjected to a total serum bilirubin (TSB) blood test based on
clinical visual assessment of the level of bilirubin. This is known to be unreliable. There is
strong evidence that visual examination alone cannot be used to assess the level of bilirubin in a
baby (see Chapter 5 on recognition). The inaccuracy of visual assessment for the detection of
bilirubin levels, particularly in babies with dark skin tones, is likely to be a major factor
responsible for the late presentation of babies with significant hyperbilirubinaemia. Therefore a

153