Neonatal screening

BMJ 1996; 312 doi: https://doi.org/10.1136/bmj.312.7024.182b (Published 20 January 1996) Cite this as: BMJ 1996;312:182

Performance is hard to monitor

  1. Francesco P Cappuccio,
  2. Matthew Hickman,
  3. Maggie Barker
  1. Senior registrar in public health medicine Epidemiologist Consultant in public health medicine Brent and Harrow Health Authority, Harrovian Business Village, Harrow, Middlesex HA1 3EX

    EDITOR,—We agree with Allison Streetly and colleagues that any additions to the established neonatal screening programme for phenylketonuria and congenital hypothyroidism should be assessed for effectiveness, including impact on the performance of the overall programme.1 But we would go further and suggest that such an assessment would prove difficult because of inherent problems with the existing neonatal screening programme, which need to be addressed first.

    Streetly and colleagues assume that 99% of babies are tested for phenylketonuria and congenital hypothyroidism, though the figure may be lower, at least in London, where coverage with Guthrie testing varies between health authorities2 3 and by ethnic group.4 Unlike with other population programmes, such as screening of women for cervical or breast cancer, immunisation, and vaccination, routine statistics on the proportion of the target population tested and a system of following up missed cases are not available.

    A local review of neonatal screening highlighted some of the problems of monitoring the performance of neonatal screening. Firstly, the community midwives of the woman's district of residence who carry out the tests (usually on day 7 after birth) and send the Guthrie cards to the laboratory also receive the results. But these arrive after day 10, when the midwife is no longer in touch with the mother. There is no mechanism for transferring the results to the primary health care team (including the health visitor), which by then is clinically responsible. Secondly, there are no arrangements for the results to be entered into the child health information system, which could be accessed by the primary health care team, and for missing cases to be identified. Instead, the community midwives work from manual registers or log books of women resident in their area who have recently given birth; these are not routinely matched against alternative records, such as the child health information system, to check that all eligible babies are tested. Thirdly, positive results are notified by the laboratory direct to the general practitioner, but no provision is made for follow up of the management and outcome of these cases.

    It is unlikely that our findings are peculiar to a single district. The comparative rarity of congenital hypothyroidism and phenylketonuria (about 1 in 10000) may perhaps explain why we are unaware of any cases that have been missed or not followed up properly. Through district committees (maternity service liaison and child health surveillance) and through the contracting process we are clarifying responsibilities for the neonatal programme and will require a system for monitoring coverage. It would be ironic if the “loss of regional responsibility for overview of programmes”1 resulted in better organisation of the neonatal screening programme. Equally, unless all districts address these fundamental problems the service may become more fragmented; thus regional guidance on how to establish an appropriate monitoring framework is necessary.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    View Abstract

    Log in

    Log in through your institution


    * For online subscription