The purpose of licensing is to ensure that medicines are examined for safety, effi-cacy and quality. Most medicines administered to adults have a product licence that outlines the particular indication, dose and route of administration. However, many medicines used for children are not licensed (i.e., do not have a product license or marketing authorization), or are used ‘off label’ (i.e., used outside the terms of the product licence for reasons such as different doses, indications, ages and routes of administration)(1). This means that use of such medicines is reliant on ‘educated guesswork’ and the ‘experiences of peers’; it is not supported by hard scientific fact. This creates a therapeutic dilemma since on the one hand children might be denied an effective medicine because it is not licensed, but on the other, we cannot be confident that we are administering a medicine which is safe and effective. Is this an acceptable way of regularly treating children? Most would agree that it is not, but we continue to do it in everyday practice where children require drug therapy.

Despite considerable concern about the use of unlicensed and ‘off label’ drugs in newborns and children, there is little information available on the extent to which these types of treatment are used. The published data(1,2) from pediatric and neonatal intensive care units are scanty and do not provide a clear idea because of the methodological bias. A recent survey of five European hospitals, which analyzed 2262 prescriptions administered to 624 children found that almost half of all drug prescriptions (1036; 46%) were either unlicensed or off label. Of these 1036, 872 were off label and 164 were unlicensed. On the whole, over half of the patients (421; 67%) received an unlicensed or off label prescription(3).

Children suffer from many acute and chronic illnesses which can be of concern for parents, nurses, and pediatricians, and the urge to act is understandable. Neonatology appears to be a particularly vulnerable field which has seen the rise and fall of a number of therapies. William Silverman(4) in his account of ‘modern parable’ classified such proclaimed therapies in the development of perinatal medicine into those that ‘lead to standard practice’, those that ‘lead to disaster’, and those that ‘mislead into fruitless byways’. The use of supplement of oxygen in 1950s for respiratory distress in preterm babies was a ‘disaster’ that contributed to an epidemic of retinopathy and blindness. This practice was only abandoned after controlled clinical trials showed that despite its life-saving potential, unrestricted use of oxygen could also cause permanent visual impairment. The risk is now minimized with modern practice of prescribing oxygen in a strictly controlled manner. There are many similar cases and anxiety remains for a number of commonly used drugs in current practice. Recent examples of concerns among neonatologists arise from adverse data related to use of Vitamin K injections showing increased incidence of leukemia(5), and Dexamethasone for chronic lung disease where the incidence of Cerebral Palsy among the recipients of this drug appears to be more than twice than in controls(6). Similarly use of Indomethacin has been shown to be effective in reducing the incidence of intra-ventricular hemorrhage but recent data from NIH (Personal Communication - Henrietta Bada, University of Tennessee, USA) suggests that the incidence of necrotizing enterocolitis and intestinal perforation is significantly higher in the treated group. Methylxanthines (Theophylline and Caffeine), one of the most commonly used drugs in the neonatal intensive care units for treatment of apnea of prematurity is another drug which lacks data regarding short term and long term safety and efficacy. If any-thing, Methylxanthines may have the potential of worsening hypoxic brain damage(7). Thus there is an urgent need to tease out these issues about short and long term risk-benefit analysis of a number of common drugs, in order to arrive at a safe decision regarding their uses. For the new treatments, it is necessary that they are tested with randomized trials which are large enough and followed up for sufficiently long time to provide a robust data on all clini-cally important end points.

On the other hand, not all off label drug use may be inappropriate. For example, drug toxicity is more likely with aminoglycosides if they are used in neonates as recommended by the manufacturers at interval of 8-12 hours rather than at longer intervals. Similarly, there cannot be many pediatricians, specially neonatologists, who are not aware that babies do experience pain and they show both short and long-term adverse physiological and behavioral responses after repeated exposure to painful stimuli. Although, further research with the effectiveness and risks of analgesic drugs in very sick preterm infants is needed, there is enough scientific evidence to justify the use of certain drugs with which clinicians have sufficient experience(8).

While this dilemma persists there are a few simple general rules that ought to govern prescribing decision. The clinicians should consider whether the patient has a condition that merits treatment, whether there is a treatment that might be expected to help, and whether the benefits of prescribing for this individual patient outweigh the risk. Before prescribing any medicine for any patient, he or she should think carefully about the potential good and the possible harm that might ensue (i.e., the benefit-risk ratio) in the light of available evidence.

The prescriber’s decision may have to be defended against a claim for negligence, and one defence is to show that the action was reasonable in the circumstances, and that a reputable group of peers would have behaved in the same way. It is also possible to draw a hierarchy of degrees of reasonableness. The more dangerous the medicine, and the flimsier the evidence on which the treatment is based, the more difficult it will be to justify a decision to prescribe it. On the contrary, it may be justifiable to prescribe certain drugs for an unlicensed condition. Examples include the prescribing of generic formulations (for which indications are not described), the use of well-established drugs for proven but unlicensed indications, and the use of drugs for conditions for which there are no other treatments (even in absence of strong evidence). Considered and reasonable decisions, especially when recorded in the casenotes, will protect both the patient and the doctor.

The interest in this subject in the UK and across Europe has resulted in the publication of European guidelines on the clinical investigation of medicinal products in children(9). The document states that ‘there is a responsibility shared by applicant (pharma-ceutical companies) and the competent authorities to ensure that children have timely access to safe and effective medicines which have accurate and scientifically justified prescribing information’. The guidelines also describe the nature of good clinical trials in children, including objectives, appropriate methods and design, subject selection, and criteria for efficacy and safety. The document does recognize the widespread concerns regarding performing clinical trials in children but also reminds that trials can be performed without undue distress if designed and carried out by investigators experienced in caring for children. The days of data sheets stating a curt ‘not recommended for use in children’ should then soon be over, to the great relief of all those concerned with the treatment of children.

Some will argue that for most of the common drugs–falling under unlicensed or off license category–there may not be even a need for further research, or the necessity to tighten their prescribing. What is needed is sensible, sustained, and constructive dialogue between the profession, the licensing authori-ties, and the manufacturers, to get drug sheets revised at regular intervals, so that they reflect all the additional information that becomes available in the years after the product first comes on the market(10).

The message is, that it is up to the pro-fession to start this ball rolling, and clearly the actions to resolve this problem have to be formulated for the individual country by its own people. The Indian problem will require an Indian solution. The recognition of the importance of studying drugs for use in children and the establishment of any initiative in this regard is to be applauded.

Sunil Sinha,
Director of Neonatology,
South Cleveland Hospital,
Middlesbrough TS4 3BW, UK.
E-mail: [email protected]

Funding : None.
Competing interests: None stated.

1. Conroy S, Mcintyre J, Choornara I. Unlicensed and off label drug use in neonates. Arch Dis Child 1999; 80: F142-F145.

2. Turner S, Gill A, Nunn T, Hewitt B, Choonara I. Use of ‘off-label’ and unlicensed drugs in pediatric intensive care unit. Lancet 1996; 347: 549-550.

3. Conroy C, Choonara I, Impicciator P, Mohn A, Arnell H, Rane A, et al. Survey of unlicensed and off lable drug use in pediatric wards in European countries. BMJ 2000; 320: 79-82.

4. Silverman WA. Retrolental fibroplasia: A modern parable. New York, Grune and Stratton. 1980; p 85.

5. Parker L, Cole M, Craft AW, Hey EN. Neonatal Vitamin K administration and childhood cancer in the north of England: Retrospective case-control study. BMJ 1998; 316: 189-193.

6. Tarnow-Modi WA, Mitra A. Postnatal dexa-methasone in preterm infants. BMJ 1999; 319: 1385-1386.

7. Schmidt B. Methylxanthine therapy in preterm infants: Sound practice, disaster, or fruitless byway? J Pediatr 1999; 135: 526-528.

8. Sabrine N, Sinha S. Pain in neonates. Lancet 2000; 355: 392-393.

9. European Agency for the Evaluation of Medicinal Products. Note for guidance on clinical investigation of medicinal product in children. EAEMP, London, 1997.

10. Hey E. Unlicensed and off label drug use in neonates. Arch Dis Child 2000; 82: F77.