This study used two complementary approaches to investigate the scope of manipulations of drugs to achieve the prescribed dose in neonatal and paediatric practice. The results provide an overview of the issues that arise when manipulations are undertaken. Although manipulations were reported more frequently from more specialist areas, such as neonatal and paediatric intensive care areas, they occurred in all paediatric in-patient areas. The manipulations identified involved a wide range of drugs.
Patient preference was the sole reason for 15% of the manipulations reported in the questionnaire. This was also found during the observational study. For example one child preferred to take half a tablet even though a liquid formulation was available because taking the liquid involved a large volume of a liquid whose taste they did not like.
The predominant concern noted by questionnaire respondents was whether the manipulated medicine would provide an accurate dose. These concerns have also been described previously with halving or quartering tablets
[8, 15–17]. The need for caution when splitting tablets with short half-lives or low therapeutic indices
 has been highlighted.
A novel finding here is that sizable number of reported and observed manipulations did not correspond to half or quarter of the dosage form. The accuracy of dividing tablets into other fractions is even more unclear that halves or quarters. The potential lack of accuracy during manipulations implied by a prescription may mean that the actual dose delivered to the patient is not known. It seems likely that prescribers are often unaware of the dosage form (and strength/concentration) that will be used to administer the dose required or the potential for inaccuracy that arises from their prescribed doses.
This study also found that tablets are manipulated by dispersion in liquid and measurement of a proportion. Highly variable dosing may occur when insoluble drugs are dispersed in water
. Dispersible tablets can also yield inconsistent doses when withdrawn from different depths of the container
Manipulations with the aim of achieving the required dose are also undertaken with capsules, sachets, suppositories, nebulisers, enemas and transdermal patches. The manipulation of these other dosage forms has not been previously investigated. The adverse effects associated with administering manipulated drugs, of any dosage form, are unknown. There is a need to conduct research about high impact manipulations, such as those that involve an active ingredient with a narrow therapeutic index or where the physicochemical properties of the active ingredient may lead to significant changes in bioavailability following a manipulation.
This study was designed to scope the nature and occurrence of manipulations and identify priorities for further research. As such, we used a purposive sampling approach for maximum variability for the observational study and a sample of convenience for the questionnaire study. This precludes quantitative generalisation of our results. The supplementary quantitative study estimated that manipulations would be required in approximately 10% of drug administration episodes in the clinical areas studied in the observational study. Nevertheless, our methods were sufficiently fine-grained and pragmatic to yield data that indicate the need for action. The dose of commonly used medicines was usually calculated from body weight and dose/kg according to standard practice
 without rounding to take account of dosage form/strength. Manipulation of the dosage form was undertaken with the aim of administering the precise dose. If other centres are using standard sources, such as BNFC, they are also likely to be conducting manipulations. The nature of the issues we have identified would not change if data were collected from more centres. We note that our methods gave complementary results. Any further research on the nature of the problem will need approaches that provide more than one perspective.
Our results demonstrate that manipulations of dosage forms are integral to paediatric practice. However, it is not entirely clear who is responsible for manipulations. There are two relevant situations, firstly, when the manipulation is conducted because a suitable dosage form is not available. This situation may arise because there is no suitable dosage form on the market, because a purchasing decision has prevented a suitable dosage form from being issued from Pharmacy or because a suitable dosage form is not available on the ward at the point of need (e.g. at night when Pharmacy staff may not be available). Secondly, the nurse makes a professional judgment to meet a patient’s preference when a dosage form is available. We have developed generic guidelines about the manipulation of dosage forms in children (available from the authors). We suggest that paediatric units should have policies and procedures to cover manipulation of dosage forms. This should ensure that there is corporate responsibility for manipulations. For example, purchasing decisions need to take account of the manipulations that may result from the choice of available products. Nurses should have access to the relevant information and the support from their institution to make professional judgments about manipulations, many of which are done on off-label or unlicensed medicines.