Development of a clinical prediction model for an ordinal outcome: the World Health Organization Multicentre Study of Clinical Signs and Etiological agents of Pneumonia, Sepsis and Meningitis in Young Infants. WHO/ARI Young Infant Multicentre Study Group.

This paper describes the methodologies used to develop a prediction model to assist health workers in developing countries in facing one of the most difficult health problems in all parts of the world: the presentation of an acutely ill young infant. Statistical approaches for developing the clinical prediction model faced at least two major difficulties. First, the number of predictor variables, especially clinical signs and symptoms, is very large, necessitating the use of data reduction techniques that are blinded to the outcome. Second, there is no uniquely accepted continuous outcome measure or final binary diagnostic criterion. For example, the diagnosis of neonatal sepsis is ill-defined. Clinical decision makers must identify infants likely to have positive cultures as well as to grade the severity of illness. In the WHO/ARI Young Infant Multicentre Study we have found an ordinal outcome scale made up of a mixture of laboratory and diagnostic markers to have several clinical advantages as well as to increase the power of tests for risk factors. Such a mixed ordinal scale does present statistical challenges because it may violate constant slope assumptions of ordinal regression models. In this paper we develop and validate an ordinal predictive model after choosing a data reduction technique. We show how ordinality of the outcome is checked against each predictor. We describe new but simple techniques for graphically examining residuals from ordinal logistic models to detect problems with variable transformations as well as to detect non-proportional odds and other lack of fit. We examine an alternative type of ordinal logistic model, the continuation ratio model, to determine if it provides a better fit. We find that it does not but that this model is easily modified to allow the regression coefficients to vary with cut-offs of the response variable. Complex terms in this extended model are penalized to allow only as much complexity as the data will support. We approximate the extended continuation ratio model with a model with fewer terms to allow us to draw a nomogram for obtaining various predictions. The model is validated for calibration and discrimination using the bootstrap. We apply much of the modelling strategy described in Harrell, Lee and Mark (Statist. Med. 15, 361-387 (1998)) for survival analysis, adapting it to ordinal logistic regression and further emphasizing penalized maximum likelihood estimation and data reduction.

Bacterial meningitis: is there a "best" antimicrobial therapy?

The introduction of several cephalosporins into pediatric practice has provided the physician with a number of choices in the treatment of neonatal and childhood meningitis. Adequate studies are available to indicate that these new drugs are as effective as traditional treatments in terms of survival and major neurologic sequelae but it is not known whether the results are worse or better as far as the incidence of more subtle neurologic changes is concerned. The advantages of the cephalosporins in treatment of childhood meningitis are that they permit single drug therapy, the risks of drug toxicity are reduced, and the problems of penicillin-tolerant pneumococci and ampicillin/chloramphenicol-resistant H. influenzae are avoided. When used in the treatment of neonatal disease, the cephalosporins have the advantage of lower toxicity than the aminoglycosides, generally making blood drug level determinations unnecessary, and are effective against strains of bacteria that have become resistant to the latter drugs.

Some aspects of the diagnosis and management of urinary tract infection in children and adolescents.

While areas of disagreement continue to exist about certain details related to the diagnosis and treatment of UTI, it is possible to devise reasonable approaches to these problems that can be shown to be effective and do not harm the patient on a short or long term basis.

The cephalosporin antibiotics in pediatric therapy.

The cephalosporins have been available for clinical use for nearly 20 years and a large number is presently marketed, including drugs with a wide range of different pharmacokinetic and microbiologic properties. While some of these agents have certain specific uses in which they excel, the cephalosporins have not replaced older antibiotics but do provide the physician with a broader range of choices for the treatment of many infections, allowing greater individualization of therapy.

Antibiotic drug therapy in the newborn.

The use of antimicrobial agents in newborn infants has always been fraught with uncertainty. Because of the immature excretory and/or metabolic process of infants, it was recognized that detailed pharmacokinetic studies were necessary before these drugs could be used safely and effectively in young infants, but it was only with experience that it became obvious that toxicity remained relatively unpredictable. Babies were found to be more resistant to certain types of adverse reactions (eg, renal damage by aminoglycosides, bacitracin, methicillin, etc) but more susceptible to others (sulfonamides, chloramphenicol, tetracyclines, nitrofurantoin, and so forth). In many instances, the toxic effects encountered could have been predicted on the basis of experimental data in animals, but in other cases, this was not possible. Thus, the use of new antimicrobial agents in babies may expose them to unpredictable dangers. For that reason, new drugs should only be administered to young infants if they clearly have a therapeutic advantage over older ones. Therapeutic trials must await adequate pharmacokinetic studies, and the investigator must be prepared to follow the treated infants (along with a group of matched control patients) for a sufficient length of time to be certain that any organ damage caused by the drug would have become clinically detectable. In some cases this requires a period of several years.