Lulu Muhe

Case-control study of the role of nutritional rickets in the risk of developing pneumonia in Ethiopian children.

BACKGROUND Pneumonia is the most important cause of morbidity and mortality in children aged under 5 years worldwide. Studies in developing countries have suggested an association between nutritional rickets and pneumonia. Since both nutritional rickets and pneumonia are common in Ethiopia, we did a case-control study to determine the role of nutritional rickets in the development of pneumonia. METHODS Cases were children younger than 5 years admitted to the Ethio-Swedish Childrens Hospital during a 5-year period with a diagnosis of pneumonia (n = 521), but data were incomplete for 21 of these and they were not included. Controls (n = 500) were matched for admission within 3 months of cases and age within 3 months and had no evidence of pneumonia. Nutritional, demographic, and clinical and radiographic data for rickets and pneumonia were collected. Matched odd ratios and logistic regression were used to test the significance of the association of rickets and pneumonia. FINDINGS Rickets was present in 210 of 500 cases compared with 20 of 500 controls (odds ratio 22.11). There were significant differences between cases and controls for family size, birth order, crowding, and months of exclusive breastfeeding (p < 0.05). After correction for these confounding factors by logistic regression, there was still a 13-fold higher incidence of rickets among children with pneumonia than among controls (13.37 [95% CI 8.08-24.22], p < 0.001). INTERPRETATION Vitamin D or calcium deficiency may be important predisposing factors for pneumonia in children aged under 5 years in developing countries. Efforts to prevent vitamin D deficiency or calcium supplementation may result in significant reductions in morbidity and mortality from pneumonia in these children.

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

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.

Predictors of neonatal sepsis in developing countries.

Background. Neonatal infections are a major cause of death worldwide. Simple procedures for identifying infants with infection that need referral for treatment are therefore of major public health importance. Methods. We investigated 3303 infants <2 months of age presenting with illness to health facilities in Ethiopia, The Gambia, Papua New Guinea and The Philippines, using a standardized approach. Historical factors and clinical signs predicting sepsis, meningitis, hypoxemia, deaths and an ordinal scale indicating severe disease were investigated by logistic regression, and the performance of simple combination rules was explored. Results. In multivariable analysis, reduced feeding ability, no spontaneous movement, temperature >38°C, being drowsy/unconscious, a history of a feeding problem, history of change in activity, being agitated, the presence of lower chest wall indrawing, respiratory rate >60 breaths/min, grunting, cyanosis, a history of convulsions, a bulging fontanel and slow digital capillary refill were independent predictors of severe disease. The presence of any 1 of these 14 signs had a sensitivity for severe disease (defined as sepsis, meningitis, hypoxemia, or radiologically proven pneumonia) of 87% and a specificity of 54%. More stringent combinations, such as demanding 2 signs from the list, resulted in a considerable loss of sensitivity. By contrast only slight loss of sensitivity and considerable gain of specificity resulted from reducing the list to 9 signs. Requiring the presence of fever and any other sign produced a diagnostic rule with extremely low sensitivity (25%). Conclusions. Physical signs can be used to identify young infants at risk of severe disease, however with limited specificity, resulting in large numbers of unnecessary referrals. Further studies are required to validate and refine the prediction of severe disease, especially in the first week of life, but there appear to be limits on the accuracy of prediction that is achievable.

Refining a probabilistic model for interpreting verbal autopsy data

Objective: To build on the previously reported development of a Bayesian probabilistic model for interpreting verbal autopsy (VA) data, attempting to improve the models performance in determining cause of death and to reassess it. Design: An expert group of clinicians, coming from a wide range geographically and in terms of specialization, was convened. Over a four-day period the content of the previous probabilistic model was reviewed in detail and adjusted as necessary to reflect the group consensus. The revised model was tested with the same 189 VA cases from Vietnam, assessed by two local clinicians, that were used to test the preliminary model. Results: The revised model contained a total of 104 indicators that could be derived from VA data and 34 possible causes of death. When applied to the 189 Vietnamese cases, 142 (75.1%) achieved concordance between the models output and the previous clinical consensus. The remaining 47 cases (24.9%) were presented to a further independent clinician for reassessment. As a result, consensus between clinical reassessment and the models output was achieved in 28 cases (14.8%); clinical reassessment and the original clinical opinion agreed in 8 cases (4.2%), and in the remaining 11 cases (5.8%) clinical reassessment, the model, and the original clinical opinion all differed. Thus overall the model was considered to have performed well in 170 cases (89.9%). Conclusions: This approach to interpreting VA data continues to show promise. The next steps will be to evaluate it against other sources of VA data. The expert group approach to determining the required probability base seems to have been a productive one in improving the performance of the model.

Revising the WHO verbal autopsy instrument to facilitate routine cause-of-death monitoring

Objective Verbal autopsy (VA) is a systematic approach for determining causes of death (CoD) in populations without routine medical certification. It has mainly been used in research contexts and involved relatively lengthy interviews. Our objective here is to describe the process used to shorten, simplify, and standardise the VA process to make it feasible for application on a larger scale such as in routine civil registration and vital statistics (CRVS) systems. Methods A literature review of existing VA instruments was undertaken. The World Health Organization (WHO) then facilitated an international consultation process to review experiences with existing VA instruments, including those from WHO, the Demographic Evaluation of Populations and their Health in Developing Countries (INDEPTH) Network, InterVA, and the Population Health Metrics Research Consortium (PHMRC). In an expert meeting, consideration was given to formulating a workable VA CoD list [with mapping to the International Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) CoD] and to the viability and utility of existing VA interview questions, with a view to undertaking systematic simplification. Findings A revised VA CoD list was compiled enabling mapping of all ICD-10 CoD onto 62 VA cause categories, chosen on the grounds of public health significance as well as potential for ascertainment from VA. A set of 221 indicators for inclusion in the revised VA instrument was developed on the basis of accumulated experience, with appropriate skip patterns for various population sub-groups. The duration of a VA interview was reduced by about 40% with this new approach. Conclusions The revised VA instrument resulting from this consultation process is presented here as a means of making it available for widespread use and evaluation. It is envisaged that this will be used in conjunction with automated models for assigning CoD from VA data, rather than involving physicians.Objective Verbal autopsy (VA) is a systematic approach for determining causes of death (CoD) in populations without routine medical certification. It has mainly been used in research contexts and involved relatively lengthy interviews. Our objective here is to describe the process used to shorten, simplify, and standardise the VA process to make it feasible for application on a larger scale such as in routine civil registration and vital statistics (CRVS) systems. Methods A literature review of existing VA instruments was undertaken. The World Health Organization (WHO) then facilitated an international consultation process to review experiences with existing VA instruments, including those from WHO, the Demographic Evaluation of Populations and their Health in Developing Countries (INDEPTH) Network, InterVA, and the Population Health Metrics Research Consortium (PHMRC). In an expert meeting, consideration was given to formulating a workable VA CoD list [with mapping to the International Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) CoD] and to the viability and utility of existing VA interview questions, with a view to undertaking systematic simplification. Findings A revised VA CoD list was compiled enabling mapping of all ICD-10 CoD onto 62 VA cause categories, chosen on the grounds of public health significance as well as potential for ascertainment from VA. A set of 221 indicators for inclusion in the revised VA instrument was developed on the basis of accumulated experience, with appropriate skip patterns for various population sub-groups. The duration of a VA interview was reduced by about 40% with this new approach. Conclusions The revised VA instrument resulting from this consultation process is presented here as a means of making it available for widespread use and evaluation. It is envisaged that this will be used in conjunction with automated models for assigning CoD from VA data, rather than involving physicians.

Child health in complex emergencies

Coordinated and effective interventions are critical for relief efforts to be successful in addressing the health needs of children in situations of armed conflict, population displacement, and/or food insecurity. We reviewed published literature and surveyed international relief organizations engaged in child health activities in complex emergencies. Our aim was to identify research needs and improve guidelines for the care of children. Much of the literature details the burden of disease and the causes of morbidity and mortality; few interventional studies have been published. Surveys of international relief organizations showed that most use World Health Organization (WHO), United Nations Childrens Fund (UNICEF), and ministry of health guidelines designed for use in stable situations. Organizations were least likely to have formal guidelines on the management of asphyxia, prematurity, and infection in neonates; diagnosis and management of children with human immunodeficiency virus (HIV) infection; active case-finding and treatment of tuberculosis; paediatric trauma; and the diagnosis and management of mental-health problems in children. Guidelines often are not adapted to the different types of health-care workers who provide care in complex emergencies. Evidence-based, locally adapted guidelines for the care of children in complex emergencies should be adopted by ministries of health, supported by WHO and UNICEF, and disseminated to international relief organizations to ensure appropriate, effective, and uniform care.

Etiology of pneumonia, sepsis and meningitis in infants younger than three months of age in Ethiopia.

METHODS Within a multicenter study coordinated by WHO, an investigation of the etiologic agents of pneumonia, sepsis and meningitis was performed among infants younger than 3 months of age seen at the Ethio-Swedish Childrens Hospital in Addis Ababa for a period of 2 years. Of the 816 infants enrolled 405 had clinical indications for investigation. RESULTS There were a total of 41 isolates from blood cultures from 40 infants. The study showed that the traditionally known acute respiratory infection pathogen Streptococcus pneumoniae was most common in this extended neonatal age group, found in 10 of 41 blood isolates. Streptococcus pyogenes was a common pathogen in this setting (9 of 41 blood isolates), whereas Salmonella group B was found in 5 of 41 isolates. Streptococcus agalactiae, which is a common pathogen in developed countries, was absent. A study of the susceptibility pattern of these organisms suggests that a combination of ampicillin with an aminoglycoside is adequate for initial treatment of these serious bacterial infections, but the combination is not optimal for the treatment of Salmonella infections. Among 202 infants on whom immunofluorescent antibody studies for viruses were performed based on nasopharyngeal aspirates, respiratory syncytial virus was found in 57 (28%) infants, and Chlamydia trachomatis was isolated in 32 (15.8%) of 203 infants.

Clinical algorithm for malaria during low and high transmission seasons

OBJECTIVES To assess the proportion of children with febrile disease who suffer from malaria and to identify clinical signs and symptoms that predict malaria during low and high transmission seasons. STUDY DESIGN 2490 children aged 2 to 59 months presenting to a health centre in rural Ethiopia with fever had their history documented and the following investigations: clinical examination, diagnosis, haemoglobin measurement, and a blood smear for malaria parasites. Clinical findings were related to the presence of malaria parasitaemia. RESULTS Malaria contributed to 5.9% of all febrile cases from January to April and to 30.3% during the rest of the year. Prediction of malaria was improved by simple combinations of a few signs and symptoms. Fever with a history of previous malarial attack or absence of cough or a finding of pallor gave a sensitivity of 83% in the high risk season and 75% in the low risk season, with corresponding specificities of 51% and 60%; fever with a previous malaria attack or pallor or splenomegaly had sensitivities of 80% and 69% and specificities of 65% and 81% in high and low risk settings, respectively. CONCLUSION Better clinical definitions are possible for low malaria settings when microscopic examination cannot be done. Health workers should be trained to detect pallor and splenomegaly because these two signs improve the specificity for malaria.

Evaluation of clinical pallor in the identification and treatment of children with moderate and severe anaemia

Summary background  Anaemia from malaria is a common problem in developing countries. Blood transfusion in developing countries is available in few hospitals. Children who are severely anaemic and may require urgent blood transfusion usually present to peripheral first‐level health facilities from where they must be referred to hospitals. Since most peripheral facilities do not determine haemoglobin levels, the decision on referral has to be made on clinical grounds.

Pneumococcal and Haemophilus influenzae meningitis in a children's hospital in Ethiopia: serotypes and susceptibility patterns

Summary Streptococcus pneumoniae and Haemophilus influenzae are responsible for most pyogenic meningitis cases in children in Ethiopia. Resistance of S. pneumoniae and H. influenzae to penicillin and chloramphenicol respectively has been reported globally. Resistance has been related to specific serotypes of S. pneumoniae or to beta‐lactamase‐producing H. influenzae strains. This study describes the serotypes/ serogroups and susceptibility pattern of the two organisms causing meningitis in Ethiopian children. There were 120 cases of meningitis caused by S. pneumoniae (46) and H. influenzae (74) over a period of 3 years (1993–95). Nineteen children died from pneumococcal and 28 from haemophilus meningitis. Penicillin‐resistant pneumococcal meningitis (4/8 = 50%) caused a greater mortality rate than penicillin‐susceptible pneumococcal meningitis (15/38 = 39%). Common serotypes accounting for 76% of S. pneumoniae were type 14, 19F, 20, 1, 18 and 5; and serotypes 14, 19F and 7 (accounting for 17% of strains) showed intermediate resistance to penicillin G. 97% of the H. influenzae isolates were type b, and in only two cases beta‐lactamase‐producing. 72% of isolates of the S. pneumoniae we identified belong to serotypes preventable by a 9‐valent vaccine. Our study highlighs the possibility of resistant pyogenic meningitis in children in Ethiopia due to emerging resistant strains of S. pneumoniae and H. influenzae isolates.