Global Research Review: Issue 12

Published on May 16, 2014
On behalf of the Global Emergency Medicine Literature Review Group

CHINA_ Developing injury prevention strategies for developing countries requires knowledge that urban and rural areas have distinct injury patterns.

Liu Q, Zhang L, Li J, Zuo D, Kong D, Shen X, Guo Y, and Zhang Q. The gap in injury mortality rates between urban and rural residents of Hubei province, China. BMC Public Health 2012; 12:180-190.

Building on previous studies that exposed rural-urban injury disparities but were limited by detailed (gender, age and injury category) demographic data collection, the current study uses data from the Hubei Province Disease Surveillance (DSP) system to demonstrate not only clear urban-rural injury mortality rate disparities but also differences in injury type, age and gender. The DSP collected detailed demographic and injury mortality data from a representative sample of the population (approximately 6 million people) from government-designated rural and urban areas using a multi-stage cluster probability sampling. Health officials checked district (urban) or municipal (rural) reported deaths daily and deaths that occurred at home were corroborated with a standardized verbal autopsy or clinical evidence. For hospital deaths, health officials verified death certificate details. These officials entered cause of death in to the database weekly, and coded cause of death based on the International Classification of Disease-10th Revision (ICD-10). The study calculated crude and adjusted injury mortality rates and 95% confidence intervals; the Chi-square or Fischer’s exact test examined rural-urban differences with a significance of p<0.01. Injury death rates for both sexes were approximately two-fold higher in rural than urban areas with crude and adjusted rates reaching significance. Overall, age-adjusted death rates for males and females, suicide, traffic-related injuries, drowning and crush injuries were significantly higher in rural areas. Furthermore, the rural residents >55years had an injury death rate three times their urban counterparts and those >65years had higher injury death rates for suicide, traffic-related injuries and drowning. Death rates for falls, poisoning, and suffocation did not differ between urban and rural residents.

This study confirms the results of previous studies that found higher injury mortality rates in rural compared to urban areas in developing and developed countries. The striking differences in injury death rates from suicide in the rural elderly versus urban elderly population, and the traffic-related injury mortality rate in rural compared to urban residents are important findings of this study. Increased risk taking behaviors, access to pesticides, decreased social welfare in rural areas, and a large urban-rural income gap are proposed by the authors as possible reasons for these disparities. This study is limited by unreported death rate of 15% and 13% in rural and urban areas respectively, possible misclassification of deaths (specifically in the area of falls), and sparse demographic data (past medical history, occupational history or socio-economic data). However, the results from this study have important implications for injury prevention policy in China. -KP, TB

GHANA_ A new decision tree for the diagnosis of P. Falciparum

Vinnemeier CD, Schwarz NG, Sarpong N, et al. Predictive value of fever and palmar pallor for P. falciparum parasitaemia in children from an endemic area. PLoS One. 2012;7(5):e36678.

This article aimed to create a clinical decision algorithm for the diagnosis of P. falciparum malaria in endemic areas. The study evaluated all children between 2-60 months of age who attended an outpatient department in Ghana. They obtained data regarding over 30 clinical symptoms, a blood count, and a thick smear. A Classification and Regression Tree (CART) model was created to create a decision tree that could be utilized to predict malaria. Palmar pallor was the most indicative with an Odds Radio (OR) of 3.06 in children, while body temperature had an OR of 2.82 and reported fever had an OR of 4.62. Two CART models were created for children 2-12months and older children including these variables among others. The CART model in younger children had a high sensitivity (97.2%), but a low specificity (22.2%) compared to the current Integrated Management of Infectious Disease (IMCI) guidelines with sensitivity and specificities of 6.7% and 99.6% respectively. In the older population, the CART model was 37.7% sensitive and 91.4% specific, compared to the IMCI-model with 55.6% and 73.4%. The CART-model had higher specificities and positive predictive values, while the IMCI-model had higher sensitivities and negative predictive values.

The main strength of the study includes the large number of variables analyzed in the preliminary analysis to help create the CART model. Additionally, the gold standard of malaria smears in all patients provided an objective measure of parasitemia. The large sample size, prospective analysis, and statistical analysis all increased the validity of the study. However, limitations include the CART model which is inherently a controversial tool and the use of subjective clinical indicators which might be biased by the providers recording the data. Nonetheless, this study further validates the IMCI guidelines and highlights the importance of clinical signs such as palmar pallor which tested well on univariate analysis. -PM, MF


KP: Kimberly Pringle, MD

TB: Torben K. Becker, MD

PM: Payal Modi, MD

MF: Mark Foran, MD, MPH

e.g. Global, Research, or India
e.g. Features, Opinion, or Research
Browse by Category
    Most Popular
      Download Latest Issue
      Issue 19