Information technology and health-care: Part ll appraisal of quality and relevance of medical information on internet

In the past few years, there has been a tremendous growth in the number of medical resources on the internet, creating new opportunities to improve decisions making among health professionals and enhancing communication in the delivery of health services. The enormity of resources has however generated some problems. Novice users may find it difficult to get quality and relevant information and may even get misled by erroneous information. This paper presents few guidelines for the users to evaluate information that they retrieve through internet. Aspects such as context, content, accessibility and methods of evaluation are presented including upstream and downstream filtering

Burn-injury is preventable: an analysis of 716 cases in burns unit

Burn trauma constitutes one of the leading causes of morbidity and mortality in our society. In order to define a target population for preventive programmes a retrospective study of 716 acutely burned patients was conducted at the burns unit of Civil Hospital Karachi from January to December 1996. The findings revealed that common victims of burn injury were females with a peak age range between 21-30 years and a mean total body surface area [TBSA] burn of 31.3%. The female to male ratio was 1.06:1 Majority [83%] of burn accidents occurred in domestic settings kitchen being the most common [53.8%] place. Inflammable liquids were the most frequent [32.7%] source of burns. The most frequent type was fire burn [67.4%] followed by scalds [20.7%], the latter were the most common [74.4%] cause of burns in children [< 10 years]. About one third [33%] of the patients suffered minor [< 10% TBSA] and 26.2% extensive [> 50% TBSA] burns. Carelessness was the most frequent [84.2%] predisposing factor. It was therefore concluded that majority of burn injuries occur due to lack of understanding the danger of certain practices prevalent in our society. Prevention appears to be the most effective way in reducing the incidence of burn injury through education of risk groups

Logistic regression analysis in predicting survival among burn patients

The study was designed to evaluate the influence of various factors on burn mortality and to develop a predictive statistical model for burn patients. Records of 670 acutely burnt patients treated at the burns unit Civil Hospital, Karachi were subjected to statistical analysis. Six variables were compared with the outcome. Out of these five [burn size, type of burn, circumstances of burn, gender and face burn and/or history of burn in closed space] showed statistical significance [p<0.001]. There was no significant relationship of age with the outcome. The significant factors were subjected to multi-factorial logistic regression analysis. Three variables [gender, type of burn and circumstances of injury] were eliminated and two independent variables [burn size and face burn and/or history of burn in closed space] were retained. Using probability equation a predictive model was developed. The predictive model was employed to predict mortality risk in present patient population and a scatter-plot developed. This model will aid to predict the chance of survival in a simplified or by medical or trained personnel in the ward setting. It would also be of assistance in assigning severity score on admission, patient and family counseling, clinical audit, comparing therapies and evaluation of new therapies