Solving a linear model of nonfatal risk behavior and injuries in school children.

A calculating method and computer program have been developed for solving an overdetermined system of linear equations. It was applied for calculation of group risk exposure of school children (10-15 years) to nonfatal injuries in sport recreation and playing (SRP). Interviewed non-injured school children were separated into non-overlapping groups of risky behavior. Data on interviewed non-injured and non-interviewed injured children were put into a linear system of equations consisting of a 4 x 3 matrix scheme. Each equation represents children of the same age and sex and consists of the percentage of interviewed children grouped into high medium or low risk behavior and of the percentage of injured children. Four methods of calculating the matrix system were evaluated and the best implemented as a new developed computer program. The resulting linear group risk factors enable the prediction of behavior outcome. Data on interviewed injured children were compared with the results obtained. The validation procedure is proposed by testing the sensitivity and robustness of the method.

Epidemiological investigation of school-related injuries in Koprivnica County, Croatia.

AIM: To assess the prevalence of injuries in elementary schools and determine specific risk groups of school-age children. METHODS: According to the 1991 census, there were 6,398 children between 7 and 14 years of age in the study area of the former Koprivnica district. During the 1992-1997 period, 354 children were injured in school. The registration of injured children was performed via structured questionnaires filled out at the emergency clinic and outpatient surgical clinic of the General Hospital in Koprivnica. The mechanism of accident and activities preceding it were categorized according to the Nordic Medico-Statistical Committee classification. Chi-square test was used to determine groups of school children at specific risk and a classification tree was made on the basis of minimum entropy values for age, sex, activity, and mechanism of injury. RESULTS: The highest injury rate of was recorded in 12-year-olds (21.7%). Upper extremities were most common site of injury (52.8%), whereas the most common type of injury was contusion (45.2%). The rate of head injuries was 3.2 times higher in younger (aged 7-10) children, whereas the rate of sports injuries was 3.5-fold higher in older (aged 11-14) children (p=0.001). Entropy classification revealed younger school-age children to be at the highest risk of contusion due to a blow from a ball, an object, or contact during sports activities. CONCLUSION: In Koprivnica County, most school-related injuries occurred during sport activities (42%) and play during recess (55%), with specific differences in age and sex.

Multiple attribute entropy classification of school-age injuries.

The data exploration task in epidemiology of school children injuries is aimed at defining specific risk groups that can be further analyzed for definition of prevention measures. These groups are selected by means of entropy classification of relevant data collected in the Koprivnica district, Croatia. The necessary pruning procedure for the entropy classification tree is proposed for: (a) entropy increase for uniformly populated classes, and (b) entropy decrease for the case of zeros in a class. The second reason for pruning is emphasized for multiple (> 3) attribute descriptions of injury events because the number of classes increases exponentially with the number of attributes. Four possible prevention groups have been found by entropy classification: (1) dislocation, sprain and strain injuries, mostly on legs of both age groups of boys (7-10 years, 11-14 years) in/around the house activities; (2) arm fractures of all boys in/around the house activities; (3) open wounds mostly of the head, in younger boys at school, and open wounds in younger boys in road traffic; and (4) contusions mostly on arms, of older girls at school and contusions in older boys in road traffic accidents. These groups comprise 57.2% of all considered cases.

Young child injury analysis by the classification entropy method.

The project 'The Register and Preventive Programs for Accidents and Injuries' enabled data collection on all the injured who sought medical aid in Koprivnica County (population 61,052), Croatia, since 1992. Children aged 1-4 years are 5.03% of the whole population of the district. Complex injury attributes were analysed. Binary attributes were classified as input: age, gender, place of injury; and output: severity of injury. A new application of information entropy was introduced and applied to the classification of injury-causes attributes. The information entropy was calculated for the classification of input attributes according to the minimum information content. The decision procedure is given as a sequential procedure separating important from unimportant causes of injury at each decision level. Thus a decision tree with increasing entropy, i.e. decreasing determinism, was obtained showing that age (0.5347 N), place (0.6062 N) and gender (0.6105 N) are measurable attributes in child injury ascertainment in a descending pattern. It was shown that this method is, at the same time, an optimal way of using an attribute decision process of injury causes classification.

Macro model prediction of elderly people's injury and death in road traffic accidents in Croatia.

The case of elderly people injured and killed in traffic accidents in Croatia is used to illustrate a prediction method. This method consists of several steps. First, the most important variables are selected. Second, to overcome intercorrelation, new variables are constructed that include the old ones. Third, the new variables and its first derivatives are used in a rank correlation method. The results show good predictive capabilities.