Automated critiquing of medical decision trees

The authors developed a decision tree-critiquing program (called BUNYAN) that identifies potential modeling errors in medical decision trees. The program's critiques are based on the structure of a decision problem, obtained from an abstract description specifying only the basic semantic categories of the model's components. A taxonomy of node and branch types supplies the primitive building blocks for representing decision trees. Bunyan detects potential problems in a model by matching general pattern expressions that refer to these primitives. A small set of general principles justifies critiquing rules that detect four categories of potential structural problems: impossible strategies, dominated strategies, unaccountable violations of symmetry, and omission of apparently reasonable strategies. Although critiquing based on structure alone has clear limitations, principled structural analysis constitutes the core of a methodology for reasoning about decision models.

A peripartum neurologic event: shooting from the hip

We have shown that a simplified model, generated quickly in response to an emergency consultation, may provide useful insights in certain situations. A more developed model was useful in verifying these insights. Because the more complex model considered a longer time horizon than the simple model, it allows us to consider questions regarding long-term benefits of aneurysm repair. When modeling any problem, the most important reason for performing decision analysis is to gain insight from analyzing the clinical setting and from constructing the model. The quantitative results are usually of only minor importance. However, our most important insights are sometimes gained by looking beyond the quantitative level to understand the interactions of various effects within the model. In this case, it was those insights that were of the greatest benefit to the patient in arriving at a decision to have cerebral arteriography.

A review of work-related fatal injuries in New Zealand 1975-84--numbers, rates and trends

Deaths resulting from work-related injuries during 1975 to 1984 in New Zealand were identified and reviewed. Nine hundred and eighty-six members of the workforce (workers) were killed at work during this period. This excluded deaths resulting from traffic-related injuries on public roads which occurred during a person's work activity and injuries which occurred whilst travelling to or from work. The estimated average work-related fatal injury rate for New Zealand was 7.2/100,000 workers/year. Variations in fatal injury rate by year of injury, age, sex and race were observed. Occupation and industry fatality rates were also estimated and gross variations by occupation were found. Those occupations with the highest estimated rates included helicopter and agricultural pilots, demolition labourers, deer cullers and commercial deer shooters. These rates are likely to be biased to some degree. Occupations with the largest number of workers killed during this 10 year period were farmers (109), fishermen (79) and forestry workers (68).

Polio: long-term problems

The most common long-term problems seen in polio are brace problems, knee recurvatum, increasing weakness due to overuse and ankle equinus. A definite increased incidence of problems is seen after the patient is more than 30 years post-polio. The basis for most of these problems is chronic mechanical strain of weak musculature and substituting ligaments. Overuse can cause increasing weakness resulting in pain and decreasing function. It is therefore important to follow polio patients closely, especially those that are more than 30 years post-polio. If signs of overuse or chronic mechanical strain are noted, treatment should not be delayed.

Effect of method of delivery on heart rate pattern in the early newborn

The relationship of fetal heart rate (FHR) to neonatal heart rate (NHR) was examined in 81 uncomplicated term (>37 weeks) deliveries. Average base line FHR during the 30 minutes before delivery was compared with baseline NHR obtained on the same infants at 5 and 10 minutes, 1, 12, and 24 hours after delivery. Fetal status and gestational age were determined by Apgar score and Dubowitz examination. Five minutes following delivery, a 30 percent increase in NHR over FHR baseline was noted in 75 neonates (93%). The NHR returned to FHR within 12 to 24 hours. There were no differences in heart rate patterns of neonates, whether delivered vaginally or by cesarean section. Neonates delivered spontaneously or with forceps also had comparable HR patterns. It appears that the events of labor and delivery are not responsible for the increased heart rate observed in the early newborn.

Ejection problems and injuries: their causes, effects and treatments, and suggestions for preventive measures

A relatively brief discussion of the problems of combat ejections in the I.A.F. is presented, including the injuries found and their prevention where possible; some problems that are known to exist and to some of which no clear-cut solutions are known. Some indications are given of what to expect when confronted with a man immediately after he has ejected from his aircraft.

Some health indicators in New Zealand adolescents: The Rotorua Lakes Study 5

This paper describes and compares health indicators in Maori and European adolescents attending a coeducational secondary school in New Zealand. No important differences were found in certain urinary constituents, haemoglobin or packed cell volume, fasting plasma glucose, serum urea nitrogen, pulse rate, age of menarche, hospital experience or tuberculin test-BCG status.