Chapter 5

With regard to trends with time in different countries, Annegers et al. (18) showed that the incidence rates for hospitalized traumatic head injury with evidence of presumed brain involvement in Olmsted County, Minnesota increased progressively during 1935-1974. For males, the increase was from 117 to 270 and for females from 46 to 116, all per 100,000 of the population. Most of the increase was in the category of mild injuries. In the period 1945-1974, the incidence rates for moderate, severe, or fatal injuries to either sex were nearly constant. Caveness (7 1) confirmed this trend for 1970-1975 for USA as a whole. For 1979-1986, MacKenzie (52) found for Maryland, USA, an increase in age-adjusted hospitalized head injury discharge rates of 3.4/100,000 population per year for all severity groups. Now, the largest increase was noted among the severest cases, increasing from 1.7 to 5 per 100,000 during the same period. Jennett et al. (21) showed that the total number of discharges after head injury increased by more than 40% in England and Wales from 1963 to 1972, almost entirely due to mild injuries. In fact, the incidence of severe injuries was slightly declining. With regard to Denmark, the main source of information for the following section is the publications on hospital discharges for the country as a whole published by the Danish National Bureau of Health (80, 81, 4-6). Comparison of the study by Als et al. (74) with that by Engberg et al. (3) showed that the annual number of TBI hospital admissions more than doubled from 1947-51 to 1986. Details for the intermediate period can only partly be derived from existing statistics. In 1960-75, statistics for hospitalized non-fatal and fatal cases (without distinction between the two categories) were presented in a series of annual publications named Medical Report I1 (80) and covering fiscal years from April 1st to March 31th. From 1979, a new series of publications appeared, now based on calendar years, but hitherto only for 1979, 1981, 1985, 1987 and 1989 (81, 4-6). Statistical information from these sources for the years 1960-89 are presented in Table 6. Separate figures for ICD 850, commotio cerebri, as opposed to more severe cases, have been collected and published only from 1972/3. Up to 1974/5, fracture diagnoses ICD 800-809 were counted as one group, encompassing other fractures beyond those of the cranium. Up to 1977, ICD 850-854 comprised only head injuries without skull fracture. From 1977, diagnosis ICD 804, fractura ossis faciei cum allis fracturis, was cancelled, and cases with skull fracture together with brain concussion/ contusion/haemorrhage were allocated to one of diagnoses ICD 8 5 0 4 as principal diagnosis if the brain damage was judged more important than the skull fracture. This led to an increase in the number of discharges under ICD 851. Probably, the introduction of CT scanning in the late 1970’s has supported the same tendency. In the period 1972-1974, the number of discharges under ICD 850 as first-listed diagnosis (uncorrected for readmissions) decreased from 15,600 to 15,000 per year. Hereafter, until 1986, it fluctuated about an average of 14,600 with a standard deviation of 2.3%. It decreased 14% from 1986 to 1988, where the number was 12,608. A special investigation for 1986 (3) showed that the fraction of readmissions within the same year was 7% for this diagnosis. With regard to more severe cases, the situation is obscured by the said alterations of diagnoses. As shown in Table 6, the increase in the number of discharges under ICD 851-854 in 1979-1989 was compensated by a decrease in the number of discharges under ICD 800-803 as first-listed diagnosis. As mentioned in Chapter 4, study of a subsample of cases discharged in 1986 with one of diagnoses ICD 851-854 ( 3 ) showed that only 4lo/o were survivors with a recent TBI; about 20% died in hospital, and 38% had been readmitted due to TBI occurring in earlier years. Because of the many changes of classifications, the development with time from 1951 to 1989 is