|
Abstract
|
1. The present investigation, the first of its kind, was designed to compare the characteristics of pedestrians 18 years of age and older fatally injured in Manhattan with those who though similarly exposed were not similarly involved. The case series consisted of 50 of the 52 such pedestrians fatally injured at known sites and times in Manhattan between May 3 and Nov. 7, 1959, and whose deaths occurred during the same period. Autopsies were performed in all cases and appropriate specimens obtained. Data with respect to those nonfatally injured were also secured.
2. The control group was obtained by visiting each accident site on a subsequent date, but on the same day of the week and at exactly the same time of day, and stopping, interviewing, and obtaining breath specimens (for alcohol) from the first 4 pedestrians of the same sex as the deceased, and 18 years of age or older. Of 212 so stopped, only 12 failed to cooperate fully.
3. The results obtained included both purely descriptive and controlled observations. Among the former: The accidents tended to occur outside of major shopping and business areas. Fifty-one per cent occurred in the 6 hours preceding midnight, 73 per cent in the 12 hours beginning at 3 p.m., and the smallest number in the 3 hours ending at 9 A. m. Forty-four per cent of those killed survived less than 6 hours. This together with the common evening and nighttime occurrence underscores the need for emphasizing the same period in the organization of emergency medical care.
4. Weather, raining vs. not raining, was not significantly associated with the occurrence of such accidents, although increased risk with rain may have been obscured by associated decreases in pedestrian density.
5. The mean age of those killed was 58.8 years; of those nonfatally injured 48.4 years; and of those interviewed at the fatal accident sites 41.6 years. The most reasonable interpretation of these findings is that there is, first, an age-associated risk of involvement per se, and, second, an age-associated risk of fatal outcome, once involved. Because of these differences and the probability that others of the characteristics studied would show associations with age, comparisons between the case and control groups included comparisons between age- and sex-matched members of both groups.
6. The members of the case group were significantly more often foreign born, less often married, and more often of lower socioeconomic status than were those in the control group. However, when these comparisons were made between cases and controls of the same age and sex, no significant differences remained. It was also found that neither Negroes nor those born in Puerto Rico, two major population groups, were significantly more or less commonly found in the case than in the control groups.
7. Only 4 of the 46 persons in the case group of known residence lived outside of Manhattan. Consequently, neither commuters nor visitors contributed substantially to the deaths studied. In addition, many of those killed were hit within relatively short distances of their residences. This can not be attributed entirely to a corresponding localization of exposure since Manhattan residents were found significantly more often in the case series than among those similarly exposed at the same sites, even when the difference in age between the two groups was taken into consideration.
8. Because postmortem specimens for alcohol analysis were employed, comparisons involving the blood alcohol concentrations of members of the case series were limited to those who had survived less than 6 hours. In this group 26 per cent showed no alcohol, 47 per cent had concentrations of 50 mg. per cent or higher, 42 per cent, 100 mg. per cent or higher, and 32 per cent, 150 mg. per cent or higher. The corresponding percentages in the much larger control group were 67 per cent (none), 16 per cent (≤ 50), 8 per cent (≤ 100), 6 per cent (≤ 150 mg. per cent), respectively. Significant differences were found whether the cases were compared with the members of the total control group, with the site-matched controls from the same accident sites, or with age- and sex-matched members of the control group. The data suggest that the risk first becomes significantly increased in the low, 10–40 mg. per cent range.
9. The members of the case group found to have either low blood alcohol concentrations or no measurable amount were significantly older than the corresponding members of the control group. In contrast, the members of the case group with high blood alcohol concentrations were not significantly older than the corresponding members of the control group. The data suggest that the case series was composed of two substantially discrete groups, each with increased risk, namely, a group of the elderly who had been drinking little or not at all, and a group of the middle aged who had been drinking heavily.
10. The importance of determining the extent to which the chronic alcoholic is overrepresented in such accidents is discussed. It is believed that such individuals were present in the case series, but controlled observations appropriate to this point were not possible. It is emphasized that fatal adult pedestrian accidents must be regarded as contributing portions of the morbidity and mortality associated with alcoholism and with advanced age. |