3.4.1. Characteristics of the injured population

Of the 5,885 people in the survey (with known age and sex) 254 (4.3 %) were reported injured (Table 13). This gives an estimated number of 5,953 cases of injury for the whole population of the island.

Table 13: Injuries by Major Age Groups

Age	Number of people	Number injured	Injury rate (‰)
0-19	3,190	79	24
20-39	1,367	84	61
40-59	751	61	81
60	577	30	51
unknown	92	2
Total	5,977	256	43

(X², p < 0.01)

The age and sex of the injured

The overall injury rate for vales and females was the same. (Figure 4). Differences in age specific injury rates were statistically significant (Table 13). The detailed graph in five year age groups (Figure 5) shows that injury rates ranged from less than 10 per thousand for the under-fives to over 120 per thousand for men aged 45-49.

Position in the household and occupation

A comparison was made between the position in the household and the occupation of those who were injured and those who were unharmed.

Table 14: Position in Household and Injury

	Injured	Not injured	Injury rate (‰)
Heads of house-holds	101	1120	84
Other household members	155	4601	33
Total population	256	5661	43

(X², p < 0.001)

Heads of households formed Just over 20 % of the population. As table 14 shows, household heads were 2 ½ times as likely as other household members to be injured.

Figure 4: Dominica: Injury Rate by Age and Sex

Figure 5: Dominica: Injury Rate by Sex and 5 Years Age Groups

About a third of household heads were 60 or older, but their injury rate was only 40 per thousand, compared with 99 per thousand and 105 per thousand for younger heads in the age groups 20-39 and 40-59 respectively. (Few household heads were under 20). (Table 15).

Table 15: Heads of Household Surveyed with Age Specific Injury Rates

INJURY RATES

age groups	Injured	Not injured	Injury rate per thousand
0-19	3	20	130
20-39	37	335	99
40-59	46	390	105
60	15	355	40
Total	101	1120	84

The difference in proportion of heads of households among injured and non injured people was significant in a1_ age groups except in those over 60 years of age. (Table 16).

Being head of a household at the time of the disaster thus seems to be a risk factor.

Table 16: Household Heads and Injury by Major Age Groups

Age group	Injured			Not injured			SND
	Head of house- hold	Total injured (all positions)	Proportion of heads of household	Head of house-hold	Total not injured population	Proportion of heads of households
0-19	3	79	3.8	20	3133	0.6	p< 0.01
20-39	37	85	43.5	335	1288	26.0	p< 0.01
40-59	46	61	75.4	390	693	56.3	p< 0.01
60 and over	15	31	48.4	355	547	64.9	p> 0.05
Total	101	256	39.5	1120	5661	19.8	p< 0.01

Table 17: Occupation and Injury

	Injured	Not injured	Injury rate per thousand
Land labourers	34	315	94*
Housewives	40	484	67**
Total population	256	5661	43

* X², p < 0.001
** X², p < 0.005

Land labourers and housewives were significantly more exposed to injury than those in other occupation groups. They were the only major occupation groups found to be significantly risky. (Table 17).

Farmers with an injury rate of 47 per thousand, were not markedly more at risk than the general population.

There was no significant difference in ages between housewives and land labourers who were injured and those who were not. It is interesting to note, however, that housewives and land labourers each formed over a quarter of those injured in the 40-59 age group.

Damage to the home

The hypothesis was that people from homes that had been badly damaged in the hurricane were more likely to have been injured than those from homes that had sustained only minor damage.

Houses were grouped into four categories according to the amount of damage of roof and walls.

Table 18: Damage to House and Injury

Damage to house	Occupants injured		Occupants not injured
	No.	%	No.	%
None or minor	63	26.0	1555	28.8
Moderate	63	26.0	1523	28.2
Severe	40	16.5	1009	18.7
Total destruction	76	31.4	1311	24.3
Total	242	100.0	5396	100.0

There was no significant difference between the injured and the uninjured persons in relation to the extent to which their houses had beer damaged (Table 18). Nonetheless, injury rates were somewhat higher for those whose houses had been completely destroyed. Nearly a third of the injured came from such homes.

As a further indicator of the severity of the impact of the hurricane or the home, damage to water and electricity supplies was examined. There was no significant association between injuries and the degree of disruption of either of these facilities.

The lack of association between damage to the home and personal injury may be partly because (as shown later) most injuries happened to pee, who were outside at the time of the disaster. Nevertheless, the amount of damage to housing, in general, did give an indication of how fiercely the hurricane had struck a particular area.

Evacuation and migration.

After the hurricane about 39 % of the population had to leave their homes for a night or longer (Table 19).

Table 19: Evacuation and Injury

JURY

	Injured		Not injured
	No.	%	No.	%
Left home	101	44.1	1995	39.0
Stayed	128	55.9	3121	61.0
Total	229	100.0	5116	100.0

Overall, there was no significant difference between the injured and the uninjured people as to whether they left or stayed at home.

Looking into the age groups of the injured persons (Table 20), it was found that the 20-39 years age group was significantly more likely to have stayed than to have left home. For the other age groups, the difference between staying and leaving was not significant.

Table 20: Evacuation and Injury by Major Age Groups

Age group	Injured who left		Injured who stayed		Total Injured
	No.	%	No.	%	No.	%
0-19	37	50.7	36	49.3	73	100.0
20-39	27*	37.5	49*	64.5	76	100.0
40-59	22	42.1	30	57.7	52	100.0
60 and over	15	53.6	13	46.4	28	100.0
Total	101		128		229

* (SND p< 0.01 )

Table 21: Length of Absence and Injury

Length of absence	Evacuees
	Injured		Not injured
	No.	%	No.	%
Less than 1 week	27	26.7	674	33.7
More than 1 week	49	48.6	773	38.7
Not yet returned	25	24.7	548	27.4
Total	101	100.0	1995	100.0

There was no significant difference in the length of absence between the injured persons and the others who had to evacuate their homes (Table 21). However. the number of injured people who were absent for more than a week was important.

Settlement types

Table 22: Type of Settlement and injury

Settlement type	Injured
	No.	%
Roseau	82	32.0
Towns	104	40.6
Villages	70	27.3
Total	256	100.0

There was no significant difference in the distribution of the injured according to the type of settlement in which they lived (Tableau 22). Injury rates were 39 per thousand for Roseau, 44 per thousand for smaller towns and 47 per thousand for the village.