Unintentional Drowning Deaths, Arizona, 1986-1996


WATER-RELATED INCIDENTS IN 1996 IN MARICOPA COUNTY

Report prepared by Timothy J. Flood, M.D., Medical Director,
Office of Chronic Disease Epidemiology


SUMMARY

INTRODUCTION

CASE DEFINITION

FINDINGS

OUTCOME (VITAL STATUS)

DISCUSSION

SUMMARY

Local fire departments in the Phoenix metropolitan area submit reports of water-related incidents that have activated the 9-1-1 system. In 1996 there were 85 serious water-related incidents that occurred in the area. Children under 5 years of age comprised 62 of the cases, 51 of which occurred in swimming pools. It is believed that prevention efforts at the community level have reduced the number of incidences over the past seven years. Two risk factors continue to dominate: lapses in parental supervision of children in pools during the summer months, and lack of pool barriers during the colder months.

INTRODUCTION

In the mid 1980's the drowning death rate of Arizona's preschoolers ranked first in the nation. Warm weather, long summers, and the presence of about 300,000 residential swimming pools make Arizona prone to water-related incidents. Furthermore, death is just one outcome of water-related incidents: in about 9% of incidents, the child survives, but with some degree of neurologic impairment.

In order to address this problem in Maricopa county (the Phoenix metropolitan area), the Drowning Prevention Coalition of Central Arizona was formed in 1988. Currently, this Coalition is comprised of municipal fire departments, hospitals, State and County health departments, community organizations, pool builders, suppliers of pool safety equipment, concerned parents, and others.

In 1988, the Arizona Department of Health Services (ADHS) established a surveillance system to monitor water-related incidents as reported by fire departments. The fire departments are first on the scene of 9-1-1 calls and are able to provide information about the event. Few, if any, incidents occur without activation of 9-1-1. Water-related incidents are reported on a standard form developed in conjunction with the fire departments. The reported data items include: demographics of the victim, the location, and the circumstances surrounding the event. The ADHS Office of Chronic Disease Epidemiology receives and analyzes these case reports.

The following report presents the data collected during 1996, and compares the findings to those in previous years. Much of the report focuses on children under five years of age, specifically on incidents occurring in swimming pools.

CASE DEFINITION

In this report a water-related incident is defined as an incident in which a fire department responded to a 9-1-1 emergency call. We include in the analysis any incident in which the victim was given CPR, not breathing, submerged, or not struggling when retrieved from the water. (Some of these cases die the same day or within a short period due to the incident; some recover fully.) We exclude from analysis any incident which did not appear to be life-threatening, for example, an incident in which a victim was struggling and did not require CPR. For consistency, one person at ADHS receives and codes the forms of each reported incident. Less than six incidents per year are questionable as to whether the incident was life-threatening.

FINDINGS

In 1996 the 9-1-1 emergency system responded to 85 serious, water-related incidents in Maricopa county among persons of all ages. The number of incidents in previous years were 105 in 1990; 102 in 1992; 112 in 1993; 75 in 1994, and 92 in 1995. Table 1 presents the distribution of the 85 incidents according to the city and the age of the victim.

Table 1

City of Incident

AGE OF VICTIM

 0-4

 5-14

15-34

35-64

  65+

Total

Chandler

7

1

0

0

1

9

Gilbert

1

0

0

1

0

2

Glendale

1

2

0

1

0

4

Mesa

12

0

3

2

0

17

Peoria

1

0

0

0

0

1

Phoenix

36

5

4

0

2

47

Paradise Valley

0

0

0

0

0

0

Rural area

1

0

0

0

0

1

Scottsdale

1

0

0

0

0

1

Tempe

2

0

0

1

0

3

All cities

62

8

7

5

3

85

 

(73%)

(9%)

(8%)

(6%)

(4%)

(100%)

Table 2 presents the body of water of the incidents according to age group. Most incidents took place in pools. Pools, either above-ground or in-ground, were involved in 69 (81%) of the 85 events. Fifty-one of the 69 incidents in pools involved children under age 5. Bathtubs (7) and buckets (3) were the next most common places for water-related incidents, all of which involved children under age 5.

Table 2

WATER BODY

AGE OF VICTIM

  0-4

  5-9

 15-34

 35-64

  65+

Total

Bath Tub

7

0

0

0

0

7

Bucket

3

0

0

0

0

3

Canal or Irrigation Ditch

0

0

2

0

0

2

Fish Pond

1

0

0

0

0

1

Pool, above ground

6

0

1

0

1

8

Pool, in ground

45

7

3

4

2

61

River or Lake

0

1

1

0

0

2

Spa

0

0

0

1

0

1

Toilet

0

0

0

0

0

0

Other

0

0

0

0

0

0

All water bodies

62

8

7

5

3

85

YOUNG CHILDREN

Children, ages 0-4 years (also called preschoolers in this report), comprised the largest proportion of persons experiencing a water-related incident. Although older individuals are equally important to consider in terms of loss, the events surrounding their incidents did not appear as preventable as those of the younger age group. The remainder of this report presents the data analyzing the findings among the 0-4 year age group. Since most incidents for this age group occurred in pools, many of the following tables are restricted to the incidents that occurred in a pool.

Figure 1 shows the number of pool related incidents over the last nine years. In 1996 the number of incidents decreased slightly to 51. This was about average for the past few years.

Figure 2 displays the distribution of cases, by month. This chart shows the effect of interventions that began in the dreadful summer of 1989. These interventions were aimed at reducing pool incidents among young children. This effort included educational messages on TV and radio, door-to-door leaflet distribution by fire departments, heightened reporting by the press, and legislation requiring pool barriers in most cities in the county.

In slight contrast to previous years, in 1996 there was a lower peak in the number of incidents between May and September. Because we have no way to measure the number of public education messages delivered, its role is speculative. However, we believe it may have had an effect in reducing the number of summertime incidents.

As presented in Table 3, the majority of the young victims of pool incidents were male (61%). Whites comprised 81% of the racial distribution. Race and ethnicity were poorly documented in 1996, causing difficulty in analysis. Of the 16 children of unknown race, 11 were Hispanic; we presume they were also White. Of the 24 children of unknown Hispanic status, 19 were recorded as White, 4 were of unknown race, and one was Black. The confusion here results from the fact that many firefighters do not consider that race and ethnic origin are two independent variables.

Assuming that all 24 of the children with unknown Hispanic status were non-Hispanic, we see that 14/51 (27%) of incidents involved Hispanic children. This correlates with 1990 census data which indicates that 26% of all children age 0-4 who reside in Maricopa county are Hispanic.

Table 3

Number

(%)

Gender

Female

Male

 

20

31

 

(39%)

(61%)

Race

Asian

Black

Amerind

Other

(Unknown)

White

 

0

3

0

3

16

29

 

(0%)

(9%)

(0%)

(9%)

(--%)

(81%)

Hispanic Origin

No

Yes

(unknown)

 

13

14

24

 

(48%)

(52%)

(--)

Table 4 presents the incidents according to the body of water and the site of the event. The most common site was the victim's own home pool (36 incidents). All seven of the bathtub incidents happened in the victim's home. Another ten incidents occurred in relative's homes, and four incidents in a friend's home.

Table 4

BODY OF WATER

FRIEND'S HOME

NEIGHBOR'S HOME

OTHER

PUBLIC OR SEMI PUBLIC POOL

RELATIVE'S HOME

UNKNOWN

VICTIM'S HOME

ALL SITES

BATHTUB

 

 

 

 

 

 

7

7

BUCKET

 

 

 

 

 

 

3

3

CANAL OR IRRIG. DITCH

 

 

 

 

 

 

 

0

FISH POND

 

 

 

 

 

 

1

1

POOL, ABOVE GROUND

 

 

 

 

1

 

5

6

POOL, IN GROUND

4

 

 

 

10

1

31

46

SUM

4

0

0

0

11

1

47

63

Table 5 presents the type of dwelling where the 51 pool incidents took place. In 31 (61%) of the cases the pool incident occurred in a single family home. Ten incidents (20%) occurred in apartments or condominiums.

Table 5

BODY OF WATER

APT./CONDO

HOTEL

OTHER

SINGLE HOME

TOWNHOUSE

TRAILER COURT

UNKNOWN

TOTAL

BATHTUB

3

 

 

4

 

 

 

7

BUCKET

 

 

 

3

 

 

 

3

CANAL OR IRRIGATION DITCH

 

 

 

 

 

 

 

0

FISH POND

 

 

 

1

 

 

 

1

POOL, ABOVE GROUND

 

 

 

6

 

 

 

6

POOL, IN GROUND

10

1

 

29

1

1

3

45

TOTAL

13

1

0

43

1

1

3

62

Figure 3 depicts the occurrence of pool-related incidents by day of the week. The most common day of occurrence varies from year to year. Sunday had the most incidents in 1990; Saturday did in 1991, 1994, 1995 and 1996; Tuesday in 1992; and Friday in 1993. Incidents occurred on every day of the week. There was no day when heightened vigilance would not have been important. The figure shows that pool incidents among these pre-school age children tend to occur slightly more often on Saturdays.

Figure 4 presents the distribution of events by hour of the day. Not surprisingly, the incidents occurred when children were likely to be awake: the peak time for an incident in this age group was between 10 a.m. and 8 p.m. However, in 1996 there was a less pronounced peak of incidents during the early afternoon compared to previous years (data not shown).

Table 6 presents information about the type of clothing being worn at the time of a pool-related incident. In 63% of the cases the children were not wearing swimming attire. Thus, more than half of the events did not occur in a swimming situation, but rather at a time when the children were not expected to be in or near the pool.

Table 6

Clothing

Number

(%)

None

Other

Swimwear

Unknown

1

30

18

2

(2%)

(59%)

(36%)

(4%)

A major goal of the Coalition is to identify and address the causes of water-related incidents in preschoolers. To assist in this effort, the personnel from the responding fire department attempt to determine the apparent circumstances surrounding each event. A member of the fire department looks for pool barriers that could prevent entry by small children, and asks about supervision at the time of the incident.

Figure 5 displays information about the supervisor of the child at the time of incident. The number of incidents in which the mother alone was supervisor has fallen steadily over the past few years. The mother was the supervisor in 13 (27%) of the incidents in 1996; the father was the supervisor in 20%. There was a combined supervisory responsibility of both mother and father identified in another 7 (14%). In 15 incidents (31%) the supervisor was someone other than the child's parent. This would seem to be higher than the amount of time that children in this young age group would spend outside the direct supervision of a parent. Thus, babysitters also need to be alert to the potential for a pool related incident to occur.

Figure 6 shows the apparent causes during the combined years 1988 through 1996. This information is further classified into events that occurred in cold and warm months. The cold months are defined as October through April and the warm months as May through September. The Figure reveals that during the cold months an absence of a barrier to the pool was the principal cause of water-related incidents. During warm months a lack or lapse of supervision was the principal cause.

The proportion of incidents attributed to gate or door failure is concerning. These are incidents in which the latch to the gate or door failed, or, more commonly, the gate was propped open. Maintenance of the integrity of the gates is obviously important.

Figure 7 also presents this data about the attributed cause over the nine-year period. Although, the data do not show consistent changes across the years, the figures do show the previously mentioned finding that a lapse in supervision in the warm months is the predominant cause, while lack of barrier in the cold months is the predominant cause. This chart is produced annually to search for trends in a single attributed cause.

OUTCOME (VITAL STATUS)

The outcome (vital status) of the children involved in a water-related incident is now tracked by the fire departments. However, this item was not determined for all victims in 1996. The fire departments have learned that at least 10 of the 62 preschoolers who experienced a water-related incident in 1996 have died, nine from the incidents in a pool, and one due to an incident in a bathtub. In addition, one child was left neurologically impaired in 1996. As mentioned, the outcomes are missing on many of the incidents, so these figures underestimate the morbidity associated with an incident. As noted in the previous report (Table 22), twenty children under age 5 died in 1996 in Maricopa county.

Figure 8 shows the death rate for children under age 5. The decline in the death rate is similar to the decline in the number of pool incidents shown in earlier graphs.

DISCUSSION

Since 1989 there has been a substantial decrease in the number of pool-related incidents among preschool children in the Phoenix metropolitan area. It is believed that the educational interventions by participants in the Drowning Prevention Coalition of Central Arizona and others especially the media have helped to reduce the magnitude of this problem. In 1996 there were 51 pool incidents, an average number for recent years. The data indicates that a large part of the problem is attributable to a lapse in supervision of these infants, often when the child already is in the pool with others.

While 51 incidents occurred in swimming pools, there were 11 incidents in other bodies of water, including seven in tubs, three in buckets, and one in a fish pond. Supervisors must remain constantly vigilant to the hazard of drowning anytime a young child is near water or has access to water.

A partial solution to pool drowning is the placement of barriers around pools. The findings in this report have indicated that the largest number of incidents occur at home in the family pool. Arizona law now requires that all homes with a child under 6 that have a pool must have a barrier between the house and pool. This law applies to pools built after June 1, 1991. However, local jurisdictions can pass laws that preempt the State law. The State law specifies that fences, motorized safety pool covers, or self-latching doors leading to the pool may be used as a barrier. There are specifications for these barriers in term of height, openings, and gate latches that are to prevent entry by small children. Barriers would appear to be most effective in reducing incidents occurring in cold months, but also might reduce some incidents occurring in warm months.

Even with the placement of barriers, there is no replacement for proper supervision. The data show us that the prevention message needs to be customized depending upon the season of the year. Supervision is particularly important in the warmer months when parents are with their children around the pool. This report identified the afternoon hours as a crucial time for watching children in order to prevent water-related incidents. The Coalition believes that a combination of a pool barrier and supervision remains the key to the prevention of drowning tragedies.

The support of the electronic media in recent years in calling attention to this problem has been most encouraging to the Coalition. It is hoped that these resources will continue in order to educate the new groups of parents that may not appreciate the risk of child drowning in Maricopa county.


Public Health Services
Office of Health Planning, Evaluation and Statistics
Arizona Center for Health Statistics
Phone: 602/542-7333; FAX: 602/542-1244