Sample Paper

FROM THE FIELD

Public Health Needs Assessments of Tutuila Island,
American Samoa, After the 2009 Tsunami
Ekta Choudhary, PhD, MPH; Tai-Ho Chen, MD; Colleen Martin, MPH; Sara Vagi, PhD, MPH;
Joseph Roth Jr, MPH; Mark Keim, MD; Rebecca Noe, RN, MPH; Seiuli Elisapeta Ponausuia, MBA;
Siitia Lemusu, MPH; Tesfaye Bayleyegn, MD; Amy Wolkin, MSPH

ABSTRACT
Objective: An 8.3 magnitude earthquake followed by tsunami waves devastated American Samoa on September
29, 2009, resulting in widespread loss of property and public services. An initial and a follow-up Community
Needs Assessment for Public Health Emergency Response (CASPER) objectively quantified disaster-affected
population needs.
Methods: Using a 2-stage cluster sampling method of CASPER, a household questionnaire eliciting information
about medical and basic needs, illnesses, and injuries was administered. To assess response efforts, percent
changes in basic and medical needs, illnesses, and injuries between the initial and follow-up CASPER were
calculated.
Results: During the initial CASPER (N = 212 households), 47.6% and 51.6% of households reported needing a
tarpaulin and having no electricity, respectively. The self-reported greatest needs were water (27.8%) and
financial help with cleanup (25.5%). The follow-up CASPER (N=207 households) identified increased vector
problems compared to pre-tsunami, and food (26%) was identified as the self-reported greatest need. As compared to the initial CASPER, the follow-up CASPER observed decreases in electricity (−78.3%), drinking water (−44.4%), and clothing (−26.6%).
Conclusion: This study highlights the use of CASPER during the response and recovery phases following a disaster. The initial CASPER identified basic needs immediately after the earthquake, whereas the follow-up CASPER
assessed effectiveness of relief efforts and identified ongoing community needs.
(Disaster Med Public Health Preparedness. 2012;6:209-216)
Key Words: community assessment, disaster, tsunami response, public health assessment, rapid needs assessment

N

atural disasters such as tsunamis can significantly damage infrastructure and cause morbidity and mortality, home damage, and population displacement. Tsunamis, which can occur
following an earthquake, are a series of ocean waves generated by any disturbance that displaces a large water
mass.1 Tsunami-related injuries may result from blunt
force trauma, and deaths may occur immediately as a
result of drowning or injuries sustained in the event aftermath. Illness may occur as a result of water source
contamination or increased vector population, and mental health problems may occur because of stress or traumatic experience related to the tsunami.2 The 2004
Asian tsunami, a particularly devastating event, resulted in more than 175 000 deaths, nearly 50 000 missing persons, and more than 1.7 million displaced persons in the Indian Ocean region.3
On September 29, 2009, an earthquake measuring 8.3
on the Richter scale struck 190 km southwest of American Samoa,4 an island territory with a population of 57 291
people.5 The resulting tsunami caused 34 deaths and widespread destruction of coastal homes, mainly in the capitol city of Pago Pago and western coast of Tutuila Island

near the town of Leone.6 The waves damaged public utilities, resulting in immediate, widespread loss of water, electricity, and sanitation throughout Tutuila Island. In addition to injuries and deaths, many families were displaced
from their homes and potentially exposed to environmental and other public health hazards.
The American Samoa Department of Health (ASDOH)
requested assistance from the US Public Health Service (USPHS) and the Centers for Disease Control and
Prevention (CDC) to assess the public health needs and
health status of the affected population. In response, a
team from the USPHS and the CDC was deployed to
assist in conducting an initial (five days after tsunami)
and follow-up (three weeks after the tsunami) Community Assessment for Public Health Emergency Response (CASPER).7 CASPER is a type of needs assessment that uses a two-stage cluster sampling method
originally developed by the World Health Organization to assess immunization coverage.8 This methodology has been used previously to assess community needs
and health status following natural disasters9,10 and provides household-based estimates of specific needs, injuries, and illnesses after a disaster.11

Disaster Medicine and Public Health Preparedness
©2012 American Medical Association. All rights reserved.

209

Public Health Needs Assessments After Tsunami

FIGURE
Map of Sampling Frame for Initial and Follow-up CASPERs Following the Earthquake-Tsunami in American Samoa, 2009.

UNITED STATES
DEPARTMENT
OF HEALTH AND
HUMAN SERVICES

Community Assessment for Public
Health Emergency Response (CASPER)
American Samoa
NOTES:

LEGEND:
Census Blocks Within
1 Mile of Flood areas

Flood Areas
of Interest

Map Produced By:

0

2.5

Miles
5

GEOSPATIAL, RESEARCH, ANALYSIS,
AND SERVICES PROGRAM
DIV OF HEALTH STUDIES, ATSDR, CDC

The objectives of the initial CASPER were to rapidly identify
basic public health needs of affected communities, special needs,
and vulnerable populations, and environmental and other health
hazards that might result in further morbidity and mortality.
The follow-up CASPER objectives were to identify ongoing and
newly emerged community public health issues and to assess
the effectiveness of the public health response following the
initial CASPER. This report summarizes and compares the findings of the initial and follow-up CASPERs conducted following the American Samoa earthquake and tsunami.

to select the clusters, represented by a single census block, in
the first sampling stage. Using probability-proportionate-tosize, based on the number of housing units in each census block,
we selected 30 census blocks (clusters) covering 16 villages from
western Fagalii to eastern Alao. For the second stage of sampling, the team selected seven households within each selected census block. Once in the field, the teams randomly chose
a starting point within the cluster and then systematically (ie,
every nth house) selected the 7 households to interview for a
goal of 210 household interviews.

METHODS
Sample Selection

Survey Instrument

Based on ASDOH’s visual surveys of Tutuila Island following
the earthquake and tsunami and geospatial data from the National Oceanic and Atmospheric Administration (NOAA),12
our sampling frame for both CASPERs included all coastal census blocks affected by 15 ft of surge water, up to one mile inland (Figure). A total of 244 census blocks with one or more
housing units was identified within our sampling frame. For both
CASPERs, a two-stage cluster sampling methodology (30 clusters, 7 households) was used. We used ESRI ArcGIS software
210

ASDOH, USPHS, and CDC jointly developed a two-page questionnaire for each CASPER. The questionnaire covered the following: medical needs; availability of medical care, food, drinking water, and clothing; and earthquake- and tsunami-related
injuries, illnesses, and preexisting chronic conditions. For the
follow-up CASPER, we adapted questions from the initial
CASPER questionnaire and added questions based on ASDOH
public health priorities for the recovery phase, including increases in disease vector problems and vector control measures. The interviewers kept tracking sheets to track response

Disaster Medicine and Public Health Preparedness
©2012 American Medical Association. All rights reserved.

VOL. 6/NO. 3

Public Health Needs Assessments After Tsunami

TABLE 1
Characteristics of Households in Affected Areas Following the Earthquake-Tsunami, American Samoa, 2009
Initial Community Needs Assessment for Public
Health Emergency Response (CASPER)

Characteristics
Residence
Feels residence is not safe
Needs tarpaulin
Utilities
No electricity
Drinking Water Source
Drinking bottled water from agency
Drinking bottled water from other
source
Drinking from stream
Other
Need clothing

% of Households
(95% CI)
(N = 212)

Projected a No.
of Households
(95% CI)

41.9 (28.8-54.9)
47.6 (35.3-59.8)

1867 (1146-2588)
1872 (1235-2509)

51.6 (38.5-69.1)
31.1 (18.9-43.3)
60.8 (50.0-71.7)
3.8 (0.0-9.7)
38.3 (24.5-52.2)

Follow-up CASPER
Projected b No.
of Households
(95% CI)

% Change b

40.4 (29.2-51.6)
47.3 (34.6-60.0)

1788 (1191-2385)
2160 (1426-2893)

−3.6
−0.6

2399 (1578-3220)

11.2 (2.6-19.8)

498 (145-851)

−78.3 c

1400 (749-2050)
2748 (2192-3304)

17.4 (8.2-26.5)
44.4 (34.2-54.6)

773 (300-1246)
2001 (1257-2746)

−17.4
−44.1

172 (0-441)

% of Households
(95% CI)
(N = 207)

1.4 (0.0-2.9)

1676 (1013-2339)

28.1 (17.98-38.30)

64 (0-138)

−63.2

1203 (753-1654)

−26.6

a Based on weighted analyses using information from 2000 census (4513 housing units with a total population of 25 700).
b Percent change = (% of households in follow-up CASPER) – (% of households in initial CASPER)/(% of households in Initial CASPER).
c Statistically significant as 95% CI does not overlap; A positive (⫹) sign indicates an increase in the response percentage, whereas a negative (−) sign suggests a decrease in the

response percentage.

rates. We used confidential referral forms to report immediate
needs pertaining to safety threats, such as chemical spills or acute
medical needs (eg, need for urgent mental health counseling).
In addition, we disseminated health education materials pertinent to the disaster and a list of public health and relief agencies to each household approached. The questionnaire was written in English; survey teams, consisting of at least one member
who spoke Samoan, translated questions into Samoan when
needed.

Data Collection
Before conducting interviews, CDC personnel trained the interview teams, which were composed of CDC and ASDOH staff.
The training included instructions on selecting the households in the second stage of sampling, administering the interviews, and completing the referral and tracking forms. CDC
also briefed the teams on Samoan culture, safety issues, field
communication, and the local incident command system (ICS)
structure. Teams conducted interviews from October 5-7, 2009,
for the initial CASPER and October 22-24, 2009, for the followup. In each household, we interviewed one adult representative, aged at least 18 years or older.

Data Analysis
The data entry and analyses were performed using Epi Info 3.5.1.
The percentage and estimates (ie, projected number of households based on weighted analysis using census 2000 data) for
questionnaire items were calculated and reported. Contact rates
were calculated by dividing the number of completed interviews by the number of all housing units where contact was attempted. Cooperation rates were calculated by dividing the number of completed interviews by the number of all housing units
where contact was made. To compare the findings of the ini-

tial and follow-up CASPERs, the differences in response percentages were calculated using percent change, with the initial CASPER as baseline.

RESULTS
Initial CASPER
The interview teams conducted 212 interviews, with a contact
rate of 77.9% and a cooperation rate of 98.1%. Thirty-two percent of households had at least one child younger than age 2 years,
and 30% of households had at least one adult family member who
was 65 years of age or older. Approximately 9% of households
included at least one pregnant woman. Table 1 shows information on residence, utilities, drinking water, and other needs. During the initial CASPER, 42% of households believed that their
house was not safe to live in, 47.6% reported needing a tarpaulin, and 51.6% reported a lack of electricity. Most households
reported using bottled drinking water from either a relief agency
(31.1%) or another source (60.8%), while a small percentage
(3.8%) reported drinking stream water. Approximately 37% of
households reported a need for clothing.
During the initial CASPER, the most frequent self-reported
symptoms or conditions were respiratory conditions (21.2%),
stress or sleep disturbance (11.8%), and stomach ache or diarrhea (9.0%). Lacerations (11.8%); bruises or contusions (6.1%);
impalement or puncture wounds (2.4%); and strains, sprains,
or dislocations (2.4%) were the most frequently reported injuries (Table 2). Nineteen percent of households reported needing medical care or supplies since the tsunami, 16.7% reported
having immediate medical needs, and 23.5% reported anticipating needs within 3 to 7 days (Table 2). Thirty-nine percent
of households reported barriers to obtaining medical care or sup-

Disaster Medicine and Public Health Preparedness
©2012 American Medical Association. All rights reserved.

211

Public Health Needs Assessments After Tsunami

plies, including financial reasons (9%) and fear of traveling after the tsunami (2.8%).
Table 3 shows households’ greatest reported needs; respondents often gave more than one response. During the initial
CASPER, the most frequently reported needs included electricity (34.0%), water (27.8%), and financial help (25.5%). Interview teams also made more than 25 medical referrals for persons in need of immediate medical care or supplies and arranged
transportation for one critically ill patient to the emergency department.

Follow-up CASPER
Three weeks after the tsunami, the interview teams conducted
the follow-up CASPER, completing 207 interviews, with a contact rate of 88% and a cooperation rate of 99%. During the follow-up CASPER, 37% of households had at least one child
younger than age 2 years, and 35.1% of households had at least
one adult family member who was 65 years of age or older. Approximately 10% of the households included at least one pregnant woman. Forty percent of households felt their homes were
not safe to live in, 47.3% reported needing a tarpaulin, and 11.2%
reported a lack of electricity (Table 1). The majority (62.8%)

of households reported using bottled water as their sole potable water source, and approximately 28% of the households
reported a need for clothing.

During the follow-up CASPER, 13.5% of households reported
respiratory symptoms, and 16% reported fever (Table 2). At
least one household member sustained lacerations (3.4%), motor vehicle injuries (2.4%), and bruises or contusions (1.4%)
in the aftermath of the tsunami. Approximately 25% of households reported needing medical care or supplies since the tsunami; 26% reported having immediate medical needs (Table 2).
Thirty-three percent of households reported having barriers to
obtaining medical care or supplies, including financial reasons
(15%) and lack of transportation (5.3%).

Table 3 shows households’ self-reported greatest needs three
weeks after the tsunami: food (26.1%), financial help with
cleanup (14%), water (13.5%), and medical care (3.9%). Interview teams also made 11 medical referrals for persons in need
of medical care or supplies and arranged transportation for one
critically ill patient to the emergency department.

TABLE 2
Illnesses, Injuries, and Medical Care Needs and Access by Household Following the Earthquake-Tsunami, American
Samoa, 2009
Initial Community Needs Assessment for Public
Health Emergency Response (CASPER)

Illness/Conditions
Respiratory (cough, flu)
Stress/sleep disturbance
Stomachache/diarrhea
Skin condition/rash/burn
Inability to perform daily tasks
Dehydration/heat stress
Injuries
Laceration, abrasion
Bruise/contusion
Impalement/puncture wound
Strain/sprain/dislocation
Motor vehicle-related or other blunt
trauma
Insect bite
Electrical injury
Burn/sunburn
Brain injury/concussion
Medical Care
Needed medical care since tsunami
Need care immediately
Cannot get medical care/supplies
Lack of transportation
No medical services available
Financial reasons
Afraid to travel

Follow-up CASPER

% of Households
(95% CI)
21.2 (14.9-27.5)
11.8 (7.2-16.3)
8.9 (4.8-13.1)
4.3 (1.6-6.9)
3.8 (1.7-5.8)
3.3 (0.4-6.2)

Projected a No.
of Households
(95% CI)
967 (632-1302)
537 (114-303)
405 (198-613)
193 (66-320)
172 (65-278)
150 (14-287)

% of Households
(95% CI)
13.5 (7.2-19.7)
5.3 (2.3-8.3)
7.2 (2.1-12.3)
2.9 (0.0-5.9)
1.9 (0.2-3.6)
3.9 (1.0-6.7)

Projected c No.
of Households
(95% CI)
610 (295-925)
236 (67-405)
322 (37-608)
129 (0-261)
86 (3-169)
172 (19-324)

%
Change b,c
−36.3
−55.1
−19.1
−32.6
−50.0
⫹18.2

11.8 (4.5-19.0)
6.1 (2.4-9.8)
2.4 (0.0-4.7)
2.4 (0.4-4.3)
1.9 (0.0-4.1)

537 (179-895)
279 (96-463)
107 (0-218)
107 (17-198)
86 (0-190)

3.4 (0.4-6.4)
1.4 (0.0-3.2)
0 (0)
1.0 (0.0-2.4)
2.4 (0.6-4.2)

150 (15-286)
64 (0-138)
0 (0)
43 (0-104)
107 (17-197)

−71.2
−77.0
−100.0
−58.3
⫹26.3

0.9 (0.0-2.2)
0.5 (0.0-1.4)
0.5 (0.0-1.4)
0.5 (0.0-1.4)

43 (0-104)
21 (0-66)
21 (0-66)
21 (0-66)

0.5 (0.0-1.4)
0 (0)
0.5 (0.0-1.4)
0 (0)

21 (0-66)
0 (0)
21 (0-66)
0 (0)

−44.4
−100.0
0
−100.0

19.0 (13.6-24.4)
16.6 (10.0-23.3)
31.2 (23.6-38.8)
6.6 (2.5-10.6)
1.4 (0.0-3.6)
9.0 (5.0-12.9)
2.8 (0.2-5.4)

854 (575-1134)
696 (408-984)
1389 (1030-1748)
301 (107-495)
64 (0-162)
406 (219-592)
129 (13-245)

24.6 (16.4-32.8)
26.2 (18.6-33.8)
33.2 (22.7-43.6)
5.3 (0.0-11.3)
0.5 (0.0-1.4)
15.0 (7.8-22.1)
0.5 (0.0-1.5)

1087 (660-1513)
1155 (746-1563)
1473 (921-2024)
236 (0-523)
21 (0-66)
666 (271-1061)
21 (0-66)

⫹29.5
⫹57.8
⫹6.4
−19.7
−64.3
⫹66.7
−82.1

a Based on weighted analyses using information from 2000 census (4513 housing units with a total population of 25 700).
b Percent change = (% of households in follow-up CASPER) – (% of households in initial CASPER)/ (% of households in Initial CASPER).
c A positive (⫹) sign indicates an increase in the response percentage, whereas a negative (−) sign suggests a decrease in the response percentage.

212

Disaster Medicine and Public Health Preparedness
©2012 American Medical Association. All rights reserved.

VOL. 6/NO. 3

Public Health Needs Assessments After Tsunami

TABLE 3
Self-Reported Greatest Needs by Household Following the Earthquake-Tsunami in American Samoa, 2009
Initial Community Needs Assessment for Public
Health Emergency Response (CASPER)

Current Greatest Need
Electricity
Water
Financial help with clean up/repairs
Food
Medications
Trash removal
Physical help with clean up/repairs
Transportation
Medical care

% of Households
(95% CI)
34.0 (54.8-77.3)
27.8 (17.2-38.4)
25.5 (14.3-36.6)
21.7 (12.3-31.0)
13.7 (6.1-21.0)
10.4 (3.3-17.4)
9.4 (3.1-15.7)
7.1 (2.6-11.5)
6.6 (0.3-2.8)

Projected a No. of
Households
(95% CI)
1531 (931-2131)
1263 (761-1764)
1144 (625-1663)
986 (511-1461)
612 (251-973)
470 (128-812)
430 (121-738)
320 (97-542)
298 (127-469)

Follow-up CASPER
% of Households
(95% CI)
2.4 (0.5-4.3)
13.5 (7.8-19.2)
14.0 (4.4-23.6)
26.1 (19.0-33.1)
1.0 (0.0-2.4)
1.0 (0.0-2.4)
0.5 (0.0-1.5)
1.0 (0.0-2.4)
3.9 (0.2-7.5)

Projected c No. of
Households
(95% CI)
107 (17-197)
601 (280-923)
623 (166-1080)
1185 (82-1550)
42 (0-104)
43 (0-104)
21 (0-66)
43 (0-104)
172 (6-337)

% Change b,c
−92.9
−51.4
−45.1
⫹20.3
−92.7
−90.4
−94.7
−85.9
−40.9

a Based on weighted analyses using information from 2000 census (4513 housing units with a total population of 25 700).
b Percent change = (% of households in follow-up CASPER) – (% of households in initial CASPER)/ (% of households in Initial CASPER).
c Statistically significant as 95% CI does not overlap. A positive (⫹) sign indicates an increase in the response percentage, whereas a negative (−) sign suggests a decrease in the

response percentage.

The follow-up CASPER assessed current vector problems compared to pre-tsunami and use of mosquito protection; these vector issues were not assessed during the first CASPER. As shown
in Table 4, households reported increases in vector problems
such as night time (which can result in malaria) (77%) and day
time (which can result in dengue fever) (68%) mosquito bites;
increases in mosquito breeding sites (50%); and increases in flies
(59%), rats (37%), and stray dogs (16%). The most commonly reported methods for preventing mosquito bites included mosquito coils (64%), aerosol insecticides (43%), and
window screens (18%). Almost 7% of households reported not
using any type of mosquito prevention method (Table 4).

Evaluation of the Initial and Follow-up CASPERs
Compared to the initial CASPER, results from the follow-up
CASPER indicated that the number of households reporting a
lack of electricity had decreased by approximately 78.3%
(Table 1). Consumption of bottled water from relief agencies
and other sources decreased by 17.4% and 44.1%, respectively, and the need for clothes decreased by 26.6%.
Table 2 compares illnesses and injuries between the initial and
follow-up CASPERs. The largest decreases in percent change
were reported for stress or sleep disturbance (55.1%) and respiratory illness (36.6%). An increase of 18.2% was observed
in dehydration or heat stress illnesses. Decreases in all injuries
were reported, with the exception of an increase in motor vehicle-related or blunt trauma (26.3%). However, none of these
percent changes were statistically significant.
Table 2 compares medical care needs captured in the initial
CASPER to those of the follow-up. Medical care needs increased by 29.5%, while a need for immediate care increased
by 57.8%. The number of households reporting an inability to
obtain medical care and supplies increased by 6.4%; the majority (66.7%) resulted from lack of financial resources.

Table 3 compares households’ greatest needs between the initial and follow-up CASPERs. Household needs for electricity
and trash removal decreased by 92.9% and 90.4%, respectively. Similarly, the needs for medications, physical help with
cleanup or repairs, and transportation decreased by 92.7%,
94.6%, and 85.9%, respectively.

DISCUSSION
To our knowledge, this study is the first to compare findings of
CASPERs conducted during the response and recovery phases
of an earthquake and tsunami disaster response. Our findings
show the extent of tsunami-related public health problems during the response and recovery phase in American Samoa.
CASPER results were used to assess the effectiveness of ASDOH
relief efforts.
Early identification of affected communities’ disasterassociated needs can guide resource management during the immediate response, especially when limited information is available about characteristics of the affected population.13-15 During
the recovery phase, assessment of response efforts and identification of emerging needs are equally important.16 Follow-up
assessments are often recommended during the recovery phase
to monitor affected communities’ changing needs.11 Only limited information is available about characteristics of the affected population.13-15 These assessments can elucidate the changing patterns of access to medical care, basic household needs,
and other health needs.
Electricity outage following a disaster is common.13 As demonstrated by results from the initial CASPER, a significant number of households in American Samoa lost electricity because
of tsunami-related damage. Immediately following the tsunami, many households in the affected area evacuated at least
one night, and a substantial number had residence-related safety
concerns. Many reported the need for additional sheltering ma-

Disaster Medicine and Public Health Preparedness
©2012 American Medical Association. All rights reserved.

213

Public Health Needs Assessments After Tsunami

TABLE 4
Percent of Households Reporting Increases in Vectors and Practices Associated With Protection From Mosquitoes Following
the Earthquake-Tsunami in the Follow-up CASPER, American Samoa, 2009
Vectors
Night time mosquito
Day time mosquito
Flies
Breeding sites for mosquitoes
Rats
Dogs
Practices
Mosquito coils
Aerosol insecticides
Window screen
Repellent
Removing breeding sites
Protective clothing

% of Households (95% CI)
77.3 (69.3-85.3)
68.1 (58.1-77.9)
58.5 (48.1-67.9)
49.8 (39.0-59.8)
36.7 (20.0-52.7)
16.4 (8.7-24.4)

Projected a No. of Households (95% CI)
3475 (2480-4470)
3070 (2142-3999)
2618 (834-3401)
2230 (1368-3093)
1642 (603-2680)
748 (380-1116)

63.8 (54.6-72.9)
42.5 (32.2-52.8)
29.5 (9.3-49.6)
18.4 (3.2-33.4)
17.4 (7.3-27.4)
6.3 (0.9-11.6)

2861 (1959-3763)
1903 (1225-2582)
1310 (214-2408)
816 (0-1646)
773 (292-1255)
279 (0-566)

Abbreviation: CASPER, Community Needs Assessment for Public Health Emergency Response.
a Based on weighted analyses using information from 2000 census (4513 housing units with a total population of 25 700).

terials, such as a tarpaulin. At the time of the initial CASPER,
four to five days post-tsunami, we estimated that 600 households in affected Tutuila areas were still displaced. Three weeks
after the tsunami, many households reported their houses were
still not safe to inhabit.
Another common concern after any natural disaster is availability of potable drinking water, as most of the population uses
municipal water for daily needs.17 At the time of the follow-up
CASPER, nearly 60% of households were still using bottled water. We also estimated that 60 households in the affected areas
on Tutuila Island were drinking water from streams, in spite of
endemic risks of water-associated infectious diseases, including leptospirosis.18
The ASDOH indicated that the number one priority of relief
agencies was to provide basic needs (such as food, water, and
shelter) to affected households. During the initial CASPER,
households’ self-reported greatest needs included electricity, water, and financial help with cleanup. ASDOH informed the
American Samoa Power Authority about the estimated number of households without public utilities, using data from the
initial CASPER. Providing information throughout the power
restoration prioritization process facilitated rapid restoration of
electric and water services. With the help of the American Red
Cross, ASDOH also provided meals to families in need. Even
though a significant decrease in need for electricity (−78.3%)
and water (−44.1%) was reported during the follow-up CASPER
three weeks after the tsunami, food and water remained the most
commonly reported greatest needs.
The follow-up CASPER identified access to medical care as an
ongoing public health need. Tutuila Island has only one hospital, and after the tsunami its medical services were limited,
and its three well-baby clinics were understaffed. Limited medical capacity and persistent post-tsunami medical needs may have
hindered access to medical services. Compared with the ini214

tial CASPER, the follow-up CASPER found an increase in the
percent of households reporting problems related to access to
medical care, including the need for immediate medical care
and inability to get medical supplies. Previous studies suggest
that the majority of disaster-related illnesses and injuries occur during the post-impact (ie, cleanup) phase.10,19 However,
these findings show an overall decrease in illnesses and injuries three weeks after the tsunami (ie, cleanup phase).

Limitations
These results are subject to several limitations. First, because
the survey was conducted only in accessible households,
these estimates may underestimate the actual impact of the
disaster. For example, houses completely destroyed or
vacated were not included in initial and follow-up
CASPERs. Second, because the estimated number of households was reported using census 2000 data, the weighted
analysis does not account for changes in the number of housing units from 2000 to 2009. Third, if participants believed
that their responses were linked to relief resources, this perception could have introduced bias.

Strengths
Despite these limitations, CASPER provided a rapid and effective method to determine the immediate needs of communities in American Samoa following the tsunami. CASPER objectively quantified disaster-affected population needs that may
have been identified anecdotally but not enumerated. CASPER
has been used in many other disaster situations to provide population-based estimates within a few days or hours after data collection. Findings were provided to ASDOH officials within 72
hours of each CASPER; this information assisted in decisions
about deploying resources to tsunami-affected areas. While determining the proportion of households with a particular need
is useful, the projection of an estimate of the number of households with a specific need is even more useful. This information provided the ASDOH disaster response with quantities

Disaster Medicine and Public Health Preparedness
©2012 American Medical Association. All rights reserved.

VOL. 6/NO. 3

Public Health Needs Assessments After Tsunami

needed in their recovery efforts (eg, estimated housing units in
need of tarpaulins or how much food/water to take to an affected area). Because the findings determined the greatest needs
of the affected population, they were also used to help identify
and prioritize needs (such as water, food, and shelter) and to
target delivery to affected communities of resources such as medication, which was especially in need at that time because of the
H1N1 pandemic, an additional strain on the ASDOH. Another problem was the limited medical resources in American
Samoa. The territory has only one 128-bed hospital, which was
overburdened after the tsunami. ASDOH used the information provided by CASPER to plan short- and long-term intervention programs, including distribution of food and water in
affected areas. Confidential referrals made during the assessment helped ASDOH address individuals’ most urgent issues
immediately by directing appropriate resources. ASDOH used
this opportunity to assess local capacity to conduct future
CASPERs and collaborated with CDC to implement disaster
epidemiology training for ASDOH staff to enhance disaster preparedness and response capacities.

CONCLUSIONS
The results of the initial CASPER highlighted environmental
and other health hazards that could lead to further morbidity
and mortality. Immediately following the tsunami, many households in affected areas left their dwelling for at least one night,
and a substantial number were projected to have residencerelated safety concerns. Fears about the safety of municipal water resulted in a boil-water advisory. Based on data from the initial CASPER, the following recommendations were made:
• Provide a continuing source of potable water, especially to
those dependent on relief agencies for water deliveries.
• Provide immediate, appropriate shelter for displaced persons.
• Provideashort-termfooddistributionprogramfordisplacedpopulations depending on food banks or shelters for meals.
The follow-up CASPER identified ongoing public health needs
for food, water, and access to medical care. Survey results also
indicated that a small number of households reported family
members with injuries and acute illnesses. Based on the follow-up CASPER findings, the CDC team made recommendations to disseminate community education materials regarding safe drinking and bathing water sources, including promoting
American Samoa Power Authority’s (ASPA) water as a safe
drinking water source, and disseminating infectious disease brochures (eg, dengue fever, H1N1, leptospirosis). CASPER provided information about the changing public health needs of
the affected community. These findings helped the ASDOH
and emergency managers to prioritize and direct resources to
address immediate and ongoing needs, as well as to assess response efforts.
The most common natural hazards experienced by American
Samoa are cyclones or hurricanes and flooding caused by
hurricanes.20 American Samoa lies just to the northern end

of Tongan Trench, one of the most active seismic areas
worldwide; therefore, the area experiences frequent earthquakes that can result in tsunamis. 20,21 While the public
health consequences of earthquakes in American Samoa
have ranged from destruction of infrastructure to mass casualties and fatalities, no earthquake fatalities were noted in
this event, probably also due in part to a well-developed
emergency operations plan, which describes a public preparedness plan during a natural disaster event. However,
before the 2009 tsunami, the last confirmed tsunami in
American Samoa occurred in 1837. 21 Response efforts
during the 2009 earthquake and tsunami post-disaster operations, including findings from initial and follow-up
CASPERs, helped ASDOH to revise and improve their hazard vulnerability assessment and emergency operations plan
for tsunamis.
Author Affiliations: Division of Environmental Hazards and Health Effects
(Drs Choudhary, Vagi, and Bayleyegen and Mss Martin, Noe, and Wolkin),
Office of the Director National Center for Environmental Health (Dr Keim),
and Office of Public Health Preparedness and Response (Dr Chen and Mr Roth),
Centers for Diseases Control and Prevention, Atlanta, Georgia; and American Samoa Department of Health, Pago Pago, American Samoa (Mss Ponausuia and Lemusu).
Correspondence: Ekta Choudhary, PhD, MPH, CDC/NCEH/HSB, 4770 Buford Hwy NE, MS F-57, Atlanta, GA 30341 (e-mail: Echoudhary@CDC.gov).
Received for publication November 21, 2011; accepted February 7, 2012.

REFERENCES
1. Boyarsky I, Shneiderman A. Natural and hybrid disasters. causes, effects and
management. Top Emerg Med. 2002;24(3):1-25.
2. Kiem M. Cyclones, tsunamis, and human health: the key role of preparedness. Oceanography (Wash DC). 2006;19(2):40-49.
3. Doocy S, Robinson C, Moodie C, Burnham G. Tsunami-related
injury in Aceh Province, Indonesia. Glob Public Health. 2009;4(2):205214.
4. US Geological Survey (USGS). Magnitude 8.1 - Samoa Islands Region.
2009. http://earthquake.usgs.gov/earthquakes/eqinthenews/2009
/us2009mdbi/. Accessed February 7, 2011.
5. US Census Bureau. 2000 Census of Population and Housing, Social, Economic, and Housing Characteristics, PHC-4-AS, American Samoa. Washington, DC: US Census Bureau; 2003.
6. Ritchie K. Villages wiped out by killer tsunami. 2009. http://www.abc.net
.au/news/2009-09-30/villages-wiped-out-by-killer-tsunami/1084422. Accessed June 22, 2012.
7. Department of Health and Human Services (DHHS), Centers for Disease Control and Prevention (CDC). Community Assessment for Public Health
Emergency Response (CASPER). Atlanta, GA: Toolkit; 2009.
8. Henderson RH, Sundaresan T. Cluster sampling to assess immunization
coverage: a review of experience with a simplified sampling method.
Bull World Health Organ. 1982;60(2):253-260.
9. Bayleyegn T, Wolkin A, Oberst K, et al. Rapid assessment of the needs
and health status in Santa Rosa and Escambia counties, Florida, after Hurricane Ivan, September 2004. Disaster Manag Response. 2006;4(1):1218.
10. Zane DF, Bayleyegn TM, Haywood TL, et al. Community assessment for
public health emergency response following Hurricane Ike—Texas, 25-30
September 2008. Prehosp Disaster Med. 2010;25(6):503-510.
11. Malilay J, Flanders WD, Brogan D. A modified cluster-sampling method
for post-disaster rapid assessment of needs. Bull World Health Organ. 1996;
74(4):399-405.

Disaster Medicine and Public Health Preparedness
©2012 American Medical Association. All rights reserved.

215

Public Health Needs Assessments After Tsunami
12. National Oceanic and Atmospheric Administration (NOAA). Center for
Tsunami Research. Tsunami Event - September 29, 2009 American Samoa. 2009. http://nctr.pmel.noaa.gov/samoa20090929-local.html. Accessed October 14, 2009.
13. Centers for Disease Control (CDC). Rapid health needs assessment following Hurricane Andrew—Florida and Louisiana, 1992. MMWR Morb
Mortal Wkly Rep. 1992;41(37):685-688.
14. Centers for Disease Control and Prevention (CDC). Surveillance for injuries and illnesses and rapid health-needs assessment following Hurricanes Marilyn and Opal, September-October 1995. MMWR Morb Mortal
Wkly Rep. 1996;45(4):81-85.
15. Centers for Disease Control and Prevention (CDC). Community
needs assessment and morbidity surveillance following an ice storm—
Maine, January 1998. MMWR Morb Mortal Wkly Rep. 1998;47(17):351354.
16. Malilay J. Public health assessments in disaster settings: recommenda-

216

17.
18.
19.

20.

21.

tions for a multidisciplinary approach. Prehosp Disaster Med. 2000;15
(4):167-172.
Noji EK. Public health issues in disasters. Crit Care Med. 2005;33(1)
(suppl):S29-S33.
Victoriano AF, Smythe LD, Gloriani-Barzaga N, et al. Leptospirosis in the
Asia Pacific region. BMC Infect Dis. 2009;9(1):147.
Centers for Disease Control and Prevention (CDC). Morbidity and mortality associated with Hurricane Floyd—North Carolina, SeptemberOctober 1999. MMWR Morb Mortal Wkly Rep. 2000;49(17):369-372.
Federal Emergency Management Agency (FEMA). American Samoa Disaster History. 2010. http://www.fema.gov/news/disasters_state.fema?id
=60. Accessed November 2, 2011.
Dunbar P, Weaver CUS. States and territories national tsunami hazard
assessment: historical record and sources for waves. Boulder, CO: National Oceanic and Atmospheric Administration and US Geological Survey; 2008:1-59.

Disaster Medicine and Public Health Preparedness
©2012 American Medical Association. All rights reserved.

VOL. 6/NO. 3