Galveston’s
Vulnerable Population Report Card
Let not the needy, O Lord, be forgotten; Nor the hope of the poor be
taken away.
BCP 1979
Two of
the core values at UTMB are a burning commitment to the health needs of
all Texans, regardless of their ability to pay, and a commitment to make
our community a better place to live and work. These values imply that
we must create a caring and healing environment, both at the University
and in its surrounding community. The 2000 Census estimated that 12,284
individuals, or 22.3% of the residents of the City of Galveston, lived
in poverty. The Director General of the World Health Organization has
argued that equity in health will result in significant reductions in
poverty.[1]
The health status of this population is largely unknown. This report
card on the vulnerable population of Galveston measures a portion of the
achievement of our commitments.
The
report cards are simply a collection of health indicators, some
qualitative, some quantitative. As such they a key step in the
Community Health Improvement Process.[2]
A major criterion for the data to be included in the Report Card is that
they correspond specifically to the Healthy People 2010[3]
(HP 2010) goals and objectives, or to goals and objectives identified by
community coalitions. This Report Card includes a narrative describing
the methodology and findings along with detailed tabulations. In
addition, the report card includes grades indicating the level of
attainment of the HP 2010 goals and objectives3.
Galveston’s Vulnerable Population Report Card is a quantitative
report on the indigent and medically vulnerable population of
Galveston. It contains data from existing databases such as the 2000
Census and, where possible, the administrative databases of the social
support agencies and the medical center. One of the major findings is
the gaps in potentially available data from these agencies. The project
relied upon guidance by a local committee of stakeholders who helped
select the relevant indicators and assisted with the analysis and
interpretation of assembled data. The project was guided by the goals
and objectives of Healthy People 20103 along with
goals identified by community coalitions. In addition the Report Card
presents extensive mapping of these indicators at the neighborhood
level. The Report Card presents in a consolidated summary of the status
of the City’s poor and vulnerable. It is a model for other communities
in the county and the State of Texas.
The
method for developing the Report Card was similar to that developed for
the Children’s Report Card[4].
Three related issues are associated with developing a report card:
- The formation of a
community coalition to produce the report card;
- The establishment of
criteria for the report card; and,
- The gathering, editing
and publishing of data, which constitutes the report card.
The formation of a
community coalition to develop a report card is not a simple
proposition, but the experience gained in the Children’s Report Card and the Elderly Report Card[5]
indicated that the community resources and interest exist in Galveston.
In addition, the City Council had formed a committee, Children, Youth
and Families Board. This board facilitated a series of focus groups
that identified many needs of the vulnerable. Finally, the United Way
of Galveston member agencies have identified additional and overlapping
needs.
The
working committee began by obtaining input from local agencies and
service organizations. Since the report card is to be population based
and quantitative, a biostatistician led the project. An epidemiologist
from the county health district was recruited to support the project.
The vulnerable population was represented by participation from several
agencies including the Galveston Partnership for Better Living, Jesse
Tree, and St. Vincent’s Episcopal House.
The most
difficult task of the committee was the empirical definition of a
vulnerable population. The committee quickly reached consensus that in
the broadest sense, this is the population of persons who have lost hope
for the future. While hopelessness cannot be quantified, there are a
number of quantitative indicators that may represent populations at risk
of hopelessness. Examples of these include out migration, poverty, lack
of education, and disability.
The
most critical policy decision concerns the criteria for inclusion of
data. The target population consists primarily of the poor who reside
in the City of Galveston. The criteria for including data are that they
are: local, comparable, believable and useful. The first criterion,
local data, means that the indicators must be available, either in fact
or in principle, at the local level. No new data were gathered because
of cost and time constraints.
The second
criterion, comparability, means that a comparison data set is
required. Three possibilities emerge. The most desirable would be time
series data to establish the direction of the indicators: are things
getting better, getting worse, or stable. The first report card will
serve as a baseline against which change can be measured.
A
second set of comparisons would be to the baselines reported by Healthy
People 20103. The list of reportable indicators will be
reduced based on what is available. Where critical indicators are found
to be unobtainable these will be included as indicators for future
development.
Another type of comparison is to other cities, the county, state, or
nation. Data are more readily available for larger governmental units.
However, care must be taken to ensure the same indicators are being
compared at the various levels.
A
fourth set of comparisons was across the neighborhoods of the City. The
City government including the City Council had established eleven
neighborhoods for the Community Development Block Grant program. We
designated three additional neighborhoods, so as to completely cover the
island. Unfortunately, these original eleven neighborhoods did not
match exactly with the tracts or block groups specified for either the
1990 Census or 2000 Census. In order to use census data at the
neighborhoods, we had to slightly modify the city’s boundaries of the
neighborhoods. This resulted in a minimal reallocation of populations.
The resulting linking of 2000 Census block groups into neighborhoods is
shown in Table 1 and Map 1. Another difficulty is that block group
level data cut across the Old Central and Central Business District
neighborhoods. As a result for most maps and tables, these two have
been combined. This will be referred to as the Old Central – Strand
neighborhood.
The third
criterion was for the indicators to be based on believable data,
that is, from reputable sources. This means that we will rely upon data
that are in the public domain and have been subjected to some form of
public review and assessment. Most of the data that met this criterion
can be obtained from government agencies, either in traditional printed
reports or from the Internet with well-documented and maintained
websites. Unfortunately, we found that most agencies either did not
have current data systems or were unwilling to release data because of
confidentiality concerns. We report on the available data and compare
it to other indicators where possible.
The major
source of data was the 2000 Census. These data began to be released in
2001 and extensive data from the ‘long form’ census was available.
Currently, extensive block level data by age, race-ethnicity and gender
are available. The data allowed the detailed mapping of poverty, need,
and disability in the City of Galveston. These are complex files that
require significant prior experience, so an expert in mapping and census
technology was included on the report card team.
The fourth
criterion is that the indicators be useful. This means that the
indicators must be readily understandable to policymakers. In addition
inferences drawn from a change or difference in an indicator should be
reasonably unambiguous. The indicators should measure a specific
process or outcome. This allows for program evaluation. Ideally, a low
rating on an indicator would suggest a program of action for
remediation.
The process
of collating and interpreting local data has a number of limitations and
potential pitfalls. The most important relate to confidentiality and
‘statistical validity.’ The confidentiality issue is largely resolved
by the decision to use publicly available data. However, it is possible
that the specificity of the report card in a particular area for the
vulnerable population may bring to light data that identifies specific
individuals. Thus, while unintentional, a presumed breach of
confidentiality can occur. Persons working on the data preparation for
the report card should be sensitive to this possibility. If a sensitive
area is breached, it is the obligation of the working group to bring it
to the attention of the larger oversight committee. In such cases the
issues will also be brought to the attention of a UTMB Institutional
Review Board for guidance. Even an accidental breach of confidentiality
can produce a negative local reaction, which could undermine the
credibility of the entire process.
The second
issue concerns ‘statistical validity’ particularly sample size. Some
indices, such as disease incidence at the local level, may be based on
so few cases that substantial variation can occur from year to year.
Aggregating several years of data can smooth this natural variation.
Such smoothing needs to be done cautiously, lest significant spikes be
overlooked. On the other hand, occasionally local populations are small
enough that complete censuses are feasible as an alternative to the
complications associated with surveys. The use of administrative data
from the Jesse Tree is an example of such a census.
Results
The
2000 Census reported a population of 57,247 in the City of Galveston
(Table 2). The reporting of race and ethnicity on the census forms is
complex. First Hispanic origin is ascertained and then one or more
races are reported. Unless otherwise specified in this report, the term
Hispanic indicates Hispanic origin regardless of race. White refers to
Non-Hispanic with only white race indicated, black refers to
Non-Hispanic with only black race indicated, and other refers to all
other races of non-Hispanic origin. Galveston’s population included
14,422 (25.2%) persons who described themselves as non-Hispanic black,
14,753 (25.8%) who described themselves as Hispanic, 25,277 (44.2%) who
described themselves as non-Hispanic white, and 2,795 4.9%) described
themselves as other races or multiple races, not of Hispanic origin.
Two thirds of the other group reported themselves to be Asian.
Galveston consists of many diverse neighborhoods. In this Report Card,
we adapted the list used by the City of Galveston’s Block Grant Program.
We added several neighborhoods such as the West End and Lake Madeline to
provide complete coverage of the island. The approximate boundaries are
shown in Table 1 and Map 1. The neighborhoods range is population size
from 1,593 (Central Business District) to 5,842 (Lasker Park) (Table
2). Because of the small size and structure of Census block groups, the
Central Business District (Strand) and Old Central were combined for
many of the summaries. All of the ethnic groups are represented in all
of the neighborhoods. Blacks are the majority population in Carver Park
and Old Central. They are the largest group in Kempner Park. Hispanics
are the plurality in Bayou Shore. All others have a plurality of
whites, with Fort Crockett, Central City, Offatts Bayou, Lake Madeline
and the West End having a white majority.
A key
element of hopelessness is the sense of being left behind. For this
reason we estimated the net migration from Galveston; which experienced
a reduction of 3.1% from the 1990 population. The changes varied from
declines of 13.7% among non-Hispanic blacks and 10.5% among non-Hispanic
whites to increases of 16.6% among Hispanics and 91.3 % among others.
The changes varied dramatically across neighborhoods. Of the 14
neighborhoods, all but 3 experienced a loss of population. The largest
neighborhood percentage declines were in Old Central (36.4%) and Offatts
Bayou (23.8%). Two neighborhoods, the Central Business District
(Strand) and the West End experienced increases in population of 64.1%
and 31.6% respectively.
Four
social indicators from the 2000 Census are shown in Table 3, where the
City of Galveston is compared to Texas City, League City, Galveston
County, Texas, and the United States. In the City of Galveston, 21.5%
of individuals live in poverty. This is the highest of the areas being
compared, which range from 4.7% in League City to 14.6% in Texas City.
Over 9% of households in the City of Galveston and Texas City report
receiving either Supplemental Security Income (SSI) or Public Assistance
Income. This compares to 7.8% for the United States, 7% for Texas, 6.8%
in the County and only 2.2% in League City. When public assistance is
compared to poverty, we see the ratio of assistance to poverty
households is 0.45. This is lower than Texas (0.5), League City (0.52)
or the United States (0.66).
Among
persons ages 25 years and older, 25.6% of the City’s population has less
schooling than a high school diploma. This is similar to Texas (24.3%)
but higher than the US (19.6%), the County (19.1%) and League City
(9.1%). The proportion of males and females with less than a high
school education is roughly the same. In terms of high education,
nearly of half of Galvestonians ages 25 and over report an education
beyond high school. This proportion is comparable to the United States
and Texas, greater than Texas City, but markedly lower than League City
(70.9%).
There
are 1,523 persons who are linguistically isolated in the City of
Galveston. Linguistic isolation means one resides in a household in
which no person 14 years or older speaks English either well or very
well. This is 6.4% of the households which is higher than the county
(3.8%) and the nation (4.1%) but lower that for Texas (7.2%).
The
Census allows individuals to report the presence of disabilities lasting
six months or more (Table 4). These include mental, self care, going
outside the home and work disabilities. In the City of Galveston one
fifth of the population reports one or more of such disabilities, this
is similar to Texas City (22.1%), Galveston County (18.3%), Texas
(18.7%) and the United States (19%). This is much less common in League
City (12.4%). The self report of a mental health disability indicates
difficulty lasting six months or more with learning, remembering or
concentrating. The rates vary between 2.4% in League City and 5.5% in
Texas City. These rates are similar to the rates of sensory
disabilities and about half the rate of physical disabilities.
Approximately one fifth of persons aged 65 years and older in Galveston
report the need assistance when going outside alone to shop or visit a
doctor’s office. This rate is essentially the same for the county,
state and nation.
While
not a direct indicator of a population’s vulnerability, housing stock is
critical for maintaining a healthy community. The City of Galveston is
noted for its older homes and in fact 22.1% of the structures were built
prior to 1940 (Table 5). This contrasts sharply with Texas City (3.4%),
League City (1.1%), Texas (5.4%) and the nation (15%). As a corollary,
the City has relatively little housing built between 1980 and 2000,
22.6%. This is similar to Texas City (23.7%), but much lower than League
City (70.8%), the county (38.3%), the state (43.3%) and the nation
(32.8%). A related marker of potential housing instability is the
proportion of renter occupied units. In the City of Galveston, 56.4% of
the units are renter occupied, which is in sharp contrast to Texas City
(36.6%), League City (22.8%), the County (33.8%), the State (36.28%) and
the Nation (33.8%)
Social Indicators –
Comparing Neighborhoods (Tables 6 through 10, Maps 4 through 7)
More
people live in Poverty in Galveston (21.5%) than in either Texas (15%)
or the Nation (12%). However, poverty is concentrated in the east end of
the island and north of Broadway (Table 6 and Maps 4 and 5). Carver
Parker (48.4%) has the highest proportion and numbers of persons (2,463)
living in poverty. Nearly two thirds of the children in Carver Park
live in poverty. The second highest concentrations of poverty are in
the East End (29.8%) and Old Central- Strand (29.4%). Again it is the
children that are the most affected with 48.4% and 62.9% of children
respectively. As one moves south of Broadway and to the west the levels
of poverty decline markedly to 13.3% in Central City, 12.7% in Fort
Crockett, 12.3% in the West End and 9.7% in the Lake Madeline
neighborhood. Put differently, these four neighborhoods have a total of
1,802 persons in poverty compared to the 2,463 in Carver Park.
Similarly, these four neighborhoods have a total of 443 children living
in impoverished families compared to 1,137 in Carver Park.
Public
assistance reported by less than half as many households as are in
poverty (9.6%). The distribution of reporting follows the distribution
of the level of poverty, 20% of the households in Carver Park receive
public assistance and 21.9% of the households in Old Central- Strand
receive public assistance (Map 6). Interestingly, some neighborhoods
have levels of poverty that are similar to the city, but have low levels
of public assistance, for example Offats Bayou has 15.4% of the
household below poverty, but only 2.4% receive public assistance. Lake
Madeline had the lowest proportion of households below poverty (9.4%),
but 6.6% of all households report receiving public assistance. Put
differently, in Offats Bayou 15.6% of poor households received
assistance. In contrast, 70% of the poor households in Lake Madeline
receive assistance.
The
Healthy People 2010 Objective is to increase high school completion to
90 percent, where the national baseline for persons 18-24 in 1998 was
85%. While exactly comparable figures are not available for 2000, for
the City of Galveston, 76.4%% of the population 25 and over had
completed high school which is lower than the national figure of 80.4%
(Table 7). This varies across neighborhoods from 57.6% in Carver Park
to 88.7% in Lake Madeline.
Approximately one fifth of Galvestonians report living with one or more
disabilities (Table 8, Map 7). In comparison with other indicators,
this is remarkably uniform across the island with of low of 16.4% in
Lake Madeline to 23.6% in Kempner Park and San Jacinto. Approximately
2,700 people report mental health disabilities (5.1%). This varies from
3.4% or 3.4% in Lindale, Offats Bayou and the West End to 7.4% in San
Jacinto. ‘Go-outside-the-home’ disability among persons aged 65 years
and older is 22% across the City and varies from 12.4% in Lake Madeline
to 33.1% in Bayou Shore.
The
Healthy People 2010 Objective 6-8 is to eliminate disparities in
employment rates between working-aged adults with and without
disabilities. The target is 82% employment of adults aged 21 through 64
with disabilities which is parity with adults without disabilities in
1994-1995. In Table 9 the proportion of the disabled in the City who
are employed is 52.6% compared to 70.3% of the non-disabled. The lowest
employment of the disabled is in Old Central - Strand (37.8%). The
highest proportion of the employment among the disabled is in Central
City and Lake Madeline 66% and 65.9% respectively.
Housing age and rental status may be indicators of housing instability,
which is associated with homelessness and poverty. In Galveston 22.1%
of the homes were built prior to 1940 and 22.6% in 1980 or later. As
shown in Table 10, these vary dramatically across the island. Carver
Park contains 70% renter occupied units and 60.6% of all units were
built after 1980, this is similar to Central City which is 78.5% rental
and 54.5% built since 1980. Kempner Park, Old Central- Strand and San
Jacinto have approximately 50 percent of the units built before 1940.
East End, Old Central and San Jacinto are more than 60.6% rental units,
while Kempner Park is 49.9% rental. No neighborhood on the island
reaches the county and national rates of 66% owner occupied.
As noted in the methods section, major effort was made to obtain data on
service delivery from the social service agencies in the city of
Galveston. By and large, almost no data were available to characterize
in detail the delivery of services, other than health care. The most
sophisticated system non governmental system was found at the Jesse
Tree, which maintains an Internet based delivery system. Table 11 shows
the cross tabulation of service contacts for ‘unduplicated
individuals.’ Approximately, 907 different people were registered at
the Jesse Tree in 2003. Of these, most (53%) were women, 51% were
between 18 and 44, and most were of unknown ethnicity. Map 8 shows the
delivery of services throughout the island, but most commonly in the
neighborhoods nearest to the Jesse Tree office in the Old Central –
Strand neighborhood.
Galveston Senior Citizen Program delivered meals to 168 addresses on
April 14, 2003. Paralleling this, the Meals on Wheels at Edgewater
program delivered meals to 72 addresses on April 1, 2003. This should
be compared to the 1,673 individuals 65 years of age and over who
reported a ‘go-outside-the-home’ disability in 2000. As with Jesse
Tree, meals were delivered primarily to neighborhoods in the center of
the island (Map 9).
Many
of the social service agencies summarize data in the number of units
delivered rather than persons served. For example between Dec 31, 2001
and January 2, 2003, Galveston Catholic Charities reports having served
4,199 clients in the City (Table 12). The specific services are also
tabulated for the county. For example 2,584 clients received service
from the food/rent program, 2,593 received food pantry assistance, and
323 received emergency financial program assistance. Limited
demographic data for the County are also available. For example
Galveston Catholic Charities report age data on 5,821 clients, more than
half of whom (3,202) are between 65 and 79. Similarly more than half of
the clients (3,044) are black, and more than half are female, and of the
2,694 with income data, most (1,765) report a household income of under
$5,000.
Data
from the Galveston County Social Services were obtained only for the
months of April and November 2002. These data summarized caseworker
interactions with clients. There were a total of 273 and 315 for the
respective months. Of these, 41.5% were at the City of Galveston
Office, 27.9% were in Texas City and 30.6% were in League City.
The
Gulf Coast Homeless Coalition has inventoried available beds for
individuals (142 beds) and families (144 beds). They have compared this
with an estimate of need, 2993 beds, which leaves a short fall of 2,707
beds. The largest estimated need is for emergency shelter (1194 beds)
but the biggest shortfall is for permanent supportive housing (930
beds).
Health
insurance is provided to indigent children in Texas through the Child
Health Insurance Program or CHIP. In the city of Galveston, for 2000
through 2002 there were 2,339 applications and 1,024 enrollees (Table
13).
The City of
Galveston is home to the University of Texas Medical Branch at
Galveston. City residents had 7,725 inpatients stays of which 9.9% were
‘self pay’ in 2002 (Table 14, Map10). The in patient rate varied from 9
per hundred residents in the West End to 19 per hundred residents in San
Jacinto. The self pay percentage varied from 4% among Lindale in
patients to 14% by Old Central- Strand (Strand) in patients.
There
were 18,465 out patient visits by residents for a rate of 32.24 per
hundred residents. The rate varied from 20 per hundred residents from
the Old Central - Strand neighborhood to 55 per hundred residents from
the East End. The East End also had the greatest absolute number of out
patient visits. The proportion of self pay visits by the Island’s
population was 12%, which varied from a low of 7% among Lake Madeline
residents and 8% among East End residents to a high of 19% of visits by
Lasker Park residents.
Report cards serve a central role in the Community Health Improvement
Process2. It is a significant, but surmountable,
challenge to find data that are comparable, believable and useful at the
local level for vulnerable populations. For counties, states and the
nation data are often accessible. For cities that are smaller than
counties, especially non-metropolitan cities, data are generally quite
limited. We found in the Children’s Report Cards that input from
a broad coalition revealed sources of data that while not definitive,
are generally quite useful. Individual members of the coalition could
not have predicted all the specific sources of data, but collectively a
rich data archive was developed. As we moved forward to the Report Card
for the Vulnerable, we found the use of community coalitions to identify
and select indicators, as well as interpret them, insured the utility of
the Report Card.
Report
cards are no better than the available data. The 2000 Census provided a
rich source of data on economic status, education and disability as well
as housing conditions. However, the data are largely self reported and
only available at ten year intervals. A more timely data system can be
obtained from the service providers such as the Jesse Tree or UTMB.
These organizations provided extensive data that could be compared
across neighborhoods. Other agencies were generally found to have
limited data that could be accessed on a timely basis. These agencies
should be encouraged to publish more extensive neighborhood specific
data. Unfortunately, for many key indications of vulnerability, such as
homelessness, data were either sparse or found to be unreliable.
With the
completion of the report card two issues remain: sustainability and
exportability to the wider community. The issue of sustainability is
critical to the process. Fielding, Sutherland and Halfon[6]
note that most report cards have been single reports. We have been able
to produce annual Children’s Report Cards for six consecutive years.
Fielding, Sutherland and Halfon6 identify broad community involvement as an
important factor for success. Annual report card production has
promoted recognition and acceptance of the report card by community
leaders as useful tools for planning and evaluation. It has been the
experience of the Children’s Report Card that the community has adopted
it as a criterion for program evaluation. This same reaction should
sustain the Report Card for the Vulnerable Populations beyond its
immediate period of funding.
The
second issue is exportability. Clearly, local knowledge is critical.
The Report Card Project is in discussion with the Gulf Coast AHEC and
the Texas City Youth at Risk Program to develop report cards similar to
those that have been developed in Galveston. In addition, the Jesse
Tree network has been extended to agencies in Texas City. This
reassures us as to the feasibility of exporting the Report Card for
Vulnerable Populations to the county and the state.
In
spite of the important limitations of data at the local level, the use
of report cards is critical to the Community Health Improvement Process2. At a
minimum, they provide quantitative boundaries within which focused
discussions about problems and solutions can take place. Ideally,
report cards identify specific areas where programs have succeeded and
other areas where attention needs to be placed.
The Report Card is an essential part of the Community Health Improvement
Process. The CHIP model is designed specific to create health caring
environments for all people who work and live in a community. When the
Report Card is completed, published and disseminated, we expect to see a
change in the character and quality of the discussions concerning the
vulnerable populations in the city of Galveston. The discussions should
become more focused and directed. This should have a positive effect on
programming planning. The clearest short-term indication of the success
of the project is its ability to affect public discussion such as in the
news media. This has occurred for the previous report cards and we
believe it will occur again.
In
conclusion, Galveston has a substantial vulnerable population. This is
reflected in part in the substantial out migration of the white and
black populations, particularly from the most impoverished
neighborhoods. The levels of poverty markedly exceed those of the
surrounding towns, county and state. Public assistance is not
correlated with levels of poverty. It is disturbing that the level of
public assistance per poor household is lower than in richer communities
such as League City and markedly below that of the United States.
Galveston has education levels comparable to the county and state, but
lower levels than the United States. Disability is similar to that of
the county, state and nation. The housing stock is relatively old and
disproportionately renter occupied. The island residents make use of
available health care facilities. Taken together these observations
indicate a need to address poverty and rental housing. In addition, the
clear documentation and coordination of service delivery is critical for
the effect support of our vulnerable population.
Footnotes
PMCH