Skip Navigation U.S. Department of Health and Human Services www.hhs.gov
Agency for Healthcare Research Quality www.ahrq.gov
Archive print banner

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: https://info.ahrq.gov. Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to www.ahrq.gov for current information.

Chapter 6 - Community Context

Introduction

The safety net is influenced by a wide variety of community characteristics in addition to those specifically related to the health care system. Population size and composition, the economy, living arrangements, and crime rates all influence the structure and functioning of communities and determine the context in which the safety net functions. We include several measures of community context in the data books, including:

  • Population size, density, and growth.
  • Age distribution.
  • Racial/ethnic distribution.
  • Indices that capture the extent to which people from different racial/ethnic and economic groups live near one another.
  • Foreign-born population.
  • Household income.
  • Unemployment.
  • Living arrangements.
  • Home ownership and housing vacancy.
  • Education.
  • Major crime rates.

Additional details on each of these measures is included in "Appendix A: Technical Information."

Return to Contents

Variation in Community Context

Population

Population size, density, and growth vary considerably across Metropolitan Statistical Areas (MSAs) and counties throughout the country. The size and density of the population affect the magnitude, distribution, and location of safety net services needed in a given area. Table 6-1 displays averages for the areas included in this data book. The MSAs included in this book range in size from 350,761 residents (Trenton, NJ) to 9,519,338 residents (Los Angeles-Long Beach, CA). The average MSA has 1.64 million residents. MSAs also differ substantially in geographic area, with a low of 269 square miles in Trenton, NJ, and a high of 39,886 square miles in the Las Vegas, NV-AZ, MSA. Population density is lowest in the South and Midwest, with 974 and 1,107 residents per square mile, respectively.

A safety net in an area facing considerable population growth is likely to address different health care needs than one in a community facing a declining population. Population growth between 1990 and 2000 was the greatest in the South and West. Figure 6-1 shows population growth rates for the MSAs included in this book, with a subset of the MSAs labeled. The Las Vegas, NV-AZ, MSA experienced by far the fastest population growth rate at 83.3 percent. Other areas—such as Phoenix, AZ, Washington, DC, Orlando, FL, and Denver, CO—also experienced substantial population growth. In contrast, the Pittsburgh, PA, metropolitan area saw a decline in its population over the past decade.

Table 6-1: Population Size, Density, and Growth by Area Type and Region

Area Average for Area
Population Size,
2000
Population per
Square Mile 1999/2000
Population Growth
1990-2000
MSA1,644,52388815.9%
Suburban County267,464703 18.2%
City322,2634,12211.4%
Northeast379,6982,7565.3%
South190,81397423.0%
Midwest256,3991,10710.9%
West379,5242,66821.9%
All Areas284,0831,74815.9%

Figure 6-1: Population Growth Metropolitan Areas, 1999-2000

Figure 6-1: Population Growth Metropolitan Areas, 1999-2000
[D] Select for Text Description.

Race/Ethnicity and Indices of Racial/Ethnic and Economic Dissimilarity

Variation in community characteristics is as great as variation in size. Across all the MSAs included in this book, 70.1 percent of the population is white, 13.9 percent black, 5.5 percent Asian, and 10.5 percent Native American, Hawaiian/Pacific Islander, two or more races, or other races (Go to Table 6-2). Within MSAs, however, racial/ethnic composition differs considerably in cities and in suburban counties. On average, 78.7 percent of the population of suburban counties is white, and 9.6 percent is black. In contrast, the population of the cities included in this book is 53.7 percent white and 22.0 percent black. Similarly, the Hispanic population is more heavily concentrated in cities than in suburban areas. Reflecting demographic patterns nationwide, the Southern areas included in this book have a larger black population than do other areas, and the Western areas have by far the largest Hispanic population.

Table 6-2: Racial/Ethnic Composition by Area Type and Region, 2000

Area Average for Area
Percent of Population White Percent of Population Black Percent of Population Asian Percent of Population Other* Percent of Population Hispanic
MSA70.1%13.9%5.5%10.5%15.1%
Suburban County78.7%9.6%4.1%7.5%10.8%
City53.7%22.0%8.0%16.3%23.4%
Northeast73.0%13.9%4.6%8.5%11.7%
South70.7%21.9%2.6%4.8%9.6%
Midwest75.1%15.6%3.1%6.2%8.2%
West64.3%6.2%10.0%19.6%26.7%
All Areas70.1%13.9%5.5%10.5%15.1%

* Includes Native American, Hawaiian/Pacific Islander, two or more races, and other.

Table 6-3 displays an index of racial dissimilarity, representing the proportion of the non-white population that would need to move in order for all ZIP Codes in an area to have equal proportions of the non-white population. On average, the racial dissimilarity index is 0.451 across all MSAs included in this book, varying from a low of 0.175 in Spokane, WA, to a high of 0.699 in Detroit, MI. While there tends to be more racial dissimilarity in cities (0.410 vs. suburban counties at 0.315), the spread between the lowest and highest areas is considerably larger in the suburbs. Variation in the racial dissimilarity index is somewhat greater in the Midwest (0.415) and West (0.415) than in the Northeast (0.324) or the South (0.397). However, the difference between the lowest and highest indices is by far the greatest in the Midwest, where Detroit, MI (0.699) and Milwaukee, WI (0.655) have considerably more racial dissimilarity than Kalamazoo, MI (0.279).

Table 6-3: Racial Dissimilarity Index (Non-White Population) by Area Type and Region, 1999

Area Range of Variation
Indexa High Low High/Low Average
MSA0.2400.6990.1753.980.451
Suburban County0.3630.5970.005119.210.315
City0.3840.6300.03219.450.410
Northeast0.3240.6130.05710.820.406
South0.397 0.6300.01157.18 0.308
Midwest0.4150.5970.005119.210.349
West0.4150.5560.03217.160.328
All Areas0.398 0.6300.005125.68 0.347

a Coefficient of variation: an index that measures the amount of variation (higher = more variation).

The economic dissimilarity index shown in Table 6-4 measures the percent of families with incomes less than $15,000 per year that would have to move in order for all ZIP Codes in an area to have an equal proportion of low-income population. The economic dissimilarity index also indicates the extent to which the need for safety net services may be spread evenly over the metropolitan area (a low value) as opposed to the extent to which the population using these services may be more concentrated in particular parts of the area (a high value). There is remarkably little difference in economic dissimilarity between cities and suburban counties. There is somewhat more variation in economic dissimilarity in the South than in other areas of the country. However, both the South and the West have a greater than 20-fold difference between the lowest and highest communities in the area. Figure 6-2 displays the economic dissimilarity index for the MSAs included in this book, with a subset of the MSAs labeled. Areas such as Newark, Philadelphia, Baltimore, Atlanta, Milwaukee, Detroit, Chicago, and Denver have the highest economic dissimilarity rates in their respective regions of the country. Those areas with the lowest economic dissimilarity indices—such as Jersey City, Scranton, Johnson City, and Modesto—tend to have very high poverty rates throughout the metropolitan area.

Table 6-4: Economic Dissimilarity Index by Area Type and Region, 1999

Area Range of Variation
Indexb High Low High/Low Average
MSA0.2420.4740.1283.700.305
Suburban County0.2860.4200.01922.150.224
City0.2800.4310.01823.340.264
Northeast0.2700.3900.0626.350.247
South0.3200.4200.01922.150.225
Midwest0.2870.4120.03711.130.252
West0.2920.4310.01823.340.232
All Areas0.2950.4310.01823.340.238

b Coefficient of variation: an index that measures the amount of variation (higher = more variation).

Figure 6-2: Economic Dissimilarity Index Metropolitan Areas, 1999

Figure 6-2: Economic Dissimilarity Index Metropolitan Areas, 1999
[D] Select for Text Description.

Figure 6-3 shows the relationship between the racial and economic dissimilarity indices. The association between the two measures is very strong, with a 0.580 correlation. Metropolitan areas such as Milwaukee, WI, Detroit, MI, and Newark, NJ, tend to be high on both measures and would require a large number of individuals from different racial and economic groups to move in order to achieve an equal geographic distribution of the population by these two characteristics.

Figure 6-3: Association Between Racial and Economic Dissimilarity Indices Metropolitan Areas, 1999

Figure 6-3: Association Between Racial and Economic Dissimilarity Indices Metropolitan Areas, 1999
[D] Select for Text Description.

Foreign-Born Population

The extent to which the safety net needs to provide services in a variety of languages to individuals from various cultures has an impact on how care is organized and the types of providers and staff needed to meet individuals' needs.

Table 6-5 shows the foreign-born population of the areas included in this book. On average, 15.7 percent of the population was born in a country other than the United States, ranging from a low of 1.2 percent in the Johnson City-Kingsport-Bristol, TN-VA, MSA to a high of 50.9 percent in Miami, FL. As with many of the other measures, this measure varies considerably by location. Cities have a considerably higher proportion of their population coming from other countries than do suburban areas (23.1 percent vs. 11.9 percent). However, the variation among suburban counties is far greater, with an index of 0.797 and a more than 175-fold difference between the highest and lowest areas. While the West has the highest proportion of foreign-born residents (22.3 percent), variation is greatest in the South, with an index of 1.035. Across all the areas included in this book, 40.6 percent of the foreign-born population comes from Latin America, 29.4 percent from Asia, and 21.8 percent from Europe (Go to Figure 6-4).

Table 6-5: Percent of Population Foreign Born by Area Type and Region, 2000

Area Range of Variation
Indexc High Low High/Low Average
MSA0.71950.9%1.2%42.1915.7%
Suburban County0.79746.4%0.3%175.6611.9%
City0.58672.1%1.6%45.9723.1%
Northeast0.77343.9%0.6%78.0316.0%
South1.03572.1%0.3%273.2111.8%
Midwest0.69921.7%0.4%55.739.5%
West0.51054.4%1.8%30.3822.3%
All Areas0.77872.1%0.3%273.2115.7%

c Coefficient of variation: an index that measures the amount of variation (higher = more variation).

Figure 6-4: Distribution of Foreign-Born Population All Ages Cities, County Residuals, and Suburban Counties, 2000

Figure 6-4: Distribution of Foreign-Born Population All Ages Cities, County Residuals, and Suburban Counties, 2000
[D] Select for Text Description.

More than 10 percent of the population in the areas included in this book speak English less than "very well" (Go to Table 6-6). Metropolitan areas such as Miami, FL, Jersey City, NJ, and Los Angeles, CA, have the highest rates among the areas included in this book. Again, the proportion is far higher in central cities (17.1 percent) than in surrounding suburbs (7.3 percent), although variation is considerably greater in the suburbs. As with the proportion foreign-born, variation in the proportion of the population speaking English less than "very well" is greatest in the South, reaching 59.3 percent in the city of Hialeah, FL. While the variation in the limited-English-proficiency population is lowest in the West, it averages the highest rate among the regions at 16.7 percent.

Table 6-6: Percent of Population Who Speak English Less Than "Very Well" by Area Type and Region, 2000

Area Range of Variation
Indexd High Low High/Low Average
MSA0.76234.7%0.8%43.0810.7%
Suburban County0.85031.9%0.2%131.217.3%
City0.60959.3%1.3%44.0617.1%
Northeast0.81236.8%0.7%52.5210.5%
South1.12159.3%0.2%243.876.2%
Midwest0.81719.4%0.6%30.146.7%
West0.58153.1%0.9%58.1116.7%
All Areas0.86259.3%0.2%243.8710.7%

d Coefficient of variation: an index that measures the amount of variation (higher = more variation).

Living Arrangements and Housing

Nearly 10 percent of the population in the areas included in this book lives alone (Go to Table 6-7). Living alone is somewhat more common in cities than in suburban counties. Among families with children, 29.1 percent are single-parent or nonmarried couple households, again with a higher rate in the cities (38.0 percent) than in the surrounding suburban areas (24.5 percent). Within cities, the prevalence of these families ranges from 9.9 percent in Naperville, IL, to 66.1 percent in Harrisburg, PA.

Table 6-7: Living Arrangements by Area Type and Region, 2000

Area Average for Area
Percent of Families
With a Single
Parent or
Nonmarried Couple
Percentage of
Population
Living Alone
MSA29.1%9.7%
Suburban County24.5%8.8%
City38.0%11.4%
Northeast29.8%10.5%
South31.0%10.0%
Midwest28.5%10.1%
West27.2%8.6%
All Areas29.1%9.7%

As Table 6-8 shows, more than 60 percent of housing in the MSAs in this book is owner-occupied, ranging from 47.9 percent in central cities to 70.9 percent in suburban counties. Housing vacancy rates average less than 5 percent but vary considerably, ranging from a low of 1.3 percent in Livonia, MI, and Anoka County, MN, to a high of 16.3 percent in St. Louis, MO.

The South and West have the greatest proportion of newer housing stock, while the Northeast and Midwest have considerably more housing that is more than 30 years old. Paralleling its tremendous population growth, the Las Vegas, NV-AZ, MSA has by far the greatest proportion of housing stock that is less than 10 years old (47.0 percent).

Unemployment, Education, and Crime

Across the areas included in this book, 5.9 percent of the population is unemployed. As shown in Figure 6-5, the unemployment rate ranges from a low of 3.0 percent in Lancaster, PA, to a high of 12.0 percent in the Fresno, CA, MSA.

Table 6-8: Housing by Area Type and Region, 2000

Area Average for Area
Percent of Housing Owner Occupied Percent of Housing Vacant Percent of Housing
Less Than 10 Years Old
Percent of Housing
Greater Than 30 Years Old
MSA62.7%4.9%15.3%52.0%
Suburban County70.9%4.3%18.8%44.7%
City47.9%5.7%9.4%64.5%
Northeast59.0%4.6%7.7%71.5%
South67.6%5.8%22.8%35.3%
Midwest68.6%4.7%14.6%57.2%
West59.4%4.2%17.0%44.6%
All Areas62.7%4.9%15.3%52.0%

Levels of education also differ throughout the country (data not shown). In the MSAs included in this book, an average of 44.4 percent of the adult population has a high school education or less, with a low of 30.1 percent in Madison, WI, and a high of 61.5 percent in Lancaster, PA.

Figure 6-5: Percent of Population Unemployed Metropolitan Areas, 2000

Figure 6-5: Percent of Population Unemployed Metropolitan Areas, 2000
[D] Select for Text Description.

The index crime rate (a measure of major crimes), shown in Table 6-9 includes the number of murders, forcible rapes, robberies, aggravated assaults, burglaries, larcenies, and auto thefts per 10,000 population. The rate averages 435.4 among all MSAs included in this book. Again, considerable variation exists across and within the types of areas described here. While cities have a higher average index crime rate, variation is greater in the suburbs. On average, crime rates are lowest in the Northeastern MSAs and highest in the Southern MSAs included in this book and vary the least in the West.

Table 6-9: Index Crime Ratee per 10,000 Population by Area Type and Region, 1999

Area Range of Variation
Indexf High Low High/Low Average
MSA0.280861.5169.45.08435.4
Suburban County0.4601248.020.560.87398.6
City0.3871387.720.567.68477.5
Northeast0.444684.144.115.53317.2
South0.4121248.023.852.43526.2
Midwest0.4931387.720.567.68433.2
West0.283729.6153.74.75434.6
All Areas0.4411387.720.567.68435.4

e The index crime rate includes the number of murders, forcible rapes, robberies, aggravated assaults, burglaries, larcenies, and auto thefts.
f Coefficient of variation: an index that measures the amount of variation (higher = more variation).

Return to Contents

How Community Context Is Related to Safety Net Performance and Population Outcomes

Table 6-10 displays the relationships among several of the community context measures and outcomes. At the place/county level, an increasing proportion of the non-white population is associated with a moderate to high increase in negative outcomes, including potentially preventable hospitalizations for all ages and negative birth outcomes. These relationships are less strong at the MSA level, where differences among areas within an individual MSA may be masked by figures for the total MSA. Higher racial and economic dissimilarity indices are generally associated with higher rates of preventable hospitalizations and negative birth outcomes, although they are associated with lower rates of lacking a usual source of care and having no physician visits in the past year. While the proportion of the population that is foreign born and the proportion speaking English less than "very well" have some associations with these outcomes, these relationships are weak, with typically only a slight to low association. This may reflect the fact that recent immigrants may be healthier than second-and third-generation residents of the United States.

Table 6-10: Association Between Community Context Measures and Outcomes (Place/County and MSA Levels)

Outcome Measure Association With Outcome Measures (R2)g
Percent of Population Non-White 2000 Racial Dissimilarity Index, 1999 Economic Dissimilarity Index, 1999 Percent of Population Foreign Born 2000 Percent of Population Speaking English Less Than "Very Well," 2000
Place/County Level Preventable Hospitalizations, Ages 0-170.259+0.190+0.024+0.064+0.050+
Preventable Hospitalizations, Ages 18-390.336+0.225+0.065+0.0000.002
Preventable Hospitalizations, Ages 40-640.404+0.221+0.032+0.020+0.044+
Late or No Prenatal Care0.246+0.191+0.065+0.025+0.046+
Low Birth Weight (Full-Term Births)0.323+0.195+0.057+0.0020.004
Preterm Births0.261+0.164+0.049+0.008-0.001
MSA Level Preventable Hospitalizations, Ages 0-170.086+0.107+0.0300.081+0.045+
Preventable Hospitalizations, Ages 18-390.037+0.210+0.211+0.0030.012
Preventable Hospitalizations, Ages 40-640.160+0.296+0.172+0.038+0.036+
Late or No Prenatal Care0.048+0.040+0.088+0.0160.009
Low Birth Weight (Full-Term Births)0.0170.137+0.172+0.0220.041-
Preterm Births0.0090.090+0.100+0.061-0.063-
No Usual Source of Care (Low Income)0.0010.175-0.117-0.0620.097+
No Physician Visit in Last Year (Low Income)0.0010.107-0.208-0.0230.071

g The higher the R2, the stronger the association. The "+" and "-" indicate the direction of the association. A "+" indicates that the outcome/performance measure increases as the factor increases, and a "-" indicates that the outcome/performance measure decreases as the factor increases.

The relationship among living arrangements, housing, and safety net outcomes is shown in Table 6-11. There are moderate positive associations between the proportion of the population living alone and each of the outcomes at the place/county level. A greater proportion of families with only one parent in the household is highly to very strongly associated with higher preventable hospitalization rates and higher rates of negative birth outcomes at the county level. All of these relationships hold at the MSA level, although they are less strong. These associations may be due to single parents and those living alone being less likely to take care of themselves, or it may represent a lesser extent of community "cohesion" in areas where these rates are high. Higher levels of owner-occupied housing are associated with better outcomes, particularly at the place/county level. In contrast, vacant or older housing stock tends to be associated with negative outcomes.

Table 6-11: Association Between Community Context Measures and Outcomes (Place/County and MSA Levels)

Outcome Measure Association With Outcome Measures (R2)h
Percent of Families With a Single Parent or Nonmarried Couple, 2000 Percent of Population Living Alone 2000 Percent of Housing Owner Occupied 2000 Percent of Housing Vacant, 2000 Percent of Housing Greater than 30 Years Old 2000
Place/County Level Preventable Hospitalizations, Ages 0-170.378+0.131+0.251-0.178+0.215+
Preventable Hospitalizations, Ages 18-390.651+0.221+0.203-0.449+0.197+
Preventable Hospitalizations, Ages 40-640.694+0.161+0.290-0.432+0.248+
Late or No Prenatal Care0.369+ 0.111+0.244-0.185+ 0.129+
Low Birth Weight (Full-Term Births)0.579+0.241+0.237-0.367+0.199+
Preterm Births0.501+0.140+0.101-0.458+0.070+
MSA Level Preventable Hospitalizations, Ages 0-170.251+0.097+0.134-0.0220.137+
Preventable Hospitalizations, Ages 18-390.336+0.138+0.0080.178+0.033+
Preventable Hospitalizations, Ages 40-640.409+0.036+0.093-0.105+0.103+
Late or No Prenatal Care0.206+0.034+0.099-0.039+0.007
Low Birth Weight (Full-Term Births)0.308+.153+0.0060.159+0.013
Preterm Births0.215+0.035+0.0180.349+0.016
No Usual Source of Care (Low Income)0.0730.229-0.0150.0080.339-
No Physician Visit in Last Year (Low Income)0.089-0.339-0.0440.0000.335-

h The higher the R2, the stronger the association. The "+" and "-" indicate the direction of the association. A "+" indicates that the outcome/performance measure increases as the factor increases, and a "-" indicates that the outcome/performance measure decreases as the factor increases.

Table 6-12 displays the final set of relationships between the community context measures and safety net outcomes. An increasing proportion of the population who are unemployed is moderately to very strongly associated with a higher rate of all of the negative outcomes studied at the place/county level. These relationships are maintained, although they are somewhat less strong, at the MSA level for all preventable hospitalization outcomes as well as for the rate of late or no prenatal care. Similar relationships exist for education, with an increasing proportion of the population having a high school education or less being associated with higher rates of negative outcomes. The relationship between the crime rate and health care outcomes generally follows the same pattern, although the relationships are less strong.

Table 6-12: Association Between Community Context Measures and Outcomes (Place/County and MSA Levels)

Outcome Measure Association With Outcome Measures (R2)i
Percent of Population Unemployed 2000 Percent of Adults With a High School Education or Less, 2000 Index Crime Rate per 10,000 Population 1999
Place/County Level Preventable Hospitalizations, Ages 0-170.302+ .251+0.038+
Preventable Hospitalizations, Ages 18-390.391+0.264+0.114+
Preventable Hospitalizations, Ages 40-640.583+0.409+0.076+
Late or No Prenatal Care0.306+0.140+0.042+
Low Birth Weight (Full-Term Births)0.318+0.186+0.085+
Preterm Births0.254+0.209+0.157+
MSA Level Preventable Hospitalizations, Ages 0-170.119+0.207+0.001
Preventable Hospitalizations, Ages 18-390.041+0.176+0.067+
Preventable Hospitalizations, Ages 40-640.265+0.340+0.016
Late or No Prenatal Care0.102+0.037+0.001
Low Birth Weight (Full-Term Births)0.0030.075+0.053+
Preterm Births0.0080.151+0.141+
No Usual Source of Care (Low Income)0.0000.0190.058
No Physician Visit in Last Year (Low Income)0.0030.0040.059

h The higher the R2, the stronger the association. The "+" and "-" indicate the direction of the association. A "+" indicates that the outcome/performance measure increases as the factor increases, and a "-" indicates that the outcome/performance measure decreases as the factor increases.

Return to Data Book Contents
Continue to Chapter 7

 

The information on this page is archived and provided for reference purposes only.

 

AHRQ Advancing Excellence in Health Care