Skip
To Content
|  | |||||||||||
 |
|
|||||||||||
|
2002-2004 Sub State Report of Substance Use & Serious Psychological Distress |
This report presents estimates of the prevalence of substance use or mental health problems in substate areas during 2002-2004 based on the National Survey on Drug Use and Health (NSDUH). NSDUH is an annual survey of the civilian, noninstitutionalized population aged 12 or older. The report marks a second time that detailed data for substate areas have been presented by the Substance Abuse and Mental Health Services Administration (SAMHSA). The first report to provide such estimates used data from the 1999-2001 surveys (Office of Applied Studies [OAS], 2005). This second report presents estimates for 22 measures of substance use or mental health problems based on the 2002-2004 NSDUHs. Estimates of underage alcohol use among persons aged 12 to 20 and binge alcohol use also are included in this report. These two reports provide a more detailed perspective on the variations in substance use rates both within and across States than was possible with prior State reports (Wright & Sathe, 2005, 2006).
Estimates are provided here for each of the 357 substate areas representing collectively the 50 States and the District of Columbia. These areas were defined by officials from each State and were typically based on the substance abuse treatment planning areas specified by the States in their applications for the Substance Abuse Prevention and Treatment (SAPT) Block Grant administered by SAMHSA. NSDUH is sponsored by SAMHSA and collects information from residents of households, residents of noninstitutionalized group quarters (e.g., shelters, rooming houses, dormitories), and civilians living on military bases. In 2002-2004, NSDUH collected data from approximately 204,000 respondents aged 12 or older and was stratified to obtain representative samples from all 50 States and the District of Columbia. The survey was planned and managed by SAMHSA's OAS. Data collection was conducted under contract with RTI International.1
Section A provides background on the survey, how substate areas were formed, the general methodological approach, and a brief discussion of the findings. A complete list of the 22 substance use measures presented in this report is given in Section B, which also provides further information on the small area estimation (SAE) methodology used to develop estimates presented in this report. Included in this report are measures of illicit drug use, alcohol use, tobacco use, substance dependence and abuse, substance use treatment need, and serious psychological distress. Section C includes tables with estimates for each of the 22 measures mentioned above and the corresponding prediction intervals (PIs). It also contains a set of national maps that show the prevalence of each outcome measure for each substate region. The rows in the tables in Section C have been ordered alphabetically by State, and the substate areas have been ordered alphabetically within each State. There are 10 separate tables, with most having a few related outcomes. Estimates for some aggregate regions are also included in these tables. Section D contains definitions of substate regions. Section E includes the population counts for persons aged 12 or older for each substate area and the combined 2002, 2003, and 2004 NSDUH sample sizes for these areas. Users may find these counts helpful in calculating the weighted average prevalence estimate for any combination of substate areas or to find counts of the number of people using a particular substance in a substate area. Section F lists the references.
In an appendix after Section F, we also present additional tables for three measures: past year cocaine use (low prevalence), past month marijuana use (moderate prevalence), and past month alcohol use (high prevalence). These tables show the region with the highest and the lowest prevalence estimate in each State and also the Bayes posterior probability for testing whether the two prevalences are significantly different from each other. For the 2002-2004 NSDUH response rates, see Appendix A of the 2003-2004 State report (Wright & Sathe, 2006).
The substate areas for each State were developed in a series of communications between SAMHSA staff and State treatment representatives in late 2005 and early 2006. The goal of the project was to provide substate-level estimates showing the geographic distribution of substance use prevalence for areas that States would find useful for treatment planning purposes.2 The final substate region boundaries were based on the recommendations of each State's staff assuming the NSDUH sample sizes provided adequate precision. Most States defined areas in terms of counties or groups of counties. A few States defined the areas in terms of census tracts. Several States also requested that we provide estimates for aggregate planning regions along with the estimates for their substate planning regions. A few of these States wanted the maps to be produced for the aggregate regions instead of their substate planning regions. Hence, for each measure in this report, maps were produced for 340 planning regions (and not for 357 regions).
These 340 substate areas used in the maps have been ranked from lowest to highest on each outcome measure and divided into seven categories designed to represent distributions that are somewhat symmetric, like a normal distribution—but flatter. Colors were assigned to all areas such that the third having the lowest prevalence are in blue, the middle third are in white, and the third with the highest prevalence are in red. The only exceptions were the three perception-of-risk outcomes, which have the highest estimates represented in blue and the lowest represented in red. To further discriminate among the areas that display relatively higher prevalence, the "highest" third has been further divided into three categories: dark red for the 15 areas with the highest estimates, medium red for the 30 areas with the next highest, and light red for the 68 areas with the next highest. The "lowest" third is categorized in a similar way using three distinct shades of blue. Due to tied values of prevalence, the number of substate regions in each category may vary a little.
A companion piece to this report includes national tables of prevalence estimates for youths aged 12 to 17, young adults aged 18 to 25, and adults aged 26 or older for all measures (for substate areas having sufficient precision), as well as separate maps and tables showing the substate estimates for each State separately for all persons aged 12 or older. This companion piece is scheduled for posting to the OAS website as its parts become available. Check the SAMHSA website at http://oas.samhsa.gov/geography.cfm for the projected date of availability.
Estimates in this report are based on hierarchical Bayes estimation methods that combine survey data with a national model. Applying this methodology to the State substance use measures has been shown to result in more precise estimates than using the sample-based results alone (Wright, 2002). The methodology used to produce estimates in this report is the same as that used to produce State estimates from the NSDUH data since 1999. Sample data have been combined across 3 years (2002-2004) in this report to improve the precision of substate area estimates. The estimate for each area is accompanied by a 95 percent PI (for more details, see Section B, Substate Region Estimation Methodology).
In addition to the substate area estimates, comparable estimates are provided for the 50 States and the District of Columbia using the same methodology. Because these estimates are based on 3 years of data, they are not directly comparable with the State estimates in earlier reports that are based on only 2 consecutive years. However, if a national map based on the substate areas for 2002-2004 is compared with one of the national maps based on 2003-2004 data, it can be seen that most of the States with significant areas in dark red correspond to States that were in the highest fifth in 2003-2004. For example, in 2003-2004, the 10 States with the highest State estimates of past month use of marijuana among persons aged 12 or older were Alaska, Maine, Massachusetts, Montana, New Hampshire, New Mexico, New York, Oregon, Rhode Island, and Vermont. Based on the 2002-2004 substate maps, seven of those States display at least one area in the highest category (dark red on the map), and the remaining three had one or more areas in the second highest category (medium red on the map).
Because the SAE methods used here tend to borrow strength from both the national model and the State-level random effects, estimates for substate areas with sample sizes that were closer to the minimum (200) tend to be shrunk more toward the corresponding State prevalence estimate than substate areas with large sample sizes. This methodology tends to cluster the small sample substate estimates around their State means. Relatively high estimates for small substate areas tend to be pulled down, while relatively low estimates tend to be pulled up. On the other hand, for areas with large sample sizes, the methodology produces estimates that are close to the weighted average of the sample data. In addition, the estimates have been made design consistent so that as the sample size for an area increases, the estimate approaches the true population value.
The prior substate report was based on the combined set of 1999-2001 surveys (OAS, 2005). In 2002, a number of methodological changes were introduced, including a $30 incentive for participating in the survey, additional training to encourage the adherence to survey protocols, and a change of the survey name from the National Household Survey on Drug Abuse to NSDUH, its current name. Collectively, these changes had significant effects on the survey responses and a significant increase in the overall response rate. The impact was a significant increase in the prevalence estimates for most substance use measures, more than could be attributed to secular trends in drug use. For example, lifetime use of any illicit drug for persons aged 12 or older went from 42 percent in 2001 to 46 percent in 2002 (OAS, 2002, 2003).
Because of the changes in the survey that took place in 2002, estimates for 2002-2004 are not comparable with estimates for 1999-2001, and it is not possible to separate the effect of the methodological changes from the true trends in substance use. Therefore, one should not conclude that any differences between estimates from 1999-2001 and 2002-2004 represent true changes. Also, there were many changes in the definitions of substate planning areas between those years that result in a greater lack of comparability between the two sets of estimates for the affected areas. The following 16 States made major changes in substate areas: Delaware, Georgia, Kansas, Maine, Maryland, Mississippi, Missouri, New Hampshire, New Jersey, New Mexico, New York, Pennsylvania, South Dakota, Texas, West Virginia, and Wyoming. For example, Maryland now has seven areas as compared with three areas specified for the 1999-2001 estimates, and none of the areas is the same for the two time periods. Another 10 States made fairly minor changes: Colorado, Hawaii, Indiana, Iowa, Louisiana, Michigan, Montana, Ohio, Oklahoma, and Wisconsin. The remaining 24 States and the District of Columbia did not change their definitions of substate areas.
Looking at the 24 States and District of Columbia that did not change any of their substate area definitions, their areas account for 149 of the total 357 areas in the Nation. If we consider marijuana, in 1999-2001, nationally 5.1 percent of all persons aged 12 or older indicated they had used it in the past month. For 2002-2004, the estimate for the same population was 6.1 percent. The difference of 1.0 percent might all be due to methodological differences, all be due to real change, or represent some combination of the two. If we focus on the period from 2001 to 2002 representing the timing of the methodological changes, the national estimates for those 2 years were 5.4 and 6.2 percent, respectively, a difference of +0.8 percent (see 2001 and 2002 detailed tables). As just discussed, the difference between the two estimates cannot be interpreted as the true change between the two periods; nevertheless, it is useful to note the impact of the national difference on individual differences at the planning area level. Of the 149 differences among the States that did not change definitions, 124 of them were positive and 25 were negative. Those differences ranged from -2.5 to +3.8 percent. The distribution of the differences across the planning areas probably results from varying degrees of the effect of the methodological changes as well as varying levels of secular change between the two time periods.
Despite the differences in the two time periods, methodological changes affecting prevalence estimates, and changes in the definitions of many of the substate planning areas, six of the areas that were in the highest group in 1999-2001 for past month use of marijuana were also in the highest group for 2002-2004. In addition, of the eight States with one or more areas in the highest group based on the 1999-2001 data, seven of them also had one or more areas with the highest rates based on the 2002-2004 data (California, District of Columbia, Hawaii, Massachusetts, Oregon, Rhode Island, and Vermont). Therefore, there is a degree of continuity in the rankings over time, even though the size of the difference for a specified area cannot be said to represent the true secular trend.
During 2002-2004, 8.1 percent of persons in the United States aged 12 or older had used an illicit drug in the past month. West Kansas and Utah County, Utah, reported the lowest rates at 5.0 percent, and the highest rate (13.6 percent) was reported by northern California (Region 1) (the 14 northernmost counties). The highest 15 areas were dispersed among 10 States, with Alaska accounting for 3 areas and Michigan accounting for 3 areas. In the top 15 areas, 11 were metropolitan areas or parts of a metropolitan area; of those 11 areas, 4 were single-county areas (U.S. Census Bureau, 2006). Of the 15 areas with the lowest rates of illicit drug use in the past month, 6 of them were from four Midwestern States: Iowa, Kansas, Nebraska, and South Dakota. Moreover, South Dakota had three eastern regions (Regions 4, 5, and 6) among those with the lowest rates of illicit drug use. It is important to note that these estimates are based on a sample, and a different sample could result in slightly different high and low areas. For example, California's Region 1 had the highest rate of past month illicit drug use (13.6 percent) of any substate area in the Nation. It can be stated with 95 percent certainty that the true value for Region 1 falls between 10.6 and 17.2 percent (based on the tables in Appendix C that include the 95 PIs). However, note that California's Region 1 may not truly have the highest rate of past month use of illicit drugs, but the probability that it is truly in the top 15 areas is 83 percent.
Marijuana is the most commonly used illicit drug, and many of the areas having a high rate of illicit drug use reported similarly high rates of marijuana use. The correlation between past month use of any illicit drug and past month marijuana use at the planning area level was 0.95. Nine of the areas in the lowest 15 for marijuana were the same as those for any illicit drug use, and 11 of the highest 15 areas for marijuana use were the same as for any illicit drug use. In 2002-2004, 6.1 percent of persons 12 or older used marijuana in the past month, and 10.7 percent of persons 12 or older had used marijuana in the past year. The rankings for past month use of marijuana were very similar to those for past year use of the substance (e.g., 11 of the highest 15 areas were the same for past month use as for past year use). Most first users of marijuana were between the ages of 12 and 25. Less than 2 percent (1.8 percent) of persons aged 12 or older had used marijuana for the first time in 2002-2004. Because new use of marijuana usually occurs by age 25 and the percentage of persons initiating it is fairly small relative to past month use of marijuana, the areas that rank the highest and lowest on these two measures tend to be different. For example, only 3 areas among the 15 highest rates for past month use of marijuana also were included in the top 15 for first use of marijuana: Southeast Alaska, Montana (Region 5), and Rhode Island's Washington County.
At the individual level, the perceived risk of using a substance is often highly correlated with whether or not a person actually uses the substance. Among persons aged 12 or older, 39.3 percent perceived a great risk of using marijuana once a month during 2002-2004. Among the 15 areas with the lowest perceived risk of occasional use of marijuana, 6 areas reported the highest level of past month marijuana use.
Nationally, 3.6 percent of persons aged 12 or older had used an illicit drug other than marijuana in 2002-2004 in the past month. Illicit drugs other than marijuana include cocaine (and crack), heroin, hallucinogens, inhalants, or any prescription-type psychotherapeutic used nonmedically. Past month use of these substances ranged from a low of 2.6 percent in Montgomery County, Maryland, to a high of 5.1 percent in the Bluegrass, Comprehend, and North Key region of Kentucky. Of the 15 areas with the highest rates of use of an illicit drug other than marijuana in the past month, 9 of them were accounted for by six States in the West: Arizona, California, Colorado, New Mexico, Oregon, and Wyoming. Also, Colorado, Kentucky, and New Mexico each had more than 1 area among the top 15. Of the 15 areas with the lowest rates, 10 occurred in five States from the Midwest: Iowa, Kansas, Minnesota, North Dakota, and South Dakota.
The national prevalence rate for the use of cocaine in the past year among all persons aged 12 or older was 2.5 percent. Because cocaine is one of the substances included in the "any illicit drug other than marijuana" category, it is useful to compare the ranking of the substate areas with respect to these two measures. The statistical correlation between these measures across the substate areas was 0.56. Among the 15 areas with the lowest rates for past month use of any illicit drug other than marijuana, 5 also were ranked in the lowest 15 for cocaine use in the past year. Similarly, only 5 areas in the group with the highest rates for cocaine also were ranked in the highest group for past month use of an illicit drug other than marijuana. The 5 areas that were common to the top 15 for both measures included the 4 areas with the highest estimates of past year cocaine use: Pima County, Arizona; Regions 2 and 7, Colorado; Region 3 (Bernalillo County), New Mexico; and Washington County, Rhode Island.
Over the 2002-2004 period, an average of 4.8 percent of all persons aged 12 or older had used a pain reliever for nonmedical use within the past year. Estimates ranged from a low of 2.4 percent in the District of Columbia (Ward 7) to 7.7 percent in the Bluegrass, Comprehend, and North Key region of Kentucky. Both Kentucky and Oregon had more than 1 area among the highest 15 prevalence rates. In the areas with the 15 lowest rates, three States (Iowa, Maryland, and South Dakota) and the District of Columbia contained more than a single area and together accounted for 13 of the 15 areas.
Alcohol is the most commonly used substance in the United States. Nationally, about half (50.4 percent) of Americans aged 12 or older reported past month use of alcohol in 2002-2004. Utah County, Utah, had the lowest rate of any area in the Nation (19.8 percent). The District of Columbia (Ward 3) had the highest rate (74.7 percent). The 15 areas with the highest rates were spread across 12 States; however, the 15 areas with the lowest rates only comprised 6 States, resulting in relatively more clustering among the areas with the lowest rates. Among the highest 15 rates of past month alcohol use, 8 areas were in States in the Northeast, 5 were in the Midwest, and 2 were in the District of Columbia.
Binge alcohol use is defined as drinking five or more drinks on the same occasion (i.e., at the same time or within a couple hours of each other) on at least 1 day in the 30 days prior to the survey. Binge alcohol use was reported by 22.8 percent of persons aged 12 or older during the period. The association of binge alcohol use with past month use of alcohol at this geographic level appeared to be low to moderate when one just looks at how many of the lowest or highest areas the two measures have in common. Eight of the lowest areas for binge use of alcohol were the same as those for alcohol use, but only four areas were the same for the highest rates. The statistical correlation among all 357 substate areas, however, was 0.72. The lowest rates encompassed all of the six planning areas in Utah and six of the seven planning areas in Tennessee. North Dakota, South Dakota, and Wisconsin accounted for the majority of the areas with the highest rates.
Perceptions of the risk of binge drinking were moderately inversely related to the actual rates of binge drinking. Areas that did not perceive much of a risk of binge drinking had high rates of past month binge use of alcohol and vice versa (the correlation was -0.65). However, among the areas with the highest rates of binge use of alcohol, only six were in the lowest group for perceptions of risk. Furthermore, among the areas with the lowest rates of binge use of alcohol, only two areas also had the highest perception of risk of its use.
The District of Columbia's Ward 7 reported the lowest rate of binge drinking (10.4 percent) among persons aged 12 to 20 (i.e., underage alcohol users). The highest rate among the 340 substate areas was in the Northeast region of North Dakota (36.1 percent). Underage binge use of alcohol was highly associated with rates of binge alcohol use in the general population aged 12 or older (the correlation was 0.78).3 However, only 8 of the areas were the same among the highest 15, and only 4 were the same among the lowest 15.
Underage alcohol use rates among the 357 substate areas also were highly correlated (0.76) with the rates of alcohol use in the general population aged 12 or older (see footnote 3). Again, only fewer than half of the 15 highest and 15 lowest rates were common to past month alcohol use and past month underage alcohol use.
In 2002-2004, about 30 percent of all persons aged 12 or older used tobacco in the past month. Tobacco is the second most commonly used substance in the United States next to alcohol. NSDUH includes a series of questions on the use of several tobacco products, including cigarettes, smokeless tobacco, cigars, and pipe tobacco. All of the substate areas with the 15 lowest rates of past month tobacco use were in California and Utah, with the exception of Montgomery County, Maryland. The lowest rate of past month tobacco use occurred in Utah County, Utah (15.2 percent). Kentucky and West Virginia accounted for 9 of the 15 areas with the highest rates of past month tobacco use. The two highest areas were in Kentucky: Communicare and River Valley (42.0 percent); and Kentucky River, Mountain and Pathways (43.4 percent).
Because cigarettes are the major tobacco product, most of the areas with high tobacco use also had high rates of cigarettes. Of the 15 areas with the lowest rates for tobacco use, 13 of them were also in the lowest 15 for cigarette use; and in the top 15, 10 of the areas were the same. During 2002-2004, the national rate of past month cigarette use was 25.5 percent. Among all persons aged 12 or older, the highest rate of past month cigarette use among persons 12 or older was in Kentucky's Kentucky River, Mountain, and Pathways region (37.5 percent). The perceived risk of heavy smoking (smoking one or more packs of cigarettes per day) was negatively correlated with past month use of cigarettes (-0.80) as expected.
NSDUH includes a series of questions to assess the prevalence of substance use disorders (i.e., dependence on or abuse of a substance) in the past 12 months. Substances include alcohol and illicit drugs, such as marijuana, cocaine, heroin, hallucinogens, and inhalants, and the nonmedical use of prescription-type psychotherapeutic drugs. These questions are used to classify persons as being dependent on or abusing specific substances based on criteria specified in the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) (American Psychiatric Association [APA], 1994). The questions on dependence ask about health and emotional problems, attempts to cut down on use, tolerance, withdrawal, and other symptoms associated with substances used. The questions on abuse ask about problems at work, home, and school; problems with family or friends; physical danger; and trouble with the law due to substance use. Dependence reflects a more severe substance problem than abuse, and persons are classified with abuse of a particular substance only if they are not dependent on that substance.
Nationally, 7.7 percent of the population aged 12 or older was classified with being dependent on or having abused alcohol in the past year in 2002-2004. Alcohol dependence or abuse levels among all the substate areas tended to be highly correlated with the levels of past month binge alcohol use (0.79). Of the top 15 substate areas for binge alcohol use, 9 areas were also in the top 15 for alcohol dependence or abuse.
In 2002–2004, past year alcohol dependence or abuse varied from a low of 5.4 percent in southern Utah (Central, Four Corners, San Juan, and Southwest region) and north central Florida (District 13) to a high of 13.5 percent in south central Wyoming (Judicial District 2). Of the 15 substate areas with the highest rates of past year alcohol dependence or abuse, most were in States in the West and Midwest (Montana, Nebraska, New Mexico, North Dakota, South Dakota, Wisconsin, and Wyoming). The District of Columbia and Rhode Island also were represented in the top 15. The District of Columbia, North Dakota, South Dakota, and Wisconsin all had more than 1 of its substate areas in the top 15. Only 4 of the top 15 substate areas for alcohol dependence or abuse were also in the top 15 for illicit drug dependence or abuse: the District of Columbia's Wards 1 and 2, Region 3 (Bernalillo County) in New Mexico, and Washington County in Rhode Island. The correlation among the substate areas between alcohol dependence or abuse and illicit drug dependence or abuse was only 0.26. In 2002-2004, 3.0 percent of persons aged 12 or older were dependent on or had abused illicit drugs in the past year. The rates among areas ranged from a low of 2.2 percent in Florida's District 13, Illinois's Region IV, Pennsylvania's Regions 17, 24, 27, 43, and 47, and South Dakota's Region 6 to 4.4 percent in Washington's Ward 8. The District of Columbia, New Mexico, and Rhode Island together accounted for 11 of the top 15 substate areas.
The national rate in 2002-2004 for past year dependence on or abuse of alcohol or illicit drugs among persons aged 12 or older was 9.3 percent. Substate areas that were ranked high for past year dependence on or abuse of alcohol also tended to be ranked high for dependence on or abuse of alcohol or illicit drugs because alcohol accounts for most of the substance dependence or abuse. For example, 11 of the top 15 areas for alcohol dependence or abuse were in the top 15 for alcohol or illicit drug dependence or abuse. The area with the highest rate of alcohol or illicit drug dependence or abuse (15.0 percent) was Judicial District 2 in Wyoming; and the Central, Four Corners, San Juan, and Southwest region in southern Utah had the lowest rate in the Nation (6.4 percent).
The definition of a person needing but not receiving treatment for an illicit drug use problem is that the person meets the criteria for abuse of or dependence on illicit drugs according to the DSM-IV, but has not received specialty treatment for an illicit drug problem in the past year. Specialty treatment is treatment received at a drug and alcohol rehabilitation facility (inpatient or outpatient), hospital (inpatient only), or mental health center. In 2002-2004, 2.7 percent of persons aged 12 or older needed treatment for an illicit drug use problem in the past year, but did not receive it. Two substate areas were tied for the lowest rate in the Nation (1.9 percent): in Maryland (Montgomery County) and in southwestern Pennsylvania (Regions 17, 24, 27, 43, and 47). The area with the highest rate in the Nation (4.0 percent) was in southwestern New Mexico (Region 5).
Only New Mexico and the District of Columbia had more than one area among those with the highest rates of needing but not receiving treatment for illicit drug use problems. The top 15 areas were not very clustered relative to other measures and represented 11 different States. Needing but not receiving treatment for illicit drug use problems had a correlation of 0.72 with past month use of any illicit drug.
In 2002-2004, the percentage of persons aged 12 or older needing but not receiving treatment for alcohol use problems was more than twice as large (7.3 percent) as the percentage for illicit drug use problems. Generally, the areas with the highest rates of untreated alcohol use problems were not the same as those areas with untreated illicit drug use problems. Only Wards 1 and 2 in the District of Columbia and Bernalillo County in New Mexico, were among the top 15 for both measures. Wyoming's Judicial District 2 had a rate of 13.7 percent, the highest in the Nation. Needing but not receiving treatment for alcohol use problems had a lower correlation (0.57) with past month alcohol use than needing but not receiving treatment for illicit drug use problems did with past month illicit drug use.
In 2002-2004, serious psychological distress (SPD) was present in 9.2 percent of persons aged 18 or older. For details on how SPD was measured in 2002, 2003, and 2004, refer to Section B.6 of this report and Section A.7 in Appendix A of the 2003-2004 SAE report (Wright & Sathe, 2006). The top 15 areas with the highest rates of SPD included 5 areas from West Virginia; the other States with more than 1 area among the top 15 were Rhode Island and Oklahoma. Southern I and III region in West Virginia had the highest rate (13.7 percent) of SPD. In the most recent State report (Wright & Sathe, 2006), the State with the highest rate for SPD was also West Virginia (12.7 percent).
In Section A, the discussion covered 22 substance use measures organized into 5 broad areas: illicit drugs, alcohol (including the two underage drinking outcomes), cigarettes, dependence and abuse and need for treatment, and SPD. The discussion was primarily limited to providing the range of rates from lowest to highest, any State-level clustering of substate areas in the lowest 15 or the highest 15 group of substate areas, and correlations among measures that are related in some way or that might appear to be related. Generally, when the "highest" or "lowest" areas have been cited, the main purpose has been to indicate the range of estimates rather than to identify the substate area because these estimates have been based on sample data fitted to a national model (as described earlier), and this uncertainty is reflected in the PIs, which can be quite large relative to the size of the estimate itself. For example, it was noted earlier that the highest estimate for past month use of any illicit drug was 13.6 percent for the northern portion of California (Region 1). The 95 percent PI for that estimate was from 10.6 to 17.2 percent. Therefore, the description of an area as lowest or highest could be interpreted more generally as having an estimate that is "among the lowest" or "among the highest." It is even more important to consider the uncertainty in the estimated prevalence estimates when "ranking" estimates within each State because the 95 percent PI can be quite large in comparison with the range of estimates within a State.
The range of estimates within a State is a function of the range of true population values across the State, the sample size of each substate area, and consequently the number of substate planning areas in a State relative to its sample size. Therefore, the 8 States with the largest population, which have an average of just over 14 substate areas, tended to have a wider range of estimates than the small States, which average just over 5.5 substate areas. The average range within the 8 most populous States across their substate areas for past month alcohol use was 17.6 percent. For past month use of cigarettes, the analogous average range was 7.1 percent; for past month marijuana use across the substate areas within each of the 8 largest States, the average range was 4.4 percent. As one can see, the substance measures with higher prevalence estimates, such as past month alcohol use (with a national rate of 50.4 percent for 2002-2004), display a greater range of estimates than those for past month cigarette use (25.5 percent for the same period) or past month marijuana use (6.1 percent for the same period). Therefore, the range within States is not only a function of the number of areas, but also a function of the prevalence rate level of the substance measure. Smaller States tend to have reduced ranges both due to having fewer substate areas within the State and because those areas with low substate sample sizes are pulled more toward the State mean prevelance rate. The average ranges within the small States for past month alcohol, cigarette, and marijuana use were 11.6, 4.4, and 2.5 percent, respectively.
The tables in Section C present the estimates and the 95 percent PIs as a guide to help determine the precision of the estimates. In an appendix at the end of Section F, we also present additional tables for three measures: past year cocaine use (low prevalence), past month marijuana use (moderate prevalence), and past month alcohol use (high prevalence). These tables show the region with the highest and the lowest estimate in each State and also the Bayes posterior probability for testing whether the two prevalences are significantly different from each other. For details on how these tests were conducted, see the appendix.
Referring to Table T1 in the appendix for past year cocaine use, Alabama's Region 1 had a low estimate of 1.9 percent and a high estimate of 2.5 percent in Region 2. These two prevalences are not statistically different at the 0.05 level of significance, as can be seen from its p value of approximately 0.15. However, the low estimate of 1.8 percent in Alaska's Rural substate area was statistically different from the high estimate (of 2.80 percent) in Alaska's Southeast area (p = 0.047). Seven of the eight most populous States displayed significantly different estimates (at the 0.05 level of significance) for cocaine in their highest and lowest areas. The average difference detected for those seven States was 1.4 percent. The difference for the other large State, Illinois, at 0.6 percent was not statistically significant.
Tables T2 and T3 for the other two measures, marijuana use and alcohol use, are similar. One can see from the tables that in moving from past year cocaine use to past month marijuana use and then to past month alcohol use, increasingly more States displayed significant differences. For past year cocaine use, of the 42 small sample States and the District of Columbia (that had an annual sample of about 900 persons), only 15 had a significant difference between the substate area with the lowest prevalence and the one with the highest prevalence rate. For past month use of marijuana, 27 of them showed significant differences, and for past month use of alcohol, 35 of the 43 had significant differences (at the 0.05 level of significance).
The precision of each estimate is partly a function of the sample size on which it is based. Substate area sample sizes range from approximately 200 persons up to 4,000 or more. For example, the estimates for Southwest Indiana are based on a sample of 203 (see Table E.1 in Section E). The 95 percent PI for the estimate of past month use of any illicit drug for Southwest Indiana (8.1 percent) is from 6.1 to 10.6 percent (see Table C.1 in Section C). The estimates for Cook County, Illinois, are based on a sample size of 4,348. The 95 percent PI for the same measure (8.8 percent) is from 7.8 to 9.9 percent. Therefore, the width of the PI for Cook County is less than half the width of the PI for Southwest Indiana because of its larger sample size, although the difference is not strictly proportional to the difference in sample sizes because the precision is also of function of the national model. Moreover, the precision is a function of the size of the prevalence estimate; therefore, if we look at past month cigarette use for the same two areas (see Table C.6 in Section C), the widths of the PIs for both areas are larger (e.g., Cook County is from 25.5 to 29.3 percent, a range of 3.8 percent, while that for Southwest Indiana is from 25.7 to 34.2 percent, a range of 8.5 percent).
|
This page was last updated on January 15, 2009. |
* Adobe™ PDF and MS Office™ formatted files require software viewer programs to properly read them.
Click here to download these FREE programs now
| Highlights | Topics | Data | Drugs | Pubs | Short Reports | Treatment | Help | OAS |
