| ![]() | |||||||||||
![]() |
|
|||||||||||
Misuse of Prescription Drugs |
An important limitation of estimates of drug use prevalence from the National Survey on Drug Use and Health (NSDUH) is that they are only designed to describe the target population of the survey—the civilian, noninstitutionalized population aged 12 or older. Although this population includes almost 98 percent of the total U.S. population aged 12 or older, it excludes some important and unique subpopulations who may have very different drug use patterns. For example, the survey excludes active military personnel, who have been shown to have significantly lower rates of illicit drug use. Also, persons living in institutional group quarters, such as prisons and residential drug use treatment centers, are not included in NSDUH, yet they have been shown in other surveys to have higher rates of illicit drug use. Also excluded are homeless persons not living in a shelter on the survey date; they are another population shown to have higher than average rates of illicit drug use.
The national estimates, along with the associated variance components, were computed using a multiprocedure package, SUDAAN® Software for Statistical Analysis of Correlated Data. SUDAAN was designed for the statistical analysis of data collected using stratified, multistage cluster sampling designs, as well as other observational and experimental studies involving repeated measures or studies subject to cluster correlation effects (RTI International, 2004). The final, nonresponse-adjusted, and poststratified analysis weights were used in SUDAAN to compute unbiased design-based drug use estimates.
The sampling error (i.e., the standard error [SE]) of an estimate is the error caused by the selection of a sample instead of conducting a census of the population. The sampling error may be reduced by selecting a large sample and/or by using efficient sample design and estimation strategies, such as stratification, optimal allocation, and ratio estimation.
With the use of probability sampling methods in NSDUH, it is possible to develop estimates of sampling error from the survey data. These estimates have been calculated in SUDAAN for all estimates presented in this report using a Taylor series linearization approach that takes into account the effects of the complex NSDUH design features. The sampling errors are used to identify unreliable estimates and to test for the statistical significance of differences between estimates.
Estimates of means or proportions, d, such as drug use prevalence estimates for a domain d, can be expressed as a ratio estimate:
, D
where d is a linear statistic estimating number of substance users in the domain, and
d is a linear statistic estimating the total number of persons in domain d (both users and nonusers). The SUDAAN software used to develop estimates and their SEs produces direct estimates of
d and
d and their SEs. The SUDAAN application also uses a Taylor series approximation method to estimate the SEs of the ratio estimate
d.
When the domain size, d, is free of sampling error, an appropriate estimate of the SE for the total number of users is
, D
This approach is theoretically correct when the domain size estimates, d, are among those forced to match their respective U.S. Bureau of the Census population projections through the weight calibration process (Chen et al., 2005). In these cases,
d is not subject to sampling error. For a more detailed explanation of the weight calibration process, see Section A.3.2 in Appendix A.
For estimated domain totals, d, where
d is not fixed (i.e., where domain size estimates are not forced to match the U.S. Bureau of the Census population projections), this formulation may still provide a good approximation if it can be assumed that the sampling variation in
d is negligible relative to the sampling variation in
d. This is a reasonable assumption for most cases in this study.
For a subset of the estimates produced from the 2002, 2003, and 2004 data, the above approach yielded an underestimate of the variance of a total because d was subject to considerable variation. In these cases, the SEs for the total estimates calculated directly within SUDAAN are reported. Using the SEs from the total estimates directly from SUDAAN does not affect the SE estimates for the corresponding proportions presented in the same sets of tables.
As has been done in past NSDUH reports, direct survey estimates produced for this study that are considered to be unreliable due to unacceptably large sampling errors are not shown in this report and are noted by asterisks (*) in the tables containing such estimates. The criteria used for suppressing all direct survey estimates were based on the relative standard error (RSE) (defined as the ratio of the SE over the estimate) on nominal sample size and on effective sample size.
Proportion estimates () within the range [0 <
< 1], rates, and corresponding estimated number of users were suppressed if
RSE[-ln ()]> 0.175 when
≤ 0.5
or
RSE[-ln(1 - )]> 0.175 when
> 0.5.
Using a first-order Taylor series approximation to estimate RSE[-ln ()] and RSE[-ln(1 -
)], the following was obtained and used for computational purposes:
> 0.175 when
≤ 0.5 D
or
> 0.175 when
> 0.5. D
The separate formulas for ≤ 0.5 and
> 0.5 produce a symmetric suppression rule (i.e., if
is suppressed, then 1 -
will be as well). This ad hoc rule requires an effective sample size in excess of 50. When 0.05 <
< 0.95, the symmetric property of the rule produces a local maximum effective sample size of 68 at
= 0.5. Thus, estimates with these values of
along with effective sample sizes falling below 68 are suppressed. See Figure B.1 for a graphical representation of the required minimum effective sample sizes as a function of the proportion estimated.
A minimum nominal sample size suppression criterion (n = 100) that protects against unreliable estimates caused by small design effects and small nominal sample sizes was employed. Prevalence estimates also were suppressed if they were close to 0 or 100 percent (i.e., if < 0.00005 or if
≥ 0.99995).
Estimates of other totals (e.g., number of initiates) along with means and rates that are not bounded between 0 and 1 (e.g., mean age at first use and incidence rates) were suppressed if the RSEs of the estimates were larger than 0.5. Additionally, estimates of the mean age at first use were suppressed if the sample size was smaller than 10 respondents. Also, the estimated incidence rate and number of initiates were suppressed if they rounded to 0.
The suppression criteria for various NSDUH estimates are summarized in Table B.1 at the end of this appendix.
This section describes the methods used to compare prevalence estimates in this report. Customarily, the observed difference between estimates is evaluated in terms of its statistical significance. Statistical significance is based on the p value of the test statistic and refers to the probability that a difference as large as that observed would occur due to random variability in the estimates if there were no difference in the prevalence estimates for the population groups being compared. The significance of observed differences in this report is generally reported at the 0.05 and 0.01 levels. When comparing prevalence estimates, the null hypothesis (no difference between prevalence estimates) was tested against the alternative hypothesis (there is a difference in prevalence estimates) using the standard difference in proportions test expressed as
, D
where 1 = first prevalence estimate,
2 = second prevalence estimate, var (
1) = variance of first prevalence estimate, var (
2) = variance of second prevalence estimate, and cov (
1,
2) = covariance between
1 and
2. In cases where significance tests between years were performed, the 2003 prevalence estimate becomes the first prevalence estimate and the 2004 estimate becomes the second prevalence estimate.
Under the null hypothesis, Z is asymptotically distributed as a normal random variable. Therefore, calculated values of Z can be referred to the unit normal distribution to determine the corresponding probability level (i.e., p value). Because the covariance term is not necessarily zero, SUDAAN was used to compute estimates of Z along with the associated p values using the analysis weights and accounting for the sample design as described in Appendix A. A similar procedure and formula for Z were used for estimated totals.
When comparing population subgroups defined by three or more levels of a categorical variable, log-linear Chi-square tests of independence of the subgroups and the prevalence variables were conducted first to control the error level for multiple comparisons. If the Chi-square test indicated overall significant differences, the significance of each particular pairwise comparison of interest was tested using SUDAAN analytic procedures to properly account for the sample design. Using the published estimates and SEs to perform independent t tests for the difference of proportions usually will provide the same results as tests performed in SUDAAN. However, where the significance level is borderline, results may differ for two reasons: (1) the covariance term is included in SUDAAN tests whereas it is not included in independent t tests, and (2) the reduced number of significant digits shown in the published estimates may cause rounding errors in the independent t tests.
Errors can occur from nonresponse, coding errors, computer processing errors, errors in the sampling frame, reporting errors, and other errors not due to sampling. These types of errors are reduced through data editing, statistical adjustments for nonresponse, close monitoring and periodic retraining of interviewers, and improvement in various quality control procedures.
Although these types of errors often can be much larger than sampling errors, measurement of most of these errors is difficult. However, some indication of the effects of some types of these errors can be obtained through proxy measures, such as response rates and from other research studies.
In 2002, 2003, and 2004, respondents received a $30 incentive in an effort to improve response rates over years prior to 2002. Of the 142,612 eligible households sampled for the 2004 NSDUH, for example, 130,130 were successfully screened for a weighted screening response rate of 90.9 percent (Table B.2). In these screened households, a total of 81,973 sample persons were selected, and completed interviews were obtained from 67,760 of these sample persons, for a weighted interview response rate of 77.0 percent (Table B.3). Weighted screening response rates for 2002 and 2003 were 90.7 percent in each survey year (Table B.2). Weighted interview response rates were 78.6 percent in 2002 and 77.4 percent in 2003 (Table B.3).
The overall weighted response rate, defined as the product of the weighted screening response rate and weighted interview response rate, was 70.0 percent in 2004. Nonresponse bias can be expressed as the product of the nonresponse rate (1-R) and the difference between the characteristic of interest between respondents and nonrespondents in the population (Pr - Pnr). Thus, assuming the quantity (Pr - Pnr) is fixed over time, the improvement in response rates in 2002 through 2004 over prior years will result in estimates with lower nonresponse bias.
Among survey participants, item response rates were above 99 percent for most drug use items. However, inconsistent responses for some items were common. Estimates of substance use from NSDUH are based on responses to multiple questions by respondents, so that the maximum amount of information is used in determining whether a respondent is classified as a drug user. Inconsistencies in responses are resolved through a logical editing process that involves some judgment on the part of survey analysts. Additionally, missing or inconsistent responses are imputed using statistical methodology. Editing and imputation of missing responses are potential sources of error.
Most drug use prevalence estimates, including those produced for NSDUH, are based on self-reports of use. Although studies have generally supported the validity of self-report data, it is well documented that these data often are biased (underreported or overreported) by several factors, including the mode of administration, the population under investigation, and the type of drug (Bradburn & Sudman, 1983; Hser & Anglin, 1993). Higher levels of bias also are observed among younger respondents and those with higher levels of drug use (Biglan, Gilpin, Rorhbach, & Pierce, 2004). Methodological procedures, such as biological specimens (e.g., urine, hair, saliva), proxy reports (e.g., family member, peer), and repeated measures (e.g., recanting), have been used to validate self-report data (Fendrich, Johnson, Sudman, Wislar, & Spiehler, 1999). However, these procedures often are impractical or too costly for community-based epidemiological studies (SRNT Subcommittee on Biochemical Verification, 2002). NSDUH utilizes widely accepted methodological practices for ensuring validity, such as encouraging privacy through audio computer-assisted self-interviewing (ACASI). Comparisons using these methods within NSDUH have been shown to reduce reporting bias (Aquilino, 1994; Turner, Lessler, & Gfroerer, 1992).
Several measurement issues are associated with the 2004 NSDUH that may be of interest and are discussed in this section. Issues that are specifically relevant to this report include the methods for measuring nonmedical use of prescription psychotherapeutic drugs, incidence, and substance dependence and abuse.
Nonmedical use in NSDUH is defined as use without a doctor's prescription or use for the experience or feeling that a drug caused. Although NSDUH collects extensive information on the nonmedical use of prescription-type pain relievers, tranquilizers, stimulants, and sedatives as therapeutic classes, the information collected about specific drugs within these classes generally is limited to use at any time in the individual's life; additional information on specific drugs is collected only for OxyContin® (a pain reliever) and methamphetamine (a stimulant). For each therapeutic class of prescription drugs, the survey instrument begins by briefly defining the category. Respondents then are shown a "pill card" displaying the names and color photographs of drugs and groups of drugs in that therapeutic class. Entries on the cards include brand-name drugs, generic drugs, and groupings of drugs sometimes from different generic categories. For each entry on the pill cards, respondents are asked whether they ever, even once, used the drug(s) when it was not prescribed for them or only for the experience or feeling it caused. In some instances, the same generic substance may be represented by multiple entries on the pill cards.
However, these data collection methods were not developed with the intent of supporting analysis of patterns of use of individual prescription drugs. The pill cards principally are designed to (a) define the therapeutic class (e.g., pain relievers) so that respondents clearly understand the types of drugs that are included, (b) aid respondents in identifying and accurately reporting nonmedical use of any drug in the class, and (c) assist respondents in identifying whether they may have misused other drugs in that therapeutic class that are not shown on the pill card, such as drugs that have recently been approved. If respondents report nonmedical use of "some other drug" within a given class (e.g., some other pain reliever), they are asked to specify the names of the other medications they have used nonmedically (referred to as "other-specify" responses). The selection of specific pharmaceuticals to be shown and their position on the pill cards was based on their prevalence as reported in prior surveys and on recommendations from National Institute on Drug Abuse (NIDA) staff with knowledge of drug abuse epidemiology and abuse liability of prescription pharmaceuticals.
As indicated in Table B.4, drug entries on each pill card are divided into two parts. The top half of each pill card depicts three groups of related drugs. These drugs were chosen by the questionnaire designers because they were the most prevalent. For each group, respondents are asked whether they have ever used any drug in the group nonmedically in their lifetime. The bottom half of the pill card shows several additional drugs. Respondents are asked first whether they ever used any of these drugs nonmedically. Persons who respond "yes" to the overall question about use of any drug at the bottom of the pill card then are asked to indicate which specific drug(s) they have used.
After responding to the top and bottom sections of the pill card, respondents are asked whether they have used any other drug(s) in the therapeutic class nonmedically. Those who respond in the affirmative are asked to type in the names of the other drug(s). These "other-specify" entries are not vetted for consistency with the therapeutic class either during the interview or in subsequent editing. For example, if a respondent typed in "diazepam" in the pain relievers module, that response was not reassigned to the tranquilizers category.
After respondents have been asked about the drugs on the pill card and nonmedical use of any other drugs in that therapeutic class, they are asked additional questions about their use of any drug in the therapeutic class. These questions include their age when they first used these drugs nonmedically, when they last used these drugs, and if respondents reported last using a drug in the past 12 months, how often they used the drug in that period. Except for OxyContin® and methamphetamine, respondents are not asked these additional questions for specific prescription drugs.
Thus, the prescription drug module uses three approaches to data collection: (1) querying separately for each group of a few related drugs, (2) querying in tandem for several drugs with follow-up to identify nonmedical use of specific drugs, and (3) asking respondents to supply names themselves. These approaches may affect the likelihood of reporting of nonmedical use of a drug, possibly with the lowest rates of reporting for write-in ("other-specify") items (approach 3), higher reporting for tandem querying for multiple drug groups with follow-up (approach 2), and the highest for separate querying for each drug group (approach 1).
Detailed information on brand names and active generic ingredients for most prescription pharmaceuticals currently approved for marketing in the United States is available in the Food and Drug Administration (FDA) Orange Book, which is updated daily and can be accessed online at http://www.fda.gov/cder/ob/default.htm (for the 26th annual edition, see FDA, Center for Drug Evaluation and Research, 2006).
In addition to the four broad classes of prescription psychotherapeutic drugs, some estimates in Chapter 3 present data for groups of drugs within a given psychotherapeutic class. These subclasses typically reflect generic chemical names, categories of drugs that are chemically related (e.g., benzodiazepines), or drugs that have similar functional properties (e.g., muscle relaxants). Subclasses of prescription psychotherapeutic drugs are described below.
Because nonmedical use of individual prescription-type psychotherapeutic drugs (except for OxyContin® and methamphetamine) is reported only for the lifetime, it is difficult to make inferences about which specific drugs currently are being misused. Trend information may not be as meaningful as it would be for current use, particularly for the 26 or older age group. Various strategies can be devised to deal with this limitation, each with its advantages and disadvantages. Tabulation of lifetime use of individual drugs among current users of any drug in the therapeutic class strengthens the implication of current use of the identified drug, but it is conclusive only if that drug is the only substance reported in the therapeutic class. However, such restriction of the analysis may reduce the coverage and introduce bias into the estimates. Another strategy is to examine use of individual drugs among persons who first used any drug in the therapeutic class in the past year. Although this strengthens the implication of current use, it excludes continuing users and may create a misimpression of relative prevalences among all users.
It also is important to note that the specific prescription-type pain relievers, tranquilizers, stimulants, or sedatives covered on the NSDUH pill cards do not represent a comprehensive list of such pharmaceuticals on the market and may not reflect fully the pharmacopoeia of specific drugs that are misused. Some specific psychotherapeutic drugs currently being misused may not be represented on the pill cards, either because they are too new to have been accommodated in the survey or have not come to the attention of drug abuse epidemiologists. The "other-specify" items in each of the therapeutic classes, however, provide a means for respondents to report all of the relevant drugs that they may have misused. Because the purpose of the pill cards, as stated above, is to communicate the general concept of the respective therapeutic classes and not necessarily to provide a comprehensive list of drugs within those classes, some drugs potentially important in current nonmedical usage patterns may be missed.
By the same token, some drugs listed on the pill cards may no longer be available, either because they have been removed from production by the U.S. Drug Enforcement Administration or due to commercial considerations, such as introduction of new, more popular medications. Examples of drugs shown on the pill card that are no longer available in the United States include methaqualone, Sopor®, or Quaalude®; Tuinal®; Placidyl®; and Preludin®. However, it is important to continue to include drugs no longer on the market because they have been misused in the past and continue to be important in reporting on lifetime use. In some cases, these drugs also may still be available on the illicit drug market, whether produced illegally in this country or smuggled from other nations.
For diseases, the incidence rate (IR) for a population is defined as the number of new cases of the disease, N, divided by the person time, PT, of exposure or
, D
where the person time of exposure is defined as the length of time a person was exposed to risk. The person time of exposure can be measured for the full period of the study or for a shorter period. The person time of exposure ends at the time of diagnosis (Greenberg, Daniels, Flanders, Eley, & Boring, 1996). Similar conventions are applied for defining the incidence of first use for a particular substance. For the purposes of this study, respondents are classified as being exposed to risk as long as they reside in the United States and have the potential to initiate use of a particular substance (i.e., once a respondent has started using a substance, he or she is no longer at risk). Because incidence is calculated for respondents belonging to specific age groups, a respondent who enters or exits a particular group during the time period of interest will have a person time of exposure equal to or less than the time he or she was a member of the age group, depending on if and when substance use was initiated. Thus, respondents with a person time of exposure less than the full time period may not have been in scope during the entire period (e.g., they did not reside in the United States or were not members of the particular age group of interest), or they began initiation of the particular substance during the time period for exposure. For example, if the time period for exposure is defined as ranging from January 1, 2003, to December 31, 2003, a respondent initiating use on January 30, 2003, would have only 1 month of exposure rather than the full year.
Beginning in 1999, the survey questionnaire allows for collection of year and month of first use for recent initiates. Month, day, and year of birth also are obtained directly or are imputed for item nonrespondents as part of the data postprocessing. Additionally, the questionnaire call record provides the date of the interview. By imputing a day of first use within the year and month of first use, a specific date can be used for estimation purposes. Exposure time can be determined in terms of days and converted to years. Beginning with the 2003 NSDUH summary report, immigrants who initiated drug use outside the United States were not included in calendar year estimates of initiation. However, those immigrants who did not initiate outside the United States were included in the analysis for the time period since they entered the United States. If respondents indicated that they were not born in the United States, they were asked to provide information regarding how long they had lived in the United States. Using this information, an entry age and date were imputed.
Having specific dates of birth, first use, and entry into the United States (for immigrants) also allows the person time of exposure during the targeted period, t, to be determined. Let the target time period for measuring incidence be specified in terms of dates; for example, the period 1998 would be specified as
t = [t1,t2) = [1 Jan 1998, 1 Jan 1999) ,
a period that includes January 1, 1998, and all days up to but not including January 1, 1999. The target age group also is defined by a half-open interval as a = [a1,a2). For example, the age group 12 to 17 would be defined by a = [12,18) for persons at least aged 12, but not yet aged 18. If person i was in age group a and residing in the United States during period t, the time and age interval referred to as the target period, Lt,a,i, is defined as follows:
, D
and
, D
where DOBi, MOBi, and YOBi, and DOEi, MOEi, and YOEi denote the day, month, and year of birth and entry to the United States, respectively.
Either this intersection will be empty (Lt,a,i = ø), or it will be designated by the half-open interval, Lt,a,i = [m1,i, m2,i, where
m1,i = Max {t1,(DOBiMOBiYOBi + a1), DOEiMOEiYOEi}
and
m2,i = Min{t2,(DOBiMOBiYOBi + a2)} .
The date of first use, tfu,d,i, also is expressed as an exact date. If an incident of first drug d use by person i in age group a occurs in time tfu,d,i ∈[m1,i, m2,i), then the indicator function Ii(d,a,t) used to count incidents of first use is set to 1, and 0 otherwise. The person time of exposure measured in years and denoted by ei(d,a,t) for a person i of age group a depends on the date of first use. If the date of first use precedes the target period (tfu,d,i < m1,i), then ei(d,a,t) = 0. If the date of first use occurs after the target period or if person i has never used drug d, then
, D
If the date of first use occurs during the target period Lt,a,i, then
, D
Note that both Ii(d,a,t) and ei(d,a,t) are set to 0 if the target period Lt,a,i is empty (i.e., person i is not in age group a during any part of time t). The incidence rate then is estimated as a weighted ratio estimate:
, D
where the wi are the analytic weights. For a more detailed explanation of the incidence methodology, see Packer, Odom, Chromy, Davis, and Gfroerer (2002).
Beginning with the 2004 NSDUH national findings report, a new measure related to incidence is being calculated and is the primary focus of that report's chapter on incidence (see Office of Applied Studies [OAS], 2005b, pp. 45-52). This measure, termed "past year initiation," refers to respondents whose date of first use of a substance, tfu,d,i, was within the year prior to their interview. Past year initiation can be viewed as an indicator variable defined as follows:
, D
where DOIi, MOIi, and YOIi denote the day, month, and year of the interview, respectively.
This measure differs from other incidence measures in that it does not refer to a particular calendar year but is rather a time period equivalent to the year prior to the interview. One additional difference to be noted is that the calculation of past year initiation does not take into account whether the respondent initiated substance use while a resident of the United States. This has little effect on past year estimates and allows for direct comparability with other standard measures of substance use because the populations of interest for the measures will be the same (i.e., both measures examine all possible respondents and do not restrict to those only initiating substance use in the United States).
One important note for both the calendar year and past year estimates of incidence is the relationship between the main categories and the subcategories of substances (e.g., illicit drugs would be a main category and inhalants and marijuana would be examples of subcategories in relation to illicit drugs). Typically, any member of a subcategory is by necessity a member of the main category (e.g., if a respondent is a past month user of a particular drug, then he or she is also a past month user of illicit drugs in general). However, this is not the case with regard to incidence statistics. Because an individual can only be an initiate of a particular substance category (main or sub) a single time, a respondent with lifetime use of multiple substances may not, by necessity, be included as an initiate of a main category, even if he or she were an initiate for a particular subcategory because his or her first initiation of other substances could have occurred earlier.
Because estimates of incidence are based on retrospective reports of age at first drug use by survey respondents, they may be subject to memory-related biases, such as recall decay and telescoping. Recall decay occurs when respondents who initiated many years ago fail to report this use and will tend to result in a downward bias in estimates for earlier years (e.g., 1960s and 1970s). Telescoping occurs, for example, when an 18-year-old respondent who first used at age 12 reports his or her age at first use as 14. Telescoping such as this will tend to result in an upward bias for estimates for more recent years.
There also is likely to be some underreporting bias due to the tendency for respondents to not report socially unacceptable behavior because of respondents' fear of disclosure. This bias is likely to have the greatest impact on recent estimates, which reflect more recent use and are based heavily on reporting by young respondents for some substances, particularly alcohol, cigarettes, and inhalants. Finally, for drug use that is frequently initiated at age 10 or younger, estimates based on retrospective reports 1 year later underestimate total incidence because 11-year-old (or younger) children are not sampled by NSDUH. Prior analyses showed that alcohol and cigarette (any use) incidence estimates could be significantly affected by this.
An evaluation of NSDUH retrospective estimates of incidence suggested that these types of bias are significant and differ by substance and length of recall (Gfroerer et al., 2004). This study showed that, for very recent time periods, such as within the past year or in the prior 2 or 3 calendar years, bias in estimates of marijuana, cocaine, alcohol, and cigarettes appears to be small, but for all other substances there is significant downward bias. Bias for all substances was shown to increase the further back in time the estimates are made, suggesting an association with the length of recall.
The past year incidence estimates are based on the data from the survey conducted that year, and they have a recall period ranging from 0 to 12 months. In other words, at the time the data are collected, the date of first drug use can be any time from today (0-month recall) to 364 days ago (12-month recall). The average length of recall for past year incidence data is 6 months. Calendar year incidence estimates have longer recall periods. For example, the recall period for 2003 calendar year incidence estimates based on the 2004 NSDUH range from 1 month (interview in January 2004, initiation in December 2003) to 23 months (interview in December 2004, initiation in January 2003), with an average recall of about 12 months. Estimates for earlier calendar years would be based on longer recall periods. This suggests that the recall bias affecting the calendar year estimates produced from one survey varies from 1 calendar year to another. For past year incidence estimates, the biases may be similar each year because the recall period is the same. Similarly, differential bias due to recall period differences could be reduced for calendar year estimates by only producing estimates for the most recent calendar year from each successive survey. For example, the 2002 calendar year estimate from the 2003 NSDUH would be expected to have a similar recall bias as the 2003 calendar year estimates from the 2004 NSDUH.
Although prior analyses and research in the literature do not provide a definitive answer to the question of which is the best approach to measure incidence with NSDUH, it is instructive to compare recent calendar year and past year incidence estimates. Exhibit B.1 shows incidence estimates based on the 2002, 2003, and 2004 NSDUHs for four drugs. If there were no bias, and no large changes over time, it would be expected that for each drug the calendar year and past year incidence estimates would be similar (but not equal). However, for all drugs, the past year incidence estimates are lower than the corresponding calendar year incidence estimates based on each survey year. In addition, a consistent pattern is evident in the calendar year estimates, in which the highest estimate generated from each survey tends to be the calendar year 2 years before the data collection period, and estimates diminish as length of recall increases. This pattern is evident in several cases in which two or more surveys are available to produce the same calendar year estimate. For alcohol, incidence estimates for the calendar year 2 years prior to each survey average 5.4 million, while estimates with a 3-year lag average 4.6 million and those with a 4-year lag average 4.3 million. Alcohol calendar year estimates based on a 1-year lag average 4.8 million, while the past year estimates average 4.1 million. Although it is unknown which estimate is closest to the true level of incidence, clearly it is not reasonable to have a continuing annual number of new users at 5.4 million when recent single-year birth cohorts in the typical age range for alcohol initiation are only about 4 to 4.5 million. This result may be due to telescoping.
Drug | Calendar Year Estimates (in Thousands) | |||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1999 | 2000 | 2001 | 2002 | 2003 | ||||||||||||
Survey Year | Survey Year | Survey Year | Survey Year | Survey Year | ||||||||||||
2002 | 2003 | 2004 | 2002 | 2003 | 2004 | 2002 | 2003 | 2004 | 2002 | 2003 | 2004 | 2002 | 2003 | 2004 | ||
N/A = not applicable. Source: SAMHSA, Office of Applied Studies, National Survey on Drug Use and Health, 2002, 2003, and 2004. |
||||||||||||||||
Alcohol | 4,520 | 4,278 | 3,922 | 5,632 | 4,770 | 4,287 | 4,548 | 5,311 | 4,708 | N/A | 4,814 | 5,324 | N/A | N/A | 5,103 | |
Cocaine | 1,083 | 877 | 852 | 1,139 | 950 | 989 | 1,160 | 1,208 | 1,093 | N/A | 1,061 | 1,100 | N/A | N/A | 1,103 | |
Cigarettes | 3,287 | 3,372 | 3,107 | 2,963 | 3,125 | 3,254 | 2,344 | 2,724 | 2,968 | N/A | 2,429 | 2,686 | N/A | N/A | 2,620 | |
Marijuana | 2,903 | 2,616 | 2,613 | 2,976 | 2,816 | 2,531 | 2,604 | 3,066 | 2,794 | N/A | 2,597 | 2,826 | N/A | N/A | 2,474 | |
Drug | Past Year Estimates (in Thousands) | |||||||||||||||
Survey Year | ||||||||||||||||
2002 | 2003 | 2004 | ||||||||||||||
Alcohol | 3,942 | 4,082 | 4,396 | |||||||||||||
Cocaine | 1,032 | 986 | 998 | |||||||||||||
Cigarettes | 1,940 | 1,983 | 2,122 | |||||||||||||
Marijuana | 2,196 | 1,973 | 2,142 |
Although it is clear that both the calendar year and the past year incidence estimates are affected by a variety of types of bias, both can provide useful epidemiological information for researchers and policymakers. Calendar year estimates, used with caution, can be analyzed to understand historical shifts in substance use as far back as the 1960s, when marijuana use began to become widespread in the United States. To track very recent shifts and patterns in incidence, however, past year incidence estimates have several important advantages and are the primary focus of this report. The main advantages are as follows:
The NSDUH computer-assisted interviewing (CAI) instrumentation in 2002, 2003, and 2004 included questions that were designed to measure dependence on and abuse of illicit drugs and alcohol. For these substances,1 dependence and abuse questions were based on the criteria in the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) (American Psychiatric Association [APA], 1994).
Specifically, for marijuana, hallucinogens, inhalants, and tranquilizers, a respondent was defined as having dependence if he or she met three or more of the following six dependence criteria:
For alcohol, cocaine, heroin, pain relievers, sedatives, and stimulants, a respondent was defined as having dependence if he or she met three or more of seven dependence criteria, including the six standard criteria listed above plus a seventh withdrawal symptom criterion. The seventh withdrawal criterion is defined by a respondent reporting having experienced a certain number of withdrawal symptoms that vary by substance (e.g., having trouble sleeping, cramps, hands tremble).
For each illicit drug and alcohol, a respondent was defined as having abused that substance if he or she met one or more of the following four abuse criteria and was determined not to be dependent on the respective substance in the past year:
Criteria used to determine whether a respondent was asked the dependence and abuse questions included responses from core substance use and frequency of substance use questions, as well as noncore substance use questions. Unknown responses in the core substance use and frequency of substance use questions were imputed. However, the imputation process did not take into account reported data in the noncore (i.e., substance dependence and abuse) CAI modules. Responses to the dependence and abuse questions that were inconsistent with the imputed substance use or frequency of substance use could have existed. Because different criteria and different combinations of criteria were used as skip logic for each substance, different types of inconsistencies may have occurred for certain substances between responses to the dependence and abuse questions and the imputed substance use and frequency of substance use as described below.
For alcohol and marijuana, respondents were asked the dependence and abuse questions if they reported substance use in the past year but did not report their frequency of substance use in the past year. Therefore, inconsistencies could have occurred where the imputed frequency of use response indicated less frequent use than required for respondents to be asked the dependence and abuse questions originally.
For cocaine, heroin, and stimulants, respondents were asked the dependence and abuse questions if they reported past year use in a core drug module or past year use in the noncore special drugs module. Thus, inconsistencies could have occurred when the response to a core substance use question indicated no use in the past year, but responses to dependence and abuse questions indicated substance dependence or abuse for the respective substance.
A respondent might have provided ambiguous information about past year use of any individual substance, in which case these respondents were not asked the dependence and abuse questions for that substance. Subsequently, these respondents could have been imputed to be past year users of the respective substance. In this situation, the dependence and abuse data were unknown; thus, these respondents were classified as not dependent on or abusing the respective substance. However, the respondent was never actually asked the dependence and abuse questions.
To maintain valid trend measurements, items in the NSDUH core modules on substance use usually are held constant from year to year. However, small refinements or additions to questions in the core modules sometimes are implemented to improve the questionnaire or obtain new information when analyses indicate that the changes will have negligible impact on the estimates for which trends are needed. In the 2004 NSDUH, new questions were added pertaining to the recency, frequency, and age at initiation of nonmedical use of the prescription-type pain reliever OxyContin®.
In addition to providing prevalence and incidence statistics on OxyContin®use in 2004, responses to these new questions also were incorporated in the 2004 editing and imputation procedures used to generate aggregate estimates on the use of any prescription-type pain reliever, any prescription-type drug, and any illicit drug.
To assess the impact of the new OxyContin®questions, 2004 estimates of drug use calculated with and without responses from the OxyContin® items were compared. Comparisons also were made against 2003 estimates to assess potential effects on trends. These comparisons were made overall and within domains defined according to the demographic and geographic variables included in the 2004 NSDUH's national findings report (OAS, 2005b). This analysis indicated that although some 2004 statistics were slightly different when calculated with and without the new OxyContin® questions, the differences were never more than 0.1 percentage point. The trends from 2002 to 2003 in overall rates of use (i.e., pain reliever use, prescription drug use, any illicit drug use, or use of any illicit drug except marijuana) were virtually the same regardless of whether or not the new OxyContin® items were included in the calculations.
The new OxyContin® item on age at first use was used to tabulate trends in incidence of use for calendar years 1965 to 2003. It is unlikely that a person could have used prescription OxyContin® nonmedically before 1995 because OxyContin® was first approved by the FDA on December 12, 1995, for prescription pain reliever use (FDA, Center for Drug Evaluation and Research, 1996) and was introduced commercially in late 1995 or early 1996. Because NSDUH data are self-reported and no restrictions are imposed on the reported date of first use, some respondents may have reported initial use prior to the date that the drug became available. In using the information on trends in calendar year incidence of OxyContin® use, data indicating use before 1995 should be viewed with caution.
Estimate | Suppress if: |
---|---|
SE = standard error; RSE = relative standard error; deff = design effect. Source: SAMHSA, Office of Applied Studies, National Survey on Drug Use and Health, 2004. |
|
Prevalence rate, ![]() |
(1) The estimated prevalence rate, ![]() (2) ![]() ![]() ![]() ![]() (3) Effective n < 68, where Effective n = ![]() (4) n < 100. Note: The rounding portion of this suppression rule for prevalence rates will produce some estimates that round at one decimal place to 0.0 or 100.0 percent but are not suppressed from the tables. |
Estimated number (numerator of ![]() |
The estimated prevalence rate, ![]() Note: In some instances when ![]() |
Mean age at first use, ![]() |
(1) RSE (![]() (2) n < 10. |
Incidence rate, ![]() |
(1) The incidence rate, ![]() (2) RSE( ![]() |
Number of initiates, ![]() |
(1) The number of initiates, ![]() (2) RSE( ![]() |
Sample Size | Weighted Percentage | |||||
---|---|---|---|---|---|---|
2002 | 2003 | 2004 | 2002 | 2003 | 2004 | |
Source: SAMHSA, Office of Applied Studies, National Survey on Drug Use and Health, 2002, 2003, and 2004. | ||||||
Total Sample | 178,013 | 170,762 | 169,514 | 100.00 | 100.00 | 100.00 |
Ineligible cases | 27,851 | 27,277 | 26,902 | 15.27 | 15.84 | 15.76 |
Eligible cases | 150,162 | 143,485 | 142,612 | 84.73 | 84.16 | 84.24 |
Ineligibles | 27,851 | 27,277 | 26,902 | 15.27 | 15.84 | 15.76 |
Vacant | 14,417 | 14,588 | 15,204 | 51.55 | 52.56 | 56.24 |
Not a primary residence | 4,580 | 4,377 | 4,122 | 17.36 | 17.07 | 15.54 |
Not a dwelling unit | 2,403 | 2,349 | 2,062 | 8.16 | 8.08 | 7.51 |
All military personnel | 289 | 356 | 282 | 1.08 | 1.39 | 1.07 |
Other, ineligible | 6,162 | 5,607 | 5,232 | 21.86 | 20.90 | 19.65 |
Eligible Cases | 150,162 | 143,485 | 142,612 | 84.73 | 84.16 | 84.24 |
Screening complete | 136,349 | 130,605 | 130,130 | 90.72 | 90.72 | 90.92 |
No one selected | 80,557 | 74,310 | 73,732 | 53.14 | 51.04 | 50.86 |
One selected | 30,738 | 30,702 | 30,499 | 20.58 | 21.46 | 21.53 |
Two selected | 25,054 | 25,593 | 25,899 | 17.00 | 18.22 | 18.53 |
Screening not complete | 13,813 | 12,880 | 12,482 | 9.28 | 9.28 | 9.08 |
No one home | 3,031 | 2,446 | 2,207 | 2.02 | 1.68 | 1.55 |
Respondent unavailable | 411 | 280 | 259 | 0.26 | 0.18 | 0.18 |
Physically or mentally incompetent | 307 | 290 | 265 | 0.20 | 0.18 | 0.17 |
Language barrier—Hispanic | 66 | 42 | 51 | 0.05 | 0.03 | 0.04 |
Language barrier—Other | 461 | 450 | 391 | 0.35 | 0.39 | 0.32 |
Refusal | 8,556 | 8,414 | 8,588 | 5.86 | 5.98 | 6.10 |
Other, access denied | 471 | 923 | 660 | 0.30 | 0.81 | 0.67 |
Other, eligible | 12 | 12 | 10 | 0.01 | 0.01 | 0.01 |
Resident < 1/2 of quarter | 0 | 0 | 0 | 0.00 | 0.00 | 0.00 |
Segment not accessible | 0 | 0 | 0 | 0.00 | 0.00 | 0.00 |
Screener not returned | 15 | 16 | 15 | 0.01 | 0.01 | 0.01 |
Fraudulent case | 479 | 6 | 14 | 0.21 | 0.00 | 0.02 |
Electronic screening problem | 4 | 1 | 22 | 0.00 | 0.00 | 0.02 |
Selected Persons | Completed Interviews | Weighted Response Rate | |||||||
---|---|---|---|---|---|---|---|---|---|
2002 | 2003 | 2004 | 2002 | 2003 | 2004 | 2002 | 2003 | 2004 | |
Source: SAMHSA, Office of Applied Studies, National Survey on Drug Use and Health, 2002, 2003, and 2004. | |||||||||
Total | 80,581 | 81,631 | 81,973 | 68,126 | 67,784 | 67,760 | 78.56% | 77.39% | 77.00% |
Age in Years | |||||||||
12-17 | 26,230 | 25,387 | 25,141 | 23,659 | 22,696 | 22,309 | 89.99% | 89.57% | 88.56% |
18-25 | 27,216 | 27,259 | 27,408 | 23,271 | 22,941 | 23,075 | 85.16% | 83.47% | 83.87% |
26 or older | 27,135 | 28,985 | 29,424 | 21,196 | 22,147 | 22,376 | 75.81% | 74.63% | 74.22% |
Gender | |||||||||
Male | 39,453 | 40,008 | 40,194 | 32,766 | 32,627 | 32,697 | 77.06% | 75.72% | 75.44% |
Female | 41,128 | 41,623 | 41,779 | 35,360 | 35,157 | 35,063 | 79.99% | 78.96% | 78.46% |
Race/Ethnicity | |||||||||
Hispanic | 10,250 | 10,753 | 11,020 | 8,692 | 8,985 | 9,218 | 80.93% | 79.55% | 79.06% |
White | 55,594 | 55,958 | 55,544 | 46,834 | 46,294 | 45,557 | 78.23% | 77.21% | 76.71% |
Black | 9,385 | 9,466 | 9,562 | 8,143 | 8,099 | 8,268 | 82.24% | 80.12% | 81.85% |
All other races | 5,352 | 5,454 | 5,847 | 4,457 | 4,406 | 4,717 | 70.50% | 69.88% | 67.21% |
Region | |||||||||
Northeast | 16,490 | 16,736 | 16,674 | 13,706 | 13,655 | 13,523 | 75.57% | 75.20% | 75.14% |
Midwest | 22,588 | 22,665 | 22,920 | 19,180 | 18,993 | 18,889 | 80.01% | 78.56% | 77.63% |
South | 24,530 | 24,725 | 24,820 | 20,900 | 20,612 | 20,807 | 79.99% | 78.38% | 78.65% |
West | 16,973 | 17,505 | 17,559 | 14,340 | 14,524 | 14,541 | 77.33% | 76.51% | 75.38% |
County Type | |||||||||
Large metropolitan | 32,294 | 36,610 | 37,103 | 26,792 | 29,759 | 30,077 | 76.85% | 75.49% | 75.72% |
Small metropolitan | 28,121 | 27,661 | 27,404 | 23,944 | 23,349 | 22,972 | 79.50% | 79.51% | 78.12% |
Nonmetropolitan | 20,166 | 17,360 | 17,466 | 17,390 | 14,676 | 14,711 | 81.38% | 79.72% | 79.23% |
Therapeutic Class | Drug Groups at Top of Pill Card1 | Drugs at Bottom of Pill Card2 |
---|---|---|
1 For drug groups shown at the top of a pill card, separate questions are asked for each drug group. 2 For drugs shown at the bottom of a pill card, all drugs are asked in tandem with follow-up to identify the specific drug(s) used. Source: SAMHSA, Office of Applied Studies, National Survey on Drug Use and Health, 2004. |
||
Pain Relievers | (1) Darvocet®, Darvon®, or Tylenol® with Codeine (2) Percocet®, Percodan®, or Tylox® (3) Vicodin®, Lortab®, or Lorcet®/Lorcet Plus® |
(4) Codeine (5) Demerol® (6) Dilaudid® (7) Fioricet® (8) Fiorinal® (9) Hydrocodone (10) Methadone (11) Morphine (12) OxyContin® (13) Phenaphen® with Codeine (14) Propoxyphene (15) SK-65® (16) Stadol® (17) Talacen® (18) Talwin® (19) Talwin® NX (20) Tramadol (21) Ultram® |
Tranquilizers | (1) Klonopin® or Clonazepam (2) Xanax®, Alprazolam, Ativan®, or Lorazepam (3) Valium® or Diazepam |
(4) Atarax® (5) BuSpar® (6) Equanil® (7) Flexeril® (8) Librium® (9) Limbitrol® (10) Meprobamate (11) Miltown® (12) Rohypnol® (13) Serax® (14) Soma® (15) Tranxene® (16) Vistaril® |
Stimulants | (1) Methamphetamine ("crank", "crystal", "ice", or "speed"), Desoxyn®, or Methedrine® (2) Prescription Diet Pills (examples given: amphetamine, Benzedrine®, Biphetamine®, Fastin®, and phentermine) (3) Ritalin® or Methylphenidate |
(4) Cylert® (5) Dexedrine® (6) Dextroamphetamine (7) Didrex® (8) Eskatrol® (9) Ionamin® (10) Mazanor® (11) Obedrin-LA® (12) Plegine® (13) Preludin® (14) Sanorex® (15) Tenuate® |
Sedatives | (1) Methaqualone, Sopor®, or Quaalude® (2) Barbiturates (examples given: Nembutal®, pentobarbital, Seconal®, secobarbital, and butalbital) (3) Restoril® or Temazepam |
(4) Amytal® (5) Butisol® (6) Chloral Hydrate (7) Dalmane® (8) Halcion® (9) Phenobarbital (10) Placidyl® (11) Tuinal® |
1 Substances include alcohol, marijuana, cocaine, heroin, hallucinogens, inhalants, pain relievers, tranquilizers, stimulants, and sedatives.
![]() This page was last updated on June 03, 2008. |
* 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 |