Who are the new users of antipsychotic medications?

Abstract

Objective: This study examined changes in the prevalence of antipsychotic medication use and the characteristics of antipsychotic users in the U.S. population between 1996 and 2005.

Methods: Data from the Medical Expenditure Panel Survey from 1996-1997 and 2004-2005 were used to examine the rate of first- and second-generation antipsychotic medication use and changes in the characteristics of users of all ages. Trends were examined in the level of use by antipsychotic users, both in terms of defined daily dose units and number of prescriptions.

Results: The rate of antipsychotic use has increased substantially between 1996-1997 and 2004-2005, but the average dose measured both by defined daily dose units and number of prescriptions has remained constant. The rapid diffusion of antipsychotic medications did not occur among individuals with schizophrenia, but rather it included substantial growth among those with newer on-label conditions (such as bipolar disorder) and a high, constant rate of off-label use. Demographic, financial, and insurance characteristics of users have remained fairly constant, with few exceptions. The average age of antipsychotic users declined during the study period, because more children were using these medications in 2004-2005. However, the gender, racial, ethnic, and insurance composition of users has been fairly stable over time.

Conclusions: The rapid diffusion of second-generation antipsychotic medications was achieved by large increases in the rate of use in certain subpopulations, most notably youths. Increasing understanding about the marginal efficacy and side-effect risks of newer and more expensive antipsychotic agents, even when prescribed as indicated, suggests that the dramatic increase in use warrants careful attention.

Figure 1

FPL=Federal Poverty Level. In 2006–7, multiple racial categories were allowed; in these years white and black both indicate no other categories were recorded. Affective disorder includes only those disorders coded in the ICD-9 category of 296 and excludes individuals with comorbid schizophrenia. Anxiety spectrum disorders exclude individuals with comorbid schizophrenia or 296-affective disorders. Some cell sizes for 1996/97 are smaller than 100 observations as noted in Appendix Table 1. * chi-square tests for discrete variables were significant at p<0.01. + comparisons are descriptive only due to small cell sizes in 1996/97.

Figure 1

FPL=Federal Poverty Level. In 2006–7, multiple racial categories were allowed; in these years white and black both indicate no other categories were recorded. Affective disorder includes only those disorders coded in the ICD-9 category of 296 and excludes individuals with comorbid schizophrenia. Anxiety spectrum disorders exclude individuals with comorbid schizophrenia or 296-affective disorders. Some cell sizes for 1996/97 are smaller than 100 observations as noted in Appendix Table 1. * chi-square tests for discrete variables were significant at p<0.01. + comparisons are descriptive only due to small cell sizes in 1996/97.

Figure 2. Prevalence of Antipsychotic Use by Characteristics

FPL=Federal Poverty Level. In 2006–7, multiple racial categories were allowed; in these years white and black both indicate no other categories were recorded. Some cell sizes for 1996/97 are smaller than 100 observations as noted in Appendix Table 2. * chi-square tests for discrete variables were significant at p<0.01. + comparisons are descriptive only due to small cell sizes in 1996/97.

Figure 2. Prevalence of Antipsychotic Use by Characteristics

FPL=Federal Poverty Level. In 2006–7, multiple racial categories were allowed; in these years white and black both indicate no other categories were recorded. Some cell sizes for 1996/97 are smaller than 100 observations as noted in Appendix Table 2. * chi-square tests for discrete variables were significant at p<0.01. + comparisons are descriptive only due to small cell sizes in 1996/97.

Figure 3. Prevalence of Antipsychotic Use by Self-reported Psychiatric Condition

Note: conditions are reported by survey respondents and are not clinical diagnoses. Affective disorders include only those in the 296 ICD-9 category and exclude those individuals who report schizophrenia (295) in the same survey year. Similarly, anxiety spectrum disorders include only conditions in the ICD-9 category of 300 and exclude those individuals who report either schizophrenia or 296 affective disorders.