This section assesses demand for SUD treatment. We begin by reviewing estimates of the prevalence of substance use and disorders at the national and state levels in 2013. We then assess recent trends in substance use and disorders.
A. Prevalence-Based Estimates of Demand
In this section we look at the prevalence of SUDs. The findings reported here are based on previously published analyses of the NSDUH. Overall, estimates of the prevalence of SUDs based on the NSDUH are probably conservative because institutionalized populations and homeless not living in shelters are excluded and survey respondents tend to underreport drug and alcohol use (Hoyt et al. 1994; Harrell 1997). The first section below addresses current prevalence of SUDs. The section that follows addresses trends in SUDs over the last decade.
1. Current Prevalence
This section initially reports national-level estimates and then addresses state-level estimates.
a. National-Level Estimates
Use of illicit drugs is generally limited to a small share of the population (Figure II.1). Both cocaine and hallucinogens had a 2 percent prevalence, inhalants had a 1 percent prevalence, and heroin less than a 1 percent prevalence. Use of marijuana and psychotherapeutics7 was much more common with a 13 percent and 6 percent prevalence rate, respectively.
Similar to usage, marijuana also had the highest percentage of abuse and dependence in 2013, with 0.5 percent abuse and 1.1 percent dependence (Figure II.2). Psychotherapeutics had the second highest abuse and dependence rates, with percentages of 0.2 percent and 0.7 percent, respectively. Heroin and cocaine both had dependence rates of 0.2 percent, while cocaine, hallucinogens, and inhalants all had abuse percentages of 0.1 percent.
|FIGURE II.1. Illicit Drug Use in Past Year among Persons Age 12 and Over, NSDUH 2013|
|SOURCE: Estimates obtained from SAMHSA 2015a, Table 7.2B.|
A little over half of people age 12 and over reported using alcohol in the previous month, but only 7 percent of the population met criteria for an alcohol use disorder in the past year (Figure II.3). Four percent of people met diagnostic criteria for alcohol abuse in 2013 but were not alcohol dependent (SAMHSA 2015a). Abuse of alcohol is defined as drinking hazardously, not fulfilling key responsibilities at work, school, or home due to drinking, legal problems associated with drinking, and drinking causing social problems (HHS Centers for Disease Control and Prevention [CDC] 2015a). Only 3 percent of the population met diagnostic criteria for alcohol dependence (SAMHSA 2015a), which is defined as an intense desire for alcohol, inability to set bounds for alcohol consumption, and repeated physical and personal issues as a result of alcohol use (CDC 2015a).
Beyond these diagnostic conditions, public health officials are concerned about other harmful patterns of drinking. Twenty-three percent of the population reported binge drinking in the past 30 days (SAMHSA 2015a), defined as consuming five or more drinks on the same occasion. Binge drinking is dangerous because it impairs brain function; results in a feeling of warmth but a loss of body heat; elevates risk for certain cancers, stroke, and liver diseases (for example, cirrhosis); can harm a developing fetus; increases the likelihood of motor-vehicle traffic crashes, violence, and other accidents; and in extreme cases, when large amounts are consumed in a brief time frame, may result in coma and death (CDC 2015a). Six percent of people reported heavy drinking (SAMHSA 2015a), that is, consuming five or more drinks on the same occasion on five or more days in the past 30 days. All heavy drinkers are binge drinkers. In addition to the health problems related to binge drinking, heavy drinking is linked to Sudden Infant Death Syndrome and risk for alcohol abuse and dependence (CDC 2015a).
|FIGURE II.2. Prevalence of Illicit Drug Abuse and Dependence in Past Year among Persons Age 12 and Over, NSDUH 2013|
|SOURCE: Estimates obtained from SAMHSA 2015a, Table 7.40B.|
|FIGURE II.3. Past Month Use and Past Year Abuse and Dependence on Alcohol among Persons Age 12 and Over, NSDUH 2013|
|SOURCE: Estimates obtained from SAMHSA 2015a, Table 7.29A, 7.29B, and Table 7.41B.
NOTE: Use is based on past month, and dependence and abuse are based on past year. Binge alcohol use is defined as drinking 5 or more drinks on the same occasion (that is, at the same time or within a couple of hours of each other) on at least 1 day in the past 30 days. Heavy drinking is defined as drinking 5 or more drinks on the same occasion on each of 5 or more days in the past 30 days; all heavy alcohol users are also binge alcohol users.
b. State-Level Estimates
State-level estimates of substance use reveal variance in patterns of usage across states. These differences are associated with several factors, such as differences in social norms, population demographics, and urbanicity.
In general, illicit drug use other than marijuana is highest in the Western and Southern states and lowest in the Midwest (Figure II.4). The 2013 NSDUH found that illicit drug usage was most common in urban areas and least common in less urbanized or rural areas (SAMHSA 2014a).
|FIGURE II.4. Illicit Drug Use Other Than Marijuana in the Past Month among Individuals Age 12 and Over, by State: Annual Average Percentages based on NSDUH 2012 and 2013|
|SOURCE: Graphic reprinted from SAMHSA 2015b.|
|ALT TEXT for FIGURE II.4, State Chart:
Marijuana use appears more prevalent in the West and Northeast, and less prevalent in the Midwest and the South (Figure II.5). Such trends may reflect, in part, local laws regarding medical marijuana use.
|FIGURE II.5. Marijuana Use in the Past Year among Individuals Age 12 or Older, by State: Percentages, Annual Averages based on NSDUH 2012 and 2013|
|SOURCE: Graphic reprinted from SAMHSA 2015b.|
|ALT TEXT for FIGURE II.5, State Chart:
In contrast to illicit drugs and marijuana, binge alcohol use appears to be more prevalent in the Midwest and less prevalent out West and in the South (Figure II.6).
|FIGURE II.6. Binge Alcohol Use in the Past Month among Individuals Age 12 or Older, by State: Percentages, Annual Averages based on NSDUH 2012 and 2013|
|SOURCE: Graphic reprinted from SAMHSA 2015b.|
|ALT TEXT for FIGURE II.6, State Chart:
Alcohol dependence patterns appear to be similar to that of binge drinking, with higher prevalence in the Midwest and lower prevalence in the South; however the West falls into higher categories for dependence than binge drinking (Figure II.7).
|FIGURE II.7. Alcohol Dependence in the Past Year among Individuals Age 12 or Older, by State: Percentages, Annual Averages based on NSDUH 2012 and 2013|
|SOURCE: Graphic reprinted from SAMHSA 2015b.|
|ALT TEXT for FIGURE II.7, State Chart:
2. Trends in SUD Prevalence
An important component in projecting demand for SUD treatment is examining recent trends in condition prevalence and considering whether these trends are likely to persist. In this section, we describe national and state-level trends in alcohol use, illicit drug use, and SUDs between 2002 and 2013, based on previously published estimates from NSDUH. Detailed tables on these trends from 2002-2013 are provided in Appendix A. We compared 2013 estimates with estimates for each year between 2002 and 2012 and report on statistically significant differences. This section initially reports national-level estimates and then addresses state-level estimates.
a. National Estimates
Based on data from the NSDUH, the proportion of individuals age 12 and older who met diagnostic criteria for an SUD8 in the past year (Figure II.8 and Appendix Table A.5) remained relatively constant from 2002-2010 (ranging from 8.8 percent to 9.4 percent) and then decreased in 2011 through 2013 (ranging from 8.0 percent to 8.5 percent). The estimate for 2013 is significantly lower than the estimates for each year between 2002 and 2010. The 2013 estimate is significantly lower than the estimates for 2002 through 2004 for both alcohol, and separately for illicit drug disorders.
|FIGURE II.8. Proportion of Individuals Age 12 and Older with Past Year Abuse or Dependence on Alcohol or Illicit Drugs, NSDUH 2002-2013|
|SOURCE: Estimates obtained from SAMHSA 2015a, Table 7.40B.|
Although there was a slight decline in current prevalence of SUDs between 2002-2010 and 2011-2013, accounting for population growth, the actual number of individuals meeting criteria for an SUD in the past 12 months in 2013 did not differ significantly from any year between 2002 and 2012 (Figure II.9 and Appendix Table A.4).
|FIGURE II.9. Number of Individuals Age 12 and Older with Past Year Abuse or Dependence on Alcohol or Illicit Drugs, NSDUH 2002-2013|
|SOURCE: Estimates obtained from SAMHSA 2015a, Table 7.40A.|
Underlying these overall trends in abuse and dependence are some shifts in substance use and opposing trends for some substances.
Use of Illicit Drugs (Appendix Tables A.1). Relative to 2011, the number of people over 12 years of age who had used illicit drugs in the past year was significantly higher in 2013 by 8.6 percent (3.3 million people). This increase was largely due to a 10.8 percent increase (3.2 million people) in use of marijuana. The number of people using heroin in 2013 was significantly higher than in each year from 2002-2008 with the exception of 2006. In contrast, use of cocaine was significantly lower in 2013 relative to each year from 2002-2009, and use of inhalants was significantly lower in 2013 relative to each year from 2002-2011.
Dependence or Abuse of Illicit Drugs (Appendix Table A.4). Although there is no significant difference between the overall number of people with illicit drug abuse or dependence in the past year between 2013 and any year from 2002-2012, there were some significant changes in the number of individuals with disorders associated with specific substances in this time period. Relative to 2010, there was a significant 31 percent decline (125,000 people) in the number of people with abuse or dependence on hallucinogens by 2013. Similarly, relative to 2009, there was also a significant 22.8 percent decline (about 250,000 people) in abuse or dependence on cocaine by 2013. The number of individuals with cocaine-related disorders ranged from 1.5 million to 1.7 million from 2002-2007, but in 2013-2014 estimates indicate only 0.9 million persons with cocaine-related disorders. In contrast, the number of individuals with heroin-related disorders ranged from 189,000 to 324,000 from 2002-2008, but in 2014 estimates indicate 586,000 individuals with heroin disorders. The number of individuals with disorders related to pain relievers ranged from 1.4 million to 1.5 million from 2002-2005. This increased to 1.9 million by 2013-2014.9
Use of Alcohol (Appendix Table A.3). Relative to 2002, by 2013 the number of people using alcohol and binge drinking in the past month increased significantly by 14.2 percent (17 million people) and 11.7 percent (6.3 million people), respectively. The number of heavy drinkers did not change significantly.
Dependence or Abuse of Alcohol (Appendix Table A.4). Relative to 2009, by 2013 the number of people with abuse of or dependence on alcohol in the past year declined by 7.8 percent (1.5 million people).
Overall, illicit drug use in the past year has increased recently; however this increase is largely due to increased use of marijuana and has not translated into a change in the number of people with illicit drug abuse or dependence. Similarly, the number of people using alcohol recently increased, but the increased use did not result in observed increases in abuse or dependence. Thus, we would not expect substantial shifts in treatment demand associated with these observed trends in use and disorders.
b. State-Level Estimates
As noted above, nationally, marijuana use had increased in recent years. This increase in use was not consistent across states. Only one region, the South, and a minority of states showed a significant increase in marijuana use based on the NSDUH's 2011-2012 and 2012-2013 regional and state-level estimates (Appendix Table A.6). The states with a significant increase in use were: Colorado, the District of Columbia, Georgia, Hawaii, Maine, Maryland, Michigan, Missouri, New Hampshire, Utah, Virginia, and Washington. States with significant decreases in use were Mississippi, New Jersey and South Dakota.
The South was also the only region to have a significant increases in illicit drug use other than marijuana between 2011-2012 and 2012-2013. At the state level, illicit drug use other than marijuana increased significantly in the District of Columbia, Georgia, Louisiana, Maryland, and North Carolina, while it decreased significantly in Idaho, Indiana, Minnesota, Montana, and West Virginia. Lastly, the South was also the only region to have a significant increase in alcohol dependence or abuse. Alcohol dependence or abuse increased in Alabama, Kansas, North Carolina, and Virginia and decreased in Alaska, Illinois, Iowa, Minnesota, Nevada, and Oklahoma.
B. Market-Based Estimate of Demand
In this section, we review the level of SUD treatment services currently provided. We address the number of individuals receiving any care, specialty care in the community, and care in an institutional criminal justice setting. Then, we describe how expenditures for SUD treatment are allocated across settings of care.
1. Number of Persons in Care
SUD treatment services are provided in a variety of settings including inpatient hospital, residential, and outpatient. SUD care can be provided in general health care settings such as primary care physician offices or community health centers; however SUD treatment is typically provided in specialty facilities or specialty care units with SUD treatment programs. This section first reports estimates from NSDUH on the number of persons receiving any care in the last year in specialty and non-specialty settings. Then it reports findings from the N-SSATS on care received in specialty settings. Finally, because incarcerated individuals are not included in the NSDUH and N-SSATS surveys, in the last section we look at care received in a criminal justice setting.
a. National Survey of Drug Use and Health
NSDUH collects data on all types of SUD treatment services ranging from self-help group services to services in an inpatient setting. According to NSDUH estimates, in 2013, 4.1 million people age 12 or older (1.5 percent of the population) received any treatment for a problem related to the use of alcohol or illicit drugs (SAMHSA 2014b). Services were provided in the following settings. (Individuals may receive services from more than one setting.)
- 2.3 million received treatment at a self-help group.
- 1.8 million received treatment at a rehabilitation facility as an outpatient.
- 1.2 million received treatment at a mental health center as an outpatient.
- 1.0 million received treatment at a rehabilitation facility as an inpatient.
- 879,000 received treatment at a hospital as an inpatient.
- 770,000 received treatment at a private doctor's office (including primary care).
- 603,000 received treatment at an emergency room.
- 263,000 received treatment at a prison or jail. (This is an undercount of the total number of people receiving treatment in these settings, as NSDUH is a community-based survey of non-incarcerated individuals.)
SUD treatment occurs predominantly in specialized facilities. Of the 4.1 million individuals who received any treatment, 2.5 million received treatment provided by a specialty SUD treatment provider. More than half of respondents who reported receiving inpatient treatment identified a rehabilitation facility as their source of inpatient care rather than a hospital. Among those receiving outpatient SUD treatment, treatment in a rehabilitation facility was over two times more frequent than treatment in a private doctor's office.
b. National Survey of Substance Abuse Treatment Services
The N-SSATS surveys all public and private specialty SUD treatment facilities in the United States including specialty SUD treatment units or programs in hospitals or mental health facilities. Solo practitioners10 and facilities serving only criminal justice clients are excluded from the N-SSATS. In contrast to the NSDUH, which contains information on individuals receiving treatment in general health care settings, the N-SSATS includes information only on clients receiving care in facilities or treatment units with treatment programs for SUDs.
In the 2013 N-SSATS, respondents were asked to report the number of inpatient and residential clients receiving care for SUDs in their facility on March 29, 2013. They were also asked to report on outpatient clients who received any SUD care at their facility in March 2013 and who were still in care at the end of March. Based on these reports, specialty facility clients were distributed as follows:
- 1,127,235 outpatient clients.
- 107,727 residential clients.
- 14,667 inpatient hospital clients.
The share of clients served in each of these settings remained similar between 2005 and 2013. Detoxification services, rehabilitative care, and medication-assisted treatment (MAT) may be provided in the inpatient, residential and outpatient setting.
According to SAMHSA (2006a) detoxification "denotes a clearing of toxins from the body of the patient who is acutely intoxicated and/or dependent on substances of abuse. Detoxification seeks to minimize the physical harm caused by the abuse of substances." Overall in March 2013, 2 percent of specialty facility clients (about 30,000 clients) were receiving detoxification services. Detoxification clients represented a larger share of clients within the inpatient and residential settings, 39 percent and 10 percent, respectively relative to the outpatient setting (1 percent) (Table II.1). However, the number of clients receiving detoxification services in the outpatient setting (13,839) was substantially larger than the numbers receiving detoxification services in the inpatient (5,768) and residential settings (10,244).
|TABLE II.1. Services Provided by Setting of Care, N-SSATS 2005, 2009 and 2013|
|Type of Care Within Setting||Number of Clients||Percent of Clients in Care Type|
|Day treatment/partial hospitalization||24,928||23,822||22,828||3||2||2|
|Methadone/buprenorphine maintenance or Vivitrol®||228,558||284,540||340,091||24||27||30|
|Clients receiving methadone, buprenorphine, or Vivitrol® treatment||240,935||308,781||382,237||100||100||100|
|Clients receiving methadone in OTP facilities||235,836||284,608||330,308||98||92||86|
|Clients receiving buprenorphine||5,099||24,173||48,148||2||8||13|
|Clients receiving Vivitrol®||N/A||N/A||3,781||N/A||N/A||1|
|SOURCE: Findings obtained from SAMHSA 2014a, Table 3.2.|
Clients not receiving detoxification services were receiving rehabilitative care. Rehabilitation services generally consist of abstinence-oriented counseling and education (Buck 2011). Many outpatient clients were receiving extended services in the outpatient setting (15 percent) including intensive treatment or day treatment/partial hospitalization. Similarly, many residential clients were receiving long-term services (65 percent). Rehabilitative care may be supplemented by MAT in any care setting. MAT is a form a pharmacotherapy that has been proven effective in alcohol and opioid dependence. In 2013, 31 percent (382,237 persons) of SUD treatment clients were receiving MAT with methadone, buprenorphine or Vivitrol® (Table II.1). This is a substantial increase from 22 percent (240,935) receiving these medications in 2005. These medications address heroin and pain reliever addiction. Vivitrol® may also be used to treat alcohol dependence.
c. Services Provided in the Criminal Justice System
The NSDUH and N-SSATS include only individuals currently living in the community. Incarcerated individuals are excluded from these surveys. Because many incarcerated individuals have SUD treatment needs, we discuss treatment for this population here.
Several studies have found a substantial percentage of prisoners are in need of SUD treatment. The National Center on Addiction and Substance Use at Columbia University (CASA 2010) reported that in 2006, 1.5 million (65 percent) of the 2.3 million inmates in United States prisons met the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) medical criteria for alcohol or other drug abuse or dependence in the year before their arrest. Similarly, Rounds-Bryant and Baker (2007) assessed both prevalence rates of substance dependence and level of treatment need among recently incarcerated prisoners in a Southeastern state in 2002 and found 72 percent were substance dependent and 46 percent needed prison-based residential treatment. Using data from the 1997 Survey of Inmates in State Correctional Facilities, a nationally representative sample of inmates, Belenko and Peugh (2007) found that one-third of male and half of female prisoners needed residential treatment. Overall, rates of co-occurring mental disorders and SUDs are over 6-10 times higher among people in the criminal justice population than people not in the system (Cropsey et al. 2011).
|TABLE II.2. Number and Percent of Incarcerated Individuals with SUDs Who Received Treatment, 2006|
|Type of Service||Federal Prison||State Prison||Local Jail||Total|
|Number (thousand)||%||Number (thousand)||%||Number (thousand)||%||Number (thousand)||%|
|Any professional treatment since admission||16.4||15.7||120.5||14.2||26.2||5.2||163.1||11.2|
|Residential facility or unit||9.2||8.8||78.1||9.2||15.6||3.1||102.9||7.1|
|Counseling by a professional||8.2||7.8||55.1||6.5||11.6||2.3||74.9||5.2|
|Other addiction-related services since admission||41.5||39.7||305.4||36.0||66.0||13.1||412.9||28.4|
|Mutual support/peer counseling||23.3||22.3||253.7||29.9||53.9||10.7||330.9||22.7|
|SOURCE: CASA'sanalysis (2010) of the Survey of Inmates in Federal Correctional Facilities (2004), Survey of Inmates in State Correctional Facilities (2004), Survey of Inmates in Local Jails (2002) (data files), and Bureau of Justice Statistics Reports, Prisoners in 2006.
NOTE: Participation in specific types of professional treatment or addiction-related services is not mutually exclusive.
According to research by CASA (2010), of the 1.5 million prison and jail inmates who met clinical diagnostic criteria for an SUD in 2006, only 11.2 percent had received professional treatment since admission (Table II.2). According to this research, care in specialized settings is associated with reduced drug use and arrests post-release, but very few criminal justice facilities (16.6 percent) offer treatment in such settings (data not shown). Moreover, few inmates actually receive evidence-based services, including access to pharmacological treatments and counseling from certified staff.
2. Total Expenditures for Care
Another way to look at demand for care is measuring the amount spent on care (since quantity of services and intensity of services are difficult to measure). The SAMHSA spending estimates initiative draws from multiple data sources to develop a comprehensive view of the allocation of SUD treatment spending nationally, by setting and type of care. The largest share of spending (Figure II.10) is for outpatient care (43 percent). Residential and inpatient care represent 34 percent and 23 percent of spending, respectively.
|FIGURE II.10. Percentage of SUD Treatment Spending by Type of Care, 2009|
|SOURCE: Mathematica analysis of SAMHSA 2013c, Table A.6.|
SUD treatment is provided in many settings (Figure II.11). The largest share of spending is allocated to specialty SUD treatment facilities (38 percent).11 Hospital-based inpatient, residential and outpatient care (each of which includes physician services provided in these settings) also represent a large share of spending (38 percent). Hospital-based care is typically provided in specialized SUD treatment units. Specialty mental health facilities also account for a substantial share of SUD treatment spending (8 percent). While these facilities are primarily focused on mental health treatment, this spending is associated with care provided through programs for individuals with co-occurring disorders or in SUD treatment units that are part of facilities primarily focused on mental health services. Psychiatrist, non-psychiatrist physician office, and other professional offices represent 14 percent of spending; services provided through primary care physicians would be represented under non-psychiatrist physician office, which represent only 2 percent of treatment spending. The final 3 percent of spending is allocated to nursing home and home health providers.
|FIGURE II.11. Percentage of SUD Treatment Spending by Setting of Care, 2009|
|SOURCE: Mathematica analysis of SAMHSA 2013c, Table A.6.|
C. The Difference Between Rates of Treatment and Prevalence--The Treatment Gap
1. What is the Size of the Treatment Gap?
There is a substantial gap between the number of people estimated to need SUD treatment and the number of people who receive any specialty treatment in a given year. According to SAMHSA (2014b) in 2013, an estimated 22.7 million United States residents needed treatment for an SUD defined as having an SUD in the past year (21.6 million) or receiving specialty treatment for an SUD in the past year although not meeting criteria for a current disorder (1.1 million). Only 18 percent (4.1 million) of these individuals received any treatment including self-help or emergency care and only 11 percent (2.5 million) received treatment at a specialty facility. According to a review by Foster (2014), this rate of treatment receipt is substantially lower than that for common health conditions such as hypertension (77 percent), diabetes (73 percent) and major depression (71 percent). However, as noted above, SUD diagnostic criteria may give rise to inflated estimates of need as compared to other estimates of need that may include further elaboration of criteria for dysfunction or harm, or self or family-perceptions of need (Wakefield and Schmitz 2015; Mechanic 2003). Nonetheless, this treatment gap is of concern to public health officials because continuation of these disorders negatively affects the health and lives of these individuals and also produces adverse consequences borne by many others in society (Bouchery et al. 2011; Bouchery and Harwood 2001). For example, these conditions are costly to the health care system, reduce economic productivity, and increase criminal justice expenditures.
2. Why is there a Treatment Gap?
|FIGURE II.12. Number of Individuals Who Received Specialty Treatment or Felt They Needed It, but did not Receive Treatment, 2004-2013|
|SOURCE: NSDUH estimates for 2004-2013 were obtained from SAMHSA 2005, SAMHSA 2006, SAMHSA 2007b, SAMHSA 2008, SAMHSA 2009, SAMHSA 2010, SAMHSA 2011a, SAMHSA 2012a, SAMHSA 2013b, SAMHSA 2014b, respectively.
NOTE: Treatment receipt is reported within the last 12 months.
According to SAMHSA (2014b) the individuals who meet criteria for an SUD but who do not receive treatment fall into three groups: (1) those who do not feel they need treatment; (2) those who feel they need treatment but do not seek treatment; and (3) those who feel they need treatment and seek it. Based on responses to the 2013 NSDUH, 95.5 percent of individuals who met criteria for an SUD but did not receive specialty treatment (20.2 million people) did not feel they needed treatment. Among the remaining small percentage (4.5 percent, or 908,000 people) who felt they needed treatment but did not get it, 65.2 percent (about 592,000 people) reported making no effort to get treatment, while 34.8 percent (about 316,000 people) reported that they did make such efforts. Figure II.12 displays trends in the number of individuals who received specialty treatment; felt they needed specialty treatment but did not seek it; and felt they needed specialty treatment and sought it but did not receive it. These numbers were relatively constant from 2004 through 2013.
According to SAMHSA (2014b), those who felt they needed treatment but did not seek care (2.7 percent of the population with an SUD) identified the following reasons for not seeking it. (Respondents could note more than one reason.)
- 40.3 percent reported they were not ready to stop using alcohol or drugs.
- 31.4 percent reported having no health coverage and could not afford cost of treatment.
- 10.7 percent reported possible negative effect on their job.
- 10.1 percent reported concern that receiving treatment might cause neighbors/community to have a negative opinion of them.
- 9.2 percent reported not knowing where to go for treatment.
- 8.0 percent reported no program available having the type of treatment needed.
Those who felt they needed SUD treatment and sought but did not receive it (1.5 percent of the population with an SUD) reported the following reasons for not receiving treatment (SAMHSA 2014b). (Respondents could identify more than one reason.)
- 37.3 percent reported having no health coverage and could not afford cost of treatment.
- 24.5 percent reported not being ready to stop using drugs or alcohol.
- 9.0 percent did not know where to go for treatment.
- 8.2 percent had health coverage but it did not cover treatment or did not cover cost.
- 8.0 percent reported it was inconvenient or they lacked transportation.
- 6.6 percent reported it might have negative effect on job.
- 6.6 percent could handle problem without treatment.
- 5.0 percent did not feel they needed treatment at the time.
Several relatively small studies have been conducted in recent years to identify barriers to receiving SUD treatment. Many of the barriers identified were consistent with reasons for not receiving treatment identified in NSDUH, including lack of health coverage or inability to afford costs (Brubaker 2013; Appel and Oldak 2007; MacMaster 2013); not being ready to stop using drugs or alcohol (Appel and Oldak 2007); social stigma (Jackson and Shannon 2012; Masson 2012); lack of transportation (Sigmon 2014; Appel and Oldak 2007); and lack of a program suited to the individual's treatment needs (MacMaster 2013; Rapp 2006).
Other studies have also identified barriers not included among the NSDUH responses. In a survey of 145 homeless men, respondents identified having no place to stay to begin treatment as a barrier (Brubaker 2013). Appel and Oldak (2007) studied barriers for injection drug users who were targets for outreach. Outreach and agency staff identified administrative application burdens, whereas common barriers clients reported were fear of treatment and anticipated negative consequences like potential loss of child custody. Rapp and others (2006) and Xu and others (2008) found fear of treatment and privacy concerns to be barriers. Jackson and Shannon (2012), in a study of pregnant women, found the most common barriers to treatment reported were family responsibilities and administrative issues (such as waiting periods, paperwork, and treatment referrals). Peterson and others (2010) found that waiting lists and requirements for photo identification, as well as concern about taking methadone, acted as barriers for methadone treatment. While NSDUH is the most nationally representative source of information about why people do not receive care, the population in need of SUD treatment varies substantially and these smaller studies may provide insight into the reasons particular subpopulations do not access care.
D. Key Factors Influencing Demand for Care
The existing literature has assessed the influence of several key factors on demand for care.
1. Increased Recognition of SUDs as Health Conditions
As part of the Vital Signs survey (Ryan et al. 2012), thought leaders were asked to identify trends that would affect the workforce over the next five years. Seventy percent identified a trend toward increasing recognition of SUDs as valid health conditions. The ACA recognizes SUDs as health conditions and expands health insurance coverage of them through several mechanisms, including identifying SUD treatment as an essential health benefit; requiring coverage of SUD screening, brief intervention, and referral to treatment (SBIRT); and promoting integration of SUD treatment into primary care and mental health treatment settings. The MHPAEA may improve recognition of SUDs by promoting parity in coverage.
The thought leaders interviewed noted that these provisions may also change perceptions on the need for professional treatment for SUDs among both affected individuals and treatment providers. Researchers from the CASA (2012) report that despite the documented benefits of screening and referral to treatment, few primary care health professionals screen for or treat SUDs, educate patients about SUDs, or refer patients with SUDs to specialty SUD treatment providers. Only 6 percent of referrals to SUD treatment come from general health care providers. Changes in the current perceptions of SUD treatment among individuals with SUDs and primary care providers may have a substantial influence on trends in demand for services.
2. Funding for SUD Treatment
The primary driver of increased SUD treatment projected from the ACA is insurance coverage expansion. However, traditionally, most SUD treatment services have been financed by state, local, and federal sources rather than by Medicaid or private health insurance. In 2009, 31 percent and 11 percent of SUD treatment spending was financed by state/local and federal sources, respectively, excluding Medicare and Medicaid spending (Figure II.13). Five percent and 21 percent of SUD treatment spending was financed by Medicare and Medicaid, respectively. Private health insurance comprised only 16 percent of SUD treatment spending in 2009 (SAMHSA 2013c). The substantial share of funding represented by other state, local, and federal sources allows many individuals who would otherwise not be able to afford treatment to receive treatment at no or a reduced cost.
|FIGURE II.13. Percentage of Spending by Type of Care, 2009|
|SOURCE: Mathematica analysis of SAMHSA 2013c, page 37.|
Although there has been extensive analysis of the relationship between insurance coverage and use of health care services in the general health field, only a small number of studies have explicitly examined the influence of insurance coverage on the receipt of SUD treatment. The studies related to general health are not applicable to SUD treatment because SUD treatment is typically provided outside the general health sector, insurance coverage may be less comprehensive, and there is social stigma attached to SUDs and receipt of treatment for SUDs. In addition, states and the Federal Government (through the use of block grants) fund SUD treatment directly. While in other areas, the receipt of health insurance may make care more affordable, in the case of SUD treatment, because of this public financing, a privately insured person may actually face higher costs due to copayments or coinsurance, or disincentives to use of SUD treatment than individuals without insurance. The findings from studies on the impact of insurance coverage on SUD treatment use are therefore mixed, with several finding that people without insurance access SUD treatment services at greater or the same rates as those who are insured. Other studies find that financial obstacles, including lack of insurance coverage are a significant barrier to treatment receipt. While public and charitable financing of SUD treatment may allow many uninsured individuals to access treatment, free or low cost services may not be available in all communities or to all individuals for whom financial barriers are an obstacle. Several studies have found that individuals with Medicaid coverage are more likely to use SUD treatment than those with private insurance. This may be due to lower coinsurance requirements under Medicaid or differences in the care management and benefit packages provided through Medicaid and private insurance plans. Additionally, on average individuals with Medicaid differ substantially from individuals with private insurance. These differences include severity of the disorders, comorbidities, prior treatment histories, and pressure from employers or legal or court pressure to get treatment (Schmidt and Weisner 2004).
Several studies have found that people without insurance access SUD treatment at greater or the same rates as individuals with insurance, however, many of these studies do not control for the differences in populations noted above. Through the 1999 National Household Survey on Drug Abuse (NHSDA--the precursor to NSDUH), Wu and Ringwalt (2005) found that uninsured young adults were more likely to use self-help or non-medical social services sector SUD services than their privately insured counterparts. Similarly, Mojtabai (2005) found through the 2002 NSDUH that the uninsured accessed SUD treatment at similar rates as the insured. Lastly, based on survey data from a sample of problem drinkers drawn from the general population and chemical dependency treatment programs, Schmidt and Weisner (2005) found that among those with alcohol disorders in the general population, individuals with Medicaid and those without insurance were as likely to enter into treatment as people with private insurance. Bouchery and colleagues (2012), using 2002-2007 NSDUH data, similarly found that uninsured individuals had similar rates of SUD treatment receipt relative to those with private insurance.
In contrast, other studies have found lack of insurance coverage or inability to pay to be significant barriers to SUD treatment receipt. Based on data from the longitudinal study, Reducing Barriers to Drug Abuse Treatment Services, Xu and colleagues (2008) found that lack of insurance coverage and inability to afford treatment were barriers to treatment entry. Likewise, Saum and colleagues (2007) concluded that having health insurance (along with other enabling factors) was influential in predicting treatment entry for cocaine-dependent women who were involved with the criminal justice system. Other researchers reporting findings from surveys of nationally representative populations or more narrow subpopulations found inability to pay or lack of insurance to be barriers to SUD treatment (Brubaker et al. 2013; MacMaster 2013; Peterson et al. 2010; SAMHSA 2014a).
Several studies have suggested that individuals with Medicaid coverage are more likely to use SUD treatment than those with private insurance. Epstein and colleagues (2004), using pooled data from the 2000 and 2001 NHSDA, found that having Medicaid, as opposed to private or no insurance, significantly increased the likelihood of specialty treatment for drug use. Larson and colleagues (2005) found that among 267 people in Massachusetts who received substance abuse counseling, access to Medicaid insurance was associated with greater access to care and higher consumer ratings of care. Bouchery and colleagues (2012), using 2002-2007 NSDUH data, similarly found that those on Medicaid had significantly higher rates of SUD treatment receipt relative to those with private insurance. Using 2008-2012 NSDUH data, Ali and colleagues (2014) found that individuals with Medicaid had a significantly higher and individuals with private insurance a significantly lower likelihood of using any SUD treatment relative to individuals with other insurance types (including Veteran's insurance or Civilian Health and Medical Program of the Uniformed Services). There was no significant difference in the likelihood of SUD treatment use between the uninsured and those with other insurance.
The mixed findings in this literature may result from the different benefit structures and treatment type coverage of the insurance and different population characteristics by insurance coverage type as noted above. For a wealthier population, insurance may make treatment at a private provider more affordable. For a lower income population, copays or coinsurance requirements might be barriers to treatment receipt even if insurance provides some coverage for services. In addition, the studies may have imprecise measures, methodological limitations and particular contexts. Many of the studies analyzed the relationship between treatment use and insurance coverage in cross-sectional data. There may be unobserved differences between populations with different insurance types, such as severity of condition and employment, which could confound the findings. We did not identify any studies using an experimental design to examine the relationship between insurance coverage and the receipt of SUD treatment. Given the mixed findings in the literature on insurance coverage, there is little basis upon which to project the change in demand for services associated with recent increases in insurance coverage for SUDs.
4. Racial and Ethnic Disparities
A number of studies highlighted racial and ethnic differences in entry into and unmet need for SUD treatment; however, the findings differ somewhat by type of SUD. For example, one study found that non-Hispanic White Americans were more likely to address substance abuse by accessing care through specialty addiction treatment facilities and less likely to access it through non-specialty facilities than were members of racial/ethnic minority groups (Lo and Cheng 2011). Asian Americans and Pacific Islanders underutilized substance abuse treatment services (Masson et al. 2013), and Asians with past-year SUDs had a higher likelihood of having an unmet need for treatment than did people who were White (Mulvaney et al. 2012); however, Asians with heavy drinking/illicit drug use had lower likelihood of unmet need (Mulvaney et al. 2012). An analysis of administrative data from the Oklahoma Department of Mental Health and Substance Abuse Services linked to other state data, such as criminal justice and employment system data, found that Black clients were significantly less likely to initiate treatment than White clients (Acevedo et al. 2012). Among people with SUDs, mental health problems, and HIV/AIDS, Blacks were less likely than Whites to utilize mental health or substance abuse treatment (Weaver et al. 2008). Another study found that Black adolescents received less specialty and informal care compared with adolescent Whites, and adolescent Latinos also received fewer informal services than adolescent Whites (Alegria et al. 2011). In contrast, in another study, Blacks with past-year disorder and with heavy drinking/illicit drug use had significantly lower likelihood of unmet need (Mulvaney et al. 2012).
While several studies suggest lower levels of access for racial/ethnic minorities, the findings are not consistent. Differences in findings are likely to be attributed to the definition of the population served and the measures of access used, which varied across the studies reviewed. Differences in how individuals from specific subgroups accessed services and the types of services they used may also contribute to differences in findings across studies.
E. Trends and Policies Affecting Future Demand for Care
In this section, we highlight trends and policies likely to have a substantial impact on demand for SUD treatment over the next five years. These include population growth, the ACA, MHPAEA, marijuana legalization, opioid use epidemic, and trends in government spending on SUD treatment.
1. Population Growth
Between 2015 and 2020 the United States youth and adult population aged 14 and older is projected to grow 4.8 percent (Table II.3). The younger age groups including individuals less than 25 are expected to shrink while the population over age 65 is expected to grow rapidly. Because prevalence rates of SUDs vary by age (SAMHSA 2014b) and are highest within age groups expected to shrink, demand for SUD treatment may not keep pace with overall population growth. In 2013 the SUD prevalence rate was highest for individuals aged 18-25 (17.3 percent), followed by individuals 26 or older (7.0 percent). Youth aged 12-17 had the lowest prevalence rate at 5.2 percent. There is concern that need for treatment among older age groups may increase over time as the baby boomers age given their higher rates of substance use relative to previous generations (SAMHSA 2013a).
|TABLE II.3. Projected Population Growth Trends, 2015-2020|
|Age Group||2015||2020||Percent Change|
|14 to 17||16,796||16,737||-0.4|
|18 to 24||31,214||30,555||-2.1|
|25 to 44||84,657||89,518||5.7|
|45 to 64||84,032||83,861||-0.2|
|65 and over||47,830||56,441||18.0|
|SOURCE: U.S. Census Bureau, Population Division, Table 3: Projections of the Population by Sex and Selected Age Groups for the United States: 2015 to 2060 (NP2014-T3), December 2014 release.|
This section lists the provisions of the ACA most likely to affect SUD treatment use. Then, it summarizes existing estimates of the ACA's effects on SUD treatment use.
a. Provisions Affecting SUD Treatment
The ACA expands coverage for SUD services through the following mechanisms:
Reduction in the Number of Uninsured. The ACA reduces the number of uninsured in several ways. First, in those states that have opted for Medicaid expansion, Medicaid eligibility is expanded to include citizens in households below 138 percent of the federal poverty level (FPL) and legal residents who meet a five-year waiting period and live in households below 138 percent of the FPL. Second, the ACA has created marketplaces for the purchase of private insurance coverage. Individuals in households with an income from 100 percent to 400 percent of FPL can qualify for subsidies to make purchasing insurance through the marketplaces more affordable. Additionally, marketplace plans can no longer deny coverage on the basis of pre-existing conditions, and treatment for pre-existing conditions begins as soon as coverage begins (HealthCare.gov 2014). Finally, the ACA has extended the age limit for dependents to remain on a parent's insurance policy to age 26 (Humphreys and Frank 2014).
Inclusion of SUD Treatment as an Essential Health Benefit. As of January 1, 2014, SUD treatment is an essential health benefit for all Americans in non-grandfathered plans in the individual and small group markets (Beronio et al. 2013). Medicaid benchmark and marketplace plans must also treat SUD treatment as an essential benefit (Barry and Huskamp 2011), with no lifetime or yearly dollar limit on services (HealthCare.gov 2014). SUD screening, brief intervention, and treatment are all included as essential benefits (Humphreys and Frank 2014). This provision will result in enhanced SUD treatment benefits for many individuals.
Service Delivery System Changes. Provisions of the ACA promote care integration, including integration of SUD treatment into primary care and mental health treatment settings. We discuss trends in integration as a supply-side trend in Section III.B.3.
Changes in Medicaid Coverage of SUD Treatment Services. States that expanded their Medicaid programs under the ACA have accepted large numbers of childless adults into their Medicaid populations, and these individuals are expected to have greater need for SUD treatment services relative to the traditional Medicaid population. We discuss Medicaid coverage changes in Section II.E.6 below.
b. Existing Estimates of Projected Impact on Demand
This section discusses existing estimates for how the ACA has affected or may in the future affect demand for SUD treatment. We begin by discussing changes associated with implementation of the ACA in the number of people who are uninsured. Then, we discuss the number of insured individuals who obtained enhanced benefits for SUD treatment through the ACA. Next, we look at available research on how obtaining insurance coverage may affect access to and use of SUD treatment. Finally, we summarize existing research on how the ACA may influence demand for SUD treatment.
c. How Many People are Newly Insured?
Many people who signed up for insurance through a marketplace or enrolled in Medicaid in 2014 might have obtained private or Medicaid insurance coverage in the absence of the ACA. Therefore, this section discusses the change in the number of uninsured Americans rather than counts of people signing up for insurance through marketplaces, as this is a better measure of the number of individuals who gained insurance as a result of the ACA. Recent surveys conducted by five entities--Rand Corporation, Commonwealth Fund, Gallup, the Urban Institute, and the CDC--indicate that between 2013 and 2014, the number of Americans without health insurance decreased by about 25 percent (8-11 million people).12 More than half of these newly insured individuals signed up for Medicaid. In addition to the changes observed between 2013 and 2014, 3-4 million people gained coverage through ACA provisions that took effect before 2014 (Sanger-Katz 2014).
Projections of the number of newly insured from the CBO and the Joint Committee on Taxation (JCT) are consistent with these observed changes. For 2014, CBO projected 12 million fewer uninsured non-elderly individuals. CBO and JCT projections through 2024 are as follows (Figure II.14):
Reductions in Uninsured. Relative to 2013 levels, the number of uninsured individuals will be reduced by 19 million non-elderly individuals by 2015, by 25 million by 2016, and by 26 million by each year from 2017 through 2024 (CBO 2014).
Marketplaces Coverage Use. Thirteen million people are projected to obtain coverage though marketplaces in 2015, 24 million in 2016 when more substantial tax penalties are implemented, and 25 million in 2017-2024 (CBO 2014).
Medicaid and CHIP Enrollment. An additional 11 million people are projected to obtain Medicaid or CHIP in 2015, 12 million in 2016 and 2017, and 13 million in 2018-2024 (CBO 2014).
The estimated reduction in the uninsured is lower than the total number of individuals estimated to obtain health insurance through the marketplaces, Medicaid, and CHIP in this period because some people will shift from employer, non-group or other coverage to marketplace, Medicaid, or CHIP coverage.
Some states are considering ways in which Medicaid eligibility expansions could help individuals exiting jails and prisons because these individuals have a higher prevalence of SUDs than the general population (Boozang et al. 2014; CASA 2010). These states are connecting Medicaid and corrections through cooperation among managed care organizations, social service organizations, and correctional facilities; building relationships among local substance abuse agencies, police, and jails; and creating protocols for crisis services, jail diversion, and safety. Boozang and colleagues (2014) reported that 20-30 percent of new Medicaid enrollees in 2014 were likely to be individuals recently released from jail or prison. Bainbridge (2012) found that 33.6 percent of inmates released annually are eligible for Medicaid coverage, with another 23.5 percent eligible for marketplace subsidies. Bainbridge also notes that several studies focused on local populations predict large numbers of recently incarcerated people being eligible for Medicaid.
|FIGURE II.14. Reduction in Uninsured and Enrollment in Marketplace and Medicaid as a Result of the ACA, 2015-2020|
|SOURCE: Estimates obtained from CBO 2014.|
Thus, the ACA has increased and is expected to continue to increase the number of individuals with insurance coverage, greatly increasing the number of individuals with SUD treatment benefits.
d. How Many Previously Insured Individuals Gained SUD Treatment Coverage?
The ACA provides enhanced SUD coverage for individuals already insured. Overall, 35.5 million insured people are expected to gain enhanced coverage for SUD treatment because of the following:
SUD Benefit Added to Plan. Those without SUD coverage but with insurance prior to the ACA (approximately 1.2 million individuals in small group plans and 3.9 million people in the individual market) will gain SUD coverage as a result of the ACA (Beronio et al. 2013).13
Existing SUD Benefits Enhanced. Due to federal parity protections, approximately 7.1 million people in individual plans and 23.3 million people in small group plans who had some coverage for SUD treatment prior to the ACA will gain enhanced SUD treatment coverage (Beronio et al. 2013).
e. How Many People Who Gain Coverage will use SUD Treatment?
A few studies have examined how the use of SUD treatment may change in response to obtaining insurance coverage. The findings from these studies range from no effect to a substantial increase. The literature we identified includes the following findings:
Coverage Mandate for Adult Children. We identified one study that examined the effects of the coverage mandate for adult children, which applied to health insurance plans whose year began on or after September 23, 2010. This study used the 2008-2012 NSDUH to study SUD treatment use among adults ages 18-25 and found no increase in use of SUD treatment after the implementation of the coverage mandate (Saloner and Cook 2014).
SAMHSA SUD Spending Projections. SAMHSA periodically develops national projections for SUD treatment spending.14 The most recent available estimates are for 2014 through 2020 (SAMHSA 2014c). SAMHSA projects SUD treatment spending will increase 35 percent from 2014 through 2020. These projections assume an additional 25 million people will gain health insurance coverage through the ACA by 2020.15 The projections also assume a 25 percent increase in SUD treatment use for the individuals who newly gain insurance coverage. The size of this increase in use associated with obtaining insurance coverage is based on findings from the Oregon Health Insurance Experiment which provided low income uninsured individuals with Medicaid coverage.16 The findings on increased treatment use from the Oregon Experiment are not specific to SUD treatment and are based on individuals transitioning to Medicaid coverage. Increases in health expenditures associated with obtaining insurance coverage may be lower for SUD treatment than for other types of health care because of substantial existing public investment in SUD treatment for low income and uninsured individuals. This projection may also overstate the ACA impact on SUD treatment use because many of the newly insured are expected to gain coverage in private insurance plans. Although studies have found individuals with Medicaid coverage use SUD treatment at greater rates than the uninsured there is little evidence suggesting that obtaining private insurance coverage is associated with a substantial increase in service use (See Section II.D.3).
Experience from State Insurance Expansions. We identified two studies of state experiences with health insurance expansions that provide evidence on the potential effects of these policy changes.
Massachusetts Coverage Expansion. This mixed-method exploratory study analyzed the experience of Massachusetts in implementing a universal health insurance law similar to the ACA in 2007. The study found stable SUD treatment use two years before and two years after program implementation. A high percentage of individuals with SUD treatment needs remained uninsured because they did not enroll in insurance despite provisions of the law; for those who became insured, copayments were a deterrent to treatment (Capoccia et al. 2012).
Experience from Three State Insurance Expansions. This study conducted case studies of three states (Maine, Massachusetts, and Vermont) that implemented insurance coverage expansions intended to achieve universal coverage with similar provisions to the ACA.17 The case study focused on publicly funded facilities only. The study found that in Massachusetts, admissions to publicly funded SUD treatment increased 20 percent between 2006 and 2008. Relative to the study discussed in the prior bullet, this study did not look as broadly at multiple data sources and policy changes implement in the same time period as the coverage expansion to provide more perspective on the sources of observed shifts.
In Maine, the number of people admitted to publicly funded SUD treatment increased 45 percent between 1999 and 2008. In Vermont, the number of people admitted to publicly funded SUD treatment doubled between 1998 and 2007. Moreover, the experience in all three states indicates that despite insurance expansion, there was still a high percentage of uninsured individuals among those in need of SUD treatment. Thus, supplemental funding sources were needed to treat this population (Morrison et al. 2010).
Overall, the evidence for how the expansion of insurance coverage will influence treatment demand is very limited. The findings from these studies range from no effect to a substantial increase in referral to treatment. SAMHSA projects SUD treatment spending will increase 35 percent from 2014 through 2020, however this estimate may be overstated because there was limited evidence specific to SUD treatment upon which to base this projection.
The MHPAEA contains several provisions that have the potential to influence the demand for SUD treatment.
a. Provisions Affecting SUD Treatment
The 2008 MHPAEA introduced parity for behavioral health treatment with medical/surgical care. Previously, employers could offer more limited coverage for SUD treatment than for other specialty treatment areas. Under the MHPAEA, large group employers with more than 50 full-time workers could no longer use such a restriction; however, this change did not impact individuals covered through the individual or small group market. This change also did not impact individuals covered through Medicaid and Medicare except those in Medicaid managed care arrangements. Another limitation is that the act does not require employers to offer SUD benefits, so parity is only required if the employer's plan covers SUD treatment (Humphreys and Frank 2014; Beronio et al. 2014). The MHPAEA identifies limitations on frequency of treatment, copayments, coinsurance, and deductibles as possible plan limitations that must be in parity (Frank et al. 2014).
b. Existing Estimates of Projected Impact on Demand
Several studies have estimated the impact of the MHPAEA or state parity legislation on demand for SUD treatment. Overall, the findings of these studies are mixed, with some studies finding no effects and others finding substantial differences in demand before and after legislation was enacted.
Oregon's 2007 Parity Law. Using enrollment and claims data, people in five Oregon commercial plans required to implement parity were compared to individuals in plans exempt from parity in Oregon and Washington. McConnell and others (2012) estimated that spending for alcohol treatment services was significantly higher among individuals in plans subject to parity provisions, relative to those exempt from parity. But there was no significant difference in drug abuse treatment service expenditures between the two groups.
Vermont's Parity Law. Rosenbach and colleagues (2003) examined claims/encounter data for Kaiser/Community Health Plan and Blue Cross/Blue Shield of Vermont, two health plans that covered almost 80 percent of the privately insured population at the time parity was implemented. Both plans made changes to their behavioral health service offering. One plan shifted from primarily using indemnity contracts to using a managed care approach. The other plan, which used managed care prior to parity, implemented an inpatient diversion program to increase use of partial hospitalization and group therapy and reduce use of inpatient care. Following implementation of Vermont's parity law, this study found a 16-29 percent decrease in the number of SUD treatment users per 1,000 members after parity was implemented.
Multistate Analyses of Parity Laws. We identified two studies that analyzed the impact of parity laws across multiple states. The first study used data from TEDS and found that parity reduced the probability that a person with an SUD treatment admission was uninsured by 2.4 percentage points. The study also found that states that mandated parity had increases in SUD treatment admissions (Dave and Muckerjee 2008). The second study used a quasi-experimental design and data from the N-SSATS surveys for 2000-2008 to compare outcomes in states that had implemented parity legislation to those with no change in parity requirements. This study found the state parity laws resulted in treatment increases of 9 percent in all specialty SUD treatment facilities and 15 percent in specialty SUD treatment facilities accepting private insurance (Wen et al. 2013).
Federal Employees Health Benefit (FEHB) Program. We identified one study that analyzed the effects of parity regulations on use of services under the FEHB program. Busch and colleagues (2013) used administrative data from FEHB and MarketScan for individuals with bipolar disorder, major depression, or adjustment disorders to compare spending for mental health and SUD treatment before and after the implementation of the 2001 FEHB parity directive. They found no difference in likelihood of service use after implementation of parity provisions for individuals enrolled in the FEHB program relative to a national sample of individuals in private insurance plans unaffected by parity.
MHPAEA Impact on Aetna-Insured Individuals. Busch and colleagues (2014) focused on the impact of the 2008 MHPAEA on Aetna-insured individuals. Aetna enrollees in ten states that had already implemented parity legislation for individuals in fully insured employer-sponsored plans were compared to individuals in self-insured plans that were newly required to implement parity under the MHPAEA. There was a modest increase in spending on SUD treatment per enrollee after implementation of parity, but no difference in treatment initiation or engagement.
Overall, the findings of these studies are mixed. Some studies found no change or a decrease is SUD treatment use. This result is not unexpected since insurers may implement care management approaches simultaneously with parity requirements. Other studies did find increased demand with estimates as high as a 15 percent increase when the analysis focused on only facilities accepting private insurance.
4. Legalization of Marijuana
Little is known about how the recent legalization of marijuana in Colorado and Washington will impact demand for SUD treatment. Lack of historical data from local marijuana markets impedes estimating the effects of this change. Legalization is likely to substantially increase consumption of marijuana; however it is unclear how legalization may affect demand for SUD treatment.
Increased Marijuana Consumption. Initially, legalization is expected to increase consumption substantially, primarily as a result of lower price (Caulkins and Lee 2012). Researchers used data from the National Epidemiologic Survey on Alcohol and Related Conditions, as well as data from NSDUH, and found that states with legal medical marijuana experienced significantly higher rates of marijuana use (Cerdá et al. 2012).
Change in Demand for Treatment. Increased consumption and potential changes in addiction rates may not directly translate into increased demand for treatment services. Historically, many referrals for treatment have come from the criminal justice system. A 2006 TEDS report found that 58 percent of admissions for marijuana treatment came from the criminal justice system (Marijuana Policy Project 2008). Thus, legalization may reduce referral to treatment. In contrast, Evans (2013) used 2010 NSDUH data to estimate how increased consumption and associated increases in addiction rates would affect treatment use. This study estimated that admissions would increase by 1.3-4.8 million. (These estimates assume between 13 million and 48 million individuals would begin using marijuana as a result of legalization and 10 percent would become dependent.)
Preliminary Findings from Recent Legalization Efforts. The availability of marijuana in Colorado has gradually changed over time. In 2009, changes in the acceptance of the state's medical marijuana law by federal and state officials led to a substantial expansion in medical marijuana use such that the number of medical marijuana cardholders went from 4,800 in 2008 to 41,000 in 2009. Next, in 2010, the Colorado Legislature passed HB-1284, which legalized medical marijuana dispensaries, marijuana cultivation operations, and manufacturers of marijuana edible products. By 2012, there were 532 licensed dispensaries in Colorado and over 108,000 registered patients. In November 2012, Colorado voters passed Amendment 64, which legalized marijuana for recreational use. This amendment allows individuals 21 years or older to grow up to six plants and possess/use/share one ounce or less. This amendment also permits marijuana retail stores, cultivation sites, edible factories and testing sites. The first recreational marijuana retail shop in Colorado opened January 1, 2014.
A study released in August 2014 (Rocky Mountain High Intensity Drug Trafficking Area 2014) looked at the impact of these policy changes in Colorado. Between 2006-2008 and 2009-2011, there was a 36 percent increase in adults' (26+ years) monthly marijuana use. However, this increase in use did not translate into an increase in demand for treatment. From 2005 through 2013, marijuana SUD treatment use from Colorado does not demonstrate a definitive trend.
Relatively few studies have examined the impact of marijuana legalization on treatment use, and the data used in the existing studies have significant limitations. Overall, the findings of these preliminary studies show a substantial increase in marijuana use associated with legalization, but the effect on treatment use is likely to be much more limited with even the possibility of a decrease in treatment use as criminal justice referrals decline.
5. Opioid Use Epidemic
Over the last decade there have been significant increases in disorders associated with pain relievers and heroin. Increases in these disorders may result in substantial increases in demand for treatment, especially MAT, in the next five years.
Reported heroin deaths have almost tripled since 2010 (HHS 2015a). As noted above, the number of individuals with heroin-related disorders ranged from 189,000 to 324,000 from 2002-2008, but in 2014 estimates indicate an increase to 586,000 individuals with heroin disorders (SAMHSA 2015e). From 2004-2014 the number of individuals receiving any treatment for a heroin-related disorder increased in parallel to the number of individuals with disorders (Figure II.15). Note that the number of individuals receiving treatment for heroin may be higher than the number of individuals meeting criteria for a disorder in the past year. This is due to receipt of treatment by individuals who no longer meet symptomatic criteria for a current disorder, but who continue to receive treatment such as methadone maintenance.
|FIGURE II.15. Number of Individuals with a Heroin Disorder and Number Receiving Any Heroin Treatment in Past Year, NSDUH 2004-2014|
|SOURCE: Number of individuals with disorders obtained from SAMHSA 2015a, Table 7.40A; Number of individuals in treatment in 2004-2014 obtained from SAMHSA 2005, SAMHSA 2006, SAMHSA 2007b, SAMHSA 2008, SAMHSA 2009, SAMHSA 2010, SAMHSA 2011a, SAMHSA 2012a, SAMHSA 2013b, SAMHSA 2014b, SAMHSA 2015a, SAMHSA 2015e, respectively.|
Since 1999, although the amount of pain reported has remained unchanged, prescribing of painkillers and deaths from prescription painkillers have nearly quadrupled (CDC 2015b; CDC 2015c). The number of individuals with disorders related to pain relievers ranged from 1.4 million to 1.5 million from 2002-2005. This increased to 1.9 million by 2014. Like heroin, the number of individuals receiving any treatment for pain reliever related disorders has increased in parallel to the number of individuals with disorders from 2004-2014 (Figure II.16).
Federal, state, and local officials and health care professionals are researching, developing and implementing initiatives targeted at reducing opioid use, abuse, overdose and related-deaths (HHS 2015e). These efforts include developing guidelines for opioid prescribing, expanding use of Prescription Drug Monitoring Programs, expanding access to and use of naloxone and increasing access to and use of MAT for opioid dependence (HHS 2015e). There is some evidence that these efforts are working, as there have been increases in the supply of physicians waivered to prescribe buprenorphine for opioid dependence, particularly in states that expanded Medicaid (Knudsen, Lofwall, Havens and Walsh 2015), as well as numbers of individuals receiving buprenorphine treatment (SAMHSA 2014a). These initiatives are intended to reverse the current trends in opioid use; however the impact of these initiatives may not be observable for several years. Thus, increases in the number of individuals with opioid-related disorders may continue in the short-run resulting in increased demand for treatment. Also, since some components of the initiative to address the opioid epidemic encourage treatment use, further increases in demand are likely to be observed when these programs are fully implemented.
|FIGURE II.16. Number of Individuals with a Pain Reliever Related Disorder and Number Receiving Any Pain Reliever Treatment in Past Year, NSDUH 2004-2014|
|SOURCE: Number of individuals with disorders obtained from SAMHSA 2015a, Table 7.40A; Number of individuals in treatment in 2004-2010 obtained from SAMHSA 2011a, Figure 7.9; Number of individuals in treatment 2011-2013 obtained from SAMHSA 2014b, Figure 7.9. Number of individuals in treatment in 2014 obtained from SAMHSA 2015e.|
Federal, state, and local government funding (not including Medicaid or Medicare spending) paid for 42 percent of all substance abuse treatment services in 2009. Medicaid and Medicare spending accounted for an additional 21 percent and 5 percent of treatment spending, respectively. Thus, shifts in government funding allocated to substance abuse treatment and decisions about SUD treatment coverage in Medicaid and Medicare can have a significant impact on total demand for SUD treatment. Thus, this section discusses key recent government funding allocations for SUD treatment, as well as government funding for prevention and workforce development that could influence the delivery of care. We first discuss federal funding, and then we address state and local funding.
a. Federal Funding
Annually, the Office of National Drug Control Policy (ONDCP) publishes estimates of federal spending intended to reduce the supply of or demand for illegal drugs, including illegal provision of alcohol. The most recent available estimates as of November 2014 are displayed in Table II.4.
|TABLE II.4. Federal Spending to Reduce Consumption of Illegal Substances|
|Spending in millions of dollars1|
|Total demand reduction||9,110.9||9,137.7||9,187.4||9,157.0||10,097.4||10,927.2|
|Total supply reduction||15,509.9||15,227.7||15,316.7||14,643.4||15,114.8||14,436.1|
|Percentage change from previous year|
|Total demand reduction||NA||0.3||0.5||-0.3||10.3||8.2|
|Total supply reduction||NA||-1.8||0.6||-4.4||3.2||-4.5|
|SOURCE: ONDCP National Drug Control Strategy Budget 2015 Highlights, March 2014.
NA = Not applicable.
After relatively little growth in spending or negative changes in spending levels for demand reduction between FY2010 and FY2013, there were substantial increases in spending to reduce demand enacted for FY2014; further increases were requested for FY2015. Below, we highlight some key spending categories.
Medicare and Medicaid Spending on SUD Treatment. The drug control budget treatment spending estimates include Medicaid and Medicare spending for SUD treatment, which increased 20 percent between FY2013 and FY2014 (from $3.62 billion to $4.30 billion) and for which an additional 16 percent increase is requested for FY2015 (to $5.07 billion). These increases are primarily associated with coverage expansions due to the ACA provisions discussed above (ONDCP 2014).
Substance Abuse Treatment Block Grant (SATBG). Funding was $1.71 billion in 2013 and rose 6.4 percent to $1.82 billion in 2014. Funding is expected to remain constant at $1.82 billion in 2015 (HHS 2015b).
Addiction Treatment within Community Health Centers. In November 2014, the HHS Health Resources and Services Administration announced that $51.3 million in ACA funding had been allocated to support establishing or expanding behavioral health services at 210 health centers in 47 states, the District of Columbia, and Puerto Rico. Earlier in 2014, HHS awarded $54.5 million in ACA funding for 223 other health centers to expand behavioral health services (HHS 2015c).
Increase Peer Professionals. SAMHSA planned to spend $10 million in FY2015 to fund a new workforce program to increase the number of trained peers, recovery coaches, mental health/addiction specialists, prevention specialists, and pre-master's-level addiction counselors (HHS 2015b). Funding was not allocated for this program in FY2015. Funding has again been requested for FY2016 (HHS 2015d).
Screening, Brief Intervention, and Referral to Treatment (SBIRT). Public Health Service evaluation funds were planned to give grants to providers to integrate SBIRT into medical treatment settings in 2015. Thirty million dollars in spending was planned for FY2015 (ONDCP 2014). Funding was not allocated for this program in FY2015. Funding has again been requested for FY2016 (HHS 2015d).
Department of Education. The U.S. Department of Education planned to increase funding for prevention activities from $39.5 million enacted in FY2014 to $102.1 million requested for FY2015. These funds would be used to improve school climates by reducing drug use and violence and improving school safety (ONDCP 2014).
b. State and Local Funding
Data on state and local SUD treatment expenditures are not systematically collected. The SAMHSA spending estimates initiative estimates state and local spending based on provider source of payment data from survey and claims datasets. State and local SUD treatment spending increased at a substantial annual rate from 1986 through 2004 (Table II.5), and the share of all SUD treatment spending represented by state and local funding increased from 27 percent to 34 percent. However, between 2004 and 2009, the rate of increase in state and local SUD treatment spending slowed and the share of SUD treatment spending represented by state and local sources declined to 31 percent (SAMHSA 2013a).
|TABLE II.5. State and Local Spending on Substance Abuse Treatment, 1986-2009|
|Total expenditures (in millions)||2,487||3,647||5,019||7,145||7,639|
|Annualized percent change since prior estimated year||NA||6.6||5.5||6.1||1.3|
|Proportion of all SUD treatment spending||27||27||34||34||31|
|SOURCE: Mathematica analysis of SAMHSA 2013c, Table A8 and page 38.
NOTE: Federal block grant funding administered by the states is not included in these estimates as this funding is considered federal spending.
NA = Not applicable.
State Medicaid agencies may play a substantial role in encouraging increased use of SUD treatment services. States that expanded their Medicaid programs under the ACA have accepted large numbers of childless adults into their Medicaid populations and these individuals are expected to have greater need for SUD treatment services relative to the traditional Medicaid population. States are required to provide SUD treatment benefits to Medicaid expansion populations in parity with physical health benefits. We found only one report that addressed this issue. Boozang et al. (2014) reviewed current and planned approaches to substance abuse benefit delivery in six states using managed care delivery systems: Arizona, Maryland, Massachusetts, New Mexico, New York, and Washington. This review found that states are substantially modifying, or replacing current approaches to SUD treatment benefit management. Some states are moving SUD treatment into integrated managed care models with physical health benefits, mental health benefits, or both. States see these arrangements as supporting better care integration. States are investing in substance abuse provider capacity particularly programs that provide less costly alternatives to inpatient care. They are also reaching out to providers with little experience with Medicaid. States are beginning to develop strategies to integrate Medicaid and Medicaid managed care organization services for individuals with SUDs with social services. They are also beginning to consider how to provide care to individuals coming out of jail or prison, who are more likely to have SUD treatment needs than the general population and may be eligible for Medicaid for the first time.
While the current available literature on state Medicaid agency response to ACA provisions is limited, Andrews et al. (2015b) noted that state Medicaid programs played an important role in transforming the mental health treatment system over the last 50 years as Medicaid took a larger role in financing mental health treatment. This study suggests that the transformations Medicaid programs encouraged in the mental health system might offer lessons for how Medicaid's increased role in addiction treatment could offer an opportunity improve the treatment system. Specifically, flexibility in Medicaid service coverage could be used to tailor services to the specific needs of the population in need of addiction treatment. Also, alternatives to inpatient care might be developed and licensure standards could be improved. The HHS Centers for Medicare and Medicaid Services (CMS) is encouraging states to make reforms and has sent a letter to State Medicaid Directors in July 2015 to "inform states of opportunities to design service delivery systems for individuals with substance use disorder (SUD), including a new opportunity for demonstration projects approved under Section 1115 of the Social Security Act (Act) to ensure that a continuum of care is available to individuals with SUD." CMS also chose SUD treatment as the first area of focus for its Innovation Accelerator Program which provides technical assistance and expert resources to states engaged in system redesign. Overall there are efforts to encourage reform, but the degree to which these efforts will transform the treatment system and encourage treatment entry is unclear. Also limited provider acceptance of Medicaid insurance may be an impediment to expanding Medicaid coverage for SUD treatment. Provider acceptance of Medicaid is addressed in Section III.B.2.
Although trends in SUD prevalence and treatment receipt have been relatively constant for the last decade, a substantial shift in these trends is likely to be observed between 2014 and 2018. While federal, state, and local funding for SUD treatment (other than through Medicaid and Medicare) are likely to remain relatively stable, the ACA could result in a substantial increase in demand for SUD treatment. Additional increases in demand may result from MHPAEA and efforts to reduce opioid use and its adverse effects. Thought leaders in the SUD treatment field identified the recent increased recognition of SUDs as a valid health condition as a key trend affecting the field. Changes in the current perceptions of SUD treatment among individuals with SUDs, integration of SUD treatment into mental health and primary care settings, and changing views on SUD screening and referral to treatment among primary care providers have the potential to influence trends in demand for services. However, the effects of these changes are likely to occur gradually because perceptions are likely to change slowly and developing integrated care models requires substantial upfront training and investments.