Remission from substance use disorders: A systematic review and meta-analysis

Highlights

• Twenty-one of 8855 studies published in the past 15 years were selected for this meta-analysis.

• Remission occurred after 17 years for up to 50% of persons with substance use disorders (SUDs).

• Aggregate SUD remission rates were 10–15% for any given year.

• Higher SUD remission rates correlated with lower sample retention and longer follow-up period.

• Results support the contention that SUDs are “chronic” or long term disorders.

Abstract

Objective

This paper presents a systematic review and meta-analysis of available evidence on remission rates for substance use disorders (SUDs), providing weighted mean estimates of SUD remission rates. The review also explores study-level characteristics that may explain variations in remission rates across studies.

Methods

A comprehensive search strategy identified studies published between 2000 and 2015 with follow-up periods of at least three years or reported lifetime remission outcomes for potential inclusion in the review. Remission was defined as not meeting diagnostic criteria for abuse or dependence for a minimum period of six months, as of final follow-up. A single-group summary meta-analysis was performed. Pooled estimated annual remission rates (PEARRs) were calculated. Meta-regression techniques and subgroup analyses were used to explore the association between study remission rates and key selected variables.

Results

Of 8855 studies identified, 21 met the eligibility criteria. The results suggested that 35.0% to 54.4% of individuals with SUDs achieved remission, and this occurred after a mean follow-up period of 17 years. The PEARRs projected few cases of SUD remission, between 6.8% and 9.1% in any given year. Studies that reported higher remission rates had longer follow-up periods, and lower sample retention rates.

Conclusions

Results support the contention that SUDs are more likely to be “chronic” or long term disorders than acute disorders for a substantial number of individuals. However, more longitudinal research is required. Treatment geared to chronicity, such as assertive community treatment and intensive case management, needs to be more readily available for SUD populations.

Introduction

Substance use disorders (SUDs), including drug and alcohol abuse and dependence, constitute a major public health issue, and cause substantial harm to individuals, families and society (Calabria et al., 2010). According to the World Health Organization, nearly 5% of the total burden of disease involves SUDs; alcohol accounts for 4% and illicit drug use 0.8% (WHO, 2004). Alcohol use is responsible for 3.6% of deaths worldwide, and illicit drug use 0.4% (Whiteford et al., 2015). SUDs are associated with health problems including cardiovascular diseases, cancers, infectious diseases, mental disorders, and with social problems such as domestic violence, criminal negligence, road accidents and suicide (Beck and Richard, 2012, Cullen et al., 2009, Von Laue et al., 2003). In the context of high SUD morbidity and mortality, efforts to improve treatments have proliferated in past decades. Numerous studies have measured outcomes, and confirmed the effectiveness of various treatments, and have demonstrated the value of self-help groups (Gerstein et al., 1997, UKATT Research Team, 2005). A recent review of 415 clinical and population studies published between 1868 and 2011 found that about half of participants had achieved remission from SUDs by final study follow-up periods (White, 2012). However, most studies measured short-term treatment outcomes (12 months), in line with the usual duration of SUD treatment (Arria and McLellan, 2012).

Whether SUDs should be understood and treated as acute or chronic diseases is the critical question that has faced the SUD community as far back as the early 19th century, but particularly in the climate of healthcare reforms since 2000. Unlike acute diseases that are usually severe and relatively short term conditions, chronic diseases are defined as diseases lasting a significant number of years, even the entire lifespan, or as diseases that include multiple episodes of relapse and remission differing in duration and clinical severity. There is no consensus around the number of years of consumption, minimum number of relapses or treatment episodes that an individual must experience in order to be labelled as a chronic SUD patient (Kelly and White, 2011, White, 2012). If SUDs are chronic diseases, then treatment needs to shift from acute to long-term care; and research should focus on longitudinal outcome studies. Many analysts have referred to SUD remission as a “natural process”, occurring after several years of addiction and often without treatment (Biernacki, 1986, Toneatto et al., 1999, Waldorf et al., 1991). This perspective supports the development of acute-care services and short-term outcome studies. Yet SUD duration is shorter and less severe among individuals who achieve remission without treatment; there is also less comorbidity, and fewer risk factors associated with these cases (Sobell et al., 1996, White, 2012). By contrast, other research has described SUDs as chronic relapsing disorders characterized by alternating cycles of remission and relapse of varying duration, but lasting several years (Brecher, 1972, Gossop et al., 2003, Hser et al., 1993, Vaillant, 1973, White et al., 2002). Under these conditions, care would need to be continuous, long-term and of variable intensity. Furthermore, similarities related to symptom recurrence and relapse rates have been identified between SUDs and chronic medical disorders such as type two diabetes, hypertension, and asthma (McLellan et al., 2000). Treatment adherence is also comparable for these populations: fewer than 30% of individuals with chronic medical conditions adhere to health-enhancing behavioral changes, for example, much like individuals with SUDs (McLellan et al., 2000).

Yet SUDs do not evolve inevitably into chronic disorders, nor does research predict with certainty under which conditions chronicity will occur (White and McLellan, 2008). The course of SUDs is highly individual, and is influenced by factors such as age of onset, intensity and duration of the disorder, cycles of remission and relapse, as well as medical and psychiatric status (El-Guebaly, 2012, White and McLellan, 2008). Family help and availability of community or health system resources also influence long-term remission (Bischof et al., 2004, Granfield and Cloud, 1999). A recent study based on syntheses of national epidemiological surveys including representative samples of substance users found that remission rates varied widely according to the primary substance used and whether substance were legal or illegal (Heyman, 2013). Numerous studies have also identified factors that positively affect the courses of SUDs, including gender (female) (Fillmore, 1987, Gomberg, 1995, Hyman et al., 2006, Liu and Kaplan, 1996, O'Brien et al., 2005, Walton et al., 2003, Weisner et al., 2003), age (older) (Chi and Weisner, 2008, Kalaydjian et al., 2009, Tuithof et al., 2013, Weisner et al., 2003), and greater participation in treatment (Best and Lubman, 2012, Dawson et al., 2012, El-Guebaly, 2012, Hser et al., 2015, Hser et al., 2003, Laudet et al., 2002, Nosyk et al., 2013, Xie et al., 2009). Research indicates that the probability of SUD remission also varies according to diagnosis, and is generally better for abuse than dependence (Hasin et al., 1997, Schuckit and Smith, 2011). Furthermore, total abstinence as opposed to controlled substance use is usually reported as more effective in achieving remission (Chi et al., 2011). Even though treatment outcome studies indicate that half or more of individuals achieved remission from SUDs in the short to medium term (Armor et al., 1978, Edwards et al., 1977, Sellman and Joyce, 1996), other research has determined that about a quarter of people with SUDs achieve continuous abstinence over the long term (Vaillant, 1988). How remission is defined, whether in terms of total abstinence or controlled substance use, how the concept is measured (e.g., no SUDs for 6 months, 12 months, or longer), and duration of follow-up (e.g., from 1 to 15 years) have a great influence on SUD outcomes (Dennis et al., 2007). Longer follow-up studies are particularly sensitive to sample retention issues, tending to inflate or under-report SUD remission rates (Simpson et al., 1997).

More accurate and updated information on remission rates and duration of consumption would allow decision-makers to develop services that are more responsive to the diverse needs of SUD populations (Arria and McLellan, 2012). While recent literature on the long-term course of SUDs, including remission rates, provides useful information, the available evidence on SUD trajectories is limited. For example, Hser et al. (2015) dealt specifically with opioid remission rates, whereas Calabria et al. (2010) studied remission rates in illicit drug use, but did not assess duration of consumption or remission in alcohol use disorder (AUD). A narrative review (Finney et al., 2013) reported SUD remission rates and duration of consumption by comparing treated and untreated individuals. The broad objective of our review was to provide a complete portrait of the long-term course of SUDs over the past 15 years. To this end, we undertook the first known systematic review and meta-analysis of studies reporting remission rates from SUDs (including both alcohol and illicit drugs) based on studies published from 2000 to 2015 that examined long-term SUD outcomes in terms of remission rates. The duration of consumption was also examined for studies that provided this information. Finally, potential sources of heterogeneity in remission rates that may explain variation across studies were explored.