Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jun 1.
Published in final edited form as: Schizophr Res. 2020 Apr 5;220:92–97. doi: 10.1016/j.schres.2020.03.053

A Single Assessment with the Brief Adherence Rating Scale (BARS) Discriminates Responders to Long-Acting Injectable Antipsychotic Treatment in Patients with Schizophrenia

Paul A Nakonezny a,b, Janet C Lindow c, T Scott Stroup d, Joseph P McEvoy e, Marvin S Swartz f, Robert A Rosenheck g, Matthew J Byerly h
PMCID: PMC7306424  NIHMSID: NIHMS1582166  PMID: 32269005

Abstract

Objective:

To determine if a single baseline adherence assessment (Brief Adherence Rating Scale [BARS]) could identify patients who are likely to respond to long-acting injectable (LAI) antipsychotic treatment.

Method:

The current secondary analysis included a sub-sample of adult outpatients (N=176) with schizophrenia or schizoaffective disorder who participated in the “A Comparison of Long-Acting Injectable Medications for Schizophrenia (ACLAIMS)” trial and had a baseline BARS assessment and a baseline and month 3 Positive and Negative Syndrome Scale (PANSS) rating. The main outcome was LAI treatment response, defined as a ≥20% decrease (baseline to month 3) on the PANSS total score. Receiver Operating Characteristic (ROC) and Area Under the Curve (AUC) analysis was conducted to determine the optimal cutpoint of baseline BARS adherence in discriminating LAI treatment response at month 3. A logistic mixed model estimated the odds of response to LAI treatment at month 3 from the optimal baseline BARS outpoint.

Results:

The ROC analysis determined that the single baseline BARS rating (cutoff ≤66%), indicating low adherence, best discriminated patients likely to respond to LAI treatment (AUC=0.603, SE=0.046, 95% binomial exact CI=0.527 to 0.676, p=0.025), with 38% sensitivity and 85% specificity. The logistic mixed model analysis revealed that patients with ≤66% BARS adherence had 3.464 times the predicted odds (95% CI=1.604 to 7.480, p=0.001) of responding to LAI treatment than those who were >66% BARS adherent.

Conclusion:

A single baseline BARS assessment discriminated response to LAI treatment suggesting it is a reasonable tool to identify candidates for LAI antipsychotic treatment.

Keywords: BARS, antipsychotic, adherence, schizophrenia, depot, injectable

1. Introduction

Non-adherence to antipsychotic medication occurs in at least 50% of persons with schizophrenia and is associated with serious consequences including exacerbation of psychotic symptoms, increased aggression toward self and others, worse prognosis, increased hospital and emergency room use, criminal justice involvement and high health care costs (Ascher-Svanum et al., 2006; Byerly et al., 2007a). Interventions for managing antipsychotic non-adherence may considerably reduce psychiatric morbidity and costs of care. This could have positive implications on the welfare of persons with schizophrenia, mitigate safety risks to themselves and others, and reduce the use of resources for acute psychotic episodes.

Long-acting injectable (LAI) antipsychotics are recommended for patients who are nonadherent with oral antipsychotic regimens in all major schizophrenia treatment guidelines (Barnes and Schizophrenia Consensus Group of British Association for Psychopharmacology, 2011; Galletly et al., 2016; Kane et al., 2003; Kishimoto et al., 2018; Kishimoto et al., 2013; Lehman et al., 2004; Miller et al., 2004; Miura et al., 2019; Velligan et al., 2010). However, these recommendations were made in the absence of specific empirical information about which patients should be targeted for LAI treatment and without clear evidence that LAIs produce superior outcomes in non-adherent patients. While some previous research has shown that LAIs and oral antipsychotics are similarly efficacious (Haddad et al., 2015; Kishimoto et al., 2014; Rosenheck et al., 2011), several randomized controlled trials (RCTs) have found that LAIs were superior to oral antipsychotics for relapse prevention (Alphs et al., 2016; Schreiner et al., 2015; Subotnik et al., 2015), especially for patients early in the course of their illness (Schreiner et al., 2015; Subotnik et al., 2015). None of the controlled trials used in previous meta-analyses or the individual RCTs, however, included patients based on the results of formal assessments of oral antipsychotic medication adherence. Indeed, the Cochrane Schizophrenia Group emphasized that such studies have not focused on populations in which LAIs might be most beneficial; specifically, those patients who are non-adherent to oral antipsychotic medication (Adams et al., 2001).

The Brief Adherence Rating Scale (BARS), which was developed and validated by Byerly et al. (Byerly et al., 2008), may have the potential to address prescribers’ challenge in identifying non-adherent patients. The BARS is a brief, clinician-administered adherence instrument. It consists of four items: three questions and a summary visual analog scale (VAS) rating (0% to 100%). The VAS rating serves as the final adherence determination. The BARS requires less than 60 minutes of staff training and each assessment takes less than 5 minutes to complete. Relative to electronic monitoring, the BARS provides valid, reliable, sensitive, and specific estimates of antipsychotic medication adherence of outpatients with schizophrenia and schizoaffective disorder (Byerly et al., 2008). These characteristics make it a promising tool to detect non-adherence in routine care in community-based settings.

The aim of this study was to determine if a single baseline adherence assessment with the BARS could identify patients who are likely to respond to long-acting injectable antipsychotic treatment.

2. Methods

2.1. Study Setting and Participants

2.1.1. ACLAIMS Trial Participants

A Comparison of Long-Acting Injectable Medications for Schizophrenia (ACLAIMS) was a multisite, parallel-group, double-blinded RCT conducted at 22 clinical sites in the United States and was affiliated with the National Institute of Mental Health-supported Schizophrenia Trials Network. The ACLAIMS trial compared the effectiveness, cost, and tolerability of paliperidone palmitate (PP) to haloperidol decanoate (HD) in patients with schizophrenia or schizoaffective disorder. A more detailed description of the full methodology and outcomes of the ACLAIMS trial have been previously reported (McEvoy et al., 2014). In brief, the ACLAIMS trial randomized 311 male and female outpatients (aged 18-65 years) with schizophrenia or schizoaffective disorder who were considered by their clinician and study psychiatrist to be at risk of relapse and of likely to benefit from a long-acting injectable antipsychotic. After baseline assessments were completed, patients were randomly assigned to up to 24 months of treatment with either paliperidone palmitate or haloperidol decanoate. Following randomization, patient outcomes were evaluated monthly during study visits as outlined in the schedule of assessments for the ACLAIMS trial (McEvoy et al., 2014). Patients, treating clinicians, and raters were blinded to treatment assignment. Each site obtained institutional review board approval, and written informed consent was obtained from all participants.

2.1.2. Current Study Participants

The participants of the current secondary analysis are the subset of participants from the ACLAIMS trial who had a BARS assessment at baseline as well as a PANSS rating at both baseline and month 3. Of the original 311 randomized participants from the ACLAIMS trial, 96 patients did not have a baseline BARS rating, 2 patients did not have a baseline PANSS rating, and 66 patients did not have a PANSS rating at month 3. Altogether, 135 patients were not assessed on all three of these ratings, which left 176 patients for this analysis. To mitigate concern of selection bias, we note that the sample of 176 patients used in the current study was similar in their demographics and clinical characteristics not only to the overall 311 randomized participants from the ACLAIMS trial (McEvoy et al., 2014) but also to the aforementioned 135 patients who were not part of this secondary analysis (Table S1).

2.2. Interventions

2.2.1. LAI Antipsychotic Medication

A loading dose strategy was used for initiation of both LAI medications. The recommended starting dose of paliperidone palmitate was 234 mg intramuscularly (IM) on day 1 followed on day 8 with 156 mg IM. The recommended standard monthly dose of paliperidone palmitate was 117 mg IM. The recommended starting dose of haloperidol decanoate was 50 mg intramuscularly on day 1 followed on day 8 with 50 mg IM. On day 28, the recommended dose of haloperidol decanoate was 75 mg IM, to be followed on day 56, and on subsequent visits every 4 weeks with 50 mg intramuscularly. However, both medications were flexibly dosed according to clinician judgment within the following ranges: PP 39-234 mg IM or HD 25-200 mg IM every 4 weeks. See McEvoy et al. (2014) for a more detailed description of the ACLAIMS trial, including pharmacologic interventions/treatments. The average starting and month 3 doses of PP and HD for the 176 evaluable participants of the current analysis are shown in Table 1.

Table 1.

Demographic and Clinical Characteristics of the Baseline BARS Adherence Cutpoint Groups and Responders versus Non-Responder to LAI Antipsychotic Treatment

Baseline BARS Adherence Cutpointb Response to LAI Treatment at Month 3c
Participant Characteristica Evaluable Sample (N=176) ≤ 66% (n=40) > 66% (n=136) p-value (FDR)d Responders (n=58) Non-Responders (n=118) p-value (FDR)d
Demograhics
Age in years, M ±SD 43.3±12.7 41.1±13.1 43.9±12.6 .21 (.40) 42.2±12.7 43.8±12.6 .44 (.75)
Sex, % (n) .17 (.40) .65 (.75)
 Male 77.3% (136) 85.0% (34) 75.0% (102) 79.3% (46) 76.3% (90)
 Female 22.7% (40) 15.0% (06) 25.0% (34) 20.7% (12) 23.7% (28)
Race, % (n) .22 (.40) .55 (.75)
 African American 52.8% (93) 45.0% (18) 55.2% (75) 56.9% (33) 50.8% (60)
 Caucasian 41.5% (73) 52.5% (21) 38.2% (52) 39.7% (23) 42.4% (50)
 Other 5.7% (10) 2.5% (01) 6.6% (09) 3.4% (02) 6.8% (08)
Clinical Characteristics
Schizophrenia ever in lifetime, % (n) 72.2% (127) 65.0% (26) 74.3% (101) .26 (.40) 65.5% (38) 75.4% (89) .17 (.48)
Schizoaffective Disorder ever in lifetime, % (n) 36.9% (65) 42.5% (17) 35.3% (48) .41 (.45) 43.1% (25) 33.9% (40) .23 (.57)
MDD in past 5 years, % (n) 26.1% (46) 15.0% (06) 29.4% (40) .06 (.20) 24.1% (14) 27.1% (32) .67 (.75)
Alcohol abuse in past 5 years, % (n) 30.7% (54) 37.5% (15) 28.7% (39) .29 (.40) 32.8% (19) 29.7% (35) .68 (.75)
Alcohol abuse in past month, % (n) 10.8% (19) 15.0% (06) 9.6% (13) .34 (.42) 10.3% (06) 11.0% (13) .89 (.89)
Drug abuse in past 5 years, % (n) 32.4% (57) 40.0% (16) 30.2% (41) .25 (.40) 34.5% (20) 31.4% (37) .68 (.75)
Drug abuse in past month, % (n) 11.9% (21) 12.5% (05) 11.7% (16) .89 (.89) 10.3% (06) 12.7% (15) .65 (.75)
Number of hospitalizations in past yeare, M±SD 1.1±1.3 1.3±1.6 1.1±1.2 .30 (.40) 1.5±1.6 0.87±1.1 .01 (.05)
Age at illness onsetf, years, M±SD 23.8±10.5 22.9±9.1 24.1± 10.8 .56 (.59) 22.7±8.4 24.4±11.3 .26 (.57)
Age at 1st antipsychotic medication, years, M±SD 26.8±9.7 25.6±7.1 27.2±10.4 .27 (.39) 25.8±7.9 27.3±10.4 .28 (.58)
LAI Treatment, % (n) .38 (.45) .59 (.75)
 Haloperidol Decanoate (HD) 51.1% (90) 45.0% (18) 52.9% (72) 48.3% (28) 52.5% (62)
 Paliperidone Palmitate (PP) 48.9% (86) 55.0% (22) 47.1% (64) 51.7% (30) 47.5% (56)
HD starting dose, mg, M±SD 49.5±12.8 52.1±19.3 48.7±10.3 .30 (.40) 52.3±19.5 48.1±7.5 .12 (.48)
HD dose at Month 3, mg, M±SD 67.8±33.6 56.3±25.9 71.0±34.9 .03 (.20) 71.6±33.5 66.1±33.7 .35 (.70)
PP starting dose, mg, M±SD 203.1±56.4 188.3±63.8 207.4±53.6 .09 (.26) 194.5±62.6 207.4±52.8 .15 (.48)
PP dose at Month 3, mg, M±SD 138.1±40.5 125.5±35.3 141.5±41.3 .05 (.20) 138.9±45.4 137.7±38.2 .86 (.89)
BARS adherence (%) at baseline, M±SD 79.5±29.6 31.1±23.8 93.7±8.7 .0001 (.002) 70.7±34.5 83.7±25.9 .01 (.05)
PANSS total at baseline, M±SD 71.6±15.7 76.1±16.5 70.3±15.3 .04 (.20) 77.1±13.9 68.9±15.9 .001 (.01)
PANSS total at month 3, M±SD 63.3±15.8 59.3±16.6 64.5±15.5 .06 (.20) 53.2±11.9 68.3±15.2 .0001 (.002)
Open in a new tab
a

The means (M) presented in this table are the sample means; SD = Standard Deviation; Race was self-reported and “Other’ race includes American Indian or Alaska Native, Asian, or Native Hawaiian or other Pacific Islander and two or more races.

b

Based on the Youden Index.

c

Response to LAI antipsychotic treatment at month 3 was operationalized as at least a 20% improvement (reduction) in PANSS total score from baseline to month 3.

d

Tested for differences between those ≤66% BARS adherent at baseline and those >66% BARS adherent at baseline and between responders and non-responders on each demographic/clinical characteristic in a separate model, and the two-tailed p-values were adjusted for multiple testing using the False Discovery Rate (FDR) described by Benjamini and Hochberg (1995).

e

Number of times hospitalized for psychiatric reasons in past year.

f

Illness was for any behavioral or emotional problem.

2.3. Measures

The study outcomes as well as schedule of assessments for the ACLAIMS trial have been previously reported in (McEvoy et al., 2014). In brief, symptom severity or psychopathology was assessed using the PANSS. The PANSS contains 30 items that assess symptoms of psychotic disorders including positive, negative, and general psychopathology (Kay et al., 1987; Kay et al., 1989). The Kay group has repeatedly demonstrated that the PANSS has strong psychometric properties when used in schizophrenia clinical research (Kay, 1990): the PANSS has established external validity (Kay et al., 1987; Kay et al., 1986; Kay and Singh, 1989), adequate construct validity (Kay et al., 1987; Kay et al., 1988), high internal reliability (Kay et al., 1987), good interrater reliability (Kay et al., 1988), and good test-retest reliability (Kay et al., 1987).

The PANSS ratings were conducted by qualified and trained clinicians (raters) who were blinded to treatment assignment. The PANSS was completed at baseline, month 1, and then every three months as outlined in the schedule of assessments for the ACLAIMS trial (McEvoy et al., 2014). PANSS ratings used in the current secondary analysis, however, were obtained only at baseline and at month 3.

2.3.1. Response to LAI Antipsychotic Treatment

The outcome of the current secondary analysis was response to LAI antipsychotic treatment at month 3, which was operationalized as a binary outcome variable and defined as at least a 20% improvement (reduction) in PANSS total score from baseline to month 3. A common response definition of at least a 20% improvement (reduction) in PANSS total score was chosen a priori. An international consensus guideline focusing on terminology for antipsychotic treatment response and resistance in schizophrenia (Howes et al., 2017) utilized at least a 20% improvement in symptoms to define response, because this percent change is the minimum that should be considered as clinically efficacious (Leucht et al., 2006). Defining response at 3 months was chosen to ensure sufficient time for response to treatment and is consistent with the same terminology guideline where a minimum of 12 weeks duration of symptoms should be allowed to evaluate antipsychotic treatment response (Howes et al., 2017).

2.3.2. Brief Adherence Rating Scale

The Brief Adherence Rating Scale, developed and validated by Byerly et al. (Byerly et al., 2008), is a brief, clinician-administered adherence instrument. It consists of four items: three questions and a visual analog scale (VAS) rating (0% to 100%). The three questions inquired about patients’ knowledge of their own medication regimen and episodes of missed medication. The three questions included: number of prescribed doses per day (question 1); number of days over the past month the patient did not take the prescribed doses (question 2); and number of days over the past month the patient took less than the prescribed doses (question 3). The VAS rating serves as the final adherence determination. The BARS has established psychometric properties (e.g., Byerly et al., 2008). Relative to electronic monitoring the BARS provides valid, reliable, sensitive, and specific estimates of oral antipsychotic medication adherence of outpatients with schizophrenia and schizoaffective disorder (Byerly et al., 2008). In the ACLAIMS trial (McEvoy et al., 2014), the BARS was completed only at baseline. Thus, BARS ratings used in the current secondary analysis were obtained at baseline. The BARS ratings were conducted by qualified and trained clinicians (raters) who were blinded to treatment assignment. The observed baseline BARS adherence ratings in the current secondary analysis ranged from no adherence 0% (n=9/176) to perfect adherence 100% (n=64/176).

2.4. Statistical Analysis

Demographic and clinical characteristics for the evaluable sample (N=176) and for both response groups were described using the sample mean and standard deviation for continuous variables and the frequency and percentage for categorical variables. To identify any differences between the characteristics of the two groups, the two-independent sample t test with the Satterthwaite method for unequal variances (continuous variables) and the chi-square test or the Fisher exact test (categorical variables) were used.

A Receiver Operating Characteristic (ROC) analysis was conducted along with the Area Under the Curve (AUC) to evaluate how well the measure of baseline BARS adherence discriminated response to LAI antipsychotic treatment at month 3 (defined as at least a 20% reduction or improvement from baseline to month 3 on the PANSS Total score). The ROC analysis determined the optimal cutpoint for the measure of baseline BARS adherence (based on the Youden Index) in discriminating response to LAI antipsychotic treatment at month 3. The AUC for the measure of BARS adherence was tested against a nominal area of 0.50 using the Z statistic. The 95% binomial exact confidence interval was reported for the AUC. Sensitivity and specificity as well as positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), and negative likelihood ratio (LR-) were also reported for the optimal cutpoint.

A logistic mixed model, while accounting for site of participants as a random effect (so as to adjust for between-site variability) and with maximum likelihood estimation, was implemented to estimate the odds of response to LAI antipsychotic treatment at month 3 from the optimal baseline BARS cutpoint (identified in the ROC analysis). Moreover, as a sensitivity analysis, we replicated the above mentioned logistic mixed model with site included as a random effect as well as with adjustment for PANSS total score at baseline, sex, race, and number of times hospitalized for psychiatric reasons during the past year. We also replicated the logistic mixed model to estimate the odds of treatment response on the positive and negative psychopathology factor domains from the PANSS.

Statistical analyses were performed using SAS software, version 9.4 (SAS Institute, Inc., Cary, NC) as well as MedCalc for Windows, version 18.11.6 (MedCalc Software, Ostend, Belgium). The procedures of PROC GLIMMIX in SAS software were used to conduct the logistic mixed model analysis. The procedures in MedCalc were used to conduct the ROC analysis. The level of significance was set at a=0.05 (two-tailed) and to address multiple testing, where applicable, p-values were adjusted using the False Discovery Rate (FDR) procedure.

3. Results

3.1. Participant Characteristics

Participant characteristics are reported in Table 1. The evaluable sample in this secondary analysis consisted of 176 participants, which included 77.3% males and 52.8% African Americans. The age range was 18 to 65 years, with an average age of 43.3 years (SD=12.7). The mean age at illness onset was 23.8 years (SD=10.5), with 72.2% of the sample having a diagnosis of schizophrenia. The average baseline and month 3 PANSS total scores were 71.6 (SD=15.7) and 63.3 (SD=15.8), respectively indicating moderately severe symptomatology. The average baseline BARS adherence was 79.5% (SD=29.6), with a range of 0% to 100%. Thirty-three percent of participants (58/176) were classified as responders to LAI treatment (those who experienced ≥ 20% reduction in PANSS total scores from baseline to month 3), while the remaining 67% (118/176) were classified as non-responders.

3.2. BARS Adherence and Response to LAI Antipsychotic Treatment

The ROC analysis determined that a single BARS rating at baseline with a cutoff ≤ 66% BARS adherence (based on the Youden Index) best discriminated those who are likely to respond to LAI antipsychotic treatment at month 3 (AUC: 0.603, SE: 0.046, 95% binomial exact CI: 0.527 to 0.676, Z=2.203, p=0.027), with 37.93% sensitivity and 84.75% specificity along with a PPV of 55.0%, NPV of 73.5%, LR+ of 2.49, and LR- of 0.73. Fifty-five percent of patients with a baseline BARS adherence rating of ≤ 66% responded to LAI antipsychotic treatment at month 3, whereas only 26% of the patients with a baseline BARS adherence rating of > 66% responded to LAI treatment at month 3 (Fisher’s exact test, p=0.001). The ROC curve is shown in Figure 1. About 23%, or 40 of the 176 participants in this secondary analysis, had a BARS adherence rating at baseline of ≤ 66%. Those that fell at or below (≤ 66%) and above (> 66%) the baseline BARS adherence threshold were similar on all demographics and on most all of the clinical characteristics

Fig. 1.

Fig. 1.

Open in a new tab

ROC Curve of the baseline BARS adherence cutpoints for Response to LAI antipsychotic treatment at month 3. The point (o) on ROC curve with the Youden index corresponds to the BARS cutpoint of ≤ 66% and associated sensitivity of 37.93% and specificity of 84.75%. AUC = 0.603.

The logistic mixed model analysis, with site included as a random effect, revealed that patients who were ≤ 66% BARS adherent at baseline had 3.464 times the predicted odds of responding to LAI antipsychotic treatment at month 3 than those who were > 66% BARS adherent at baseline (odds ratio = 3.464, 95% CI = 1.604 to 7.480, Wald χ2=10.14, p=0.001). Moreover, in a sensitivity analysis, after adjustment for PANSS total score at baseline, sex, race, and number of times hospitalized for psychiatric reasons during past year, with site included as a random effect, the logistic mixed model revealed that the predicted odds of responding to LAI antipsychotic treatment at month 3 persisted (adjusted odds ratio = 3.153, 95% CI = 1.398 to 7.114, Wald χ2=7.76, p=0.005).

When we replicated the analysis separately on the positive and negative psychopathology factor domains (from the PANSS), the logistic mixed model revealed that patients who were ≤ 66% BARS versus those who were > 66% BARS adherent had greater predicted odds of responding to LAI antipsychotic treatment at month 3 on the positive psychopathology scale (odds ratio = 2.749, 95% CI = 1.275 to 5.928, Wald χ2=6.75, p=0.009), but not on the negative psychopathology scale (odds ratio = 1.546, 95% CI = 0.687 to 3.478, Wald χ2=1.12, p=0.289), than those who were > 66% BARS adherent.

4. Discussion

The current study examined if, in patients with schizophrenia, a single baseline adherence assessment with the BARS could discriminate likely responders of long-acting injectable antipsychotic treatment. Patients with schizophrenia who are ≤ 66% BARS adherent (the optimal outpoint based on the Youden index) are about 3 times more likely to respond to LAI antipsychotic treatment than those who are > 66% BARS adherent.

A single assessment with the BARS at ≤ 66% adherence discriminates patients who are likely to respond to LAI treatment, with a modest degree of accuracy (AUC=0.603, 95% Cl: 0.527 to 0.676) and with a modest degree of positive predictive value (PPV=55%). The positive predictive value indicates the likelihood (55%) that a schizophrenia patient with a BARS adherence rating at baseline of ≤ 66% would actually be a responder to LAI treatment. Thus, in our case, 55% of all LAI responders with a BARS adherence ≤ 66% are true positives (i.e., actual LAI responders). Our positive likelihood ratio (LR+) of 2.49 also indicates an increased likelihood (or probability) that the patient is indeed an LAI responder based on a single BARS rating at baseline with a cutoff ≤ 66%. The relatively low sensitivity (38%) of the single BARS assessment indicates that the test misses a considerable portion of responders who are negative on the test (i.e., responders who are “adherent” as defined by BARS of > 66%). This potential shortcoming is mitigated in part because clinicians generally would not provide LAI to adherent patients. A high specificity (85%) indicates the BARS is very unlikely to waste resources by providing LAI to patients who will not benefit (just 15% of non-responders would receive LAI based on BARS outpoint generated by a single BARS assessment). High specificity is what clinicians and community mental health centers would desire in the setting of prescribing LAIs.

Long-acting injectable antipsychotics are underutilized in the United States despite their recommendation as a non-adherence intervention by major treatment guidelines (Barnes and Schizophrenia Consensus Group of British Association for Psychopharmacology, 2011; Galletly et al., 2016; Kane et al., 2003; Lehman et al., 2004; Miller et al., 2004; Velligan et al., 2010). A plausible contribution to the under-utilization of LAIs is the inability of prescribers to identify non-adherence in their patients, which is the patient population most likely to benefit from LAI preparations. A study conducted by our group (Byerly et al., 2007b) found, that relative to the “gold standard” of electronically monitored adherence, patients and prescribers under-estimated antipsychotic non-adherence (57% vs. 5% and 7%, respectively) (Byerly et al., 2007b). Another study found that about 94% of schizophrenia patients with low-to-moderate (≤ 70%) adherence levels, identified by prescription fill data, had high (≥ 71%) prescriber-estimated adherence (Stephenson et al., 2012). These findings, in part, corroborate prior reports that prescribers are usually unable to detect antipsychotic medication non-adherence when it occurs in their patients, and suggests that such non-detection may be an important factor in the under-use of adherence interventions in the United States, particularly those that require prescribing, such as LAIs. Even when aware of non-adherence via electronic monitoring (EM), prescribers rarely recommend the use of an LAI to address identified non-adherence—doing so in just 6% of non-adherent cases (Nakonezny et al., 2013). Instead, prescribers generally opt for adherence interventions such as having a patient visit with a case manager about the non-adherence (Nakonezny et al., 2013), which has no greater guideline-recommended support than the use of LAIs as an intervention for non-adherence (Lehman et al., 2004; Velligan et al., 2010).

While the study showed promising results, there were a few possible limitations. A potential limitation to this study’s findings is that the study population may have been more likely to be nonadherent than the general clinical population because the eligibility criteria required participants to be at risk of efficacy failure based on a history of medication noncompliance, significant substance abuse, or both. Lastly, this study used the definition for response to treatment (20% reduction in symptoms) that is commonly used in clinical trial, but this may not be considered a satisfactory change by many patients.

In conclusion, there is a need to address prescribers’ challenge in identifying non-adherence. Until now, answers to this challenge have been hindered by a lack of a feasible, valid, and reliable adherence assessment method capable of identifying schizophrenia patients’ ongoing adherence to antipsychotic medication in the community setting as well as helping to identify responders to LAI treatment. In this study, patients with a pattern of oral antipsychotic noncompliance (identified by the BARS) were more likely to show an improvement in symptoms over the 3 months of treatment with assured LAI medication than were patients who had consistent compliance with oral antipsychotics. The BARS appears to offer a reasonable screening tool for those who are candidates for LAI treatment.

Supplementary Material

1

Acknowledgements

Role of the funding source

Funding for this study was provided by NIMH (R01-MH081234). The funding source did not play a role in the study design, implementation, interpretation of results, preparation of manuscript drafts, or the publication of this paper.

Conflicts of Interest

PAN and MJB have received research grant funding from Otsuka Pharmaceutical. TSS was an investigator in a clinical trial sponsored by Auspex Pharmaceuticals (now Teva Pharmaceutical Industries) and participated in a continuing medical education presentation supported by an independent educational grant from Intra-Cellular Therapies Inc. All other authors declare that they have no conflicts of interest.

Abbreviations:

LAI

long-acting injectable antipsychotic treatment

BARS

Brief Adherence Rating Scale

PANSS

Positive and Negative Syndrome Scale

AUC

Area Under the Curve

ROC

Receiver Operating Characteristic

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Adams CE, Fenton MK, Quraishi S, David AS, 2001. Systematic meta-review of depot antipsychotic drugs for people with schizophrenia. Br J Psychiatry 179, 290–299. [DOI] [PubMed] [Google Scholar]
  2. Alphs L, Mao L, Lynn Starr H, Benson C, 2016. A pragmatic analysis comparing once-monthly paliperidone palmitate versus daily oral antipsychotic treatment in patients with schizophrenia. Schizophr Res 170(2–3), 259–264. [DOI] [PubMed] [Google Scholar]
  3. Ascher-Svanum H, Faries DE, Zhu B, Ernst FR, Swartz MS, Swanson JW, 2006. Medication adherence and long-term functional outcomes in the treatment of schizophrenia in usual care. J Clin Psychiatry 67(3), 453–460. [DOI] [PubMed] [Google Scholar]
  4. Barnes TR, Schizophrenia Consensus Group of British Association for Psychopharmacology, 2011. Evidence-based guidelines for the pharmacological treatment of schizophrenia: recommendations from the British Association for Psychopharmacology. J Psychopharmacol 25(5), 567–620. [DOI] [PubMed] [Google Scholar]
  5. Byerly MJ, Nakonezny PA, Lescouflair E, 2007a. Antipsychotic medication adherence in schizophrenia. Psychiatr Clin North Am 30(3), 437–452. [DOI] [PubMed] [Google Scholar]
  6. Byerly MJ, Nakonezny PA, Rush AJ, 2008. The Brief Adherence Rating Scale (BARS) validated against electronic monitoring in assessing the antipsychotic medication adherence of outpatients with schizophrenia and schizoaffective disorder. Schizophr Res 100(1–3), 60–69. [DOI] [PubMed] [Google Scholar]
  7. Byerly MJ, Thompson A, Carmody T, Bugno R, Erwin T, Kashner M, Rush AJ, 2007b. Validity of electronically monitored medication adherence and conventional adherence measures in schizophrenia. Psychiatr Serv 58(6), 844–847. [DOI] [PubMed] [Google Scholar]
  8. Galletly C, Castle D, Dark F, Humberstone V, Jablensky A, Killackey E, Kulkarni J, McGorry P, Nielssen O, Tran N, 2016. Royal Australian and New Zealand College of Psychiatrists clinical practice guidelines for the management of schizophrenia and related disorders. Aust N Z J Psychiatry 50(5), 410–472. [DOI] [PubMed] [Google Scholar]
  9. Haddad PM, Kishimoto T, Correll CU, Kane JM, 2015. Ambiguous findings concerning potential advantages of depot antipsychotics: in search of clinical relevance. Curr Opin Psychiatry 28(3), 216–221. [DOI] [PubMed] [Google Scholar]
  10. Howes OD, McCutcheon R, Agid O, de Bartolomeis A, van Beveren NJ, Birnbaum ML, Bloomfield MA, Bressan RA, Buchanan RW, Carpenter WT, Castle DJ, Citrome L, Daskalakis ZJ, Davidson M, Drake RJ, Dursun S, Ebdrup BH, Elkis H, Falkai P, Fleischacker WW, Gadelha A, Gaughran F, Glenthoj BY, Graff-Guerrero A, Hallak JE, Honer WG, Kennedy J, Kinon BJ, Lawrie SM, Lee J, Leweke FM, MacCabe JH, McNabb CB, Meltzer H, Moller HJ, Nakajima S, Pantelis C, Reis Marques T, Remington G, Rossell SL, Russell BR, Siu CO, Suzuki T, Sommer IE, Taylor D, Thomas N, Ucok A, Umbricht D, Walters JT, Kane J, Correll CU, 2017. Treatment-Resistant Schizophrenia: Treatment Response and Resistance in Psychosis (TRRIP) Working Group Consensus Guidelines on Diagnosis and Terminology. Am J Psychiatry 174(3), 216–229. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Kane JM, Leucht S, Carpenter D, Docherty JP, Expert Consensus Panel for Optimizing Pharmacologic Treatment of Psychotic, D., 2003. The expert consensus guideline series. Optimizing pharmacologic treatment of psychotic disorders. Introduction: methods, commentary, and summary. J Clin Psychiatry 64 Suppl 12, 5–19. [PubMed] [Google Scholar]
  12. Kay SR, 1990. Positive-negative symptom assessment in schizophrenia: psychometric issues and scale comparison. Psychiatr Q 61(3), 163–178. [DOI] [PubMed] [Google Scholar]
  13. Kay SR, Fiszbein A, Opler LA, 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull 13(2), 261–276. [DOI] [PubMed] [Google Scholar]
  14. Kay SR, Opler LA, Fiszbein A, 1986. Significance of positive and negative syndromes in chronic schizophrenia. Br J Psychiatry 149, 439–448. [DOI] [PubMed] [Google Scholar]
  15. Kay SR, Opler LA, Lindenmayer JP, 1988. Reliability and validity of the positive and negative syndrome scale for schizophrenics. Psychiatry Res 23(1), 99–110. [DOI] [PubMed] [Google Scholar]
  16. Kay SR, Opler LA, Lindenmayer JP, 1989. The Positive and Negative Syndrome Scale (PANSS): rationale and standardisation. Br J Psychiatry Suppl(7), 59–67. [PubMed] [Google Scholar]
  17. Kay SR, Singh MM, 1989. The positive-negative distinction in drug-free schizophrenic patients. Stability, response to neuroleptics, and prognostic significance. Arch Gen Psychiatry 46(8), 711–718. [DOI] [PubMed] [Google Scholar]
  18. Kishimoto T, Hagi K, Nitta M, Leucht S, Olfson M, Kane JM, Correll CU, 2018. Effectiveness of Long-Acting Injectable vs Oral Antipsychotics in Patients With Schizophrenia: A Meta-analysis of Prospective and Retrospective Cohort Studies. Schizophr Bull 44(3), 603–619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kishimoto T, Nitta M, Borenstein M, Kane JM, Correll CU, 2013. Long-acting injectable versus oral antipsychotics in schizophrenia: a systematic review and meta-analysis of mirror-image studies. J Clin Psychiatry 74(10), 957–965. [DOI] [PubMed] [Google Scholar]
  20. Kishimoto T, Robenzadeh A, Leucht C, Leucht S, Watanabe K, Mimura M, Borenstein M, Kane JM, Correll CU, 2014. Long-acting injectable vs oral antipsychotics for relapse prevention in schizophrenia: a meta-analysis of randomized trials. Schizophr Bull 40(1), 192–213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lehman AF, Lieberman JA, Dixon LB, McGlashan TH, Miller AL, Perkins DO, Kreyenbuhl J, American Psychiatric A, Steering Committee on Practice, G., 2004. Practice guideline for the treatment of patients with schizophrenia, second edition. Am J Psychiatry 161(2 Suppl), 1–56. [PubMed] [Google Scholar]
  22. Leucht S, Kane JM, Etschel E, Kissling W, Hamann J, Engel RR, 2006. Linking the PANSS, BPRS, and CGI: clinical implications. Neuropsychopharmacology 31(10), 2318–2325. [DOI] [PubMed] [Google Scholar]
  23. McEvoy JP, Byerly M, Hamer RM, Dominik R, Swartz MS, Rosenheck RA, Ray N, Lamberti JS, Buckley PF, Wilkins TM, Stroup TS, 2014. Effectiveness of paliperidone palmitate vs haloperidol decanoate for maintenance treatment of schizophrenia: a randomized clinical trial. JAMA 311(19), 1978–1987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Miller A, Hall CS, Buchanan RW, Buckley PF, Chiles JA, Conley RR, Crismon ML, Ereshefsky L, Essock SM, Finnerty M, Marder SR, Miller DD, McEvoy JP, Rush AJ, Saeed SA, Schooler NR, Shon SP, Stroup S, Tarin-Godoy B, 2004. The Texas Medication Algorithm Project antipsychotic algorithm for schizophrenia: 2003 update. J Clin Psychiatry 65(4), 500–508. [DOI] [PubMed] [Google Scholar]
  25. Miura G, Misawa F, Kawade Y, Fujii Y, Mimura M, Kishimoto T, 2019. Long-Acting Injectables Versus Oral Antipsychotics: A Retrospective Bidirectional Mirror-Image Study. J Clin Psychopharmacol 39(5), 441–445. [DOI] [PubMed] [Google Scholar]
  26. Nakonezny PA, Byerly MJ, Pradhan A, 2013. The effect of providing patient-specific electronically monitored antipsychotic medication adherence results on the treatment planning of prescribers of outpatients with schizophrenia. Psychiatry Res 208(1), 9–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Rosenheck RA, Krystal JH, Lew R, Barnett PG, Fiore L, Valley D, Thwin SS, Vertrees JE, Liang MH, Group CSPR, 2011. Long-acting risperidone and oral antipsychotics in unstable schizophrenia. N Engl J Med 364(9), 842–851. [DOI] [PubMed] [Google Scholar]
  28. Schreiner A, Aadamsoo K, Altamura AC, Franco M, Gorwood P, Neznanov NG, Schronen J, Ucok A, Zink M, Janik A, Cherubin P, Lahaye M, Hargarter L, 2015. Paliperidone palmitate versus oral antipsychotics in recently diagnosed schizophrenia. Schizophr Res 169(1–3), 393–399. [DOI] [PubMed] [Google Scholar]
  29. Stephenson JJ, Tunceli O, Gu T, Eisenberg D, Panish J, Crivera C, Dirani R, 2012. Adherence to oral second-generation antipsychotic medications in patients with schizophrenia and bipolar disorder: physicians’ perceptions of adherence vs. pharmacy claims. Int J Clin Pract 66(6), 565–573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Subotnik KL, Casaus LR, Ventura J, Luo JS, Hellemann GS, Gretchen-Doorly D, Marder S, Nuechterlein KH, 2015. Long-Acting Injectable Risperidone for Relapse Prevention and Control of Breakthrough Symptoms After a Recent First Episode of Schizophrenia. A Randomized Clinical Trial. JAMA Psychiatry 72(8), 822–829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Velligan DI, Weiden PJ, Sajatovic M, Scott J, Carpenter D, Ross R, Docherty JP, 2010. Strategies for addressing adherence problems in patients with serious and persistent mental illness: recommendations from the expert consensus guidelines. J Psychiatr Pract 16(5), 306–324. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1582166-supplement-1.xls (29.5KB, xls)

RESOURCES