Adherence to augmentation therapy for the treatment of major depressive disorder

ABSTRACT

Background

Inadequate response to antidepressant medication is common. Often, adjunctive pharmacotherapy or psychotherapy is recommended.

Objective

To measure adherence to adjunctive pharmacotherapy and psychotherapy among individuals with major depressive disorder (MDD).

Methods

Retrospective cohort study of individuals with MDD on antidepressant monotherapy who added adjunctive pharmacotherapy and/or psychotherapy. Medication adherence was measured by proportion of days covered (PDC) with optimal adherence defined as PDC≥0.80 and psychotherapy adherence defined by count of visits (optimal 8+ visits). Factors associated with optimal adherence were assessed by logistic regression.

Results

Among 218,192 individuals with adjunctive therapy, 185,349 added pharmacotherapy and 32,843 added psychotherapy. In the subsequent 12 months, 36.2% and 54.9% achieved optimal adherence to adjunctive pharmacotherapy and psychotherapy, respectively. Adherence to adjunctive pharmacotherapy was associated with adding psychotherapy, index antidepressant adherence, medical comorbidities, and MDD severity codes. Adherence to adjunctive psychotherapy was associated with adding another medication, previous psychiatry visit and psychiatric comorbidities.

Conclusion

Adjunctive psychotherapy appears under-utilized and adherence to adjunctive therapy was low. Low adherence to adjunctive therapy reinforces challenges in managing MDD. That a second adjunctive therapy enhanced adherence to the initial adjunctive therapy indicates an opportunity to explore alternative adjunctive therapies.

KEYWORDS:

• Major depressive disorder
• adherence
• adjunctive therapy
• psychotherapy
• medication

1. Introduction

Ranked as a leading cause of disability, major depressive disorder (MDD) is a common, chronic episodic condition associated with elevated use of healthcare services [1], reduced productivity, inability to work [2], increased morbidity and mortality [3,4], diminished health-related quality of life [5] and increased risk of suicide [6].

For most individuals with MDD, the initial treatment option is antidepressant pharmacotherapy. There are numerous antidepressant drug classes, which have comparable therapeutic effectiveness. (APA) Despite this array of available treatment options, more than half of MDD patients experience a lack of adequate response with the initial therapy selected [7,8]. In these patients, care often evolves to include augmentation – the addition of a second prescription drug class or of psychotherapy to the treatment regimen – both of which have been shown to be equally effective augmentation strategies [9].

Medication adherence is a significant contributor to treatment response. Low rates of adherence to antidepressant therapy are common [10] and are associated with lack of treatment response [11]. However, much less is known about adherence to pharmacotherapy and psychotherapy used to augment existing therapy. This population represents an important sub-group – one that appears motivated to engage in treatment, having undertaken one course of antidepressant therapy and initiated a second or even third treatment option.

Recently published articles on adherence to augmentation therapy among patients with MDD have focused exclusively on augmentation with atypical antipsychotics [12,13] however, atypical antipsychotics are just one of numerous therapeutic augmentation options. The objective of this research study was to improve our understanding of adherence to adjunctive pharmacotherapy or psychotherapy among patients with MDD and to identify the factors associated with higher levels of adherence.

2. Methods

This research was a retrospective observational cohort study of individuals with MDD who augmented antidepressant monotherapy with either a second antidepressant medication (‘Adjunctive Pharmacotherapy’), psychotherapy (‘Adjunctive Psychotherapy’), or both a second medication and psychotherapy (‘Dual Adjunctive Therapy’). The research protocol received an exempt determination from the Advarra Institutional Review Board.

2.1. Identification and selection of study participants

The sampling frame was the national sample of individuals identifiable in a commercially-available database of medical and pharmacy insurance claims licensed from Clarivate (www.clarivate.com). The study period was 1 January 2014 through 31 December 2019. The case-finding period was 1 July 2014 through 31 December 2018 and was used to identify the adjunctive therapy initiation date (index). The baseline and follow-up periods were the six months prior to and 12 months following index.

Eligible adults had two or more outpatient claims (or one inpatient claim) associated with MDD diagnosis codes (ICD-9-CM (296.2x or 296.3x) or ICD-10-CM (F32.x, F33.x) excluding MDD with psychotic features) on claims separated by at least 30 days; evidence of adjunctive therapy in the period between 1 July 2014 and 31 December 2018; claim(s) for antidepressant medication (alpha-2 receptor antagonist, MAOI, serotonin modulators, selective serotonin reuptake inhibitor (SSRI), selective norepinephrine reuptake inhibitor (SNRI), tricyclic, tetracyclic, or other antidepressant (primarily, bupropion)) before and after initiation of adjunctive therapy, defined as 50% adherence in the 56 days (eight weeks) immediately prior to and following the initiation of adjunctive therapy; evidence of continual insurance benefits eligibility defined by at least one medical or pharmacy claim in each calendar quarter in the baseline and follow-up periods; and had no evidence of psychotherapy in the three months prior to index. Study participants were excluded if they had a paid insurance claim by Medicare or two or more claims associated with bipolar or other depression disorders, dementia, psychosis, schizophrenia, or dissociative and conversion disorders.

The eligible study population was assigned to one of two cohorts: adjunctive pharmacotherapy or adjunctive psychotherapy. The Adjunctive Pharmacotherapy cohort had at least 50% adherence to adjunctive pharmacotherapy in the eight weeks following index and the Adjunctive Psychotherapy population had at least one additional psychotherapy claim in the 56 days following index. Adjunctive pharmacotherapy included the addition of a second antidepressant or an atypical antipsychotic. Adjunctive psychotherapy was defined by CPT codes 90832, 90834, 90837, 90839, 90840, 90845, 90847, 90847, 90853, and 96152.

2.2. Statistical analysis

Demographic variables included age, sex, state of residence, and insurance. Psychiatric comorbidities, measured across the 18-month study period, included generalized anxiety disorder, substance-related disorders, insomnia, alcohol-related disorders, PTSD, dysthymic disorder, and panic disorder. The highest prevalence medical comorbidities included hypertension, lower back pain, hyperlipidemia, obesity, COPD, fatigue, osteoarthritis, diabetes, asthma, and anemia. MDD severity level (mild, moderate, severe, in remission, unspecified) was derived from ICD diagnosis codes in the period prior to index. Individuals with multiple severity classifications were assigned the most severe diagnosis. A psychiatrist visit was defined as a visit to a provider with a psychiatry specialty taxonomy code any time in the 18-month study period.

Adherence measures included discontinuation (a 60-day gap without a subsequent claim following the last available days’ supply) and proportion of days covered (PDC) defined as the total days’ supply associated with paid pharmacy claims in the follow-up period divided by 365 days with adherent defined as PDC≥0.80. Further, beginning on the index date, using the days’ supply associated with each pharmacy claim, and assuming that each day of supply was consumed consecutively, we determined if an individual had drug available on a given day. By summing across all individuals, we calculated and then graphed the expected proportion of individuals with available adjunctive pharmacotherapy drug by day throughout the follow-up period. Adherence to adjunctive psychotherapy was the count of unique claims for psychotherapy service during the follow-up period. Optimal adherence was defined as eight or more sessions, the median number of documented visits in the follow-up period.

Frequencies and percentages were calculated for categorical variables and means and standard deviations for continuous variables. Differences between categorical variables were tested with Chi-squared tests. Differences between continuous variables were tested with two-sided t-tests for normally distributed variables and Wilcoxon Mann-Whitney U-tests for skewed variables. Two separate multivariate logistic regression models were used to identify factors associated with: 1) achieving optimal adherence to augmentation therapy among individuals who augmented with pharmacotherapy and 2) achieving optimal adherence to augmentation therapy among individuals who augmented with psychotherapy.

The threshold for significance was set at 0.05. All analyses were performed with SAS software, v9.4, SAS Institute Inc., Cary, NC, USA.

3. Results

A total 218,192 met eligibility criteria: 189,091 augmented with pharmacotherapy and 33,516 augmented with psychotherapy. (Figure 1) The population was predominantly female (74.1%), had an average age of 47.8 years, commercially-insured (71.5%), and included residents from all 50 states and the District of Columbia. Of the 35.1% of participants who had an MDD claim indicating severity, ‘moderate’ was most common (57.4%) followed by ‘severe’ (28.0%), ‘mild’ (10.5%), and ‘in remission’ (4.1%). The most common psychiatric comorbidities were generalized anxiety disorder (20.8%), substance use disorder (20.5%), and insomnia (15.9%). (Table 1)

Figure 1. Flow diagram for identification and selection of study participants.

Table 1. Select demographic, comorbidity, and MDD-related healthcare characteristics.

		Total			Adjunctive Pharmacotherapy			Adjunctive Psychotherapy
		N =n	218,192%		N =n	185,349%		N =n	32,843%	p-value
Gender
	Female	161,700	74.1%		137,936	74.4%		23,764	72.4%	<0.01
	Male	56,492	25.9%		47,413	25.6%		9,079	27.6%
Age
	18–30	28,322	13.0%		21,394	11.5%		6,928	21.1%	<0.01
	31–49	79,835	36.6%		66,637	36.0%		13,198	40.2%
	50–64	96,937	44.4%		85,162	45.9%		11,775	35.9%
	65+	13,033	6.0%		12,094	6.5%		939	2.9%
	Unknown/Other	65	0.0%		62	0.0%		3	0.0%
MDD (most severe, baseline)
	Not specified	141,524	64.9%		129,124	69.7%		12,400	37.8%	<0.01
	Mild	8,068	3.7%		5,885	3.2%		2,183	6.6%
	Moderate	44,029	20.2%		31,267	16.9%		12,762	38.9%
	Severe	21,481	9.8%		16,479	8.9%		5,002	15.2%
	In remission	3,090	1.4%		2,594	1.4%		496	1.5%
Index antidepressant**
	SSRI	106,634	48.9%		87,096	47.0%		19,538	59.5%	<0.01
	SNRI	42,919	19.7%		36,675	19.8%		6,244	19.0%
	Bupropion	29,746	13.6%		25,853	13.9%		3,893	11.9%
	Serotonin modulators	22,028	10.1%		20,388	11.0%		1,640	5.0%
	Tricyclics/tetracyclics	13,325	6.1%		12,193	6.6%		1,132	3.4%
	Alpha 2s	3,456	1.6%		3,077	1.7%		379	1.2%
	MAOI	84	0.0%		67	0.0%		17	0.1%
Insurance
	Commercial	156,026	71.5%		132,958	71.7%		23,068	70.2%	<0.01
	Medicaid	62,166	28.5%		52,391	28.3%		9,775	29.8%
Adjunctive Pharmacotherapy
	SSRI	-	-		44,889	24.2%		-	-
	Antidepressants, other	-	-		37,116	20.0%		-	-
	Serotonin modulators	-	-		35,217	19.0%		-	-
	SNRI	-	-		23,829	12.9%		-	-
	Atypical antipsychotics	-	-		19,162	10.3%		-	-
	Tricyclics/tetracyclics	-	-		17,908	9.7%		-	-
	Alpha 2 receptor agonist	-	-		6,480	3.5%		-	-
	MAOI	-	-		748	0.4%		-	-
Dual Adjunctive Therapy*
	Yes	9,585	4.4%		4,876	2.6%		4,709	14.3%	<0.01
	No	208,607	95.6%		180,473	97.4%		28,134	85.7%
Psychiatric comorbidity
	Generalized anxiety disorder	103,036	47.2%		83,719	45.2%		19,317	58.8%	<0.01
	Substance-related disorders	44,693	20.5%		38,306	20.7%		6,387	19.4%	<0.01
	Insomnia	34,647	15.9%		31,538	17.0%		3,109	9.5%	<0.01
	Alcohol-related disorders	14,289	6.5%		11,212	6.0%		3,077	9.4%	<0.01
	Post-traumatic stress disorder	13,660	6.3%		9,411	5.1%		4,249	12.9%	<0.01
	Dysthymic disorder	12,313	5.6%		9,494	5.1%		2,819	8.6%	<0.01
	Panic attacks/disorder	10,993	5.0%		8,489	4.6%		2,504	7.6%	<0.01
Medical comorbidities
	Hypertension	87,503	40.1%		77,351	41.7%		10,152	30.9%	<0.01
	Lower back pain	77,403	35.5%		68,023	36.7%		9,380	28.6%	<0.01
	Hyperlipidemia	75,795	34.7%		66,949	36.1%		8,846	26.9%	<0.01
	Obesity	59,272	27.2%		51,035	27.5%		8,237	25.1%	<0.01
	COPD	51,131	23.4%		45,010	24.3%		6,121	18.6%	<0.01
	Fatigue	48,857	22.4%		42,633	23.0%		6,224	19.0%	<0.01
	Osteoarthritis	42,481	19.5%		37,761	20.4%		4,720	14.4%	<0.01
	Diabetes	41,356	19.0%		36,462	19.7%		4,894	14.9%	<0.01
	Asthma	27,760	12.7%		23,865	12.9%		3,895	11.9%	<0.01
	Anemia	26,986	12.4%		23,633	12.8%		3,353	10.2%	<0.01

* defined as adding both a 2nd medication (either an antidepressant or antipsychotic) and psychotherapy following an

initial antidepressant medication regimen.

** MDD: major depressive disorder; COPD: chronic obstructive pulmonary

disease; SSRI: selective serotonin reuptake inhibitor; SNRI: serotonin and norepinephrine reuptake inhibitor;

MAOI: monoamine oxidase inhibitor.

3.1. Pharmacotherapy vs. psychotherapy

Individuals who augmented with psychotherapy were substantially different than those who augmented with pharmacotherapy, being younger (43.6 vs. 48.5 years; p < 0.01), more likely to be male (27.6% vs. 25.6%; p < 0.01), and to have MDD severity documented at baseline (62.2% vs 30.3%; p < 0.01). Dual augmentation was relatively infrequent, with only 2.6% of the pharmacotherapy group adding psychotherapy during follow up and only 14.3% who augmented with psychotherapy, adding a 2^nd drug to the treatment regimen. The psychotherapy group had higher rates of generalized anxiety disorder (58.8% vs. 45.2%; p < 0.01); post-traumatic stress disorder (12.9% vs. 5.1%; p < 0.01); alcohol-related disorders (9.4% vs. 6.0%; p < 0.01); dysthymic disorder (8.6% vs. 5.1%; p < 0.01); and panic attacks (7.6% vs. 4.6%; p < 0.01). The pharmacotherapy group had higher rate of insomnia (17.0% vs. 9.5%; p < 0.01) and substance use disorder (20.7% vs. 19.4%; p < 0.01). (Table 2)

Table 2. Demographic characteristics and comorbidities among eligible participants with MDD, by adjunctive therapy.

			Pharmacotherapy							Psychotherapy
			Adherent			Non-Adherent				Adherent			Non-Adherent
			N =n/(mean)	67,148%/(SD)		N =n/(mean)	118,201%/(SD)	p-value		N =n/(mean)	18,024%/(SD)		N =n/(mean)	14,819%/(SD)	p-value
Gender
	Female		49,682	74.0%		88,253	74.7%	<0.01		12,972	72.0%		10,792	72.8%	ns
	Male		17,466	26.0%		29,948	25.3%			5,052	28.0%		4,027	27.2%
Age (mean, years)			49.8	(13.0)		47.7	(13.4)	<0.01		43.2	(13.5)		44.1	(13.9)	<0.01
Index antidepressant**
	SSRI		29,651	44.2%		57,445	48.6%	<0.01		10,667	59.2%		8,871	59.9%	<0.01
	SNRI		12,699	18.9%		23,976	20.3%			3,351	18.6%		2,893	19.5%
Serotonin modulators			8,814	13.1%		11,574	9.8%			895	5.0%		745	5.0%
Alpha 2 receptor agonist			1,178	1.8%		1,899	1.6%			215	1.2%		164	1.1%
Tricyclic or tetracyclic			5,043	7.5%		7,150	6.0%			630	3.5%		502	3.4%
Antidepressant, other			9,739	14.5%		16,114	13.6%			2,256	12.5%		1,637	11.0%
	MAOI		24	0.0%		43	0.0%			10	0.1%		7	0.0%
Dual Adjunctive Therapy*															<0.01
	Yes		2,972	4.4%		1,904	1.6%	<0.01		3,403	18.9%		1,306	8.8%
	No		64,176	95.6%		116,297	98.4%			14,621	81.1%		13,513	91.2%
Psychiatry visit															<0.01
	Yes		13,184	19.6%		26,325	22.3%	<0.01		7,479	41.5%		5,584	37.7%
	No		53,964	80.4%		91,876	77.7%			10,545	58.5%		9,235	62.3%
Psychiatric
Generalized anxiety disorder			11,743	17.5%		22,425	19.0%	<0.01		6,400	35.5%		4,828	32.6%	<0.01
Panic attacks/disorder			2,739	4.1%		5,750	4.9%	<0.01		1,395	7.7%		1,109	7.5%	ns
Post-traumatic stress disorder			3,032	4.5%		6,379	5.4%	<0.01		2,628	14.6%		1,621	10.9%	<0.01
	Dysthymic disorder		3,205	4.8%		6,289	5.3%	<0.01		1,665	9.2%		1,154	7.8%	<0.01
	Insomnia		10,841	16.1%		20,697	17.5%	<0.01		1,622	9.0%		1,487	10.0%	<0.01
Alcohol-related disorders			3,389	5.0%		7,823	6.6%	<0.01		1,787	9.9%		1,290	8.7%	<0.01
Substance-related disorders			12,471	18.6%		25,835	21.9%	<0.01		3,505	19.4%		2,882	19.4%	ns
Medical
	Hypertension		29,396	43.8%		47,955	40.6%	<0.01		5,228	29.0%		4,924	33.2%	<0.01
	Lower back pain		24,063	35.8%		43,960	37.2%	<0.01		4,905	27.2%		4,475	30.2%	<0.01
	Hyperlipidemia		26,143	38.9%		40,806	34.5%	<0.01		4,565	25.3%		4,281	28.9%	<0.01
	Obesity		19,529	29.1%		31,506	26.7%	<0.01		4,339	24.1%		3,898	26.3%	<0.01
	COPD		16,509	24.6%		28,501	24.1%	<0.05		3,192	17.7%		2,929	19.8%	<0.01
	Fatigue		14,917	22.2%		27,716	23.4%	<0.01		3,295	18.3%		2,929	19.8%	<0.01
	Osteoarthritis		14,419	21.5%		23,342	19.7%	<0.01		2,472	13.7%		2,248	15.2%	<0.01
	Diabetes		14,483	21.6%		21,979	18.6%	<0.01		2,536	14.1%		2,358	15.9%	<0.01
	Asthma		8,658	12.9%		15,207	12.9%	ns		2,098	11.6%		1,797	12.1%	ns
	Anemia		8,561	12.7%		15,072	12.8%	ns		1,778	9.9%		1,575	10.6%	<0.05

* defined as adding both a 2nd medication (either an antidepressant or antipsychotic) and psychotherapy following an

initial antidepressant medication regimen.

** MDD: major depressive disorder; COPD: chronic obstructive pulmonary

disease; SSRI: selective serotonin reuptake inhibitor; SNRI: serotonin and norepinephrine reuptake inhibitor;

MAOI: monoamine oxidase inhibitor

3.2. Adherence to augmentation therapy

The most frequent augmentation pharmacotherapy class was SSRIs (20.6%) followed by other antidepressants (primarily, bupropion) (17.0%), serotonin modulators (16.1%) and SNRIs (10.9%). (Table 1) After 12 months, 50.4% of the SSRI group remained on therapy, followed by SNRI (46.8%), other antidepressants (42.2%), and atypical antipsychotics (32.2%). (Figure 3) The overall rate of adherence to augmentation pharmacotherapy was substantially higher among those who also initiated psychotherapy (64.0% on therapy at 12 months compared to 41.4%) compared to those who did not. (Figure 2)

Figure 2. Adherence to adjunctive pharmacotherapy comparing individuals who did and did not add psychotherapy.

Figure 3. Adherence to adjunctive pharmacotherapy by therapeutic class.

Among the 32,843 individuals who augmented with psychotherapy, 92.8% of the visits were individual therapy with the individual and/or family present, followed by group psychotherapy (4.5%), family psychotherapy (1.6%), health and behavioral assessment (0.8%), and psychotherapy for crisis (0.2%). Overall, the median number of psychotherapy visits was 8 (IQR: 4–15); with individuals who added a 2^nd drug to the therapeutic regimen in the follow-up period with a median of 13 visits (IQR: 7–22). Individuals who added a 2^nd pharmacotherapy to their regimen remained in psychotherapy significantly longer than those who did not: 68% remained in psychotherapy as of 180 days post-augmentation compared to only 48% of those who did not add a 2^nd drug (Figure 4).

Figure 4. Time (in days) from psychotherapy augmentation event to final psychotherapy visit.

3.3. Factors associated with adherence

Given the study sample size, nearly all factors were significantly associated with adherence to pharmacotherapy and with completing eight or more psychotherapy sessions. (Table 2) After adjustment, the strongest positive predictor of adherence to adjunctive pharmacotherapy was adherence to index antidepressant therapy (OR: 3.10; 95%CI: 3.02, 3.19)), followed by initiating psychotherapy (OR: 3.03; 95%CI: 2.84, 3.22), age relative to the youngest age group of 18–30 years (65+ years – OR: 1.39; 95%CI: 1.32, 1.46; and 50–64 years – OR: 1.29; 95%CI: 1.24, 1.33); and 31–49 years – OR 1.15; 95%CI: 1.11, 1.19), and the highest coded MDD severity level – relative to severity unspecified (moderate – OR: 1.09; 95%CI: 1.06, 1.12 and severe – OR: 1.07; 95% CI: 1.03, 1.11). (Table 3) Compared to SSRIs, augmentation with any other drug class was associated with significantly lower odds of adherence and each psychiatric comorbidity was independently associated with a lower odds of achieving adherence to augmentation therapy.

Table 3. Logistic regression results for factors associated with adherence to adjunctive therapy.

			Adjunctive Pharmacotherapy				Adjunctive Psychotherapy
Characteristics	Level		O.R.		95% CI		O.R.		95% CI
Age
	18–30		1.0				1.0
	31–49		1.15		(1.11, 1.19)		-
	50–64		1.29		(1.24, 1.33)		-
	65+		1.39		(1.32, 1.46)		0.76		(0.66, 0.88)
Sex
	Male		1.0				1.0
	Female		0.94		(0.92, 0.96)		0.95		(0.90, 0.99)
Insurance							-
	Medicaid		1.0
	Commercial		0.95		(0.93, 0.98)		-
Dual Adjunctive Therapy*
	No		1.0				1.0
	Yes		3.03		(2.84, 3.22)		2.35		(2.19, 2.52)
Adjunctive Therapy**
	SSRI		1.0				N/A
	SNRI		0.83		(0.81, 0.86)
Serotonin modulator			0.36		(0.35, 0.37)
Alpha-2 receptor agonist			0.34		(0.32, 0.36)
Tricyclics and tetracyclics			0.36		(0.35, 0.38)
Antidepressant, other			0.59		(0.57, 0.61)
Atypical antipsychotics			0.46		(0.44, 0.47)
Adherence to index ADT
	≤0.59 PDC		1.0				1.0
	0.6–0.79 PDC		2.01		(1.96, 2.07)		1.06		(1.00, 1.13)
	0.8+ PDC		3.10		(3.02, 3.19)		1.13		(1.07, 1.20)
Psychiatry visit	No		1.0				1.0
	Yes		0.95		(0.92, 0.97)		1.09		(1.04, 1.15)
Comorbidities***
Substance use disorder			0.89		(0.86, 0.91)		-
Alcohol use disorder			0.85		(0.82, 0.89)		1.13		(1.05, 1.23)
	Anxiety		0.95		(0.93, 0.97)		-
Post-traumatic stress disorder			0.94		(0.90, 0.99)		1.34		(1.25, 1.44)
	Panic attacks		0.93		(0.88, 0.98)		-
Dysthymic disorder			0.91		(0.87, 0.95)		1.24		(1.14, 1.34)
	Hypertension		1.03		(1.01, 1.06)		0.88		(0.83, 0.93)
	Low back pain		0.96		(0.94, 0.98)		0.88		(0.84, 0.93)
	Hyperlipidemia		1.08		(1.06, 1.11)		0.92		(0.87, 0.98)
	Obesity		1.10		(1.07, 1.12)		-
	COPD		1.04		(1.01, 1.07)		0.93		(0.87, 0.98)
Most Recent MDD Diagnosis Severity
	Unspecified		1.0				1.0
	Severe		1.07		(1.03, 1.11)		1.12		(1.04, 1.20)
	Moderate		1.09		(1.06, 1.12)		1.15		(1.09, 1.21)
	Mild		ns				1.2		(1.09, 1.31)
	In remission		ns				1.2		(1.00, 1.45)

* defined as adding both a 2nd medication (either an antidepressant or antipsychotic) and psychotherapy following an initial antidepressant medication regimen. ** MDD: major depressive disorder; COPD: chronic obstructive pulmonary disease; SSRI: selective serotonin reuptake inhibitor; SNRI: serotonin and norepinephrine reuptake inhibitor; MAOI: monoamine oxidase inhibitor; PDC: proportion of days covered*** Odds ratio for a comorbidity is in comparison with absence of each condition.

After adjustment, fewer factors were associated with the odds of achieving optimal psychotherapy adherence. The strongest positive predictor of adherence to adjunctive psychotherapy was the addition of a 2^nd pharmacotherapy during the follow-up period (OR: 2.35; 95%CI: 2.19, 2.52), followed by selected psychiatric comorbidities including anxiety (OR: 1.34), dysthymic disorder (OR: 1.24), and alcohol use disorder (OR: 1.13). The oldest age group (65+ years) and select medical comorbidities were associated with lower odds of achieving optimal psychotherapy adherence. (Table 3)

4. Discussion

In this population of nearly 220,000 individuals with MDD, adherence to adjunctive pharmacotherapy was poor, and continuation of psychotherapy was low. However, individuals who initiated a 2^nd pharmacotherapy treatment and psychotherapy achieved significantly higher levels of adherence to both augmentation therapies. The likelihood of achieving optimal adherence to adjunctive pharmacotherapy varied by therapeutic class, with the highest adherence among individuals who added SSRIs and SNRIs. Of potentially more significance is the association of adherence with psychiatric comorbidity. Each measured psychiatric comorbidity was associated with lower levels of adherence to augmentation pharmacotherapy while PTSD, alcohol-use disorder, and dysthymic disorder were associated with higher levels of psychotherapy use.

Based on our results, augmenting antidepressant therapy with a 2^nd drug is the preferred therapeutic option among those who do not adequately respond to initial treatment. The results of this study are relevant to any clinician who manages individuals with MDD. It is estimated that more than 50% of individuals have an inadequate response to the initial therapy. Among these individuals, providers have various therapeutic options, including changing dose, switching medications, and adding a medication. Given the frequency of augmentation, it is important for clinicians to recognize the high rate of non-adherence, even among a population willing to attempt additional therapeutic options. Our results suggest the importance of monitoring progress.

Despite the prevalence of augmentation, little has been published on adherence to adjunctive pharmacotherapy using real-world data. Recently published results focused only on atypical antipsychotic medications as augmentation therapy. Yan, et al [12]. reported poor adherence among four specific antipsychotic medications – with PDCs ranging from 20.7% to 29.6% and Broder, et al [13]. reported six-month adjusted PDC rates for adjunctive atypical antipsychotic use between 55.5 and 58.9%.

The concept of adherence to psychotherapy is difficult to measure in administrative data. There is no consensus methodology for measuring adherence to psychotherapy visits. Gaspar, et al [14]., defined adherence as four visits in the first month; however, the population studied was more severe and not representative of the broader population of individuals with MDD.

Despite this challenge, it is worth highlighting that several factors that predict adherence to pharmacotherapy also predict non-adherence (or lower rates of) psychotherapy. Older age and several medical comorbidities are associated with higher adherence to pharmacotherapy but less persistence with psychotherapy, while psychiatric care and several psychiatric comorbidities are associated with greater use of psychotherapy but lower adherence to pharmacotherapy.

As defined by the ICD code selected, the severity of MDD appears to play an important role in adherence to augmentation therapy. This finding is difficult to interpret because of the substantial number of codes selected that define severity as ‘unspecified.’ However, our results suggest that selecting an ICD-10 code that includes a severity assessment may be useful for identifying future adherence to therapy. These results are consistent with other studies that indicate ICD-10 coding of MDD severity is clinically valuable, predicting both the risk of relapse and of suicide [15] and that use severity classifications from validated screening questionnaires to predict treatment response [16].

Our results document relatively low rates of adherence to adjunctive therapy, whether pharmaco- or psychotherapeutic. These results may provide information to providers interested in identifying individuals at highest risk of non-adherence and who may benefit from a more intensive care model. More promising, our results indicate that adding a psychotherapeutic intervention to adjunctive pharmacotherapy has a positive influence on medication adherence. These results may represent the cumulative therapeutic effect of complementary interventions or may reflect a selection bias, selecting for individuals and providers with a commitment to therapeutic engagement. In practice, it likely reflects some of each. Teasing out the independent effect of the drug’s therapeutic effect from the impact of an individual’s or provider’s behavior is a significant and long-standing challenge. Nonetheless, given the relatively low rate of psychotherapy [17,18] and the well-documented emotional and practical patient barriers to psychotherapy (i.e. time commitment, cost, stigma) [19,20], these results suggest that alternative therapeutic options – complementary to pharmacotherapy – may have significant benefits to the individual. Finally, if these results hold true, the potential benefits of increased adherence may extend from improved clinical outcomes to improved financial outcomes (lower rates of inpatient and emergency services use).

4.1. Limitations

The study sample represents individuals who elect to add adjunctive pharmacotherapy and/or psychotherapy to an initial antidepressant medication. These results only apply to this population and may not be generalizable to individuals with MDD, overall. Next, the data source does not include insurance benefits eligibility, and may omit some healthcare utilization. This risk is mitigated by the required ongoing use of care, measured by submitted and remitted claims data but cannot be discounted. Further, the study can only infer the intent of treatment and cannot determine reasons for discontinuation. The population has recent diagnoses of MDD, and prescription drug use recommended by guidelines; however, claims cannot confirm the intent for the adjunctive therapeutic intent or the reasons for discontinuation which are well documented [21]. The analysis assumes that adjunctive therapy should persist throughout the 12-month follow up period.

5. Conclusions

Use of psychotherapy as an adjunctive therapy appears under-utilized and adherence to adjunctive therapy, whether medication or psychotherapy, was low. High rates of augmentation coupled with low rates of adherence to augmentation reinforce the challenges faced by providers and individuals in the management of MDD. The insight that an additional therapy appears to enhance adherence to the initial adjunctive therapy indicates an opportunity to explore the synergistic effect of alternative adjunctive therapies, including mHealth and related solutions.

Declarations of interest

J Liberman, C Ruetsch and P Rui are employees of Health Analytics, LLC which was funded to conduct the study. F Forma was an employee of Otsuka Pharmaceutical Development & Commercialization, Inc. at the time the study was conducted. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Author contributions

F Forma: Study conception and design; interpretation of results; critical review of manuscript; final approval of submitted version.

J Liberman: Study conception and design; interpretation of results; critical review of manuscript; final approval of the submitted version.

C Ruetsch: Study conception and design; interpretation of results; critical review of manuscript; final approval of the submitted version.

P Rui: Data management and analysis; critical review of manuscript; final approval of the submitted version.

All authors approved the final version for publication and agree to be accountable for all aspects of the work.

Acknowledgments

The authors would like to thank Margaret L. Stinstrom for her administrative support.

Additional information

Funding

This work was supported by Otsuka Pharmaceutical Development & Commercialization, Inc.