Does knowledge of patient non-compliance change prescribing behavior in the real world? A claims-based analysis of patients with serious mental illness

Figures & data

Table 1 Summary of outcome variables

Outcome	Definition
Increased dose	To detect dose increase we examined daily doses (strength^*quantity/days supply) of filled prescriptions of the first atypical drug. We flagged patients if they had any daily dose increase.
Augmented with another atypical drug	To be classified as an augmenter with another antipsychotic, we required that the patient had two fills of a new antipsychotic, and the initiation of the new antipsychotic occurred prior to the last fill of the first atypical prescription.
Augmented with non- atypical SMI drug	Augmenting with another SMI drug was defined similarly to augmenting with another antipsychotic except we imposed an additional restriction that the new augmenting agent not be used in the 6 months prior to initiating an atypical. Specifically, we required that: the patient had two fills of a new class, the initiation of the new class occurred prior to the last fill of the first atypical prescription, and the patient had no claims for the new class during the 6 months prior to the time of first atypical fill.
Switched to another atypical drug	To be classified as a switcher, we required that: the patient discontinued the first atypical drug, that is, he had no supply for at least 30 days (no filled prescription for the first atypical molecule for at least 30 days after the last prescription ran out [last fill date + days supply]), and the patient filled a new antipsychotic molecule during the period of no supply defined earlier (within 30 days from the day the last prescription of the first atypical molecule ran out [last fill + days supply]).
Switched to LAI formulation of same molecule	Individuals considered as using an LAI had a fill for a long-acting formulation of the first atypical which occurred after the date of first oral atypical fill.

Note: Additional details on drugs tracked for these definitions are listed in Tables S1 and S2.

Abbreviations: LAI, long-acting injectable; SMI, serious mental illness.

Table 2 Attrition table

Criteria	Included N	Excluded N	%
Patients with at least one of the following ICD-9 diagnosis codes: 295.xx, 296.xx, 301.13, 300.4x, 311.xx	15,082,765
Oral atypical antipsychotic prescribed	1,899,983	13,182,782	87.40%
≥18 years of age at time of prescription	1,546,256	353,727	2.35%
Continuous enrollment for 12 months after first atypical fill	891,555	654,701	4.34%
≥1 inpatient or ≥2 outpatient claims with a diagnosis code for SMI during the 6 months pre- and 12-month post-first atypical fill period	664,304	227,251	1.51%
Not in another payer population in a prior year	661,634	2,670	0.02%
No more than one atypical oral on date of first atypical fill	649,979	11,655	0.08%
Atypical drug prescriptions have a day supply ≤90	610,094	39,885	0.26%
First atypical drug prescription has a day supply >0 on date of first atypical fill	607,933	2,161	0.01%
Continuous enrollment for 6 months prior to first atypical fill	423,118	199,845	1.32%
Patients with no atypical antipsychotics in the 6 months prior to first atypical fill (LAI)	419,965	3,153	0.02%

Notes: Percents represent proportion of initial sample: Patients with at least one of the following codes occurring any time from January 1, 2008 through December 31, 2014: ICD-9 295.xx, 296.xx, 301.13, 300.4x, 311.xx.

Abbreviations: LAI, long-acting injectable; SMI, serious mental illness.

Table 3 Descriptive statistics of the study variables for non-adherent (PDC<0.8) patients with SMI stratified by history of noncompliance status

	History of non- compliance (N=4,033)		No history of non- compliance (N=282,216)		Difference
	Na	% within subgroupa	Na	% within subgroupa	Difference in mean/rate	P-value
Age (years), mean/SD	38.7	14.4	42.5	15.3	−3.8	<0.001
Charlson Comorbidity Index, mean/SD	0.77	1.47	0.42	1.01	0.35	<0.001
Female	2,136	53.0%	186,287	66.0%	−13.0%	<0.001
Diagnosis type
Bipolar disorder	2,180	54.1%	114,321	40.5%	13.6%	<0.001
Major depressive disorder	2,872	71.2%	209,715	74.3%	−3.1%	<0.001
Schizophrenia	1,740	43.1%	30,137	10.7%	32.4%	<0.001
Payer type
Commercial	1,249	31.0%	179,264	63.5%	−32.5%	<0.001
Medicaid	2,670	66.2%	84,069	29.8%	32.6%	<0.001
Medicare	114	2.8%	18,883	6.7%	−3.9%	<0.001
Outcome
Increased dose of first atypical drug	1,056	26.2%	63,399	22.5%	3.7%	<0.001
Augmented with another atypical drug	109	2.7%	4,780	1.7%	1.0%	<0.001
Augmented with non-atypical SMI drug	1,378	34.2%	94,459	33.5%	0.7% 0.176
Switched to another atypical drug	72,313	25.6%	1,759	43.6%	−18.0%	<0.001
Switched to LAI formulation of same molecule	552	0.2% 58	1.4%	−1.2%	<0.001

Notes:

a Unless otherwise indicated.

Abbreviations: LAI, long-acting injectable; SMI, serious mental illness; PDC, proportion of days covered.

Figure 1 Predicted treatment patterns for non-adherent new user patients (PDC <0.8) with SMI.

Abbreviations: LAI, long-acting injectable; SMI, serious mental illness; PDC, proportion of days covered.

Table 4 Differences in predicted treatment patterns between patients with and without documented history of non-compliance as measured in sensitivity analyses

	Increased dose of first atypical drug	Augmented with another atypical drug	Augmented with non- atypical SMI drug	Switched to another atypical drug	Switched to LAI formulation of same molecule
Baseline analysis (all patients) (N = 286,249)
Difference in predicted value	2.28%	−0.26%	−0.54%	8.14%	0.05%
P value	0.004	0.035	0.526	<0.001	0.008
Schizophrenia patients (N = 31,877)
Difference in predicted value	1.05%0.332	−1.92%<0.001	−1.58%0.175	10.50%<0.001	1.08%<0.001
P value
Major depressive disorder patients (N = 212,587)
Difference in predicted value	3.80%	0.30%	2.71%	14.85%	0.12%
P value	<0.001	0.122	0.003	<0.001	0.002
Bipolar disorder patients (N = 116,501)
Difference in predicted value	3.38%	0.48%	0.72%	15.32%	0.31%
P value	<0.001	0.124	0.498	<0.001	<0.001
Patients with Medicare coverage (N = 18,997)
Difference in predicted value	3.11%	0.33%	−1.32%	4.92%	0.52%
P value	0.431	0.759	0.753	0.253	0.265
Patients with Medicaid coverage (N = 86,739)
Difference in predicted value	1.50%	−0.81%	−1.83%	6.15%	0.13%
P value	0.073	<0.001	0.0505	<0.001	0.016
Patients with commercial coverage (N = 180, 513)
Difference in predicted value	3.46%	0.03%	2.06%	12.87%	0.02%
P value	0.005	0.888	0.127	0.353	<0.001
Baseline analysis (history of non-compliance checked within 12 months of first fill)
Difference in predicted value	2.28%	−0.26%	−0.54%	8.14%	0.05%
P value	0.004	0.035	0.526	<0.001	0.008
Sensitivity analysis (history of non-compliance checked any time before or after atypical fill)
Difference in predicted value	0.30%	−0.13%	−1.00%	3.92%	0.03%
P value	0.436	0.059	0.018	<0.001	<0.001
Sensitivity analysis (history of non-compliance checked within 6 months of first fill)
Difference in predicted	0.82%	−0.08%	−0.47%	9.43%	0.03%
P value	0.349	0.582	0.637	<0.001	0.012
Sensitivity analysis with no weights
Difference in predicted value	2.14%	−0.34%	−0.70%	7.90%	0.03%
P value	0.0016	<0.001	0.3498	<0.001	0.0026
Propensity score matched analysis
Difference in predicted value	3.14%	−0.82%	−0.51%	8.75%	0.71%
P value	0.005	0.069	0.674	<0.001	<0.001

Abbreviations: LAI, long-acting injectable; SMI, serious mental illness.

Table S1 Antipsychotic medications included in the analysis

Drugs	Available routes of administration
	Oral	LAI	SAI
Atypical
Aripiprazole	x	x	x
Asenapine	x
Brexpiprazole	x
Cariprazine	x
Clozapine	x
Iloperidone	x
Lurasidone	x
Olanzapine	x	x	x
Paliperidone	x	x
Pimavanserin	x
Quetiapine	x
Risperidone	x	x
Ziprasidone	x		x
Typical
Chlorpromazine	x		x
Fluphenazine	x	x	x
Haloperidol	x	x	x
Loxapine	x		x
Molindone	x
Perphenazine	x		x
Pimozide	x
Prochlorperazine	x		x
Thioridazine	x
Thiothixene	x
Trifluoperazine	x		x

Notes: The list of drugs was drawn from the Redbook therapeutic class “Pscyhother,Tranq/Antipsychotic” therapeutic class (thercls=70). NDC codes for each atypical molecule were obtained from the US Food and Drug Administration (FDA) Drugs@FDA and the Centers for Medicare and Medicaid Services (CMS) HCPCS databases; LAI and SAI as existing routes of administration were identified based on drug labels.

Abbreviations: LAI, long-acting injectable; SAI, short-acting injectable; HCPCS, Healthcare Common Procedure Coding System; NDC, National Drug Codes.

Table S2 Other SMI drug classes of interest for augmenting

Class
Antidepressants
Monoamine oxidase inhibitors (MAOIs)
Selective serotonin reuptake inhibitors (SSRI)
Serotonin-norepinephrine reuptake inhibitors (SNRI)
Tetracyclic tricyclic
Mood stabilizers
Benzodiazepines

Notes: NDC codes for each molecule were obtained from the US Food and Drug Administration (FDA) Drugs@FDA and the Centers for Medicare and Medicaid Services (CMS) HCPCS databases.

Abbreviations: SMI, serious mental illness; NDC, National Drug Codes; HCPCS, Healthcare Common Procedure Coding System.

Table S3 Descriptive statistics of study variables for all new starter patients

Variables	Non-adherent		Adherent		All
	N	%	N	%	N	%
N	286,249		133,703		419,952
Age (years), mean/SD	42.45	15.33	46.83	15.34	43.84	15.47
Female	188,423	65.8%	84,758	63.4%	273,181	65.1%
Charlson Comorbidity Index, mean/SD	0.42	1.02	0.44	1.04	0.43	1.03
Diagnosis type
Schizophrenia	31,877	11.1%	30.123	22.5%	62,000	14.8%
Major depressive disorder	212,587	74.3%	84,187	63.0%	296,774	70.7%
Bipolar disorder	116,501	40.7%	52,959	39.6%	169,460	40.4%
Payer type
Medicare	18,997	6.6%	11,488	8.6%	30,485	7.3%
Medicaid	86,739	30.3%	62,639	46.8%	149,378	35.6%
Commercial	180,513	63.1%	59,576	44.6%	240,089	57.2%
Outcomes
Increased dose of first atypical drug	64,455	22.5%	63,043	47.2%	127,498	30.4%
Augmented with another atypical drug	4,889	1.7%	12,050	9.0%	16,939	4.0%
Augmented with non-atypical SMI drug	95,837	33.5%	73,109	54.7%	168,946	40.2%
Switched to another atypical drug	74,072	25.9%	7,595	5.7%	81,667	19.4%
Switched to LAI formulation of same molecule	610	0.2%	115	0.1%	725	0.2%
Had any pharmacotherapy change (dose increase/augment or switch)	167,829	58.6%	103,838	77.7%	271,667	64.7%

Abbreviations: LAI, long-acting injectable; SMI, serious mental illness.

Table S4 Logit regression results for all non–adherent new starter patients

	Increased dose of first atypical drug	Augmented with another atypical drug	Augmented with non−atypical SMI drug	Switched to another atypical drug	Switched to LAI formulation of same molecule
Documented non– compliance	0.13***	−0.22*	−0.02	0.40***	0.74***
	(0.04, 0.21)	(−0.45, 0.01)	(−0.10, 0.05)	(0.32, 0.48)	(0.38,1.09)
Age range, years
25–34	−0.02	−0.05	0.07***	0.00	−0.31***
	(−0.06, 0.01)	(−0.15, 0.05)	(0.04, 0.09)	(−0.03, 0.03)	(−0.54, −0.08)
35–44	−0.01	0.08	0.09***	−0.02	−0.75***
	(−0.04, 0.02)	(−0.02, 0.18)	(0.06, 0.12)	(−0.05, 0.01)	(−1.02, −0.49)
45–54	−0.09***	0.15***	0.07***	−0.06***	−0.58***
	(−0.12, −0.06)	(0.06, 0.25)	(0.04, 0.09)	(−0.09, −0.03)	(−0.83, −0.34)
55–64	−0.18***	0.16***	−0.02	−0.13***	−0.56***
	(−0.22, −0.15)	(0.05, 0.28)	(−0.05, 0.01)	(−0.16, −0.10)	(−0.89, −0.23)
65–74	−0.15***	0.15	0.03	−0.01	−0.69**
	(−0.25, −0.05)	(−0.11, 0.41)	(−0.06, 0.11)	(−0.10, 0.08)	(−1.35, −0.04)
75+	0.01	0.34**	0.01	0.16***	−0.74*
	(−0.09, 0.11)	(0.07, 0.62)	(−0.08, 0.10)	(0.07, 0.25)	(−1.55, 0.08)
Female	−0.00	0.05	0.01	0.09***	−0.37***
	(−0.02, 0.02)	(−0.02, 0.12)	(−0.01, 0.03)	(0.06, 0.11)	(−0.58, −0.16)
Charlson Comorbidity Index	0.00	0.04**	−0.05***	0.00	−0.09
	(−0.01, 0.02)	(0.01, 0.08)	(−0.06, −0.04)	(−0.01, 0.01)	(−0.20, 0.03)
Schizophrenia diagnosis	0.22***	1.43***	0.07***	1.01***	2.94***
	(0.19, 0.26)	(1.35, 1.51)	(0.03, 0.10)	(0.98, 1.04)	(2.65, 3.24)
Bipolar disorder diagnosis	0.25***	0.53***	0.37***	0.65***	0.06
	(0.23, 0.27)	(0.46, 0.60)	(0.35, 0.39)	(0.63, 0.67)	(−0.15, 0.27)
Commercial insurance	0.04	−0.19	0.06	−0.14***	−1.22***
	(−0.05, 0.13)	(−0.43, 0.04)	(−0.01, 0.14)	(−0.22, −0.06)	(−1.84, −0.59)
Medicaid	−0.03	0.29**	0.09**	−0.03	0.38
	(−0.12, 0.05)	(0.07, 0.52)	(0.01, 0.16)	(−0.11, 0.05)	(−0.19, 0.95)
Constant	−1.33***	−4.76***	−0.95***	−1.42***	−1.41***
	(−1.42, −1.24)	(−5.01, −4.51)	(−1.03, −0.87)	(−1.50, −1.33)	(−1.49, −1.32)
N	286,249	286,249	286,249	286,249	286,249
Pseudo R–squared	0.00403	0.0500	0.00788	0.0319	0.0350

Notes: Robust 95% CI in parentheses,

*** P<0.01,

** P<0.05,

* P<0.1. Reference diagnosis is major depressive disorder; reference insurance coverage is Medicare; reference age group is 18–24.

Abbreviations: N, number; LAI, long-acting injectable; SMI, serious mental illness.