Treatment patterns, healthcare utilization, and costs of patients with treatment-resistant depression initiated on Esketamine intranasal spray and covered by US commercial health plans

Major depressive disorder (MDD) affects an estimated 8.4% of US adults, and roughly 30% of treated patients with MDD—about 1.1% of the adult population—experience treatment-resistant depression (TRD), typically defined as inadequate response to two or more adequate antidepressant trials. TRD is associated with substantially worse clinical and economic outcomes than non-TRD MDD: lower remission and response after multiple trials, higher rates of comorbid physical and mental conditions (for example cardiometabolic disease and anxiety), greater healthcare use, and notable productivity losses. The authors cite an annual societal burden estimate of $43.8 billion attributable to TRD and report that patients with TRD take on average 35.8 work-loss days per year, translating into higher work-loss rates compared with employees without MDD or without TRD. This study set out to describe real-world use of esketamine (ESK) intranasal spray and associated healthcare resource use and costs among patients with TRD in the United States following ESK approval for TRD in 2019. Specifically, the investigators aimed to characterise treatment patterns (session counts, dosing, timing), and to compare acute-care utilisation and costs in the 6 months before versus the 6 months after ESK initiation, including analyses in subgroups with cardiometabolic conditions, pain, anxiety disorder, and substance use disorder (SUD). The focus was on patients treated at REMS-certified centres and covered by US commercial or Medicare supplemental plans.

This was a descriptive retrospective cohort study using de-identified US health insurance claims from the IBM MarketScan Commercial and Medicare Supplemental Databases covering January 2015 through October 2020. The index window began on 5 March 2019 (ESK approval date for TRD); the index date was each patient's first ESK claim. A 6-month baseline period before the index and a 6-month follow-up period beginning on the index date were used for comparisons. Analyses were restricted to patients with continuous health plan eligibility for the baseline and follow-up windows. Patients entered the TRD cohort if they had evidence of TRD prior to ESK initiation, operationalised as pharmacy claims for at least two unique antidepressants during the same major depressive episode (MDE). An MDE was defined as a period without a gap of 6 months without MDD diagnoses or antidepressant claims. Individuals with any ESK pharmacy or medical claim prior to 5 March 2019 or before evidence of TRD were excluded. Four non‑mutually exclusive comorbidity subgroups were specified: cardiometabolic (metabolic and cardiovascular conditions), pain, anxiety disorder, and SUD; subgroup membership required at least one relevant diagnosis or procedure claim in the baseline period. ESK treatment sessions were identified by claims on distinct dates and characterised by number, frequency (days between sessions), and dose based on NDC and HCPCS codes. The authors framed sessions according to the label: an induction phase of 8 sessions (twice weekly for 28 days), a first maintenance phase comprising sessions 9–12 (once weekly), and a second maintenance phase beginning at the 13th session to be individualised. Outcomes included ESK treatment patterns and payer-perspective medical costs and acute-care healthcare resource utilisation (inpatient and emergency department [ED] visits), reported per patient per 6 months or per 100 patients per 6 months where stated. Costs were inflation-adjusted to 2020 US dollars using the medical care component of the US Consumer Price Index. All analyses were descriptive; continuous variables were summarised by means, standard deviations and medians, and categorical variables by frequencies and proportions. The extracted text notes supplementary coding lists but does not reproduce them.

A total of 269 patients met inclusion criteria for the TRD cohort. Subgroup sizes were 123 in the cardiometabolic subgroup, 144 in the pain subgroup, 189 in the anxiety disorder subgroup, and 58 in the SUD subgroup. The mean age in the TRD cohort was 44.6 years and 62.5% were female. Within the cardiometabolic subgroup, 56.1% had a cardiovascular diagnosis in the baseline period; hypertension was the most common condition (53.7%). ESK treatment patterns showed a mean (median) of 11 treatment sessions in the TRD cohort; subgroup means ranged from 11 (cardiometabolic) to 13 (SUD). Overall, 61.3% of patients completed at least 8 sessions, the number that corresponds to the induction phase per label; subgroup completion rates ranged from 60.2% (cardiometabolic) to 72.4% (SUD). The mean time to complete 8 sessions was 56.9 days (median 38.0 days), and the median interval between sessions during induction ranged from 5 to 8 days across cohorts—longer than the twice-weekly (about 28-day) induction interval recommended on the label. Forty point five percent of patients completed at least 12 sessions (first maintenance phase), with subgroup rates varying from 36.6% to 44.4%. The mean (median) time from the 9th to the 12th session was 24.0 (21.0) days overall and median days between sessions during this phase ranged from 6 to 8 days; the label recommends once-weekly dosing in this phase. Regarding dose, most patients initiated at the recommended 56‑mg dose (78.1% overall) and by the 12th session the majority had titrated to 84 mg (88.1% overall). Comparisons of a 6‑month post-initiation period versus the 6‑month pre-initiation period showed generally lower mean all-cause and mental health-related acute-care medical costs after ESK initiation. For example, mean mental health-related inpatient costs fell from $3,480 ± $13,328 (median $0) pre-initiation to $3,262 ± $16,666 (median $0) post-initiation (mean difference -$218). Mean mental health-related ED costs decreased from $608 ± $2,525 to $269 ± $1,143 (mean difference -$339). However, mean all-cause inpatient costs increased post-initiation by $547, which the authors attribute to non-mental-health inpatient costs (+$765) observed in eight patients, related to circulatory and respiratory system diagnoses and general symptoms. Healthcare resource utilisation trends paralleled the cost findings. The mean number of all-cause inpatient admissions per-100-patients-per-6-months increased by 1.9 overall, driven by a +2.6 increase in non-mental-health inpatient admissions seen in a small subset of patients. By contrast, mean mental health-related inpatient admissions decreased by 0.7 admissions overall, with larger reductions in the comorbidity subgroups (largest reported reduction -13.8 in the SUD subgroup). Mean all-cause ED visits per-100-patients-per-6-months decreased by 17.9 visits overall and by as much as 48.2 in the SUD subgroup. Mental health-related ED visits decreased across cohorts, with reductions ranging up to 26.4 visits in the pain subgroup. Outpatient visits and outpatient medical costs increased post-initiation across the TRD cohort and subgroups, which the authors note likely reflect ESK administrations.

Joshi and colleagues interpret their descriptive findings to indicate that most patients with TRD who initiated ESK in routine care completed the induction number of sessions, but commonly at dosing intervals longer than those recommended on the product label. In the 6 months after initiation there were reductions in mental health-related inpatient costs in the TRD cohort and most comorbidity subgroups (except for the SUD subgroup in one inpatient-cost comparison), together with reductions in all-cause and mental health-related ED costs across the cohort and subgroups. The investigators suggest these cost reductions were driven by fewer mental health-related inpatient admissions and fewer ED visits. The study team discusses plausible reasons for the observed longer-than-recommended dosing intervals: logistical challenges such as scheduling or rescheduling appointments, difficulty arranging transport to REMS-certified treatment centres, coordination of time off work, and the limited availability of REMS-certified centres on specific days—missing a single session could therefore extend the dosing interval. They also note that progression to maintenance is a clinical decision and not all patients who complete induction enter maintenance, and that clinician judgement and patient preference may influence dosing schedules. Although this analysis did not measure clinical effectiveness, the authors comment that adherence to the recommended dosing schedule could plausibly affect outcomes and merits further investigation. The authors additionally observe that patients with comorbid conditions had larger baseline burdens and may therefore exhibit larger absolute reductions in acute-care utilisation and costs following ESK initiation. The paper lists several limitations acknowledged by the investigators. The available data covered only about 20 months of ESK use in the US and the analytic sample—especially some comorbidity subgroups—was small; a 6‑month follow-up window was used to preserve sample size but longer follow-up with more recent data is needed to confirm trends. TRD was identified using pharmacy claims rather than detailed clinical information, although prior authorisation requirements for ESK make misclassification less likely. Pharmacy claims do not guarantee medication ingestion and ESK pharmacy claim dates may not equal administration dates, so session timing may be misestimated. Results may not generalise to uninsured populations or those without commercial/Medicare supplemental coverage. The required 6‑month follow-up may introduce immortal time bias, and pre/post comparisons are susceptible to the influence of time-varying events on observed associations. The authors recommend further research to assess the clinical impact of real-world ESK dosing patterns and to explore strategies to improve adherence to recommended treatment schedules.

In this descriptive claims-based analysis, most patients with TRD initiated on esketamine intranasal spray—including those with cardiometabolic, pain, anxiety or substance use comorbidities—completed induction treatment, but commonly at intervals longer than the label recommendation, suggesting practical challenges in routinely maintaining appointments. Observed reductions in mental health-related inpatient and ED utilisation and costs after ESK initiation were accompanied by higher outpatient visits and costs, likely reflecting ESK administration; the authors call for additional research with larger samples and longer follow-up to confirm these findings and to evaluate the clinical implications of real-world dosing patterns.