Adherence to Antipsychotics and Job Acquisition in Patients with First-episode Schizophrenia: A 10-year Follow-up Using the Korean National Health Information Database

Technical Summary: Adherence to Antipsychotics and Job Acquisition in First-Episode Schizophrenia

Problem Statement
Schizophrenia significantly impairs social and occupational functioning, with global unemployment rates among affected individuals ranging from 70% to 90%. While existing literature suggests that antipsychotic medication adherence improves clinical outcomes (e.g., symptom remission, reduced relapse) and behavioral metrics, the specific impact of adherence on subsequent job acquisition remains uncertain. Previous studies have yielded mixed results; for instance, a prospective cohort study by Ascher-Svanum et al. (2006) did not find a statistically significant association between adherence status and unemployment, though the directionality suggested potential harm from non-adherence. A critical gap exists in the literature regarding the prospective relationship between adherence and incident job acquisition (hiring) specifically among individuals who are unemployed at the time of diagnosis, rather than analyzing changes in employment prevalence across mixed cohorts.

Methodology
This study utilized a 10-year longitudinal cohort (January 1, 2014 – December 31, 2023) drawn from the Korean National Health Information Database (NHID), which covers the entire Korean population.

• Cohort Selection: The study identified 6,499 individuals aged 25–54 years diagnosed with first-episode schizophrenia (ICD-10 codes F20–F29) who were non-employed at the index date. Individuals with prior diagnoses of organic mental disorders, substance use disorders, or those who had been prescribed antipsychotics or diagnosed with schizophrenia in the three years preceding the index date were excluded.
• Exposure Definition: Adherence to antipsychotics was measured using the Proportion of Days Covered (PDC). Participants were classified as having "high adherence" (PDC ≥ 0.8) or "low adherence" (PDC < 0.8). Dose intensity was calculated using Defined Daily Doses (DDD).
• Outcome Definition: The primary outcome was job acquisition, operationalized as the transition to "employee-insured" status in the National Health Insurance system.
• Statistical Analysis: The researchers employed multivariable Cox proportional hazards models to estimate the association between adherence levels and time-to-employment. The models were adjusted for age, sex, income quintile, prior job experience, clinical severity (hospitalization status and length of stay), antipsychotic dose intensity, and concurrent use of other psychotropic medications (antiepileptics, antidepressants, etc.). Subgroup and interaction analyses were conducted to examine effect modification by demographic and clinical factors.

Key Results

• Cohort Characteristics: Among the 6,499 participants (mean age 39.6 years; 56.0% female), 2,961 were classified as high-adherence and 3,538 as low-adherence. Over the follow-up period, 44.8% of the high-adherence group and 38.0% of the low-adherence group secured employment.
• Primary Association: High adherence to antipsychotics was associated with a significantly greater likelihood of job acquisition compared to low adherence. In the fully adjusted Model 3, the Adjusted Hazard Ratio (AHR) was 1.17 (95% CI, 1.07–1.29).
• Subgroup Analysis: A statistically significant interaction was found by sex. The association was stronger in men (AHR, 1.29; 95% CI, 1.13–1.46) than in women (AHR, 1.07; 95% CI, 0.96–1.20). However, no significant heterogeneity was observed across age groups, socioeconomic status (income), or antipsychotic dose intensity categories.
• Other Factors: Prior job experience was a strong predictor of employment (AHR ~2.46), while longer hospital stays (≥90 days) were associated with reduced likelihood of job acquisition.

Key Contributions

1. Incidence-Based Framework: Unlike previous studies that analyzed changes in employment prevalence across mixed cohorts, this study focused exclusively on individuals unemployed at baseline, utilizing a time-to-event (incidence) approach to assess job acquisition.
2. Large-Scale National Data: The study leveraged a nationally representative sample from the Korean NHID, providing robust statistical power and generalizability within the Korean context.
3. Comprehensive Covariate Adjustment: The analysis controlled for a wide range of confounders, including disease severity indicators (hospitalization), dose intensity, and concurrent polypharmacy, offering a more isolated view of the adherence-employment relationship.
4. Sex-Specific Insights: The study identified that the beneficial effect of adherence on employment is more pronounced in men, a nuance not previously highlighted in this specific context.

Significance and Claims
The authors claim that their findings provide evidence supporting the integration of adherence-focused pharmacologic management with tailored vocational support to promote functional recovery in schizophrenia. The study suggests that maintaining high adherence to antipsychotics is beneficial for job attainment among individuals with first-episode schizophrenia, independent of sociodemographic and medical conditions.

The authors explicitly state that this is the first study to document such findings using an incidence-framework among jobless individuals at study entry based on a nationally representative Korean sample. They conclude that these results inform public health and clinical approaches, suggesting that education and support programs aimed at improving medication adherence could empower this marginalized population to better function as members of society.

Limitations Acknowledged
The paper notes several limitations: adherence was calculated based on prescription records rather than actual intake; disease severity could not be measured directly due to the nature of claims data; the exploration of polypharmacy was limited; and the observational design precludes definitive causal inference due to potential unmeasured confounding.