Factorial Structure of the Patient Health Questionnaire-9 in Long-Term Survivors of Severe COVID-19

The COVID-19 pandemic continues to cast a long shadow, with a significant portion of survivors experiencing persistent health challenges, collectively known as Long COVID (LC). Among these lingering effects, depression has emerged as a prominent concern, yet robust psychometric evidence for screening instruments in this specific population, particularly within Latin America, remains scarce. A recent study published in Frontiers in Psychology sheds light on the validity of the widely used Patient Health Questionnaire-9 (PHQ-9) for assessing depression in individuals who have survived severe COVID-19, revealing a predominantly unidimensional structure of the scale in this group.

The research, conducted by a team of Colombian scientists, aimed to address the critical gap in understanding the psychometric properties of the PHQ-9 among long-term survivors of severe COVID-19. Their findings suggest that while the PHQ-9 can be a useful screening tool, its interpretation might need careful consideration, especially regarding the symptom of fatigue.

The Lingering Impact of Severe COVID-19
The initial wave of the COVID-19 pandemic, which began in late 2019 and escalated globally in early 2020, overwhelmed healthcare systems worldwide. Survivors of severe forms of the illness, often requiring hospitalization and intensive care, have since faced a spectrum of long-term health issues. These post-acute sequelae of SARS-CoV-2 infection, commonly referred to as Long COVID, can manifest in various ways, including neurological, respiratory, cardiovascular, and psychological symptoms.

Estimates indicate that a substantial percentage of COVID-19 patients experience depressive symptoms during follow-up, with some studies reporting persistence for up to two years post-infection. The complex interplay of biological, psychological, and social factors contributing to these neuropsychiatric manifestations is still being unraveled. Given this context, effective and validated screening tools are crucial for early identification and intervention.

The PHQ-9: A Widely Used Screening Tool
The Patient Health Questionnaire-9 (PHQ-9) is a nine-item self-report questionnaire widely adopted for screening and assessing the severity of depressive symptoms. It measures the frequency of nine diagnostic criteria for major depressive disorder over the past two weeks. Despite its common use, the factorial structure of the PHQ-9 has been a subject of debate, with studies supporting both unidimensional and two-dimensional models (typically separating somatic from cognitive-affective symptoms). The consistency of these findings across diverse populations, however, remains unclear.

Study Design and Methodology
The Colombian study enrolled 177 patients who had been hospitalized with severe COVID-19 at a tertiary care center. These individuals were followed for an average of 20 months post-discharge. The PHQ-9 was administered as part of a comprehensive battery of psychiatric assessments. The researchers employed both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) to evaluate competing structural models of the PHQ-9.

Key Findings: A Predominantly Unidimensional Structure
The analysis revealed that both one-factor and two-factor models of the PHQ-9 demonstrated acceptable statistical fit to the data. However, the internal consistency of the unidimensional model was superior, with a Cronbach’s alpha of 0.81 and McDonald’s omega of 0.86, compared to lower reliability estimates for the two-factor solution.

Crucially, when the researchers attempted to fit two-factor models based on existing theoretical frameworks, they found that the item loadings did not align coherently with established depression constructs. Furthermore, confirmatory factor analysis indicated that the two-factor models did not offer a significant improvement in fit over the simpler one-factor model. The interfactor correlations in the two-factor models were consistently high, approaching or exceeding unity, suggesting that the somatic and cognitive-affective dimensions were not empirically separable in this cohort of long-term severe COVID-19 survivors.

The Fatigue Item: A Notable Observation
A particularly interesting finding was the consistently lower factor loading of the "fatigue" item (item 4) across both exploratory and confirmatory analyses. This item exhibited the lowest loading in the unidimensional model and a weak loading in the two-factor models, while the scale’s internal consistency improved when this item was removed.

This observation is significant because fatigue is a hallmark symptom of Long COVID, particularly in individuals who experienced severe illness and intensive care unit (ICU) stays. The study noted that 57.6% of their cohort required ICU admission. The persistence of fatigue for up to 20 months post-infection, as reported in other studies, could lead to an overlap with depressive symptomatology. The PHQ-9 item on fatigue might, therefore, capture aspects of Long COVID-related fatigue that are distinct from core depressive symptoms, complicating its interpretation within a purely depression-focused framework.

Implications for Clinical Practice and Future Research
The study’s findings have important implications for how the PHQ-9 is used in the follow-up of severe COVID-19 survivors, especially in Latin America, where mental health screening can face unique challenges related to stigma, access to care, and socioeconomic inequities.

The confirmation of a predominantly unidimensional structure suggests that the PHQ-9 can be interpreted as measuring a general level of depressive symptomatology in this population. This aligns with the practical needs of screening in resource-limited settings. However, the weak performance of the fatigue item underscores the need for caution. Clinicians using the PHQ-9 should be aware that a high score on this item might reflect not only depression but also the pervasive fatigue associated with Long COVID.

The researchers emphasize the need for future criterion validation studies. This would involve comparing PHQ-9 scores against a structured clinical psychiatric assessment, considered the gold standard for diagnosing depression. Such validation could help determine if the current PHQ-9 cut-off scores, commonly used for screening and severity classification, are appropriate for this specific patient group or if recalibration is necessary. The study also highlighted that women in their cohort were more likely to screen positive for depressive symptoms at both commonly used thresholds (≥7 and ≥10), suggesting potential sex-based differences in symptom presentation or screening tool performance that warrant further investigation.

Addressing a Critical Knowledge Gap
This research contributes to a growing body of literature on the long-term health consequences of COVID-19. By focusing on a Latin American cohort, it addresses a critical gap, as much of the existing research on Long COVID psychometrics has originated from North America and Europe. The study’s authors noted that cultural and structural factors in Latin America can influence both symptom expression and the application of screening tools, making context-specific validation essential.

Limitations and Future Directions
The study acknowledges several limitations, including the absence of a structured clinical psychiatric interview to serve as a gold standard for diagnostic accuracy, which restricts the derivation of population-specific cutoffs. Furthermore, the study was conducted at a single center, potentially introducing survivorship and participation biases. The sample size, while adequate for the tested unidimensional model, might have been insufficient to detect subtle differences between competing models or to perform robust multi-group analyses, such as testing for factorial invariance across sex. Future research with larger, multi-center cohorts is needed to confirm these findings and explore the complex interplay of Long COVID symptoms, pre-existing vulnerabilities, and mental health outcomes.

Conclusion
In conclusion, the study by Rivas and colleagues provides valuable psychometric evidence for the PHQ-9 in long-term survivors of severe COVID-19 in Colombia. The findings support a predominantly unidimensional interpretation of the scale, offering a parsimonious approach to screening for depressive symptoms in this population. However, the observed weakness of the fatigue item signals a potential overlap with Long COVID-related fatigue, underscoring the importance of careful interpretation and the urgent need for further validation against clinical diagnostic standards. This research is a significant step towards improving the accurate identification and management of mental health challenges faced by individuals recovering from severe COVID-19.