Exploratory Factor Analysis for Validating Traditional Chinese Syndrome Patterns of Chronic Atrophic Gastritis

Yin Zhang, A X Zhou, Yue Liu, Ying Zhao, Li Zhang, Leilei Sun, Shiyu Du, Qiang Yang, Xin Song, Chaoyang Liang, Xia DingEvidence-based 2016 · DOI

TL;DR

This study uses Exploratory Factor Analysis (EFA) to identify and validate six core Traditional Chinese Medicine (TCM) syndrome patterns associated with Chronic Atrophic Gastritis (CAG). By analyzing diagnostic data from over 100 patients, the research provides a data-driven framework for understanding the complex pathogenesis of CAG within the TCM paradigm, aiming to optimize treatment strategies.

What Problem Does This Paper Address?

Chronic Atrophic Gastritis (CAG) is a significant precursor to gastric cancer, and Traditional Chinese Medicine (TCM) has a long history of treating it. However, a lack of consensus exists regarding the specific TCM syndrome patterns and core pathogenesis of CAG across different studies. This ambiguity hinders the standardization of diagnosis and the optimization of treatment strategies. This paper aims to address this by applying a statistical method, Exploratory Factor Analysis (EFA), to systematically identify and validate the key TCM syndrome patterns associated with CAG, thereby providing a more robust foundation for clinical practice and research.

Methods

The researchers employed Exploratory Factor Analysis (EFA), a statistical technique used for dimensionality reduction and identifying underlying latent variables. The process involved:

  1. Data Collection: Information was gathered from over 100 CAG patients across four hospitals in China. This data included general patient information, disease history, and crucially, findings from the four diagnostic methods of TCM: looking (observation), listening/smelling, asking (symptom inquiry), and palpating (pulse diagnosis).
  2. Variable Selection: High-frequency diagnostic variables (those appearing in at least 10% of cases) were selected, and then further refined to 40 variables with a frequency of over 21% for EFA.
  3. Suitability Testing: The Kaiser-Meyer-Olkin (KMO) test and Bartlett's test of sphericity were used to ensure the data was suitable for factor analysis.
  4. Factor Extraction and Rotation: Principal component analysis was used to extract common factors. A varimax rotation was then applied to simplify the factor structure and make the interpretation of factor loadings clearer.
  5. Syndrome Pattern Identification: The extracted common factors were then interpreted by TCM gastroenterology experts and combined to establish six distinct syndrome patterns.

Datasets

The study utilized a dataset of 120 participants diagnosed with Chronic Atrophic Gastritis (CAG) who underwent gastrointestinal endoscopy. The data was collected through participant surveys conducted between September 2010 and October 2012 across four medical centers in Beijing and Xiamen, China. No new benchmark datasets were introduced; the study relied on collected clinical data.

Key Insights and Ideas

  • Data-Driven Syndrome Validation: The study demonstrates the power of EFA in moving beyond anecdotal evidence to statistically validate TCM syndrome patterns for a specific disease (CAG). This provides a more objective basis for diagnostic frameworks.
  • Core Pathogenesis Identified: The research pinpoints a combination of six key TCM syndrome patterns as the core pathogenesis of CAG: qi deficiency, qi stagnation, blood stasis, phlegm turbidity, heat, and yang deficiency. This offers a comprehensive view of the disease's underlying mechanisms in TCM.
  • Strategic Treatment Implications: The identified syndrome patterns directly inform treatment strategies. The findings suggest that while all six patterns are important, qi deficiency, qi stagnation, phlegm turbidity, and heat are more prevalent. Yang deficiency, though less frequent overall, can be intense when present. This nuanced understanding can guide the prioritization of therapeutic approaches.
  • Organ Involvement: The study also links these syndrome patterns to specific organ locations (liver, spleen, stomach), further refining the diagnostic and therapeutic approach within TCM.

Key Results

  • Six Syndrome Patterns: EFA successfully extracted six core syndrome patterns for CAG: qi deficiency, qi stagnation, blood stasis, phlegm turbidity, heat, and yang deficiency.
  • Prevalence of Patterns:
    • When considering all syndrome patterns, qi deficiency, qi stagnation, blood stasis, phlegm turbidity, and heat were highly prevalent (76.7%-84.2%), while yang deficiency was less so (42.5%).
  • However, when looking at primary syndrome patterns, qi deficiency, qi stagnation, phlegm turbidity, heat, and yang deficiency showed higher frequencies (15.8%-20.8%) compared to blood stasis (8.3%).
  • Core Pathogenesis: The study concludes that the core pathogenesis of CAG is a combination of these six patterns.
  • Treatment Recommendations: Based on these findings, TCM treatment strategies for CAG should focus on regulating qi, activating blood, resolving turbidity, clearing heat, removing toxins, and warming yang.

Limitations and Future Work

  • Selection Bias: The authors acknowledge that the study's reliance on participants from hospitals in only two cities might introduce selection bias, limiting the generalizability of the findings to CAG patients across all of China.
  • Data Mining Limitations: EFA, while powerful, has limitations in capturing complex multicollinearity between variables. The "dimension effect" might lead to a loss of information.
  • Need for Validation: While the statistical tests (KMO and Bartlett's) indicated EFA was acceptable, the authors suggest the results need validation with other supervised data mining methods.
  • Future Work: The authors recommend future multicentered studies with larger sample sizes to verify these conclusions and further explore the complex relationships within TCM diagnostic variables. They also suggest exploring other data mining techniques to gain a more comprehensive understanding of the disease and its syndrome patterns.

Summary updated

Abstract

<i>Background.</i> Traditional Chinese medicine (TCM) has long been used to treat chronic atrophic gastritis (CAG). The aim of the present study was to evaluate the TCM syndrome characteristics of CAG and its core pathogenesis so as to promote optimization of treatment strategies. <i>Methods.</i> This study was based on a participant survey conducted in 4 hospitals in China. Patients diagnosed with CAG were recruited by simple random sampling. Exploratory factor analysis (EFA) was conducted on syndrome extraction. <i>Results.</i> Common factors extracted were assigned to six syndrome patterns: <i>qi</i> deficiency, <i>qi</i> stagnation, blood stasis, phlegm turbidity, heat, and <i>yang</i> deficiency. Distribution frequency of all syndrome patterns showed that <i>qi</i> deficiency, <i>qi</i> stagnation, blood stasis, phlegm turbidity, and heat excess were higher (76.7%-84.2%) compared with <i>yang</i> deficiency (42.5%). Distribution of main syndrome patterns showed that frequencies of <i>qi</i> deficiency, <i>qi</i> stagnation, phlegm turbidity, heat, and <i>yang</i> deficiency were higher (15.8%-20.8%) compared with blood stasis (8.3%). <i>Conclusions.</i> The core pathogenesis of CAG is combination of <i>qi</i> deficiency, <i>qi</i> stagnation, blood stasis, phlegm turbidity, heat, and <i>yang</i> deficiency. Therefore, treatment strategy of herbal prescriptions for CAG should include herbs that regulate <i>qi</i>, activate blood, resolve turbidity, clear heat, remove toxin, and warm yang.
BibTeX
@inproceedings{zhang2016exploratory,
  author = {Yin Zhang and A X Zhou and Yue Liu and Ying Zhao and Li Zhang and Leilei Sun and Shiyu Du and Qiang Yang and Xin Song and Chaoyang Liang and Xia Ding},
  title = {{Exploratory Factor Analysis for Validating Traditional Chinese Syndrome Patterns of Chronic Atrophic Gastritis}},
  booktitle = {Evidence-based},
  year = {2016},
  doi = {10.1155/2016/6872890},
}
Figure