|
Back to 2017 Posters
CLASSIFICATION OF PATIENTS WITH SYMPTOMS OF GASTROESOPHAGEAL REFLUX DISEASE USING LATENT CLASS ANALYSIS
Harish G. Medapati*, Nandakishore Maroju
Surgery, Jawaharlal Institute of Postgraduate Medical Education & Research, PONDICHERRY, PUDUCHERRY, India
Aims and Objectives:
To identify distinct classes of Gastro Esophageal Reflux DISEASE (GERD) based on patient symptom profile using latent class analysis and to define the association of GERD symptoms with patient characteristics.
Materials & Methods:
Its an observational study. Data from 1050 patients attending surgery and medicine outpatient departments with complaints of atleast one typical symptom (Heartburn/Regurgitation) with in one week of presentation were interviewed about their baseline data (Age, Gender, Demographic data, Physical data and substance abuse history) and symptomatology (both typical and atypical symptoms). Patients with only atypical symptoms or with ulcer disease were excluded. Sample size was calculated using the formula 2n where n is number of variables studied. A total of 10 variable symptoms which includes both typical and atypical were studied. These variables include daytime heartburn and regurgitation, night time heartburn and regurgitation, Nausea related symptoms( nausea, vomiting and eructation) , Early satietry and dysphagia, respiratory symptoms( cough and breathlessness), Atypical precordial pain, ENT symptoms (throat pain, hoarseness), and Sleep disturbances ( difficulty in falling asleep, Nightmares, Nocturnal awakening). Statistical analysis was done by performing a factor analysis with principle component to extract latent classes. The significance between symptoms and baseline data, between the symptoms and baseline data was calculated by chi square test.
Results:
It was found that GERD can be grouped under 4 LATENT CLASSES which include Class 1 (Typical GERD symptoms), Class 2 (Nausea associated symptoms), Class 3 (Laryngeal symptoms), Class 3 ( Dysphagia-Chest pain complex) From baseline data it was found that GERD is more prevalent in the urban population and is associated with alcohol intake among male patients. It was found that in relation with gender, the symptoms like night time regurgitation, breathlessness and atypical precordial pain were more significant in the male gender and nausea in female gender. It was also found that patients who have history of alcohol consumption have higher association with symptoms like Regurgitation and Sleep disturbances.
Conclusions:
Presentation of GERD in our geographical region is different from the published literature from other regions of the world. Latent class analysis can classify patients with symptoms of GERD. Further studies are required to assess the reproducibility of this classification and also to find out the contribution of these classifications in the management and follow-up of the patients with GERD.
Back to 2017 Posters
|
