This research was conducted at North Celebes, Indonesia. Participants were recruited using the multi-stage sampling method, which began in July 2019. North Celebes Province was conveniently selected in the first stage sampling, which had the highest prevalence of gouty arthritis. North Celebes province is stratified it into 15 regions. In the second stage, two regions of fifteen regions were randomly selected, including urban and rural areas. Tomohon City is the one urban area that has been able to engage in this partnership. Seven Community Health Service (CHS) centers in Tomohon City have been randomly chosen in the final level. Four qualified professional nurses collected all the data. The data source of diagnoses was their medical records in the Public Community Health Services (CHS) of Tomohon City. The sample calculation utilized G-Power 3.1 software with the alpha level (α) of 0.01, power (1-ß) of 0.80, and the effect size is 0.9, the total estimated sample size required 31 in each group. An estimated 20% dropout rate increases; the sample size is 36 for each group.
Variable description 1.1.1. Participants Characteristic Blood pressure and BMI were examined in a standardized medical examination. Demographic data were collected by using a self-designed questionnaire. The questionnaire included the demographic characteristics of the participant, including age, gender, marital status, educational level, income, occupation, family gout history, gouty arthritis drug consumption, alcohol consumption, and smoking status. The translated food frequency questionnaire was employed with Cronbach's alpha as 0.95.
1.1.2. Pain The Visual Analog Scale (VAS) instrument has 10 centimeters length. The standard scale anchored by "no pain" marked as a score of 0 and the "worst imaginable pain" marked as a score of 10 cm. The higher score identifies greater pain intensity. The reliability test showed that VAS was good (r = 0.94, p = 0.001). The reliability test of the VAS instrument has been done in Indonesia with the Cronbach's alpha as 0.89.
1.1.3. Joint mobility The range of knee joint motion was calculated using an International Standard Orthopedic Measurement (ISOM) goniometric. Its method using the Sagittal Frontal Transverse Rotation (SFTR) system to an accuracy of 10. Goniometry provides an objective assessment tool to measure the range of motion. The validity and reliability of Goniometry are 0.98 and 0.97, respectively. The validity score of the Indonesian version is 0.97, and the Cronbach alpha as 0.51.
1.1.4. Stress, anxiety, depression The Depression Anxiety Stress Scale (DASS-21) questionnaire is used to measure the negative states of three mental health conditions: depression, anxiety, and stress using a self-report 4-point Likert scale. Data was collected by reflects the thoughts, feelings, and behavior. The result will be defined by the total scores of responses from the 7-item subscale of this questionnaire. The Cronbach's alphas of Indonesian translated is 0.72 to 0.87.
1.1.5. Quality of Life (QOL) 36 Health Survey (SF-36) questionnaire was employed to estimate QOL. It included Physical Component Summary (PCS), Mental Component Summary (MCS). Total QOL range from 0 to 100. The Indonesia version of the SF-36 questionnaire has tested the reliability with the Cronbach's alphas 0.7.
1.1.6. Physical activity (MET-h/week) The physical activity level was investigated as metabolic equivalent of task (MET)-hr/week by using three questions based on the modified Physical Activity Guideline from the Advisory Committee for Americans and the Godin Leisure-Time Exercise Questionnaire. The investigator then categorized participants based on the exercise type (that is divided into the following categories: mild (e.g., yoga, bowling, or floor-sweeping), moderate (e.g., gym, baseball, or badminton), and strenuous exercise (e.g., hiking, soccer, or running)), duration in minutes and exercise intensity they presented during a typical week. The investigator multiplied the amount of mild, moderate, and strenuous exercise hours and duration by 3, 5, and 9, respectively. By summing the exercise pattern of the separate things, the investigator measured total weekly PA in arbitrary units.
The per-protocol method was employed to deal with the missing data. Each participant is allowed to immediately report the adverse event during the intervention.
The Chi-square and t-test were employed to analyze the homogeneity of subject characteristics between the experimental and control group. The Generalized Estimating Equation (GEE) model was applied to compare the differences pain, joint mobility, stress, anxiety, depression, QOL (included PCS and MCS), physical activity (MET-h/week) within one-month intervention between the intervention and control group. The interaction between group and time was also further analyzed to determine if the mean differences in their pain, joint mobility, stress anxiety, and depression, QOL, and physical activity between groups differed significantly over time. p\<0.05 was applied as statistically significant. All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) vers. 25.0 (Chicago, IL, USA). Besides, Structural Equation Modeling (SEM) was also employed to analyze the mediating effect between variables through the Analysis of Moment Structures (AMOS) vers. 24. Then the investigator evaluate the measurement and structural models by utilizing the goodness fit index.
