응답 일치 여부와 면접 대상자의 인구학적 특성의 연관성: 2014년지역사회건강조사 자료 활용

Abstract

Objectives: The response concordance between interviewer-assisted interviews and telephone survey was introduced in the Community Health Survey for the quality control and examining the association between the response concordance and related variables is required. Methods: Total 22,665 were participated in the 2014 Community Health Survey, Korea. Multivariate logistic regression analysis was performed on the response concordance for the five questions; self-rated health status (SRH), car driving, sleep time, health check-ups, and doctors diagnosed hypertension (DDH). Response concordance was investigated by gender, age (19-39, 40-64, 65-74, ≥75), residential areas (urban, rural), comorbidity, and interval period (<6 days, 6 days ≤) defined as the different days between interviewer-assisted interviews and telephone survey. Results: Women (12,658, 55.9%), urban residents (15,486, 68.3%), age 40-64 years (10,732, 47.4%), and comorbidities (6,911, 30.5%) were predominant among study population. The association of SRH were observed for women [1.08 (95% confidence interval [CI], 1.02-1.14)] and age group 40-64 [0.79 (95% CI, 0.73-0.85)], 65-74 [0.60 (95% CI, 0.55-0.66)], ≥75 [0.43 (95% CI, 0.38-0.48)], rural [0.84 (95% CI, 0.79-0.89)], and comorbidities [0.90 (95% CI, 0.78-1.04)]. Gradual decrease in car driving was observed by the age groups 40-64 [1.43 (95% CI, 1.17-1.74)], 65-74 [1.78 (95% CI, 1.32-2.40)], and ≥75 [1.87 (95% CI, 1.29-2.70)]. The variables showing significant association with the interval period (6 days ≤) were SRH [0.88 (95% CI, 0.83-0.93)] and sleep time [0.85 (95% CI, 0.77-0.93)], and rural area residents showed lower response concordance with SRH [0.84 (95% CI, 0.79-0.89)], sleep time [0.64 (95% CI, 0.58-0.70)], health check-ups [0.90 (95% CI, 0.83- 0.99)], and DDH [0.71 (95% CI, 0.64-0.80)] except for car driving [1.21 (95% CI, 0.90-1.26)]. Conclusions: The factors associated with the response concordance varied and having comprehensive understanding about the survey is necessary to assess the quality of the data.