(The) predictive model of depression in rural elderly
This descriptive study aims to develop a nursing protocol to prevent from and reduce depression experienced by elderly in rural areas and provide fundamental data on a nursing intervention program by investigating into their depression estimation factors and building and reviewing an optimal predictive model. Subjects in this study are elderly over 65 residing in 30 rural areas in Gyeongsangbuk-do and Gangwon-do. Data were collected from May 1, 2011 to July 30, 2011, and a total number of 461 interviews done by researchers and research assistants, excluding 39 insincere interviews from 500 interviews collected, were used as final subjects for an analysis. Depression for elderly residing in rural areas as a dependent variable was measured by using Cho et al. (1999)’s 15 questions from Short Form of Geriatric Depression Scale Korean version(SGDS-K), adopted from 15 questions from Short Form Geriatric Depression Scale(SFGDS) developed by Sheikh and Yesavage(1986). 33 chronic diseases were suggested as factors affecting depression and measured by asking prevalance and doctors’ diagnosis based on Korean Standard Statistical Classification of Diseases (National Statistical Office, 2010) by dividing into human circulatory organ system, endocrine system, musculoskeletal system, respiratory system, sensory system, digestive system, urogenital system, and malignant neoplasm. Cognitive function was measured by Lee et al.(2002)’s Mini-Mental Status Examination in the Korean Version of the CERAD Assessment Packet(MMSE-KC), which was adopted from Mini-Mental State Examination(MMSE) developed by Folstein and etc(1975). Self-worth was measured by Chun(1974)’s 10 questions and 4-grade measure, which were adopted from Rosenberg(1965). Exercise capacity and ability to take care of oneself were measured by using Lee Yoon-hwan and etc(2002)’s physical function measure developed to evaluate exercise capacity for seniors residing in rural areas. Participation in agricultural activities was measured through one occupational question on availability of working to make an earning at least for the last week and one question on occupational type suggested by Korean Standard Industrial Classification(2007, Korean Statistical Office), while participation in social activities were measured based on 2 questions on availability of participation in social gathering and degree of participation. Last, support exchange among friends or neighbors was measured by referring to 8 questions from 2008 Survey on Seniors(Korea Institute for Health and Social Affairs, 2009). Data collected were analyzed by using SPSS WIN 18 and a statistical program, real numbers and percentage were derived, and Chi-square Test was conducted for factorial differences between a depression group and a non-depression group. Also, to draw out an optimal model estimating depression for seniors residing in rural areas, an estimation model was established through Decision Tree Analysis employing SPSS Modeler 14.2 Statistical Program, and study results are as follows. 1. The degree of depression for seniors in rural areas as test subjects was an average 5.99 from scale 0∼15, which was not a very high score average for depression, but it was found that depression for 169 seniors(36.7%) with a score higher than 8 is a common health problem for them. 2. Depression for seniors residing in rural areas showed a significant difference in age, level of education, economic status, number of chronic diseases, cognitive function, self-worth, exercise capacity, ability to take care of oneself, participation in agricultural activities, and participation in social activities. In particular, depression between subjects with less than 2 chronic diseases and with more than 3 chronic diseases showed a significant difference, and type of chronic diseases between musculoskeletal, malignant neoplasm disease and sensory disease. 3. Exercise capacity, self-worth, economic status, participation in social activities, cognitive function, participation in agricultural activities, and sex were indicated as important factor for estimating depression among seniors residing in rural areas through a logistic regression analysis. 4. Taking a look at a depression estimation model for seniors residing in rural areas by using Decision Tree Analysis as a final method, exercise capacity, self-worth, participation in agricultural activities, participation in social activities, cognitive function, and sex were understood as estimation factors. Based on them, subjects’ depression estimation model was indicated through the following 4 courses. First are male subjects with exercise capacity under 35 and self-worth under 20. Second are female subjects with exercise capacity under 35 and self-worth under 20. Third are subjects with exercise capacity under 35 and self-worth over 20, and those not participating in agricultural activities. Fourth are subjects with exercise capacity over 35, self-worth under 20, and those with reduced cognitive function not participating in social activities. 5. An accuracy model for this study’s model is 78.59%, error rate 21.41%, sensitivity rate 53.33%, and specificity rate 93.16%. In conclusion, to prevent from depression experienced by seniors residing in rural areas, exercise capacity and ability to take care of oneself should be reinforced to increase activities and participation. This study’s depression estimation model for seniors living in rural areas will be expected to induce a theoretical development of a situation-producing theory that can apply to actual nursing practice sites. This, in turn, will be used as theoretical basis for developing a systematic knowledge system for nursing and for developing a protocol that prevents from depression for elderly living in rural areas, thereby contributing to advanced depression prevention for elderly.