| Fall is a common and severe problem for the elder population. The occurrence of fall, injury and economic burden caused by fall is growing with the age increased in old people over 65 years old. Fall in inpatients is also the most common adverse events, and accounts for 38% of adverse events occurred in hospital. Foreign reports showed that the incidence of fall was 2.3-7.0 per 1,000 in elderly inpatients. Due to illness, change of environment and caregivers, fall risk factors increased. Once fall took place, patients suffering increased, also the patient's recovery delayed, the hospitalization expense increased, and might also bring with legal problems. Therefore, fall prevention has become more and more important to medical personnel. Fall intervention measures include the patient and family education, the environmental safety guarantee, drug administration,and nursing guardianship. However, the first step is to identify the risk of falling. According to the clinical nursing documents code by Guangdong Health Department, Fall risk assessment Scale (Chinese version Morse Fall assessment Scale, MFS) is used as fall risk assessment Scale for residents of elderly patients. But in literature the domestic used status of MFS was not reported. In clinics, each of fall risk assessment Scale has its own characteristics. Before the application in different countries, different RACES or different people, the reliability and validity must be tested. This research aims at introducing the original license under MFS, performing investigation in Guangzhou to the elderly patients hospitalized for investigation and fall history, and factors affecting the use of MFS, discusses its feasibility and validity of the application in our country, proves the correct choice of fall risk assessment tool for clinical nurse, and provides effective prevent fall prevention and intervention for elderly patients.This research was a cross-section investigation. The questionnaire was adopted for investigation and assessment.From August to December in 2009, with convenient sampling method,325 cases of elderly patients were selected from three 3A hospital in Guangzhou. Inpatient Areas included acute and chronic ward, surgical ward, medical ward and high-ranking official ward.319 valid questionnaires were collected from 163 males and 156 females, of all with the Han nationality and with age 60-100 years old, average 79.06±7.3 years. Cases were selected with over 60 years old, conscious, able to walk (including the use of auxiliary appliance, without obvious physiological walking, without defect or major disease, without understanding disorders, completed action as ordered and informed consent).Scales selected according to the research purpose:Chinese version Morse Scale Fall Assessment Scale (MFS), Berg Balance Scale (BBS), Bathel Ability Index Scale used to assess the daily activities (BI), Timing Stand-Walk Test used to assess activity ability (TUGT), and simple intelligence test scale (AMT).This study was done by two registered nurses of our Rehabilitation Department. Training for the nurses was done before tests were performed. No difference could be found in the understanding and test results of 35 patients in forecasting try, then the research was formally started.3 months after the first testing, the fall incidence of patients was inquired by telephone, whether lost to follow-up or death within the three months should be eliminated.Statistical analysis methods included:according to the fall incidence in nearly three months, patients were divided into two groups, x2 analysis was adopted to show the differences of fall incidence between the two groups. Independent-Samples T test was applied to test the score difference between TUGT, AMT, MFS, BBS, BI scales. Logistic regression method using binary classification analysis was used to analyze the factors affecting patient falls. Cronbach's Alpha coefficient was used to evaluate the reliability of MFS. Pearson correlation analysis and two related sample non-parameter test analysis were applied to show the reliability between the testers. Independent-Samples T test was adopted to analyze the validity of MFS. Content validity was showed by the scale total score and correlation analysis. With sensitivity and specificity of the ROC curves, MFS analysis and prediction effect was presented. All data analysis was performed using the statistical analysis software SPSS13.0, and a=0.05 was considered as significant level.1. The initial investigation cases were 321. Three months later, inquiry by telephone,2 cases died because of illness, the effective rate of the study is 99.4%. Acceptance of the cases'scale is high.2. Fall history survey for senile inpatients and the effect of society, illness factors. Urinary incontinence or frequency, use of auxiliary appliance, professional and cultural level before retiring, care of life were statistically significant between the two groups with or without fall history in nearly three months (P< 0.05). Higher incidence of falls was seen in the patients with stress urinary incontinence or frequency symptoms than those without symptoms. Higher incidence of falls was also seen in the patients using auxiliary walkers than those not using. Fall incidence rates differed according to professional before retiring:highest in knowledge workers (39.3%), followed by senior cadres (33.6%), farmers (27.3%) and workers (17.3%). Due to the different culture levels, fall incidence rates were also different:for university degree (52.4%), followed by high school or technical school education (41.7%), primary education (25.4%), illiteracy (27.7%, junior high school (18.4%). For the different life care:highest fall incidence rate for nursing care (40.6%), followed by children care (28%), by themselves (21.2%), by spouse (17.9%). In the use of TUGT, MFS, BBS, BI scale, scores of the balance of patient, evaluation of daily activities, sit-stank ability, fall history, disease factors in the two groups was statistically significant (P< 0.05), while cognitive aspects were not statistically significant. Three variables included educational level, used auxiliary appliance, and MFS scores were selected into the Logistic regresstion model classification. x2=137.101, P=0.000, the logistic formaula is statistically significant (P< 0.05). R2=0.804. It suggested that the 3 factors accounted for 80.4% of all the reasons leading to fall incidence of patients.3. The reliability evaluation between the two testers.35 patients were evaluated by two testers with MFS. The rank correlation coefficient between the testers was over 0.7 (P< 0.01), showing that the testing results of two testers were closely related with a statistical significance. By two related sample non-parameter test, scale scores of two testers were analyzed, and no significant differences were found between the two testers (P> 0.05).4. The internal consistency of the scale. Cronbach's Alpha was 0.102, Cronbach's correction coefficient was 0.206. It was found that the removal of intravenous infusion or the heparin lock use entry, Cronbach's coefficient of the scale increased to 0.444.5. Content validity. Content validity was tested with related analysis of the total score of six items. The results showed that the correlation coefficient between the total scores of 6 entries fluctuated between 0.153-0.624 (P< 0.01), with the most closely related entry as the use of auxiliary appliance,0.624, followed by walking gait 0.615, and the lowest correlation coefficient for intravenous infusion or the use of heparin.6. Discrimination validity. Using the falling history in nearly a year as grouping variable, the MFS total and each item scores of two groups were analyzed by Independent-Samples T test. Except the item of more than one medical diagnosis (no significant differences, P=0.566), the scores of each other item and total entries all showed significant difference (P< 0.01).7. Forecasting effect. In the 319 cases of elderly patients,68 cases had a fall incidence within three months after the initial evaluation, fall occurrence rate was 21.32%. The sensitivity and specificity of diagnosis in the source scale were 98.5% and 40.2% respectively when the boundary value recommended as 45 points. In our research, the boundary value was recommended as 55 points, and the sensitivity and specificity were 86.8% and 67.3% respectively.1. Three variables including educational level, use of auxiliary appliance, and MFS score were selected into the Logistic egression model classification. x2=137.101, P=0.000, the logistic formaula is statistically significant (P< 0.05). R2=0.804. It suggested that the 3 factors accounted for 80.4% of all the reasons leading to fall incidence of patients. BBS or BI scale low score may be indirect influencing factors. It's very important to evaluate the fall history and fall risk for elderly patients when they are admitted in hospital, which helps nurse to adopt interventions and prevent fall incidence.2. MFS is simple and less time-consuming, and easily accepted by testers and senile patients. The reliability evaluation between the two testers is better. Discrimination validity and forecasting effect for the two groups (with or without fall history) are also good. But the internal consistency of the scale is low, and the content validity is unbalanced.3. It's suggested that the scale be modified according the characteristics of senile patients in our country. It will make the scale better for clinical nurse to prevent falling of patients.
|
PDF Full Text Request