| BACKGROUND AND OBJECTIVEFall is one of the most common adverse events in hospitalized patients,which may lead to adverse outcomes for the patients,prolonged hospital stays,and increased medical expenses.Health professionals should identify patients at risk of falling and take effective interventions to prevent patients from falling.A fall risk assessment is an effective way to assess and identify people at high risk of falling.The Morse Fall Scale(MFS)is the most widely used fall assessment scale in China.However,due to the heterogeneous clinical characteristics of hospitalized patients,falls often occur in hospitalized patients who have been classified as with a low or moderate risk of falling,suggesting that the MFS may be not sensitive enough to predict the risk of fall in hospitalized patients.Therefore,this study aimed to(1)retrospectively analyze the characteristics of patients who fell during hospitalization in tertiary hospitals,(2)test the applicability of the MFS in hospitalized patients using statistical analysis based on the classical test theory and the item response theory,and(3)explore risk factors for falls in gastroenterology inpatients and the predictive efficacy of the Morse Fall Scale in gastroenterology inpatients by conducting a case-control study.METHODSA retrospective analysis of 200 hospitalized patients falling from January 2012 to September 2017 in a large size Tertiary comprehensive hospital in Guangzhou was conducted.The information about the fall event and the Morse Fall Scale score before the patient fell were analyzed and statistical analysis was performed.The reliability and construct validity of the MFS were analyzed by classical test theory,and the reliability and construct validity of the MFS were tested by Rasch analysis based on item response theory.A case-control design was adopted.Thirty gastroenterological inpatients who fell from May 2017 to December 2020 in a tertiary hospital were set as a case group.Ninety non-fall patients who were admitted to the same department during the same period were set as a control group.Two groups were matched in a 1:3 ratio based on age,gender,and main diagnosis.Multivariate logistic regression analysis was used to screen out the risk factors for falls in inpatients in the gastroenterology department,and the predictive efficacy of MFS and total risk factor score for falls in gastroenterology department patients were tested by sensitivity,specificity,the area under the curve of ROC.RESULTS(1)During the study period,a total of 200 cases of inpatient falls occurred,and the average age of the fall patients was 61.63±16.86 years(range:16-93 years).96(48%)fall patients were younger than 65 years,65(32.5%)were at low risk by MFS,62(31.0%)were at intermediate risk,and 73(36.5%)were at high risk.The first diagnosis of fall patients was mostly tumor(35%),followed by circulatory system disease(23.5%).With MFS score of 45 as the cut-off value,the patients with falls were divided into low-and medium-risk groups and high-risk groups,and the item responses of the two groups were compared.There was no difference between the two groups only in the item score of "whether there is more than one medical diagnosis"(P>0.05),and the scores of other items were significantly different(P<0.001).59.8%of the patients in the low-and medium-risk group and 76.7%of the patients in the high-risk group used 2 or more fall-related high-risk drugs 24 hours before the occurrence of falls,and the difference between the groups was statistically significant(χ2=5.942,P=0.049);the proportion of patients in the low-and medium-risk group with activity restricted to a bed/chair(7.9%)was significantly lower than that in the high-risk group(23.3%;χ2=11.098,P=0.004).The high-incidence time period for falls was the night shift(22:00-8:00;low-and medium-risk group:40.9%;high-risk group:49.3%;P>0.05),and the high-incidence place for falls was the ward(low-and medium-risk group:51.2%;high-risk group:52.0%;P>0.05).Overall,57%of the patients suffered from different degrees of damage after falling,and there was no difference between the groups in terms of damage after falls in different risk groups(P>0.05).(2)The psychometric test of MFS based on the classical test theory and item response theory shows that:The Cronbach’s a of the MFS is 0.390,and each item of the scale has a low to moderate correlation with the total score.The most relevant correlation is "intravenous infusion and indwelling needle",which is 0.697.The scale extracts a total of three common factors.Rasch analysis:The first dimension of the scale explained 44.8%of the variation,and the second dimension explained 13.4%of the variation,with a eigenvalue ratio of 4.87/1.45.The Person Reliability in hospitalized falls and Item Reliability of the MFS were 0.23 and 0.98;the Person Seperation and Item Seperation are 0.54 and 7.11.The Item-Person diagram indicates that the items of the scale are in a mismatched state with the subjects.In addition,the threshold spacing of most items of this scale is less than 1.4 Logits,suggesting that item options are not discriminative enough.(3)A case-control study of inpatients in gastroenterology showed that:Diagnostic accuracy of the MFS for predicting falls in gastroenterology inpatients was low(sensitivity 0.667,specificity 0.578,AUC 0.652,Youden index 0.245).Multivariate logistic regression analysis showed that hyponatremia(OR=18.907,95%CI 2.929-122.032),hypokalemia(OR=5.322,95%CI 1.340-21.135),use of two or more fall risk-increasing drugs 24h before falls(OR=7.105,95%CI 2.002-25.215)and systolic blood pressure<110 mmHg(OR=4.748,95%CI 1.374-16.406),Moderate to severe low hemoglobin(OR=3.378,95%CI 0.816-13.990)were risk factors of falls in the patients.The total score of risk factors predicted falls with a sensitivity0.800,specificity 0.844,Youden index 0.644,AUC 0.871 in gastroenterology patients.There was a statistical difference between the two AUCs(Z=3.182,P<0.001).CONCLUSION(1)This study found that more than half of hospitalized fall patients were rated as low-to-moderate risk of falls by MFS,nearly half of hospitalized patients with falls were younger than 65 years old,and falls were also high in non-elderly hospitalized patients.The high incidence of falls is at night,and the common places are wards and toilets.The main causes of falls are unstable center of gravity,dizziness/vertigo,general or lower limb weakness,and environmental factors.Clinical medical staff should pay attention to improving the fall risk assessment strategy for inpatients,further optimize the fall risk assessment tool for inpatients,and pay attention to the high frequency period,place and potential causes of falls.(2)The internal consistency of the MFS was poor,and the items of the scale did not match well with the characteristics of inpatient falls.In addition,the threshold spacing of most item options is too narrow,and the ability to discriminate subjects is weak.It is recommended that researchers revise fall risk assessment tools that are more suitable for hospitalized patients according to the characteristics of inpatient falls.(3)Hyponatremia,hypokalemia,concomitant use of fall risk-increasing drugs and low systolic blood pressure,Moderate to severe low hemoglobin were risk factors for falls in gastroenterological patients.The MFS has a low diagnostic accuracy in predicting fall risks in gastroenterological patients.Further revision of the scale is needed for gastroenterological patients. |
PDF Full Text Request