Identifying factors associated with post-hospital falls in older patients: a territory-wide cohort study.

OBJECTIVES: Post-hospital falls impose a substantial healthcare burden on older adults, yet contributing factors remain inadequately examined. This study aimed to investigate underinvestigated factors associated with post-hospital falls. STUDY DESIGN: Retrospective territory-wide cohort study. METHODS: We examined the electronic medical records of patients aged ≥65 who were discharged from public hospitals in Hong Kong (2007-2018). During the 12 months following discharge, participants were monitored to identify falls based on diagnosis codes or clinical notes from inpatient episodes, the emergency department (ED) visits, and death records. Falls were categorized into two groups: those only requiring ED visits and those requiring hospitalizations. Binary logistic and multinomial logistic regressions examined the associated factors for post-hospital falls and subcategories of falls, respectively. RESULTS: Among 606,392 older patients, 28,593 (4.71%; 95% CI = 4.66%-4.77%) experienced falls within 12 months after discharge. Of those, 8438 (29.5%) only required ED visits, and 20,147 (70.5%) required hospitalizations. Discharge from non-surgical wards, length of stay over two weeks, receiving the Geriatric Day Hospital and Rehabilitation Day Program, advancing age, being female, having more comorbidities, taking more fall risk increasing drugs, previous admission for falls, and living in Hong Kong Island were associated with increased fall risk. Receiving allied health service or nurse service was associated with reduced risk. The same factors were more associated with falls requiring hospitalizations rather than falls only requiring ED visits. CONCLUSIONS: Older patients with identified factors were particularly vulnerable to post-hospital falls leading to rehospitalizations. Fall risk assessment and tailored prevention should prioritize this group.

Investigating Risk Factors for Falls among Community-Dwelling Older Adults According to WHO's Risk Factor Model for Falls.

OBJECTIVES: The World Health Organization developed the Risk Factor Model for Falls to describe fall risks in a comprehensive manner. However, there was a lack of study adopting such framework in quantifying falls risk from different factors in a single model. Therefore, this study examined the risk factors from four domains in the Risk Factor Model for Falls among older adults. DESIGN: Secondary data analysis of 10-year assessment records of the Minimum Data Set-Home Care instrument. SETTING: Hong Kong. PARTICIPANTS: 89,100 community-dwelling adults aged 65 and over who first applied for publicly funded long-term care services from 2005 to 2014. MEASUREMENTS: The Minimum Data Set-Home Care instrument was used to ascertain older adults' care needs and match them with appropriate services. Additionally, meteorological records from the same period were extracted from the Hong Kong Observatory. The logistic regression model was used to examine risk factors and their associations with falls. RESULTS: In total 70 factors were included in the analysis, of which 37 were significantly associated with falls. Behavioral risk factors generally had greater odds ratios of falling, as compared with biological, socioeconomic, and environmental factors. Out of all significant factors, functional status, alcohol drinking, and locomotion outdoors had the largest odds ratios of falling. CONCLUSION: Behavioral risk factors for falls are of remarkable influence yet are modifiable among older adults. Hence, falls prevention programs may need to prioritize addressing these factors.

Role of insulin in the anti-apoptotic effect of glucose-insulin-potassium in rabbits with acute myocardial ischemia and reperfusion.

OBJECTIVE: To study the effects of glucose-insulin-potassium (GIK) cocktail on cardiac myocyte apoptosis and cardiac functional recovery following myocardial ischemia/reperfusion (MI/R), and to further determine the role of insulin in the GIK-induced cardioprotective effect in vivo . METHODS: Forty eight male rabbits were subjected to 40 min MI followed by R for 3 h and were randomly received one of the following treatments: saline, GIK (glucose: 150 g/L, insulin: 60 U/L and KCl: 80 mmol/L), or insulin (n = 16 in each group) at 1 ml x kg(-1) x h(-1), beginning 30 min before MI and continuing throughout the 3 h-reperfusion. Blood glucose, electrolytes, arterial blood pressure and left ventricular pressure (LVP) were monitored throughout the experiment. Plasma creatine kinase (CK) and lactate dehydrogenase (LDH) activity were measured spectrophotometrically. Myocardial infarction and myocardial apoptosis (both DNA laddering and TUNEL analysis) were determined in a blinded manner. RESULTS: MI/R caused significant cardiac dysfunction and myocardial apoptosis (both strong DNA ladder formation and TUNEL-positive staining). Compared with vehicle, GIK-treated rabbits showed protection against MI/R as evidenced by reduced myocardial infarction (19.7% +/- 2.6% vs . 26.8% +/- 3.3% of vehicle, n = 10, P < 0.05), marked decrease in DNA fragmentation and apoptotic index (11.0% +/- 2.1% vs . 20.1% +/- 3.1% of vehicle, n = 6, P < 0.01), significant decrease of plasma CK and LDH and improved recovery of cardiac systolic/diastolic function at the end of R. Treatment with insulin alone decreased blood glucose significantly but still exerted cardioprotective effects comparable with that of GIK. CONCLUSIONS: GIK exerts cardioprotective effects against postischemic myocardial injury and improves cardiac functional recovery in vivo . Insulin, mainly through the anti-apoptotic effect, plays a key role in the GIK-elicited myocardial protection in MI/R.

The effect of Isoptin SR on blood pressure, heart function and hypertrophy of left ventricle of hypertensive patients.

Isoptin SR was used in 65 essential hypertensive patients. 240mg Isoptin SR (German Knoll Pharmaceutical company) per day was used in group A (35 cases) for 6 weeks, and in group B (30 cases) for 24 weeks, BP in group A and B decreased obviously in the first and second week after treatment. Marked effective rate and total effective rate were 65.7% and 74.3% respectively in group A, and 70.0% and 83.3% in group B. There were no significant changes of HR in group A before and after treatment while there was a decrease of HR in group B (P < 0.001). Left atrium (LA) decreased after 6 weeks in group A with an obvious increase of E/A ratio. There was significant decrease in LA, increase in E/A ratio, regression of IVS, LVPW and LVMI, but no changes in SV, CO, LVEF after 24 weeks of treatment of Isoptin SR in group B. E/A ratio was very sensitive and occurred earliest both in group A and group B. Side effects such as headache, dizziness, constipation, insomnia, peripheral oedema, sinus bradycardia occurred mainly in the first week of treatment. These symptoms disappeared gradually in the course of continued administration of Isoptin SR.

Correlation of serum lipids, lipoproteins, lipid peroxide products and metals with coronary heart disease.

Serum lipids, lipoproteins, malondialdelyde (MDA) and metal levels were determined in 86 patients with coronary heart disease (CHD) proved angiographically and 33 controls subjects. Serum concentrations of TC, LDL-C, AI, MDA and Cu were significantly elevated and serum HDL-C, Zn and Mg contents were decreased markedly in patients. Correlation analyses indicated that the severity of coronary arterial lesions was related positively to serum TC, LDL-C, AI, MDA and Cu levels, and inversely to HDL-C levels; both the serum Cu and MDA contents were related positively to TC and LDL-C levels. These data suggest that serum Cu and MDA might have effects on the extent of CA lesions during the progress of atherosclerosis in patients with CHD.

Correlations of blood lipids and lipoproteins with coronary artery disease documented by angiography.

A study of the relationships between serum TC, TG and HDL.C and selective coronary arteriography was carried out on 117 patients who were divided into groups according to the extent of artery stenosis. Normal control group consisted of 153 healthy subjects. There were no statistical differences in TC, TG, HDL.C, LDL.C, TC/HDL.C, LDL.C/HDL.C and TC-HDL.C/HDL.C between normal control group and the group with normal coronary arteriogram. LDL.C, TC/HDL.C, LDL.C/HDL.C, TC HDL.C/HDL.C rose and HDL.C decreased as the degree of coronary artery stenosis and the extent of stenosis increased, besides the medium and severe stenosis group. Analyses based on the correlation coefficients indicate that 3 compound indexes (TC-HDL.C/HDL.C, TC-HDL.C, LDL.C/HDL.C) are better in assessing CAD than single index such as LDL.C, HDL.C and TC. The results of our study showed that the 3 compound indexes might be regarded as important risk factors for CAD.