Assessment of fracture risk tools in care home residents

Ihama, Felix Emwinghama

Abstract

Introduction
Fragility fractures are common in care home residents. National guidelines recommend risk assessment to allow initiation of prophylactic measures. Currently available risk assessment tools have been tested in community dwelling adults but not in care home residents. It is possible that one or more of the existing tools are also practicable in this population.

Aim
The aim of this project was to identify fracture risk assessment tools which are usable in care home residents and to determine which is the most suitable for use in this population.

Objectives
1. To conduct a systematic literature review of existing fragility fracture risk assessment tools and select those that can be used in care home residents.
2. To develop a composite questionnaire which can be used to test the identified fragility fracture risk assessment tools in a care home population. This will be done by using the covariates in each tool to design a questionnaire, the acceptability of which will be assessed by consultation visits to two care homes to aid its refinement.
3. To undertake an observational pilot study of the fragility fracture risk assessment in a cohort of care home residents
4. To design a Clinical Algorithm.

Methods
1. A literature search was performed by a combination of electronic and manual literature searches and studies of assessment tools potentially usable in a care home population were selected and assessed based on content and quality criteria. The search was updated on 12/08/2019
2. A questionnaire was designed based on information from the literature review and tested by a cross-sectional survey in two care homes in Staffordshire.
3. A cohort observational study was conducted using the above questionnaire in 18 care homes in Boston, Lincolnshire, England.

Results
1. In the systematic review, 33 fragility tools were identified and four were potentially practicable in care home residents. These were: FRAX, QFractureScores, Garvan nomogram and Body Mass Index (BMI). The updated search identified a fifth-measure micro ribonucleic acid (miRNA). However, this was not implemented.
2. A composite questionnaire and information leaflet were designed and refined following feedback gleaned from the consultation visits.
3. In the feasibility study 217 (35%) participants out of 618 residents in the 18 care homes were enrolled. Out of the 217 participants, 147 (68%) had mental capacity and only 70 (32%) did not. This was because there was difficulty in obtaining informed consent from the consultees in residents without mental capacity.
4. Low BMI and history of dementia were identified as the risk factors for falls, fractures and combined falls and fractures in the cohort. Charlson Commorbidity Index predicted mortality (p= 0.034) and a score of = 36% was identified as the threshold for identifying participants who would not benefit from treatment. These three variables were used to design a Clinical Algorithm.
The fragility tools were easy to use given that the average duration for assessment was between 1 and 2 minutes. BMI of 25kg/m² or less had the highest sensitivity of 74.5% for falls. The sensitivity and specificity of FRAX and Garvan nomogram were not calculated because neither tool predicted falls, fractures or combined falls and fractures. The odds ratios for the prediction of the outcomes were as follows: FRAX falls 1.003, SE 0.011 (p=0.813), fractures 1.027, SE 0.024 (p=0.267), combined falls and fractures 1.027, SE 0.024 (p=0.267); QFractureScores falls 1.007 SE 0.005 (p=.160), fractures 1.024, SE 0.011 (p=0.036), combined falls and fractures 1.024, SE 0.011 (p=0.036); Garvan nomogram fall, 1.010, SE 0.005 (p=0.054), fractures 1.021, SE 0.011 (p=0.0620, combined falls and fractures 1.021, SE 0.011 (p=0.062); BMI falls 0.952, SE 0.021 (p=0.015), fractures 0.868, SE 0.073 (p=0.024), combined falls and fractures 0.868, SE 0.073 (p=0.024). Of the 10 incident fractures, 40% occurred in the participants who had dementia.

Conclusions
The systematic literature review identified many fragility risk assessment tools, but only four were potentially practicable in a care home population. Recruitment to the observational study was restricted mainly to residents who possessed mental capacity, because it was difficult to obtain consultee consent in this setting.
Although the fragility tools were easy to use, generally they had poor screening performance for the prediction of falls. BMI of 25kg/m² had the highest sensitivity. BMI was the best predictor of falls, fractures and combined falls and fractures but the associations were weak. QFractureScores was a predictor of fractures and combined falls and fractures. Neither FRAX nor Garvan nomogram were predictors of these outcomes. Of the 10 incident fractures, 40% were observed in participants who had dementia despite the small representation of this group, thus dementia is a strong risk factor for fractures in this cohort.
A fully powered and representative study is unlikely to be feasible, if individual consent is required, as the majority of care home residents do not have mental capacity, and legal representative consent is difficult to obtain in this setting. The results of this thesis suggest that BMI and dementia are strong predictors of falls and fractures. An algorithm was then designed using these to guide selection of suitable residents for treatment.

Citation

Ihama, F. E. (2019). Assessment of fracture risk tools in care home residents. (Thesis). Keele University. Retrieved from https://keele-repository.worktribe.com/output/415215

• Thesis Type: Thesis
• Publicly Available Date: May 26, 2023
• Public URL: https://keele-repository.worktribe.com/output/415215
• Award Date: 2019-12