Our previous study (ISAC Protocol 18_186R) based on CPRD Gold and linked databases showed that people with intellectual disabilities (ID) are at markedly higher risk of fractures, particularly of the hip, than age and sex matched people without ID. These are devastating events, which can cause persistent disability for the individual, and high health and social care costs. Such fractures are often preventable, being largely due to osteoporosis (thinning of the bones), a very treatable condition. People with ID do not generally have their fracture risk assessed, and when they do, standard scores underestimate their risk, hence opportunities for prevention are missed. Therefore, we developed a new score for assessing fracture risk, IDFracture, specifically tailored to these patients.
In the proposed study, we will validate our IDFracture risk score in the CPRD Aurum database to ensure that it is robust for use in clinical practice. We will then investigate which of the following strategies is most cost-effective for finding ID people at risk of fracture so that they can be offered fracture prevention treatments: 1) Evaluate fracture risk according to current methods 2) Use IDFracture in all patients from age 40, with bone mineral density scan (DXA) in those near an intervention threshold 3) Perform DXA in all patients aged 40 and above, with follow up according to result.
Our study is funded by the National Institute of Health Research within its Translating Research into Policy scheme. Its results are expected to inform health policy at national level.
Unpublished results from our previous study (ISAC Protocol 18_186R) in CPRD Gold linked to HES show higher incidence of major osteoporotic (MOP) fractures (vertebra, shoulder, wrist, hip) and of hip fracture in adults with intellectual disabilities [ID] (n= 27706) compared to age and sex matched adults without ID (n= 139033).
We found that the current fracture risk calculator (QFracture) underestimated risk in ID patients.
We developed a risk score (IDFracture) estimating the 10-year risk of MOP and of hip fracture for ID adults 30-79 years old.
Validate IDFracture in the Aurum database
Determine the most cost-effective risk assessment method for MOP and for hip fractures in ID adults
Inform policy for osteoporotic fracture risk assessment
Incidence of MOP and of hip fracture within 10-years of the index date
Cost-effectiveness of 3 different strategies to determine risk of MOP and of hip fracture
Validation of IDFracture risk scores in the Aurum database with full linkage to HES and IMD datasets (to ensure complete capture of events and covariates).
For cost-effectiveness analyses we will use the subset aged 40-79 years.
We will use a Markov model with an annual transition cycle projecting life-long incidence of fractures and death. The model will be run assuming three strategies:
• Current strategy, using QFracture for risk calculation
• Risk assessment by IDFracture in all patients from age 40 years, with bone mineral density scan (DXA) in those in the region of an intervention threshold.
• DXA in all patients from age 40 (follow up according to result)
For each strategy, total lifetime costs and outcomes plus incremental cost-effectiveness ratio (ICER) will be calculated against the next most effective strategy. Main analyses will be done from NHS perspective. Impact of fracture on health-related quality of life will be taken from the literature.
Health Outcomes to be Measured:
Major osteoporotic fracture, hip fracture within 10-years of index date.
Cost-effectiveness of each of the following strategies to estimate risk of MOP and of hip fracture
• Current strategy as recommended by the National Institute for Health and Care Excellence (NICE) using QFracture for risk score calculation (1-3)
• Risk assessment by IDFracture (4) in all patients from age 40 years, with bone mineral density scan (DXA) in those in the region of an intervention threshold (1, 5)
• To perform DXA in all patients from age 40
HES Accident and Emergency;HES Admitted Patient Care;HES Diagnostic Imaging Dataset;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation