Secondary relevance to vitamin D This paper mainly addresses another topic and also reports a vitamin D analysis, result or scientific synthesis.
This paper mainly addresses another topic and also reports a vitamin D analysis, result or scientific synthesis.
Why this paper is included
Supporting source evidence:
This increased to 38% and 53% in the 16 and 48‐weeks post‐fracture, respectively. 24% already had a prescription for vitamin D/calcium which increased to 73% at 48‐weeks post‐fracture.
Source: Abstract. View original paper
Relevance classification does not establish study quality, causation or editorial approval.
Study design Observational study.
Observational study
Observes exposures and outcomes without assigning the exposure; the subtype is not established here.
Meta-analysis unclear · Blinding unclear / not reported · Human research reported
How this paper was classified
Automatic source-based classification; awaiting editorial verification. This is not a quality score.
For this retrospective observational study, we used routinely‐collected general practice (GP) data within the Clinical Practice Research Datalink (CPRD) GOLD together with Office for National Statistics (ONS) mortality data.
Source: Methods.
Patients were included if they received a diagnosis of parkinsonism prior to or during the study period.
Source: Methods.
What the study types mean
- Systematic review
- Uses explicit methods to identify, assess and summarise research on a defined question.
- Scoping review
- Maps the extent and characteristics of research on a topic.
- Narrative review
- Discusses existing research; systematic-review methods have not been established by this classification.
- Review / evidence synthesis
- Reviews existing research; the exact review methods need further assessment.
- Randomised controlled trial
- Assigns participants or groups to comparison interventions by chance.
- Non-randomised intervention study
- Examines an intervention without random allocation.
- Cohort study
- Examines exposures and outcomes over time within a defined group.
- Case-control study
- Compares people with an outcome against people without it, examining exposures.
- Cross-sectional study
- Examines a population at a particular point in time.
- Observational study
- Observes exposures and outcomes without assigning the exposure; the subtype is not established here.
- Case report
- Describes an individual case.
- Case series
- Describes a group of individual cases.
- Secondary analysis
- Analyses existing study data; this paper is not automatically another independent trial.
- Laboratory study
- Investigates biological processes in a laboratory setting.
- Qualitative study
- Examines experiences or meanings using qualitative methods.
- Genetic instrumental-variable study
- Uses genetic variation as an instrument to investigate relationships; this is not a randomised treatment trial.
- Study protocol
- Describes planned methods and is not evidence that the planned study has reported results.
- Multiple reported designs
- More than one design is explicitly reported; examine the supporting methods.
- Study design awaiting verification
- The available source does not establish a sufficiently clear study design.
Meta-analysis combines numerical results from studies. Blinding describes who was unaware of treatment allocation. Neither is a separate quality grade. Study design, vitamin D relevance and document format are independent filters.
Full studies
Mind the Gap: Predictors of Osteoporosis Treatment Following Fragility Fracture in Parkinsonism
Observational study · Movement Disorders Clinical Practice · 2026
Original publication title
Mind the Gap: Predictors of Osteoporosis Treatment Following Fragility Fracture in Parkinsonism
© 2026 The Author(s). Movement Disorders Clinical Practice published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.
Original publication · Publisher record · CC BY 4.0
Source wording retained. Formatting, section navigation and display-title punctuation adapted for this research record.
At a glance
Abstract: What objectives does the paper state?
Our objective was to study the extent to which osteoporosis was treated, and predictors of treatment in a large representative cohort with parkinsonism.
Source: ObjectivesStudy methods
What methods does the paper describe?
Data Source
For this retrospective observational study, we used routinely‐collected general practice (GP) data within the Clinical Practice Research Datalink (CPRD) GOLD together with Office for National Statistics (ONS) mortality data. In 2013, CPRD contained 4.4 million active patients, representing 6.9% of the UK population from a network of 674 GP practices, with an additional 6.9 million inactive patients which includes those who have died or are no longer contributing data. 11 Medical events are recorded within the clinical and referral files using Read codes, and prescription medications are recorded using British National Formulary (BNF) coding.
CPRD therefore offers a unique opportunity to study the osteoporosis treatment gap in a large representative cohort of individuals with parkinsonism who have experienced at least one fragility fracture across the UK, where the treatment gap is defined as those who require treatment but are not receiving it within the given population.
Data Extraction
All individuals with a Read code for parkinsonism (excluding drug‐induced parkinsonism, Table A1) in the CPRD clinical and/or referral files who contributed data between January 1, 2010 and December, 31 2019 were extracted. Patients were included if they received a diagnosis of parkinsonism prior to or during the study period. Patients were excluded if: they were censored out of CPRD prior to the study period or the censor date was prior to the first instance of a Read code for parkinsonism, since this suggested a likely data error; they had a Read code for parkinsonism prior to age 35; or there was no available data on primary care prescriptions (CPRD therapy data).
This is because Read codes prior to age 35 potentially relate to a coding error, and the lack of prescription data indicates that there is likely to be further missing data.
Study Population
The index date was the date of the first fragility fracture which occurred in the study period (January 1, 2010: December 31, 2019) for patients with prevalent parkinsonism, registered at a practice contributing data deemed to be of research quality, which is determined by the recording and consistency of variables including practice registration date, transfer out date and date of birth. 12 Patients were followed‐up until they received a prescription for bone protective medication up to a maximum of 48‐weeks; or until they were censored, whichever came first.
This was repeated considering prescriptions for vitamin D and/or calcium supplementation. The censor date was defined as the earliest of: date of death; the last day of data collection at a given GP practice; the date of transfer to another practice without CPRD data‐linkage, or the study end date. If patients received a prescription prior to their fragility fracture, they were not followed up beyond the index date.
In this study, fragility fracture sites considered were hip, vertebrae, wrist/distal radius, humerus, rib, pelvis and unspecified osteoporotic fractures. We followed the methods described by Watson et al 2017 to form a list of Read codes indicating the presence of a fragility fracture (fracture codes) (Table A2). 13 The list was reviewed by two independent clinicians and any discrepancies were agreed by a third clinician. These Read codes were then used to identify patients who experienced at least one fragility fracture during the study period.
Fragility fractures were grouped into fracture events, and the initial fracture event within the study period was retained in the analysis as the index fracture event. Exact duplicates were removed from the dataset. To avoid double counting of fracture events, if two or more fracture codes related to the same fracture site or if an osteoporotic fracture at an unspecified site occurred less than 31 days apart, they were considered to be the same fracture event. Fracture events could consist of more than one specified fracture site if they occurred less than 31 days apart, and information about each fracture present at the event was retained.
Prescription Data
Prescription data relating to either bone protective medications, or vitamin D and/or calcium supplementation were extracted from the CPRD therapy file using the BNF chapters 060602 (bone protective medications), 090604 (vitamin D supplementation), and 09050101 (calcium supplementation). Additionally, the medication descriptions were searched for relevant terms (Table A3), as some drugs were listed in other BNF chapters. Medications were cross‐referenced with the NICE guidelines, and drugs which were not licensed for use in osteoporosis, calcium or vitamin D deficiency, were removed.
Pamidronate was excluded from this list as it is typically used in the context of malignancy. All prescriptions from the CPRD therapy file refer to drugs prescribed/dispensed in primary care for patients in all settings (own home, assisted living, care home etc).
Using index fragility fracture date, we determined whether patients were prescribed bone protective medications or vitamin D and/or calcium supplements before, or whether treatment was initiated after the fracture event. Patients were considered on treatment at the fracture event if prescription for oral bisphosphonates, strontium ranelate or teriparatide were received up to 3 months before; or if denosumab or zoledronate were documented in their medical records up to 9 or 18 months before respectively. These treatment windows were chosen using the standard dosing regimen for denosumab and zoledronate allowing for potential delays in treatment administration.
Based on the UK Fracture Liaison Service gold‐standard, stating that appropriate osteoporosis drug treatment should be commenced within 16‐weeks of a fragility fracture, 14 we analyzed 16‐week time intervals following the fracture event to determine adherence to this standard.
Incomplete Follow‐Up
We considered, within each 16 week time interval, how many patients remained within the study and therefore could have possibly received a prescription. Patients no longer remained within the study if they were censored out (eg, death, transfer out) prior to receiving a prescription. This was considered separately for the bone protective and supplementation prescription data. The patients remaining within each time interval are reported and used as the denominator to calculate the proportion on treatment within each time window.
The total number of patients censored from both the bone protective medications and supplementation data, and the reasons for censorship, are also described.
Bone protective medication and supplementation prescription data were considered separately, since these may not be initiated concurrently, potentially resulting in a patient being censored after prescription of only one medication type.
Statistical Analysis
Stata 18 was used for the analysis. 15 Data are described using mean/standard deviation, and the incidence of fractures is reported per 1000 person‐years with 95% confidence intervals. Logistic regression was used to obtain the prescription probability within a clinically meaningful time window and to test potential predictors of receiving prescriptions including: gender, age (at the mid‐point of the year of birth to the index fragility fracture date), PD duration at the time of fracture (calculated from the date at which the first diagnostic code for parkinsonism was applied), fracture site, Cambridge Multimorbidity Score (CMS) and whether the patient was resident in a care home at the time of fracture.
The CMS quantifies multimorbidity using a count of 37 conditions based on diagnostic and prescription codes. 16 Minimally adjusted models were determined using directed acyclic graphs (Figure A1). Two sensitivity analyses were conducted, restricting the analysis to: (i) patients with an initial diagnosis of PD; (ii) patients with 48‐weeks of data before and after the index fracture to test whether data availability affected any observed relationships.
Included parkinsonism read codes
| Description (as listed in CPRD medical.txt file) | Medcode | Read code |
|---|---|---|
| Parkinson's disease | 4321 | F12.00 |
| Parkinsonism with OH | 8956 | F130300 |
| Dementia in Parkinson's disease | 9509 | Eu02300 |
| Parkinson's disease NOS | 14,912 | F12z.00 |
| Parkinsonism in diseases classified elsewhere | 86,062 | Fyu2200 |
| Paralysis agitans | 1691 | F120.00 |
| Vascular parkinsonsim | 100,128 | F124.00 |
| Lewy body disease | 7572 | F116.00 |
| [X]Lewy body dementia | 26,270 | Eu02500 |
| Multiple system atrophy | 22,454 | F174.00 |
| Shy‐Drager syndrome | 35,839 | F130500 |
| Progressive supranuclear palsy | 9385 | F24y000 |
| Progressive supranuclear ophthalmoplegia | 40,553 | F130400 |
| Steele‐Richardson Oszewski syndrome | 93,910 | F24y012 |
| Steele Richardson Oszewski syn | 49,034 | F24y011 |
| Steele‐Richardson‐Oszewski syndrome | 7037 | F24y200 |
| Supranucelar paralysis | 29,515 | F36y000 |
| Corticobasal degeneration | 107,650 | F11y200 |
| Cerebral degeneration in Parkinson's disease | 96,860 | F11x900 |
Included fragility fracture read codes.
| Description (as listed in CPRD medical.txt file) | Medcode | Read code |
|---|---|---|
| Hip | ||
| Hip fracture | 1994 | S30.11 |
| Fracture of neck of femur | 2225 | S30.00 |
| Closed fracture of proximal femur, pertrochanteric | 5301 | S302.00 |
| Prmy open red+int fxn prox femoral # + screw/nail+plate device | 5742 | 7K1D000 |
| Closed reduction of fracture of hip | 6660 | 7K1L400 |
| Closed fracture of femur, intertrochanteric | 8648 | S302400 |
| Cls red+int fxn proximal femoral # + screw/nail device alone | 8719 | 7K1J000 |
| DHS‐Dynamic hip screw primary fixation of neck of femur | 9792 | 7K1D01E |
| Hip fracture NOS | 10,570 | S30y.11 |
| Dynamic hip screw primary fixation of neck of femur | 12,544 | 7K1D01F |
| Cls # prox femur, subcapital, Garden grade unspec. | 17,019 | S300500 |
| H/O: hip fracture | 17,936 | 14G7.00 |
| Closed fracture of neck of femur NOS | 18,273 | S30y.00 |
| Cls # proximal femur, trochanteric section, unspecified | 19,117 | S302000 |
| Open fracture proximal femur, subcapital, Garden grade III | 23,803 | S301800 |
| Closed fracture of unspecified proximal femur | 24,276 | S30w.00 |
| Prim open reduct # neck femur & op fix‐Richards screw | 24,493 | 7K1D01A |
| Pertrochanteric fracture | 28,965 | S304.00 |
| Closed fracture proximal femur, subtrochanteric | 29,145 | S302200 |
| Prmy open red+int fxn prox femoral # + screw/nail device alone | 33,624 | 7K1D600 |
| Closed fracture proximal femur, subcapital, Garden grade II | 33,957 | S300700 |
| Closed fracture proximal femur, subcapital, Garden grade IV | 34,078 | S300900 |
| Closed fracture proximal femur, subcapital, Garden grade I | 34,351 | S300600 |
| Prmy open red+int fxn prox fem # + screw/nail+intramed device | 34,764 | 7K1D700 |
| Primary int fxn(no red) prox fem # + screw/nail device alone | 35,004 | 7K1J500 |
| Closed fracture head of femur | 36,391 | S300400 |
| Closed fracture proximal femur, subcapital, Garden grade III | 36,599 | S300800 |
| Open fracture of neck of femur NOS | 38,054 | S30z.00 |
| Closed fracture proximal femur, transcervical | 38,489 | S300.00 |
| Primary int fxn(no red) prox fem # + screw/nail+plate device | 38,856 | 7K1J700 |
| Open fracture proximal femur, subcapital, Garden grade unspec | 38,878 | S301500 |
| Open fracture of femur, intertrochanteric | 39,396 | S303400 |
| Cls # prox femur, intracapsular section, unspecified | 39,984 | S300000 |
| Cl red intracaps fract neck femur fix − Smith‐Petersen nail | 40,999 | 7K1J012 |
| Primary int fxn(no red) prox fem # + scrw/nail+intramed device | 44,594 | 7K1J600 |
| Cls # of proximal femur, pertrochanteric section, NOS | 44,735 | S302z00 |
| Closed fracture proximal femur, intertrochanteric, two part | 45,141 | S302100 |
| Prim open reduct # neck femur & op fix − Neufield nail plate | 46,258 | 7K1D018 |
| Primary cls red+int fxn prox fem # + screw/nail+plate device | 46,959 | 7K1JD00 |
| Closed fracture proximal femur, other transcervical | 49,209 | S300y00 |
| Opn # proximal femur, intracapsular section, unspecified | 50,727 | S301000 |
| Cls # proximal femur, intertrochanteric, comminuted | 51,216 | S302300 |
| Closed fracture, base of neck of femur | 51,861 | S300311 |
| Open fracture proximal femur, subcapital, Garden grade IV | 51,999 | S301900 |
| Closed fracture proximal femur, basicervical | 52,194 | S300300 |
| Prim op red # nck femur and op fix‐Charnley compression screw | 52,395 | 7K1D012 |
| Cl red intracaps frac neck femur fix‐Garden cannulated screw | 53,670 | 7K1J011 |
| Prim cls rd + int fxn prox fem # + screw/nail+intramdulry device | 54,819 | 7K1JC00 |
| Primary cls red+int fxn prox fem # + screw/nail device alone | 55,386 | 7K1JB00 |
| Prim open red # neck femur and op fix‐McLaughlin nail plate | 56,568 | 7K1D017 |
| Cls red+int fxn prox femoral # + Richard's cannulat hip screw | 57,514 | 7K1J013 |
| Prim open reduct # neck femur and op fix‐Ross Brown nail | 57,884 | 7K1D01B |
| Prim op red # nck femur and op fix‐Zickel intramed nail plate | 57,889 | 7K1D01D |
| Open fracture of unspecified proximal femur | 58,642 | S30x.00 |
| Prim open reduct # neck femur and op fix‐Blount nail plate | 58,817 | 7K1D011 |
| Open fracture proximal femur, subcapital, Garden grade I | 60,885 | S301600 |
| Open fracture of proximal femur, pertrochanteric | 61,733 | S303.00 |
| Closed fracture proximal femur, transcervical, NOS | 62,966 | S300z00 |
| Prim open reduct # neck femur and op fix‐Pugh nail plate | 65,536 | 7K1D019 |
| Closed fracture proximal femur, midcervical section | 65,690 | S300200 |
| Open fracture proximal femur, subcapital, Garden grade II | 67,394 | S301700 |
| Open # of proximal femur, trochanteric section, unspecified | 67,633 | S303000 |
| Closed fracture of femur, subcapital | 68,229 | S300y11 |
| Open fracture proximal femur, other transcervical | 68,668 | S301y00 |
| Primary external fixation (without reduction) prox femoral # | 70,018 | 7K1K300 |
| Open fracture of proximal femur, pertrochanteric, NOS | 70,479 | S303z00 |
| Open fracture proximal femur, subtrochanteric | 71,282 | S303200 |
| Open fracture of femur, subcapital | 73,234 | S301y11 |
| Open fracture proximal femur, transcervical | 73,981 | S301.00 |
| Closed fracture‐subluxation, hip joint | 93,374 | S4E2.00 |
| Prim open reduct # neck femur and op fix‐Holt nail | 94,714 | 7K1D014 |
| Prim op red # nck femur and op fix‐Deyerle multiple hip pin | 97,337 | 7K1D013 |
| Open fracture proximal femur, intertrochanteric, comminuted | 97,971 | S303300 |
| Open fracture base of neck of femur | 100,771 | S301311 |
| Open fracture proximal femur, intertrochanteric, two part | 101,567 | S303100 |
| Primary cls reduction+external fixation proximal femoral # | 102,313 | 7K1K500 |
| Prim open reduct # neck femur and op fix‐Jewett nail plate | 105,352 | 7K1D015 |
| Prim op red frac neck fem op fix us prox fem nail antirotatn | 105,803 | 7K1DE00 |
| Remanip intracap fract neck fem and fix using nail or screw | 107,358 | 7K1Y000 |
| Prim open reduct # neck femur and op fix‐Massie nail plate | 107,932 | 7K1D016 |
| Open fracture proximal femur, transcervical, NOS | 112,712 | S301z00 |
| Open fracture proximal femur, midcervical section | 112,836 | S301200 |
| Open fracture proximal femur, basicervical | 113,426 | S301300 |
| Closed reduction of intracapsular # NOF internal fixat DHS | 39,322 | 7K1Jd00 |
| Vertebral fractures | ||
| Collapse of vertebra NOS | 2793 | N331.12 |
| Closed fracture lumbar vertebra | 3888 | S104.00 |
| Collapse of vertebra due to osteoporosis NOS | 4013 | N331L00 |
| Fracture of vertebra without spinal cord lesion | 4409 | S10.12 |
| Fracture of thoracic vertebra | 5381 | S15.00 |
| Collapse of lumbar vertebra due to osteoporosis | 5841 | N331J00 |
| Closed fracture lumbar vertebra, wedge | 8266 | S104100 |
| Collapse of thoracic vertebra | 9319 | N331F00 |
| Fracture of lumbar vertebra | 10,990 | S10B000 |
| Collapse of lumbar vertebra | 11,543 | N331G00 |
| Other specified closed fracture thoracic vertebra | 11,770 | S102y00 |
| Fracture of lumbar spine and pelvis | 12,406 | S10B.00 |
| Osteoporosis + pathological fracture thoracic vertebrae | 12,673 | N331900 |
| Closed fracture of unspecified cervical vertebra | 15,613 | S100000 |
| Collapse of thoracic vertebra | 15,837 | N331011 |
| Fatigue fracture of vertebra | 16,895 | N1y1.00 |
| Collapse of cervical vertebra | 17,008 | N331E00 |
| Osteoporosis + pathological fracture lumbar vertebrae | 17,377 | N331800 |
| Collapse of thoracic vertebra due to osteoporosis | 19,048 | N331K00 |
| H/O: vertebral fracture | 19,235 | 14G8.00 |
| Collapse of lumbar vertebra | 23,686 | N331111 |
| Closed fracture of seventh cervical vertebra | 24,672 | S100700 |
| Closed fracture thoracic vertebra | 27,404 | S102.00 |
| Closed fracture of fifth cervical vertebra | 27,575 | S100500 |
| Closed fracture of sixth cervical vertebra | 27,654 | S100600 |
| Fracture of first cervical vertebra | 28,133 | S10A000 |
| Closed fracture thoracic vertebra, wedge | 28,524 | S102100 |
| C6 vertebra closed fracture without spinal cord lesion | 33,503 | S100611 |
| C2 vertebra closed fracture without spinal cord lesion | 33,967 | S100211 |
| Fracture of second cervical vertebra | 34,403 | S10A100 |
| C5 vertebra closed fracture without spinal cord lesion | 34,873 | S100511 |
| C7 vertebra closed fracture without spinal cord lesion | 38,053 | S100711 |
| Collapsed vertebra NOS | 38,728 | N331D00 |
| Closed fracture thoracic vertebra not otherwise specified | 41,138 | S102z00 |
| Closed fracture of fourth cervical vertebra | 41,548 | S100400 |
| C1 vertebra closed fracture: no spinal cord lesion | 42,561 | S100111 |
| Osteoporotic vertebral collapse | 44,386 | N331.14 |
| Collapse of cervical vertebra due to osteoporosis | 45,736 | N331H00 |
| Osteoporosis + pathological fracture cervical vertebrae | 48,772 | N331A00 |
| C3 vertebra closed fracture without spinal cord lesion | 52,699 | S100311 |
| Closed fracture cervical vertebra, wedge | 53,337 | S100H00 |
| Closed fracture of third cervical vertebra | 60,593 | S100300 |
| Open fracture lumbar vertebra, wedge | 65,302 | S105100 |
| C4 vertebra closed fracture without spinal cord lesion | 67,358 | S100411 |
| [X]Collapsed vertebra in diseases classified elsewhere | 73,900 | Nyu6700 |
| [X]Fracture of other specified cervical vertebra | 94,638 | Syu1500 |
| Collapse of cervical vertebra | 110,999 | N331C11 |
| Wrist/distal radius | ||
| Wrist fracture: closed | 203 | S234.11 |
| Closed reduction of fracture of wrist | 5951 | 7K1LM00 |
| Fracture at wrist and hand level | 8056 | S242.00 |
| Wrist fracture: open | 10,022 | S235.11 |
| Closed fracture‐subluxation of the wrist | 17,921 | S4C2.00 |
| Fracture‐dislocation or subluxation of wrist | 18,614 | S4C.00 |
| Closed fracture dislocation of wrist | 34,429 | S4C0.00 |
| Open fracture dislocation wrist | 42,844 | S4C1.00 |
| [X]Fracture of other and unspecified parts of wrist and hand | 53,068 | Syu6500 |
| Open fracture‐subluxation of the wrist | 59,219 | S4C3.00 |
| Closed fracture distal radius, extra‐articular, other type | 19,058 | S234D00 |
| Closed fracture distal radius, intra‐articular, die‐punch | 44,844 | S234C00 |
| Closed fracture distal radius, intra‐articular, other type | 28,293 | S234E00 |
| Closed fracture of the distal radius, unspecified | 1742 | S234200 |
| Open fracture distal radius, extra‐articular other type | 53,698 | S235D00 |
| Open fracture distal radius, intra‐articular other type | 63,588 | S235E00 |
| Open fracture distal radius, intra‐articular, die‐punch | 104,070 | S235C00 |
| Humerus | ||
| Fracture of humerus | 517 | S22.00 |
| Fracture of upper end of humerus | 2101 | S226.00 |
| Closed reduction of fracture of humerus | 6379 | 7K1LF00 |
| Open fracture of the proximal humerus | 9420 | S221.00 |
| Fracture of humerus NOS | 10,382 | S22z.00 |
| Closed fracture proximal humerus, greater tuberosity | 11,044 | S220300 |
| Closed fracture of the proximal humerus | 11,222 | S220.00 |
| Closed fracture proximal humerus, neck | 11,313 | S220100 |
| Closed fracture of humerus NOS | 19,186 | S222000 |
| Closed fracture of humerus, shaft | 27,886 | S222100 |
| Closed fracture proximal humerus, head | 28,739 | S220400 |
| Closed fracture proximal humerus, four part | 29,137 | S220700 |
| Fracture of shaft of humerus | 30,659 | S227.00 |
| Closed fracture of proximal humerus, anatomical neck | 33,489 | S220200 |
| Closed fracture of humerus, shaft or unspecified part | 36,464 | S222.00 |
| Closed fracture of proximal humerus not otherwise specified | 38,353 | S220z00 |
| Closed fracture proximal humerus, three part | 40,330 | S220600 |
| Closed fracture of proximal humerus, unspecified part | 44,721 | S220000 |
| Open fracture of proximal humerus not otherwise specified | 45,275 | S221z00 |
| Open fracture proximal humerus, greater tuberosity | 48,239 | S221300 |
| Closed fracture of humerus, upper epiphysis | 52,406 | S220500 |
| Open fracture proximal humerus, neck | 53,622 | S221100 |
| Open fracture of proximal humerus, unspecified part | 53,688 | S221000 |
| Closed fracture of humerus, shaft or unspecified part NOS | 61,378 | S222z00 |
| Open fracture proximal humerus, head | 70,486 | S221400 |
| Open fracture proximal humerus, three part | 70,604 | S221600 |
| Open fracture proximal humerus, four part | 70,653 | S221700 |
| Open fracture of proximal humerus, anatomical neck | 71,207 | S221200 |
| Rib | ||
| Closed fracture rib | 280 | S120.00 |
| Closed fracture of rib, unspecified | 7831 | S120000 |
| Fracture of rib | 9688 | S127.00 |
| Multiple fractures of ribs | 10,696 | S127000 |
| Closed fracture multiple ribs | 11,004 | S120900 |
| Rib fracture NOS | 11,378 | S12z.11 |
| Closed fracture of one rib | 16,494 | S120100 |
| Multiple rib fractures | 17,107 | S29.12 |
| Closed fracture of rib(s) NOS | 28,244 | S120z00 |
| Cough fracture of ribs | 28,538 | S127100 |
| Closed fracture of four ribs | 34,197 | S120400 |
| Closed fracture of two ribs | 40,533 | S120200 |
| Open fracture of two ribs | 44,826 | S121200 |
| Open fracture rib | 48,224 | S121.00 |
| Closed fracture of five ribs | 55,424 | S120500 |
| Closed fracture of three ribs | 56,384 | S120300 |
| Closed fracture of six ribs | 65,484 | S120600 |
| Closed fracture of eight or more ribs | 68,652 | S120800 |
| Open fracture multiple ribs | 71,452 | S121900 |
| Open fracture of rib, unspecified | 73,613 | S121000 |
| Open fracture of seven ribs | 73,956 | S121700 |
| Closed fracture of seven ribs | 90,494 | S120700 |
| Open fracture of rib(s) NOS | 101,560 | S121z00 |
| Open fracture of one rib | 112,701 | S121100 |
| Open fracture of four ribs | 113,739 | S121400 |
| Pelvis | ||
| Fracture or disruption of pelvis | 738 | S13.00 |
| Closed fracture pelvis, multiple pubic rami‐stable | 6667 | S132100 |
| Closed fracture pelvis, single pubic ramus | 7004 | S132000 |
| Other or multiple closed fracture of pelvis NOS | 11,639 | S134z00 |
| Fracture of lumbar spine and pelvis | 12,406 | S10B.00 |
| Closed fracture pelvis, coccyx | 14,834 | S108.00 |
| Closed fracture of pelvis NOS | 28,375 | S13y.00 |
| Other or multiple closed fracture of pelvis | 33,961 | S134.00 |
| Closed fracture pelvis, ischium | 34,708 | S134100 |
| Closed fracture‐dislocation of pelvis | 34,910 | S4J0100 |
| Closed fracture pelvis, multiple pubic rami‐unstable | 46,592 | S132200 |
| Unspecified site | ||
| Fragility fracture due to unspecified osteoporosis | 11,503 | N331M00 |
| H/O: fragility fracture | 18,731 | 14G6.00 |
| [X]Unspecified osteoporosis with pathological fracture | 18,825 | NyuB800 |
| Idiopathic osteoporosis with pathological fracture | 27,597 | N331600 |
| Osteoporosis of disuse with pathological fracture | 33,526 | N331300 |
| Postmenopausal osteoporosis with pathological fracture | 38,395 | N331B00 |
| Postoophorectomy osteoporosis with pathological fracture | 39,334 | N331200 |
| Drug‐induced osteoporosis with pathological fracture | 46,894 | N331500 |
| [X]Other osteoporosis with pathological fracture | 57,301 | NyuB000 |
| Postsurgical malabsorption osteoporosis with path fracture | 68,019 | N331400 |
| Fragility fracture | 93,497 | N331N00 |
| Minimal trauma fracture due to unspecified osteoporosis | 93,705 | N331M11 |
Search Terms for prescriptions
| Search term | Target |
|---|---|
| Bisphosphon | Bisphosphonate/Bisphosphonic acid |
| Disphosphon | Diphosphonate/Diphosphonic acid |
| Alendron | Alendronate/Alendronic acid |
| MK‐217 | Alendronic acid (Alternative notation) |
| Ibandron | Ibandronate/Ibandronic acid |
| Risedron | Risedronate/Risedronic acid |
| Zolendron | Zoledronate/Zolendronic acid |
| Etidron | Etidronate/Etidronic acid |
| EHDP | Etidronate (alternative notation) |
| Strontium ranelate | Strontium ranelate |
| Sodium clondronate | Sodium clondronate |
| Tiludron | Tilundronate/Tilundronic acid |
| Denosumab | Denosumab |
| AMG 162 | Denosumab (alternative notation) |
| Romosozumab | Romosozumab |
| AMG 785 | Romosozumab (alternative notation) |
| Teriparatide | Teriparatide |
| Human parathyroid hormone | Human parathyroid hormone |
| Abaloparatide | Abaloparatide |
| BNF chapters | |
| 060602 | Bone protective medications |
| 090604 | Vitamin D supplementation |
| 09050101 | Calcium supplementation |
Prescriptions for Bone Protective Medications and Supplementation at Indexed Fracture Event
Results: Prescriptions for Bone Protective Medications and Supplementation at Indexed Fracture Event
Of those with a fragility fracture event, 225 people (12.3%) were on any bone protective medication prior to their fracture. A further 486 people were started on bone protective medication within 48 weeks following their fracture, therefore cumulatively 52.9% were in receipt of a prescription within 48 weeks (Fig. 2A and Table A4). Four hundred and forty two (24.2%) were prescribed vitamin D and/or calcium supplementation prior to their fracture, and a further 651 people were prescribed supplementation up to 48 weeks after their fracture, therefore cumulatively 73.2% were in receipt of a prescription within 48 weeks (Fig. 2B and Table A5).
Stacked bar charts illustrating the prescription of (A) bone protective medication (B) vitamin D/calcium supplementation prior to and following a fragility fracture.
Table of bone protective medication prescriptions over 48 weeks following a fragility fracture
| Timeline | Censored from study (N) | Number of participants a | Oral bisphosphonate | Strontium ranelate | Denosumab | Zoledronate | All |
|---|---|---|---|---|---|---|---|
| Before fracture 1 | 0 | 1823 | 217 | 6 | 2 | 0 | 225 |
| (Percentage) | (11.9%) | (0.3%) | (0.1%) | (0.0%) | (12.3%) | ||
| Up to 16 weeks after | 236 | 1587 | 572 | 17 | 5 | 3 | 597 |
| (Percentage) | (36.0%) | (1.1%) | (0.3%) | (0.2%) | (37.6%) | ||
| Up to 32 weeks after | 369 | 1454 | 653 | 18 | 9 | 3 | 683 |
| (Percentage) | (44.9%) | (1.2%) | (0.6%) | (0.2%) | (47.0%) | ||
| Up to 48 weeks after | 479 | 1344 | 678 | 18 | 12 | 3 | 711 |
| (Percentage) | (50.4%) | (1.3%) | (0.9%) | (0.2%) | (52.9%) |
Note: This table displays the cumulative frequency of each medication type and all bone protective medications, starting before the fracture event followed by 16 week intervals.
The number of people with available data within each time window for bone protective medication.
Table of vitamin D and/or calcium supplement prescriptions over 48 weeks following a fragility fracture
| Timeline | Censored from study (N) | Number of participants a | Supplementation b |
|---|---|---|---|
| Before fracture 1 | 0 | 1823 | 442 |
| (Percentage) | (24.2%) | ||
| Up to 16 weeks after | 181 | 1642 | 951 |
| (Percentage) | (57.9%) | ||
| Up to 32 weeks after | 269 | 1554 | 1054 |
| (Percentage) | (67.8%) | ||
| Up to 48 weeks after | 330 | 1493 | 1093 |
| (Percentage) | (73.2%) |
Note: This table displays the cumulative frequency of supplementation (vitamin D and/or calcium) starting before the fracture event followed by 16 week intervals.
The number of people with available data within each time window for supplementation (Vitamin D and/or calcium).
Supplementation refers to a prescription being received for vitamin D and/or calcium.
Results: Incomplete Follow‐Up
A total of 479 patients were censored from the bone protective medication data within 48 weeks of a fragility fracture, due to 123 deaths, 53 patients reaching the last point of data collection from the practice, 207 transferring to a GP practice without CPRD linkage, and 96 reaching the end of the study period prior to a prescription being issued (Figure A2). Of the 1598, who were not prescribed bone protective medication prior to their fragility fracture, the mean follow‐up time was 1.2 years (SD = 1.6 years, range 0 days to 9.7 years).
With respect to vitamin D and/or calcium prescriptions, a total of 330 patients were censored within 48 weeks of a fragility fracture, due to 92 deaths, 37 patients reaching the last point of data collection from the practice, 149 transferring to a GP practice without CPRD linkage, and 52 reaching the end of the study period prior to a prescription being issued (Figure A3). Of the 1381, who were not prescribed vitamin D and/or calcium prior to their fragility fracture, the mean follow‐up time was 0.9 years (SD = 1.5 years, range 0 days to 9.7 years).
Source: Incomplete Follow‐Up Source: Incomplete Follow‐UpResults: Predictors of Bone Protective Prescriptions
We explored demographic factors and fracture site as predictors of bone protective prescriptions being issued (Table 2). Being female (OR 2.06 [95% CI (1.69 to 2.51)], p < 0.001) and being aged 75 to 79 (OR 1.82 [95% CI (1.40 to 2.37)], p < 0.001), 80 to 84 (OR 1.67 [95% CI (1.28 to 2.17)], p < 0.001) or over 85 (OR 1.63 [95% CI (1.24 to 2.16)], p = 0.001), increased the odds of bone protective medication prescription. Residing in a care home reduced the odds of bone protective medication prescription (OR 0.64 [95% CI (0.43 to 0.96)], p = 0.03).
Sustaining a vertebral fracture (OR 1.84 [95% CI (1.38 to 2.45)], p < 0.001), or having unspecified fragility fracture recorded (OR 1.76 [95% CI (1.33 to 2.33)], p < 0.001), when compared to sustaining a hip fracture, also increased the odds of bone protective medication prescription, whereas sustaining a rib fracture (OR 0.28 [95% CI (0.16 to 0.49)], p < 0.001) reduced the odds.
Unadjusted logistic regression models to look at factors predicting bone protective medication prescriptions
| Unadjusted odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Gender | |||
| Male | ref | 1823 | |
| Female | 2.06 (1.69 to 2.51) | <0.001 | |
| Age | |||
| 35 to 74 | ref | 1823 | |
| 74 to 79 | 1.82 (1.40 to 2.37) | <0.001 | |
| 80 to 84 | 1.67 (1.28 to 2.17) | <0.001 | |
| 85+ | 1.63 (1.24 to 2.16) | 0.001 | |
| PD duration | |||
| Less than 1 year | ref | 1823 | |
| 1 to 2.4 years | 0.78 (0.59 to 1.03) | 0.08 | |
| 2.5 to 4 years | 0.80 (0.61 to 1.05) | 0.11 | |
| 5 to 9 years | 0.79 (0.59 to 1.05) | 0.10 | |
| 10+ years | 0.76 (0.54 to 1.07) | 0.12 | |
| Fracture type | |||
| Hip | ref | 1823 | |
| Vertebral | 1.84 (1.38 to 2.45) | <0.001 | |
| Wrist | 0.68 (0.44 to 1.05) | 0.08 | |
| Humerus | 0.80 (0.57 to 1.12) | 0.19 | |
| Rib | 0.28 (0.16 to 0.49) | <0.001 | |
| Pelvis | 1.14 (0.71 to 1.85) | 0.58 | |
| Unspecified | 1.76 (1.33 to 2.33) | <0.001 | |
| Multiple Fractures | 1.30 (0.62 to 2.71) | 0.49 | |
| Cambridge multimorbidity score | |||
| 0/1 | ref | 1823 | |
| 2 | 0.82 (0.51 to 1.33) | 0.42 | |
| 3 | 0.83 (0.53 to 1.31) | 0.43 | |
| 4 | 0.91 (0.58 to 1.43) | 0.68 | |
| 5 | 0.93 (0.59 to 1.46) | 0.74 | |
| 6 | 0.95 (0.59 to 1.52) | 0.83 | |
| 7+ | 0.88 (0.56 to 1.38) | 0.57 | |
| Care home residence | |||
| Not living in care home | ref | 1823 | |
| Living in care home | 0.64 (0.43 to 0.96) | 0.03 | |
When adjusting for age and gender, vertebral fractures (OR 2.06 [95% CI (1.53 to 2.77)], p < 0.001) had stronger association with receiving a prescription whereas unspecified fracture (OR 1.57 [95% CI (1.17 to 2.10)], p < 0.001), and rib fractures (OR 0.34 [95% CI (0.19 to 0.60)], p < 0.001) both had a slightly attenuated relationship with receiving a prescription for bone protective medication versus hip fracture. When adjusting for age, multi‐morbidity (CMS) and gender, living in a care home (OR 0.58 [95% CI (0.38 to 0.88)], p = 0.01) had a stronger association of decreasing the odds of receiving a prescription than in the univariate model.
Multimorbidity still did not change the odds of receiving a prescription after adjusting for age (Table 3). The direction of these associations was not altered when restricting the analysis to those with an initial diagnosis of PD or those with 48 weeks of data before and after the index fracture, although it did weaken the association between prescriptions and care home residency (Table A6 and A7).
Minimally adjusted logistic regression models to look at factors predicting bone protective medication prescriptions
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (unadjusted) | |||
| Hip | ref | 1823 | |
| Vertebral | 1.84 (1.38 to 2.45) | <0.001 | |
| Wrist | 0.68 (0.44 to 1.05) | 0.08 | |
| Humerus | 0.80 (0.57 to 1.12) | 0.19 | |
| Rib | 0.28 (0.16 to 0.49) | <0.001 | |
| Pelvis | 1.14 (0.71 to 1.85) | 0.58 | |
| Unspecified | 1.76 (1.33 to 2.33) | <0.001 | |
| Multiple Fractures | 1.30 (0.62 to 2.71) | 0.49 | |
| Fracture type adjusted for age | |||
| Hip | ref | 1823 | |
| Vertebral | 1.98 (1.48 to 2.65) | <0.001 | |
| Wrist | 0.75 (0.49 to 1.17) | 0.21 | |
| Humerus | 0.82 (0.58 to 1.16) | 0.26 | |
| Rib | 0.31 (0.18 to 0.55) | <0.001 | |
| Pelvis | 1.15 (0.71 to 1.86) | 0.57 | |
| Unspecified | 1.84 (1.39 to 2.44) | <0.001 | |
| Multiple fractures | 1.29 (0.62 to 2.71) | 0.50 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 1823 | |
| Vertebral | 2.06 (1.53 to 2.77) | <0.001 | |
| Wrist | 0.60 (0.38 to 0.93) | 0.02 | |
| Humerus | 0.72 (0.50 to 1.02) | 0.06 | |
| Rib | 0.34 (0.19 to 0.60) | <0.001 | |
| Pelvis | 0.99 (0.61 to 1.61) | 0.96 | |
| Unspecified | 1.57 (1.17 to 2.10) | <0.001 | |
| Multiple fractures | 1.23 (0.58 to 2.61) | 0.59 | |
| Cambridge multimorbidity score (unadjusted) | |||
| 0/1 | ref | 1823 | |
| 2 | 0.82 (0.51 to 1.33) | 0.42 | |
| 3 | 0.83 (0.53 to 1.31) | 0.43 | |
| 4 | 0.91 (0.58 to 1.43) | 0.68 | |
| 5 | 0.93 (0.59 to 1.46) | 0.74 | |
| 6 | 0.95 (0.59 to 1.52) | 0.83 | |
| 7+ | 0.88 (0.56 to 1.38) | 0.57 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 1823 | |
| 2 | 0.77 (0.47 to 1.25) | 0.29 | |
| 3 | 0.76 (0.48 to 1.21) | 0.25 | |
| 4 | 0.81 (0.51 to 1.28) | 0.37 | |
| 5 | 0.81 (0.51 to 1.28) | 0.36 | |
| 6 | 0.82 (0.51 to 1.32) | 0.41 | |
| 7+ | 0.77 (0.49 to 1.21) | 0.26 | |
| Care home residency (unadjusted) | |||
| No | ref | 1823 | |
| Yes | 0.64 (0.43 to 0.96) | 0.03 | |
| Care home residency adjusted for age | |||
| No | ref | 1823 | |
| Yes | 0.58 (0.39 to 0.88) | 0.01 | |
| Care home residency adjusted for age and gender | |||
| No | ref | 1823 | |
| Yes | 0.58 (0.38 to 0.88) | 0.01 | |
| Care home residency adjusted for age, gender and CMS | |||
| No | ref | 1823 | |
| Yes | 0.58 (0.38 to 0.88) | 0.01 | |
Sensitivity analysis restricted to those with an initial diagnosis of Parkinson's disease
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (unadjusted) | |||
| Hip | ref | 1521 | |
| Vertebral | 1.79 (1.31 to 2.43) | < 0.001 | |
| Wrist | 0.64 (0.40 to 1.02) | 0.06 | |
| Humerus | 0.72 (0.50 to 1.04) | 0.08 | |
| Rib | 0.25 (0.13 to 0.46) | < 0.001 | |
| Pelvis | 1.00 (0.57 to 1.73) | 0.99 | |
| Unspecified | 1.69 (1.24 to 2.30) | < 0.001 | |
| Multiple fractures | 1.38 (0.61 to 3.12) | 0.44 | |
| Fracture type adjusted for age | |||
| Hip | ref | 1521 | |
| Vertebral | 1.91 (1.40 to 2.61) | < 0.001 | |
| Wrist | 0.72 (0.45 to 1.16) | 0.18 | |
| Humerus | 0.74 (0.51 to 1.08) | 0.12 | |
| Rib | 0.28 (0.15 to 0.53) | < 0.001 | |
| Pelvis | 1.00 (0.57 to 1.75) | 1.00 | |
| Unspecified | 1.79 (1.31 to 2.44) | < 0.001 | |
| Multiple fractures | 1.40 (0.61 to 3.19) | 0.42 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 1521 | |
| Vertebral | 2.04 (1.48 to 2.81) | < 0.001 | |
| Wrist | 0.56 (0.34 to 0.91) | 0.02 | |
| Humerus | 0.64 (0.44 to 0.94) | 0.02 | |
| Rib | 0.31 (0.16 to 0.58) | < 0.001 | |
| Pelvis | 0.85 (0.48 to 1.49) | 0.56 | |
| Unspecified | 1.53 (1.11 to 2.10) | 0.01 | |
| Multiple Fractures | 1.31 (0.56 to 3.02) | 0.53 | |
| Cambridge multimorbidity score unadjusted | |||
| 0/1 | ref | 1521 | |
| 2 | 0.80 (0.48 to 1.36) | 0.42 | |
| 3 | 0.80 (0.48 to 1.33) | 0.39 | |
| 4 | 0.88 (0.53 to 1.45) | 0.61 | |
| 5 | 0.95 (0.58 to 1.57) | 0.85 | |
| 6 | 0.97 (0.58 to 1.63) | 0.92 | |
| 7+ | 0.87 (0.53 to 1.43) | 0.58 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 1521 | |
| 2 | 0.74 (0.44 to 1.27) | 0.28 | |
| 3 | 0.74 (0.44 to 1.23) | 0.24 | |
| 4 | 0.79 (0.48 to 1.31) | 0.36 | |
| 5 | 0.84 (0.50 to 1.40) | 0.49 | |
| 6 | 0.83 (0.49 to 1.41) | 0.50 | |
| 7+ | 0.76 (0.46 to 1.26) | 0.29 | |
| Care home residence (Unadjusted) a | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.75 (0.48 to 1.17) | 0.20 | |
| Care home residence adjusted for age | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.68 (0.44 to 1.07) | 0.10 | |
| Care home residence adjusted for age and gender | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.69 (0.44 to 1.08) | 0.11 | |
| Care home residence adjusted for age, gender and CMS | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.68 (0.43 to 1.08) | 0.10 | |
There are now 94 (6.2%) of patients living in a care home.
Sensitivity analysis restricted to those with 48‐weeks of data before and after the index fracture
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (Unadjusted) | |||
| Hip | ref | 765 | |
| Vertebral | 1.64 (1.05 to 2.56) | 0.03 | |
| Wrist | 0.58 (0.32 to 1.05) | 0.07 | |
| Humerus | 0.67 (0.39 to 1.13) | 0.13 | |
| Rib | 0.22 (0.10 to 0.48) | <0.001 | |
| Pelvis | 0.56 (0.25 to 1.27) | 0.17 | |
| Unspecified | 1.57 (1.05 to 2.36) | 0.03 | |
| Multiple Fractures | 1.31 (0.37 to 4.63) | 0.67 | |
| Fracture type adjusted for age | |||
| Hip | ref | 765 | |
| Vertebral | 1.73 (1.10 to 2.71) | 0.02 | |
| Wrist | 0.62 (0.34 to 1.14) | 0.13 | |
| Humerus | 0.69 (0.40 to 1.17) | 0.17 | |
| Rib | 0.24 (0.11 to 0.52) | <0.001 | |
| Pelvis | 0.57 (0.25 to 1.30) | 0.18 | |
| Unspecified | 1.61 (1.07 to 2.42) | 0.02 | |
| Multiple Fractures | 1.36 (0.38 to 4.84) | 0.64 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 765 | |
| Vertebral | 1.78 (1.12 to 2.82) | 0.01 | |
| Wrist | 0.50 (0.27 to 0.94) | 0.03 | |
| Humerus | 0.60 (0.35 to 1.03) | 0.06 | |
| Rib | 0.26 (0.12 to 0.57) | < 0.001 | |
| Pelvis | 0.48 (0.21 to 1.10) | 0.08 | |
| Unspecified | 1.44 (0.95 to 2.19) | 0.09 | |
| Multiple Fractures | 1.30 (0.36 to 4.69) | 0.69 | |
| Cambridge multimorbidity score unadjusted | |||
| 0/1 | ref | 765 | |
| 2 | 0.66 (0.29 to 1.48) | 0.31 | |
| 3 | 0.85 (0.40 to 1.81) | 0.67 | |
| 4 | 1.02 (0.48 to 2.16) | 0.96 | |
| 5 | 0.85 (0.39 to 1.84) | 0.68 | |
| 6 | 1.16 (0.53 to 2.56) | 0.71 | |
| 7+ | 1.08 (0.51 to 2.29) | 0.84 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 765 | |
| 2 | 0.60 (0.26 to 1.37) | 0.23 | |
| 3 | 0.79 (0.36 to 1.70) | 0.54 | |
| 4 | 0.90 (0.42 to 1.93) | 0.78 | |
| 5 | 0.76 (0.34 to 1.66) | 0.48 | |
| 6 | 0.97 (0.43 to 2.17) | 0.93 | |
| 7+ | 0.91 (0.42 to 1.95) | 0.80 | |
| Care home residence (unadjusted) a | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.54 (0.24 to 1.18) | 0.12 | |
| Care home residence adjusted for age | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.48 (0.22 to 1.07) | 0.07 | |
| Care home residence adjusted for age and gender | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.47 (0.21 to 1.05) | 0.07 | |
| Care home residence adjusted for age, gender and CMS | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.44 (0.20 to 1.00) | 0.05 | |
There are now 31 (4.1%) of patients living in a care home.
Discussion
Despite a significantly increased risk of sustaining fragility fractures, bone‐health and osteoporosis management is often overlooked when caring for people with parkinsonism. To understand the extent to which osteoporosis is treated in line with national guidelines, we investigated whether either bone protective medications, vitamin D and/or calcium supplements were prescribed. This was based on primary care records before and after people experienced fragility fractures. We identified treatment gaps and investigated potential predictors of prescriptions being made within 48 weeks following a fracture.
We identified a fracture rate of 27.3 per 1000 person years which aligns with the current literature. 6 , 17 Consistent with our previous findings 18 and with the results of the Parkinson's UK National Audit, 9 we identified a substantial treatment gap. Only 38% of this real‐world population who sustained a fragility fracture were receiving treatment within the recommended 16 week period. Increasing the treatment window to 48 weeks post fracture only increased this proportion to 53%. More people were prescribed vitamin D and/or calcium supplementation within 16 (58%) and 48 weeks (73%).
Being female, aged >75 years and having sustained either a vertebral or unspecified fragility fracture increased the odds of receiving a prescription for bone protective medications. Conversely, people with parkinsonism who had sustained rib fracture(s), or were residing in a care home were less likely to be prescribed bone protective medication. The direction of these associations was maintained when restricting the analysis to those with: more complete data (Table A7); or an initial diagnosis of PD (Table A6).
Due to initial diagnostic uncertainty, it is possible that some PD diagnoses may subsequently have been revised to an atypical parkinsonism. 19
Whilst demonstrating this striking and stark treatment gap, we have considered additional factors which may have impacted this finding. Using CPRD data it is possible to see which patients were prescribed bone protective medication in primary care. There is insufficient data to reliably determine whether bone health was considered and then no treatment was initiated (appropriately), adherence to prescribed treatments, or whether patients had received intravenous zoledronate via secondary care. There are several reasons why osteoporosis treatment may have been deemed inappropriate despite a patient sustaining a fragility fracture, including: concerns over polypharmacy; upper gastrointestinal disorders; renal impairment; patient preferences; bisphosphonate treatment holidays; life‐expectancy; concerns over side‐effects; or fracture risk assessment concluding treatment wasn’t warranted. 20 , 21 These legitimate reasons why treatments are sometimes not initiated will contribute to an over‐estimation of the reported treatment gap.
We might hypothesize that individuals with greater co‐morbidity burden are less likely to receive a prescription, but this was not evident in our data.
It is unclear to what extent the inability to capture treatment with zoledronate affects the treatment gap during the study period (2010 to 2019). Data from the National Hip Fracture Database (NHFD) indicates that in 2020 only 9% of patients were started on zoledronate following a hip fracture, 22 , 23 and in 2021 NOGG guidance was updated to recommend zoledronate as a first‐line treatment. 21 , 24 Prior to 2021, the recommendation was to use zoledronate if first‐line treatments, alendronate and risedronate, were not tolerated or contraindicated. 25 , 26 However, given that people with PD commonly experience difficulty swallowing, 27 perhaps oral bisphosphonates were deemed inappropriate and zoledronate was considered a suitable alternative in some instances.
Therefore, the inability to capture zoledronate may contribute to an over‐estimation of the identified treatment gap, however this gap is unlikely to be fully explained by this.
We excluded missing data from the treatment gap analysis. However, it is also important to consider those who did not receive a prescription before the last collection date for the practice, the end of the study period or who transferred to a practice without CPRD follow‐up. It is possible that these patients were later provided osteoporosis treatment. Amongst patients who died within 48 weeks and did not receive bone protective medication or supplementation, this may indicate a clinical decision that treatment was unlikely to be beneficial due to limited life expectancy.
When exploring demographic factors as predictors for bone protective medication prescription, it was unsurprising that being female compared to male increased the odds of prescriptions (OR 2.06 [95% CI (1.69 to 2.51)], p<0.001), given the known gender disparities in the treatment of osteoporosis. Recent work has demonstrated that in vertebral fracture cohorts the odds ratio for treatment was 1.82 favoring women 28 and, at hospital discharge following a hip fracture, only 2.7% (n = 3/110) of men compared to 27% (n = 60/253) of women were on anti‐resorptive treatment in a single center study in the United States. 29
Fracture type also affected the odds of receiving a prescription for bone protective medications. Vertebral fractures (OR 2.06 [1.53 to 2.77], p < 0.001) and unspecified osteoporotic fractures (OR 1.57 [1.17 to 2.10], p < 0.001) when compared to hip fracture increased the odds, whereas rib fractures (OR 0.34 [0.19 to 0.60], p < 0.001) when compared to hip fractures decreased the odds of prescription. We hypothesize that this could be explained by the different inpatient treatment received, although this cannot be addressed using CPRD data.
In the UK, rib fractures, unlike hip fractures, are more variably managed in different inpatient locations besides geriatrics, including orthopedics and general surgery 30 where there may be less familiarity with the use of osteoporosis treatments potentially leading to a discrepancy in the recommendation for bone protective medications. The finding that those with vertebral versus hip fracture have increased odds of receiving a prescription reflects findings from the US that following a vertebral fracture there is the largest increase in osteoporosis diagnosis and treatment rates compared to other fracture types. 31
Care home residency also reduced the odds of prescription after adjustment for age, gender and number of comorbidities (OR 0.58 [0.38 to 0.88], p = 0.01). However, the evidence for this was weakened when restricting the analysis to those with an initial diagnosis of PD or to those with 48 weeks of data before and after the index fracture. This is most likely explained by the small number of patients who were resident in a care home in these sub‐groups (Table A6 and A7) therefore widening the confidence interval.
This result is consistent with findings from US care home residents with dementia and high fracture risk, where only 11.6% received fracture prevention medications. 32 The authors also highlighted a potential lack of awareness about the appropriateness of treatment, as 10% of those receiving hospice care and 11% of individuals who were not dependent for activities of daily living were receiving medication for fracture prevention reflecting over and under‐treatment respectively. 32 Whilst the appropriateness of treatment could not be assessed in our study, it is important to consider that treatment for osteoporosis should be considered on a case‐by‐case basis accounting for several patient factors regardless of whether the patient resides in a care home.
Sensitivity analysis restricted to those with 48‐weeks of data before and after the index fracture
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (Unadjusted) | |||
| Hip | ref | 765 | |
| Vertebral | 1.64 (1.05 to 2.56) | 0.03 | |
| Wrist | 0.58 (0.32 to 1.05) | 0.07 | |
| Humerus | 0.67 (0.39 to 1.13) | 0.13 | |
| Rib | 0.22 (0.10 to 0.48) | <0.001 | |
| Pelvis | 0.56 (0.25 to 1.27) | 0.17 | |
| Unspecified | 1.57 (1.05 to 2.36) | 0.03 | |
| Multiple Fractures | 1.31 (0.37 to 4.63) | 0.67 | |
| Fracture type adjusted for age | |||
| Hip | ref | 765 | |
| Vertebral | 1.73 (1.10 to 2.71) | 0.02 | |
| Wrist | 0.62 (0.34 to 1.14) | 0.13 | |
| Humerus | 0.69 (0.40 to 1.17) | 0.17 | |
| Rib | 0.24 (0.11 to 0.52) | <0.001 | |
| Pelvis | 0.57 (0.25 to 1.30) | 0.18 | |
| Unspecified | 1.61 (1.07 to 2.42) | 0.02 | |
| Multiple Fractures | 1.36 (0.38 to 4.84) | 0.64 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 765 | |
| Vertebral | 1.78 (1.12 to 2.82) | 0.01 | |
| Wrist | 0.50 (0.27 to 0.94) | 0.03 | |
| Humerus | 0.60 (0.35 to 1.03) | 0.06 | |
| Rib | 0.26 (0.12 to 0.57) | < 0.001 | |
| Pelvis | 0.48 (0.21 to 1.10) | 0.08 | |
| Unspecified | 1.44 (0.95 to 2.19) | 0.09 | |
| Multiple Fractures | 1.30 (0.36 to 4.69) | 0.69 | |
| Cambridge multimorbidity score unadjusted | |||
| 0/1 | ref | 765 | |
| 2 | 0.66 (0.29 to 1.48) | 0.31 | |
| 3 | 0.85 (0.40 to 1.81) | 0.67 | |
| 4 | 1.02 (0.48 to 2.16) | 0.96 | |
| 5 | 0.85 (0.39 to 1.84) | 0.68 | |
| 6 | 1.16 (0.53 to 2.56) | 0.71 | |
| 7+ | 1.08 (0.51 to 2.29) | 0.84 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 765 | |
| 2 | 0.60 (0.26 to 1.37) | 0.23 | |
| 3 | 0.79 (0.36 to 1.70) | 0.54 | |
| 4 | 0.90 (0.42 to 1.93) | 0.78 | |
| 5 | 0.76 (0.34 to 1.66) | 0.48 | |
| 6 | 0.97 (0.43 to 2.17) | 0.93 | |
| 7+ | 0.91 (0.42 to 1.95) | 0.80 | |
| Care home residence (unadjusted) a | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.54 (0.24 to 1.18) | 0.12 | |
| Care home residence adjusted for age | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.48 (0.22 to 1.07) | 0.07 | |
| Care home residence adjusted for age and gender | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.47 (0.21 to 1.05) | 0.07 | |
| Care home residence adjusted for age, gender and CMS | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.44 (0.20 to 1.00) | 0.05 | |
There are now 31 (4.1%) of patients living in a care home.
Sensitivity analysis restricted to those with an initial diagnosis of Parkinson's disease
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (unadjusted) | |||
| Hip | ref | 1521 | |
| Vertebral | 1.79 (1.31 to 2.43) | < 0.001 | |
| Wrist | 0.64 (0.40 to 1.02) | 0.06 | |
| Humerus | 0.72 (0.50 to 1.04) | 0.08 | |
| Rib | 0.25 (0.13 to 0.46) | < 0.001 | |
| Pelvis | 1.00 (0.57 to 1.73) | 0.99 | |
| Unspecified | 1.69 (1.24 to 2.30) | < 0.001 | |
| Multiple fractures | 1.38 (0.61 to 3.12) | 0.44 | |
| Fracture type adjusted for age | |||
| Hip | ref | 1521 | |
| Vertebral | 1.91 (1.40 to 2.61) | < 0.001 | |
| Wrist | 0.72 (0.45 to 1.16) | 0.18 | |
| Humerus | 0.74 (0.51 to 1.08) | 0.12 | |
| Rib | 0.28 (0.15 to 0.53) | < 0.001 | |
| Pelvis | 1.00 (0.57 to 1.75) | 1.00 | |
| Unspecified | 1.79 (1.31 to 2.44) | < 0.001 | |
| Multiple fractures | 1.40 (0.61 to 3.19) | 0.42 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 1521 | |
| Vertebral | 2.04 (1.48 to 2.81) | < 0.001 | |
| Wrist | 0.56 (0.34 to 0.91) | 0.02 | |
| Humerus | 0.64 (0.44 to 0.94) | 0.02 | |
| Rib | 0.31 (0.16 to 0.58) | < 0.001 | |
| Pelvis | 0.85 (0.48 to 1.49) | 0.56 | |
| Unspecified | 1.53 (1.11 to 2.10) | 0.01 | |
| Multiple Fractures | 1.31 (0.56 to 3.02) | 0.53 | |
| Cambridge multimorbidity score unadjusted | |||
| 0/1 | ref | 1521 | |
| 2 | 0.80 (0.48 to 1.36) | 0.42 | |
| 3 | 0.80 (0.48 to 1.33) | 0.39 | |
| 4 | 0.88 (0.53 to 1.45) | 0.61 | |
| 5 | 0.95 (0.58 to 1.57) | 0.85 | |
| 6 | 0.97 (0.58 to 1.63) | 0.92 | |
| 7+ | 0.87 (0.53 to 1.43) | 0.58 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 1521 | |
| 2 | 0.74 (0.44 to 1.27) | 0.28 | |
| 3 | 0.74 (0.44 to 1.23) | 0.24 | |
| 4 | 0.79 (0.48 to 1.31) | 0.36 | |
| 5 | 0.84 (0.50 to 1.40) | 0.49 | |
| 6 | 0.83 (0.49 to 1.41) | 0.50 | |
| 7+ | 0.76 (0.46 to 1.26) | 0.29 | |
| Care home residence (Unadjusted) a | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.75 (0.48 to 1.17) | 0.20 | |
| Care home residence adjusted for age | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.68 (0.44 to 1.07) | 0.10 | |
| Care home residence adjusted for age and gender | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.69 (0.44 to 1.08) | 0.11 | |
| Care home residence adjusted for age, gender and CMS | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.68 (0.43 to 1.08) | 0.10 | |
There are now 94 (6.2%) of patients living in a care home.
What strengths and limitations does the paper describe?
This study is strengthened by the ability to assess the management of osteoporosis in a large, representative cohort of people with Parkinsonism across the United Kingdom. By using CPRD data it is possible to include previously under‐represented groups such as those who are older, with more advanced parkinsonism, care home residents, as well as people across the spectrum of parkinsonian disorders. 33 This is of particular importance given that people with atypical parkinsonism have greater fracture risk, alongside an osteoporosis treatment gap, 34 , 35 and there is often diagnostic uncertainty. 19
It was beyond the scope of this paper to ascertain whether patients were adherent to prescribed treatment or whether health care professional follow‐up was provided to maximize adherence. 14 Adherence to osteoporosis treatment is a well‐documented issue; persistence to oral bisphosphonates amongst post‐menopausal women is reported to be only 28% at 12 months. 36 Therefore, despite prescriptions being made it is likely that many patients were still without treatment following a fragility fracture.
When determining if patients were on treatment prior to fracture, for oral bisphosphonates patients were considered “on‐treatment” if they received a prescription in the 3 months prior to their fracture. However, this could mis‐classify participants if they were on treatment holiday after previously being prescribed oral bisphosphonates. It was also likely that patients being given zoledronate are mis‐categorized as “untreated” as it is not administered in the primary care setting and is therefore not reliably recorded within the CPRD data.
Furthermore, patients could have been misclassified as “not taking supplementation” if they are using over‐the‐counter vitamin D and/or calcium. However over‐the‐counter formulations may be sub‐standard in comparison to licensed medical products. 37 All of which could contribute to the observed treatment gaps.
As previously discussed, it is also possible that treatment options were considered and deemed inappropriate or declined by patients; however, this information is not routinely coded in CPRD data. It is also not possible using CPRD data to determine the circumstances surrounding the fracture. Whilst we included fracture sites typically associated with fragility fractures, it is possible that some of these fractures occurred from a trauma greater than a fall from standing height and are also contributing to an over‐estimation of the treatment gap as they are not indicative of osteoporosis.
Additionally, this population includes all people with parkinsonism who experienced a fracture between 2010 and 2019; in this study we consider the treatment decisions made after only the initial fracture which occurred within this period. However, it is possible that patients had multiple fracture events within this period or had fractures prior to 1st January 2010. Furthermore, the data available to us for analysis pre‐dates the updated NOGG guidance which mentions Parkinson's as an osteoporosis risk factor, 21 and our guidance on managing bone‐health in parkinsonism. 18 , 38 However, the treatment gap identified remains consistent with our 2025 findings that less than one third of people with parkinsonism are treated in‐line with the NOGG guidelines. 18
There remains a significant osteoporosis treatment gap within the parkinsonian population. Men with parkinsonism, those living in a care home, and those who sustain rib fractures are at increased odds of not receiving a prescription for bone protective medication following a fracture. Addressing this gap might reasonably be expected to prevent future fragility fractures and the associated serious and negative sequalae in this vulnerable and at‐risk population.
Source: Strengths and Limitations Source: Strengths and Limitations Source: Strengths and Limitations Source: Strengths and Limitations Source: Strengths and Limitations Source: Strengths and LimitationsAuthor Roles
(1) Research project: A. Conception, B. Organization, C. Execution, (2) Statistical analysis: A. Design, B. Execution, C. Review and critique; (3) Manuscript preparation: A. Writing of the first draft, B. Review and critique.
K.C.N.: 1A, 1B, 1C, 2A, 2B, 3A.
E.J.H.: 1A, 1B, 1C, 2A, 2C, 3B.
E.T.: 1A, 1B, 1C, 2A, 2C, 3B.
Source: Author Roles Source: Author Roles Source: Author Roles Source: Author RolesResults
We identified 21,792 people with parkinsonism, of whom 211 were excluded (n = 1 and n = 136 due to censoring before the study period or first parkinsonism code respectively; n = 33 due to missing prescription data; n = 41 due to age Fig. 1). The mean duration of available data was 3.1 years (SD = 2.5 years, range 1 day to 10.0 years). The mean age of patients at entry to the cohort was 74.7 years (SD = 9.9).
Flowchart showing the inclusion/exclusion of people from study population.
There were 1823 (8.4%) people with parkinsonism who had at least one fragility fracture during the study period. The incidence rate of fragility fractures was 27.3 per 1000 person‐years (95% CI = 26.1 to 28.6). The mean age at the time of fracture was 78.9 (SD = 7.79) years. Of those who experienced a fracture within the study period, hip fractures were the most common, experienced by 813 people (44.6%), followed by vertebral fractures in 255 people (14.0%). Within a 31‐day period, which we assumed to relate to the same fracture event, only 29 (1.6%) people experienced two different fractures and 1 (0.1%) experienced three (Table 1).
Demographic characteristics of the parkinsonian cohort identified in clinical practice research datalink (CPRD) 2010 to 2019, and the fracture subgroup
| Full cohort | Fracture subgroup | |
|---|---|---|
| Mean (SD) or N (%) | Mean (SD) or N (%) | |
| N | 21,581 | 1823 |
| Age upon entry to cohort a | 74.7 (9.8) | 76.5 (8.1) |
| Sex | ||
| Male | 12,942 (60.0%) | 751 (41.2%) |
| Female | 8639 (40.0%) | 1072 (58.8%) |
| PD duration (years) | 1.4 (3.5) | 1.9 (3.7) |
| Diagnosis | ||
| Parkinson's disease | 17,897 (82.9%) | 1521 (83.4%) |
| Dementia with Lewy bodies | 1716 (8.0%) | 122 (6.7%) |
| Vascular Parkinsonism | 929 (4.3%) | 78 (4.3%) |
| Progressive supranuclear palsy | 374 (1.7%) | 37 (2.0%) |
| Parkinson's disease dementia | 206 (1.0%) | 15 (0.8%) |
| Multiple system atrophy | 166 (0.8%) | 10 (0.5%) |
| Corticobasal degeneration | 65 (0.3%) | 4 (0.2%) |
| Parkinsonism | 12 (0.1%) | 0 |
| Other b | 216 (1.0%) | 36 (2.0%) |
| Years of available CPRD data | 3.1 (2.5) | 4.3 (2.6) |
| Fragility fracture present in study period | ||
| None | 19,758 (91.6%) | Not reported |
| One or more | 1823 (8.4%) | Not reported |
| How many individual fractures occurred in this first event? | ||
| 1 Fracture | Not reported | 1793 (98.4%) |
| 2 Fractures | Not reported | 29 (1.6%) |
| 3 Fractures | Not reported | 1 (0.1%) |
| Age at fracture | Not reported | 78.9 (7.8) |
| Fracture site c | ||
| Hip | Not reported | 813 (44.6%) |
| Vertebral | Not reported | 255 (14.0%) |
| Wrist | Not reported | 125 (6.9%) |
| Humerus | Not reported | 197 (10.8%) |
| Rib | Not reported | 115 (6.3%) |
| Pelvis | Not reported | 78 (4.3%) |
| Unspecified | Not reported | 269 (14.8%) |
This is age on January 1, 2010 for those diagnosed with Parkinson's before the study period began, or age on the day of Parkinson's diagnosis if diagnosed during the study period.
This includes the Read codes for paralysis agitans, Parkinson's orthostatic hypotension and cerebral degeneration PD, all of which indicate Parkinsonism.
These frequencies reflect the number of participants whom experienced a fracture at each site rather than the number of fractures.
Prescriptions for Bone Protective Medications and Supplementation at Indexed Fracture Event
Of those with a fragility fracture event, 225 people (12.3%) were on any bone protective medication prior to their fracture. A further 486 people were started on bone protective medication within 48 weeks following their fracture, therefore cumulatively 52.9% were in receipt of a prescription within 48 weeks (Fig. 2A and Table A4). Four hundred and forty two (24.2%) were prescribed vitamin D and/or calcium supplementation prior to their fracture, and a further 651 people were prescribed supplementation up to 48 weeks after their fracture, therefore cumulatively 73.2% were in receipt of a prescription within 48 weeks (Fig. 2B and Table A5).
Stacked bar charts illustrating the prescription of (A) bone protective medication (B) vitamin D/calcium supplementation prior to and following a fragility fracture.
Incomplete Follow‐Up
A total of 479 patients were censored from the bone protective medication data within 48 weeks of a fragility fracture, due to 123 deaths, 53 patients reaching the last point of data collection from the practice, 207 transferring to a GP practice without CPRD linkage, and 96 reaching the end of the study period prior to a prescription being issued (Figure A2). Of the 1598, who were not prescribed bone protective medication prior to their fragility fracture, the mean follow‐up time was 1.2 years (SD = 1.6 years, range 0 days to 9.7 years).
With respect to vitamin D and/or calcium prescriptions, a total of 330 patients were censored within 48 weeks of a fragility fracture, due to 92 deaths, 37 patients reaching the last point of data collection from the practice, 149 transferring to a GP practice without CPRD linkage, and 52 reaching the end of the study period prior to a prescription being issued (Figure A3). Of the 1381, who were not prescribed vitamin D and/or calcium prior to their fragility fracture, the mean follow‐up time was 0.9 years (SD = 1.5 years, range 0 days to 9.7 years).
Predictors of Bone Protective Prescriptions
We explored demographic factors and fracture site as predictors of bone protective prescriptions being issued (Table 2). Being female (OR 2.06 [95% CI (1.69 to 2.51)], p < 0.001) and being aged 75 to 79 (OR 1.82 [95% CI (1.40 to 2.37)], p < 0.001), 80 to 84 (OR 1.67 [95% CI (1.28 to 2.17)], p < 0.001) or over 85 (OR 1.63 [95% CI (1.24 to 2.16)], p = 0.001), increased the odds of bone protective medication prescription. Residing in a care home reduced the odds of bone protective medication prescription (OR 0.64 [95% CI (0.43 to 0.96)], p = 0.03).
Sustaining a vertebral fracture (OR 1.84 [95% CI (1.38 to 2.45)], p < 0.001), or having unspecified fragility fracture recorded (OR 1.76 [95% CI (1.33 to 2.33)], p < 0.001), when compared to sustaining a hip fracture, also increased the odds of bone protective medication prescription, whereas sustaining a rib fracture (OR 0.28 [95% CI (0.16 to 0.49)], p < 0.001) reduced the odds.
Unadjusted logistic regression models to look at factors predicting bone protective medication prescriptions
| Unadjusted odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Gender | |||
| Male | ref | 1823 | |
| Female | 2.06 (1.69 to 2.51) | <0.001 | |
| Age | |||
| 35 to 74 | ref | 1823 | |
| 74 to 79 | 1.82 (1.40 to 2.37) | <0.001 | |
| 80 to 84 | 1.67 (1.28 to 2.17) | <0.001 | |
| 85+ | 1.63 (1.24 to 2.16) | 0.001 | |
| PD duration | |||
| Less than 1 year | ref | 1823 | |
| 1 to 2.4 years | 0.78 (0.59 to 1.03) | 0.08 | |
| 2.5 to 4 years | 0.80 (0.61 to 1.05) | 0.11 | |
| 5 to 9 years | 0.79 (0.59 to 1.05) | 0.10 | |
| 10+ years | 0.76 (0.54 to 1.07) | 0.12 | |
| Fracture type | |||
| Hip | ref | 1823 | |
| Vertebral | 1.84 (1.38 to 2.45) | <0.001 | |
| Wrist | 0.68 (0.44 to 1.05) | 0.08 | |
| Humerus | 0.80 (0.57 to 1.12) | 0.19 | |
| Rib | 0.28 (0.16 to 0.49) | <0.001 | |
| Pelvis | 1.14 (0.71 to 1.85) | 0.58 | |
| Unspecified | 1.76 (1.33 to 2.33) | <0.001 | |
| Multiple Fractures | 1.30 (0.62 to 2.71) | 0.49 | |
| Cambridge multimorbidity score | |||
| 0/1 | ref | 1823 | |
| 2 | 0.82 (0.51 to 1.33) | 0.42 | |
| 3 | 0.83 (0.53 to 1.31) | 0.43 | |
| 4 | 0.91 (0.58 to 1.43) | 0.68 | |
| 5 | 0.93 (0.59 to 1.46) | 0.74 | |
| 6 | 0.95 (0.59 to 1.52) | 0.83 | |
| 7+ | 0.88 (0.56 to 1.38) | 0.57 | |
| Care home residence | |||
| Not living in care home | ref | 1823 | |
| Living in care home | 0.64 (0.43 to 0.96) | 0.03 | |
When adjusting for age and gender, vertebral fractures (OR 2.06 [95% CI (1.53 to 2.77)], p < 0.001) had stronger association with receiving a prescription whereas unspecified fracture (OR 1.57 [95% CI (1.17 to 2.10)], p < 0.001), and rib fractures (OR 0.34 [95% CI (0.19 to 0.60)], p < 0.001) both had a slightly attenuated relationship with receiving a prescription for bone protective medication versus hip fracture. When adjusting for age, multi‐morbidity (CMS) and gender, living in a care home (OR 0.58 [95% CI (0.38 to 0.88)], p = 0.01) had a stronger association of decreasing the odds of receiving a prescription than in the univariate model.
Multimorbidity still did not change the odds of receiving a prescription after adjusting for age (Table 3). The direction of these associations was not altered when restricting the analysis to those with an initial diagnosis of PD or those with 48 weeks of data before and after the index fracture, although it did weaken the association between prescriptions and care home residency (Table A6 and A7).
Minimally adjusted logistic regression models to look at factors predicting bone protective medication prescriptions
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (unadjusted) | |||
| Hip | ref | 1823 | |
| Vertebral | 1.84 (1.38 to 2.45) | <0.001 | |
| Wrist | 0.68 (0.44 to 1.05) | 0.08 | |
| Humerus | 0.80 (0.57 to 1.12) | 0.19 | |
| Rib | 0.28 (0.16 to 0.49) | <0.001 | |
| Pelvis | 1.14 (0.71 to 1.85) | 0.58 | |
| Unspecified | 1.76 (1.33 to 2.33) | <0.001 | |
| Multiple Fractures | 1.30 (0.62 to 2.71) | 0.49 | |
| Fracture type adjusted for age | |||
| Hip | ref | 1823 | |
| Vertebral | 1.98 (1.48 to 2.65) | <0.001 | |
| Wrist | 0.75 (0.49 to 1.17) | 0.21 | |
| Humerus | 0.82 (0.58 to 1.16) | 0.26 | |
| Rib | 0.31 (0.18 to 0.55) | <0.001 | |
| Pelvis | 1.15 (0.71 to 1.86) | 0.57 | |
| Unspecified | 1.84 (1.39 to 2.44) | <0.001 | |
| Multiple fractures | 1.29 (0.62 to 2.71) | 0.50 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 1823 | |
| Vertebral | 2.06 (1.53 to 2.77) | <0.001 | |
| Wrist | 0.60 (0.38 to 0.93) | 0.02 | |
| Humerus | 0.72 (0.50 to 1.02) | 0.06 | |
| Rib | 0.34 (0.19 to 0.60) | <0.001 | |
| Pelvis | 0.99 (0.61 to 1.61) | 0.96 | |
| Unspecified | 1.57 (1.17 to 2.10) | <0.001 | |
| Multiple fractures | 1.23 (0.58 to 2.61) | 0.59 | |
| Cambridge multimorbidity score (unadjusted) | |||
| 0/1 | ref | 1823 | |
| 2 | 0.82 (0.51 to 1.33) | 0.42 | |
| 3 | 0.83 (0.53 to 1.31) | 0.43 | |
| 4 | 0.91 (0.58 to 1.43) | 0.68 | |
| 5 | 0.93 (0.59 to 1.46) | 0.74 | |
| 6 | 0.95 (0.59 to 1.52) | 0.83 | |
| 7+ | 0.88 (0.56 to 1.38) | 0.57 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 1823 | |
| 2 | 0.77 (0.47 to 1.25) | 0.29 | |
| 3 | 0.76 (0.48 to 1.21) | 0.25 | |
| 4 | 0.81 (0.51 to 1.28) | 0.37 | |
| 5 | 0.81 (0.51 to 1.28) | 0.36 | |
| 6 | 0.82 (0.51 to 1.32) | 0.41 | |
| 7+ | 0.77 (0.49 to 1.21) | 0.26 | |
| Care home residency (unadjusted) | |||
| No | ref | 1823 | |
| Yes | 0.64 (0.43 to 0.96) | 0.03 | |
| Care home residency adjusted for age | |||
| No | ref | 1823 | |
| Yes | 0.58 (0.39 to 0.88) | 0.01 | |
| Care home residency adjusted for age and gender | |||
| No | ref | 1823 | |
| Yes | 0.58 (0.38 to 0.88) | 0.01 | |
| Care home residency adjusted for age, gender and CMS | |||
| No | ref | 1823 | |
| Yes | 0.58 (0.38 to 0.88) | 0.01 | |
Table of bone protective medication prescriptions over 48 weeks following a fragility fracture
| Timeline | Censored from study (N) | Number of participants a | Oral bisphosphonate | Strontium ranelate | Denosumab | Zoledronate | All |
|---|---|---|---|---|---|---|---|
| Before fracture 1 | 0 | 1823 | 217 | 6 | 2 | 0 | 225 |
| (Percentage) | (11.9%) | (0.3%) | (0.1%) | (0.0%) | (12.3%) | ||
| Up to 16 weeks after | 236 | 1587 | 572 | 17 | 5 | 3 | 597 |
| (Percentage) | (36.0%) | (1.1%) | (0.3%) | (0.2%) | (37.6%) | ||
| Up to 32 weeks after | 369 | 1454 | 653 | 18 | 9 | 3 | 683 |
| (Percentage) | (44.9%) | (1.2%) | (0.6%) | (0.2%) | (47.0%) | ||
| Up to 48 weeks after | 479 | 1344 | 678 | 18 | 12 | 3 | 711 |
| (Percentage) | (50.4%) | (1.3%) | (0.9%) | (0.2%) | (52.9%) |
Note: This table displays the cumulative frequency of each medication type and all bone protective medications, starting before the fracture event followed by 16 week intervals.
The number of people with available data within each time window for bone protective medication.
Table of vitamin D and/or calcium supplement prescriptions over 48 weeks following a fragility fracture
| Timeline | Censored from study (N) | Number of participants a | Supplementation b |
|---|---|---|---|
| Before fracture 1 | 0 | 1823 | 442 |
| (Percentage) | (24.2%) | ||
| Up to 16 weeks after | 181 | 1642 | 951 |
| (Percentage) | (57.9%) | ||
| Up to 32 weeks after | 269 | 1554 | 1054 |
| (Percentage) | (67.8%) | ||
| Up to 48 weeks after | 330 | 1493 | 1093 |
| (Percentage) | (73.2%) |
Note: This table displays the cumulative frequency of supplementation (vitamin D and/or calcium) starting before the fracture event followed by 16 week intervals.
The number of people with available data within each time window for supplementation (Vitamin D and/or calcium).
Supplementation refers to a prescription being received for vitamin D and/or calcium.
Sensitivity analysis restricted to those with an initial diagnosis of Parkinson's disease
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (unadjusted) | |||
| Hip | ref | 1521 | |
| Vertebral | 1.79 (1.31 to 2.43) | < 0.001 | |
| Wrist | 0.64 (0.40 to 1.02) | 0.06 | |
| Humerus | 0.72 (0.50 to 1.04) | 0.08 | |
| Rib | 0.25 (0.13 to 0.46) | < 0.001 | |
| Pelvis | 1.00 (0.57 to 1.73) | 0.99 | |
| Unspecified | 1.69 (1.24 to 2.30) | < 0.001 | |
| Multiple fractures | 1.38 (0.61 to 3.12) | 0.44 | |
| Fracture type adjusted for age | |||
| Hip | ref | 1521 | |
| Vertebral | 1.91 (1.40 to 2.61) | < 0.001 | |
| Wrist | 0.72 (0.45 to 1.16) | 0.18 | |
| Humerus | 0.74 (0.51 to 1.08) | 0.12 | |
| Rib | 0.28 (0.15 to 0.53) | < 0.001 | |
| Pelvis | 1.00 (0.57 to 1.75) | 1.00 | |
| Unspecified | 1.79 (1.31 to 2.44) | < 0.001 | |
| Multiple fractures | 1.40 (0.61 to 3.19) | 0.42 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 1521 | |
| Vertebral | 2.04 (1.48 to 2.81) | < 0.001 | |
| Wrist | 0.56 (0.34 to 0.91) | 0.02 | |
| Humerus | 0.64 (0.44 to 0.94) | 0.02 | |
| Rib | 0.31 (0.16 to 0.58) | < 0.001 | |
| Pelvis | 0.85 (0.48 to 1.49) | 0.56 | |
| Unspecified | 1.53 (1.11 to 2.10) | 0.01 | |
| Multiple Fractures | 1.31 (0.56 to 3.02) | 0.53 | |
| Cambridge multimorbidity score unadjusted | |||
| 0/1 | ref | 1521 | |
| 2 | 0.80 (0.48 to 1.36) | 0.42 | |
| 3 | 0.80 (0.48 to 1.33) | 0.39 | |
| 4 | 0.88 (0.53 to 1.45) | 0.61 | |
| 5 | 0.95 (0.58 to 1.57) | 0.85 | |
| 6 | 0.97 (0.58 to 1.63) | 0.92 | |
| 7+ | 0.87 (0.53 to 1.43) | 0.58 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 1521 | |
| 2 | 0.74 (0.44 to 1.27) | 0.28 | |
| 3 | 0.74 (0.44 to 1.23) | 0.24 | |
| 4 | 0.79 (0.48 to 1.31) | 0.36 | |
| 5 | 0.84 (0.50 to 1.40) | 0.49 | |
| 6 | 0.83 (0.49 to 1.41) | 0.50 | |
| 7+ | 0.76 (0.46 to 1.26) | 0.29 | |
| Care home residence (Unadjusted) a | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.75 (0.48 to 1.17) | 0.20 | |
| Care home residence adjusted for age | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.68 (0.44 to 1.07) | 0.10 | |
| Care home residence adjusted for age and gender | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.69 (0.44 to 1.08) | 0.11 | |
| Care home residence adjusted for age, gender and CMS | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.68 (0.43 to 1.08) | 0.10 | |
There are now 94 (6.2%) of patients living in a care home.
Sensitivity analysis restricted to those with 48‐weeks of data before and after the index fracture
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (Unadjusted) | |||
| Hip | ref | 765 | |
| Vertebral | 1.64 (1.05 to 2.56) | 0.03 | |
| Wrist | 0.58 (0.32 to 1.05) | 0.07 | |
| Humerus | 0.67 (0.39 to 1.13) | 0.13 | |
| Rib | 0.22 (0.10 to 0.48) | <0.001 | |
| Pelvis | 0.56 (0.25 to 1.27) | 0.17 | |
| Unspecified | 1.57 (1.05 to 2.36) | 0.03 | |
| Multiple Fractures | 1.31 (0.37 to 4.63) | 0.67 | |
| Fracture type adjusted for age | |||
| Hip | ref | 765 | |
| Vertebral | 1.73 (1.10 to 2.71) | 0.02 | |
| Wrist | 0.62 (0.34 to 1.14) | 0.13 | |
| Humerus | 0.69 (0.40 to 1.17) | 0.17 | |
| Rib | 0.24 (0.11 to 0.52) | <0.001 | |
| Pelvis | 0.57 (0.25 to 1.30) | 0.18 | |
| Unspecified | 1.61 (1.07 to 2.42) | 0.02 | |
| Multiple Fractures | 1.36 (0.38 to 4.84) | 0.64 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 765 | |
| Vertebral | 1.78 (1.12 to 2.82) | 0.01 | |
| Wrist | 0.50 (0.27 to 0.94) | 0.03 | |
| Humerus | 0.60 (0.35 to 1.03) | 0.06 | |
| Rib | 0.26 (0.12 to 0.57) | < 0.001 | |
| Pelvis | 0.48 (0.21 to 1.10) | 0.08 | |
| Unspecified | 1.44 (0.95 to 2.19) | 0.09 | |
| Multiple Fractures | 1.30 (0.36 to 4.69) | 0.69 | |
| Cambridge multimorbidity score unadjusted | |||
| 0/1 | ref | 765 | |
| 2 | 0.66 (0.29 to 1.48) | 0.31 | |
| 3 | 0.85 (0.40 to 1.81) | 0.67 | |
| 4 | 1.02 (0.48 to 2.16) | 0.96 | |
| 5 | 0.85 (0.39 to 1.84) | 0.68 | |
| 6 | 1.16 (0.53 to 2.56) | 0.71 | |
| 7+ | 1.08 (0.51 to 2.29) | 0.84 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 765 | |
| 2 | 0.60 (0.26 to 1.37) | 0.23 | |
| 3 | 0.79 (0.36 to 1.70) | 0.54 | |
| 4 | 0.90 (0.42 to 1.93) | 0.78 | |
| 5 | 0.76 (0.34 to 1.66) | 0.48 | |
| 6 | 0.97 (0.43 to 2.17) | 0.93 | |
| 7+ | 0.91 (0.42 to 1.95) | 0.80 | |
| Care home residence (unadjusted) a | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.54 (0.24 to 1.18) | 0.12 | |
| Care home residence adjusted for age | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.48 (0.22 to 1.07) | 0.07 | |
| Care home residence adjusted for age and gender | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.47 (0.21 to 1.05) | 0.07 | |
| Care home residence adjusted for age, gender and CMS | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.44 (0.20 to 1.00) | 0.05 | |
There are now 31 (4.1%) of patients living in a care home.
Abstract: Conclusions
There is an osteoporosis treatment gap in Parkinson's rendering an already high‐risk population at further risk of fracture. Men and those with rib fractures are two groups who are least likely to receive potentially beneficial treatment.
Source: ConclusionsAuthors’ abstract
Abstract
Background
Fracture risk is increased in Parkinson's yet this risk is often not addressed.
Objectives
Our objective was to study the extent to which osteoporosis was treated, and predictors of treatment in a large representative cohort with parkinsonism.
Methods
The clinical practice research datalink (CPRD) GOLD data contains anonymized UK primary care data. We identified fragility fractures (hip, vertebrae, wrist/distal radius, humerus, rib, pelvis and unspecified osteoporotic) amongst prevalent parkinsonism patients between 2010 and 2019, and ascertained whether bone protective medications and/or vitamin D/calcium supplements were prescribed in the pre‐fracture period or in the 48‐weeks post‐fracture. Logistic regression was used to determine predictors of prescriptions.
Results
There were 21,581 people with parkinsonism (mean age 74.7 ± 9.8 years, 40% female), with a mean of 3.1 ± 2.5 years of available data. One thousand eight hundred twenty‐three experienced at least one fragility fracture. Prior to a fragility fracture 12% had a prescription for bone protective medication. This increased to 38% and 53% in the 16 and 48‐weeks post‐fracture, respectively. 24% already had a prescription for vitamin D/calcium which increased to 73% at 48‐weeks post‐fracture. Female gender (OR = 2.06 [1.69 to 2.51], p < 0.001) and sustaining a vertebral fracture (OR = 2.06 [1.53 to 2.77], p < 0.001) increased the odds of bone protection prescriptions.
However, sustaining a rib fracture (OR = 0.34 [0.19 to 0.60], p < 0.001) or residing in a care home (OR = 0.58 [0.38 to 0.88], p = 0.01) decreased the odds.
Conclusions
There is an osteoporosis treatment gap in Parkinson's rendering an already high‐risk population at further risk of fracture. Men and those with rib fractures are two groups who are least likely to receive potentially beneficial treatment.
Read the full paper
In Parkinson's disease (PD), lower bone mineral density (BMD) and increased incidence of falls lead to an increased fragility fracture risk. 1 , 2 Compared to age and gender‐matched controls, people with PD are at 2‐fold increased risk of osteoporotic fracture, and 3‐fold increased risk for hip fracture. 3 Furthermore, people with PD are more likely to develop complications 4 , 5 or have difficulty regaining mobility 6 following hip fracture and mortality is over twice that of age‐matched controls. 7
In Parkinson's care bone health and osteoporosis prevention and treatment are often overlooked. In their 2020 UK‐wide national audit, Parkinson's UK reported that, of 1131 patients, 73% did not have an up‐to‐date fracture risk assessment 8 ; and that only 18.5% of respondents recalled osteoporosis or fracture risk being discussed with specialists. 9 The lack of attention to fracture risk is further compounded by the fact that people with PD under‐estimate their fracture risk. 10
In the UK, according to the gold‐standard fracture liaison services (FLS), following a fragility fracture patients are expected to have a bone health assessment to determine if they are: low risk, and therefore offered life‐style advice to minimize their risk of future fractures; or high risk, and are therefore offered osteoporosis drug treatment.
Since bone health is often overlooked, we assessed the extent to which these treatment guidelines were followed in a population with parkinsonism following a fragility fracture. We sought to: (1) understand the extent to which bone protective medications and calcium/vitamin D were prescribed within 16 and 48‐weeks of a fragility fracture; (2) identify demographic and clinical factors which predict prescriptions being made using logistic regression.
Methods
Data Source
For this retrospective observational study, we used routinely‐collected general practice (GP) data within the Clinical Practice Research Datalink (CPRD) GOLD together with Office for National Statistics (ONS) mortality data. In 2013, CPRD contained 4.4 million active patients, representing 6.9% of the UK population from a network of 674 GP practices, with an additional 6.9 million inactive patients which includes those who have died or are no longer contributing data. 11 Medical events are recorded within the clinical and referral files using Read codes, and prescription medications are recorded using British National Formulary (BNF) coding.
CPRD therefore offers a unique opportunity to study the osteoporosis treatment gap in a large representative cohort of individuals with parkinsonism who have experienced at least one fragility fracture across the UK, where the treatment gap is defined as those who require treatment but are not receiving it within the given population.
Data Extraction
All individuals with a Read code for parkinsonism (excluding drug‐induced parkinsonism, Table A1) in the CPRD clinical and/or referral files who contributed data between January 1, 2010 and December, 31 2019 were extracted. Patients were included if they received a diagnosis of parkinsonism prior to or during the study period. Patients were excluded if: they were censored out of CPRD prior to the study period or the censor date was prior to the first instance of a Read code for parkinsonism, since this suggested a likely data error; they had a Read code for parkinsonism prior to age 35; or there was no available data on primary care prescriptions (CPRD therapy data).
This is because Read codes prior to age 35 potentially relate to a coding error, and the lack of prescription data indicates that there is likely to be further missing data.
Study Population
The index date was the date of the first fragility fracture which occurred in the study period (January 1, 2010: December 31, 2019) for patients with prevalent parkinsonism, registered at a practice contributing data deemed to be of research quality, which is determined by the recording and consistency of variables including practice registration date, transfer out date and date of birth. 12 Patients were followed‐up until they received a prescription for bone protective medication up to a maximum of 48‐weeks; or until they were censored, whichever came first.
This was repeated considering prescriptions for vitamin D and/or calcium supplementation. The censor date was defined as the earliest of: date of death; the last day of data collection at a given GP practice; the date of transfer to another practice without CPRD data‐linkage, or the study end date. If patients received a prescription prior to their fragility fracture, they were not followed up beyond the index date.
In this study, fragility fracture sites considered were hip, vertebrae, wrist/distal radius, humerus, rib, pelvis and unspecified osteoporotic fractures. We followed the methods described by Watson et al 2017 to form a list of Read codes indicating the presence of a fragility fracture (fracture codes) (Table A2). 13 The list was reviewed by two independent clinicians and any discrepancies were agreed by a third clinician. These Read codes were then used to identify patients who experienced at least one fragility fracture during the study period.
Fragility fractures were grouped into fracture events, and the initial fracture event within the study period was retained in the analysis as the index fracture event. Exact duplicates were removed from the dataset. To avoid double counting of fracture events, if two or more fracture codes related to the same fracture site or if an osteoporotic fracture at an unspecified site occurred less than 31 days apart, they were considered to be the same fracture event. Fracture events could consist of more than one specified fracture site if they occurred less than 31 days apart, and information about each fracture present at the event was retained.
Prescription Data
Prescription data relating to either bone protective medications, or vitamin D and/or calcium supplementation were extracted from the CPRD therapy file using the BNF chapters 060602 (bone protective medications), 090604 (vitamin D supplementation), and 09050101 (calcium supplementation). Additionally, the medication descriptions were searched for relevant terms (Table A3), as some drugs were listed in other BNF chapters. Medications were cross‐referenced with the NICE guidelines, and drugs which were not licensed for use in osteoporosis, calcium or vitamin D deficiency, were removed.
Pamidronate was excluded from this list as it is typically used in the context of malignancy. All prescriptions from the CPRD therapy file refer to drugs prescribed/dispensed in primary care for patients in all settings (own home, assisted living, care home etc).
Using index fragility fracture date, we determined whether patients were prescribed bone protective medications or vitamin D and/or calcium supplements before, or whether treatment was initiated after the fracture event. Patients were considered on treatment at the fracture event if prescription for oral bisphosphonates, strontium ranelate or teriparatide were received up to 3 months before; or if denosumab or zoledronate were documented in their medical records up to 9 or 18 months before respectively. These treatment windows were chosen using the standard dosing regimen for denosumab and zoledronate allowing for potential delays in treatment administration.
Based on the UK Fracture Liaison Service gold‐standard, stating that appropriate osteoporosis drug treatment should be commenced within 16‐weeks of a fragility fracture, 14 we analyzed 16‐week time intervals following the fracture event to determine adherence to this standard.
Incomplete Follow‐Up
We considered, within each 16 week time interval, how many patients remained within the study and therefore could have possibly received a prescription. Patients no longer remained within the study if they were censored out (eg, death, transfer out) prior to receiving a prescription. This was considered separately for the bone protective and supplementation prescription data. The patients remaining within each time interval are reported and used as the denominator to calculate the proportion on treatment within each time window.
The total number of patients censored from both the bone protective medications and supplementation data, and the reasons for censorship, are also described.
Bone protective medication and supplementation prescription data were considered separately, since these may not be initiated concurrently, potentially resulting in a patient being censored after prescription of only one medication type.
Statistical Analysis
Stata 18 was used for the analysis. 15 Data are described using mean/standard deviation, and the incidence of fractures is reported per 1000 person‐years with 95% confidence intervals. Logistic regression was used to obtain the prescription probability within a clinically meaningful time window and to test potential predictors of receiving prescriptions including: gender, age (at the mid‐point of the year of birth to the index fragility fracture date), PD duration at the time of fracture (calculated from the date at which the first diagnostic code for parkinsonism was applied), fracture site, Cambridge Multimorbidity Score (CMS) and whether the patient was resident in a care home at the time of fracture.
The CMS quantifies multimorbidity using a count of 37 conditions based on diagnostic and prescription codes. 16 Minimally adjusted models were determined using directed acyclic graphs (Figure A1). Two sensitivity analyses were conducted, restricting the analysis to: (i) patients with an initial diagnosis of PD; (ii) patients with 48‐weeks of data before and after the index fracture to test whether data availability affected any observed relationships.
Results
We identified 21,792 people with parkinsonism, of whom 211 were excluded (n = 1 and n = 136 due to censoring before the study period or first parkinsonism code respectively; n = 33 due to missing prescription data; n = 41 due to age Fig. 1). The mean duration of available data was 3.1 years (SD = 2.5 years, range 1 day to 10.0 years). The mean age of patients at entry to the cohort was 74.7 years (SD = 9.9).
Flowchart showing the inclusion/exclusion of people from study population.
There were 1823 (8.4%) people with parkinsonism who had at least one fragility fracture during the study period. The incidence rate of fragility fractures was 27.3 per 1000 person‐years (95% CI = 26.1 to 28.6). The mean age at the time of fracture was 78.9 (SD = 7.79) years. Of those who experienced a fracture within the study period, hip fractures were the most common, experienced by 813 people (44.6%), followed by vertebral fractures in 255 people (14.0%). Within a 31‐day period, which we assumed to relate to the same fracture event, only 29 (1.6%) people experienced two different fractures and 1 (0.1%) experienced three (Table 1).
Demographic characteristics of the parkinsonian cohort identified in clinical practice research datalink (CPRD) 2010 to 2019, and the fracture subgroup
| Full cohort | Fracture subgroup | |
|---|---|---|
| Mean (SD) or N (%) | Mean (SD) or N (%) | |
| N | 21,581 | 1823 |
| Age upon entry to cohort a | 74.7 (9.8) | 76.5 (8.1) |
| Sex | ||
| Male | 12,942 (60.0%) | 751 (41.2%) |
| Female | 8639 (40.0%) | 1072 (58.8%) |
| PD duration (years) | 1.4 (3.5) | 1.9 (3.7) |
| Diagnosis | ||
| Parkinson's disease | 17,897 (82.9%) | 1521 (83.4%) |
| Dementia with Lewy bodies | 1716 (8.0%) | 122 (6.7%) |
| Vascular Parkinsonism | 929 (4.3%) | 78 (4.3%) |
| Progressive supranuclear palsy | 374 (1.7%) | 37 (2.0%) |
| Parkinson's disease dementia | 206 (1.0%) | 15 (0.8%) |
| Multiple system atrophy | 166 (0.8%) | 10 (0.5%) |
| Corticobasal degeneration | 65 (0.3%) | 4 (0.2%) |
| Parkinsonism | 12 (0.1%) | 0 |
| Other b | 216 (1.0%) | 36 (2.0%) |
| Years of available CPRD data | 3.1 (2.5) | 4.3 (2.6) |
| Fragility fracture present in study period | ||
| None | 19,758 (91.6%) | Not reported |
| One or more | 1823 (8.4%) | Not reported |
| How many individual fractures occurred in this first event? | ||
| 1 Fracture | Not reported | 1793 (98.4%) |
| 2 Fractures | Not reported | 29 (1.6%) |
| 3 Fractures | Not reported | 1 (0.1%) |
| Age at fracture | Not reported | 78.9 (7.8) |
| Fracture site c | ||
| Hip | Not reported | 813 (44.6%) |
| Vertebral | Not reported | 255 (14.0%) |
| Wrist | Not reported | 125 (6.9%) |
| Humerus | Not reported | 197 (10.8%) |
| Rib | Not reported | 115 (6.3%) |
| Pelvis | Not reported | 78 (4.3%) |
| Unspecified | Not reported | 269 (14.8%) |
This is age on January 1, 2010 for those diagnosed with Parkinson's before the study period began, or age on the day of Parkinson's diagnosis if diagnosed during the study period.
This includes the Read codes for paralysis agitans, Parkinson's orthostatic hypotension and cerebral degeneration PD, all of which indicate Parkinsonism.
These frequencies reflect the number of participants whom experienced a fracture at each site rather than the number of fractures.
Prescriptions for Bone Protective Medications and Supplementation at Indexed Fracture Event
Of those with a fragility fracture event, 225 people (12.3%) were on any bone protective medication prior to their fracture. A further 486 people were started on bone protective medication within 48 weeks following their fracture, therefore cumulatively 52.9% were in receipt of a prescription within 48 weeks (Fig. 2A and Table A4). Four hundred and forty two (24.2%) were prescribed vitamin D and/or calcium supplementation prior to their fracture, and a further 651 people were prescribed supplementation up to 48 weeks after their fracture, therefore cumulatively 73.2% were in receipt of a prescription within 48 weeks (Fig. 2B and Table A5).
Stacked bar charts illustrating the prescription of (A) bone protective medication (B) vitamin D/calcium supplementation prior to and following a fragility fracture.
Incomplete Follow‐Up
A total of 479 patients were censored from the bone protective medication data within 48 weeks of a fragility fracture, due to 123 deaths, 53 patients reaching the last point of data collection from the practice, 207 transferring to a GP practice without CPRD linkage, and 96 reaching the end of the study period prior to a prescription being issued (Figure A2). Of the 1598, who were not prescribed bone protective medication prior to their fragility fracture, the mean follow‐up time was 1.2 years (SD = 1.6 years, range 0 days to 9.7 years).
With respect to vitamin D and/or calcium prescriptions, a total of 330 patients were censored within 48 weeks of a fragility fracture, due to 92 deaths, 37 patients reaching the last point of data collection from the practice, 149 transferring to a GP practice without CPRD linkage, and 52 reaching the end of the study period prior to a prescription being issued (Figure A3). Of the 1381, who were not prescribed vitamin D and/or calcium prior to their fragility fracture, the mean follow‐up time was 0.9 years (SD = 1.5 years, range 0 days to 9.7 years).
Predictors of Bone Protective Prescriptions
We explored demographic factors and fracture site as predictors of bone protective prescriptions being issued (Table 2). Being female (OR 2.06 [95% CI (1.69 to 2.51)], p < 0.001) and being aged 75 to 79 (OR 1.82 [95% CI (1.40 to 2.37)], p < 0.001), 80 to 84 (OR 1.67 [95% CI (1.28 to 2.17)], p < 0.001) or over 85 (OR 1.63 [95% CI (1.24 to 2.16)], p = 0.001), increased the odds of bone protective medication prescription. Residing in a care home reduced the odds of bone protective medication prescription (OR 0.64 [95% CI (0.43 to 0.96)], p = 0.03).
Sustaining a vertebral fracture (OR 1.84 [95% CI (1.38 to 2.45)], p < 0.001), or having unspecified fragility fracture recorded (OR 1.76 [95% CI (1.33 to 2.33)], p < 0.001), when compared to sustaining a hip fracture, also increased the odds of bone protective medication prescription, whereas sustaining a rib fracture (OR 0.28 [95% CI (0.16 to 0.49)], p < 0.001) reduced the odds.
Unadjusted logistic regression models to look at factors predicting bone protective medication prescriptions
| Unadjusted odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Gender | |||
| Male | ref | 1823 | |
| Female | 2.06 (1.69 to 2.51) | <0.001 | |
| Age | |||
| 35 to 74 | ref | 1823 | |
| 74 to 79 | 1.82 (1.40 to 2.37) | <0.001 | |
| 80 to 84 | 1.67 (1.28 to 2.17) | <0.001 | |
| 85+ | 1.63 (1.24 to 2.16) | 0.001 | |
| PD duration | |||
| Less than 1 year | ref | 1823 | |
| 1 to 2.4 years | 0.78 (0.59 to 1.03) | 0.08 | |
| 2.5 to 4 years | 0.80 (0.61 to 1.05) | 0.11 | |
| 5 to 9 years | 0.79 (0.59 to 1.05) | 0.10 | |
| 10+ years | 0.76 (0.54 to 1.07) | 0.12 | |
| Fracture type | |||
| Hip | ref | 1823 | |
| Vertebral | 1.84 (1.38 to 2.45) | <0.001 | |
| Wrist | 0.68 (0.44 to 1.05) | 0.08 | |
| Humerus | 0.80 (0.57 to 1.12) | 0.19 | |
| Rib | 0.28 (0.16 to 0.49) | <0.001 | |
| Pelvis | 1.14 (0.71 to 1.85) | 0.58 | |
| Unspecified | 1.76 (1.33 to 2.33) | <0.001 | |
| Multiple Fractures | 1.30 (0.62 to 2.71) | 0.49 | |
| Cambridge multimorbidity score | |||
| 0/1 | ref | 1823 | |
| 2 | 0.82 (0.51 to 1.33) | 0.42 | |
| 3 | 0.83 (0.53 to 1.31) | 0.43 | |
| 4 | 0.91 (0.58 to 1.43) | 0.68 | |
| 5 | 0.93 (0.59 to 1.46) | 0.74 | |
| 6 | 0.95 (0.59 to 1.52) | 0.83 | |
| 7+ | 0.88 (0.56 to 1.38) | 0.57 | |
| Care home residence | |||
| Not living in care home | ref | 1823 | |
| Living in care home | 0.64 (0.43 to 0.96) | 0.03 | |
When adjusting for age and gender, vertebral fractures (OR 2.06 [95% CI (1.53 to 2.77)], p < 0.001) had stronger association with receiving a prescription whereas unspecified fracture (OR 1.57 [95% CI (1.17 to 2.10)], p < 0.001), and rib fractures (OR 0.34 [95% CI (0.19 to 0.60)], p < 0.001) both had a slightly attenuated relationship with receiving a prescription for bone protective medication versus hip fracture. When adjusting for age, multi‐morbidity (CMS) and gender, living in a care home (OR 0.58 [95% CI (0.38 to 0.88)], p = 0.01) had a stronger association of decreasing the odds of receiving a prescription than in the univariate model.
Multimorbidity still did not change the odds of receiving a prescription after adjusting for age (Table 3). The direction of these associations was not altered when restricting the analysis to those with an initial diagnosis of PD or those with 48 weeks of data before and after the index fracture, although it did weaken the association between prescriptions and care home residency (Table A6 and A7).
Minimally adjusted logistic regression models to look at factors predicting bone protective medication prescriptions
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (unadjusted) | |||
| Hip | ref | 1823 | |
| Vertebral | 1.84 (1.38 to 2.45) | <0.001 | |
| Wrist | 0.68 (0.44 to 1.05) | 0.08 | |
| Humerus | 0.80 (0.57 to 1.12) | 0.19 | |
| Rib | 0.28 (0.16 to 0.49) | <0.001 | |
| Pelvis | 1.14 (0.71 to 1.85) | 0.58 | |
| Unspecified | 1.76 (1.33 to 2.33) | <0.001 | |
| Multiple Fractures | 1.30 (0.62 to 2.71) | 0.49 | |
| Fracture type adjusted for age | |||
| Hip | ref | 1823 | |
| Vertebral | 1.98 (1.48 to 2.65) | <0.001 | |
| Wrist | 0.75 (0.49 to 1.17) | 0.21 | |
| Humerus | 0.82 (0.58 to 1.16) | 0.26 | |
| Rib | 0.31 (0.18 to 0.55) | <0.001 | |
| Pelvis | 1.15 (0.71 to 1.86) | 0.57 | |
| Unspecified | 1.84 (1.39 to 2.44) | <0.001 | |
| Multiple fractures | 1.29 (0.62 to 2.71) | 0.50 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 1823 | |
| Vertebral | 2.06 (1.53 to 2.77) | <0.001 | |
| Wrist | 0.60 (0.38 to 0.93) | 0.02 | |
| Humerus | 0.72 (0.50 to 1.02) | 0.06 | |
| Rib | 0.34 (0.19 to 0.60) | <0.001 | |
| Pelvis | 0.99 (0.61 to 1.61) | 0.96 | |
| Unspecified | 1.57 (1.17 to 2.10) | <0.001 | |
| Multiple fractures | 1.23 (0.58 to 2.61) | 0.59 | |
| Cambridge multimorbidity score (unadjusted) | |||
| 0/1 | ref | 1823 | |
| 2 | 0.82 (0.51 to 1.33) | 0.42 | |
| 3 | 0.83 (0.53 to 1.31) | 0.43 | |
| 4 | 0.91 (0.58 to 1.43) | 0.68 | |
| 5 | 0.93 (0.59 to 1.46) | 0.74 | |
| 6 | 0.95 (0.59 to 1.52) | 0.83 | |
| 7+ | 0.88 (0.56 to 1.38) | 0.57 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 1823 | |
| 2 | 0.77 (0.47 to 1.25) | 0.29 | |
| 3 | 0.76 (0.48 to 1.21) | 0.25 | |
| 4 | 0.81 (0.51 to 1.28) | 0.37 | |
| 5 | 0.81 (0.51 to 1.28) | 0.36 | |
| 6 | 0.82 (0.51 to 1.32) | 0.41 | |
| 7+ | 0.77 (0.49 to 1.21) | 0.26 | |
| Care home residency (unadjusted) | |||
| No | ref | 1823 | |
| Yes | 0.64 (0.43 to 0.96) | 0.03 | |
| Care home residency adjusted for age | |||
| No | ref | 1823 | |
| Yes | 0.58 (0.39 to 0.88) | 0.01 | |
| Care home residency adjusted for age and gender | |||
| No | ref | 1823 | |
| Yes | 0.58 (0.38 to 0.88) | 0.01 | |
| Care home residency adjusted for age, gender and CMS | |||
| No | ref | 1823 | |
| Yes | 0.58 (0.38 to 0.88) | 0.01 | |
Discussion
Despite a significantly increased risk of sustaining fragility fractures, bone‐health and osteoporosis management is often overlooked when caring for people with parkinsonism. To understand the extent to which osteoporosis is treated in line with national guidelines, we investigated whether either bone protective medications, vitamin D and/or calcium supplements were prescribed. This was based on primary care records before and after people experienced fragility fractures. We identified treatment gaps and investigated potential predictors of prescriptions being made within 48 weeks following a fracture.
We identified a fracture rate of 27.3 per 1000 person years which aligns with the current literature. 6 , 17 Consistent with our previous findings 18 and with the results of the Parkinson's UK National Audit, 9 we identified a substantial treatment gap. Only 38% of this real‐world population who sustained a fragility fracture were receiving treatment within the recommended 16 week period. Increasing the treatment window to 48 weeks post fracture only increased this proportion to 53%. More people were prescribed vitamin D and/or calcium supplementation within 16 (58%) and 48 weeks (73%).
Being female, aged >75 years and having sustained either a vertebral or unspecified fragility fracture increased the odds of receiving a prescription for bone protective medications. Conversely, people with parkinsonism who had sustained rib fracture(s), or were residing in a care home were less likely to be prescribed bone protective medication. The direction of these associations was maintained when restricting the analysis to those with: more complete data (Table A7); or an initial diagnosis of PD (Table A6).
Due to initial diagnostic uncertainty, it is possible that some PD diagnoses may subsequently have been revised to an atypical parkinsonism. 19
Whilst demonstrating this striking and stark treatment gap, we have considered additional factors which may have impacted this finding. Using CPRD data it is possible to see which patients were prescribed bone protective medication in primary care. There is insufficient data to reliably determine whether bone health was considered and then no treatment was initiated (appropriately), adherence to prescribed treatments, or whether patients had received intravenous zoledronate via secondary care. There are several reasons why osteoporosis treatment may have been deemed inappropriate despite a patient sustaining a fragility fracture, including: concerns over polypharmacy; upper gastrointestinal disorders; renal impairment; patient preferences; bisphosphonate treatment holidays; life‐expectancy; concerns over side‐effects; or fracture risk assessment concluding treatment wasn’t warranted. 20 , 21 These legitimate reasons why treatments are sometimes not initiated will contribute to an over‐estimation of the reported treatment gap.
We might hypothesize that individuals with greater co‐morbidity burden are less likely to receive a prescription, but this was not evident in our data.
It is unclear to what extent the inability to capture treatment with zoledronate affects the treatment gap during the study period (2010 to 2019). Data from the National Hip Fracture Database (NHFD) indicates that in 2020 only 9% of patients were started on zoledronate following a hip fracture, 22 , 23 and in 2021 NOGG guidance was updated to recommend zoledronate as a first‐line treatment. 21 , 24 Prior to 2021, the recommendation was to use zoledronate if first‐line treatments, alendronate and risedronate, were not tolerated or contraindicated. 25 , 26 However, given that people with PD commonly experience difficulty swallowing, 27 perhaps oral bisphosphonates were deemed inappropriate and zoledronate was considered a suitable alternative in some instances.
Therefore, the inability to capture zoledronate may contribute to an over‐estimation of the identified treatment gap, however this gap is unlikely to be fully explained by this.
We excluded missing data from the treatment gap analysis. However, it is also important to consider those who did not receive a prescription before the last collection date for the practice, the end of the study period or who transferred to a practice without CPRD follow‐up. It is possible that these patients were later provided osteoporosis treatment. Amongst patients who died within 48 weeks and did not receive bone protective medication or supplementation, this may indicate a clinical decision that treatment was unlikely to be beneficial due to limited life expectancy.
When exploring demographic factors as predictors for bone protective medication prescription, it was unsurprising that being female compared to male increased the odds of prescriptions (OR 2.06 [95% CI (1.69 to 2.51)], p<0.001), given the known gender disparities in the treatment of osteoporosis. Recent work has demonstrated that in vertebral fracture cohorts the odds ratio for treatment was 1.82 favoring women 28 and, at hospital discharge following a hip fracture, only 2.7% (n = 3/110) of men compared to 27% (n = 60/253) of women were on anti‐resorptive treatment in a single center study in the United States. 29
Fracture type also affected the odds of receiving a prescription for bone protective medications. Vertebral fractures (OR 2.06 [1.53 to 2.77], p < 0.001) and unspecified osteoporotic fractures (OR 1.57 [1.17 to 2.10], p < 0.001) when compared to hip fracture increased the odds, whereas rib fractures (OR 0.34 [0.19 to 0.60], p < 0.001) when compared to hip fractures decreased the odds of prescription. We hypothesize that this could be explained by the different inpatient treatment received, although this cannot be addressed using CPRD data.
In the UK, rib fractures, unlike hip fractures, are more variably managed in different inpatient locations besides geriatrics, including orthopedics and general surgery 30 where there may be less familiarity with the use of osteoporosis treatments potentially leading to a discrepancy in the recommendation for bone protective medications. The finding that those with vertebral versus hip fracture have increased odds of receiving a prescription reflects findings from the US that following a vertebral fracture there is the largest increase in osteoporosis diagnosis and treatment rates compared to other fracture types. 31
Care home residency also reduced the odds of prescription after adjustment for age, gender and number of comorbidities (OR 0.58 [0.38 to 0.88], p = 0.01). However, the evidence for this was weakened when restricting the analysis to those with an initial diagnosis of PD or to those with 48 weeks of data before and after the index fracture. This is most likely explained by the small number of patients who were resident in a care home in these sub‐groups (Table A6 and A7) therefore widening the confidence interval.
This result is consistent with findings from US care home residents with dementia and high fracture risk, where only 11.6% received fracture prevention medications. 32 The authors also highlighted a potential lack of awareness about the appropriateness of treatment, as 10% of those receiving hospice care and 11% of individuals who were not dependent for activities of daily living were receiving medication for fracture prevention reflecting over and under‐treatment respectively. 32 Whilst the appropriateness of treatment could not be assessed in our study, it is important to consider that treatment for osteoporosis should be considered on a case‐by‐case basis accounting for several patient factors regardless of whether the patient resides in a care home.
Strengths and Limitations
This study is strengthened by the ability to assess the management of osteoporosis in a large, representative cohort of people with Parkinsonism across the United Kingdom. By using CPRD data it is possible to include previously under‐represented groups such as those who are older, with more advanced parkinsonism, care home residents, as well as people across the spectrum of parkinsonian disorders. 33 This is of particular importance given that people with atypical parkinsonism have greater fracture risk, alongside an osteoporosis treatment gap, 34 , 35 and there is often diagnostic uncertainty. 19
It was beyond the scope of this paper to ascertain whether patients were adherent to prescribed treatment or whether health care professional follow‐up was provided to maximize adherence. 14 Adherence to osteoporosis treatment is a well‐documented issue; persistence to oral bisphosphonates amongst post‐menopausal women is reported to be only 28% at 12 months. 36 Therefore, despite prescriptions being made it is likely that many patients were still without treatment following a fragility fracture.
When determining if patients were on treatment prior to fracture, for oral bisphosphonates patients were considered “on‐treatment” if they received a prescription in the 3 months prior to their fracture. However, this could mis‐classify participants if they were on treatment holiday after previously being prescribed oral bisphosphonates. It was also likely that patients being given zoledronate are mis‐categorized as “untreated” as it is not administered in the primary care setting and is therefore not reliably recorded within the CPRD data.
Furthermore, patients could have been misclassified as “not taking supplementation” if they are using over‐the‐counter vitamin D and/or calcium. However over‐the‐counter formulations may be sub‐standard in comparison to licensed medical products. 37 All of which could contribute to the observed treatment gaps.
As previously discussed, it is also possible that treatment options were considered and deemed inappropriate or declined by patients; however, this information is not routinely coded in CPRD data. It is also not possible using CPRD data to determine the circumstances surrounding the fracture. Whilst we included fracture sites typically associated with fragility fractures, it is possible that some of these fractures occurred from a trauma greater than a fall from standing height and are also contributing to an over‐estimation of the treatment gap as they are not indicative of osteoporosis.
Additionally, this population includes all people with parkinsonism who experienced a fracture between 2010 and 2019; in this study we consider the treatment decisions made after only the initial fracture which occurred within this period. However, it is possible that patients had multiple fracture events within this period or had fractures prior to 1st January 2010. Furthermore, the data available to us for analysis pre‐dates the updated NOGG guidance which mentions Parkinson's as an osteoporosis risk factor, 21 and our guidance on managing bone‐health in parkinsonism. 18 , 38 However, the treatment gap identified remains consistent with our 2025 findings that less than one third of people with parkinsonism are treated in‐line with the NOGG guidelines. 18
There remains a significant osteoporosis treatment gap within the parkinsonian population. Men with parkinsonism, those living in a care home, and those who sustain rib fractures are at increased odds of not receiving a prescription for bone protective medication following a fracture. Addressing this gap might reasonably be expected to prevent future fragility fractures and the associated serious and negative sequalae in this vulnerable and at‐risk population.
Author Roles
(1) Research project: A. Conception, B. Organization, C. Execution, (2) Statistical analysis: A. Design, B. Execution, C. Review and critique; (3) Manuscript preparation: A. Writing of the first draft, B. Review and critique.
K.C.N.: 1A, 1B, 1C, 2A, 2B, 3A.
E.J.H.: 1A, 1B, 1C, 2A, 2C, 3B.
E.T.: 1A, 1B, 1C, 2A, 2C, 3B.
Disclosures
Ethical Compliance Statement: This study is based in part on data from the Clinical Practice Research Datalink obtained under license from the UK Medicines and Healthcare products Regulatory Agency (protocol number 20_000060). However, the interpretation and conclusions contained in this report are those of the author's alone. The authors confirm that the approval of an institutional review board and patient consent were not required for this work. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.
Funding Sources and Conflict of Interest: This study was funded by the Gatsby Foundation (GAT3676). EJH is HEFCE funded for her academic work. ET is funded by a National Institute for Health and Care Research Academic Clinical Lectureship. The authors declare that there are no conflicts of interest relevant to this work.
Financial Disclosures for the Previous 12 Months: EJH is HEFCE funded by University of Bristol for her academic work and has received research funding from the NIHR, the British Geriatrics Society, the Gatsby Foundation, the Alzheimer's Society, Royal Osteoporosis Society, the Dunhill Society, and Parkinson's UK. She has received travel support, honoraria, and/or sat on advisory boards for Kyowa Kirin, AbbVie, Luye, the CME institute, Ever, Simbec Orion, the Neurology Academy, and Bial. ET is funded by a National Institute for Health and Care Research Academic Clinical Lectureship.
KCN is funded by the University of Bristol.
Acknowledgments
The authors acknowledge Savannah Haworth's role in assessing Read codes for clinically relevant fragility fractures.
Data Availability Statement
This study is based in part on data from the Clinical Practice Research Datalink obtained under license from the UK Medicines and Healthcare products Regulatory Agency. However, the interpretation and conclusions contained in this report are those of the author's alone.
References
Directed acyclic graphs for the minimally adjusted logistic regressions. This directed acyclic graph was developed using the DAGitty software1 to visually represent the causal and non‐causal associations that informed the minimally adjusted logistic regression models. 11 Textor J, van der Zander B, Gilthorpe MS, Liskiewicz M, Ellison GT. Robust causal inference using directed acyclic graphs: the R package “dagitty.” Int J Epidemiol. 2016;45(6):1887 to 94.
Reasons for incomplete follow‐up: bone protective medications.
Reasons for incomplete follow‐up: vitamin D and/or calcium supplements.
Included parkinsonism read codes
| Description (as listed in CPRD medical.txt file) | Medcode | Read code |
|---|---|---|
| Parkinson's disease | 4321 | F12.00 |
| Parkinsonism with OH | 8956 | F130300 |
| Dementia in Parkinson's disease | 9509 | Eu02300 |
| Parkinson's disease NOS | 14,912 | F12z.00 |
| Parkinsonism in diseases classified elsewhere | 86,062 | Fyu2200 |
| Paralysis agitans | 1691 | F120.00 |
| Vascular parkinsonsim | 100,128 | F124.00 |
| Lewy body disease | 7572 | F116.00 |
| [X]Lewy body dementia | 26,270 | Eu02500 |
| Multiple system atrophy | 22,454 | F174.00 |
| Shy‐Drager syndrome | 35,839 | F130500 |
| Progressive supranuclear palsy | 9385 | F24y000 |
| Progressive supranuclear ophthalmoplegia | 40,553 | F130400 |
| Steele‐Richardson Oszewski syndrome | 93,910 | F24y012 |
| Steele Richardson Oszewski syn | 49,034 | F24y011 |
| Steele‐Richardson‐Oszewski syndrome | 7037 | F24y200 |
| Supranucelar paralysis | 29,515 | F36y000 |
| Corticobasal degeneration | 107,650 | F11y200 |
| Cerebral degeneration in Parkinson's disease | 96,860 | F11x900 |
Included fragility fracture read codes.
| Description (as listed in CPRD medical.txt file) | Medcode | Read code |
|---|---|---|
| Hip | ||
| Hip fracture | 1994 | S30.11 |
| Fracture of neck of femur | 2225 | S30.00 |
| Closed fracture of proximal femur, pertrochanteric | 5301 | S302.00 |
| Prmy open red+int fxn prox femoral # + screw/nail+plate device | 5742 | 7K1D000 |
| Closed reduction of fracture of hip | 6660 | 7K1L400 |
| Closed fracture of femur, intertrochanteric | 8648 | S302400 |
| Cls red+int fxn proximal femoral # + screw/nail device alone | 8719 | 7K1J000 |
| DHS‐Dynamic hip screw primary fixation of neck of femur | 9792 | 7K1D01E |
| Hip fracture NOS | 10,570 | S30y.11 |
| Dynamic hip screw primary fixation of neck of femur | 12,544 | 7K1D01F |
| Cls # prox femur, subcapital, Garden grade unspec. | 17,019 | S300500 |
| H/O: hip fracture | 17,936 | 14G7.00 |
| Closed fracture of neck of femur NOS | 18,273 | S30y.00 |
| Cls # proximal femur, trochanteric section, unspecified | 19,117 | S302000 |
| Open fracture proximal femur, subcapital, Garden grade III | 23,803 | S301800 |
| Closed fracture of unspecified proximal femur | 24,276 | S30w.00 |
| Prim open reduct # neck femur & op fix‐Richards screw | 24,493 | 7K1D01A |
| Pertrochanteric fracture | 28,965 | S304.00 |
| Closed fracture proximal femur, subtrochanteric | 29,145 | S302200 |
| Prmy open red+int fxn prox femoral # + screw/nail device alone | 33,624 | 7K1D600 |
| Closed fracture proximal femur, subcapital, Garden grade II | 33,957 | S300700 |
| Closed fracture proximal femur, subcapital, Garden grade IV | 34,078 | S300900 |
| Closed fracture proximal femur, subcapital, Garden grade I | 34,351 | S300600 |
| Prmy open red+int fxn prox fem # + screw/nail+intramed device | 34,764 | 7K1D700 |
| Primary int fxn(no red) prox fem # + screw/nail device alone | 35,004 | 7K1J500 |
| Closed fracture head of femur | 36,391 | S300400 |
| Closed fracture proximal femur, subcapital, Garden grade III | 36,599 | S300800 |
| Open fracture of neck of femur NOS | 38,054 | S30z.00 |
| Closed fracture proximal femur, transcervical | 38,489 | S300.00 |
| Primary int fxn(no red) prox fem # + screw/nail+plate device | 38,856 | 7K1J700 |
| Open fracture proximal femur, subcapital, Garden grade unspec | 38,878 | S301500 |
| Open fracture of femur, intertrochanteric | 39,396 | S303400 |
| Cls # prox femur, intracapsular section, unspecified | 39,984 | S300000 |
| Cl red intracaps fract neck femur fix − Smith‐Petersen nail | 40,999 | 7K1J012 |
| Primary int fxn(no red) prox fem # + scrw/nail+intramed device | 44,594 | 7K1J600 |
| Cls # of proximal femur, pertrochanteric section, NOS | 44,735 | S302z00 |
| Closed fracture proximal femur, intertrochanteric, two part | 45,141 | S302100 |
| Prim open reduct # neck femur & op fix − Neufield nail plate | 46,258 | 7K1D018 |
| Primary cls red+int fxn prox fem # + screw/nail+plate device | 46,959 | 7K1JD00 |
| Closed fracture proximal femur, other transcervical | 49,209 | S300y00 |
| Opn # proximal femur, intracapsular section, unspecified | 50,727 | S301000 |
| Cls # proximal femur, intertrochanteric, comminuted | 51,216 | S302300 |
| Closed fracture, base of neck of femur | 51,861 | S300311 |
| Open fracture proximal femur, subcapital, Garden grade IV | 51,999 | S301900 |
| Closed fracture proximal femur, basicervical | 52,194 | S300300 |
| Prim op red # nck femur and op fix‐Charnley compression screw | 52,395 | 7K1D012 |
| Cl red intracaps frac neck femur fix‐Garden cannulated screw | 53,670 | 7K1J011 |
| Prim cls rd + int fxn prox fem # + screw/nail+intramdulry device | 54,819 | 7K1JC00 |
| Primary cls red+int fxn prox fem # + screw/nail device alone | 55,386 | 7K1JB00 |
| Prim open red # neck femur and op fix‐McLaughlin nail plate | 56,568 | 7K1D017 |
| Cls red+int fxn prox femoral # + Richard's cannulat hip screw | 57,514 | 7K1J013 |
| Prim open reduct # neck femur and op fix‐Ross Brown nail | 57,884 | 7K1D01B |
| Prim op red # nck femur and op fix‐Zickel intramed nail plate | 57,889 | 7K1D01D |
| Open fracture of unspecified proximal femur | 58,642 | S30x.00 |
| Prim open reduct # neck femur and op fix‐Blount nail plate | 58,817 | 7K1D011 |
| Open fracture proximal femur, subcapital, Garden grade I | 60,885 | S301600 |
| Open fracture of proximal femur, pertrochanteric | 61,733 | S303.00 |
| Closed fracture proximal femur, transcervical, NOS | 62,966 | S300z00 |
| Prim open reduct # neck femur and op fix‐Pugh nail plate | 65,536 | 7K1D019 |
| Closed fracture proximal femur, midcervical section | 65,690 | S300200 |
| Open fracture proximal femur, subcapital, Garden grade II | 67,394 | S301700 |
| Open # of proximal femur, trochanteric section, unspecified | 67,633 | S303000 |
| Closed fracture of femur, subcapital | 68,229 | S300y11 |
| Open fracture proximal femur, other transcervical | 68,668 | S301y00 |
| Primary external fixation (without reduction) prox femoral # | 70,018 | 7K1K300 |
| Open fracture of proximal femur, pertrochanteric, NOS | 70,479 | S303z00 |
| Open fracture proximal femur, subtrochanteric | 71,282 | S303200 |
| Open fracture of femur, subcapital | 73,234 | S301y11 |
| Open fracture proximal femur, transcervical | 73,981 | S301.00 |
| Closed fracture‐subluxation, hip joint | 93,374 | S4E2.00 |
| Prim open reduct # neck femur and op fix‐Holt nail | 94,714 | 7K1D014 |
| Prim op red # nck femur and op fix‐Deyerle multiple hip pin | 97,337 | 7K1D013 |
| Open fracture proximal femur, intertrochanteric, comminuted | 97,971 | S303300 |
| Open fracture base of neck of femur | 100,771 | S301311 |
| Open fracture proximal femur, intertrochanteric, two part | 101,567 | S303100 |
| Primary cls reduction+external fixation proximal femoral # | 102,313 | 7K1K500 |
| Prim open reduct # neck femur and op fix‐Jewett nail plate | 105,352 | 7K1D015 |
| Prim op red frac neck fem op fix us prox fem nail antirotatn | 105,803 | 7K1DE00 |
| Remanip intracap fract neck fem and fix using nail or screw | 107,358 | 7K1Y000 |
| Prim open reduct # neck femur and op fix‐Massie nail plate | 107,932 | 7K1D016 |
| Open fracture proximal femur, transcervical, NOS | 112,712 | S301z00 |
| Open fracture proximal femur, midcervical section | 112,836 | S301200 |
| Open fracture proximal femur, basicervical | 113,426 | S301300 |
| Closed reduction of intracapsular # NOF internal fixat DHS | 39,322 | 7K1Jd00 |
| Vertebral fractures | ||
| Collapse of vertebra NOS | 2793 | N331.12 |
| Closed fracture lumbar vertebra | 3888 | S104.00 |
| Collapse of vertebra due to osteoporosis NOS | 4013 | N331L00 |
| Fracture of vertebra without spinal cord lesion | 4409 | S10.12 |
| Fracture of thoracic vertebra | 5381 | S15.00 |
| Collapse of lumbar vertebra due to osteoporosis | 5841 | N331J00 |
| Closed fracture lumbar vertebra, wedge | 8266 | S104100 |
| Collapse of thoracic vertebra | 9319 | N331F00 |
| Fracture of lumbar vertebra | 10,990 | S10B000 |
| Collapse of lumbar vertebra | 11,543 | N331G00 |
| Other specified closed fracture thoracic vertebra | 11,770 | S102y00 |
| Fracture of lumbar spine and pelvis | 12,406 | S10B.00 |
| Osteoporosis + pathological fracture thoracic vertebrae | 12,673 | N331900 |
| Closed fracture of unspecified cervical vertebra | 15,613 | S100000 |
| Collapse of thoracic vertebra | 15,837 | N331011 |
| Fatigue fracture of vertebra | 16,895 | N1y1.00 |
| Collapse of cervical vertebra | 17,008 | N331E00 |
| Osteoporosis + pathological fracture lumbar vertebrae | 17,377 | N331800 |
| Collapse of thoracic vertebra due to osteoporosis | 19,048 | N331K00 |
| H/O: vertebral fracture | 19,235 | 14G8.00 |
| Collapse of lumbar vertebra | 23,686 | N331111 |
| Closed fracture of seventh cervical vertebra | 24,672 | S100700 |
| Closed fracture thoracic vertebra | 27,404 | S102.00 |
| Closed fracture of fifth cervical vertebra | 27,575 | S100500 |
| Closed fracture of sixth cervical vertebra | 27,654 | S100600 |
| Fracture of first cervical vertebra | 28,133 | S10A000 |
| Closed fracture thoracic vertebra, wedge | 28,524 | S102100 |
| C6 vertebra closed fracture without spinal cord lesion | 33,503 | S100611 |
| C2 vertebra closed fracture without spinal cord lesion | 33,967 | S100211 |
| Fracture of second cervical vertebra | 34,403 | S10A100 |
| C5 vertebra closed fracture without spinal cord lesion | 34,873 | S100511 |
| C7 vertebra closed fracture without spinal cord lesion | 38,053 | S100711 |
| Collapsed vertebra NOS | 38,728 | N331D00 |
| Closed fracture thoracic vertebra not otherwise specified | 41,138 | S102z00 |
| Closed fracture of fourth cervical vertebra | 41,548 | S100400 |
| C1 vertebra closed fracture: no spinal cord lesion | 42,561 | S100111 |
| Osteoporotic vertebral collapse | 44,386 | N331.14 |
| Collapse of cervical vertebra due to osteoporosis | 45,736 | N331H00 |
| Osteoporosis + pathological fracture cervical vertebrae | 48,772 | N331A00 |
| C3 vertebra closed fracture without spinal cord lesion | 52,699 | S100311 |
| Closed fracture cervical vertebra, wedge | 53,337 | S100H00 |
| Closed fracture of third cervical vertebra | 60,593 | S100300 |
| Open fracture lumbar vertebra, wedge | 65,302 | S105100 |
| C4 vertebra closed fracture without spinal cord lesion | 67,358 | S100411 |
| [X]Collapsed vertebra in diseases classified elsewhere | 73,900 | Nyu6700 |
| [X]Fracture of other specified cervical vertebra | 94,638 | Syu1500 |
| Collapse of cervical vertebra | 110,999 | N331C11 |
| Wrist/distal radius | ||
| Wrist fracture: closed | 203 | S234.11 |
| Closed reduction of fracture of wrist | 5951 | 7K1LM00 |
| Fracture at wrist and hand level | 8056 | S242.00 |
| Wrist fracture: open | 10,022 | S235.11 |
| Closed fracture‐subluxation of the wrist | 17,921 | S4C2.00 |
| Fracture‐dislocation or subluxation of wrist | 18,614 | S4C.00 |
| Closed fracture dislocation of wrist | 34,429 | S4C0.00 |
| Open fracture dislocation wrist | 42,844 | S4C1.00 |
| [X]Fracture of other and unspecified parts of wrist and hand | 53,068 | Syu6500 |
| Open fracture‐subluxation of the wrist | 59,219 | S4C3.00 |
| Closed fracture distal radius, extra‐articular, other type | 19,058 | S234D00 |
| Closed fracture distal radius, intra‐articular, die‐punch | 44,844 | S234C00 |
| Closed fracture distal radius, intra‐articular, other type | 28,293 | S234E00 |
| Closed fracture of the distal radius, unspecified | 1742 | S234200 |
| Open fracture distal radius, extra‐articular other type | 53,698 | S235D00 |
| Open fracture distal radius, intra‐articular other type | 63,588 | S235E00 |
| Open fracture distal radius, intra‐articular, die‐punch | 104,070 | S235C00 |
| Humerus | ||
| Fracture of humerus | 517 | S22.00 |
| Fracture of upper end of humerus | 2101 | S226.00 |
| Closed reduction of fracture of humerus | 6379 | 7K1LF00 |
| Open fracture of the proximal humerus | 9420 | S221.00 |
| Fracture of humerus NOS | 10,382 | S22z.00 |
| Closed fracture proximal humerus, greater tuberosity | 11,044 | S220300 |
| Closed fracture of the proximal humerus | 11,222 | S220.00 |
| Closed fracture proximal humerus, neck | 11,313 | S220100 |
| Closed fracture of humerus NOS | 19,186 | S222000 |
| Closed fracture of humerus, shaft | 27,886 | S222100 |
| Closed fracture proximal humerus, head | 28,739 | S220400 |
| Closed fracture proximal humerus, four part | 29,137 | S220700 |
| Fracture of shaft of humerus | 30,659 | S227.00 |
| Closed fracture of proximal humerus, anatomical neck | 33,489 | S220200 |
| Closed fracture of humerus, shaft or unspecified part | 36,464 | S222.00 |
| Closed fracture of proximal humerus not otherwise specified | 38,353 | S220z00 |
| Closed fracture proximal humerus, three part | 40,330 | S220600 |
| Closed fracture of proximal humerus, unspecified part | 44,721 | S220000 |
| Open fracture of proximal humerus not otherwise specified | 45,275 | S221z00 |
| Open fracture proximal humerus, greater tuberosity | 48,239 | S221300 |
| Closed fracture of humerus, upper epiphysis | 52,406 | S220500 |
| Open fracture proximal humerus, neck | 53,622 | S221100 |
| Open fracture of proximal humerus, unspecified part | 53,688 | S221000 |
| Closed fracture of humerus, shaft or unspecified part NOS | 61,378 | S222z00 |
| Open fracture proximal humerus, head | 70,486 | S221400 |
| Open fracture proximal humerus, three part | 70,604 | S221600 |
| Open fracture proximal humerus, four part | 70,653 | S221700 |
| Open fracture of proximal humerus, anatomical neck | 71,207 | S221200 |
| Rib | ||
| Closed fracture rib | 280 | S120.00 |
| Closed fracture of rib, unspecified | 7831 | S120000 |
| Fracture of rib | 9688 | S127.00 |
| Multiple fractures of ribs | 10,696 | S127000 |
| Closed fracture multiple ribs | 11,004 | S120900 |
| Rib fracture NOS | 11,378 | S12z.11 |
| Closed fracture of one rib | 16,494 | S120100 |
| Multiple rib fractures | 17,107 | S29.12 |
| Closed fracture of rib(s) NOS | 28,244 | S120z00 |
| Cough fracture of ribs | 28,538 | S127100 |
| Closed fracture of four ribs | 34,197 | S120400 |
| Closed fracture of two ribs | 40,533 | S120200 |
| Open fracture of two ribs | 44,826 | S121200 |
| Open fracture rib | 48,224 | S121.00 |
| Closed fracture of five ribs | 55,424 | S120500 |
| Closed fracture of three ribs | 56,384 | S120300 |
| Closed fracture of six ribs | 65,484 | S120600 |
| Closed fracture of eight or more ribs | 68,652 | S120800 |
| Open fracture multiple ribs | 71,452 | S121900 |
| Open fracture of rib, unspecified | 73,613 | S121000 |
| Open fracture of seven ribs | 73,956 | S121700 |
| Closed fracture of seven ribs | 90,494 | S120700 |
| Open fracture of rib(s) NOS | 101,560 | S121z00 |
| Open fracture of one rib | 112,701 | S121100 |
| Open fracture of four ribs | 113,739 | S121400 |
| Pelvis | ||
| Fracture or disruption of pelvis | 738 | S13.00 |
| Closed fracture pelvis, multiple pubic rami‐stable | 6667 | S132100 |
| Closed fracture pelvis, single pubic ramus | 7004 | S132000 |
| Other or multiple closed fracture of pelvis NOS | 11,639 | S134z00 |
| Fracture of lumbar spine and pelvis | 12,406 | S10B.00 |
| Closed fracture pelvis, coccyx | 14,834 | S108.00 |
| Closed fracture of pelvis NOS | 28,375 | S13y.00 |
| Other or multiple closed fracture of pelvis | 33,961 | S134.00 |
| Closed fracture pelvis, ischium | 34,708 | S134100 |
| Closed fracture‐dislocation of pelvis | 34,910 | S4J0100 |
| Closed fracture pelvis, multiple pubic rami‐unstable | 46,592 | S132200 |
| Unspecified site | ||
| Fragility fracture due to unspecified osteoporosis | 11,503 | N331M00 |
| H/O: fragility fracture | 18,731 | 14G6.00 |
| [X]Unspecified osteoporosis with pathological fracture | 18,825 | NyuB800 |
| Idiopathic osteoporosis with pathological fracture | 27,597 | N331600 |
| Osteoporosis of disuse with pathological fracture | 33,526 | N331300 |
| Postmenopausal osteoporosis with pathological fracture | 38,395 | N331B00 |
| Postoophorectomy osteoporosis with pathological fracture | 39,334 | N331200 |
| Drug‐induced osteoporosis with pathological fracture | 46,894 | N331500 |
| [X]Other osteoporosis with pathological fracture | 57,301 | NyuB000 |
| Postsurgical malabsorption osteoporosis with path fracture | 68,019 | N331400 |
| Fragility fracture | 93,497 | N331N00 |
| Minimal trauma fracture due to unspecified osteoporosis | 93,705 | N331M11 |
Search Terms for prescriptions
| Search term | Target |
|---|---|
| Bisphosphon | Bisphosphonate/Bisphosphonic acid |
| Disphosphon | Diphosphonate/Diphosphonic acid |
| Alendron | Alendronate/Alendronic acid |
| MK‐217 | Alendronic acid (Alternative notation) |
| Ibandron | Ibandronate/Ibandronic acid |
| Risedron | Risedronate/Risedronic acid |
| Zolendron | Zoledronate/Zolendronic acid |
| Etidron | Etidronate/Etidronic acid |
| EHDP | Etidronate (alternative notation) |
| Strontium ranelate | Strontium ranelate |
| Sodium clondronate | Sodium clondronate |
| Tiludron | Tilundronate/Tilundronic acid |
| Denosumab | Denosumab |
| AMG 162 | Denosumab (alternative notation) |
| Romosozumab | Romosozumab |
| AMG 785 | Romosozumab (alternative notation) |
| Teriparatide | Teriparatide |
| Human parathyroid hormone | Human parathyroid hormone |
| Abaloparatide | Abaloparatide |
| BNF chapters | |
| 060602 | Bone protective medications |
| 090604 | Vitamin D supplementation |
| 09050101 | Calcium supplementation |
Table of bone protective medication prescriptions over 48 weeks following a fragility fracture
| Timeline | Censored from study (N) | Number of participants a | Oral bisphosphonate | Strontium ranelate | Denosumab | Zoledronate | All |
|---|---|---|---|---|---|---|---|
| Before fracture 1 | 0 | 1823 | 217 | 6 | 2 | 0 | 225 |
| (Percentage) | (11.9%) | (0.3%) | (0.1%) | (0.0%) | (12.3%) | ||
| Up to 16 weeks after | 236 | 1587 | 572 | 17 | 5 | 3 | 597 |
| (Percentage) | (36.0%) | (1.1%) | (0.3%) | (0.2%) | (37.6%) | ||
| Up to 32 weeks after | 369 | 1454 | 653 | 18 | 9 | 3 | 683 |
| (Percentage) | (44.9%) | (1.2%) | (0.6%) | (0.2%) | (47.0%) | ||
| Up to 48 weeks after | 479 | 1344 | 678 | 18 | 12 | 3 | 711 |
| (Percentage) | (50.4%) | (1.3%) | (0.9%) | (0.2%) | (52.9%) |
Note: This table displays the cumulative frequency of each medication type and all bone protective medications, starting before the fracture event followed by 16 week intervals.
The number of people with available data within each time window for bone protective medication.
Table of vitamin D and/or calcium supplement prescriptions over 48 weeks following a fragility fracture
| Timeline | Censored from study (N) | Number of participants a | Supplementation b |
|---|---|---|---|
| Before fracture 1 | 0 | 1823 | 442 |
| (Percentage) | (24.2%) | ||
| Up to 16 weeks after | 181 | 1642 | 951 |
| (Percentage) | (57.9%) | ||
| Up to 32 weeks after | 269 | 1554 | 1054 |
| (Percentage) | (67.8%) | ||
| Up to 48 weeks after | 330 | 1493 | 1093 |
| (Percentage) | (73.2%) |
Note: This table displays the cumulative frequency of supplementation (vitamin D and/or calcium) starting before the fracture event followed by 16 week intervals.
The number of people with available data within each time window for supplementation (Vitamin D and/or calcium).
Supplementation refers to a prescription being received for vitamin D and/or calcium.
Sensitivity analysis restricted to those with an initial diagnosis of Parkinson's disease
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (unadjusted) | |||
| Hip | ref | 1521 | |
| Vertebral | 1.79 (1.31 to 2.43) | < 0.001 | |
| Wrist | 0.64 (0.40 to 1.02) | 0.06 | |
| Humerus | 0.72 (0.50 to 1.04) | 0.08 | |
| Rib | 0.25 (0.13 to 0.46) | < 0.001 | |
| Pelvis | 1.00 (0.57 to 1.73) | 0.99 | |
| Unspecified | 1.69 (1.24 to 2.30) | < 0.001 | |
| Multiple fractures | 1.38 (0.61 to 3.12) | 0.44 | |
| Fracture type adjusted for age | |||
| Hip | ref | 1521 | |
| Vertebral | 1.91 (1.40 to 2.61) | < 0.001 | |
| Wrist | 0.72 (0.45 to 1.16) | 0.18 | |
| Humerus | 0.74 (0.51 to 1.08) | 0.12 | |
| Rib | 0.28 (0.15 to 0.53) | < 0.001 | |
| Pelvis | 1.00 (0.57 to 1.75) | 1.00 | |
| Unspecified | 1.79 (1.31 to 2.44) | < 0.001 | |
| Multiple fractures | 1.40 (0.61 to 3.19) | 0.42 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 1521 | |
| Vertebral | 2.04 (1.48 to 2.81) | < 0.001 | |
| Wrist | 0.56 (0.34 to 0.91) | 0.02 | |
| Humerus | 0.64 (0.44 to 0.94) | 0.02 | |
| Rib | 0.31 (0.16 to 0.58) | < 0.001 | |
| Pelvis | 0.85 (0.48 to 1.49) | 0.56 | |
| Unspecified | 1.53 (1.11 to 2.10) | 0.01 | |
| Multiple Fractures | 1.31 (0.56 to 3.02) | 0.53 | |
| Cambridge multimorbidity score unadjusted | |||
| 0/1 | ref | 1521 | |
| 2 | 0.80 (0.48 to 1.36) | 0.42 | |
| 3 | 0.80 (0.48 to 1.33) | 0.39 | |
| 4 | 0.88 (0.53 to 1.45) | 0.61 | |
| 5 | 0.95 (0.58 to 1.57) | 0.85 | |
| 6 | 0.97 (0.58 to 1.63) | 0.92 | |
| 7+ | 0.87 (0.53 to 1.43) | 0.58 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 1521 | |
| 2 | 0.74 (0.44 to 1.27) | 0.28 | |
| 3 | 0.74 (0.44 to 1.23) | 0.24 | |
| 4 | 0.79 (0.48 to 1.31) | 0.36 | |
| 5 | 0.84 (0.50 to 1.40) | 0.49 | |
| 6 | 0.83 (0.49 to 1.41) | 0.50 | |
| 7+ | 0.76 (0.46 to 1.26) | 0.29 | |
| Care home residence (Unadjusted) a | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.75 (0.48 to 1.17) | 0.20 | |
| Care home residence adjusted for age | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.68 (0.44 to 1.07) | 0.10 | |
| Care home residence adjusted for age and gender | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.69 (0.44 to 1.08) | 0.11 | |
| Care home residence adjusted for age, gender and CMS | |||
| Not living in care home | ref | 1521 | |
| Living in care home | 0.68 (0.43 to 1.08) | 0.10 | |
There are now 94 (6.2%) of patients living in a care home.
Sensitivity analysis restricted to those with 48‐weeks of data before and after the index fracture
| Odds ratio (95% CI) | N | p‐value | |
|---|---|---|---|
| Fracture type (Unadjusted) | |||
| Hip | ref | 765 | |
| Vertebral | 1.64 (1.05 to 2.56) | 0.03 | |
| Wrist | 0.58 (0.32 to 1.05) | 0.07 | |
| Humerus | 0.67 (0.39 to 1.13) | 0.13 | |
| Rib | 0.22 (0.10 to 0.48) | <0.001 | |
| Pelvis | 0.56 (0.25 to 1.27) | 0.17 | |
| Unspecified | 1.57 (1.05 to 2.36) | 0.03 | |
| Multiple Fractures | 1.31 (0.37 to 4.63) | 0.67 | |
| Fracture type adjusted for age | |||
| Hip | ref | 765 | |
| Vertebral | 1.73 (1.10 to 2.71) | 0.02 | |
| Wrist | 0.62 (0.34 to 1.14) | 0.13 | |
| Humerus | 0.69 (0.40 to 1.17) | 0.17 | |
| Rib | 0.24 (0.11 to 0.52) | <0.001 | |
| Pelvis | 0.57 (0.25 to 1.30) | 0.18 | |
| Unspecified | 1.61 (1.07 to 2.42) | 0.02 | |
| Multiple Fractures | 1.36 (0.38 to 4.84) | 0.64 | |
| Fracture type adjusted for age and gender | |||
| Hip | ref | 765 | |
| Vertebral | 1.78 (1.12 to 2.82) | 0.01 | |
| Wrist | 0.50 (0.27 to 0.94) | 0.03 | |
| Humerus | 0.60 (0.35 to 1.03) | 0.06 | |
| Rib | 0.26 (0.12 to 0.57) | < 0.001 | |
| Pelvis | 0.48 (0.21 to 1.10) | 0.08 | |
| Unspecified | 1.44 (0.95 to 2.19) | 0.09 | |
| Multiple Fractures | 1.30 (0.36 to 4.69) | 0.69 | |
| Cambridge multimorbidity score unadjusted | |||
| 0/1 | ref | 765 | |
| 2 | 0.66 (0.29 to 1.48) | 0.31 | |
| 3 | 0.85 (0.40 to 1.81) | 0.67 | |
| 4 | 1.02 (0.48 to 2.16) | 0.96 | |
| 5 | 0.85 (0.39 to 1.84) | 0.68 | |
| 6 | 1.16 (0.53 to 2.56) | 0.71 | |
| 7+ | 1.08 (0.51 to 2.29) | 0.84 | |
| Cambridge multimorbidity score adjusted for age | |||
| 0/1 | ref | 765 | |
| 2 | 0.60 (0.26 to 1.37) | 0.23 | |
| 3 | 0.79 (0.36 to 1.70) | 0.54 | |
| 4 | 0.90 (0.42 to 1.93) | 0.78 | |
| 5 | 0.76 (0.34 to 1.66) | 0.48 | |
| 6 | 0.97 (0.43 to 2.17) | 0.93 | |
| 7+ | 0.91 (0.42 to 1.95) | 0.80 | |
| Care home residence (unadjusted) a | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.54 (0.24 to 1.18) | 0.12 | |
| Care home residence adjusted for age | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.48 (0.22 to 1.07) | 0.07 | |
| Care home residence adjusted for age and gender | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.47 (0.21 to 1.05) | 0.07 | |
| Care home residence adjusted for age, gender and CMS | |||
| Not living in care home | ref | 765 | |
| Living in care home | 0.44 (0.20 to 1.00) | 0.05 | |
There are now 31 (4.1%) of patients living in a care home.
Publication and source details
Original title: Mind the Gap: Predictors of Osteoporosis Treatment Following Fragility Fracture in Parkinsonism
Katie C. Naylor; Emily J. Henderson; Emma Tenison
Movement Disorders Clinical Practice · 2026
* Correspondence to: Katie C. Naylor, University of Bristol, Canynge Hall, Bristol, BS8 2PS, UK; E‐mail: katie.naylor@bristol.ac.uk© 2026 The Author(s). Movement Disorders Clinical Practice published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.https://creativecommons.org/licenses/by/4.0/This is an open access article under the terms of the http://creativecommons.org/licenses/by/4.0/ License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Gatsby Charitable Foundation10.13039/501100000324 GAT3676
Repository JATS snapshot; version as supplied
Source retrieved: 2026-11-10T23:13:09.049507+00:00
DOI: 10.1002/mdc3.70652 · PMC13339595