Surgical plating versus closed reduction for fractures in the distal radius in older patients: a cost‐effectiveness analysis from the hospital perspective

Abstract

Background

Given the cost differential between surgical and non‐surgical management of distal radius fractures, we aimed to evaluate the cost‐effectiveness of surgical compared with non‐surgical treatment of distal radius fractures in a cohort of older patients.

Methods

This evaluation was conducted alongside the combined randomized and observational study of surgery for fractures of the distal radius in the elderly (CROSSFIRE) trial (ACTRN 12616000969460) which compared surgical (open reduction and internal fixation using volar‐locking plate (VLP) fixation) and non‐surgical (closed fracture reduction and cast immobilization (CR)) treatment for displaced distal radius fractures in patients ≥60 years. Cost‐effectiveness was assessed from the perspective of the public hospital funder. Hospital records from a sub‐sample of participants were used to estimate costs. Outcomes were patient‐reported wrist pain and function questionnaire (PRWE) scores and quality adjusted life years (QALYs) calculated using the EuroQoL five‐dimension five‐level tool (EQ‐5D‐5L).

Results

From 166 participants (81 surgical, 85 non‐surgical), costs were obtained for 56 (29 surgical, 27 non‐surgical). The mean costs for VLP fixation were Australian dollars (AUD) 6668 (95% CI $4857 to $8479) compared to AUD 3343 (95% CI $1304 to $5381) for CR. The incremental cost‐effectiveness ratios (ICER) to achieve a 1‐point improvement in the PRWE were AUD 375, AUD 1736 and AUD 1126 at 3, 12 and 24 months for VLP compared with CR. At 12 months, the cost effectiveness was dominated by CR (lower cost and better QoL) whereas at 24 months, the incremental cost per QALY gained by VLP was AUD 1 946 127.

Conclusion

In the treatment of distal radius fractures in patients ≥60 years, VLP fixation was not cost‐effective compared with CR from the perspective of hospital funders.

Question What is the relative cost‐effectiveness of volar‐locking plate (VLP) fixation compared with closed reduction and cast immobilization (CR) in the treatment of displaced distal radius fractures in patients aged 60 years and older from the perspective of hospital funders? Findings in this economic evaluation of a randomized clinical trial, the mean costs for VLP fixation were AUD 6668 (95% CI 4857 to 8479) compared with AUD 3343 (95% CI 1304 to 5381) for closed reduction. At 12 months, the cost‐effectiveness was dominated by CR (lower cost and better quality of life) whereas at 24 months, the incremental cost per QALY gained by VLP was AUD 1 946 127. Meaning in the treatment of distal radius fractures in patients aged 60 years and older, VLP fixation was not cost‐effective compared with CR.

Introduction

Wrist fractures are common in older people and the incidence is increasing. ¹ In Australia, it is estimated that the number of osteoporotic wrist fractures in people older than 50 years will increase to over 25 000 in 2022, and most of these will be in people older than 65 years. ² Treatment represents a significant cost to health systems. In 2017 in Australia, there were almost 23 500 wrist fractures in people older than 50 years. ³ Total direct treatment costs were estimated to be over Australian dollars (AUD) 160 million with mean treatment costs per patient estimated to range from AUD 4607 (men 50–69 years) to AUD 8736 (women 70+ years). ³

The two most common treatments for wrist fracture are non‐surgical treatment by closed reduction and cast immobilization (CR), and surgical treatment by open reduction and volar locking plate internal fixation (VLP fixation). ⁴ Surgical treatment produces better anatomical alignment ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ and earlier recovery of function than non‐surgical treatment ¹⁰ , ¹¹ , ¹² but by 12 months post‐treatment, surgical treatment offers no clinically important advantage in pain and function. ⁶ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ Evidence regarding quality‐of‐life suggests that, although VLP fixation is associated with higher quality adjusted life years (QALYs) than CR at 12 months in a general adult population, ¹⁴ , ¹⁵ no clinically important between‐group treatment differences in quality‐of‐life scores at 12 months are evident in older populations. ¹⁰ , ¹¹ , ¹² , ¹⁶ Current evidence suggests that VLP fixation may be cost‐effective in the treatment of wrist fractures in a general adult population ¹⁴ , ¹⁵ but not in an older (>60 years) population. ¹⁶

There is significant practice variation in the management of wrist fractures in older people. ⁴ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ Given that costs for surgical treatment are as much as 10 times the cost for non‐surgical treatment, ⁸ there is potential to substantially reduce direct treatment costs by avoiding unnecessary surgery if patient outcomes are comparable. The aim of this study was to conduct a cost‐effectiveness analysis (CEA) based on our recent randomized controlled trial ¹¹ , ²¹ of surgical compared with non‐surgical treatment of wrist fractures in a cohort of older participants.

Methods

Study design

This analysis was a trial‐based CEA, based on the CROSSFIRE trial (ACTRN 12616000969460). The methods and statistical analysis for the clinical trial were reported separately. ²² , ²³ Briefly, the trial was a combined randomized controlled trial (RCT) and observational study where participants aged ≥60 years with fractures of the distal radius were randomized to VLP, or CR and cast immobilization. The age cut‐off of 60 years was chosen as appropriate by clinicians involved in the original RCT. Patients who declined participation in the RCT were included in the parallel observational study. The trial was conducted in Australia and New Zealand between December 2016 and December 2018. The evaluation was based on participants in the CROSSFIRE RCT.

Health outcomes

Outcomes from the trial used in the CEA were the patient‐rated wrist evaluation (PRWE) scores, and the EuroQoL five‐dimension five‐level tool (EQ‐5D‐5L) ²⁴ scores. The 12 and 24‐month outcomes were previously reported. ¹¹ , ²¹

The primary outcome of the CROSSFIRE trial was the PRWE questionnaire administered 12 (±1) months after injury. The PRWE is a wrist‐specific 15‐item patient‐reported measure of pain and function reported as a continuous 0–100 point score with higher scores indicating poorer outcomes. ²⁵ The PRWE minimal clinically important difference (MCID) is 14 points. ²⁶ Secondary outcomes included the PRWE score at 3 months and 24 months and quality of life (QoL) measured using the EQ‐5D‐5L at baseline and at 3, 12 and 24 months after fracture. Utility values for the EQ‐5D‐5L were calculated using the UK value set ²⁷ and used to calculate QALYs. Baseline EQ‐5D‐5L scores were obtained at recruitment in retrospect of the participants wrist injury. Retrospective measurement of QoL has been validated in orthopaedics. ²⁸

Resource use (costs)

Hospital resource use included emergency department (ED), admitted services and outpatient clinics. For this CEA, a sub‐sample of the RCT participants was used to estimate total costs associated with utilization of hospital services. From 19 recruitment sites for the RCT, we had ethics approval to access medical records at three particular sites and this defined the sub‐sample used.

All trial participants attended the ED. Patients randomized to VLP fixation were admitted for surgery and patients randomized to CR were generally treated in the ED. If CR participants were admitted, then in‐patient costs were included. Patient costs for ED attendance were estimated from the urgency related group (URG) codes URG58 (non‐admitted, injury, triage 4) for CR and URG34 (admitted to same hospital, injury, triage 4) for VLP fixation. Outpatient clinic usage for the RCT cohort was based on clinical expert opinion of the number of required follow‐up visits. CR patients were assumed to require two medical consultations under the Tier 2 non‐admitted services, one for the procedure and one as an out‐patient (Table 1). This is likely to over‐estimate average costs for CR patients as the first consult should be covered by URG58, however, some hospitals may also apply a Tier 2 non‐admitted service for treatment within the ED. VLP fixation participants were assumed to require only one medical consult under the Tier 2 outpatient service. While there may be variation between individuals in the utilization of other outpatient services such as physiotherapy, it was assumed that they were the same for CR and VLP fixation groups. The 2018–2019 Tier‐2 non‐admitted costs were used ²⁹ to estimate total outpatient clinic costs for each participant.

Table 1.

Aggregate in and out‐patient hospital costs (AUD) ²⁹

Arm	Emergency Presentation	Medical consult for procedure	Surgical procedure	Clinic	Other admissions
Volar locking plate (VLP) fixation	URG34 (admitted to same hospital, injury, triage 4) $816	NA	DRG I74A $8.021 DRG 174B $3469 DRG 19A $16 198 DRG 19B $8573	Tier 2 non‐admitted orthopaedic consult 20.29 $223	DRGs recorded for subset of 29 patients
Closed reduction (CR)	URG58 (non‐admitted, injury, triage 4) $411	Tier 2 non‐admitted orthopaedic consult 20.29 $223	NA	Tier 2 non‐admitted orthopaedic consult 20.29 $223	DRGs recorded for subsets of 27 patients

To capture hospital admission costs during the follow‐up period of the CROSSFIRE study, detailed hospital records were obtained for a sub‐sample of RCT participants and used to estimate in‐patient hospital service utilization. The hospital records covered all admissions and therefore capture complications associated with both CR and VLP fixation patients. Aggregated costs for hospital admissions were estimated using the Australian refined diagnostic related groups (AR‐DRGs) obtained from the patients' hospital records. ²⁹ All of the URG, AR‐DRGs and Tier2 clinic codes are activity‐based and assigned aggregate costs by the Independent Hospital Pricing Authority. ²⁹ These include costs for general medical, specialist, nursing, allied health and administrative time, consumables and pharmaceuticals provided by the hospital relevant to that activity.

Economic evaluation

The evaluation was undertaken from the perspective of the public hospital system and did not include costs incurred outside of public hospitals or costs borne by patients. The time horizon was the study follow‐up period of 24 months. Given the short timeframe, no discounting of costs or outcomes was applied. The CEA was based on PRWE scores at 3, 12 and 24 months and the cost utility analysis (CUA) as costs per QALY at 12 and 24 months using EQ‐5D‐5L utility scores at baseline, 3, 12 and 24 months. Given the short time frame (i.e. less than 12 months), costs per QALY were not calculated at 3 months. The assessments compared the incremental costs and benefits (QALYs or PRWE scores) of VLP fixation compared with CR.

Statistical analysis

All analyses relating to the CEA were completed using STATA V 16.1. ³⁰ This economic analysis followed the Consolidated Health Economic Evaluation Reporting Standards checklist ³¹ .

In the RCT sub‐sample, bootstrap sampling with replacement was used to generate 1000 estimates of the mean costs for in‐patient treatment by VLP fixation and by CR. Costs for the complete RCT cohort were then estimated by imputing missing values. In the base case, missing values of CR and VLP fixation treatment groups were replaced with randomly selected values from the 1000 bootstrapped mean costs for respective treatment groups.

Incremental costs were calculated from the average costs and incremental outcomes from the RCT, specifically QALYs at 12 and 24 months and PRWE scores at 3, 12 and 24 months. As lower PRWE scores indicate increasing benefit the sign of the PRWE values were reversed to align with QALYs. A positive incremental value for QALY and PRWE scores indicate benefit from VLP fixation compared to CR. Incremental costs and outcomes were then used to derive incremental cost‐effectiveness ratios (ICERs) which were expressed as the incremental cost to achieve one extra QALY or one extra point on the PRWE (i.e. 1‐point improvement).

Sensitivity analysis

Sensitivity analysis included a conservative cost case whereby missing values in the VLP fixation group were replaced by the lower 5th percentile of the bootstrap mean VLP fixation costs and missing values for the CR group by the upper 95th percentile of the bootstrap mean CR costs. This imputation method aimed to produce lower‐than‐average costs in the VLP fixation group and higher‐than‐average costs in the CR group and has been termed the ‘conservative cost case’ favouring VLP over CR. A probabilistic sensitivity analysis of costs and outcomes was also undertaken by boot strap sampling of the base case and the conservative cost case to produce cost effectiveness scatter plots of 1000 mean estimates of incremental costs and outcomes for VLP fixation compared to CR.

Results

Participant characteristics

There were 166 participants in the RCT, 81 randomized to VLP fixation and 85 to CR. Baseline characteristics of the two treatment groups were similar (Table 2). The CEA was conducted based on hospital admissions records of a sub‐sample of 56 (34%) of the 166 RCT participants (29 VLP fixation and 27 CR). In this sub‐sample, baseline EQ‐5D‐5L utility scores were higher in the CR group, compared with the VLP fixation group. Otherwise, baseline characteristics of the cost sample were comparable to the full RCT sample and not different between treatment groups (Table 2).

Table 2.

Baseline characteristics for RCT participants and subgroup of patients with available cost estimates medical records

		RCT		Cost subgroup
Baseline measure		CR (n = 85)	VLP fixation (n = 81)	CR (n = 27)	VLP fixation (n = 29)	P‐value
Age (years), mean (range)		71.3 (60–90)	70.5 (60–86)	70.9 (60–84)	67.5 (60–81)	0.051
Female n (%)		75 (88.2%)	70 (86.4%)	24 (88.9%)	25 (86.2%)	0.762
Fracture type n (%)	23A	49 (58.3%)	55 (67.9%)	13 (48.2%)	20 (69.0%)	0.114
	23C	35 (41.7%)	26 (32.1%)	14 (51.9%	9 (31.0%)
Co‐morbidities, n (%)	Diabetes? (Yes)	9 (11%)	10 (12%)	2 (7.4%)	5 (17.2%)	0.266
	Smoker? (Yes)	3 (4%)	1 (1%)	4 (14.8%)	4 (13.8%)	0.913
	Glucocorticoid treatment? (Yes)	10 (12%)	10 (12%)	0	0	N/A
	Osteoporosis treatment? (Yes)	10 (12%)	10 (12%)	6 (22.2%)	2 (6.9%)	0.101
Treatment preference n (%)	Surgery	5 (5.9%)	10 (12.7%)	4 (14.8%)	3 (10.3%)	0.467
	Closed reduction	24 (28.2%)	25 (31.6%)	7 (25.9%)	12 (41.4%)
	No preference	56 (65.9%)	44 (55.7%)	16 (59.3%)	14 (48.3%)
EQ‐5D‐5L, mean (SD)	Index	0.89 (0.14)	0.85 (0.19)	0.93 (0.10)	0.81 (0.24)	0.037
	EQ‐VAS	83.6 (14.4)	81.1 (17.4)	81.9 (12.3)	82.6 (13.6)	0.844

Cost‐effectiveness

The mean costs for the sub‐sample of RCT participants were AUD 6668 (95% CI $4896 to $8439) for VLP fixation and AUD 3343 (95% CI $1304 to $5381) for CR. Base case costs for the complete RCT cohort using imputed random bootstrap means were similar (Table 3).

Table 3.

Estimation of costs of public hospital services for RCT participants treated with CR or VLP fixation

	CR		VLP fixation
Costs	n	Cost AUD (95% CI)	n	Cost AUD (95% CI)
Cost sample	29	3343 (1304 to 5381)	27	6668 (4896 to 8439)
Base case† – missing imputation – random bootstrap mean	80	3320 (2594 to 3866)	78	6756 (6092 to 7420)
Conservative case‡	80	4250 (3805 to 4695)	78	5782 (5118 to 6447)

†Missing value imputation – random bootstrap means for VLP and CR. ‡Missing value imputation – lower 5% of bootstrap means for VLP and upper 95% of bootstrap means for CR. CR, non‐surgical closed reduction; VLP, volar‐locking plate.

Patient‐reported wrist pain and function and QoL improved across three timepoints for participants in both groups. There were no clinically important differences between treatment groups for PRWE or for QALYs. The clinical outcomes from the cost sample were similar to those from the RCT with the exception of PRWE at 12 months which was higher in the cost sample (Table 4).

Table 4.

Outcomes and ICER at three timepoints post‐treatment services for RCT participants treated with CR or VLP fixation

	3 months		12 months		24 months
Mean (95% CI)	CR	VLP fixation	CR	VLP fixation	CR	VLP fixation
Sample QALYs			0.523 (0.472 to 0.574)	0.485 (0.429 to 0.541	1.430 (1.308 to 1.553)	1.348 (1.208 to 1.488)
RCT QALYs			0.523 (0.489 to 0.556)	0.517 (0.486 to 0.548)	1.353 (1.263 to 1.443)	1.355 (1.261 to 1.450)
Incremental QALY†			−0.005 (−0.014 to 0.025)		0.002 (−0.128 to 0.132)
ICER ‐ base case ($/QALY)			Dominated†		$1,946 127.00
ICER ‐ conservative cost case ($/QALY)			Dominated†		$777 733.60
Sample PRWE	40.8 (32.5 to 49.1)	33.2 (22.4 to 44.0)	24.7 (15.1 to 34.3)	25.9 (16.0 to 35.8)	20.9 (7.5 to 34.3)	14.9 (5.6 to 24.2)
RCT PRWE	37.1 (32.2 to 42.0)	28.1 (22.9 to 33.3)	21.5 (16.2 to 26.9)	19.8 (15.1 to 24.5)	15.8 (11.0 to 20.5)	13.6 (9.4 to 17.8)
Incremental PRWE†	9.0 (1.8 to 16.2)		1.7 (−5.4 to 8.8)		2.2 (−4.2 to 8.6)
ICER ‐ base case ($/1‐point difference in PRWE)	$386.52		$2141.72		$1647.62
ICER ‐ conservative cost case ($/1‐point difference in PRWE)	$161.86		$884.51		$671.91

†Mean difference for RCT.

The base case ICER for VLP fixation compared to CR for a 1‐point PRWE improvement at 3, 12 and 24 months were AUD 375, AUD 1736 and AUD 1126. The incremental QALYs for VLP compared with CR were very small at 12 (−0.005) and 24 months (0.002) with resulting base case ICERs for one QALY being either dominated by CR at 12 months (CR had higher QALYs and lower costs) or very large at 24 months (Table 4).

Sensitivity analysis

For the conservative cost case, the VLP fixation cost was AUD 5782 ($5118 to $6447) and AUD 4250 ($3805 to $4695) for CR (Table 3). ICERs for the conservative cost case were AUD 162 at 3 months, AUD 885 at 12 months and AUD 672 at 24 months, per 1‐point PRWE improvement. Incremental cost per QALY gained at 12 months was dominated by CR (less costly and had higher QALYs). The conservative estimate for the incremental cost per QALY gained at 24 months post‐treatment was AUD 777,733.60 (Table 4).

Uncertainty in the estimates of incremental costs of PRWE scores and QALYs are shown in Figures 1 and 2. The probability that incremental PRWE scores for VLP fixation were clinically meaningful at 3 months was 10% and 12% for the base case and conservative cost cases respectively. The probability of a clinically meaningful improvement in PRWE at 12 and 24 months was less than 0.5% for the base case and conservative cost case. The probability of VLP fixation being more costly and less effective was 0.5% and 0.2% for PRWE at 3 months, 28% and 26% for PRWE at 12 months, and 19% and 22% for PRWE at 24 months for the base and conservative cases respectively. The QALY cost‐effectiveness scatter plots show an approximately equal spread of positive and negative incremental QALYs at 12 and 24 months (Fig. 2).

Fig. 1.

Incremental cost‐effectiveness planes. (a) Incremental cost‐effectiveness for PRWE score at three timepoints for VLP fixation compared to CR. (b) Incremental cost‐effectiveness for QALY at 12 and 24 months for VLP fixation compared to CR.

Fig. 2.

Incremental cost‐effectiveness planes for QALY at 12 and 24 months for VLP fixation compared to CR.

Discussion

Main findings

In the treatment of displaced distal radius fractures in patients ≥60 years, VLP fixation was associated with higher costs and no clinically meaningful benefit in patient‐reported wrist function at 3, 12 and 24 months. Similarly, the cost utility analysis showed no meaningful benefits in QALYs at 12 and 24 months with either very high ICER values (24 months) or VLP fixation being dominated by CR (12 months). Overall, the cost‐effectiveness analysis indicated that the higher costs of VLP fixation were not cost‐effective from the perspective of hospital funders and the ICERs did not approach any accepted willingness to pay thresholds. The small probability of VLP fixation providing a clinically meaningful benefit in wrist function at 3 months post‐treatment (10% base case and 12% conservative case) was offset by a higher probability that VLP fixation would be more costly and have poorer wrist function than CR at 12 months (28% for base case and 26% for conservative case) and 24 months (19% for base case and 22% for conservative case).

Comparison with similar studies

Two secondary analyses ¹⁴ , ¹⁵ of the internal plate fixation versus plaster in complete extra‐articular distal radius fractures (VIPER) and volar internal plate fixation versus plaster in complete articular distal radius fractures (VIPAR) RCTs ³² , ³³ compared the costs of VLP fixation with CR in the treatment of type A and type C distal radius fractures in a population aged 18 to 75 years. Each study found VLP fixation produced better function and QoL at 12 months than CR. Both studies included direct medical and non‐medical costs as well as indirect costs (e.g. lost productivity). For type A fractures, treatment costs were lower with VLP fixation. ¹⁴ For type C fractures, while the direct medical costs were higher with surgical treatment, VLP fixation was found to be cost‐effective, especially for patients in paid employment. ¹⁵ Our study included patients with either type A or C fractures, however, all were aged ≥60 years. Our economic analysis was limited to hospital costs and we found that VLP fixation was unlikely to be cost‐effective from the perspective of the public hospital funder.

Another comparable study was the secondary economic analysis of The Wrist and radius injury surgical trial (WRIST) that followed type A distal radius fracture patients aged >60 years treated non‐surgically (CR) or surgically (including VLP fixation, external fixation and percutaneous pinning). ¹⁶ The authors concluded that CR was the most cost‐effective treatment in this age group. Our study included a similar cohort (we also included type C fractures) and arrived at similar conclusions.

Strengths and limitations

A strength of our study was the availability of baseline QoL data for all trial participants. However, we were only able to access detailed medical records and therefore public hospital costs for a sub‐set of the trial participants. Although the sample QALYs and the RCT QALYs differed at 12 months, the risk of selection bias was minimal as this sub‐sample were broadly comparable to the full RCT cohort based on baseline characteristics and other patient‐reported outcomes. In addition, cost estimates were in accord with published hospital treatment costs for distal radius fractures in older patients in Australia ³ , ³⁴ and we captured a similar proportion of hospital costs for each treatment group. To address this uncertainty, we undertook sensitivity analyses that included a conservative case for the cost‐effectiveness analysis that was likely to underestimate costs for the VLP fixation group and overestimate costs for the CR treatment arm. In addition, utilization of outpatient services was based on assumptions of minimal requirements for the two procedures and did not include additional or differential use of physiotherapy or other clinics. Indeed, in the analysis of 12 months outcomes, we found significantly higher therapy utilization in the VLP fixation group (72% vs. 54%; RR, 1.32; 95% CI, 1.04–1.69; P = 0.02), providing another reason why overall costs to the hospital for VLP fixation are likely to be underestimated. ¹¹

Our study was based on certain generalities, for example, patients aged ≥60 years. We acknowledge that clinical decision‐making is based on individual patient characteristics and that our findings cannot be generalized to all patients or periods beyond 12 months but adds context to an individual decision‐making process. Another limitation was that our study was undertaken from the hospital perspective and did not include other costs that might be important in shared clinical decision‐making, such as costs arising from co‐payments, travel or lost productivity to the patient and their family or carers. The costs of hospital services were estimated based on a subset of trial patients and did not cover all hospitals. As noted, the cost sample was representative of the RCT participants but there may be between‐site differences based on practice patterns that are not captured in our cost estimates based on activity‐based aggregate costs. For example, different lengths of stay or different outpatient clinic usage could introduce uncertainty that has not been fully captured and may affect generalizability of reported costs to clinical practice.

Conclusion

In the treatment of displaced distal radius fractures in patients 60 years and older, VLP fixation was associated with higher costs and no benefit in patient‐reported wrist function at 3, 12 and 24 months. Similarly, the cost utility analysis showed small and clinically unimportant benefit of VLP fixation over CR in QALYs at 12 and 24 months. Compared with CR, VLP fixation was not cost‐effective from the perspective of hospital funders.

Declarations

Trial registration

The trial was registered on 22nd July 2016 with The Australian and New Zealand Clinical Trials Registry (ANZCTR Number; ACTRN12616000969460) and can be viewed at http://www.ANZCTR.org.au/ACTRN12616000969460.aspx. The trial protocol was published in June 2017 (doi: 10.1136/bmjopen‐2017‐016100) and the statistical analysis plan was published in July 2020 (doi.org/10.1186/s13063‐020‐4228‐0). The 12‐month outcomes were published in January 2021 (doi:10.1001/jamasurg.2020.5672) and the 24‐month outcomes were published in April 2022 (doi:10.1001/jamasurg.2022.0809).

Patient and public involvement

The Consumer Advisory Group of the Australia and New Zealand Musculoskeletal (ANZMUSC) Clinical Trial Network reviewed the protocol and the study was endorsed by ANZMUSC. Separately, three older wrist fracture patients (who were not study participants) were interviewed and provided feedback on what post‐treatment information was most relevant and important to older wrist fracture patients. This was used to develop the printed participant information (Appendix 1).

Conflict of interest

None declared.

Ethical approval

CROSSFIRE was approved by The Hunter New England Human Research Ethics Committee (HNEHREC Reference No: 16/02/17/3.04). All study participants provided informed consent.

Funding information

Grant funding has been received from NHMRC Project Grant (2016, APP1098550). Ian A. Harris, Justine Naylor and Kirsten Howard were authors on the grant application. Also, project funding was received from The Australian Orthopaedic Association Research Foundation, AO Trauma Asia Pacific and The Lincoln Foundation. Funders played no role in study design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Author contributions

Martin Howell: Conceptualization; data curation; formal analysis; investigation; methodology; software; validation; writing – original draft; writing – review and editing. Andrew Lawson: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; validation; writing – original draft; writing – review and editing. Justine Naylor: Conceptualization; data curation; funding acquisition; methodology; project administration; supervision; validation; writing – review and editing. Kirsten Howard: Conceptualization; funding acquisition; methodology; supervision; validation; writing – review and editing. Ian Harris: Conceptualization; data curation; funding acquisition; methodology; project administration; supervision; validation; writing – review and editing.

Data availability statement

Deidentified participant‐level data set and statistical code will be made available for collaborative research projects, on request of the chief investigator.