Purpose: We aimed to preliminarily validate a newly developed system, the radiographic union scale for tibial (RUST) fracture healing. We hypothesized that RUST would demonstrate better inter-rater reliability than assessment of the number of cortices bridged and correlate with functional outcomes at least as strongly as surgeon’s assessment of cortical bridging.

Method: Three blinded orthopaedic trauma surgeons independently assigned a RUST score and a number of cortices bridged by callus (zero to four) to each set of AP and lateral radiographs at each follow up period. RUST is scored from four (definitely not healed) to 12 (definitely healed) based on the presence or absence of callus and of a visible fracture line at the total of four cortices visible.

Results: For 549 sets of reviewed radiographs, inter-rater reliability for RUST scores were found to be substantially higher than for assessment of the number of cortices bridged (intra-class correlation coefficient=0.84; 95% CI, 0.80–0.87 versus kappa = 0.73; 95% CI, 0.64 – 0.81, respectively). Both methods of assessing radiographic healing were strongly correlated with weight-bearing status (r and ρ> 0.50), moderately correlated with patient-reported functional recovery and the SF-36 Physical Functioning component scores (r and ρ> 0.30), and minimally correlated with HUI Mark II scores, return to work, and the SF-36 Role Physical component and Physical Component Summary scores (r and ρ> 0.10). Neither assessment was correlated with patient-reported pain scores. All correlations were similar for RUST and the number of cortices bridged.

Conclusion: This study provides preliminary evidence that RUST can be used as a valid and reliable alternative assessment of tibial fracture healing.

Purpose: To investigate

if adding the prospect of co-authorship to a survey’s final paper would increase, and

Method: We identified orthopaedic surgeons through the internet-based Orthopaedic Trauma Association member list. All surgeons received the same questionnaire. In a factorial randomized, controlled fashion, they were allocated

to receive or not receive an additional cover page promising co-authorship of the survey’s final paper if they filled in and returned the survey (an “academic incentive”), and

Results: For 429 surveyed surgeons, six-week response rates were similar for surgeons in the incentive – and no-incentive groups (36.8% vs. 35.4%, respectively, p=0.39). Similarly, response rates did not significantly differ between emailed and faxed surgeons (32.9% vs. 39.9%, respectively, p=0.13). The mean time to response seemed shorter in the incentive-group than in the no-incentive group (p=0.058).

Conclusion: We cannot recommend promising co-authorship to increase the response rates of surveys to orthopaedic surgeons. Additionally, emailed and faxed surveys yielded statistically similar response rates, leaving the decision regarding what modality to employ to time and money constraints.

Purpose: Estimating recruitment for clinical trials is vital to ensuring the feasibility of larger multi-centre trials. We compared estimates of potential recruitment from a prospective eight-week screening study and a retrospective chart review across sites participating in three fracture management trials.

Method: During the planning phase of two multi-centre, randomized controlled trials regarding the operative treatment of hip (two studies) and tibial shaft (one study) fractures, 74 clinical sites provided estimates of the annual recruitment rate both retrospectively (based on chart reviews) and prospectively. The prospective estimate was generated by screening all incoming patients for eligibility in the concerning trial, without actually enrolling any patient, for eight weeks. These prospective and retrospective estimates were correlated with each other (for 74 sites) and with actual one-year recruitment rates in the definitive trial (for nine sites).

Results: On average, a centre’s prospective estimate was only slightly lower than its retrospective estimate (3.1 patient-difference, p=0.64). Both predictions were substantial overestimations of recruitment in the definitive trial; only 31% (95% confidence interval: 28%–35%) of retrospectively estimated patients and 34% (95% confidence interval: 30%–37%) of prospectively estimated patients were recruited in the definitive trials (p< 0.001 and p=0.001 for both overestimations, respectively). The overall costs of conducting retrospective chart reviews and prospective screening studies in 65 sites were $68,107 ($CAN) and $153,725 ($CAN), respectively.

Conclusion: Compared to relatively simple and inexpensive chart reviews, prospectively screening for eligible patients at clinical sites did not result in more accurate predictions of accrual in large randomized controlled trials.