Two One-sided Tests Research Articles

Using microwave-accelerated digestion instead of dry ashing during sodium analysis of low-moisture, part-skim mozzarella

Mineral analysis of cheese products is a labor- and time-intensive process. Pretreatment using dry or atmospheric wet ashing to decompose organic matter before measurement on a spectrometer can take more than 24 h to complete. Samples of low-moisture, part-skim mozzarella (LMPS) cheese were digested using microwave-accelerated digestion (MAD) and conventional dry ashing. A comparison of 23 samples, in duplicate, was performed to determine equivalency. A two one-sided test (TOST) analysis showed that the confidence intervals were within acceptable differences for standard method variation (less than 70 mg/100 g per International Dairy Federation method 119:2007). Bias was observed for the quantification of sodium in microwave-digested samples, averaging 14 mg/100 g higher. Based on the TOST criteria, we concluded that the methods are equivalent.

Retrospective analysis of the long-term effect of subgingival air polishing in supportive periodontal therapy.

Glycine powder air polishing (GPAP) procedure has become popular. Aim of the analysis was to compare the clinical outcomes during supportive periodontal therapy (SPT) of subgingival application of GPAP with those using sole conventional mechanical debridement (SC). Over a median SPT period of 5.3years (re-evaluation through last observation), the GPAP cohort (n=263) received supra- and subgingival biofilm removal with GPAP. Supragingival calculus was removed using curets and sonic scalers here. Patients in the SC cohort (n=264) were treated with sonic scalers, curets and rubber cup polishing only. Changes in, that is pocket probing depth (PPD) and furcation involvement were assessed retrospectively. A bootstrapping equivalence testing method in line with the principle of the two one-sided tests (TOST) procedure was used to compare clinical outcomes. The GPAP procedure was statistically equivalent to SC regarding the number of sites with stable PPDs (83.3%; IQR 68.8%, 91.0% vs. 84.0%; IQR 77.8%, 90.0%). However, in the GPAP cohort, a trend towards deterioration in furcation status (no equivalence) was noted. In periodontal maintenance, the use of GPAP instead of mechanical plaque removal does not improve the clinical outcome. It seems to be contraindicated to treat furcation defects with GPAP only.

New Model-Based Bioequivalence Statistical Approaches for Pharmacokinetic Studies with Sparse Sampling.

In traditional pharmacokinetic (PK) bioequivalence analysis, two one-sided tests (TOST) are conducted on the area under the concentration-time curve and the maximal concentration derived using a non-compartmental approach. When rich sampling is unfeasible, a model-based (MB) approach, using nonlinear mixed effect models (NLMEM) is possible. However, MB-TOST using asymptotic standard errors (SE) presents increased type I error when asymptotic conditions do not hold. In this work, we propose three alternative calculations of the SE based on (i) an adaptation to NLMEM of the correction proposed by Gallant, (ii) the a posteriori distribution of the treatment coefficient using the Hamiltonian Monte Carlo algorithm, and (iii) parametric random effects and residual errors bootstrap. We evaluate these approaches by simulations, for two-arms parallel and two-period, two-sequence cross-over design with rich (n = 10) and sparse (n = 3) sampling under the null and the alternative hypotheses, with MB-TOST. All new approaches correct for the inflation of MB-TOST type I error in PK studies with sparse designs. The approach based on the a posteriori distribution appears to be the best compromise between controlled type I errors and computing times. MB-TOST using non-asymptotic SE controls type I error rate better than when using asymptotic SE estimates for bioequivalence on PK studies with sparse sampling.

Does it matter how we evaluate HRQOL? Longitudinal comparison of the EORTC QLQ-C30/QLQ-OG25 and FACT-E.

To determine whether EORTC QLQ-C30/QLQ-OG25 and FACT-E compared longitudinally provide similar reflections of health-related quality of life (HRQOL). Eighty-six esophageal cancer patients treated with curative intent, scheduled to complete both questionnaires at baseline and post-treatment time points until 36months. A generalized estimating equation model utilizing a Gaussian family compared instruments longitudinally. The two-one-sided-test (TOST) method assessed equivalence between the instruments. Trajectories for social domain and overall quality of life differed significantly between instruments. Also, FACT-G's functional well-being post-treatment returns to baseline 3-6months earlier than the EORTC QLQ-C30's role functioning subscale, suggesting measurement of different components. Trajectories for physical and esophageal symptom subscales are similar and are deemed equivalent. Emotional domains are comparable and bear little resemblance to the physical domain trajectories indicating reflection of emotional experience rather than a physical proxy. EORTC QLQ-C30 subscales have a trajectory similar to its physical functioning scale except for the emotional and esophageal symptoms scales. Overall HRQOL in both instruments showed a consistent return to baseline/pre-treatment levels by 6months post-treatment. Overall HRQOL recovers earlier after curative-intent treatment than previously reported despite persistence of physical symptoms, with a consistent return to pre-treatment levels by 6months after treatment. This supports the concept that HRQOL is not primarily defined by physical function. Based on this longitudinal comparison, FACT-E provides a more multidimensional assessment of HRQOL. Curative intent treatment for esophageal cancer has adverse effects on HRQOL but despite intense treatment, overall HRQOL recovers within 6months.

A comparative study of TOST and UMPT procedures for evaluating dispersion equivalence

ABSTRACT The two one-sided tests (TOST) procedure is widely applied for assessing bioequivalence between two different drugs in clinical studies. The same notion can be applied to construct TOST method for establishing dispersion equivalence. In addition to the TOST extension, a uniformly most power test (UMPT) for equivalence in variability has been described in the literature. This article presents analytic and numerical comparisons to clarify the limitation and strength of the two procedures. The UMPT approach has comparatively superior control of Type I errors and demands more involved computations than the TOST method. To enhance the utility of the recommended UMPT procedure, a full set of computer algorithms are constructed to implement the described hypothesis testing and power calculation. Optimal sample size determinations are also developed for planning equivalence designs under different allocation and cost concerns. The suggested procedures are demonstrated with the data regarding the dispersion comparability between two measuring devices of breaking strengths in dental medicine.

Efficient model-based bioequivalence testing.

The classical approach to analyze pharmacokinetic (PK) data in bioequivalence studies aiming to compare two different formulations is to perform noncompartmental analysis (NCA) followed by two one-sided tests (TOST). In this regard, the PK parameters area under the curve (AUC) and $C_{\max}$ are obtained for both treatment groups and their geometric mean ratios are considered. According to current guidelines by the U.S. Food and Drug Administration and the European Medicines Agency, the formulations are declared to be sufficiently similar if the $90\%$ confidence interval for these ratios falls between $0.8$ and $1.25 $. As NCA is not a reliable approach in case of sparse designs, a model-based alternative has already been proposed for the estimation of $\rm AUC$ and $C_{\max}$ using nonlinear mixed effects models. Here we propose another, more powerful test than the TOST and demonstrate its superiority through a simulation study both for NCA and model-based approaches. For products with high variability on PK parameters, this method appears to have closer type I errors to the conventionally accepted significance level of $0.05$, suggesting its potential use in situations where conventional bioequivalence analysis is not applicable.

Equivalence Testing and the Second Generation P-Value


 
 
 To move beyond the limitations of null-hypothesis tests, statistical approaches have been developed where the observed data are compared against a range of values that are equivalent to the absence of a meaningful effect. Specifying a range of values around zero allows researchers to statistically reject the presence of effects large enough to matter, and prevents practically insignificant effects from being interpreted as a statistically significant difference. We compare the behavior of the recently proposed second generation p-value (Blume, D’Agostino McGowan, Dupont, & Greevy, 2018) with the more established Two One-Sided Tests (TOST) equivalence testing procedure (Schuirmann, 1987). We show that the two approaches yield almost identical results under optimal conditions. Under suboptimal conditions (e.g., when the confidence interval is wider than the equivalence range, or when confidence intervals are asymmetric) the second generation p-value becomes difficult to interpret. The second generation p-value is interpretable in a dichotomous manner (i.e., when the SGPV equals 0 or 1 because the confidence intervals lies completely within or outside of the equivalence range), but this dichotomous interpretation does not require calculations. We conclude that equivalence tests yield more consistent p-values, distinguish between datasets that yield the same second generation p-value, and allow for easier control of Type I and Type II error rates.
 
 

Statistical Considerations on Clinical Efficacy Studies of Biosimilar for PMDA Submission.

In general, similar to FDA and other health authorities, the PMDA requires clinical efficacy study(ies) to evaluate equivalence between a reference biological product and a Biosimilar product for new drug applications. Even if an identical clinical efficacy study is included in both of PMDA and FDA submissions, the coefficients of confidence interval (CI) used for comparison with the equivalence margins could be different between the two submissions (e.g., 95% CI vs. 90% CI). In this article, we will focus on clinical efficacy studies of Biosimilar products and provide an overview of the two one-sided tests (TOST) and the type I error rate for equivalence design. Then, we summarize published PMDA review reports of Biosimilar products in terms of the coefficients of CI and other elements of the primary endpoints, and explain some Japanese guidelines of Biosimilar and Statistics behind the difference between PMDA and FDA submissions. In addition, we discuss how to use statistical methods correctly and efficiently for PMDA submissions.

Assessing Biogeography of Coffee Rust Risk in Brazil as Affected by the El Niño Southern Oscillation.

The El Niño Southern Oscillation (ENSO) is an oceanic-atmospheric phenomenon influencing worldwide weather and climate. Its occurrence is determined by the sea surface temperature (SST) anomaly of the 3.4 Niño region in the Pacific Ocean (5°N-5°S, 120°-170°W). El Niño (EN), Neutral (NT), and La Niña (LN) are the three possible phases of ENSO, respectively, for warm, normal, and cold SST anomaly. As in other regions around the world, weather in Brazil is influenced by ENSO phases. The country is the major coffee producer in the world, and production is strongly influenced by weather conditions, which affect plant yield, harvest quality, and interactions with pests and diseases. Coffee leaf rust (CLR), caused by the fungus Hemileia vastatrix, is a major cause of coffee yield and quality losses in Brazil, and requires fungicide spray applications every season. Because CLR is highly influenced by weather conditions, it is possible to use weather variables to simulate its progress during the cropping cycle. Therefore, the aims of this study were to estimate CLR infection rate based on a validated empirical model, which has daily minimum air temperature and relative humidity as inputs, and to assess the extent of ENSO influence on the annual risk of this disease at 45 sites in Brazil. Cumulative infection rates (CIR) were estimated daily from October to June of each growing season and location, based on the prevailing ENSO phase. Differences between the extreme phases (EN-LN) were assessed by the Two-One-Sided-Tests (TOST) method. Analysis of data from eight sites, located mainly in Paraná State, provided evidence of CIR differences between EN and LN phases (G1). Evidence of no difference of CIR between EN and LN was found in 18 sites (G2), whereas 19 sites showed no evidence of differences (G3) due to relatively large variation of CIR within the same ENSO phase. The G1 sites are located mostly in Southern Brazil, where ENSO exerts a well-defined influence on rainfall regime. In contrast, the G2 sites are mainly in Minas Gerais State, which is characterized as a transition region for ENSO influence on rainfall. The G3 sites are located between the northern region of Minas Gerais State and southern region of Bahia State, which is characterized by a subhumid climate that is usually very dry during winter, and where rainfall can vary up to 300% from one year to another, influencing relative humidity and resulting in a high CIR variability. Therefore, ENSO had a well-defined influence on CIR only in Paraná State, a region with minor importance for coffee production in Brazil. No ENSO influence was found in more northerly zones where the majority of Brazilian coffee is produced. This is the first evidence of ENSO-linked regional impact on the risk of coffee rust.

Assessing Agreement Between Two Methods of Quantitative Measurements: Exact Test Procedure and Sample Size Calculation

Assessment of agreement between measurement methods is an important problem in medical practice and research. Several two one-sided tests (TOST) procedures have been proposed to evaluate whether two methods have sufficient agreement. These TOST techniques of limits of agreement or normal percentiles are natural generalizations of the favorable TOST procedure of mean differences. But their fundamental behavior of Type I error control has not been examined. This article describes an improved and exact approach for declaring agreement between two measurement methods when the range of central proportion of paired differences is within the designated threshold interval around zero. Theoretical examinations and numerical investigations are provided to justify the advantages of the proposed approach and to illustrate the shortcomings of the existing TOST procedures for assessing agreement between two methods. To facilitate the suggested agreement appraisal in design planning, the corresponding power and sample size formulas are also derived. Both SAS and R algorithms are presented for conducting agreement test, power calculation, and sample size determination in method comparison studies.

Comparison of the effects of perineural or intravenous dexamethasone on low volume interscalene brachial plexus block: a randomised equivalence trial

BackgroundEfforts to prolong interscalene block (ISB) analgesia include the use of local anaesthetic adjuvants such as dexamethasone. Previous work showing prolonged block duration suggests that both perineural and intravenous (i.v.) routes can both prolong analgesia. The superiority of either route is controversial given the design of previous studies. As perineural dexamethasone is an off-label use, anaesthesiologists should be fully informed of the clinical differences, if any, on block duration. This study was designed to test whether perineural vs i.v. dexamethasone administration are equivalent. MethodsWe randomised 182 eligible patients scheduled for arthroscopic shoulder surgery to receive low-dose ISB (0.5% ropivacaine 5 ml) with perineural or i.v. dexamethasone 4 mg. Subjects, anaesthesiologists, and research personnel were blinded. All subjects also received a standardised general anaesthetic and multimodal analgesia. The primary outcome was duration of analgesia analysed as an equivalence outcome (2 h equivalency margin) using the two one-sided test (TOST) method. ResultsFor the primary outcome, duration of analgesia, and perineural and i.v. administration of dexamethasone were not equivalent. The upper and lower bounds of the 90% confidence interval were 1 h (P=0.12) and –2.5 h (P=0.01), respectively. The observed difference in mean block duration was not clinically relevant (0.75 h longer for i.v. dexamethasone). There were no other clinically significant differences between groups. ConclusionIn the context of low-volume ISB with ropivacaine, perineural and i.v. dexamethasone were not equivalent in terms of their effects on block duration. However, there were no clinically significant differences in outcomes, and there is no advantage of perineural over intravenous dexamethasone. www.clinicaltrials.gov registrationNCT02322242.

221. Intravenous and oral tranexamic acid are equivalent at reducing blood loss in thoracolumbar spinal fusion: a prospective randomized trial phase II

BACKGROUND CONTEXT The use of antifibrinolytic agents such as tranexamic acid (TXA) to decrease operative blood loss and allogenic blood transfusions is well documented in the literature. While evidence supports the use of intravenous (IV) and topical formulations of TXA in spine surgery, the use of oral (PO) TXA has not been studied. PURPOSE The objective of the study is to compare perioperative blood loss in patients undergoing elective posterior thoracolumbar fusion who were treated with IV vs PO TXA. STUDY DESIGN/SETTING A prospective randomized trial of patients enrolled at a university affiliated tertiary medical center between February 2017 and October 2018. PATIENT SAMPLE Enrolled patients were randomly assigned between the 2 treatment groups of IV and oral TXA using a random number algorithm. 90 patients received IV TXA and 81 patients received PO TXA. OUTCOME MEASURES The primary outcome was the reduction of hemoglobin. Secondary outcomes included calculated blood loss, drain output, postoperative transfusion, complications, and length of hospital stay. METHODS A total of 171 patients undergoing thoracolumbar fusion were randomized to receive 1.95g of PO TXA 2 hours preoperatively or 2g IV TXA (1g before incision and 1g before wound closure) intraoperatively. The sample was further stratified into 3 categories based on number of levels fused (1-2 level fusions, 3-5, and >5). Equivalence analysis was performed with a two one-sided test (TOST). A P-value of RESULTS Patient demographic factors were similar between groups except for age, weight and BMI. The mean reduction of hemoglobin was similar between IV and PO groups (3.48 g/dL vs 3.19 g/dL, respectively; P = 0.004, equivalence). Similarly, the calculated blood loss was equivalent (1274 mL vs 1206 mL, respectively; P = 0.001, equivalence). A total of 21 patients (23%) in IV TXA group received a transfusion compared with 10 patients in PO TXA group (13%) (P = 0.07). Two patients (2% and 3% in IV and PO, respectively) in each group experienced a DVT/PE (P = 0.90). CONCLUSIONS Patients treated with IV and PO TXA experienced the same perioperative blood loss after spinal fusions. Given its lower cost, PO TXA represents an excellent alternative to IV TXA in patients undergoing elective posterior thoracolumbar fusion and may improve health care cost-efficiency in the studied population. FDA DEVICE/DRUG STATUS Unavailable from authors at time of publication.

Accuracy and Reproducibility of Linear and Angular Measurements in Virtual Reality: a Validation Study.

The purpose of this experimental study is to validate linear and angular measurements acquired in a virtual reality (VR) environment via a comparison with the physical measurements. The hypotheses tested are as follows: VR linear and angular measurements (1) are equivalent to the corresponding physical measurements and (2) achieve a high degree of reproducibility. Both virtual and physical measurements were performed by two raters in four different sessions. A total of 40 linear and 15 angular measurements were acquired from three physical objects (an L-block, a hand model, and a dry skull) via the use of fiducial markers on selected locations. After both intra- and inter-rater reliability were evaluated using inter-class coefficient (ICC), equivalence between virtual and physical measurements was analyzed via paired t test and Bland-Altman plots. The accuracy of the virtual measurements was further estimated using two one-sided tests (TOST) procedure. The reproducibility of virtual measurements was evaluated via ICC as well as the repeatability coefficient. Virtual reality measurements were equivalent to physical measurements as evidenced by a paired t test with p values of 0.413 for linear and 0.533 for angular measurements and Bland-Altman plots in all three objects. The accuracy of virtual measurements was estimated to be 0.5mm for linear and 0.7° for angular measurements, respectively. Reproducibility in VR measurements was high as evidenced by ICC of 1.00 for linear and 0.99 for angular measurements, respectively. Both linear and angular measurements in the VR environment are equivalent to the physical measurements with high accuracy and reproducibility.

Development of Sound Field Audiometry System for Small Audiometric Booths and Comparison of Its Equivalence With Traditional System.

ObjectivesSound field (SF) audiometry tests are usually conducted in audiometric booths measuring greater than 2×2 m in size. However, most private ENT clinics carry about 1×1-m-sized audiometric booths, making SF audiometry testing difficult to perform. The aims of this study were to develop an SF audiometry system for use in smaller audiometric booths and compare its performance with traditional system.MethodsThe newly developed SF audiometry system can yield an SF signal at a distance of about 30 cm from the subject’s ears. Its height can be adjusted according to the subject’s head height. We compared SF hearing results between the new SF system and the traditional SF audiometry system in 20 adults with normal hearing (40 ears) and 24 adults with impaired hearing levels (38 ears) who wore hearing aids. Comparative parameters included warble tone audiometry threshold, a speech reception threshold (SRT), and a speech discrimination score (SDS). For statistical analysis, paired t-test was used. The equivalence of both SF systems was tested using two one-sided test (TOST) with a margin of 5 dB (normal hearing participants) and 10 dB (hearing aids wearing participants).ResultsAmong participants with normal hearing, warble tone hearing thresholds of 0.5, 1, 2, and 4 kHz, average values of these four frequencies, and SRT were similar between the two systems (all P>0.05). Participants with hearing aids showed similar warble tone threshold and SRT (P>0.05) in both systems except for threshold of 4 kHz (P=0.033). SDS was significantly higher in the newly developed system (P<0.05). TOST results showed equivalent SF audiometry results using either system.ConclusionAudiometric results of the newly developed SF audiometry system were equivalent to those of a traditional system. Therefore, the small SF audiometry system can be used at small audiometric booths present in most private ENT clinics.

Intravenous and Oral Tranexamic Acid Are Equivalent at Reducing Blood Loss in Thoracolumbar Spinal Fusion: A Prospective Randomized Trial.

A prospective randomized trial of patients enrolled at a university affiliated tertiary medical center between February and December 2017. To compare perioperative blood loss in patients undergoing elective posterior thoracolumbar fusion who were treated with intravenous (IV) versus oral (PO) tranexamic acid (TXA). The use of antifibrinolytic agents such as TXA to decrease operative blood loss and allogenic blood transfusions is well documented in the literature. While evidence supports the use of IV and topical formulations of TXA in spine surgery, the use of PO TXA has not been studied. Eighty-three patients undergoing thoracolumbar fusion were randomized to receive 1.95 g of PO TXA 2 hours preoperatively or 2 g IV TXA (1 g before incision and 1 g before wound closure) intraoperatively. The sample was further stratified into three categories based on number of levels fused (1-2 level fusions, 3-5, and >5). The primary outcome was the reduction of hemoglobin. Secondary outcomes included calculated blood loss, drain output, postoperative transfusion, complications, and length of hospital stay. Equivalence analysis was performed with a two one-sided test (TOST). A P-value of <0.05 suggested equivalence between treatments. Fourty three patients received IV TXA and 40 patients received PO TXA. Patient demographic factors were similar between groups except for body mass index (BMI). The mean reduction of hemoglobin was similar between IV and PO groups (3.36 g/dL vs. 3.43 g/dL, respectively; P = 0.01, equivalence). Similarly, the calculated blood loss was equivalent (1235 mL vs. 1312 mL, respectively; P = 0.02, equivalence). Eight patients (19%) in IV TXA group received a transfusion compared with five patients in PO TXA group (13%) (P = 0.44). One patient (2% and 3% in IV and PO, respectively) in each group experienced a deep venous thrombosis/pulmonary embolism (P = 0.96). Patients treated with IV and PO TXA experienced the same perioperative blood loss after spinal fusions. Given its lower cost, PO TXA represents an excellent alternative to IV TXA in patients undergoing elective posterior thoracolumbar fusion and may improve healthcare cost-efficiency in the studied population. 1.

Are prostate cancer patients with BRCA1 and BRCA2 mutations safe for active surveillance?

19 Background: Active surveillance (AS) is recommended as a treatment option for men presenting with low risk (Gleason 3+3) and some intermediate risk (Gleason 3+4) prostate cancer. BRCA1 or 2 germline mutations have been implicated in prostate cancer pathogenesis. It is unknown if germline BRCA1/2 mutations in AS candidacy are associated with more aggressive histologic grade, higher stage or other worse genetic alterations, such as RB1 and p53 deletions. Methods: We analyzed sequencing data from 498 men who underwent radical prostatectomy from The Cancer Genome Atlas (TCGA) data set. The primary outcome was the difference in the proportions of AS candidates among subjects with and without BRCA homodeletions. Tests for differences in the proportions were conducted using Fisher’s Exact Test. Equivalence tests for proportions of AS candidates were conducted using the two-one sided tests (TOST) method. As a secondary outcome we studied the associated coincident mutations in the men with BRCA1 and BRCA2 homodeletions. Results: Forty-one men (8%) of the cohort had homodeletion of BRCA1 or BRCA2. Ten men (2%) had complete loss of BRCA1 while 31 (6%) had loss of BRCA2. Rates of candidacy for AS based on histology and stage (defined as stage T2, Gleason 6) are not different between subjects with and without BRCA 1 or 2 homodeletions, within an equivalence margin of 10 percentage points. These findings are similar when the AS criteria are modified to add Gleason 3+4 subjects. Fifty percent of men with organ confined (pT2), 3+3 and 3+4 prostate cancer with BRCA1 or BRCA2 homodeletions had concomitant RB1 deletions compared with 16.5% of the entire cohort (p = 0.002). This was primarily driven by BRCA2 deletions co-occurrent with RB1 deletions (log OR: 2.4, p &lt; 0.001). Twenty-nine percent of men from this group had concomitant p53 deletions compared to 7.5% of the entire cohort (p = 0.004). Conclusions: Men with prostate cancer and BRCA1 or BRCA2 homodeletions present with similar stage and grade tumors than men without these deletions. Despite having low or low intermediate grade histology, BRCA1 and BRCA2 deleted tumors are enriched with deletions in RB1 and TP53, both of which are associated with more aggressive phenotypes and treatment resistance.

Auspicious or suspicious — Does religiosity really promote elder well-being? Examining the belief-as-benefit effect among older Japanese

Recent findings suggest that the belief-as-benefit effect (BABE) — the positive association between religiosity and health/well-being — is a spurious correlation voided by personality traits. The current paper investigates the cross-sectional relationships among personality, religiosity and psychological well-being in an older adult sample randomly-selected from Tokyo, Japan. Correlation and Hierarchical Regression Modelling (HRM)—with a two one-sided test (TOST) of equivalence—is utilized. The standard BABE correlation is reproduced. However, HRM utilizing a meaningful benchmark of effect (β ≥ .15) largely neuters the result after controlling for trait agreeableness and conscientiousness. Religiosity does remain statistically related to the purpose in life and positive relationships sub-scales; though it explains just a sliver of variance in both instances. Compared to religiosity, agreeableness, conscientiousness and education level were more substantial and consistent well-being predictors. Whether religiosity auspices the psychological health of older Japanese adults thus remains to be established.

A new statistical method to test equivalence: an application in male and female eastern bluebird song

Understanding complex natural systems requires approaches with minimal statistical limitations and biases. Previously, however, the fields of behaviour, ecology and evolution generally have been restricted to evaluating differences between statistical distributions. This framework can be very powerful. Unfortunately, it has also created a bias in these, and many other, fields towards publications focusing only on phenomena for which there are statistically significant differences. However, there is a wide range of questions that would benefit from reversing the existing paradigm and testing, instead, for equivalence. We have adapted the two one-sided test (TOST), also known as equivalence testing, from pharmaceutical science to be applicable to behavioural and ecological studies. We created a repeated measures analysis that allows researchers to statistically examine similarities between distributions. We compared song structure in male and female eastern bluebirds, Sialia sialis, as a case study for this method. We failed to find significant differences between male and female songs via a more traditional test, repeated measures ANOVA. Therefore, no definitive conclusion could be drawn about the similarities or differences in song structure. However, our repeated measures equivalence test showed that, based on five standard measures of song variation, male and female eastern bluebirds sing statistically equivalent songs. Our study highlights the presence of complex female song in a temperate songbird species. Additionally, we provide a new statistical test useful for expanding the statistical toolbox to assess new behavioural and ecological questions, and to help counteract publication bias in our fields.

Friday, September 28, 2018 9:00 AM–10:00 AM best papers Friday: 153. Intravenous and oral tranexamic acid are equivalent at reducing blood loss in thoracolumbar spinal fusion: a prospective randomized trial

BACKGROUND CONTEXT Due to the significant risks and complications associated with blood loss and allogeneic transfusions, identifying safe and effective ways to minimize blood loss during spine surgery is crucial. Tranexamic acid (TXA) is effective at reducing blood loss and transfusion rate in orthopaedic surgery. Although intravenous (IV) TXA has been shown to be successful in patients undergoing spine fusion, no studies in the current literature have investigated the efficacy of its low cost counterpart, oral (PO) TXA. PURPOSE The purpose of this prospective randomized trial was to determine if PO TXA is equivalent to IV TXA in reducing blood loss in spinal fusion. STUDY DESIGN/SETTING This study was a single-center, randomized, prospective trial designed to test equivalence between IV and oral TXA in the setting of thoracolumbar fusion between February and December 2017. PATIENT SAMPLE Enrolled patients were randomly assigned between the two treatment groups of IV and oral TXA using a random number algorithm. A total of 44 patients received IV TXA and 45 patients received PO TXA. OUTCOME MEASURES The primary outcome was reduction of hemoglobin. In addition, calculated blood loss, drain output, postoperative transfusion, complications, and length of hospital stay were investigated. METHODS In this prospective randomized trial, 89 patients undergoing thoracolumbar fusion were randomized to receive 1.95g of PO TXA 2hours preoperatively or 2g IV TXA (1g before incision and 1g before wound closure) intraoperatively. Because the number of fusion levels can influence blood loss, the sample was sub-categorized by the number of vertebral levels arthrodesed (1–2 levels, 3–5 levels, >5 levels). Power analysis determined that 30 patients were needed in each group. Equivalence analysis was performed with a two one-sided test (TOST). A P-value of RESULTS Patient demographic factors were similar between groups except for BMI. The mean reduction of hemoglobin was similar between IV and PO groups (3.35g/dL vs. 3.18g/dL, respectively; P=.016, equivalence). Similarly, the calculated blood loss was equivalent (1233mL vs. 1227mL, respectively; P=.027, equivalence). Eight patients (18%) in IV TXA group received a transfusion compared to six patients in PO TXA group (13%) (P=.53). One patient (2% each) in each group experienced a DVT/PE (P=.99). CONCLUSIONS Intravenous and PO TXA are both effective in reducing the expensive perioperative costs associated with blood loss and transfusions after spinal fusions. Oral TXA provides equivalent reductions in blood loss at a greatly reduced cost compared to the IV formulation. As the incidence of spinal fusions increases in the near future, adopting an oral regimen of TXA can drastically improve cost-efficiency for our health care system.

Equivalence Testing for Psychological Research: A Tutorial

Psychologists must be able to test both for the presence of an effect and for the absence of an effect. In addition to testing against zero, researchers can use the two one-sided tests (TOST) procedure to test for equivalence and reject the presence of a smallest effect size of interest (SESOI). The TOST procedure can be used to determine if an observed effect is surprisingly small, given that a true effect at least as extreme as the SESOI exists. We explain a range of approaches to determine the SESOI in psychological science and provide detailed examples of how equivalence tests should be performed and reported. Equivalence tests are an important extension of the statistical tools psychologists currently use and enable researchers to falsify predictions about the presence, and declare the absence, of meaningful effects.