Do Common Risk Adjustment Methods Do Their Job Well if Center Effects are Correlated With the Center-Specific Mean Values of Patient Characteristics?

BACKGROUND: Direct and indirect standardization are well-established approaches to performing risk adjustment when comparing outcomes between healthcare providers. However, it is an open question whether they work well when there is an association between the center effects and the distributions of the patient characteristics in these centers. OBJECTIVES AND METHODS: We try to shed further light on the impact of such an association. We construct an artificial case study with a single covariate, in which centers can be classified as performing above, on, or below average, and the center effects correlate with center-specific mean values of a patient characteristic, as a consequence of differential quality improvement. Based on this case study, direct standardization and indirect standardization-based on marginal as well as conditional models-are compared with respect to systematic differences between their results. RESULTS: Systematic differences between the methods were observed. All methods produced results that partially reflect differences in mean age across the centers. This may mask the classification as above, on, or below average. The differences could be explained by an inspection of the parameter estimates in the models fitted. CONCLUSIONS: In case of correlations of center effects with center-specific mean values of a covariate, different risk adjustment methods can produce systematically differing results. This suggests the routine use of sensitivity analyses. Center effects in a conditional model need not reflect the position of a center above or below average, questioning its use in defining the truth. Further empirical investigations are necessary to judge the practical relevance of these findings.

Prospective Back Pain Trajectories or Retrospective Recall-Which Tells Us Most About the Patient?

In patients with low back pain (LBP), a visually identified retrospective pain trajectory often mismatches with a trajectory derived from prospective repeated measures. To gain insight into the clinical relevance of the 2 trajectory types, we investigated which showed a higher association with clinical outcomes. Participants were 724 adults seeking care for LBP in Danish chiropractic primary care. They answered weekly short-message-services on pain intensity and frequency over 52 weeks, which we translated into 8 trajectory classes. After 52 weeks, participants selected a retrospective visual pain trajectory from the same 8 trajectory classes. Clinical outcomes included disability, back/leg pain intensity, back beliefs, and work ability. The patient-selected pain trajectory classes were more strongly associated with clinical outcomes than the short-message-service trajectory classes at baseline, at follow-up, and with outcome changes between baseline and follow-up. This held across all 5 clinical outcomes, with the strongest associations observed at week 52 and the weakest at baseline. Patients' retrospective assessment of their LBP is more strongly associated with their clinical status than their prospective assessments translated into trajectory classes. This suggests that retrospective assessments of pain trajectories may provide valuable information not captured by prospective assessments. Researchers collecting prospective pain data should know that the captured pain trajectories are not strongly reflected in patients' perceptions of clinical status. Patients' retrospective assessments seem to offer an interpretation of their pain course that is likely more clinically relevant in understanding the perceived impact of their condition than trajectories based on repeated measures. PERSPECTIVE: Prospective pain data inadequately reflect patients' clinical status. Retrospective assessments provide a more clinically valuable understanding of the impact of their condition.

Anchor-based minimal important difference values are often sensitive to the distribution of the change score.

PURPOSE: Anchor-based studies are today the most popular approach to determine a minimal important difference value for an outcome variable. However, a variety of construction methods for such values do exist. This constitutes a challenge to the field. In order to distinguish between more or less adequate construction methods, meaningful minimal requirements can be helpful. For example, minimal important difference values should not reflect the intervention(s) the patients are exposed to in the study used for construction, as they should later allow to compare interventions. This requires that they are not sensitive to the distribution of the change score observed. This study aims at investigating to which degree established construction methods fulfil this minimal requirement. METHODS: Six constructions methods were considered, covering very popular and recently suggested methods. The sensitivity of MID values to the distribution of the change score was investigated in a simulation study for these six construction methods. RESULTS: Five out of six construction methods turned out to yield MID values which are sensitive to the distribution of the change score to a degree that questions their usefulness. Insensitivity can be obtained by using construction methods based solely on an estimate of the conditional distribution of the anchor variable given the change score. CONCLUSION: In future the computation of MID values should be based on construction methods avoiding sensitivity to the distribution of the change score.

Variability in human tooth cementum thickness reflecting functional processes.

OBJECTIVE: The aim of this study was to investigate the thickness of acellular extrinsic fibre cementum (AEFC) at four root positions of anterior and posterior teeth with special focus on functional aspects. Furthermore, the correlations between cementum thickness and chronological age and sex are investigated. BACKGROUND: While numerous studies confirm continuous cementum apposition with age, masticatory forces as well as physiological and orthodontically induced tooth movements also have the potential to affect tooth cementum thickness. MATERIALS AND METHODS: Undecalcified teeth were embedded in resin and transverse-sectioned in the cervical third of the root. Two sections per root were selected, and digital images at four positions were obtained (mesial, distal, oral, and vestibular) using light microscopy. The AEFC thickness of 99 teeth (anterior = 66, posterior = 33, male = 54, female = 45) were measured in both sections. The differences in mean values between root positions and the association of root position variation with tooth type, age, sex, and subject as well as the overall effects of age and sex were analysed using a mixed model. RESULTS: First incisors and canines showed the greatest mean AFEC thickness, in contrast to premolars which had the lowest values. Differences were found across the four root positions, with a pattern varying considerably between anterior and posterior teeth and between maxilla and mandible in the anterior teeth. An interaction between root position and subject pointed to the existence of an individual component in the variation of AEFC thickness across the four root positions. There was an age trend with an almost linear increase in cementum thickness of 1 µm per year. Overall, females tended to exhibit a significantly lesser AEFC thickness compared to males. CONCLUSIONS: Distinct differences in the pattern of thickness values across the four root positions in anterior and posterior teeth support the assumption that the AEFC is strongly affected by functional processes. In addition to sex-specific differences and age-related trends, the root position variation of AEFC thickness varies from individual to individual.

Exploring usual care for patients with low back pain in primary care: a cross-sectional study of general practitioners, physiotherapists and chiropractors.

OBJECTIVES: To explore the elements and composition of care provided by general practitioners (GPs), physiotherapists (PTs) and chiropractors (DCs) to patients with low back pain (LBP). DESIGN: Observational study. SETTING: Primary care setting, Denmark. PARTICIPANTS: Primary care clinicians (GPs, PTs and DCs) in the Region of Southern Denmark were invited to register consecutive adult patient visits with LBP as the primary complaint. PRIMARY OUTCOME MEASURES: Clinicians reported care elements provided to patients with LBP. Elements varied due to professional differences (eg, prescriptive rights). Data were descriptively analysed, on group and individual levels, for frequency and combination of care elements, and practice patterns were explored with latent class analysis. RESULTS: The clinicians (33 GPs, 67 PTs and 43 DCs with a median experience of 15 years and 59% were females) registered 3500 patient visits. On average, the visits involved patients aged 51 years, and 51% were with females. The frequencies of common care elements across professions were information (42%-56% of visits between professions) and advice (56%-81%), while other common elements for GPs were pain medication (40%) and referrals to PTs (36%), for PTs, use of exercises (81%) and for DCs, use of manual therapy (96%). Substantial variation was observed within professions and distinct practice patterns, with different focuses of attention to information and advice versus exercise and manual therapy, were identified for PTs and DCs. CONCLUSIONS: These data indicate substantial variation in the care elements provided by GPs, PTs and DCs to LBP patients. The compositions of care and practice patterns identified challenge the understanding of usual care as a uniform concept and professions as homogeneous groups. Strategic use of particular care elements in different parts of treatment courses is indicated. Longitudinal data and qualitative enquiry are needed to assess if or how care is tailored to individual patients.

Potential effect modifiers for treatment with chiropractic manipulation versus sham manipulation for recurrent headaches in children aged 7-14 years: development of and results from a secondary analysis of a randomised clinical trial.

BACKGROUND: A recent randomized controlled trial (RCT) investigating the effect of chiropractic manipulation in 199 children aged 7-14 years with recurrent headaches demonstrated a significant reduction of number of days with headache and a better global perceived effect (GPE) in the chiropractic manipulation group compared to a sham manipulation group. However, potential modifiers for the effectiveness of chiropractic manipulation of children with recurrent headaches have never been identified. The present study is a secondary analysis of data from that RCT and will investigate potential effect modifiers for the benefit of chiropractic manipulation for children with headache. METHODS: Sixteen potential effect modifiers were identified from the literature and a summary index was prespecified based on clinical experience. Relevant variables were extracted from baseline questionnaires, and outcomes were obtained by means of short text messages. The modifying effect of the candidate variables was assessed by fitting interaction models to the data of the RCT. In addition, an attempt to define a new summary index was made. RESULTS: The prespecified index showed no modifying effect. Four single variables demonstrated a treatment effect difference of more than 1 day with headache per week between the lower and the upper end of the spectrum: intensity of headache (p = 0.122), Frequency of headache (p = 0.031), sleep duration (p = 0.243), and Socioeconomic status (p = 0.082). Five variables had a treatment effect difference of more than 0.7 points on the GPE scale between the lower and the upper end of the spectrum: Frequency of headache (p = 0.056), Sport activity (p = 0.110), Sleep duration (p = 0.080), History of neck pain (p = 0.011), and Headache in the family (0.050). A new summary index could be constructed giving highest weight to History of neck pain and Headache in the family and Frequency of headache. The index suggests a difference of about 1 point in GPE between low and high values of the index. CONCLUSION: Chiropractic manipulation offers a moderate benefit for a broad spectrum of children. However, it cannot be excluded that specific headache characteristics, family factors, or a history of neck pain may modify the effect. This question must be addressed in future studies. TRIAL REGISTRATION: ClinicalTrials.gov (Albers et al in Curr Pain Headache Rep 19:3-4, 2015), identifier NCT02684916, registered 02/18/2016-retrospectively registered.

Gwet's AC1 is not a substitute for Cohen's kappa - A comparison of basic properties.

Gwet's AC1 has been proposed as an alternative to Cohen's kappa in evaluating the agreement between two binary ratings. This approach is becoming increasingly popular, and researchers have been criticized for still using Cohen's kappa. However, a rigorous discussion of properties of Gwet's AC1 is still missing. In this paper several basic properties of Gwet's AC1 are investigated and compared with those of Cohen's kappa, in particular the dependence on the prevalence of positive ratings for a given agreement rate and the behaviour in case of no association or maximal disagreement. Both approaches compare the observed agreement rate with a comparative number. Cohen's kappa uses an expected agreement rate as comparator, whereas Gwet's AC1 uses an expected disagreement rate. Consequently, for a fixed agreement rate, Gwet's AC1 increases with increasing difference of the prevalence of positive ratings from 0.5. In contrast, Cohen's kappa decreases. Gwet's AC1 can take positive and negative values in the case of no association between the two raters, whereas Cohen's kappa is 0. Due to these fundamental differences, Gwet's AC1 should not be seen as a substitute for Cohen's kappa. In particular, the verbal classification of kappa values by Landis & Koch should not be applied to Gwet's AC1.

Do Visual Pain Trajectories Reflect the Actual Course of Low Back Pain? A Longitudinal Cohort Study.

Different trajectories of low back pain (LBP) have been identified prospectively using repeated measures. For these trajectories to inform clinical practice, they must be available in the clinical consultation. Therefore, identified LBP trajectories have been translated into visual pain trajectories (VPTs) that allow people with LBP, at the time of consult, to reflect upon their pain experience and identify the VPT that best categorizes their pain course. We have limited knowledge regarding the extent to which a chosen VPT reflects the prospectively experienced trajectory. Thus, we explored the distribution of pain intensity and pain pattern characteristics (from prospective pain trajectory data) within the retrospectively chosen VPT classes. We enrolled patients with LBP from Danish chiropractic practice. Using SMS, participants (n = 719) scored their pain weekly on an 11-point numerical rating scale for 52 weeks. At week 52, participants identified 1 of 8 VPTs that reflected their perceived back pain trajectory during the preceding year. We found that the chosen VPTs reflected pain intensity, but that pain patterns (episodic, fluctuating, and persistent) were not systematically recognized, and the experienced course varied substantially amongst participants within the same VPT. The VPTs are related to some aspects of the experienced LBP course but are not a proxy for the SMS-measured trajectories. Reasons for apparent mismatches between the experienced course of LBP and VPT recall warrant further investigation. PERSPECTIVE: Self-reported back pain trajectories reflected pain intensities obtained through weekly SMS tracking over a year, but participants' recall did not reflect the pain patterns (episodes and fluctuations) discovered prospectively. Clinicians can use self-reported pain trajectories to facilitate a dialog about pain experience, but not as a proxy for prospective measures.

Effectiveness assessment of non-pharmaceutical interventions: lessons learned from the COVID-19 pandemic.

Effectiveness of non-pharmaceutical interventions (NPIs), such as school closures and stay-at-home orders, during the COVID-19 pandemic has been assessed in many studies. Such assessments can inform public health policies and contribute to evidence-based choices of NPIs during subsequent waves or future epidemics. However, methodological issues and no standardised assessment practices have restricted the practical value of the existing evidence. Here, we present and discuss lessons learned from the COVID-19 pandemic and make recommendations for standardising and improving assessment, data collection, and modelling. These recommendations could contribute to reliable and policy-relevant assessments of the effectiveness of NPIs during future epidemics.

Understanding the Influence of Patient Factors on Accuracy and Decision-Making in a Diagnostic Accuracy Study with Multiple Raters-A Case Study from Dentistry.

In diagnostic accuracy studies, the test of interest is typically applied only once in each patient. This paper illustrates some possibilities that arise when diagnoses are carried out by a sufficiently large number of multiple raters. In a dental study, sixty-one examiners were asked to diagnose 49 tooth areas with different grades of tissue loss (minor, moderate, and advanced) to decide whether dentine was exposed (positive status) or not (negative status). The true status was determined by histology (reference). For each tooth, the rate of correct decisions reflecting the difficulty to diagnose this tooth and the positive rate reflecting the perception of the tooth by the raters was computed. Meta-analytical techniques were used to assess the inter-tooth variation and the influence of tooth-specific factors on difficulty or perception, respectively. A huge variation in diagnostic difficulty and perception could be observed. Advanced tissue loss made diagnoses more difficult. The background colour and tissue loss were associated with perception and may hint to cues used by the raters. The use of multiple raters in a diagnostic accuracy study allows detailed investigations which make it possible to obtain further insights into the decision-making process of the raters.

Description of recurrent headaches in 7-14-year-old children: Baseline data from a randomized clinical trial on effectiveness of chiropractic spinal manipulation in children with recurrent headaches.

BACKGROUND: Headaches in children are poorly described and diagnosing can be challenging. Objectives are: (1) to describe headache characteristics and child characteristics, (2) to explore whether data can suggest a more diverse way to categorize headaches than traditionally. METHODS: Baseline data for a clinical trial included a questionnaire and a physical screening. Children's characteristics and detailed description of headache symptoms were provided. Children were classified for migraine or tension-type-headache based on questionnaire data reported by children and parents. This required to apply slightly modified classification criteria and a "non-classifiable" group was added. Severity and symptoms, related to the migraine versus tension type distinction, were investigated to define a migraine-tension-type-index. RESULTS: 253 children were included. Mean pain intensity was 5.9/10. Over 2/3 of the children had headache for > 1 year, and > 50% for several days/week. Half of the children were non-classifiable, 22% were classified as migraine and 23% as tension-type headache. A migraine-tension-type-index was constructed and describes a continuous spectrum rather than two distinct groups. CONCLUSIONS: Children with recurrent headaches are often severely affected. A questionnaire-based classification appeared feasible to distinguish between migraine and tension-type headaches in children but leaving many children unclassified. A migraine-tension-type-index can be generated allowing to regard the traditional distinction as a continuum (including mixed headache), and potentially serving as an instrument to improve headache management. Trial registration ClinicalTrials.gov, identifier NCT02684916.

Response Monitoring in Metastatic Breast Cancer: A Prospective Study Comparing 18F-FDG PET/CT with Conventional CT.

This study aimed to compare contrast-enhanced CT (CE-CT) and 18F-FDG PET/CT for response monitoring in metastatic breast cancer using the standardized response evaluation criteria RECIST 1.1 and PERCIST. The objective was to examine whether progressive disease was detected systematically earlier by one of the modalities. Methods: Women with biopsy-verified metastatic breast cancer were enrolled prospectively and monitored using combined CE-CT and 18F-FDG PET/CT every 9-12 wk to evaluate response to first-line treatment. CE-CT scans and RECIST 1.1 were used for clinical decision-making without accessing the 18F-FDG PET/CT scans. At study completion, 18F-FDG PET/CT scans were unmasked and assessed according to PERCIST. Visual assessment was used if response criteria could not be applied. The modality-specific time to progression was defined as the time from the baseline scan until the first scan demonstrating progression. Paired comparative analyses for CE-CT versus 18F-FDG PET/CT were applied, and the primary endpoint was earlier detection of progression by one modality. Secondary endpoints were time to detection of progression, response categorization, visualization of changes in response over time, and measurable disease according to RECIST and PERCIST. Results: In total, 87 women were evaluable, with a median of 6 (1-11) follow-up scans. Progression was detected first by 18F-FDG PET/CT in 43 (49.4%) of 87 patients and first by CE-CT in 1 (1.15%) of 87 patients (P < 0.0001). Excluding patients without progression (n = 32), progression was seen first on 18F-FDG PET/CT in 78.2% (43/55) of patients. The median time from detection of progression by 18F-FDG PET/CT to that of CE-CT was 6 mo (95% CI, 4.3-6.4 mo). At baseline, 76 (87.4%) of 87 patients had measurable disease according to PERCIST and 51 (58.6%) of 87 patients had measurable disease according to RECIST 1.1. Moreover, 18F-FDG PET/CT provided improved visualization of changes in response over time, as seen in the graphical abstract. Conclusion: Disease progression was detected earlier by 18F-FDG PET/CT than by CE-CT in most patients, with a potentially clinically relevant median 6-mo delay for CE-CT. More patients had measurable disease according to PERCIST than according to RECIST 1.1. The magnitude of the final benefit for patients is a perspective for future research.

Comparing a Sensor for Movement Assessment with Traditional Physiotherapeutic Assessment Methods in Patients after Knee Surgery-A Method Comparison and Reproducibility Study.

Wearable sensors offer the opportunity for patients to perform a self-assessment of their function with respect to a variety of movement exercises. Corresponding commercial products have the potential to change the communication between patients and physiotherapists during the recovery process. Even if they turn out to be user-friendly, there remains the question to what degree the numerical results are reliable and comparable with those obtained by assessment methods traditionally used. To address this question for one specific recently developed and commercially available sensor, a method comparison study was performed. The sensor-based assessment of eight movement parameters was compared with an assessment of the same parameters based on test procedures traditionally used. Thirty-three patients recovering after arthroscopic knee surgery participated in the study. The whole assessment procedure was repeated. Reproducibility and agreement were quantified by the intra class correlation coefficient. The height of a one-leg vertical jump and the number of side hops showed high agreement between the two modalities and high reproducibility (ICC > 0.85). Due to differences in the set-up of the assessment, agreement could not be achieved for three mobility parameters, but even the correlation was only fair (r < 0.5). Knee stability showed poor agreement. Consequently, the use of the sensor can currently only be recommended for selected parameters. The variation in degree of agreement and reproducibility across different parameters clearly indicate the need for developing corresponding guidance for each new sensor put onto the market.

The methodologies to assess the effectiveness of non-pharmaceutical interventions during COVID-19: a systematic review.

Non-pharmaceutical interventions, such as school closures and stay-at-home orders, have been implemented around the world to control the spread of SARS-CoV-2. Their effectiveness in improving health-related outcomes has been the subject of numerous empirical studies. However, these studies show fairly large variation among methodologies in use, reflecting the absence of an established methodological framework. On the one hand, variation in methodologies may be desirable to assess the robustness of results; on the other hand, a lack of common standards can impede comparability among studies. To establish a comprehensive overview over the methodologies in use, we conducted a systematic review of studies assessing the effectiveness of non-pharmaceutical interventions between January 1, 2020 and January 12, 2021 (n = 248). We identified substantial variation in methodologies with respect to study setting, outcome, intervention, methodological approach, and effectiveness assessment. On this basis, we point to shortcomings of existing studies and make recommendations for the design of future studies.

The concordance of signals based on irregular incremental lines in the human tooth cementum with documented pregnancies: Results from a systematic approach.

BACKGROUND AND OBJECTIVE: There is evidence from previous studies that pregnancies and diseases are recorded in the tooth cementum. This study aims to assess the degree of concordance between signals based on irregular incremental lines (ILs) and reported pregnancies. MATERIAL AND METHODS: 23 recent and 24 archaeological human teeth with known birth history were included in this investigation. 129 histological sections of tooth roots were assessed for irregularities in appearance and width using a standardized protocol. Similarity of observed irregularities at the section level allowed us to define signals at the tooth level. The sensitivity of signals to detect pregnancies was determined and related to the signal prevalence. RESULTS: Pregnancy signals were frequently visually observed. However, applying a standardized process we could only reach signal sensitivities to identify pregnancies up to 20 percentage points above chance level. CONCLUSIONS: Based on a standardized and reproducible method it could be confirmed that some pregnancies leave visible signals in the tooth cementum. The results show the potential of the tooth cementum to support reconstruction of life courses in paleopathology. However, it seems that not all pregnancies affect the cementogenesis in such a way that irregular ILs are identifiable. Further research is needed to better understand which type of pregnancies and other conditions are recorded in the tooth cementum.

Dose-response relationship of in vivo ambulatory load and mechanosensitive cartilage biomarkers-The role of age, tissue health and inflammation: A study protocol.

OBJECTIVE: To describe a study protocol for investigating the in vivo dose-response relationship between ambulatory load magnitude and mechanosensitive blood markers of articular cartilage, the influence of age, cartilage tissue health and presence of inflammation on this relationship, and its ability to predict changes in articular cartilage quality and morphology within 2 years. DESIGN: Prospective experimental multimodal (clinical, biomechanical, biological) data collection under walking stress and three different load conditions varied in a randomized crossover design. EXPERIMENTAL PROTOCOL: At baseline, equal numbers of healthy and anterior cruciate ligament injured participants aged 20-30 or 40-60 years will be assessed clinically and complete questionnaires regarding their knee health. Biomechanical parameters (joint kinetics, joint kinematics, and surface electromyography) will be recorded while performing different tasks including overground and treadmill walking, single leg balance and hopping tasks. Magnetic resonance images (MRI) of both of knees will be obtained. On separate stress test days, participants will perform a 30-minute walking stress with either reduced (80% body weight (BW)), normal (100%BW) or increased (120%BW) load. Serum blood samples will be taken immediately before, immediately after, 30, 120 and 210 minutes after the walking stress. Concentration of articular cartilage blood biomarkers will be assessed using enzyme linked immunosorbent assays. At 24-month follow-up, participants will be again assessed clinically, undergo an MRI, complete questionnaires, and have a blood sample taken. CONCLUSION: The study design provides a standardized set up that allows to better understand the influence of ambulatory load on articular cartilage biomarkers and thereby extend current knowledge on in vivo cartilage metabolism and mechanosensitivity. Further, this study will help to elucidate the prognostic value of the load-induced cartilage biomarker response for early articular cartilage degeneration. TRIAL REGISTRATION: The protocol was approved by the regional ethics committee and has been registered at clinicaltrials.gov (NCT04128566).

The impact of specific fracture characteristics of low-energy fractures of the pelvis on mortality.

BACKGROUND: Pelvic fractures in older patients are associated with relevant morbidity and mortality. Both might be determined by fracture morphology and/or patient characteristics. The aim of this project is to investigate the prognostic value of specific fracture characteristics with respect to overall survival and to compare it with an established classification system. METHODS: Retrospective analysis of patients ≥ 60 years, treated conservatively for a CT-scan verified, low-energy pelvic ring fracture between August 2006 and December 2018. Survival data was available from patients' charts and cantonal or national registries. The prognostic value of fracture characteristic describing the anterior and posterior involvement of the pelvic ring was investigated. This analysis was repeated after patients were stratified into a high-risk vs a low-risk group according to patient characteristic (age, gender, comorbidities, mobility, living situation). This allowed to assess the impact of the different fracture morphologies on mortality in fit vs. frail senior patients separately. RESULTS: Overall, 428 patients (83.4% female) with a mean age of 83.7 years were included. Two thirds of patients were still living in their home and mobile without walking aid at baseline. In-hospital mortality was 0.7%, overall, one-year mortality 16.9%. An independent and significant association of age, gender and comorbidities to overall survival was found. Further, the occurrence of a horizontal sacral fracture as well as a ventral comminution or dislocation was associated with an increased mortality. The effect of a horizontal sacral fracture was more accentuated in low-risk patients while the ventral fracture components showed a larger effect on survival in high-risk patients. CONCLUSION: Specific fracture characteristics may indicate a higher risk of mortality in conservatively treated patients with a low-energy pelvic ring fracture. Hence, they should be taken into account in future treatment algorithms and decisions on patient management.

Sample size recalculation based on the prevalence in a randomized test-treatment study.

BACKGROUND: Randomized test-treatment studies aim to evaluate the clinical utility of diagnostic tests by providing evidence on their impact on patient health. However, the sample size calculation is affected by several factors involved in the test-treatment pathway, including the prevalence of the disease. Sample size planning is exposed to strong uncertainties in terms of the necessary assumptions, which have to be compensated for accordingly by adjusting prospectively determined study parameters during the course of the study. METHOD: An adaptive design with a blinded sample size recalculation in a randomized test-treatment study based on the prevalence is proposed and evaluated by a simulation study. The results of the adaptive design are compared to those of the fixed design. RESULTS: The adaptive design achieves the desired theoretical power, under the assumption that all other nuisance parameters have been specified correctly, while wrong assumptions regarding the prevalence may lead to an over- or underpowered study in the fixed design. The empirical type I error rate is sufficiently controlled in the adaptive design as well as in the fixed design. CONCLUSION: The consideration of a blinded recalculation of the sample size already during the planning of the study may be advisable in order to increase the possibility of success as well as an enhanced process of the study. However, the application of the method is subject to a number of limitations associated with the study design in terms of feasibility, sample sizes needed to be achieved, and fulfillment of necessary prerequisites.