The built-in selection bias of hazard ratios formalized using structural causal models.

It is known that the hazard ratio lacks a useful causal interpretation. Even for data from a randomized controlled trial, the hazard ratio suffers from so-called built-in selection bias as, over time, the individuals at risk among the exposed and unexposed are no longer exchangeable. In this paper, we formalize how the expectation of the observed hazard ratio evolves and deviates from the causal effect of interest in the presence of heterogeneity of the hazard rate of unexposed individuals (frailty) and heterogeneity in effect (individual modification). For the case of effect heterogeneity, we define the causal hazard ratio. We show that the expected observed hazard ratio equals the ratio of expectations of the latent variables (frailty and modifier) conditionally on survival in the world with and without exposure, respectively. Examples with gamma, inverse Gaussian and compound Poisson distributed frailty and categorical (harming, beneficial or neutral) distributed effect modifiers are presented for illustration. This set of examples shows that an observed hazard ratio with a particular value can arise for all values of the causal hazard ratio. Therefore, the hazard ratio cannot be used as a measure of the causal effect without making untestable assumptions, stressing the importance of using more appropriate estimands, such as contrasts of the survival probabilities.

Bias of the additive hazard model in the presence of causal effect heterogeneity.

Hazard ratios are prone to selection bias, compromising their use as causal estimands. On the other hand, if Aalen's additive hazard model applies, the hazard difference has been shown to remain unaffected by the selection of frailty factors over time. Then, in the absence of confounding, observed hazard differences are equal in expectation to the causal hazard differences. However, in the presence of effect (on the hazard) heterogeneity, the observed hazard difference is also affected by selection of survivors. In this work, we formalize how the observed hazard difference (from a randomized controlled trial) evolves by selecting favourable levels of effect modifiers in the exposed group and thus deviates from the causal effect of interest. Such selection may result in a non-linear integrated hazard difference curve even when the individual causal effects are time-invariant. Therefore, a homogeneous time-varying causal additive effect on the hazard cannot be distinguished from a time-invariant but heterogeneous causal effect. We illustrate this causal issue by studying the effect of chemotherapy on the survival time of patients suffering from carcinoma of the oropharynx using data from a clinical trial. The hazard difference can thus not be used as an appropriate measure of the causal effect without making untestable assumptions.

Comparison of analysis methods and design choices for treatment-by-period interaction in unidirectional switch designs: a simulation study.

BACKGROUND: Due to identifiability problems, statistical inference about treatment-by-period interactions has not been discussed for stepped wedge designs in the literature thus far. Unidirectional switch designs (USDs) generalize the stepped wedge designs and allow for estimation and testing of treatment-by-period interaction in its many flexible design forms. METHODS: Under different forms of the USDs, we simulated binary data at both aggregated and individual levels and studied the performances of the generalized linear mixed model (GLMM) and the marginal model with generalized estimation equations (GEE) for estimating and testing treatment-by-period interactions. RESULTS: The parallel group design had the highest power for detecting the treatment-by-period interactions. While there was no substantial difference between aggregated-level and individual-level analysis, the GLMM had better point estimates than the marginal model with GEE. Furthermore, the optimal USD for estimating the average treatment effect was not efficient for treatment-by-period interaction and the marginal model with GEE required a substantial number of clusters to yield unbiased estimates of the interaction parameters when the correlation structure is autoregressive of order 1 (AR1). On the other hand, marginal model with GEE had better coverages than GLMM under the AR1 correlation structure. CONCLUSION: From the designs and methods evaluated, in general, parallel group design with a GLMM is, preferred for estimation and testing of treatment-by-period interaction in a clustered randomized controlled trial for a binary outcome.

Non-inferiority testing for qualitative microbiological methods: Assessing and improving the approach in USP 1223.

The United States Pharmacopoeia (USP) presents two approaches for showing non-inferiority of an alternate qualitative microbiological method versus a compendial method. One approach compares the positive rates for the alternate and compendial methods at one spike level, while the other one compares multiple most probable number (MPN) estimates from a multi-spike design using a t-test. In this paper, we discuss these approaches under certain assumptions and propose a third approach that can be used for both single and multiple dilutions, which we call the generalized MPN (gMPN) approach. Simulations, using Poisson distributed numbers of microorganisms in test samples, confirm that the USP approach based on rates is not suitable, that the USP approach based on MPNs is appropriate for non-inferiority, but the gMPN approach outperforms the MPN-based approach and is therefore recommended.

Association between markers of immune response at hospital admission and COVID-19 disease severity and mortality: A meta-analysis and meta-regression.

BACKGROUND: To determine the utility of admission laboratory markers in the assessment and prognostication of coronavirus disease-2019 (COVID-19), a systematic review and meta-analysis were conducted on the association between admission laboratory values in hospitalized COVID-19 patients and subsequent disease severity and mortality. MATERIAL AND METHODS: Searches were conducted in MEDLINE, Pubmed, Embase, and the WHO Global Research Database from December 1,2019 to May 1, 2020 for relevant articles. A random effects meta-analysis was used to calculate the weighted mean difference (WMD) and 95% confidence interval (95% CI) for each of 27 laboratory markers. The impact of age and sex on WMDs was estimated using meta-regression techniques for 11 markers. RESULTS: In total, 64 studies met the inclusion criteria. The most marked WMDs were for neutrophils (ANC) at 3.82 × 109 /L (2.76, 4.87), lymphocytes (ALC) at -0.34 × 109 /L (-0.45, -0.23), interleukin-6 (IL-6) at 32.59 pg/mL (23.99, 41.19), ferritin at 814.14 ng/mL (551.48, 1076.81), C-reactive protein (CRP) at 66.11 mg/L (52.16, 80.06), D-dimer at 5.74 mg/L (3.91, 7.58), LDH at 232.41 U/L (178.31, 286.52), and high sensitivity troponin I at 90.47 pg/mL (47.79, 133.14) when comparing fatal to nonfatal cases. Similar trends were observed comparing severe to non-severe groups. There were no statistically significant associations between age or sex and WMD for any of the markers included in the meta-regression. CONCLUSION: The results highlight that hyper inflammation, blunted adaptive immune response, and intravascular coagulation play key roles in the pathogenesis of COVID-19. Markers of these processes are good candidates to identify patients for early intervention and, importantly, are likely reliable regardless of age or sex in adult patients.

A latent-class heteroskedastic hurdle trajectory model: patterns of adherence in obstructive sleep apnea patients on CPAP therapy.

BACKGROUND: Sleep apnea patients on CPAP therapy exhibit differences in how they adhere to the therapy. Previous studies have demonstrated the benefit of describing adherence in terms of discernible longitudinal patterns. However, these analyses have been done on a limited number of patients, and did not properly represent the temporal characteristics and heterogeneity of adherence. METHODS: We illustrate the potential of identifying patterns of adherence with a latent-class heteroskedastic hurdle trajectory approach using generalized additive modeling. The model represents the adherence trajectories on three aspects over time: the daily hurdle of using the therapy, the daily time spent on therapy, and the day-to-day variability. The combination of these three characteristics has not been studied before. RESULTS: Applying the proposed model to a dataset of 10,000 patients in their first three months of therapy resulted in nine adherence groups, among which 49% of patients exhibited a change in adherence over time. The identified group trajectories revealed a non-linear association between the change in the daily hurdle of using the therapy, and the average time on therapy. The largest difference between groups was observed in the patient motivation score. The adherence patterns were also associated with different levels of high residual AHI, and day-to-day variability in leakage. CONCLUSION: The inclusion of the hurdle model and the heteroskedastic model into the mixture model enabled the discovery of additional adherence patterns, and a more descriptive representation of patient behavior over time. Therapy adherence was mostly affected by a lack of attempts over time, suggesting that encouraging these patients to attempt therapy on a daily basis, irrespective of the number of hours used, could drive adherence. We believe the methodology is applicable to other domains of therapy or medication adherence.

How did governmental interventions affect the spread of COVID-19 in European countries?

BACKGROUND: To reduce the transmission of the severe acute respiratory syndrome coronavirus 2 in its first wave, European governments have implemented successive measures to encourage social distancing. However, it remained unclear how effectively measures reduced the spread of the virus. We examined how the effective-contact rate (ECR), the mean number of daily contacts for an infectious individual to transmit the virus, among European citizens evolved during this wave over the period with implemented measures, disregarding a priori information on governmental measures. METHODS: We developed a data-oriented approach that is based on an extended Susceptible-Exposed-Infectious-Removed (SEIR) model. Using the available data on the confirmed numbers of infections and hospitalizations, we first estimated the daily total number of infectious-, exposed- and susceptible individuals and subsequently estimated the ECR with an iterative Poisson regression model. We then compared change points in the daily ECRs to the moments of the governmental measures. RESULTS: The change points in the daily ECRs were found to align with the implementation of governmental interventions. At the end of the considered time-window, we found similar ECRs for Italy (0.29), Spain (0.24), and Germany (0.27), while the ECR in the Netherlands (0.34), Belgium (0.35) and the UK (0.37) were somewhat higher. The highest ECR was found for Sweden (0.45). CONCLUSIONS: There seemed to be an immediate effect of banning events and closing schools, typically among the first measures taken by the governments. The effect of additionally closing bars and restaurants seemed limited. For most countries a somewhat delayed effect of the full lockdown was observed, and the ECR after a full lockdown was not necessarily lower than an ECR after (only) a gathering ban.

Evaluation of 10 years of parainfluenza virus, human metapneumovirus, and respiratory syncytial virus infections in lung transplant recipients.

Respiratory tract infection with pneumoviruses (PVs) and paramyxoviruses (PMVs) are increasingly associated with chronic lung allograft dysfunction (CLAD) in lung transplant recipients (LTRs). Ribavirin may be a treatment option but its effectiveness is unclear, especially with respect to infection severity. We retrospectively analyzed 10 years of PV/PMV infections in LTRs. The main end points were forced expiratory volume in 1 second (FEV1 ) at 3 and 6 months postinfection, expressed as a percentage of pre-infection FEV1 and incidence of new or progressed CLAD 6 months postinfection. A total of 139 infections were included: 88 severe infections (63%) (defined as >10% FEV1 loss at infection) and 51 mild infections (37%) (≤10% FEV1 loss). Overall postinfection CLAD incidence was 20%. Associations were estimated on postinfection FEV1 for ribavirin vs no ribavirin (+13.2% [95% CI: 7.79; 18.67]) and severe vs mild infection (-11.1% [95% CI: -14.76; -7.37]). Factors associated with CLAD incidence at 6 months were ribavirin treatment (odds ratio (OR [95% CI]) 0.24 [0.10; 0.59]), severe infection (OR [95% CI] 4.63 [1.66; 12.88]), and mycophenolate mofetil use (OR [95% CI] 0.38 [0.14; 0.97]). These data provide valuable information about the outcomes of lung transplant recipients with these infections and suggests possible associations of ribavirin use and infection severity with long-term outcomes. Well-designed prospective trials are needed to confirm these findings.

Calcitonin testing for detection of medullary thyroid cancer in people with thyroid nodules.

BACKGROUND: Thyroid nodules are very common in general medical practice, but rarely turn out to be a medullary thyroid carcinoma (MTC). Calcitonin is a sensitive tumour marker for the detection of MTC (basal calcitonin). Sometimes a stimulation test is used to improve specificity (stimulated calcitonin). Although the European Thyroid Association's guideline advocates calcitonin determination in people with thyroid nodules, the role of routine calcitonin testing in individuals with thyroid nodules is still questionable. OBJECTIVES: The objective of this review was to determine the diagnostic accuracy of basal and/or stimulated calcitonin as a triage or add-on test for detection of MTC in people with thyroid nodules. SEARCH METHODS: We searched CENTRAL, MEDLINE, Embase and Web of Science from inception to June 2018. SELECTION CRITERIA: We included all retrospective and prospective cohort studies in which all participants with thyroid nodules had undergone determination of basal calcitonin levels (and stimulated calcitonin, if performed). DATA COLLECTION AND ANALYSIS: Two review authors independently scanned all retrieved records. We extracted data using a standard data extraction form. We assessed risk of bias and applicability using the QUADAS-2 tool. Using the hierarchical summary receiver operating characteristic (HSROC) model, we estimated summary curves across different thresholds and also obtained summary estimates of sensitivity and specificity at a common threshold when possible. MAIN RESULTS: In 16 studies, we identified 72,368 participants with nodular thyroid disease in whom routinely calcitonin testing was performed. All included studies performed the calcitonin test as a triage test. Median prevalence of MTC was 0.32%. Sensitivity in these studies ranged between 83% and 100% and specificity ranged between 94% and 100%. An important limitation in 15 of the 16 studies (94%) was the absence of adequate reference standards and follow-up in calcitonin-negative participants. This resulted in a high risk of bias with regard to flow and timing in the methodological quality assessment. At the median specificity of 96.6% from the included studies, the estimated sensitivity (95% confidence interval (CI)) from the summary curve was 99.7% ( 68.8% to 100%). For the median prevalence of MTC of 0.23%, the positive predictive value (PPV) for basal calcitonin testing at a threshold of 10 pg/mL was 7.7% (4.9% to 12.1%). Summary estimates of sensitivity and specificity for the threshold of 10 pg/mL of basal calcitonin testing was 100% (95% CI 99.7 to 100) and 97.2% (95% CI 95.9 to 98.6), respectively. For combined basal and stimulated calcitonin testing, sensitivity ranged between 82% and 100% with specificity between 99% and 100%. The median specificity was 99.8% with an estimated sensitivity of 98.8% (95% CI 65.8 to 100) . AUTHORS' CONCLUSIONS: Both basal and combined basal and stimulated calcitonin testing have a high sensitivity and specificity. However, this may be an overestimation due to high risk of bias in the use and choice of reference standard The value of routine testing in patients with thyroid nodules remains questionable, due to the low prevalence, which results in a low PPV of basal calcitonin testing. Whether routine calcitonin testing improves prognosis in MTC patients remains unclear.

Does Intraoperative Cell Salvage Reduce Postoperative Infection Rates in Cardiac Surgery?

OBJECTIVE: Primary outcome was the risk for infections after cell salvage in cardiac surgery. DESIGN: Data of a randomized controlled trial on cell salvage and filter use (ISRCTN58333401). SETTING: Six cardiac surgery centers in the Netherlands. PARTICIPANTS: All 716 patients undergoing elective coronary artery bypass grafting, valve surgery, or combined procedures over a 4-year period who completed the trial. INTERVENTIONS: Postoperative infection data were assessed according to Centre of Disease Control and Prevention/National Healthcare Safety Network surveillance definitions. MEASUREMENTS AND MAIN RESULTS: Fifty-eight (15.9%) patients with cell salvage had infections, compared with 46 (13.1%) control patients. Mediation analysis was performed to estimate the direct effect of cell salvage on infections (OR 2.291 [1.177;4.460], p = 0.015) and the indirect effects of allogeneic transfusion and processed cell salvage blood on infections. Correction for confounders, including age, seks and body mass index was performed. Allogeneic transfusion had a direct effect on infections (OR = 2.082 [1.133;3.828], p = 0.018), but processed cell salvage blood did not (OR = 0.999 [0.999; 1.001], p = 0.089). There was a positive direct effect of cell salvage on allogeneic transfusion (OR = 0.275 [0.176;0.432], p < 0.001), but a negative direct effect of processed cell salvage blood (1.001 [1.001;1.002], p < 0.001) on allogeneic transfusion. Finally, there was a positive direct effect of cell salvage on the amount of processed blood. CONCLUSIONS: Cell salvage was directly associated with higher infection rates, but this direct effect was almost completely eliminated by its indirect protective effect through reduced allogeneic blood transfusion.

Latent variable models for harmonization of test scores: A case study on memory.

Combining data from different studies has a long tradition within the scientific community. It requires that the same information is collected from each study to be able to pool individual data. When studies have implemented different methods or used different instruments (e.g., questionnaires) for measuring the same characteristics or constructs, the observed variables need to be harmonized in some way to obtain equivalent content information across studies. This paper formulates the main concepts for harmonizing test scores from different observational studies in terms of latent variable models. The concepts are formulated in terms of calibration, invariance, and exchangeability. Although similar ideas are present in measurement reliability and test equating, harmonization is different from measurement invariance and generalizes test equating. In addition, if a test score needs to be transformed to another test score, harmonization of variables is only possible under specific conditions. Observed test scores that connect all of the different studies, are necessary to be able to test the underlying assumptions of harmonization. The concepts of harmonization are illustrated on multiple memory test scores from three different Canadian studies.

Gender Differences in Developing Biomarker-Based Major Depressive Disorder Diagnostics.

The identification of biomarkers associated with major depressive disorder (MDD) holds great promise to develop an objective laboratory test. However, current biomarkers lack discriminative power due to the complex biological background, and not much is known about the influence of potential modifiers such as gender. We first performed a cross-sectional study on the discriminative power of biomarkers for MDD by investigating gender differences in biomarker levels. Out of 28 biomarkers, 21 biomarkers were significantly different between genders. Second, a novel statistical approach was applied to investigate the effect of gender on MDD disease classification using a panel of biomarkers. Eleven biomarkers were identified in men and eight in women, three of which were active in both genders. Gender stratification caused a (non-significant) increase of Area Under Curve (AUC) for men (AUC = 0.806) and women (AUC = 0.807) compared to non-stratification (AUC = 0.739). In conclusion, we have shown that there are differences in biomarker levels between men and women which may impact accurate disease classification of MDD when gender is not taken into account.

Additional filtering of blood from a cell salvage device is not likely to show important additional benefits in outcome in cardiac surgery.

BACKGROUND: Several authors and manufacturers of cell salvage devices recommend additional filtering of processed blood before transfusion. There is no evidence to support this practice. Therefore, we compared the clinical outcome and biochemical effects of cell salvage with or without additional filtering. STUDY DESIGN AND METHODS: The patients, scheduled for coronary artery bypass grafting, valve replacement, or combined procedures were part of our randomized multicenter factorial study of cell salvage and filter use on transfusion requirements (ISRCTN 58333401). They were randomized to intraoperative cell salvage or cell salvage plus additional WBC depletion filter. We compared the occurrence of major adverse events (combined death/stroke/myocardial infarction) as primary outcome and minor adverse events (renal function disturbances, infections, delirium), ventilation time, and length of stay in the intensive care unit and hospital. We also measured biochemical markers of organ injury and inflammation. RESULTS: One hundred eighty-nine patients had cell salvage, and 175 patients had cell salvage plus filter and completed the study. Demographic data, surgical procedures, and amount of salvaged blood were not different between the groups. There was no difference in the primary outcome with a risk of 6.3% (95% confidence interval [CI], 3.34-11.25) in the cell salvage plus filter group versus 5.8% (95% CI, 3.09-10.45) in the cell salvage group, a relative risk of 1.08 (95% CI, 0.48- 2.43]. There were no differences in minor adverse events and biochemical markers between the groups. CONCLUSION: The routine use of an additional filter for transfusion of salvaged blood is unlikely to show important additional benefits.

Variance Components Models for Analysis of Big Family Data of Health Outcomes in the Lifelines Cohort Study.

Large multigenerational cohort studies offer powerful ways to study the hereditary effects on various health outcomes. However, accounting for complex kinship relations in big data structures can be methodologically challenging. The traditional kinship model is computationally infeasible when considering thousands of individuals. In this article, we propose a computationally efficient alternative that employs fractional relatedness of family members through a series of founding members. The primary goal of this study is to investigate whether the effect of determinants on health outcome variables differs with and without accounting for family structure. We compare a fixed-effects model without familial effects with several variance components models that account for heritability and shared environment structure. Our secondary goal is to apply the fractional relatedness model in a realistic setting. Lifelines is a three-generation cohort study investigating the biological, behavioral, and environmental determinants of healthy aging. We analyzed a sample of 89,353 participants from 32,452 reconstructed families. Our primary conclusion is that the effect of determinants on health outcome variables does not differ with and without accounting for family structure. However, accounting for family structure through fractional relatedness allows for estimating heritability in a computationally efficient way, showing some interesting differences between physical and mental quality of life heritability. We have shown through simulations that the proposed fractional relatedness model performs better than the standard kinship model, not only in terms of computational time and convenience of fitting using standard functions in R, but also in terms of bias of heritability estimates and coverage.

A comparison of spiking experiments to estimate the detection proportion of qualitative microbiological methods.

The detection proportion of a qualitative microbiological test method is the probability to detect a single micro-organism. A general expression for the moment estimator of the detection proportion is provided. It depends on the distribution of the spikes used in a validation study through its moment-generating function. Several forms of spiking experiments are compared on their estimation performance using simulations and assuming a generalized Poisson distribution (GPD) for the spikes. The optimal design, which minimizes the mean squared error of our proposed moment estimator, depends on the dispersion parameter of the GPD. The design that uses just one spiked solution instead of multiple solutions is optimal for Poisson and overdispersed Poisson and it is robust against distributions for the spikes.

Posaconazole therapeutic drug monitoring in clinical practice and longitudinal analysis of the effect of routine laboratory measurements on posaconazole concentrations.

Posaconazole is indicated for prophylaxis and treatment of invasive aspergillosis. Therapeutic drug monitoring (TDM) of posaconazole is used to optimise drug exposure. The aim of this study was to analyse and describe the TDM practices and exposure of posaconazole tablets. Patients who received posaconazole for treatment or prophylaxis of fungal infections were included in the study. The following therapeutic window was defined: if concentration was low (<0.7 mg/L for prophylaxis or < 1.5 mg/L for treatment) or high (>3.75 mg/L), the hospital pharmacist provided the physician with dosage advice, which implementation to patient care was analysed. A longitudinal analysis was performed to analyse if different confounding variables had an effect on posaconazole concentrations. Forty-seven patients were enrolled resulting in 217 posaconazole trough concentrations. A median of 3 (IQR 1-7) samples was measured per patient. The median concentration was 1.7 mg/L (IQR 0.8-2.7) for prophylaxis and 1.76 mg/L (IQR 1.3-2.3) for treatment. Overall, 78 posaconazole concentrations were out of the therapeutic window. For 45 (54%) of these concentrations, a dosage change was recommended. In the longitudinal analysis, the laboratory markers and patient baseline variables did not have an effect on posaconazole concentrations. Adequate posaconazole exposure was shown in 64% (affected 28 patients) of the measured concentrations. TDM practice of posaconazole can be improved by increasing the implementation rate of dose recommendation by a multidisciplinary antifungal stewardship team.

Incidence of Massive Transfusion and Overall Transfusion Requirements During Lung Transplantation Over a 25-Year Period.

OBJECTIVE: To establish the incidence of massive transfusion and overall transfusion requirements during lung transplantation, changes over time, and association with outcome in relation to patient complexity. DESIGN: Retrospective cohort study. SETTING: University hospital. PARTICIPANTS: All 514 adult patients who underwent transplantation from 1990 until 2015. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Patient records and transfusion data, divided into 5-year intervals, were analyzed. The incidence of massive transfusion (>10 units of red blood cells [RBCs] in 24 h) was 27% and did not change over time, whereas the median (interquartile range) transfusion requirement in the whole cohort decreased from 8 (5-12) to 3 (0-10) RBCs (p < 0.001). In patients transplanted from the intensive care unit, the incidence of massive transfusion increased over time from 25% to 54% (p = 0.04) and median transfusion requirements from 4.5 (3-8.5) units to 14.5 (5-26) units of RBCs (p = 0.03). Multivariable analysis showed that circulatory support, pulmonary hypertension, re-transplantation, cystic fibrosis, Eisenmenger syndrome, bilateral transplantation, and low body mass index were associated with massive transfusion. Patients with massive transfusion had more primary graft dysfunction grade III at 0, 24, 48, and 72 hours (p < 0.001), higher 30-day mortality (13% v 4%; p < 0.001), and lower 5-year survival (hazard ratio 3.67 [95% confidence interval 1.72-7.85]; p < 0.001). CONCLUSION: The incidence of massive transfusion did not change over time, whereas transfusion requirements in the whole cohort decreased. In patients transplanted from the intensive care unit, massive transfusion and transfusion requirements increased. Massive transfusion was associated with poor outcome.

Optimal unidirectional switch designs.

Stepped wedge designs and delayed start designs can all be considered as special cases of the so-called unidirectional switch design. This paper provides optimal proportions of clusters that are allocated to switch patterns in a unidirectional switch design to minimize the asymptotic variance of the treatment effect estimator. This unique optimal design applies to certain cross-sectional and longitudinal variance component models. When the intraclass correlation coefficient is zero, the optimal unidirectional switch design coincides with the classic (cluster) parallel group design. The optimal unidirectional switch design is more efficient than the optimal stepped wedge design and delayed start designs. Compared with the uniform unidirectional switch design, the efficiency gain of the optimal unidirectional switch design can be substantial, but it depends on the intraclass correlation and the cluster size. We also showed that augmenting the optimal stepped wedge design with pure control pattern is more efficient than the optimal stepped wedge design. In addition, robust minimax design for unidirectional switch design, delayed start design, and stepped wedge design are provided.

A preference-based item response theory model to measure health: concept and mathematics of the multi-attribute preference response model.

BACKGROUND: A new patient-reported health measurement model has been developed to quantify descriptions of health states. Known as the multi-attribute preference response (MAPR) model, it is based on item response theory. The response task in the MAPR is for a patient to judge whether hypothetical health-state descriptions are better or worse than his/her own health status. METHODS: In its most simple form MAPR is a Rasch model where for each respondent on the same unidimensional health scale values are estimated of their own health status and values of the hypothetical comparator health states. These values reflect the quality or severity of the health states. Alternatively, the respondents are offered health-state descriptions that are based on a classification system (e.g., multi-attribute) with a fixed number of health attributes, each with a limited number of levels. In the latter variant, the weights of the levels of the attributes in the descriptive system, which represents the range of the health states, are estimated. The results of a small empirical study are presented to illustrate the procedures of the MAPR model and possible extensions of the model are discussed. RESULTS: The small study that we conducted to illustrate the procedure and results of our proposed method to measure the quality of health states and patients' own health status showed confirming results. CONCLUSIONS: This paper introduces the typical MAPR model and shows how it extends the basic Rasch model with a regression function for the attributes of the health-state classification system.

Motor development in infancy is related to cognitive function at 4 years of age.

AIM: Evidence is accumulating that motor and cognitive development are interrelated. This study investigates associations between motor development in infancy and cognitive function at 4 years of age. METHOD: This study is part of the Groningen prospective cohort study on the development of children born after assisted reproductive techniques such as in vitro fertilization. The study group consisted of 223 children (119 males, 104 females) born to subfertile couples (median gestational age 39.6wks, range 30-43wks). Motor behaviour was assessed with the Infant Motor Profile (IMP) at 4 months, 10 months, and 18 months. IQ was evaluated at 4 years with the Kaufman Assessment Battery for Children, Second Edition. Latent class growth modelling was used to analyse relations between IMP and IQ scores. RESULTS: Infants with low total IMP scores had an IQ that was 8.9 points lower at 4 years than infants with typical IMP scores (95% confidence interval 3.6-14.1). Also, low scores in the domains of variation and performance were associated with a lower IQ at 4 years of age, by 6.1 points and 13.2 points respectively. INTERPRETATION: Motor development in relatively low-risk infants is associated with cognition at 4 years of age. In particular, low motor variation and performance are associated with a lower IQ at 4 years of age. WHAT THIS PAPER ADDS: Motor development in infancy is directly related to cognitive function at 4 years of age. This relationship is most pronounced for infant motor variation and performance. The size of the differences in IQ is clinically relevant.