Multinomial N-mixture models for removal sampling.

Multinomial N -mixture models are commonly used to fit data from a removal sampling protocol. If the mixing distribution is negative binomial, the distribution of the counts does not appear to have been identified, and practitioners approximate the requisite likelihood by placing an upper bound on the embedded infinite sum. In this paper, the distribution which underpins the multinomial N -mixture model with a negative binomial mixing distribution is shown to belong to the broad class of multivariate negative binomial distributions. Specifically, the likelihood can be expressed in closed form as the product of conditional and marginal likelihoods and the information matrix shown to be block diagonal. As a consequence, the nature of the maximum likelihood estimates of the unknown parameters and their attendant standard errors can be examined and tests of the hypothesis of the Poisson against the negative binomial mixing distribution formulated. In addition, appropriate multinomial N -mixture models for data sets which include zero site totals can also be constructed. Two illustrative examples are provided.

A systematic approach to optimize electronic health record medication alerts in a health system.

PURPOSE: The effectiveness of a systematic, streamlined approach to optimize drug-drug interaction alerts in an electronic health record for a health system was studied. METHODS: An 81-week quasi-experimental study was conducted to evaluate interventions made to medication-related clinical decision-support (CDS) alerts. Medication-related CDS alerts were systematically reduced using a multi disciplinary healthcare committee. The primary endpoint was weekly overall, modification, and acknowledgement rates of medication alerts after drug-drug interaction reclassification. Secondary endpoints included sub analysis of types of medication alerts (drug-drug interaction and duplicate therapy alerts) and alert use by providers (pharmacist and prescribers). Data was analyzed using interrupted time series regression analysis. RESULTS: After implementation of the new alert system, total number of weekly inpatient alerts decreased from 68,900 (66,300-70,900) and 50,300 (48,600-53,600) in the postintervention period (p < 0.001). The perentage of alerts acknowledged weekly increased from 11.8% (IQR, 11.4-12.1%) in the preintervention period to 13.7% (IQR, 13.3-14.0%) in the postintervention period (p < 0.001). The percentage of alerts that were modified also increased from 5.0% (IQR, 4.9-5.3%) in the preintervention period to 7.3% (IQR, 7.0-7.6%) in the postintervention period (p < 0.001). Both increases were primarily seen with pharmacists versus other healthcare professionals (p < 0.001). CONCLUSION: A committee-led systematic approach to optimizing drug-drug interactions facilitated a significant decrease in the overall number of alerts and an increase in both medication alert acknowledgement and modification rates.

Maximum likelihood estimation for N-mixture models.

The focus of this article is on the nature of the likelihood associated with N-mixture models for repeated count data. It is shown that the infinite sum embedded in the likelihood associated with the Poisson mixing distribution can be expressed in terms of a hypergeometric function and, thence, in closed form. The resultant expression for the likelihood can be readily computed to a high degree of accuracy and is algebraically tractable. Specifically, the likelihood equations can be simplified to some advantage, the concentrated likelihood in the probability of detection formulated and problematic cases identified. The results are illustrated by means of a simulation study and a real world example. The study is extended to N-mixture models with a negative binomial mixing distribution and results similar to those for the Poisson case obtained. N-mixture models with mixing distributions which accommodate excess zeros and, separately, with a beta-binomial distribution rather than a binomial used to model the intra-site counts are also investigated. However the results for these settings, while computationally attractive, do not provide insight into the nature of the maximum likelihood estimates.

Start-up designs for response-adaptive randomization procedures with sequential estimation.

Response-adaptive randomization procedures are appropriate for clinical trials in which two or more treatments are to be compared, patients arrive sequentially and the response of each patient is recorded before the next patient arrives. However, for those procedures that involve sequential estimation of model parameters, start-up designs are commonly required in order to provide initial estimates of the parameters. In this paper, a suite of such start-up designs for two treatments and binary patient responses are considered and compared in terms of the numbers of patients required in order to give meaningful parameters estimates, the number of patients allocated to the better treatment, and the bias in the parameter estimates. It is shown that permuted block designs with blocks of size 4 are to be preferred over a wide range of parameter values. For the case of two treatments, normal responses and selected start-up procedures, a design incorporating complete randomization followed appropriately by repeats of one of the treatments yields the minimum expected number of patients and is to be preferred.

Sequential designs for ordinal phase I clinical trials.

Sequential designs for phase I clinical trials which incorporate maximum likelihood estimates (MLE) as data accrue are inherently problematic because of limited data for estimation early on. We address this problem for small phase I clinical trials with ordinal responses. In particular, we explore the problem of the nonexistence of the MLE of the logistic parameters under a proportional odds model with one predictor. We incorporate the probability of an undetermined MLE as a restriction, as well as ethical considerations, into a proposed sequential optimal approach, which consists of a start-up design, a follow-on design and a sequential dose-finding design. Comparisons with nonparametric sequential designs are also performed based on simulation studies with parameters drawn from a real data set.

Interventions to reduce dosing errors in children: a systematic review of the literature.

Children are a particularly challenging group of patients when trying to ensure the safe use of medicines. The increased need for calculations, dilutions and manipulations of paediatric medicines, together with a need to dose on an individual patient basis using age, gestational age, weight and surface area, means that they are more prone to medication errors at each stage of the medicines management process. It is already known that dose calculation errors are the most common type of medication error in neonatal and paediatric patients. Interventions to reduce the risk of dose calculation errors are therefore urgently needed. A systematic literature review was conducted to identify published articles reporting interventions; 28 studies were found to be relevant. The main interventions found were computerised physician order entry (CPOE) and computer-aided prescribing. Most CPOE and computer-aided prescribing studies showed some degree of reduction in medication errors, with some claiming no errors occurring after implementation of the intervention. However, one study showed a significant increase in mortality after the implementation of CPOE. Further research is needed to investigate outcomes such as mortality and economics. Unit dose dispensing systems and educational/risk management programmes were also shown to reduce medication errors in children. Although it is suggested that 'smart' intravenous pumps can potentially reduce infusion errors in children, there is insufficient information to draw a conclusion because of a lack of research. Most interventions identified were US based, and since medicine management processes are currently different in different countries, there is a need to interpret the information carefully when considering implementing interventions elsewhere.

Rising incidence of type 2 diabetes in children in the U.K.

OBJECTIVE: To estimate the incidence of type 2 diabetes in children <17 years of age and to investigate the relationship of diabetes with increasing childhood obesity in the U.K. and the Republic of Ireland (ROI). RESEARCH DESIGN AND METHODS: Active monthly reporting of cases by consultant pediatricians occurred through the framework of the British Pediatric Surveillance Unit, with additional reports from specialist diabetes nurses. All children <17 years of age and diagnosed by their clinician as having non-type 1 diabetes from 1 October 2004 to 31 October 2005 were included. RESULTS: A total of 168 confirmed cases of non-type 1 diabetes were reported, resulting in a national incidence (excluding the ROI) of 1.3 x 100,000(-1) x year(-1). Of these, 40% were diagnosed with type 2 diabetes giving a minimum incidence of 0.53 x 100,000(-1) x year(-1). Children of ethnic minorities were greatly overrepresented, with those of black and South-Asian origin (England data only) having an incidence of 3.9 and 1.25 x 100,000(-1) x year(-1), respectively, compared with 0.35 x 100,000(-1) x year(-1) in those defined as white. Of those diagnosed with type 2 diabetes, 95% were overweight and 83% obese according to International Obesity Task Force guidelines. Eighty-four percent had a family history of type 2 diabetes. CONCLUSIONS: Type 2 diabetes is still less common than type 1 diabetes in U.K. children. However, compared with previous prevalence data, the frequency of type 2 diabetes appears to be increasing. Incidence among ethnic minorities is far higher than in whites, as previously described in the U.S. Increased adiposity and family history of type 2 diabetes were strongly associated with the diagnosis of type 2 diabetes in U.K. children.

Sequential designs for logistic phase I clinical trials.

Both parametric and nonparametric sequential designs and estimation methods are implemented in phase I clinical trials. In this article, we take a systematic approach, consisting of a start-up design, a follow-on design, a sequential dose-finding design, and an estimation method, to find an efficient estimate of the maximum tolerated dose under the assumption that the dose-response curve has a true underlying logistic distribution. In particular, for the problem of the nonexistence of the maximum likelihood estimates of the logistic parameters, a constraint on the probability of an undetermined maximum likelihood estimator (MLE) is incorporated into the parametric sequential designs. In addition, this approach can also be extended to incorporate ethical considerations, which prohibit an administered dose from exceeding the maximum acceptable dose. Comparison based on simulation studies between the systematic designs and nonparametric designs are described both for continuous dose spaces and discrete dose spaces, respectively.

Health-related quality of life of children with vision impairment or blindness.

The aims of the study were to describe the functional ability, health status, and health-related quality of life (HRQL) of young children with a vision impairment or blindness (VI/BL) and to examine the effect of different types of ophthalmic condition and the presence of other impairments or systemic disorders. A cross-sectional community survey of children aged 3 to 8 years with VI/BL was conducted in four areas of England using the Health Utilities Index Mark 3 system. Seventy-nine children (47 males, 32 females; mean age 6 y 2 mo [SD 1 y 6 mo]) met the selection criteria: 43% had a visual pathway condition, 38% a condition of the eye, and 19% nystagmus alone; and 61% had additional impairments/disorders. Eighty per cent had functional limitations on at least two of the following attributes: vision, hearing, speech, cognition, ambulation, dexterity, emotion, and pain. Forty-four per cent had functional limitations on four or more attributes. Children with nystagmus alone had significantly higher mean HRQL utility scores (0.80 [SD 0.26]) than children with a condition of the eye (0.45 [SD 0.33]), who, in turn, had higher scores than children with a visual pathway condition (0.05 [SD 0.33], p=0.002). Children with an isolated VI/BL had significantly higher mean scores (0.73 [SD 0.21]) than those with additional impairments/disorders (0.09 [SD 0.34], p<0.001). These findings underline the need for a broad assessment of each child with VI/BL and a multidisciplinary approach to care.

The management of children with chronic fatigue syndrome-like illness in primary care: a cross-sectional study.

BACKGROUND: Most studies on children with chronic fatigue syndrome (CFS)/myalgic encephalomyelitis (ME) have been undertaken in tertiary care and little is known about their management in primary care. AIM: To describe the characteristics of patients aged 5-19 years with CFS-like illness in primary care and to examine how GPs investigate and manage patients. DESIGN OF STUDY: Descriptive retrospective questionnaire study. SETTING: Sixty-two UK GP practices in the MRC General Practice Research Framework (GPRF). METHOD: One hundred and twenty-two practices were approached; 62 identified 116 patients consulting a GP with severe fatigue lasting over 3 months. Practice nurses and GPs completed questionnaires from medical notes and patients completed postal questionnaires. RESULTS: Ninety-four patients were considered by a clinical panel, blind to diagnosis, to meet the Oxford CFS criteria with a fatigue duration of 3 months. Seventy-three per cent were girls, 94% white, mean age was 12.9 years and median illness duration 3.3 years. GPs had principal responsibility for 62%. A diagnosis of CFS/ME was made in 55%, 30% of these within 6 months. Fifty per cent had a moderate illness severity. Paediatric referrals were made in 82% and psychiatric referrals in 46% (median time of 2 and 13 months respectively). Advice given included setting activity goals, pacing, rest and graded exercise. CONCLUSIONS: Patient characteristics are comparable to those reported in tertiary care, although fewer are severe cases. GPs have responsibility for the majority of patients, are diagnosing CFS/ME within a short time and applying a range of referral and advice strategies.

Cost-effectiveness of digital photographic screening for retinopathy of prematurity in the United Kingdom.

OBJECTIVES: To compare the cost-effectiveness of alternative methods of screening for retinopathy of prematurity (ROP) in the United Kingdom, including the existing method of indirect ophthalmoscopy by ophthalmologists and digital photographic screening by nurses. METHODS: A decision tree model was used to compare five screening modalities for the UK population of preterm babies, using a health service perspective. Data were taken from published sources, observation at a neonatal intensive care unit (NICU), and expert judgment. RESULTS: We estimated that use of standard digital cameras by nurses in NICUs would cost more than current methods (pound 371 compared with pound 321 per baby screened). However, a specialist nurse visiting units with a portable camera would be cheaper (pound 172 per baby). These estimates rely on nurses capturing and interpreting the images, with suitable training and supervision. Alternatively, nurses could capture the images then transmit them to a central unit for interpretation by ophthalmologists, although we estimate that this would be rather more expensive (pound 390 and pound 201, respectively, for NICU and visiting nurses). Sensitivity analysis was used to examine the robustness of estimates. CONCLUSIONS: It is likely that there is an opportunity to improve the efficiency of the ROP screening program. We estimate that screening by specialist nurses trained in image capture and interpretation using portable digital cameras is a cost-effective alternative to the current program of direct visualization by ophthalmologists. This option would require the development of a suitable portable machine. Direct comparative research is strongly needed to establish the accuracy of the various screening options.

The evolution of placental mammal body sizes: evolutionary history, form, and function.

The unimodal, right-skewed distribution, most frequently identified in contemporary descriptions of placental mammal body size distributions, masks an underlying multidistribution structure; a long-term evolutionary process that has generated a concatenation of two or three frequency distributions specific to locomotory modes (plantigrade, digitigrade and unguligrade). The Afrotropical assemblages are bimodal, with a tendency towards trimodality, whereas the Nearctic assemblage is unimodal. However, mixtures of two and three normal distributions fitted the Nearctic data well, suggesting a multidistribution structure masked by disproportionate species numbers within locomotory modes. Differences in proportional species numbers within modes between assemblages may reflect the evolutionary history of form and function. However, common interassemblage predictions of such proportions in contemporary distributions may be disguised by the relative severity of the Pleistocene megafaunal extinction (patterns supported by the fossil record), geographical scale, and taxonomic composition. A species gap occurs at body sizes around 1 kg at the interface between the largest plantigrade mammals and the smallest digitigrade mammals, coincident with the minimum interspecific variance of basal metabolic rate. In terms of the evolution of the optimal body size in the trade-off between mortality and production, there may be good historical and evolutionary reasons why we should not expect optimization to produce the same results in different zoogeographical assemblages. Moreover, the evolution of diverse mammalian forms and functions, especially with respect to predator-prey interactions and diet, render a single body size optimum untenable in the search for an energetic definition of fitness.

Bayesian optimal designs for Phase I clinical trials.

A broad approach to the design of Phase I clinical trials for the efficient estimation of the maximum tolerated dose is presented. The method is rooted in formal optimal design theory and involves the construction of constrained Bayesian c- and D-optimal designs. The imposed constraint incorporates the optimal design points and their weights and ensures that the probability that an administered dose exceeds the maximum acceptable dose is low. Results relating to these constrained designs for log doses on the real line are described and the associated equivalence theorem is given. The ideas are extended to more practical situations, specifically to those involving discrete doses. In particular, a Bayesian sequential optimal design scheme comprising a pilot study on a small number of patients followed by the allocation of patients to doses one at a time is developed and its properties explored by simulation.

Competing designs for phase I clinical trials: a review.

Phase I clinical trials are typically small, uncontrolled studies designed to determine a maximum tolerated dose of a drug which will be used in further testing. Two divergent schools have developed in designing phase I clinical trials. The first defines the maximum tolerated dose as a statistic computed from data, and hence it is identified, rather than estimated. The second defines the maximum tolerated dose as a parameter of a monotonic dose-response curve, and hence is estimated. We review techniques from both philosophies. The goal is to present these methods in a single package, to compare them from philosophical and statistical grounds, to hopefully clear up some common misconceptions, and to make a few recommendations. This paper is not a review of simulation studies of these designs, nor does it present any new simulations comparing these designs.