The FACET (Formoterol and Corticosteroid Establishing Therapy) study established that there is a clear clinical benefit in adding formoterol to budesonide therapy in patients who have persistent symptoms of asthma despite treatment with low to moderate doses of an inhaled corticosteroid. We combined the clinical results from the FACET study with an expert survey on average resource use in connection with mild and severe asthma exacerbations in the U.K., Sweden and Spain. The primary objective of this study was to assess the health economics of adding the inhaled long-acting beta2-agonist formoterol to the inhaled corticosteroid budesonide in the treatment of asthma. The extra costs of adding the inhaled beta2-agonist formoterol to the corticosteroid budesonide in asthmatic patients in Sweden were offset by savings from reduced use of resources for exacerbations. For Spain the picture was mixed. Adding formoterol to low dose budesonide generated savings, whereas for moderate doses of budesonide about 75% of the extra formoterol costs could be recouped. In the U.K., other savings offset about half of the extra cost of formoterol. All cost-effectiveness ratios are within accepted cost-effectiveness ranges reported from previous studies. If productivity losses were included, there were net savings in all three countries, ranging from Euro 267-1183 per patient per year. In conclusion, adding the inhaled, long-acting beta2-agonist formoterol to low-moderate doses of the inhaled corticosteroid budesonide generated significant gains in all outcome measures with partial or complete offset of costs. Adding formoterol to budesonide can thus be considered to be cost-effective.
To establish normal reference values for prostate-specific antigen (PSA) in a Swedish population we investigated 878 healthy men, 56-75 years of age. They were randomly selected from a population of 9171 males in this group. Cancer of the prostate was excluded by digital rectal examination. When digital rectal examination was suspicious for carcinoma of the prostate and/or serum PSA > 4 micrograms l-1, fine-needle aspiration biopsy was performed. Central values, values of variance and reference limits were defined by a non-parametric method in four age groups. A strong positive correlation between PSA values and age was found and the variance increased with age. The relationship between PSA value and age was non-linear. For the age group 56-60 the upper reference limit (95th percentile) was 4.6 micrograms l-1 (confidence interval, CI: 3.9-5.5). For the age groups 61-65, 66-70 and 71-75 the corresponding values were 4.4 (3.8-5.2), 7.6 (6.5-8.9) and 8.4 micrograms l-1 (7.2-9.8) respectively. For the age groups studied the increment over time of the PSA value was 2-8% per year depending on age, with an average increment per year over 15 years of 4.3%. Overall, 11% of our reference sample had a serum PSA level > 4 micrograms l-1. We consider our study population to be representative for a normal Swedish male population in these age groups.
We present a general algorithm for the detection of genomic variants using the Illumina iSelect platform. The Illumina iSelect platform is designed to detect SNPs, but our algorithm allows for the detections of more general forms of variations, including copy number polymorphisms and microsatellites. The algorithm does not rely on a priori information of the type of polymorphism being studied and is designed to genotype call a large number of individuals simultaneously. The algorithm proceeds by initially normalizing intensity and correcting for batch effects. Then each marker is clustered using a modified Gaussian mixture model where we account for variances in the expression of an individuals and the variance measured in bead level intensities of a probe/marker pair. Finally, these clusters are used to determine genotypes. The algorithm was then run on a dataset of 35,000 Icelandic individuals.
Data in health research are frequently structured hierarchically. For example, data may consist of patients nested within physicians, who in turn may be nested in hospitals or geographic regions. Fitting regression models that ignore the hierarchical structure of the data can lead to false inferences being drawn from the data. Implementing a statistical analysis that takes into account the hierarchical structure of the data requires special methodologies. In this paper, we introduce the concept of hierarchically structured data, and present an introduction to hierarchical regression models. We then compare the performance of a traditional regression model with that of a hierarchical regression model on a dataset relating test utilization at the annual health exam with patient and physician characteristics. In comparing the resultant models, we see that false inferences can be drawn by ignoring the structure of the data.
Gaussian process (GP) models are widely used in disease mapping as they provide a natural framework for modeling spatial correlations. Their challenges, however, lie in computational burden and memory requirements. In disease mapping models, the other difficulty is inference, which is analytically intractable due to the non-Gaussian observation model. In this paper, we address both these challenges. We show how to efficiently build fully and partially independent conditional (FIC/PIC) sparse approximations for the GP in two-dimensional surface, and how to conduct approximate inference using expectation propagation (EP) algorithm and Laplace approximation (LA). We also propose to combine FIC with a compactly supported covariance function to construct a computationally efficient additive model that can model long and short length-scale spatial correlations simultaneously. The benefit of these approximations is computational. The sparse GPs speed up the computations and reduce the memory requirements. The posterior inference via EP and Laplace approximation is much faster and is practically as accurate as via Markov chain Monte Carlo.
Time course of the prevalent dental diseases may be satisfactorily approximated by the Gauss equation, permitting a compact and universal representation of epidemiological date. The interpretation of equation parameters is useful from a theoretical and practical viewpoints.
Harbor seals in Iliamna Lake, Alaska, are a small, isolated population, and one of only two freshwater populations of harbor seals in the world, yet little is known about their abundance or risk for extinction. Bayesian hierarchical models were used to estimate abundance and trend of this population. Observational models were developed from aerial survey and harvest data, and they included effects for time of year and time of day on survey counts. Underlying models of abundance and trend were based on a Leslie matrix model that used prior information on vital rates from the literature. We developed three scenarios for variability in the priors and used them as part of a sensitivity analysis. The models were fitted using Markov chain Monte Carlo methods. The population production rate implied by the vital rate estimates was about 5% per year, very similar to the average annual harvest rate. After a period of growth in the 1980s, the population appears to be relatively stable at around 400 individuals. A population viability analysis assessing the risk of quasi-extinction, defined as any reduction to 50 animals or below in the next 100 years, ranged from 1% to 3%, depending on the prior scenario. Although this is moderately low risk, it does not include genetic or catastrophic environmental events, which may have occurred to the population in the past, so our results should be applied cautiously.
The objectives were to compare perinatal mortality (PNM) in the Faroes and Denmark while accounting for the high birth weights in the Faroes, and to discuss methodological aspects related to this task. We applied conventional methods employing absolute birth weight standards, and the Wilcox-Russell way of comparing relative birth weights. During 1977-85 perinatal mortality (PNM) in the Faroes was 14.7 (98 cases) per 1,000 births, and 1.57 times higher than that in Denmark. Conventional method: birth weight-standardised risk ratio for PNM in the Faroes v Denmark was 1.95; the risk ratio declined with increasing birth weight. Wilcox-Russell model: the risk tended to be more uniformly increased across the birth weight distribution when babies with same relative birth weights were compared; the residual component of the birth weight distribution (i.e. the excess of observed births in the lower tail beyond what could be predicted by a Gaussian distribution) was 2.1% in the Faroes and 3.6% in Denmark, which does not fit with the model assumption that the size of the residual component is a strong determinant of a population's PNM.
We examined 429 women, aged 20-80 years, randomly selected from the population register to establish normal values for bone mineral density (BMD) in Swedish women. BMD of the spine and hip was measured by dual-energy X-ray absorptiometry (DEXA; Hologic QDR 1000) and in the forearm by single photon absorptiometry (SPA; Molsgaard ND-1100). The recalled age of menarche was negatively correlated to BMD at all ages. There was no significant change in BMD from 20-49 years at any site except a slight decline at Ward's triangle. Bone loss was rapid at all sites during the first decade after menopause. Thereafter, BMD declined slowly in the trochanter and total hip but more rapidly in the forearm, femoral neck, and Ward's triangle. BMD in the spine even increased in the eighth decade probably due to osteoarthritis. The average change in forearm BMD during the 15 perimenopausal years comprising mean age for menopause +/- 2 SD (43-57 years) was -0.4% per year in premenopausal females and -1.6% per year in postmenopausal females. The corresponding annual percental change was, for the spine, +0.2 and -1.7; neck, -0.7 and -1.7; trochanter, +0.5 and -1.5; and Ward's triangle, -0.1% and -2.2%, respectively. Our normal values for lumbar spine BMD prior to menopause did not differ from published values or the manufacturer's normal values; however, our spine BMD values for the first decade after menopause were significantly lower (approximately 10%) than in other studies. Our femoral neck BMD values for younger women were, like those of several other groups, significantly lower than the manufacturer's normal values, but our sample of young women in this study was small. The prevalence of osteoporosis, if defined as t score
Commingling and segregation of age-sex-adjusted systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial blood pressure (MBP) were examined in 1,560 individuals from 374 French-Canadian nuclear families. After correction for skewness, evidence in favor of two commingled distributions was found for SBP in the combined data (parents and offspring) and in parents, but not in offspring. Segregation analysis (using the computer program POINTER) suggested that a multifactorial contribution to all three phenotypes was greater in offspring than in parents, which could be the result of either polygenic or shared environmental components relevant to sibships, or both. Statistical evidence was found for a major effect on SBP. However, Mendelian transmission of the major effect was rejected, and no transmission of the major effect (equal tau's) was not. This is just the opposite to what would be expected if the major effect was due to a major gene, and it would ordinarily be considered as sufficient evidence to refute a major gene effect on SBP. However, the commingling in parents but not in offspring (who are all below 26 years of age), and the finding of equal transmission probabilities (nearly equal to 1), are compatible with an alternative interpretation. It is possible that there is a real major gene effect on SBP but that the genotype for elevated SBP has not yet expressed itself in the offspring as they have not yet gone through the risk period. Accordingly, this possibility needs to be evaluated further in additional studies involving older offspring.
Cites: Circulation. 1967 Dec;36(6):980-86061343
Cites: Hum Hered. 1971;21(6):523-425149961
Cites: Am J Hum Genet. 1974 Jul;26(4):489-5034842773
Cites: Am J Hum Genet. 1978 Jan;30(1):46-58623102
Cites: Am J Epidemiol. 1980 Feb;111(2):142-556965561
Cites: Am J Hum Genet. 1980 Jul;32(4):620-57395873