Features associated with cardiac abnormalities in systemic lupus erythematosus.

OBJECTIVES: To determine the prevalence of echocardiographic abnormalities and identify associated clinical and laboratory features in a large systemic lupus erythematosus (SLE) cohort. METHODS: Patients fulfilling ACR criteria for SLE underwent a transthoracic echocardiogram (TTE) between January 2005 and June 2006. Variables used as potential correlates included age, sex, ethnicity, lupus duration, lupus disease activity (SLEDAI), cumulative damage (SLICC/ACR damage index (DI)), arterial hypertension, diabetes, current smoking, medication use and laboratory data. Multivariate logistic regression was used to examine the association between TTE abnormalities and potential determinants. RESULTS: For the 217 subjects with a TTE performed during the study, the main abnormalities were of the mitral valve (37.3%) and included thickening (25.4%) and insufficiency (25.8%). Other findings included pulmonary artery pressure (PAP) ≥ 30( )mm( )Hg (10.1%), pericardial effusion (4.6%), hypokinesis (2.8%), and aortic insufficiency (3.7%). In multivariate analysis, mitral insufficiency was associated with the use of corticosteroids (OR 2.90; 95%CI 1.42-5.94) and hypokinesis with angiotensin-converting enzyme inhibitors (12.89; 1.06-157.18). Elevated PAP was associated with age (1.04; 1.01-1.07) and with DI (1.20; 1.01-1.42). CONCLUSION: Valvular abnormalities are frequent in patients with SLE, with mitral valve lesions occurring in over one third. TTE screening may be indicated in patients with SLE, especially for those with identified risk factors such as corticosteroid use.

Association of the Charlson comorbidity index with mortality in systemic lupus erythematosus.

OBJECTIVE: To investigate whether comorbidity as assessed by the Charlson Comorbidity Index (CCI) is associated with mortality in a long-term followup of systemic lupus erythematosus (SLE) patients. METHODS: Data were collected from 499 SLE patients attending the Lupus Clinic at the McGill University Health Center, Montreal, Quebec, Canada, and 170 SLE patients from the Department of Rheumatology at Lund University Hospital, Lund, Sweden. This included data on comorbidity, demographics, disease activity, the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI), and antiphospholipid antibody syndrome (APS). Variables were entered into a Cox proportional hazards survival model. RESULTS: Mortality risk in the Montreal cohort was associated with the CCI (hazard ratio [HR] 1.57 per unit increase in the CCI, 95% confidence interval [95% CI] 1.18-2.09) and age (HR 1.04 per year increase in age, 95% CI 1.00-1.09). The CCI and age at diagnosis were also associated with mortality in the Lund cohort (CCI: HR 1.35, 95% CI 1.13-1.60; age: HR 1.09, 95% CI 1.05-1.12). Furthermore, the SDI was associated with mortality in the Lund cohort (HR 1.40, 95% CI 1.19-1.64), while a wide CI for the estimate in the Montreal cohort prevented a definitive conclusion (HR 1.20, 95% CI 0.97-1.48). We did not find a strong association between mortality and sex, race/ethnicity, disease activity, or APS in either cohort. CONCLUSION: In this study, comorbidity as measured by the CCI was associated with decreased survival independent of age, lupus disease activity, and damage. This suggests that the CCI may be useful in capturing comorbidity for clinical research in SLE.

Surveillance of systemic autoimmune rheumatic diseases using administrative data.

There is growing interest in developing tools and methods for the surveillance of chronic rheumatic diseases, using existing resources such as administrative health databases. To illustrate how this might work, we used population-based administrative data to estimate and compare the prevalence of systemic autoimmune rheumatic diseases (SARDs) across three Canadian provinces, assessing for regional differences and the effects of demographic factors. Cases of SARDs (systemic lupus erythematosus, scleroderma, primary Sjogren's, polymyositis/dermatomyositis) were ascertained from provincial physician billing and hospitalization data. We combined information from three case definitions, using hierarchical Bayesian latent class regression models that account for the imperfect nature of each case definition. Using methods that account for the imperfect nature of both billing and hospitalization databases, we estimated the over-all prevalence of SARDs to be approximately 2-3 cases per 1,000 residents. Stratified prevalence estimates suggested similar demographic trends across provinces (i.e. greater prevalence in females-versus-males, and in persons of older age). The prevalence in older females approached or exceeded 1 in 100, which may reflect the high burden of primary Sjogren's syndrome in this group. Adjusting for demographics, there was a greater prevalence in urban-versus-rural settings. In our work, prevalence estimates had good face validity and provided useful information about potential regional and demographic variations. Our results suggest that surveillance of some rheumatic diseases using administrative data may indeed be feasible. Our work highlights the usefulness of using multiple data sources, adjusting for the error in each.

Polymyalgia rheumatica prevalence in a population-based sample.

OBJECTIVE: To determine polymyalgia rheumatica (PMR) prevalence using population-based administrative data, and to estimate the error associated with case ascertainment approaches when using these databases. METHODS: Cases were ascertained using physician billing and hospitalization data from the province of Manitoba (population 1.1 million). Focusing on the population age >/=45 years, we compared 3 different case definition algorithms and also used statistical methods that accounted for imperfect case ascertainment to estimate the prevalence and the properties of the ascertainment algorithms. A hierarchical Bayesian latent class regression model was developed that also allowed us to assess differences across patient demographics (sex and region of residence). RESULTS: Using methods that account for the imperfect nature of both billing and hospitalization databases, we estimated the prevalence of PMR in women age >/=45 years to be lower in urban areas (754.5 cases/100,000; 95% credible interval [95% CrI] 674.1-850.3) compared with rural areas (1,004 cases/100,000; 95% CrI 886.3-1,143). This regional trend was also seen in men age >/=45 years, where the prevalence was estimated at 273.6 cases/100,000 (95% CrI 219.8-347.6) in urban areas and 380.7 cases/100,000 (95% CrI 311.3-468.1) in rural areas. Billing data appeared more sensitive in ascertaining cases than hospitalization data, and a large proportion of diagnoses was made by physicians other than rheumatologists. CONCLUSION: These data suggest a higher prevalence of PMR in rural versus urban regions. Our approach demonstrates the usefulness of methods that adjust for the imperfect nature of multiple information sources, which also allow for estimation of the sensitivity of different case ascertainment approaches.

Scleroderma prevalence: demographic variations in a population-based sample.

OBJECTIVE: To estimate the prevalence of systemic sclerosis (SSc) using population-based administrative data, and to assess the sensitivity of case ascertainment approaches. METHODS: We ascertained SSc cases from Quebec physician billing and hospitalization databases (covering approximately 7.5 million individuals). Three case definition algorithms were compared, and statistical methods accounting for imperfect case ascertainment were used to estimate SSc prevalence and case ascertainment sensitivity. A hierarchical Bayesian latent class regression model that accounted for possible between-test dependence conditional on disease status estimated the effect of patient characteristics on SSc prevalence and the sensitivity of the 3 ascertainment algorithms. RESULTS: Accounting for error inherent in both the billing and the hospitalization data, we estimated SSc prevalence in 2003 at 74.4 cases per 100,000 women (95% credible interval [95% CrI] 69.3-79.7) and 13.3 cases per 100,000 men (95% CrI 11.1-16.1). Prevalence was higher for older individuals, particularly in urban women (161.2 cases per 100,000, 95% CrI 148.6-175.0). Prevalence was lowest in young men (in rural areas, as low as 2.8 cases per 100,000, 95% CrI 1.4-4.8). In general, no single algorithm was very sensitive, with point estimates for sensitivity ranging from 20-73%. CONCLUSION: We found marked differences in SSc prevalence according to age, sex, and region. In general, no single case ascertainment approach was very sensitive for SSc. Therefore, using data from multiple sources, with adjustment for the imperfect nature of each, is an important strategy in population-based studies of SSc and similar conditions.

Estimating the prevalence of polymyositis and dermatomyositis from administrative data: age, sex and regional differences.

OBJECTIVE: To estimate the prevalence of polymyositis and dermatomyositis using population-based administrative data, the sensitivity of case ascertainment approaches and patient demographics and these parameters. METHODS: Cases were ascertained from Quebec physician billing and hospitalisation databases (approximately 7.5 million beneficiaries). Three different case definition algorithms were compared, and statistical methods were also used that account for imperfect case ascertainment, to generate estimates of disease prevalence and case ascertainment sensitivity. A hierarchical Bayesian latent class regression model was developed to assess patient characteristics with respect to these parameter estimates. RESULTS: Using methods that account for the imperfect nature of both billing and hospitalisation databases, the 2003 prevalence of polymyositis and dermatomyositis was estimated to be 21.5/100,000 (95% credible interval (CrI) 19.4 to 23.9). Prevalence was higher for women and for older individuals, with a tendency for higher prevalence in urban areas. Prevalence estimates were lowest in young rural men (2.7/100,000, 95% CrI 1.6 to 4.1) and highest in older urban women (70/100,000, 95% CrI 61.3 to 79.3). Sensitivity of case ascertainment tended to be lower for older versus younger individuals, particularly for rheumatology billing data. Billing data appeared more sensitive in ascertaining cases in urban (vs rural) regions, whereas hospitalisation data seemed most useful in rural areas. CONCLUSIONS: Marked variations were found in the prevalence of polymyositis and dermatomyositis according to age, sex and region. These methods allow adjustment for the imperfect nature of multiple data sources and estimation of the sensitivity of different case ascertainment approaches.

Prevalence of Schistosoma japonicum infection among animals in fifty villages of Samar province, the Philippines.

In the Philippines, there is a need to understand the contribution of different domestic and wild animals in transmitting Schistosoma japonicum infection to humans better. The current study describes variation in animal S. japonicum prevalence across 50 endemic villages of Samar Province, the Philippines. A total of 50 villages were selected, 25 with predominantly rain-fed farms and 25 with some irrigation system. At least 35 cats, dogs, pigs, and water buffaloes each were randomly selected and 30 rat traps were set in each village. Fecal samples were collected for up to three consecutive days for each species. The Danish Bilharziasis Laboratory method (DBL method) was used to determine S. japonicum infection status. A hierarchical logistic regression model with clustering by village and with adjustment for measurement error of the DBL method was used to estimate the prevalence of infection per village and species. Stool samples were collected from 23.4% (1189), 28.6% (1274), 36.3% (1899), and 49.4% (873) of the censused dogs, cats, pigs, and water buffaloes, respectively, and from 663 rats. The adjusted prevalence of S. japonicum infection varied greatly across villages ranging from 1.6% (95% Bayesian Credible Interval: 0.1%-10.2%) to 86.3% (65.9%-97.8%) for dogs, from 0.1% (0%-2.1%) to 21.7% (4.7%-51.2%) for cats, from 0.01% (0.0%%-1.0%) to 18.4% (7.1%-34.7%) for pigs, from less than 0.1% (0.0%-1.2%) to 72.5% (46.0%-97.4%) for water buffaloes, and from 0.7% (0.0%-9.0%) to 95.4% (77.2%-99.9%) for rats. This is the most comprehensive study of animal S. japonicum infection conducted to date. Our results show that, unlike what has been reported in China, very few water buffaloes were infected whereas rats and dogs show high prevalence proportions of infection. This, combined with significant village-to-village variation in prevalence of S. japonicum infection, suggest possible different transmission dynamics of the infection in the Province of Samar in the Philippines and China.

The persistence of anticardiolipin antibodies is associated with an increased risk of the presence of lupus anticoagulant and anti-beta2-glycoprotein I antibodies.

OBJECTIVE: We studied antiphospholipid antibodies (aPL) in blood samples from a cohort of individuals followed for thrombosis to determine whether the persistent presence of anticardiolipin antibodies (aCL) is associated with a greater likelihood of having lupus anticoagulant and/or anti-beta2-glycoprotein I antibodies (LA/abeta2GPI). METHODS: Blood samples from 353 individuals who had been tested for aCL on at least two occasions were tested for abeta2GPI and LA. Two groups were defined: aCL-persistent, who tested aCL-positive on at least two occasions, and aCL non-persistent, who tested aCL-positive on fewer than two occasions. Multivariate logistic regressions were performed using LA/abeta2GPI, LA and abeta2GPI as outcome variables and the percentage of aCL-positive tests as the predictor variable, adjusted for age, gender, family history of cardiovascular disease (CVD), systemic lupus erythematosus (SLE), smoking and number of venous (VT) and arterial thromboses (AT). RESULTS: Sixty-eight (19%) individuals were aCL persistent and 285 (81%) were aCL non-persistent. LA/abeta2GPI was found in 36 (53%) of the aCL persistent group and 38 (13%) of the aCL non-persistent group. The two groups were similar for age, gender and smoking. Family history of CVD, SLE, VT and AT were more frequent in the aCL persistent group. Multivariate analyses revealed that odds ratios for LA/abeta2GPI, LA and abeta2GPI were 1.34 [95% confidence interval (CI) = 1.22-1.47], 1.36 (95% CI = 1.24-1.50) and 1.47 (95% CI = 1.31-1.65) respectively for each 10% increase in aCL-positive tests vs 0% positive tests. CONCLUSION: Persistence of aCL positivity is associated with an increased risk of LA/abeta2GPI.

Effect of delay in initiating radiotherapy for patients with early stage breast cancer.

AIMS: For early stage breast cancer, a standard treatment option is partial mastectomy followed by radiation treatment. The 5-year risk of local recurrence ranges from 6-9%. Variable waiting times for radiation treatment of breast cancer in our institution provided an opportunity to evaluate the impact of waiting time on the risk of local recurrence. MATERIALS AND METHODS: Between January 1988 and December 1989, 482 patients with stage I and II breast cancer were treated with radiotherapy in our institution. Information on prognostic factors, such as age, tumour size, histological grade, number of positive lymph nodes and margins of resection, was abstracted from their charts. The interval between date of surgery and date of initial radiation treatment was noted. Dates of local recurrence, metastasis and deaths were recorded. RESULTS: At 5 years, the local recurrence rate was 8%, the metastatic rate 12% and the 'cause-specific' survival rate 90%. In univariate analysis, the risk of local recurrence was associated with younger age, higher histological grade and time to radiation treatment. In the multivariate analysis, the effect of time to radiation treatment on the risk of local recurrence was equivocal. CONCLUSION: Delay in radiation treatment may be associated with an increased risk of local recurrence of breast cancer, at least in our centre. Future research is needed on a larger data set to more accurately estimate the effect of time to radiation treatment on the risk of local recurrence.

Predictors of quality of life 6 months and 1 year after acute myocardial infarction.

BACKGROUND: Quality of life (QOL) is an increasingly important outcome measure after hospital admission for acute myocardial infarction (AMI). However, the ability to adjust these outcomes for differences between compared groups of patients is limited because the predictors of QOL after AMI are unknown. METHODS: To identify any clinical, demographic, and psychosocial characteristics of patients at admission that were independent predictors of QOL 6 months and 1 year after AMI, we measured physical and mental QOL (Short Form-36 Physical and Mental Component summary scores) and overall QOL (EuroQol health perception scale) in a prospective cohort of 587 patients admitted at 10 hospitals in Quebec. A set of plausible multivariate linear regression models was created for each outcome measure with use of the Bayesian Information Criterion. RESULTS: Mean physical, mental, and overall QOL scores corresponding to the time immediately before admission (baseline) were 45 (95% confidence interval [CI] 44-46), 47 (95% CI 46-48), and 70 (95% CI 68-72), respectively. By 1 year, mean physical, mental, and overall QOL scores were close to baseline (45 [95% CI 44-46], 48 [95% CI 47-49], and 73 [95% CI 71-74], respectively). The predictors of physical, mental, and overall QOL were similar at 6 months and 1 year. Important predictors of physical QOL were the corresponding score at baseline, age, and previous bypass surgery (beta coefficients at 1 year: 5 [per 10-point difference in baseline score], -1 [per 10-year age difference], 5.3; 95% CIs 4 to 5, -2 to -1, -9.2 to -1.3, respectively). Predictors of mental QOL were the corresponding score at baseline and depression (beta coefficients at 1 year: 3 [per 10-point difference in baseline score], -3 [per 10-point difference in depression score]; 95% CIs 2 to 4, -5 to -2, respectively). Predictors of overall QOL included the corresponding score at baseline and age (beta at 1 year: 2 [per 10-point score difference], -3 [per 10-year age difference]; 95% CIs 1 to 3, -4 to -1, respectively). Depression was also a predictor of impaired physical and overall QOL at 6 months (beta at 6 months: -1.6 [per 10-point score difference], -5.4 [per 10-point score difference]; 95% CIs -2.9 to -0.4, -7.7 to -3.2, respectively). No variables related to treatments received in-hospital were found in the most clinically relevant models. CONCLUSIONS: These results suggest that age and psychosocial characteristics at baseline are the most important predictors of QOL after AMI. Other clinical characteristics and treatments received in-hospital do not appear to strongly affect patients' long-term perceptions of QOL.

Iron stores and coronary artery disease: a clinical application of a method to incorporate measurement error of the exposure in a logistic regression model.

Rates of coronary artery disease (CAD) increase sharply after menopause. We examined the hypotheses that high iron stores, as measured by plasma ferritin levels, are a risk factor for CAD and that the increase in iron stores after menopause is at least in part responsible for the rise in CAD in women. We also investigated measurement error of plasma ferritin using a Bayesian conditional independence model and incorporated it into the estimation of the odds ratio (OR) for males. Cases had >/=1 coronary artery stenosis >/=70%. Controls had no visible coronary lesions on angiography. The median plasma ferritin level was 48 mg/L (interquartile range: 28 to 86) among 244 cases and 45 mg/l (24 to 85) among 140 controls. The multivariate analyses among females, males, and females and males combined did not support an association between plasma ferritin levels and CAD (OR for one unit change in log ferritin 1.01, 95% CI 0.71-1.44, OR 0.95, 95% CI 0.66-1.37 and OR 0.95, 95% CI 0.75-1.21, respectively). Accounting for the measurement error of ferritin in males slightly improved the precision of the estimate of the OR but did not unmask an association (OR: 0.94, 95% CI 0.69-1.30). We conclude that high ferritin levels before or after menopause are not associated with CAD. Measurement error might be considered in situations where a one-time measurement is assumed to be representative of long-term exposure.

Learning from discordance in patient and physician global assessments of systemic lupus erythematosus disease activity.

OBJECTIVE: Differences have been described between patient and physician assessments of well being in several chronic illnesses, and these differences may affect outcome. Disagreement may lead to dissatisfaction and to behaviors with dangerous consequences. We describe and identify predictors of patient-physician differences on ratings of disease activity in systemic lupus erythematosus (SLE). METHODS: Data collected on 154 patients included age, education, disease duration, and patient and physician global assessments of lupus activity on a 10 cm visual analog scale (VAS), the Health Assessment Questionnaire (HAQ), the Medical Outcome Study Short-Form 36 (SF-36), the Systemic Lupus Disease Activity Index (SLEDAI), the Systemic Lupus Activity Measure (SLAM-R), and the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI). Multiple linear regression models were performed using patient VAS scores, physician VAS scores, and patient minus physician VAS scores as the dependent variables, and age, disease duration, selected SF-36 and SLAM-R subscales, and SDI as independent variables. RESULTS: Patients were 90% female and 80% Caucasian, with a mean education of 13 +/- 2.8 years and a mean age of 43.1 +/- 13.6 years. The overall mean disease duration was 10.5 +/- 7.8 years. Physicians overscored patients by 2.5 cm in 6% of the cases and patients overscored physicians in 16% of the cases. The best multivariate model to predict overall differences included SF-36 mental health and SLAM-R kidney scores. CONCLUSION: Patient-physician differences may result from a divergence in focus. Patients score lupus activity based on their psychological status, while physicians rely more heavily on the physical effect of the disease.

Outcomes of total hip and knee replacement: preoperative functional status predicts outcomes at six months after surgery.

OBJECTIVE: To determine whether patients with knee or hip osteoarthritis (OA) who have worse physical function preoperatively achieve a postoperative status that is similar to that of patients with better preoperative function. METHODS: This study surveyed an observational cohort of 379 consecutive patients with definite OA who were without other inflammatory joint diseases and were undergoing either total hip or knee replacement in a US (Boston) and a Canadian (Montreal) referral center. Questionnaires on health status (the Short Form 36 and Western Ontario and McMaster Universities Osteoarthritis Index) were administered preoperatively and at 3 and 6 months postoperatively. Physical function and pain due to OA were deemed the most significant outcomes to study. RESULTS: Two hundred twenty-two patients returned their questionnaires. Patients in the 2 centers were comparable in age, sex, time to surgery, and proportion of hip/knee surgery. The Boston group had more education, lower comorbidity, and more cemented knee prostheses. Patients undergoing hip or knee replacement in Montreal had lower preoperative physical function and more pain than their Boston counterparts. In patients with lower preoperative physical function, function and pain were not improved postoperatively to the level achieved by those with higher preoperative function. This was most striking in patients undergoing total knee replacement. CONCLUSION: Surgery performed later in the natural history of functional decline due to OA of the knee, and possibly of the hip, results in worse postoperative functional status.

Can health utility measures be used in lupus research? A comparative validation and reliability study of 4 utility indices.

OBJECTIVE: To assess validity and reliability of 4 utility indices in patients with systemic lupus erythematosus (SLE). METHODS: Twenty-five patients with stable SLE underwent assessment of disease activity [Systemic Lupus Disease Activity Measure (SLAM-R) and SLE Disease Activity Index (SLEDAI)] and damage [Systemic Lupus Collaborating Clinics/American College of Rheumatology Damage Index (SLICC/ACR DI)] and completed a health survey [Medical Outcome Survey Short Form-36 (SF-36)] and 4 utility measures: the visual analog scale (VAS), the time trade-off (TTO), the standard gamble (SG), and the McMaster Health Utilities Index Mark 2 (HUI2). To assess validity, Pearson's correlations were calculated between the SF-36 subscales and the utility measures. To assess reliability, intraclass correlations or kappa coefficients were calculated between first and second assessments, performed from 2 to 4 weeks apart, in patients without important clinical change in disease activity. RESULTS: Disease activity from a SLAM-R varied from 0 to 14 (median = 4) and SLEDAI from 0 to 18 (median = 0). All subscales of the SF-36 correlated well with the VAS [lowest r = 0.56, 95% confidence interval (CI) (0.17, 0.80)] and poorly with the SG [maximum r = 0.41, CI (-0.01, 0.70); minimum r = 0.10, CI (-0.32, 0.50)]. The subscales of bodily pain (r = 0.56), mental health (r = 0.45), physical functioning (r = 0.62), role-emotional (r = 0.47), social functioning (r = 0.49) and vitality (r = 0.44) correlated significantly with TTO. All subscales correlated significantly [lowest r = 0.48, CI (0.09, 0.75)] with the HUI2 index of pain. Intraclass correlations for the VAS (ICC = 0.67) and TTO (ICC = 0.60) were good. They were fair for the SG (ICC = 0.45). The kappa coefficient was poor (0.32) for the HUI attribute of pain, but varied from fair (0.46) to excellent (0.88) for the remaining attributes. Regression analysis showed that a model incorporating the SLAM-R value and SF-36 subset of mental health was a good predictor of VAS and TTO utility measures. CONCLUSION: The VAS, TTO, and to some extent, the HUI2, when compared with the SF-36 health survey, are valid and reliable measures to assess health related quality of life in a group of patients with SLE and may be useful for future research in this population. On the basis of these results the usefulness of the SG is questionable in these patients.

Taking account of between-patient variability when modeling decline in Alzheimer's disease.

The pattern of deterioration in patients with Alzheimer's disease is highly variable within a given population. With recent speculation that the apolipoprotein E allele may influence rate of decline and claims that certain drugs may slow the course of the disease, there is a compelling need for sound statistical methodology to address these questions. Current statistical methods for describing decline do not adequately take into account between-patient variability and possible floor and/or ceiling effects in the scale measuring decline, and they fail to allow for uncertainty in disease onset. In this paper, the authors analyze longitudinal Mini-Mental State Examination scores from two groups of Alzheimer's disease subjects from Palo Alto, California, and Minneapolis, Minnesota, in 1981-1993 and 1986-1988, respectively. A Bayesian hierarchical model is introduced as an elegant means of simultaneously overcoming all of the difficulties referred to above.

Change-point analysis of neuron spike train data.

In many medical experiments, data are collected across time, over a number of similar trials, or over a number of experimental units. As is the case of neuron spike train studies, these data may be in the form of counts of events per unit of time. These counts may be correlated within each trial. It is often of interest to know if the introduction of an intervention, such as the application of a stimulus, affects the distribution of the counts over the course of the experiment. In such investigations, each trial generates a sequence of data that may or may not contain a change in distribution at some point in time. Each sequence of integer counts can be viewed as arising from a Poisson process and are therefore independently distributed or as an integer-valued time series that allows for correlations between these counts. The main aim of this paper is to show how the ensemble of sample paths may be used to make inference about the distribution of the instantaneous times of change in a given population. This will be accomplished using a Bayesian hierarchical model for these change-points in time. A bonus of these models is they also allow for inference about the probability of a change in each unit and the magnitude of the effects, if any. The use of such change-point models on integer-valued time series is illustrated on neuron spike train data, although the methods can be applied to other situations where integer-valued processes arise.

Bayesian and mixed Bayesian/likelihood criteria for sample size determination.

Sample size estimation is a major component of the design of virtually every experiment in medicine. Prudent use of the available prior information is a crucial element of experimental planning. Most sample size formulae in current use employ this information only in the form of point estimates, even though it is usually more accurately expressed as a distribution over a range of values. In this paper, we review several Bayesian and mixed Bayesian/likelihood approaches to sample size calculations based on lengths and coverages of posterior credible intervals. We apply these approaches to the design of an experiment to estimate the difference between two binomial proportions, and we compare results to those derived from standard formulae. Consideration of several criteria can contribute to selection of a final sample size.