John E. Kolassa

Series approximation methods in statistics

This study covers the uses of series approximation techniques in statistics and is intended to give advanced graduate students in statistics an introduction to various expansions. Its aims are to provide a rigorous development of these expansions, with attention paid to regularity conditions and uniformity, and to survey applications of these series. Topics include characteristic functions and the Berry-Esseen Theorem, Edgeworth series, saddlepoint series for densities and distribution functions, multivariate expansions, conditional distribution approximations, and applications to likelihood ratio and maximum likelihood statistics. The use of a computerized algebra program for performing many of the calculations in this volume is demonstrated. The author presupposes a knowledge of undergraduate real and complex analysis, but otherwise the book is essentially self-contained. It is based on lecture courses given by the author, and will also be of use to researchers seeking a sophisticated account of an area of recent research activity.

Acute Respiratory Tract Infection in Daycare Centers for Older Persons

OBJECTIVE: To evaluate the rate of specific pathogens and clinical syndromes associated with acute respiratory tract infections (ARTI) in frail older persons attending daycare.

Carbamazepine treatment of agitation in nursing home patients with dementia: a preliminary study.

OBJECTIVE: To determine the effects of carbamazepine versus placebo on ratings of behavior in agitated nursing home patients with dementia.

Primary prevention of childhood lead exposure: A randomized trial of dust control.

Background. Dust control is recommended as one of the primary strategies to prevent or control childrens exposure to residential lead hazards, but the effect of dust control on childrens blood lead levels is poorly understood. Objective. To determine the effectiveness of dust control in preventing childrens exposure to lead, as measured by blood lead levels, during their peak age of susceptibility. Design. A randomized, controlled trial. Setting. Rochester, NY. Participants. A total of 275 urban children were randomized at 6 months of age, of whom 246 (90%) were available for the 24-month-old follow-up visit. Interventions. Children and their families were randomly assigned to an intervention group (n = 140), which received cleaning equipment and up to eight visits by a dust control advisor, or a control group (n = 135). Outcome Measures. Geometric mean blood lead levels and prevalence of elevated blood lead levels (ie, >10 μg/dL, 15 μg/dL, and 20 μg/dL). Results. At baseline, childrens geometric mean blood lead levels were 2.9 μg/dL (95% confidence interval [CI] = 2.7, 3.1); there were no significant differences in characteristics or lead exposure by group assignment, with the exception of water lead levels. For children in the intervention group, the mean number of visits by a dust control advisor during the 18-month study period was 6.2; 51 (36%) had 4 to 7 visits, and 69 (49%) had 8 visits. At 24 months of age, the geometric mean blood lead was 7.3 μg/dL (95% CI = 6.6, 8.2) for the intervention group and 7.8 μg/dL (95% CI = 6.9, 8.7) for the control group. The percentage of children with a 24-month blood lead ≥10 μg/dL, ≥15 μg/dL, and ≥20 μg/dL was 31% versus 36%, 12% versus 14%, and 5% versus 7% in the intervention and control groups, respectively. Conclusions. We conclude that dust control, as performed by families and in the absence of lead hazard controls to reduce ongoing contamination from lead-based paint, is not effective in the primary prevention of childhood lead exposure. blood lead, lead-contaminated house dust, randomized trial, children, environmental exposure, lead poisoning, primary prevention, prevention.

The "common cold" in frail older persons: impact of rhinovirus and coronavirus in a senior daycare center.

OBJECTIVE: To evaluate the incidence and impact of rhino‐virus and Coronavirus infections in older persons attending daycare.

Evaluation of a handwashing intervention to reduce respiratory illness rates in senior day-care centers

To decrease respiratory infections in senior day care, staff were educated on viral transmission and the value of hand washing. Fanny packs with alcohol foam supplemented hand washing and were alternated monthly between centers. Infection rates were unchanged with alcohol foam use. The intervention years infection rate was significantly lower than the previous 3 years, suggesting a benefit of education.

Hepatitis B immunization of healthy elderly adults: relationship between naïve CD4+ T cells and primary immune response and evaluation of GM-CSF as an adjuvant.

The efficacy of granulocyte–macrophage colony-stimulating factor (GM-CSF) to enhance the primary immune response to hepatitis B vaccine was studied in healthy elderly with young volunteers included as controls in this double-blind, placebo-controlled trial of GM-CSF as an immune adjuvant. Na¨ıve T-helper cells (CD4+CD45RA+) were determined at baseline. Forty-five healthy elderly (average age, 74 years) and 37 healthy young controls (average age, 28 years) were randomized. Hepatitis B vaccine was administered at 0, 1, and 6 months. GM-CSF as a single injection of either 80 μg or 250 μg with the first and second doses of hepatitis B vaccine. In this trial GM-CSF did not enhance antibody responses. However, the antibody responses were dramatically different between these two groups: 35/35 young developed a protective titer versus 19/45 elderly (P < 0.0001). In addition, the mean logarithm of anti-hepatitis B antibody level in the 35 young who completed the study was 3.17 (log mIU/ml) but only 2.21 in the 19 elderly responders (P < 0.0001). Na¨ıve T-helper cells differed significantly between the two groups: the mean percentage of CD4+CD45RA+ T cells was 47.9% versus 35.0% (P < 0.0001) in the young and elderly volunteers respectively. Na¨ıve T cells also differed significantly between elderly who did or did not respond to HBV (39.9% vs. 31.7%, P = 0.039). Using linear regression, age, and percent na¨ıve, CD4 T cells were determined to significantly influence the anti-hepatitis B antibody response, but sex and dose of GM-CSF did not. For a two-parameter model: logarithm of antibody titer = (−0.038 × age in years) + (0.031 × % na¨ıve CD4T cells) + 2.68; adjusted r2 = 0.605 and P < 0.0001. However, age had a larger effect than na¨ıve CD4 T cells, i.e., in comparing young and elderly groups the log antibody titer decreased by 1.73 due to the increase in age but only 0.40 due to the decrease in na¨ıve CD4 T cells. Thus, there was a large effect of age that could not be explained by the quantitative change in the na¨ıve T-helper cells.

Medical Practice with Nursing Home Residents: Results from the National Physician Professional Activities Census

OBJECTIVE: The study describes the prevalence of medical nursing home practice. Further, it examines the extent to which physician characteristics and local county health care resources predict nursing home involvement. This information is relevant to evaluating and devising strategies that address the future provision of medical care in institutionalized long‐term care.

Approximate Conditional Inference in Exponential Families via the Gibbs Sampler

Abstract This article presents the Gibbs-Skovgaard algorithm for approximate frequentist inference. The method makes use of the double saddlepoint approximation of Skovgaard to the conditional cumulative distribution function of a sufficient statistic given the remaining sufficient statistics. This approximation is then used in the Gibbs sampler to generate a Markov chain. The equilibrium distribution of this chain approximates the joint distribution of the sufficient statistics associated with the parameters of interest conditional on the observed values of the sufficient statistics associated with the nuisance parameters. This Gibbs-Skovgaard algorithm is applied to the cases of logistic and Poisson regression.

Response of C-Reactive Protein and Serum Amyloid A to Influenza A Infection in Older Adults

Influenza epidemics are associated with significant morbidity and mortality in the elderly, with a substantial proportion of deaths due to cardiovascular events. Elevations of acute-phase proteins have been associated with an increased risk of atherosclerotic events. Therefore, serum amyloid A (SAA) and C-reactive protein (CRP) were measured during influenza illness and 4 weeks later in 7 young persons, 15 elderly outpatients, and 36 hospitalized adults. Striking elevations were seen in mean acute SAA and CRP levels in all groups, but hospitalized patients had the highest levels (SAA, 503 vs. 310 microg/mL [P=.006]; CRP, 120 vs. 34 microg/mL [P<.001]). The presence of dyspnea, wheezing, and fever was also associated with high CRP levels. Influenza infection is associated with significant elevations of SAA and CRP levels in elderly patients, especially those who require hospitalization. It is possible that direct effects of CRP may exacerbate preexisting atherosclerotic lesions and may help explain cardiovascular events associated with acute influenza.