Estimation of Parameters in Random Effect Models with Incidence Matrix Uncertainty

Random effect models have been widely applied in many fields of research. However, models with uncertain design matrices for random effects have been little investigated before. In some applications with such problems, an expectation method has been used for simplicity. This method does not include the extra information of uncertainty in the design matrix is not included. The closed solution for this problem is generally difficult to attain. We therefore propose an two-step algorithm for estimating the parameters, especially the variance components in the model. The implementation is based on Monte Carlo approximation and a Newton-Raphson-based EM algorithm. As an example, a simulated genetics dataset was analyzed. The results showed that the proportion of the total variance explained by the random effects was accurately estimated, which was highly underestimated by the expectation method. By introducing heuristic search and optimization methods, the algorithm can possibly be developed to infer the 'model-based' best design matrix and the corresponding best estimates.

Association between levels of pentraxin 3 and incidence of chronic kidney disease in the elderly

OBJECTIVE: Higher levels of the novel inflammatory marker pentraxin 3 (PTX3) predict cardiovascular mortality in patients with chronic kidney disease (CKD). Yet, whether PTX3 predicts worsening of kidney function has been less well studied. We therefore investigated the associations between PTX3 levels, kidney disease measures and CKD incidence. METHODS: Cross-sectional associations between serum PTX3 levels, urinary albumin/creatinine ratio (ACR) and cystatin C-estimated glomerular filtration rate (GFR) were assessed in two independent community-based cohorts of elderly subjects: the Prospective Investigation of the Vasculature in Uppsala Seniors (PIVUS, n = 768, 51% women, mean age 75 years) and the Uppsala Longitudinal Study of Adult Men (ULSAM, n = 651, mean age 77 years). The longitudinal association between PTX3 level at baseline and incident CKD (GFR <60 mL( ) min(-1)  1.73 m(-) ²) was also analysed (number of events/number at risk: PIVUS 229/746, ULSAM 206/315). RESULTS: PTX3 levels were inversely associated with GFR [PIVUS: B-coefficient per 1 SD increase -0.16, 95% confidence interval (CI) -0.23 to -0.10, P < 0.001; ULSAM: B-coefficient per 1 SD increase -0.09, 95% CI -0.16 to -0.01, P < 0.05], but not ACR, after adjusting for age, gender, C-reactive protein and prevalent cardiovascular disease in cross-sectional analyses. In longitudinal analyses, PTX3 levels predicted incident CKD after 5 years in both cohorts [PIVUS: multivariable odds ratio (OR) 1.21, 95% CI 1.01-1.45, P < 0.05; ULSAM: multivariable OR 1.37, 95% CI 1.07-1.77, P < 0.05]. CONCLUSIONS: Higher PTX3 levels are associated with lower GFR and independently predict incident CKD in elderly men and women. Our data confirm and extend previous evidence suggesting that inflammatory processes are activated in the early stages of CKD and drive impairment of kidney function. Circulating PTX3 appears to be a promising biomarker of kidney disease.