ML and REML estimation in survival analysis with time dependent correlated frailty

✦ Synopsis

In the study of multiple failure times for the same subjects, for example, recurrent infections for patients with a given disease, there are often subject effects, that is, subjects have different risks that cannot be explained by known covariates. Standard methods, which ignore subject effects, lead to overestimation of precision. The frailty model for subject effects is better, but can be insufficient, because it assumes that subject effects are constant over time. Experience has shown that the dependence between different time periods often decreases with distance in time. Such a model is presented here, assuming that the frailty is no longer constant, but time varying, with one value for each spell. The main example is a first-order autoregressive process. This is applied to a data set of 128 patients with chronic granulomatous disease (CGD), participating in a placebo controlled randomized trial of gamma interferon ( -IFN), suffering between 0 and 7 infections. It is shown that the time varying frailty model gives a significantly better fit than the constant frailty model.

Saliva and serum samples were collected

Saliva and serum samples were collected from eight healthy volunteers every two hours during a 26-hour period. Melatonin concentrations were measured by radioimmunoassay after chloroform extraction using radioiodinated melatonin as a tracer. Five of the subjects had high serum melatonin levels at night (peak levels higher than 75 pg/ml); in three subjects the highest serum melatonin concentration was 20-40 pg/ml. All subjects had low levels (P <0.001, was obtained for all detectable value pairs (n= 73). The regression and correlation coefficients were almost equal for the peak values of melatonin and during the rising and descending phases of the secretion patterns. However, no significant correlation was found between low daytime salivary and serum concentrations when calculated separately. In the five high-secretors the melatonin levels in saliva reflected reliably the changes in serum, but in the three low-secretors the correlation between salivary and serum melatonin was not significant. The proportion of melatonin found in saliva decreased with increasing serum melatonin levels. Circadian rhythm parameters were estimated by single cosinor analysis. The acrophases did not differ significantly within a subject in the concomitant measurements of serum and salivary melatonin. The measurements of salivary melatonin levels seem valid for studies on melatonin rhythms, but the melatonin concentrations measured in saliva do not always consistently reflect the absolute concentrations in blood.

✍ Maija-Liisa Laakso; Tarja Porkka-Heiskanen; Aino Alila; Dag Stenberg; Gunnar Joh 📂 Article 📅 1990 🏛 John Wiley and Sons 🌐 English ⚖ 622 KB