Search

Objectives: Longitudinal mediation effects can be estimated with mixed effects models. Mixed effects models are versatile, as they accommodate the estimation of contemporaneous, lagged, time-independent, and time-dependent effects. However, the inclusion of time lags and time interactions in mixed effects models for longitudinal mediation analysis has received little attention. This article demonstrates how time lags and time interactions in mixed effects models affect the interpretation of longitudinal mediation effect estimates. Study Design and Setting: We used a data example from the Amsterdam Growth and Health Longitudinal Study to illustrate how the inclusion of time lags and time interactions in mixed effects models may affect the size and interpretation of longitudinal mediation effect estimates. Results: The chosen time lags between the determinant, mediator, and outcome influenced the size and interpretation of the mediation effect estimates. Furthermore, time interactions can be used to model linear or nonlinear development of the mediation effects over time. Conclusion: The inclusion of time lags and time interactions should be considered when estimating longitudinal mediation effects based on mixed effects models, as this enables the estimation of lagged and time-dependent effects.

Error in Table In the original publication, there was a mistake in Table 1 as published [1]. The number of patients with death was swapped between the one-hour administration group and the push administration group. The corrected Table 1 appears below. The authors state that the scientific conclusions are unaffected. This correction was approved by the Academic Editor. The original publication has also been updated.

In the original publication, study group author name “on behalf of Netherlands Brain Bank” was mistakenly included in the author name and it should be omitted. The original article has been corrected.

Additional file 1.

Additional file 6. Observer OPTIONMCC scores with and without use of patient preparatory tool (intervention group).

Additional file 4. Baseline characteristics informal caregivers.

Additional file 4.

Background: White matter (WM) lesions and brain atrophy are present early in multiple sclerosis (MS). However, their spatio-temporal relationship remains unclear. Methods: Yearly magnetic resonance images were analysed in 387 patients with a first clinical demyelinating event (FCDE) from the 5-year REFLEXION study. Patients received early (from baseline; N = 258; ET) or delayed treatment (from month-24; N = 129; DT) with subcutaneous interferon beta-1a. FSL-SIENA/VIENA were used to provide yearly percentage volume change of brain (PBVC) and ventricles (PVVC). Yearly total lesion volume change (TLVC) was determined by a semi-automated method. Using linear mixed models and voxel-wise analyses, we firstly investigated the overall relationship between TLVC and PBVC and between TLVC and PVVC in the same follow-up period. Analyses were then separately performed for: the untreated period of DT patients (first two years), the first year of treatment (year 1 for ET and year 3 for DT), and a period where patients had received at least 1 year of treatment (stable treatment; ET: years 2, 3, 4, and 5; DT: years 4 and 5). Results: Whole brain: across the whole study period, lower TLVC was related to faster atrophy (PBVC: B = 0.046, SE = 0.013, p < 0.001; PVVC: B = −0.466, SE = 0.118, p < 0.001). Within the untreated period of DT patients, lower TLVC was related to faster atrophy (PBVC: B = 0.072, SE = 0.029, p = 0.013; PVVC: B = −0.917, SE = 0.306, p = 0.003). A similar relationship was found within the first year of treatment of ET patients (PBVC: B = 0.081, SE = 0.027, p = 0.003; PVVC: B = −1.08, SE = 0.284, p < 0.001), consistent with resolving oedema and pseudo-atrophy. Voxel-wise: overall, higher TLVC was related to faster ventricular enlargement. Lower TLVC was related to faster widespread atrophy in year 1 in both ET (first year of treatment) and DT (untreated) patients. In the second untreated year of DT patients and within the stable treatment period of ET patients (year 4), faster periventricular and occipital lobe atrophy was associated with higher TLVC. Conclusions: WM lesion changes and atrophy occurred simultaneously in early MS. Spatio-temporal correspondence of these two processes involved mostly the periventricular area. Within the first year of the study, in both treatment groups, faster atrophy was linked to lower lesion volume changes, consistent with higher shrinking and disappearing lesion activity. This might reflect the pseudo-atrophy phenomenon that is probably related to the therapy driven (only in ET patients, as they received treatment from baseline) and “natural” (both ET and DT patients entered the study after a FCDE) resolution of oedema. In an untreated period and later on during stable treatment, (real) atrophy was related to higher lesion volume changes, consistent with increased new and enlarging lesion activity.

Additional file 7. Observer OPTIONMCC scores for geriatricians, patients and informal caregivers (subgroup analysis of eight geriatricians).

Additional file 3. Adapted control preference scale.

Additional file 1. Patient preparatory tool.

Additional file 5. Evaluation of use of the patient preparatory tool in the intervention group (n = 108).

Additional file 3.

Objectives: The researchers aimed to (1) explore the occurrence of psychological resilience in the face of a major life stressor and conflict in older residents of long-term care facilities (LTCFs), and (2) identify factors associated with resilience in this population. Design: Longitudinal cohort study using the Dutch InterRAI-LTCF cohort. Setting and participants: Older residents (≥60 years old) of 21 LTCFs in the Netherlands. Methods: The researchers selected 2 samples of residents who had at least 2 assessments surrounding (1) an incident major life stressor, or (2) incident conflict with other resident or staff. A resilient outcome was operationalized as not having clinically meaningful mood symptoms at the post-stressor assessment and equal or fewer mood symptoms at the post-stressor relative to the pre-stressor assessment. The researchers used 2 resilience outcomes per stressor: 1 based on observer-reported mood symptoms and 1 based on self-reported mood symptoms. The most important factors from among 21 potential resilience factors for each of the 4 operationalizations of resilience were identified using a backward selection procedure with 2-level generalized estimating equations analyses. Results: Forty-eight percent and 50% of residents were resilient in the face of a major life stressor, based on observer-reported (n = 248) and self-reported (n = 211) mood, respectively. In the face of conflict, 26% and 51% of the residents demonstrated resilience, based on the observer-reported (n = 246) and self-reported (n = 183) mood, respectively. Better cognitive functioning, a strong and supportive relationship with family, participation in social activities, and better self-reported health were most strongly associated with resilience in the face of a major life stressor. Better communicative functioning, absence of psychiatric diagnoses, a strong and supportive relationship with family, not being lonely, social engagement, and not reminiscing about life were most strongly associated with resilience in the face of conflict. Conclusions and Implications: Factors with a social aspect appear to be particularly important to psychological resilience in older LTCF residents, and provide a potential target for intervention in the LTCF setting.

Additional file 5.

Additional file 2.