Effectiveness of family meetings for family caregivers on delaying time to nursing home placement of dementia patients: a randomized trial

Background

Interventions relieving the burden of caregiving may postpone or prevent patient institutionalization. The objective of this study was to determine whether a family meetings intervention was superior to usual care in postponing nursing home placement of patients with dementia.

Methods

A randomized multicenter trial was conducted among 192 patients with a clinical diagnosis of dementia living at home at enrolment and their primary family caregiver. Dyads of caregivers and patients were randomized to the family meetings intervention (n = 96) or usual care (n = 96) condition. The intervention consisted of two individual sessions with the primary caregiver and four family counseling sessions that included family members and friends. The primary outcome measure was the time until institutionalization of the patient. Intention-to-treat as well as per protocol analyses were performed. Survival analyses were carried out to evaluate the effectiveness of the intervention.

Results

During 18 months follow-up 23 of 96 relatives with dementia of caregivers in the intervention group and 18 of 96 relatives with dementia of caregivers in the usual care group were institutionalized. No significant difference between the intervention and the usual care group was found in time until institutionalization (adjusted hazard ratio (HR) 1.46, 95% confidence interval (CI) 0.78 to 2.74). The per-protocol analysis revealed no significant effect either (adjusted HR 0.57, 95% CI 0.21 to 1.57), although the number of placements among the adherers was relatively low (9.4%). A subgroup effect was found for patients’ age, with a significantly higher risk of institutionalization for ‘younger’ patients in the intervention group compared with the usual care group (adjusted HR = 4.94, 95% CI 1.10 to 22.13).

Conclusion

This family meetings intervention for primary caregivers of patients with dementia did not postpone patient institutionalization more than usual care.

Trial Registration: Controlled-Trials.com ISRCTN90163486

     

Benchmarking protein classification algorithms via supervised cross-validation

Development and testing of protein classification algorithms are hampered by the fact that the protein universe is characterized by groups vastly different in the number of members, in average protein size, similarity within group, etc. Datasets based on traditional cross-validation (k-fold, leave-one-out, etc.) may not give reliable estimates on how an algorithm will generalize to novel, distantly related subtypes of the known protein classes. Supervised cross-validation, i.e., selection of test and train sets according to the known subtypes within a database has been successfully used earlier in conjunction with the SCOP database. Our goal was to extend this principle to other databases and to design standardized benchmark datasets for protein classification. Hierarchical classification trees of protein categories provide a simple and general framework for designing supervised cross-validation strategies for protein classification. Benchmark datasets can be designed at various levels of the concept hierarchy using a simple graph-theoretic distance. A combination of supervised and random sampling was selected to construct reduced size model datasets, suitable for algorithm comparison. Over 3000 new classification tasks were added to our recently established protein classification benchmark collection that currently includes protein sequence (including protein domains and entire proteins), protein structure and reading frame DNA sequence data. We carried out an extensive evaluation based on various machine-learning algorithms such as nearest neighbor, support vector machines, artificial neural networks, random forests and logistic regression, used in conjunction with comparison algorithms, BLAST, Smith-Waterman, Needleman-Wunsch, as well as 3D comparison methods DALI and PRIDE. The resulting datasets provide lower, and in our opinion more realistic estimates of the classifier performance than do random cross-validation schemes. A combination of supervised and random sampling was used to construct model datasets, suitable for algorithm comparison.

The datasets are available at http://hydra.icgeb.trieste.it/benchmark.     

The dcr gene family of Desulfovibrio: Implications from the sequence of dcrH and phylogenetic comparison with other mcp genes

Desulfovibrio vulgaris Hildenborough contains a family of genes for methyl-accepting chemotaxis proteins (MCPs). Here we report the complete sequence of the gene for Desulfovibrio chemoreceptor H (dcrH). The deduced amino acid sequence of DcrH protein, which has an enlarged N-terminal, ligand binding domain, indicates a structure similar to that of other MCPs. Comparison of the sequences for DcrA, determined earlier, and DcrH indicated that similarity is essentially limited to the C-terminal excitation region. The dcr gene family differs, in this respect, from mcp gene families in other eubacteria (e.g. Escherichia coli and Bacillus subtilis), where MCPs share significant homology throughout their C-terminal signal transduction domains. This may point to an ancient evolutionary origin of the dcr gene family, which is widely distributed throughout the genus Desulfovibrio. The evolutionary origin of mcp genes was traced by comparing nucleotide sequences for the excitation region that is common to all MCPs. Phylogenetic analysis of sequences for thirty mcp genes from nine eubacterial and one archaebacterial species suggested that multiplication of mcp genes has occurred at least twice since the eubacteria diverged from the archaebacteria.Nucleotide accession number: The nucleotide sequence reported in this paper has been entered into GenBank under accession number U30319. Phone: 403-220-6388. Fax: 403-289-9311. Electronic mail address: voordouw@acs.ucalgary.ca.     

Dark recovery of UV-irradiated phage T1 I. A minor recovery effect whose exclusion permits the study of survival kinetics under presumably repairless conditions

The survival of UV-irradiated phage T1 is much lower in excision repair-deficient than in excision repair-proficient E. coli cells, due to lack of “host cell reactivation” (HCR). An additional decrease in phage survival occurs when repair-deficient (HCR⁻) host cells have been exposed to UV doses from 3000–10 000 erg mm⁻² of 254 nm UV-radiation prior to infection. The observed effect is attributed to loss of a minor phage recovery process, which requires neither the bacterial excision repair nor the bacterial REC repair system. This type of recovery is little affected by caffeine or acriflavine at concentrations that preclude HCR completely. Its full inhibition by UV-irradiation of the cells requires an approximately 8 times larger dose than complete inhibition of HCR.

In heavily preirradiated cells, the T1 burst size is extremely small and multiplicity reactivation is considerably less extensive than in unirradiated cells. Presumably the survival of singly infecting T1 in these cells reflects absence of any type of repair. The observed phage sensitivity and shape of the curve are compatible with the expectation for completely repairless conditions. The mechanism underlying the minor recovery is not known; theoretical considerations make a phage REC repair mechanism seem likely.     

Herbivore trampling as an alternative pathway for explaining differences in nitrogen mineralization in moist grasslands

Studies addressing the role of large herbivores on nitrogen cycling in grasslands have suggested that the direction of effects depends on soil fertility. Via selection for high quality plant species and input of dung and urine, large herbivores have been shown to speed up nitrogen cycling in fertile grassland soils while slowing down nitrogen cycling in unfertile soils. However, recent studies show that large herbivores can reduce nitrogen mineralization in some temperate fertile soils, but not in others. To explain this, we hypothesize that large herbivores can reduce nitrogen mineralization in loamy or clay soils through soil compaction, but not in sandy soils. Especially under wet conditions, strong compaction in clay soils can lead to periods of soil anoxia, which reduces decomposition of soil organic matter and, hence, N mineralization. In this study, we use a long-term (37-year) field experiment on a salt marsh to investigate the hypothesis that the effect of large herbivores on nitrogen mineralization depends on soil texture. Our results confirm that the presence of large herbivores decreased nitrogen mineralization rate in a clay soil, but not in a sandy soil. By comparing a hand-mown treatment with a herbivore-grazed treatment, we show that these differences can be attributed to herbivore-induced changes in soil physical properties rather than to above-ground biomass removal. On clay soil, we find that large herbivores increase the soil water-filled porosity, induce more negative soil redox potentials, reduce soil macrofauna abundance, and reduce decomposition activity. On sandy soil, we observe no changes in these variables in response to grazing. We conclude that effects of large herbivores on nitrogen mineralization cannot be understood without taking soil texture, soil moisture, and feedbacks through soil macrofauna into account.     

Unravelling enzymatic discoloration in potato through a combined approach of candidate genes,QTL, and expression analysis

Enzymatic discoloration (ED) of potato tubers was investigated in an attempt to unravel the underlying genetic factors. Both enzyme and substrate concentration have been reported to influence the degree of discoloration and as such this trait can be regarded as polygenic. The diploid mapping population C × E, consisting of 249 individuals, was assayed for the degree of ED and levels of chlorogenic acid and tyrosine. Using this data, Quantitative Trait Locus (QTL) analysis was performed. Three QTLs for ED have been found on parental chromosomes C3, C8, E1, and E8. For chlorogenic acid a QTL has been identified on C2 and for tyrosine levels, a QTL has been detected on C8. None of the QTLs overlap, indicating the absence of genetic correlations between these components underlying ED, in contrast to earlier reports in literature. An obvious candidate gene for the QTL for ED on Chromosome 8 is polyphenol oxidase (PPO), which was previously mapped on chromosome 8. With gene-specific primers for PPO gene POT32 a CAPS marker was developed. Three different alleles (POT32-1, -2, and -3) could be discriminated. The segregating POT32 alleles were used to map the POT32 CAPS marker and QTL analysis was redone, showing that POT32 coincides with the QTL peak. A clear correlation between allele combinations and degree of discoloration was observed. In addition, analysis of POT32 gene expression in a subset of genotypes indicated a correlation between the level of gene expression and allele composition. On average, genotypes having two copies of allele 1 had both the highest degree of discoloration as well as the highest level of POT32 gene expression.     

A simple, robust and semi-automated parasite egg isolation protocol

Large-scale parasite quantification is required for improving our understanding of the epidemiology and genetics of host-parasite interactions. We describe a protocol that uses a low-density salt solution for flotation and centrifugation of nematode eggs. Subsequently, sucrose flotation and precipitation are used to obtain clear egg preparations. Most traditional quantification protocols such as the McMaster technique are unsuited for the standardized processing of large numbers of samples and the analysis of large amounts of feces per sample. Consequently, they are suited only for small-scale surveys. Our protocol, which can be used to analyze up to 6 g of feces, results in clear egg preparations that are concentrated in wells of a microtiter plate and that are suited for digital recording and automated counting. Starting from a fecal suspension in the first flotation solution to a digital recording requires approximately 40 min per 24 samples.