Refine
Year of publication
Document Type
- Article (12)
- Postprint (3)
- Review (3)
- Working Paper (1)
Is part of the Bibliography
- yes (19)
Keywords
- meta-analysis (19) (remove)
Institute
- Department Sport- und Gesundheitswissenschaften (4)
- Institut für Biochemie und Biologie (4)
- Department Psychologie (3)
- Strukturbereich Kognitionswissenschaften (3)
- Institut für Ernährungswissenschaft (2)
- Patholinguistics/Neurocognition of Language (2)
- Center for Economic Policy Analysis (CEPA) (1)
- Extern (1)
The large literature that aims to find evidence of climate migration delivers mixed findings. This meta-regression analysis i) summarizes direct links between adverse climatic events and migration, ii) maps patterns of climate migration, and iii) explains the variation in outcomes. Using a set of limited dependent variable models, we meta-analyze thus-far the most comprehensive sample of 3,625 estimates from 116 original studies and produce novel insights on climate migration. We find that extremely high temperatures and drying conditions increase migration. We do not find a significant effect of sudden-onset events. Climate migration is most likely to emerge due to contemporaneous events, to originate in rural areas and to take place in middle-income countries, internally, to cities. The likelihood to become trapped in affected areas is higher for women and in low-income countries, particularly in Africa. We uniquely quantify how pitfalls typical for the broader empirical climate impact literature affect climate migration findings. We also find evidence of different publication biases.
Research synthesis on simple yet general hypotheses and ideas is challenging in scientific disciplines studying highly context-dependent systems such as medical, social, and biological sciences. This study shows that machine learning, equation-free statistical modeling of artificial intelligence, is a promising synthesis tool for discovering novel patterns and the source of controversy in a general hypothesis. We apply a decision tree algorithm, assuming that evidence from various contexts can be adequately integrated in a hierarchically nested structure. As a case study, we analyzed 163 articles that studied a prominent hypothesis in invasion biology, the enemy release hypothesis. We explored if any of the nine attributes that classify each study can differentiate conclusions as classification problem. Results corroborated that machine learning can be useful for research synthesis, as the algorithm could detect patterns that had been already focused in previous narrative reviews. Compared with the previous synthesis study that assessed the same evidence collection based on experts' judgement, the algorithm has newly proposed that the studies focusing on Asian regions mostly supported the hypothesis, suggesting that more detailed investigations in these regions can enhance our understanding of the hypothesis. We suggest that machine learning algorithms can be a promising synthesis tool especially where studies (a) reformulate a general hypothesis from different perspectives, (b) use different methods or variables, or (c) report insufficient information for conducting meta-analyses.
This study examined the effectiveness of psychological interventions for severe health anxiety (SHA) regarding somatic symptoms (SS) and health anxiety (HA). The databases Web of Science, EBSCO, and CENTRAL were searched on May 15, 2019, May 16, 2019, and August 5, 2019, respectively. Eighteen randomized controlled trials (N = 2,050) met the inclusion criteria (i.e., hypochondriasis, illness anxiety disorder or somatic symptom disorder with elevated HA being assessed with validated interviews: use of standardized outcome measures). Two reviewers independently evaluated the studies' risk of bias using the Revised Cochrane Risk-of-Bias Tool for randomized trials (RoB-2) tool. Overall, psychological interventions were significantly more effective than waitlist, treatment-as-usual, or placebo post-treatment (g(SS) = 0.70, g(HA) = 1.11) and at follow-up (g(SS) = 0.33, g(HA)= 0.70). CBT outperformed other psychological interventions or pharmacotherapy for HA post- treatment (Hedge's g(HA) = 0.81). The number of sessions did not significantly predict the effect sizes. In sum, psychological interventions were effective for SHA, but the generalizability of the results for SS is limited, because only two high-quatity trials contributed to the comparison.
Research synthesis on simple yet general hypotheses and ideas is challenging in scientific disciplines studying highly context-dependent systems such as medical, social, and biological sciences. This study shows that machine learning, equation-free statistical modeling of artificial intelligence, is a promising synthesis tool for discovering novel patterns and the source of controversy in a general hypothesis. We apply a decision tree algorithm, assuming that evidence from various contexts can be adequately integrated in a hierarchically nested structure. As a case study, we analyzed 163 articles that studied a prominent hypothesis in invasion biology, the enemy release hypothesis. We explored if any of the nine attributes that classify each study can differentiate conclusions as classification problem. Results corroborated that machine learning can be useful for research synthesis, as the algorithm could detect patterns that had been already focused in previous narrative reviews. Compared with the previous synthesis study that assessed the same evidence collection based on experts' judgement, the algorithm has newly proposed that the studies focusing on Asian regions mostly supported the hypothesis, suggesting that more detailed investigations in these regions can enhance our understanding of the hypothesis. We suggest that machine learning algorithms can be a promising synthesis tool especially where studies (a) reformulate a general hypothesis from different perspectives, (b) use different methods or variables, or (c) report insufficient information for conducting meta-analyses.
This systematic review investigated how successful children/adolescents with poor literacy skills learn a foreign language compared with their peers with typical literacy skills. Moreover, we explored whether specific characteristics related to participants, foreign language instruction, and assessment moderated scores on foreign language tests in this population. Overall, 16 studies with a total of 968 participants (poor reader/spellers:n = 404; control participants:n = 564) met eligibility criteria. Only studies focusing on English as a foreign language were available. Available data allowed for meta-analyses on 10 different measures of foreign language attainment. In addition to standard mean differences (SMDs), we computed natural logarithms of the ratio of coefficients of variation (CVRs) to capture individual variability between participant groups. Significant between-study heterogeneity, which could not be explained by moderator analyses, limited the interpretation of results. Although children/adolescents with poor literacy skills on average showed lower scores on foreign language phonological awareness, letter knowledge, and reading comprehension measures, their performance varied significantly more than that of control participants. Thus, it remains unclear to what extent group differences between the foreign language scores of children/adolescents with poor and typical literacy skills are representative of individual poor readers/spellers. Taken together, our results indicate that foreign language skills in children/adolescents with poor literacy skills are highly variable. We discuss the limitations of past research that can guide future steps toward a better understanding of individual differences in foreign language attainment of children/adolescents with poor literacy skills.
This systematic review investigated how successful children/adolescents with poor literacy skills learn a foreign language compared with their peers with typical literacy skills. Moreover, we explored whether specific characteristics related to participants, foreign language instruction, and assessment moderated scores on foreign language tests in this population. Overall, 16 studies with a total of 968 participants (poor reader/spellers:n = 404; control participants:n = 564) met eligibility criteria. Only studies focusing on English as a foreign language were available. Available data allowed for meta-analyses on 10 different measures of foreign language attainment. In addition to standard mean differences (SMDs), we computed natural logarithms of the ratio of coefficients of variation (CVRs) to capture individual variability between participant groups. Significant between-study heterogeneity, which could not be explained by moderator analyses, limited the interpretation of results. Although children/adolescents with poor literacy skills on average showed lower scores on foreign language phonological awareness, letter knowledge, and reading comprehension measures, their performance varied significantly more than that of control participants. Thus, it remains unclear to what extent group differences between the foreign language scores of children/adolescents with poor and typical literacy skills are representative of individual poor readers/spellers. Taken together, our results indicate that foreign language skills in children/adolescents with poor literacy skills are highly variable. We discuss the limitations of past research that can guide future steps toward a better understanding of individual differences in foreign language attainment of children/adolescents with poor literacy skills.
Questions Has plant species richness in semi-natural grasslands changed over recent decades? Do the temporal trends of habitat specialists differ from those of habitat generalists? Has there been a homogenization of the grassland vegetation? Location Different regions in Germany and the UK. Methods We conducted a formal meta-analysis of re-survey vegetation studies of semi-natural grasslands. In total, 23 data sets were compiled, spanning up to 75 years between the surveys, including 13 data sets from wet grasslands, six from dry grasslands and four from other grassland types. Edaphic conditions were assessed using mean Ellenberg indicator values for soil moisture, nitrogen and pH. Changes in species richness and environmental variables were evaluated using response ratios. Results In most wet grasslands, total species richness declined over time, while habitat specialists almost completely vanished. The number of species losses increased with increasing time between the surveys and were associated with a strong decrease in soil moisture and higher soil nutrient contents. Wet grasslands in nature reserves showed no such changes or even opposite trends. In dry grasslands and other grassland types, total species richness did not consistently change, but the number or proportions of habitat specialists declined. There were also considerable changes in species composition, especially in wet grasslands that often have been converted into intensively managed, highly productive meadows or pastures. We did not find a general homogenization of the vegetation in any of the grassland types. Conclusions The results document the widespread deterioration of semi-natural grasslands, especially of those types that can easily be transformed to high production grasslands. The main causes for the loss of grassland specialists are changed management in combination with increased fertilization and nitrogen deposition. Dry grasslands are most resistant to change, but also show a long-term trend towards an increase in more mesotrophic species.
We evaluated the effectiveness and acceptability of metacognitive interventions for mental disorders. We searched electronic databases and included randomized and nonrandomized controlled trials comparing metacognitive interventions with other treatments in adults with mental disorders. Primary effectiveness and acceptability outcomes were symptom severity and dropout, respectively. We performed random-effects meta-analyses. We identified Metacognitive Training (MCTrain), Metacognitive Therapy (MCTherap), and Metacognition Reflection and Insight Therapy (MERIT). We included 49 trials with 2,609 patients. In patients with schizophrenia, MCTrain was more effective than a psychological treatment (cognitive remediation, SMD = -0.39). It bordered significance when compared with standard or other psychological treatments. In a post hoc analysis, across all studies, the pooled effect was significant (SMD = -0.31). MCTrain was more effective than standard treatment in patients with obsessive-compulsive disorder (SMD = -0.40). MCTherap was more effective than a waitlist in patients with depression (SMD = -2.80), posttraumatic stress disorder (SMD = -2.36), and psychological treatments (cognitive-behavioural) in patients with anxiety (SMD = -0.46). In patients with depression, MCTherap was not superior to psychological treatment (cognitive-behavioural). For MERIT, the database was too small to allow solid conclusions. Acceptability of metacognitive interventions among patients was high on average. Methodological quality was mostly unclear or moderate. Metacognitive interventions are likely to be effective in alleviating symptom severity in mental disorders. Although their add-on value against existing psychological interventions awaits to be established, potential advantages are their low threshold and economy.
Physical activity protects from incident anxiety: A meta-analysis of prospective cohort studies
(2019)
Background Prospective cohorts have suggested that physical activity (PA) can decrease the risk of incident anxiety. However, no meta-analysis has been conducted. Aims To examine the prospective relationship between PA and incident anxiety and explore potential moderators. Methods Searches were conducted on major databases from inception to October 10, 2018 for prospective studies (at least 1 year of follow-up) that calculated the odds ratio (OR) of incident anxiety in people with high PA against people with low PA. Methodological quality was assessed using the Newcastle-Ottawa Scale (NOS). A random-effects meta-analysis was conducted and heterogeneity was explored using subgroup and meta-regression analysis. Results Across 14 cohorts of 13 unique prospective studies (N = 75,831, median males = 50.1%) followed for 357,424 person-years, people with high self-reported PA (versus low PA) were at reduced odds of developing anxiety (adjusted odds ratio [AOR] = 0.74; 95% confidence level [95% CI] = 0.62, 0.88; crude OR = 0.80; 95% CI = 0.69, 0.92). High self-reported PA was protective against the emergence of agoraphobia (AOR = 0.42; 95% CI = 0.18, 0.98) and posttraumatic stress disorder (AOR = 0.57; 95% CI = 0.39, 0.85). The protective effects for anxiety were evident in Asia (AOR = 0.31; 95% CI = 0.10, 0.96) and Europe (AOR = 0.82; 95% CI = 0.69, 0.97); for children/adolescents (AOR = 0.52; 95% CI = 0.29, 0.90) and adults (AOR = 0.81; 95% CI = 0.69, 0.95). Results remained robust when adjusting for confounding factors. Overall study quality was moderate to high (mean NOS = 6.7 out of 9). Conclusion Evidence supports the notion that self-reported PA can confer protection against the emergence of anxiety regardless of demographic factors. In particular, higher PA levels protects from agoraphobia and posttraumatic disorder.
Background: Population-specificity of exploratory dietary patterns limits their generalizability in investigations with type 2 diabetes incidence.
Objective: The aim of this study was to derive country-specific exploratory dietary patterns, investigate their association with type 2 diabetes incidence, and replicate diabetes-associated dietary patterns in other countries.
Methods: Dietary intake data were used, assessed by country-specific questionnaires at baseline of 11,183 incident diabetes cases and 14,694 subcohort members (mean age 52.9 y) from 8 countries, nested within the European Prospective Investigation into Cancer and Nutrition study (mean follow-up time 6.9 y). Exploratory dietary patterns were derived by principal component analysis. HRs for incident type 2 diabetes were calculated by Prentice-weighted Cox proportional hazard regression models. Diabetes-associated dietary patterns were simplified or replicated to be applicable in other countries. A meta-analysis across all countries evaluated the generalizability of the diabetes-association.
Results: Two dietary patterns per country/UK-center, of which overall 3 dietary patterns were diabetes-associated, were identified. A risk-lowering French dietary pattern was not confirmed across other countries: pooled HRFrance per 1 SD: 1.00; 95% CI: 0.90, 1.10. Risk-increasing dietary patterns, derived in Spain and UK-Norfolk, were confirmed, but only the latter statistically significantly: HRSpain: 1.09; 95% CI: 0.97, 1.22 and HRUK-Norfolk: 1.12; 95% CI: 1.04, 1.20. Respectively, this dietary pattern was characterized by relatively high intakes of potatoes, processed meat, vegetable oils, sugar, cake and cookies, and tea. Conclusions: Only few country/center-specific dietary patterns (3 of 18) were statistically significantly associated with diabetes incidence in this multicountry European study population. One pattern, whose association with diabetes was confirmed across other countries, showed overlaps in the food groups potatoes and processed meat with identified diabetes-associated dietary patterns from other studies. The study demonstrates that replication of associations of exploratory patterns with health outcomes is feasible and a necessary step to overcome population-specificity in associations from such analyses.