Meta-analysis of individual patient data from randomised trials: a review of methods used in practice

Background: Meta-analyses based on individual patient data (IPD) are regarded as the gold standard for systematic reviews. However, the methods used for analysing and presenting results from IPD meta-analyses have received little discussion. Methods We review 44 IPD meta-analyses published during the years 1999–2001. We summarize whether they obtained all the data they sought, what types of approaches were used in the analysis, including assumptions of common or random effects, and how they examined the effects of covariates. Results: Twenty-four out of 44 analyses focused on time-to-event outcomes, and most analyses (28) estimated treatment effects within each trial and then combined the results assuming a common treatment effect across trials. Three analyses failed to stratify by trial, analysing the data is if they came from a single mega-trial. Only nine analyses used random effects methods. Covariate-treatment interactions were generally investigated by subgrouping patients. Seven of the meta-analyses included data from less than 80% of the randomized patients sought, but did not address the resulting potential biases. Conclusions: Although IPD meta-analyses have many advantages in assessing the effects of health care, there are several aspects that could be further developed to make fuller use of the potential of these time-consuming projects. In particular, IPD could be used to more fully investigate the influence of covariates on heterogeneity of treatment effects, both within and between trials. The impact of heterogeneity, or use of random effects, are seldom discussed. There is thus considerable scope for enhancing the methods of analysis and presentation of IPD meta-analysis.

Composition differences between organic and conventional meat: a systematic literature review and meta-analysis

Demand for organic meat is partially driven by consumer perceptions that organic foods are more nutritious than non-organic foods. However, there have been no systematic reviews comparing specifically the nutrient content of organic and conventionally produced meat. In this study, we report results of a meta-analysis based on sixty-seven published studies comparing the composition of organic and non-organic meat products. For many nutritionally relevant compounds (e.g. minerals, antioxidants and most individual fatty acids (FA)), the evidence base was too weak for meaningful meta-analyses. However, significant differences in FA profiles were detected when data from all livestock species were pooled. Concentrations of SFA and MUFA were similar or slightly lower, respectively, in organic compared with conventional meat. Larger differences were detected for total PUFA and n-3 PUFA, which were an estimated 23 (95 % CI 11, 35) % and 47 (95 % CI 10, 84) % higher in organic meat, respectively. However, for these and many other composition parameters, for which meta-analyses found significant differences, heterogeneity was high, and this could be explained by differences between animal species/meat types. Evidence from controlled experimental studies indicates that the high grazing/forage-based diets prescribed under organic farming standards may be the main reason for differences in FA profiles. Further studies are required to enable meta-analyses for a wider range of parameters (e.g. antioxidant, vitamin and mineral concentrations) and to improve both precision and consistency of results for FA profiles for all species. Potential impacts of composition differences on human health are discussed.

A preliminary meta-analysis of SME ebusiness journal publications: current trends and future research opportunities moving beyond adoption factor studies

Previous meta-analyses of SME-eBusiness journal research focuses on analysing adoption factors, pre-2000 articles and a small number of journals. This paper departs from this research by analysing 100 articles published between 2003 and 2006 in 41 journals on the basis of the research approaches employed, countries and eBusiness technologies studied, and research objectives focused upon. The paper presents preliminary insights into current major research trends based on this analysis, such as the predominant focus on adoption factor by many studies. It also identifies future research opportunities, and proposes a research agenda which aims to progress SME-eBusiness research beyond adoption factor studies by outlining research objectives to help SMEs overcome barriers and exploit drivers.

Physical activity for cancer survivors : a meta-analysis of randomised controlled trials

Objective To systematically evaluate the effects of physical activity in adult patients after completion of main treatment related to cancer. Design Meta-analysis of randomised controlled trials with data extraction and quality assessment performed independently by two researchers. Data sources Pubmed, CINAHL, and Google Scholar from the earliest possible year to September 2011. References from meta-analyses and reviews. Study selection Randomised controlled trials that assessed the effects of physical activity in adults who had completed their main cancer treatment, except hormonal treatment. Results There were 34 randomised controlled trials, of which 22 (65%) focused on patients with breast cancer, and 48 outcomes in our meta-analysis. Twenty two studies assessed aerobic exercise, and four also included resistance or strength training. The median duration of physical activity was 13 weeks (range 3-60 weeks). Most control groups were considered sedentary or were assigned no exercise. Based on studies on patients with breast cancer, physical activity was associated with improvements in insulin-like growth factor-I, bench press, leg press, fatigue, depression, and quality of life. When we combined studies on different types of cancer, we found significant improvements in body mass index (BMI), body weight, peak oxygen consumption, peak power output, distance walked in six minutes, right handgrip strength, and quality of life. Sources of study heterogeneity included age, study quality, study size, and type and duration of physical activity. Publication bias did not alter our conclusions. Conclusions Physical activity has positive effects on physiology, body composition, physical functions, psychological outcomes, and quality of life in patients after treatment for breast cancer. When patients with cancer other than breast cancer were also included, physical activity was associated with reduced BMI and body weight, increased peak oxygen consumption and peak power output, and improved quality of life.

Economic evaluation of adult rehabilitation: a systematic review and meta-analysis of randomized controlled trials in a variety of settings

Objectives:
To report if there is a difference in costs from a societal perspective between adults receiving rehabilitation in an inpatient rehabilitation setting versus an alternative setting. If there are cost differences, to report whether opting for the least expensive program setting adversely affects patient outcomes.

Data Sources:
Electronic databases from the earliest possible date until May 2011. All languages were included.

Study Selection
Multiple reviewers identified randomized controlled trials with a full economic evaluation that compared adult inpatient rehabilitation with an alternative. There were 29 included trials with 6746 participants.

Data Extraction
Multiple observers extracted data independently. Trial appraisal included a risk of bias assessment and a checklist to report the strength of the economic evaluation.

Data Synthesis:
Results were synthesized using standardized mean differences (SMDs) and meta-analyses for the primary outcome of cost. The Grading of Recommendations Assessment, Development, and Evaluation was applied to assess for risk of bias across studies for meta-analyses. There was high-quality evidence that cost was significantly reduced for rehabilitation in the home versus inpatient rehabilitation in a meta-analysis of 732 patients poststroke (pooled SMD [δ]=−.28; 95% confidence interval [CI], −.47 to −.09), without compromise to patient outcomes. Results of individual trials in other patient groups (orthopedic, rheumatoid arthritis, and geriatric) receiving rehabilitation in the home or community were generally consistent with the meta-analysis. There was moderate quality evidence that cost was significantly reduced for inpatient rehabilitation (stroke unit) versus general acute care in a meta-analysis of 463 patients poststroke (δ=.31; 95% CI, .15–.48), with improvement to patient outcomes. These results were not replicated in 2 individual trials with a geriatric and a mixed cohort, where costs did not differ between general acute care and inpatient rehabilitation. Three of the 4 individual trials, inclusive of a stroke or orthopedic population, reported less cost for an intensive inpatient rehabilitation program compared with usual inpatient rehabilitation. Sensitivity analysis included a health service perspective and varied inflation rates with no change to the significant findings of the meta-analyses.

Conclusions:
Based on this systematic review and meta-analyses, a single rehabilitation service may not provide health economic benefits for all patient groups and situations. For some patients, inpatient rehabilitation may be the most cost-effective method of providing rehabilitation; yet, for other patients, rehabilitation in the home or community may be the most cost-effective model of care. To achieve cost-effective outcomes, the ideal combination of rehabilitation services and patient inclusion criteria, as well as further data for nonstroke populations, warrants further research.

Fundamental Movement Skill Interventions in Youth: A Systematic Review and Meta-analysis

BACKGROUND
Fundamental movement skill (FMS) proficiency is positively associated with physical activity and fitness levels. The objective of this study was to systematically review evidence for the benefits of FMS interventions targeting youth.

METHODS
A search with no date restrictions was conducted across 7 databases. Studies included any school-, home-, or community-based intervention for typically developing youth with clear intent to improve FMS proficiency and that reported statistical analysis of FMS competence at both preintervention and at least 1 other postintervention time point. Study designs included randomized controlled trials (RCTs) using experimental and quasi-experimental designs and single group pre-post trials. Risk of bias was independently assessed by 2 reviewers.

RESULTS
Twenty-two articles (6 RCTs, 13 quasi-experimental trials, 3 pre-post trials) describing 19 interventions were included. All but 1 intervention were evaluated in primary/elementary schools. All studies reported significant intervention effects for ≥1 FMS. Meta-analyses revealed large effect sizes for overall gross motor proficiency (standardized mean difference [SMD] = 1.42, 95% confidence interval [CI] 0.68–2.16, Z = 3.77, P < .0002) and locomotor skill competency (SMD = 1.42, 95% CI 0.56–2.27, Z = 3.25, P = .001). A medium effect size for object control skill competency was observed (SMD = 0.63, 95% CI 0.28–0.98, Z = 3.53, P = .0004). Many studies scored poorly for risk of bias items.

CONCLUSIONS
School- and community-based programs that include developmentally appropriate FMS learning experiences delivered by physical education specialists or highly trained classroom teachers significantly improve FMS proficiency in youth.

A systematic review and meta-analysis of diabetes and risk of physical disability and functional impairment - protocol

Background
Diabetes and increased age are known risk factors for physical disability. With the increasing prevalence of diabetes within our aging population, the future burden of disability is expected to increase. To date, there has not been a pooled estimate of the risk for disability associated with diabetes or its precursor states, impaired glucose tolerance and impaired fasting glucose. We aim to conduct a systematic review and meta-analysis of the association between prediabetes and diabetes with disability, and quantify the risk of association.

Methods/design
We will search for relevant studies in Medline via Pubmed, Embase, Cochrane library and Cumulative Index to Nursing and Allied Health Literature (CINAHL), as well as scan reference lists from relevant reviews and publications included in our review. We will review all publications that include studies on human adults (18 years and older) where information is included on diabetes status and at least one measure of disability (Activities of Daily Living (ADL), Instrumental ADL (IADL) or functional/mobility limitation), and where a risk association is available for the relationship between diabetes and/or prediabetes with disability, with reference to those without diabetes.

We will further conduct a meta-analysis to pool estimates of the risk of disability associated with prediabetes and diabetes. Sensitivity analysis will be conducted to assess for publication bias and study quality.Findings from this systematic review and meta-analysis will be widely disseminated through discussions with stake-holders, publication in a peer-reviewed journal and conference presentation.

Diabetes and risk of physical disability in adults: a systematic review and meta-analysis

Background
According to previous reports, the risk of disability as a result of diabetes varies from none to double. Disability is an important measure of health and an estimate of the risk of disability as a result of diabetes is crucial in view of the global diabetes epidemic. We did a systematic review and meta-analysis to estimate this risk.

Methods
We searched Ovid, Medline, Embase, Cochrane Library, and Cumulative Index to Nursing and Allied Health Literature up to Aug 8, 2012. We included studies of adults that compared the risk of disability—as measured by activities of daily living (ADL), instrumental activities of daily living (IADL), or mobility—in people with and without any type of diabetes. We excluded studies of subpopulations with specific illnesses or of people in nursing homes. From the studies, we recorded population characteristics, how diabetes was diagnosed (by doctor or self-reported), domain and definition of disability, and risk estimates for disability. We calculated pooled estimates by disability type and type of risk estimate (odds ratio [OR] or risk ratio [RR]).

Results
Our systematic review returned 3224 results, from which 26 studies were included in our meta-analyses. Diabetes increased the risk of mobility disability (15 studies; OR 1·71, 95% CI 1·53—1·91; RR 1·51, 95% CI 1·38—1·64), of IADL disability (ten studies; OR 1·65, 95% CI 1·55—1·74), and of ADL disability (16 studies; OR 1·82, 95% CI 1·63—2·04; RR 1·82, 95% CI 1·40—2·36).

Interpretation
Diabetes is associated with a strong increase in the risk of physical disability. Efforts to promote healthy ageing should account for this risk through prevention and management of diabetes.

Funding
Monash University, Baker IDI Bright Sparks Foundation, Australian Postgraduate Award, VicHealth, National Health and Medical Research Council, Australian Research Council, Victorian Government.

Suicide by occupation: systematic review and meta-analysis

Background Previous research has shown that those employed in certain occupations, such as doctors and farmers, have an elevated risk of suicide, yet little research has sought to synthesise these findings across working-age populations. Aims To summarise published research in this area through systematic review and meta-analysis. Method Random effects meta-analyses were used to calculate a pooled risk of suicide across occupational skill-level groups. Results Thirty-four studies were included in the meta-analysis. Elementary professions (e.g. labourers and cleaners) were at elevated risk compared with the working-age population (rate ratio (RR) = 1.84, 95% CI 1.46-2.33), followed by machine operators and deck crew (RR = 1.78, 95% CI 1.22-2.60) and agricultural workers (RR = 1.64, 95% CI 1.19-2.28). Results suggested a stepwise gradient in risk, with the lowest skilled occupations being at greater risk of suicide than the highest skill-level group. Conclusions This is the first comprehensive meta-analytical review of suicide and occupation. There is a need for future studies to investigate explanations for the observed skill-level differences, particularly in people employed in lower skill-level groups.

Protocol for a collaborative meta-analysis of 5-HTTLPR, stress, and depression

Background: Debate is ongoing about what role, if any, variation in the serotonin transporter linked polymorphic region (5-HTTLPR) plays in depression. Some studies report an interaction between 5-HTTLPR variation and stressful life events affecting the risk for depression, others report a main effect of 5-HTTLPR variation on depression, while others find no evidence for either a main or interaction effect. Meta-analyses of multiple studies have also reached differing conclusions.

Methods/Design: To improve understanding of the combined roles of 5-HTTLPR variation and stress in the development of depression, we are conducting a meta-analysis of multiple independent datasets. This coordinated approach utilizes new analyses performed with centrally-developed, standardized scripts. This publication documents the protocol for this collaborative, consortium-based meta-analysis of 5-HTTLPR variation, stress, and depression.

Study eligibility criteria: Our goal is to invite all datasets, published or unpublished, with 5-HTTLPR genotype and assessments of stress and depression for at least 300 subjects. This inclusive approach is to minimize potential impact from publication bias.

Data sources: This project currently includes investigators from 35 independent groups, providing data on at least N = 33,761 participants.  The analytic plan was determined prior to starting data analysis. Analyses of individual study datasets will be performed by the investigators who collected the data using centrally-developed standardized analysis scripts to ensure a consistent analytical approach across sites. The consortium as a group will review and interpret the meta-analysis results.

Discussion: Variation in 5-HTTLPR is hypothesized to moderate the response to stress on depression. To test specific hypotheses about the role of 5-HTTLPR variation on depression, we will perform coordinated meta-analyses of de novo results obtained from all available data, using variables and analyses determined a priori. Primary analyses, based on the original 2003 report by Caspi and colleagues of a GxE interaction will be supplemented by secondary analyses to help interpret and clarify issues ranging from the mechanism of effect to heterogeneity among the contributing studies. Publication of this protocol serves to protect this project from biased reporting and to improve the ability of readers to interpret the results of this specific meta-analysis upon its completion.

Peripheral brain-derived neurotrophic factor in schizophrenia and the role of antipsychotics: meta-analysis and implications

It has been postulated that schizophrenia (SZ) is related to a lower expression of brain-derived neurotrophic factor (BDNF). In the past few years, an increasing number of divergent clinical studies assessing BDNF in serum and plasma have been published. It is now possible to verify the relationship between BDNF levels and severity of symptoms in SZ as well as the effects of antipsychotic drugs on BDNF using meta-analysis. The aims of this study were to verify if peripheral BDNF is decreased in SZ, whether its levels are correlated with positive and negative symptomatology and if BDNF levels change after antipsychotic treatment. This report consists of two distinct meta-analyses of peripheral BDNF in SZ including a total of 41 studies and more than 7000 participants: (1) peripheral BDNF levels in serum and plasma were moderately reduced in SZ compared with controls. Notably, this decrease was accentuated with the disease duration. However, the extent of peripheral BDNF level decrease did not correlate with the severity of positive and negative symptoms. (2) In plasma, but not serum, peripheral BDNF levels are consistently increased after antipsychotic treatment irrespective of the patient's response to medication. In conclusion, there is compelling evidence that there are decreased levels of peripheral BDNF in SZ, in parallel to previously described reduced cerebral BDNF expression. It remains unclear whether these systemic changes are causally related to the development of SZ or if they are merely a pathologic epiphenomenon.Molecular Psychiatry advance online publication, 30 September 2014; doi:10.1038/mp.2014.117.

Adipokines as emerging depression biomarkers: a systematic review and meta-analysis

Adiponectin, leptin and resistin may play a role in the pathophysiology of major depressive disorder (MDD). However, differences in peripheral levels of these hormones are inconsistent across diagnostic and intervention studies. Therefore, we performed meta-analyses of diagnostic studies (i.e., MDD subjects versus healthy controls) and intervention investigations (i.e., pre-vs. post-antidepressant treatment) in MDD. Adiponectin (N=1278; Hedge's g=-0.35; P=0.16) and leptin (N=893; Hedge's g=-0.018; P=0.93) did not differ across diagnostic studies. Meta-regression analyses revealed that gender and depression severity explained the heterogeneity observed in adiponectin diagnostic studies, while BMI and the difference in BMI between MDD individuals and controls explained the heterogeneity of leptin diagnostic studies. Subgroup analyses revealed that adiponectin peripheral levels were significantly lower in MDD participants compared to controls when assayed with RIA, but not ELISA. Leptin levels were significantly higher in individuals with mild/moderate depression versus controls. Resistin serum levels were lower in MDD individuals compared to healthy controls (N=298; Hedge's g=-0.25; P=0.03). Leptin serum levels did not change after antidepressant treatment. However, heterogeneity was significant and sample size was low (N=108); consequently meta-regression analysis could not be performed. Intervention meta-analyses could not be performed for adiponectin and resistin (i.e., few studies met inclusion criteria). In conclusion, this systematic review and meta-analysis underscored that relevant moderators/confounders (e.g., BMI, depression severity and type of assay) should be controlled for when considering the role of leptin and adiponectin as putative MDD diagnostic biomarkers.

Problem gambling and intimate partner violence : a systematic review and meta-analysis

This study provides a systematic review of the empirical evidence related to the association between problem gambling and intimate partner violence (IPV). We identified 14 available studies in the systematic search (8 for victimisation only, 4 for perpetration only and 2 for both victimisation and perpetration). Although there were some equivocal findings, we found that most of the available research suggests that there is a significant relationship between problem gambling and being a victim of IPV. There was more consistent evidence that there is a significant relationship between problem gambling and perpetration of IPV. Meta-analyses revealed that over one third of problem gamblers report being victims of physical IPV (38.1%) or perpetrators of physical IPV (36.5%) and that the prevalence of problem gambling in IPV perpetrators is 11.3%. Although the exact nature of the relationships between problem gambling and IPV is yet to be determined, the findings suggest that less than full employment and clinical anger problems are implicated in the relationship between problem gambling and IPV victimization and that younger age, less than full employment, clinical anger problems, impulsivity, and alcohol and substance use are implicated in the relationship between problem gambling and IPV perpetration. The findings highlight the need for treatment services to undertake routine screening and assessment of problem gambling, IPV, alcohol and substance use problems, and mental health issues and provide interventions designed to manage this cluster of comorbid conditions. Further research is also required to investigate the relationship between problem gambling and violence that extends into the family beyond intimate partners.

Barriers to primary care clinician adherence to clinical guidelines for the management of low back pain: protocol of a systematic review and meta-synthesis of qualitative studies.

INTRODUCTION: Low back pain is the highest ranked condition contributing to years lived with disability, and is a significant economic and societal burden. Evidence-based clinical practice guidelines are designed to improve quality of care and reduce practice variation by providing graded recommendations based on the best available evidence. Studies of low back pain guideline implementation have shown no or modest effects at changing clinical practice. OBJECTIVES: To identify enablers and barriers to adherence to clinical practice guidelines for the management of low back pain. METHODS AND ANALYSIS: A systematic review and meta-synthesis of qualitative studies that will be conducted and reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement guidelines. Eight databases will be searched using a priori inclusion/exclusion criteria. Two independent reviewers will conduct a structured review and meta-synthesis, and a third reviewer will arbitrate where there is disagreement. This protocol has been registered on PROSPERO 2014. ETHICS AND DISSEMINATION: Ethical approval is not required. The systematic review will be published in a peer-reviewed journal. The review will also be disseminated electronically, in print and at conferences. Updates of the review will be conducted to inform and guide healthcare translation into practice. TRIAL REGISTRATION NUMBER: PROSPERO 2014:CRD42014012961. Available from http://www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42014012961.

C-reactive protein is increased in schizophrenia but is not altered by antipsychotics : meta-analysis and implications

The inflammatory hypothesis of schizophrenia (SZ) posits that inflammatory processes and neural-immune interactions are involved in its pathogenesis, and may underpin some of its neurobiological correlates. SZ is the psychiatric disorder causing the most severe burden of illness, not just owing to its psychiatric impairment, but also owing to its significant medical comorbidity. C-reactive protein (CRP) is a commonly used biomarker of systemic inflammation worldwide. There are some conflicting results regarding the behaviour of CRP in SZ. The aims of this study were to verify whether peripheral CRP levels are indeed increased in SZ, whether different classes of antipsychotics divergently modulate CRP levels and whether its levels are correlated with positive and negative symptomatology. With that in mind, we performed a meta-analysis of all cross-sectional studies of serum and plasma CRP levels in SZ compared to healthy subjects. In addition, we evaluated longitudinal studies on CRP levels before and after antipsychotic use. Our meta-analyses of CRP in SZ included a total of 26 cross-sectional or longitudinal studies comprising 85 000 participants. CRP levels were moderately increased in persons with SZ regardless of the use of antipsychotics and did not change between the first episode of psychosis and with progression of SZ (g=0.66, 95% confidence interval (95% CI) 0.43 to 0.88, P<0.001, 24 between-group comparisons, n=82 962). The extent of the increase in peripheral CRP levels paralleled the increase in severity of positive symptoms, but was unrelated to the severity of negative symptoms. CRP levels were also aligned with an increased body mass index. Conversely, higher age correlated with a smaller difference in CRP levels between persons with SZ and controls. Furthermore, CRP levels did not increase after initiation of antipsychotic medication notwithstanding whether these were typical or atypical antipsychotics (g=0.01, 95% CI -0.20 to 0.22, P=0.803, 8 within-group comparisons, n=713). In summary, our study provides further evidence of the inflammatory hypothesis of SZ. Whether there is a causal relationship between higher CRP levels and the development of SZ and aggravation of psychotic symptoms, or whether they are solely a marker of systemic low-grade inflammation in SZ, remains to be clarified.