Advanced Search

Completed Systematic Reviews




Screening for Elevated Blood Lead Levels in Childhood [Entered Retrospectively]


Public Project Complete

Statistics: 25 Studies, 6 Key Questions, 1 Extraction Form,
Date Published: Sep 20, 2019 09:42AM
Description: Background: In 2006, the United States Preventive Services Task Force (USPSTF) found insufficient evidence to recommend for or against routine screening for elevated blood lead levels in asymptomatic children aged 1 to 5 who are at increased risk for lead poisoning (I recommendation), and recommended against routine screening for those at average risk (D recommendation). Purpose: To synthesize evidence on the effects of screening, testing, and treatment for elevated blood lead level in children aged five and under in the primary care setting, in order to update a prior USPSTF review on screening for elevated blood lead levels in childhood. Data Sources: Cochrane CENTRAL and Cochrane Database of Systematic Reviews (through June 2018), and Ovid MEDLINE (1946 to June 2018), reference lists, and surveillance through December 5, 2018. Study Selection: English-language trials and observational studies of screening effectiveness, test accuracy, benefits and harms of screening and interventions in asymptomatic children five and under. Data Extraction: One investigator abstracted details about study design, patient population, setting, screening method, follow up, and results. Two investigators independently applied pre specified criteria to rate study quality using methods developed by the USPSTF. Discrepancies were resolved through consensus. Data Synthesis (Results): A total of 22 studies were included in this review (N=10,449). No studies directly evaluated clinical benefits or harms of screening versus not screening children for elevated blood lead levels. More than one positive answer on the 5-item 1991 Centers for Disease Control and Prevention (CDC) screening questionnaire was associated with a pooled sensitivity of 48 percent (95% confidence interval [CI], 31.4 to 65.6%) and specificity of 58 percent (95% CI, 39.9 to 74.0%) for identifying children with a venous blood level >10µg/dL (5 studies, N = 2,265). Adapted versions of the CDC questionnaire did not demonstrate improved accuracy. Capillary blood lead testing demonstrated sensitivity of 87 percent to 91 percent and specificity >90 percent, compared with venous measurement (4 studies, N = 1,431). Counseling and nutritional interventions or residential lead hazard control techniques did not reduce blood lead concentrations in asymptomatic children, but studies were few and had methodological limitations (7 studies, N = 1,419). A trial of dimercaptosuccinic acid (DMSA) chelation therapy found reduced blood lead levels in children at one week to one year but not at 4.5 to 6 years (N = 780), while another trial found no effect at one- and six-months (N = 39). Seven-year followup assessments showed no effect on neuropsychological development; a small deficit in linear growth (height difference at 7 years in treated patients 1.17cm; 95% CI, 0.41 to 1.93); and poorer cognitive outcomes reported as the Attention and Executive Functions sub-score of the Developmental Neuropsychological Assessment (NEPSY) (unadjusted difference -1.8; 95% CI, -4.5 to 1.0, adjusted P = 0.045) in children treated with DMSA chelation. Limitations: Limited to English-language articles; quality and applicability of studies were limited due to study design, poor reporting of statistical outcomes, and loss to follow up. Studies were lacking on the effectiveness of screening or effectiveness of treatments in reducing elevated blood lead levels or improving health outcomes in children. There was no direct evidence on the harms of screening children for elevated blood lead levels. Conclusions: Evidence on the benefits and harms of screening children for elevated lead levels is lacking. Screening questionnaires are not accurate for identifying children with elevated blood lead levels. Capillary blood testing is slightly less accurate than venous blood levels for identification of elevated blood levels. Treatment studies of chelating agents, often combined with environmental or household interventions, were not associated with sustained effects on blood level levels but were associated with harms.
Contributor(s): Amy G. Cantor, MD, MPH Rob Hendrickson, MD Ian Blazina, MPH Jessica Griffin, MS Sara Grusing, BA Marian S. McDonagh, PharmD
Funding Source: This report is based on research conducted by the Pacific Northwest Evidence-based Practice Center (EPC) under contract to the Agency for Healthcare Research and Quality (AHRQ), Rockville, MD (Contract No. HHSA290201500009I, Task Order No. 7)
Methodology Description: Study Selection: English-language trials and observational studies of screening effectiveness, test accuracy, benefits and harms of screening and interventions in asymptomatic children five and under. Data Extraction: One investigator abstracted details about study design, patient population, setting, screening method, follow up, and results. Two investigators independently applied pre specified criteria to rate study quality using methods developed by the USPSTF. Discrepancies were resolved through consensus.

Zoom Preview | Show Downloadable Content

Phenotypes and body mass in women with PCOS identified in Referral vs. Unselected populations: systematic review and meta-analysis


Public Project Complete

Statistics: 43 Studies, 2 Key Questions, 1 Extraction Form,
Date Published: Aug 01, 2019 12:55PM
Description: Objective: To compare the prevalence of PCOS phenotypes and obesity of PCOS women seen in the clinical (referred) setting vs. those identified in general population.
Contributor(s): Daria Lizneva, Richard Kirubakaran, Katerina Mykhalchenko, Larisa Suturina, Chernukha Galina, Michael P. Diamond, Ricardo Azziz.
Funding Source: Funds from the Career Development Award (MD Medical Group) were used to support D.L. throughout the manuscript preparation. No other funding recourses were identified.
Methodology Description: Objective: To compare the prevalence of polycystic ovary syndrome (PCOS) phenotypes and obesity among patients detected in referral versus unselected populations. Design: Systematic review and meta-analysis. Setting: Not applicable. Patient(s): Thirteen thousand seven hundred ninety-six reproductive-age patients with PCOS, as defined by the extended 2003 Rotterdam criteria. Intervention(s): Review of PUBMED, EMBASE, and Cochrane Library, 2003–2016. Only observational studies were included. Data were extracted using a web-based, piloted form and combined for meta-analysis. Main Outcome Measure(s): PCOS phenotypes were classified as follows: phenotype A, clinical and/or biochemical hyperandrogenism (HA) + oligo-/anovulation (OA) + polycystic ovarian morphology (PCOM); phenotype B, HA+OA; phenotype C, HA+PCOM; and phenotype D, OA+PCOM. Result(s): Forty-one eligible studies, reporting on 43 populations, were identified. Pooled estimates of detected PCOS phenotype prevalence were consequently documented in referral versus unselected populations, as [1] phenotype A, 50% (95% confidence interval [CI], 46%–54%) versus 19% (95% CI, 13%–27%); [2] phenotype B, 13% (95% CI, 11%–17%) versus 25% (95% CI, 15%–37%); [3] phenotype C, 14% (95% CI, 12%–16%) versus 34% (95% CI, 25–46%); and [4] phenotype D, 17% (95% CI, 13%–22%) versus 19% (95% CI, 14%– 25%). Differences between referral and unselected populations were statistically significant for phenotypes A, B, and C. Referral PCOS subjects had a greater mean body mass index (BMI) than local controls, a difference that was not apparent in unselected PCOS. Conclusion(s): The prevalence of more complete phenotypes in PCOS and mean BMI were higher in subjects identified in referral versus unselected populations, suggesting the presence of significant referral bias.

Zoom Preview | Show Downloadable Content

Decision Aids for Cancer Screening and Treatment


Public Project Complete

Statistics: 75 Studies, 2 Key Questions, 2 Extraction Forms,
Date Published: May 23, 2019 03:00PM
Description: Background: Many health decisions about screening and treatment for cancers involve uncertainty or tradeoffs between the expected benefits and harms. Patient decision aids have been developed to help health care consumers and their providers identify the available alternatives and choose the one that aligns with their values. It is unclear whether the effectiveness of decision aids for decisions related to cancers differs by people’s average risk of cancer or by the content and format of the decision aid.; Objectives: We sought to appraise and synthesize the evidence assessing the effectiveness of decision aids targeting health care consumers who face decisions about cancer screening or prevention, or early cancer treatment (Key Question 1), particularly with regard to decision aid or patient characteristics that might function as effect modifiers. We also reviewed interventions targeting providers for promotion of shared decision making using decision aids (Key Question 2).; Data sources: We searched MEDLINE®, Embase®, the Cochrane Central Register of Controlled Trials (CENTRAL), PsycINFO®, and the Cumulative Index to Nursing and Allied Health Literature (CINAHL®) from inception to the end of June 2014.; Review methods: For Key Question 1, we included randomized controlled trials comparing decision aid interventions among themselves or with a control. We included trials of previously developed decision aids that were delivered at the point of the actual decision. We predefined three population groups of interest based on risk or presence of cancer (average cancer risk, high cancer risk, early cancer). The assessed outcomes pertained to measurements of decisional quality and cognition (e.g., knowledge scores), attributes of the decision-making process (e.g., Decisional Conflict Scale), emotion and quality of life (e.g., decisional regret), and process and system-level attributes. We assessed for effect modification by population group, by the delivery format or content of the decision aid or other attributes, or by methodological characteristics of the studies. For Key Question 2, we included studies of any intervention to promote patient decision aid use, regardless of study design and outcomes assessed.; Results: Of the 16,669 screened citations, 87 publications were eligible, corresponding to 83 (68 trials; 25,337 participants) and 5 reports for Key Questions 1 and 2, respectively. Regarding the evolution of the decision aid format and content over time, more recent trials increasingly studied decision aids that were more practical to deliver (e.g., over the Internet or without human mediation) and more often clarified preferences explicitly. Overall, participants using decision aids had higher knowledge scores compared with those not using decision aids (standardized mean difference, 0.23; 95% credible interval [CrI], 0.09 to 0.35; 42 comparison strata with 12,484 participants). Compared with not using decision aids, using decision aids resulted in slightly lower decisional conflict scores (weighted mean difference of -5.3 units [CrI, -8.9 to - 1.8] on the 0-100 Decisional Conflict Scale; 28 comparison strata; 7,923 participants). There was no difference in State-Trait Anxiety Inventory scores (weighted mean difference = 0.1; 95% CrI, -1.0 to 0.7 on a 20-80 scale; 16 comparison strata; 2,958 participants). Qualitative synthesis suggested that patients using decision aids are more likely to make informed decisions and have accurate risk perceptions; further, they may make choices that best agree with their values and may be less likely to remain undecided. Because there was insufficient, sparse, or no information about effects of decision aids on patient-provider communication, patient satisfaction with decision-making process, resource use, consultation length, costs, or litigation rates, a quantitative synthesis was not done. There was no evidence for effect modification by population group, by the delivery format or content of the decision aid or other attributes, or by methodological characteristics of the studies. Data on Key Question 2 were very limited.; Conclusions: Cancer-related decision aids have evolved over time, and there is considerable diversity in both format and available evidence. We found strong evidence that cancer-related decision aids increase knowledge without adverse impact on decisional conflict or anxiety. We found moderate- or low-strength evidence that patients using decision aids are more likely to make informed decisions, have accurate risk perceptions, make choices that best agree with their values, and not remain undecided. This review adds to the literature that the effectiveness of cancer-related decision aids does not appear to be modified by specific attributes of decision aid delivery format, content, or other characteristics of their development and implementation. Very limited information was available on other outcomes or on the effectiveness of interventions that target providers to promote shared decision making by means of decision aids.
Contributor(s): Gaelen Adam Thomas Trikalinos L. Susan Wieland Anja Zgodic Evangelia Ntzani
Funding Source: AHRQ
Methodology Description: Eligible Studies for Key Question 1: We included randomized controlled trials comparing use of patient decision aids with other patient decision aids or with no decision aid intervention. We included trials of mature patient decision aids delivered at the point of the actual decision. We excluded trials about hypothetical treatment decisions. For example, we excluded hypothetical questions about early cancer treatment in people not yet diagnosed with cancer, or trials about cancer screening among people who would not be typical screening candidates. We predefined three populations of interest, based on risk or presence of cancer. The first population included people without cancer who are at average risk and face decisions about cancer screening (whether or how to be screened). The second population included people without cancer but with high risk of cancer, e.g., because they are suspected or known to have a hereditary cancer-related condition, such as the Lynch or von Hippel-Lindau syndromes, or are carriers of deleterious BRCA gene mutations. This group may face decisions about further diagnostic workup or about undergoing preventive interventions. The third population included patients diagnosed with early cancer, defined as being at a stage with favorable prognosis (typically local disease only) and where interventions have curative intent (e.g., stage IIa or lower for prostate cancer). We accepted the individual study claims for the definition of early cancer. When a study used an alternative cancer staging, we adjudicated an early cancer stage using information for the National Cancer Institute site. We included only studies in people who were legally able to make decisions for themselves or an underage minor. We followed the IPDAS collaboration and previous systematic reviews in defining decision aid-based interventions as, at a minimum, (1) informing about available options and the expected associated benefits and harms, and (2) incorporating at least implicit clarification of the decisionmaker’s values.3,4 ; Eligible Studies for Key Question 2: For the second Key Question, we included comparative studies informing on the effectiveness of interventions for promoting shared decision making to providers caring for the populations discussed for the first Key Question, specifically provider-targeted interventions to increase shared decision making with the use or increased use of a decision aid. Because so few studies have been done on this topic, eligible designs included randomized and cluster- randomized trials, nonrandomized studies with concurrent comparators, before-after studies, and interrupted time series studies.

Zoom Preview | Show Downloadable Content

Early Intensive Intervention for Children with Autism Spectrum Disorder: A Systematic Review Update


Public Project Complete

Statistics: 80 Studies, 1 Key Question, 1 Extraction Form,
Date Published: Mar 25, 2019 05:51AM
Description: The objective of this study was to systematically review studies of early intensive behavioral and developmental approaches that may improve outcomes for children with ASD.
Contributor(s): Amy S. Weitlauf, Ph.D. Melissa L. McPheeters, Ph.D., M.P.H. Brittany Peters, M.D. Nila Sathe, M.A., M.L.I.S. Rebekah Travis, Psy.D. Rachel Aiello, Ph.D. Edwin Williamson, M.D. Jeremy Veenstra-VanderWeele, M.D. Shanthi Krishnaswami, M.B.B.S., M.P.H. Rebecca Jerome, M.L.I.S., M.P.H. Zachary Warren, Ph.D.
Funding Source: This project was supported by the Agency for Healthcare Research and Quality (contract number: HHSA290201200009I)
Methodology Description: Our primary literature search employed three databases: MEDLINE® via the PubMed interface, PsycINFO® (psychology and psychiatry literature), and the Educational Resources Information Clearinghouse. Our search strategies used a combination of subject heading terms appropriate for each database and key words relevant to ASD (e.g., autism, Asperger). We limited searches to the English language and literature published since the development of the 2011 AHRQ review on management of ASD. Two reviewers independently assessed studies against predetermined inclusion/exclusion criteria. Two reviewers independently extracted data regarding participant and intervention characteristics, assessment techniques, and outcomes and assigned quality and strength of evidence ratings using predetermined methods.

Zoom Preview | Show Downloadable Content

The shortcomings of cluster-randomized controlled trials for the assessment of complex interventions in general practices: a systematic review


Public Project Complete

Statistics: 29 Studies, 1 Key Question, 1 Extraction Form,
Date Published: Mar 25, 2019 05:50AM
Description: None Provided
Contributor(s): Professor Andrea Siebenhofer-Kroitzsch, Institute of General Practice, Goethe Univ. Frankfurt, Germany Dr Christiane Muth, Institute of General Practice, Goethe Univ. Frankfurt, Germany Dr Birgit Fullerton, Institute of General Practice, Goethe Univ. Frankfurt, Germany Professor Paul Glasziou, Centre for Research in Evidence-Based Practice, Bond University, Australia Professor Elaine Beller, Centre for Research in Evidence-Based Practice, Bond University, Australia Mrs Sarah Thorning, Centre for Research in Evidence-Based Practice, Bond University, Australia Professor Andrea Berghold, Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Austria Dr Klaus Jeitler, Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Austria Ms Gudrun Pregartner, Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Austria Ms Jennifer Engler, Institute of General Practice, Goethe Univ. Frankfurt, Germany Dr Gudrun Klein, Institute of General Practice, Goethe Univ. Frankfurt, Germany Mr Michael Paulitsch, Institute of General Practice, Goethe Univ. Frankfurt, Germany
Funding Source: This review will be funded by German Federal Ministry of Education and Research (FKZ: 01KG1504); approval date: 05/2015
Methodology Description: None Provided

Zoom Preview | Show Downloadable Content