Jamie H Thompson

Low-Dose Aspirin for Prevention of Morbidity and Mortality From Preeclampsia: A Systematic Evidence Review for the U.S. Preventive Services Task Force

Preeclampsia is a leading cause of maternal death, affecting 2% to 8% of pregnancies globally (1, 2). It affected 3.8% of U.S. deliveries in 2010, and the rate of severe preeclampsia has increased over the past 3 decades (3). Perinatal mortality is nearly 2 times higher in pregnancies affected by preeclampsia (4), with 12% of maternal deaths due to the condition (5). Serious illness is more common, with more than one third of serious maternal morbidity and 15% of preterm births related to preeclampsia (6, 7). Preeclampsia is defined as hypertension (blood pressure 140/90 mm Hg) and proteinuria (presence of 0.3 g of protein in a 24-hour period) observed during the second half of pregnancy (>20 weeks of gestation) (8, 9). It is also classified as having severe features with any of the following: blood pressure above 160/110 mm Hg, thrombocytopenia, impaired liver function, renal insufficiency, pulmonary edema, or cerebral or visual disturbances (9). Preeclampsia with or without severe features can evolve rapidly into eclampsia or the hemolysis, elevated liver enzymes, and low platelets syndrome, sometimes leading to systemic complications and maternal death (10, 11). Poor perinatal health outcomes are associated with preeclampsia, primarily due to increased risk for intrauterine growth restriction (IUGR) or medically initiated preterm delivery. Once preeclampsia develops, the only effective treatment is delivery, with serious neonatal harms when remote from term (<34 weeks of gestation). Current understanding of preeclampsia pathophysiology suggests that it may be a collection of syndromes with different precipitating factors and outcomes (12). Early in pregnancy, aberrations in placental development can result in placental ischemia and release of inflammatory and oxidative stress factors into the maternal bloodstream. In addition, even with normal placentation, preexisting hypertension, diabetes, and other inflammatory conditions (such as lupus) may activate systemic inflammatory and oxidative stress processes, as can twin or higher-order pregnancies. Accurate prediction of who will develop preeclampsia and have serious complications is not currently possible (1315). The most consistent predictors of high risk are previous preeclampsia, certain medical conditions (diabetes, chronic hypertension, renal disease, autoimmune diseases, and the antiphospholipid syndrome), and multifetal pregnancy (16). Moderately elevated risk for preeclampsia is associated with nulliparity (first birth), advanced maternal age (40 years), between-pregnancy interval of more than 10 years, high body mass index (35 kg/m2), and family history of preeclampsia (mother or sister). Risk factors with less consistent evidence include changes in paternity between pregnancies, history of migraine headaches (17, 18), and asthma (17, 1922). Predictive models combining various biomarkers, patient risk factors, and clinical readings hold promise but are not yet sufficiently validated for clinical use (10, 2325). Previous comprehensive systematic reviews have found antiplatelets (primarily low-dose aspirin) to be beneficial for the prevention of preeclampsia among women at heightened risk (26, 27). We conducted this systematic review to support the U.S. Preventive Services Task Force (USPSTF) in updating its 1996 recommendation, which is no longer active. Methods Detailed methods are outlined in our full evidence report (28). This review addressed 3 key questions (Appendix Figure 1). First, is low-dose aspirin effective for reducing adverse maternal and perinatal health outcomes among women at increased risk for preeclampsia? Second, is low-dose aspirin effective for preventing preeclampsia among women at increased risk for the condition? Third, are there harms to the woman and fetus associated with aspirin use during pregnancy? Appendix Figure 1. Analytic framework and key questions. ARDS = acute respiratory distress syndrome; HELLP = hemolysis, elevated liver enzymes, and low platelets. *Abbreviated list of health outcomes. See Appendix Table 2 for a full list. Data Sources and Searches In addition to considering all studies from the previous USPSTF review, we performed a comprehensive search of MEDLINE, PubMed, the Database of Abstracts of Reviews of Effects, and the Cochrane Central Register of Controlled Trials for studies published between January 2006 and 1 June 2013. We also examined the reference lists from existing systematic reviews to identify potentially eligible studies, including an individual-patient data (IPD) meta-analysis published by the Perinatal Antiplatelet Review of International Studies (PARIS) Collaboration (27) and a 2007 Cochrane review (26). We searched ClinicalTrials.gov for ongoing trials (May 2013). Between the last search date and this publication, we actively monitored published literature for potentially important new trials or other large observational studies directly relevant to our key questions; none were identified. Study Selection Two investigators independently reviewed abstracts and full-text articles for inclusion according to predetermined criteria. We resolved discrepancies through consensus with a third investigator. To evaluate benefits of aspirin prophylaxis, we included any study that used a risk selection approach aimed at achieving a sample of women at high risk for preeclampsia. The trials could define risk on the basis of medical history, pregnancy characteristics, or clinical measurements known to be associated with risk for the condition. Although preeclampsia occurs more often in first births than in subsequent ones, prevalence rates are relatively low (approximately 4%) compared with other high-risk groups. Because aspirin treatment based only on this risk factor has not been supported, trials with nulliparity as the sole risk factor were not included for evaluation of benefits. We used broader inclusion criteria to identify possible harms of aspirin exposure during pregnancy. The trials of women at high risk were combined with trials of women at low or average risk exposed to daily low-dose aspirin. Large prospective observational studies were also included to assess harms but were not included in pooled analyses. We included interventions that compared patients receiving 50 to 150 mg of aspirin with a placebo or no treatment group and excluded studies of nonaspirin antiplatelet medications or aspirin combined with another active substance. We also excluded studies that we rated as poor-quality on the basis of the USPSTF quality rating standards (29) and studies not published in English. Data Extraction and Quality Assessment Two investigators critically appraised all included studies independently using the USPSTFs design-specific criteria (29), which we supplemented with the National Institute for Health and Care Excellence methodology checklists (30) and the Newcastle-Ottawa Scale (31). According to the USPSTF criteria, a good-quality study met all prespecified standards. A fair-quality study did not meet (or it was unclear whether it met) at least 1 criterion, but it also had no known limitation that could invalidate its results. A poor-quality study had a single fatal flaw or multiple important limitations that could seriously bias its results. Discrepancies were resolved through discussion of identified limitations and consultation with a third investigator, if necessary. One investigator extracted study details and results, and a second investigator reviewed the abstracted information. Data Synthesis and Analysis We used the metan procedure in Stata, version 11.2 (StataCorp, College Station, Texas), for all reported meta-analyses and the metaan procedure for sensitivity analyses (32). For dichotomous outcomes, we entered the number of events and nonevents and estimated pooled random-effects risk ratios by using the DerSimonianLaird method for all outcomes, except those in which fewer than 10% of the participants had the event (33), for which we used a fixed-effects MantelHaenszel model (34). We also included prediction intervals in forest plots of random-effects models, which provided an estimate of where the effect size from 95% of newly conducted trials would fall, assuming that the between-study variability in the included trials held for new trials (35). The prediction intervals are shown on the forest plots by the horizontal lines that extend from the diamond representing the 95% CI of the pooled estimate. Potential sources of heterogeneity in effect size by aspirin timing, dosage, and preeclampsia risk determination were identified a priori and explored using meta regression and visual inspection of sorted forest plots. We used the I 2 and chi-square statistics to assess statistical heterogeneity. To evaluate small-study effects, we examined funnel plots and used the Begg or Peter test depending on the outcome distribution (36, 37). We used profile likelihood estimation to conduct sensitivity analyses for the pooled effects because the DerSimonianLaird method can overestimate CI precision in meta-analysis, particularly when fewer than 10 studies or when smaller studies with few events are pooled (38). Role of the Funding Source This study was funded by the Agency for Healthcare Research and Quality (AHRQ) under a contract to support the work of the USPSTF. Members of the USPSTF and the AHRQ medical officer assisted in the development of the reviews scope. Approval from AHRQ was required before the manuscript could be submitted for publication, but the authors are solely responsible for its content and the decision to submit it for publication. Results Our literature search yielded 544 unique citations. From these, we reviewed the full text of 75 articles. Twenty-three studies (27 articles) met our inclusion criteria (Appendix Figure 2 and Appendix Table 1). Appendix Figure 2. Summary of evidence search and selection. The diagram excludes 51 RefMan (Thomson Reuters, Philadelph

Supplemental Screening for Breast Cancer in Women With Dense Breasts: A Systematic Review for the U.S. Preventive Services Task Force

Dense breasts are defined by mammographic appearance. The American College of Radiologys (ACRs) Breast Imaging Reporting and Data System (BI-RADS) classifies breasts as almost entirely fatty (BI-RADS category a), scattered areas of fibroglandular density (category b), heterogeneously dense (category c), or extremely dense (category d). About 27.6 million (43%) women aged 40 to 74 years in the United States have dense breasts; most of these are classified as category c (1). Higher breast density is associated with decreased mammographic sensitivity and specificity and also with increased breast cancer risk. The relative hazard of breast cancer for women with dense breasts ranged from 1.50 (women aged 65 to 74 years) to 1.83 (women aged 40 to 49 years) in an analysis of 1169248 women enrolled in the Breast Cancer Surveillance Consortium (unpublished data). Increased breast density has been associated with hormone replacement therapy use, younger age, and lower body mass index (2). Data on breast density and race or ethnicity are limited. In the United States, Asian women have higher breast density (3) but lower than average incidence of breast cancer (4). Increased breast density is not associated with higher breast cancer mortality among women with dense breasts diagnosed with breast cancer, after adjustment for stage and mode of detection (5). Supplemental breast cancer screening with additional screening modalities has been proposed to improve the early detection of breast cancers. No clinical guidelines explicitly recommend use of supplemental breast cancer screening on women with dense breasts (69), but as of September 2015, 24 states had enacted legislation requiring that women be notified of breast density with their mammography results; 9 more states are considering mandatory notification (10) (Appendix Table 1). Most states require specific language distinguishing dense (BI-RADS c and d) from nondense breasts, and 4 states require that insurers cover subsequent examinations and tests for women with dense breasts (1114). Federal legislation requiring breast density notification is pending (15). Appendix Table 1. Breast Density Legislation in the United States This report summarizes a systematic review of current evidence on the reproducibility of BI-RADS breast density determinations and on test performance characteristics and outcomes of supplemental screening of women with dense breasts by using hand-held ultrasonography (HHUS), automated whole-breast ultrasonography (ABUS), breast magnetic resonance imaging (MRI), and digital breast tomosynthesis (DBT). Mandatory reporting laws frame notification of women as dense/nondense, so this review focused on this categorization. Methods The review protocol included an analytic framework with 4 key questions (KQs) (Appendix Figure 1). Detailed methods, including search strategies, detailed inclusion criteria, and excluded studies, are available in the full evidence report (16). Appendix Figure 1. Analytic framework. BI-RADS = Breast Imaging Reporting and Data System; DCIS = ductal carcinoma in-situ; KQ = key question; MRI = magnetic resonance imaging. Data Sources and Searches MEDLINE, PubMed, EMBASE, and the Cochrane Library were searched for relevant English-language studies published between January 2000 and July 2015. We reviewed reference lists from retrieved articles and references suggested by experts. Study Selection Two investigators independently reviewed abstracts and full-text articles for inclusion according to predetermined criteria (E.P.W. and J.H.T. for KQ 1, J.M. and J.J.F. for KQs 2 to 4). Included studies examining the reproducibility of BI-RADS breast density categorization focused on asymptomatic women aged 40 years or older undergoing digital or film mammography. Included studies on supplemental screening with HHUS, ABUS, MRI, or DBT reported outcomes for asymptomatic women with dense breasts aged 40 years and older. In studies that focused primarily on women at high risk for breast cancer (including those with preexisting breast cancer or high-risk breast lesions [such as ductal carcinoma in situ, atypical hyperplasia, and lobular carcinoma in situ], BRCA mutations, familial breast cancer syndromes, or previous chest-wall radiation) and studies that included women with nondense breasts, we analyzed the relevant subset when available in the publication or provided by the authors. A priori inclusion criteria limited studies on BI-RADS reproducibility to fair- or good-quality randomized, controlled trials; cohort studies; or test sets involving multiple blind readings by at least 3 readers. Studies on test performance characteristics and outcomes of supplemental screening modalities were limited to fair- or good-quality randomized, controlled trials; cohort studies; or diagnostic accuracy studies with reference standards applied to all participants. We examined sensitivity, specificity, positive predictive values (PPVs), negative predictive values (NPVs), and available clinical outcomes (including cancer detection rates, recall rates, and biopsy rates). We defined recall as the need for any additional diagnostic testing after supplemental screening, including imaging and biopsy. Data Extraction and Quality Assessment Two investigators (E.P.W. and J.H.T. for KQ 1, J.M. and J.J.F. for KQs 2 to 4) critically appraised all included studies independently using the U.S. Preventive Services Task Forces (USPSTFs) design-specific criteria (17), supplemented with the National Institute for Health and Clinical Excellence methodology checklists (18) and the Quality Appraisal Tool for Studies of Diagnostic Reliability (19). According to USPSTF criteria, a good-quality study generally met all prespecified criteria; fair-quality studies did not meet all criteria but had no important limitations. Poor-quality studies had important limitations that could invalidate results (inadequate or biased application of reference standard; population limited to very high-risk patients). Data Synthesis and Analysis When available or provided by the authors, results of supplemental screening for subgroups of women with dense breasts were extracted; we excluded those with other risk factors for breast cancer. We calculated the sensitivity and specificity of the supplemental breast screening tests for women with negative mammography results. Only cancers detected by the supplemental test after negative mammography results and cancers found at interval follow-up were included. Hence, the values reported represent the sensitivity and specificity for detection of additional cancer in women with negative mammography findings. Similarly, we defined cancer detection rates, recall rates, and biopsy rates to include only those cancer cases, recalls, and biopsies related to supplemental screening after negative results on mammography. Meta-analysis was not performed because there were few good-quality studies. Role of the Funding Source This research was funded by the Agency for Healthcare Research and Quality (AHRQ) under a contract to support the work of the USPSTF. The investigators worked with USPSTF members to develop and refine the scope, analytic frameworks, and KQs. AHRQ had no role in study selection, quality assessment, synthesis, or development of conclusions. AHRQ provided project oversight; reviewed the draft report; and distributed the draft for peer review, including to representatives of professional societies and federal agencies. AHRQ performed a final review of the manuscript to ensure that the analysis met methodological standards. The investigators are solely responsible for the content and the decision to submit the manuscript for publication. Results The literature search yielded 2067 unique citations; 128 full-text articles considered potentially relevant were reviewed to identify 24 unique studies meeting inclusion criteria (Appendix Figure 2). Table 1 (2043) provides the characteristics of included studies. No studies addressed the effect of supplemental screening (compared with women without supplemental screening) on breast cancer morbidity or mortality. Appendix Figure 2. Summary of evidence search and selection. KQ = key question. Table 1. Characteristics of Included Studies Accuracy and Reliability of BI-RADS Density Determination Absent a gold standard for breast density, studies could not evaluate the accuracy of BI-RADS density determinations. Five studies reported repeated assignment of categorical BI-RADS breast density classification by the same or different radiologists, altogether including more than 440000 women, almost all with data from 2 sequential screening mammograms. To reflect current U.S. practice, we included only studies based on the BI-RADS density categories. The 3 largest studies were set in the United States. Two used data from the Breast Cancer Surveillance Consortium (20, 22), and the third presented findings from community radiologists conducting repeated readings of a large screening test set (24). Two other small studies (not discussed here) were based on mammographic screening programs in Spain (21) and Italy (23). All United Statesbased studies reflected community practice by use of clinical readings from community screening programs or test set readings by practicing community radiologists without additional training. Overall, group prevalence of BI-RADS density ratings was similar across initial and subsequent examinations among community radiologists (Appendix Table 2), but there was greater disagreement at the individual level. On subsequent screening examinations, approximately 1 in 5 women (23%) was placed in a different BI-RADS density category (a, b, c, d) by the same radiologist, while approximately 1 in 3 was categorized differently when a different radiologist read the subsequent examination result (Table 2). Considering clinical interpretations that combine categories (dense representing those with BI-RAD

Behavioral Counseling and Pharmacotherapy Interventions for Tobacco Cessation in Adults, Including Pregnant Women: A Review of Reviews for the U.S. Preventive Services Task Force

Cigarette smoking and exposure to smoke result in more than 480000 premature deaths in the United States every year, along with substantial illness (1, 2). Despite considerable progress in tobacco control over the past 50 years, in 2013, an estimated 17.8% of U.S. adults (3) and 15.9% of pregnant women aged 15 to 44 years were current cigarette smokers (4). Many tools are available to help smokers quit, including counseling by health care providers, telephone- and print-based interventions, computer and text-messaging interventions, and pharmacologic agents (that is, nicotine replacement therapy [NRT], bupropion hydrochloride sustained release [bupropion], and varenicline). In 2009, the U.S. Preventive Services Task Force (USPSTF) reaffirmed its 2003 recommendation that clinicians ask all adults about tobacco use and provide interventions for cessation for those who use tobacco products (grade A recommendation) (5). The original USPSTF recommendation (2003) and reaffirmation (2009) were based on the Public Health Services clinical practice guidelines on treating tobacco use and dependence (6, 7). Because there were no plans to update the Public Health Service report, we undertook the current review to assess the benefits and harms of behavioral and pharmacologic interventions for tobacco cessation in adults, including pregnant women, to assist the USPSTF in updating its 2009 recommendation. Because of the rapid increase in the use of electronic nicotine delivery systems (ENDS) and the vigorous debate about the public health effect of these devices and their role in smoking cessation (813), our review also synthesized the primary trial evidence on the efficacy and safety related to this technology as a means for quitting conventional smoking. Methods We relied primarily on a review of reviews method for this update. We did not replicate quality rating or data abstraction for original studies or replicate review-specific analyses. However, we decided a priori to conduct a de novo search for primary evidence related to the effectiveness and safety of ENDS. In addition, we did a bridge search for evidence related to pharmacotherapy interventions among pregnant women because of the limited number of studies included in the available systematic reviews and the length of time that had elapsed since their last search dates. We developed an analytic framework and 3 key questions with input from the USPSTF (Appendix Figure 1). The final version of the framework and key questions reflects both USPSTF and public input. The full report provides detailed methods (14). Appendix Figure 1. Analytic framework. KQ = key question. Data Sources and Searches We searched the following databases for relevant reviews from January 2009 to 1 August 2014: PubMed, PsycInfo, Cochrane Database of Systematic Reviews, Health Technology Assessment database, and Database of Abstracts of Reviews of Effects of the Centre for Reviews and Dissemination. We also searched the following organizational Web sites: the Agency for Healthcare Research and Quality, the British Medical Journal Clinical Evidence (through 7 August 2013), the Canadian Agency for Drugs and Technologies in Health, Guide to Community Preventive Services, the Institute of Medicine, the National Institute for Health and Clinical Excellence, the National Health Service Health Technology Assessment Programme, and the Surgeon General. We supplemented our searches with suggestions from experts. We searched PubMed for primary evidence related to ENDS through 1 March 2015 and for pharmacotherapy interventions among pregnant women through 15 August 2014 (the full report outlines the search strategies for these 2 searches [14]). Study Selection Two investigators independently reviewed all identified abstracts and dually reviewed full-text articles against prespecified eligibility criteria (14). We resolved disagreements through discussion. We included systematic reviewswith or without meta-analysisthat examined the effectiveness of interventions for tobacco cessation for adults, including pregnant women, and were linked to primary care or took place in a general adult population. We excluded nonsystematic meta-analyses and narrative reviews. We also excluded reviews that focused on reduction of tobacco harms, interventions for relapse prevention, or cessation medications that were not approved by the U.S. Food and Drug Administration as first-line medications for cessation (such as nortriptyline). We included only the most recent version of updated reviews. We outlined separate selection criteria when considering primary evidence related to ENDS and pharmacotherapy among pregnant women, as described in the full report (14). Data Extraction and Quality Assessment At least 2 independent reviewers rated the quality of all included systematic reviews using a slightly modified version of the Assessment of Multiple Systematic Reviews tool (15, 16) (see the full report for modifications and methods for determining the overall quality rating of individual reviews [14]). We excluded all poor-quality studies (17). One reviewer completed primary data abstraction, and a secondary reviewer checked all data for accuracy and completeness. Data Synthesis and Analysis When we found several fair- and good-quality reviews that met the inclusion criteria in a given population and intervention subgroup, we applied criteria (Appendix Table 1) to identify 1 or more reviews that represented the most current and applicable evidence to serve as the basis for the main findings (called primary reviews). We reviewed the remaining reviews for complementary or discordant findings. When we encountered discordant bodies of evidence, we sought explanations for these differences by examining the eligibility criteria and included studies within each review. Appendix Table 1. Criteria for Choosing the Primary Existing Systematic Reviews We used the pooled point estimates presented in the included reviews when appropriate. We did not reanalyze any of the individual study evidence. We evaluated the appropriateness of meta-analytic procedures and used our technical judgment to interpret pooled analyses accounting for limitations or concerns around heterogeneity, statistical approaches (18, 19), and other factors. Role of the Funding Source This review was funded by the Agency for Healthcare Research and Quality. Agency staff provided technical oversight for the project. Liaisons from the USPSTF helped resolve issues around the reviews scope but were not involved in its conduct. Results We reviewed 638 abstracts and 114 full-text reviews for possible inclusion (Appendix Figure 2). We identified 54 systematic reviews that met our eligibility criteria (2073), and 22 of these served as the basis for the primary findings (Table 1). In general, results across all included reviews were consistent within each population and intervention grouping. Our results are organized by outcomes and subcategories by population and interventions. Eleven of the 54 included reviews synthesized evidence on interventions among specific subpopulations of adults (such as persons with depression and young adults) that are not included here but appear in detail in the full report (14). Appendix Figure 2. Summary of evidence search and selection. * 2 studies included both adults and pregnant women. Reviews can be counted in multiple intervention areas. Table 1. Characteristics of Included Systematic Reviews (n =54), by Population, Intervention, and Last Search Date Behavioral Interventions Among Adults Eleven reviews served as primary reviews examining the effects of behavioral interventions for smoking cessation among the general adult population (Table 1) (21, 22, 31, 37, 55, 58, 60, 61, 67, 71, 78). Health and Cessation Outcomes Data on health outcomes after behavioral interventions were limited to 1 study (79) that was reported in 1 review (58) (Table 2). This study reported no statistically significant differences in rates of total mortality, coronary disease mortality, and lung cancer incidence and mortality at 20-year follow-up among men at high risk for cardiorespiratory disease (n=1445) (80). However, at 33-year follow-up, there were significantly fewer deaths from respiratory illnesses among participants who received an intervention than control participants (58). Table 2. Summary of Evidence for the General Adult Population Several behavioral interventions increased smoking cessation at 6 months or more, including physician- (58) and nurse-delivered (55) counseling interventions, tailored self-help print materials (37), and telephone counseling (60), when compared with minimal intervention or usual care (Table 2 and Appendix Table 2). Smokers who were offered cessation advice by a physician, for example, were 76% more likely to have quit at 6 months or more than those who received no advice or usual care (risk ratio [RR], 1.76 [95% CI, 1.58 to 1.96]; I 2=40%; 28 trials; n=22239) (58). Both minimal and intensive advice (>20 minutes, additional materials beyond a brochure, or >1 follow-up visit) showed statistically significant increases in cessation rates when compared with control participants who did not receive advice. Direct comparisons between intensive and minimal advice in 15 trials suggested that more intensive advice offered a significant advantage (RR, 1.37 [CI, 1.20 to 1.56]; I 2=32%; 15 trials; n=9775) (58). Appendix Table 2. Summary of Smoking Abstinence Results From Reviews of Behavioral Counseling and Pharmacotherapy Interventions for Smoking Cessation Among Adults, by Type of Intervention A separate meta-analysis of 38 randomized, controlled trials (RCTs) done among more than 15000 smokers found a small relative benefit of adjunctive behavioral support to pharmacotherapy when compared with pharmacotherapy alone (RR, 1.16 [CI, 1.09 to 1.24]) (61). Cessation rates were relatively high in both the intervention (21.4%) and control (18

Preeclampsia Screening: Evidence Report and Systematic Review for the US Preventive Services Task Force

Importance Preeclampsia is a complex disease of pregnancy with sometimes serious effects on maternal and infant morbidity and mortality. It is defined by hypertension after 20 weeks’ gestation and proteinuria or other evidence of multisystem involvement. Objective To systematically review the benefits and harms of preeclampsia screening and risk assessment for the US Preventive Services Task Force. Data Sources MEDLINE, PubMed, and Cochrane Central Register of Controlled Trials databases from 1990 through September 1, 2015. Surveillance for new evidence in targeted publications was conducted through October 5, 2016. Study Selection English-language trials and observational studies, including externally validated prediction models, of screening effectiveness, benefits, and harms from routine preeclampsia screening during pregnancy. Data Extraction and Synthesis Independent dual review of article abstracts and full texts against a priori inclusion criteria. Meta-analysis was not performed because of clinical and statistical heterogeneity of included studies. Main Outcomes and Measures Maternal and infant health outcomes, including eclampsia, stroke, stillbirth, preterm birth, and low birth weight; screening and risk prediction test performance; harms of screening and risk assessment. Results Twenty-one studies (13 982 participants) were included. No studies directly compared the effectiveness of preeclampsia screening in a screened population vs an unscreened population; 1 US trial (n = 2764) found no difference in benefits or harms with fewer prenatal visits but was underpowered for rare, serious outcomes. For harms, a before-after comparison cohort noninferiority study of urine protein screening for specific indications compared with routine screening (n = 1952) did not identify harms with fewer urine screening tests. Four studies (n = 7123) reported external validation performance of 16 risk prediction models, 5 of which had good or better discrimination (c statistic >0.80) for prediction of preeclampsia, and positive predictive values of 4% in the largest, most applicable validation cohorts. Calibration was not reported despite being a key model performance measure. There were no studies of urine screening test performance conducted in asymptomatic primary care populations; 14 studies of protein urine test performance among women being evaluated for suspected preeclampsia (n = 1888) had wide-ranging test accuracy (sensitivity, 22%-100%; specificity, 36%-100%) and high statistical and clinical heterogeneity in tests used, eligibility criteria, and proteinuria prevalence (8.7%-93.8%). Conclusions and Relevance Evidence to estimate benefits and harms of preeclampsia screening and the test performance of different screening approaches over the course of pregnancy was limited. Externally validated risk prediction models had limited applicability and lacked calibration and clinical implementation data needed to support routine use. Further research is needed to better inform risk-based screening approaches and improve screening strategies, given the complex pathophysiology and clinical unpredictability of preeclampsia.

An approach to addressing subpopulation considerations in systematic reviews: the experience of reviewers supporting the U.S. Preventive Services Task Force

BackgroundGuideline developers and other users of systematic reviews need information about whether a medical or preventive intervention is likely to benefit or harm some patients more (or less) than the average in order to make clinical practice recommendations tailored to these populations. However, guidance is lacking on how to include patient subpopulation considerations into the systematic reviews upon which guidelines are often based. In this article, we describe methods developed to consistently consider the evidence for relevant subpopulations in systematic reviews conducted to support primary care clinical preventive service recommendations made by the U.S. Preventive Services Task Force (USPSTF).Proposed approachOur approach is grounded in our experience conducting systematic reviews for the USPSTF and informed by a review of existing guidance on subgroup analysis and subpopulation issues. We developed and refined our approach based on feedback from the Subpopulation Workgroup of the USPSTF and pilot testing on reviews being conducted for the USPSTF. This paper provides processes and tools for incorporating evidence-based identification of important sources of potential heterogeneity of intervention effects into all phases of systematic reviews. Key components of our proposed approach include targeted literature searches and key informant interviews to identify the most important subpopulations a priori during topic scoping, a framework for assessing the credibility of subgroup analyses reported in studies, and structured investigation of sources of heterogeneity of intervention effects.ConclusionsFurther testing and evaluation are necessary to refine this proposed approach and demonstrate its utility to the producers and users of systematic reviews beyond the context of the USPSTF. Gaps in the evidence on important subpopulations identified by routinely applying this process in systematic reviews will also inform future research needs.

Supplemental Screening for Breast Cancer in Women With Dense Breasts: A Systematic Review for the U.S. Preventive Services Task ForceSupplemental Breast Cancer Screening in Women With Dense Breasts

Dense breasts are defined by mammographic appearance. The American College of Radiologys (ACRs) Breast Imaging Reporting and Data System (BI-RADS) classifies breasts as almost entirely fatty (BI-RADS category a), scattered areas of fibroglandular density (category b), heterogeneously dense (category c), or extremely dense (category d). About 27.6 million (43%) women aged 40 to 74 years in the United States have dense breasts; most of these are classified as category c (1). Higher breast density is associated with decreased mammographic sensitivity and specificity and also with increased breast cancer risk. The relative hazard of breast cancer for women with dense breasts ranged from 1.50 (women aged 65 to 74 years) to 1.83 (women aged 40 to 49 years) in an analysis of 1169248 women enrolled in the Breast Cancer Surveillance Consortium (unpublished data). Increased breast density has been associated with hormone replacement therapy use, younger age, and lower body mass index (2). Data on breast density and race or ethnicity are limited. In the United States, Asian women have higher breast density (3) but lower than average incidence of breast cancer (4). Increased breast density is not associated with higher breast cancer mortality among women with dense breasts diagnosed with breast cancer, after adjustment for stage and mode of detection (5). Supplemental breast cancer screening with additional screening modalities has been proposed to improve the early detection of breast cancers. No clinical guidelines explicitly recommend use of supplemental breast cancer screening on women with dense breasts (69), but as of September 2015, 24 states had enacted legislation requiring that women be notified of breast density with their mammography results; 9 more states are considering mandatory notification (10) (Appendix Table 1). Most states require specific language distinguishing dense (BI-RADS c and d) from nondense breasts, and 4 states require that insurers cover subsequent examinations and tests for women with dense breasts (1114). Federal legislation requiring breast density notification is pending (15). Appendix Table 1. Breast Density Legislation in the United States This report summarizes a systematic review of current evidence on the reproducibility of BI-RADS breast density determinations and on test performance characteristics and outcomes of supplemental screening of women with dense breasts by using hand-held ultrasonography (HHUS), automated whole-breast ultrasonography (ABUS), breast magnetic resonance imaging (MRI), and digital breast tomosynthesis (DBT). Mandatory reporting laws frame notification of women as dense/nondense, so this review focused on this categorization. Methods The review protocol included an analytic framework with 4 key questions (KQs) (Appendix Figure 1). Detailed methods, including search strategies, detailed inclusion criteria, and excluded studies, are available in the full evidence report (16). Appendix Figure 1. Analytic framework. BI-RADS = Breast Imaging Reporting and Data System; DCIS = ductal carcinoma in-situ; KQ = key question; MRI = magnetic resonance imaging. Data Sources and Searches MEDLINE, PubMed, EMBASE, and the Cochrane Library were searched for relevant English-language studies published between January 2000 and July 2015. We reviewed reference lists from retrieved articles and references suggested by experts. Study Selection Two investigators independently reviewed abstracts and full-text articles for inclusion according to predetermined criteria (E.P.W. and J.H.T. for KQ 1, J.M. and J.J.F. for KQs 2 to 4). Included studies examining the reproducibility of BI-RADS breast density categorization focused on asymptomatic women aged 40 years or older undergoing digital or film mammography. Included studies on supplemental screening with HHUS, ABUS, MRI, or DBT reported outcomes for asymptomatic women with dense breasts aged 40 years and older. In studies that focused primarily on women at high risk for breast cancer (including those with preexisting breast cancer or high-risk breast lesions [such as ductal carcinoma in situ, atypical hyperplasia, and lobular carcinoma in situ], BRCA mutations, familial breast cancer syndromes, or previous chest-wall radiation) and studies that included women with nondense breasts, we analyzed the relevant subset when available in the publication or provided by the authors. A priori inclusion criteria limited studies on BI-RADS reproducibility to fair- or good-quality randomized, controlled trials; cohort studies; or test sets involving multiple blind readings by at least 3 readers. Studies on test performance characteristics and outcomes of supplemental screening modalities were limited to fair- or good-quality randomized, controlled trials; cohort studies; or diagnostic accuracy studies with reference standards applied to all participants. We examined sensitivity, specificity, positive predictive values (PPVs), negative predictive values (NPVs), and available clinical outcomes (including cancer detection rates, recall rates, and biopsy rates). We defined recall as the need for any additional diagnostic testing after supplemental screening, including imaging and biopsy. Data Extraction and Quality Assessment Two investigators (E.P.W. and J.H.T. for KQ 1, J.M. and J.J.F. for KQs 2 to 4) critically appraised all included studies independently using the U.S. Preventive Services Task Forces (USPSTFs) design-specific criteria (17), supplemented with the National Institute for Health and Clinical Excellence methodology checklists (18) and the Quality Appraisal Tool for Studies of Diagnostic Reliability (19). According to USPSTF criteria, a good-quality study generally met all prespecified criteria; fair-quality studies did not meet all criteria but had no important limitations. Poor-quality studies had important limitations that could invalidate results (inadequate or biased application of reference standard; population limited to very high-risk patients). Data Synthesis and Analysis When available or provided by the authors, results of supplemental screening for subgroups of women with dense breasts were extracted; we excluded those with other risk factors for breast cancer. We calculated the sensitivity and specificity of the supplemental breast screening tests for women with negative mammography results. Only cancers detected by the supplemental test after negative mammography results and cancers found at interval follow-up were included. Hence, the values reported represent the sensitivity and specificity for detection of additional cancer in women with negative mammography findings. Similarly, we defined cancer detection rates, recall rates, and biopsy rates to include only those cancer cases, recalls, and biopsies related to supplemental screening after negative results on mammography. Meta-analysis was not performed because there were few good-quality studies. Role of the Funding Source This research was funded by the Agency for Healthcare Research and Quality (AHRQ) under a contract to support the work of the USPSTF. The investigators worked with USPSTF members to develop and refine the scope, analytic frameworks, and KQs. AHRQ had no role in study selection, quality assessment, synthesis, or development of conclusions. AHRQ provided project oversight; reviewed the draft report; and distributed the draft for peer review, including to representatives of professional societies and federal agencies. AHRQ performed a final review of the manuscript to ensure that the analysis met methodological standards. The investigators are solely responsible for the content and the decision to submit the manuscript for publication. Results The literature search yielded 2067 unique citations; 128 full-text articles considered potentially relevant were reviewed to identify 24 unique studies meeting inclusion criteria (Appendix Figure 2). Table 1 (2043) provides the characteristics of included studies. No studies addressed the effect of supplemental screening (compared with women without supplemental screening) on breast cancer morbidity or mortality. Appendix Figure 2. Summary of evidence search and selection. KQ = key question. Table 1. Characteristics of Included Studies Accuracy and Reliability of BI-RADS Density Determination Absent a gold standard for breast density, studies could not evaluate the accuracy of BI-RADS density determinations. Five studies reported repeated assignment of categorical BI-RADS breast density classification by the same or different radiologists, altogether including more than 440000 women, almost all with data from 2 sequential screening mammograms. To reflect current U.S. practice, we included only studies based on the BI-RADS density categories. The 3 largest studies were set in the United States. Two used data from the Breast Cancer Surveillance Consortium (20, 22), and the third presented findings from community radiologists conducting repeated readings of a large screening test set (24). Two other small studies (not discussed here) were based on mammographic screening programs in Spain (21) and Italy (23). All United Statesbased studies reflected community practice by use of clinical readings from community screening programs or test set readings by practicing community radiologists without additional training. Overall, group prevalence of BI-RADS density ratings was similar across initial and subsequent examinations among community radiologists (Appendix Table 2), but there was greater disagreement at the individual level. On subsequent screening examinations, approximately 1 in 5 women (23%) was placed in a different BI-RADS density category (a, b, c, d) by the same radiologist, while approximately 1 in 3 was categorized differently when a different radiologist read the subsequent examination result (Table 2). Considering clinical interpretations that combine categories (dense representing those with BI-RAD

Participatory Research to Advance Colon Cancer Prevention (PROMPT): Study protocol for a pragmatic trial

BACKGROUND Colon cancer is the second leading cause of cancer deaths in the United States. The Participatory Research to Advance Colon Cancer Prevention (PROMPT) study is a collaboration between two research institutions and a federally qualified health center (FQHC). The study seeks to raise colon cancer screening rates using a direct-mail fecal immunochemical testing (FIT) and reminder program in an FQHC serving a predominantly Latino population in California. METHODS PROMPT is a pragmatic trial enrolling 16 clinics. The study will test automated and live prompts (i.e., alerts, reminders) to a direct-mail FIT program in two phases. In Phase I, we tailored and defined intervention components for the pilot using a community-based participatory research approach called boot camp translation. We then plan to conduct a three-arm patient-randomized comparative effectiveness trial in two pilot clinics to compare 1) automated prompts, 2) live prompts, and 3) a combination of automated plus live prompts to alert and remind patients to complete screening. In Phase II, the adapted best practice intervention will be spread to additional clinics within the FQHC (estimated population 27,000) and assessed for effectiveness. Patient and staff interviews will be conducted to explore receptivity to the program and identify barriers to implementation. DISCUSSION This pragmatic trial applies innovative approaches to engage diverse stakeholders and will test the effectiveness and spread of a direct-mail plus reminder program. If successful, the program will provide a model for a cost-effective method to raise colon cancer screening rates among Latino patients receiving care in FQHCs. TRIAL REGISTRATION National Clinical Trial (NCT) Identifier NCT03167125.

Abstract A13: Using boot camp translation to design a system-based intervention to improve rates of colon cancer screening using fecal immunochemical testing among Latino patients in community health centers

Introduction: Colon cancer is the second-leading cause of cancer death in the United States, and screening rates are disproportionately low among Latinos. In 2015, only 63% of eligible adults, and 50% of Latinos, were up to date with colon cancer screening recommendations. One factor thought to contribute to the low screening rate is that patient-facing health information for Latinos is difficult to understand and patients face challenges in taking appropriate health action. As part of the Participatory Research to Advance Colon Cancer Prevention (PROMPT) study that seeks to raise rates of colon cancer screening in a Latino-serving community health center in California, we used boot camp translation (BCT), a validated community-based participatory strategy, to elicit input from diverse stakeholders and refine the messages and format of colon cancer screening reminders for a clinic-based direct mail fecal immunochemical test (FIT) program. Methods: We adapted BCT to engage Latino patients and clinic staff in this research. Eligible patient participants were Latino, aged 50 to 75 years, able to speak English or Spanish, and willing to participate in a 6-hour in-person meeting and three 30-minute follow-up phone calls. Materials were developed in English and Spanish, and separate sessions were held for English- and Spanish-speaking participants. As part of the in-person sessions, a national bilingual colon cancer expert delivered a presentation on colon health, cancer screening, and messages to improve screening participation, specific to Latino populations. Following the presentation, BCT experts facilitated brainstorming sessions to obtain feedback on the presented information, followed by an interactive small-group session where participants reviewed sample written materials and reminder messages using various modalities (e.g., text, letter, automated and live calls). We asked participants to consider what information about colon cancer screening is important to share with other patients, what the best methods are to share these messages, and the frequency with which these messages should be delivered to patients to encourage FIT completion. Participants then engaged in a hands-on exercise to share input about the timing and frequency of reminder delivery. Results from the exercise were used to define the intervention for the PROMPT pilot. Results: A total of 25 adults participated in an in-person session (12 in the English-language session; 13 in the Spanish-language session). Participants were mostly clinic patients (84%) and the majority were female (80%). Among the patient participants, 57% were enrolled in Medicaid, and 67% reported an annual household income of less than

Supplemental Screening for Breast Cancer in Women With Dense Breasts

20,000. Key themes from the sessions included increasing awareness about colon cancer and screening options, stressing the urgency of screening, and using personalized messages such as “I” statements in letters or human voices on automated phone call reminders. Both sessions noted the importance of receiving an alert (automated or live) before the FIT kit is mailed, and of receiving a reminder within 2 weeks of FIT kit mailing. Conclusions: Our BCT process allowed English- and Spanish-speaking Latino patients to directly inform which approaches get tested in the pilot study by refining message content and selecting their modality and timing to encourage patients who are mailed a FIT to complete it and mail it back. Using BCT, we successfully incorporated participant feedback to design culturally relevant health messages to promote FIT testing among patients served by community clinics. Citation Format: Jamie Thompson, Melinda Davis, LeAnn Michaels, Jennifer Rivelli, Marta Castro, Anne Escaron, Brittany Younger, Melissa Castillo, Sacha Reich, Gloria Coronado. Using boot camp translation to design a system-based intervention to improve rates of colon cancer screening using fecal immunochemical testing among Latino patients in community health centers [abstract]. In: Proceedings of the Tenth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2017 Sep 25-28; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2018;27(7 Suppl):Abstract nr A13.

Behavioral Counseling and Pharmacotherapy Interventions for Tobacco Cessation in Adults, Including Pregnant Women

Dense breasts are defined by mammographic appearance. The American College of Radiologys (ACRs) Breast Imaging Reporting and Data System (BI-RADS) classifies breasts as almost entirely fatty (BI-RADS category a), scattered areas of fibroglandular density (category b), heterogeneously dense (category c), or extremely dense (category d). About 27.6 million (43%) women aged 40 to 74 years in the United States have dense breasts; most of these are classified as category c (1). Higher breast density is associated with decreased mammographic sensitivity and specificity and also with increased breast cancer risk. The relative hazard of breast cancer for women with dense breasts ranged from 1.50 (women aged 65 to 74 years) to 1.83 (women aged 40 to 49 years) in an analysis of 1169248 women enrolled in the Breast Cancer Surveillance Consortium (unpublished data). Increased breast density has been associated with hormone replacement therapy use, younger age, and lower body mass index (2). Data on breast density and race or ethnicity are limited. In the United States, Asian women have higher breast density (3) but lower than average incidence of breast cancer (4). Increased breast density is not associated with higher breast cancer mortality among women with dense breasts diagnosed with breast cancer, after adjustment for stage and mode of detection (5). Supplemental breast cancer screening with additional screening modalities has been proposed to improve the early detection of breast cancers. No clinical guidelines explicitly recommend use of supplemental breast cancer screening on women with dense breasts (69), but as of September 2015, 24 states had enacted legislation requiring that women be notified of breast density with their mammography results; 9 more states are considering mandatory notification (10) (Appendix Table 1). Most states require specific language distinguishing dense (BI-RADS c and d) from nondense breasts, and 4 states require that insurers cover subsequent examinations and tests for women with dense breasts (1114). Federal legislation requiring breast density notification is pending (15). Appendix Table 1. Breast Density Legislation in the United States This report summarizes a systematic review of current evidence on the reproducibility of BI-RADS breast density determinations and on test performance characteristics and outcomes of supplemental screening of women with dense breasts by using hand-held ultrasonography (HHUS), automated whole-breast ultrasonography (ABUS), breast magnetic resonance imaging (MRI), and digital breast tomosynthesis (DBT). Mandatory reporting laws frame notification of women as dense/nondense, so this review focused on this categorization. Methods The review protocol included an analytic framework with 4 key questions (KQs) (Appendix Figure 1). Detailed methods, including search strategies, detailed inclusion criteria, and excluded studies, are available in the full evidence report (16). Appendix Figure 1. Analytic framework. BI-RADS = Breast Imaging Reporting and Data System; DCIS = ductal carcinoma in-situ; KQ = key question; MRI = magnetic resonance imaging. Data Sources and Searches MEDLINE, PubMed, EMBASE, and the Cochrane Library were searched for relevant English-language studies published between January 2000 and July 2015. We reviewed reference lists from retrieved articles and references suggested by experts. Study Selection Two investigators independently reviewed abstracts and full-text articles for inclusion according to predetermined criteria (E.P.W. and J.H.T. for KQ 1, J.M. and J.J.F. for KQs 2 to 4). Included studies examining the reproducibility of BI-RADS breast density categorization focused on asymptomatic women aged 40 years or older undergoing digital or film mammography. Included studies on supplemental screening with HHUS, ABUS, MRI, or DBT reported outcomes for asymptomatic women with dense breasts aged 40 years and older. In studies that focused primarily on women at high risk for breast cancer (including those with preexisting breast cancer or high-risk breast lesions [such as ductal carcinoma in situ, atypical hyperplasia, and lobular carcinoma in situ], BRCA mutations, familial breast cancer syndromes, or previous chest-wall radiation) and studies that included women with nondense breasts, we analyzed the relevant subset when available in the publication or provided by the authors. A priori inclusion criteria limited studies on BI-RADS reproducibility to fair- or good-quality randomized, controlled trials; cohort studies; or test sets involving multiple blind readings by at least 3 readers. Studies on test performance characteristics and outcomes of supplemental screening modalities were limited to fair- or good-quality randomized, controlled trials; cohort studies; or diagnostic accuracy studies with reference standards applied to all participants. We examined sensitivity, specificity, positive predictive values (PPVs), negative predictive values (NPVs), and available clinical outcomes (including cancer detection rates, recall rates, and biopsy rates). We defined recall as the need for any additional diagnostic testing after supplemental screening, including imaging and biopsy. Data Extraction and Quality Assessment Two investigators (E.P.W. and J.H.T. for KQ 1, J.M. and J.J.F. for KQs 2 to 4) critically appraised all included studies independently using the U.S. Preventive Services Task Forces (USPSTFs) design-specific criteria (17), supplemented with the National Institute for Health and Clinical Excellence methodology checklists (18) and the Quality Appraisal Tool for Studies of Diagnostic Reliability (19). According to USPSTF criteria, a good-quality study generally met all prespecified criteria; fair-quality studies did not meet all criteria but had no important limitations. Poor-quality studies had important limitations that could invalidate results (inadequate or biased application of reference standard; population limited to very high-risk patients). Data Synthesis and Analysis When available or provided by the authors, results of supplemental screening for subgroups of women with dense breasts were extracted; we excluded those with other risk factors for breast cancer. We calculated the sensitivity and specificity of the supplemental breast screening tests for women with negative mammography results. Only cancers detected by the supplemental test after negative mammography results and cancers found at interval follow-up were included. Hence, the values reported represent the sensitivity and specificity for detection of additional cancer in women with negative mammography findings. Similarly, we defined cancer detection rates, recall rates, and biopsy rates to include only those cancer cases, recalls, and biopsies related to supplemental screening after negative results on mammography. Meta-analysis was not performed because there were few good-quality studies. Role of the Funding Source This research was funded by the Agency for Healthcare Research and Quality (AHRQ) under a contract to support the work of the USPSTF. The investigators worked with USPSTF members to develop and refine the scope, analytic frameworks, and KQs. AHRQ had no role in study selection, quality assessment, synthesis, or development of conclusions. AHRQ provided project oversight; reviewed the draft report; and distributed the draft for peer review, including to representatives of professional societies and federal agencies. AHRQ performed a final review of the manuscript to ensure that the analysis met methodological standards. The investigators are solely responsible for the content and the decision to submit the manuscript for publication. Results The literature search yielded 2067 unique citations; 128 full-text articles considered potentially relevant were reviewed to identify 24 unique studies meeting inclusion criteria (Appendix Figure 2). Table 1 (2043) provides the characteristics of included studies. No studies addressed the effect of supplemental screening (compared with women without supplemental screening) on breast cancer morbidity or mortality. Appendix Figure 2. Summary of evidence search and selection. KQ = key question. Table 1. Characteristics of Included Studies Accuracy and Reliability of BI-RADS Density Determination Absent a gold standard for breast density, studies could not evaluate the accuracy of BI-RADS density determinations. Five studies reported repeated assignment of categorical BI-RADS breast density classification by the same or different radiologists, altogether including more than 440000 women, almost all with data from 2 sequential screening mammograms. To reflect current U.S. practice, we included only studies based on the BI-RADS density categories. The 3 largest studies were set in the United States. Two used data from the Breast Cancer Surveillance Consortium (20, 22), and the third presented findings from community radiologists conducting repeated readings of a large screening test set (24). Two other small studies (not discussed here) were based on mammographic screening programs in Spain (21) and Italy (23). All United Statesbased studies reflected community practice by use of clinical readings from community screening programs or test set readings by practicing community radiologists without additional training. Overall, group prevalence of BI-RADS density ratings was similar across initial and subsequent examinations among community radiologists (Appendix Table 2), but there was greater disagreement at the individual level. On subsequent screening examinations, approximately 1 in 5 women (23%) was placed in a different BI-RADS density category (a, b, c, d) by the same radiologist, while approximately 1 in 3 was categorized differently when a different radiologist read the subsequent examination result (Table 2). Considering clinical interpretations that combine categories (dense representing those with BI-RAD