Engaging stakeholders in pharmacoepidemiology research: Current state and recommendations.

PURPOSE: Given current efforts to enhance patient-centered care and shared decision-making, the International Society of Pharmacoepidemiology Workgroup on Patient Engagement assessed patient and other stakeholder engagement in pharmacoepidemiology research and provides recommendations for the field. METHODS: A systematic review used MEDLINE and EMBASE to identify published literature from 2005 to 2016 addressing how stakeholders-patients, caregivers, and others-assisted researchers conducting pharmacoepidemiologic research. Three pairs of Workgroup members screened titles and abstracts to select articles for full-text review and analysis. Two Workgroup members abstracted the following data: research focus, characterization and role of stakeholders, and type(s) of engagement strategy employed. Data were summarized descriptively. RESULTS: We identified 5717 references for abstract screening. Of these, 69 met the criteria for full-text screening, and 11 were selected for data abstraction. Of these 11 studies, seven focused on the development of a research agenda and eight had stakeholders react or advise on an aspect of the study. Although patients were the most commonly identified stakeholders, advocacy groups and health care professionals were also frequently identified. Some studies reported the engagement of other stakeholders, including local government or policy experts. Engagement strategies varied, with five studies using more than one strategy. Studies often did not indicate the involvement of stakeholders in developing the study design or with implementation. CONCLUSIONS: Currently, few pharmacoepidemiology publications mention patient or other stakeholder engagement in the design, analysis, or reporting of research. This suggests that there are opportunities to expand stakeholder engagement and/or increase the transparency of reporting stakeholder engagement.

Underutilization and disparities in access to EGFR testing among Medicare patients with lung cancer from 2010 - 2013.

BACKGROUND: Tumor testing for mutations in the epidermal growth factor receptor (EGFR) gene is indicated for all newly diagnosed, metastatic lung cancer patients, who may be candidates for first-line treatment with an EGFR tyrosine kinase inhibitor. Few studies have analyzed population-level testing. METHODS: We identified clinical, demographic, and regional predictors of EGFR & KRAS testing among Medicare beneficiaries with a new diagnosis of lung cancer in 2011-2013 claims. The outcome variable was whether the patient underwent molecular, EGFR and KRAS testing. Independent variables included: patient demographics, Medicaid status, clinical characteristics, and region where the patient lived. We performed multivariate logistic regression to identify factors that predicted testing. RESULTS: From 2011 to 2013, there was a 19.7% increase in the rate of EGFR testing. Patient zip code had the greatest impact on odds to undergo testing; for example, patients who lived in the Boston, Massachusetts hospital referral region were the most likely to be tested (odds ratio (OR) of 4.94, with a 95% confidence interval (CI) of 1.67-14.62). Patient demographics also impacted odds to be tested. Asian/Pacific Islanders were most likely to be tested (OR 1.63, CI 1.53-1.79). Minorities and Medicaid patients were less likely to be tested. Medicaid recipients had an OR of 0.74 (CI 0.72-0.77). Hispanics and Blacks were also less likely to be tested (OR 0.97, CI 0.78-0.99 and 0.95, CI 0.92-0.99), respectively. Clinical procedures were also correlated with testing. Patients who underwent transcatheter biopsies were 2.54 times more likely to be tested (CI 2.49-2.60) than those who did not undergo this type of biopsy. CONCLUSIONS: Despite an overall increase in EGFR testing, there is widespread underutilization of guideline-recommended testing. We observed racial, income, and regional disparities in testing. Precision medicine has increased the complexity of cancer diagnosis and treatment. Targeted interventions and clinical decision support tools are needed to ensure that all patients are benefitting from advances in precision medicine. Without such interventions, precision medicine may exacerbate racial disparities in cancer care and health outcomes.

User testing of an adaptation of fishbone diagrams to depict results of systematic reviews.

BACKGROUND: Summary of findings tables in systematic reviews are highly informative but require epidemiological training to be interpreted correctly. The usage of fishbone diagrams as graphical displays could offer researchers an effective approach to simplify content for readers with limited epidemiological training. In this paper we demonstrate how fishbone diagrams can be applied to systematic reviews and present the results of an initial user testing. METHODS: Findings from two systematic reviews were graphically depicted in the form of the fishbone diagram. To test the utility of fishbone diagrams compared with summary of findings tables, we developed and pilot-tested an online survey using Qualtrics. Respondents were randomized to the fishbone diagram or a summary of findings table presenting the same body of evidence. They answered questions in both open-ended and closed-answer formats; all responses were anonymous. Measures of interest focused on first and second impressions, the ability to find and interpret critical information, as well as user experience with both displays. We asked respondents about the perceived utility of fishbone diagrams compared to summary of findings tables. We analyzed quantitative data by conducting t-tests and comparing descriptive statistics. RESULTS: Based on real world systematic reviews, we provide two different fishbone diagrams to show how they might be used to display complex information in a clear and succinct manner. User testing on 77 students with basic epidemiological training revealed that participants preferred summary of findings tables over fishbone diagrams. Significantly more participants liked the summary of findings table than the fishbone diagram (71.8% vs. 44.8%; p < .01); significantly more participants found the fishbone diagram confusing (63.2% vs. 35.9%, p < .05) or indicated that it was difficult to find information (65.8% vs. 45%; p < .01). However, more than half of the participants in both groups were unable to find critical information and answer three respective questions correctly (52.6% in the fishbone group; 51.3% in the summary of findings group). CONCLUSIONS: Fishbone diagrams are compact visualizations that, theoretically, may prove useful for summarizing the findings of systematic reviews. Initial user testing, however, did not support the utility of such graphical displays.

Preferences for patient medication information: what do patients want?

This study investigated respondent preferences on how best to display patient medication information (PMI) that accompanies prescription medications to promote comprehension and appropriate usage. The authors identified 30 individuals diagnosed with select immune disorders, 30 with other chronic diseases, and 30 from the general public and had them review one of two PMI handouts that varied by format, organization, and content. The authors explored preferences for the PMI handout using one-on-one interviews. The authors analyzed the qualitative data to identify relevant themes and patterns using NVivo9 qualitative software. The majority of respondents noted that the formats of the two PMI handouts were more informative than those they currently receive from the pharmacist, with a preference for the 2-column, segmented design. However, respondent PMI preferences varied by age, education, and health status. Patients need simpler and more concise drug information to make better decisions about their health. Current PMI handouts are dense and complex, which can be confusing and not reader friendly. To improve PMI understandability and usefulness, the U.S. Food and Drug Administration is working with stakeholders, consumer advocates, and academics. Findings from this study may help inform future development of more user-friendly PMI.

Communicating quantitative risks and benefits in promotional prescription drug labeling or print advertising.

PURPOSE: Under the Food, Drug, and Cosmetic Act, all promotional materials for prescription drugs must strike a fair balance in presentation of risks and benefits. How to best present this information is not clear. We sought to determine if the presentation of quantitative risk and benefit information in drug advertising and labeling influences consumers', patients', and clinicians' information processing, knowledge, and behavior by assessing available empirical evidence. METHODS: We used PubMed for a literature search, limiting to articles published in English from 1990 forward. Two reviewers independently reviewed the titles and abstracts for inclusion, after which we reviewed the full texts to determine if they communicated risk/benefit information either: (i) numerically (e.g., percent) versus non-numerically (e.g., using text such as "increased risk") or (ii) numerically using different formats (e.g., "25% of patients", "one in four patients", or use of pictographs). We abstracted information from included articles into standardized evidence tables. The research team identified a total of 674 relevant publications, of which 52 met our inclusion criteria. Of these, 37 focused on drugs. RESULTS AND CONCLUSIONS: Presenting numeric information appears to improve understanding of risks and benefits relative to non-numeric presentation; presenting both numeric and non-numeric information when possible may be best practice. No single specific format or graphical approach emerged as consistently superior. Numeracy and health literacy also deserve more empirical attention as moderators.

Comparative effectiveness research using electronic health records: impacts of oral antidiabetic drugs on the development of chronic kidney disease.

PURPOSE: Little is known about the comparative effects of common oral antidiabetic drugs ([OADs] metformin, sulfonylureas, or thiazolidinediones [THZs]) on chronic kidney disease (CKD) outcomes in patients newly diagnosed with type 2 diabetes (T2DM) and followed in community primary care practices. Electronic health records (EHRs) were used to evaluate the relationships between OAD class use and incident proteinuria and prevention of glomerular filtration rate decline. METHODS: A retrospective cohort study on newly diagnosed T2D cases requiring OADs documented in the EHRs of two primary care networks between 1998 and 2009 was conducted. CKD outcomes were new-onset proteinuria and estimated GFR (eGFR) falling below 60 ml/min/1.73 m(2). OAD exposures defined cohorts. Hazard ratios represent differential CKD outcome risk per year of OAD class use. RESULTS: A total of 798 and 977 patients qualified for proteinuria and eGFR outcome analyses, respectively. With metformin as the reference group, sulfonylurea exposure trended toward association with an increased risk of developing proteinuria ([adjusted hazard ratio; 95% CI] 1.27; 0.93, 1.74); proteinuria risk associated with THZ exposure (1.00; 0.70, 1.42) was similar to metformin. Compared with metformin, sulfonylurea exposure was associated with an increased risk of eGFR reduction to <60 ml/min/1.73 m(2) (1.41; 1.05, 1.91). THZ exposure (1.04; 0.71, 1.50) was not associated with change in the risk of eGFR decline. CONCLUSIONS: In a primary care population, metformin appeared to decrease the risk of CKD development compared with sulfonlyureas; risks of CKD development between metformin and THZs were similar. EHR use in pharmacotherapy comparative effectiveness research creates specific challenges and study limitations.

The incident user design in comparative effectiveness research.

Comparative effectiveness research includes cohort studies and registries of interventions. When investigators design such studies, how important is it to follow patients from the day they initiated treatment with the study interventions? Our article considers this question and related issues to start a dialogue on the value of the incident user design in comparative effectiveness research. By incident user design, we mean a study that sets the cohort's inception date according to patients' new use of an intervention. In contrast, most epidemiologic studies enroll patients who were currently or recently using an intervention when follow-up began. We take the incident user design as a reasonable default strategy because it reduces biases that can impact non-randomized studies, especially when investigators use healthcare databases. We review case studies where investigators have explored the consequences of designing a cohort study by restricting to incident users, but most of the discussion has been informed by expert opinion, not by systematic evidence.

Initial and subsequent therapy for newly diagnosed type 2 diabetes patients treated in primary care using data from a vendor-based electronic health record.

BACKGROUND: Diabetes is a leading cause of death and disability, and its prevalence is increasing. When diet fails, patients with type 2 diabetes mellitus (T2DM) are prescribed oral hypoglycemics for glycemic control. Few studies have explored initial use or change from initial oral hypoglycemic therapy in the primary care setting. We aimed to describe the utilization of initial oral hypoglycemics among newly diagnosed patients with diabetes from 1998-2009 and changes from initial to subsequent therapy among patients prescribed older oral hypoglycemic agents using electronic health records. METHODS: This observational cohort study used electronic health records from newly diagnosed patients with T2DM between 1 January 1998 and 31 March 2009 at two large health systems in the USA. Oral hypoglycemics included older (biguanide, sulfonylurea, and thiazolidinedione) and newer agents (incretin mimetic agents, alpha-glucosidase inhibitors, and D-phenylalanine derivatives). Multinomial regression models were fit to evaluate initial older oral hypoglycemic medication. We used incidence density sampling and conditional logistic regression models to evaluate predictors of regimen change. RESULTS: Most patients were treated from the biguanide class of oral hypoglycemics (67%), but there were differences in initial prescribing by age and race. HbA1c (Odds Ratio for HbA1c 7.0-8.9 vs < 7.0, 5.87 [95% Confidence Interval: 3.62-9.52]; Odds Ratio for HbA1c ≥ 9 vs < 7.0, 20.25 [95% Confidence Interval: 8.32-49.29] and Black people (Odds Ratio, 0.29 [95% Confidence Interval: 0.14, 0.60]) versus White people were associated with regimen change in the adjusted analysis. CONCLUSIONS: Clinical and demographic characteristics influence choice and duration of initial oral hypoglycemic treatment as well as regimen changes.

Clinical heterogeneity in systematic reviews and health technology assessments: synthesis of guidance documents and the literature.

OBJECTIVES: The aim of this study was to synthesize best practices for addressing clinical heterogeneity in systematic reviews and health technology assessments (HTAs). METHODS: We abstracted information from guidance documents and methods manuals made available by international organizations that develop systematic reviews and HTAs. We searched PubMed® to identify studies on clinical heterogeneity and subgroup analysis. Two authors independently abstracted and assessed relevant information. RESULTS: Methods manuals offer various definitions of clinical heterogeneity. In essence, clinical heterogeneity is considered variability in study population characteristics, interventions, and outcomes across studies. It can lead to effect-measure modification or statistical heterogeneity, which is defined as variability in estimated treatment effects beyond what would be expected by random error alone. Clinical and statistical heterogeneity are closely intertwined but they do not have a one-to-one relationship. The presence of statistical heterogeneity does not necessarily indicate that clinical heterogeneity is the causal factor. Methodological heterogeneity, biases, and random error can also cause statistical heterogeneity, alone or in combination with clinical heterogeneity. CONCLUSIONS: Identifying potential modifiers of treatment effects (i.e., effect-measure modifiers) is important for researchers conducting systematic reviews and HTAs. Recognizing clinical heterogeneity and clarifying its implications helps decision makers to identify patients and patient populations who benefit the most, who benefit the least, and who are at greatest risk of experiencing adverse outcomes from a particular intervention.

The identification of pregnancies within the general practice research database.

BACKGROUND: The United States is moving toward active drug safety surveillance using sources such as administrative claims and electronic medical records, but use of these data for studying teratogenicity has been challenging, as they typically do not allow for the easy identification of pregnancies. Our goal was to develop and validate an algorithm for the identification of pregnancies in the general practice research database (GPRD) that could be used to study pregnancy outcomes. METHODS: The algorithm identified pregnancies in women 15-45-year-old that were pregnant between 1 January 1987 and 31 December 2006. We identified live births, stillbirths, and spontaneous and elective terminations within a woman's record. We validated the algorithm using the additional clinical details maternity (ACDM) file and de-identified free-text records. RESULTS: We analyzed 16,035,394 records from 3,093,927 individuals and identified 383,184 women who had a total of 580,356 pregnancies. There were 415,221 full-term live births, 3080 pre- or post-term births, 1834 multi-fetus deliveries, 86,408 spontaneous abortions or miscarriages, 72 164 elective terminations, and 1649 stillbirths or fetal deaths. A marker of pregnancy care was identifiable for 86.3% of the 580,356 pregnancies. The internal validation steps indicated that the algorithm produced consistent results with the ACDM file. CONCLUSIONS: We were successful in identifying a large number of pregnancies in the GPRD. Our use of a hierarchical approach to identify pregnancy outcomes builds upon the methods suggested in previous work, while implementing additional steps to minimize potential misclassification of pregnancy outcomes.

The adverse effects of comorbid pain on depression outcomes in primary care patients: results from the ARTIST trial.

OBJECTIVES: To explore the effect of pain symptoms and improvements in pain on depression outcomes. METHODS: We analyzed data from A Randomized Trial Investigating SSRI Treatment (ARTIST), a randomized longitudinal effectiveness study comparing selective serotonin reuptake inhibitors (SSRIs) for the treatment of depression in primary care (n = 573). Depression outcome at month 6, defined as remission, partial response, and nonresponse using the Symptom Checklist-20, was the primary outcome. RESULTS: Compared to patients with no pain at baseline, those with severe pain were less likely to achieve remission (OR = 0.11, 95% CI 0.05-0.25) and partial response (OR = 0.24, 95% CI 0.10-0.59) vs nonresponse. Patients with moderate pain were less likely to achieve remission vs nonresponse (OR = 0.25, 95% CI 0.13-0.48). Patients with early improvement in pain were more likely to achieve remission (OR = 1.90, 95% CI 1.03-3.49). Accounting for missing data with last observation carried forward or multiple imputation yielded similar results. CONCLUSION: Pain symptoms are present in the majority of depressed primary care patients beginning antidepressant therapy. Pain symptoms are associated with worse depression outcomes, while improvement in pain is associated with significantly better depression outcomes. Attention to comorbid pain may be important in enhancing depression care.

The DRD3 rs6280 polymorphism and prevalence of tardive dyskinesia: a meta-analysis.

To elucidate a widely suspected but inconclusive association between rs6280 in the dopamine receptor 3 gene (DRD3) and prevalence of tardive dyskinesia (TD), we conducted a meta-analysis of studies obtained in a systematic search of several bibliographic systems. We conducted several analyses of funnel plot asymmetry, overall heterogeneity, and study characteristics in analyses analogous to general, dominant and recessive inheritance models with the prevalence odds ratio (POR) as the measure of association. Thirteen eligible studies were identified with publication dates between 1997 and 2008. Evidence of funnel plot asymmetry was discerned in the dominant and general model analyses, but not in the recessive model analysis. Stratified analyses indicated that publication year, TD assessment method (Schooler-Kane criteria or other) and TD assessment frequency (single or repeated) were important study characteristics associated with heterogeneous PORs across studies. Studies conducted among patients with older age, fewer women or European (compared with Asian) ancestry reported stronger average PORs. Summary POR estimates under the dominant and general inheritance models were not warranted due to funnel plot asymmetry and heterogeneity. Under the recessive model, the summary estimate was POR = 0.93 (95% confidence interval: 0.70-1.23). We conclude that there is no or little association between DRD3 rs6280 polymorphisms and prevalence of TD.

Did Arkansas' Medicaid Patient-Centered Medical Home Program Have Spillover Effects on Commercially Insured Enrollees?

Patient-centered medical homes are increasingly being implemented by state Medicaid programs to incentivize high-quality, coordinated care and ultimately lower health care spending. This study examined whether the Arkansas Medicaid Patient-Centered Medical Home Program's practice-wide transformation activities had spillover effects on commercial beneficiaries. We used difference-in-differences to compare utilization and expenditures of commercially insured enrollees as their practices received Medicaid patient-centered medical home certification on a rolling basis between 2014 and 2016. We found a 5.7% increase in outpatient visits and 13% higher expenditures among early adopting practices. Even without associated reductions in costly emergency department visits or inpatient hospital admissions, decisionmakers should not lose sight of the potential value of increased engagement in and coordination of professional services for a population with high unmet health needs. Our results also emphasize that states can leverage Medicaid to spur system-wide transformation, and the investments generate spillover effects beyond those covered directly by Medicaid.

Early impact of the implementation of Medicaid episode-based payment reforms in Arkansas.

OBJECTIVE: To evaluate episode-based payments for upper respiratory tract infections (URI) and perinatal care in Arkansas's Medicaid population. STUDY SETTING: Upper respiratory infection and perinatal episodes among Medicaid-covered individuals in Arkansas and comparison states from fiscal year (FY) 2011 to 2014. STUDY DESIGN: Cross-sectional observational analysis using a difference-in-difference design to examine outcomes associated with URI and perinatal episodes of care (EOC) from 2011 to 2014. Key dependent variables include antibiotic use, emergency department visits, physician visits, hospitalizations, readmission, and preventive screenings. DATA COLLECTION: Claims data from the Medicaid Analytic Extract for Arkansas, Mississippi, and Missouri from 2010 to 2014 with supplemental county-level data from the Area Health Resource File (AHRF). PRINCIPAL FINDINGS: The URI EOC reduced the probability of antibiotic use (marginal effect [ME] = -1.8, 90% CI: -2.2, -1.4), physician visits (ME = 0.6, 90% CI: -0.8, -0.4), improved the probability of strep tests for children diagnosed with pharyngitis (ME = 9.4, 90% CI: 8.5, 10.3), but also increased the probability of an emergency department (ED) visit (ME = 0.1, 90% CI: 0.1, 0.2), relative to the comparison group. For perinatal EOCs, we found a reduced probability of an ED visit during pregnancy (ME = 0.1, 90% CI: -0.2, -0.0), an increased probability of screening for HIV (ME = 6.2, 90% CI: 4.0, 8.5), chlamydia (ME = 9.5, 90% CI: 7.2, 11.8), and group B strep-test (ME = 2.6, 90% CI: 0.5, 4.6), relative to the comparison group. Predelivery and postpartum hospitalizations also increased (ME = 1.2, 90% CI: 0.4, 2.0; ME = 0.4, 90% CI: 0.0, 0.8, respectively), relative to the comparison group. CONCLUSION: Upper respiratory infection and perinatal EOCs for Arkansas Medicaid beneficiaries produced mixed results. Aligning shared savings with quality metrics and cost-thresholds may help achieve quality targets and disincentivize over utilization within the EOC, but may also have unintended consequences.

IGF-I and IGFBP-3 polymorphisms in relation to circulating levels among African American and Caucasian women.

Circulating insulin-like growth factor-I (IGF-I) and IGF-binding protein-3 (IGFBP-3) levels have been associated with common diseases. Although family-based studies suggest that genetic variation contributes to circulating IGF-I and IGFBP-3 levels, analyses of associations with multiple IGF-I and IGFBP-3 single nucleotide polymorphisms (SNP) have been limited, especially among African Americans. We evaluated 30 IGF-I and 15 IGFBP-3 SNPs and estimated diplotypes in association with plasma IGF-I and IGFBP-3 among 984 premenopausal African American and Caucasian women. In both races, IGFBP-3 rs2854746 (Ala32Gly) was positively associated with plasma IGFBP-3 (CC versus GG mean difference among Caucasians, 631 ng/mL; 95% confidence interval, 398-864; African Americans, 897 ng/mL; 95% confidence interval, 656-1,138), and IGFBP-3 diplotypes with the rs2854746 GG genotype had lower mean IGFBP-3 levels than reference diplotypes with the CG genotype, whereas IGFBP-3 diplotypes with the CC genotype had higher mean IGFBP-3 levels. IGFBP-3 rs2854744 (-202 A/C) was in strong linkage disequilibrium with rs2854746 in Caucasians only, but was associated with plasma IGFBP-3 in both races. Eight additional IGFBP-3 SNPs were associated with >or=5% differences in mean IGFBP-3 levels, with generally consistent associations between races. Twelve IGF-I SNPs were associated with >or=10% differences in mean IGF-I levels, but associations were generally discordant between races. Diplotype associations with plasma IGF-I did not parallel IGF-I SNP associations. Our study supports that common IGFBP-3 SNPs, especially rs2854746, influence plasma IGFBP-3 levels among African Americans and Caucasians but provides less evidence that IGF-I SNPs affect plasma IGF-I levels.

Hypoactive sexual desire disorder in postmenopausal women: quality of life and health burden.

OBJECTIVES: To describe the health-related quality of life (HRQOL) implications of hypoactive sexual desire disorder (HSDD) in a national sample of postmenopausal women ages 30­70. METHODS: The Nationwide Survey of Female Sexual Health, a random-digit telephone survey of US households, collected information on female sexual function, demographic characteristics, HRQOL, and the presence of specific medical disorders from 1189 naturally or surgically postmenopausal women in stable relationships of ≥3 months duration. HSDD was defined as <40 on the Profile of Female Sexual Function© scale and <60 on the Personal Distress Scale©. Short Form-12 Health Survey (SF-12) summary and domain scores, and EuroQol (EQ-5D) index score and dimensions were compared with population-based norms for healthy individuals and selected chronic conditions. RESULTS: HSDD was associated with significant HRQOL decrements, with the largest SF-12 score differences in mental health (HSDD: 45.4 [standard error 1.9] vs. no HSDD: 51.0 [0.6], P < 0.01), vitality (HSDD: 47.7 [1.3] vs. no HSDD: 52.0 [0.7], P < 0.01), social function (HSDD: 47.3 [1.4] vs. no HSDD: 50.9 [0.7], P < 0.05), and bodily pain (HSDD: 41.4 [2.2] vs. no HSDD: 46.7 [0.9], P < 0.05). EQ-5D index was 0.08 points lower (HSDD: 0.76 [0.03] vs. no HSDD: 0.84 [0.02], P < 0.05) for those with HSDD compared with those without. HSDD was associated with a 0.1-point decrement in naturally menopausal women (HSDD: 0.78 [0.03] vs. no HSDD 0.88 [0.01], P < 0.01). Women with HSDD showed more HRQOL impairment than healthy population norms but were similar to adults with other chronic conditions such as diabetes and back pain. CONCLUSIONS: Women with HSDD showed substantial impairment in HRQOL. Given a prevalence of 6.6% to 12.5% among US women, HSDD represents an important burden on quality of life.

Reflections on the use of electronic health record data for clinical research.

The adoption of electronic health records (EHRs) offers the potential to improve the delivery, quality, and continuity of clinical care, but widespread use has not yet occurred. In this article, we describe our use of clinical (production) data that were derived from outpatient and inpatient visits at a university teaching hospital for clinical research, a use for which the data and their structure were not originally designed. Similar data exist at many outpatient and inpatient clinical facilities, and we believe that our insights are relevant to electronically captured medical data regardless of their origin. We describe the approaches taken to ensure compliance with the Health Insurance Portability and Accountability Act (HIPAA) and to leverage the vast stores of structured and unstructured data that are currently underused. We conclude by reflecting on what we would have done differently and by making recommendations to streamline the process.

Unintended Effects of Communicating About Drug Safety Issues: A Critical Review of the Literature.

Communications about the safety and effectiveness of human drugs can influence patients' and prescribers' perceptions and behaviors, which in turn can affect the public's health more broadly. We conducted a critical review of the literature on the unintended effects from communicating information to the public about safety issues with prescription and over-the-counter drugs. We searched PubMed for peer-reviewed studies published from 1990 to 2017 where study authors reported probable unintended effects of communicating drug safety. The types of communications included in these studies were news reports, direct-to-consumer advertisements, and those released by government agencies. Among the 26 studies identified, the most commonly reported unintended effects were decreased drug use or discontinuation. Other unintended effects included spillover to populations not targeted by the communications (e.g., discontinuation of antidepressants among adults following communications concerning use among youth), shifts in clinical diagnoses (e.g., fewer diagnoses of depression), increased use of alternative therapies, and other undesirable behaviors (e.g., possible increased suicide attempts because antidepressants were discontinued). Limitations to the literature include the inability to establish causation or to isolate the effects of multiple communication sources and messages. Further, because the intended effect of many communications was not known, our study was limited by challenges in defining some effects as unintended. Most studies used health insurer claims data to identify unintended effects of communications, which provide an incomplete picture; few used self-reported or other methodologies that could help illuminate the reasons underlying the effects observed in the claims data. Best practices for communicating about the potential benefits and harms of drugs in a manner that minimizes negative unintended effects are needed to protect and improve public health.

Comparing record linkage software programs and algorithms using real-world data.

Linkage of medical databases, including insurer claims and electronic health records (EHRs), is increasingly common. However, few studies have investigated the behavior and output of linkage software. To determine how linkage quality is affected by different algorithms, blocking variables, methods for string matching and weight determination, and decision rules, we compared the performance of 4 nonproprietary linkage software packages linking patient identifiers from noninteroperable inpatient and outpatient EHRs. We linked datasets using first and last name, gender, and date of birth (DOB). We evaluated DOB and year of birth (YOB) as blocking variables and used exact and inexact matching methods. We compared the weights assigned to record pairs and evaluated how matching weights corresponded to a gold standard, medical record number. Deduplicated datasets contained 69,523 inpatient and 176,154 outpatient records, respectively. Linkage runs blocking on DOB produced weights ranging in number from 8 for exact matching to 64,273 for inexact matching. Linkage runs blocking on YOB produced 8 to 916,806 weights. Exact matching matched record pairs with identical test characteristics (sensitivity 90.48%, specificity 99.78%) for the highest ranked group, but algorithms differentially prioritized certain variables. Inexact matching behaved more variably, leading to dramatic differences in sensitivity (range 0.04-93.36%) and positive predictive value (PPV) (range 86.67-97.35%), even for the most highly ranked record pairs. Blocking on DOB led to higher PPV of highly ranked record pairs. An ensemble approach based on averaging scaled matching weights led to modestly improved accuracy. In summary, we found few differences in the rankings of record pairs with the highest matching weights across 4 linkage packages. Performance was more consistent for exact string matching than for inexact string matching. Most methods and software packages performed similarly when comparing matching accuracy with the gold standard. In some settings, an ensemble matching approach may outperform individual linkage algorithms.

Validation of neural tube defects in the full featured--general practice research database.

BACKGROUND: The General Practice Research Database (GPRD) has been used to identify associations between pregnancy medication exposures and birth defects, but experts have argued that databases such as this one cannot provide detailed information for the valid identification of complicated congenital anomalies. Our objective was to determine if the GPRD could be used to identify cases of neural tube defects (NTDs). METHODS: First, we created algorithms for anencephaly, encephalocele, meningocele, and spina bifida and used them to identify potential cases. We used the algorithms to identify 217 potential NTD cases in either a child's or a mother's record. We validated cases by querying general practitioners (GPs) via questionnaire. Where cases of NTD were identified in the mother's record, in addition to confirming the diagnosis, we asked the GPs if the diagnosis was for the mother or that of her fetus or offspring. RESULTS: Two hundred seventeen cases were identified, and 165 GP questionnaires were returned. We validated an NTD diagnosis for 117 cases, giving our algorithms a positive predictive value (PPV) of 0.71. The PPVs varied by NTD type: 0.81 for anencephaly, 0.83 for cephalocele, 0.64 for meningocele, and 0.47 for spina bifida. CONCLUSIONS: Our identification algorithm was useful in identifying three of the four types of NTDs studied. Additional information is necessary to accurately identify cases of spina bifida.