Kishore L. Jayakumar

Evidence synthesis activities of a hospital evidence‐based practice center and impact on hospital decision making

BACKGROUND Hospital evidence-based practice centers (EPCs) synthesize and disseminate evidence locally, but their impact on institutional decision making is unclear. OBJECTIVE To assess the evidence synthesis activities and impact of a hospital EPC serving a large academic healthcare system. DESIGN, SETTING, AND PARTICIPANTS Descriptive analysis of the EPCs database of rapid systematic reviews since EPC inception (July 2006-June 2014), and survey of report requestors from the EPCs last 4 fiscal years. MEASUREMENTS Descriptive analyses examined requestor and report characteristics; questionnaire examined report usability, impact, and requestor satisfaction (higher scores on 5-point Likert scales reflected greater agreement). RESULTS The EPC completed 249 evidence reviews since inception. The most common requestors were clinical departments (29%, n = 72), chief medical officers (19%, n = 47), and purchasing committees (14%, n = 35). The most common technologies reviewed were drugs (24%, n = 60), devices (19%, n = 48), and care processes (12%, n = 31). Mean report completion time was 70 days. Thirty reports (12%) informed computerized decision support interventions. More than half of reports (56%, n = 139) were completed in the last 4 fiscal years for 65 requestors. Of the 64 eligible participants, 46 responded (72%). Requestors were satisfied with the report (mean = 4.4), and agreed it was delivered promptly (mean = 4.4), answered the questions posed (mean = 4.3), and informed their final decision (mean = 4.1). CONCLUSIONS This is the first examination of evidence synthesis activities by a hospital EPC in the United States. Our findings suggest hospital EPCs can efficiently synthesize and disseminate evidence addressing a range of clinical topics for diverse stakeholders, and can influence local decision making.

Trends in US dermatology residency and fellowship programs and positions, 2006 to 2016

hyperkeratosis alternated with basketweave hyperkeratosis, whereas in VEN, agranulosis alternated with a normal granular layer. None of the biopsies of nevus sebaceus, lichen striatus, nevus comedonicus, linear lichen planus, or linear lichen planus pigmentosus showed focal changes. Alternating histologic changes were seen in only 3 nonmosaic cases (4.41%) (Darier disease in 1 and Dowling-Degos disease in 2). Overall, focal changes with skipping of epidermis was seen significantly more frequently in blaschkoid dermatoses (in 37.7% vs in 4.41% [P\ .001]). Roughly one-third of the blaschkoid dermatoses in our study showed an alternating histologic pattern, which is significantly greater than in nonmosaic dermatoses. Of these, linear EHK and ILVEN showed the pattern most consistently. In a large study of 167 epidermal nevi that included several variants such as EHK, psoriasiform ILVEN, porokeratosis-like, focal acantholytic dyskeratosis, and nevus comedonicus, alternating histologic changes were reported only for ILVEN. Recently, Ross et al described this finding in mosaic EHK. The basis of this histologic pattern is not clear. Actinic keratosis shows an alternating pattern of parakeratosis due to the sparing of follicular structures, whereas the cornoid lamellae in porokeratotic eccrine ostial and dermal duct nevus overlie acrosyringia; however, a relationship with adnexal structures was not seen in any of our cases. Mosaicism is characterized by mutant and nonmutant cell lines in an organism. Our findings raise the question whether this coexistence of 2 cell populations extends to the tissue level even within lesional skin in mosaic disorders. Gene and expression studies may help answer this question. Also, why only some blaschkoid dermatoses show this distinctive histologic pattern is not clear. Our study is limited by the retrospective study design and the relatively small number of individual entities.

Differences between recruitment advertisements for academic and private practice dermatologists

To the Editor: A shortage of academic dermatologists persists in the United States. Recruitment strategies for positions in academia and private practice may differ and contribute to this shortage. The first phase of physician recruitment often involves outreach to prospective applicants through communication media. About half of physicians use online recruitment sources during their job search. We compared recruitment advertisements for academic and private practice dermatologists to investigate differences. Analysis was conducted on all dermatologist advertisements on the official American Academy of Dermatology online career center, AADCareerCompass.org, over 2 days in February 2017. Advertisements were coded for employer, length, employment status, duration, patient population, clinical responsibilities, location description, schedule and work environment, and compensation and benefits. A position was classified as academic if the employer was a health system or hospital with a dermatology residency program, a physician organization that was part of such a health system or hospital, or if a research requirement was specified. Distinct positions advertised in a single listing were coded separately. Advertisements for governmental positions, fellowships, and practice sales were excluded. Statistical comparisons were conducted using t and Fisher exact tests in Stata software (StataCorp, College Station, TX). This study was deemed exempt by the University of Pennsylvania Institutional Review Board. In total, 391 advertisements were analyzed (Table I), with 10.5% (41/391) for academic positions, similar to the percentage of dermatologists who practice in academic settings (8.0%). Recruitment advertisements for academic positions were substantially longer than those for private practice (mean 321.1 vs. 194.6 words, P \ .01), but were less specific with details, including compensation amount or range (0% vs. 9.4%, P 1⁄4 .04), signing bonus (0% vs. 17.7%, P 1⁄4 .001), and cosmetic opportunities (29.3% vs. 49.4%, P 1⁄4 .02). Thirty-one private practice advertisements specified an annual compensation amount or range, with nearly all figures (28/31) ranging from

Expanding epidemiology and biostatistics curricula in undergraduate medical education to promote evidence-based practice

400,000 to

The Potential Data-Generating Role of a Federal Junk Food and Sugar-Sweetened Beverage Tax

600,000. Twenty-three advertisements specified a paid time off duration, with 4 to 6 weeks being most common (19/23). Our findings suggest that academic advertisements provide fewer details about position characteristics than private practice ones. Though academic advertisements were significantly longer with respect to word count, they emphasized general institutional information about hospital facilities, programs, and rankings rather than position-specific characteristics. The reasons for this lack of detail are unclear, but may include standardization of advertisement language across positions within an academic center, offering of benefits commensurate with academic rank and experience, flexibility in clinic schedule, or employer perception that certain characteristics are relatively less attractive. Among the general population, detailed recruitment advertisements have been associated with higher intention to apply, perceptions of organization attributes, applicanteorganization fit, organizational attractiveness, and advertisement truthfulness. Consequently, lower level of detail in advertisements for academic dermatologists may reduce interest and impede recruitment for these positions. Our study has limitations. First, generalizability may be limited, as we examined a single website. Second, dermatologists may use myriad resources, including referrals, physician recruiters, and mailings. Future studies should compare advertised position characteristics to verified characteristics, and clarify position-specific factors motivating applicants to practice in academic versus private practice settings.

Balancing patient care with profitability: Ethical considerations

As predatory journals emerge,1 the pace of scientific discovery accelerates, and data from electronic health records offer the promise of learning from clinical practice,2 it is becoming increasingly important for physicians to develop the skills to appraise new research evidence. Evidence evaluation and application, part of the Accreditation Council for Graduate Medical Education Core Competency of Practice-Based Learning and Improvement, require proficiency in clinical epidemiology and biostatistics. Yet, the majority of resident physicians lack confidence in understanding relatively basic concepts critical to appraising the medical literature, resulting in a desire for additional …

Preventing Publication Misrepresentation Among Residency and Fellowship Applicants

We read with interest the article by Pomeranz et al. on the legal and administrative feasibility of implementing a federal junk food and sugar-sweetened beverage (SSB) tax. As the authors indicate, such a tax could improve diets on an unprecedented scale. In addition, this tax could generate data to advance research and evidence-based policies. Surveys are traditionally used in epidemiological studies of food and beverage consumption, but population-level data may offer more comprehensive and reliable evidence regarding such trends. If designed and implementedeffectively, the junk food andSSB tax could provide an invaluable tool to public health researchers, epidemiologists, and policymakers seeking to understand the effects of the tax andother interventions on consumption over time. Excise tax revenues are correlated with, or even proportional to, consumer demand for a particular good or service. We recently demonstrated this approach with indoor tanning service excise tax revenues to estimate national trends in the prevalence of indoor tanning. Similarly, revenue data could be used to approximate junk food and SSB consumption over time. Alternatively, tax legislation could mandate reporting of the quantity of junk food and SSBs sold. As mentioned by the authors, an excise tax could be levied anywhere along the supply chain, from suppliers and manufacturers to distributors and retailers. As the end of this chain, retail sales are the best approximation of junk food and SSB consumption. Thus, a point-of-sale tax may provide more reliable and timely data on consumption than a manufacturer tax, which would tax junk food and SSBs not yet purchased by consumers. Moreover, retailers could more easily report the geographic location of sales, permitting determination of regional consumption patterns. With this information, researchers could analyze the effects of various state and local public health interventions. Despite these advantages, a point-of-sale tax might be more administratively complex given the higher market fragmentation in the grocery industry than in food manufacturing. Regardless of which entities are taxed, food sales in different product or nutrition categories would ideally be reported separately to provide more detailed consumption data. We commend the authors for advancing the discussion on how innovative tax policies could improve public health. A federal junk food and SSB tax has the potential to simultaneously deter unhealthy behaviors among consumers and illuminate consumption trends as a means of informing policy.

US Medical Licensing Examination Step 2 Clinical Knowledge Score in Dermatology Resident Selection

Dr Lone owns a busy solo dermatology practice in a rural area. There is a shortage of dermatologists in the region, with no other dermatologic care for 100 miles. Since starting his practice 15 years ago, Dr Lone has viewed treating patients insured byMedicaid as part of his social mission as a physician; 50% of his patients are on Medicaid. He is not currently accepting new patients. Unfortunately, the practice has recently fallen on hard times financially. A private equity firm has offered to invest in Dr Lone’s practice, which would not only ensure the practice’s survival but also enable him to accept new patients by reducing his administrative burden. However, this investment is conditional on Dr Lone terminating services to his existing Medicaid patients (and replacing them with new, non-Medicaid patients). If Dr Lone declines the offer and continues to serve all his current Medicaid patients indefinitely, the sustainability of the practice will be in doubt.

Henry Radcliffe Crocker—From the Elephant Man to the Textbook

O ver the last 2 decades, misrepresentation of publications has been documented among residency and fellowship applicants in multiple specialties. Misrepresentation includes claiming authorship of a nonexistent article, falsely claiming authorship of an existing article, and elevating author rank. This unethical behavior violates the integrity and trust instilled in the medical profession; both are essential for not only the care of patients but also the dissemination of knowledge. This publication misrepresentation also may distort match outcomes. Adopting stricter documentation requirements for application submission would enhance applicant accountability and likely reduce misrepresentation. For published journal articles, the Electronic Residency Application Service (ERAS) requires the applicant to input standard citation information, and provides an optional field for the PubMed Identifier (PMID). However, the addition of this optional field to ERAS has not reduced the incidence of unverifiable journal articles. A more effective strategy would be to require applicants to upload a PDF of the article, thereby holding applicants more directly accountable and impeding misrepresentation. In addition, data entry fields could be added for the URL and the digital object identifier (DOI), which provides a permanent link to recently published articles. It would be reasonable to require completion of at least 1 of these electronic records, as it would be extremely rare for a recently published journal article to lack all of them. For unpublished journal articles, ERAS requires applicants to describe the publication status as submitted, provisionally accepted, accepted, or in press. These categories are somewhat subjective. For example, it is unclear whether a journal’s request for major or minor revisions constitutes provisionally accepted, or what threshold differentiates in press from accepted. In addition, ERAS does not request any mandatory or optional documentation to verify these assertions. Definitions and examples of publication status should be provided to encourage applicants to report the status accurately, and applicants should be required to upload acceptance, revision, or submission messages from the relevant journal. Though ideally applicants would not be tempted to misrepresent their applications in the first place, these practical reforms could help prevent misrepresentation and restore integrity to the application process.

Applying Feedback Lessons to Online Medical Question Banks

This survey analysis assesses the use of USMLE Step 2 CK scores in resident selection in US dermatology residency programs.