win$newsletter - american academy of pediatricsassociation of pulmonary critical care and sleep...
TRANSCRIPT
!
WiN$Newsletter!
January$2019!
In#This#Issue#
!!
• WiN#news!!
• Upcoming#Opportunities#for#WiN#!!
• Recommended#Reading!!
!
!
!
!
!
!
!
!
!
WiN#News:##Dr.! Barbara! Warner! has! been! named! Division! Chief! at! Washington!
University!and!Dr.!Cynthia!Blanco!at!University!of!Texas!in!San!Antonio.!
Drs.! Warner! and! Blanco! step! up! to! increase! the! number! of! women!
Division! Chiefs,! but! women! still! are! <20%! of! all! Neonatal! Perinatal!
Division!Chiefs.!They!join:!
!
Billie!Short!(George!Washington!U)!
Nicole!Dobson!(Army)!
Lisa!Peterson!(Navy)!
DeeOAnn!Pillars!(U.!Ill)!
Laura!Haneline!(Indiana!U)!
Cynthia!Bearer!(Maryland)!
Stella!Kourembanas!(Harvard)!
Terrie!Inder!(Brigham)!
Renate!Savich!(Mississippi)!
Deb!Campbell!(Montefiore)!
Annemarie!Stroustrup!(Icahn)!
Janell!Fuller!(U!of!New!Mexico)!
Michele!Walsh!(Rainbow!Babies!and!Children’s!Hospital)!
Seetha!Shankaran!(Wayne!State)!
Susan!Guttentag!(Vanderbilt)!
Sandra!Juul!(University!of!Washington)!
Lourdes!Garcia!(Puerto!Rico)!
Elba!SimonOFayard!(Loma!Linda!University)!
Agnes!Sierocka!!!(Walter!Reed!National!Military!Medical!Center)!
Shanthy!Sridhar!(Stony!Brook!School!of!Medicine)!
Joan!Richardson!(UTMB,!Galveston)!
and!Karen!Hendricks!Munoz!(Medical!College!of!Virginia)!
!
!
There! are! about! 112! neonatology! divisions.! Women! are! still!
underrepresented!at!this!level!of!leadership!!More!women!at!the!table,!
in! part,! will! help! generate! ideas! and! facilitate! discussion! regarding!
gender! disparity! and! encourage! improved! sustainability! of! women! in!
medicine.!We!laud!the!dedication!of!these!women!and!their!mentoring!
through! modeling.! We! look! forward! to! highlighting! further!
achievements!among!women!in!the!field.!
!
!
!
Upcoming$Opportunities$ for$WiN!
Please! save! the! date:! The!
Workshop! on! NeonatalO
Perinatal! Practice!
Strategies! from!
03/29/2019! to! 03/31/2019!
and! the! WiN! meeting! and!
events! 03/31/2019O
04/01/2019.! There! will! be!
presentations! on! the! WiN!
health/wellness! survey!
results! and! comments! by!
Dena!Hubbard! followed!by!
breakout! groups! with!
discussion! on! how! to!
improve! health,! wellness!
and!prevent!burnout.!
!$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
!
WiN$Trainee$Travel$Awards$to$Scottsdale$WiN$meeting$
Through!the!section!SoNPM!we!are!excited!to!grant!10!$1000!travel!
scholarships!to!female!TECaN!trainee!members!to!attend!the!WiN!
meeting!in!Scottsdale.!!!Please!send!your!CV!and!briefly!answer!the!
following!2!questions:$Why!do!you!want!to!attend!the!WiN!meeting?!
After!attending!the!meetingOhow!will!you!work!with!WiN!and!others!to!
accomplish!our!goals!for!yourself!and!others?!
!
Please!respond!to!https://www.surveymonkey.com/r/WiNTravelGrants2019 !expressing!your!interest!to!attend.!The!applications!are!due!by!Feb!15
th.!
Recommended$Reading$The!departments!of!pulmonary!critical!care!medicine!recently!
released!a!report!addressing!gender!inequality!in!academic!medicine.!!
This!panel!comes! up! with! recommendations! to! address:! ! inequities! in! gender!climate,! disproportionate! burden! of! family! responsibilities,! lack! of!women!in!senior!academic!and!leadership!
positions,!poor!retention!of!senior!women!and!lack!of!gender!equality!
in!compensation.!The!PDF!of!this!article!is!attached!with!the!newsletter.!
The!podcast!associated!with!this!article!is!available!at!the!link!below:!
https://s3.amazonaws.com/static.thoracic.org/podcast/AnnalsATSO interviews/GenderOInequalityOinOAcademicOMedicine.mp3!!
A! recent! study! in! Journal! of! Perinatology! by! Horowitz! et! al! titled:!Neonatologist!salary:!factors,!equity!and!gender.!!
The! study! finds! that! female! neonatologists! have! a! significantly! lower!
salary!than!male!counterparts,!even!when!adjusting!for!factors!such!as!
hours! worked! and! years! post! fellowship.! Unadjusted,! they! found! a!
$60,000! gender! gap! in!median! salary! for! female!neonatologists.!After!
corrections,!this!difference!was!reduced!to! just!under!$9500!per!year.!!
The!PDF!is!attached!at!the!end!of!the!newsletter.!
Contact#Us!
!The$WiN$Steering$Committee!
• Christiane#Dammann,!Tufts!University,!CDammann@tuftsO
nemc.org!!
• Marilyn#Escobedo,!University!of!Oklahoma,!MarilynO
• Renate#Savich,#University!of!Mississippi,[email protected]!
• Shazia#Bhombal,#Stanford!University,[email protected]!
• Krithika#Lingappan,#Texas!Children’s!Hospital,[email protected]!
• Clara#Song,#University!of!Oklahoma!(Social!Media!Chair)[email protected];!Twitter!ID:!@songMD!
• Kim#Lee,!Medical!University!of!South!Carolina!(website!Co4Chair)[email protected]!
• Rita#M.#Ryan,#Medical!University!of!South!Carolina!(website!Co4Chair)[email protected]!
!
!
Twitter:!https://twitter.com/WomenNeo!@WomenNeo!
Facebook:!https://www.facebook.com/groups/WomenNeo/!
WORKSHOP REPORT
Addressing Gender Inequality in Our Disciplines: Report fromthe Association of Pulmonary, Critical Care, and SleepDivision ChiefsCarey C. Thomson1,2, Kristin A. Riekert3, Carol K. Bates4, Anupam B. Jena5, Zea Borok6, Jennifer W. McCallister7,Lynn M. Schnapp8, Vibha N. Lama9, Monica Kraft10, Stephanie D. Davis11, Patricia Finn12, Shannon S. Carson13,James M. Beck14, Charles A. Powell15, Lynn T. Tanoue16, Naftali Kaminski16, and Anne E. Dixon17; on behalf of theAssociation of Pulmonary Critical Care and Sleep Division Directors, American Thoracic Society1Division of Pulmonary and Critical Care Medicine and Department of Medicine, Mount Auburn Hospital, Cambridge, Massachusetts;2Harvard Medical School, Boston, Massachusetts; 3Division of Pulmonary and Critical Care Medicine, Department of Medicine,Johns Hopkins University, Baltimore, Maryland; 4General Medicine and Primary Care, Beth Israel Deaconess Medical Center andAssociate Dean for Faculty Affairs, Harvard Medical School, Cambridge, Massachusetts; 5Department of Health Care Policy, HarvardMedical School, and Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts; 6Division of Pulmonary, CriticalCare and Sleep Medicine, Department of Medicine, University of Southern California, Los Angeles, California; 7Pulmonary, Critical Care,and Sleep Medicine, Department of Medicine, the Ohio State University, Columbus, Ohio; 8Pulmonary, Critical Care, Allergy and SleepMedicine, Department of Medicine, Medical University of South Carolina, Charleston, South Carolina; 9Pulmonary and Critical Care,Department of Medicine, University of Michigan, Ann Arbor, Michigan; 10Pulmonary and Critical Care Medicine, Department of Medicine,College of Medicine, University of Arizona Health Sciences, Tucson, Arizona; 11Division of Pediatric Pulmonology, Allergy, and SleepMedicine, Department of Pediatrics, Riley Children’s Hospital, Indiana University School of Medicine, Indianapolis, Indiana; 12Departmentof Medicine, University of Illinois at Chicago, Chicago, Illinois; 13Division of Pulmonary Diseases and Critical Care Medicine, Department ofMedicine, University of North Carolina School of Medicine, Chapel Hill, North Carolina; 14Pulmonary Sciences and Critical Care Medicine,Department of Medicine, University of Colorado, Aurora, Colorado; 15Pulmonary, Critical Care and Sleep Medicine, Department ofMedicine, Icahn School of Medicine at Mount Sinai, New York, New York; 16Pulmonary, Critical Care and Sleep Medicine, Department ofMedicine, Yale School of Medicine, New Haven, Connecticut; and 17Pulmonary and Critical Care Medicine, Department of Medicine,University of Vermont, Burlington, Vermont
Abstract
Despite the increasing proportion of women in U.S. medicalschools, there are relatively few women in leadership positions,and a number of recent publications have highlighted manyfactors that could contribute to gender inequity and inequality inmedicine. The Association of Pulmonary, Critical Care, and SleepDivision Directors, an organization of Division Directors fromacross the United States, convened a workshop to review data andobtain input from leaders on the state of gender equity in our field.The workshop identified a number of factors that could contribute
to gender inequality and inequity: gender climate (includingimplicit and perceived biases); disproportionate familyresponsibilities; lack of women in leadership positions; poorretention of women; and lack of gender equality in compensation.The panel members developed a roadmap of concreterecommendations for societies, leaders, and individuals thatshould promote gender equity to achieve gender equality andimprove retention of women in the field of pulmonary, criticalcare, and sleep medicine.
Keywords: gender; equality; equity; pulmonary
(Received in original form April 13, 2018; accepted in final form August 28, 2018 )
Correspondence and requests for reprints should be addressed to Carey C. Thomson, M.D., M.P.H., Mount Auburn Hospital–Pulmonary and Critical Care, DOB419, Mount Auburn Hospital, Cambridge, MA 02138. E-mail: [email protected].
Ann Am Thorac Soc Vol 15, No 12, pp 1382–1390, Dec 2018Copyright © 2018 by the American Thoracic SocietyDOI: 10.1513/AnnalsATS.201804-252ARInternet address: www.atsjournals.org
1382 AnnalsATS Volume 15 Number 12| December 2018
ContentsMethodsWorkshop FindingsGender ClimateDisproportionate Burden of FamilyResponsibilities
Lack of Women in Senior AcademicRank or Leadership Positions
Lack of Retention of Senior Women:the “Leaky Pipeline”
Lack of Gender Equality inCompensation
Summary of Recommendations forPromoting Gender Equity
Conclusions
Many recent publications have highlightedgender inequality in academic medicine(1–13). Although more than half of medicalschool students are women, exceeding menfor the first time in 2017, only 34–35% ofassociate professors and 22–25% of fullprofessors are women. In some specialties,women constitute fewer than 20% ofassociate or full professors (1), and genderequality remains an issue across disciplinesand regions of the country. The AmericanThoracic Society (ATS) supportedestablishing the Association of Pulmonary,Critical Care, and Sleep Directors(APCCSD) to foster common knowledgeand best practices regarding divisionorganization, leadership, and management.The APCCSD identified gender inequalityand equity as an important commonchallenge in academic medicine, and onethat likely contributes to the higher rates ofphysician burnout among women (14). Thegoal of gender equality requires a processfor promoting gender equity, a processthat promotes fairness according to therespective needs of a group (in this case,women). Although these recommendationsare particularly pertinent to the UnitedStates, our hope is that these insights will bevaluable to programs outside the UnitedStates as well.
Methods
The three workshop chairs (C.C.T., N.K.,and A.E.D.) organized a full-day workshopin conjunction with the ATS InternationalConference in Washington, D.C., in May2017. Although this was a workshop fromthe APCCSD and most panel members werefrom this association, we also includedleaders at the training program,departmental, and decanal levels, currentand past ATS executive committee andassembly leaders, and a research expert ingender equality and pay disparity inmedicine. The panel included 12 womenand 4 men from a diverse range of regions,
ethnicities, public and private institutions,genderual orientations, and faculty ranks.The aim of this workshop was to review therecent literature on gender equity andequality in medicine, to examine practicesand beliefs of leaders in our discipline, andto provide recommendations and strategiesfor promoting gender equity at theindividual, institutional, and professionalsociety levels. Although chairs andparticipants recognized that inequities inmedicine can also be based on or magnifiedby other important factors, such as race,sexual orientation, gender identification,disability, and ethnicity, the workshopfocused only on gender.
The workshop included an evaluationof literature and three presentations: 1) areport on in-depth qualitative phoneinterviews with division and departmentpulmonary and critical care leaders; 2) aperspective from an associate dean forfaculty affairs with expertise on the topic;and 3) an overview of recent literature ongender equality in medicine from a researchexpert on the topic. Each presentation wasfollowed by open discussion, and individualexperiences were shared regarding eachtopic. The workshop concluded with adiscussion to develop recommendations foradvancing gender equity at the individual,institutional, and medical society levels.
Workshop Findings
In a recent qualitative analysis ofsemistructured phone interviews conductedby the Association of American MedicalColleges (AAMC) of senior leaders at 24U.S. medical schools, five themes emerged asbeing associated with gender disparity inacademic medicine and frame our workshoprecommendations:
1) Gender climate – factors contributing togender inequities including implicit andperceived bias
2) Disproportionate burden of familyresponsibilities among women faculty (15)
3) Lack of women at senior academic rankor in leadership positions
4) Lack of retention of women (the “leakypipeline”)
5) Lack of gender equality in compensation.
These themes also arose during thequalitative interviews with Pulmonary,Critical Care, and Sleep Medicine (PCCSM)leaders and, in this article, quotes areintegrated with the research literature tohighlight the relevance to PCCSM.We begineach section with a review of relevantliterature and include specifics on PCCSMwhere data are available.
Gender ClimateThe panel discussed issues related toperceived disparities in gender climate,including perceived bias, implicit bias,consideration for leadership, academic andnetworking opportunities, sponsorshipand mentorship differences, fundingopportunities, compensation, andorganizational support. We did not addressthe issue of sexual harassment; this topicis of sufficient importance that it requiressole focus beyond the scope of thisdocument.
Workshop members reviewedliterature related to perceived bias inresearch and employment opportunities.We noted studies that reported that 70%of female versus 22% of male CareerDevelopment (K) award recipientsperceived gender bias, and 66% of womenand 10% of men reported experiencinggender bias in their academic environments(16). Gender bias in hiring was striking inour review of a 2012 Yale study, in whichresearchers reported statistically significantgender bias on ratings of competence,strength of candidacy for hire, andmentoring the candidate when comparingthe same application materials labeled withmale and female names (17). Bias also existsin student evaluations, which are used tomake decisions regarding employment,promotion, and tenure (18). Women alsoexperience bias when introduced for
WORKSHOP REPORT
Workshop Report 1383
medical presentations: a recent studyshowed that women were introduced bytheir professional titles for medical grandrounds in only 49% of cases compared with72% in men (19).
These examples illustrate factors thatcontribute to a poor gender climate forequity in opportunity, mentorship, and,ultimately, promotion and compensation.Many workshop contributors stated thatwomen perceive gender bias in theirworkplace settings, from entry-levelthrough senior leadership, and suggestthat training in implicit bias may play arole in uncovering bias that may impactwomen’s experiences and opportunities:“Faculty members who’ve participated [inunconscious bias training] have said thatthey’ve learned quite a bit. I think it hasinfluenced how search committees look atcandidates and search for candidates.”
The panel drafted a series of specificrecommendations for action by professionalsocieties, institutional leadership, and theindividual to address inequities in gender
climate (Table 1). These recommendationsinclude surveys to document current status,formal policies, appropriate representation,formal training programs, programs toincrease awareness of inequities, andadvocacy and support programs.
Disproportionate Burden ofFamily ResponsibilitiesFemale physicians report disproportionateresponsibility for caregiving and familyresponsibilities, which can impact jobsatisfaction, time available for work-relatedactivities outside of business hours, mentalhealth, and burnout. Women have a higherrate of depression as early as internshipcompared with men, which is associatedwith work–family conflict (20–22). In arecent survey of surgeons, women weresignificantly more likely to have a spousewho worked full time and be responsiblefor the majority of household and familyresponsibilities, with the exception offinancial matters (23). In a survey ofNational Institutes of Health K awardees,
86% of women versus 45% of men werepartnered to a full-time employee. Womenspent 8.5 hours more per week on familyand household responsibilities, and werenearly four times more likely to take timeoff from work for childcare disruptionscompared with male respondents (15, 24).Not all hospitals or practices offer medicalleave, and many fields of medicine offerlittle to no flexibility (25). One-third tothree-quarters of physician mothers statedthat they were discriminated against as aresult of pregnancy, maternity leave, orbreastfeeding (26–28). For example,maternity leave is extracted from time-limited research blocks rather than clinicaleffort, which slows academic progress.Many of these factors are potentiallymodifiable with careful attention andtransparency. Some survey participantsnoted that they became more aware of thefamily responsibilities of junior facultyafter gender equality discussions at facultyretreats, and remarked, “After the retreat[where equality issues were discussed], I’ve
Table 1. Panel recommendations to address inequities in gender climate
Intervention Professional Society Organizational Leader Individual
Survey Administer a standard survey Survey group; disseminateresults and action items
Advocate for, and respond to, institutionalsurvey
Policies Develop and distribute policies onsexual harassment and establishzero tolerance policies aroundthese issues
Advocate for institutionalombudsperson as facultyresource
Know your organizational resourcesfor support
Representation Maintain proportional representation ofwomen on committees (includingawards and nominating committees),in leadership and editorial roles, onguideline panels, and in presentationplatforms
Invite equal distribution ofwomen speakers and visitingprofessors
Assess the proportion of women onfaculty, in senior faculty ranks, andin leadership positions in yourorganization or professional society
Training Disseminate best practices in genderequality for leaders, programorganizers, editorial boards, anddivision director and programdirector groups
Establish a formal process forprofessional introductions
Seek out and participate in local andnational training opportunities
Provide training in implicit bias toorganizational leaders, committeeleaders, editorial boards, anddivision and program directorbodies
Provide training in implicit biasfor search committees
Awareness Provide programming and women’sgroups to address issues specific towomen
Provide objective criteria–basedcompensation
Assess the gender climate, includinginstitutional- and professional society–level policies important to women, anddiscuss these issues with female facultyin those organizations
Advocacy Support workshops and discussionsrelated to gender equality
Assess and work toward genderequality in leadership positions
Participate in institutional and nationalwomen’s groups that build awarenessand advocate for gender equality andimproved work–life balance
Support Provide opportunities for women tomentor other women through formalnetworking and mentoring platforms
Advocate for promotion ofqualified women faculty
WORKSHOP REPORT
1384 AnnalsATS Volume 15 Number 12| December 2018
been more mindful and understanding ofthe schedule the younger faculty have, bothmen and women.[if I try to get a hold ofsomeone in the morning and can’t] insteadof immediately going to ‘well, they aretaking it easy today’ I immediately thinkthat it takes time to get kids [ready].”
The panel drafted a series of specificrecommendations for action by professionalsocieties, institutional leadership, and theindividual to address this disproportionateburden, with explicit proposals to addresschildcare, breastfeeding, and leave policies(Table 2).
Lack of Women in Senior AcademicRank or Leadership PositionsWomen continue to be underrepresented inleadership positions in academic medicine.Women hold less than 25% of seniorleadership roles in academic medicine, fewerthan 25% of women have been promoted tofull professor, and few women hold majordivision and departmental leadership roles.The AAMC reports that women constituteonly 24% of all division chiefs, 24% of vicechairs, 15% of department chairs, and 16%of medical school deans (1). Women areunderrepresented in pulmonary and criticalcare medicine. The AAMC reported in 2015–2016 that only 31.7% of adult pulmonary andcritical care medicine fellowship trainees werewomen, despite an equal-to-increasedproportion of women in medical school. Inpediatrics, 56.6% of pulmonary and 62.5% ofcritical care trainees are women. Despite this,women constitute only 20% of adultpulmonary directors and 15% of critical caretraining program directors (14). Womendirect only 15% of the adult divisions and23% of the pediatric divisions represented inthe current membership of the APCCSD,resulting in few women role models. Thisgender disparity is similar to some otherspecialties (15).
The paucity of women on editorialboards, and institutional and professionalsociety committees, may limit access to firstand senior author publications, researchfunding, and awards. A 2017 review ofclinical practice guidelines found thatwomen constituted only 13% of authors ofguidelines in critical care and 25% of all 413clinical practice guidelines (29). Similarly,first and last authorship was lower amongwomen, mirroring the proportion of femalereviewers and editorial board members (30).The proportion of women in major editorialroles (editor, deputy editor, and associateeditor) of the three ATS journals is17%, 21%, and 40%. As women areunderrepresented in the editorial process,same-sex preference (homophily) may playa role in the likelihood of publication bywomen (6). Homophily may also play a rolein appointments to leadership roles, giventhat there is a disproportionate lack offemale division chiefs, chairs, and deans.Less involvement in these activities reducesopportunity for networking, futureacademic work, promotion, and leadership.These issues further highlight the need formore women among editorial panels,guideline panels, award committees,leadership positions, and promotion andsearch committees (31).
Barriers to networking exist. Womenmay have difficulty forming connections ininformal networks that are often maledominated. These networks provide forumsto discuss academic work and professionalopportunities, and lead to collaborations.Even when women are able to networkthrough mentoring relationships, our panelacknowledged that they are more likely to beasked to fulfill traditional “helping roles,”which are often time consuming and notlinked to, and can even slow, academicadvancement (32). Women also needleadership roles that lead to other
opportunities (33). Workshop membersfound this to be an important area of focusfor division leaders, and that they couldmake a difference in sponsoring women andpromoting the advancement of women.
Research in both business and academicmedicine suggests that women lacksponsorship more than mentorship, and thatprograms targeting only mentorship werenot effective at advancing the careers ofwomen (34–36). Sponsorship differs frommentorship in that the sponsor is a seniorinfluential leader who publically supports andadvocates for their more junior colleague(34). In the business world, influential malementors are “sponsoring” theirmalementees,leading to more opportunity for leadershipand promotion (35, 36). A recent review of42 studies found a similar pattern in academicmedicine (37). A lack of women in seniorleadership roles to serve as sponsors mayinhibit career advancement of other women.Indeed, in the survey of our own specialtyleadership, we found a paucity of women tointerview. Mentors and department leadersmay need coaching to learn how to sponsorwomen (38). For example, sponsorship bysenior leaders and department chairs iscrucial for identifying candidates forleadership and promotion. This has beencited as an issue in corporate climates (39),and may also be reflected in medicine, as wasnoted in our panel discussion.
Another factor our panel identified asimpeding career advancement was thatfewer women apply for leadership positionsor nominate themselves for honors andawards. We recognized several potentialmediating influences, including the fact thatwomen are more likely to expect thatexcellent work will be noticed and rewarded(sometimes referred to as “the tiarasyndrome”) (40). As one interviewed leadernoted, “If there is a new job [many womenthink] ‘if somebody really thinks I’m good
Table 2. Panel recommendations to address disproportionate burden of family responsibilities
Intervention Professional Society Organizational Leader Individual
Advocate forchildcare services
Provide childcare or informationon local services at nationalmeetings
Advocate for workplace childcarefacilities
Share the impact of childcare on worklife, advocate for support, andassess dependent care benefits
Breast feeding Designate and publicize space atnational meetings
Advocate for designated spacefor faculty and trainees
Assess leave policies and structuralsupport for breastfeeding,including organizational and statelaws before signing a contract
Leave policies Ensure organization has leavepolicies
WORKSHOP REPORT
Workshop Report 1385
enough they would ask me to apply’.” Thelack of senior female sponsors or mentors,disproportionate family responsibilities, andless desire to relocate were all identified byour panel as significant barriers to women inleadership. The care of both aging parentsand children plays a factor for women intheir consideration of relocating or takingon a leadership role that would addadditional responsibilities (41, 42). Leaderswho recognize the lack of women in seniorleadership roles have the capacity to increasethe representation of women in leadership.One participant noted, “When I started [at anew institution a few years ago], females inleadership positions, such as directors, sectionleaders, managers.I don’t think any of themwere women. So I made abrupt changes.Idid that rather tacitly because of grossimbalances.” One further consideration isthat the current gender imbalance inleadership places a “burden” on a limitedpool of senior women related to requests fortheir time. The push for equal genderrepresentation on committees, the increasednumber of junior women seeking seniorwomen mentors, and the request for senior
women to be more visible within theirinstitutions are all positive outcomes;however, these are typically acts ofcitizenship that are uncompensated. Thelack of women at the senior rank or inleadership positions dictates that theseresponsibilities, which senior women shouldand do accept, are disproportionatelydistributed among relatively few individuals,compared with their male counterparts.Methods to track and reward thesenoncompensated activities may beparticularly helpful to female faculty (43).
The panel drafted a series of specificrecommendations for action by professionalmedical societies, institutional leadership,and the individual to address the lack ofwomen in senior academic rank orleadership positions that can be addressedwith training, advocacy, and networking,and also with specific plans forimplementation (Table 3).
Lack of Retention of Senior Women:the “Leaky Pipeline”Historically, attrition has been named as acause for the “leaky pipeline,” leading to the
lower percentage of female senior faculty orwomen in leadership positions. Althoughnew faculty positions increased amongwomen in 2013–2014 by 4%, there was a5% increase in the percentage of womenleaving faculty positions (1). Workshopleaders believed that the leaky pipelinewas not the only significant factor in theunderrepresentation of women in leadershipor at higher faculty ranks. One explanationfor fewer women in leadership roles may bea difference in career goals, which maycontribute to the leaky pipeline. Aninterviewee suggested that there were fewerwomen interested in careers as physicianscientists: “[Years ago] fewer women haddesired or had the goal of achieving, not thatthey weren’t capable, but the climate 20–30years ago of who wanted to be a professor orphysician-scientist was different.” However,a 2016 survey of career goals of K awardrecipients revealed that men and womenwere similar in their desire to teach, conducthigh-quality research, publish, and provideexcellent patient care. Men were more likelyto desire higher compensation and anational or international reputation as an
Table 3. Panel recommendations to address fewer women in senior academic ranks or leadership positions
Intervention Professional Society Organizational Leader Individual
Training Provide specific leadershiptraining for women
Provide training for women and resourcesto assist in advancing their careers,including promotion criteria andaligning composition of work withthose criteria, training in negotiationskills, and emphasizing the importanceof networking, mentorship, andsponsorship
Seek knowledge about facultydevelopment and skills-basedtraining for women, includingnegotiation skills, grant writing,manuscript and curriculum vitaepreparation, mentorship, andsponsorship
Provide training for senior leaders on theirroles as mentors and sponsors
Seek leadership training and apply forleadership roles in your organizationand at the society level
Provide training in implicit bias to faculty,search committees, leaders, and hiringprofessionals
Advocacy Advocate for inclusion ofwomen in leadershippositions
Advocate for women faculty for promotionand leadership positions
Understand the criteria for promotionin your academic track
Networking Inclusion of women onpanels, in awards and oncommittees
Support networking at institutional andnational level
Network at the professional societylevel to actively seek opportunitieson committees, guidelines, andother professional activities
Implementation Establish process forinclusion
Establish a process to assess facultydistribution of clinical andadministrative roles and responsibilitiesby gender, develop mechanisms totrack these responsibilities, and avoidassigning “busy work”disproportionately to women
Survey faculty regarding job satisfaction,gender climate, and work–lifebalance
WORKSHOP REPORT
1386 AnnalsATS Volume 15 Number 12| December 2018
expert, whereas women were more likely toseek work–life balance (10). Althoughpersonal considerations may be a factorin some decisions, our workshop panelbelieved that lack of organizational supportfor women, implicit bias, lack of parityin advancement and leadership, andequality in compensation, are potentiallymodifiable factors that lead to disparities inretention.
The panel drafted a series of specificrecommendations for action by professionalsocieties, institutional leadership, and theindividual to address this leaky pipeline.These recommendations include increasingrepresentation and recognition, along withfiscal and structural support (Table 4).
Lack of Gender Equalityin CompensationAcross all occupations, there is a 20% paygap between men and women in the UnitedStates. This disparity is lowest at 10%among younger workers, and increases toover 25% for older workers. This disparitywill impact not only current economicsecurity, but also potential retirement ageand longer-term financial health (44).Multiple studies demonstrate inequalitiesin compensation among male and femalephysicians (3–6, 45, 46), ranging from 16%to 37% (45). Although there are claims thatdisparities in compensation are due tovariability in faculty rank (44), this mayrepresent bias because academicadvancement is slower for women. Other
possible factors may include specialty,number of hours worked per week, yearson faculty, including time away forparenting, and clinical productivity (3, 4).Female physicians may also spend moretime with each patient, which may relate topatient expectations (47). However, thegender pay gap persists even aftercontrolling for these factors (45–48), anddoes not appear to be related to quantity orquality of clinical work. Two 2017 studiesdocumented lower 30-day readmissionsand mortality and better surgical outcomesamong patients treated by women (48, 49).Pay gaps exist for program directors (50)and for physician scientists aftercontrolling for important covariates(51, 52).
Jena and colleagues (3, 4) analyzedequality in compensation by gender,controlling for age, years since training,rank of medical school attended, andnumber of scientific publications,including authorship position, grantfunding, clinical trial participation, andclinical productivity (as measured bybilling to Medicare). Women were paid lessthan men after controlling for thesevariables. Specialties with the largestadjusted salary differences includedorthopedic surgery, surgical subspecialties,neurology, cardiology, hematology/oncology, and obstetrics–gynecology,though there were gaps in almost everyfield. There were too few responses inPCCSM to draw conclusions regarding
compensation differences. Our workshopparticipants, however, felt there were clearcompensation gaps in our field, and someleaders had worked to successfully removethese gaps in their divisions upon assuminga leadership role. In our workshop, Dr. Jenastated that the absence of salary differentialwithin a specific faculty rank does notimply that salary gaps are absent. If womenface barriers to promotion to the higherfaculty ranks, then the main mechanism bywhich salary gender inequality may arise isin differential promotion to higher facultyranks, not differential salary by genderwithin the same faculty rank. Women weresignificantly underrepresented at theprofessorship academic rank, regardlessof specialty, when controlling for thesefactors.
Members of the panel and seniorleaders we interviewed agreed that startingsalaries were often lower for women thanmen due to a lack of negotiation amongwomen at hiring. Women were identified asbeing less likely to negotiate a higher salaryupon initial appointment, and this disparitycreated a difference that is challengingto rectify over time (7). The panelacknowledged that candidates most likely“get what they negotiate and not what theydeserve,” and this was a strong factor instarting salaries that differ between men andwomen. As one leader we interviewed stated,“Women who get offers with [the institution]are less interested in negotiating [comparedwith men].” Another leader stated, “What it
Table 4. Panel recommendations to address poor retention of senior women
Intervention Professional Society Organizational Leader Individual
Representation Seek women as nominees forawards and society-basedleadership positions
Mandate gender equality duringfaculty recruitment, especiallyfor leadership positions
Seek leadership roles, including servingon search committees and promotioncommittees where one can influencedecisionmaking around gender equality
Ensure proportionaterepresentation on guidelinepanels and platformpresentations
Populate search committeeswith a focus on gender andethnic diversity and providetraining in implicit bias
Consider requesting senior leaders tosponsor you
Recognition Highlight the accomplishmentsof women
Provide tenure-track extensions Apply for awards, funding, and structuralsupport available through yourorganization or specialty society.Advocate for other qualified women.
Fiscal support Assess the proportion of womenwho receive society-basedfunding
Provide funding opportunitiesfor women and institutionalsupport for research
Structural support Develop and support structuresthat focus on these issues
Provide flexible work structureswhen possible, especially duringcritical times for working mothers
WORKSHOP REPORT
Workshop Report 1387
takes to be a good negotiator is not seenas a trait women should have. So it isperceived more negatively when women[negotiate] than when men do it.” It was alsorecognized that institutions with uniformcompensation policies are less likely todifferentiate starting salaries on the basis ofgender.
Several interviewees and panelistsperceived that women were less likely torecognize negotiating as an option (“Theyimplicitly trust that what is offered is the finalsay, so they either take it or leave it.mostoften they take it because they’re prettycommitted by the time they sit down for thediscussion of salary.”) and undervalue theirworth (“It is hard for a woman to say ‘I amworth this much’ or ‘I merit this or that’”). Inaddition, some women accept lower salarieswith promises of higher compensation later.This factor may be partially explained by“the tiara syndrome,” discussed previouslyhere. The imposter syndrome may also leadto women feeling that they do not “deserve”more in compensation despite their merits(53). This feeling may impact initialnegotiations. Many panel members statedthat men were more likely to relocate foranother opportunity, and that there weremore frequent attempts on these groundsfor men to negotiate raises. As discussedpreviously, female physicians are more likelythan their male counterparts to be partneredto someone working full time, and may beless able to move. As one person stated,“Women stay put.[they are] less willing tomove so miss the salary bump you get when
you move to another place.easiest way toget a raise is to go to another institution evenif it is a lateral move.”
One participant acknowledged that,when he assumed a division chief role, therewere no female leaders and there werecompensation-based disparities at everyrank. He rectified this by establishing acommittee to review salaries and leadershiproles, establishing objective criteria forpromotion and compensation, and formingbalanced search committees for new facultythat actively sought female applicants.He also began hiring faculty withoutnegotiation, offering a transparent and flatbase salary for any new hire, predefined byfaculty rank and role, a practice endorsedby other workshop participants. Withthis directive approach, disparities incompensation by rank and step wereeliminated within 8 years with the exceptionof one highly specialized role.
Some leaders believe that bias is lesslikely to occur within a system that rewardsrelative value units; however, our panelacknowledged that clinical assignmentscould be impacted by gender, especially ininterventional fields and critical care, andbelieved that these could account for salarygaps within our divisions as well. Moreover,presently, there are not comparative targetsfor administrative, committee, and other“nonclinical” work assigned to an academicphysician that impact compensation. Inaddition, our panel acknowledged thatwomen are asked more frequently toparticipate in committee work and
“uncompensated” activities. A balancebetween compensation for quantity andquality of work is lacking.
The panel drafted a series of specificrecommendations for action by professionalsocieties, institutional leadership, and theindividual to address knowledge about payscales and gender pay gaps, training innegotiation, advocacy, and structuralsupport (Table 5).
Summary ofRecommendations forPromoting Gender Equity
There are a number of ways professionalmedical societies can promote genderequity. Societies are uniquely positionedto provide opportunities for academiccurrency, awards, and leadershipopportunities—key metrics for academicpromotion—which are linked to academicadvancement, opportunities for leadership,compensation, and, potentially, retention. Itis important that professional societiesassess the distribution of women on keycommittees, including nominating andawards committees. At the ATS, 3 womenwere selected to give the prestigiousAmberson lecture in 61 years, and 9 womenreceived the Trudeau medal in 92 years. Thesituation did not improve until the 21stcentury—only 3 women received the medaland 1 delivered the Amberson lecture.Women have been award recipients between7% and 23% of the time over the past
Table 5. Panel recommendations to address lack of gender equality in compensation
Intervention Professional Society Organizational Leader Individual
Training/knowledge
Provide standards for salary data byspecialty to improve the individual’sability to negotiate an appropriatesalary based on their skill and rankand for the organizational leaders tobenchmark offers
Assess the faculty by rank andgender and then compensate thembased on predefined and transparentcriteria for clinical, academic, andadministrative activities for thedivision and department
Know national pay scales andmarket rate for your role
Offer programming that provides trainingin negotiation skills for graduatingfellows, junior and senior faculty
Pursue training in negotiationaround offers and compensation
Advocacy Support and encourage the promotionof qualified women
Understand how to complete acounter offer
Explore positions at other academicinstitutions to improve leadershipopportunities and compensation
Structuralsupport
Create a structure for transparentcriteria-based compensation thatsupports gender equality
Review offers and contracts withmentors and advisors beforesigning
Standardize recruitment packages byrank and leadership position
Negotiate administrative time andsupport for your work
WORKSHOP REPORT
1388 AnnalsATS Volume 15 Number 12| December 2018
15 years. Women have served as the ATSpresident 17% of the time over the past30 years, improving to 27% in the past15 years. Having a critical mass of womenin leadership roles will allow more juniorprofessionals to identify role models andpotential sponsors. Societies can alsosponsor events to showcase women,and provide training in mentorship,sponsorship, leadership, promotion,work–life balance, and negotiation.
A first step is to acknowledge thatgender inequality exists and impacts theperceived climate. If organizational leadersdo not perceive that a disparity exists, thenthe path to equality and retention ofwomen will be a distant reality. A HarvardBusiness Review article highlighted that“leaders who are serious about drivinginclusion need to audit hiring, evaluation,and promotion processes and activelydrive out bias. They need to realize that
they themselves cast big shadows onthese issues, and if they are not deeplymindful of their impact, they willreinforce the status quo in a myriadof ways” (54).
To promote gender equality at theindividual level, the panel membersrecommend that individuals shouldunderstand the recommendations for theinstitution and their professional societiesto support their professional development.Although some of these recommendationsmay seem specific to women faculty, theyare targeted also at men. We considerthe promotion of gender equality anurgent priority of both men and women.
Conclusions
Gender inequality in academic medicinewas not created in a vacuum; it is the result
of the combined effects of social,organizational, and cultural biases. Toestablish gender equality, the individual,the organization and its leadership, andmedical specialty societies need to take anactive role in removing barriers to thatequality and building structures topromote equity. As PCCSM division chiefsand departmental chairs from across thecountry, we have outlined a roadmap ofconcrete recommendations at each level toachieve gender equality in academicPCCSM. We believe that this is a worthy,necessary, and achievable goal, and onethat will need to be reassessed withperiodic future workshops addressing thisissue. n
Author disclosures are available with the textof this article at www.atsjournals.org.
References
1 Association of American Medical Colleges. The state of women inacademic medicine: the pipeline and pathways to leadership, 2015–2016. 2016 [accessed 2017 Sep 1]. Available from: https://www.aamc.org/members/gwims/statistics.
2 Lautenberger DM, Dandar VM, Raezer CL, Sloane RA. The state ofwomen in academic medicine: pipelines of pathways in leadership.Washington, DC: Association of American Medical Colleges.AAMC Publications. 2014 [accessed 2018 Oct 12]. Available from:https://members.aamc.org/eweb/upload/The%20State%20of%20Women%20in%20Academic%20Medicine%202013-2014%20FINAL.pdf.
3 Jena AB, Olenski AR, Blumenthal DM. Sex differences in physiciansalary in US public medical schools. JAMA Intern Med 2016;176:1294–1304.
4 Ly DP, Seabury SA, Jena AB. Differences in incomes of physicians in theUnited States by race and sex: observational study. BMJ 2016;353:i2923.
5 Freund KM, Raj A, Kaplan SE, Terrin N, Breeze JL, Urech TH, et al.Inequities in academic compensation by gender: a follow-up tothe National Faculty Survey Cohort Study. Acad Med 2016;91:1068–1073.
6 Carr PL, Gunn C, Raj A, Kaplan S, Freund KM. Recruitment, promotion,and retention of women in academic medicine: how institutions areaddressing gender disparities. Womens Health Issues 2017;27:374–381.
7 Bates C, Gordon L, Travis E, Chatterjee A, Chaudron L, Fivush B, et al.Striving for gender equity in academic medicine careers: a call toaction. Acad Med 2016;91:1050–1052.
8 Sege R, Nykiel-Bub L, Selk S. Sex differences in institutional support forjunior biomedical researchers. JAMA 2015;314:1175–1177.
9 Raj A, Carr PL, Kaplan SE, Terrin N, Breeze JL, Freund KM. Longitudinalanalysis of gender differences in academic productivity amongmedical faculty across 24 medical schools in the United States. AcadMed 2016;91:1074–1079.
10 Jones RD, Griffith KA, Ubel PA, Stewart A, Jagsi R. A mixed-methodsinvestigation of the motivations, goals, and aspirations of male andfemale academic medical faculty. Acad Med 2016;91:1089–1097.
11 Rochon PA, Davidoff F, Levinson W. Women in academic medicineleadership: has anything changed in 25 years? Acad Med 2016;91:1053–1056.
12 Sklar DP. Women in medicine: enormous progress, stubbornchallenges. Acad Med 2016;91:1033–1035.
13 Lautenberger D, Raezer CL, Bunton S. The underrepresentation ofwomen in leadership positions at USmedical schools.AAMCAnalysisin Brief 2015;15(2) [accessed 2017 Sep 1]. Available from: https://www.aamc.org/download/425020/data/february2015theunderrepresentationofwomeninleadershippositionsa.pdf.
14 Puffer JC, Knight HC, O’Neill TR, Rassolian M, Bazemore AW, PetersonLE, et al. Prevalence of burnout in board certified family physicians.J Am Board Fam Med 2017;30:125–126.
15 Carr PL, Gunn CM, Kaplan SA, Raj A, Freund KM. Inadequate progressfor women in academic medicine: findings from the National FacultyStudy. J Womens Health (Larchmt) 2015;24:190–199.
16 Jagsi R, Griffith KA, Jones RD, Stewart A, Ubel PA. Factors associatedwith success of clinician-researchers receiving career developmentawards from the National Institutes of Health: a longitudinal cohortstudy. Acad Med 2017;92:1429–1439.
17 Moss-Racusin CA, Dovidio JF, Brescoll VL, GrahamMJ, Handelsman J.Science faculty’s subtle gender biases favor male students. Proc NatlAcad Sci USA 2012;109:16474–16479.
18 Mitchell K, Martin J. Gender bias in student evaluations. PS Polit SciPolit 2018;51(3):648–652.
19 Files JA, Mayer AP, Ko MG, Friedrich P, Jenkins M, Bryan MJ, et al.Speaker introductions at internal medicine grand rounds: forms ofaddress reveal gender bias. J Womens Health (Larchmt) 2017;26:413–419.
20 Jagsi R. Gender differences in time spent on parenting and domesticresponsibilities. Ann Intern Med 2014;161:534.
21 Beckett L, Nettiksimmons J, Howell LP, Villablanca AC. Do familyresponsibilities and a clinical versus research faculty position affectsatisfaction with career and work–life balance for medical schoolfaculty? J Womens Health (Larchmt) 2015;24:471–480.
22 Guille C, Frank E, Zhao Z, Kalmbach DA, Nietert PJ, Mata DA, et al.Work–family conflict and the sex difference in depression amongtraining physicians. JAMA Intern Med 2017;177:1766–1772.
23 Baptiste D, Fecher AM, Dolejs SC, Yoder J, Schmidt CM, Couch ME,et al. Gender differences in academic surgery, work–life balance,and satisfaction. J Surg Res 2017;218:99–107.
24 Jolly S, Griffith KA, DeCastro R, Stewart A, Ubel P, Jagsi R. Genderdifferences in time spent on parenting and domestic responsibilitiesby high-achieving young physician-researchers. Ann Intern Med2014;160:344–353.
WORKSHOP REPORT
Workshop Report 1389
25 MacVane CZ, Fix ML, Strout TD, Zimmerman KD, Bloch RB, Hein CL.Congratulations, you’re pregnant! now about your shifts . . . : the stateof maternity leave attitudes and culture in EM. West J Emerg Med2017;18:800–810.
26 Jagsi R, Tarbell NJ, Weinstein DF. Becoming a doctor, starting afamily—leaves of absence from graduate medical education. N Engl JMed 2007;357:1889–1891.
27 Adesoye T, Mangurian C, Choo EK, Girgis C, Sabry-Elnaggar H, Linos E;Physician Moms Group Study Group. Perceived discriminationexperienced by physician mothers and desired workplace changes: across-sectional survey. JAMA Intern Med 2017;177:1033–1036.
28 Sowa A. Doctors who are moms say they face discrimination at work.Shots from NPR. May 15 2017 [accessed 2018 Feb 1]. Available from:https://www.npr.org/sections/health-shots/2017/05/15/528168990/doctors-who-are-moms-say-they-face-discrimination-at-work.
29 Mehta S, Burns KEA, Machado FR, Fox-Robichaud AE, Cook DJ, CalfeeCS, et al. Gender parity in critical care medicine. Am J Respir Crit CareMed 2017;196:425–429.
30 Erren TC, Gross JV, Shaw DM, Selle B. Representation of women asauthors, reviewers, editors in chief, and editorial board members at 6general medical journals in 2010 and 2011. JAMA Intern Med 2014;174:633–635.
31 Helmer M, Schottdorf M, Neef A, Battaglia D. Gender bias in scholarlypeer review. eLife 2017;6:e21718.
32 Yedidia MJ, Bickel J. Why aren’t there more women leaders in academicmedicine? The views of clinical department chairs. Acad Med 2001;76:453–465.
33 Haslam SA, Ryan MK. The road to the glass cliff: differences in theperceived suitability of men and women for leadership positionsin succeeding and failing organizations. Leadersh Q 2008;19:530–546.
34 Travis EL, Doty L, Helitzer DL. Sponsorship: a path to the academicmedicine C-suite for women faculty? Acad Med 2013;88:1414–1417.
35 Ibarra H, Carter NM, Silva C. Why men still get more promotions thanwomen. Harv Bus Rev 2010;88:80–85, 126.
36 Carter NM, Silva C. Mentoring: necessary but insufficient foradvancement. Catalyst 2010 [accessed 2018 Feb 1]. Available from:https://www.catalyst.org/system/files/Mentoring_Necessary_But_Insufficient_for_Advancement_Final_120610.pdf.
37 Sambunjak D, Straus SE, Marusic A. Mentoring in academic medicine: asystematic review. JAMA 2006;296:1103–1115.
38 Geist LJ, Cohen MB. Commentary: mentoring the mentor: executivecoaching for clinical departmental executive officers. AcadMed 2010;85:23–25.
39 Faniko K, Ellemers N, Derks B, Lorenzi-Cioldi F. Nothing changes, really:why women who break through the glass ceiling end up reinforcing it.Pers Soc Psychol Bull 2017;43:638–651.
40 Fitzpatrick TA, Curran CR. Waiting for your coronation: a career-limitingtrap. Nurs Econ 2014;32:162–165.
41 Berg S. Pay inequity, flex schedules top concerns for women physicians.AMA Wire Sep 27, 2017 [accessed 2018 Feb 1]. Available from:https://wire.ama-assn.org/ama-news/pay-inequity-flex-schedules-top-concerns-women-physicians.
42 Rabin RC. Health care? Daughters know all about it. New York Times;2017 May 11 [accessed 2018 Oct 12]. Available from: https://www.nytimes.com/2017/05/11/well/family/health-care-daughters-know-all-about-it.html.
43 Fassiotto M, Simard C, Sandborg C, Valantine H, Raymond J. Anintegrated career coaching and time-banking system promotingflexibility, wellness, and success: a pilot program at StanfordUniversity School of Medicine. Acad Med 2018;93:881–887.
44 American Association of University Women. The simple truth about thegender pay gap. 2018 Spring [accessed 2018 Mar 1]. Available from:https://www.aauw.org/aauw_check/pdf_download/show_pdf.php?file=The_Simple_Truth.
45 Grisham S. Physician compensation report 2017 Apr 5.Medscape 2017[accessed 2017 May 15]. Available from: https://www.medscape.com/slideshow/compensation-2017-overview-6008547.
46 AAMC. 2016 physician specialty data report: executive summary.2014 [updated 2016; accessed 2017 May 15]. Availablefrom: https://www.aamc.org/download/471786/data/2016physicianspecialtydatareportexecutivesummary.pdf.
47 Linzer M, Harwood E. Gendered expectations: do they contribute to highburnout among female physicians? J Gen Intern Med 2018;33:963–965.
48 Tsugawa Y, Jena AB, Figueroa JF, Orav EJ, Blumenthal DM, Jha AK.Comparison of hospital mortality and readmission rates for Medicarepatients treated by male vs female physicians. JAMA Intern Med2017;177:206–213.
49 Wallis CJ, Ravi B, Coburn N, Nam RK, Detsky AS, Satkunasivam R.Comparison of postoperative outcomes among patients treated bymale and female surgeons: a population basedmatched cohort study.BMJ 2017;359:j4366.
50 Willett LL, Halvorsen AJ, McDonald FS, Chaudhry SI, Arora VM. Genderdifferences in salary of internal medicine residency directors: anational survey. Am J Med 2015;128:659–665.
51 Jagsi R, Griffith KA, Stewart A, Sambuco D, DeCastro R, Ubel PA.Gender differences in the salaries of physician researchers. JAMA2012;307:2410–2417.
52 Jagsi R, Motomura AR, Griffith KA, Rangarajan S, Ubel PA. Sexdifferences in attainment of independent funding by careerdevelopment awardees. Ann Intern Med 2009;151:804–811.
53 Clance PR, Imes SA. The imposter phenomenon in high achievingwomen: dynamics and therapeutic intervention. PsychotherapyTheory, Research and Practice 1978;15:241–247.
54 Warner T, King M. To address gender bias at your company, start withteams. Harvard Business Review 2017 Jan 27 [accessed 2017 Sep 1].Available from: https://hbr.org/2017/01/to-address-gender-bias-at-your-company-start-with-teams.
WORKSHOP REPORT
1390 AnnalsATS Volume 15 Number 12| December 2018
Journal of Perinatologyhttps://doi.org/10.1038/s41372-018-0304-7
FEATURE
Neonatologist salary: factors, equity and gender
Eric Horowitz1 ● Henry A. Feldman2● Renate Savich3
Received: 7 September 2018 / Revised: 27 November 2018 / Accepted: 3 December 2018© Springer Nature America, Inc. 2019
AbstractObjective Physician compensation has been found to be influenced by gender, academic affiliation, specialty, productivity,and time in practice. This study explores their impact in the field of neonatology to inform institutional strategic planningand decisions by current and future practitioners.Study design A voluntary anonymous survey was distributed to members of the American Academy of Pediatrics Sectionon Neonatal-Perinatal Medicine with a 15% response rate. The survey contained questions assessing clinician characteristics,work environment, and professional productivity. Statistical analysis was done using JMP Pro 14.0.0 by SAS.Results Median salary was $256,000 (interquartile range, $213,608–315,000). Generalized linear model found that yearspost fellowship, academic affiliation, gender, practice location, professional duties, and clinical team member types inde-pendently influenced expected salary.Conclusion Several factors influence the expected compensation of this cohort of neonatologists, even after adjustments fordifferences in clinician characteristics, work environment, and productivity.
Introduction
With more than a tripling of hospital-based employment ofspecialists over the past 10 years, greater understanding intobenchmarking compensation, determining fair marketvalue, and the factors that influence compensation hasbecome increasingly important [1]. In addition, it isimportant to identify and address modifiable disparities insalaries. There are limited data currently available on neo-natologist compensation. While resources like Salary.com[2], Doximity [3], American Medical Group Association[4], Association of American Medical Colleges [5], Hos-pital and Healthcare Compensation Service [6], and Medi-cal Group Management Association [7] offer some insights,their data typically contain small numbers of neonatologistsor a specific subgroup (academic) of neonatologists. These
often-referenced resources, however, can guide decisionsthat may not best reflect the factors influencing the com-pensation of a neonatologist, nor their fair market value. Tobetter understand the factors that influence compensationand fair market value of a neonatologist, we conducted anational survey of the current members of the AmericanAcademy of Pediatrics Section on Neonatal-PerinatalMedicine (AAP SoNPM).
Methods
Instrument
A 43-item anonymous Qualtrics questionnaire was specifi-cally developed for this study and approved by the DukeUniversity Medical Center Institutional Review Board(IRB) for IRB exemption. The instrument utilized questionsto characterize neonatologists, describe the practice setting,and determine professional workload.
The original anonymous 25-item instrument for the2014 study, n= 929 and data not included in this analy-sis, was pretested for readability and comprehensibilityon a sample of 7 neonatologists from across the UnitedStates. Face validity was good, and any misunderstood orambiguous items were rewritten. The 2016 questionnaireused in this study expanded on the 2014 questionnaire
* Eric [email protected]
1 Division of Neonatology, Duke University Medical Center,Durham, NC, USA
2 Institutional Centers for Clinical and Translational Research,Boston Children’s Hospital, Boston, MA, USA
3 Department of Pediatrics, University of Mississippi MedicalCenter, Jackson, MS, USA
1234
5678
90();,:
1234567890();,:
and was again pretested with a convenience sample of10 neonatologists from across the United States toassess face validity of previous and new questions.
Their pretest responses were not included in the finaldata set.
Study setting and participants
This study took place via a national anonymous Qualtricssurvey distributed to all members of the AAP Section onNeonatal-Perinatal Medicine from March to April 2016.The Section on Neonatal-Perinatal Medicine is a 3226voluntary-member subsection of the American Academy ofPediatrics, and home organization for subspecialists in thefield of Neonatal-Perinatal Medicine. This group represents79% of the total US American Board of PediatricsNeonatal-Perinatal Medicine Delegates (4078 in 2015–2016). All participation was voluntary, and completion ofthe survey was demonstration of consent.
Data analysis
Statistical analysis of the responses was completed with JMPPro 14.0 (SAS, Cary, NC, 2018). For univariate analysis,results are reported as mean unless skewed data distribution,which was then reported as median. Salary showed a skeweddistribution and was log-transformed for analysis with ageneralized linear model, including continuous, nominal,ordinal, and binomial dummy variables as potential influences.Effect sizes of final model are reported as percentage differ-ence, calculated as 100% × exp(regression coefficient)−1).
Due to the highly variable and nuanced federal and statetax implications, all financial terms are reported as pre-tax.All financial interpretations should be made with potentialtax implications in mind.
Results
Overall, 492 responses were obtained from the 3226 activemembers of the AAP SoNPM, a response rate of 15.3%.The general description of the characteristics of therespondents is in Tables 1–3. A total of 366 respondentswere evaluated in the generalized linear model afterexclusion of 126 responders: not board certified/eligible orpracticing in the United States (21), working part-time orper diem (29), and lacked compensation data (76).
Neonatologist characteristics
As shown in Table 1, neonatologists responding to thesurvey were well distributed over their years in practice.Most were employed by a health system (49%), while 36%were in private practice, and 14% were employed by thegovernment. The majority of respondents were affiliated
Table 1 Characteristics of full-time board eligible/certifiedneonatologists in the United States
Variable Categories N (%)
Years postfellowshipa
<5 Years 75 (20)
5–10 Years 56 (15)
10–15 Years 43 (12)
15–20 Years 34 (9)
20–25 Years 53 (14)
>25 Years 105 (29)
Years in currentpracticea
<5 Years 128 (35)
5–10 Years 50 (14)
10–15 Years 52 (14)
15–20 Years 32 (9)
20–25 Years 32 (9)
>25 Years 72 (20)
Practice typeb Health systememployee
180 (49)
Private practice 131 (36)
Government 50 (14)
Other 4 (1)
Academic trackb Academic 237 (65)
Non-academic 129 (35)
Academic ranka Instructor 8 (4)
Assistant 89 (40)
Associate 51 (23)
Professor 75 (34)
Gendera Female 168 (47)
Male 192 (53)
Raceb Asian 59 (17)
Black/AfricanAmerican
15 (4)
White 252 (75)
Other 12 (4)
Ethnicityb Hispanic/Latino 19 (5)
Medical trainingb American medicalgraduate
286 (80)
Internationalmedical graduate
71 (20)
Compensationc–median (IQR)
Salary ($) 256,000 (213,608–315,000)
Bonus ($) 7200 (0–26,500)
Moonlighting ($) 20,000 (8000–36,500)
Total cashcompensation ($)
280,000 (225,000–355,750)
Data were collected as aordinal variable; bnominal variable; ccontin-uous variable; and ddummy ordinal variable (0 or 1). Distributionaround median listed as interquartile range (IQR)
E. Horowitz et al.
with an academic institution (65%), white (71%), andAmerican medical graduates (80%). The median base salarywas $256,000 with 51% of respondents receiving someamount of grant support.
Of the evaluated characteristics, years post fellowship,academic affiliation, and gender were all found to havesignificant independent impacts on salary in our generalizedlinear model. For every 5 years post fellowship, compen-sation increased by 2.71% (p < 0.001). Academic affiliationreduced annual compensation by an average of 5.86% (p <0.001). With regard to gender, being female reduced theaverage annual compensation by 3.68% (p < 0.001). Over a35-year career, assuming a 1.53% annual salary growth rateas seen in this cross-sectional sample, this gender effectcould mean a net loss in pay of over $430,000 for womencompared to men. If the annual pay deficit had beeninvested in a pre-tax retirement account that grew by 6%annually over the 35-year career, the total lost pre-taxearnings would grow from $430,000 to over $1,250,000 inpotential retirement savings.
With a median salary of $256,000 per year, the medianbonus was found to be $7200 per year. The binary dummyvariable for receiving an annual bonus included in thegeneralized linear model, however, found receiving a bonusdecreased base compensation by 3.48% or approximate$8900 (p= 0.002).
Practice description
Geographic distribution of respondents was not found to bestatistically different by chi-square test from that of the2016 data from the American Board of Pediatrics for boardcertified/eligible neonatologists practicing in the UnitedStates (p= 0.07). This study had a distribution of: GreatLakes (OH, MI, IN, IL, WI, MN): 19%; Mid-Atlantic (WV,VA, DE, MD, DC, PA, NJ): 20%; North Central (IA, MO,KS, NE, SD, ND): 7%; Northeast (ME, NH, VT, MA, CT,NY, RI): 15%; Northwest (MT, WY, ID, OR, WA): 3%;South Central (TX, OK, AR): 8%; Southeast (SC, GA, FL,AL, MS, LA, TN, KY): 14%; Southwest (AZ, UT, CO,NM, NV): 4%; and West (CA, AK, HI): 10%
As summarized in Table 2, most practices were locatedin large central (48%) or medium (36%) metropolitan areas(as defined by National Center for Health Statistics classi-fication). Of these, 75% had 7 neonatologists or more withintheir group, and 90% worked with neonatal nurse practi-tioners. The average delivery volume affiliated with theirprimary institution was 3000 per year, and 78% of the unitshad 25 or more specialized neonatal beds.
Geographic region and county size had a strong impacton compensation. Practicing in the Northeast or Mid-Atlantic regions reduced compensation by 6.72% (p <0.001) and 6.12% (p < 0.001), respectively. Living in theNorth Central region, however, positively influenced com-pensation by 5.00% (p= 0.02). Compared to other countydensities, large central metropolitan locations reducedexpected salary by 4.44% (p < 0.001).
The types of providers that comprise the clinical teamwere also found to impact compensation. Working withphysician assistants was associated with an increase ofcompensation by 4.02% (p= 0.004). The impact seen withneonatal hospitalists of a 1.97% (p= 0.10) reduction incompensation needs further examination. With a p value of0.43, working with neonatal nurse practitioners was notincluded in our model as it did not impact compensationand is common in most neonatal intensive care units.
Professional workload
As professional service revenues shift from fee-for-servicemodels to capitated and bundled payment models, anunderstanding of professional productivity is essential tobest forecast a budget and develop contractual expectations.
Table 3 summarizes professional workload found in oursurvey. On average, the respondents to this survey providedclinical service for 24 weeks per year, worked an average of65 h per week while on service, and 43 h per week when noton service. On further clarification, annual clinical time wasbroken down into an average of 86 weekdays, 40 week-nights, 22 weekend days, and 15 weekend nights. While
Table 2 Practice description
Variable Categories N (%)
County typea Large central metropolitan(>1,000,000 people)
175 (48)
Medium metropolitan(>250,000 people)
130 (36)
Small metropolitan(>50,000 people)
56 (15)
Neonatologists ingroupa
<3 15 (4)
3–6 76 (21)
7–10 80 (22)
11–14 56 (15)
>14 139 (38)
Clinical team membersb Neonatal hospitalist 107 (29)
Neonatal nurse practitioner 328 (90)
Physician assistant 95 (26)
Institutional volumec Births–median (IQR) 3000 (2000–4650)
Capacity of primaryunita
<25 50 (22)
25–50 87 (38)
51–75 57 (25)
76–100 26 (11)
>100 12 (5)
Data were collected as aordinal variable; bnominal variable; andccontinuous variable
Neonatologist salary: factors, equity and gender
weeks of service was not found to have a significant impacton compensation in our model, Table 4, weekday coveragedid. For each day of weekday clinical service, compensationwas increased by 0.05% (p < 0.001).
Call type did differ between in-house call, home call,both in-house and home call, or no call requirements. In-house call had a significant impact on compensation, raisingit by 3.35% (p= 0.005). The other types of call coveragewere not found to have a significant impact.
The average number of critical care patients billed forper day was associated with a 0.18% (p= 0.07) increase incompensation. Intensive and non-critical care patient acuityvolumes did not have a strong association, and hence werenot included in the generalized linear model.
Other professional duties were also found to impactcompensation. While weeks of administrative time wasfound to increase compensation by 0.24% (p < 0.001),weeks of medical education decreased compensation by0.26% (p= 0.001). Time dedicated for research was notfound to have an independent impact.
Compensation
There is a wide distribution of cash compensation in ourstudy (Fig. 1). The overall median cash value of base salary,bonus, moonlighting, and total cash compensation were$256,000 (interquartile range (IQR), $213,608–315,000),$7200 (IQR, $0–26,500), $20,000 (IQR, $8000–36,500),and $280,000 (IQR, $225,000–335,750), respectively.When broken down by academic affiliation, the medianvalues for base salary, bonus, and total cash compensationwere $240,000 (IQR, $205,488–300,000), $4107 (IQR, $0–20,000), $17,000 (IQR, $8250–36, 500), and $252,400(IQR, $216,134–315,500), respectively, for academicallyaffiliated neonatologists, and $296,800 (IQR, $238,000–348,950), $20,000 (IQR, $0–50,000), $20,000 ($7850–38,750), and $330,000 (IQR, $270,000–430,000), respec-tively, for not academically affiliated neonatologists.
Supplemental compensation
One-fifth of full-time neonatologists did some type ofsupplemental clinical work. This accounted for a median of159 (IQR, 72–379) h and $20,000 (IQR, $8000–36,500) peryear. These duties were divided over level 2 (39%), level 3(46%), and level 4 (14%) units.
Predictors of compensation
Results from the generalized linear model (Table 4) show thefactors found to independently predict base salary. At anestimated annual reduction of $15,000 (6%), both academicaffiliation and practicing in the Northeast had the greatestnegative impacts. Living in the North Central region andworking with physician assistants had the greatest positiveimpact on compensation at approximately $13,000 (5.00%)and $10,000 (4.02%), respectively. As to be expected, over
Table 3 Profession workload
Variable Categories
Duties (weeks/year)a –median (IQR)
Clinical time 24 (15–36)
Research time 1 (0–10)
Administrative time 7 (1.75–15)
Medical education 2 (0–6)
Other 0 (0–0)
Clinical time (shifts/year)a–median (IQR)
Weekday days 86 (57–125)
Weekday nights 40 (24–60)
Weekend days 22 (13–30)
Weekend nights 15 (10–24)
Hours/weeka–mean (SD) Clinical service 65 (17)
Non-clinical service 43 ± 23
Rounding scheduleb–N (%) Some days on/somedays off
61 (17)
1-Week block 80 (22)
2-Week block 122 (33)
3-Week block 41 (11)
Other 62 (17)
Call typeb–N (%) In-house 124 (34)
From home 106 (29)
Both in-house andfrom home
128 (35)
Do not take call 8 (2)
Average daily roundingcensusa–median (IQR)
Total 20 (15–25)
Critical care 6 (3–10)
Intensive care 8 (5–12)
Non-critical care 3 (0–6)
Estimates of wRVUc–
median/annual billable days(IQR)
Work RVU 8709 (5591–13,146)
Professional Revenued ($)–median/annual billable days(IQR)
Government payerrate
439,117(282,935–663,752)
Provide locums,moonlighting, or perdiemb–N (%)
Yes 71 (20)
Data were collected as acontinuous variable and bnominal variablecEstimate of wRVU was derived for each respondent using thefollowing equation: (reported average daily census of critical carepatients × 7.99)+ (reported average daily census of intensive carepatients × 2.55)+ (reported average daily census of non-critical carepatients × 1.38)dProfessional revenue was derived for each respondent using thefollowing equation: (reported average daily census of critical carepatients × $402.44)+ (reported average daily census of intensive carepatients × $128.30)+ (reported average daily census of non-criticalcare patients × $72.44)
E. Horowitz et al.
the course of a career, however, time since completing fel-lowship has the greatest impact on expected compensation ata predicted $7000 (2.71%) for every 5 years post fellowship.While these factors may be modifiable to some degree, theinnate factor of gender, however, independently predicted aloss of approximately $9400 (3.68%) per year for women.
Discussion
Neonatologists have varied practice environments andprofessional obligations. This survey begins to provideinsight on factors that contribute to the fair market value forindividuals within this profession. It examined character-istics of the neonatologist, practice location, and profes-sional workload. Our results identified eight key factors thathad a significant impact on compensation. These included:time since training, amount of clinical, administrative, andmedical education time, acuity of patients, provider mix ofclinical care team, academic affiliation, location of thepractice, and gender.
The associations of factors on compensation, such as timesince training, amount of in-house clinical time, and acuity ofpatients, may not be surprising. Duration since training is anindicator of successful time in the work force, and eligibilityfor repeated raises and seniority. In-house clinical timereflects potential revenue generation through patient billings
and the need to be compensated for lesser desirable in-housecall. While more neonatologists are becoming health systememployees (49%, in this study), and therefore as salariedemployees their compensation may not be directly reliant onrevenue generation, clinical time and its relation to compen-sation is likely a point of negotiation during contract devel-opment. Further, the association of critical care patients withincreasing compensation is likely related to expected revenuegeneration and commensurate available funds forcompensation.
The impact of location on compensation, however, maynot be as straight forward. Due to increased cost of living inlarge central metropolitan areas, compensation would beexpected to increase. Our data, however, suggest theopposite. We found on the order of nearly $11,000/year lossin compensation for practicing in a more urban setting. Forneonatologists, this observation is supported by the con-cepts of non-cash compensation and social capital [8, 9].Most training programs exist in larger metropolitan areas.During the training of a neonatologist, they develop a socialcapital network. In short, they place a value on theirknowledge of the current system and location in which theyfind themselves. This has economic utility and a non-cashvalue. Combine this with the trend since the 1970s of younghighly educated workers preferring to live in larger cities, anewly trained neonatologist has another factor drawingthem to seek their first employment in an urban setting. As
Table 4 Generalized linearmodel–factors influencing basecompensation
Factors Impact (%) Impact ($)a P value
Region–North Centralb 5.00 12,813 0.02
Work with physician assistantsb 4.02 10,286 0.004
In-house callb 3.35 8579 0.005
Years post fellowship (5-year blocks)c 2.71 6927 <0.001
Administrative time–weeks/yearc 0.24 612 <0.001
Daily rounding–critical care patientsc 0.18 452 0.07
Clinical time–weekdays (daytime)a,c 0.05 125 <0.001
Medical education–weeks/yearc −0.26 −661 0.001
Work with neonatal hospitalistsb −1.97 −5030 0.10
Eligibility for annual bonusb −3.48 −8911 0.002
Gender (female vs male)d −3.68 −9425 <0.001
Large central metropolitan countyb −4.44 −11,359 <0.001
Academic (vs non-academic)d −5.86 −14,996 <0.001
Region–Mid-Atlanticb −6.12 −15,673 <0.001
Region–Northeastb −6.72 −17,193 <0.001
R2 adjusted= 0.45556
All correlations of estimates are between −0.3 and 0.3, except for Work with Physician Assistants andRegion–Northeast which had a value of −0.348aImpact ($) calculated from Impact (%) ×median compensation ($256,000)bVariable analyzed as a binomial dummy variable with potential value of 0 or 1cVariable analyzed as continuousdVariable analyzed as nominal binomial
Neonatologist salary: factors, equity and gender
newly trained individuals command a lower compensationpoint, their desire to stay in large cities may provide com-petitive pressures that reduce compensation in these areas.Then, by staying in this setting, the effects of social capitalplay a non-tangible non-cash amenity in retention andaugmenting compensation.
Regional variations, however, are less easily explained, butare found in other reports [3, 10]. Medscape also found thatphysicians living in the North Central region of the countrywere among the highest paid, while those living in theNortheast and Mid-Atlantic regions were lower compensated.While their survey did not account for specialty or profes-sional productivity differences, in our study, these regionalvariations persisted after multivariate analysis. The next ver-sion of our questionnaire will begin to explore these factors.
Many physicians feel that those in academic medicine earnless than their peers without academic affiliation. Doximityalso found this true among its members [3]. Our study furtherconfirms this belief. After correcting for other factors in ourgeneralized linear model, academic affiliation independentlypredicts a reduction in compensation by nearly 6% or about$15,000 per year, even after adjusting for workload andweeks of service. In univariate analysis, those with academicaffiliation earned a median compensation of $240,000 (IQR,$205,000–300,000), while those not affiliated with an aca-demic institution had median income of $297,000 (IQR,$238,000–349,000). Academic salaries may be further effec-tively reduced given the requirements for teaching, scholarly,and research activities required for promotion, and necessaryserving on committees in an academic institution which arenon-compensated and non-revenue generating. Further sur-veys will attempt to more robustly capture work responsi-bilities outside of clinical care provided by both academic andprivate practice neonatologists.
Of most concern however, this study further supports thegrowing evidence that female physicians have a significantlylower salary than male counterparts, even when adjusting forfactors such as hours worked and years post fellowship [11–14]. Unadjusted, we found a $60,000 gender gap in mediansalary for female neonatologists. After correcting for factorsdescribed in Table 4, this difference was reduced to just under$9500 per year. While some strategies have been proposed toaddress these sex differences [15], equity in compensationmay be a long time coming unless we continue to assess thisdifference, the contributing factors, and draw attention to thisdisparity. It is important that medicine address these inequitiesin salary due to gender not only for those now entering thework force, but for those already in practice for several yearswho have already encountered several years of salary dis-parities due to gender.
The increase in compensation seen with administrativeduties may reflect the value placed on social networks. Hortonet al. [16] describe how social capital and networks affordpotential benefits as they create connections and these con-nection facilitate access to a broader source of information at alower cost, and improve its quality, relevance, and timeliness.This is good for the bottom line. The nearly equal, butopposite in direction, impact of medical education time(teaching) on predicted compensation, however, may speak toa different economic utility on this aspect of the professionalcareer of a neonatologist. Medical education effort was anindependent factor for lower predicted compensation, evenwhen controlling for academic affiliation, research time, andclinical time and productivity. While it is critical to have adedicated pool of academic physicians to train the next gen-eration of doctors, the decrease in salary associated with thesework efforts may discourage excellent teachers from partici-pating in this critical role in academic institutions.
Fig. 1 Components of compensation and total cash compensation forfull-time neonatologists.Box plots show median, first and third quar-tiles, maximum and minimum values, and outliers to demonstrate
distribution of base compensation, annual bonus, moonlighting, andtotal annual cash compensation from primary employer by academicaffiliation status and years post fellowship
E. Horowitz et al.
Interestingly, provider types within the care team had abearing on predicted compensation. Having physician assis-tants on the team had a statistically significant influence,suggesting a nearly $10,000 increase in compensation forthose neonatologists working alongside them. While not sta-tistically significant (p= 0.10), working with neonatal hos-pitalists reduced predicted compensation. Working with nursepractitioners was not found to have a significant influence (p>> 0.20). These factors may reflect the cost of employingthese other clinical providers, their availability for hire, ortheir perceived value to the clinical care team.
The limitations of this study include the response rate of15.3%, although similar to other survey studies of com-pensation. In addition, these data were collected by anon-ymous self-report which does not allow for validation ofdata obtained or provide a means for follow-up to obtainadditional information or clarification. This can lead torecall and self-selection bias.
The strengths of this study include broad representationof different neonatologist career paths and practice types,anonymity of responses that fostered increased granularityof compensation data, and inclusion of multiple careercharacteristics to best discover potential variables to opti-mize multivariate analysis. Further, the responses were notfound to have statistically different geographic and genderdistribution than that reported by the American Board ofPediatrics for the same time period [17].
Conclusion
This compensation survey provides the first comprehensivelook at factors influencing neonatologist compensation/cost andprovides a valuable resource to those seeking to understandtheir best fair market value and to administrators trying to bestunderstand the costs of maternal–child strategic plans. Ulti-mately, this information will aid in increasing the transparencyaround compensation and health care costs, and may aid infuture work force planning for this subspecialty in pediatrics.
Acknowledgements We would like to thank the American Academyof Pediatrics Section on Neonatal-Perinatal Medicine for allowingaccess to its membership to conduct this work. We would also like tothank the members of the Section on Neonatal-Perinatal Medicine fortheir willing participation.
Author contributions EH conceptualized and designed the study,designed the data collection instrument, collected data, carried outinitial analysis, drafted the initial manuscript, and reviewed andrevised the manuscript. HAF reviewed the data collection tool, criti-cally reviewed statistical analysis, and critically reviewed the manu-script for important intellectual content. RS reviewed the datacollection tool, and critically reviewed the manuscript for importantintellectual content. All authors approved the final manuscript assubmitted and agree to be accountable for all aspects of the work.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict ofinterest.
Publisher’s note: Springer Nature remains neutral with regard tojurisdictional claims in published maps and institutional affiliations.
References
1. Kocher R, Sahni NR. Hospitals’ race to employ physicians--thelogic behind a money-losing proposition. N Engl J Med.2011;364:1790–3.
2. Salary.com. Neonatologist Compensation. 2017. http://swz.salary.com/SalaryWizard/Physician-Pediatric-Neonatology-Salary-Details.aspx. Accessed 17 July 2017.
3. Doximity. Doximity Salary Map. 2017. https://www.doximity.com/careers#4/38.34/-96.94. Accessed 17 July 2017
4. Association AMG. AMGA 2015 Medical Group Compensationand Productivity Survey. 2017. https://www.amga.org/store/detail.aspx?id=COMPSRV_2015. Accessed September 15, 2018.
5. Colleges AoAM. Report on Medical School Faculty Salaries.2017. https://services.aamc.org/dsportal2/index.cfm?fuseaction=login.login&thread=jump.FSSREPORTS&appname=FSSREPORTS&frompermissionscheck=true. Accessed Sep-tember 15, 2018.
6. Service HaHC. Physician Salary and Benefits Report. 2017.https://www.hhcsinc.com/hcs-reports.html. Accessed September15, 2018.
7. Association MGM. Physician Compensation and ProductivityReport. 2017. http://www.mgma.com/industry-data/mgma-surveys-reports/physician-compensation-and-production-survey.Accessed September 15, 2018.
8. Blyler MaC RW. Dynamic capabilities, social capital, and rentappropriation: ties that split pies. Strateg Manag J. 2003;24:677–86.
9. Peri G. Young people, skills and cities. CESifo Working Paper610. Center for Economic Studies & Ifo Institute for EconomicResearch Poschingerstr. 5, 81679 Munich, Germany,2011.
10. Grisham S. Medscape Physician Compensation Report. 2017.http://www.medscape.com/slideshow/compensation-2017-overview-6008547#9. Accessed September 15, 2018.
11. Jagsi R, Griffith KA, Stewart A, Sambuco D, DeCastro R, UbelPA. Gender differences in the salaries of physician researchers.JAMA. 2012;307:2410–7.
12. Jena AB, Olenski AR, Blumenthal DM. Sex differences in phy-sician salary in US public medical schools. JAMA Intern Med.2016;176:1294–304.
13. Kaplan SH, Sullivan LM, Dukes KA, Phillips CF, Kelch RP,Schaller JG. Sex differences in academic advancement. Resultsof a national study of pediatricians. N Engl J Med.1996;335:1282–9.
14. Jagsi R, Griffith KA, Stewart A, Sambuco D, DeCastro R, UbelPA. Gender differences in salary in a recent cohort of early-careerphysician-researchers. Acad Med. 2013;88:1689–99.
15. Byington CL, Lee V. Addressing disparities in academic medi-cine: moving forward. JAMA. 2015;314:1139–41.
16. Horton J, Millo, Y, and Serafeim, G. Resources or power?Implications of social networks on compensation and firm per-formance. J Bus FInance Account. 2012;39:399–426.
17. American Board of Pediatrics Inc. 2015-2016 Workforce Data.Chapel Hill: American Board of Pediatrics, Inc.; 2016.
Neonatologist salary: factors, equity and gender