post for guru answers only econ ct3

© 2016 American Society for Healthcare Risk Management of the American Hospital Association

Published online in Wiley Online Library (wileyonlinelibrary.com) • DOI: 10.1002/jhrm.21219

JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 35, NUMBER 414

By Hanan Edrees , DrPH, MHSA , Douglas M. Brock , PhD , Albert W. Wu , MD, MPH , Patricia I. McCotter , RN, JD, CPC , Ron Hofeldt , MD , Sarah E. Shannon , PhD, RN , Thomas H. Gallagher , MD , and Andrew A. White , MD

The experiences of risk managers in providing emotional support for health care workers after adverse events

I N T R O D U C T I O N

Medical error resulting in patient harm is a global issue. 1,2 While the emotional impact of adverse events on patients and families is typically apparent, the health care professionals involved often suffer in silence. 3,4 Health care profes- sionals who are emotionally traumatized following unanticipated adverse events are referred to as the “second victims.” 3 These individuals may experience feelings of anxiety, depression, anger, fear, shame, self-doubt, guilt, isolation, sleeping difficulties, and disturbing thoughts and memories. 5–10 Although sec- ond victims may seek support from their colleagues or relatives, 2,3 most do not receive formal psychological support. If second victims do not cope with these emotions effectively, it can lead to absenteeism, low morale, frustration, and lack of concentration when providing care to other patients. 11

After a serious adverse event, risk managers’ primary responsibilities are to con- duct an investigation into root causes, reduce the risk and loss to the organi- zation, and guide communication with the patient or family. 12,13 As part of this process, risk managers interview the involved clinicians to collect critical information related to the event. In these interactions, they commonly become acutely aware of the clinicians’ needs for emotional support 14 or referral to additional resources. 15 Risk managers may find these encounters challenging because asking health care workers to recall adverse events may trigger distress that exacerbates, rather than alleviates, the need for emotional support. 15,16

Risk managers often meet with health care workers who are emo- tionally traumatized following adverse events. We surveyed mem- bers of the American Society for Health care Risk Management (ASHRM) about their training, experience, competence, and comfort with providing emotional support to health care workers. Although risk managers reported feeling comfortable and competent in pro- viding support, nearly all respondents prefer to receive additional training. Risk managers who were comfortable listening to and supporting health care workers were more likely to report prior training. Health care organizations implementing second victim support programs should not rely solely on risk managers to pro- vide support, rather engage and train interested risk managers and provide them with opportunities to practice.

Care of the Care Provider

15DOI: 10.1002/jhr m AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 35, NUMBER 4

To mitigate the distress experienced by second victims, some organizations, including the Medically Induced Trauma Support Services (MITSS), University of Missouri, Johns Hopkins Hospital, and Brigham and Women’s Hospital, have established dedicated clinician support programs. 17 These include peer support programs, mental health and counseling services, and employee assistance programs. 18,19 Patient safety leaders have embraced these programs, and national organizations have called for all health care institu- tions to create systems to support clinicians after adverse events. 20,21 Although most clinicians would like to receive support after adverse events, they may not voice their needs to risk managers due to a belief that support may not be effective. 10 Improving how risk managers recognize and sup- port second victims represents an important opportunity to increase utilization of support services.

Although there is growing information about the preva- lence and characteristics of provider support programs, little is known about risk managers’ preparation or level of comfort in providing psychological support to distressed health care workers. 13,15 , 22–24 We developed and adminis- tered a survey to explore risk managers’ self-efficacy and experience with providing support to clinicians following an adverse event.

M E T H O D S

Setting and survey sample

The study design was a cross-sectional survey. Between May 14 and July 30, 2013, invitations to participate in a confidential web survey were sent electronically to 5272 members of the American Society for Healthcare Risk Management (ASHRM). ASHRM is a personal member- ship group of the American Hospital Association (AHA), with approximately 5800 members at the time of the sur- vey. Since this analysis focused on the attitudes and experi- ences of US risk managers, we excluded ASHRM members who self-identified as students, lived outside of the United States, did not report holding a job in risk management, or did not have risk management certification (ARM or CPHRM). Participation was encouraged through 3 e-mail reminders. Respondents were also entered into a drawing to win one of four $50 gift cards. The final analytic sample included certified risk managers and risk management staff.

Survey content

The survey was based on responses to semistructured interviews with a convenience sample of 5 experts in patient safety, risk management, and the emotional impact of adverse events. These experts included 3 co- authors (PM, RH, AWW) and 2 senior directors of risk management departments at academic hospitals. Items were refined after pilot testing with risk managers. The survey was administered using REDCap version 5.3, a secure web-based survey instrument (Vanderbilt University, Nashville TN). Definitions of adverse event,

health care worker, wellness program , and employee assistance program were provided at the beginning of the survey. The definitions used have been previously published. 16

The survey asked respondents for details describing their experience with supporting health care workers involved in adverse events, and their training to provide this support. Items assessing provider support included reported frequen- cy, comfort level, and perceived competence using a 4-point Likert scale (“strongly disagree,” “disagree,” “agree,” “strong- ly agree”). The primary outcome variables were comfort in listening to the second victim, comfort in supporting the second victim, and perceived competence in providing sup- port. Personal characteristics were collected, including age, gender, educational attainment, institutional role, and time spent working in health care. Data collected simultane- ously on characteristics of the provider support programs at respondents’ health care organizations were published in a previous study, 16 and key variables regarding these programs were included in this article, such as mechanisms and bar- riers to access support programs, referral mechanisms, location of the program within the institution, training for employees who provide support, program funding, leader- ship, legal protection, perception of programs’ effectiveness in identifying distressed health care workers, and barriers to establishing and initiating a program.

Analysis

Survey responses were excluded from analysis if the respondent aborted the survey prior to completion of the initial demographic information section (17 participants). All other responses were included in analyses.

Descriptive statistics included means and standard devia- tions for continuous variables and percentages for categor- ical variables. Questions with Likert response scales were dichotomized by agreement or disagreement (strongly agree/agree vs. strongly disagree/disagree). Analyses of categorical variables used Pearson chi-square, and Fisher’s exact tests as appropriate. All tests were two-tailed and a p -value less than .05 was considered significant. Analyses of ordinal associations used Spearman rank-ordered cor- relations. Other comparisons were limited to plausible associations. Analyses were performed using SPSS Version 19.0 (IBM, Armonk, NY).

Multivariate logistic regression was used to explore the relationships between risk manager characteristics and self-efficacy, and between facility type and self-efficacy. This was done using forward selection with maximum likelihood estimators. Analysis of variance (ANOVA) was used to examine the differences among facilities in risk managers’ reports of self-efficacy. Post-hoc analyses, apply- ing a Bonferroni contrast ( p < .05) to reduce risk of Type I error, were conducted to explore paired differences. We analyzed the relationships of risk managers who reported having been trained in providing support with their reported self-efficacy and frequency in providing support.

JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 35, NUMBER 4 DOI: 10.1002/jhr m16

We also looked at the associations between reported self- efficacy and risk manager characteristics and facility type. We further explored whether facility type was associated with risk managers’ comfort in listening, comfort in sup- porting, and competence in providing support.

R E S U LT S

Respondent characteristics

A total of 641 of the 5272 eligible ASHRM members completed the survey, for an overall response rate of 12.2%. We excluded 62 non–risk managers, leaving an analytic sample of 579. Among responding individu- als, 89.1% were female, with a mean age of 53.4 years ( Table 1 ). Respondents worked in several different set- tings, with the largest group (38.2%) employed by private hospitals within a health care network. The respondents possessed considerable institutional and personal experi- ence; the average time spent working in health care was 28.3 years, and respondents had worked at their current facility an average of 12.4 years. Respondents also reported extensive formal training and certification. Approximately four-fifths (82.4%) of survey respondents reported having a degree. Of those reporting a degree, 80.0% reported a clinical degree (eg, MD, RN, BSN, MSN). Nurses con- stituted the vast majority (97.6%) of respondents who reported a clinical degree. Risk management certifica- tions (CPHRM, ARM, LHRM) were held by a majority (59.2%) of the total number of survey respondents.

Existence, characteristics, and efficacy of organizational support programs

As previously described, a majority of risk managers (75.3%) reported the presence of a support program at their institution. Programs were most commonly located within the organization’s employee assistance program (EAP), followed by the risk management department ( Table 1 ). In these support programs, 90.1% involved EAP counselors, followed by risk managers and pastoral care staff, who participated at 54.8% and 47.0% of sites, respectively. Among support programs, 39.9% offered training to personnel who support health care workers and 14.6% had a manual to guide support providers. A total of 29.7% of risk managers reported that the program at their institution was very effective, 54.1% somewhat effective, 10.3% somewhat ineffective, and 4.2% very ineffective, at supporting health care workers.

Risk managers’ perceived self-efficacy in providing emotional support

Nearly 90% of risk managers reported that they person- ally provided emotional support to health care workers involved in adverse events during the last year, although the frequency varied widely. The most common category for providing emotional support was 1–6 times per year (49.4%), followed by provision on a monthly (21.9%),

or weekly (13.3%) basis, rarely or never (11.7%), or daily (3.6%) ( Table 1 ). The proportion of respon dents who reported they provided support on a weekly or daily basis did not differ significantly between those from organizations with or without a support program ( p = ns).

Risk managers reported high levels of comfort in their listening and support roles: 94.1% strongly agreed or agreed that they “feel comfortable listening to the health care worker seeking support,” and 93.2% strongly agreed or agreed that they “feel comfortable supporting the health care worker seeking support.” Most (85.7%) risk managers strongly agreed or agreed they were “competent in providing support to health care workers after adverse events.” Self-reported comfort listening to health care workers, comfort supporting health care workers, and competence in providing support were moderately and significantly positively correlated with the frequency with which respondents reported providing emotional support ( R = .30, .31, .35, respectively, all p < .001).

Risk managers’ training in providing support

About half of risk managers (52.2%, n = 302) reported they had received training on how to provide emotional sup- port to health care workers. The respondents who received training reported various modes of education. Lecture was the most common modality (64.6%), followed by books or articles (55.6%), role-play (50.7%), and live or video demonstration (42.4%). Approximately 14.6% of these respondents indicated that they received another form of training, which included formal coursework or an advanced degree in counseling or psychology (50%), experience in providing support as part of the job (31.3%), or training as part of a clinical degree or clinical rotation (12.5%).

Relationships between perceived self-efficacy and facility type and respondent characteristics

Risk managers’ perceived self-efficacy and training in providing support. Of the 302 risk managers who reported receiving training in providing support, 96.6% agreed they were competent to train others to provide emo- tional support. However, 92.4% also “strongly agreed” or “agreed” they would benefit from receiving additional training. Of the risk managers who reported having been trained in providing support, approximately 49.7% pro- vided support 1–6 times a year, 21.5% provided support on a monthly basis, 16.9% offered support weekly, 7.6% rarely or never offered support, and 4.3% offered support daily ( Table 2 ).

Risk managers’ perceived self-efficacy and facility type. Overall, facility type was not related to risk manager per- ceived competency in providing support.

Risk managers’ perceived self-efficacy and respondent char- acteristics. Within the subset of risk managers (93.2%, n = 536) who agreed they felt comfortable providing

DOI: 10.1002/jhr m AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 35, NUMBER 4 17

Table 1: Respondent and Program Characteristics Primary Variables Measures/Interview Questions & Scales N (%) a Respondent characteristics Age (mean years, Std Dev) 53.4 (8.11)

Gender ( n = 569) Female 507 (89.1%) Male 62 (10.9%)

Primary Work Facility Characteristics ( n = 579) Private hospital in a network 221 (38.2%) Private unaffiliated hospital 84 (14.5%) University-based hospital 72 (12.4%) Rural/Critical access hospital 40 (6.9%) County or state hospital 32 (5.5%) Private multispecialty group 24 (4.1%) Federal or Veterans Affairs hospital 3 (0.6%) Ambulatory surgery center 3 (0.6%) Other 99 (17.1%)

Support program characteristics Existence of Support Program ( n = 579) Yes 436 (75.3%) In the process of developing a program 43 (7.4%) No 65 (11.2%)

Administration location of support programs ( n = 436) Employee assistance program 392 (89.9%) Risk management 127 (29.1%) Wellness program 113 (25.9%) Quality Improvement/Patient Safety 59 (13.5%) Otherb 271 (65.2%)

Outcome Variables Frequency of providing support 1–6 times per year ( n = 579) 286 (49.4%)

Monthly basis 127 (21.9%) Weekly basis 77 (13.3%) Rarely or never 68 (11.7%) Daily basis 21 (3.63%)

Training experience Received training ( n = 579) Yes 302 (52.2%)

Type of training received ( n = 302) Lecture 195 (64.6%) Books or articles 168 (55.6%) Role-play 153 (50.7%) Live or video demonstration 128 (42.4%) Other forms of training 44 (14.6%)

Perceived self-efficacy “I feel comfortable listening to the health care worker seeking support.” ( n = 579)

Strongly agree/Agree 545 (94.1%) “I feel comfortable supporting the health care worker seeking support.” ( n = 575)

Strongly agree/Agree 536 (93.2%) “I am competent in providing support to health care workers after adverse events.” ( n = 574)

Strongly agree/Agree 492 (85.7%)

a Total responses may not sum to 100% due to rounding and nonresponse. bMedical director or executive leadership, nursing or patient care services, claims management, an individual department, pro- fessional liability insurer program external to the organization, spiritual care/pastoral care/chaplain services, human resources, peer support program, external mental health service, social work, physician advocate resource, departmental manager.

JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 35, NUMBER 4 DOI: 10.1002/jhr m18

support to health care workers in distress, 54.9% had received training to support others ( p < .001), and 56.2% agreed or strongly agreed they were able to train others ( p < .001). Despite feeling comfortable with providing support to health care workers, 94.4% of risk managers agreed or strongly agreed they valued additional training or experience in providing support ( p = .002).

Two multivariate logistic regression analyses were used to select significant predictors of risk manager comfort listen- ing to health care workers in distress and competence in providing support. Each regression model included the fol- lowing independent variable: respondent age, gender, years at their institution, years in health care, holding a clinical degree, having received support training, frequency of pro- viding support, ability to train others, belief in the value of training, and training role. Table 3 shows the significant predictors of comfort listening to health care workers and competence in providing support. Risk managers who felt comfortable listening to health care workers in distress were significantly more likely to report that they received training on how to emotionally support others ( p = .028), frequently provided support ( p = .004), felt confident in training others ( p = .027), and valued additional training to support others ( p = .001). Risk managers who indicated that they were competent in providing support reported providing support more frequently ( p = .000), confidence in training others ( p = .000), and would value additional training to support others ( p = .000).

D I S C U S S I O N

Many health care workers suffer in silence after unan- ticipated adverse events. There is growing consensus that organizations should offer timely and easily accessible support for these health care workers. 4,20 Given that risk managers are among the first to interact with these individuals after serious adverse events, a deeper under- standing of how they are prepared for these conversations could inform recommendations about their participation in a support program. However, little is known about

risk managers’ preparation for supporting second vic- tims. Our study provides novel insight into risk manag- ers’ training, experience, self-assessed competence, and comfort with providing emotional support to health care workers.

In this large national survey, the majority of risk manag- ers reported providing emotional support to health care workers after adverse events at least once a year, and approximately 40% at least monthly. Nearly all felt com- fortable and confident in doing so. However, only half of the respondents had received formal training in this type of counseling, and most agreed they would benefit from more education. The fact that most of the respondents provide support six or fewer times a year may indicate that many risk managers either do not routinely pro- vide support to all health care workers involved in the claims they manage, that second victims are not seeking support from risk managers, or a combination of these factors. We found an association between prior training in providing emotional support and a greater frequency of providing support. This suggests that one strategy for improving support for second victims is for health care institutions to train selected risk managers and encourage them to routinely practice providing support. Our find- ing that risk managers provide support to second victims relatively infrequently does not seem to be explained by personal discomfort with listening or providing support.

Although we hypothesized that environmental factors, such as hospital type or availability of an institutional sup- port program, might be associated with self-assessed com- petency at providing support to second victims, we instead found significant associations with personal characteristics. For example, risk managers who reported high levels of self-efficacy in offering support to second victims were those that had a clinical background, had been trained in providing support, and provided support more frequently. This finding may indicate that risk managers’ clinical experiences allow them to effectively relate to second victims or perhaps patient care delivery fosters empathic communication skills that enhance the ability to support

Table 2: Characteristics of Risk Manager Who Have Been Trained to Provide Emotional Support ( n = 302) “How often do you provide emotional support to clinicians involved in adverse events?” n (%)

1–6 times a year 150 (49.7%) Monthly 65 (21.5%) Weekly 51 (16.9%) Rarely or never 23 (7.6%) Daily 13 (4.3%)

I would be able to train others to provide emotional support to health care workers after adverse events. Strongly agree/Agree 216 (72.0%)

I would value additional training or experience in responding to health care workers seeking support. Strongly agree/Agree 279 (92.4%)

DOI: 10.1002/jhr m AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 35, NUMBER 4 19

second victims. Self-assessed competency in supporting second victims of adverse events or medical errors was also related to comfort levels in listening to second victims. These findings suggest that providing training and experi- ence may help to increase risk managers’ self-efficacy and effectiveness in responding to second victims’ needs. This is consistent with studies that highlight the importance of training support providers. 19,21,25 Ultimately, the effective- ness of training for support programs should be assessed by systematic data collection from second victims.

Implications

Health care organizations should support health care workers after adverse events to promote staff well-being, retention, and provider readiness to give high-quality care. Many health care institutions have invested in support services, but existing resources are generally underused. 26,27 Organizations should define the role of risk managers in the support delivery system. One option is to provide support services independently of risk management and managerial functions. This approach allows risk managers to focus their practice and addresses a barrier to utilization of support ser- vices; some health care workers may find it uncomfortable to receive support from risk managers. However, our find- ings confirm that most risk managers have experience and comfort with providing emotional support to health care workers after adverse events, which could be an organiza- tional asset. Furthermore, risk managers may encounter sit- uations in which immediate emotional support is required

and referral would not be appropriate. Our finding that risk managers vary widely in how often they support second victims suggests the need to provide formal training for all risk managers in providing psychological first aid 28,29 and basic counseling skills. Additionally, risk managers should be aware of other existing support services and be quick to refer to them as needed.

Some health care workers may perceive potential con- flicts of interest for the risk manager given their primary responsibility for conducting investigations within their organizations. Therefore, coordination is needed within organizations to delineate the roles and responsibilities of these different entities, including EAP services. 30 Future efforts and research should clarify the roles of risk manag- ers, EAP counselors, and dedicated second-victim support program resources for responding to the needs of health- care workers. Even though there is relatively little pub- lished literature on the structure of existing and potential support programs, some programs indicate the impor- tance of including EAPs as part of the support process and note that training is critical for an effective support program. 19,26,31 Moreover, published literature highlights the lack of standardization in support training and in training EAP management. 32

Little is known about the most effective training modal- ity for increasing risk managers’ self-efficacy in provid- ing emotional support to second victims. However, it is notable that most risk managers received training through

Table 3: Multivariate Logistic Regression: Predictors of Risk Manager Reported Comfort Listening to and Competence in Providing Support to Second Victims Risk Managers Who Felt Comfortable Listening to Health care Workers

Predictors* Odds Ratio p -value 95% CI Received training 3.75 p =.028 1.15–12.19 Frequently provided support to health care workers 2.93 p =.004 1.41–6.06 Able to train others 2.28 p =.027 1.10–4.76 Value additional training to support others 3.42 p =.001 1.68–7.00 –2 Log likelihood 123.1 Nagelkerke R 2 0.31 n 381

Risk Managers Who Indicated That They Were Competent in Providing Support

Predictors* Odds Ratio p -value 95% CI Received training 5.71 p = .000 2.32–14.03 Frequently provided support to health care workers 2.95 p = .000 1.76–4.92 Able to train others 4.52 p = .000 2.52–8.14 –2 Log likelihood 125.3 Nagelkerke R 2 0.35 n 378

*Model includes only statistically significant predictors.

JOURNAL OF HEALTHCARE RISK MANAGEMENT • VOLUME 35, NUMBER 4 DOI: 10.1002/jhr m20

lectures, which may not include practice of these com- plex communication skills. It may be helpful to develop training to the level of support likely to be offered by risk managers. For example, at Johns Hopkins, these include: Stage 1—basic emotional first aid; Stage 2—ongoing counseling services; and Stage 3—medical treatment. Stage 1 focuses on offering immediate and basic emotional sup- port to those in distress; this is similar to what risk manag- ers and peers may provide currently on a less systematic basis to second victims. Comprehensive training should be available for Stage 1, and this training can be offered to all of those who provide basic support to individuals within the organization, including peer clinicians and other work- ers. Referrals for Stage 2 services can be offered through groups offering formal counseling, and referrals for Stage 3 services can be addressed through existing medical systems.

Limitations

Our study has important limitations. First, the low response rate indicates that results could be subject to nonresponse bias. However, the basic demographic char- acteristics of our sample were representative of ASHRM’s membership. Second, the results may be subject to recall and social desirability bias, in particular with respect to reports of self-efficacy at providing support. We do not have any independent measures, such as from direct observation or survey of second victims, that might be used to validate our measures. Third, the survey did not ask about other factors that might contribute to variability in support practices, such as accessibility of risk managers or institutional protocols guiding them to refer to other support services rather than providing support themselves.

C O N C L U S I O N S

This exploratory survey study confirmed that risk man- agers are a potential source of emotional support for health care professionals following adverse events that cause patient harm. Our results also highlight the critical need for further identification of specific training needs and best practices. Future studies can focus on detailed processes to develop effective organizational support programs and specific training for those providing emo- tional support, including risk managers. Because second victims may not seek support due to the belief that it is not effective, training programs for risk managers should be coupled with efforts to track outcomes for clinicians and to market the utility of support services. It is time for health care professionals who have been a part of harmful adverse events to no longer suffer in silence but to share their pain with trained and compassionate professionals, like risk managers, who nurture these wounded caregivers.

F U N D I N G

Funding was provided by the Agency for Healthcare Research and Quality (Grant R18HS019531), as well as funding for REDCap through the National Institutes of

Health (UL RR025014). The funding sources had no role in the design, conduct, or reporting of the study or in the decisions to submit the manuscript for publication.

R E F E R E N C E S

1. Schoen C , Osborn R , Huynh PT , et al. Taking the pulse of health care systems: experiences of patients with health problems in six countries . Health Aff . 2005 ; 10 : 1377 .

2. Leape LL. Error in medicine . JAMA . 1994 ; 272 ( 23 ): 1851 – 1857 .

3. Wu AW. Medical error: the second victim: the doctor who makes the mistake needs help too . BMJ . 2000 ; 320 ( 7237 ): 726 .

4. Denham CR. TRUST: the 5 rights of the second vic- tim . J Patient Saf . 2007 ; 3 ( 2 ): 107 – 119 .

5. Rassin M , Kanti T , Silner D . Chronology of medica- tion errors by nurses: accumulation of stresses and PTSD symptoms . Issues Ment Health Nurs . 2005 ; 26 : 873 – 876 .

6. Newman MC. The emotional impact of mistakes on family physicians . Arch Fam Med . 1996 ; 5 : 71 – 75 .

7. Joesten L , Cipparrone N , Okuno-Jones S , DuBose ER . Assessing the perceived level of institu- tional support for the second victim after a patient safety event . J Patient Saf . 2014 ; 11 : 21 .

8. Paparella S. Caring for the caregiver: Moving beyond the finger pointing after an adverse event . J Emerg Nurs . 2011 ; 37 ( 3 ): 263 – 265 .

9. O’Beirne M , Sterling P , Palacios-Derflingher L , Hohman S , Zwicker K . Emotional impact of patient safety incidents on family physicians and their office staff . J Am Board Fam Med . 2012 ; 25 ( 2 ): 177 – 183 .

10. Waterman AD , Garbutt J , Hazel E , et al. The emo- tional impact of medical errors on practicing physi- cians in the United States and Canada . Jt Comm J Qual Patient Saf . 2007 ; 33 ( 8 ): 467 – 476 .

11. Bell SK , Moorman DW , Delbanco T . Improving the patient, family, and clinician experience after harmful events: The “when things go wrong’’ curriculum . Acad Med . 2010 ; 85 : 1010 – 1017 .

12. Perry DG , Bokar V . Different roles, same goal: risk and quality management partnering for patient safety. By the ASHRM Monographs Task Force . J Healthc Risk Manag. 2007 ; 27 ( 2 ): 17 – 25 . doi: 10.1002/ jhrm.5600270205

13. Wu AW , Lipshutz AK , Pronovost PJ . Effectiveness and efficiency of root cause analysis in medicine . JAMA . 2008 ; 299 ( 6 ): 685 – 687 .

DOI: 10.1002/jhr m AMERICAN SOCIETY FOR HEALTHCARE RISK MANAGEMENT • VOLUME 35, NUMBER 4 21

14. Wu AW , Steckelberg RC . Medical error, incident investigation and the second victim: doing better but feeling worse? BMJ Qual Saf . 2012 ; 21 ( 4 ): 267 – 270 .

15. Gallagher TH , Brundage G , Summy EA , et al. National survey: risk managers’ attitudes and experi- ences regarding patient safety and error disclosure . J Healthc Risk Manag . 2006 ; 26 ( 3 ): 11 – 16 .

16. White AA , Brock DM , McCotter PI , et al. Risk man- agers’ descriptions of programs to support second vic- tims after adverse events . J Healthc Risk Manag . 2015 ; 34 ( 4 ): 30 – 40 .

17. Edrees H , Federico F . Supporting clinicians after medical error . BMJ . 2015 ; 350:h1982 .

18. MITSS tools: tools for building a clinician and staff support program. MITSS (Medically Induced Trauma Support Services) Web site. http://www.mitsstools .org/tool-kit-for-staff-support-for-healthcare- organizations.html . Published 2014.

19. Scott SD , Hirschinger LE , Cox KR , et al. Caring for our own: deploying a systemwide second victim rapid response team . Jt Comm J Qual Patient Saf . 2010 ; 36 ( 5 ): 233 – 240 .

20. Sentinel event policy & procedures: comprehen- sive accreditation manual for hospitals . The Joint Commission Web site. http://www.jointcommission .org/assets/1/6/CAMH_24_SE_all_CURRENT.pdf . Published 2015.

21. Safe practices for better healthcare—2010 update: a consensus report. National Quality Forum Web site. http://www.qualityforum.org/Publications/2010/04/ Safe_Practices_for_Better_Healthcare_–_2010_ Update.aspx . Published 2010.

22. White AA , Waterman AD , McCotter P , Noyle DJ , Gallagher TH . Supporting health care workers after medical error: considerations for health care leaders . JCOM . 2008 ; 15 ( 5 ): 240 – 247 .

23. Carr S. Disclosure and apology: what’s missing? Medically Induced Trauma Support Services (MITSS). http://www.mitss.org/MITSS_WhatsMissing.pdf . Published 2009.

24. van Pelt F. Peer support: Healthcare professionals sup- porting each other after adverse medical events . Qual Saf Health Care . 2008 ; 17 : 249 – 252 .

25. McDonald TB , Helmchen LA , Smith KM , et al. Responding to patient safety incidents: the “seven pillars .” Qual Saf Health Care . 2010 ; 19 ( 6 ): e11 .

26. Hu YY , Fix ML , Hevelone ND , et al. Physicians’ needs in coping with emotional stressors: the case for peer support . Arch Surg . 2012 ; 147 ( 3 ): 212 – 217 .

27. Scott SD , Hirschinger LE , Cox KR , McCoig M , Brandt J , Hall LW . The natural history of recovery for the healthcare provider “second victim” after adverse patient events . Qual Saf Health Care . 2009 ; 18 ( 5 ): 325 – 330 .

28. Everly GS Jr , Flynn BW . Principles and practice of acute psychological first aid after disasters . In: Everly GS Jr , Parker CL , eds. Mental Health Aspects of Disaster: Public Health Preparedness and Response . Baltimore, MD : Johns Hopkins Center for Public Health Preparedness ; 2005 : 68 – 76 .

29. McCabe OL , Everly GS Jr , Brown LM , et al. Psychological first aid: a consensus-derived, empiri- cally supported, competency-based training model . Am J Public Health . 2014 ; 104 ( 4 ): 621 – 628 .

30. Cagliano AC , Grimaldi S , Rafele C . A systemic meth- odology for risk management in healthcare sector . Saf Sci . 2011 ; 49 ( 5 ): 695 – 708 .

31. Pratt S , Kenney L , Scott SD , Wu AW . How to develop a second victim support program: a toolkit for health care organizations . Jt Comm J Qual Patient Saf . 2012 ; 38 ( 5 ): 235 – 240 .

32. Rothermel S , Slavit W , Finch RA , Marlo K , Dan D . Center for Prevention and Health Services . An employer’s guide to employee assistance programs: recommendations for strategically defining, integrating and measuring employee assistance programs . Washington, DC : National Business Group on Health ; 2008 .

A B O U T T H E A U T H O R S

Hanan Edrees, DrPH, MHSA, is a researcher, Department of Health Policy & Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, and Quality Manager at Ministry of National Guard-Health Affairs, Saudi Arabia. Douglas M. Brock, PhD, is Associate Professor, Department of Family Medicine and MEDEX Northwest, University of Washington School of Medicine, Seattle, Washington. Albert W. Wu, MD, MPH, is Professor of Health Policy and Management and Medicine, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland. Patricia I. McCotter, RN, JD, CPC, is Director, Patient Safety Innovations and Provider Support, Physicians Insurance A Mutual Company/Experix, Seattle, Washington. Ron Hofeldt, MD, is Director, Physician Affairs, Physicians Insurance A Mutual Company/Experix, Seattle, Washington. Sarah E. Shannon, PhD, RN, is Associate Professor of Nursing, Department of Biobehavioral Nursing and Health Systems, University of Washington, Seattle. Thomas H. Gallagher, MD, is Professor, Department of Medicine and Department of Bioethics & Humanities, University of Washington School of Medicine, Seattle. Andrew A. White, MD, is Assistant Professor of Medicine, Department of Medicine, University of Washington School of Medicine, Seattle.