Introduction

In the last three decades, moral distress has garnered the attention of researchers, because it has become a central and dominant issue in healthcare.^1,2 The term “moral distress” was originally coined by the nurse-philosopher, Andrew Jameton, in 1984, and defined as a situation that arises “when one knows the right thing to do, but institutional constraints make it nearly impossible to pursue the right course of action.”^3(p6) Studies show that moral distress has biological, psychological, and stress-related consequences on healthcare workers and these adverse effects undermine the provision of optimal care for patients.^4,5,6  An understanding of the predictive relationship between moral distress and the propensity for moral disengagement can potentially increase the knowledge base for healthcare organizations and residency faculty to address these negative phenomena. It could provide the scientific evidence that hospital administrators, directors of residency programs, medical faculty, and policymakers need to develop appropriate mechanisms to effectively address the issues of moral distress and the propensity for moral disengagement among residents. Furthermore, it could generate the appropriate language and clarity residents need to articulate distressing situations they may have experienced and thought they were isolated and personal problems. It also empowers them to seek help when they experience morally distressing situations, knowing fully that moral distress and disengagement have negative impacts among healthcare professionals. Moreover, moral distress can lead healthcare providers to become desensitized to the moral dimensions of their work or even to leave their profession.⁷  

Moral distress is a common phenomenon in the daily training and experience of residents.⁸ The common causes of moral distress are clinical, organizational, interpersonal, and personal.^9,10,11 Evidence indicates that unaddressed moral distress during residency education, not only contributes to career dissatisfaction and ineffective patient care, but also leads to detachment, loss of empathy,¹² burnout, depersonalization,^13,14 moral numbness, blunting, and denial.^15,16 Such negative consequences and feelings have not only been conceptually identified as precursors to dehumanizing attitudes in medicine,^17,18,19 but could also lead to moral disengagement among healthcare workers.^20,21,22

Historically, some physicians and other healthcare professionals have engaged in inhumane treatment and torture of war detainees, prisoners, and terror suspects, as was the case in Nazi Germany, during the Second World War.^23,24,25. Similarly, abuse of prisoners by physicians and allied health workers was seen in Apartheid South Africa,^26,27, and in Iraq, especially during the U.S.-led invasion of 2003.^28,29 Instances of dehumanization and victim blame abound in health care when health professionals treat patients “less like persons and more like objects.”^30(p176)  Such morally disengaging actions impact patient safety and quality of care as well as engender healthcare inequality.^31,32 The rationale for bringing in moral disengagement into the analysis of moral distress in this study is because both are negative phenomena that adversely impact healthcare workers and the care they provide to patients. 

Moral distress has been loosely associated with the propensity for moral disengagement.^33,34 Moral disengagement refers to the “social-cognitive process in which people disengage their internalized moral standards from external immoral behaviors in order to persuade themselves that the deviant behaviors are actually morally permissible”.^35,36(p330) Some of the research that explored the possible connection between moral distress and the propensity for moral disengagement among healthcare workers, has largely focused on the negative consequences of being morally desensitized in a clinical setting, without delving into the causes of the moral disengagement.^37,38,39 Of these studies, only a few have examined morally disengaged nurses and found they violated ethical standards and engaged in counterproductive workplace behaviors. The studies demonstrate that the more nurses disengage from their moral self-sanctions (i.e. individual moral standards that serve as guides and deterrents for conduct), the more likely they exhibit unruly behavior to colleagues and disregard organization norms.^40,41 When individuals disengage from their moral standards, they can behave inhumanely and still believe they are doing what is right.⁴² The research on morally disengaged healthcare workers is limited to the nursing profession, and not much is known about morally disengaged medical residents who experience moral distress during residency training. The present exploratory study intends to fill the gap and contribute to the body of literature, by examining the relationship between moral distress and the propensity for moral disengagement among medical residents and the role ethical climate plays in moderating the relationship between both negative phenomena. The researcher hypothesized that the frequency and intensity of the experience of moral distress (MD) among U.S medical residents is predictive of the propensity for moral disengagement (PMD), and the perception of the hospital’s ethical climate (EC) moderates this predictive relationship. Ethical climate is defined as “those shared perceptions of organizational practices related to ethical decision-making and reflection, and includes issues of power, trust and human interactions within an organization.”^43(p345) Studies indicate that a positive ethical climate has the propensity to encourage ethical behavior, teamwork, job satisfaction, and retention more than a negative ethical climate.^44,45,46 Also, perception of the hospital’s ethical climate has been shown to have an impact on the experience of moral distress by healthcare workers.^47,48,49

The study was guided by the following research questions:

RQ1: To what extent, if any, does the frequency and intensity of moral distress predict the propensity for moral disengagement among U.S. residents?

H1: The frequency and intensity of moral distress are statistically significant predictors of the propensity for moral disengagement among U.S. residents. 

RQ2: Does the perception of the ethical climate of the hospital moderate the predictive relationship between moral distress and the propensity for moral disengagement among U.S. residents?

H2: The perception of the ethical climate of the hospital does moderate the predictive relationship between moral distress and the propensity for moral disengagement among U.S. residents.

Methods

Study Design and Subjects

This study used a cross-sectional, correlational design to examine the relationship between moral distress and the propensity for moral disengagement, by testing whether the frequency and intensity of moral distress is predictive of moral disengagement among U.S medical residents, and the role of ethical climate in moderating the relationship between both phenomena. Online surveys were sent to 150 residents at fourteen residency programs in the Eastern and Mid-Western United States in 2018. Surgical residents were not part of the study. An email containing the survey link hosted on Qualtrics was sent to the residents describing the research, the length of time to complete, confidentiality, and the consent form. The residents were advised to discontinue the survey if they felt any emotional discomfort with any of the questions. Weekly reminders were sent to them and no remuneration was given for participation.

Survey Instruments

Moral Distress: The study used the 21-item Moral Distress Scale-Revised (MDS-R), by Hamric and colleagues.⁵⁰ The scale assesses the frequency and intensity of an individual’s experience of moral distress. The items were scored by frequency (how often the situation arises), and level of disturbance/intensity (how disturbing the situation is when it occurs). The frequency scale ranges from 0 (never) to 4 (very frequently), and the intensity scale, 0 (none) to 4 (great extent). The frequency and intensity scores for each item of the scale were summed separately. To generate a composite score, the frequency score and intensity score for each item were multiplied and added together (frequency score x intensity score which ranges from 0 to 16). Then, the final score of moral distress was obtained by summing each item’s frequency and intensity score which has a range of 0-336.^51,52 Adequate reliability and evidence of construct validity were appropriately demonstrated for MDS-R as an instrument for measuring moral distress in different health care settings within the hospital.⁵³

Moral Disengagement: The study measured the propensity for moral disengagement with the 16-item Propensity to Moral Disengage Scale (MDS), designed by Moore et al.⁵⁴ The items were measured on a 9-point Likert scale, ranging from “strongly disagree” to “strongly agree”. The scale measured the eight mechanisms outlined by Bandura⁵⁵ that are activated for moral disengagement, namely: moral justification, euphemistic labeling, advantageous comparison, displacement of responsibility, diffusion of responsibility, disregard, or distortion of consequences, dehumanization, and attribution of blame.  The total score was calculated according to the author’s guidelines by summing across the items and calculating its average. Higher scores indicate a higher propensity to morally disengage in accord with Bandura’s notion of moral disengagement. The instrument has been validated extensively in research.^56,57

Ethical Climate: The study used the Ethics Environment Questionnaire (EEQ) developed by McDaniel⁵⁸ that offers a measure of ethics in healthcare services. This is a 20-item questionnaire measured on a 5-point Likert scale with response choices ranging from Strongly Agree (5) to Strongly Disagree (1). Following the author's guidelines, an overall score was obtained by an average of the summative score taken with the assigned points on the scale being: Strongly Agree =5; Agree = 4; Undecided =3; Disagree =2; Strongly Disagree =1. The overall mean score of EEQ was 3.1 out of 5.0 and the individual mean scores ranged from 1.8 to 4.8, however, the item raw score was from 1.0 to 5.0. The construct validity, criterion validity, and predictive validity were assessed and established.⁵⁹

Sociodemographic Characteristics: Participants were asked about their program, year of residency, gender, age, race, marital status, level of ethics education, the highest level of education completed, and where they went to medical school (U.S. or abroad).

Sample Size and Statistical Power: A sample size analysis was conducted using G*Power 3.1.9.2.  The a priori sample size analysis indicated that for a two-tailed bivariate correlation, a sample size of at least 84 participants is necessary.  The a priori sample size analysis for multiple linear regression with a medium effect size (f2 = .15) and two predictors a sample size of at least 68 participants is necessary.

IRB Approval: Approval for the study was granted by Eastern University’s Institutional Review Board.

Data Analysis

The data was entered into SPSS version 25 for data management and analysis. The primary outcomes measured were MD, PMD, and EC. The scores for moral distress frequency (MD-F), moral distress intensity (MD-I), MD composite score, PMD, and EC were created using the scoring instructions set forth by the respective authors. The descriptive statistics for the dataset and sociodemographic characteristics were calculated. Pearson correlation analysis, linear regression analysis, and moderation analysis were performed to examine the predictive relationship between MD and PMD and the moderating role of the EC.

Results

Of the 150 residents who received the survey, 101 responded representing a 67.33% response rate. Five incomplete responses were eliminated, leaving a final tally of 96 complete responses with 47 residents in internal medicine (49%), 24 residents in family medicine (25%), 9 residents in emergency medicine and family medicine (9%), 8 residents in emergency medicine and internal medicine (8%), 3 residents in emergency medicine (3%), 2 residents in critical care (2%), 1 resident in dermatology (1%) and 2 residents did not check any program. Sample demographics are presented in Tables 1a and b:

Table 1a

Frequencies and Percentages for Categorical Demographic Items

Table 1b

Frequencies and Percentages for Categorical Demographic Items

Table 2

Means and Standard Deviations for MD, MD-F, MD-I, EC, and PMD

Reliability: Reliability analyses for MD, MD-F, MD-I, EC, and PMD indicated that all the scales exhibited greater than acceptable reliability, with coefficients above 0.7. EC (α = 0.80), MD (α = 0.96), MD-F (α = 0.97), MD-I (α = 0.94), and PMD (α = 0.99).

Correlation Analysis between MD (MD-F and MD-I) and PMD

The correlation coefficient (r_p = .93) indicated a statistically significant relationship between MD-F and PMD.  Also, the correlation coefficient (r_p = .59) indicated a positive relationship was found between MD-I and PMD, p < .001. These results show a positive relationship between MD-F and PMD as well as MD-I and PMD. Table 3 shows the analysis.

Table 3

Pearson Correlation Matrix between MD-F and PMD and between MD-I and PMD

Note 1: For MD-F vs PMD. The critical values are 0.27, 0.35, and 0.44 for significance levels .05, .01, and .001 respectively.

Note 2: For MD-I vs PMD. The critical values are 0.28, 0.36, and 0.45 for significance levels .05, .01, and .001 respectively.

Linear Regression Analysis with MD-F and MD-I Predicting PMD

The linear regression analysis was statistically significant, F(2, 47) = 152.71, p < .001. The R² was .87. When MD-F and MD-I were assessed as individual predictors of PMD.  MD-I was not a statistically significant predictor of PMD, B = 0.00, t(47) = 0.43, p = .671, while MD-F was a statistically significant predictor of PMD, B = 0.08, t(47) = 13.53, p < .001, indicating MD-F contributed more to the variance in PMD than MD-I.  This result partially supports H1 because when combined, both MD-F and MD-I are significant predictors of PMD, but as individual predictors, MD-F predicted PMD more than MD-I.   Table 4 shows the results:

Table 4

Results for Linear Regression with MD-F and MD-I Predicting PMD

Unstandardized Regression Equation: PM.Disengagement = -0.55 + 0.08*M.Distress-Frequency + 0.00*M.Distress-Intensity

Moderation Analysis

The moderation analysis for the second research question did not indicate any moderation effect and the hypothesis was rejected.  The results show that the perception of EC does not moderate the predictive relationship between MD and PMD. Table 5 shows the results:

Table 5

Moderation Analysis Table with PMD Predicted by MD Moderated by EC

Discussion

This exploratory study was the first to examine the predictive relationship between moral distress and the propensity for moral disengagement and the moderating role of ethical climate in a population of medical residents. The correlation between moral distress and the propensity for moral disengagement shows that the frequent occurrence of moral distress and the intensity associated with it predisposes residents to morally disengage. This study yields new insight into a common phenomenon in residency and challenges the administration and leadership of health care institutions, especially residency directors, to recognize the negative impact on residents and put mechanisms in place to address it whenever it occurs. In contrast, the moderation analysis, which examined whether the perception of the hospital’s ethical climate moderated the predictive relationship between moral distress and the propensity for moral disengagement was not supported.   

The findings are consistent with a previous study,⁶⁰ that found a correlation between moral distress in internal medicine residents and depersonalization, which is a precursor to moral disengagement.^61,62,63 However, this study’s high level of moral distress is inconsistent with a prior study by Abbasi et al⁴⁸ that found a moderate level of moral distress among Iranian physicians.  The mean moral distress scores for the residents in this study, MD-F, 51.61 [SD = 24.98], and MD-I 59.12 [SD = 19.08] were higher than the mean in the study by Abbasi et al which showed a mean frequency score of 2.94 and an intensity score of 2.14 (SD = 0.80).⁶⁵  Even though both studies used the adult physician scale (MDS-R) by Hamric,⁶⁶ Abbasi et al⁶⁷ modified the scale to reflect the Iranian situation, and also included the scale for pediatricians.  Furthermore, the disparity in the levels of moral distress in the two studies may be due to the difference in samples: residents versus physicians.  Residents tend to grapple with the medical power hierarchy as they practice medicine under a supervisor or attending physician, which might exacerbate their moral distress level^68,69 whereas, physicians have some level of independence and autonomy to refrain from what they feel is contrary to their moral compass.

It was anticipated that the ethical climate would moderate the predictive relationship between moral distress and the propensity for moral disengagement, but the data did not support it. No other study was found that examined the moderating role of the ethical climate in the predictive relationship between moral distress and the propensity for moral disengagement. The lack of moderation could be because of the unique and insular culture of residency that may shield it from the broader hospital’s ethical climate. Residency programs are insulated by their unwritten codes and ethos that sometimes preclude the wider organizational climate from permeating through.70 Residents may see the hospital or organization as a temporary place of training before moving on to find employment elsewhere, which may affect the way they interact with the organizational culture and conform with the ethical norms.

The strong hierarchical medical structure conditions residents to accept their position at the bottom of the power structure such that they are expected to accept ‘orders’ and not to challenge authority. Moreover, residents may believe that their career progression concerning reference letters is dependent on attending physicians, consultants, and faculty and are willing to endure forms of humiliation and dehumanization for fear of angering superiors.⁷¹ As such, they devise ways of surviving within the bubble of a residency culture that is insulated from the wider organizational ethical climate.^72,73,74  This explanation is compatible with a previous study⁷⁵ that identified the lower status of residents in the hierarchy of medical care team as a source of moral distress, especially when there is a disagreement between those residents and the senior staff.

Residency is a crucial time to nurture the next generation of physicians such that factors that imperil physical, psychological, and emotional well-being ought to be addressed. Moral distress should be addressed in residency before it leads to moral disengagement. If this is not addressed, a self-perpetuating cycle emerges, because “morally blunted residents who become morally blunted attending physicians may disproportionately engage in negative mentoring, thereby exacerbating and perpetuating this problem.”^76(p396) As such, it is necessary to break the cycle of indifference towards moral distress in residency programs because unaddressed moral distress not only affects the residents and the care they provide to patients but also preserves the culture of inaction and denial of a pervasive issue.  There is also a need to change the culture of stoicism that is endemic in the medical community that perceives any admission of feelings of distress during residency as a sign of weakness. 

Recommendations

The findings of this study have implications for the leadership of health care institutions that have residency programs.  Evidence shows that the frequency and intensity of moral distress predispose medical residents to moral disengagement.  The challenge facing the administration and leadership of health care institutions, especially residency directors, is how to put mechanisms in place to address moral distress whenever it occurs.  This is necessary to break the cycle of indifference that has dogged the issue of moral distress in residency programs. 

There is a need for a change in the culture of stoicism that is endemic in the medical community.  There is an unchallenged belief that medicine is a tough profession and only tough people survive the rigors of medical training.  Consequently, an admission of any feelings of distress during medical training is perceived as a sign of weakness.  Leaders, hospital administrators, and residency directors should openly recognize and discuss the reality of moral distress in residency, as is evident in medical literature and the findings of this study.  It should become a priority for leaders in national meetings and conferences.  Such moves would elevate the awareness of this phenomenon as a problem in need of solutions.  It would be difficult to initiate and sustain a cultural change in the medical community about moral distress without the visionary direction and commitment of those in leadership.  Leaders do not need to force any structural readjustments to effect this change.  A good start would be for leaders to recognize and prioritize the threat that moral distress poses to the emotional and physical wellbeing of residents and patients, just as many in the medical community are currently grappling with the issue of physician burnout.

In line with the need to counteract the medical culture of stoicism, the Association of American Medical Colleges (AAMC) should require a more robust ethics curricula to be integrated into training and clinical rotations of third- and fourth-year medical students.  Emphasis should be placed on situations that negatively impact the physical and emotional health of prospective physicians, such as moral distress, burnout, and stress.  Already, the Accreditation Council for Graduate Medical Education (ACGME), which is responsible for the accreditation of residency programs, requires hospitals and medical centers to have ethics curricula and a mechanism in place for residents to freely discuss ethically and emotionally challenging experiences they encounter during residency. But some residency programs do not give this important matter the seriousness it deserves. Given the importance of the emotional health of residents to themselves and their patients, there is a need to intensify and recalibrate this effort at the national level.

The 1992 mandate of the Joint Commission on the Accreditation of Healthcare Organizations (JCAHO) requires all JCAHO-approved hospitals put in place some mechanism for discerning and addressing ethical concerns,77 in line with this mandate, every ethics committee should create a subcommittee or team tasked with the responsibility of helping residents discuss their experiences of moral distress in a safe and supportive manner.  The team should be interdisciplinary and must include at least one subject matter expert on moral distress.  Such a team would make the residents comfortable to freely discuss their feelings in a smaller setting as opposed to a full ethics committee, which residents might find overwhelming or intimidating.  Some medical centers have instituted moral distress consult service alongside clinical ethics consult service to give moral distress the undivided attention it deserves.78

There may be situations in some hospitals where the ethics committee is either non-existent or dysfunctional.  A viable alternative would be to create a Confidential Support Group (CSG) that offers residents a confidential forum to discuss sensitive and morally distressing experiences that they may not feel comfortable discussing with their immediate supervisors or human resources. The hospital’s administration, residency faculty, and ACGME would appoint members of this group based on knowledge and expertise in addressing residents’ moral distress and wellness. The CSG should be neutral so that residents can feel free to contact any member of the group at any time to discuss issues. The issues discussed would be kept strictly confidential within the support group and members of the group would help provide reasonable services to assist the residents in their distress.

Furthermore, hospitals should institute ethics programs such as grand rounds, didactic sessions, and weekly ethics teaching rounds with emphasis on the emotional health of residents.  Residency directors, faculty, and attending physicians who mentor residents should be encouraged to attend these programs.  Including mentors in these programs is essential to reorient them on the reality and impact of moral distress. As Berger noted, “unfortunately, most physician mentors had poor support for their own moral distress during their training experiences… and may not fully appreciate its impact or even validate its existence.”79(p396) All stakeholders in the medical power structure need to recognize and embrace this phenomenon and do all they can to help and support residents who experience distress.  Shared and collective responsibility of all stakeholders can create awareness on this issue plaguing residency programs.

Limitations

There are some limitations to this study. First, the relatively small sample of residents and the sampling from selected hospitals (excluding urban university medical centers) in the Midwestern and Eastern United Stated limit the generalizability of the finding to the entire population of residents in the United States. Second, the hospitals selected for the present study did not have a complete range of medical specialties, pediatrics and surgery were excluded which would also affect generalizability. Finally, the study relied on the memory and recollection of residents for morally distressing experiences which may be prone to bias. Future studies should explore the correlation between the variables in a longitudinal setting. Also, further investigation should be done with other healthcare professionals. Finally, the study should be replicated with residents in the South and West Coast of the United States.

Conclusions

This study explored the concept of moral distress, an everyday phenomenon in residency, and found it to predictive of moral disengagement. The study provides an empirical ground for the connection between moral distress and moral disengagement but does not serve as definitive social science support for the conceptual and psychological assumption.  It shows that a correlation exists between the two phenomena using social science research methods.  Correlations are different from causations.  Correlations can make predictions but do not imply causation. The study raises the awareness that moral distress and moral disengagement are phenomena that negatively impact the physical and emotional health of residents and ought to be addressed to enhance the well-being of residents.