MORAL DISTRESS in the health care setting, a situation in which one believes he or she knows the appropriate ethical action to take but is unable to take that action (Jameton, 1984), has received substantial attention in the past decade. This phenomenon is increasingly recognized as a serious problem among nurses, physicians, and other health care providers because of its apparent link to burnout and intention to leave (Meltzer & Huckabay, 2004). Some institutions are now devising education programs and committees for addressing moral distress with their personnel (Rogers, Babgi, & Gomez, 2008). However, the experiences of moral distress among advanced practice nurses, specifically nurse practitioners (NPs), are understudied in comparison with staff nurses and physicians. Nurse practitioners are distinctive in that they have previous experience as registered nurses (RNs) and additional education to diagnose and treat patients. Without an understanding of moral distress among advanced practice nurses, including NPs, health care organizations' efforts to address this phenomenon may not be as effective as hoped. The purpose of this study was to measure moral distress among a population of NPs, specifically those in the emergency department (ED), and to examine relationships between moral distress and intention to leave as well as level of practice independence.

REVIEW OF THE LITERATURE

Moral Distress

Ethical dilemmas are encountered in all aspects of health care. An ethical or moral dilemma is a situation in which there are at least two morally justifiable solutions to the problem and not one of these solutions is wholly adequate (Beauchamp & Childress, 2009). Moral distress is different from an ethical dilemma in an important way because it involves a situation in which there appears to be a clear, ethically justifiable solution, but resolution happens to be impossible to implement (Jameton, 1984, 1993).

Moral distress has been studied in nurses, particularly critical care nurses, but fewer studies describe this phenomenon among NPs and physicians. Moral distress scores are consistently higher among nurses than among physicians (Hamric & Blackhall, 2007; Hamric, Borchers, & Epstein, 2012; Knifed, Goyal, & Bernstein, 2010), but NP moral distress levels have not been measured. In addition, some studies suggest that nurses' moral distress levels rise with increasing years in a current position (Elpern, Covert, & Kleinpell, 2005; Hamric et al., 2012). However, this crescendo effect has not been seen among physicians (Knifed et al., 2010). Interestingly, the root causes of moral distress among critical care physicians and nurses appear to be similar including following families' wishes in continuing life support when it is not in patients' best interest and initiating lifesaving measures that they feel only prolong death (Hamric et al., 2012). Nurse practitioners are unique in their health care role. They possess qualities of both nursing and medical practices in that they have experience as RNs and education in diagnosis and treatment. Moral distress of NPs has been previously characterized through qualitative studies (Godfrey & Smith, 2002; Laabs, 2007; Viens, 1995) and through the use of instruments designed to measure ethics stress (Ulrich et al., 2006; Ulrich & Soeken, 2005; Ulrich, Soeken, & Miller, 2003) ethical dilemmas, and ethical knowledge (Laabs, 2005, 2007, 2012). Ulrich and her colleagues used an ethical conflict instrument to examine ethical dilemmas (Ulrich et al., 2006; Ulrich & Soeken, 2005; Ulrich, Soeken, & Miller, 2003), and Laabs used both qualitative methods and a questionnaire specifically developed to measure ethical stressors and knowledge of the NPs she researched (Laabs, 2005, 2007, 2012). In summary, these studies suggest that NPs experience moral distress when they are constrained by insurance companies, clinic policies, lack of patient compliance, and pressure to see more patients (Godfrey & Smith, 2002; Laabs, 2005, 2007; Viens, 1995). This study is the first to report moral distress among NPs using an instrument measuring moral distress.

Level of Practice Independence

Level of practice independence is a cornerstone of advanced practice nursing, and NPs are increasingly becoming part of the solution of keeping health care accessible to Americans (Kleinpell, Hudspeth, Scordo, & Magdic, 2012). For NPs, level of practice independence includes collaborative practice and authority to treat patients with a physician, using advanced practice skills and knowledge, and self-directed medical judgments (Cajulis & Fitzpatrick, 2007; Dempster, 1990; Maylone, Ranieri, Griffin, McNulty, & Fitzpatrick, 2011; Ulrich & Soeken, 2005). Several research studies have analyzed level of practice independence of NPs, also referred to as "autonomy" in the literature (Bahadori & Fitzpatrick, 2009; Cajulis & Fitzpatrick, 2007; Dempster, 1990; Ulrich & Soeken, 2005). Practice independence is different from scope of practice because it can be a personal perspective of one's clinical practice, whereas scope of practice is dictated by the state's nurse practice act (National Governors Association, 2012).

Emergency department NPs are a unique group within advanced practice nursing as ED NPs practice primary care medicine in addition to caring for critically ill patients. Serious gaps in current knowledge are the extent to which practice independence, moral distress, and intent to leave exist and are associated among ED NPs.

Intent to Leave

Turnover of medical staff is a problem experienced by health care systems all over the world (Aiken et al., 2012). Both moral distress and level of practice independence have been cited as reasons nurses and nurse practitioners leave their positions and the profession (De Milt, Fitzpatrick, & McNulty, 2011; Hamric et al., 2012). Fitzpatrick, Campo, Graham, and Lavandero (2010) found 41% of critical care nurses and NPs surveyed stated they intended to leave their current position due to lack of empowerment. De Milt, Fitzpatrick, and McNulty (2010) reported 27% of NPs surveyed at a national conference indicated their intent to leave their current position and 5% intended to leave the nursing profession for practice independence concerns. Common threads for intention to leave among nurses and NPs include feelings of powerlessness, lack of independence, and moral distress regarding their occupation satisfaction (De Milt et al., 2010; Fitzpatrick et al., 2010; Goldman & Tabak, 2010; Laabs, 2005, 2007). Hamric and Blackhall (2007) reported 45% of the nurses surveyed at one of the study hospitals left a position or considered leaving a position because of moral distress. In a more recent study, 49% of the nurses surveyed left or considered leaving a position related to their experience of moral distress (Hamric et al., 2012). Thus, in this era of need to retain and recruit more health care workers to meet the needs of our increasing national population, it is important to investigate reasons for dissatisfaction, and concerns of moral distress related to level of practice independence experienced by NPs and their associated impact on the intention to leave current positions or the profession.

SPECIFIC AIMS

The specific aims of this research study were to investigate moral distress among ED NPs and to examine relationships between moral distress, level of practice independence, and intent to leave their position.

METHODS

Design

This study applied a cross-sectional correlational design using survey methods.

Sample and Setting

Emergency department NPs were identified using convenience sampling from the Emergency Nurses Association (ENA) master mailing list by credential and position description (A. Daleo, ENA research department, personal communication, February 2013). Of the 34,000 ENA members, approximately 800 are NPs (S. Russell, Infocus Marketing representative, personal communication, February 2013). Sample inclusion criteria consisted of NPs currently working or have worked in an ED setting or having worked in a civilian ED or a military installation ED.

Study Power

Power calculations for t tests, using the statistical software nQuery, were based on results from a recent study of moral distress and intent to leave in intensive care unit RNs and physicians (Hamric et al., 2012). Because little is known about moral distress in ED NPs, power calculations for this study were run twice, using the RN results and then the MD results from Hamric et al.'s (2012) study. On the basis of these findings, a sample size of 200 would have more than 99% power in a t test at a significance level of 0.05 to detect a difference in the Moral Distress Scale-Revised (MDS-R) means between ED NPs who had considered leaving or who had left a position due to moral distress and ED NPs who had never considered leaving. For the correlation between moral distress and practice independence, a sample of 200 would have 81% power to detect a medium-small correlation of 0.20. Preparing for the logistic regression, Harrell (2001) suggests no more than nine or 10 independent variables in the model determined by the number of cases in the regression. For a logistic regression model of leaving one's job on moral distress and other independent variables, 108 in this study considered leaving or left and 96 never considered leaving.

Procedures

Approval for this study was granted from the University of Virginia Investigational Review Board (SBSIRB # 2012-0354-00) and the Institute of Emergency Nursing Research within the ENA. Study packets including measures, a postage-paid return envelope, and an introductory letter were mailed to 788 ED NPs identified from the ENA mailing list. Reminder postcards were sent at 2 and 3 weeks after the first mailing. Data from returned surveys were entered into the statistical software package IBM SPSS 21.0 for analysis.

Instruments

Moral Distress Scale-Revised

The MDS-R has been successfully tested with nurses and physicians (Hamric et al., 2012). This 21-item Likert scale survey assesses both the frequency (0 = never; 4 = very frequently) and level of disturbance (0 = not disturbing; 4 = very disturbing) of common morally distressing events. Total MDS-R scores were obtained by multiplying the frequency and disturbance scores for each item and then summing those products (range = 0-336). Higher scores indicate higher levels of moral distress. When tested among nurses and physicians in adult and pediatric intensive care settings, the MDS-R was found to have a Cronbach's [alpha] of 0.89 for nurses, 0.67 for physicians, and 0.88 overall (Hamric et al., 2012).

Intent to Leave

The MDS-R includes two additional categorical items asking respondents about their current and past intent to leave their position due to moral distress. These items are separate from the MDS-R score. Yes and no questions of this type are considered reliable as general perspective questions (Patrician, 2004).

Dempster Practice Behavior Scale

The Dempster Practice Behavior Scale (DPBS) is a scale to determine the participant's level of practice independence behavior (Bahadori & Fitzpatrick, 2009; Cajulis & Fitzpatrick, 2007; De Milt et al., 2010; Dempster, 1990, 2011). It has 30 items and uses a 5-point Likert scale measuring four different aspects of level of practice independence of NPs: readiness, empowerment, actualization, and valuation. Item responses range from 1 (not at all true) to 5 (extremely true), and total scores may range from 30 to 150. A higher score on the scale reflects a higher level of practice independence. A Cronbach's [alpha] of 0.95 was reported testing the sample of 569 practicing RNs (Dempster, 1990).

RESULTS

Settings and Sample

Of the 788 questionnaires sent, 246 were returned (31% response rate). Ten respondents were excluded from the sample because of failure to complete the questionnaires and another 29 cases were excluded for missed item responses. Twenty-six subjects had more than 10% missing data from the MDS-R, and another three subjects had more than three missing responses from the DPSB, leaving a final sample of 207. Descriptive statistics and inferential statistics were performed using the final sample (Pallant, 2010). As shown in Table 1, the sample consisted of primarily White women older than 45 years with 10 years or less of ED NP experience.

Table 1. Demographic data (

Moral Distress

MDS-R scores were calculated using the guidelines provided by A. Hamric (personal communication, March 2011). Questionnaires with more than 10% missing data were omitted from analysis. If a respondent had one or more missing values, but less than 10% in all, a score was computed from the responses present and then standardized to the 0-336 scoring range. MDS-R scores ranged from 0 to 224, with a mean of 74.4 (SD = 39.6). Cronbach's [alpha] coefficient was 0.85. A significant difference (p = 0.006) was found between female and male respondent mean MDS-R scores (77.7 vs. 59.2). Items from the MDS-R ranked in descending order to identify the most common causes of moral distress among ED NPs are shown in Table 2.

Table 2. Top six reported item scores in the MDS-R, for ED NPs (

Level of Practice Independence

DPBS scores were computed by adding the item responses together (J. Dempster, personal communication, March 2011). Subjects with more than three missing item responses out of 30 were dropped from the final analysis. If a respondent had one to three missing values, a score was computed from the responses present and then standardized to the 30-150 scoring range. A total of 207 DPBS scores were analyzed. The five negative statement questions within the DPBS were reverse coded when entered into SPSS. DPBS scores ranged from 81 to 149 (M = 127.6; SD = 12.1). Cronbach's [alpha] coefficient was 0.84. There was no significant difference in mean scores between female and male respondents. The highest median DPBS item scores with the smallest interquartile ranges are shown in Table 3.

Table 3. Top four item scores reported in the DPBS, for ED NPs (

MDS-R and DPBS Relationship

Pearson's correlation coefficient testing used to estimate the strength of the relationship between the MDS-R and the DPBS showed no significant correlation between the two variables, r = -0.071, n = 207, p = 0.312. Multiple linear regression was used to determine whether the measure DPBS was predictive of scores on the MDS-R after controlling for age, gender, facility type, NP type, years of NP practice, and years of ED practice. The total variance explained by the model was 8.9% (adjusted R² = 0.05, p = 0.028). Two control measures neared statistical significance: gender (p = 0.065) and years of NP practice (p = 0.08). The DPBS measure was not found to be a significant predictor of MDS-R scores (p = 0.854).

Intent to Leave

Two questions at the end of the MDS-R instrument asked respondents: (1) Have they ever considered leaving their current position due to moral distress? and (2) Are they considering leaving their position now due to moral distress? The results found nearly half (47%) of respondents affirmed that they had never considered leaving their current position, 27% reported they had considered leaving, and 25% stated they did leave a position. Noticeable differences were found between women and men with 63% of male respondents compared with 43% of females reporting that they had never considered leaving a position, although the differences were not significant (p = 0.08) (see Table 4 for frequencies and proportions of intent to leave). When comparing current consideration of leaving, no significant difference was found between males and females (p = 0.44). Although gender differences in nurses regarding intent to leave because of moral distress have not been reported in past studies, 19% of the study population was male ED NPs, a higher percentage of male nursing professionals than other studies (Hamric et al., 2012). Mean MDS-R scores varied significantly for different lifetime intentions to leave one's position, with higher mean MDS-R scores found for those who had considered leaving or had left a position due to moral distress, and lower mean scores for those who had never considered leaving a position due to moral distress (p < 0.001, Welch's test of equality of means). Similar results were obtained when male and female respondents were separately analyzed (see Table 4).

Table 4. Intention to leave one's position, with mean MDS-R scores, for ED NPs (

Logistic regression was performed to identify predictors of intent to leave (see Table 5). The dependent variable, intent to leave, was coded into three levels: 1 = Yes, left, or 2 = Yes, considered leaving, and 0 = Never considered leaving due to moral distress. Levels 1 and 2 were recoded into one category, yes, left or yes, considered leaving, and no, never considered leaving was coded into another category, creating a dichotomous variable. Independent variables included age, gender, years of NP practice, years of ED practice, MDS-R score, DPBS score, type of NP specialty (FNP [family nurse practitioner], ACNP [acute care nurse practitioner]; reference category: CNS [clinical nurse specialist]/other), and facility size (small, large; reference category: medium). The full model was found to be statistically significant, [chi]² (10), p < 0.001, indicating the model was able to distinguish between ED NPs who reported they had considered leaving or had left and those who had never considered leaving. As a whole, the model explained between 24.7% (Cox and Snell R²) and 32.9% (Nagelkerke R²) of the variance in the intention to leave and correctly identified 69.8% of the cases. However, only one of the independent variables made a statistically significant contribution to the model: MDS-R (p < 0.001). With an odds ratio of 1.034 for the MDS-R score, the model estimated that an increase of 1 point in the MDS-R score corresponded to a 1.034 increased odds of having left or having considered leaving, when controlling for the other factors in the model. An increase of 10 points in the MDS-R corresponded to a 1.397 increased odds of having left or having considered leaving.

Table 5. Logistic regression of intention to leave one's position due to moral distress, for ED NPs (

Outlier Analysis

Analyzing the study sample, separate analyses were calculated using the 29 removed cases (12%). Comparisons between this subset and the final sample found no significant differences in gender or ethnicity (female, 80%; male, 19%; non-Hispanic, 97%; Hispanic, 2.5%); however, the subset reported significantly fewer years of NP practice (p = 0.02) than the study sample. A chi-square test for independence indicated a significant association between having no score on either the MDS-R or the DPBS (too few items were answered, thus no score) with intent to leave, [chi]² (n = 29), p = 0.007, with 24% of the removed cases considering leaving now, compared with 32% of the final sample. No other significant differences were found between the samples when comparing NP type, facility type, or years in ED.

DISCUSSION

Previous studies have shown that nurses experience fairly high levels of moral distress, with physicians reporting moral distress but to a lesser extent than nurses (Corley, Elswick, Gorman & Clor, 2001; Elpern et al., 2005; Hamric & Blackhall, 2007; Hamric et al., 2012). When compared with previous investigations, the present study found lower moral distress scores among ED NPs than nurses, slightly higher moral distress scores than physicians (Hamric & Blackhall, 2007; Hamric et al., 2012). The two most morally distressing situations for ED NPs were witnessing poor patient care resulting from poor staff communication and working with less competent nurses and colleagues. These differ from the most morally distressing situations for nurses reported in another study, which were family wishes to continue futile life support and initiating extensive lifesaving actions that prolong death (Hamric et al., 2012). The root causes of moral distress found in this study are different from other studies concerning moral distress in NPs, including situations that result in psychological stress such as insurance constraints, productivity demands, and patient treatment noncompliance, which are not consistent with the definition of moral distress.

The results from this study also found that in general ED NPs report feeling empowered in their practice evidenced by a high mean DPBS score (M = 127.6) and mild to moderate moral distress scores (M = 74.4). Although the literature suggests that practice independence could result in moral distress, this was not found to be a cause of moral distress in the current study. This difference may be a result of changes nationwide in nurse practice acts, which are allowing NPs to practice more independently, or with looser ties to collaborative physicians where moral dilemmas encountered in practice.

Although ED NPs in this study obtained high scores on the DPBS related to level of practice independence, they also obtained moderate scores on the MDS-R. Moderate moral distress may be unique to ED NPs resulting from psychological stress of emergency medicine practice and the stress of making independent clinical decisions. Although the definition of moral distress was included in the instrument instructions, it is possible that ED NPs considered both their experiences with moral stress and work stress when completing the MDS-R. Previous investigators, who have examined moral distress among NPs, have found psychological stressors including making clinical decisions in practice are often cited as moral distressing experiences (Laabs, 2005, 2007; Ulrich et al., 2006).

Higher levels of moral distress among ED NPs in the present study were found to be associated with considering or having left their current practice, consistent with previous research with moral distress among nurses and physicians (Hamric & Blackhall, 2007; Hamric et al., 2012). This suggests a need to address situations likely to result in moral distress for ED NPs to improve job satisfaction and retention. Because ED NPs reported somewhat different morally distressing situations than those previously reported among nurses and physicians, unit-specific analysis may be helpful in determining sources of moral distress to develop strategies to address concerns and improve retention.

Limitations

There were several limitations within this study. First, study participants consisted only of ENA members; therefore, the findings may not be representative of feelings of moral distress and level of practice independence for most ED NPs. Similarly, the actual survey response rate was only 31%; therefore, the findings may not represent those of the majority of ED NPs who are members of ENA. A higher response rate may have resulted if an incentive for questionnaire completion was offered. The MDS-R instrument, although scoring valid and reliable in assessing moral distress, is very difficult to answer. Twenty-six cases were lost to more than 10% missing data and it was noted that respondents who failed to complete all of the questionnaire items often did not answer the second half of questions regarding frequency. More work on making this instrument easier to answer needs to be considered. Perhaps, transfer to a web-based format that requires completing both the frequency and level of disturbance components of a question before advancing might decrease missing data occurrences. Another limitation is that the sample consisted primarily of White women over the age of 40 years. Finally, cross-sectional, correlational studies cannot determine causal relationships between variables.

CONCLUSIONS

This study found approximately 25% of participants stated they left a position because of moral distress. This finding is significant for considering strategies to retain competent ED NPs, given current ED staffing gaps and census trends. Although no significant relationship was found between practice independence and participant moral distress scores, this study did find that situations associated with moral distress are different for ED NPs when compared with physicians or nurses. Differences may result from the unique role of the ED NP whose scope of practice bridges nursing with medicine. More studies comparing nurses, physicians, and NPs and their responses to moral distress in the clinical setting would be beneficial in understanding the implication of moral distress to each health care provider.

IMPLICATIONS FOR FUTURE RESEARCH

The findings from this study lend support to further research in identifying causes of moral distress and the intention to leave in ED NPs especially given current ED workforce needs and efforts to retain competent providers. This study found the MDS-R shows promise as an indicator of moral distress of health care providers. Further research should attempt to recruit a more demographically diverse group, to determine whether there are other situations that are morally distressing for ED NPs resulting in intention to leave their jobs or the profession. A better understanding of moral distress among emergency care providers may also help improve the ED work environment and reduce risks of burnout and compassion fatigue among all ED providers.

REFERENCES