We delineated the process of retrieval and selection of studies in
. We captured 863 records at initial search stage in PubMed (n = 438), Embase (n = 328), and CNKI (n = 97) and added additional 3 records from application studies, and 846 records remained after removing duplicates using the literature management software EndNote X7. We further excluded 786 records after carefully checking the title and abstract, and thus, 60 items were included to be checked in full-text. Finally, 16 eligible studies
12
,
24
-
38
including 10 original and 6 revised instruments were included for the final analysis after excluding 44 studies due to several reasons including unrelated to topic (n = 29), validation of instruments in different populations (n = 3), cross-cultural adaptation and validation of instruments (n = 7), validation of modified instrument (n = 1), and development and validation of other instruments such as moral distress map (n = 4).
Qualitative Summaries of All Instruments
Moral distress scale and revisions
Moral distress scale (MDS) is the first instrument which was initially developed to measure the frequency and intensity of MD in intensive care nurses in 2001 by Corley and colleagues.
Original MDS was designed to have 32 items with 3 factors structure on a 7 points Likert scale. The psychometric properties of MDS were examined among 214 intensive care nurses from American hospitals, and test result indicated a Cronbach’s alpha of .98 for individual responsibility, .82 for not in patient’s best interest, and .84 for deception, respectively. After validated, MDS has been extensively used to detect the frequency and intensity of MD in intensive care nurses.
3
,
39
For example, Rice and colleagues performed a study to determine the prevalence and contributing factors of MD in 260 medical and surgical nurses, and found that MD is common and can be elicited from different types of situations encountered in the work environment.
Considering the limitations of the original version, Corley and colleagues reported a modified version in 2005.
Additional 6 items were added in the modified version, and a content validity index of 1.0 was obtained for 38 items. After employed this modified scale in 106 registered nurses, a Cronbach’s alpha of .98 was produced. To date, this modified scale has also been extensively implemented in empirical studies. For example, Elpern et al performed a study to assess the level of nurses’ MD working in the medical intensive care unit, and found a high level of MD.
Fruet et al adapted MDS and then assess the applicability of this version in the context of nursing in hemato-oncology services in Brazil in 2017.
Finally, 26 questions were included in this modified version with 3 factors structure including lack of competence in the team, denial of the nursing role as the patient’s advocate, and disrespect for the patient’s autonomy. Psychometric test suggested a Cronbach’s alpha of .98. To date, however, this revised version of MDS has not been validated and employed in further studies.
In 2017, Badolamenti et al modified MDS and developed a brief instrument (MDS-11) to determine MD.
Factor analysis of 347 nurses generated a 2-factor structure with 11 items. Psychometric test suggested a high reliability with Cronbach’s alpha of .823 for the futility dimension and .756 for the potential damage dimension, respectively. However, this version was not presently validated and utilized in further studies.
In 2010, for the purpose of proposed a reliable and valid MD assessment tool which can be used in multiple health care settings and with multiple disciplines, Hamric and colleagues adapted MDS to cover more distressing situations and shortened to include 21 items, which were evaluated using a 5-point Likert scale.
Authors utilized this scale in 169 registered nurses and 37 physicians to demonstrate its reliability and construct validity. As the first generic instrument of MD for healthcare professionals, moral distress scale - revised (MDS-R) has 3 parallel versions for nurses, physicians, and other health care providers, and has also been extensively utilized by other subsequent survey or comparative studies,
42
-
44
as well as been extensively validated in other countries including Turkey,
Italy,
32
,
46
Australia,
China,
Korea,
Brazil,
Sweden,
and Iran.
52
,
53
To date, the MDS-R has 6 versions, designed for different healthcare provider populations: adult-nurse, adult-physician, adult-other, pediatric-nurse, pediatric-physician, and pediatric-other.
In 2015, Shoorideh expanded the MDS-R to specifically use in intensive care units.
This scale was developed with the content analysis approach based on the published theoretical frameworks including Jameton’s conceptualization of MD, House and Rizzo’s role conflict theory, and Rokeach’s value theory. The initial scale was structured with 3-factor including 30 items, which were answered at 0 to 4 Likert scale format, with 0 indicating none and 4 indicating great extent. The content validity index was .98, .95, and .96 for relevance, clarity, and simplicity, respectively. Cronbach alpha of .96 also indicated a highl reliability. Unfortunately, this instrument has not previously been extensively employed in subsequent empirical studies.
Although the aim of the MDS-R and corresponding expanded version are to detect MD in healthcare professionals or specific populations, some root causes were not covered by them.
54
-
56
As a result, Epstein et al constructed and validated an updated instrument, measure of moral distress for healthcare professionals (MMD-PH), to detect the MD in healthcare professionals.
Application in 653 professionals demonstrate its reliability and validity with a Cronbach’s alpha of .93. These authors therefore recommend to replace the MDS-R with the MMD-HP as a generic instrument of MD in healthcare professionals. However, further studies should be designed to extensively validate MMD-PH as a preferred tool of measuring MD in healthcare professionals because several limitations existed in the methodology study.
Certainly, the validation in different cultural settings should also be considered.
Moral distress measures for specific populations
The parallel version MDS primarily focused on pediatric setting has been used in clinical practice, and the psychometric properties of revised version of the original MDS-PV has also been validated in Iran
and Sweden.
Moreover, Professor Corley also developed moral distress scale neonatal-pediatric version (MDS-NPV), which which has also been employed in neonatal intensive care unit nurses
although the Corley’s original version was not published. In 2014, Lazzarin et al revised and employed MDS for nurses in pediatric context (MDS-PV),
which was initially developed by Corley et al based on MDS. In their empirical study, Lazzarin et al shorten items from initial 38 to 33 which was answered with 0 to 6 Likert scale, and produced a highly reliability with a Cronbach’s alpha of .959. Meanwhile, Professor Corley also confirmed the equivalence between adapted version and the original version. The psychometric properties of MDS-NPV in nurses working other pediatric settings rather than intensive care should be further investigated.
In 2017, Muccio et al initially created a MDS for correctional nurses (MDS-CN) based on MDS-R, which was subsequently employed and validated by Lazzari and colleagues in 2019.
This scale was structured with unidimensiona with 20 items, which should be indicated using the numerical number from 0 to 5 format. Content validity index, Cronbach’s alpha, and test-retest reliability was 99.0, .91, and .99 in original version respectively, and the subsequent study of 238 correctional nurses confirmed its reliability with a Cronbach’s alpha of .89.
Eizenberg et al found that systematic examination for measurement equivalence of all applicable instruments has not been reported, and thus they developed a culture-sensitive MD questionnaire for nurses working in across working settings in 2008.
Authors firstly elicit the culture specific themes based on focus group interview, and then tested the psychometric properties of questionnaire in 179 nurses from a variety of work settings. The psychometric analysis suggested a Cronbach’s alpha of more than .79, with a Cronbach’s alpha of .851, .791, and .804 for the first, second, and the third factor, respectively. Although as the first generic instrument of measuring MD in clinical nurses, no published study utilized it to measure MD. Therefore, further studies are needed to evaluate the measure in different cultural settings.
Because MDS was not administered in psychiatric setting, Ohnishi et al developed the MDS for psychiatric nurses (MDS-P) through combining the 24 items derived from the MDS with and then employed it in Japanese psychiatric nurses in 2010.
MDS-P consists of 3 factors structure with 15 items in a 7-Likert scale. In this scale, the frequency and intensity were all marked from 0 to 6, and the score was positively associated with the severe levels of MD. A total of 369 psychiatric nurses from 6 Japanese hospitals were enrolled to respond the scale, and subsequent psychometric analysis suggested a Cronbach’s alpha of .90 for the whole scale, with .85, .82, and .79 for factor 1, 2, and 3 structure, respectively. As an instrument specific to psychiatric nurses, further studies should be designed to test its psychometric properties across cultural settings.
Moral Distress in Dementia Care Survey (MDDCS) tool was created by Awosoga and colleagues in 2018, with the purpose of specifically examine MD among nursing staff caring for the unique population with dementia.
The team conducted an exploratory sequential mixed method to generate item pool, and then enrolled 389 sample of consisting of registered nurses, licensed practical nurses, and healthcare aides to test the psychometric properties. After completing factor analysis, authors got a 3-factor structure covering 55 items on a 5-Likert scale. The construct validity was demonstrated by the CFA method. Psychometric evaluation indicated a Cronbach’s alpha of .94, .92, .93, and .83 for the frequency of MD, severity, effects, and mitigating factors, respectively.
Other moral distress measures for nurses
Sporrong et al found that, in 2006, several instruments can be used to closely measure MD, however the scope of application of these instruments is limited.
Thus, they developed a 9-item MD questionnaire in 2006 to cover this gap,
which can be called as moral distress questionnaire for clinical nurse (MDQ-CN). And then, authors tested the validity and reliability of this questionnaire in 259 staff members working in 200 departments and 59 pharmacies, and obtained a Cronbach’s alpha of .78 for factor 1 and a Cronbach’s alpha of .62 for factor 2, respectively. Although this study enrolled numerous professionals with diverse working settings to validate the reliability and validity of the questionnaire, inadequate root causes limited the application value. Moreover, no further empirical study to determine the psychometric properties.
Brazilian scale of moral distress in nurses was initially developed by Ramos and colleagues in 2017 in order to specifically detect the frequency and intensity of MD.
The initial version included 57 questions but did not performed factor analysis. Since then, original authors performed a methodology study to test the psychometric properties of this scale.
Factor analysis shorten items from 57 to 49 and thus generated a 6-factor structure including (a) acknowledgment, power, and professional identity; (b) safe and qualified care; (c) defense of values and rights; (d) work conditions; (e) ethical infractions; and (f) work teams. Psychometric test indicated a high reliability with a Cronbach’s alpha of .980 for the whole scale. Moreover, these authors also investigated the application of this scale in the primary health care setting and also obtained a high reliability.
Moral distress thermometer (MDT) is a tool of rapidly measuring MD and tracking changes in MD over time, as well as quantitatively the time-specific level of MD, which was developed by Wocial and colleagues in 2012.
Authors analyzed the data from 529 nurses to demonstrate the convergent validity and concurrent validity of MDT, and then indicated that MDT has great potential as a screening tool for use in research, evaluating the effectiveness of interventions designed to decrease a nurse’s level of MD. To date, some empirical studies have employed this instrument to examine the level of MD in school nurses
62
,
63
and intensive care unit nurses.
64
,
65
However, the application in other clinical nurses or cultural settings has not yet been reported.
Other measures of moral distress and related topics
We also identified another 4 instruments including Moral distress risk scale (MDRS),
MD assessment questionnaire (MDAQ),
ethics stress scale (ESS),
and stress of conscience questionnaire (SCQ),
which were all also designed to measure the level of MD in nurses.
Moral distress risk scale (MDRS) was created by Schaefer et al in 2019 in order to offer a tool that could be useful beyond hospital care settings.
After identified 53 items, the team enrolled 268 nurses from hospitals and primary healthcare setting to perform factor analysis and psychometric evaluation. Finally, this scale was developed with 7 factors covering 30 items on a 4-Likert scale (1 indicates never and 4 represents always). Psychometric test indicated a good internal consistency of .913. Compared to previous instruments which were all built based on causes, MDRS was developed at the basis of risk factors which may reduce the negative impact of subjectivity of MD on the measurement result. Thus, considering this advantage, further studies should be performed to validate the psychometric properties of MDRS in other working and cultural settings.
The MDAQ, which was established in 2002 by Hanna, can measure MD from type, intensity, frequency, and duration of the experience. According to published literature,
this questionaire can be utilized across disciplines; however psychometric evaluation for it has not yet been performed.
In 2000, Raines and Tymchuk developed a novel and self-administered instrument, which was named as ESS, to measures the individual’s perception for MD.
In this scale, 56 questions were designed and the first 52 questions rated on a Likert-type scale from 1 (agree strongly) to 5 (disagree strongly). Questions 53 through 56 are designed to provide additional information to the researcher and are intended to be answered in various ways. After developed this scale, Raines employed it in emprical study and obtained a content validity of .89 (
P
< .05), a test-retest reliability of .82 (
P
< .005), and a Cronbach α of .87. Unfortunately, this scale can not be extensively validated or utilized in subsequent studies.
67
,
68
SCQ was developed by Glasberg in 2007 with the purpose of measuring stress arising from a disturbed conscience.
The questionnaire was structured with 2 parts covering 9 items. In part 1, the respondent was asked to indicate the stressful situation with a 6-Likert scale from 0 (indicating never) to 5 (indicating every day). In part 2, the respondent was required to indicate the level of bad conscience based on the degree of guilt for every situation in the part 1 on a 10 cm visual analogue scale (VAS) from 0 (indicating no, not at all) to 10 (indicating yes, gives me very bad conscience).
In this questionnaire, higher total scores represents higher the perceived stress level. Psychometric evaluation indicated a Cronbach’s alpha of .83, which also demonstrated in empirical study.
Moreover, SCQ was also revalidated by Ahlin and colleagues in 2012 to be valid for Swedish settings.