Moral distress is a type of moral conflict that is experienced when a practitioner recognizes the right course of action but is unable to act ethically because of factors beyond his or her control (Jameton, 1984). When a practitioner is unable to follow the ethical values and professional standards of his or her profession, moral distress may occur (Varcoe, Pauly, Webster, & Storch, 2012). Moral distress is an issue that is gaining attention among the health care professions (Hamric, Borchers, & Epstein, 2012).

Moral distress was first recognized and investigated in the profession of nursing; however, in recent years other health care professions have begun to explore this phenomenon (Austin, Rankel, Kagan, Bergum, & Lemermeyer, 2005; Carpenter, 2010; Penny, Ewing, Hamid, Shutt, & Walter, 2014; Schwenzer & Wang, 2006; Sporrong, Höglund, & Arnetz, 2006; Whitehead, Herbertson, Hamric, Epstein, & Fisher, 2015; Wiggleton et al., 2010). Moral conflicts can arise when there are disagreements between health care team members, between health care team members and health care organizations, or between health care team members and the wishes of the client. Examples of situations that might give rise to moral distress in occupational therapy include being expected to continue services even after the client has met all therapy goals to meet productivity standards or being directed to discontinue services because of concerns about third-party payment.

For some practitioners, these conflicts are just a part of their daily work; for others, the initial distress does not dissipate, but rather a reactive distress sets in that has lingering consequences (Jameton, 1993). Moral distress has been found to contribute to professional burnout, progressive insensitivity to ethical patient care, and staff turnover (Corley, 2002; Epstein & Hamric, 2009; Hamric & Blackhall, 2007; Hamric et al., 2012). If occupational therapy practitioners experience moral distress, as emerging evidence suggests they do (Penny et al., 2014; Slater & Brandt, 2009), then it is important to develop an instrument that will enable researchers to measure the effectiveness of strategies designed to mitigate the distress and prevent staff burnout.

The Moral Distress Scale (MDS) was created to measure the experience of moral distress in the profession of nursing (Corley, Elswick, Gorman, & Clor, 2001). Hospital-based nurses were asked to rate their experience of moral distress in 32 situations. Mean scores were calculated for each item indicating the level of moral distress the situation caused. Hamric et al. (2012)  revised the original MDS into a shorter instrument applicable to settings other than hospitals and to professions other than nursing, the Moral Distress Scale–Revised (MDS–R). The MDS–R is based on the theory that moral distress arises from three root causes: (1) clinical situations, (2) internal constraints, and (3) external constraints. In clinical situations, service-based circumstances generate moral distress, such as a clash of duties, a disregard for ethical or professional standards, and competing or overlapping roles. Internal constraints relate to attributes of the health care practitioner, including having low assertiveness or not feeling empowered to speak up in challenging situations. External constraints relate to factors specific to work in health care organizations, such as reimbursement systems, institutional policies, work norms, and power structures.

The MDS–R consists of items describing 21 situations and an additional two empty spaces where participants can write and rate situations not included that caused them moral distress. Several parallel versions of the scale were developed with modification in the wording of items for different contexts. Four experts rated the items for agreement with the three root causes of moral distress, and an interrater agreement of 88% was reported (Hamric et al., 2012).

Penny et al. (2014)  used a version of the MDS–R, the Moral Distress Scale–Revised–Other Health Provider Adult (MDS–R–OHPA), to survey occupational therapists in two practice settings to investigate whether they encountered moral distress. The researchers selected geriatric and physical disability practice settings because they share similar types of clients and reimbursement providers, which are different from other practice settings such as pediatrics (e.g., early intervention, school based) or behavioral health. Participants reported moderate levels of moral distress. One limitation identified was the wording of the items, which, although intended for other health care providers, are firmly rooted in nursing practice—for example, Item 9, “Assist a physician who in my opinion is providing incompetent care”; Item 11, “Witness medical students perform painful procedures on patients solely to increase their skill”; and Item 14, “Witness increasing doses of sedatives/opiates given to an unconscious patient that I believe could hasten the patient’s death.” Penny et al. recommended that “in order to accurately measure the levels of moral distress experienced by occupational therapists, there is a need to develop an instrument made up of scenarios familiar to occupational therapists” (p. 390). Therefore, the purpose of the current study was to describe the development of the Moral Distress Scale–Revised–Occupational Therapy–Adult Settings (MDS–R–OT[A]) and to describe how the content validity of this new instrument was established.

Furr and Bacharach (2014)  defined content validity as “the degree to which the content of a measure truly reflects the full domain of the construct for which it is being used, no more and no less” (p. 206). This definition captures three important elements: (1) that the measure is designed to assess an underlying theoretical construct, (2) that the measure contains only items that are relevant to the intended construct, and (3) that the measure includes a full range of items relevant to the construct. The presence of irrelevant items or lack of a full range of items reduces the validity of an instrument.

In the allied health literature, work by Lynn (1986)  and Davis (1992)  has provided the foundation many researchers use to establish content validity for new instruments. Lynn proposed a content validity index (CVI), which is the proportion of experts who give the item a rating of 3 or 4 on a 4-point Likert scale of relevance. The CVI of the full instrument is then the proportion of total items judged to have content validity (Lynn, 1986). Using standard error calculations, Lynn showed the impact of different levels of agreement on the CVI. From this work, the threshold of 80% agreement became the widely used criterion for an instrument to be said to possess content validity (Davis, 1992).

Although considerable agreement exists about how to calculate the item content validity index (I–CVI; i.e., the proportion of agreement of item relevance), the scale content validity index (S–CVI) can be calculated in two ways, and resulting differences typically are not discussed when content validity is reported (Polit & Beck, 2006). One way to calculate the S–CVI requires universal agreement among all raters; the other is to calculate average agreement among raters. Universal agreement is more applicable when there are few raters (5 or less), and average agreement is more applicable when there are many raters (≥10) and the opportunity for disagreement is greater.

Grove, Burns, and Gray (2013)  suggested that evidence for content validity comes from three sources: “the literature, representatives of the relevant populations, and content experts” (p. 394). Rubio, Berg-Weger, Tebb, Lee, and Rauch (2003)  described two types of experts: (1) content experts, who have published on the topic, and (2) lay experts, or “people for whom the topic is most salient” (p. 96). The use of lay experts ensures that the measure has relevance to the intended population and may lead to the inclusion of items not judged as important by content experts. The theoretical foundation of the MDS–R was demonstrated by Hamric et al. (2012); therefore, the relevance of the modified instrument to a new population of occupational therapists was most salient and the focus of this study.

This project occurred in two stages. The first stage was the creation of the MDS–R–OT[A], and the second stage was assessment of its content validity. Before starting the research process, we obtained institutional review board approval, and the lead author sought and was granted permission to make changes to the MDS–R–OHPA (A. B. Hamric, personal communication, July 9, 2013).

To create the MSD–R–OT[A], we made use of unsolicited written comments provided on questionnaires returned during the study by Penny et al. (2014) . Of the 224 returned questionnaires, 78 contained comments (34.8%). The comments fell into three broad categories: (1) general suggestions or commentary about the MDS–R–OHPA, (2) suggestions or commentary about the individual items, and (3) new or different situations in which respondents had experienced moral distress. The process we followed to make changes to the instructions and items on the MDS–R–OHPA is illustrated in Figure 1.

Changes made to the instructions were designed to improve clarity and thereby improve the reliability of the data generated. First, we changed the general description of moral distress to the definition first used by Jameton (1984)  and added a contemporary definition of moral distress. Second, to improve readability, we split the instructions into two paragraphs. Finally, to remove ambiguity and a mindset toward moral distress, we changed the wording of the instruction in the MDS–R–OHPA that directed respondents to rate their level of disturbance even if they had not experienced the situation; respondents were instructed to select 0 (never) as appropriate for both frequency and level of disturbance.

Before changes could be made to the items that make up the MDS–R–OHPA, the first task was to align each item on the MDS–R–OHPA with its root cause. This task was needed because of our desire to maintain the distribution of items related to each root cause and because the records of which item corresponded to which root cause were not available (A. B. Hamric, personal communication, September 21, 2014). Using the descriptions of the root causes published in Hamric et al. (2012), we independently categorized the 21 items on the MDS–R–OHPA into one of three root causes: (1) clinical situations (Items 2, 3, 4, 5, 6, 7, 9, 12, 13, 14, 17, 19, 20, and 21), (2) internal constraints (Items 8, 10, and 15), and (3) external constraints (Items 1, 11, 16, and 18).

The next step was to identify which items were to be replaced. Items were marked for replacement if the mean frequency of the item response was <0.54 (range = 1.88 to 0.19) or a minimum of four respondents indicated the item was not applicable to their clinical practice (Penny et al., 2014). In addition to Items 9, 11, and 14, listed earlier as being strongly related to nursing practice, the following items were marked for replacement:

• Item 3, “Follow the family’s wishes to continue life support even though I believe it is not in the best interest of the patient”

• Item 4, “Initiate extensive life-saving actions when I think they only prolong death”

• Item 7, “Continue to participate in care for a hopelessly ill person who is being sustained on a ventilator, when no one will make a decision to withdraw support”

• Item 12, “Participate in care that does not relieve the patient’s suffering because the physician fears that increasing the dose of pain medication will cause death”

• Item 13, “Follow the physician’s request not to discuss the patient’s prognosis with the patient or family.”

All 8 items marked for replacement had as their major root cause clinical situations except Item 11, which had as its major root cause external constraints.

Guided by principles of Grounded Theory (Corbin & Strauss, 2008), we identified 14 themes in the 78 written comments in Penny et al. (2014) ’s data. In order of number of comments in each theme, these themes were quality of care, management, insensitivity of staff, quality of life, insurance-driven therapy, incorrect documentation, productivity, insurance companies, Medicare abuse, issue addressed in another item that may need rephrasing, distress related to the patient’s family, legal obligations, shortage of staff, and Medicare guidelines. We selected the eight themes with the most frequent comments to guide creation of eight new items for the MDS–R–OT[A].

The theme with the most comments was related to concerns about the quality of care provided. Comments typical of this theme included “Knowing that poor-quality care caused injury/sores” and “Staff who don’t or can’t follow through on recommendations from therapy.” The new item created to address this theme (replacing Item 13) was, “Be told to ignore situations in which poor quality of care might contribute to the development of a preventable condition.”

The theme with the second most frequent comments related to concerns about directives from management and included statements such as “Productivity standards interfere with quality of care” and “Strongly being encouraged to see a patient for a certain length of treatment time.” The new item created to address this theme (replacing Item 3) was, “Be expected to follow unrealistic directives from administration, which diminishes quality of client care.”

Insensitivity of staff was the next theme identified. This theme addressed issues of prognosis, injury, or illness caused by the insensitivity of the staff providing the care—for example, “Witness unnecessary levels of patient/family distress after receiving a grim prognosis from a doctor who didn’t empathize” and “Watching a patient slowly slide out of his wheelchair and a nurse/assistant not paying attention or walking away.” The new item created to address this theme (replacing Item 7) was, “Work with health care team members who demonstrate insensitivity toward clients, families, or staff.”

The next theme was related to quality of life and addressed situations in which therapy services would not improve the quality of life of clients—for example, “Patients who are terminally ill being put through inpatient rehab when it is not likely to help or increase participation in life” and “Be unable to provide care I would consider to enhance comfort (bed, wheelchair positioning, splint care) for hospice patients because they are on ‘comfort care.’” The new item created to address this theme (replacing Item 14) was, “Witness an intervention or the withholding of an intervention that negatively impacts a client’s safety or quality of life.”

The theme of insurance-driven therapy was derived from comments that discussed pressure to provide therapy services regardless of the client’s need—for example, “Being expected to obtain as many billable units as possible per patient, regardless of patient needs” or “Requested to treat patients for an extended amount of time per session to meet higher Resource Utilization Group reimbursement levels.” The new item created to address this theme (replacing Item 9) was, “Be expected to obtain as many billable units as possible per client, regardless of their individual needs.”

The next theme related to incorrect documentation. These comments addressed situations in which respondents either witnessed incorrect documentation or were asked to document incorrectly—for example, “Be expected to document more treatment time with patients than is given, to document transport and documentation time as treatment time” and “Other therapists not treating for time billed.” The new item created to address this theme (replacing Item 4) was, “Witness situations in which team members incorrectly bill for services.”

The seventh theme related to productivity; comments suggested pressure to increase caseload or meet productivity standards—for example, “Strongly encouraged to see a patient for a certain length of treatment time despite the need for it” and “Productivity standards from administrators even with patients who are not able to tolerate therapy. Only worried about money.” The new item created to address this theme (replacing Item 12) was, “Be expected to treat and write documentation for more clients than time allows.”

The final theme was insurance companies; comments identified related to the limitations on services driven by the companies—for example, “Decrease treatment session due to Medicare cuts” and “Unable to provide care due to lack of insurance.” The new item created to address this theme (replacing Item 11) was, “Be unable to provide optimal therapy services to clients due to limited insurance coverage or insurance cutoffs.”

We modified the remaining items on the MDS–R–OHPA for clarity of wording and applicability to occupational therapy practice using terminology in the Occupational Therapy Practice Framework: Domain and Process (3rd ed.; American Occupational Therapy Association [AOTA], 2014). The only item that underwent no modification from the MDS–R–OHPA to the MDS–R–OT[A] was Item 1, “Provide less than optimal care due to pressures from administrators or insurers to reduce costs.” The final step was to analyze each item, new or modified, to ensure it corresponded to the major root cause identified in the initial analysis.

The second stage of this research project involved mailing the MDS–R–OT[A] and a demographic questionnaire to a random selection of 400 occupational therapists who were members of AOTA. Two hundred indicated geriatrics (N = 3,789), and 200 indicated physical disabilities (N = 6,012) as their primary area of special interest.

A total of 122 responses were received (response rate = 30.5%), and of these, 114 contained fully completed questionnaires. The majority of the participants were female (89.5%), non-Hispanic (94.7%), and White (87.7%) and held either a bachelor’s (38.6%) or master’s (56.1%) degree. Respondents reported an average age of 43.00 yr (standard deviation [SD] = 12.75) and an average of 15.56 yr (SD = 11.84) practicing as an occupational therapist. Participants were equally distributed across the four U.S. Census regions (Midwest = 28.9%, Northeast = 21.9%, South = 21.1%, and West = 28.1%), χ²(3) = 2.28, p = .516. In terms of practice areas, 41.2% reported working in a geriatric setting, 45.6% in a physical disability setting, and 13.2% in a combination of these two settings.

For each item on the MDS–R–OT[A], participants were asked to respond yes or no to the following question: Is the item relevant to occupational therapists working in adult settings? The item content validity index was calculated as the percentage of yes responses for each item (Table 1). Results show both agreement and disagreement on the relevance of the items. I–CVIs were 75% or above for all items except 2, 15, and 19. I–CVIs were within 5% of each other for geriatric and physical disability settings on 11 items (4, 6, 7, 8, 11, 13, 14, 16, 19, 20, and 21). I–CVIs differed by 5% or more favoring geriatric settings on six items (1, 3, 5, 9, 17, and 18) and favoring physical disability settings on four items (2, 10, 12, and 15). The scale content validity was calculated as the average I–CVI percentage of agreement for all 21 items. The S–CVI was 81.8% (geriatric = 81.5%, physical disability = 80.8%, multiple settings = 85.7%).

Inspection of the data in Table 1 revealed that three items (2, 15, and 19) did not meet a minimum I–CVI threshold of 70% (Lynn, 1986). Item 2 (69.9%), “Witness health care providers giving ‘false hope’ or false information to a client or family member,” was identified as having low relevance in geriatric settings (I–CVI = 61.7%) and more relevance in physical disability settings (I–CVI = 74.5%). Providing false hope is a consistent theme in the moral distress literature; thus, a decision was made to keep this item. It is theorized that the values of occupational therapy encourage practitioners to provide hope even in the face of seemingly insurmountable obstacles, so this may be less of a source of distress for occupational therapists compared with other health care professionals.

Item 15 (67.9%) is “Observe an ethical issue and have the involved staff member or someone in a position of authority request I do nothing.” The autonomous role of the professional occupational therapist was thought to be the cause for the relatively low relevance. We further modified the wording to make this item more specific to the practice of occupational therapy and the supervisory support provided: “Witness a violation of a standard of practice or a code of ethics and not feel sufficiently supported to report the violation.”

Lastly, Item 19 (60.7%) is “Avoid taking action in situations in which patients have not been given adequate information regarding their plan of care to ensure informed consent and engagement in therapy.” Although the issue of adequate informed consent is another major theme in the moral distress literature, most of this literature comes from the field of nursing, in which the overlapping responsibilities of the physician and the nurse might create tensions related to informed consent. In the profession of occupational therapy, it is the occupational therapist who has the responsibility for communicating the occupational therapy treatment plan and obtaining informed consent from the client, which could be the reason for the low I–CVI score in this study. After reviewing comments from 26 respondents who provided additional comments in this study and the unsolicited comments from the 2014 study, we created a new item to replace Item 19: “Provide services that do not fall within my scope of practice because of staff shortages in other areas or disciplines.”

Two final steps completed the creation of the MDS–R–OT[A]. The first was to review the written comments and recommendations from the respondents. Small changes were made to the wording of four items to expand but not alter the item concept; for example, Item 17, “Work with team members or health care providers who are not as competent as client care requires,” became “Work with team members or health care providers who are not as competent or experienced as client care requires,” and Item 21, “Work with team members or health care providers who give care that I consider unsafe,” became “Work with team members or health care providers who give care that I consider unsafe or unethical.” The last step was to review the items and confirm that they corresponded to the intended major root cause. The complete MDS–R–OT[A] is reprinted in the supplemental appendix (available online at http://otjournal.net; navigate to this article, and click on “Supplemental”).

The purpose of this study was to describe the development of the MDS–R–OT[A] and to establish the content validity of a version of the MDS–R with application to occupational therapy. Using a Grounded Theory approach, we replaced items in the MDS–R–OHPA that lacked relevance to occupational therapy practice while maintaining the major root causes of moral distress identified by Hamric et al. (2012) . Eight of the 21 items on the MDS–R–OHPA were replaced, and 12 of the remaining items were modified to align with occupational therapy practice and terminology.

The second stage of this study involved testing the content validity of the MDS–R–OT[A]. Three reasons led to the decision to use practicing occupational therapists rather than content experts to assess the content validity. First, during the creation of the MDS–R, the instrument was assessed by content experts for its theoretical foundation. Second, moral distress is an emerging field of study in occupational therapy, and few, if any, content experts are available within the discipline. Third, involving the intended population, in this case occupational therapists, in the assessment of content validity is supported by the literature (Grove et al., 2013; Rubio et al., 2003). Additionally, each practitioner is his or her own expert when it comes to moral distress; thus, practitioners are in the best position to assess the content validity of the MDS–R–OT[A]. In this study, an overall S–CVI of 81.8% was found, which meets the threshold of 80% for acceptable scale content validity (Davis, 1992; Grant & Davis, 1997; Polit & Beck, 2006).

A review of the items that received low I–CVI scores identified three items that did not meet a base threshold of 70% for content validity. One item was retained because of its high conceptual relevance as a cause of moral distress, one item was reworded to improve its relevance, and one item was replaced with a new item derived from comments received from respondents in this study and in Penny et al. (2014) . The finding that six items were deemed more relevant by therapists working in geriatric settings and four items were deemed more relevant by therapists working in physical disability settings suggests that the MDS–R–OT[A] may have the ability to differentiate between levels of moral distress experienced by occupational therapists working in those two settings.

The main limitation of this study was the decision to include only practicing occupational therapists and not occupational therapy assistants in the sample. Although we made this decision because it is the occupational therapist who is responsible for the therapy services delivered, it cannot be assumed that occupational therapy assistants do not experience moral distress. Indeed, occupational therapy assistants may experience moral distress in ways that are different from occupational therapists; thus, further research is needed in this area. A second limitation was the focus on just geriatric and physical disability settings, which limits the use of the MDS–R–OT[A] to these two groups; again, further research is needed for application in other settings. Another limitation is that content validity was established for occupational therapists working only in U.S. settings. Researchers interested in moral distress in countries other than the United States need to consider differences between the U.S. and their own health care systems.

The findings of this study have several important implications for occupational therapy research and practice:

• Modifying an instrument from another discipline is a viable approach for researchers seeking to measure emerging constructs in occupational therapy.

• Establishing instrument content validity does not need to be limited to expert opinion but can, and perhaps should, include the opinion of representatives from the intended population.

• The MDS–R–OT[A] possesses content validity and is ready for further investigation of its psychometric properties.

• That high I–CVIs were found for most items on the MDS–R–OT[A] suggests that moral distress has relevance to practicing occupational therapists.

• When practitioners encounter an ethical conflict, they need both cognitive problem-solving skills and emotional coping skills to resolve the conflict. Occupational therapy educators can provide opportunities for students to acquire both types of skills.

At the time of the study, all authors were affiliated with Alvernia University, Reading, PA. The research was conducted at Alvernia University, Reading, PA.