Edited by: Danny Horesh, Bar-Ilan University, Israel
Reviewed by: Matt R. Judah, Old Dominion University, United States; Michal Finklestein, Zefat Academic College, Israel; Jacob Y. Stein, Tel Aviv University, Israel
*Correspondence: Harold G. Koenig,
This article was submitted to Psychopathology, a section of the journal Frontiers in Psychiatry
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Experiences during combat have long been known to cause internal moral or ethical conflicts (
MI has been distinguished from posttraumatic stress disorder (PTSD), which may occur alongside it (
One reason that MI has received increasing attention over the past decade is the possibility that it may block successful treatment of PTSD, one of the most common mental disorders in Veterans and Active Duty Military (ADM) (
Systematic research has shown that MI is common among Veterans with PTSD symptoms. One study reported at least one MI symptom of significant severity in over 90% of 373 Veterans (59% with five or more such symptoms) (
The purpose of this study was to review measures used to assess MI that clinicians may use for screening and behavioral health investigators for conducting research in current and former military personnel. This review focused on scales that assess single or only a few dimensions of MI (guilt, shame, difficulty forgiving, loss of meaning, moral objections, and transgressions) and those that more comprehensively assess multiple aspects of this construct. In order to be comprehensive, we have included measures that address only one or two aspects of MI (e.g., transgressions, guilt, and shame). However, we do not believe that those measures are assessing the construct of MI as a unique phenomenon, but only assess certain dimensions of MI and are therefore incomplete in themselves.
Measures are distinguished in terms of whether they assess morally injurious events (experiences in war that cannot be changed) or MI symptoms (feelings about those events that can be altered by therapeutic interventions), or both events and symptoms. Reviewed are studies using these scales for the first time to assess MI in Veterans (including original validation studies) and later studies that have used those scales in military populations. Based on this review, recommendations are made on the best measures to use depending on the clinician’s or researcher’s goal. Treatments for MI are also briefly discussed.
The review focused on studies that developed or used measures of MI to examine health outcomes in present and former military personnel. Because the emphasis was on “moral injury,” this term was included either alone or with the keywords “Active Duty Military,” “Veterans,” “measuring,” and “screening.” The Boolean operators “and”/“or” were used between search terms to reduce the number of articles to those meeting the inclusion and exclusion criteria for this review. Inclusion criteria were 1) quantitative measurement of MI (scales including more than one item), 2) assessment of Veterans or ADM, 3) quantitative measurement of health outcomes, and 4) publication in a peer-reviewed academic journal in the English language. Excluded were literature reviews, dissertations, book chapters, letters to the editor, case reports, and qualitative studies.
The search strategy involved four stages. The first stage involved a search of the literature between 1980 and April 3, 2018, using the databases PubMed, PsychInfo, and Google Scholar. Second, the titles of promising articles were reviewed to identify studies that appeared to meet the inclusion criteria. Third, abstracts of these articles were reviewed. Finally, the full texts of articles that passed the initial screens were retrieved and examined more closely to ensure that inclusion and exclusion criteria were met. Each of the three co-authors independently conducted the review, screened relevant articles, and then agreed by consensus on the articles that met the criteria above.
Flow diagram of selection of studies (PRISMA chart).
The search term “moral injury” alone identified 62 articles in PubMed and 160 articles in PsychInfo, which represented the total number of articles identified by the three reviewers (all reviews were independently conducted in March and early April 2018). Given the number of articles in those two databases were relatively few, all were screened. When the keyword “moral injury” was used to search the Google Scholar database, however, over 5,000 articles were retrieved. To narrow down the search based on study inclusion criteria, the terms “Veterans,” “Active Duty Military,” “measuring,” and “screening” were added to the search term “moral injury” reducing the number of articles to 446, all of which were screened. Thus, search of the three databases identified 728 possible studies. Two additional studies were identified (known by the authors to be published soon), increasing the total to 730. Of those, 118 looked promising enough to download the full text articles and review them more carefully for inclusion criteria. Of those, 42 were eliminated for failing to meet inclusion and exclusion criteria leaving 76 eligible records. After excluding 34 duplicates, this resulted in the final 42 studies for this review. Most of these (93%) were published in 2013 or later, and 78% were published in 2017 or 2018, underscoring the recent attention paid to this topic.
Of the 42 studies, 17 studies developed or used previously published measures that assessed only one or two aspects of MI (e.g., guilt, shame, or forgiveness), and five studies reported the development of scales that assessed multiple dimensions of MI (
Characteristics of studies developing or using scales to assess moral injury (ordered by year of publication) (n = 22).
Reference (abbreviation) | Design | Population Studied | Events vs. Symptoms | Moral Injury Dimension | No. Items (Rating) | Source Scale | Psychometrics |
---|---|---|---|---|---|---|---|
Henning and Frueh ( |
CS | 40 Veterans with PTSD | Symptoms only | Guilt | 15 (1 vs. 0) | Authors | α = .78 |
Stein et al. ( |
CS | 122 ADM | Event Categories | MI by self |
2 (1 vs. 0) | Authors | kappa = .74-.90 |
Gray et al. ( |
NRCT | 44 Marines | Cognitions and beliefs | Trauma-related cognitions | 33 (1-7) | Posttraumatic Cognitions Inventory | — |
Nash et al. ( |
CS | 533 Marines |
Events and symptoms | Transgressions by self, others, and betrayal | 9 (1-6) | Authors | 2 factors (F) |
Bryan et al. ( |
CS | 69 ADM | Symptoms only | Guilt | 6 (0-4) | Personal | α = .85 |
Shame | 10 (0-4) | Feelings Questionnaire (PFQ-2) | α = .86 | ||||
Ritov et al. ( |
CS | 147 ADM (Israeli) | Symptoms (moral response to events) | Moral objections | 4 (1-7) | Authors | α = .83 |
Currier et al. ( |
CS | 131 Veterans 82 Veterans | Events and symptoms | Betrayal, moral violations, guilt, others | 19 (1-4) | Authors | 1 factor |
Bryan et al. ( |
CS | 474 ADM or Veterans | Symptoms only | Self-forgiveness | 6 (1-7) | Heartland Forgiveness Scale | α = .84 |
Hijazi et al. ( |
CS | 167 Veterans | Symptoms only | Wrongdoing | 5 (0-4) | Trauma-Related Guilt Inventory (TRGI) subscale | α = .78 |
Crocker et al. ( |
CS | 127 Veterans | Symptoms only | Shame | 24 (0-4) | Internalized Shame Scale; | α = .96 |
Guilt | 32 (0-4) | TRGI | α = .87-.91 | ||||
Campbell ( |
CS | 378 Sailors |
Symptoms only | Maladaptive shame regulation | 6 by 4 | Author | α = .89 |
Yan ( |
CS | 100 Veterans | Events only | Combat experiences (aftermath of battle) | 30 (1 vs. 0) | Deployment Risk & Resilience Inventory (DRRI) | α = .85-.86 |
Dennis et al. ( |
CS | 603 Veterans | Events and symptoms | Atrocities committed |
6 (1-5) |
Vietnam Stress Invent. |
α = .87 |
Frankfurt et al. ( |
CS | 190 Veterans | Events and symptoms | Transgressive acts |
8 (1 vs. 0) |
Clinician |
K = .72 |
Lancaster ( |
CS | 161 Veterans | Events and symptoms | Transgressions/betrayal |
6 (1-6) |
MIES (partial) |
— |
Maguen et al. ( |
RCT | 33 Veterans with PTSD | Symptoms and beliefs | Maladaptive beliefs about killing | 55 (1-5) | Author (Killing Cognitions Scale) | — |
Currier et al. ( |
CS | 286 Veterans |
Symptoms only | Self-directed, Other-directed (shame, guilt, betrayal, etc.) | 17 (1-5) | Authors | 2 factors |
Koenig et al. ( |
CS | 214 Veterans 213 Veterans (with PTSD symptoms) | Symptoms only | Guilt, shame, moral concerns, betrayal, religious struggles, loss of faith, loss of meaning, loss of trust, difficulty forgiving, self-condemnation | 45 (1-10) | Items from multiple established scales, and study authors | 1-2 factors per subscale Overall α = .92 Test-retest α = .91 |
Koenig et al. ( |
CS | 214 Veterans 213 Veterans (as above) | Symptoms only | Same as above MISS-M-LF | 10 (1-10) | Based on MISS-M-LF | 1 item/scale Overall α = .73 Test-retest α = .87 |
Nazarov et al. ( |
CS | 4854 ADM (Canadian) | Events only | Potential moral injury events (PMIEs) | 3 (1 vs. 0) | US/Canada Combat Experiences Scale | None reported |
Bryan et al. ( |
CS | 930 ADM | Symptoms only | Anger outward, hostility inward, shame, guilt, sorrow; | 15 (1-5) | Differential Emotions Scale-IV | α = .85-.93 |
low cohesion | 5 (1-5) | DRRI-2 | α = .91 | ||||
Currier et al. ( |
CS | 1124 Veterans | Symptoms only | Religious/spiritual struggles | 22 (1-5) | Religious and Spiritual Struggles Scale | α = or >.90 |
CS, cross-sectional; ADM, Active Duty Military;
Other studies in which moral injury scales in
Reference | Design | Population Studied | Events vs. Symptoms | MI Dimension | No. Items (Rating) | Source Scale | Psychometrics |
---|---|---|---|---|---|---|---|
Bryan et al. ( |
CS | 97 ADM | Symptoms only | Guilt | 6 (0-4) | PFQ-2 | α = .85 |
Bryan et al. ( |
CS | 151 ADM | Events and symptoms | Transgressions by self, by others, and betrayal | 9 (1-6) | MIES | 3 factors reported α’s > .79 reported |
Currier et al. ( |
CS | 131 Veterans | Events and symptoms | Betrayal, moral violations, guilt, others | 19 (1-4) | MIQ-M | — |
Bryan et al. ( |
CS | 464 ADM or Veterans | Symptoms only | Guilt | 6 (0-4) | PFQ-2 | α = .85 |
Bryan et al. ( |
CS | 151 ADM 935 ADM | Events and symptoms | Transgressions by self, by others, and betrayal) | 9 (1-6) | MIES | 3 factors demonstrated α’s = .83-.96 |
Wisco et al. ( |
CS | 564 Veterans | Events and symptoms | Transgressions by self, by others, and betrayal | 9 (1-6) | MIES | 3 factors reported α = .88 (total) |
Lancaster and Erbes ( |
CS | 182 Veterans | Symptoms only | Shame | 10 (0-4) | PFQ-2 | α = .92 |
Guilt | 5 (0-4) | α = .88 | |||||
Ferrell et al. ( |
CS | 37 Veterans | Events and symptoms | Betrayal, moral violations, guilt, others | 19 (1-4) | MIQ-M | — |
Currier et al. ( |
CS | 125 Veterans | Events and symptoms | Betrayal, moral violations, guilt, others | 19 (1-4) | MIQ-M | — |
Evans et al. ( |
CS | 155 Veterans | Events and symptoms | Transgressions, by self, by others, and betrayal | 9 (1-6) | MIES | 3 factors reported α = .91 |
Houtsma et al. ( |
CS | 522 ADM | Events and symptoms | Transgressions by self, by others, and betrayal | 9 (1-6) | MIES | 3 factors reported α’s = .75-.94 |
Jordan et al. ( |
CS | 867 Marines | Events and symptoms | Transgressions by self and betrayal | 7 (1-6) | MIES (partial) | 2 factors reported α’s = .84-.93 |
Martin et al. ( |
CS | 562 ADM | Symptoms only | Betrayal | 3 (1-6) | MIES (partial) | 1 factor reported α = .86 |
Cunningham et al. ( |
CS | 988 Veterans with PTSD | Symptoms only | Guilt (hindsight bias, wrongdoing, lack of justification) | 22 (0-4) | TRGI cognitions | α = .91 |
Yeterian et al. (planned) ( |
RCT | 186 Veterans | Symptoms only | Guilt |
32 (0-4) |
TRGI |
— |
Dedert et al. ( |
CS | 50 Veterans | Symptoms only | Guilt (hindsight bias, wrongdoing, lack of justification) | 18 (0-4) | TRGI cognitive subscales | — |
Volk and Koenig ( |
CS | 103 ADM w PTSD symptoms | Symptoms only | 10 MI symptom categories | 45 (1-10) | MISS-M-LF | α = .92 |
Norman et al. ( |
CS | 254 ADM | Symptoms only | Guilt (hindsight, bias, wrongdoing, lack of justification) | 22 (0-4) | TRGI cognitions | — |
Koenig et al. ( |
CS | 373 Veterans w PTSD symptoms | Symptoms only | 10 MI symptom categories | 45 (1-10) | MISS-M-LF | α = .92 |
Zerach and Levi-Belz ( |
CS | 191 Israeli combat Veterans | Events and symptoms | Transgressions by self, by others, and betrayal | 9 (1-6) |
MIES |
— |
The majority of studies used scales that assessed only one or two dimensions of MI in Veterans and ADM. These studies either a) reported the development of a new scale or b) used previously published scales or subscales that had assessed specific aspects of MI in non-military populations (discussed below by year of publication). We include these scales for background only in this comprehensive review.
Regarding studies reporting the development of a
Stein and colleagues conducted structured clinical interviews with 122 active duty Army personnel, who had experienced traumatic events during their military service (
Ritov and colleagues developed a 4-item scale assessing “moral objections” (MO) to commands given by superior officers (
Campbell reported the development of a scale assessing “shame,” called the Military Compass of Shame Scale (M-CoSS) (
Lancaster administered two 5-item subscales from the 15-item State Shame and Guilt Scale (
Finally, Maguen and colleagues conducted a RCT examining effects of the Impact of Killing (IOK) intervention in 33 combat Veterans with PTSD (
Rather than examine MI using a new scale, several studies have used scales or subscales from existing measures originally published and validated in non-military populations or used for purposes other than examining MI. Gray and associates conducted an open trial (without a control group) examining Adaptive Disclosure Therapy (ADT) in 44 active duty Marines (
Bryan (CJ) and colleagues administered the 6-item guilt and 10-item shame subscales of the Personal Feelings Questionnaire (
Bryan (AO) and colleagues administered the six-item self-forgiveness subscale from the Heartland Forgiveness Scale (
Next, Hijazi and colleagues administered the 5-item “wrongdoing” subscale from the 32-item Trauma-Related Guilt Inventory (TRGI) (
In another study assessing guilt and now also shame, Crocker and colleagues examined whether these indicators of MI mediated the relationship between PTSD symptom severity and aggression in 127 U.S. Veterans returning from deployment to the Middle East (
Yan administered the Combat Experiences (CE) and Aftermath of Battle (AB) subscales from the Deployment Risk and Resilience Inventory (DRRI-1) (
Likewise, Dennis and colleagues examined the relationship between PIES and mental health outcomes in 603 U.S. combat Veterans seeking mental health services for PTSD (
Frankfurt and associates asked questions on commission of transgressive acts (PMIEs) from the Clinician Administered PTSD Scale-IV (
In one of the few studies of military personnel outside of the U.S., Nazarov and colleagues examined the relationship between PMIEs, PTSD, and depressive symptoms in 4,854 Canadian ADM (reserve ADM deployed to Afghanistan and members of the regular armed forces) (
Bryan (CJ) and colleagues administered five three-item subscales of the Differential Emotions Scale-IV (
Finally, Currier and colleagues examined Veterans’ preferences for incorporating spirituality into therapies for treating PTSD or major depression (
Of the 22 studies, five were designed to assess multiple dimensions of MI in Veterans or ADM. Two of the five scales measure a combination of events and symptoms, and three scales measure MI symptoms alone. We describe each of these measures below.
The greatest strength and the greatest weakness of the MIES is that it measures both the occurrence of transgressive events and the symptoms associated with those events. Including events that might be the cause of MI symptoms makes it excellent as a screening measure, since it identifies specific events that might be the target of interventions. The inclusion of events, however, means that the MIES might be less useful in intervention studies that seek to assess change in MI symptoms over time, in that the inclusion of MI events in the MIES that cannot change complicates the assessment of MI symptom change in response to treatment.
The MISS-M-LF (long form) was designed for use in Veterans and ADM with PTSD symptoms. The measure assesses 10 dimensions of MI that capture both the psychological and the spiritual or religious (S/R) symptoms of this construct. Each dimension of the MISS-M-LF was intentionally chosen based on the definitions for MI reported in the literature. Psychological symptoms assessed include guilt (4 items), shame (2 items), betrayal (3 items), moral concerns (3 items), loss of meaning and purpose (4 items), difficulty forgiving (7 items), loss of trust (4 items), and self-condemnation (10 items). S/R symptoms assessed include religious struggles (six items) and loss of religious faith and hope (two items). Items making up the scale were derived primarily from existing scales published in the literature. All items are rated on a scale from 1 to 10 (total score range 45 to 450).
To ensure that items with strong face validity for a particular dimension ended up on the subscale assessing that dimension, EFA and CFA were conducted at the subscale level rather than at the item level. A sample of 427 Veterans and ADM with PTSD symptoms was randomly split into two groups. EFA was performed on an original pool of 54 items in the first half of the sample (n = 214). EFA identified one or two factors per dimension and reduced the total number of items to 45 when only those items with factor loadings ≥ 0.45 were retained. The factor structure for each dimension was then independently verified using CFA in the second half of the sample (n = 213). The final MISS-M-LF had high internal reliability (α = 0.92) and test–retest reliability [intraclass correlation (ICC) = 0.91]. Discriminant validity was demonstrated by relatively weak correlations with S/R measures, community activities, and indicators of physical health; convergent validity was indicated by strong correlations with symptoms of PTSD, anxiety, and depression (r’s ranging from 0.56 to 0.62). The MISS-M-LF is the first multidimensional scale that measures both the psychological and S/R symptoms of MI, and because it measures symptoms alone, the scale can be used for tracking symptom severity in clinical practice and for conducting research that examines treatments for MI in Veterans and ADM that target MI symptoms.
In order to create a shorter measure that might facilitate its use by clinicians and researchers, an abbreviated version of the MISS-M was developed (
The MIES is currently the most frequently used multidimensional measure in the literature that assesses PMIEs and MI symptoms, followed by the MIQ-M (
Scales measuring events, symptoms, and events and symptoms.
Events Only | Symptoms Only | Events and Symptoms |
---|---|---|
Event Categories ( |
Combat Guilt Scale ( |
Moral Injury Events Scale ( |
Vietnam Stress Inventory | Posttraumatic Cognitions Inventory ( |
Moral Injury Questionnaire ( |
(atrocities exposure subscale) ( |
Personal Feelings Questionnaire | Deployment Risk & |
Moral Objections Scale ( |
(guilt and shame subscales) ( |
Resilience Inventory ( |
Clinician PTSD Scale-IV | Heartland Forgiveness Scale) | |
(transgressive acts subscale) ( |
(self-forgiveness subscale) ( |
|
Transgressive Acts Scale ( |
Trauma Related Guilt Inventory ( |
|
Combat Experiences Scale ( |
Internalized Shame Scale ( |
|
Military Compass of Shame Scale ( |
||
State Shame and Guilt Scale ( |
||
Killing Cognitions Scale ( |
||
Expressions of Moral Injury Scale ( |
||
Moral Injury Symptoms Scale-LF ( |
||
Moral Injury Symptoms Scale-SF ( |
||
Differential Emotions Scale-IV ( |
||
Religious & Spiritual Struggles Scale ( |
Moral injury is a term now used widely in clinical discussions and research studies involving Veterans and ADM personnel (
This is the first comprehensive review of MI measures developed specifically for use in current or former military personnel. We described the development of these measures, their psychometric properties, and their relationship to mental health outcomes such as PTSD, anxiety, depression, and suicide risk. These measures assess PMIEs or transgressions, current symptoms of moral conflict over those events, or both events and symptoms. Some scales measure either one or two aspects of MI, whereas others assess multiple dimensions. Because some measures are new (published within the past 12 months), clinicians and researchers have had little opportunity to use them outside of the original validation studies, underscoring the need for future studies.
Nevertheless, it is becoming increasingly clear that MI is a syndrome associated with much distress and comorbidity, making it necessary for clinicians treating Veterans or ADM and for those doing research in these populations to be aware of both earlier and more recent measures. This is particularly important because of the role that MI may play in the pathway that leads from war trauma to the development and maintenance of PTSD (
Further epidemiological research is necessary to determine whether and how MI affects PTSD (and related co-morbidity) over time and how MI is affected by these conditions, all of which requires longitudinal studies have yet to be done. However, given the high prevalence of MI among Veterans and military personnel with PTSD and the frequent lack of recognition by clinicians, it may be important to start now to identify those with significant MI symptoms through screening (
The field, however, is moving fast. Despite knowing relatively little about MI or how it relates to PTSD over time, researchers are now developing and testing interventions to treat some aspects of MI in both Veterans and ADM (
Thus, many of the MI measures above will be useful for both clinicians working with patients and researchers designing and implementing research studies. However, none of the measures reviewed here was created using a gold standard methodology, such as by starting with representative focus groups to collect a comprehensive list of all possible symptoms, behaviors, affects, and cognitions that might possibly be a result (and component) of MI, and then see what correlates with what, letting the data create the symptom clusters. The EMIS goes a long way in this regard, although possibly not far enough. Without doing such heavy lifting involved in the discovery of symptoms clusters from a much larger pool, researchers cannot be sure that they’ve got the right measure that comprehensively assesses this concept. The development of measures driven solely by statistical grouping, on the other hand, may not be the ideal solution either, since the face validity of items guided by theory should also play some role in determining items for a comprehensive measure of any new construct. That too cannot be ignored.
A number of limitations should be considered when interpreting the results of this review. First, not examined here were MI scales designed to assess symptoms resulting from traumatic experiences occurring outside of the military, such as trauma from assault, rape, or natural or man-made disasters. This may not have always been indicated in the scales. For example, the MIQ-M specifies that MIEs must have occurred in the context of wartime deployment, whereas other measures are not as clear in that instruction. Second, this review was also limited by not including all studies that measured various dimensions of MI (e.g., guilt, shame, difficulty forgiving, self-condemnation, and loss of meaning or trust), particularly those that did not include the term “moral injury” in the title, abstract, or full text of the article (an inclusion criterion for this review). The relative recency of the term “moral injury” likely contributed to missing such studies. However, conducting a review that separately examined each possible dimension of MI (indicated by a wide range of terms) would have gone beyond the scope of this paper. Third, and perhaps most concerning, the present authors developed two of the measures discussed in this review (MISS-M-LF and MISS-M-SF), thus introducing the possibility of bias in study description, particularly since these two measures are recommended for use (see below). In order to address this bias, the authors have described the other three multidimensional measures as comprehensively and accurately as possible, especially the only other “pure” MI symptom measure, the EMIS-M. Despite these efforts, readers should be aware that this bias may have colored our descriptions of these measures. Finally, the scales reviewed here (even those assessing PMIEs) did not always identify the exact circumstances in which Veterans or ADM experienced their trauma, i.e., whether this occurred while fighting in combat, during deployment but not combat, or either before or after returning from deployment, and the specific nature of the trauma (assault, rape, etc.), which clinicians will need to explore beyond simply administering a scale.
As noted earlier, we have included measures in this review that address only certain aspects of MI (e.g., transgressions, guilt, and shame). These measures, in our opinion, are not assessing the complete phenomenon of MI, but rather only certain dimensions of this construct. For this reason, we recommend the use of multidimensional measures that go beyond measuring guilt and shame and are more likely to capture MI as the unique phenomenon that experts in the field now describe (see above). However, given the limitations noted above, these recommendations should be viewed as strictly preliminary rather than instructive.
As always, the scale chosen will depend on the purpose of the clinician or investigator. Multidimensional scales that assess
The two shortest scales for clinicians are the 9-item MIES (
While the recognition of inner conflict over moral transgressions in former or current military personnel has increased during the past decade, many clinicians and researchers may not know how to measure or treat these injuries. There is growing evidence that MI in Veterans and ADM is associated with adverse mental health states, including PTSD, depression, anxiety, and risk of suicide, and may block the treatment of these conditions unless also addressed. We identified 42 studies in this review that used scales to assess one or more aspects of MI as currently defined. Among those studies, 17 reported the use of scales that assessed only one or two dimensions of MI, while five studies reported the development and psychometric properties of scales assessing multiple dimensions. These measures assess morally injurious events, symptoms that result from the events, or both events and symptoms. Measures that assess both events and consequences are assessing the morally injurious event and the symptoms that the event may cause. Some events may not result in symptoms, whereas some symptoms assessed may not result from the morally injurious event. Therefore, when clinicians are using these scales to screen for MI, a clinical interview will be necessary to clarify which MI symptoms may have followed the acknowledged event, and which MI symptoms may have other causes (possibly prior traumas during youth or adulthood).
In comparing the comprehensiveness, internal consistency, and validity across the five multidimensional measures, the 45-item MISS-M-LF (and shorter 10-item MISS-M-SF) is probably the most comprehensive, assessing 10 dimensions of MI including both psychological and spiritual aspects. With regard to internal consistency and reliability, all five scales have solid psychometric properties, although the 17-item EMIS-M has perhaps the best internal reliability (alphas exceeding 0.92) and test–retest reliability (ICCs in the 0.74 to 0.80 range), as well as strong concurrent validity with PTSD symptoms (r = 0.69–0.73), depression (r = 0.58–0.65), and loss of meaning (r = 0.69), established in large samples. However, except for the 9-item MIES and 19-item MIQ-M, the newer scales have not yet been used in many studies (as noted above), so the performance of these scales (MISS-M-LF, MISS-M-SF, and EMIS-M) in other populations and settings still needs to be demonstrated.
Multidimensional scales that assess both events and symptoms (MIES and MIQ-M) are recommended for clinicians who wish to screen Veterans and military personnel for MI and for researchers who wish to conduct observational studies on this syndrome. Multidimensional scales that assess symptoms only (MISS-M-LF, MISS-M-SF, and EMIS-M), however, are recommended for clinicians and researchers wishing to track change in MI symptoms with treatment. Future longitudinal studies are needed to identify cutoff scores and clinically significant change scores for these measures. Likewise, clinical trials are needed to determine whether treatments directed at MI not only reduce MI symptoms but also impact the many adverse mental health outcomes that have been associated with it.
HK is a researcher and psychiatrist at Duke University Medical Center in Durham, NC, USA. He contributed to the literature review and is the main author of this article. NY is a researcher and psychiatrist at the Medical College of Georgia and Charlie Norwood Veterans Administration Medical Center in Augusta, GA. He contributed to the literature review and the writing and editing of this paper. MP is a researcher and psychologist at the University of Maryland. She contributed to the literature review and the writing and editing of this paper. In addition, HK, NY, and MP all made important intellectual contributions to this article.
This research was not supported by a grant from a funding agency in the commercial, public, or not-for-profit sectors. The study received no funding from any outside funding bodies. The study authors’ time was covered by their individual departments.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.