Page 62 - JSOM Summer 2019

In 2016, more than 5 years after initial publications about CSF From distilling the RSES, De La Rosa et al. developed and
45
and GAT implementation, Vie et al. reported the “Initial validated a four-item instrument intended to measure resil-
41
42
Validation of the US Army Global Assessment Tool,” in which ience in military populations. Researchers sampled several
40,000 soldiers participated. In their report, the theoretical participant groups with different instruments: the RSES, the
basis for the GAT is discussed in regard to the development of PTSD Checklist (military) scale (PCL-M), brief resilience scale,
the CSF program. CSF is “framed by positive psychology” and burnout measure, and quality-of-life burnout subscale. To
45
the GAT assesses positive individual personality traits, posi- develop the four-item instrument, four participant samples of
tive emotions, positive psychological attributes, and positive various sizes were used; civilian personnel and active-duty per-
mental health outcomes. Many elements of the GAT draw sonnel were included as participants in the last sample group
42
from previous work of one of its principal architects, positive (sample 4), whereas sample groups 1–3 excluded civilians and
psychology guru Martin Seligman. Specifically, “the GAT maintained participant homogeneity, a plus for internal va-
42
45
assesses positive emotions, meaning, and personal attributes lidity. Sample 1 in that study had 1,448 participants, whereas
(i.e., optimism) that contribute to a full life” such as good cit- sample 4 had 68 participants. Granular psychometric prop-
45
izenship and other admirable character traits. 42 erties of each scale used in development of the four-item scale
were not discussed before descriptions of their use in the study.
Developers creating the GAT elected to not use some ele- The reduction by De La Rosa et al. of the RSES from a 22-
ments of more-established resilience scales and instead sub- item to a four-item instrument was shown by them to be psy-
stituted and interchanged psychosocial fitness with resilience chometrically valid, yet its application as a tool to measure
45
in regard to assessing Servicemembers’ strengths; however, resilience was not explained fully, because they primarily used
the rationale for specific substitutions is not described. The proxy measures of stress responses.
42
GAT was reportedly developed from extant validated scales
measuring coping, pessimism, work satisfaction, loneliness, Resilience measured in nontherapeutic contexts
flexibility, and so forth, but developers excluded specific mea- Originally developed for civilian clinical practice to deter-
sures of friendship and social supports. Vie et al. omitted those mine baseline resilience and therapeutic results to improve
measures “because of the different [dichotomous] response coping mechanisms, the Connor-Davidson Resilience Scale
formats.” They referenced the work of Paul Bartone, 32,34,35 (CD-RISC) is built on factors of individual competency, trust,
42
principal architect of the Dispositional Resilience Scale (DRS), tolerance, effects of stress on personal strength, assimilation
used widely in military resilience research, as inspiration for of change, relationship stability, locus of control, and effects
GAT construction, but specific elements of Bartone’s DRS of individual spirituality. To evaluate the efficacy of master
37
(e.g., commitment, control, and challenge) 34,35 are not immedi- resilience training (MRT) program integration, Carr et al.
46
ately visible within the GAT’s structure nor subscales. 42 administered the CD-RISC to more than 200 military partic-
ipants before and after resilience training. In 2013, Carr et
Resilience measured as a response to stress al. combined the CD-RISC with an instrument they created
In 2016, Vyas et al., of the Naval Center for Combat and and found overall resilience declined in their selected sample.
16
46
Operational Stress Control (part of the Operational Stress MRT did not improve stress mitigation or, by proxy, coping;
Control and Readiness [OSCAR] resilience program), retro- resilience training significantly decreased resilience and morale
spectively examined 2,171 Servicemembers’ mental health re- (p = .007) . The CD-RISC has high internal reliability (Cron-
46
cords from 2009 to 2013 to determine those Servicemembers’ bach’s α = .89) and is often used as a standard against which
resilience from proxy measurements found in the Response to new instruments measuring resilience are designed. 46
Stressful Events Scale (RSES). Their interpretation of results
16
indicated that improving resilience in Servicemembers by even Within similar nontherapeutic contexts, Hernandez et al.
47
20% would significantly reduce statistical risks of PTSD, de- studied 245 military registered nurses (RNs) and medical per-
pression, and subsequently save the DoD approximately $600 sonnel with the CD-RISC; reliability of the CD-RISC ranged
million or more in health care costs. No discussion occurred from .89 to .94, though that was their only discussion of
47
16
regarding the RSES’ specific psychometric properties, nor was the CD-RISC’s psychometric properties in that study. Via the
a rationale presented for choosing an instrument that mea- CD-RISC, Hernandez et al. found weak statistical connections
sures responses to stressful events versus using an instrument among stress, mental health stigma, and resilience in their
that specifically measures resilience. Using odds calculations sample of military clinicians, though stigma and resilience
16
and logistic regression, investigators estimated substantial were determined to be negatively associated factors. In that
47
cost savings but then disclose that the population of partici- study, military RNs (n = 141), who are officers, had higher
pants was drawn from military mental health clinics, possibly measured resilience and reported more mental health stigma
skewing the underlying resilience and psychopathology of the (p < .05) than did enlisted military medical technicians (n =
group, and thus the results. 16 104), though enlisted technicians reported significantly higher
stress than did RNs (p < .05). 47
Resilience and human responses to stress were similarly inter-
changed by Johnson et al. in their 2011 study of resilience Resilience as a dispositional personality trait
43
with 870 Servicemembers and veterans. Johnson et al. also Bartone’s original 22-item DRS instrument has been used
proposed that resilience is found by a complete absence of widely to measure resilience in military populations. Bar-
34
PTSD. However, main constructs of the RSES (e.g., social tone et al. also administered the DRS-15 (shortened to 15
43
48
support, personal faith, positivity, cognitive flexibility, self- questions) to 7,555 Norwegian soldiers while measuring
efficacy, coping, and hardiness) are disconnected from diag- self-reported alcohol use. They reported that subscale analy-
nostic criteria for PTSD (e.g., traumatic exposure, persistent ses found acceptable DRS internal consistency (α coefficients
re-experiencing, avoidance, anhedonia, sustained hypervigi- for commitment = .77; control = .68; challenge = .69). Those
lance and/or aggression, creating functional impairments. 44 findings were consistent with psychometric results obtained
60 | JSOM Volume 19, Edition 2 / Summer 2019