Psychiatry 77(2) Summer 2014
199
Disaster Mental Health Workforce Capacity Reifels et al.
Examining Disaster Mental Health Workforce Capacity Lennart Reifels, Lucio Naccarella, Grant Blashki, and Jane Pirkis
Objective: We examined the capacity of the disaster mental health workforce in Victoria, Australia, to provide the three evidence-supported intervention types of psychological first aid, skills for psychological recovery, and intensive mental health treatments. Method: Utilizing data from a cross-professional, state-level disaster mental health workforce survey (n = 791), we developed composite capacity indicators (CCI) for each intervention and performed logistic regression analyses to examine key predictors of disaster mental health workforce capacity. Results: CCI profiles highlighted significant gaps in the disaster mental health capacity of Victorian providers, with only 32–42% able to deliver current best practice interventions. Key predictors of workforce capacity common and unique to interventions were highlighted. Conclusions: Key strategies to raise Victoria’s disaster mental health workforce capacity should focus on targeted multilevel training in best practice interventions, creation of practice opportunities, and structural provider support/engagement. CCIs focused on best practice interventions provide a methodology for rapid workforce capacity assessment that can facilitate disaster preparedness planning, capacity building, and delivery of quality disaster mental health services. Lennart Reifels, Ph.D., and Jane Pirkis, Ph.D., are affiliated with the Centre for Mental Health, Melbourne School of Population and Global Health, The University of Melbourne. Lucio Naccarella, Ph.D., is affiliated with the Australian Health Workforce Institute, Melbourne School of Population and Global Health, The University of Melbourne. Grant Blashki, M.D., is affiliated with the Nossal Institute for Global Health, The University of Melbourne. This study was funded through the Natural Disaster Resilience Grant Scheme of the Office of the Emergency Services Commissioner in Victoria and the Australian Attorney-General’s Department. The authors wish to thank Rachel Sore, MSc, and Matthew Spittal, Ph.D., for providing valuable statistical advice in relation to the paper, as well as the following agencies which facilitated the conduct of the research: the Victorian Departments of Health and Human Services, the Australian Centre for Posttraumatic Mental Health, the Melbourne Sustainable Society Institute, the Australian Association of Social Workers, the Australian Psychological Society, Occupational Therapy Australia, the Royal Australian & New Zealand College of Psychiatrists, the Australian College of Mental Health Nurses, the Australian Nursing Federation (Victoria Branch), General Practice Victoria, the Australian Counselling Association, the Victorian Council of Churches, Beyondblue: The National Depression and Anxiety Initiative and the Australian Red Cross. An earlier version of this paper was presented at the XIII Conference of the European Society for Traumatic Stress Studies in Bologna, Italy, in June 2013. Address correspondence to Lennart Reifels, Centre for Mental Health, Melbourne School of Population and Global Health, The University of Melbourne, Carlton, 3010 Victoria, Australia. E-mail: [email protected]
© 2014 Washington School of Psychiatry
200
Disaster Mental Health Workforce Capacity
TABLE 1. The Australian Disaster Mental Health Response Framework Psychological First Aid Purpose Population-level support of common distress responses in the immediate aftermath Core Principles:
Skills for Psychological Recovery
Intensive Mental Health Treatments
Purpose Support of individuals with mild to moderate sub-clinical levels of distress
Purpose Treatment of minority experiencing significant mental health conditions
Modules:
Treatments:
Promote safety
Gathering information and prioritizing assistance
Exposure treatments
Promote calming
Problem-solving skills
Cognitive therapy
Promote self-efficacy
Promoting positive activities
Goal setting/activity scheduling
Promote connectedness
Managing reactions
Managing anger
Promote hope
Promoting helpful thinking
Treatment of complicated grief
Promote help
Rebuilding healthy social connections
Managing comorbidity Pharmacotherapy
Providers Disaster relief workers, volunteers, community leaders, generic workers
Providers Primary care providers, including: family physicians, allied health professionals, counselors, welfare staff
Providers Specialist mental health care staff, including: Psychologists, psychiatrists
Note. Adapted from Bryant et al. (2009) and the Australian Psychological Society (2010).
Large-scale natural disasters can have a profound and differential impact on the mental health of affected populations and may therefore require enhanced mental health responses (Norris, Friedman, & Watson, 2002). Vital to the provision of enhanced mental health services is a workforce with the capacity to deliver current best practice interventions. Recent advances in Australian disaster responses involved the implementation of a three-tiered disaster mental health intervention framework in support of affected populations that was underpinned by national expert consensus and first employed in response to Australia’s worst fire disaster, the Victorian Black Saturday Bushfires (Australian Psychological Society, 2010; Bryant et al., 2009). This framework comprised three evidence-supported interventions (Table 1): Psychological First Aid (PFA), a primary psychosocial prevention/support model for common disaster distress responses (Wade et al., 2013); Skills for Psychological Recovery (SPR), an intermediate skill-building/
prevention model for individuals experiencing mild to moderate distress levels (Forbes et al., 2010); and intensive Mental Health Treatments (MHT), evidence-based core interventions for severe mental health issues commonly experienced by a minority of survivors (Australian Centre for Posttraumatic Mental Health, 2007). Consistent with a modern public mental health approach to disaster mental health care (Watson, Brymer, & Bonanno, 2011) and underpinned by a stepped care model, the framework was operationalized through a training agenda that targeted disaster volunteers, allied, general and mental health professionals. Despite these advances, little is known about the capacity of the Victorian disaster mental health workforce to deliver best practice interventions. Moreover, intrinsic obstacles to the efficient and timely delivery of quality disaster mental health services remain. These concern provider adherence to ineffective interventions; the need for provider mobilization/training/accreditation; the influx/management of spontaneous volun-
Reifels et al. 201
teers; and varied practical/logistical barriers to disaster response involvement. One of the key challenges to effective disaster response planning therefore concerns the lack of systematic data on the capacity of the multifaceted workforce which can provide best practice interventions in disaster-affected communities (Reifels et al., 2013). In order to inform future disaster workforce planning, this study constituted the first systematic state-level examination of disaster mental health workforce capacity in Victoria, Australia, and the world. Workforce capacity was primarily examined at a provider level and conceived of as a multidimensional concept centered on two key aspects, namely, the ability of providers to deliver best practice disaster mental health interventions (Table 1) and effectively participate in disaster responses. METHOD
Sample We utilized data from a state-level survey (N = 791) of the Victorian disaster mental health workforce (Reifels, Naccarella, Blashki, & Pirkis, 2012) with the following sample characteristics: mean age 51.7 years (range 22–82, SD 11.7); female (71.6%); bilingual (10.0%); tertiary qualified (82.2%); including psychologists (25.3%), nurses (18.5%), counselors (14.5%), social workers (14.3%), pastoral care workers (13.4%), teachers (6.7%), community workers (3.2%), and other professions (24.5%); in metropolitan (63.1%) or regional Victoria (33.9%); with paid employment (86.4%) and/or volunteering roles (37.4%); primarily involving direct service delivery (78.6%); in work settings including private practice (26.8%), NGOs (21.0%), pastoral care services (16.8%), psychiatric hospitals (6.7%), clinical mental health services (6.6%), general hospitals
(5.3%), schools (4.6%), primary care clinics (4.4%), and other contexts (34.8%). Composite Capacity Indicators Composite workforce capacity indicators (CCI) were developed for each key intervention comprising six equally weighted variables reflective of desirable provider characteristics: 1. Training completion (yes = 1/no = 0) 2. Interest to provide intervention: at least “moderate” (yes = 1/no = 0) 3. Confidence to provide intervention: at least “moderate” (yes = 1/no = 0) 4. Experience providing intervention (yes = 1/no = 0) 5. Capacity to participate in disaster mental health responses: “moderate” (yes = 1/no = 0) for at least one delivery mode: –Face-to-face service delivery –Telemental health –Outreach to affected areas –One-off visits to affected areas –Temporary relocation 6. Understanding of disaster aspects: “reasonable” (yes = 1/no = 0) for at least three disaster aspects: –Impact of disasters on affected individuals –Impact of disasters on affected communities –Mental health consequences of disasters –Disaster preparedness –Disaster recovery –Emergency management –Organizational emergency plan –Local area emergency plan –Regional/state emergency plans
202
Disaster Mental Health Workforce Capacity
CCI summary scores (range 0–6 points) enabled the creation of binary outcome variables that indicated either the presence (threshold 4–6 points) or absence (0–3 points) of capacity to provide each intervention. Data Analysis Exploratory logistic regression analyses performed in SPSS v20 on three binary outcome variables examined predictors of disaster mental health workforce capacity through a process of stepwise-backward elimination (Wald, p < 0.15 entry, p < 0.01 retention). Predictor variables included provider: age, gender, language, professional background, qualification level, organizational affiliation, work location, setting, role, experience, primary activity, work hours, disaster response barriers and enablers. Further details on the range, scales and means of predictor variables are available from the authors. RESULTS
CCI Profiles Mean CCI scores for PFA, SPR and MHT interventions were 3.39 (SD 1.72), 3.07 (SD 1.58) and 2.97 (SD 1.62), respectively. Application of the capacity threshold (4–6 points) demonstrated that only 41.8% of respondents had the required capacity to provide PFA, 31.6% to provide SPR, and 32.0% to provide MHT interventions. Almost one-half (46.5%) lacked capacity to provide any of these interventions, while others had capacity to provide one (19.2%), two (14.5%), or three (19.0%) interventions.
Predictors of Disaster Mental Health Workforce Capacity Logistic regression models in Table 2 outline statistically significant predictors (p < 0.01) of disaster mental health workforce capacity for each intervention, with composite indicators as binary outcome variables. Tests of resulting models against constant-only models were statistically reliable, as indicated for PFA by c² (7, n = 689) = 138.77, p < .001, for SPR c² (10, n = 688) = 176.01, p < .001 and MHT c² (10, n = 690) = 246.10, p < .001. Thus, all three sets of predictors reliably distinguished the presence and absence of capacity at respective levels. Nagelkerke’s R2 of .245 (SPR), .317 (PFA), and .420 (MHT) indicated weak to moderately strong relationships between prediction and grouping. Prediction success overall was 70.5% (SPR), 76.7% (PFA), and 79.3% (MHT). Key predictors of PFA capacity (with odds ratios in brackets) were: organizational affiliation with the Victorian Council of Churches (3.19); direct service provision (1.84); a need for SPR training (1.59); work experience with disaster survivors (1.37); and a need for peer support (1.20). Key predictors of SPR capacity included: bilingual providers (2.28); direct service provision (2.08); psychology background (1.79); the need for other training (1.78); work experience with disaster survivors (1.56); and a need for MHT training (1.34). Key predictors of MHT capacity included: psychology (3.49); or counseling backgrounds (2.45); limited role flexibility (1.81); work experience in mental health (1.58); or with disaster survivors (1.44); older age (1.35); and required involvement opportunities (1.23). Overall effects of organizational affiliation and professional backgrounds were congru-
Reifels et al. 203
TABLE 2. Logistic Regression Models and Predictors of Disaster Mental Health Workforce Capacity for Each Intervention Type OR
95% CI
p-value
Psychological First Aid (PFA) Work experience—Disaster survivors
1.37
1.19, 1.57
0.000
Primary role activity—Direct service provision
1.84
1.18, 2.85
0.007
Organizational affiliation—Victorian Council of Churches
3.19
2.05, 4.95
0.000
Key barrier—Lack of familiarity with DMH interventions
0.42
0.27, 0.65
0.000
Key enabler—PFA training
0.57
0.45, 0.72
0.000
Key enabler—SPR training
1.59
1.25, 2.01
0.000
Key enabler—Peer support
1.20
1.05, 1.36
0.008
Constant
0.17
0.000
Skills for Psychological Recovery (SPR) Professional background—Psychologist
1.79
1.18, 2.71
0.006
Languages other than English
2.28
1.24, 4.17
0.008
Work experience—Disaster survivors
1.56
1.34, 1.82
0.000
Average weekly working hours
0.88
0.82, 0.94
0.000
Primary role activity—Direct service provision
2.08
1.24, 3.50
0.006
Organizational affiliation—Australian Red Cross
0.07
0.02, 0.26
0.000
Key barrier—Lack of dedicated DMH training
0.48
0.31, 0.75
0.001
Key enabler—PFA training
0.73
0.62, 0.86
0.000
Key enabler—MHT training
1.34
1.14, 1.57
0.000
Key enabler—Other training
1.78
1.27, 2.48
0.001
Constant
0.10
0.000
Intensive Mental Health Treatments (MHT) Age
1.35
1.12, 1.62
0.001
Professional background—Psychologist
3.49
2.24, 5.44
0.000
Professional background—Counselor
2.45
1.39, 4.32
0.002
Professional background—Pastoral care worker
0.17
0.06, 0.46
0.000
Work experience—People with mental illness
1.58
1.32, 1.89
0.000
Work experience—Disaster survivors
1.44
1.22, 1.70
0.000
Organizational affiliation—Australian Red Cross
0.05
0.01, 0.37
0.003
Key barrier—Lack of familiarity with DMH interventions
0.46
0.28, 0.76
0.002
Key barrier—Limited scope or flexibility in existing role
1.81
1.21, 2.73
0.004
Key enabler—Opportunities for involvement
1.23
1.07, 1.40
0.003
Constant
0.00
0.000
Note. OR = odds ratio; CI = confidence interval; DMH = disaster mental health. Providers affiliated with the Victorian Council of Churches and the Australian Red Cross were primarily organized disaster response volunteers. Variables are coded as follows: Disaster mental health capacity: 0 = no, 1 = yes; Work experience: 1 = no experience, 2 = a little, 3 = quite experienced, 4 = considerably, 5 = very experienced; Primary role activity: 0 = no, 1 = yes; Organizational affiliation: 0 = no, 1 = yes; Key barrier: 0 = no, 1 = yes; Key enabler: 1 = not a priority, 2 = low priority, 3 = moderate priority, 4 = high priority, 5 = very high priority; Professional background: 0 = no, 1 = yes; Languages other than English: 0 = no, 1 = yes; Average weekly working hours: 1 = 0, 2 = 1–5, 3 = 6–10, 4 = 11–15, … 11 = 46–50, 12 = > 50; Age (in years): 1 = 21–30, 2 = 31–40, … 7 = 81–90.
204
Disaster Mental Health Workforce Capacity
ent with agency core business and anticipated levels of provider involvement. The lack of familiarity with disaster mental health interventions, need for PFA training and high weekly working hours were all associated with decreased capacity. DISCUSSION
Resulting CCI profile data highlighted significant gaps in the disaster mental health capacity of Victorian providers in view of the three-tiered intervention framework, which have implications for a more strategic approach to future disaster workforce planning and capacity building. Only 32–42% of providers had capacity to deliver current best practice interventions. While not every provider will deliver each intervention, these findings need to be considered suboptimal from disaster preparedness and quality of care perspectives. Key predictors of provider capacity to deliver interventions included previous disaster work experience and direct service roles. Victorian Council of Churches volunteers were significantly more likely to have PFA capacity, reflective of recent concerted efforts to up-skill this provider group, while psychologists and counselors were more likely to have SPR and/or MHT capacity. Bilingual providers were most likely to have SPR capacity, while increasing provider age was uniquely predictive of MHT capacity. Other predictors highlighted key barriers to effective disaster response participation, including the lack of familiarity with disaster mental health interventions and dedicated training. PFA training needs generally indicated lesser, and advanced or other
training needs greater, PFA and SPR capacity. Intervention-specific predictors highlighted the importance of peer support as an enabler of PFA capacity (Creamer et al., 2012); the impact of higher weekly working hours on reduced SPR capacity; and structural barriers and enablers of MHT capacity, including limited role flexibility and involvement opportunities. Key strategies to elevate existing capacity levels of Victorian providers should therefore focus on targeted multilevel training in best practice interventions, creation of practice opportunities, and structural provider support. Disaster mental health training and support programs need to be evidencebased (in content and delivery), scalable, and continuously available for widely dispersed provider groups. There is a need to better utilize existing MHT/SPR providers who are engaged in non-disaster roles. Limitations The study utilized cross-sectional selfreport data and limited organization- or system-level workforce capacity indicators. CCIs benefit from future expert/concept validation. CONCLUSIONS
CCIs focused on best practice interventions provide a methodology for rapid workforce capacity assessment that can facilitate disaster preparedness planning with a view to establishing the conditions for efficient delivery of quality disaster mental health services.
Reifels et al. 205
REFERENCES Australian Centre for Posttraumatic Mental Health. (2007). Australian guidelines for the treatment of adults with acute stress disorder and posttraumatic stress disorder. Melbourne: Australian Centre for Posttraumatic Mental Health. Australian Psychological Society (Producer). (2010). Guidelines for provision of psychological support to people affected by the 2009 Victorian bushfires. Retrieved from http://www.psychology. org.au/Assets/Files/Bushfires-Guidelines-provision-psychological-support.pdf Bryant, R. A., Creamer, M., Forbes, D., Raphael, B., Gordon, R., & Coghlan, A. (2009). Scoping document for mental health response to Victorian bushfires. Discussion paper prepared for Victorian Bushfire Cross-Professional Education and Training Working Group. Creamer, M. C., Varker, T., Bisson, J., Darte, K., Greenberg, N., Lau, W., . . . Forbes, D. (2012). Guidelines for peer support in high-risk organizations: An international consensus study using the delphi method. Journal of Traumatic Stress, 25(2), 134-141. doi:10.1002/jts.21685 Forbes, D., Fletcher, S., Wolfgang, B., Varker, T., Creamer, M., Brymer, M. J., . . . Bryant, R. A. (2010). Practitioner perceptions of Skills for Psychological Recovery: A training program for health practitioners in the aftermath of the Victorian bushfires. Australian and New Zealand Jour-
nal of Psychiatry, 44(12), 1105-1111. doi:10.310 9/00048674.2010.513674 Norris, F. H., Friedman, M. J., & Watson, P. J. (2002). 60,000 disaster victims speak: Part II. Summary and implications of the disaster mental health research. Psychiatry, 65(3), 240-260. doi:10.1521/psyc.65.3.240.20169 Reifels, L., Bassilios, B., Forbes, D., Creamer, M., Wade, D., Coates, S., . . . Pirkis, J. (2013). A systematic approach to building the mental health response capacity of practitioners in a post-disaster context. Advances in Mental Health, 11(3), 246-256. doi:10.5172/jamh.2013.11.3.246 Reifels, L., Naccarella, L., Blashki, G., & Pirkis, J. (2012). The Victorian Disaster Mental Health Workforce Capacity Survey: Final project report. Melbourne: The University of Melbourne. Wade, D., Varker, T., Coates, S., Fitzpatrick, T., Shann, C., & Creamer, M. (2013). A mental health training program for community members following a natural disaster. Disaster Health, 1(1), 9-12. doi:10.4161/dh.22658 Watson, P. J., Brymer, M. J., & Bonanno, G. A. (2011). Postdisaster psychological intervention since 9/11. American Psychologist, 66(6), 482494. doi:10.1037/a0024806
Copyright of Psychiatry: Interpersonal & Biological Processes is the property of Guilford Publications Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
199
Disaster Mental Health Workforce Capacity Reifels et al.
Examining Disaster Mental Health Workforce Capacity Lennart Reifels, Lucio Naccarella, Grant Blashki, and Jane Pirkis
Objective: We examined the capacity of the disaster mental health workforce in Victoria, Australia, to provide the three evidence-supported intervention types of psychological first aid, skills for psychological recovery, and intensive mental health treatments. Method: Utilizing data from a cross-professional, state-level disaster mental health workforce survey (n = 791), we developed composite capacity indicators (CCI) for each intervention and performed logistic regression analyses to examine key predictors of disaster mental health workforce capacity. Results: CCI profiles highlighted significant gaps in the disaster mental health capacity of Victorian providers, with only 32–42% able to deliver current best practice interventions. Key predictors of workforce capacity common and unique to interventions were highlighted. Conclusions: Key strategies to raise Victoria’s disaster mental health workforce capacity should focus on targeted multilevel training in best practice interventions, creation of practice opportunities, and structural provider support/engagement. CCIs focused on best practice interventions provide a methodology for rapid workforce capacity assessment that can facilitate disaster preparedness planning, capacity building, and delivery of quality disaster mental health services. Lennart Reifels, Ph.D., and Jane Pirkis, Ph.D., are affiliated with the Centre for Mental Health, Melbourne School of Population and Global Health, The University of Melbourne. Lucio Naccarella, Ph.D., is affiliated with the Australian Health Workforce Institute, Melbourne School of Population and Global Health, The University of Melbourne. Grant Blashki, M.D., is affiliated with the Nossal Institute for Global Health, The University of Melbourne. This study was funded through the Natural Disaster Resilience Grant Scheme of the Office of the Emergency Services Commissioner in Victoria and the Australian Attorney-General’s Department. The authors wish to thank Rachel Sore, MSc, and Matthew Spittal, Ph.D., for providing valuable statistical advice in relation to the paper, as well as the following agencies which facilitated the conduct of the research: the Victorian Departments of Health and Human Services, the Australian Centre for Posttraumatic Mental Health, the Melbourne Sustainable Society Institute, the Australian Association of Social Workers, the Australian Psychological Society, Occupational Therapy Australia, the Royal Australian & New Zealand College of Psychiatrists, the Australian College of Mental Health Nurses, the Australian Nursing Federation (Victoria Branch), General Practice Victoria, the Australian Counselling Association, the Victorian Council of Churches, Beyondblue: The National Depression and Anxiety Initiative and the Australian Red Cross. An earlier version of this paper was presented at the XIII Conference of the European Society for Traumatic Stress Studies in Bologna, Italy, in June 2013. Address correspondence to Lennart Reifels, Centre for Mental Health, Melbourne School of Population and Global Health, The University of Melbourne, Carlton, 3010 Victoria, Australia. E-mail: [email protected]
© 2014 Washington School of Psychiatry
200
Disaster Mental Health Workforce Capacity
TABLE 1. The Australian Disaster Mental Health Response Framework Psychological First Aid Purpose Population-level support of common distress responses in the immediate aftermath Core Principles:
Skills for Psychological Recovery
Intensive Mental Health Treatments
Purpose Support of individuals with mild to moderate sub-clinical levels of distress
Purpose Treatment of minority experiencing significant mental health conditions
Modules:
Treatments:
Promote safety
Gathering information and prioritizing assistance
Exposure treatments
Promote calming
Problem-solving skills
Cognitive therapy
Promote self-efficacy
Promoting positive activities
Goal setting/activity scheduling
Promote connectedness
Managing reactions
Managing anger
Promote hope
Promoting helpful thinking
Treatment of complicated grief
Promote help
Rebuilding healthy social connections
Managing comorbidity Pharmacotherapy
Providers Disaster relief workers, volunteers, community leaders, generic workers
Providers Primary care providers, including: family physicians, allied health professionals, counselors, welfare staff
Providers Specialist mental health care staff, including: Psychologists, psychiatrists
Note. Adapted from Bryant et al. (2009) and the Australian Psychological Society (2010).
Large-scale natural disasters can have a profound and differential impact on the mental health of affected populations and may therefore require enhanced mental health responses (Norris, Friedman, & Watson, 2002). Vital to the provision of enhanced mental health services is a workforce with the capacity to deliver current best practice interventions. Recent advances in Australian disaster responses involved the implementation of a three-tiered disaster mental health intervention framework in support of affected populations that was underpinned by national expert consensus and first employed in response to Australia’s worst fire disaster, the Victorian Black Saturday Bushfires (Australian Psychological Society, 2010; Bryant et al., 2009). This framework comprised three evidence-supported interventions (Table 1): Psychological First Aid (PFA), a primary psychosocial prevention/support model for common disaster distress responses (Wade et al., 2013); Skills for Psychological Recovery (SPR), an intermediate skill-building/
prevention model for individuals experiencing mild to moderate distress levels (Forbes et al., 2010); and intensive Mental Health Treatments (MHT), evidence-based core interventions for severe mental health issues commonly experienced by a minority of survivors (Australian Centre for Posttraumatic Mental Health, 2007). Consistent with a modern public mental health approach to disaster mental health care (Watson, Brymer, & Bonanno, 2011) and underpinned by a stepped care model, the framework was operationalized through a training agenda that targeted disaster volunteers, allied, general and mental health professionals. Despite these advances, little is known about the capacity of the Victorian disaster mental health workforce to deliver best practice interventions. Moreover, intrinsic obstacles to the efficient and timely delivery of quality disaster mental health services remain. These concern provider adherence to ineffective interventions; the need for provider mobilization/training/accreditation; the influx/management of spontaneous volun-
Reifels et al. 201
teers; and varied practical/logistical barriers to disaster response involvement. One of the key challenges to effective disaster response planning therefore concerns the lack of systematic data on the capacity of the multifaceted workforce which can provide best practice interventions in disaster-affected communities (Reifels et al., 2013). In order to inform future disaster workforce planning, this study constituted the first systematic state-level examination of disaster mental health workforce capacity in Victoria, Australia, and the world. Workforce capacity was primarily examined at a provider level and conceived of as a multidimensional concept centered on two key aspects, namely, the ability of providers to deliver best practice disaster mental health interventions (Table 1) and effectively participate in disaster responses. METHOD
Sample We utilized data from a state-level survey (N = 791) of the Victorian disaster mental health workforce (Reifels, Naccarella, Blashki, & Pirkis, 2012) with the following sample characteristics: mean age 51.7 years (range 22–82, SD 11.7); female (71.6%); bilingual (10.0%); tertiary qualified (82.2%); including psychologists (25.3%), nurses (18.5%), counselors (14.5%), social workers (14.3%), pastoral care workers (13.4%), teachers (6.7%), community workers (3.2%), and other professions (24.5%); in metropolitan (63.1%) or regional Victoria (33.9%); with paid employment (86.4%) and/or volunteering roles (37.4%); primarily involving direct service delivery (78.6%); in work settings including private practice (26.8%), NGOs (21.0%), pastoral care services (16.8%), psychiatric hospitals (6.7%), clinical mental health services (6.6%), general hospitals
(5.3%), schools (4.6%), primary care clinics (4.4%), and other contexts (34.8%). Composite Capacity Indicators Composite workforce capacity indicators (CCI) were developed for each key intervention comprising six equally weighted variables reflective of desirable provider characteristics: 1. Training completion (yes = 1/no = 0) 2. Interest to provide intervention: at least “moderate” (yes = 1/no = 0) 3. Confidence to provide intervention: at least “moderate” (yes = 1/no = 0) 4. Experience providing intervention (yes = 1/no = 0) 5. Capacity to participate in disaster mental health responses: “moderate” (yes = 1/no = 0) for at least one delivery mode: –Face-to-face service delivery –Telemental health –Outreach to affected areas –One-off visits to affected areas –Temporary relocation 6. Understanding of disaster aspects: “reasonable” (yes = 1/no = 0) for at least three disaster aspects: –Impact of disasters on affected individuals –Impact of disasters on affected communities –Mental health consequences of disasters –Disaster preparedness –Disaster recovery –Emergency management –Organizational emergency plan –Local area emergency plan –Regional/state emergency plans
202
Disaster Mental Health Workforce Capacity
CCI summary scores (range 0–6 points) enabled the creation of binary outcome variables that indicated either the presence (threshold 4–6 points) or absence (0–3 points) of capacity to provide each intervention. Data Analysis Exploratory logistic regression analyses performed in SPSS v20 on three binary outcome variables examined predictors of disaster mental health workforce capacity through a process of stepwise-backward elimination (Wald, p < 0.15 entry, p < 0.01 retention). Predictor variables included provider: age, gender, language, professional background, qualification level, organizational affiliation, work location, setting, role, experience, primary activity, work hours, disaster response barriers and enablers. Further details on the range, scales and means of predictor variables are available from the authors. RESULTS
CCI Profiles Mean CCI scores for PFA, SPR and MHT interventions were 3.39 (SD 1.72), 3.07 (SD 1.58) and 2.97 (SD 1.62), respectively. Application of the capacity threshold (4–6 points) demonstrated that only 41.8% of respondents had the required capacity to provide PFA, 31.6% to provide SPR, and 32.0% to provide MHT interventions. Almost one-half (46.5%) lacked capacity to provide any of these interventions, while others had capacity to provide one (19.2%), two (14.5%), or three (19.0%) interventions.
Predictors of Disaster Mental Health Workforce Capacity Logistic regression models in Table 2 outline statistically significant predictors (p < 0.01) of disaster mental health workforce capacity for each intervention, with composite indicators as binary outcome variables. Tests of resulting models against constant-only models were statistically reliable, as indicated for PFA by c² (7, n = 689) = 138.77, p < .001, for SPR c² (10, n = 688) = 176.01, p < .001 and MHT c² (10, n = 690) = 246.10, p < .001. Thus, all three sets of predictors reliably distinguished the presence and absence of capacity at respective levels. Nagelkerke’s R2 of .245 (SPR), .317 (PFA), and .420 (MHT) indicated weak to moderately strong relationships between prediction and grouping. Prediction success overall was 70.5% (SPR), 76.7% (PFA), and 79.3% (MHT). Key predictors of PFA capacity (with odds ratios in brackets) were: organizational affiliation with the Victorian Council of Churches (3.19); direct service provision (1.84); a need for SPR training (1.59); work experience with disaster survivors (1.37); and a need for peer support (1.20). Key predictors of SPR capacity included: bilingual providers (2.28); direct service provision (2.08); psychology background (1.79); the need for other training (1.78); work experience with disaster survivors (1.56); and a need for MHT training (1.34). Key predictors of MHT capacity included: psychology (3.49); or counseling backgrounds (2.45); limited role flexibility (1.81); work experience in mental health (1.58); or with disaster survivors (1.44); older age (1.35); and required involvement opportunities (1.23). Overall effects of organizational affiliation and professional backgrounds were congru-
Reifels et al. 203
TABLE 2. Logistic Regression Models and Predictors of Disaster Mental Health Workforce Capacity for Each Intervention Type OR
95% CI
p-value
Psychological First Aid (PFA) Work experience—Disaster survivors
1.37
1.19, 1.57
0.000
Primary role activity—Direct service provision
1.84
1.18, 2.85
0.007
Organizational affiliation—Victorian Council of Churches
3.19
2.05, 4.95
0.000
Key barrier—Lack of familiarity with DMH interventions
0.42
0.27, 0.65
0.000
Key enabler—PFA training
0.57
0.45, 0.72
0.000
Key enabler—SPR training
1.59
1.25, 2.01
0.000
Key enabler—Peer support
1.20
1.05, 1.36
0.008
Constant
0.17
0.000
Skills for Psychological Recovery (SPR) Professional background—Psychologist
1.79
1.18, 2.71
0.006
Languages other than English
2.28
1.24, 4.17
0.008
Work experience—Disaster survivors
1.56
1.34, 1.82
0.000
Average weekly working hours
0.88
0.82, 0.94
0.000
Primary role activity—Direct service provision
2.08
1.24, 3.50
0.006
Organizational affiliation—Australian Red Cross
0.07
0.02, 0.26
0.000
Key barrier—Lack of dedicated DMH training
0.48
0.31, 0.75
0.001
Key enabler—PFA training
0.73
0.62, 0.86
0.000
Key enabler—MHT training
1.34
1.14, 1.57
0.000
Key enabler—Other training
1.78
1.27, 2.48
0.001
Constant
0.10
0.000
Intensive Mental Health Treatments (MHT) Age
1.35
1.12, 1.62
0.001
Professional background—Psychologist
3.49
2.24, 5.44
0.000
Professional background—Counselor
2.45
1.39, 4.32
0.002
Professional background—Pastoral care worker
0.17
0.06, 0.46
0.000
Work experience—People with mental illness
1.58
1.32, 1.89
0.000
Work experience—Disaster survivors
1.44
1.22, 1.70
0.000
Organizational affiliation—Australian Red Cross
0.05
0.01, 0.37
0.003
Key barrier—Lack of familiarity with DMH interventions
0.46
0.28, 0.76
0.002
Key barrier—Limited scope or flexibility in existing role
1.81
1.21, 2.73
0.004
Key enabler—Opportunities for involvement
1.23
1.07, 1.40
0.003
Constant
0.00
0.000
Note. OR = odds ratio; CI = confidence interval; DMH = disaster mental health. Providers affiliated with the Victorian Council of Churches and the Australian Red Cross were primarily organized disaster response volunteers. Variables are coded as follows: Disaster mental health capacity: 0 = no, 1 = yes; Work experience: 1 = no experience, 2 = a little, 3 = quite experienced, 4 = considerably, 5 = very experienced; Primary role activity: 0 = no, 1 = yes; Organizational affiliation: 0 = no, 1 = yes; Key barrier: 0 = no, 1 = yes; Key enabler: 1 = not a priority, 2 = low priority, 3 = moderate priority, 4 = high priority, 5 = very high priority; Professional background: 0 = no, 1 = yes; Languages other than English: 0 = no, 1 = yes; Average weekly working hours: 1 = 0, 2 = 1–5, 3 = 6–10, 4 = 11–15, … 11 = 46–50, 12 = > 50; Age (in years): 1 = 21–30, 2 = 31–40, … 7 = 81–90.
204
Disaster Mental Health Workforce Capacity
ent with agency core business and anticipated levels of provider involvement. The lack of familiarity with disaster mental health interventions, need for PFA training and high weekly working hours were all associated with decreased capacity. DISCUSSION
Resulting CCI profile data highlighted significant gaps in the disaster mental health capacity of Victorian providers in view of the three-tiered intervention framework, which have implications for a more strategic approach to future disaster workforce planning and capacity building. Only 32–42% of providers had capacity to deliver current best practice interventions. While not every provider will deliver each intervention, these findings need to be considered suboptimal from disaster preparedness and quality of care perspectives. Key predictors of provider capacity to deliver interventions included previous disaster work experience and direct service roles. Victorian Council of Churches volunteers were significantly more likely to have PFA capacity, reflective of recent concerted efforts to up-skill this provider group, while psychologists and counselors were more likely to have SPR and/or MHT capacity. Bilingual providers were most likely to have SPR capacity, while increasing provider age was uniquely predictive of MHT capacity. Other predictors highlighted key barriers to effective disaster response participation, including the lack of familiarity with disaster mental health interventions and dedicated training. PFA training needs generally indicated lesser, and advanced or other
training needs greater, PFA and SPR capacity. Intervention-specific predictors highlighted the importance of peer support as an enabler of PFA capacity (Creamer et al., 2012); the impact of higher weekly working hours on reduced SPR capacity; and structural barriers and enablers of MHT capacity, including limited role flexibility and involvement opportunities. Key strategies to elevate existing capacity levels of Victorian providers should therefore focus on targeted multilevel training in best practice interventions, creation of practice opportunities, and structural provider support. Disaster mental health training and support programs need to be evidencebased (in content and delivery), scalable, and continuously available for widely dispersed provider groups. There is a need to better utilize existing MHT/SPR providers who are engaged in non-disaster roles. Limitations The study utilized cross-sectional selfreport data and limited organization- or system-level workforce capacity indicators. CCIs benefit from future expert/concept validation. CONCLUSIONS
CCIs focused on best practice interventions provide a methodology for rapid workforce capacity assessment that can facilitate disaster preparedness planning with a view to establishing the conditions for efficient delivery of quality disaster mental health services.
Reifels et al. 205
REFERENCES Australian Centre for Posttraumatic Mental Health. (2007). Australian guidelines for the treatment of adults with acute stress disorder and posttraumatic stress disorder. Melbourne: Australian Centre for Posttraumatic Mental Health. Australian Psychological Society (Producer). (2010). Guidelines for provision of psychological support to people affected by the 2009 Victorian bushfires. Retrieved from http://www.psychology. org.au/Assets/Files/Bushfires-Guidelines-provision-psychological-support.pdf Bryant, R. A., Creamer, M., Forbes, D., Raphael, B., Gordon, R., & Coghlan, A. (2009). Scoping document for mental health response to Victorian bushfires. Discussion paper prepared for Victorian Bushfire Cross-Professional Education and Training Working Group. Creamer, M. C., Varker, T., Bisson, J., Darte, K., Greenberg, N., Lau, W., . . . Forbes, D. (2012). Guidelines for peer support in high-risk organizations: An international consensus study using the delphi method. Journal of Traumatic Stress, 25(2), 134-141. doi:10.1002/jts.21685 Forbes, D., Fletcher, S., Wolfgang, B., Varker, T., Creamer, M., Brymer, M. J., . . . Bryant, R. A. (2010). Practitioner perceptions of Skills for Psychological Recovery: A training program for health practitioners in the aftermath of the Victorian bushfires. Australian and New Zealand Jour-
nal of Psychiatry, 44(12), 1105-1111. doi:10.310 9/00048674.2010.513674 Norris, F. H., Friedman, M. J., & Watson, P. J. (2002). 60,000 disaster victims speak: Part II. Summary and implications of the disaster mental health research. Psychiatry, 65(3), 240-260. doi:10.1521/psyc.65.3.240.20169 Reifels, L., Bassilios, B., Forbes, D., Creamer, M., Wade, D., Coates, S., . . . Pirkis, J. (2013). A systematic approach to building the mental health response capacity of practitioners in a post-disaster context. Advances in Mental Health, 11(3), 246-256. doi:10.5172/jamh.2013.11.3.246 Reifels, L., Naccarella, L., Blashki, G., & Pirkis, J. (2012). The Victorian Disaster Mental Health Workforce Capacity Survey: Final project report. Melbourne: The University of Melbourne. Wade, D., Varker, T., Coates, S., Fitzpatrick, T., Shann, C., & Creamer, M. (2013). A mental health training program for community members following a natural disaster. Disaster Health, 1(1), 9-12. doi:10.4161/dh.22658 Watson, P. J., Brymer, M. J., & Bonanno, G. A. (2011). Postdisaster psychological intervention since 9/11. American Psychologist, 66(6), 482494. doi:10.1037/a0024806
Copyright of Psychiatry: Interpersonal & Biological Processes is the property of Guilford Publications Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
