Stepped collaborative care for pain and posttraumatic stress disorder after major trauma: a randomized controlled feasibility trial

Figures & data

Table 1. Risk criteria for PTSD or pain.

Outcome	Risk criteria
PTSD	Three or more: Pre-existing PTSD diagnosis at admission Any other psychiatric diagnosis at admission Any substance use disorder diagnosis or positive blood alcohol content or illicit drug test on admission Current or prior tobacco use Intentional injury Intensive care unit admission with mechanical ventilation* One or more prior trauma hospital admissions A death occurred in the injury event Funding source: compensable injury* Non-male gender Non-white ethnicity Aged 35–65 years* Socioeconomic disadvantage (e.g., unemployed, at risk of job loss, living in a disadvantaged neighborhood [i.e., Index of Relative Socioeconomic Advantage and Disadvantage state decile <5])*
Pain	One or more: Neuropathic injury Prior psychiatric or substance use condition Prior chronic pain Prior treatment for opioid-addiction disorder Intense post-injury or post-surgical pain (e.g., persistent intense pain requiring breakthrough analgesia beyond expected level for injury, acute pain consultations for severe pain, or discharge is delayed due to complex pain) High post-injury opioid consumption (i.e., ≥90 mg/day oral morphine, long acting opioid prescription) High levels of pain-related distress

* These risk criteria were added or adapted based on Australian research of risk factors for poor mental health or pain outcomes after injury.

Table 2. Intervention design, consistent with the TIDieR checklist.

TIDieR domain	Description of the TIDieR component to be documented
The rationale, theory, or goal of the elements essential to the intervention	Pain and PTSD often co-occur after traumatic injury, and there is a need for early interventions targeting both outcomes. To prevent the persistence of mental health symptoms following traumatic injury there is strong evidence of broad population reach and effectiveness of stepped collaborative care case management interventions [14]. However, similar models of care have not been developed to prevent persistent pain, except in the context of post-surgical pain [23–27]. These interventions are effective because they include: Screening to identify people at risk of persistent symptoms and targeted intervention delivery only towards those who are at risk Ongoing assessment to identify when interventions need to be enhanced or optimized (i.e., “stepped” up) Tailoring of strategies and supports to the individual’s needs and preferences while also considering their resources to access or pay for supports in the community Ensuring people have appropriate primary care and allied health supports in the community to provide ongoing care. The goal of the intervention is to prevent persistent pain and mental health symptoms, particularly posttraumatic stress disorder.
Name of intervention	The enhanced Trauma Survivor Outcome Support (TSOS) intervention with integrated elements from the Toronto Transitional Pain Service (TPS)
Intervention materials	The TSOS intervention manual formed the primary basis of the intervention, which is available from the TSOS principal investigator Professor Zatzick. This was enhanced with strategies outlined in the design of the TPS, which is available from the TPS investigators Professors Katz and Clarke. The main therapeutic components of the intervention were: Elements of cognitive behavioral therapy, behavioral activation, relaxation, grounding strategies and psychoeducation The use of evidence-based smartphone applications to help participants to use self-management strategies (e.g., PTSD Coach) A motivational interviewing counselling approach for people with low engagement or risk behaviors, particularly for those engaging in substance use. Advocacy and support to access rehabilitation or therapy in the community, or through the relevant compensation scheme if they had a traffic or work-related injury.
Who provided the intervention	An experienced social worker who received training and weekly supervision in all aspects of the treatment (i.e., the adapted intervention manual for the TSOS intervention [17,18]), and the additional pain management processes in line with the TPS model of care [23] and acute pain management guidelines [29]. The broader clinical team included a supervising psychiatrist, a neuropsychologist, health psychologist and a pain physician.
Mode and location of intervention delivery	Participants were predominantly supported remotely through telephone or video calls. Written or printed resources were sent to participants using email or mail. Face to face case management support was also provided when appropriate and allowed in the context of COVID-19 physical distancing requirements, such as when participants were already scheduled to attend the hospital for outpatient appointments. The intervention was provided individually to the injured person, but could also include outreach to key family members and collaborative care with their healthcare providers in the community. Case management notes were recorded electronically in a REDCap database, which were discussed during case management supervision (weekly) and case conferencing meetings (every 3 weeks).
Duration, intensity and dose of intervention	Case management interactions were initially provided weekly or fortnightly, typically for 30–60 min per contact, depending on the participant’s symptoms, needs and preferences. Over time contacts were reduced in frequency for participants who reported that they no longer experiencing significant problems to the case manager, or in routine follow-up interviews; however, participants could always contact the case manager if they had new issues requiring support. The case manager met with the health psychologist and lead investigator for 30 min every week, and a 1-h case conferencing session was held every 3 weeks.
Tailoring of the intervention	Intervention components were fully tailored to the needs and preferences of each participant. A “stepped” care approach was used where the persistence or escalation of symptoms prompted the delivery of more regular or intensive supports. Decisions about therapeutic strategies or escalation of care were discussed at the routine case conferencing meetings or at ad hoc meetings if needed.
Adverse events	Any adverse events were recorded in case management notes and reported to the ethics committee and other parties, where relevant, to ensure the ongoing safety and wellbeing of participants.

Figure 1. Participant inclusion flow chart.

*Note. Some patients met multiple exclusion criteria, LTFU indicates participants completely lost to follow up from the respective time-point; + exclusion of patients aged 65 years and older injured in low falls was introduced in March 2020.

A CONSORT flow chart that shows the number of trauma patients admitted to hospital who were ineligible and eligible for the study, and the total number who enrolled and completed follow up interviews at 1, 3 and 6-months in the intervention group and control group.

Table 3. Sample characteristics at baseline.

		Intervention group (N = 17)	Enhanced usual care (N = 13)
Demographic, health and prior trauma
Age (years), Med (Q1, Q3)		54 (41, 61)	47 (33, 51)
Age group (years), n (%)	18–34	4 (24)	5 (38)
	35–54	6 (35)	5 (38)
	55+	7 (41)	3 (23)
Gender, n (%)	Male	10 (59)	8 (62)
	Female	7 (41)	5 (38)
Ethnicity, n (%)	Anglo Celtic/Caucasian	11 (73)	8 (73)
	Other (European, Asian, Maori)	4 (27)	3 (27)
Prior Health (EQ VAS), Med (Q1, Q3)		96 (87, 100)	90 (80, 95)
Prior trauma (number), Med (Q1, Q3)		4 (3, 7)	6 (3, 6)
Injury characteristics
Blame someone for the injury at baseline, n (%)		9 (53)	10 (77)
Blame self		6 (40)	4 (44)
Blame another		1 (6.7)	3 (33)
Injury Severity Score, M (SD)		17 (10)	17 (5)
Cause of injury, n (%)	Road trauma & motorized vehicle impacts	7 (41)	7 (54)
	Falls	3 (18)	4 (31)
	Other	2 (15)	7 (41)
Fund, n (%)	Medicare	9 (53)	4 (31)
	Transport Accident Commission	5 (29)	5 (38)
	Workcover	3 (18)	4 (31)
Hospital LOS (days), Med (Q1, Q3)		10 (7, 16)	6 (5, 11)
Risk of developing PTSD or pain
Pain risks, n (%)	One or more risk present	4 (24)	6 (46)
Total PTSD or pain risks, n (%)	<4/36 risks	9 (53)	5 (38)
	≥4/36 risks	8 (47)	8 (62)
Total PTSD risks, n (%)	<3/10 risks	7 (41)	6 (46)
	≥3/10 risks	10 (59)	7 (54)
Demographic risks, n (%)	Middle-aged	11 (65)	8 (62)
	Low SES or job at jeopardy	5 (29)	7 (54)
Baseline symptoms
PTSD (PCL), Med (Q1, Q3)		35 (26, 50)	35 (25, 51)
Depression (PHQ-9), Med (Q1, Q3)		9 (3, 12)	17 (11, 19)
Pain (PEG), M (SD)		6.57 (2.28)	6.49 (2.01)
Suicidal (PHQ-9, Item 9), n (%)	Yes	5 (29)	6 (46)

Abbreviations. BAC: blood alcohol content; LOS: length of stay; M: mean; Med: median; n: number; PCL: PTSD Checklist; PEG: Pain severity; interference with Enjoyment of life and General activity; PHQ: Patient Health Questionnaire; PTSD: Posttraumatic Stress Disorder; Q: quartile; SD: standard deviation; SES: socioeconomic status; VAS: visual analogue scale.

Table 4. Intervention processes, N = 17.

		Number of encounters with process^a
	Number (%) of participants who received each process ≥ once	Mean (SD)	Med [Q1, Q3]
Care coordination
Any care coordination	17 (100.0)	5.80 (4.29)	5 [3, 8]
Primary care provider	6 (35.3)	1.17 (0.41)	1 [1, 1]
Psychology or psychiatry	14 (82.4)	3.57 (2.50)	3 [1, 6]
Rehabilitation	10 (58.8)	1.80 (1.03)	2 [1, 2]
Pain specialist	5 (29.4)	1.40 (0.89	1 [1, 1]
Social services	2 (11.8)	4.00 (4.24)	4 [1, 7]
Addiction services	2 (11.8)	1.50 (0.71)	2 [1, 2]
Medication management
Any medication-related support	16 (94.1)	3.13 (3.01)	2 [1, 3]
Current medications discussed	12 (70.6)	2.17 (1.70)	2 [1, 3]
A change in medications discussed	11 (64.7)	2.36 (2.16)	1 [1, 3]
CBT elements, any
Any CBT elements	14 (82.4)	5.00 (4.33)	4 [1, 7]
Identifying solutions or strategies, including behavior activation	13 (76.5)	3.85 (2.88)	4 [1, 6]
Psychoeducation	7 (41.2)	2.14 (1.86)	1 [1, 3]
Relaxation or grounding	9 (52.9)	2.11 (1.90)	2 [1, 2]
MI elements	5 (29.4)	4.4 (2.70)	4 [3, 6]

^a Only including cases who had one or more session that included the respective element.

Abbreviations. CBT: Cognitive Behavioral Therapy; MI: motivational interviewing.

Figure 2. Willingness for treatment for mental health or pain symptoms, 95% confidence intervals estimated with 100 bootstrapped samples.

A panel of three line charts shows the willingness for treatments at baseline, 1-month, 3-months and 6-months into the study. The charts report the average willingness of participants in the intervention group and control group to see a mental health professional in the first panel, to take mental health medications in panel two and to take pain medications in panel three.

Figure 3. Percentage of participants satisfied or very satisfied with treatment over time.

A panel of three line charts shows participant ratings of their satisfaction with treatment at baseline, 1-month, 3-months and 6-months into the study. The charts report the average satisfaction of participants in the intervention group and control group with overall treatment in the first panel, with mental health treatment in panel two and with pain treatment in panel three.

Figure 4. PCL-C scores for the control and intervention group over time in all participants, and in participants with high baseline PTSD symptoms.

A panel of two line charts shows average PTSD symptoms on the PCL-C at baseline, 1-month, 3-months and 6-months into the study. The chart in the first panel reports data for all participants in the intervention group and control group, and the chart in the second panel reports data for participants in each group that had clinically elevated PTSD symptoms at baseline.

Figure 5. Pain (PEG) scores between the control and intervention group over time in all participants, and participants with higher baseline symptoms.

A panel of two line charts shows average pain scores on the PEG at baseline, 1-month, 3-months and 6-months into the study. The chart in the first panel reports data for all participants in the intervention group and control group, and the chart in the second panel reports data for participants in each group that had clinically elevated pain at baseline.

Giummarra MJ, Lennox A, Dali G, et al. Early psychological interventions for posttraumatic stress, depression and ­anxiety after traumatic injury: a systematic review and meta-analysis. Clin Psychol Rev. 2018;62:11–36.

 PubMed Web of Science ®Google Scholar

Zatzick D, Jurkovich G, Heagerty P, et al. Stepped collaborative care targeting posttraumatic stress disorder symptoms and comorbidity for US trauma care systems: a randomized clinical trial. JAMA Surg. 2021;156(5):430–474.

 PubMed Web of Science ®Google Scholar

Zatzick DF, Russo J, Darnell D, et al. An effectiveness-implementation hybrid trial study protocol targeting posttraumatic stress disorder and comorbidity. Implementation Sci. 2015;11(1):58.

 Google Scholar

Katz J, Weinrib A, Fashler SR, et al. The Toronto General Hospital Transitional Pain Service: development and implementation of a multidisciplinary program to prevent chronic postsurgical pain. J Pain Res. 2015;8:695–702.

 PubMed Web of Science ®Google Scholar

Australian and New Zealand College of Anaesthetists, Faculty of Pain Medicine. Acute pain management: scientific evidence. Melbourne (Australia): ANZCA and FPM; 2015.

 Google Scholar