Cheung 2014.

Study characteristics
Methods	Study design: RCT Study grouping: parallel group Unit of randomisation: individuals Power (power sample size calculation, level of power achieved): Based on a small effect size of Cohen’s d = 0.2 for the outcome measures, and an attrition rate of 30% during the 6‐month follow‐up, sample sizes of 259 in each arm could achieve a power of 0.80 to detect a significant difference. A total sample of 518 was needed (Machin 1997); however, interest from the AMS was higher than expected and the total sample size immediately pre‐training (n = 802) was higher than the sample size that was needed Imputation of missing data: no imputation of missing data; per‐protocol analysis (only participants who completed allocated intervention, i.e. without 2 participants in IG who did not complete PFA intervention) and available case analysis (only participants for whom outcomes were obtained)
Participants	Country: China (including Hong Kong) Setting: AMS of the Hong Kong Special Administrative Region Age: mean = 37.38 (SD = 11.78) years Sample size (randomised): 918 Sex: 412 women, 391 men Comorbidity (mean (SD) of respective measures in indicated, if available) at baseline: general psychopathology in GHQ‐28: IG = 0.67 (0.31), CG = 0.60 (0.32); psychological distress in DASS‐21: IG = 0.80 (0.57), CG = 0.76 (0.57); distress from traumatic exposure IES‐R: IG = 0.68 (0.64), CG = 0.60 (0.67); significantly higher depression and anxiety symptoms than normative sample of university students (P < 0.001); significantly lower stress symptoms than normative sample (P < 0.001) Population description: members of AMS of the Hong Kong Special Administrative Region, a government division responsible for providing voluntary supplementary medical and health service in times of community emergency; voluntary first responders of the AMS with and without previous trauma exposure Inclusion criteria: 1) first responders, including fire fighters, police, ambulance officers, rescuers and auxiliary medical personnel; 2) with and without previous trauma exposure (see appendix of the publication, information from trial registration) Exclusion criteria: interested individuals with psychiatric history or current diagnosis of psychiatric disorders (see appendix of the publication, information from trial registration) Attrition (withdrawals and exclusions): between randomisation and pre‐intervention assessment: 116 withdrawals (IG = 63, CG = 53); between pre‐ and post‐intervention (during training): 2 withdrawals (IG = 2); 67 withdrawals between post‐intervention and 3‐month follow‐up (IG = 29; CG = 38); 109 withdrawals between 3‐month follow‐up and 6‐month follow‐up (IG = 45; CG = 64); completion rate for total trial (i.e. from pre‐intervention assessment to 6‐month follow‐up): IG = 80%, CG = 75% Reasons for missing data: not specified
Interventions	Intervention: PFA (n = 458) delivery: face‐to‐face; group sessions; didactic lecture, group discussions, simulation role‐play exercises providers: all sessions conducted by author, who is registered clinical psychologist in Hong Kong and has frontline experience in offering psychological support to disaster survivors duration of treatment period and timing: 1‐day 7‐hour training description: content developed and based on 8 core actions: 1) CONTACT AND ENGAGEMENT: how to approach people in need 2) SAFETY AND COMFORT: emphasises principles of safety and comfort of the individuals and protection of survivors from additional traumatic experiences 3) STABILISATION: describes stabilisation and grounding techniques for calming emotionally overwhelmed survivors 4) INFORMATION GATHERING: gathering of necessary information about the current situations and services available 5) PRACTICAL ASSISTANCE: highlights how to offer practical assistance and discuss with individuals what they can do for themselves 6) CONNECTION WITH SOCIAL SUPPORTS: connect individuals with their social support 7) INFORMATION ON COPING: to some individuals, knowing the normal stress reactions and learning some relaxation skills helped them cope with abnormal situations 8) LINKAGE WITH COLLABORATIVE SERVICES: for severely disturbed people, 8th core action is about referrals and links to existing services in the community for long‐term follow‐up; 3 simulation role play exercises with scenarios relevant to Hong Kong situation to practice core actions; discussion of responder’s self‐care and taking care of each other in the field PART 1 (120 minutes): pre‐programme assessment; welcome and introduction; introduction of PFA knowledge on disaster mental health; PFA core action 1. contact and engagement; PFA core action 2. safety and comfort; PFA core action 3. stabilisation; scenario‐based simulation role‐play exercise: flooding in a fishing village PART 2 (90 minutes): PFA core action 4. information gathering: current needs; PFA core action 5. practical assistance; scenario‐based simulation role‐play exercise: fire disaster happened in a 20‐storey residential building in downtown PART 3 (90 minutes): PFA core action 6. connection with social supports; PFA core action 7. information on coping; PFA core action 8. linkage with collaborative services; PART 4 (120 minutes): self and team care; scenario‐based simulation role‐play exercise: airport disaster; post‐programme assessment compliance: 393/395 completed the intervention, n = 2 dropouts (see flow chart) integrity of delivery: not specified economic information (intervention cost, changes in other costs as result of intervention): not specified Theoretical basis: Chinese translation of the PFA: Field operation guide 2nd edition by National Child Traumatic Stress Network National Center for Post‐traumatic Stress Disorder (Brymer 2006) Control: wait‐list control (n = 460)
Outcomes	Outcomes collected and reported: Primary outcomes: actual helping behaviour ‐ single items for psychological support to people affected in emergency and details of service ‐ not reported (numbers of participants who engaged in providing actual psychological support during time points too small for statistical analyses) self‐efficacy ‐ 13‐item self‐efficacy scale ‐ reported knowledge of PFA and disaster mental health ‐ self‐developed scale ‐ reported Secondary outcomes: general psychopathology ‐ GHQ‐28 psychological distress ‐ DASS‐21 distress from exposure to trauma ‐ IES‐R resilience ‐ Conner‐Davidson Resilience Scale coping with stress (adaptive and maladaptive coping) ‐ Brief Coping Orientation to Problems Experienced life satisfaction ‐ Satisfaction with Life Scale social support ‐ Multidimensional Scale of Perceived Social Support Time points measured and reported: 1) pre‐intervention; 2) post‐intervention (only IN IG); 3) 3‐month follow‐up (3‐months post‐intervention); 6) 6‐month follow‐up (6 months post‐intervention) Adverse events: not specified
Notes	Contact with authors: no correspondence required Study start/end date: April 2011 to November 2011 Funding source: Chinese University of Hong Kong (CUHK) Direct Grant for Research #2009.2.041; Student Research Grant (see trial registration) Declaration of interest: not specified Ethical approval needed/obtained for study: ethics approvals from Survey and Behavioral Research Ethics Committee and Clinical Research Ethics Committee Comments by authors: registered at the CUHK Centre for Clinical Trials, Clinical Trials Registry (CUHK‐CCT00278) Miscellaneous outcomes by the review authors: dissertation Correspondence: Yee Lai Eliza Cheung, School of Public Health and Primary Care, CUHK; eliza.cheung@cuhk.edu.hk; chair: Prof. Yeung Shan Samuel Wong; supervisor: Prof. Ying Yang Emily Chan; The Jockey Cluc School of Public Health and Primary Care, CUHK, Ngan Shing St, Sha Tin, Hongkong; emily.chan@cuhk.edu.hk
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Quote: "Computerized randomization was conducted using SPSS." Quote: "Participants were then randomly assigned to the waitlist control group or the PFA group using random numbers generated by SPSS." Quote: "Means and standard deviations of outcome variables across conditions at Time 1, Time 1b (post‐training), Time 2 (3‐month follow‐up), and Time 3 (6‐month follow‐up) were presented in Table 9. Daggers denote significant differences of the baseline scores between intervention and control group" Quote: "At Time 1, no significant differences were found between intervention and control groups for age, gender, income, occupations, education, marital status, and previous training in post‐disaster psychological interventions" Quote: "Meanwhile, significant difference was found for trauma history between intervention and controls groups, χ²=63.40, p<.001. There was 67.8% of the intervention group reported prior traumatic experience while 39.8% of the control did that. Intervention group reported proportionately more prior traumatic experience than control group despite randomization." Judgement comment: The investigators describe a random component in the sequence generation process (computerised randomisation) and there is verified baseline comparability between groups for most sociodemographic characteristics except for trauma history/prior traumatic experience between groups; no significant baseline differences between groups in most outcome variables (psychological distress, distress from exposure to trauma, maladaptive coping, resilience and social support); however, significant baseline differences in self‐efficacy, knowledge of PFA and disaster mental health, general psychopathology, adaptive coping and life satisfaction
Allocation concealment (selection bias)	Unclear risk	Judgement comment: insufficient information about allocation concealment to permit judgement of ‘Low risk’ or ‘High risk’
Blinding of participants and personnel (performance bias) Subjective outcomes	High risk	Judgement comment: blinding of participants and personnel probably not done (face‐to‐face intervention) and the outcome is likely to be influenced by lack of blinding
Blinding of outcome assessment (detection bias) Subjective outcomes	High risk	Quote: "Investigators and authorized research assistants were responsible for data collection and consolidation. Main researcher who was also the trainer in the Psychological first aid training had access only to the anonymous dataset for further analyses due to the protection of participants’ anonymity." Judgement comment: insufficient information about blinding of outcome assessment (paper‐pencil and online questionnaires; unclear if investigators and research assistants responsible for data collection were blinded); however, due to potential performance bias (no blinding of participants), the review authors judge that the participants' responses to questionnaires may be affected by the lack of blinding (i.e. knowledge and beliefs about intervention they received)
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote: "Figure 4. Algorithm of the randomized controlled study" Quote: "Among the 458 participants assigned to training intervention, 395 of them joined 9 identical training sessions on the training days from April to June 2011. A total of 393 completed pre‐post questionnaires (Appendix VI) were received at Time 1 and Time 1b, 364 participants completed the 3‐month follow‐up questionnaire at Time 2, and 319 at 6‐ month follow‐up at Time 3. Among the waitlist control group, 407 out of 460 filled the questionnaires and 53 withdrew from the study before it starts and 369 participants completed the 3‐month follow‐up questionnaire. At 6‐month follow up, 305 filled in the questionnaire (Figure 4 for the flow of the study). The completion rate of intervention arm is 80% while that of control group is 75%." Judgement comment: reasons for missing data likely to be related to true outcome with imbalance in missing data between groups (over study course: IG: n = 139; CG: n = 155); reasons for missing data in groups not stated; available‐case analysis (only participants for whom outcomes were obtained; see Table 9) and per‐protocol analysis (only participants who completed allocated intervention, i.e. without n = 2 participants in IG who did not complete PFA intervention)
Selective reporting (reporting bias)	Low risk	Judgement comment: trial registration available (registered at the CUHK Centre for Clinical Trials,Clinical Trials Registry (Appendix V); CUHK_CCT00278); all of the study’s prespecified (primary and secondary) outcomes that are of interest in the review have been reported in the prespecified way; only actual psychological support provided/actual helping behaviour was not analysed and could therefore not be reported (too small numbers for statistical analyses)