Figure presents the recruitment flow for the HRC evaluation. A total of 784 ANFs were lodged with NRMA Insurance of which 345 claims met the initial inclusion criteria and agreed to participate in the evaluation. Another 28 participants were excluded for various reasons after the allocation to the two groups (see Figure ). Of the 317 injured people 231 injured people completed the initial survey (73% participation rate) and at 7 months post injury data were obtained from 186 participants. The analyses were carried out on these 186 participants. There was no difference in rate of follow up between the two groups. Participants who dropped out during the trial were younger compared to those who were followed up at 7 months (38 years (SD = 15) versus 45 years (SD = 17), p = 0.01). Twice as many females dropped out at 7 months compared to males (23% versus 13%, F- test p = 0.08). Twice as many participants who were employed (either full time or part time) prior to the accident dropped out at 7 months compared to participants not employed prior to injury (n = 7, 12%, F-test p = 0.06). There was no difference in pre-injury health and reported general health 1 month after injury between those who dropped out and those who were followed up at 7 months.
Demographic and social characteristics of participants in the two groups are shown in Table . Overall there were no differences in baseline characteristics between the two groups.
Participant Characteristics at baseline
Seventy percent of participants were employed pre-injury of which the vast majority were white collar workers. Five percent of these workers were employed in modified duties prior to the injury. This means that these workers performed duties which were modified from the usual job description to accommodate the employee and there could be reduction in hours, other than normal duties, or exclusion of some duties.
Fifty five percent of injuries were whiplash associated disorders (WAD), followed by multiple soft tissue injuries (12.7%). The two groups were similar with regard to injury category, the number of injuries sustained, and severity of injury using the MAIS system. See Table .
At baseline the self-rated average percentage recovery was 58% (SD = 27%), which increased to 78% (SD = 24%) at 7 months follow-up with no difference between the two groups (T-test, p = 0.41). The majority (87%) reported that pain was present at baseline and the average pain level was 4.5 (SD ± 2.6). At 7 months 57% reported pain with average pain level of 2.7 (SD ± 2.8) with no difference between the two groups (T-test, p = 0.92).
Quality of life
Table shows SF-12 norm-based scores for all participants at baseline compared to the Australian general population which has a mean score of 50 and standard deviation of 10. Both groups improved on health scores between baseline and 7 months post injury. We found that there was a significant difference for the domain of perceived health limitation for physical activities between the two groups with the intervention group improving more than the control group over time (p = .033).
SF-12 and HADS scores Mean (SD) health outcomes and change from baseline to 7 months post injury
Anxiety and depressive symptoms
At baseline, anxiety and depression symptoms were prevalent across both teams (Table ). The number of participants who screened positive for anxiety or depression was reduced by 14% by 7 months post injury. This resulted in 31% of participants screening positive for anxiety and 19% for depression. In terms of 'caseness' (any participant with a score greater than 8), a marginally significant difference was found between the teams at 7 months post injury with regard to depression (p = 0.044), with the intervention group having a lower number of participants reporting depression compared to the control group.
Number and percentage of participants with mild, moderate, and severe anxiety and depression at baseline and 7 months post injury
Return to work and usual activities
Eighty two percent of participants who were employed pre-injury (n = 130), had already returned to work at baseline, either in full capacity or modified duties (Table ). At 7 months there were only seven participants (5%) who were still incapacitated. The two groups were similar in their return to work rate. There were a significantly higher number of participants in the control group who reported they had not yet returned to their usual activities at 7 months, compared to the intervention group (F-test, p = 0.005).
Employment status, return to work and return to usual activities
Type of claims
The 186 ANFs were converted to a personal injury claim in 56 cases, and more frequently in the control group (41% versus 25%; F-test, p = 0.04). Forty three personal injury claims included an economic loss component with no difference between the two groups (F-test, p = 0.14). Twenty one percent of the ANFs became legally represented with no difference between the two groups (F-test, p = 0.19). For finalised claims, all except two cases were ANF claims. For open claims the majority (86%) of claims were converted to personal injury claims. Legal representation was correlated to injury severity as measured with the MAIS score from NRMA Insurance (Pearson 0.21, p < 0.001), number of injuries (Pearson 0.44, p < 0.001), level of pain at baseline (Pearson 0.27, p < 0.001), and the level of satisfaction with NRMA Insurance (Pearson -0.18, p = 0.01). The more severe the injury, the more injuries, the higher level of pain and the lower level of satisfaction were related to a higher chance of participants with legal representation.
Predictors of health and return to usual activities
Because almost all participants had returned to work no further analyses related to work were performed. Table shows the variables that were associated with return to usual activities at 7 months. Demographic (age, gender, marital status) and socio-economic factors (level of education, occupational group, pre-injury employment status or satisfaction with employment) were not found to be associated with return to usual activities.
Univariate analysis of predictors for return to usual activities at 7 months post-injury
Stepwise backward conditional logistic regression analysis with return to usual activities at 7 months as dependent variable showed that expected longer duration of return to usual activities (OR 0.42, 95% CI 0.26-0.68), good baseline general health (OR 2.60, 95% CI 1.05-6.45), and having a claim including an economic loss component (OR 0.22, 95% CI 0.06-0.84) explained 39.4% of the variation.
Table shows the variables that were associated with good general health at 7 months (measured with the first question of the SF-12). No association was found for injury severity measured with the MAIS, BMI, smoking or having pain at baseline.
Univariate analysis of predictors for good or excellent general health at 7 months post-injury
Stepwise backward conditional logistic regression analysis with good or excellent general health at 7 months (measured with the first question of the SF-12) as dependent variable showed that baseline general health (OR 4.01, 95% CI 1.57-10.25), together with age (OR 0.96, 95% CI 0.93-0.98), and HADs depression scores (OR 0.86, 95% CI 0.78-0.95) remained significant predictors for general health at 7 months and could predict 84.9% of the cases.