There is a substantial body of literature supporting the view that access to health compensation — notably health care cover and income support — is associated with poor recovery after injury.1-5 While these findings are disturbing, it must be recognised that most of these studies are based on a workers compensation population. It may be that the relationship between work-related injury and health outcomes is mediated by factors other than compensation,6 and these factors may or may not be relevant to other compensation populations.

In one of the few studies to examine the relationship between compensation and recovery in a non-workers-compensation sample, Gabbe and colleagues7 studied a large group of orthopaedic patients in a “no-fault” motor vehicle accident (MVA) compensation scheme. After controlling for differences between groups in factors such as age, injury severity, head injury status, injury group and discharge destination, the study found that compensable patients were more likely than non-compensable patients to report moderate-to-severe disability in both physical and mental health domains at 12 months after injury. Compensable patients were also less likely to have returned to work at 12 months. Thus, the study appears to support the notion that access to compensation results in poorer health outcomes.

Most compensation studies to date are limited by their failure to control for many factors that may affect the relationship between compensation and recovery. These factors include pre-injury psychiatric history, pre-injury disability, prior exposure to traumatic events, and income prior to injury. Moreover, no study to date has examined the stressfulness of interactions with the compensation agency. It may be that stressful interactions with the agency influence the relationship between compensation and health outcomes.

In the state of Victoria, injury patients involved in MVAs are covered by the no-fault compensation scheme of the Transport Accident Commission (TAC). This scheme pays expenses associated with medical treatment, rehabilitation services, disability services, income assistance and household support services.

The aim of our study was to investigate the effect of access to MVA compensation on recovery outcomes at 24 months after injury. We sought to replicate and extend the study by Gabbe et al7 and other studies by:

• Comparing health outcomes of injury patients covered by the Victorian MVA compensation scheme with those of non-compensable patients. We used two different methods for identifying eligibility for compensation: examination of medical records (using a similar method to that of Gabbe et al7), and asking patients at 24-month follow-up to identify any compensation scheme they had accessed;

• Controlling for many factors that could contribute to between-group differences, including pre-injury disability, pre-injury quality of life, prior trauma, psychiatric history, demographic factors, acute psychological response to injury, and injury characteristics; and
• Identifying whether interactions with the compensation agency affected recovery outcomes.

Initially our findings appeared to replicate those of Gabbe and colleagues7 in suggesting that access to a no-fault MVA compensation scheme was associated with poor health outcomes. When we looked at eligibility for compensation based on hospital medical records, we found that patients who were MVA-compensable had poorer mental health outcomes at 24 months than those who were non-compensable, and were also less likely to have returned to their pre-injury work level.

However, further analysis suggested that the relationship between compensation and health outcomes is far more complex. When we examined access to compensation, we found that that the non-compensable group included patients who did access other forms of compensation. When we removed these groups from the non-compensable group, the differences in outcomes between MVA-compensable and non-compensable patients all but disappeared. Owing to the low cell sizes in the “other compensation” groups, we could not analyse their contribution to between-group differences identified in the first set of analyses. However, it would seem that the inclusion, in the non-compensable group, of patients receiving other forms of compensation meant that people with no compensation at all appeared to have better health outcomes than they actually did. At very least, these results suggest that studies examining compensable health outcomes need to have a rigorous methodology and to ensure that any patients classified as non-compensable have not in fact accessed other forms of compensation.

In our study, the MVA-compensable group did have significantly higher levels of anxiety than the non-compensable group, but this appeared to be explained by the level of stress experienced in dealing with the compensation agency. Stress in dealing with compensation agencies can arise from having to undergo numerous assessments, delays in receiving funds, and the often adversarial relationship between client and organisation. Our study suggests that stressful interactions of this kind can affect mental health outcomes and that interactions between claimants and compensation agencies should be as constructive and positive as possible. This may require increased awareness, education and training for claims officers and others who have direct contact with claimants.

Our study provides important baseline data for future compensation studies. It shows that there are differences between compensable and non-compensable groups that need to be considered when conducting similar research. For example, the MVA-compensable patients in our study were more likely to have a history of psychiatric problems and traumatic events than the non-compensable group. As both these factors may predispose patients to developing poor mental health after traumatic events, in not controlling for these variables we may have mistakenly attributed poor mental health outcomes to compensation status rather than prior vulnerability. Future studies exploring the relationship between health outcomes and compensation need to plan carefully to control for these variables.

A limitation of our study was that pre-injury quality of life and pre-injury disability were measured retrospectively (ie, after the participants had been injured). This may have biased their reporting of pre-injury health status. Secondly, the variables we measured may not have captured some important complexities that exist. Thirdly, at 24-month follow-up, patients may have misclassified themselves in terms of their access to compensation, and this may have influenced our results. Finally, the results for MVA-compensable patients covered by a no-fault scheme (as in our study) may not be generalisable to patients covered by other schemes. Furthermore, the participants in our study were adults under the age of 70 years with moderate-to-severe injuries, and the results may not be applicable to all people who have had an MVA (including children and less severely injured patients).

In conclusion, our study does not support earlier findings that access to health compensation is associated with poor recovery after injury. The relationship between access to compensation and health outcomes is highly complex, and studies that aim to investigate this relationship should address this complexity.

1 Flowchart of progression of patients through our study

TBI = traumatic brain injury.