Early Identification and Intervention to Prevent Disability in Injured Workers Michael Nicholas, PhD, MAPS, FFPMANZCA(hon) Pain Management Research Institute, University of Sydney at Royal North Shore Hospital
Background Australasian work injury data on those who have not returned to work by 6-months postinjury indicate that ~80% of those who do not feel ready to RTW relate this to their injury/pain. (Campbell Research & Consulting. 2005/06 ). Non-RTW by 6-months risks long-term unemployment, disability and suffering What could we do to prevent this?
Australian data (Cohen, Nicholas, Blanch: J Occ Health Safety Aust N.Z, 2000) Medical management of pain from neck/arm and back injuries in first 2 years (Workers Compensation cases, NSW) Three groups identified: (1) RTW with no pain; (2) RTW with pain; (3) no RTW - all in pain No major medical status differences between pain groups (working, or not)
Despite that Pain group not working received 2x amount (same) treatment cf. working pain group Pain group not working had multiple psychosocial factors vs other groups. But, even when noted, psychosocial factors rarely addressed
Today s message is not new
Risk factors for disability after injury Higher pain levels History of similar injuries/pain Poor perceptions of general health Catastrophic beliefs about pain/injury Psychological distress/depression Pain behaviours Number of pain sites Employment status (loss of job) Longer absence from work due to injury/pain Job dissatisfaction Low social support Expectations of delayed recovery/rtw [Main et al., 2008]
Increased risk for poor RTW after injury Biological Personal and environmental Factors (Psychosocial) Environmental (systemic) (Main et al., 2008) Red flags Orange flags Yellow flags Blue flags Black flags Serious pathology Co-morbidity Depression PTSD Unhelpful (eg. avoidant) coping strategies (eg. resting) Emotional distress Passive role in recovery Overly solicitous carers Perceived low social support at wk; Perceived unpleasant work Low job satisfaction Perception of excessive demands Legislative criteria for compensation Nature of workplace (eg. heavy work) Threats to financial security
Fear-Avoidance model for acute/sub-acute pain (Vlaeyen et al., 1995) Injury Recovery Avoidance Depression Disuse Disability Pain experience Graduated Confrontation Pain-related fear/hypervigilance No fear Pain catastrophizing Negative affectivity Threatening illness information
Recent evidence: Sydney primary care study Inception cohort study of 973 patients presenting to primary care with LBP < 2 weeks duration Follow up at 6 weeks, 3 months, and 12 months (< 3% dropout) Sampled three dimensions of recovery: return to work, interference with function due to pain, and pain status Henschke et al. BMJ (2008)
1 year Three pictures of recovery from LBP 1.0 0.8 0.6 0.4 0.2 1.0 0.8 0.6 0.4 0.2 1.0 0.8 0.6 0.4 0.2 Cumulative probability of reduced work status 6 weeks 3 months 6 months 9 months 1 year 6 weeks 3 months 6 months 9 months 1 year 6 weeks 3 months 6 months 9 months Cumulative probability of still having disability Cumulative probability of still having pain Normal work status No disability Pain-free
Yellow flags (adverse psychosocial prognostic factors) Older age More intense pain Longer duration of low back pain More days of reduced activity Patient reports feeling depressed Patient believes pain is likely to persist Compensable low back pain Henschke et al. (2008)
1 year 9 months 6 months 3 months 6 weeks 1.0 0.8 0.6 0.4 0.2 6 yellow flags 5 yellow flags 4 yellow flags 3 yellow flags 2 yellow flags 1 yellow flags 0 yellow flags Cumulative probability of remaining unrecovered
Concepts of risk in injury management Increased chances of poor outcomes (eg. prolonged disability and delayed return to work) (Main, 2002), Opportunity for secondary prevention But, we need to bear in mind: Level of analysis: individual vs population levels Uncertainty over how the presence of multiple risk factors might interact (eg. threshold levels versus cumulative impact) Unmeasured factors (can t measure everything)
Other contributors to outcomes The efficacy (and quality) of the treatment Social, economic and occupational factors (Blue and Black flags) (see Shaw, Linton et al. J Occ Rehab 2009). Note also: many individuals with persisting, work-related pain return to work with little or no help and in the face of personal and workplace variables that seem to be major obstacles to others (eg. Pransky et al., 2005). The mere presence of risk factors cannot be assumed to imply poor outcomes are inevitable at the individual level. But they do increase the chances of problems and if no action is taken, the risks could increase
Screening for modifiable risk factors balance effort and value Screening should be directed towards identifying potentially modifiable risk factors. It should be expected to enable those performing the screening to screen in (those who merit further investigation) and to screen out (those who do not need further investigation at this stage). By its pragmatic nature, screening is more likely to occur when it is relatively brief and requires few resources. This requires a balance to be sought between effort and value.
Practical considerations Timing - Different factors may be important at different stages (eg. initiation, first onset, continuation and consequences of disease or illness) (Von Korff et al., 2002). Timing early means small number amongst many so harder than later (many cases and easier but too late?) Over vs under-inclusive screening (better to capture almost all than miss some) Which measures to use? Which methods (questionnaire, file check, interview)?
Current examples ACC (NZ) screening questions ASAP, then f/u interview to check details if indicated as in at risk range Workcover (NSW) screening at 4-6 weeks if not progressing (left to providers to do) NRMA (NSW) screening by telephone ASAP, if risk factors action plan (insurer initiated)
In Victoria: VWA Alignment of Clinical Framework to Psychology Key Concepts September 2005
In New Zealand Pain Management Services Agreement Operational Guidelines April 2008
In NSW (2008)
Örebro Musculoskeletal Pain Screening Questionnaire (ÖMPSQ) (Linton and Halldèn,1998) 25 item, self-administered questionnaire 21 items rated on a 0-10 scale Items cover emotional state, fears, beliefs, coping strategies, expectations about RTW, job satisfaction, current activity levels. Scores range 0-210
Short Örebro Musculoskeletal Pain Screening Questionnaire (SÖMPSQ) (Linton, Nicholas et al., in review)
In occupational sample, >12 days sick leave Receiver Operating Characteristic (ROC) Kumla 100 80 Sensitivity 60 40 20 ÖMPSQ ÖMPSQ short form 0 0 20 40 60 80 100 100-Specificity
A recent Sydney pilot study Dr Garry Pearce (Occ Physician) and A/Prof Michael Nicholas presented the results at the joint meeting of Faculties of Occ and Rehab Medicine in Adelaide (May, 2008)
Early use of ÖMPSQ at Concord Hospital, NSW (Pearce, McGarity, Nicholas, Linton, Peat, 2008) Two year study with hospital employees making injury claims Modified ÖMPSQ: 13 item scale ÖMPSQ given when claim submitted (ie. generally within 48 hrs of injury) 2 phases in study: Phase 1: usual care, ÖMPSQ data not examined until RTW Three groups identified high, medium, low scorers High scorers reporting more pain, more distress, expectations of delayed RTW Phase 2: Additional interventions offered to high score (high risk) group Costs obtained from insurer (for each case in both phases)
Preliminary cost findings with Concord ÖMPSQ study Costs, from insurer, when claims closed (~ 1 yr). ÖMPSQ scores (at time of claim) Ave. cost of claims (at closure) Low $4,878 Medium $6,240 High $17,178 Key points: 1) Psychosocial factors present at time of claim 2) Psychological sequelae are treatable..
Intervention (phase 2 of Concord study) High Risk (scores >85) * Independent Rehabilitation Provider within 2 weeks * Clinical Psychological assessment and treatment within 2 3 weeks. * Independent Medical Assessment within 1 month * Independent Physiotherapy Assessment after 6 weeks. * File review by Rehabilitation Medical Specialist if not returned to work within 4 weeks Medium risk (70 84) Usual care + clinical psychologist Low risk (<69) Usual care
RESULTS: Comparison between Control and Intervention Cohorts CONTROL GROUP INTERVENT GROUP CONTROL GROUP INTERVENT GROUP RISK CATEGORY % % $ COST $ COST LOW 47 51 4,878 4,898 MEDIUM 31 29 6,240 6,752 HIGH 22 19 17,178 12,847 [A 25% saving]
Selecting those with high psychosocial risk factors in subacute pain and intervening before chronicity sets in (controlled studies from 2000) Study Intervention & Outcomes (bold) Comment Linton & Andersson, 2000 Loisel et al., 2002 Van den Hout et al. 2003 Gatchel, et al. 2003 Linton et al., 2005 Schiltenwolf et al., 2005 6 x 2-hr grp sessions with Clin. Psychologist + Rehab > Information + Rehab (on lost time from work) All interventions achieved gains, but comprehensive Sherbrooke model (combined occupational and clinical interventions) had fewer days on benefits. (RTW) + Graded activities (behavioural principles) + problem-solving training > Graded activities + education (on longer-term work status) + high risk acute patients in functional restoration group (CBT approach) >a treatment-as-usual group. (on indices of disability; work, healthcare utilization, medication use and self-reported pain). + CBT grp = CBT + exercise grp >> minimal tmt grp (examination, reassurance, advice on activities). (lost time) + The addition of the behavioural therapy for dealing with stress and problems generally seems to have added significantly to exercise/activity program. (Lost time) + + Verbeek et al., 2002 Jelema et al., 2005 Hay et al., 2005 Many similarities in content of control grp and treatment grp. No difference between grps on disability & RTW outcome (both improved). Psychosocial intervention = standard care (both by GP only) (on disability) CBT (pain management) and manual therapy (+ home exercise) achieved similar results (disability) (-) Low distress in both groups (-) Low level of psychosocial risk factors at baseline (-) Low distress initially Sullivan et al., 2006 Psychosocial risk factors reduced in both groups (Physio + CBT vs Physio only), but catastrophizing reduced more in combined group. Combined group had better RTW 4-wks after end of treatment. Reduced catastrophizing associated with better outcomes
Interventions based on identifying specific risk/prognostic factors Sullivan et al. Phys Ther. 2006; 86(1):8-18. Whiplash patients (6 wks off work) selected on basis of pronounced psychosocial risk profiles (Scores > 50 th percentile on at least 1 scale, compared to normative dataset) Catastrophizing. The Pain Catastrophizing Scale (PCS). 50 th percentile: 20. Fear of movement The Tampa Scale for Kinesophobia (TSK). or reinjury. 50 th percentile: 39. Perceived disability. The Pain Disability Index (PDI). 50 th percentile: 37.
Implications When psychosocial risk/prognostic factors low, usual care is sufficient (Usual care seems effective in uncomplicated cases of LBP Jallema et al. Pain 2006) When psychosocial risk/prognostic factors high, interventions targeting these aspects often more effective than usual care
All injuries and treatments occur in a context
Franche et al. (2005) J. Occ. Rehab Workplace intervention strategies Strength of Evidence for (less) Work loss Early contact with the worker by the workplace Moderate Work accommodation offer Strong Contact between healthcare provider Strong and the workplace RTW coordination Moderate Super-numerary replacements Insufficient
What about the insurer? Could the beliefs held by Claims and IMA (injury management advisers) staff be important? What do they believe?
A reduction in pain is necessary before a person can start resuming normal functioning. 40% 35% 30% 25% 20% 15% Insurance company claims managers & IMAs 10% 5% 0% Totally Agree - 0 1 2 3 4 Totally Disagree - 5
A reduction in pain is necessary before a person can start resuming normal functioning. Before & After: WA Workers Comp Conference, 2007 1. Strongly Agree 2. Agree 3. Somewhat Agree 4. Neutral 5. Somewhat Disagree 6. Disagree 7. Strongly Disagree 1 2 3 4 5 6 7 Before After - 2% 0% 1% 1% 3% 19% 74% 2% 6% 10% 1% 12% 38% 31%
What if the community accepted the idea that being active despite pain was OK? Buchbinder et al. Spine 2001;26:2535 2542 Population-based, state-wide public health intervention to alter beliefs about back pain and its medical management. N = 4730 interviewed 2.5 yrs apart; 2556 GPs interviewed 2 yrs apart. 1 state (Victoria) = intervention, another state (NSW) = control
Buchbinder et al, BMJ, 2003
General Practitioners behaviour Derived from responses to a case study with sub-acute LBP presented by Buchbinder et al. Response No tests ordered Prescription of bed rest Advice on exercise Advice on work modification Victorian Drs (vs NSW) More likely not to order tests Less likely to support bed rest More likely to support exercise More likely to advise change
Buchbinder s Findings In Victoria: Decline in claims for back pain, rates of days off, and costs of medical management (~ $65m) In NSW: No change Conclusion: Changing community s beliefs about back pain led to change in behaviour of patients and the behaviour of their GPs.
What about therapists? Biomedically orientated therapists tend to see activity despite pain as potentially harmful and to recommend avoidance of such activities. In contrast, biopsychosocially oriented therapists tend to see activity despite pain as helpful and to encourage activity and RTW
Fear-Avoidance model for acute/sub-acute pain (Vlaeyen et al., 1995) Injury Recovery Avoidance Depression Disuse Disability Pain experience Graduated Confrontation Pain-related fear/hypervigilance Treatment Provider factor No fear Pain catastrophizing Negative affectivity Threatening illness information
Treatment of injured workers needs to consider contextual contributors
Choice for the injured worker? Pain persisting, no major pathology Seek return to normal activities, despite pain Pain management Seek pain relief first. Avoid painful activities Acute, symptom management
In sum Most injured workers recover and RTW Those at risk of delayed RTW can often be identified very soon after injury Interventions aimed at helping this at risk group are available and can be effective But there are obstacles to negotiate An informed, collaborative approach is advised