- Slides: 42
Pre-Employment Exams A Spinal Focus Spinal Intrinsic Risk Factors Dr David Mc. Grath 2008 www. drdavidmcgrath. com. au
Questions • What are They ? • How Bad are they ? • Can we find them ? • Should we find them ?
Possible Risk Factors • Current Pain • Old Injuries • Congenital or Acquired Deformities • Anthropometric Extremes • Age • Gender • Level of Fitness • Strength and Robustness • Imaged Pathology
Assessing the Risk (hypothetical non specific risk factor OR=17 ) Risk Present Risk Absent Bad Outcome 5 Good Outcome 5 5 85
Deductions • With a prevalence of 10%, and an absolute risk contribution of ½, we need approximately 100 examinations to reduce, the adverse employment outcomes by 50% (by not employing or effective risk management )
Detection (sensitivity 80%, specificity 77% ) Risk Factor Present Absent Diagnosed 8 20 Not 2 Diagnosed 70
The Risk Adjusted Company (Perfect Intervention=All those identified with risk factor prevented from bad outcome NNT=2) First number indicates employees with risk factor Second number indicates, employees without risk factor Bad Outcome Risk 0+1=1 Diagnosed & Intervention (n=28) No Risk Factor 1+4=5 Diagnosed (n=72) Good Outcome 8+19=27 1+66=67
Risk Intervention Effectiveness (Compared to natural company history with 50% bad outcome) NNT=1/(6/8 -4/8)=4 Those with risk factor Intervention Bad Outcome 2 No 4 Intervention Good Outcome 6 4
Company after Realistic Intervention (NNT=4) Bad Outcome Good Outcome Risk Intervention 2+1=3 (n=28) 6+19=25 No Risk Intervention (n=72) 1+66=67 1+4=5
Non-Hiring Option (28 at-risk diagnosed and not hired, leading to 72 remaining ) (8 with risk factor excluded, 20 non risk employees excluded ) (72 remaining employees, 2 have risk factor, 70 don’t ) 50% of the 2=1 and 5(1/2) % of 70=4 have bad outcome Bad Outcome Good Outcome Screening and Discarding 1+4=5 1+66=67 Without Screening 5+5=10 5+85=90
After Hiring another 28 in 139 exams total (excluding 39 candidates, only 11 of which have the risk factor, while 28 don’t ) Bad Outcome Good Outcome After Screening 1. 5+6=7. 5 No Screening 5+5=10 1. 5+91=92. 5 5+85=90
Comparing Outcomes Bad Outcome Good Outcome No Screening 10 90 Screening & Intervention Screening & Rejection 8 (NNT=4) 6 (perfect Rx) 7. 5 92 94 92. 5
Remember • Odds Ratio=17 (ie relatively high) • Detection se=80% sp=77% (ie good) • NNT=4 Risk Intervention (ie good) • Interventions for 28% (20% no risk) • Exams=139% with rejection strategy • Rejected applicants without risk=28 (20% of • candidates) Rejected applicants with risk=11 (8% candidates, but 10% prevalence)
Summary • With Detection/Risk Intervention strategy, we examine 100% employees, 20% less bad outcome, and intervene on 28% employees (20% of which don’t have risk) • With Detection/Don’t hire policy, we have, 25 % less bad outcome, and examine, 139 % prospective employees, reject 20% with no risk
Things To Consider • Cost of Exams • Stigma to rejected applicants • Cost of interventions • Cost and Significance of Bad Outcomes
Armed Forces • Cost of Training High • Cost of Intervention High • Cost of Bad Outcome High • Cost of Exams relatively Cheap • Stigma of Rejection ? ?
Trial Of Fire • Relax initial entry criteria • Boot Camp survival test (many compo claims arise from this period) • Chance of re-entry if fitness or strength is limiting factor • Not generally available to high end, service skill occupations e. g. pilots, aircrew
Spines • What’s Worth Looking For ? • What’s the best detection method ? • What’s the cost/benefit ? For either the non hire or risk intervene option • What are technical, legal, ethical, social limitations
Recurrent Back Pain • LBP >30 days during past year, increased risk OR=4. 8 long lasting BP OR=3. 3 Leg pain OR=5. 9 Medical Discharge (Hestbaek 2005) n=1711 Danish Military Recruits Conscription • Generally this factor is thought to have good Se, but poor Sp. Also poor Reliability.
Muscle Strength • Strength testing alone has no predictive value for work place injuries (Harbin G 2005) n=2, 482 Food factory study • Likely discrimination against women, certain ethnic groups, and handicapped, using the Detect/Don’t hire policy
Job Matching • With employees who had matching physical capacity, to inherent requirements of job, incidence of injuries was 3% • With employees, without matching capacity, incidence of injuries was 33% NNT=3 in this instance (Harbin G 2005) Factory Workers • 38% physical theoretical mismatch in an industry with a high incidence of LBP (Pedersen DM 2005) Utah Mechanics
Scoliosis etc • Kyphosis/lordosis (side plane) • Skewed pelvis • Scoliosis (frontal plane) • Rotoscoliosis (front, axial) • Sparse reliable, valid data, on occupation outcome measures • OR=3. 0 LBP adolescents (Kovac’s 2005)
Congenital Malformations • Dysraphism (usually detected at birth) • Dysegmentalism (sacralisation, lumbarisation, accessory articulations ) • No valid data
Disc Disease • Isolated Disc Resorption (significant loss of disc height) • Bulges • Prolapses • V. E. P Osteophytic Outgrowth • Sparse, unreliable data
Scheuermann’s Disease • The definition, has relaxed from the original thoracic kyphosis deformity (to a number of vertebral end plate deformities) • No reliable /valid data on risk for various occupations
Isthmic Spondylolitheses • No reliable /valid data on occupational risk • Overall risk perceived to be low
Intervertebral Canal Pathology • Congenital or Acquired Spinal Stenosis • Spinal Cord Pathology • Other Spinal canal Pathology • No Reliable or Valid data for occupations
Muscular Imbalance • Signs of muscular Irritability • Associated geometry deformity • Associated poor dynamic range of specific joints • Low reliability (inter or intra observer ) • No valid data • Assume similar to scoliosis ? (OR=3 LBP)
Body Mass Index • Mild positive association, increasing for longer pain duration • Positive association, unlikely to be causal, as correlation disappears with monozygotic twins (Leboeuf Yde 1999 ) n=29, 424 twins “Probable, weight increase with chronicity”
Co-Morbidity • Positive association of LBP and headache and asthma (Hestbaek L 2006 ) n=9, 567 Age cohort 12 -22 The presence of two other disorders increased the probability of LBP considerably Frail subgroup drops from 60% to 25 % at age 22. “a common origin should be considered”
Smoking • Association between smoking and LBP (Leboeuf-Yde 1999 ) The association is not likely to be causal, as there is no dose response relationship, and disappears with twin study
LBP Adolescents • Strong correlation between adolescent LBP and adult LBP OR=4 • 8 year follow up study • N=10, 000 • Dose Response Relationship (Hestbaek L 1999 Danish Study) “Future Research, should focus on young Population “
LBP Schoolchildren • LBP not related to heredity • Scoliosis related to heredity • Strong association between pain in bed or upon arising and LBP. OR=13 • LBP and Scoliosis OR=3 • LBP and Leg length difference OR=1. 3 • LBP and sport practice more than 2/week OR=1. 2 (Kovacs FM 2005 Spanish cross sectional study n=16, 394)
LBP and Schoolchildren (continued) • No Association for LBP and alcohol intake, cigarette smoking, BMI, book transport method, hours of leisure sitting. • Point prevalence (7 day period) was 17% boys and 33% girls.
Spondylosis • Not valuable as a diagnoses • The population attributable risk is around 15%, as an association with LBP, and thus can be viewed as an intrinsic risk factor.
Gender • Women more at risk of developing chronic LBP OR=2. 65 Military Physical training OR=2. 49 Military Occupation OR=2. 91 Off Duty Activities OR=0. 05 Sporting Activity OR=3. 17 Overall (Strowbridge NR 2005) n=928 new cases English Military Prospective Study
Intelligence and Education • In one study, LBP in military recruits, intelligence protects OR=5, while having parents with high education was slightly negative OR=1. 9 (Hestbaek L 2005)
Multiple Minor Risk Factors • A little bit of this, a little bit of that • Do risk factors, add or compound ? A slightly bad neck and a crook back • No valid studies
What to Do ? • History (detailed or cursory ) • History + Exam + XR (whole spine or Lx) • History +Exam+ XR + Further Imaging • History + Exam + Functional Matching (quantitative or qualitative job/physical characteristics ) • None of the Above
Strategies for Less than Ideal Math's • Identify Risk Factors as an awareness promoting exercise • • and Institute an early reporting system Consider, job matching trial (recall Harbin NNT=3) and/or Aggressive early intervention (secondary prevention) Avoiding, costly intervention on false positives, non effective intervention on true positives, or the stigma of non-hire (true or false positive) Employment exams as an Insurance policy, against fraudulent or excessive claims ?
Things To Re-Consider • Cost of Exams • Stigma to rejected applicants • Cost of interventions (primary or secondary) • Cost and Significance of Bad Outcomes • Potential Role of Exams as Part of Employee Care Program
Questions & Lively Discussion • Are other areas of the body better off ? • How reliable/valid are cardiovascular risk factors ? • How to better integrate intrinsic and extrinsic risk factors ? • Most diseases are multi-risk generated ?