Use of sensorineural tests in a large volume of medico-legal compensation claims for HAVS

BACKGROUND: The Department of Trade and Industry set up a contract to examine miners and ex-miners claiming compensation for hand-arm vibration syndrome (HAVS). They had been exposed to hand-transmitted vibration. Over 100,000 have now been examined using sensorineural tests as part of that process. AIMS: The purpose of this paper was to examine the internal consistency of the vibrotactile threshold test (VTT) and the thermal aesthesiometry test (TA). METHODS: In 18 centres across the UK, nurses were trained to perform the measurement of VTT and TA in a controlled standardized manner. These tests were to aid the staging of the neurological component of the Stockholm Workshop Scales. The staging of this component was modified by dividing stage 2SN into 2SN (early) and 2SN (late). The test results and an automatic neurological staging were presented to the examining doctor following the clinical examination. The results of these sensorineural tests were held on a central database. The results and analysis of the first 57,000 tests are reported. RESULTS: The correlations within the VTT and TA scores were consistent with reliable measures. However, these correlations were not of such strength as to allow reliance on the results of a single test when making an assessment of the severity of neurological damage. CONCLUSIONS: Different end organs and nerve fibres should be tested when making an assessment of damage in the sensorineural component of HAVS. The correlations demonstrated for the VTT and TA suggest that they are of value in assessing these claimants and would be for other vibration exposed workers.     

The diagnostic value of finger systolic blood pressure and cold-provocation testing for the vascular component of hand-arm vibration syndrome in health surveillance

BACKGROUND: Hand-arm vibration syndrome (HAVS) is a complex condition with vascular, sensorineural and musculoskeletal components. A number of quantitative tests have been used for assisting in the diagnosis of HAVS and grading disease severity. AIMS: To investigate and compare the diagnostic value of finger systolic blood pressure (FSBP) and rewarming of finger skin temperature (FST) following cold-provocation testing, in the assessment of vascular HAVS. METHODS: Twenty-four individuals with vascular HAVS (Stockholm Workshop stage 2 or 3V) and 22 control subjects underwent FSBP measurements at 30, 15 and 10 degrees C and monitoring of FST following immersion of the hands in water at 15 degrees C for 5 min. RESULTS: There was a significant reduction in median FSBP% in the vascular HAVS group in the change in FSBP from 30 to 15 degrees C adjusted for brachial blood pressure (FSBPC%). There was no difference in the median time for FST to rewarm by 4 degrees C between HAVS cases and controls. The sensitivity and specificity of FSBP to discriminate between the groups varied between 44 and 61% and 91 and 95%, respectively. The sensitivity and specificity for the time for FST to rewarm by 4 degrees C were 71 and 77%. CONCLUSIONS: There is little evidence that the described form of finger rewarming after cold-provocation testing is a useful diagnostic test for vascular HAVS, although it may have some moderate influence in ruling out vascular problems. Based on our data, the FSBP may also have limited use in confirming a positive diagnosis of vibration-induced vascular problems. The higher specificity of the FSBP test suggests it may have some value in ruling out the vascular component of HAVS. The data from this study do not confirm the diagnostic power of FSBP for the vascular component of HAVS reported by a few other investigators.     

Hand-arm vibration syndrome in South African gold miners

BACKGROUND: Hand-arm vibration syndrome (HAVS) is associated with the use of hand-held vibrating tools. Affected workers may experience symptoms of tingling, numbness, loss of grip strength and pain. Loss of dexterity may impair everyday activities, and potentially increase the risk of occupational accidents. Although high vibration levels (up to 31 m/s(2)) have been measured in association with rock drills, HAVS has not been scientifically evaluated in the South African mining industry. AIMS: The aim of this study was to determine the prevalence and severity of HAVS in South African gold miners, and to identify the tools responsible. METHODS: A cross-sectional study was conducted in a single South African gold-mine. Participants were randomly selected from mineworkers returning from annual leave, comprising 156 subjects with occupational exposure to vibration, and 140 workers with no exposure. Miners who consented to participate underwent a clinical HAVS assessment following the UK Health and Safety Laboratory protocol. RESULTS: The prevalence of HAVS in vibration-exposed gold miners was 15%, with a mean latent period of 5.6 years. Among the non-exposed comparison group, 5% had signs and symptoms indistinguishable from HAVS. This difference was statistically significant (P < 0.05). All the cases of HAVS gave a history of exposure to rock drills. CONCLUSIONS: The study has diagnosed the first cases of HAVS in the South African mining industry. The prevalence of HAVS was lower than expected, and possible explanations for this may include a survivor population, and lack of vascular symptom reporting due to warm-ambient temperatures.     

Hand-arm vibration syndrome with concomitant arterial thrombosis in the hands

BACKGROUND: Hand-arm vibration syndrome (HAVS) refers to the vascular, neurological and musculoskeletal effects that may occur in workers with prolonged exposure to vibrating tools. Hypothenar hammer syndrome (HHS) is a lesion of the ulnar artery at the level of the hamate bone secondary to single or repeated episodes of trauma to the hypothenar eminence. The literature suggests that digital arterial thrombosis and HHS may be associated with the use of vibrating tools. AIM: This study will familiarize investigators with the range of vascular abnormalities seen in workers using vibrating tools, and highlight the importance of screening for arterial thrombosis in the hand when assessing hand-arm vibration-exposed patients. METHODS: In the patients referred to our clinic for HAVS assessment, three were identified during the period 2001 to 2004 who had vascular occlusions in the hands in addition to HAVS. In addition to standardized HAVS vascular investigations, all three patients had arteriograms based on a significantly positive Allen's test. RESULTS: All three cases had documented HAVS based on vascular testing. Arteriograms revealed a spectrum of severity of arterial thromboses from severe HHS, to occlusion limited to the digital arteries. CONCLUSION: Our study reports three cases of HAVS with concomitant HHS and/or digital artery thrombosis. These findings support previous reports of an association between HAVS and vascular thrombosis in the hands. Screening for arterial occlusive problems in the hands should be included in the HAVS work up.     

Upper limb disability in HAVS cases--how does it relate to the neurosensory or vascular elements of HAVS?

AIMS: Hand-arm vibration syndrome (HAVS) consists of vascular, neurosensory and musculoskeletal components, characterized by symptoms that include Raynaud's phenomenon, tingling and numbness in the hands. However, there has been little published data on the effects of HAVS on the capability to carry out normal daily tasks. We have investigated the application of the widely-used disability, arm, shoulder and hand (DASH) disability questionnaire that reflects functionality problems in the upper extremities, as well as symptoms, in a HAVS cohort. METHODS: The cohort consisted of 118 males who, as a part of their health surveillance, had been referred for further assessment and Stockholm workshop staging. This process involved medical interview, physical examination and quantitative tests covering neurosensory function, manual dexterity and handgrip strength. RESULTS: The relationship between DASH outcome metric and a combination of quantitative tests reflecting a range of abnormalities found in HAVS, supports the validity of this questionnaire in HAVS studies. The data suggest that HAVS cases have a greater level of upper extremity disability compared with a general population. The study confirms that disability in HAVS is very largely related to sensorineural Stockholm workshop staging, rather than vascular staging. Any influence of vibration-induced Raynaud's phenomenon on upper extremity disability is related to the frequency of blanching attacks rather than their extent across the digits. CONCLUSION: This study strengthens the importance of identifying and preventing the exacerbation of the neurosenory component of HAVS, that unlike the blanching attacks of the vascular component does not have such an obvious pathognomic signal.     

A comparison between two methods of aesthesiometric assessment in patients with hand-arm vibration syndrome.

Hand-arm vibration syndrome (HAVS) is caused by prolonged exposure to vibration. The diagnosis and assessment of disease severity are subjective at present. The aim of this study was to determine sensorineural dysfunction in patients with HAVS using two methods of aesthesiometric assessment. We recruited three groups of age-matched subjects: 20 subjects diagnosed as having HAVS, 15 manual workers and 15 sedentary workers. We measured both two-point discrimination and depth sense perception using an aesthesiometer. We found that the two-point discrimination wheel was more accurate than the depth sense perception wheel at detecting levels of sensorineural dysfunction in subjects with HAVS. The increased sensitivity of the two-point disc would suggest that it should be used in preference to the depth sense disc for the assessment of sensorineural dysfunction in patients with HAVS.     

A screening questionnaire for HAVS?

AIM: To define the diagnostic power of simple questions most applicable for a hand-arm vibration syndrome (HAVS) assessment screening questionnaire. METHOD: Using a binary logistic regression we analysed 365 physician led HAVS health surveillance assessments to identify which questions could form the basis of a screening questionnaire. RESULTS: Four sensorineural related questions regarding tingling and numbness in response to the cold and after using vibrating tools, and two vascular-related questions focusing on the patient's fingers going white on exposure to cold and numbness during an attack of whiteness were identified. CONCLUSIONS: Questions of high sensitivity for screening subjects for the vascular and neurosensory components of HAVS were identified, which can be used to identify those requiring further clinical investigation and functional testing.     

Assessment of the hand-arm vibration syndrome: thermometry, plethysmography and the Stockholm Workshop Scale

BACKGROUND: The Stockholm Workshop Scale (SWS) provides a staging scheme for hand-arm vibration syndrome (HAVS) based on subjective history. Cold provocation finger thermometry and plethysmography are commonly used objective tests for the vascular component of HAVS. AIM: To examine the correlation between the cold provocation tests and SWS vascular stage. A secondary goal was to evaluate the correlation between cold provocation finger plethysmography and thermometry testing. METHODS: Patients investigated for HAVS at St Michael's Hospital, Toronto, Ontario, were subjected to the same protocol including a questionnaire, clinical assessment and objective testing. Spearman correlation coefficients were calculated for the vascular tests with the SWS and for the vascular tests themselves. Logistic regression models controlled for age, smoking, use of vasoactive medications and time since last vibration exposure. RESULTS: One hundred and thirty-nine patients investigated for HAVS consented to participate in the study. The correlation coefficients for plethysmography (rho = 0.14) and thermometry (rho = 0.18) with the SWS were not statistically significant. Plethysmography and thermometry results were significantly correlated (rho = 0.47, P < 0.001). Logistic regression showed plethysmography and thermometry to weakly predict SWS vascular stage (OR 1.5 and 1.3, respectively). None of the potential confounders had a significant effect in the models. CONCLUSION: The results of plethysmography and thermometry did not significantly correlate with SWS vascular stage in this study. The objective tests did correlate with each other, suggesting that they are reliable measures of similar phenomena likely related to underlying vascular pathology.     

Accidental blood and body fluid exposure among doctors

Aim To study the epidemiology and time trends of blood and bodyfluids (BBF) exposures among hospital doctors. Methods A 3-year study was carried out using data from the ExposurePrevention Information Network of four teaching hospitals inthe UK. Results One hundred and seventy-five cases of BBF exposuresin doctors were reported over the 3-year study period. Eighty-one(46%) occurred in senior doctors and 94 (54%) in junior doctors.Junior doctors had a higher rate of BBF exposures compared tosenior doctors: 13 versus 4 incidents per 100 person-years,respectively (relative risk 3, 95% confidence interval 2–4).The most frequent setting for BBF exposures among senior doctorswas the operating theatre/recovery (59%). Among junior doctors,it was the patient room (48%). The commonest original reasonfor use of sharps by junior doctors was the taking of bloodsamples (42%). Among senior doctors, it was suturing (41%). Conclusion While ongoing training efforts need to be directedtowards both junior and senior doctors, our data suggest thatjunior doctors are at higher risk of BBF exposures and may needparticular attention in prevention strategies. An improvementin the safety culture in teaching hospitals can be expectedto reduce the number of BBF exposures.     

Diagnostic criteria and staging of hand-arm vibration syndrome in the United Kingdom

In the United Kingdom the diagnosis of Hand-arm Vibration Syndrome varies depending on the purpose of that diagnosis. The criteria differ in three situations. More than 100,000 miners and ex-miners with claims for HAVS have been examined using a Medical Assessment Process which included the use of standardised tests. This contract is unique but it has had significant effects on the two other processes. The Industrial Injuries Disablement Benefit Scheme provides a benefit that can be paid to an employed earner because of an accident or Prescribed Disease. New recommendations have been published to remove the anomalies in the present format for assessing HAVS. If implemented the new scheme will recognise the Stockholm Workshop Scales and workers with neurological problems will also be compensated. The Health and Safety Executive will issue new guidance in the near future on the hazards of hand-arm vibration. Health surveillance in the workplace will be fundamental and the HSE propose a tiered approach with levels 1 to 5. Specialist occupational nurses and doctors with training in the diagnosis and assessment of HAVS will be needed for levels 3 and 4. Only at this level may a diagnosis of HAVS be made. The Medical Assessment Process has demonstrated that it is possible to examine a large number of claimants in a standardised manner. The IIAC and HSE recommendations contain very important improvements on the existing positions in the UK and it must be hoped that they will be implemented in the near future.     

A critique of a UK standardized test of finger rewarming after cold provocation in the diagnosis and staging of hand-arm vibration syndrome

BACKGROUND: Accurate diagnosis and staging of hand-arm vibration syndrome (HAVS) is important in health surveillance of vibration-exposed workers and the substantial number of related medico-legal cases. The measurement of the rewarming rate of fingers after cold provocation to the hands (CPT) has been suggested as a useful test in diagnosing HAVS. AIM: To investigate the diagnostic value of a standardized version of the CPT test using a 15 degrees C cold challenge for 5 min applied in the recent compensation assessment of UK miners. METHODS: Analysis of a subset of UK miners assessed at our unit, together with data from a small repeatability study of the standardized CPT in normal subjects. RESULTS: Rewarming time in the CPT was significantly lower in those subjects classified as vascular Stockholm stage 0 compared with Stockholm stages 1-3 combined, but did not discriminate between the stages of abnormality. Using the suggested cut-off in the CPT test, the sensitivity and specificity were calculated as 43 and 78%, respectively. Receiver operator characteristic analysis suggested that the rewarming time of highest accuracy gave a sensitivity of 66% and specificity of 59%. In 10 miners who reported unilateral hand blanching, there was no significant difference in CPT measurements between blanching and non-blanching hands. Repeat CPT measurements in normal subjects suggested mean differences of 52 and 107 s for each hand, and the Bland-Altman coefficient of repeatability was approximately 600 s for all fingers. CONCLUSION: Single application of this standardized CPT test may have limited value in diagnosing the vascular component of HAVS in an individual.     

Mortality of UK oil refinery and petroleum distribution workers, 1951-2003

BACKGROUND: The health of UK petroleum industry workers has been monitored for many years. AIM: To identify any long-term adverse health outcomes from occupational exposures in this industry. METHODS: The mortality (1951-2003) and cancer morbidity (1971-2003) experienced by cohorts of 28,555 oil refinery workers and 16,477 petroleum distribution workers has been investigated. Study subjects were all those males first employed in the period 1946-74 at one of eight UK oil refineries or 476 UK petroleum distribution centres; all subjects had a minimum of 12 months employment with some employment after 1 January 1951. Observed numbers of cause-specific deaths and site-specific cancer registrations were compared with expectations based on national mortality and cancer incidence rates. RESULTS: Standardized mortality ratios (SMRs) were significantly <100 for all causes both in oil refinery workers (Obs 11,156, SMR 89) and in petroleum distribution workers (Obs 7320, SMR 96). Significantly elevated SMRs were shown in oil refinery workers for cancer of the pleura (mesothelioma) (Obs 64, SMR 261) and melanoma (Obs 48, SMR 168). Significantly elevated SMRs were not found in petroleum distribution workers for any site of cancer. Significantly elevated standardized registration ratios (SRRs) were only shown in oil refinery workers and for cancer of the pleura (mesothelioma) (Obs 115, SMR 274), melanoma (Obs 85, SMR 129) and other skin cancer (Obs 983, SRR 117). CONCLUSIONS: The only findings that showed clear evidence of an occupational cancer hazard were those for mesothelioma in oil refinery workers.     

Improving the quality of peak flow measurements for the diagnosis of occupational asthma

INTRODUCTION: Serial measurements of peak expiratory flow (PEF) are recommended in the evidence-based review list as the first stage in objective confirmation of occupational asthma. Different centres have reported widely different success in obtaining records of sufficient data quantity for diagnosis. We investigated different methods of instruction and determined the return rate and quality of the resulting record for the diagnosis of occupational asthma. METHODS: Consecutive new referrals were recruited from a specialized occupational lung disease clinic and requested to carry out serial PEFs for the assessment of suspected occupational asthma. Requests to carry out the records were either from written postal instructions or personal instruction from a PEF specialist. Record quality received from other clinicians was also analysed separating those using dedicated occupational forms, and those submitting on graph type forms. RESULTS: The postal return rate was 56% and the personal rate 85%. The number of records fulfilling all the data quality criteria were similar in the postal and personal groups (55 and 59%, respectively). Pre-existing records from other clinics plotted from graph charts (fulfilling all criteria) were only adequate in 23%, compared with 61% adequate for pre-existing records plotted from occupational forms. Failure of the record to contain consecutive work periods of > or =3 workdays was the most common failure. CONCLUSION: The return rate of PEFs for diagnosing occupational asthma is better when patients have been given specific instructions from a PEF specialist and the data quantity better when recorded on a dedicated form.     

The analysis of sensitivity, specificity, positive predictive value and negative predictive value of cold provocation thermography in the objective diagnosis of the hand-arm vibration syndrome

The diagnosis of digital artery vasospasm in the hand-arm vibration syndrome (HAVS) is clinically based, and the need for an accurate objective test to support the diagnosis has been highlighted. This study aims to analyse the potential of cold provocation thermography (CPT) to fulfill this role. CPT was performed on two groups of subjects: 10 controls and 21 patients with Raynaud's phenomenon (RP) secondary to HAVS. After taking a pre-cooling image, patients donned latex gloves and immersed their hands in water at a temperature of 5 degrees C for 1 min. The patients removed their hands from the water and discarded the gloves, and further images were taken every 30 s for 10 min. On each image, the temperatures of the tip and base were analysed for each digit. The sensitivity, specificity, positive and negative predictive values for fingertip temperatures only, fingertip and fingerbase temperatures combined, and fingertip temperature, fingerbase temperature and temperature gradient combined were determined. Patients with RP secondary to HAVS demonstrated significantly lower finger tip and base temperatures and lower digital temperature gradients at all time intervals when compared with controls (P < 0.01, Student's t-test). CPT has good sensitivity, specificity, positive predictive value and negative predictive value; it strongly supports the clinical diagnosis of digital vasospasm.     

The effect on attendance rates for sickness absence assessment of a revised employee information sheet

BACKGROUND: An increasingly high standard of ethical practice is expected of all doctors and in particular those not providing treatment services. AIMS: This case-control study investigated the effect on non-attendance rates for first sickness absence appointments of a new employee information sheet sent to staff from the two largest departments, education and social services, of a large UK local authority. METHOD: An information sheet detailing the ethical standards applying to a local authority occupational health service was developed and sent to all employees referred by their managers as part of the sickness absence management procedures, along with details of their appointment. RESULTS: A reduction of approximately 1/3 in the rate of non-attendance was noted in the intervention group, which was statistically significant for social services referrals. CONCLUSION: This approach may improve the efficiency and effectiveness of occupational health services, as well as helping to achieve the informed consent of employees undergoing occupational health assessment.     

The sensitivity and specificity of thermometry and plethysmography in the assessment of hand-arm vibration syndrome

Background Finger plethysmography and thermometry are objectivemeasures used to assess the vascular aspect of hand–armvibration syndrome (HAVS). Research to date shows poor correlationbetween these tests and Stockholm Workshop Scale (SWS) vascularstage. Clinicians, researchers and compensation boards requireobjective means to diagnose and quantify HAVS. Aims To define the specificity and sensitivity of thermometryand plethysmography using the SWS as the reference criterion.A secondary goal was to consider cut points for the tests optimizingsensitivity and specificity. Methods A cross-sectional analysis was conducted on HAVS patientsseen at an occupational medicine specialty clinic. Plethysmographyand thermometry were analyzed using SWS vascular stage as theoutcome variable. Logistic regression controlled for age, smokingand time since last vibration exposure and use of vasoactivemedications. The sensitivity and specificity of the combinedtests were calculated using varying cut points. Results A total of 139 patients consented to participate inthe study. Plethysmography stage 1 or greater showed the highestsensitivity (sensitivity 94% and specificity 15%). Specificitywas optimized combining plethysmography stage 3 and thermometrystage 3 (specificity 98% and sensitivity 23%). Maximal diagnosticaccuracy was achieved by plethysmography alone setting the criteriafor a positive test as being stage 1 or greater (70%). Conclusions Neither plethysmography nor thermometry either aloneor in combination demonstrated sufficient sensitivity and specificityto serve as an objective correlate for SWS vascular stage. Allcombinations of plethysmography and thermometry showed a lowerspecificity than sensitivity indicating that the SWS may beless sensitive in detecting vascular pathology than the objectivetests.     

A cohort mortality and cancer incidence survey of recent entrants (1982-91) to the UK rubber industry: findings for 1983-2004

AIM: To monitor the occurrence of cancer in a recently defined cohort of UK rubber workers. METHODS: A cohort of 8651 male and female workers from 41 UK rubber factories has been enumerated. All employees had a minimum of 12 months employment and were first employed at one of the participating factories in the period 1982-91. Mortality and cancer incidence data for the period 1983-2004 were compared with expected values based on appropriate national rates. RESULTS: Mortality from lung cancer was close to expectation for males [observed 22, standardized mortality ratio (SMR) 93] and females (observed 2, SMR 70). Mortality from stomach cancer was also unexceptional in males (observed 4, SMR 86) and females (observed 0, SMR 0). Although based on small numbers, significantly elevated mortality was shown for multiple myeloma in males (observed 5, SMR 385) and females (observed 2, SMR 952). All seven of these latter deaths occurred in workers from the general rubber goods (GRG) sector. CONCLUSIONS: The findings should be treated with caution as they relate to a relatively early period of follow-up. Nevertheless, they hold out the prospect that the elevated SMRs for stomach and lung cancers reported for historical cohorts of UK rubber workers will not be present in more recent cohorts. The elevated occurrence of multiple myeloma may represent no more than a chance finding. Alternatively, these findings may reflect the presence of an unrecognized occupational cancer hazard in parts of the GRG sector of the UK rubber industry.     

CPD for GPs using the THOR-GP website

BACKGROUND: An estimated 3200 UK general practitioners (GPs) practise occupational medicine on a sessional basis. AIM: To assess the educational needs of GPs practising occupational medicine and participating in The Health and Occupational Reporting (THOR) network. METHODS: A questionnaire survey of GPs participating in a national reporting scheme, recording occupational ill-health from general practice (THOR-GP). The questionnaire used scales derived from the syllabus for the Diploma of the Faculty of Occupational Medicine to assess the use of the THOR-GP website for continuing professional development (CPD). Questions were also asked concerning the attitudes and experience of these doctors to CPD in occupational medicine. RESULTS: The response rate was 73% (213/291). Only 22% of responders used the THOR-GP website for CPD. Lack of time was the most frequently cited reason for not using the site. The topics provided on the website which were rated least interesting also appeared as requests for further information in questionnaire returns. CONCLUSION: Online learning has the potential to fulfil the needs of GPs practising occupational medicine. The designers of material for online learning should actively manage and modify the material available in response to educational needs. Further research is required into the clinical and business outcomes of online learning for these doctors.     

Burnout and patient care in junior doctors in Mexico City

Background Burnout is known to occur in public service workersleading to a reduction in effectiveness at work. Aim To estimate the prevalence of burnout in junior doctorsand its impact on patient care. Methods A cross-sectional study of junior doctors at three hospitalsin Mexico City was conducted. Measures used included the MaslachBurnout Inventory (MBI), measuring depersonalization (DP), emotionalexhaustion (EE) and personal achievement (PA), a questionnaireabout patient care practices and attitudes and one on sociodemographiccharacteristics. Logistic regression analysis was used to assessthe association between burnout and suspected risk factors. Results A total of 312 junior doctors participated (responserate 65%). In total, 57% were male and the average age was 28.Average scores in MBI subscales were EE: 18.2, DP: 6.9 and PA:37.6. Burnout prevalence was 40% (126). Junior doctors withburnout were more likely to report suboptimal patient care practicesoccurring monthly (OR 5.5; 95% CI 2.7–11.2) and weekly(OR 5.2; 95% CI 1.6–16.3). The logistic regression modelfor burnout included shifts lasting >12 h, current depression,former major depression, first- or second-year junior doctors,male gender and single status. Conclusions Burnout was most strongly associated with shifts>12 h and with both current and previous depression. Reportedsuboptimal patient care was also associated with working shiftsof 12 h. Burnout may be adversely affecting junior doctors’health and their patients’ care.     

Ill-health retirement: a survey of decision making by occupational physicians working for local authority fire and police services

AIMS: To investigate the effect of qualifications, experience and service on ill-health retirement decisions made by police force and fire brigade medical advisers. METHODS: Doctors advising UK fire and police authorities attending a conference were given 16 case summaries from experienced fire and police advisers and asked to predict the outcome in terms of return to work or ill-health retirement. The answers were scored by assessing whether the outcome predicted by the participant agreed with the case setter. Correlation coefficients were calculated and a kappa coefficient was determined for assessing inter-participant variation. RESULTS: Forty-seven doctors attended, 44 participated and the responses of 39 (86%) were analysed. Over half the doctors participating had held their post for less than 5 years. The majority (>50%) agreed with the case setter in 11 case summaries and in five cases the agreement was strong (>75%). The majority disagreed in five cases, with < or =80% dissenting in three cases. There was wide variation in the inter-participant variation (kappa = -0.333 to +0.5) but the total agreed score correlated significantly (P < 0.05) with experience and service as a fire or police adviser (kappa = 0.33) and ill-health retirement agreement correlated significantly with experience (kappa = 0.49). There was no correlation between the level of qualification and total correct decisions. CONCLUSIONS: Whilst we found general agreement overall, our results suggest that there is inconsistency in ill-health retirement decision making by police and fire service medical advisers in certain cases. Experience appears to improve the likelihood of agreement in decision making.