Occupational Medicine Advance Access published online on October 3, 2008
Occupational Medicine, doi:10.1093/occmed/kqn128
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The use of evidence-based clinical tools in occupational medicine
3 Rembrandt Close, Wokingham, Berkshire RG41 3BL, UK
Correspondence to: Kevin Bailey. Tel: +44 1189791055; fax: +44 845 7090431; e-mail: kevinbailey{at}doctors.org.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
Background Little is known of UK occupational physicians' usage of screening questionnaires in assessment.
Aim To determine, among members of the Society of Occupational Medicine (SOM), the degree of awareness and the extent of usage of 12 previously validated screening questionnaires and two educational interventions.
Method A cross-sectional self-report survey of a random sample of 400 members of the SOM.
Results The response rate was 54% (216). Awareness was good especially for the disease-specific questionnaires. However, no usage (0%) exceeded low usage (1–50%) for half the survey instruments. For three instruments, the converse applied and for four instruments no usage equalled low usage. The main reasons for non-usage were lack of availability and lack of time.
Conclusions The most used and familiar instruments were the Back Book, Mini-Mental State Examination and Numerical Rating Scale. Over half the respondents also used the Hospital Anxiety and Depression Scale, Alcohol Use Disorder Identification Test and the remaining pain scales at least to some degree.
Keywords Occupational physicians; physician behaviour; questionnaire; screening
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
Little is known of UK occupational physicians' daily clinical practice. A systematic review of evaluative research conducted in occupational health services found that occupational health consultations, occupational rehabilitation and use of screening questionnaires have hardly been studied at all [1]. Reasons for using questionnaires are debated. However, they may help standardize clinical assessment and prove useful in defending one's clinical opinion (for example when a general practitioner holds a contrary opinion on fitness to return to work) or in a court of law or employment tribunal. Secondly, evidence-based research and practice rely on a quantifiable database which in turn rely on standardized assessment. Thirdly, clinical assessment should cover general health status (quality of life and psychological distress). In view of their potential benefits, the aim of this study was therefore to determine the level of awareness and usage among occupational physicians of a number of screening questionnaires and two educational interventions previously validated in primary care. Secondary aims included determining whether any relationship existed between usage pattern and demographic variables.
The study questionnaires selected are used for the assessment of disease-specific states (anxiety, depression, cognitive impairment, alcohol use disorder and pain assessment) and general health status as shown in Table 1.
|
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
A literature search was conducted using MEDLINE (1950 present), EMBASE (1974 present), CINAHL (1982 present) and PsychINFO (1950 present), ‘Occupational and Environmental Medicine’ issues from January 1994, ‘Occupational Medicine’ issues from 1993 and clinical guidelines from the National Screening Committee, the National Institute for Clinical Excellence and the Scottish Intercollegiate Guidelines Network.
A sample of 400 members of the UK Society of Occupational Medicine (SOM) was selected randomly using ‘Research Randomizer’, an online random sampling programme.
After receiving ethical approval from the West Midlands Multi-centre Research Ethics Committee, the definitive questionnaire was distributed over two mail outs with a covering letter explaining the purpose of the survey, guaranteeing anonymity and assuring that data collected would remain confidential. Data from all questionnaires were inputted and verified by the author. Seventy per cent of the sample responded in the comments section which contained two tick boxes (‘lack of availability’ or ‘insufficient time’) and space for bespoke comments.
The study hypotheses were Hypothesis 1 (descriptive): ‘occupational physicians are not aware of, and do not use, these instruments’ and Hypothesis 2 (comparative): ‘there is no difference between the demography of occupational physicians and their awareness and usage patterns of the instruments’. For the second hypothesis, P values were all two tailed and their level of significance was set at 0.05. The age/gender data are categorical and were analysed using the chi-square test. The variables of proportion of time spent in occupational health clinical consultations and qualifications are ordinal and Kendall’s tau-b analysis is an appropriate test for analysing trends for ordinal data.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
The study response rate was 54%. Six questionnaires were unusable because of omissions leaving a total of 210 questionnaires that were suitable for analysis. Seventy-three per cent of respondents were male and 27% female. The mean age of respondents was 51.1 years with a standard deviation of 5.1 years.
Postgraduate qualifications included none (11%), diploma (18%), advanced diploma (2%), Associateship of the Faculty of Occupational Medicine (AFOM) (21%), Membership of the Faculty of Occupational Medicine (MFOM) (26%) and Fellowship of the Faculty of Occupational Medicine (FFOM) (21%).
In terms of clinical consultations, 5% (11 respondents) did no clinical consultations, 58% spent up to half their day (1–50%) doing them and 37% spent 51–100% of their day in occupational health clinical consultations.
A breakdown of awareness of the survey instruments by respondents is shown in Table 2.
|
More than half the respondents were aware of 10 out of the 14 survey instruments and over three quarters were aware of five of them. From most to least familiar, the order was HADS (>90% aware); BECK Depression Inventory/Back Book (both >80%); GHQ/VAS (over three quarters); AUDIT/NRS (nearly two-thirds) and finally the Goldberg scale, MMSE and the VRS (>50%). Less than half the respondents were aware of the health-related quality of life scales, SF 36 (44%) and SF 12 (42%) or the MHI (43%), and the least well-known instrument was the Whiplash Book (39%).
A breakdown of usage of the survey instruments by respondents is shown in Table 3.
|
A high usage pattern was uncommon. Less than 10% of respondents used any survey instrument >50% of the time except for the Back Book (20% respondents). The commonest pattern was either no usage (0%) or low usage (1–50% of the time). For half the instruments, namely the Goldberg, BECK, Whiplash Book and general health measures (GHQ, MHI and the quality of life scales SF 36 and SF 12), no usage clearly exceeded low usage. This may reflect a perceived lack of relevance of quality of life measures to occupational medicine compared to the other more disease-specific instruments. Low usage clearly exceeded no usage for the MMSE, NRS and the Back Book and low usage equalled no usage for the HADS, AUDIT, VAS and VRS. Low awareness predicted low usage for the general measures of health (SF 36, SF 12 and MHI) but not for the HADS and BECK. Despite more than four-fifths of respondents being aware of the BECK, over two-thirds of them did not use it.
Table 4 summarizes the reasons given for not using the survey instruments.
|
The lack of availability and insufficient time were commonly cited reasons. The commonest bespoke reasons were a negative attitude towards questionnaires especially in regard to their value relative to clinical assessment, uncertainty regarding the validity and clinical relevance to daily practice in occupational medicine, irrelevance to the current job and lack of awareness of them or training in their use.
The second, Hypothesis 2 was comparative, and the first step explored the associations between awareness and demographic variables as shown in Table 5. Qualifications were classified as nil, diploma, advanced diploma, AFOM, MFOM and FFOM. Higher qualification was significantly associated with greater awareness for the MHI, Goldberg, GHQ, SF 36 and SF 12 but the reverse was true for the MMSE. Greater awareness was significantly associated with male gender for the MHI and with female gender for the BECK. Awareness of the BECK was significantly greater in the younger age group (<52 years of age).
|
The second step of the comparative data analysis compared usage of the survey instruments (0, 1–50 and 51–100%) with demographic variables as shown in Table 6. Lower qualification was significantly associated with greater usage for the MMSE, MHI, SF 36 and SF 12. Older respondents (>52 years of age) were significantly more likely to use the GHQ and MHI. Female respondents were significantly more likely to use the Back Book.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
Most occupational physicians are aware of the HADS, BECK, Back Book, GHQ and VAS but a minority are aware of the quality of life scales (SF 36/SF 12) and the Whiplash Book. Overall, the questionnaires are either not used (notably the general health measures) or used less than half the time although usage is higher for the HADS, AUDIT, VAS and VRS. The MMSE was more familiar to, and used to a greater degree by, those with less postgraduate occupational medicine qualifications who may be more involved in primary care where its use is recommended for the assessment of suspected dementia [2]. The age of some questionnaires (MHI and GHQ) may account for the greater awareness and usage pattern in older respondents. The MHI was developed >30 years ago [3], and in the 1970s, the GHQ was the most widely used psychiatric screening test in the UK [4]. However, the BECK is even older [5] but was more familiar to younger respondents perhaps possibly because it is taught in some occupational medicine diploma courses.
There is no satisfactory explanation to account for most of the statistical associations between awareness/usage pattern and demographic variables which might be explained by Type 1 (false positive) error which occurs when a true null hypothesis is rejected. In any statistical test, the probability of Type 1 error is equivalent to the level of significance set; in this case 5%. In addition, the greater the number of associations studied between non-associated variables, the more frequently a significant statistical finding occurs. This may explain the greater awareness of the BECK and greater usage of the Back Book among female respondents, greater awareness of the MHI among male respondents and greater awareness but lower usage of the SF 36, SF 12 and MHI among respondents with higher qualifications.
The limitations of the study are firstly, the questionnaire used in the survey has not been previously validated although it was piloted among a group of occupational physicians before its distribution. Secondly, the low-usage (1–50% of the time) and high-usage (51–100% of the time) categories were broad and it is not possible to quantify further the degree of usage within each category.
Thirdly, as the response rate for the survey was 54%, there was a possibility of non-response bias. Unfortunately, it was not possible to collect demographic information on non-responders to compare with responders because, in order to encourage frank responses and reduce self-report bias, the anonymity of all respondents was protected.
Fourthly, it has been acknowledged in many studies that physicians may overestimate their clinical practice behaviour when self-reporting [6, 7]. Despite a guarantee assuring anonymity in the survey, it is possible that responses reflect a more idealized version of practice rather than what actually takes place thus leading to self-report bias.
Finally, as the sample for this survey was drawn from members of the SOM whose membership is voluntary, any conclusions drawn from the study relate specifically to the target population of members of the SOM and may not extend to all occupational physicians in the UK.
Lack of endorsement from authoritative bodies in occupational medicine was one reason some respondents gave for not using the survey instruments. All self-reporting questionnaires are subjective and many ask about physical symptoms such as fatigue (for example the BDI and the Goldberg scale but not the HADS or GHQ 12) and this might artificially inflate scores if these symptoms are due to concomitant physical illness. Cost considerations apply for HADS, GHQ, SF 36, SF 12 and BDI-11. Although copyright specifically does not apply to the other survey questionnaires. Finally, completion times range from <5 min for HADS and AUDIT [8,9] to up to 10 min for the BDI, MMSE, SF 36, MHI and GHQ 60 take [2].
However, should the survey instruments ever form part of evidence-based guidelines, there may be ethical as well as legal reasons to use them. Indeed, some of them already receive mention in current authoritative UK guidelines. Lack of endorsement from authoritative bodies in occupational medicine was that one reason some respondents gave for not using the survey instruments. If the Faculty of Occupational Medicine were to publish an evidence-based review of the role of screening questionnaires and educational interventions, as it has on back pain and hand–arm vibration syndrome, there is a reasonable expectation it might influence the daily practice of occupational physicians.
Key points
|
|
Conflicts of interest |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
None declared.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConflicts of interestReferences |
|---|
- Hulshof CTJ, Verbeek JHAM, van Dijk FJ, van der Weide WE, Braam IT. Evaluation research occupational health services: general principles and a systematic review of empirical studies. Occup Environ Med (1999) 56:361–377.
[Abstract/Free Full Text] - National Institute for Clinical Excellence. Dementia: Supporting People with Dementia and Their Carers in Health and Social Care, National Clinical Practice Guideline Number 42, National Collaborating Centre for Mental Health Commissioned by the Social Care Institute for Excellence National Institute for Health and Clinical Excellence. London: National Institute for Clinical Excellence, 2006.
- Veit CT, Ware J. The structure of psychological distress and well-being in general populations. J Consult Clin Psychol (1983) 51:730–742.[CrossRef][Web of Science][Medline]
- Winston M, Smith J. A trans-cultural comparison of four psychiatric case finding instruments in a Welsh community. Soc Psychiatry Psychiatr Epidemiol (2000) 35:569–575.[CrossRef][Web of Science][Medline]
- Beck AT, Ward CH, Mendelson M, et al. An inventory for measuring depression. Arch Gen Psychiatry (1961) 4:53–63.
- Tunis SR, Hayward RSA, Wilson MC, Mock J, Erbaugh J. Internists' attitudes about clinical practice guidelines. Ann Intern Med (1994) 120:956–963.
[Abstract/Free Full Text] - Adams AS, Soumerai SB, Lomas J, Ross-Degnan D. Evidence of self-report bias in assessing adherence to guidelines. Int J Qual Health Care (1999) 11:187–192.
[Abstract/Free Full Text] - Scottish Intercollegiate Guidelines Network. The Management of Harmful Drinking in Alcohol Dependence in Primary Care. Guideline Number 74. London: Scottish Intercollegiate Guidelines Network, 2003.
- Mulrow CD, Williams JW, Genety MB, Ramirez G, Montiel OM, Kerber C. Case-finding instruments for depression in primary care settings. Ann Int Med (1995) 122:913–921.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||