Occupational Medicine 2010;60:354–357

Advance Access publication on 22 January 2010 doi:10.1093/occmed/kqp195

The predictive capacity of declared musculoskeletal

disorder at pre-employment screening
S. Ryan
Weybridge Health Ltd, 14 Oatlands Close, Weybridge, Surrey, KT13 9ED, UK.
Correspondence to: S. Ryan, Weybridge Health Ltd, 14 Oatlands Close, Weybridge, Surrey, KT13 9ED, UK. Tel/fax: +44

Downloaded from https://academic.oup.com/occmed/article-abstract/60/5/354/1382219 by guest on 10 January 2019

(0)1932840196; e-mail: drseanryan@btinternet.com
...................................................................................................................................................................................

Background This study was undertaken at a major UK airport in response to management concerns about levels of
sickness absence and impaired work capacity in the security department in order to investigate the
possibility of achieving reductions through changes in pre-employment screening procedures.
...................................................................................................................................................................................
Aims To determine the predictive capacity of a declared musculoskeletal disorder (MSD) at pre-employment
screening.
...................................................................................................................................................................................

Methods This study was based on the pre-employment health declarations of security workers. Work outcomes
measured included sickness absence rates (including total and MSD-related absence), work restric-
tion and ill-health retirement rates. Pre-employment questionnaires (PEQs) were analysed for 594
individuals who joined as security staff from 1993–2002. Those who declared an MSD were compared
with those who did not using data from the human resources department and the occupational health
records.
...................................................................................................................................................................................

Results Subjects who disclosed MSDs at pre-employment had significantly higher rates of subsequent total
absence, MSD-related absence and work restrictions compared with those who did not. Potential
confounders included sex, smoking, age and years in employment.
...................................................................................................................................................................................

Conclusions Pre-employment health screening can predict subsequent work outcomes. It is suggested that greater
scrutiny is applied to declarations of an MSD at pre-employment screening in the potentially adverse
manual handling environment of airport security work and that more advice is given to management
about future risks of sickness absence and impaired work capacity.
...................................................................................................................................................................................

Key words Absence; musculoskeletal; pre-employment.

...................................................................................................................................................................................

Introduction true, then the value of such information is called into

Sickness absence and impaired work capability cause fi-
nancial losses and operational difficulties for organiza-
tions. Recognizing which employees are more likely to
have sickness absence or impaired work capability be-
cause of ill-health will help employers to make informed
choices at the pre-employment stage, which should result
in increased economic efficiency.
A high percentage of employees are passed fit for work
without additional comment at pre-employment health
assessments [1,2]. Evaluating the predictive capacity of
the pre-employment questionnaire (PEQ) permits assess-
ment of its potential value as an effective tool to identify
applicants at increased future risk of sickness absence or
impaired work capability. If the information declared at
pre-employment screening is predictive of future work ca-
pability, this may enable evidence-based pre-employment
guidelines to be developed. However, if the converse is

The Author 2010. Published by Oxford University Press on behalf of the Society of Occupational Medicine.
All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
doubt.
Airport security work includes manual handling tasks
involving repetitive frisking, lifting luggage and ‘paddling’
(the process of sweeping luggage along a conveyor with
the arm). There has been little research on the specific
occupational health risks for airport security workers
and this has been confined to issues such as hearing loss
[3] and respiratory symptoms [4]. This study was under-
taken in response to the employer’s impression that there
was an excessive incidence of sickness absence and im-
paired work capability resulting in restriction of work ac-
tivities in the airport security department. However, there
was no comparative data analysis available to support this
view. The main aim of the study was to determine whether
a declared previous history of a musculoskeletal disorder
(MSD) at pre-employment screening predisposed to
a greater subsequent level of sickness absence and/or work
restrictions.
 S. RYAN: PREDICTIVE CAPACITY OF DECLARED MSD AT PRE-EMPLOYMENT SCREENING 355

Methods the hypothesis to be tested. STATA software was used in

This study was undertaken as part of an MSc degree at
the University of Manchester Centre for Occupational
and Environmental Health and the study proposal was re-
viewed by an Ethics Committee Meeting in 2002 and
considered to be ethically sound.
The members of the cohort consisted of new starters
joining the airport security department from January
1993 to December 2002. Lists for each of these years were
obtained from the human resources (HR) department.
Information on sex, date of birth, cigarette smoking, al-
these calculations to reveal a power in excess of 80%.
Results
The cohort consisted of 291 men and 307 women. Of
these, only four individuals had missing PEQs. The
mean age at start of work was 37.4 years with a stan-
dard deviation of 11.1 years. The age range was 17–55
years.
Analysis of the members by starting year is shown in

Downloaded from https://academic.oup.com/occmed/article-abstract/60/5/354/1382219 by guest on 10 January 2019

cohol consumption, height, weight and declaration of an Table 1. All PEQs were analysed for declarations
MSD was obtained from the PEQ in the occupational concerning MSDs (n 5 135). Restriction rates, total
health records (OHRs), held in the occupational health absence rates and MSD absence rates were all signifi-
department. Information on relocations or ill-health re- cantly higher in subjects who had declared an MSD than
tirement was found elsewhere in the OHR. Information in non-declarers (n 5 459; Table 2). Potential confound-
was extracted from the PEQs and if applicable, dates ers consisted of sex, smoking, age and years in employ-
of leaving were also recorded. Employee years of exposure ment. There was no significant difference in these
were calculated from the start dates in HR department factors between the two groups, using the Mann–Whitney
records, in conjunction with any record of leaving date test. Other factors such as body mass index and alcohol
in the OHR, if applicable. The HR department also pro- intake were not prognostic of work outcomes. Retirement
vided lists of employees who had received ill-health retire- rates were not significantly different between the two
ment during this period and of all employees who were groups.
restricted in their job roles or relocated to other areas
as well as sickness absence data for each subject. Each epi-
sode had a health condition category as a cause. Informa- Discussion
tion on outcomes was entered into the SPSS database. This study showed that a questionnaire tool at the pre-
This consisted of the following: employment stage was predictive of future absence and
• Sickness absence for MSDs (‘MSD absence’) work restriction rates in this group of workers. MSD de-
• Sickness absence for all causes (‘Total absence’) clarer absence and restriction rates may have been higher
• Ill-health retirement because of the greater vulnerability of these individuals to
• Restrictions on work activities recorded. the manual-handling environment in the security depart-
ment. The lack of difference in retirement rates may be
Restriction rates were calculated by counting a re- due to the small numbers of retirements in the cohort,
stricted employee as one in the numerator (regardless which amounted to nine in total.
of number and type of restrictions) and the number Poole [5] reviewed the literature on predicting sick-
of years worked to two decimal places as the denomina- ness absence at the pre-employment stage with a view
tor. The retirement rate was calculated as a numerator to providing evidence-based guidelines for assessment
of either zero, for a non-retired employee or one for a re- of conditions likely to influence future sickness absence
tired employee, with the number of years worked as the
denominator. Sickness absence rates were calculated in
days divided by years worked. Descriptive statistics
were calculated for sickness absence rates per employee Table 1. Number of new starters in cohort by year
years of exposure for MSD declarer and non-declarer
sections of the cohort. Restriction and ill-health retire- Year Number % of total
ment rates were calculated for declarer and non-
declarer groups. 1993 11 2
1994 13 2
Comparison in relation to the distribution of con-
1995 12 2
founders in the two groups was made using SPSS soft- 1996 30 5
ware. The non-parametric Mann–Whitney test was 1997 34 6
used for two independent samples in each comparison 1998 71 12
to derive a Z-value to give a corresponding P-value. Power 1999 61 10
2000 151 25
calculations prior to this study used the assumptions that 2001 96 16
there would be 500 subjects, of which a quarter were de- 2002 119 20
clarers. A difference in sickness absence rates was used as
356 OCCUPATIONAL MEDICINE
Table 2. Declared MSD classifications
MSD declarers (n 5 135), mean
(95% CIs)
Total absence (days/year) 13.3 (8.5–18.2)
MSD absence (days/year) 2.3 (1.4–3.1)
Restriction rate (restrictions/year) 0.05 (0.03–0.08)
Retirement rate (retirement/year) 0.004 (0.0008–0.008)
rates, noting that ‘Very few prospective studies of risk
factors for sickness absence have been undertaken’.
Lucey [6], in a more recent study, used the risk table
developed by Poole and showed that those in the high-
risk group had a significantly higher level of sickness
absence in the years following pre-employment assess-
ment. Given that such predictions can be made, do pre-
employment assessments result in communication of
this element of risk to prospective employers, for in-
stance in terms of a statement such as ‘fit with an in-
creased risk of sickness absence’? In an audit of
National Health Service pre-employment assessments,
Lucey [1] found that 98.5% applicants were classified
as fit, with additional comments qualifying fitness made
to management in only 1.5% of cases. No applicant was
declared unfit. Moshe et al [2] found similarly high ac-
ceptance rates in a study of white-collar workers and
labourers. Additionally, they concluded that an occupa-
tional physician-evaluated questionnaire was as effective
as a medical examination in terms of assessing fitness.
Whittaker and Aw [7] in a further audit showed the
most common form of assessment was by questionnaire
alone. There was wide variation in mode of assessment
and in restriction and rejection rates.
This study and others confirm that it is possible to
make predictions from questionnaire evaluation,
but the evidence of high rates of unqualified state-
ments of fitness suggests that this information is
not being used significantly in recommendations to
employers.
It was assumed in data collection that all declarations
were honest and not withheld because of a perceived
prejudice to successful job application. However, the
purpose of the study was to evaluate the predictive
capacity of a ‘declared’ condition on the questionnaire
and not whether the presence of an objectively
confirmed MSD influenced subsequent employment
outcomes.
In this study, MSDs were grouped together crudely
and no distinction was made between different types of
MSD. Further research is required to investigate the sub-
sequent impact of the different types of MSD declared in
this occupational group. Different occupational group
Non-declarers P-value of
(n 5 459), mean difference
(95% CIs)
5.1 (4.1–6.1) ,0.001
0.9 (0.4–1.4) ,0.05
0.01 (0.005–0.02) ,0.01
0.0021 (0.00035–0.0038) NS
Downloaded from https://academic.oup.com/occmed/article-abstract/60/5/354/1382219 by guest on 10 January 2019
studies have addressed specific MSDs such as back pain
and its predictors [8], and shoulder pain prognostic fac-
tors for MSD absence [9], but more research in specific
MSDs and airport work is required.
In this occupational group, the PEQ has been shown
to be a useful predictive tool. This study pointed to the
conclusion that as a group, those declaring MSDs at the
pre-employment stage will have significantly higher sub-
sequent sickness absence and work restriction rates
when undertaking airport security work than non-
declarers. Further research is needed to investigate
the predictive capacity of the different ‘types’ of MSD
with regard to anatomical site, functional impairment
and previous treatment, before predictions of work out-
comes can be made in respect of MSD declarers individ-
ually. This will help to identify at risk individuals in this
potentially adverse manual-handling environment.
According to research, a high level of unqualified state-
ments of fitness are made as a result of pre-employment
health screening and this seems to indicate a failure to
identify and communicate to prospective employers im-
portant information regarding the risk of future sickness
absence.
Key points
• As a group those who declare musculoskeletal dis-
orders at the pre-employment stage will have sig-
nificantly higher subsequent sickness absence and
work restriction rates when undertaking airport se-
curity work than non-declarers.
• Further research is required to investigate the pre-
dictive capacity of different types of musculo-
skeletal disorder declared at this stage in this
occupational group.
• There is evidence of a current failure to identify
and communicate to prospective employers im-
portant information regarding the risk of future
sickness absence in job applicants.
 S. RYAN: PREDICTIVE CAPACITY OF DECLARED MSD AT PRE-EMPLOYMENT SCREENING 357
Acknowledgements
I would like to thank University of Manchester Centre for Oc-
cupational and Environmental Health for help with this study.
Conflicts of interest
None declared.
References
1. Lucey S. Audit of pre-placement assessments undertaken in
an NHS Trust. Occup Med (Lond) 2008;58:512–514.
2. Moshe S, Shilo M, Yagev Y, et al. Comparison of three
methods of pre-employment medical evaluations. Occup
Med (Lond) 2008;58:46–51.
3. Hong OS, Chen S, Conrad KM. Noise induced hearing loss
among male airport workers in Korea. AAOHN J 1998;
46:67–75.
4. Tunnicliffe WS, O’Hickey SP, Fletcher TJ, Miles TF,
Burge PS, Ayres JG. Pulmonary function and respiratory
symptoms in a population of airport workers. Occup Environ
Med 1999;56:118–123.
5. Poole CJM. Can sickness absence be predicted at the pre-
placement health assessment? Occup Med (Lond) 1999;49:
337–339.
6. Lucey S. Can pre-placement health assessments predict sub-
sequent sickness absence? Occup Med (Lond) 2008;58:
355–360.
7. Whittaker S, Aw TC. Audit of pre-employment assessments
by occupational health departments in the National Health
Downloaded from https://academic.oup.com/occmed/article-abstract/60/5/354/1382219 by guest on 10 January 2019
Service. Occup Med (Lond) 1995;45:75–80.
8. Smedley J, Egger P, Cooper C, Coggon D. Prospective cohort
study of predictors of incident low back pain in nurses. Br
Med J 1997;314:1225–1228.
9. Burdorf A, Naaktgeboren B, Post W. Prognostic factors
for musculoskeletal absence and return to work among
welders and metal workers. Occup Environ Med 1998;55:
490–495.