Post on 03-Jun-2018
transcript
8/13/2019 Thor Trends 12
1/92
1
Time trends in the incidence of work-related ill-health in the UK, 1996-2011: estimation from
THOR surveillance data
Carder M, McNamee R,Hussey L, Turner S, Agius R
Centre for Occupational and Environmental Health, Health Sciences Group,School of Community Based Medicine, Faculty of Medical and Human
Sciences, the University of Manchester
http://www.medicine.manchester.ac.uk/oeh/research/thor/
September 2012
8/13/2019 Thor Trends 12
2/92
2
EXECUTIVE SUMMARY
BACKGROUND: This report describes the latest analyses of case reports
collected by The Health and Occupation Research (THOR) network to
investigate the trend in the physician reported incidence of work-related
illness (WRI) over time. Case reports of WRI from the three THOR schemes
currently supported by HSE were analysed: skin disease reported to
EPIDERM by dermatologists (1996-2011), respiratory disease reported to
SWORD by chest physicians (1999-2011), and (any) WRI reported to THOR-
GP by general practitioners (2006-2011).
METHOD: The methodology employed remained unchanged from that
described previously. The statistical software package STATA was used to fit
longitudinal, negative binomial (i.e. over-dispersed) Poisson models with
random effects. The dependent variable was the number of actual cases,
including zeros, per reporter per month. Variables were also included to
represent season, reporter type (core or sample), and first month/s as a
new reporter. Changes in the population base were accounted for by
including an offset variable representing the UK working population for each
year. Change in incidence was measured in two different ways related to the
treatment of calendar time in the model: a non-parametric approach
containing separate indicators for different years (2011 as the reference year)
and a parametric approach with continuous time variable measured on a
scale of years.
RESULTS: Case reports of skin disease to EPIDERM were predominantly
contact dermatitis (CD) (81%) with smaller proportions of neoplasia (12%) and
other skin diagnoses. Overall the addition of a further year of data (2011) had
little impact on the annual average change in incidence of work-related skin
disease which remained similar to those previously reported i.e. indicating a
small but significant yearly reduction in incidence. The updated results were
-3.4% (95% CIs:-3.9, -2.8) for all skin diagnoses, -3.3% (95% CIs:-3.9,-2.6) for
CD, -5.6% (95% CIs:-7.5, -3.6) for contact urticaria and -2.6% (95% CIs:-4.1,
-1.0) for neoplasia. Relative rates by year, however, suggested a larger
8/13/2019 Thor Trends 12
3/92
8/13/2019 Thor Trends 12
4/92
4
with previous observations regarding the annual average estimated change in
incidence of the categories of work-related illness investigated here. For
dermatologists and chest physicians, the reported incidence was observed
either to be falling (contact dermatitis, contact urticaria, neoplasia, asthma,
mesothelioma) or to remain relatively unchanged over time (benign pleural
plaques, pneumoconiosis). In addition to simply observing these trends,
recent THOR work has begun to investigate to what extent the observed
changes are the result of changes in legislation, regulatory activity and market
forces, relating to specific substances such as latex, chromium and
glutaraldehyde. The application of the MLM methodology to investigate
interventions such as these, has proved to be a useful tool and further work is
anticipated in this area.
The estimated annual change in incidence based on data from GPs also
remained relatively unchanged with the addition of the 2011 data, with a
decrease in incidence over time observed for all the disease groups
investigated. However, the impact of the changes in sampling methodology,
resulting from the cessation of funding to GPs for participation, on the
estimates of trend is yet to be fully understood.
It is also possible that the trends in GPs as well as in specialists have been
influenced by reporter fatigue. A large body of work has been carried out to
date investigating this issue within THOR with some evidence of fatigue
present for some (but not all) physician groups and types (core or sample) as
suggested by an increase in nil returns, non-response and excess zeros over
time. The implications of these findings are under review.
CONCLUSION: Annual reports describing the trend in incidence of work-
related illness, as reported by medical physicians to THOR, have been
provided to HSE since 2006. To date, this has generated a large and
invaluable body of work not only providing annual updates of the trends but
also describing ongoing developments in the methodology, including
investigations of the important issue of reporter fatigue. Some of the observed
trends have remained relatively unchanged with the addition of each
8/13/2019 Thor Trends 12
5/92
5
successive year of data and are in accordance with those expected as a
result of Government initiatives (for example, the decline in incidence of
asthma, contact dermatitis and contact urticaria), with further work by this
group investigating these trends in relation to specific sectors or agents. For
others, in particular the asbestos related diseases, the observed trends have
shown a greater degree of variation and may, at least in part, reflect a change
in management and referral patterns rather than a true trend. Ongoing work,
including benchmarking with other data sources will help clarify these issues
further. Since 2009, the annual trends reports have also analysed data
reported by GPs. However, these GP data should be interpreted with caution
until issues such as the impact of reporting activity and reporter fatigue are
better understood. Work continues to improve the THOR trends methodology
to enable these and other partially resolved issues to be investigated further.
8/13/2019 Thor Trends 12
6/92
6
CONTENTS
Page number1 BACKGROUND 10
2 METHOD 10
3 RESULTS 133.1 OVERVIEW OF SCHEMES 133.1.1 EPIDERM 133.1.2 SWORD 133.1.3 THOR-GP 143.2 TIME TRENDS BY DISEASE CATEGORY 143.2.1 TOTAL WORK-RELATED ILLNESS 153.2.2 WORK-RELATED SKIN DISEASE 173.2.3 WORK-RELATED RESPIRATORY DISEASE 37
3.2.4 WORK-RELATED MUSCULOSKELETAL DISORDERS 513.2.5 WORK-RELATED MENTAL ILL-HEALTH 56
4 DISCUSSION 60
5 CONCLUSION 67
8/13/2019 Thor Trends 12
7/92
7
LIST OF TABLES
Pagenumber
1 Data period for trends analyses 11
2 Categories of illness included in the analyses 123 Average annual percentage change in risk in total work-
related ill-health16
4 Relative rates by year, with 95% comparison intervals,total work-related ill-health (2011 estimate =1)
16
5 Average annual percentage change in reportedincidence in work-related skin disease
19
6 Relative rates by year, with 95% comparison intervals,total skin disease (2011 estimate =1)
20
7 Relative rates by year, with 95% comparison intervals,all contact dermatitis (2011 estimate =1)
23
8 Relative rates by year, with 95% comparison intervals,allergic contact dermatitis (2011 estimate =1)
25
9 Relative rates by year, with 95% comparison intervals,irritant contact dermatitis (2011 estimate =1)
27
10 Relative rates by year, with 95% comparison intervals,mixed contact dermatitis (2011 estimate =1)
29
11 Relative rates by year, with 95% comparison intervals,contact urticaria (2011 estimate =1)
31
12 Relative rates by year, with 95% comparison intervals,neoplasia (2011 estimate =1)
33
13 Relative rates by year, with 95% comparison intervals,other skin* (2011 estimate =1)
35
14 Average annual percentage change in reportedincidence in work-related respiratory disease
38
15 Relative rates by year, with 95% comparison intervals,total respiratory disease (2011 estimate =1)
39
16 Relative rates by year, with 95% comparison intervals,asthma (2011 estimate =1)
41
17 Relative rates by year, with 95% comparison intervals,mesothelioma (2011 estimate =1)
43
18 Relative rates by year, with 95% comparison intervals
(2011 estimate =1), benign pleural plaques
45
19 Relative rates by year, with 95% comparison intervals,pneumoconiosis (2011 estimate =1)
47
20 Relative rates by year, with 95% comparison intervals,other* respiratory disease (2011 estimate =1)
49
21 Average annual percentage change in reportedincidence in work-related musculoskeletal disorders
51
22 Relative rates by year, with 95% comparison intervals,total musculoskeletal disorders (2011 estimate = 1)
52
23 Relative rates by year, with 95% comparison intervals,upper limb disorders (2011 estimate = 1)
53
24 Relative rates by year, with 95% comparison intervals,spine/back disorders (2011 estimate = 1)
54
8/13/2019 Thor Trends 12
8/92
8
25 Relative rates by year, with 95% comparison intervals,lower limb disorders (2011 estimate = 1)
55
26 Average annual percentage change in reportedincidence in work-related mental ill-health
56
27 Relative rates by year, with 95% comparison intervals,
total mental ill-health (2011 estimate = 1)
57
28 Relative rates by year, with 95% comparison intervals,anxiety and depression (2011 estimate = 1)
58
29 Relative rates by year, with 95% comparison intervals,other work stress (2011 estimate = 1)
59
LIST OF FIGURES
Page
number1 Relative rates by year (2011 estimate = 1), with 95%
comparison intervals, total work-related ill-health16
2 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, total skin disease
21
3 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, all contact dermatitis
24
4 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, allergic contact dermatitis
26
5 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, irritant contact dermatitis
28
6 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, mixed contact dermatitis
30
7 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, contact urticaria
32
8 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, neoplasia
34
9 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, other (than contact dermatitis)skin
36
10 Relative rates by year (2011 estimate = 1), with 95%
comparison intervals, total respiratory disease
40
11 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, asthma
42
12 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, mesothelioma
44
13 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, benign pleural plaques
46
14 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, pneumoconiosis
48
15 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, other respiratory disease
50
16 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, total musculoskeletal disorders
52
8/13/2019 Thor Trends 12
9/92
9
17 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, upper limb disorders
53
18 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, upper limb disorders
54
19 Relative rates by year (2011 estimate = 1), with 95%
comparison intervals, lower limb disorders
55
20 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, total mental ill-health
57
21 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, anxiety and depression
58
22 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, other work stress
59
LIST OF APPENDICESPage number
A Descriptive analyses 71
8/13/2019 Thor Trends 12
10/92
10
1. BACKGROUND
This report describes the latest analyses of case reports collected by The
Health and Occupation Research (THOR) network1to investigate the trend in
incidence of work-related illness (WRI) over time. This report should not be
viewed as a stand-alone report but should be viewed in conjunction with
previous reports submitted to the Health and Safety Executive (HSE) which
provide much fuller detail regarding the THOR network, the rationale behind
the study and advances in the methodology over time2-8. For the present
study, data from the three THOR schemes currently supported by HSE were
analysed: case reports of work-related skin disease reported to EPIDERM by
dermatologists (1996-2011), case reports of work-related respiratory disease
reported to SWORD by chest physicians (1999-2011), and case reports of
(any) WRI reported to THOR-GP by general practitioners (2006-2011).
2. METHOD
The methodology employed in this study has been described in full in
previous reports2-8and only a brief summary is provided here. The data period
and diagnostic groups analysed are shown in Tables 1 and 2. Analyses were
based on all reporters in each scheme combined, and separately (where
appropriate) for core (i.e. physicians reporting every month) or sample (i.e.
physicians reporting for one randomly selected month per year) reporting
groups. A brief history of the individual schemes, including any significant
changes to the proportion of core to sample reporters over time, was provided
in the trends report submitted to HSE in 20112. The only substantial change in
2011 was for THOR-GP, which moved from 80% sample reporting in 2010 to
100% sample reporting in 2011. For the present report, analyses of THOR-GP
data were based on all reporters (i.e. core and sample combined) only, with
no separate analysis by reporter type (core only or sample only). Core only
analyses were not carried out because core reporting stopped in December
2010, with core only trends for the period 2006-2010 reported on in the trends
report submitted in 20112
. Sample only analyses were not carried out because
prior to 2010 only a small proportion (4%) of GPs reported on a sample basis;
8/13/2019 Thor Trends 12
11/92
8/13/2019 Thor Trends 12
12/92
12
Table 2 Categories of illness included in the analyses
Clinical specialist THOR-GP
All WRI - Yes
Total skin Yes YesContact dermatitis (CD) Yes -
Allergic CD Yes -
Irritant CD Yes -
Mixed CD Yes -
Neoplasia Yes -
Contact urticaria Yes -Other skina Yes -
Total respiratory Yes -Asthma Yes -
Mesothelioma Yes -Benign pleural disease Yes -
Pneumoconiosis Yes -Other respiratory diseaseb Yes -
Total musculoskeletal - YesUpper limb disordersc - YesSpine/back disorders - Yes
Lower limb disorderse - Yes
Total mental ill-health - YesAnxiety and depression - Yes
Other work stress - YesaOther than contact dermatitis
bOther than asthma, mesothelioma, benign pleural disease or pneumoconiosis
cHand/wrist/arm, shoulder and elbow
dNeck/thoracic spine, lumbar spine/trunk
eHip/knee, ankle/foot
8/13/2019 Thor Trends 12
13/92
13
3. RESULTS
3.1 OVERVIEW OF SCHEMES
An overview of the reporting activity of the physicians participating in
EPIDERM, SWORD, and THOR-GP is provided in Appendix A and briefly
described below.
3.1.1 EPIDERM
A total of 415 dermatologists have been enrolled in EPIDERM during the
study period with 93% actively participating at least once (i.e. either returning
cases or declaring I have nothing to report this month). 2011 saw a small
decrease in the overall number of physicians in EPIDERM (from 181 in 2010
to 175 in 2011) and a decrease in the average number of active reporters per
month (26 per month in 2010 to 21 per month in 2011). However, the average
cases per active reporter remained little changed between the two years (2.4
in 2010 and 2.2 in 2011). Reporters to EPIDERM are predominantly sample
(86% in 2011) but core reporters report more cases per active reporter per
month (3.7) compared to sample (0.9). Case reports to EPIDERM continue tobe predominantly contact dermatitis (81% of total cases) with smaller
proportions of neoplasia (12%) and other skin diagnoses.
3.1.2 SWORD
Active participation in SWORD during the study period was similar to
EPIDERM with 94% of the 798 chest physicians enrolled during this periodactively reporting at least once. Although the total number of reporters in
SWORD fell slightly between 2010 and 2011 (474 to 460) the average number
of active reporters per month was similar (37 in 2010 and 35 in 2011) and the
average number of cases per active reporter actually slightly increased (1 in
2010 to 1.2 in 2011). Similar to EPIDERM, the smaller group of chest
physicians reporting as core reported more cases per active reporter per
month than chest physicians reporting as sample. The majority of the actual
diagnoses (44%) reported to SWORD during the study period were benign
8/13/2019 Thor Trends 12
14/92
14
pleural plaques. Of the remaining cases 20% were mesothelioma, 19%
asthma, 8% pneumoconiosis, and 12% other respiratory disease.
3.1.3 THOR-GP
THOR-GP differs to EPIDERM and SWORD in that at the start of the study
period (2006) all GPs reported to THOR-GP on a core basis whilst at the end
of the study period (2011) reporting was exclusively sample, with the majority
of the switch from core to sample occurring in 2010. Active participation during
this period was lower than observed for SWORD and EPIDERM (76% of the
699 GPs ever enrolled). In contrast to the clinical specialists, GPs participating
as sample physicians reported on average more cases per active reporter per
month than GPs participating as core reporters. Thus, although the switch to
100% sample meant that there was a substantial decrease in the average
number of active reporters per month between 2010 and 2011 (47 to 15), the
average number of cases per active reporter month actually increased
between 2010 and 2011 (0.9 to 1.4). Musculoskeletal and mental ill-health
case reports comprised the majority (52% and 32%, respectively) of the cases
reported to THOR-GP with smaller proportions of skin (10%), respiratory (2%)
and other diagnoses (4%).
3.2 TIME TRENDS BY DISEASE CATEGORY
This report continues with the approach first adopted in the trends report
submitted to HSE in September 20103, in that the statistical uncertainty
(confidence intervals) in the graphs illustrating time trends are presented in
such a way as to allow the reader to assess the significance of the difference
between any two years. This approach suggested by the HSE liaison officer
when steering the research follows the method described by Firth and de
Menezes10 which assigns a confidence (or comparison) interval to the
reference category (2011 in the present analyses) and reduces the width of
the confidence (comparison) intervals of non-reference categories in such a
way that all pairwise comparisons between years can validly be made using
these adjusted confidence intervals.
8/13/2019 Thor Trends 12
15/92
15
3.2.1 TOTAL WORK-RELATED ILLNESS
The average annual percentage change in risk of total WRI, as reported to
THOR-GP is shown in Table 3 whilst the relative rates by year are shown in
Table 4 and Figure 1. For the period 2006-2011, the average annual decrease
in incidence was -12.5% (95% CIs: -15, -10). This remains little changed from
the last reported estimate (for the period 2006-2010) which was very similar at
-12.3% (95% CIs: -14.8, -9.7). The graphs showing relative rates by year for
THOR-GP suggest the slight rise in incidence observed in 2010 has been
followed by a fall in 2011. However, it should be noted that the confidence
intervals are wider in 2010 and (especially) 2011, (reflecting the move to a
greater proportion of sample reporters) and overlap earlier years.
8/13/2019 Thor Trends 12
16/92
16
Table 3 Average annual percentage change in risk in total work-related illness
ESTIMATED % CHANGE(95% CONFIDENCE INTERVAL)
Reporters Year (continuous) THOR-GP
All 2006-2011 -12.5 (-15, -10)
Table 4 Relative rates by year, with 95% comparison intervals, totalwork-related ill-health (2011 estimate =1)
Relative rates (95% comparison interval)
THOR-GP
Reporter Group Year
All 2006 1.74 (1.61, 1.88)
2007 1.39 (1.30, 1.48)
2008 1.30 (1.20, 1.40)
2009 1.06 (0.98, 1.14)2010 1.23 (1.09, 1.38)
2011 1.00 (0.83, 1.20)
Figure 1 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, total work-related ill-health
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.82
2006 2007 2008 2009 2010 2011
Relativerate
Year
THORGP, all reporters
8/13/2019 Thor Trends 12
17/92
17
3.2.2 WORK-RELATED SKIN DISEASE
The average annual percentage change in risk of work-related skin disease,
as reported by dermatologists and GPs is shown in Table 5 whilst the relative
rates by year are shown in Tables 6 to 13 and Figures 2 to 9.
For both dermatologists and GPs, the addition of a further year of data (2011)
had little impact on the annual average change in incidence of work-related
skin disease which remained at -3.4% (95% CIs: -3.9, -2.8) for dermatologists
and increased slightly from -5.3% (2006-2010) to -5.7% (95% CI: -13.2, 2.3)
for GPs (2006-2011). Analyses restricted to case reports from dermatologists
during 2006-2011 (thus enabling a direct comparison with GP case reports)
suggested a steeper decrease in incidence for GPs compared to
dermatologists. Although the annual average change in incidence remained
little changed after the addition of 2011 data, the graphs showing relative
rates by year suggest a large decrease in incidence for dermatologists
between 2010 and 2011 compared to changes in previous years. Analyses
based on reporter type (i.e. core only or sample only) suggest that this
decrease occurred amongst the core reporters rather than the sample
reporters. The graph showing relative rates by year for THOR-GP suggeststhe slight increase observed for 2010 has been followed by a further increase
in 2011. However, trends based on GP data should be viewed with caution as
results are based on relatively few cases (confidence intervals are overlapping
for all years).
The predicted trend for contact dermatitis (CD) was very similar to that
observed for total work-related skin disease. For dermatologists, in addition toall contact dermatitis combined, further analyses by type of CD were carried
out. As reported previously, a steeper decrease in the incidence of allergic CD
was observed (-5.4% (95% CIs: -6.3, -4.6)) compared to irritant CD (-1.2 (95%
CIs: -2.1, -0.3)) or mixed CD (-2.2% (95% CIs: -3.5, -0.8)). Restricting the
analyses to cases reported by core reporters had little effect on the observed
trend for these CD sub-categories. For sample only analyses however, the
percentage decrease in allergic CD and irritant CD were very similar (2-3%).
8/13/2019 Thor Trends 12
18/92
18
The annual average change in incidence of both urticaria and neoplasia
(investigated for dermatologists only) also remained little changed by the
addition of the 2011 data with the previously observed variation between
reporter type (decrease in incidence observed for core reporters and increase
for sample reporters) still apparent. Of interest, the large fall in incidence
between 2010 and 2011 observed for CD was similarly observed for
neoplasia, with the drop again appearing to be largely driven by a decrease in
case reports from core reporters. However, it should be noted that the
confidence intervals for 2011 are very wide and overlap previous years,
making it difficult to draw any firm conclusions.
For the purpose of this study, the group other skin referred to all non CD
cases (and therefore included all of the groups already discussed, other then
CD). As expected from the individual group analyses, an overall downward
trend was predicted for this group. Some variation between core and sample
reporters was observed for EPIDERM, but this was probably largely driven by
the (already discussed) core and sample differences for neoplasia.
8/13/2019 Thor Trends 12
19/92
19
Table 5 Average annual percentage change in reported incidence inwork-related skin disease
a) All reporters
ESTIMATED % CHANGE(95% CONFIDENCE INTERVAL)
Year(continuous)
EPIDERM THOR-GP
Total skin 1996-2011 -3.4 (-3.9, -2.8) /
2006-2011 -2.0 (-4.3, 0.4) -5.7 (-13.2, 2.3)
Contact dermatitis (CD) 1996-2011 -3.3 (-3.9, -2.6) /
Allergic CD 1996-2011 -5.4 (-6.3, -4.6) /
Irritant CD 1996-2011 -1.2 (-2.1,-0.3) /
Mixed CD 1996-2011 -2.2 (-3.5,-0.8) /
Urticaria 1996-2011 -5.6 (-7.5, -3.6) /
Neoplasia 1996-2011 -2.6 (-4.1, -1.0) /Other* skin 1996-2011 -3.2 (-4.3, -2.0) /
b) Core reporters
ESTIMATED % CHANGE(95% CONFIDENCEINTERVAL)
Year (continuous) EPIDERM
Total skin 1996-2011 -3.6 (-4.2, -3.0)
Contact dermatitis (CD) 1996-2011 -3.3 (-4.0, -2.7)
Allergic CD 1996-2011 -6.0 (-6.9, -5.0)
Irritant CD 1996-2011 -1.1 (-2.0, -0.1)
Mixed CD 1996-2011 -2.3 (-3.7, -0.9)
Urticaria 1996-2011 -6.2 (-8.2, -4.2)
Neoplasia 1996-2011 -4.6 (-6.2, -3.0)Other* skin 1996-2011 -4.5 (-5.7, -3.3)
c) Sample reporters
ESTIMATED % CHANGE(95% CONFIDENCE INTERVAL)
Year(continuous)
EPIDERM
Total skin 1996-2011 -1.3 (-3.1, 0.5)
Contact dermatitis (CD) 1996-2011 -2.7 (-4.6, -0.6)
Allergic CD 1996-2011 -1.8 (-4.5, 1.0)
Irritant CD 1996-2011 -2.5 (-5.3, 0.5)
Mixed CD 1996-2011 -0.3 (-4.7, 4.3)
Urticaria 1996-2011 1.2 (-6.4, 9.5)
Neoplasia 1996-2011 6.7 (2.8, 10.6)Other* skin 1996-2011 3.9 (0.8, 7.1)*Other than contact dermatitis
8/13/2019 Thor Trends 12
20/92
20
Table 6 Relative rates by year, with 95% comparison intervals, totalskin disease (2011 estimate =1)
Relative rates (95% comparison interval)
EPIDERM THOR-GP
Reporter Group Year
All 1996 1.71 (1.58, 1.86) /
1997 1.82 (1.69, 1.96) /
1998 1.67 (1.54, 1.80) /
1999 1.71 (1.59, 1.85) /
2000 1.57 (1.45, 1.71) /
2001 1.45 (1.34, 1.56) /
2002 1.41 (1.30, 1.52) /
2003 1.41 (1.31, 1.52) /
2004 1.30 (1.20, 1.40) /
2005 1.32 (1.22, 1.43) /
2006 1.16 (1.07, 1.26) 0.98 (0.79, 1.23)
2007 1.22 (1.12, 1.32) 0.91 (0.74, 1.11)
2008 1.13 (1.03, 1.24) 0.93 (0.76, 1.14)
2009 1.30 (1.19, 1.42) 0.76 (0.60, 0.96)
2010 1.22 (1.11, 1.35) 0.81 (0.58, 1.12)
2011 1.00 (0.89, 1.13) 1.00 (0.58, 1.73)
Core 1996 1.80 (1.65, 1.96) /
1997 1.92 (1.78, 2.07) /
1998 1.75 (1.62, 1.89) /
1999 1.79 (1.65, 1.93) /
2000 1.68 (1.54, 1.82) /
2001 1.51 (1.39, 1.64) /
2002 1.44 (1.33, 1.56) /
2003 1.49 (1.38, 1.61) /
2004 1.34 (1.23, 1.45) /2005 1.36 (1.25, 1.48) /
2006 1.17 (1.07, 1.27) /
2007 1.26 (1.16, 1.38) /
2008 1.19 (1.08, 1.30) /
2009 1.34 (1.23, 1.47) /
2010 1.26 (1.13, 1.39) /
2011 1.00 (0.88, 1.14) /
Sample 1996 1.00 (0.73, 1.37) /
1997 1.22 (0.93, 1.61) /
1998 1.14 (0.89, 1.46) /
1999 1.23 (0.98, 1.53) /2000 0.99 (0.78, 1.26) /
2001 1.09 (0.87, 1.37) /
2002 1.26 (0.99, 1.61) /
2003 0.90 (0.68, 1.18) /
2004 1.05 (0.82, 1.35) /
2005 1.05 (0.82, 1.34) /
2006 1.21 (0.96, 1.51) /
2007 0.93 (0.72, 1.20) /
2008 0.78 (0.57, 1.07) /
2009 1.04 (0.78, 1.39) /
2010 1.00 (0.74, 1.35) /
2011 1.00 (0.73, 1.38) /Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
21/92
21
Figure 2 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, total skin disease
a) EPIDERM
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, all reporters
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, core reporters
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, sample reporters
8/13/2019 Thor Trends 12
22/92
22
b) THOR-GP
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.82
2006 2007 2008 2009 2010 2011
Relativerate
Year
THORGP, all reporters
8/13/2019 Thor Trends 12
23/92
23
Table 7 Relative rates by year, with 95% comparison intervals, allcontact dermatitis (2011 estimate =1)
Relative rates (95% comparison interval)
EPIDERM
Reporter Group Year
All 1996 1.64 (1.50, 1.80)
1997 1.76 (1.62, 1.90)
1998 1.64 (1.51, 1.78)
1999 1.67 (1.54, 1.81)
2000 1.48 (1.36, 1.62)
2001 1.35 (1.24, 1.47)
2002 1.41 (1.30, 1.53)
2003 1.38 (1.26, 1.50)
2004 1.25 (1.15, 1.36)
2005 1.25 (1.14, 1.37)
2006 1.10 (1.00, 1.20)
2007 1.15 (1.05, 1.26)
2008 1.10 (1.00, 1.22)
2009 1.29 (1.17, 1.41)
2010 1.23 (1.11, 1.36)
2011 1.00 (0.88, 1.13)
Core 1996 1.69 (1.53, 1.86)
1997 1.83 (1.68, 1.99)
1998 1.71 (1.57, 1.86)
1999 1.70 (1.56, 1.85)
2000 1.55 (1.41, 1.70)
2001 1.41 (1.29, 1.55)
2002 1.43 (1.31, 1.56)
2003 1.44 (1.32, 1.57)
2004 1.29 (1.19, 1.41)2005 1.28 (1.16, 1.40)
2006 1.12 (1.01, 1.23)
2007 1.19 (1.08, 1.32)
2008 1.15 (1.04, 1.28)
2009 1.33 (1.20, 1.46)
2010 1.26 (1.14, 1.41)
2011 1.00 (0.88, 1.14)
Sample 1996 1.14 (0.81, 1.59)
1997 1.21 (0.89, 1.63)
1998 1.17 (0.88, 1.56)
1999 1.40 (1.11, 1.78)2000 1.04 (0.80, 1.36)
2001 0.95 (0.72, 1.24)
2002 1.32 (1.01, 1.73)
2003 0.88 (0.65, 1.21)
2004 0.90 (0.66, 1.22)
2005 1.05 (0.80, 1.39)
2006 1.00 (0.76, 1.31)
2007 0.78 (0.58, 1.06)
2008 0.71 (0.50, 1.02)
2009 0.91 (0.65, 1.27)
2010 0.93 (0.66, 1.31)
2011 1.00 (0.70, 1.44)Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
24/92
24
Figure 3 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, all contact dermatitis
a) EPIDERM
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, all reporters
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, core reporters
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, sample reporters
8/13/2019 Thor Trends 12
25/92
25
Table 8 Relative rates by year, with 95% comparison intervals,allergic contact dermatitis (2011 estimate =1)
Relative rates (95% comparison interval)
EPIDERM
Reporter Group Year
All 1996 2.39 (2.09, 2.73)
1997 2.77 (2.48, 3.10)
1998 2.29 (2.03, 2.58)
1999 2.27 (2.01, 2.55)
2000 2.26 (2.00, 2.55)
2001 1.84 (1.61, 2.09)
2002 2.12 (1.89, 2.39)
2003 2.06 (1.83, 2.32)
2004 1.63 (1.44, 1.85)
2005 1.65 (1.45, 1.89)
2006 1.55 (1.35, 1.77)
2007 1.35 (1.16, 1.56)
2008 1.35 (1.16, 1.58)
2009 1.33 (1.14, 1.56)
2010 1.34 (1.14, 1.58)
2011 1.00 (0.82, 1.22)
Core 1996 2.64 (2.30, 3.04)
1997 3.07 (2.73, 3.45)
1998 2.59 (2.30, 2.93)
1999 2.51 (2.21, 2.84)
2000 2.53 (2.23, 2.87)
2001 2.11 (1.85, 2.41)
2002 2.25 (1.99, 2.54)
2003 2.26 (1.99, 2.55)
2004 1.79 (1.57, 2.03)2005 1.73 (1.50, 1.99)
2006 1.63 (1.41, 1.88)
2007 1.42 (1.21, 1.66)
2008 1.44 (1.23, 1.70)
2009 1.44 (1.22, 1.69)
2010 1.39 (1.17, 1.66)
2011 1.00 (0.80, 1.24)
Sample 1996 1.17 (0.74, 1.83)
1997 1.34 (0.91, 1.96)
1998 0.91 (0.59, 1.38)
1999 1.08 (0.74, 1.57)2000 1.00 (0.70, 1.44)
2001 0.64 (0.42, 0.99)
2002 1.48 (1.05, 2.09)
2003 1.07 (0.72, 1.58)
2004 0.82 (0.53, 1.28)
2005 1.18 (0.82, 1.69)
2006 1.06 (0.73, 1.54)
2007 0.88 (0.58, 1.33)
2008 0.80 (0.50, 1.27)
2009 0.72 (0.43, 1.21)
2010 0.93 (0.57, 1.50)
2011 1.00 (0.61, 1.64)Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
26/92
26
Figure 4 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, allergic contact dermatitis
0
0.5
1
1.5
2
2.5
3
3.5
4
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, all reporters
0
0.5
1
1.5
2
2.5
3
3.5
4
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, core reporters
0
0.5
1
1.5
2
2.5
3
3.5
4
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, sample reporters
8/13/2019 Thor Trends 12
27/92
27
Table 9 Relative rates by year, with 95% comparison intervals,irritant contact dermatitis (2011 estimate =1)
Relative rates (95% comparison interval)
EPIDERM
Reporter Group Year
All 1996 1.27 (1.12, 1.44)
1997 1.18 (1.05, 1.33)
1998 1.23 (1.10, 1.38)
1999 1.29 (1.15, 1.45)
2000 1.05 (0.92, 1.20)
2001 1.01 (0.89, 1.14)
2002 1.05 (0.93, 1.19)
2003 1.01 (0.89, 1.14)
2004 0.90 (0.80, 1.02)
2005 1.03 (0.90, 1.17)
2006 0.94 (0.83, 1.07)
2007 1.06 (0.94, 1.20)
2008 0.98 (0.86, 1.12)
2009 1.12 (0.98, 1.27)
2010 1.17 (1.03, 1.34)
2011 1.00 (0.86, 1.17)
Core 1996 1.29 (1.13, 1.48)
1997 1.21 (1.07, 1.37)
1998 1.22 (1.08, 1.38)
1999 1.29 (1.14, 1.46)
2000 1.06 (0.93, 1.22)
2001 0.99 (0.86, 1.13)
2002 1.04 (0.91, 1.18)
2003 1.01 (0.89, 1.15)
2004 0.90 (0.79, 1.03)2005 1.02 (0.90, 1.17)
2006 0.95 (0.83, 1.08)
2007 1.10 (0.97, 1.24)
2008 1.01 (0.88, 1.16)
2009 1.15 (1.01, 1.31)
2010 1.21 (1.05, 1.38)
2011 1.00 (0.85, 1.18)
Sample 1996 0.76 (0.41, 1.39)
1997 0.86 (0.51, 1.44)
1998 1.37 (0.95, 1.98)
1999 1.29 (0.90, 1.84)2000 0.95 (0.63, 1.43)
2001 1.24 (0.87, 1.76)
2002 1.30 (0.87, 1.95)
2003 0.94 (0.60, 1.49)
2004 0.89 (0.57, 1.40)
2005 1.04 (0.67, 1.60)
2006 1.01 (0.66, 1.55)
2007 0.70 (0.43, 1.16)
2008 0.69 (0.39, 1.23)
2009 0.71 (0.40, 1.27)
2010 0.78 (0.45, 1.37)
2011 1.00 (0.59, 1.70)Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
28/92
28
Figure 5 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, irritant contact dermatitis
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.61.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, all reporters
0
0.20.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, core reporters
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, sample reporters
8/13/2019 Thor Trends 12
29/92
29
Table 10 Relative rates by year, with 95% comparison intervals,mixed contact dermatitis (2011 estimate =1)
Relative rates (95% comparison interval)
EPIDERM
Reporter Group Year
All 1996 1.28 (1.04, 1.58)
1997 1.41 (1.18, 1.69)
1998 1.29 (1.07, 1.56)
1999 1.28 (1.06, 1.55)
2000 1.10 (0.89, 1.35)
2001 1.00 (0.82, 1.24)
2002 1.24 (1.04, 1.49)
2003 1.28 (1.07, 1.53)
2004 1.28 (1.08, 1.51)
2005 1.06 (0.87, 1.29)
2006 0.84 (0.68, 1.05)
2007 0.99 (0.81, 1.21)
2008 0.91 (0.73, 1.12)
2009 1.22 (0.99, 1.49)
2010 0.97 (0.77, 1.22)
2011 1.00 (0.78, 1.28)
Core 1996 1.39 (1.11, 1.73)
1997 1.53 (1.27, 1.84)
1998 1.42 (1.16, 1.73)
1999 1.38 (1.13, 1.68)
2000 1.22 (0.99, 1.50)
2001 1.18 (0.95, 1.45)
2002 1.35 (1.12, 1.63)
2003 1.45 (1.22, 1.74)
2004 1.39 (1.17, 1.65)2005 1.16 (0.95, 1.42)
2006 0.91 (0.73, 1.13)
2007 1.13 (0.91, 1.40)
2008 0.99 (0.79, 1.24)
2009 1.39 (1.13, 1.70)
2010 0.99 (0.77, 1.27)
2011 1.00 (0.76, 1.31)
Sample 1996 0.57 (0.25, 1.33)
1997 0.74 (0.39, 1.43)
1998 0.63 (0.34, 1.15)
1999 0.64 (0.37, 1.12)2000 0.47 (0.25, 0.89)
2001 0.25 (0.10, 0.61)
2002 0.67 (0.34, 1.31)
2003 0.33 (0.14, 0.80)
2004 0.70 (0.37, 1.32)
2005 0.52 (0.26, 1.05)
2006 0.54 (0.28, 1.05)
2007 0.24 (0.09, 0.65)
2008 0.40 (0.18, 0.91)
2009 0.24 (0.08, 0.74)
2010 0.89 (0.47, 1.70)
2011 1.00 (0.53, 1.90)Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
30/92
30
Figure 6 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, mixed contact dermatitis
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.61.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, all reporters
0
0.20.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, core reporters
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, sample reporters
8/13/2019 Thor Trends 12
31/92
31
Table 11 Relative rates by year, with 95% comparison intervals,contact urticaria (2011 estimate =1)
Relative rates (95% comparison interval)
EPIDERM
Reporter Group Year
All 1996 2.36 (1.66, 3.36)
1997 3.33 (2.58, 4.30)
1998 2.65 (1.98, 3.55)
1999 2.59 (1.91, 3.51)
2000 3.59 (2.80, 4.60)
2001 2.28 (1.70, 3.06)
2002 2.64 (2.02, 3.45)
2003 2.82 (2.18, 3.67)
2004 2.14 (1.60, 2.86)
2005 3.05 (2.33, 3.99)
2006 1.89 (1.35, 2.63)
2007 1.84 (1.29, 2.61)
2008 1.16 (0.75, 1.80)
2009 1.15 (0.74, 1.79)
2010 1.60 (1.08, 2.39)
2011 1.00 (0.59, 1.68)
Core 1996 2.65 (1.85, 3.80)
1997 3.82 (2.96, 4.94)
1998 3.00 (2.24, 4.03)
1999 2.87 (2.10, 3.92)
2000 4.20 (3.27, 5.38)
2001 2.59 (1.92, 3.49)
2002 2.98 (2.28, 3.91)
2003 3.26 (2.52, 4.22)
2004 2.36 (1.76, 3.16)2005 3.46 (2.64, 4.53)
2006 1.89 (1.32, 2.69)
2007 2.02 (1.41, 2.89)
2008 1.19 (0.75, 1.89)
2009 1.18 (0.74, 1.86)
2010 1.70 (1.13, 2.57)
2011 1.00 (0.58, 1.73)
Sample 1996 0.89 (0.21, 3.72)
1997 0.83 (0.20, 3.48)
1998 0.92 (0.29, 2.95)
1999 1.22 (0.48, 3.09)2000 0.69 (0.22, 2.22)
2001 0.71 (0.22, 2.30)
2002 0.75 (0.18, 3.08)
2003 0.34 (0.05, 2.43)
2004 0.92 (0.29, 2.91)
2005 0.73 (0.18, 2.97)
2006 1.88 (0.75, 4.74)
2007 0.81 (0.20, 3.30)
2008 0.83 (0.20, 3.42)
2009 1.02 (0.25, 4.27)
2010 0.98 (0.24, 4.10)
2011 1.00 (0.24, 4.19)Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
32/92
32
Figure 7 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, contact urticaria
0
1
2
3
4
5
6
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, all reporters
0
1
2
3
4
5
6
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, core reporters
0
1
2
3
4
5
6
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, sample reporters
8/13/2019 Thor Trends 12
33/92
33
Table 12 Relative rates by year, with 95% comparison intervals,neoplasia (2011 estimate =1)
Relative rates (95% comparison interval)
EPIDERM
Reporter Group Year
All 1996 1.67 (1.35, 2.06)
1997 1.58 (1.29, 1.94)
1998 1.24 (1.00, 1.55)
1999 1.32 (1.06, 1.64)
2000 1.46 (1.20, 1.79)
2001 1.52 (1.28, 1.80)
2002 1.21 (0.99, 1.47)
2003 1.27 (1.05, 1.54)
2004 1.17 (0.96, 1.43)
2005 1.10 (0.90, 1.34)
2006 1.07 (0.88, 1.30)
2007 1.25 (1.02, 1.54)
2008 1.07 (0.83, 1.38)
2009 1.33 (1.05, 1.67)
2010 1.07 (0.78, 1.45)
2011 1.00 (0.57, 1.74)
Core 1996 4.08 (3.28, 5.08)
1997 3.78 (3.07, 4.66)
1998 2.87 (2.28, 3.61)
1999 3.29 (2.63, 4.11)
2000 3.49 (2.83, 4.31)
2001 3.37 (2.80, 4.05)
2002 2.60 (2.13, 3.19)
2003 2.79 (2.31, 3.36)
2004 2.45 (2.01, 3.00)2005 2.48 (2.03, 3.04)
2006 2.04 (1.64, 2.53)
2007 2.54 (2.03, 3.19)
2008 2.30 (1.78, 2.99)
2009 2.50 (1.94, 3.23)
2010 1.86 (1.30, 2.67)
2011 1.00 (0.30, 3.31)
Sample 1996 0.37 (0.15, 0.92)
1997 0.77 (0.42, 1.42)
1998 0.55 (0.30, 1.01)
1999 0.42 (0.22, 0.79)2000 0.62 (0.36, 1.06)
2001 1.06 (0.68, 1.66)
2002 0.91 (0.52, 1.58)
2003 0.82 (0.47, 1.42)
2004 1.06 (0.66, 1.70)
2005 0.66 (0.36, 1.19)
2006 1.52 (1.02, 2.25)
2007 1.34 (0.86, 2.08)
2008 0.87 (0.48, 1.59)
2009 1.42 (0.84, 2.42)
2010 1.28 (0.74, 2.21)
2011 1.00 (0.54, 1.85)Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
34/92
34
Figure 8 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, neoplasia
0
0.5
1
1.5
2
2.5
3
3.5
44.5
5
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, all reporters
0
0.51
1.5
2
2.5
3
3.5
4
4.5
5
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, core reporters
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, sample reporters
8/13/2019 Thor Trends 12
35/92
35
Table 13 Relative rates by year, with 95% comparison intervals, otherskin* (2011 estimate =1)
Relative rates (95% comparison interval)
EPIDERM
Reporter Group Year
All 1996 2.19 (1.84, 2.60)
1997 2.22 (1.90, 2.59)
1998 1.95 (1.66, 2.30)
1999 1.98 (1.68, 2.34)
2000 2.17 (1.86, 2.53)
2001 1.96 (1.69, 2.28)
2002 1.55 (1.32, 1.83)
2003 1.69 (1.45, 1.98)
2004 1.59 (1.35, 1.86)
2005 1.87 (1.61, 2.17)
2006 1.66 (1.43, 1.93)
2007 1.78 (1.51, 2.09)
2008 1.28 (1.04, 1.58)
2009 1.52 (1.25, 1.84)
2010 1.43 (1.15, 1.79)
2011 1.00 (0.71, 1.41)
Core 1996 3.40 (2.86, 4.03)
1997 3.31 (2.82, 3.88)
1998 2.85 (2.40, 3.38)
1999 3.11 (2.63, 3.69)
2000 3.39 (2.89, 3.97)
2001 2.85 (2.44, 3.33)
2002 2.18 (1.84, 2.58)
2003 2.44 (2.08, 2.87)
2004 2.16 (1.83, 2.55)2005 2.73 (2.34, 3.18)
2006 2.06 (1.75, 2.44)
2007 2.42 (2.04, 2.87)
2008 1.76 (1.41, 2.20)
2009 1.98 (1.61, 2.42)
2010 1.85 (1.45, 2.36)
2011 1.00 (0.63, 1.58)
Sample 1996 0.56 (0.30, 1.06)
1997 0.97 (0.60, 1.55)
1998 0.80 (0.52, 1.22)
1999 0.57 (0.36, 0.91)2000 0.62 (0.40, 0.97)
2001 0.95 (0.64, 1.41)
2002 0.86 (0.54, 1.36)
2003 0.69 (0.43, 1.11)
2004 1.04 (0.70, 1.54)
2005 0.71 (0.45, 1.13)
2006 1.57 (1.13, 2.20)
2007 1.14 (0.77, 1.70)
2008 0.80 (0.48, 1.35)
2009 1.27 (0.81, 2.00)
2010 1.06 (0.64, 1.73)
2011 1.00 (0.60, 1.67)Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model *Other than contact dermatitis
8/13/2019 Thor Trends 12
36/92
36
Figure 9 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, other (than contact dermatitis) skin
0
0.5
1
1.5
2
2.5
3
3.5
4
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, all reporters
0
0.5
1
1.5
2
2.5
3
3.5
4
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, core reporters
0
0.5
1
1.5
2
2.5
3
3.5
4
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
EPIDERM, sample reporters
8/13/2019 Thor Trends 12
37/92
37
3.2.3 WORK-RELATED RESPIRATORY DISEASE
The average annual percentage change in risk of work-related respiratory
disease, as reported by chest physicians to SWORD is shown in Table 14
whilst the relative rates by year are shown in Tables 15 to 20 and Figures 10to 15.
For total respiratory disease, the graphs showing relative rates by year
suggest a slight increase in incidence between 2010 and 2011 (although
confidence intervals are overlapping for the two years) with the overall
average annual percentage decrease in incidence reducing slightly from
-3.6% reported upon last year for the period 1999-2010 to -3.3% (95% CIs:
-4.2,-2.4) for the period 1999-2011, with little variation between core and
sample reporters.
As observed previously, the annual average decrease in the incidence of
asthma was greater than that observed for total respiratory disease at -7.7%
(95% CIs: -9.2, -6.2) with a sharper decline observed for sample compared to
core reporters. Similarly, the addition of the 2011 case reports had little impact
on the overall observed trends for the other respiratory diagnoses investigated
with a smaller (than asthma), but still significant downward trend in the
incidence of mesothelioma (annual average decrease of -2.8% (95% CIs:
-4.4,-1.1), and relatively flat trends for benign pleural plaques,
pneumoconiosis and other respiratory disease (which includes rhinitis,
allergic alveolitis, lung cancer, inhalation accidents and other respiratory
diagnoses not already specified) over the study period.
8/13/2019 Thor Trends 12
38/92
38
Table 14 Average annual percentage change in reported incidence inwork-related respiratory disease
a) All reporters
ESTIMATED % CHANGE
(95% CONFIDENCEINTERVAL)
Year (continuous) SWORD
Total respiratory 1999-2011 -3.3 (-4.2, -2.4)
Asthma 1999-2011 -7.7 (-9.2, -6.2)
Mesothelioma 1999-2011 -2.8 (-4.4, -1.1)
Benign pleural plaques 1999-2011 -0.7 (-2.0, 0.6)
Pneumoconiosis 1999-2011 -0.3 (-2.9, 2.3)Other* respiratory disease 1999-2011 0.5 (-1.5, 2.6)
b) Core reporters
ESTIMATED % CHANGE(95% CONFIDENCEINTERVAL)
Year (continuous) SWORD
Total respiratory 1999-2011 -3.7 (-4.7, -2.6)
Asthma 1999-2011 -7.4 (-9.0, -5.7)
Mesothelioma 1999-2011 -1.6 (-4.0, 0.8)
Benign pleural plaques 1999-2011 -1.7 (-3.2, -0.1)
Pneumoconiosis 1999-2011 0.7 (-2.4, 3.9)Other* respiratory disease 1999-2011 0.3 (-2.1, 2.8)
c) Sample reporters
ESTIMATED % CHANGE(95% CONFIDENCEINTERVAL)
Year (continuous) SWORD
Total respiratory 1999-2011 -2.5 (-4.1, -0.9)
Asthma 1999-2011 -9.1 (-12.7, -5.3)
Mesothelioma 1999-2011 -4.0 (-6.3, -1.6)
Benign pleural plaques 1999-2011 1.8 (-0.8, 4.5)
Pneumoconiosis 1999-2011 -3.2 (-7.8, 1.7)Other* respiratory disease 1999-2011 1.1 (-2.8, 5.2)
8/13/2019 Thor Trends 12
39/92
39
Table 15 Relative rates by year, with 95% comparison intervals, totalrespiratory disease (2011 estimate =1)
Relative rates (95% comparison interval)
SWORD
Reporter Group Year
All 1999 1.35 (1.23, 1.47)
2000 1.30 (1.19, 1.41)
2001 1.28 (1.17, 1.39)
2002 1.31 (1.20, 1.43)
2003 1.30 (1.20, 1.42)
2004 1.22 (1.12, 1.33)
2005 1.16 (1.06, 1.27)
2006 1.07 (0.98, 1.18)
2007 0.94 (0.85, 1.05)
2008 1.01 (0.91, 1.13)
2009 0.96 (0.86, 1.07)
2010 0.92 (0.82, 1.04)
2011 1.00 (0.89, 1.12)
Core 1999 1.35 (1.22, 1.50)
2000 1.34 (1.21, 1.48)
2001 1.37 (1.25, 1.51)
2002 1.44 (1.30, 1.59)
2003 1.47 (1.34, 1.61)
2004 1.36 (1.24, 1.50)
2005 1.22 (1.10, 1.34)
2006 1.10 (0.99, 1.22)
2007 0.92 (0.81, 1.05)
2008 1.00 (0.88, 1.14)
2009 0.94 (0.82, 1.09)
2010 0.95 (0.83, 1.10)2011 1.00 (0.87, 1.15)
Sample 1999 1.40 (1.19, 1.65)
2000 1.24 (1.05, 1.47)
2001 1.07 (0.89, 1.29)
2002 1.03 (0.85, 1.24)
2003 0.89 (0.73, 1.09)
2004 0.87 (0.71, 1.06)
2005 1.08 (0.90, 1.28)
2006 1.05 (0.88, 1.26)
2007 0.99 (0.82, 1.19)
2008 1.01 (0.84, 1.22)2009 0.98 (0.80, 1.19)
2010 0.84 (0.68, 1.05)2011 1.00 (0.81, 1.23)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
40/92
40
Figure 10 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, total respiratory disease
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.61.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, all reporters
0
0.2
0.40.6
0.8
1
1.2
1.4
1.6
1.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
R
elativerate
Year
SWORD, core reporters
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.61.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, sample reporters
8/13/2019 Thor Trends 12
41/92
41
Table 16 Relative rates by year, with 95% comparison intervals,asthma (2011 estimate =1)
Relative rates (95% comparison interval)
SWORD
Reporter Group Year
All 1999 2.56 (2.21, 2.98)
2000 1.92 (1.64, 2.26)
2001 2.04 (1.74, 2.38)
2002 2.10 (1.78, 2.48)
2003 2.06 (1.74, 2.43)
2004 1.92 (1.61, 2.29)
2005 1.72 (1.43, 2.07)
2006 1.63 (1.37, 1.94)
2007 1.13 (0.91, 1.40)
2008 1.21 (0.98, 1.49)
2009 0.93 (0.73, 1.18)
2010 0.95 (0.75, 1.22)
2011 1.00 (0.78, 1.28)
Core 1999 2.34 (1.97, 2.77)
2000 1.74 (1.45, 2.09)
2001 2.02 (1.71, 2.39)
2002 2.09 (1.75, 2.50)
2003 2.09 (1.75, 2.50)
2004 1.97 (1.64, 2.38)
2005 1.69 (1.38, 2.07)
2006 1.51 (1.25, 1.84)
2007 1.08 (0.85, 1.36)
2008 1.13 (0.90, 1.43)
2009 0.89 (0.68, 1.16)
2010 0.90 (0.69, 1.18)2011 1.00 (0.77, 1.31)
Sample 1999 4.53 (3.26, 6.30)
2000 3.33 (2.28, 4.85)
2001 2.10 (1.30, 3.40)
2002 2.19 (1.36, 3.54)
2003 1.79 (1.08, 2.98)
2004 1.54 (0.89, 2.67)
2005 2.04 (1.28, 3.25)
2006 2.65 (1.70, 4.13)
2007 1.58 (0.91, 2.74)
2008 1.89 (1.13, 3.17)2009 1.21 (0.62, 2.33)
2010 1.44 (0.77, 2.70)2011 1.00 (0.47, 2.12)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
42/92
42
Figure 11 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, asthma
0
1
2
3
4
5
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, all reporters
0
1
2
3
4
5
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
R
elativerate
Year
SWORD, core reporters
0
1
2
3
4
5
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, sample reporters
8/13/2019 Thor Trends 12
43/92
43
Table 17 Relative rates by year, with 95% comparison intervals,mesothelioma (2011 estimate =1)
Relative rates (95% comparison interval)
SWORD
Reporter Group Year
All 1999 1.40 (1.19, 1.65)
2000 1.41 (1.21, 1.65)
2001 1.40 (1.20, 1.64)
2002 1.39 (1.18, 1.63)
2003 1.34 (1.14, 1.57)
2004 1.18 (1.01, 1.39)
2005 1.04 (0.87, 1.24)
2006 1.04 (0.87, 1.25)
2007 1.32 (1.09, 1.61)
2008 1.30 (1.06, 1.59)
2009 1.16 (0.93, 1.45)
2010 1.08 (0.86, 1.36)
2011 1.00 (0.78, 1.28)
Core 1999 1.15 (0.92, 1.43)
2000 1.28 (1.05, 1.56)
2001 1.26 (1.03, 1.53)
2002 1.28 (1.05, 1.56)
2003 1.27 (1.05, 1.55)
2004 1.16 (0.96, 1.41)
2005 0.96 (0.77, 1.19)
2006 1.02 (0.81, 1.29)
2007 1.15 (0.87, 1.52)
2008 1.47 (1.12, 1.94)
2009 1.22 (0.91, 1.63)
2010 1.03 (0.75, 1.41)2011 1.00 (0.72, 1.39)
Sample 1999 1.75 (1.36, 2.24)
2000 1.56 (1.20, 2.02)
2001 1.58 (1.22, 2.05)
2002 1.48 (1.12, 1.95)
2003 1.34 (1.01, 1.77)
2004 1.10 (0.81, 1.49)
2005 1.13 (0.84, 1.51)
2006 1.04 (0.76, 1.42)
2007 1.54 (1.18, 2.00)
2008 1.12 (0.82, 1.51)2009 1.11 (0.80, 1.54)
2010 1.17 (0.84, 1.62)2011 1.00 (0.69, 1.45)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
44/92
44
Figure 12 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, mesothelioma
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.61.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, all reporters
0
0.2
0.40.6
0.8
1
1.2
1.4
1.6
1.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
R
elativerate
Year
SWORD, core reporters
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.61.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, sample reporters
8/13/2019 Thor Trends 12
45/92
45
Table 18 Relative rates by year, with 95% comparison intervals (2011estimate =1), benign pleural plaques
Relative rates (95% comparison interval)
SWORD
Reporter Group Year
All 1999 0.92 (0.80, 1.06)
2000 1.10 (0.98, 1.25)
2001 1.07 (0.94, 1.21)
2002 1.09 (0.95, 1.24)
2003 1.14 (1.02, 1.29)
2004 1.03 (0.92, 1.16)
2005 1.09 (0.98, 1.22)
2006 0.95 (0.84, 1.08)
2007 0.88 (0.76, 1.02)
2008 0.94 (0.81, 1.10)
2009 0.89 (0.75, 1.05)
2010 1.01 (0.86, 1.18)
2011 1.00 (0.85, 1.18)
Core 1999 0.98 (0.84, 1.14)
2000 1.21 (1.05, 1.39)
2001 1.23 (1.08, 1.40)
2002 1.24 (1.08, 1.43)
2003 1.30 (1.14, 1.48)
2004 1.15 (1.01, 1.31)
2005 1.14 (1.01, 1.29)
2006 0.92 (0.80, 1.06)
2007 0.89 (0.75, 1.06)
2008 0.89 (0.74, 1.07)
2009 0.88 (0.72, 1.08)
2010 1.06 (0.88, 1.29)2011 1.00 (0.82, 1.22)
Sample 1999 0.95 (0.70, 1.27)
2000 0.92 (0.69, 1.24)
2001 0.67 (0.48, 0.95)
2002 0.72 (0.51, 1.02)
2003 0.74 (0.53, 1.02)
2004 0.77 (0.56, 1.05)
2005 1.17 (0.92, 1.49)
2006 1.18 (0.92, 1.52)
2007 0.89 (0.67, 1.19)
2008 1.09 (0.84, 1.41)2009 0.91 (0.68, 1.23)
2010 0.88 (0.64, 1.20)2011 1.00 (0.74, 1.36)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
46/92
46
Figure 13 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, benign pleural plaques
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.61.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, all reporters
0
0.2
0.40.6
0.8
1
1.2
1.4
1.6
1.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
R
elativerate
Year
SWORD, core reporters
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.61.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, sample reporters
8/13/2019 Thor Trends 12
47/92
47
Table 19 Relative rates by year, with 95% comparison intervals,pneumoconiosis (2011 estimate =1)
Relative rates (95% comparison interval)
SWORD
ReporterGroup
Year
All 1999 0.84 (0.66, 1.06)
2000 0.79 (0.62, 1.00)
2001 0.74 (0.58, 0.94)
2002 0.65 (0.50, 0.85)
2003 0.70 (0.54, 0.89)
2004 0.56 (0.42, 0.74)
2005 0.68 (0.54, 0.87)
2006 0.77 (0.60, 0.98)
2007 0.53 (0.38, 0.74)
2008 0.69 (0.50, 0.95)
2009 0.84 (0.62, 1.15)
2010 0.67 (0.48, 0.95)
2011 1.00 (0.75, 1.33)
Core 1999 0.74 (0.56, 0.98)
2000 0.78 (0.61, 1.02)
2001 0.67 (0.51, 0.89)
2002 0.70 (0.52, 0.94)
2003 0.74 (0.57, 0.96)
2004 0.64 (0.48, 0.85)
2005 0.69 (0.52, 0.91)
2006 0.80 (0.62, 1.05)
2007 0.47 (0.31, 0.71)
2008 0.68 (0.47, 0.98)
2009 0.89 (0.63, 1.27)
2010 0.75 (0.51, 1.11)
2011 1.00 (0.71, 1.41)
Sample 1999 1.31 (0.84, 2.03)
2000 0.78 (0.44, 1.38)
2001 1.04 (0.63, 1.71)
2002 0.41 (0.18, 0.92)
2003 0.47 (0.22, 0.99)
2004 0.23 (0.08, 0.60)
2005 0.68 (0.38, 1.22)
2006 0.64 (0.35, 1.17)
2007 0.66 (0.37, 1.18)2008 0.71 (0.39, 1.29)
2009 0.70 (0.37, 1.30)
2010 0.43 (0.20, 0.96)2011 1.00 (0.58, 1.73)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
8/13/2019 Thor Trends 12
48/92
48
Figure 14 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, pneumoconiosis
0
0.5
1
1.5
2
2.5
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, all reporters
0
0.5
1
1.5
2
2.5
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
R
elativerate
Year
SWORD, core reporters
0
0.5
1
1.5
2
2.5
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, sample reporters
8/13/2019 Thor Trends 12
49/92
49
Table 20 Relative rates by year, with 95% comparison intervals,other* respiratory disease (2011 estimate =1)
Relative rates (95% comparison interval)
SWORD
Reporter Group Year
All 1999 0.92 (0.72, 1.18)
2000 0.90 (0.71, 1.14)
2001 0.93 (0.73, 1.17)
2002 1.02 (0.80, 1.29)
2003 0.97 (0.77, 1.22)
2004 1.12 (0.92, 1.38)
2005 0.81 (0.64, 1.03)
2006 1.07 (0.87, 1.32)
2007 0.77 (0.60, 1.00)
2008 1.07 (0.84, 1.36)
2009 1.22 (0.98, 1.52)
2010 0.81 (0.61, 1.07)
2011 1.00 (0.78, 1.29)
Core 1999 1.14 (0.86, 1.51)
2000 1.01 (0.76, 1.34)
2001 1.04 (0.78, 1.38)
2002 1.29 (0.98, 1.70)
2003 1.24 (0.96, 1.59)
2004 1.42 (1.14, 1.78)
2005 0.98 (0.75, 1.28)
2006 1.40 (1.11, 1.76)
2007 0.93 (0.69, 1.26)
2008 1.35 (1.03, 1.77)
2009 1.42 (1.09, 1.85)
2010 1.08 (0.79, 1.46)2011 1.00 (0.72, 1.39)
Sample 1999 0.53 (0.32, 0.88)
2000 0.68 (0.43, 1.06)
2001 0.72 (0.47, 1.13)
2002 0.52 (0.31, 0.87)
2003 0.52 (0.31, 0.88)
2004 0.62 (0.39, 0.98)
2005 0.52 (0.32, 0.87)
2006 0.50 (0.30, 0.83)
2007 0.49 (0.29, 0.81)
2008 0.59 (0.36, 0.96)2009 0.82 (0.54, 1.25)
2010 0.36 (0.19, 0.69)2011 1.00 (0.65, 1.53)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model*Other than those specified above
8/13/2019 Thor Trends 12
50/92
50
Figure 15 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, other respiratory disease
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.61.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, all reporters
0
0.2
0.40.6
0.8
1
1.2
1.4
1.6
1.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
R
elativerate
Year
SWORD, core reporters
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.61.8
2
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Relativerate
Year
SWORD, sample reporters
8/13/2019 Thor Trends 12
51/92
51
3.2.4 WORK-RELATED MUSCULOSKELETAL DISORDERS
The average annual percentage change in reported incidence of work-related
musculoskeletal disorders (MSDs), as reported by GPs (THOR-GP) is shown
in Table 21 whilst the relative rates by year are shown in Tables 22 to 25 and
Figures 16 to 19. Data from GPs suggested a downward trend in the
incidence of total work-related MSDs in the order of -15.4% (95% CIs: -18.6,
-12.0) per year. For comparison, the annual average decrease based on data
for 2006-2010 was -15.2% (95% CIs: -18.5, -11.6). The pattern for upper limb
disorders was very similar to that observed for total MSDs, whilst for
spine/back disorders generally a steeper annual decrease was observed. A
downward trend in the incidence of lower limb disorders was also observed
(although this was not statistically significant). The graphs showing relativerates by year show a sharp drop in incidence for THOR-GP between 2006
and 2007 with rates continuing to fall thereafter.
Table 21 Average annual percentage change in reported incidence inwork-related musculoskeletal disorders
ESTIMATED % CHANGE(95% CONFIDENCEINTERVAL)
Reportergroup
Year (continuous) THOR-GP
All Total musculoskeletal 2006-2011 -15.4 (-18.6, -12.0)
Upper limb 2006-2011 -15.2 (-19.5, -10.7)
Spine/back 2006-2011 -19.2 (-24.1, -14.1)Lower limb 2006-2011 -8.7 (-17.4, 0.9)
8/13/2019 Thor Trends 12
52/92
52
Table 22 Relative rates by year, with 95% comparison intervals, totalmusculoskeletal disorders (2011 estimate = 1)
Relative rates (95% comparisoninterval)
Reporter Group THOR-GP
All 2006 2.51 (2.27, 2.78)
2007 1.85 (1.69, 2.03)
2008 1.74 (1.57, 1.92)
2009 1.47 (1.33, 1.63)
2010 1.27 (1.07, 1.52)2011 1.00 (0.77, 1.31)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
Figure 16 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, total musculoskeletal disorders
0
0.5
1
1.5
2
2.5
3
3.5
4
2006 2007 2008 2009 2010 2011
Relativerate
Year
THOR-GP, all reporters
8/13/2019 Thor Trends 12
53/92
53
Table 23 Relative rates by year, with 95% comparison intervals,upper limb disorders (2011 estimate = 1)
Relative rates (95% comparisoninterval)
Reporter Group THOR-GP
All 2006 2.98 (2.58, 3.44)
2007 2.23 (1.95, 2.55)
2008 1.97 (1.70, 2.30)
2009 1.87 (1.61, 2.18)
2010 1.48 (1.20, 1.84)2011 1.00 (0.64, 1.56)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
Figure 17 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, upper limb disorders
0
0.5
1
1.5
2
2.5
3
3.5
4
2006 2007 2008 2009 2010 2011
Relativerate
Year
THOR-GP, all reporters
8/13/2019 Thor Trends 12
54/92
54
Table 24 Relative rates by year, with 95% comparison intervals,spine/back disorders (2011 estimate = 1)
Relative rates (95%comparison interval)
Reporter Group THOR-GP
All 2006 2.68 (2.30, 3.14)
2007 2.01 (1.75, 2.30)
2008 1.86 (1.53, 2.26)
2009 1.30 (1.07, 1.58)
2010 1.27 (0.99, 1.63)2011 1.00 (0.63, 1.59)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
Figure 18 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, spine/back disorders
0
0.5
1
1.5
2
2.5
3
3.5
4
2006 2007 2008 2009 2010 2011
Relativerate
Year
THOR-GP, all reporters
8/13/2019 Thor Trends 12
55/92
55
Table 25 Relative rates by year, with 95% comparison intervals, lowerlimb disorders (2011 estimate = 1)
Relative rates (95%comparison interval)
ReporterGroup
THOR-GP
All 2006 1.64 (1.22, 2.21)
2007 1.10 (0.83, 1.44)2008 1.27 (0.97, 1.66)
2009 1.16 (0.88, 1.54)
2010 0.99 (0.69, 1.43)2011 1.00 (0.48, 2.10)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
Figure 19 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, lower limb disorders
0
0.5
1
1.5
2
2.5
3
3.5
4
2006 2007 2008 2009 2010 2011
Relativerate
Year
THOR-GP, all reporters
8/13/2019 Thor Trends 12
56/92
56
3.2.5 WORK-RELATED MENTAL ILL-HEALTH
The average annual percentage change in reported incidence of work-related
mental ill-health, as reported by GPs (THOR-GP) is shown in Table 26 whilst
the relative rates by year are shown in Tables 27 to 29 and Figures 20 to 22.
Overall an average annual decrease in the incidence of total mental ill-health
was observed of -9.2% (95% CIs: -13.5, -4.7). This compared to a decrease of
-8.8% (95% CIs: -13.1, -4.1) observed previously (2006-2010). However, the
graphs showing relative rates by year suggest the previously observed
increase in incidence in 2010 has been followed by a decrease in incidence in
2011 (although it must be noted that the confidence intervals for the 2010 and
2011 estimates overlap the other years). A similar pattern was seen for
anxiety and depression, and other work stress.
Table 26 Average annual percentage change in reported incidence inwork-related mental ill-health
ESTIMATED % CHANGE(95% CONFIDENCE
INTERVAL)Reportergroup
Year (continuous) THOR-GP
All Total mental ill-health 2006-2011 -9.2 (-13.5, -4.7)
Anxiety and depression 2006-2011 -7.7 (-14.2, -0.6)Other work stress 2006-2011 -10.7 (-15.9, -5.2)
8/13/2019 Thor Trends 12
57/92
57
Table 27 Relative rates by year, with 95% comparison intervals, totalmental ill-health (2011 estimate = 1)
Relative rates (95%comparison interval)
ReporterGroup
THOR-GP
All 2006 1.43 (1.26, 1.63)
2007 1.29 (1.15, 1.45)2008 1.17 (1.02, 1.35)
2009 0.92 (0.79, 1.06)
2010 1.41 (1.21, 1.65)2011 1.00 (0.71, 1.40)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
Figure 20 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, total mental ill-health
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
2006 2007 2008 2009 2010 2011
Relativerate
Year
THOR-GP, all reporters
8/13/2019 Thor Trends 12
58/92
58
Table 28 Relative rates by year, with 95% comparison intervals,anxiety and depression (2011 estimate = 1)
Relative rates (95%comparison interval)
ReporterGroup
THOR-GP
All 2006 1.54 (1.26, 1.87)
2007 1.46 (1.22, 1.73)2008 1.21 (1.00, 1.48)
2009 1.08 (0.88, 1.33)
2010 1.68 (1.29, 2.19)2011 1.00 (0.63, 1.59)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
Figure 21 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, anxiety and depression
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
2006 2007 2008 2009 2010 2011
Re
lativerate
Year
THOR-GP, all reporters
8/13/2019 Thor Trends 12
59/92
59
Table 29 Relative rates by year, with 95% comparison intervals, otherwork stress (2011 estimate = 1)
Relative rates (95%comparison interval)
ReporterGroup
THOR-GP
All 2006 1.36 (1.16, 1.60)
2007 1.18 (1.02, 1.36)2008 1.16 (1.00, 1.35)
2009 0.81 (0.68, 0.96)
2010 1.16 (0.93, 1.44)2011 1.00 (0.70, 1.43)
Models adjusted for reporter type (where appropriate), season and harvestingPopulation offset included in the model
Figure 22 Relative rates by year (2011 estimate = 1), with 95%comparison intervals, other work stress
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
2006 2007 2008 2009 2010 2011
Relativerate
Year
THOR-GP, all reporters
8/13/2019 Thor Trends 12
60/92
60
4 DISCUSSION
This is the latest report describing temporal trends in incidence of WRI as
reported by physicians to the THOR surveillance network. For the current
report, data from NHS hospital based specialist consultants reporting to
EPIDERM (skin) and SWORD (respiratory), and from GPs reporting to THOR-
GP (all WRI) were used to estimate trends for the period 1996-2011
(EPIDERM), 1999-2011 (SWORD) and 2006-2011 (THOR-GP). HSE funding
for data collection for the other main extant THOR scheme, OPRA
(occupational physicians), ceased at the end of 2010, with trends for that
period (2006-2010) reported upon previously2.
There have been no changes to the methodology reported on here in respect
of trends calculations since the 2011 trends report was submitted and
changes made at that point and earlier are discussed in full in previous
reports2-7. However, one important issue that was discussed in the 2011
report that warrants further comment here was the substantial change to the
THOR-GP sampling methodology, which moved from >90% core (reporting on
a monthly basis) prior to 2010 to 78% sample (reporting one randomly chosen
month per year) in 2010. By way of background it should be restated thatwhen THOR-GP was launched in 2005 it started generating reports at a far
higher level than specified in the contract (so as to pump prime the database).
The data collection in 2010 was as per the original contract. Moreover, since
HSE funding was limited, the participating GPs were then no longer paid an
honorarium for these reports. Therefore, for the present round of analyses, a
further shift occurred with 100% of the GPs reporting in 2011 reporting on a
sample basis. Initial investigations of case reporting in THOR-GP suggestedthat physicians reporting as sample in 2010 and 2011 reported, on average,
three times as many cases (in any one month) than when reporting as core
reporters in 2009. Work is currently underway to investigate this rate
difference between core and sample. We had previously reported the same
phenomenon with OPRA11 but with a much smaller magnitude. One
hypothesis that has been put forward is that some of the cases reported by
GPs might not be truly incident. It is likely that a patient will make more
frequent visits to the GP than to other physicians, possibly for further sickness
8/13/2019 Thor Trends 12
61/92
61
certification. Therefore a GP may report a case in their reporting month that
they first saw in clinic with the same problem in a previous month (i.e.
harvesting). This would produce an overestimate in rates as such a case is
prevalent, rather than incident. This is more likely to happen for those only
reporting one month a year (sample) compared to continuous (core) reporters.
However, further work is required to understand this issue fully.
As discussed previously2the approach adopted in the MLM analyses was to
treat a physician as a new reporter if they changed from core to sample (or
vice versa). GPs reporting for the first time as sample reporters in 2010 would
therefore only have two data points and those reporting as sample for the first
time in 2011 would only have one data point and would therefore (in theory)
contribute less to the overall trend. Analyses reported upon previously for the
period 2006-2010 did appear to confirm this with estimates from the THOR-
GP MLM in which sample data were excluded being very similar to those in
which all data were included. Core reporting stopped at the end of 2010 so
this exercise was not repeated in the current round of analyses. It would also
be informative to investigate trends based on sample data only but at present
numbers are too small to yield any meaningful results (63% of sample
reporters have two data points, with a further 30% having one data point only).In view of the small numbers and until the issue of change in time sampling
pattern is more fully understood, the THOR-GP trend estimates need to be
interpreted with caution.
THOR trends methodology has progressively addressed relevant artefacts
(e.g. changing denominators). The main residual issue which has been much
discussed within the trends project is the issue of reporter fatigue. Initialinvestigations primarily focussed on determining whether the proportion of
responses that are zero returns (i.e. declaring I have nothing to report) or the
proportion of non-response had increased over time (both of which could be
an indication of reporter fatigue)12, 13. In brief, for the specialist schemes,
results suggested some evidence that the proportion of both non-response
and zero returns increased with membership time, whilst for THOR-GP, the
proportion of zero returns but not non-response increased with membership
time. Following this it was recognised that further work in this area was
8/13/2019 Thor Trends 12
62/92
62
required, and as such, additional HSE funding was secured to address this
and other issues relating to THOR trends and incidence. This comprised 6
work packages, 2 of which related to specifically to the issue of fatigue14-16.
The first of these built on the earlier work investigating changes in the
probability of a zero-return or non-response over time. The other work
package relevant to this issue investigated whether there was evidence of
fatigue manifesting as an excess of zeros in the THOR data and if so, how
this could best be modelled. The results suggested evidence of zero-inflation
in some schemes/reporter groups (in EPIDERM both core and sample
reporters and in OPRA sample reporters only) but not others (THOR-GP,
SWORD, and OPRA core reporters). However, an excess of zeros would only
impact on the trends analysis if there was further evidence that the
percentage of false zeros changed over time and such evidence was found for
EPIDERM sample reporters only.
As in previous reports, the estimates presented here have not been adjusted
for the potential effects of fatigue. The latest body of work regarding this issue
has only fairly recently been completed, with the findings and their
implications for future analyses still under discussion within the THOR team
and in the THOR steering group. However, it is planned that the 2013 trendsreport will provide further updates regarding this issue.
An abridged commentary by category of illness is provided in the following
sections.
SKIN (EPIDERM and THOR-GP): The primary source of THOR data of work-
related skin disease is reports from dermatologists to EPIDERM, with trendsin incidence for this group first described in a report submitted to HSE in
20057. Overall, the addition of each successive year of data has not greatly
affected the estimated average annual change in incidence of total work-
related skin disease which has remained at around a 3% decrease per year.
However, the plots showing relative rates have suggested some variation in
incidence between years. Most notable in the current round of analyses was
the relatively large drop in incidence between 2010 and 2011 compared to
changes between other years. This may reflect (at least in part) changes in
8/13/2019 Thor Trends 12
63/92
63
reporting methods between the two years and has therefore to be treated very
cautiously. Prior to 2011, dermatologists could submit their case reports to
EPIDERM using either the traditional paper based report card (which would
be sent to them in the post prior to their reporting month) or electronically via a
web form. For 2011 however, it was decided not to send the report card in the
post as usual but to instead email the physicians an electronic version of the
card which they could then complete and return (with the option to report via a
web form also still available). This was both an attempt to increase the
proportion of physicians reporting electronically (with the obvious benefits of
reducing transcription errors etc) and of reducing the costs incurred through
the printing and postage of the traditional report cards (HSE funding for 2011
EPIDERM data collection had not been secured at the time). However,
response rates for 2011 were notably lower than in previous years, particularly
for core reporters, and feedback from the physicians suggested that they
found the traditional report cards much more practical for their day to day
practice. In view of this, the cards were reinstated for reporting in 2012.
The trend in incidence observed for dermatologist reported contact dermatitis
(CD) was very similar to that observed for skin overall, unsurprising given that
case reports of CD comprise the majority of the skin reports. The largerdecrease in incidence observed for allergic CD compared to irritant CD has
been discussed in previous reports2,3with the suggestion that it may reflect
steps such as the reduction in use of powdered latex gloves following
Government interventions (introduced between 1996 and 1998) aiming to
reduce exposure to latex. This theory is further corroborated by the observed
change in the incidence of (dermatologist reported) contact urticaria (CU)
(which would also be expected to decrease if the latex interventions weresuccessful) which shows a decline over a similar period. Indeed, as the quality
and quantity of the data increases, it has been possible for THOR to apply the
trends methodology to investigate the impact of Government (or other)
interventions aiming to reduce the incidence of WRI attributed to specific
agents such as latex and chromium17, 18.
The other main sub-group of skin diagnoses reported by dermatologists to
EPIDERM was neoplasia, for which a smaller (compared to CD and CU), but
8/13/2019 Thor Trends 12
64/92
64
still statistically significant, annual decrease in incidence was observed.
However, there was a large drop in the number of case reports of neoplasia
(particularly from core physicians) in 2011 compared to previous years,
reflected in the results by the wide confidence intervals around the 2011
estimate. This drop in 2011 may partly reflect the overall decline in response
rates arising from the cessation of the paper based report card. However,
different trends have been observed between reporter type (core and sample)
previously2, with core data suggesting a decrease in incidence and sample
data suggesting an increase in incidence. This may be because core
EPIDERM reporters main area of expertise has historically been and probably
still is in occupational contact dermatitis, so skin neoplasia referrals may be
triaged to other (sample) reporters.
A decrease in the incidence of GP reported work-related skin disease was
also observed which, although not statistically significant, was larger than that
seen for dermatologists over the same time period. However, skin reports
comprise a relatively small proportion of the total cases reported to THOR-GP,
and this coupled with the (as yet unknown) impact on the trend estimate of the
previously discussed changes in sampling frequency (i.e. the move from
predominantly core to predominantly sample), means it is important to becautious when interpreting the trend estimates for THOR-GP.
RESPIRATORY (SWORD): Trends in incidence based on data from
respiratory physicians have also been described since the initial report
submitted to HSE in 20067. Compared to dermatologists however, the addition
of each successive year of data appears to have had more of an impact on
the annual average estimated change in incidence of total work-relatedrespiratory disease from an initial decrease of around 1% per year (based on
data for the period 1999-2004) to a decrease of around 3% observed in the
current round of analyses. The greater variability (compared to skin) may
reflect differences in case mix (and changes in case mix over time) between
the two groups of reporters. Whilst case reports to EPIDERM are
predominantly CD and neoplasia (and have been throughout the study
period), case reports to SWORD encompass a wider diagnostic range with the
proportion of the total cases attributed to each diagnosis exhibiting some
8/13/2019 Thor Trends 12
65/92
65
variation throughout the study period. Therefore, for respiratory disease (as
reported by chest physicians), it is probably more informative to look at
changes in incidence for the individual respiratory sub-groups.
Regarding asthma, the average annual percentage change in reported
incidence has remained at about a 7% decrease per year with much of this
decrease appearing to have been driven by a fall in incidence in earlier years,
with a relatively flat trend apparent since 2007. As observed for skin disease,
this decrease in incidence may, in part, reflect the introduction of Government
(and other) initiatives targeting this disease category. Recent work has begun
to investigate the impact of changes in legislation, regulatory activity and
market forces on the incidence of workplace asthma in the UK19.
In addition to asthma, changes in the incidence of the (primarily) asbestos
related diseases, namely, mesothelioma, benign pleural plaques and
pneumoconiosis were also (separately) investigated. For all three groups, an
overall decrease in incidence was observed (but this was only statistically
significant for mesothelioma). The reported decline in the incidence of