Investigating tiredness in Australian general practice. Do pathology tests help in diagnosis?

Abstract

Tiredness is a common presentation in general practice for which pathology tests are commonly ordered. Our aim was to study their utilisation for tiredness.
We examined an integrated database which contains the medical records for 58,139 patients and their 696,518 associated general practitioner encounters. Three hundred and forty-two patients and their 1652 associated encounters were randomly selected out of 12,291 patients and their 26,748 associated encounters that had mentioned tiredness (or a synonym).
One hundred and eighty-one patients (53%) had at least one pathology test ordered at any time in their episode of care. Patients over 60 years of age, patients who consulted their GP more than once and patients without comorbidity were more likely to have a pathology test ordered. Only 12 patients (3%) had a significant clinical diagnosis based on an abnormal pathology test.
Pathology testing for patients presenting with tiredness is high. Most tests do not yield a significant clinical diagnosis.

Full text

Page 1

Reprinted from Australian Family Physician Vol. 32, No. 8, August 2003 • 663
A
practice, it is rarely associated with serious
disease.1–3It is the seventeenth most fre-
quent reason for a patient to consult an
Australian general practitioner4and the
fourth most common problem associated
with pathology testing. The rate of ordering
increased from 300 test orders per 100 prob-
lems in 1998–1999 to 355 in 2001–2002.4–7
Nevertheless, tiredness does not rank
among the GP reported 30 most commonly
managed problems in general practice.4–7
The diagnostic challenge for the GP is to
detect physical causes and serious disease
without excessive medical investigation.
Tiredness can indicate a number of
physical or psychological disorders. Almost
lthough tiredness is a common
symptom among patients in general
any medical condition can cause tiredness.
Psychosocial illnesses are the most
common causes.8–11About half of consulta-
tions with Australian GPs for tiredness
result in a pathology test (the most
common being a full blood examination).12
Yet confirmation of diagnosis is rare.13–15
There is limited Australian research
about the appropriateness of pathology
tests for tiredness, or the impact investi-
gations have on patient management. In
Britain, 33% of patients with tiredness
had one or more abnormal result from a
pathology test, but only 9% were clini-
cally important.13
We aimed to describe the use of
pathology ordering in patients who present
to the Australian GP with tiredness.
Methods
We retrospectively examined case notes
audited over seven years (1 April 1994 to
25 April 2001) of patients enrolled in six
general practices in South Australia,
Western Australia, Victoria and the
Australian Capital Territory with com-
puterised medical record systems. Clinical
information from participating practices
has been consolidated into a comput-
erised database (Medic-GP) of all patient
clinical encounters including consultation,
pathology investigations, diagnostic
imaging and medications prescribed.16
Anonymity of patient records was pre-
served because patient names were
replaced by record numbers. Patients in
the Medic-GP database are representa-
Investigating tiredness in Australian general practice
Do pathology tests help in diagnosis?
Angela Gialamas,Justin J Beilby,Nicole L Pratt,Rhys Henning,John E Marley,John F Roddick
Angela Gialamas, BHSc, is Project Officer, Department of General Practice, The University of Adelaide, South Australia.
Justin J Beilby, MBBS, MD, FRACGP, is Professor of General Practice, Department of General Practice,
The University of Adelaide, South Australia.
Nicole L Pratt, BSc, is a statistician, Department of General Practice, The University of Adelaide, South Australia.
Rhys Henning, MBBS, FRACGP, is General Practitioner Consultant, Rhylyn Pty Ltd.
John E Marley, MD, MBChB, DA, FRACGP, FACRRM, is Pro Vice-Chancellor, The University of Newcastle, New South
Wales.
John F Roddick, BSc, PhD, FACS, is Professor of Informatics and Engineering, Flinders University, South Australia.
INTRODUCTION Tiredness is a common presentation in general practice for which pathology tests are commonly
ordered. Our aim was to study their utilisation for tiredness.
METHODS We examined an integrated database which contains the medical records for 58 139 patients and their 696
518 associated general practitioner encounters. Three hundred and forty-two patients and their 1652 associated
encounters were randomly selected out of 12 291 patients and their 26 748 associated encounters that had mentioned
tiredness (or a synonym).
RESULTS One hundred and eighty-one patients (53%) had at least one pathology test ordered at any time in their episode
of care. Patients over 60 years of age, patients who consulted their GP more than once and patients without comorbidity
were more likely to have a pathology test ordered. Only 12 patients (3%) had a significant clinical diagnosis based on
an abnormal pathology test.
CONCLUSIONPathology testing for patients presenting with tiredness is high. Most tests do not yield a significant clinical diagnosis.

Page 2

664 • Reprinted from Australian Family Physician Vol. 32, No. 8, August 2003
tive of Australian general practice atten-
dees, both by age and sex.17
We searched the Medic-GP database
for all patient visits using the key terms:
tiredness, fatigue, weariness, weakness,
lethargy or malaise, and individually
checked them. Patients under five years
of age were excluded.
Trained data entry staff coded the rel-
evant details (such
comorbidities, other symptoms, pathology
investigations) using a controlled vocabu-
lary list, to an episode of care (defined as
the period surrounding all care from start
of the tiredness to its finish, or the end of
the relevant care).18
Of the 58 139 patients included on the
database there were 12 291 patients with
the mention of at least one of the specified
problems. The prevalence of tiredness was
21% of all patients included on the data-
base. All records mentioning tiredness,
fatigue, weariness, weakness, lethargy or
malaise were randomly sorted. We aimed
to select 300 patients from the database
presenting to their GP with tiredness.
Previous studies of similar size were able to
describe the patient group and find statisti-
cally and practically
associations.13–15A random number genera-
tor was used to select 345 patients and their
as patient
important
1656 associated GP encounters for the case
note review but of these, three did not have
sex specified in their case notes. As sex is
an important factor when considering the
management of tiredness these three
patients were excluded from all analyses
and this paper reports on 342 patients and
their 1652 associated GP encounters.
Discrete variables were compared
using the Pearson x2 test. We used logis-
tic regression to determine the possible
predictors of a patient having a pathology
test ordered (age, sex, number of visits
and whether there were any comorbidi-
ties associated with the patient).
Results
Two hundred and twenty (64%) were
female; the mean age was 44 years (range
6–90 years). The mean number of consul-
tations for all patients with tiredness was
4.8 (SD: 9.8), (range from 1–74 consulta-
tions) and 189 patients (55%) consulting
once for tiredness. Female patients con-
sulted the GP significantly more
frequently (p=0.014) (Table 1). Those
over 60 years of age consulted the GP
with tiredness significantly more than
younger patients (p=0.009).
There were 161 (47%) patients who
had no pathology ordered at any time;
181 (53%) who had at least one. At least
one pathology test was ordered for 125
(57%) females compared to 56 (46%)
males (p=0.053). Patients Ž60 years were
significantly more likely to have a pathol-
ogy test ordered (p<0.001), and to have
their first pathology test on a visit other
than their first visit, than those aged
30–59 and Ž29 years (p=0.002).
Most, 121 (67%) patients, had a
pathology test ordered in the first visit, 25
(14%) not until a second visit, 8 (4%) at a
third visit, and 27 (15%) after that. Sex
Table 2. Association between age, sex, number of visits and comorbidity, of patients with tiredness having
a pathology test ordered
n%Total *Adjusted
odds ratio
1.1
–
–
2.4
2.8
7.2
–
4.0
–
95% CIUnadjusted
odds ratio
1.5
–
–
2.0
2.9
5.0
–
1.7
–
95% CI
Sex Female
Male
<29 years
30–59 years
>60 years
1 visit
>2 visits
No
Yes
220
122
92
163
87
189
153
87
255
64
36
27
48
25
55
45
25
75
342
–
342
–
–
342
–
342
–
0.7
–
–
1.3
1.4
4.2
–
2.2
–
1.9
–
–
4.3
5.5
12.5
–
2.3
–
0.9
–
–
1.1
1.6
3.1
–
1.1
–
2.4
–
–
3.3
5.4
8.0
–
2.9
–
Age
Number of visits
Comorbidities
* Adjusted for sex, age, number of visits and comorbidities
Table 1. Frequency of consultation for tiredness by sex
Number of visits to the GP
1 visit
2–4 visits
Ž5 visits
Total
Female
109 (49%)
70 (32%)
41 (19%)
220 (100%)
Male
80 (65%)
29 (24%)
13 (11%)
122 (100%)
Total
189 (55%)
99 (29%)
54 (16%)
342 (100%)

Page 3

Reprinted from Australian Family Physician Vol. 32, No. 8, August 2003 • 665
(38, 68% males compared to 83, 66%
females) had no significant effect on
having a test ordered at the first visit.
Logistic regression for independent
predictors showed that patients aged over
60 years of age were 2.8 times more likely
to have a pathology test compared to
younger patients; those who consulted the
GP more than once were 7.2 times more
likely to have a pathology test, and those
without a comorbidity were 4.0 times more
likely to have a pathology test (Table 2).
There were 1183 pathology tests
ordered. Most patients had between one
and six tests (Table 3).
Table 3. Type of pathology tests ordered
Pathology testn%n abnormal pathology resulting
in significant clinical diagnosis
4
4
2
2
2
0
2
0
0
16
Full blood examination
Urea, electrolytes and creatinine
Thyroid function test
Liver function test
Blood glucose test
Erythrocyte sedimentation rate
Iron studies
B12
Folate
Total
316
214
139
129
109
103
85
54
34
1183
26.7%
18.1%
11.7%
10.9%
9.2%
8.7%
7.2%
4.6%
2.9%
100%
Table 4. Number of tests performed per patient that were normal or abnormal by typed test
Number of abnormal tests for each patient
12
nn
18–
94
34
–1
14–
311
22
––
6–
––
–1
––
5–
22
11
39–
Typed test Number of times tests
performed per patient
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
1
0
n
3
n
–
4
n
–
Total
n
235
21
9
3
150
15
10
1
115
7
2
1
102
6
5
103
Full blood examination 217
8
1
–
136
1
–
–
109
7
–
1
97
2
1
64
1
2
–
–
6
–
–
–
1
–
–
–
2
–
–
–
–
–
–
1
–
–
–
–
–
–
–
–
Urea, electrolytes,
creatinine
Thyroid function test*
Liver function test
Erythrocyte
sedimentation rate
Iron studies1
2
3
1
2
1
1
45
4
1
96
2
54
34
880
9
3
–
7
1
–
–
122
–
1
4
–
–
–
–
31
–
–
–
–
–
–
–
12
–
–
–
–
–
–
–
1
54
8
5
103
3
54
34
1046
Blood glucose
B12
Folate
Total
* Note: Eight patients had an abnormal thyroid function test, but only two patients had a new and active diagnosis of thyroid disease. (The other six
had a past history of thyroid disease being monitored by the GP)

Page 4

666 • Reprinted from Australian Family Physician Vol. 32, No. 8, August 2003
Of the 1183 pathology tests ordered
there were only 1046 corresponding test
results recorded. The laboratory records
of 137 tests were incomplete, and there-
fore could not be coded; 880 (84%) tests
were normal with only 166 (16%) deemed
abnormal (Table 4). Only 12 patients (4%
of 342 patients sampled) had a significant
clinical diagnosis made because of an
abnormal pathology test, two of which
were separate diagnoses (renal failure
and diabetes mellitus, nephropathy and
hepatitis). Overall the diagnoses based on
an abnormal pathology test included
anaemia (3), diabetes mellitus (2), renal
failure (2), glandular fever (1), goiter (1),
hepatitis (1), HIV infection (1),
hypokalaemia (1), hypothyroidism (1),
and nephropathy (1). Of patients who had
no pathology ordered, 16 (5%) had a sig-
nificant clinical diagnosis identified, and
51 (15%) had no clinical diagnosis identi-
fied in their episode of care.
Discussion
Was the sample representative? The GPs
were self selected and providing patient
medical records for more than 10 years.
It seems unlikely that information bias
could have influenced the results.
We showed that the majority of
patients have a pathology test ordered on
the first visit to the GP, although only a
small number have a new diagnosis
derived from an abnormal pathology test.
Those with no pathology ordered were as
likely to be given a new diagnosis as those
who had pathology ordered.
Our findings confirm previous work
elsewhere.13–15The three most likely diag-
noses seem to be anaemia, diabetes
mellitus and hypothyroidism; and the
most useful pathology tests are the full
blood examination, blood glucose test and
the urea, electrolytes and creatinine test.
We could identify no Australian guide-
lines for investigating tiredness in general
practice. The Dutch College of General
Practitioners recommends postponing
blood tests for patients presenting with
vague medical conditions for one month.19
More research to elucidate the predic-
tors of patients with important medical
problems is required to begin the process
of defining appropriate guidelines and
decision support tools for the problem of
tiredness in Australian general practice.
Acknowledgments
Thanks to both Heather McElroy who
provided statistical advice and comment and
Gillian Laven in the Department of General
Practice, The University of Adelaide who
provided editorial comment. Thanks to the
participating GPs and the Commonwealth
Department of Health and Aged Care for the
financial support. Ethics approval was
obtained from the University of Adelaide.
Conflict of interest: none declared.
References
1.De Rijik A, Schreurs K G, Bensing J.
General practitioners’ attributions of
fatigue. Soc Sci Med 1998; 47:487–496.
2.Murtagh J. Making fatigue less tiresome.
Med J Aust 1996; 164:580–581.
3.Ridsdale L. Tired all the time. BMJ 1991;
303:1490–1491.
4. Britt H, Miller GC, Knox S, et al. General
practice activity in Australia 2001-2002.
Canberra: Australian Institute of Health
and Welfare, 2002.
5.Britt H, Sayer G P, Miller G C, et al.
General practice activity in Australia
1998-1999. Canberra: Australian Institute
of Health and Welfare, 1999.
6. Britt H, Miller G C, Charles J, et al.
General practice activity in Australia
1999-2000. Canberra: Australian Institute
of Health and Welfare, 2000.
7.Britt H, Miller G C, Knox S, et al. General
practice activity in Australia 2000-2001.
Canberra: Australian Institute of Health
and Welfare, 2001.
8.David A, Pelosi A, McDonald E, et al.
Tired, weak or in need of rest: fatigue
among general practice attenders. BMJ
1990; 301:1199–1202.
9. Knottnerus J, Knipschild P G, van Wersch J
W J, Sijstermanns A H J. Unexplained
fatigue and hemoglobin: A primary care
study. Can Fam Physician 1986;
32:1601–1604.
10. Llewelyn M. Assessing the fatigued patient.
Br J Hosp Med 1996; 55:125–129.
11. Ridsdale L, Evans A, Jerrett W, Mandalia S,
Osler K, Vora H. Patients who consult with
tiredness: Frequency of consultation, per-
ceived causes of tiredness and its
association with psychological distress. Br
J Gen Pract 1994; 44:413–416.
12. Britt H, Miller G C, McGeechan K, Sayer
G P. Pathology ordering by general practi-
tioners in Australia 1998. AIHW Cat. No.
GEP 4. Canberra: Department of Health
and Aged Care, 1999.
13. Ridsdale L, Evans A, Jerrett W, Mandalia S,
Osler K, Vora H. Patients with fatigue in
general practice: A prospective study. BMJ
1993; 307:103–106.
14. Jerrett W. Lethargy in general practice. The
Practitioner 1981; 225:731–737.
15. Morrison J. Fatigue as a presenting com-
plaint in family practice. J Fam Pract 1980;
10:795–801.
16. Beilby J J, Marley J E, Walker D C,
Duszynski K M. Medic-GP: An example of
an observational, pharmacoepidemiologi-
cal database. Pharmacoepidemiol Drug Saf
2001; 9:S34.
17. Beilby J J, Marley J E, Walker D C,
Chamberlain N L, Burke M. Effect of
changes in antibiotic prescribing on
patient outcomes in a community setting:
a natural experiment in Australia. Clin
Infect Dis 2002; 34:55–64.
18. Lamberts H, Hofmans-Okkkes I. Episode of
care: A core concept in family practice. J
Fam Pract 1996; 42:161–167.
19. Dinant G J, van Wijk M A M, Janssens H J
E M, et al. NHG-standaard bloedonder-
zoek. Algemene principes en uitvoering in
eigen beheer. (Dutch College of General
Practitioners guidelines for blood tests.
General principles and testing in own
practice.) Huisarts Wet 1994; 37:202–211.
Implications of this study
for general practice
• Most use of pathology is
unproductive.
• There are no guidelines about its
best approach to investigation in
Australian general practice.
• Only 5% of patients with tiredness
have a clinical diagnosis derived
from an abnormal pathology test.
• The most useful pathology tests are:
– full blood examination
– blood glucose test
– urea, electrolytes and creatinine
test.
Correspondence
Email: angela.gialamas@adelaide.edu.au
AFP