Original article
Illustrations reduce contamination of midstream
1
Department of Emergency
Medicine, Princess Alexandra
Hospital, Brisbane,
Queensland, Australia
2
School of Medicine, The
University of Queensland,
Brisbane, Queensland,
Australia
3
Department of Chemical
Pathology, Princess Alexandra
Hospital, Brisbane,
Queensland, Australia
Correspondence to
Dr Robert Eley, Department of
Emergency Medicine, Princess
Alexandra Hospital, 199
Ipswich Rd, Woolloongabba,
Brisbane, QLD 4102, Australia;
r.eley@uq.edu.au
Received 10 November 2015
Revised 15 January 2016
Accepted 31 January 2016
Published Online First
18 February 2016
To cite: Eley R, Judge C,
Knight L, et al. J Clin Pathol
2016;69:921–925.
urine samples in the emergency department
Robert Eley,1,2 Chantelle Judge,1 Lisette Knight,1 Goce Dimeski,2,3 Michael Sinnott1,2
ABSTRACT
Aims Urinalysis provides direction in diagnosis and
treatment of patients in the emergency department (ED).
Midstream urine (MSU) collection from female patients
has a high contamination rate. Verbal instruction by
nurses to patients reduces contamination but is
inconsistent owing to lack of time and professional
knowledge. This study aimed to determine if an
alternative mode of instruction requiring minimal staff
input may be effective.
Methods A pseudorandomised controlled trial was
undertaken with 240 female patients for whom
urinalysis was clinically required. No change was made
to normal practice with regards to verbal instruction.
Prior to collecting their sample the intervention cohort
received illustrated instruction on how to collect a clean
uncontaminated MSU sample. The control cohort
received no illustrated instruction. Compared outcomes
were rate of contamination on urinalysis, defined as 10
or more epithelial cells per high power field, and
answers to a structured patient questionnaire.
Results Contamination rate was reduced from 40% to
25% by the intervention. According to patient survey
responses, verbal collection instructions were seldom
given and the actions of hand washing, cleaning with a
towelette, and voiding then stopping were significantly
higher in the intervention group. The illustrations were
well received by over 95% of patients and were
considered to be clear and effective especially for
patients with reading difficulties and/or from a non-
English speaking background.
Conclusions Illustrated urine collection instructions
were well accepted by female ED patients, improved the
rate of proper MSU collection and reduced the rate of
urinalysis contamination in the ED.
INTRODUCTION
Urinary tract infections are reported to be the
second most common infections1 affecting up to
15% of women in any given year2 and 50% of
women during their lifetime.3 Urine is the most fre-
quent sample presented to microbiology laborator-
ies,4 and is probably the most common
microbiology procedure where the specimen is self-
collected by the patient.5
Urinalysis provides vital information in the clin-
ical management of patients in the emergency
department (ED). Contamination rates of urine
samples are high especially in women.6 Sample
contamination can lead to diagnostic ambiguity or
incorrect diagnosis and inappropriate treatment.
This in turn may lead to poorer patient outcomes,
and increases the misuse of antibiotics and overall
resistance.
Eley R, et al. J Clin Pathol 2016;69:921–925. doi:10.1136/jclinpath-2015-203504
The need to repeat samples incurs additional
cost, prolongs time for diagnosis and treatment and
can increase patient anxiety and time spent in the
ED.7 8
Contamination of urine may not be completely
avoidable; however the Q-Probe study5 concluded
little progress had been made since a similar study
a decade earlier.9
Contamination rates vary by institution, collec-
tion, storage and transport of samples and by defin-
ition of what constitutes a contaminated sample.
Contamination has been defined variously includ-
ing numbers of squamous cells visible by micros-
copy, mixed growth by culture or varying numbers
of isolates in different concentrations of colony
forming units.1 5 10
BACKGROUND
Over 12 000 urine specimens originate from the
adult ED at the study hospital annually. A review of
the results of 2000 ED-sourced midstream urine
(MSU) samples indicated contamination by squa-
mous epithelial cells (≥10 cells per field) in 41.5%
of samples from women and 5.4% of men.
One of the potential reasons for contamination is
poor patient technique resulting in large part from
inadequate instruction to the patient. The literature
on communication processes between health pro-
fessions and patients conclude that the likelihood
of success in collection of medical samples increases
with careful explanation.11–13
Written instructions are reported to complement
verbal instructions;14 however there is a paucity of
confirmatory research data. In fact a Cochrane
Review of the medical literature12 reported that
there were only two well controlled trials15 16
which compared instruction method in a medical
context. Those trials occurred with patients at dis-
charge and showed that combined verbal and
written information appeared to improve knowl-
edge and satisfaction.
With respect to urine collection, verbal instruc-
tion alone to patients has shown to be lacking on
two fronts. First, two studies have shown that the
knowledge of many nurses in urine collection tech-
nique was lacking and they require training or
retraining in this technique.14 17 Second, the sus-
tainability of the approach has been questioned. A
recent Californian study evaluated a nursing MSU
instruction protocol for the ED.18 The protocol
which prescribed a series of verbal instructions to
patients had immediate positive effect and
increased steps such as hand washing over those
patients not receiving instruction. However, despite
widespread dissemination of information to the
nurses, 40% of the time instructions were not
921
 Original article
given to patients. The authors concluded that ‘what really
happens’ in an ED is that competing priorities resulted in the
failure to deliver instruction. Those results were in agreement
with the view of Lifshitz and Kramer19 who reported that
instructing patients verbally on the MSU clean catch technique
was time-consuming and was likely not to be sustainable.
Several other studies have compared instruction method
within the context of urine collection with mixed results. In the
research laboratory environment there was no statistical differ-
ence between written and verbal instruction.20 Other studies
support this view.19 21 In contrast provision of written and/or
verbal instructions has some success in reducing contamination
in several other studies7 20 21 including samples collected from
EDs where written instructions were distributed to patients.5
The conclusion from these studies is that verbal communica-
tion while effective is not sustainable and the effectiveness of
written instruction is not absolute. A more effective method
which will work in the ED environment is needed.
At the study hospital instructions for collection of MSU are
only provided verbally. No specific training of nursing and
medical staff is provided; instead reliance is given on the experi-
ence and communication skills of the nurses or doctors.
Preliminary observational research complemented by conversa-
tions with nursing and medical staff indicated that verbal
instruction to patients was inconsistent in frequency and
content.
Recognising that in reality this was unlikely to change, this
study was designed to assess whether illustrations depicting the
step-by-step process for collecting a urine sample would be
effective in decreasing the contamination rate in female ED
patients. Instructions were provided for an MSU sample as this
is the most frequently used patient-collected method whereby
the first portion of urine which has the highest rate of contamin-
ation is voided.
Design
A pseudorandomised control trial undertaken with 240 female
patients, comparing pathology results and answers to a struc-
tured questionnaire between those patients provided with an
illustrated instruction (figure 1) and those with no instruction.
Participants
Participants were 18 years and over who presented to the ED,
and because of their presenting symptoms suggestive of infec-
tion (eg, flank pain, painful urination, fever) were clinically
required to provide a clean catch MSU sample for diagnostic
purposes. Patients who were unable to provide the sample in
this way and provided a sample via other methods such as
indwelling urinary catheter and suprapubic catheter were
excluded.
Participants were alternately provided with an envelope con-
taining either the illustrated instruction (figure 1; Illustration
group=I) or a piece of paper advising them to return their
sample to the nurse (No illustration group=NI). Both envelopes
contained a non-alcohol based towelette. In order to mimic the
usual situation as closely as possible the nurses were not pro-
vided with any additional instruction other than to give the
envelope at the same time as the sample container. Nurses were
therefore free to give verbal instruction as per their normal
practice.
After patients had returned their sample to the nurse they
were approached by one of two investigators (CJ or LK) and
both groups consented to complete an investigator-administered
questionnaire.
922
As per standard procedure the urine sample was sent to the
pathology department for analysis which included the quantifi-
cation of epithelial cells. A high number of epithelial cells is
indicative of contamination of the sample with cells from the
distal urethra and/or perineum. This is frequently associated
with false-positive urine culture results, as these epithelial cells
can carry large numbers of bacteria. For the purposes of this
study contamination was defined simply as the presence of 10
or more squamous epithelial cells per high power field as a sur-
rogate of bacterial of contamination.
Questionnaire
The questionnaire contained questions requiring either answer
of yes/no or a response on a Likert-type scale. The questions
determined (A) memory of any verbal instructions given by the
nurse, and (B) the process by which they collected their sample.
Skip logic was applied such that if the response to Q1 “Did the
nurse tell you how to collect your sample?” was ‘No’ then
Questions 2–5 on recall of what was said were omitted.
Questions 7–11 were only administered to those patients who
received the illustrated instructions. After completion of the
questionnaire the NI group participants were given a copy of
the illustrations and both groups were provided with the oppor-
tunity to offer comment on them.
Ethics approval
The study received approval from the ethics committees of the
hospital and university.
Statistical analysis
Statistical analysis was performed using two-proportion z test
for proportions and χ2 test for categorical variables.
RESULTS
Table 1 presents the pathology results. Based on epithelial cells
of ≥10 group I had fewer contaminated samples (25%) than did
the NI group (40%). There was no difference in the contamin-
ation rate regardless of whether they received verbal instruction
from the nurse.
In the I group there were 16 UTIs reported based on culture
results. Of these five were qualified as ‘possible UTI’ owing to
the sample being contaminated with epithelial cells. In the NI
group the figures were 15 and 8, respectively.
In response to the question “if you had received this instruc-
tion sheet would it have been clearer to you as to how to give
the sample” 91 of the 120 NI participants said ‘Yes’.
All participants were offered the opportunity to make add-
itional comments. Of the 123 who did so 65 referred specific-
ally to the clarity and effectiveness of the illustrations with
phrases such as ‘instructions were good’, ‘they made it clearer’,
‘pretty thorough’, ‘well detailed’, ‘very helpful’. Examples of
other comments are
▸ “Beneficial in waiting room as more discrete way of giving
MSU instructions”;
▸ “That’’s brilliant, easier to understand”;
▸ “Verbal instructions are rushed with many people in the
room. Written is much more helpful”.
Six participants noted that the illustrations would be useful
for people who had difficulty reading or for those who are from
a non-English speaking background.
▸ “Pictures [will be] good for non-English speaking people or
those with poor eyesight”.
“Information sheet must be given automatically to
non-English speaking people”
Eley R, et al. J Clin Pathol 2016;69:921–925. doi:10.1136/jclinpath-2015-203504

Figure 1 Illustration of midstream urine collection technique.
However 10 patients had negative feedback of the
illustrations.
▸ “Too graphic”;
▸ “Verbal instructions should be enough”;
▸ “The detail in the instructions are a little over the top”.
Responses to the questionnaire (table 2) indicated that almost
half of the patients did not receive any verbal instruction from
the nurses. More NI patients than I recollected being told to
wash their hands but fewer to void their first urine into the toilet.
Steps the subjects recalled performing during sample collec-
tion are presented in table 3. Of six steps queried, three were
reported to be performed statistically more often by the I
group.
Eley R, et al. J Clin Pathol 2016;69:921–925. doi:10.1136/jclinpath-2015-203504
Original article
There were no differences in the rate between NI subjects
who recalled receiving verbal instructions and those who did
not.
While all the intervention patients received instruction of
what to do with the towelette, some control group patients
reported that they had used the towelette to wipe their hands or
the toilet seat. The effect of the towelette was shown to be non-
significant when it was removed for the last 40 of the NI group.
DISCUSSION
There is only one prior study evaluating instructions for urine
collection in an ED environment and it was a verbal interven-
tion.18 The authors showed that adherence to a new protocol
923
 Original article

Table 1 Pathology results Table 3 Steps patients recalled performing during collection
Illustrations Per cent No illustrations Per cent Steps Illustrations Per cent No illustrations Per cent

Epithelial cells Wash hands 101 84 64 53*

≤10 90 75 73 60* Clean with towelette 96 80 42/80† 52*
11–50 20 17 28 23 Void then stop 118 98 99 82*
51–100 10 8 19 16 Half fill container 118 98 108 90
*p<0.05. Finish voiding in toilet 119 99 109 90
Wash hands 119 99 115 96
*p<0.05.
†Towelettes were not given to that last 40 in the NI group.
for provision of verbal instruction was unsustainable and identi- NI, no illustration.
fied a need for pragmatic ED studies on the impact of urine col-
lection instructions. These conclusions are an endorsement of
our study to determine a practical approach to improve sample Overall the study demonstrated that these low-cost illustra-
collection. tions were effective. It is noted however that some, albeit a very
It is clear that verbal instruction cannot be relied upon. In our low, proportion of patients found the graphic illustrations dis-
own department only a little over half of the patients recalled tasteful. We concede that this is a small, unavoidable conse-
receiving verbal instruction. Interestingly, while only 50% of the quence of such illustrations.
intervention group stated that they received verbal instruction
61% of the control group did so. This latter figure is exactly the CONCLUSION
same as that reported previously.18 It is possible that some inter- In an ED setting, where nurse verbal urine collection instruc-
vention participants forgot about the verbal instruction after tions often were not recalled, simple illustrated instructions
receipt of the illustrations. Factors affecting this may be related improved patient performance of MSU collection steps and
to the workload of nurses, however there also may be reticence reduced urine contamination.
in describing the process in a public area.5
The value of using a towelette was questioned with no differ-
ence seen in any parameters to those patients in the NI group
Take home messages
who used the towelette for its intended purpose. A small or neg-
ligible effect of perineal cleaning has also been seen in other
studies21 22 with parting of the labia reported to be a more ▸ The time-poor and busy environment of the emergency
important step.18 department creates challenges to ensure patient-collected
We used squamous epithelial cells from the skin surface or urine samples are contaminant-free.
from the outer urethra as a surrogate marker for contamination ▸ Verbal and written interventions are either unsustainable or
and demonstrated a reduction in the intervention group. provide barriers to many patients.
Another measure of contamination may have provided different ▸ Illustrated instructions are an inexpensive, easily applied and
results, however our data are consistent with a previous study acceptable method of providing information to patients in
showing that written instructions resulted in lower urine con- the emergency setting to reduce contamination rate of
tamination in ED patients.5 midstream urine samples.
Disposable, patient-held devices, designed to automatically
obtain a midstream collection, are on the market, but it is
unclear whether such devices reduce contamination, particularly
Handling editor Slade Jensen
in a clinical real world setting.6 10 Furthermore they are rela-
tively expensive and unlikely to be adopted soon for use in our Acknowledgements The authors thank Dr Joan Faoagali and Angela Payne for
supporting this study and David Toohey for drawing the illustrations.
public health system.
Contributors All of the named authors contributed to the design of the study,
data collection and analysis. RE drafted this article, which was then commented on
by the other authors.
Table 2 Recall of nurse instruction for a midstream urine sample Funding This study received funding from the Queensland Emergency Medicine
No Research Foundation grant number EMSS-12-183.
Illustrations Per illustrations Competing interests None declared.
Question (n=120) cent (n=120) Per cent
Patient consent Obtained.
Received no verbal 60 50 47 39 Ethics approval Metro South HREC.
instruction
Provenance and peer review Not commissioned; externally peer reviewed.
Received verbal instruction 60 50 73 61
Recall of what was told by nurse
REFERENCES
Wash hands 39 65 60 82* 1 Car J. Urinary tract infections in women: diagnosis and management in primary
Clean with towelette 32 53 24 32 care. BMJ 2006;332:94–7.
Void then stop 54 90 54 74* 2 Foxman B. Epidemiology of urinary tract infections: incidence, morbidity and
Half fill container 52 86 58 79 economic costs. Am J Med 2002;113(Suppl 1A):5S–13S.
3 Better Health Victoria. Urinary tract infections. Better Health Channel: State of
Finish voiding in toilet 51 85 57 78 Victoria, 2012. https://www.betterhealth.vic.gov.au/health/conditionsandtreatments/
Wash hands again 38 63 48 65 urinary-tract-infections-uti (accessed 2 Nov 2015).
*p<0.05. 4 Turner D, Little P, Raftery J, et al. Cost effectiveness of management strategies for
urinary tract infections: results from randomised controlled trial. BMJ 2010;340:c346.

924 Eley R, et al. J Clin Pathol 2016;69:921–925. doi:10.1136/jclinpath-2015-203504


5 Bekeris LG, Jones BA, Walsh MK, et al. Urine culture contamination: a College of 13
American Pathologists Q-Probes study of 127 laboratories. Arch Pathol Lab Med
2008;132:913–17. 14
6 Curtis J, Perry K, Bower L. Evidence review: clean Catch® midstream urine collection 15
device. London: Centre for Evidence-based Purchasing, 2008.
7 Jackson SR, Dryden M, Gillett P, et al. A novel midstream urine-collection device
reduces contamination rates in urine cultures amongst women. BJU Int 16
2005;96:360–4.
8 Wilson P, Hay-Smith J, Nygaard I, et al. Adult Conservative Management. In:
Abrams P, Cardozo L, Khoury S, et al, eds. Incontinence, Vol 2: Management. 3rd 17
International Consultation on Incontinence. Plymouth, UK: Health Publication Ltd,
2005:857–964.
9 Valenstein P, Meier F. Urine culture contamination: a College of American
Pathologists Q-Probes study of contaminated urine cultures in 906 institutions. Arch 18
Pathol Lab Med 1998;122:123–9.
10 Collier S, Matjiu F, Jones G, et al. A prospective study comparing 19
contamination rates between a novel mid-stream urine collection device (Peezy)
and a standard method in renal patients. J Clin Pathol 2014;67:139–42. 20
11 Jarrett N, Payne S. A selective review of the literature on nurse-patient
communication: has the patient’s contribution been neglected? J Adv Nurs
1995;22:72–8. 21
12 Johnson A, Sandford J, Tyndall J. Written and verbal information versus verbal
information only for patients being discharged from acute hospital settings to home: 22
systematic review. Health Educ Res 2005;20:423–9.
Eley R, et al. J Clin Pathol 2016;69:921–925. doi:10.1136/jclinpath-2015-203504
Original article
McCabe C. Nurse–patient communication: an exploration of patients’ experiences.
J Clin Nurs 2004;13:41–9.
Gilbert R. Taking a midstream specimen of urine. Nurs Times 2005;102:22–3.
Isaacman DJ, Purvis K, Gyuro J, et al. Standardized instructions: do they improve
communication of discharge information from the emergency department? Pediatrics
1992;89(Pt 2):1204–8.
Jenkins HML, Blank V, Miller K, et al. A randomized single-blind evaluation of a
discharge teaching book for pediatric patients with burns. J Burn Care Rehabil
1996;17:49–61.
Tosif S, Baker A, Oakley E, et al. Contamination rates of different urine
collection methods for the diagnosis of urinary tract infections in young children:
an observational cohort study. J Paediatr Child Health 2012;48:
659–64.
Frazee BW, Frausto K, Cisse B, et al. Urine collection in the emergency department:
what really happens in there? West J Emerg Med 2012;13:401.
Lifshitz E, Kramer L. Outpatient urine culture: does collection technique matter?
Arch. Int Med 2000;160:2537–40.
Fisher LA, Johnson TS, Porter D, et al. Collection of a clean voided urine specimen:
a comparison among spoken, written, and computer-based instructions. Am J Public
Health 1977;67:640–4.
Blake DR, Doherty LF. Effect of perineal cleansing on contamination rate of
mid-stream urine culture. J Pediatr Adolesc Gynecol 2006;19:31–4.
Leisure MK, Dudley SM, Donowitz LG. Does a clean catch urine sample reduce
bacterial contamination? N Engl J Med 1993;328:289.
925