Inhaler technique in Turkish people with poor English: a case of ...

Pharm World Sci (2006) 28:107–114
DOI 10.1007/s11096-006-9019-5
RESEARCH ARTICLE
Inhaler technique in Turkish people with poor English: a
case of information discrimination?
Larry Goodyer Æ Imogen Savage Æ
Zeynep Dikmen
Received: 23 March 2006 / Accepted: 29 March 2006 / Published online: 4 July 2006
Ó Springer Science+Business Media B.V. 2006
Abstract
Objectives (a) To compare metered dose inhaler
(MDI) technique in users with poor English and
fluent English, (b) to evaluate two interventions: a
translated patient information leaflet (PIL) plus
support from an translator (PIL + verbal) and a
multimedia touch screen system (MTS) using
video clips and own-language instruction.
Methods (a) Inhaler technique was videotaped
and key steps rated blind for 105 fluent Englishspeakers
(FE) and 69 Turkish-speakers with poor
English (EP). (b) The EP group was randomised
to receive information by MTS (n = 34) or
PIL + verbal (n = 35). Inhaler technique was
videotaped before and after information.
Main outcome measures (a) Global inhaler
technique; (b) breathing-in time; (c) co-ordination
of inspiration and inhaler actuation.
L. Goodyer Æ I. Savage (&)
Department of Pharmacy, King’s College London,
150 Stamford Street, London SE1 9NN, UK
e-mail: imogen.savage@ulsop.ac.uk
Z. Dikmen
Tower Hamlets Health Strategy Group, Oxford
House, Derbyshire Street, London E2 6HG, UK
I. Savage
Department of Policy and Practice, School of
Pharmacy, London University, 29–39 Brunswick
Square, London WC1N 1AX, UK
Results Global technique, co-ordination and
breath-holding were all significantly worse in MDI
users with poor English. Only 17% of that group
had adequate technique compared to over half
(62%) of FE. The EP group were significantly less
likely than the FE group to report ever seeing the
practice nurse about their asthma. After information,
global technique was rated as improved in
50% of the MTS group compared to 28% of those
given a translated PIL. A further six people (17%)
in the PIL group improved after subsequent verbal
advice in their own language. Both information
methods significantly increased inhaler shaking
and mouthpiece checking, but co-ordination only
improved in a small number of people.
Conclusions: The study suggests that Turkishspeaking
MDI users with poor English may be
disadvantaged in terms of access to medicines
information in the UK. The acceptability of
pharmacy-based support services for this, and
other specific language groups should be explored.
Multimedia offers an alternative to a translator for
brief explanations, particularly for first-time users,
but improving poor co-ordination requires individualised
‘‘hands on’’ teaching from health professionals.
Keywords English speakers Æ Inhaler technique Æ
Language ability Æ Medicines information Æ
Metered dose inhalers Æ Patient information
leaflet Æ Turkish speakers
123

108 Pharm World Sci (2006) 28:107–114
Introduction
Good information on how to take or use medicines
is important if people are to get the most
benefit from them. In the European Union,
comprehensive leaflets for patients are now
mandatory [1], and all medicine packs should
contain a leaflet as a package insert, produced by
the manufacturer. In the United Kingdom, these
leaflets are written in English. Translation support
for patients is available in hospitals and primary
care, but provision is patchy and services
over-stretched.
People whose first language is not English (non-
EFL) may therefore, have difficulty in accessing
information about their medicines. This puts them
at a significant disadvantage, yet there has been
relatively little research [2, 3] in this area.
Asthma is one example of a condition where
the inability to understand and use printed
information (functional illiteracy) could compromise
effective management. The metered dose
inhaler (MDI) is the first choice device for most
people with asthma but these devices are not easy
to use well [4–7] and many patients have problems.
If the inhaler is pressed before the person
starts to breathe in, then a good portion of the
released dose will be lost. It is also important that
the user breathes in deeply, and then holds their
breath, to maximise drug delivery to the lungs.
Good instruction can help to overcome these
difficulties, [8, 9] particularly in relation to a poor
hand/lung co-ordination. However, regular review
takes staff time [10] and is not universally available.
For many users in the UK, the first and only
information they get on how to use their inhaler is
the patient information leaflet (PIL) inside the
pack. People with poor English reading skills may
not be able to understand this information, and
could use their inhaler incorrectly, resulting in
sub-optimal dosing and sub-optimal response.
The first aim of this study was to test this
hypothesis by comparing inhaler technique in
MDI users with poor English (EP) and people
whose first language was English (FE). For the
EP group, we chose to study people of Turkish
origin, as previous work [11] in Haringey, London
had suggested the language barrier was a significant
factor in medicines compliance.
123
The second aim was to evaluate two interventions:
a translated PIL plus verbal support from a
translator, and a multimedia touch screen system
(MTS) which provides the same information
using video clips of an actual demonstration
with own-language voice-over instruction. It was
hypothesised that the MTS system might reduce
the need for translator support.
Method
This was a single blind randomised controlled
trial with two groups: MTS and a translated leaflet
plus support from a translator fluent in Turkish
and English (PIL + verbal).
Patient allocation was done using random
number tables. Patients did not know in advance
which information method they would receive.
The MTS system covered the material shown
in Table 1 with additional information on posture.
The subject could interact with the program
by selecting to replay a step, or move on to the
next one.
The leaflet was produced by one of the authors
(ZD) and was an exact translation of the current
version of Allen and Hanbury’s PIL for the
Ventolin MDI.
Patient recruitment
All recruits gave their informed consent to the
study, which was approved by the relevant local
ethics committees.
Fluent English-speaking MDI users of non-
Turkish origin who had been using an inhaler for
more than 3 months were recruited through GP
practices in Haringey and Camden, and assessed
in the surgery. Reading ability was not formally
assessed, but the recruitment method provided
strong evidence that every volunteer had at least
average literacy. Participants were recruited by
letter and had to reply in writing, providing some
initial information. Their responses could be validated
against their handwriting on the consent
form which was completed in the surgery before
inhaler technique was assessed.
Initially, Turkish-speakers (Turkish first language;
not fluent in English) were recruited
through GP surgeries in Camden but this was not

Pharm World Sci (2006) 28:107–114 109
Table 1 Information points
h Take cover off and check the mouthpiece is clean
h Shake the inhaler well
h Hold the inhaler upright with thumb on the base, below
the mouthpiece
h Breathe out normally
h Close lips firmly around the mouthpiece
h Start to breathe in, then press down on top of the inhaler.
Keep on breathing in steadily and deeply
h Take inhaler from mouth
h Hold breath for as long as is comfortable
successful; of 100 invited to attend only 10
accepted. A ‘‘snowball’’ convenience sampling
method was therefore used, using contacts in
social clubs and informal networks in Tower
Hamlets and Camden. Subjects were assessed
mainly in their own homes. Written and spoken
skills in English and Turkish were assessed using a
language proficiency test devised by the Language
Department at King’s College London.
This included being asked to read and explain an
English phrase and to answer the question: ‘‘what
medicines do you have now?’’
Table 2 shows population data on literacy and
birthplace for the three study areas.
Study procedure
Sessions for the EP group were run by a Turkish
pharmacist (ZD) who acted as both researcher
and translator. The procedure for the English
group has been reported elsewhere [12]. Demographic
and treatment-related information (see
Table 3) was collected before each patient was
Table 2 Area demography
Area Total
population
% born outside
UK/Eire
Haringey 216,507 34.4% 5.3
Tower Hamlets 196,106 33.4% 10.6
Camden 198,020 33.6% 5.1
London 7,172,091 24.9% 4.9
England 49,138,831 8.3% 4.0
% very low
literacy*
*Data from Census 2001 and Basic Skills Agency
Very low literacy: difficulties in coping with everyday
reading and numbers; need intensive help to reach
minimum standard. Includes people with learning
difficulties
assessed. As the time per patient was limited, only
basic information was collected. Participants were
not asked about their educational history, or for
details of their current job. Access to GP records
was only possible for the EP group. Asthma
severity in the two groups could therefore not be
compared.
Recruits gave written consent for a videotape
to be made. They then demonstrated to the
camera the use of the MDI, without a spacer
device, using a placebo canister placed into their
own holder. Technique was demonstrated twice;
before and after viewing information.
Patients were allowed to use the program, or
study the leaflet for as long as they wished before
repeating the demonstration. The researcher/
translator identified areas where technique could
be improved, then spent up to 15 min discussing
this with the user. A final video recording was
then made.
This allowed the effect of the information
alone to be separated from the combined effect of
PIL + translator support.
Assessment of inhaler technique
Videotaped demonstrations were digitised, compressed
to MPEG format and transferred to
CD-ROM. They were subsequently assessed by
two pharmacists with experience in respiratory
medicine. These assessors did not attend experimental
sessions and had no contact with patient
volunteers. They were therefore, blind as to the
information format, but not as to the patient’s
probable ethnic origin.
First, the assessors were asked to play the clip
through without stopping and to rate global
technique as poor, adequate or good. This would
probably approximate to a ‘‘quick check’’ of
technique performed in the clinic. Poor technique
was defined as any of the following:
– pressing the canister before inspiration
– visible ‘‘mist’’ escaping from the mouth
– no significant inhalation observed
– not releasing a dose at all
Assessors then assessed technique in detail using
a checklist. The individual steps shown in
Table 1 were scored as correct or incorrect.
123

110 Pharm World Sci (2006) 28:107–114
Table 3 Subject details
Inhaler-shaking (counts) and length of inspiration
(seconds) were assessed directly from the videotape
recordings. The variable ‘‘breathing-in’’ was
defined as the time interval between starting to
breathe in and exhaling.
Rater reliability
Intra-rater
Demonstrations from a purposive sample of 12
patients, chosen to illustrate the range of good
and poor technique, were rated twice. There was
good agreement between the two data sets
(Global technique: Chi-square df 4 P < 0.002;
co-ordination: kappa 0.795 P < 0.0001).
Inter-rater
All kappa scores for individual observations were
positive and comparison of all observations yielded
a median kappa score of 0.5 (quartile range
0.32–0.75). Whilst this demonstrates only a modest
level of agreement, assessing breathing patterns
was often a problem. ‘‘Difficult to assess’’
subjects were equally distributed between groups.
Statistical analysis
Chi-square test (categorical variables) and independent
t-tests (continuous normally distributed
123
variables) were used to compare baseline data in
the English and Turkish groups.
Global technique ratings before and after
information were compared and subjects categorised
as improved, no change, or worse.
Chi-square test was used for differences in proportions.
The sign test was used for checklist
variables; the paired t-test for inhaler-shaking
frequency and breathing-in time.
Results
Fluent English
N = 105
Turkish-speaking
N =69
Age (years)
Mean +/) sd
Age group
51 +/) 20.9 42 +/) 17.5

Pharm World Sci (2006) 28:107–114 111
Approximately, a quarter of both language
groups said they had not received initial instruction
from a health professional on how to use
their inhalers. There was no difference in use of
spacing devices: approximately a third of both
groups used these for at least some of their doses.
The EP group were significantly less likely to report
seeing the practice nurse about their asthma
than the FE group. Nearly half (45%) the EP
group said they never saw the nurse, compared
with 25% of the English-speaking group. (Chi
square = 6.8, df 1, P < 0.01).
Tables 4 and 5 compare inhaler technique.
Turkish-speakers performed better than Englishspeakers
in terms of shaking the inhaler and were
more likely to check the mouthpiece. However,
there was a marked difference in global scores for
inhaler technique, in co-ordination, and in breathholding,
all of which were significantly worse in
the group with poor English.
Impact of interventions in group with poor
English
Thirty-four Turkish patients with EP were
randomised to the MTS group and thirty-five to
the PIL/verbal group. Groups were well matched
for age, gender and baseline (before information)
inhaler technique ratings. Only 4 of the 35 people
in the MTS group had used a computer. Despite
this, they needed little instruction on use of the
system.
Changes in inhaler technique after information
are presented in Tables 6–8. Half the MTS group
were rated as improved compared to 28% of
those given a translated leaflet. A further six
people (17%) improved after subsequent verbal
advice in their own language.
Table 4 Global technique assessments in fluent Englishspeakers
and Turkish-speakers with poor English
Global technique rating
Poor Adequate Good Total
Fluent English 36 48 11 95
(37.9%) (50.5%) (11.6%)
Poor English 57 12 0 69
(83%) (17%) –
Chi-square = 30.36 df 2 P < 0.0001
Many subjects had shallow breathing which
made accurate assessment difficult. In some cases,
they failed to take any perceptible inhalation and
five failed to release a dose. Own-language verbal
instruction after reading the PIL produced small
but statistically significant improvement in
breath-holding.
Inhaler-shaking and mouthpiece-checking
improved significantly in both groups after information.
The proportion of patients with correct
co-ordination also increased in both groups but
numbers were small and changes did not reach
statistical significance.
Multiple quick actuation of the inhaler was
observed in 21% MTS and 20% PIL/verbal
before instruction. This problem was resolved in
most subjects by all modes of instruction.
Discussion
The inhaler users with FE and EP in this study
came from three London boroughs with diverse
mix of ethnic backgrounds and languages.
Approximately, a third of the local population in
each borough had been born outside the UK. All
participants were volunteers who were interested
in having an ‘‘inhaler technique check’’. The FE
group could all read English; most of the EP
group could not.
The FE and EP groups were recruited by different
methods, and assessed in different settings.
We did not take an education history, but the
broad occupational profile of the two groups was
clearly different, with more retired people in the
English group and more women looking after the
home in the Turkish group.
Inhaler technique in the FE group was in line
with previous reports, [5–9] with around a third
graded as poor and two-thirds having some
co-ordination deficiencies. Global technique in
the EP group was almost universally rated as
poor, with some extreme co-ordination problems
observed. It was not possible to compare asthma
severity in the two groups, as the researchers only
had access to GP records for patients recruited
through surgeries. However, the reported prescribing
of large volume spacer devices (roughly a
third of both groups said they used a spacer)
123

112 Pharm World Sci (2006) 28:107–114
Table 5 Comparison of
key steps in technique in
fluent English-speakers
and Turkish-speakers
with poor English
Shakes inhaler
(% correct)
suggests that GP perception of clinical need was
similar in the two groups.
The majority of participants in both FE and EP
groups said they remembered a demonstration of
how to use their inhaler when it was first prescribed.
In addition, they would all have received
the manufacturer’s PIL, which is included in
every inhaler pack. These mandatory leaflets,
defined by EU regulations, act as a reference
source for the patient and family and carers and
are intended to complement and reinforce any
verbal advice from health professionals. PILs for
asthma inhalers provide illustrations of the key
steps in correct use, but these need to be viewed
in conjunction with the text to be fully understood.
It is therefore, reasonable to propose that the
inability to read and understand the written
information on how to use their inhaler may have
made it more difficult for the EP group to develop
good inhaler technique. Cultural differences in
the use of primary care preventive services would
have perpetuated this information deficit. Our EP
sample was hard to recruit through surgeries, and
also much less likely than the FE group to use
non-acute services such as nurse-run asthma
clinics. There were therefore, fewer opportunities
for nurses and GPs to pick up and correct misunderstandings
about technique.
Table 6 Global technique change after intervention in
poor English-speakers
n shakes
(sd)
Checks
mouthpiece
(% correct)
Co-ordination
(% correct)
Breathing-in
time(secs)
Fluent English 59(59%) 4.7 (5.9) 4 (4%) 28 (29.5%) 7.8 (4.4)
n = 100 n =99 n = 100 n =95 n =52
Poor English 38 (55%) 7.0 (4.1) 10 (14.5%) 8 (11.6%) 4.8 (6.3)
n =69 n =40 n =69 n =69 n =52
Difference ns P < 0.05 P < 0.01 P < 0.02 ns
MTS PIL + translator
After
PIL
After
PIL + verbal
Worse than baseline 5 (15%) 5 (16%) 4 (11%)
No change 12 (35%) 18 (56%) 16 (48%)
Better than baseline 17 (50%) 9 (28%) 15 (45%)
123
Poor reading skills are not just a problem for
people whose first language is not English.
Nationally 4 in 10 adults have been estimated[13]
to have difficulties in coping with everyday literacy
and numeracy requirements. A recent survey
[14] found that 15% of 127 consecutive patients
(125 Caucasian) attending a rheumatoid arthritis
clinic were functionally illiterate. These patients
did not have more severe disease than patients
who could read information about their condition,
but were more anxious and visited hospital
more often.
Motivation to read and the perceived authority
of the information source are also factors. All of
our volunteers with FE were able to read, yet a
proportion also had poor technique. FE asthmatics[15]
do not always look at PILs, or see
them as important. They place a higher value on
face-to-face individualised advice.
Properly blinded studies of interventions to
improve inhaler technique have rarely been
attempted. The present study has attempted to
overcome the problem of bias by filming patients
using their inhalers and then digitising the films
for ‘‘blind’’ assessment. Deciding if chest movement
had occurred was sometimes very difficult,
and this may explain the relatively low level of
agreement between assessors.
Bearing this weakness in mind, brief exposure
to own-language information presented via multimedia
and a translated leaflet with translator
support appeared to be equally effective in
changing what some inhaler users did, at least in
the short term.
Approximately, half the subjects had improved
their technique after information given by either
method. The effect of written translated information
alone was smaller; only one quarter of
subjects improved after a leaflet.

Pharm World Sci (2006) 28:107–114 113
Both information methods resulted in small but
statistically significant improvements in some
individual technique steps, although co-ordination,
the most important aspect of technique, was only
improved in a small number of people.
Conclusions
There were demographic differences between our
two language groups, and this is an acknowledged
weakness in this study. However, we do not
believe that these differences alone explain why
inhaler technique in the people with poor English
was so much worse than in the FE users of the
same inhaler (Ventolin). Discrimination in terms
of access to the written information supplied inside
every inhaler pack also needs to be considered.
The interventions we tested suggested that a
translated PIL, used alone, is of less benefit than
the same leaflet used together with own-language
verbal advice. This supports others [3, 16] who
have concluded that it is simplistic to propose that
written translation alone is the answer to the
language barrier when providing medicines
information. The Turkish inhaler-users in our
study were mostly literate in their mother tongue
but the assumption that if you can speak a language
then you can read it is not necessarily
true[13], either for native English speakers or for
foreigners.
A broad portfolio of methods, including pictograms,
audiotapes and computer and web-based
multimedia (available through terminals in surgeries,
pharmacies and other public information
points) needs to be developed if all patients are to
Table 7 Categorical technique variables (paired data; sign test)
Technique
measure
Checks mouthpiece
(n correct)
Shakes inhaler
(n correct)
Co-ordination
(n correct)
Multimedia group Leaflet + translator group
N cases Before
info
After
MTS
P
value
have equal access to good medicines information.
Lack of computer experience need not be a barrier.
The majority of Turkish-speaking subjects
and elderly English subjects had little or no
experience but found the method both easy and
enjoyable.
The ability to offer patients a choice of language
makes multimedia systems useful educational
tools but they are not a panacea to the
problem of poor inhaler technique. Our volunteers
were long-term inhaler users, with deeply
ingrained habits. The impact of information on
their inhaler technique was modest at best, and
co-ordination did not improve in the majority of
patients tested.
Greater benefits could be expected in people
using inhalers for the first time, and a multimedia
touch-screen computer could offer a convenient
alternative to the use of a translator for the brief
explanations normally given in surgery or pharmacy.
However, the findings demonstrate that
information alone is rarely enough to change
behaviour. To improve poor co-ordination, the
inhaler user first needs to see and understand
what he is doing wrong, and then to practise until
the correct method becomes automatic. This
learning process needs practical one-to-one
teaching from someone who can demonstrate
correct MDI technique. In our study, the video
camera was a very useful teaching and learning
tool.
This study suggests that a significant proportion
of inhaler users with poor English may never
attend GP asthma check clinics, and only visit the
surgery when they become ill. However, all
inhaler users must present their prescriptions to a
pharmacist. The acceptability and feasibility of
N cases Before
info
After
PIL
P
value
After PIL +
verbal advice
P
value
34 8 (23%) 16 (47%) 0.05 35 2 (6%) 1 (3%) ns 11 (31%) 0.004
34 16 (47%) 24 (72%) 0.04 35 22 (62%) 16 (46%) ns 35 (100%) 0.001
34 4 (12%) 11 (35%) ns 35 4 (11%) 6 (18%) ns 10 (28%) ns
123

114 Pharm World Sci (2006) 28:107–114
Table 8 Continuous technique variables (paired data; paired t-test)
Technique measure Multimedia group
Inhaler shaking
(n times)
Breathing-in time
(secs)
Inhaler shaking
(n times)
Breathing-in time
(secs)
pharmacy-based services targeted at specific
language groups should be investigated.
Acknowledgements The authors would like to thank
everyone who took part in the study, or provided facilities
and support. The study was funded by the Department of
Health, and the North London Primary Care Research
Network (Nocten).
Possible conflicts of interest None.
References
N
cases
Before
info
After
MTS
1. Directive 2001/83/EC of the European Parliament and
of the Council of 6 November 2001 on the Community
code relating to medicinal products for human use.
Official Journal L 311, 28 November 2001:p20
2. Shaw J, Hemming MP, Hobson JD, Marty JJ (1977)
Comparison of English and non-English speaking
outpatients comprehension of medications. Aust J
Hosp Pharm 7(3):116–119
3. Jesson J, Sadler S, Pocock R, Jepson M (1995) Ethnic
minority consumers of community pharmaceutical
services. Int J Pharm Pract 3:129–132
4. Paterson LC, Crompton GK (1976) Use of pressurised
aerosols by asthmatic patients. BMJ 1:76–77
5. Hilton S (1990) An audit of inhaler technique among
asthma patients of 34 general practitioners. Br J Gen
Pract 40:505–506
6. Nimmo CJ, Chen DN, Martinusen SM, Ustad TL (1993)
Assessment of patient acceptance and inhalation technique
of a pressurized aerosol inhaler and two breathactuated
devices. Ann Pharmacother 27:922–927
123
Difference
(95% cl)
34 3.0 6.2 +3.2 (1–5.4) 0.06
26 3.9 4.5 +0.6 ()1.4–2.6) ns
Leaflet + translator group
N
cases
Before
info
After
PIL
Difference
(95% cl)
p
p After PIL +
verbal advice
Difference
(95% cl)
34 5.0 5.7 )0.68 ()0.1–2.5) ns 8.0 +3 (1–4.9) 0.001
17 1.8 3.2 +1.3 ()0.5–2.7) ns 5.4 +3.6 (2–5) 0.001
7. Larsen JS, Hahn M, Ekholm B, Wick KA (1994).
Evaluation of conventional press-and-breathe metered-dose
inhaler technique in 501 patients. J Asthma
31:193–199
8. Self TE, Brooks JB, Lieberman P, Ryan MR (1983).
The value of demonstration and the role of the pharmacist
in teaching the correct use of pressurised
bronchodilators. Can Med Assoc J 62:3–21
9. Dickinson J, Hutton S, Atkin A, Jines K (1992)
Reducing asthma morbidity in the community: the
effect of a targeted nurse-run asthma clinic in an
English general practice. Respir Med 91:634–640
10. Martys C (1992). Asthma care in Darley Dale: GP
audit. BMJ 304:758–760
11. Gul V, Mackenzie M (1993). Medical counselling and
patient compliance in Turkish and Greek-speaking
communities in Haringey and Enfield. J R Soc Health
113(6):286–287
12. Savage I, Goodyer L (2003). Providing information on
metered dose inhaler technique: is multimedia as
effective as print? Fam Pract 20(5):552–557
13. Basic Skills Agency benchmark statistics: http://
www.bsa.broadshow.net. Accessed 08/04/2003
14. Gordon M-M, Hampson R, Capell HA, Madhok R
(2002) Illiteracy in RA patients as determined by the
rapid estimate of adult literacy in medicine (REALM)
score. Rheumatology 41:750–754
15. Raynor DK, Savage I, Knapp P, Henley J (2004) We
are the experts: people with asthma talk about their
medicine information needs. Patient Educ Couns
53:167–174
16. Tuffnell DJ, Nuttall K, Raistrick J, Jackson TL (1994)
Use of translated written material to communicate
with non-English speaking patients. BMJ 309:992
p