www.nature.

com/npjpcrm
All rights reserved 2055-1010/14

ARTICLE OPEN

Computer decision support systems for asthma:

a systematic review
Patricia Matui1, Jeremy C Wyatt2, Hilary Pinnock1, Aziz Sheikh1 and Susannah McLean1

BACKGROUND: Increasing use of electronic health records offers the potential to incorporate computer decision support systems
(CDSSs) to prompt evidence-based actions within routine consultations.
AIM: To synthesise the evidence for the use of CDSSs by professionals managing people with asthma.
MATERIALS AND METHODS: We systematically searched Medline, Embase, Health Technology Assessment, Cochrane and Inspec
databases (1990 to April 2012, no language restrictions) for trials, and four online repositories for unpublished studies. We also
wrote to authors. Eligible studies were randomised controlled trials of CDSSs supporting professional management of asthma.
Studies were appraised (Cochrane Risk of Bias Tool) and findings synthesised narratively.
RESULTS: A total of 5787 articles were screened, and eight trials were found eligible, with six at high risk of bias. Overall, CDSSs for
professionals were ineffective. Usage of the systems was generally low: in the only trial at low risk of bias the CDSS was not used at
all. When a CDSS was used, compliance with the advice offered was also low. However, if actually used, CDSSs could result in closer
guideline adherence (improve investigating, prescribing and issuing of action plans) and could improve some clinical outcomes.
The study at moderate risk of bias showed increased prescribing of inhaled steroids.
CONCLUSIONS: The current generation of CDSSs is unlikely to result in improvements in outcomes for patients with asthma
because they are rarely used and the advice is not followed. Future decision support systems need to align better with professional
workflows so that pertinent and timely advice is easily accessible within the consultation.
npj Primary Care Respiratory Medicine (2014) 24, Article number: 14005; doi:10.1038/npjpcrm.2014.5; published online 20 May 2014

INTRODUCTION Inclusion criteria

The Global Initiative for Asthma estimates that 300 million people We used the PICOS (Participants, Intervention, Comparator, Outcomes,
worldwide have asthma.1 Prevalence rates as high as 32% have Study design) strategy for describing trials in which we were interested:
been recorded in the United Kingdom and Australia,2 and the
prevalence is increasing in many parts of the world.3–5 Despite Participants. As this study is a review of the evidence, the study
participants were de facto the health professionals using CDSSs who
evidence-based guidelines,1,6–9 there is consistent evidence that
were caring for people with asthma—i.e., doctors, nurses and others
asthma is suboptimally controlled, resulting in unnecessary (e.g., physiotherapists).
morbidity, loss of school and workdays, and high costs for
countries.9–11 There are 250,000 asthma-related deaths each year.1 Intervention. We adopted Wyatt et al.’s definition of CDSs as ’active
There are many reasons why guidelines are poorly implemen- knowledge systems which use two or more items of patient data to
ted, including physician’s lack of knowledge or inertia of generate case-specific advice.’17 Haynes and Wilczynski similarly described
practice.12,13 As electronic health records are now the norm in such systems as ‘information technology which matches characteristics of
many parts of the world,14,15 it is feasible to provide professionals individual patients to a computerised knowledge base’, with software
with computer decision support systems (CDSSs) to prompt algorithms generating patient-specific information in the form of
evidence-based actions within routine consultations, potentially recommendations.18 There are various levels of sophistication for CDSSs,
improving professional adherence to guidelines. from reminders to enter specific data, prescribe certain drugs/vaccines or
Our systematic review aimed to synthesise the evidence for the provide an asthma action plan, to a system retrieving patient asthma
information from an electronic health-care record and providing a critique
use of CDSSs by professionals managing people with asthma. We
on the intended clinical action. Systems were included if they used patient
were primarily interested in the effectiveness of CDSSs in data to generate case-specific asthma advice. Systems relating only to the
improving patient outcomes, but also sought to investigate task of asthma diagnosis or those exclusively providing patients with
process measures of guideline adherence and practical usage of support for self-management were excluded.
the system.
Comparator. The comparator was ‘usual care’, specifically without the use
of a CDSS.
MATERIALS AND METHODS
Our protocol is registered with the PROSPERO international prospective Outcomes. Our primary interest was in the impact of CDSSs on clinical
register of systematic reviews (CRD 42012002412). We followed the asthma control. In line with recommended guidelines,19 we included
methodology described in the Cochrane Handbook for Systematic Reviews outcomes that reflected current control (including asthma-related quality
of Interventions.16 of life) and frequency of asthma exacerbations (including frequency of the

1
Allergy and Respiratory Research and eHealth Research Groups, Centre for Population Health Sciences, The University of Edinburgh, Edinburgh, UK and 2Yorkshire Centre for
Health Informatics, University of Leeds, Leeds, UK.
Correspondence: S McLean (susannah.mclean@ed.ac.uk)
Received 18 October 2013; revised 24 January 2014; accepted 31 January 2014

© 2014 Primary Care Respiratory Society UK/Macmillan Publishers Limited

 Computer decision support systems for asthma
P Matui et al
2
general practitioner’s asthma visits, emergency department asthma visits Quality of reporting of trials
and asthma hospitalisations). We assessed the risk of bias in each trial using the seven-criteria approach
We were also interested in the process by which CDSSs might impact described in the Cochrane Handbook for Systematic Reviews of
asthma control, both practical usage issues (e.g. the proportion of Interventions.16 Overall, each study was rated as follows: A: low risk of
professionals who actually used the CDSS, the numbers of alerts issued bias—no bias found; B: moderate risk of bias—one criterion for risk of bias;
and the impact on time within the consultation) as well as process C: high risk of bias—more than one criterion for risk of bias.
measures reflecting enhanced guideline adherence (e.g. changes in
treatment, in tests ordered and in the proportion of patients with asthma
Synthesis of results
action plans).
We anticipated considerable heterogeneity in the populations studied, and
in the interventions and the outcomes reported in the trials precluding
Study design. All reports of randomised controlled trials of CDSSs used by meta-analysis of data. Instead, we planned to undertake a narrative
health-care professionals for patients with asthma, in any language, synthesis based on our theoretical model of how such computer systems
published and unpublished, were eligible for inclusion. No other study are expected to exert their effects (see Figure 1). The expectation is that, in
designs were included. a linked causal chain, CDSSs will impact process outcomes, which, in turn,
will impact clinical outcomes. The theory underpinning their effectiveness
is that relevant reminders and recommendations during a consultation will
Information sources and search strategy influence clinicians’ behaviour and thereby improve guideline adherence
We searched Medline, Embase, Cochrane Central Register of Controlled as measured by process outcomes (e.g. more rational ordering of
Trials, Health Technology Assessment and Inspec (engineering) data- investigations, prescribing of treatment and use of asthma action plans).
bases from 1990 to April 2012 with the terms listed in Supplementary Implementation of evidence-based practices will consequently be
Appendix 1. We wrote to experts and authors of all included studies measureable in clinical outcomes for asthma patients, such as fewer
requesting additional relevant studies. We searched for ongoing and exacerbations, emergency department attendances and hospitalisations.
unpublished trials on the following websites: https://portal.nihr.ac.uk/
Pages/NRRArchive.aspx, www.clinicaltrials.gov, www.controlled-trials.com
RESULTS
and www.anzctr.org.au.
Study selection
Figure 2 is the PRISMA flow diagram. From 5,787 titles, eight
Study selection studies were selected,20–27 seven in English and one in Spanish.26
Two authors (PM and SM) independently screened titles and abstracts, One study had two reports.24,28 None of the experts we contacted
assessing them against the inclusion criteria. The full text of each identified any additional eligible studies. We found nine ongoing
potentially eligible paper was reviewed by both authors to decide whether and eight unpublished trials (Supplementary Appendix 2).
the study should be included. Disagreements were resolved by discussion We excluded a small group of studies from the early 1990s of
and, if necessary, arbitration of a third researcher (HP, AS or JCW). computerised theophylline dose calculators because they
addressed a specific problem in emergency care and have already
been evaluated in a Cochrane review.29
Data collection and abstraction
Using a piloted data extraction form, PM and SM independently extracted
the following data from included trials: country, setting, funding, study
Study characteristics
design, health-care professionals, patient population, features of the CDSS See Table 1 for details of study characteristics. Most studies were
intervention, description of the control group, outcome measures and any cluster randomised controlled trials20–26 in primary care in the
adverse effects. Extraction tables were compared, and discussed with a UK21,25 or the Netherlands.23,24 Two studies randomised practices
third researcher (HP, AS or JW) arbitrating in the event of unresolved to receive a CDSS for asthma prescribing or a system for angina or
disagreement. cholesterol prescribing.21,24 The practices providing data on

Facilitators Stage Barriers

Financial incentives Guideline Perceived lack of credibility of

development guideline
changes in treatment protocols

Smooth integration of software Software Prototype does not work in practice

into electronic health record development

User-friendly design, Practical Usability issues: too much non-

easy (potentially automatic) aspects specific information, difficulty
access to relevant advice of use of activating CDSS, e.g., separate
CDSS screen, long time to load,
password protection

Relevant recommendations Process Clinicians choose not to implement

customised to clinical context outcomes recommended changes

Measureable improvement Health

in patients’ conditions outcomes

Figure 1. Theoretical model showing how a computer decision support system can improve asthma outcomes.

npj Primary Care Respiratory Medicine (2014) 14005 © 2014 Primary Care Respiratory Society UK/Macmillan Publishers Limited
 Computer decision support systems for asthma
P Matui et al
3
Medline Embase Health technology Cochrane INSPEC
1,252 3,163 assessment 1,398 161
52

6,026 titles

239 duplicates
10 Ongoing or deleted
unpublished trials from:
5,787 titles
https://portal.nihr.ac.uk/
Pages/NRRArchive.aspx 5,745 titles excluded
and by title and abstract
www.clinicaltrials.gov, screening
www.controlled-
trials.com 41 titles
www.anzctr.org.au
1 Study protocol only
15 not RCTs
10 not CDSSs
4 theophylline calculators
8 Studies included
2 others
(described in 9 articles)

Figure 2. PRISMA flow diagram.

angina and cholesterol prescribing were unaware that their (usual Process outcomes
care) asthma prescribing data were control data for the parallel Changes in tests ordered. Eccles et al.,21 McCowan et al.25 and
asthma study. Plaza et al.26 all reported that the systems made no difference in the
Six of the systems were integrated into an electronic health rates of ordering spirometry, X-rays, allergy tests or blood tests. Bell
record:20,22–24,26,27 one was partly integrated21 and one was a et al.20 reported an increase in spirometry requests at intervention
stand-alone system.25 Five of the studies20,21,23,26,27 explicitly practices from 15 to 24%, whereas there was a decrease at control
reported that the system gave prescribing advice and reminders. practices from 8 to 1%. In Kuilboer et al.,23 peak expiratory flow rate
One system concentrated solely on the prescribing of influenza and spirometry tests were ordered more often in the intervention
vaccine for ‘at-risk’ children.27 Four studies were based on asthma group, in patients over 11 years of age. In a four-arm trial, Tierney
management guidelines.21,25–27 One system included a complex et al.27 reported that between 39 and 50% of patients received the
risk prediction algorithm,25 and one system ‘critiqued’ the doctor’s suggestion to obtain pulmonary function tests.
intended management plan and made recommendations.23
Changes in treatments. Eccles et al.,21 the only trial at low risk of
Risk of bias within studies bias, found no difference in asthma-related prescribing as a result
Table 2 lists the quality assessment: most studies were rated at of the intervention. Martens et al.24 demonstrated an increase in
high risk of bias. The study by Eccles et al.21 was rated at low risk the prescribing of inhaled corticosteroids to 44% of asthma
of bias and that by Martens et al.24 at moderate risk of bias. patients (95% confidence interval (CI), 30–56%) in the intervention
group, compared with 27% (95% CI, 14–47%) in the control group.
In the trial by Bell et al.,20 there was a highly significant
Effectiveness of CDSSs
(P = 0.006) difference between the rate of prescribing inhaled
The impact of CDSS on process, usage and clinical outcomes is corticosteroids in the subgroup of urban intervention practices
detailed in Table 3. It was anticipated that usage and process compared with urban control practices. Urban and suburban
outcomes would influence clinical outcomes as reflected in our practices were analysed separately in the cluster controlled trial
model (Figure 1). because of marked baseline differences in patient population: the
urban practices had more severe asthma.
Practical aspects of CDSS use Kuilboer et al.23 demonstrated a significant reduction in the
In the study by Eccles et al.,21 the median number of activations of prescribing of cromoglyate in a post hoc analysis. Plaza et al.26
the system per practice was zero. In that by Kuilboer et al.,23 demonstrated a doubling of treatment conforming to guidelines,
10,863 visits generated 10,532 decision support comments, from 18 to 34% (P = 0.02). Vaccination rates increased in both arms
but the doctor waited for the critique only 22% of the time, of the Fiks trial with no significant differences.22 McCowan et al.25
and then read only a third of them. In Tierney et al.’s study,27 found no difference in asthma-related prescribing between the
doctors complied with a third of the systems’ suggestions. trial arms due to the intervention. Tierney et al.27 reported on
Bell et al.20 reported that the CDSS was used 70% of the time. treatment suggestions for both asthma and chronic obstructive
In the study by Fiks et al.,22 the vaccine alerts were only active pulmonary disease. For example, across the four arms of the
during 27% of visits. Tierney trial, between 5 and 9% of patients received the suggest-

© 2014 Primary Care Respiratory Society UK/Macmillan Publishers Limited npj Primary Care Respiratory Medicine (2014) 14005
 Computer decision support systems for asthma
P Matui et al
4
Table 1. Characteristics of studies

Author (country) Study design Participants and setting Age Time scale Intervention Control
(years)

Bell et al.20 (USA) Cluster RCT 12 clusters: 12 primary 0–18 12 months CDSS embedded in an The control group
care practices, 19,450 6 months prior to electronic health record experienced educational
patients study start (EHR) in the form of alerts programme for
clinicians and reminders based on professionals. It also had
participated in an expert asthma guidelines. access to the data entry
educational This included a data entry and all documentation
programme, tool, standardised tools but only passively,
12 months of documentation for asthma without alerts and
intervention severity classification, reminders.
standardised drug and
spirometry order sets and
an asthma control plan.
There was also an
educational programme for
professionals.
Eccles et al.21 (UK) Cluster RCT with 60 clusters: 60 primary ⩾ 18 24 months CDSS offered suggestions Controls received
2 × 2 incomplete care practices, 1,129 12 months for management (including intervention for angina,
block design patients baseline period, prescribing) depending on while the asthma
12 months the chosen clinical scenario intervention group was
intervention and requested the entry of the control from the
relevant information. angina group as a
strategy to balance the
Hawthorne effect.
Fiks et al.22 (USA) Cluster RCT 20 clusters: 20 5–19 6 months CDSS was an EHR-based Described as routine
practices, 6,110 All intervention influenza vaccination alert care.
patients system. Influenza vaccine
alerts appeared
prominently at the top of
the computer screen in
bold and highlighted text
whenever the electronic
health record was opened
for a study subject who was
due for this vaccine. Also a
link was provided to
simplify vaccine ordering.
Kuilboer et al.23 Cluster RCT 40 clusters: 32 primary All 10 months ‘AsthmaCritic’, the CDSS, Described as usual care.
(The Netherlands) care practices with a 5 months baseline relied solely on the existing
total of 40 GPs, each period, 5 months data in the EHR. Once data
control practice with a intervention related to the visit was
mean of 4,933 control entered, the system
and 4,865 intervention evaluated whether the
patients patient had asthma or
COPD, reviewed the
physician’s treatment of
asthma and COPD, and
generated feedback. In this
way, the doctor made the
decisions and the CDSS
‘critiqued’ these decisions.
Martens et al.24,28 Cluster RCT with 53 clusters, 14 All 12 months CDSS was part of a One group that received
(The Netherlands) an incomplete practices with a 6 months computer-reminder system prescription reminders
block design total of 53 GPs intervention, integrated into the EHR as a for cholesterol-lowering
6 months data prescribing module. When drugs served as controls
collection the GP prescribed a drug for the other group that
the decision support received CDSS for
system was activated and antibiotics, asthma and
provided information COPD, and vice versa.
specific to the patient (e.g.,
age and gender) and the
prescribed drug. The GP
was obliged to enter a
diagnosis code which the
CDSS would check and use
to issue relevant reminders.
McCowan et al.25 Cluster RCT 40 clusters: 40 All 6 months ‘Asthma Crystal Byte’ was a The control group had
(UK) practices, 477 patients No baseline data stand-alone decision no knowledge of the
support system with intervention and had to
management guidelines report parallel data on
for asthma that aimed to the same number of
improve the quality of the patients as were
consultation. It included recruited to the
risk prediction software intervention group.
and printed asthma
management plans.

npj Primary Care Respiratory Medicine (2014) 14005 © 2014 Primary Care Respiratory Society UK/Macmillan Publishers Limited
 Computer decision support systems for asthma
P Matui et al
5
Table. 1. (Continued )

Author (country) Study design Participants and setting Age Time scale Intervention Control
(years)

Plaza et al.26 Cluster RCT 20 clusters: 10 ⩾ 14 12 months CDSS providing patient- The control group
(Spain) pulmonologists and 6 months baseline tailored recommendations worked as normal but
10 GPs, 198 patients and 2 sessions of based on the GINA recorded additional data
educational guidelines enabled for comparison.
programme for clinicians to establish the
clinicians, severity of asthma
12 months according to the GINA
intervention classification, from relevant
inputs such as PEFR,
symptom frequency,
quantity of corticosteroids
and the clinician’s
professional opinion. Then
the CDSS would
recommend medications
according to the GINA
guidelines. There were also
education programmes for
clinician and patients,
teaching inhaler technique
and general information
about the condition of
asthma.
Tierney et al.27 2 × 2 factorial 4 clusters: 4 hospital- ⩾ 18 36 months CDSS generated care Care suggestions were
(USA) randomisation of based academic 28 months suggestions based on still generated by the
patients practices with 25 recruitment and agreed guidelines. These CDSS but were not
faculty general baseline, include performing displayed to the
internists and over 100 8 months pulmonary function tests, physician or pharmacists
internal medicine intervention giving influenza and caring for patients in the
residents, 1 full-time pneumococcal control group.
and 9 part-time vaccinations, prescribing
pharmacists, 706 advice and encouraging
patients smoking cessation. These
suggestions were
presented on doctors’
workstations or were
printed under a list of
active medications that
doctors received along with
the patient’s paper chart
when he/she presented for
usual care.
Abbreviations: CDSS, computer decision support system; COPD, chronic obstructive pulmonary disease; GINA, The Global Initiative for Asthma; GP, general
practitioner; RCT, randomised controlled trial.

ion to ‘start inhaled corticosteroids.’ However, only 11–30% of the scales,33,34 but found a significant result only in one subdomain,
physicians or pharmacists complied with this suggestion. possibly due to multiple testing.

Clinical outcomes Frequency of asthma exacerbations. Two studies reported exacer-

bation rates. In the study by Plaza et al.,26 exacerbation rates were
Asthma symptoms. Three studies reported asthma
symptoms.21,25,26 Eccles and coworkers30 reported that the CDSS not significantly different between the control and intervention
had no effect on the validated Newcastle Asthma Symptoms groups: mean exacerbations, 1.3 (s.e.) = 1.2) in the control group
Questionnaire (mean difference − 0.6 (95% CI, − 2.1 to 0.9)).21 and 0.5 (s.e. = 0.3) in the intervention group (Wilcoxon P = 0.2).
Plaza et al.26 reported that asthma daytime symptoms, but McCowan et al.25 reported that in the intervention group 12/147
not night-time symptoms, were significantly reduced in the patients had exacerbations compared with 57/330 in the control
intervention group compared with the control group (Wilcoxon group: control patients were approximately twice as likely to
P o0.02). McCowan et al.25 reported no significant differences in experience an exacerbation as were intervention patients (odds
asthma symptoms between the intervention and control groups ratio 0.4, 95% CI, 0.2–0.9, after adjustment for clustering). The
(odds ratio 0.3, 95% CI, 0.03–3.3), although this study was denominators were different because of study dropouts.
underpowered.
Unscheduled health-care utilisation. McCowan et al.25 reported
Asthma-related quality of life. Three studies reported asthma- significantly fewer unscheduled general practitioner consultations
related quality of life.21,26,27 The study by Eccles et al.,21 a trial at in the intervention group in comparison with the control group
low risk of bias, reported no effect on the validated Asthma (odds ratio 0.6, 95% CI, 0.4–0.95). Four studies reported no
Quality of Life Questionnaire.31 Plaza and coworkers32 reported differences in the frequency of asthma-related visits to the general
quality of life using the Spanish version of the St George’s practitioner.20,22,23,26
Respiratory Questionnaire and found significant improvement in Two studies reported no significant difference between the
all domains (activity P = 0.002, symptoms P = 0.003, impact intervention and control groups in emergency department visits
P = 0.001).26 Tierney et al.27used two different quality-of-life or hospitalisations.25,27 The absolute numbers were close to zero.

© 2014 Primary Care Respiratory Society UK/Macmillan Publishers Limited npj Primary Care Respiratory Medicine (2014) 14005
 6
Table 2. Risk of bias summary table

Trial Selection bias Allocation concealment Performance bias Detection bias Attrition bias Selective reporting Other Quality
bias

Bell et al.20 Yes—there were ethnic No allocation Yes—there was no Unclear—no mention Unclear as to how many of Unclear—no pre- No C—high risk
differences between concealment blinding for users of blinding of the patients enroled at each published protocol.
suburban and rural outcome assessors practice remained in the
practices; however, trial—pragmatic design,
clustering would have denominator quite flexible,
helped to control for withdrawals not reported.
this
Eccles et al.21 No—minimised by No allocation No—GPs were acting No—data collectors No—attrition rates were No—a pre-published No A—low risk
computerised concealment as controls for the were blinded to the presented and balanced; protocol-outlined
randomisation of other block status of practice there were 31 intervention plan for data analysis
practices in a cluster practices and 29 control and embedded case
design practices who completed the study and economic
study and two withdrawals. evaluation.
Fiks et al.22 Unclear—no details of No allocation Yes—no blinding, Unclear—no mention No—attrition fairly Unclear—possibility No C—high risk

npj Primary Care Respiratory Medicine (2014) 14005

randomisation concealment clinicians were aware of blinding of balanced—no patients of post hoc analysis of
that their software outcome assessors withdrew; however, there vaccination rates to
either did or did not was fluctuation in the achieve higher
have the alerts numbers of patients, as may rates—no pre-
be expected in such a large published analysis
cohort. protocol.
Kuilboer et al.23 No—randomisation No allocation Yes—there was no Unclear—no mention No—flow diagram explains Unclear—no pre- No C—high risk
performed with a table concealment blinding for GP users of blinding of why patients dropped out or published protocol.
of random numbers by outcome assessors withdrew. No attrition at
P Matui et al
Computer decision support systems for asthma

a researcher who was practice level.

blinded to the identity
of practices
Martens et al.24,28 Unclear—no details of Yes—GPs blinded as No—GPs did not Unclear—no mention No—attrition was fairly Unclear—no pre- No B—moderate
randomisation to whether they were know that they were of blinding of balanced but resulted in the published protocol. risk
assessed on acting as controls for outcome assessors study being underpowered.
treatment of the other block Reasons for attrition were
cholesterol or asthma given.
and COPD
McCowan et al.25 No—randomisation No allocation Yes—there was no Unclear—no mention No—attrition was Unclear—no pre- No C—high risk
using random number concealment blinding of GPs of blinding of unbalanced and although published protocol.
sequence and outcome assessors most practices gave some
performed reasons this resulted in the
independently of the study being underpowered
project administration and intention-to-treat
team analysis was impossible due
to insufficient information.
Plaza et al.26 No—randomisation No allocation Unclear, not reported Unclear—no mention No—clinician withdrawals Unclear—no pre- No C—high risk
using SAS (statistics concealment of blinding of reported (2/22) due to published protocol.
programme). Patients outcome assessors administrative problems,
were recruited as they patient withdrawals also
came for consultation reported in diagram.
Tierney et al.27 No—randomisation by No allocation Yes—there was no Unclear—no mention No—flow diagram explains Yes—no pre- No C—high risk
flip coin, then switching concealment for blinding of GPs of blinding of why patients dropped out or published protocol
to equal numbers of professionals or outcome assessors withdrew. Attrition appeared and post hoc analysis
consultations per arm patients to be fairly balanced. of power calculation.
by a researcher blinded
to allocation.
Abbreviations: COPD, chronic obstructive pulmonary disease; GP, general practitioner.

© 2014 Primary Care Respiratory Society UK/Macmillan Publishers Limited

 Table 3. Effectiveness of CDSS: process outcomes—guideline adherence

Study Risk of Practical aspects of CDSS use Process outccomes Clinical outcomes Interpretation
bias

Eccles Low For both groups the median No significant difference in drugs Overall effect of the CDSS on symptom score The design of this British study incorporated two arms, each
et al.21 number of active interactions prescribed for asthma before and was non-significant: the parameter estimate controlling for the other. The study was a cluster design, with
was zero. The number of alerts after introduction of CDSS. No from analysis of covariance of scale was − 0.62 practices as the unit of randomisation. Practices investigating
was approximately zero significant difference in lung (95% CI is − 2.12 to 0.88).28 CDSS-driven care for angina provided usual care control data for
function assessment before and No effect on quality of life was measured on the the asthma CDSS care practices, and vice versa. In addition, the
after OR 0.94 (0.67–1.33) validated Juniper’s Asthma Quality of Life study was very large, with 62 practices across the UK, and so
Questionnaire (AQLQ).30 No differences in GP results should have been robust. This trial demonstrated very
visit rate; OR = 0.94 (0.81–1.08) clearly that CDSS will not be used by clinicians if it is not
integrated with their usual workflow. The median usage of the
CDSS in this study was zero and there were no differences in
consultation rates, process outcomes or clinical outcomes, which
were carefully measured.
Martens Medium GPs did not have a choice to 44% of the intervention group were No clinical outcomes reported This Dutch study consisted of 14 general practices in a cluster
et al.24,28 decide if the CDSS was to be prescribed according to the randomised controlled trial. As in the Eccles study, two arms of
activated recommendations compared with the study acted as controls for each other. One arm was given a
27% of the control group among CDSS to guide on antibiotic, asthma and COPD prescribing, and
patients with mildly persistent the other received CDSS for cholesterol prescribing. This design
asthma minimises the impact of performance bias and the Hawthorne
effect and has therefore contributed to it being rated as only at
moderate risk of bias. The study was underpowered (the actual
variation was larger than values used to estimate study power),
which may have contributed to the non-significant results.
Bell et al.20 High No difference between groups Controller medication prescribed No differences in GP visits Although this US study was graded at high risk of bias, it did
in the rate at which the CDSS more often in urban intervention have a recognisable cluster design in which steps were taken to
was used (70% of the time practices compared with urban try to randomise the baseline differences of poverty and

© 2014 Primary Care Respiratory Society UK/Macmillan Publishers Limited

during the intervention control practices (P = 0.006) and ethnicity in the different urban versus suburban practices. This
periods) NSD in suburban practices. Increase study demonstrated that CDSS could improve the adherence to
in spirometry in intervention sites guidelines for prescribing, test ordering and use of asthma
from 15 to 24% and decrease in action plans. No clinical improvements were measured or
control sites from 8 to 1% reported in this trial and a major confounder was the
(P = 0.003). Number of asthma plans introduction of asthma care-related pay-per-performance
filed in suburban intervention incentives during the time period of this trial (though this
practices increased compared with applied to both groups).
P Matui et al

suburban controls (P = 0.03). NSD in

urban practices
Fiks et al.22 High Influenza vaccine alerts were Vaccination rates increased by 3.8% No differences in GP visits This American study investigated the impact of CDSS for
active at only 27% of visits at control practices and by 4.8% at reminding clinicians to give children with asthma an influenza
intervention sites vaccination. The rate of increase in vaccination was not
significantly different across the control and intervention groups
as the rate increased in both groups. In interpreting this study it
should be remembered that there are many influences on the
uptake rate of vaccination, including whether a child is acutely
unwell or not at the time they attend the clinic, and the health
beliefs of the child and their parents.
Kuilboer High The doctor waited for the result Some evidence for a decrease in No differences in GP visits except in one of the This trial provides some evidence of the effectiveness of CDSS in
et al.23 of the CDSS analysis in 22% of cromoglycate prescriptions in one four age brackets, but risk of multiple testing terms of its impact on guideline adherence. There were
10,863 visits. 10,532 comments of four age brackets, but no other appreciable increases in the ordering of peak expiratory flow
were produced and 32% of significant changes. More tests rates and spirometry. In addition, there was some evidence that
these were read by doctors. The were ordered among the CDSS doctors were more likely to change their prescribing of
Computer decision support systems for asthma

CDSS took on average 31.7 s to group, but this difference was not cromoglycate with the CDSS; however, there were no changes
analyse the record. The median always significant for the other drugs in the guideline (deptropine, antihistamines
time spent by the doctor and oral bronchodilators)—probably because the general
reading comments was 9 s practitioners rarely prescribed these drugs anyway. Also
(25th percentile = 4 s, 75th measured were changes in the coding of the record: doctors
percentile = 48 s) recorded more data in a more structured fashion. It was
reported that only a third of the comments were read by
doctors. The explanation for this may be that the CDSS provided
asthma-related comments irrespective of the reason for the visit.

npj Primary Care Respiratory Medicine (2014) 14005

7
 8
Table. 3. (Continued )

Study Risk of Practical aspects of CDSS use Process outccomes Clinical outcomes Interpretation
bias

McCowan High Usually less than 10 min to fill in There was no difference in the Reported no significant differences in asthma From an initial 46 UK practices who registered to undertake the
et al.25 the template and generate the proportions of patients in the symptoms between the intervention and trial only 12 control practices and 5 intervention practices
advice according to a nested different categories of maintenance control groups (odds ratio 0.31, 95% CI, completed the trial. A significant number from the intervention
study prescribing according to the British 0.03–3.32) practices had problems installing and using the software at the
asthma guidelines. No difference in In the CDSS intervention group, 12/147 patients trial initiation. The CDSS was apparently partially effective in that
PEFRs ordered. No difference in had exacerbations and in the control group there were significantly fewer exacerbations of asthma among
proportion with action plans 57/330 patients had exacerbations; OR = 0.43 intervention patients. However, the majority of outcomes
(95% CI, 0.21–0.85) after adjusting for clustering. (symptoms, inhaler technique and measurement of peak flow)
Therefore control patients were approximately were not statistically significantly different between control and
twice as likely to experience an exacerbation as intervention arms. This is on the basis of those who completed
intervention patients the trial; the data were not analysed by intention-to-treat
Significantly fewer patients initiated GP analysis.
consultations in the intervention group; OR 0.59
(0.37–0.95)
No difference in emergency department visits:
OR = 0 (0–9.16)

npj Primary Care Respiratory Medicine (2014) 14005

No difference in asthma hospitalisations; OR = 0
(0–3.44)
Plaza et al.26 High Not reported 17.9 of control and 34% of The number of patients with symptoms during This Spanish study reported randomising groups (clusters) to
intervention patients conformed to the day in the intervention group was either the intervention or the control arm. It was a small study
strict treatment guidelines significantly less than that in the control group with only 10 doctors in each arm. There were two components
(Wilcoxon P = 0.0240). No difference (Wilcoxon P o 0.02). There was no difference to the intervention: the CDSS and an asthma education
in spirometry rates, X-rays allergy or between the groups in terms of nocturnal programme for nurses based on the GINA guidelines. This study
blood tests symptoms (Wilcoxon P = 0.1). Exacerbation produced significant improvements in the measures of the St
rates were not significantly different between George’s quality of life questionnaire. Daytime symptoms and
the control and the intervention groups. Quality exacerbations also improved but night-time symptoms did not.
P Matui et al
Computer decision support systems for asthma

of life reported using the validated Spanish This study clearly demonstrated a link between significantly
version of the St George’s Respiratory higher prescribing in the intervention arm of long-acting beta-
Questionnaire (SGRQ)31 showed significant agonists (especially formoterol) and leukotriene antagonists as
improvement by more than the minimally per the guidelines and improved short-term outcomes (within
important difference of four points in all 6 months). There was no significant difference in the rate of
domains (activity P = 0.002, symptoms P = 0.003, prescribing of inhaled steroids, oral steroids, anticholinergics or
impact P = 0.001) cromoglycate. This intervention was applied over a winter to
No difference in GP visits (P>0.1) spring period, which may have been a confounding factor in a
No difference in emergency department visits seasonal condition such as asthma.
(P = 0.0888)
No difference in asthma hospitalisations (P>0.1)
Tierney High 87–95% of consultations 5–9% of patients received the The authors reported that patients with asthma This study had four arms: one control and three intervention.
et al.27 resulted in the generation of a suggestion to ‘start inhaled in the pharmacist intervention arm of the trial The intervention arms consisted of physician CDSS intervention,
suggestion; doctors complied corticosteroids.’ 11–0% of clinicians had significantly (P o 0.05) improved scores in pharmacist CDSS intervention and both physician and
with only 32–37% of who received this suggestion the emotion subscale and that this was the only pharmacist intervention. There were no significant differences
suggestions adhered to it. Pulmonary function significant result following analysis of between the four study groups in adherence to the care
tests: 6% of the 39% in the control covariance of quality of life scores. It seems suggestions. However, the care suggestions were also generated
group and between 6 and 12% of likely that this result may be significant only as a for the control patients—only that they were on paper, not on
40–50% in the three intervention result of multiple testing the computer. Adherence to care suggestions for the control
groups who received the No difference in emergency department vists or arm varied from 9 to 71%. Adherence to care suggestions for the
suggestion adhered to it asthma hospitalisations physician and pharmacist arm was from 12 to 91%. Overall, there
was no clear pattern. It may be surmised that as the adherence
to suggestions was very variable and frequently less than 50%
this may explain why no significant differences were found in
the quality of life and asthma control questionnaires.

Figures in brackets represent 95% confidence intervals.

Abbreviations: CDSS, computer decision support system; CI, confidence interval; COPD, chronic obstructive pulmonary disease; GINA, The Global Initiative for Asthma; GP, general practitioner; NR, not reported;
NSD, no significant difference; OR, odds ratio.

© 2014 Primary Care Respiratory Society UK/Macmillan Publishers Limited

 Computer decision support systems for asthma
P Matui et al
9
DISCUSSION suitable basis for a full description. Future research should
Main findings substantiate our theoretical model (Figure 1), which we suggest
We found eight relevant trials, four of which reported clinical as a possible useful framework. In terms of the logical chain from
measures of asthma control.21,25–27 The key finding was that CDSSs usage to process outcomes to clinical outcomes, Bell et al.20
for health-care professionals were ineffective in improving patient demonstrated that usage rates have an impact on process
outcomes because the systems were rarely used,21–23 and there was outcomes, and Plaza et al.26 demonstrated the impact of process
low compliance with the advice when it was issued.23,27 However, outcomes on health outcomes. However, we feel that further
when systems are used, clinical outcomes can improve.20,25 research is required to evidence this model more thoroughly.

Strengths and limitations of this study Conclusions

A strength of this review is its robust search strategy. We used the Our review suggests that current CDSSs are unlikely to result in
Cochrane-suggested terminology for asthma and randomised improved outcomes in asthma because they are rarely used and
controlled trials, and drew on our eHealth research group’s the advice not followed. A key challenge in the future design of
inclusive search terms for CDSS.35 Nevertheless, we may have decision support systems lies in the better integration and
missed some relevant studies, and the list of ongoing trials alignment with professional workflows such that they are adopted
suggests that more evidence may be available in due course. into routine practice.
In contrast to the methodology used by the recent McMaster
group series of reviews in which improvement was considered to
have occurred if >50% of the selected outcomes showed ACKNOWLEDGEMENTS
benefit,18,36–41 we report specific clinical, usage and process We acknowledge the support of Lisette van den Bemt (PM’s supervisor from Radboud
outcomes from each trial to explain why the systems were having University Nijmegen).
an effect or not.
We did not perform a meta-analysis as populations and
outcomes across trials were too heterogeneous. Descriptions of CONTRIBUTIONS
interventions were often poorly described, which may have PM and SM, with AS, HP and JCW, wrote the protocol and undertook the
limited our interpretation of the findings. searches, selection of studies and extraction of data. All authors contributed to
the interpretation of the findings. PM and SM wrote the initial draft of the
Interpretation of findings in relation to previously published work paper. All the authors contributed to and have approved the final text. SM is the
study guarantor.
Our review focuses on asthma as a clinically important area for
CDSSs. A crucial observation was that the systems were rarely
used.13,21,42 Usage was not considered in the recent McMaster COMPETING INTERESTS
group’s meta-regression,43 although this is clearly fundamental to
All authors have completed the ICMJE uniform disclosure form (www.icmje.org/
understanding the reasons for lack of effect, and should be a
crucial focus for development if systems are to improve patient coi_disclosure.pdf). AS serves on the World Health Organization's Health and
outcomes above the 15–31% impact on outcomes reported by the Information Technology for Patient Safety Expert Working Groups and is an
McMaster group in a series of reviews.36–41 Usage rate of the adviser to NHS Connecting for Health's Evaluation Programme. He is a
systems should be a core standard for reporting trials of CDSSs. consultant to ALK and Phadia, and has received support from Napp, Pfizer and
The McMaster group’s meta-regression explored the features of Cheisi for research advice. AS is Joint Editor-in-Chief of, and HP is an Associate
CDSSs associated with system ‘effectiveness’. They found editor of, the PCRJ; neither were involved in the editorial review of, nor the
(1) stand-alone programs, (2) advice directed at both health-care decision to publish, this article. HP has spoken for AstraZeneca, Boehringer
practitioners and patients, (3) requiring users to enter an Ingelheim, Chiesi, GlaxoSmithKline, Pfizer and Teva and undertaken advisory
explanation for any overrides of system recommendations and group work for Chiesi. The remaining authors declare no conflict of interest.
(4) developers’ involvement in trials to be associated with better
patient outcomes. Poor integration (as in a stand-alone program),
however, risks clinicians avoiding using the system as in Eccles FUNDING
et al.21 The issue, however, is complex as advice presented at the There was no specific funding for this research. HP was supported by a Primary
time of care does not always predict success, possibly because Care Research Career Award from the Chief Scientist’s Office of the Scottish
practitioners become overwhelmed by such integrated alerts that Government at the time of work on this review.
interrupt their workflow.43
Our recent analysis of recordings of general practice consulta-
tions emphasised the importance of the timing of alerts in the REFERENCES
context of prescribing safety CDSSs.44 The practitioner, negotiat- 1 Global Initiative for Asthma. The global strategy for asthma management and
ing with the patient, makes decisions regarding drugs and prevention, 2012. Available from http://www.ginasthma.org (accessed September
management throughout the consultation when information 2013).
about allergies, sensitivities, interactions and guideline recom- 2 Patel S, Jarvelin M, Little M. Systematic review of worldwide variations of the
prevalence of wheezing symptoms in children. Environ Health 2008; 7: 57.
mendations might be useful. Provision of information during the
3 Beasley R for the International Study of Asthma and Allergies in Childhood
final computer-based task of generating the prescription can (ISAAC) Steering Committee. Worldwide variation in prevalence of symptoms of
frustrate clinicians, who then override the alerts. Integration with asthma, allergic rhinoconjunctivitis, and atopic eczema. Lancet 1998; 351:
workflow requires a detailed study of the consultation process. 1225–1232.
4 Anandan C, Nurmatov U, van Schayck OC, Sheikh A. Is the prevalence of asthma
Implications for future research, policy and practice declining? Systematic review of epidemiological studies. Allergy 2009; 65:
152–167.
A detailed description of the CDSS intervention under investiga- 5 Gupta R, Sheikh A, Strachan DP, Anderson HR. Burden of allergic disease in the
tion is essential to providing insight into what promotes a well- UK: secondary analyses of national databases. Clin Exp Allergy 2004; 34: 520–526.
used and effective system that can inform future development. 6 National Heart, Lung and Blood Institute. Guidelines for the diagnosis and man-
Taxonomies and frameworks such as those described by agement of asthma, 2007. Available from http://www.nhlbi.nih.gov/guidelines/
Kawamoto et al.,45 Garg et al.46 or Berlin et al.47 may provide a asthma/ (accessed September 2013).

© 2014 Primary Care Respiratory Society UK/Macmillan Publishers Limited npj Primary Care Respiratory Medicine (2014) 14005
 Computer decision support systems for asthma
P Matui et al
10
7 British Thoracic Society/Scottish Intercollegiate Guideline Network. British 29 Durieux P, Trinquart L, Colombet I, Niès J, Walton R, Rajeswaran A et al. Computerized
guideline on the management of asthma, 2012. Available from http://www.sign. advice on drug dosage to improve prescribing practice. Cochrane Database Syst Rev
ac.uk/guidelines/fulltext/101/index.html (accessed September 2013). 2008, CD002894. doi:10.1002/14651858.CD002894.pub2.
8 Canadian Thoracic Society. CTS 2012 Guideline Update: diagnosis and manage- 30 Steen N, Hutchinson A, McColl E, Eccles MP, Hewison J, Meadows KA et al.
ment of asthma in preschoolers, children and adults. Available from http://www. Development of a symptom based outcome measure for asthma. BMJ 1994; 309:
ncbi.nlm.nih.gov/pmc/articles/PMC3373283 (accessed September 2013). 1065–1068.
9 Rabe KF, Adachi M, Lai CKW, Soriano JB, Vermeire PA, Weiss KB et al. Worldwide 31 Juniper EF, Buist AS, Cox FM, Ferrie PJ, King DR. Validation of a standardized
severity and control of asthma in children and adults: the global asthma insights version of the Asthma Quality of Life Questionnaire. Chest 1999; 115: 1265–1270.
and reality surveys. J Allergy Clin Immunol 2004; 114: 40–47. 32 Ferrer M, Alonso J, Prieto L, Plaza V, Monsó E, Marrades R et al. Valididty and
10 Haughney J, Barnes G, Partridge M, Cleland J. The Living & Breathing study: a reliability of the St George's Respiratory Questionnaire after adaptation to a
study of patients' views of asthma and its treatment. Prim Care Resp J 2004; 13: different language aand culture: the Spanish example. Eur Respir J 1996; 9:
28–35. 1160–1166.
11 Stallberg B, Lisspers K, Hasselgren M, Janson C, Johansson G, Svardsudd K. 33 Guyatt GH, Berman LB, Townsend M, Pugsley WO, Chanbers LW. A measure of
Asthma control in primary care in Sweden: a comparison between 2001 and 2005. quality of life for clinical trials in chronic lung disease. Thorax 1987; 42: 773–778.
Prim Care Respir J 2009; 18: 279–286. 34 Juniper E, Guyatt GH, Ferrie PJ, Griffith LE. Measuring quality of life in asthma.
12 Cabana MD, Rand CS, Powe NR, Wu AW, Abboud PC, Rubin HR. Why don't phy- Am Rev Resp Dis 1993; 147: 832–838.
sicians follow clinical practice guidelines? A framework for improvement. JAMA 35 Car J, Black A, Anandan C, Cresswell K, Pagliari C, McKinstry B et al., The Impact of
1999; 282: 1458–1465. eHealth on the Quality & Safety of Healthcare. Connecting for Health Evaluation
13 Wiener-Oglilvie S, Pinnock H, Huby G, Sheikh A, Partridge MR, Gillies J. Do prac- Programme 001 Report. University of Birmingham: Birmingham, UK, 2011.
tices comply with key recommendations of the British Asthma Guideline? If not, 36 Roshanov PS, Misra S, Gerstein HC, Garg AX, Sebaldt RJ, Mackay JA et al., for the
why not? Prim Care Respir J 2007; 16: 369–377. CCDSS SystematicReviewTeam. Computerized clinical decision support systems
14 Robertson A, Cresswell K, Takian A, Petrakaki D, Crowe S, Cornford T et al. for chronic disease management: a decision-maker-researcher partnership
Implementation and adoption of nationwide electronic health records in sec- systematic review. Implement Sci 2011; 6: 92.
ondary care in England: qualitative analysis of interim results from a prospective 37 Sahota N, Lloyd R, Ramakrishna A, Mackay JA, Prorok JC, Weise-Kelly L et al., for
national evaluation. BMJ 2010; 341: c4564. the CCDSS SystematicReviewTeam.. Computerized clinical decision support
15 Sheikh A, Cornford T, Barber N, Avery A, Takian A, Lichtner V et al. Implementation systems for acute care management: a decision-maker-researcher partnership
and adoption of nationwide electronic health records in secondary care in Eng- systematic review of effects on process of care and patient outcomes. Implement
land: final qualitative results from prospective national evaluation in "early Sci 2011; 6: 91.
adopter" hospitals. BMJ 2011; 343: d6054. 38 Nieuwlaat R, Connolly SJ, Mackay JA, Weise-Kelly L, Navarro T, Wilczynski NL et al.,
16 Higgins J, Green S. Cochrane Handbook for Systematic Reviews of Interventions,, for the CCDSS SystematicReviewTeam. Computerized clinical decision support
Version, 5.1.0 edn. The Cochrane Collaboration, 2011. http://www.cochrane.org/ systems for therapeutic drug monitoring and dosing: a decision-maker-researcher
training/cochrane-handbook (accessed September 2013). partnership systematic review. Implement Sci 2011; 6: 90.
17 Wyatt J, Spiegelhalter D. Field trials of medical decision-aids: potential problems 39 Hemens BJ, Holbrook A, Tonkin M, Mackay JA, Weise-Kelly L, Navarro T et al., for
and solutions. Proc Annu Symp Comput Appl Med Care 1991, 3–7. the CCDSS SystematicReviewTeam. Computerized clinical decision support sys-
18 Haynes RB, Wilczynski NLfor the CDSS SystematicReviewTeam. Effects of com- tems for drug prescribing and management: a decision-maker-researcher part-
puterized clinical decision support systems on practitioner performance and nership systematic review. Implement Sci 2011; 6: 89.
patient outcomes: methods of a decision-maker-researcher partnership 40 Roshanov PS, You JJ, Dhaliwal J, Koff D, Mackay JA, Weise-Kelly L et al., for the
systematic review. Implement Sci 2010; 5: 12. CCDSS SystematicReviewTeam. Can computerized clinical decision support sys-
19 Reddel HK, Taylor DR, Bateman ED, Boulet LP, Boushey HA, Busse WW et al. An tems improve practitioners' diagnostic test ordering behavior? A decision-maker-
Official American Thoracic Society/European Respiratory Society Statement: researcher partnership systematic review. Implement Sci 2011; 6: 88.
asthma control and exacerbations: standardizing endpoints for clinical asthma 41 Souza NM, Sebaldt RJ, Mackay JA, Prorok JC, Weise-Kelly L, Navarro T et al., for the
trials and clinical practice. Am J Respir Crit Care Med 2009; 180: 59–99. CCDSS SystematicReviewTeam. Computerized clinical decision support systems
20 Bell LM, Grundmeier R, Localio R, Zorc J, Fiks AG, Zhang X et al. Electronic health for primary preventive care: a decision-maker-researcher partnership systematic
record-based decision support to improve asthma care: a cluster-randomized trial. review of effects on process of care and patient outcomes. Implement Sci 2011; 6:
Pediatrics 2010; 125: e770–e777. 87.
21 Eccles M, McColl E, Steen N, Rousseau N, Grimshaw J, Parkin D et al. Effect of 42 Ryan D, Price D, Musgrave SD, Malhotra S, Lee AJ, Ayansina D et al. Clinical and
computerised evidence based guidelines on management of asthma and angina cost effectiveness of mobile phone supported self monitoring of asthma: multi-
in adults in primary care: cluster randomised controlled trial. BMJ 2002; 325: 941. centre randomised controlled trial. BMJ 2012; 344: e1756.
22 Fiks AG, Hunter KF, Localio AR, Grundmeier RW, Bryant-Stephens T, Luberti AA 43 Roshanov PS, Fernandes N, Wilczynski JM, Hemens BJ, You JJ, Handler SM et al.
et al. Impact of electronic health record-based alerts on influenza vaccination for Features of effective computerised clinical decision support systems: meta-
children with asthma. Pediatrics 2009; 124: 159–169. regression of 162 randomised trials. BMJ 2013; 346: f657.
23 Kuilboer MM, van Wijk MA, Mosseveld M, van der Does E, de Jongste JC, 44 Hayward J, Thomson F, Milne H, Buckingham S, Sheikh A, Fernando B et al. 'Too
Overbeek SE et al. Computed critiquing integrated into daily clinical practice much, too late': mixed methods multi-channel video recording study of compu-
affects physicians' behavior—a randomized clinical trial with AsthmaCritic. terized decision support systems and GP prescribing. J Am Med Inform Assoc 2013;
Methods Inf Me 2006; 45: 447–454. 7: 7.
24 Martens JD, van der Weijden T, Severens JL, de Clercq PA, de Bruijn DP, Kester AD 45 Kawamoto K, Houlihan CA, Balas EA, Lobach DF. Improving clinical practice using
et al. The effect of computer reminders on GPs' prescribing behaviour: a cluster- clinical decision support systems: a systematic review of trials to identify features
randomised trial. Int J Med Inform 2007; 76: S403–S416. critical to success. BMJ 2005; 330: 765–768.
25 McCowan C, Neville RG, Ricketts IW, Warner FC, Hoskins G, Thomas GE. Lessons 46 Garg AX, Adhikari NKJ, McDonald H, Rosas-Arellano MP, Devereaux PJ, Beyene J
from a randomized controlled trial designed to evaluate computer decision et al. Effects of computerized clinical decision support systems on practitioner
support software to improve the management of asthma. Inform Health Social performance and patient outcomes: a systematic review. JAMA 2005; 293:
Care 2001; 26: 191–201. 1223–1238.
26 Plaza V, Cobos A, Ignacio-Garcia JM, Molina J, Bergoñón S, García-Alonso F et al. 47 Berlin A, Sorani M, Sim I. A taxonomic description of computer-based clinical
Cost-effectiveness of an intervention based on the Global INitiative for Asthma decision support systems. J Biomed Inform 2006; 39: 656–667.
(GINA) recommendations using a computerized clinical decision support system:
a physicians randomized trial [Spanish]. Med Clin 2005; 124: 201–206.
27 Tierney WM, Overhage JM, Murray MD, Harris LE, Zhou XH, Eckert GJ et al. Can This work is licensed under a Creative Commons Attribution-
computer-generated evidence-based care suggestions enhance evidence-based NonCommercial-NoDerivatives 4.0 International License. The images
management of asthma and chronic obstructive pulmonary disease? A rando- or other third party material in this article are included in the article’s Creative Commons
mized controlled trial. Health Services Res 2005; 40: 477–497. license, unless indicated otherwise in the credit line; if the material is not included under
28 Martens JD, van der Aa A, Panis B, van der Weijden T, Winkens RA, Severens JL. the Creative Commons license, users will need to obtain permission from the license
Design and evaluation of a computer reminder system to improve prescribing holder to reproduce the material. To view a copy of this license, visit http://
behaviour of GPs. Stud Health Technol Inform 2006; 124: 617–623. creativecommons.org/licenses/by-nc-nd/4.0/

Supplemental Information accompanies this paper on the npj Primary Care Respiratory Medicine website (http://www.nature.com/npjpcrm)

npj Primary Care Respiratory Medicine (2014) 14005 © 2014 Primary Care Respiratory Society UK/Macmillan Publishers Limited