APPENDIX 22
Joint memorandum by Andrea Jones, Dr Angie
Hart, Dr Flis Henwood, and Chloe Gerhardt (MS 28)
1. INTRODUCTION
This memorandum describes the relevant findings
from a two-year research project funded by the Department of Health
(DoH) under its initiative on Information and Communication Technologies.
The purpose of this research was to explore the professional and
public acceptability of Electronic Patient Records (EPRs) however
some of the findings have a bearing on the problem of poor national
data collection in maternity.
The research was carried out in two stages.
The first was a national survey of Heads of Midwifery in England
(in 2001) which had a response rate of 74% (146 maternity units).
The second stage was in-depth case studies in four maternity services,
qualitative and ethnographic in approach, involving interviews,
focus groups, observations (with 340 individuals, mainly from
midwifery, medical staff, IT and administrative staff, and clients)
and analysis of documentation.
A full report research report is due for submission
to DoH in March 2003. Published articles based on this research
are listed at the end.
2. RELEVANT FACTS
ABOUT HOW
MATERNITY SERVICES
CURRENTLY KEEP
RECORDS AND
COLLECT DATA
2.1 The use of paper records
— Paper records are considered by practitioners
to be more important then computerised records in the vast majority
of services (97%).
— The main records are the paper Client
Held Record (CHR) for ante-natal and post-natal care, and the
paper Hospital record for labour.
— There is considerable duplication
of data gathering in the majority of services (88%) both between
paper systems and with computerised systems and the burden of
this duplication falls mainly on midwives.
2.2 Computerised maternity information systems
(MISs) and their use
— Maternity Information Systems are
used in 65% of services in England.
— Four suppliers provide 58% of these
MISs (Euroking, Protos, Stork, Ciconia/SMMIS).
— Their functionality, age, cost and
usability vary enormously nationally.
— The vast majority of data entry is
carried out by Midwives.
— Most of their training is conducted
by other Midwives, called "cascade" training.
— Two thirds (62%) of these computerised
MISs are "interfaced" in some way with other hospital
information systems—meaning data can be transferred from
one system to another without being re-typed in. The majority
of these (80%) were interfaced with the hospital Patient Administration
System (PAS) or Hospital Information System (HIS), meaning that,
in theory, demographic data should not need re-typing in to both
hospital management reporting systems AND the departmental maternity
system, although the case studies revealed that this was not always
the case.
— Only 13% of these MISs were interfaced
in any way with other information systems outside the hospital
(for example to GP surgeries).
2.3 Maternity Information Systems (MISs) and
their role in data collection and reporting
— Services reported as many advantages
as disadvantages of having a MIS.
— Producing statistics for audit and
evaluation of service was given by Heads of Midwifery as the main
reason for having a MIS. However, difficulty getting data in and
out was also reported as the main disadvantage of having one,
alongside finding them time-consuming to use (mainly for midwives).
— There appears to be no one off-the-shelf
MIS which produces the best data reporting: suppliers have tended
to tailor their systems to individual service requirements based
on cost, and different services have different age systems, so
some services have better reporting facilities than others.
2.4 Maternity Staff Dedicated to Data Collection
— Responsibility for data reporting
lies with maternity managers.
— Few maternity managers have the IT
skills required to extract data from computerised MISs and this
task is usually delegated, to an IT specialist midwife or to staff
from the IT department.
— Only one third of services have an
IT specialist midwife.
— Of those that do, their grade and
hours vary from E grade and a few hours a week to G grade, full
time.
— Many of these staff combine other
(mainly clinical and training) responsibilities with their information
quality and information reporting responsibilities.
— Where the post is full time and the
midwives employed are more senior, the computerised systems function
considerably more effectively to produce meaningful clinical data.
2.5 Conformance to national standards—the
National Maternity Record (NMR) and the Maternity Care Data Dictionary
(MCDD)
Two national initiatives in maternity have attempted
to help standardise data gathering in services. The NMR has been
available to services since 1999 and was set up to design a paper
antenatal record initially which would be comprehensive, useable
by all HCP and encourage good record keeping and communication
practice . . . (and) optimise features that contribute to effective
clinical care during pregnancy and childbirth, and that can be
supported by evidence or expert opinion' (National Maternity Record
Project 1999). The initiative has been backed by the RCOG and
the RCM.
— Only 27% (n=39) of services reported
using the NMR.
— The two main reasons respondents
gave for using it were: the quality of the notes (better/easier)
(17/39); wanting to conform to national standard and to help the
process of women transferring between units (14/39).
— The most common reasons for not using
it were: it was not accepted and/or liked and/or found to be difficult
to use by health care professionals (HCP) (34/107); it was not
complete (26/107); clinicians prefered their own or regional notes
and/or don't reflect local standards/priorities (21/107).
The MCCD was a national attempt to standardise
maternity terms which has been running since November 1999. The
original intention was:
"To determine the information
that needs to be recorded to support the planning and delivery
of maternity care for both the mother and child, and wherever
possible, to standardise this data. This will enable standardised
and consistent recording of data relating to maternity and childbirth,
for women and infants, within EPR systems in all affected NHS
organisations." (NHSIA 1999)
— Half of the respondents in our survey
had not heard of the MCDD (71=49%).
— Of those who had, only 2 services
had made use of it.
3. KEY FACTORS
IN IMPROVING
NATIONAL DATA
RECORDING
The following points are main research findings
followed by our recommendations.
3.1 One third of maternity services are
still using entirely manual systems to record and report data.
RECOMMENDATION: resources should be made available
in those services without a MIS to ensure they have appropriate
IT systems for gathering data and reporting.
3.2 There is a lack of standard reports
from Maternity Information Systems (MISs). Suppliers tend to customise
systems for each service, largely dictated by the services and
related to trying to limit the cost of the system, and suppliers
have no remit to promote national data standards. The Maternity
Care Data Dictionary experience shows that the process of agreeing
clinical standard terms is not technical in nature, rather a difficult,
contested social process which requires careful management.
RECOMMENDATION: the work of agreeing standard
clinical terms within maternity should be funded nationally and
the project management of it should be linked to the evolving
maternity NSF; the same people involved with this initiative should
also work with suppliers and representatives of maternity staff
within the NHS to specify standard maternity data reports, which
services can then require MIS suppliers to provide when they procure
systems.
3.3 There is a lack of interfacing between
Trust-wide management reporting systems (which produce the NHS
Trusts statistics) and computerised maternity information systems.
This results in an unnecessary burden of duplication of data entry
falling on midwives and takes time from clinical care. It also
results in data becoming "trapped" in one system (the
MIS) and unable to be shared with either locally or nationally.
RECOMMENDATION: the NHSIA efforts to ensure
that information systems bought by the NHS can share data should
be made a national priority; maternity services should be required
to ensure this "interoperability" when they purchase
systems; NHS Trust IT and maternity services should prioritise
resolving outstanding interfacing problems, some of which are
not technical or funding in cause.
3.4 There is poor national sharing of experiences
of MISs within the NHS. Services procuring new systems spend insufficient
time with other maternity services learning about their experiences
and, therefore, poor elements of system design or use are often
repeated from one service to another.
RECOMMENDATION: maternity service managers and
IT Midwives should be made more aware of where they can get help
with procurement of IT systems; the new NHSIA system of accreditation
of suppliers (outlined in `Delivering 21st Century IT Support
for the NHS') should include MIS suppliers; services should budget
appropriately for information sharing across NHS Trusts during
system procurement; and services should ensure continuity of staff
taking part in inter-Trust learning during system procurements.
3.5 Procurements of MISs are insufficiently
focussed on how systems report rather than how data is enteredand
limitations and problems are often not realised until after the
systems has been procured, or even installed.
RECOMMENDATION: services need more support from
qualified staff skilled in both IT and clinical practice during
the procurement process; support should be available from organisations
and individuals that have more of a national picture and specialise
in procurement, rather than maternity services having to rely
on their own IT staff alone, so that there is some consistency
in checking that MISs report appropriately to meet national data
requirements.
3.6 Many maternity managers treat service
level information as a relatively low priority, and national data
even more so. This is for a range of reasons including the relatively
low priority given to it by the Senior Management within their
NHS Trusts.
RECOMMENDATION: maternity information should
be made a higher priority for maternity managers.
3.7 Managers cite supplier costs for improving
and upgrading systems amongst the reasons for retaining poor performing
systems and systems which do not reflect current, national data
collection needs. Many of these costs are relatively small amounts
of money (for example £5,000 annually) but are not built
into to supplier contracts at the time of procurement of systems.
Poor system performance, in turn, has a demotivating effect on
midwifery staff entering data and can lead to a spiral of poor
output and poor data entry.
RECOMMENDATION: service managers need to recognise
that on-going, annual costs are required to maintain and maximise
use of MISs, not just the one-off cost of buying the system; funding
for maintaining and upgrading IT systems has to be ring-fenced
to avoid being absorbed by general managerial pressures, especially
if national data collection requirements are to be kept pace with;
representatives of the midwifery profession (eg Royal College
of Midwives) should do more to urge midwives to be critical and
challenge midwifery managers where they are entering data into
system for which they do not see, or have not been demonstrated,
any benefits to professional practice or patient care.
3.8 There is a lack of knowledge and understanding
of the importance of national data amongst frontline maternity
clinical staff, especially midwives, who are the principal data
entry staff.
RECOMMENDATION: Midwifery education should include
greater emphasis on national data requirements and the role of
information in improving standards of care.
3.9 Lack of time for even mandatory clinical
training for midwives in some services means MIS training, and
particularly an understanding how the clinical information they
input gets used, takes a very low priority.
RECOMMENDATION: issues of midwifery recruitment
and workload pressures notwithstanding, services with MIS should
improve the quality of IT training and midwives understanding
and use of their own, existing systems (for example, midwives
should be able to access data about their own practice for audit
purposes).
3.10 There is a conflict of priorities for
frontline midwives between the paper record, which they see as
the most important record (legally, professionally and for the
client) and computerised records, of which they do not see the
benefits in many services. This is particularly so in maternity
where the paper Client Held Record plays an important role in
informing and empowering their clients—something which maternity
policy has as its' heart (Changing Childbirth 1993). This is a
key aspiration of the government for a use-focused NHS (Shifting
the Balance of Power 2001)
RECOMMENDATION: ensure that IT policy acknowledges
the vital role of paper records, both now and for the foreseeable
future; services should do more to share the output of MISs with
their midwifery staff (for example, midwives should be more able
to access data about their own practice for audit purposes); IT
policy should acknowledge that the necessity of retaining both
paper and computerised records means that increased use of IT
does not save time for midwives (even if it does for other maternity
staff such as managers or doctors); there should be more communication
of the legal status of paper V computerised records to frontline
clinical staff.
3.11 There is a lack of support for the
hybrid clinical/IT staff in services, who play a crucial role
in the quality of maternity data. Where these hybrid roles exist
they are not valued as highly as other, solely clinical, specialisms.
RECOMMENDATION: NHSIA should work with UKCC
and Royal College of Midwives to improve the status and career
structure for such roles.
3.12 The hybrid clinical and IT posts lack
continuity. Posts are often created for procurement and installation
of systems but few are permanent and this jeopardises the longer-term
project of sustaining meaningful information collection.
RECOMMENDATION: services will need to fund these
roles more continuously if information is to become more of a
priority issue.
4. THE IMPACT
OF CURRENT
POLICIES ON
ELECTRONIC PATIENT
RECORDS (EPRS)
ON MATERNITY
DATA COLLECTION
In relation to current policies to increase
the use of Electronic Patient Records (Information for Health
1998, Building the Information Core 2001) our research suggests
that IT professionals continue to dominate decision-making about
the procurement of the hospital-wide EPR systems, within NHS Trusts.
Such "top down" approaches mean it is unlikely that
clinical data will be any more of a priority than it has been
in the past within NHS Trust level decision-making. It is also
the case that there is little understanding within the IT professionals
in the NHS of the governments imperatives about use involvement,
despite the policy claims that future IT should be patient-centred
(Building the Information Core 2001). Our case study findings
suggest that there is a risk that EPR strategy and implementation
is failing to acknowledge that is has to develop alongside user-centred
systems like the Client Held Record, rather than seek to replace
them. If IT policy and practitioners do not combine the interests
of national data collection and client-centred record keeping,
there will continue to be a conflict of priorities for midwives
at the frontline, in which data collection will obviously take
second place.
Our case study work also suggests that individual
NHS Trusts have been very much pursuing their own solutions with
different EPR suppliers, although this looks set to change with
the NHSIA stepping in to ensure standard specifications for EPR
suppliers, in 2002 (Delivering 21st Century IT Support for the
NHS, July 2002). These specifications—especially those relating
to the ability of new systems to share and report clinical data—have
important implications for the medium to long term future of data
collection in all services, including maternity.
In the EPR "hi-flyer" case study we
undertook the experience was that the Trust-wide managerial information
needs and the clinical functionality offered by such new systems
(such as reporting blood test results and prescribing online)—whilst
the latter were welcomed and used by clinical staff—dominated
at the expense of the departmental, maternity information needs.
The consequences were that a service which had previously experienced
very satisfactory region-wide maternity data ended being unable
to produce even data to serve their own service needs, let alone
national data returns.
RECOMMENDATION: key figures within Department
of Health responsible for national maternity data collection should
be included in EPR policy forums, to ensure that EPR initiatives
do not fragment the information base for maternity even further.
PUBLICATIONS BASED
ON EARLY
STAGES OF
THIS RESEARCH
Hart, A., Henwood, F. and Jones, A. "Views
of Heads of Midwifery on electronic patient records". British
Journal of Midwifery Volume 11, Number 1, January 2003, 53-57.
Jones, A., Henwood, F. and Hart, A. "Electronic
patient records: the view from maternity." British Journal
of Midwifery Volume 10, Number 10, October 2002, 635-639.
Jones, A., Henwood, F. and Hart, A. "EPRs
and maternity services—the challenge of client held records
and the blurring of boundaries" in Healthcare Computing 2002,
Guildford: British Computer Society, 2002.
February 2003
| 
