APPENDIX
Independent Midwives Association Database
Project
NHS maternity statistics are collected in hospital
by means of computer programmes recording each woman's episode
of maternity care. Typically, the same programme will be used
for collecting both hospital and home-based statistics. Such programmes
are designed to record details of high-intervention hospital deliveries.
They are not designed to pick out the differences between home
and hospital care, nor to highlight those elements of home-based
care which may be having a positive impact upon the outcomes.
There has been until now in the UK no system of recording low
intervention home-based midwifery practice in its own right. With
the aim of addressing this absence of data the Independent Midwives
Association (IMA) set up a database in 2002 to collect practice
details and outcomes from all its members.
Women who book with independent midwives (IMs)
are typically searching for low intervention personalised maternity
care. IM caseload practice offers care on a one-to-one basis,
focusing on establishing the woman's unique needs for care in
this particular pregnancy, and on addressing and supporting these.
Independent midwives work as the lead health professional, offering
women a complete package of antenatal, intrapartum and postnatal
care. Most of the midwifery care is located in the woman's home,
including attendance at births. Throughout the perinatal period
such care includes giving the woman as much control of her own
situation as possible within the boundaries of safe professional
practice.
The Database
The IMA database has been designed to capture
a wide range of information about the mother, the baby and the
care given. There are several hundred fields in total, many of
which can accommodate multiple answers for the same case or baby.
The options (for example drugs used for pain relief) are numerically
coded to avoid spelling errors, and to assist in analysis. The
interpretation is displayed to the data-entry operator to permit
immediate checking. There is always an option for "other"
answers to be entered as text.
The data entry form (filled in by the midwives
themselves) contains over 280 questions, though not all are applicable
to every case. Nearly 40 of them can have multiple answers (for
example several methods of resuscitation might be used for a particular
baby, and all are recorded).
The database has a system of passwords to protect
confidentiality, and to prevent entered data being erased in error.
Each record is uniquely identifiable, and has a checking system
to ensure that duplication does not accidentally occur. The project
is registered under Data Protection legislation.
The original design of the dataform took into
account the need for comparability with UK national maternity
statistics. Documents studied for this purpose include:
HES Data Manual, Version 3.1, November 1996.
The International Classification of Diseases
and Related Health Problems (ICD-10).
DoH Statistical Bulletin: NHS Maternity Statistics,
England: 1989-90 to 1994-95.
DoH Statistical Bulletin: NHS Maternity Statistics,
England: 1995-96 to 1997-98.
National Statistics Primary Set of Outputs and
Questions—Demographic information, Household Composition.
National Statistics Primary Set of Outputs and
Questions—occupation, employment status and socio-economic
classification.
A Steering Group was set up towards the end
of 2002 to assist with the furtherance of the IMA Database Project.
The group includes Professor Alison MacFarlane, Lead Researcher,
Department of Nursing and Midwifery, City University, Mr Lesz
Lancucki, Department of Health, Ms Mary Newburn, Head of Policy
Research, National Childbirth Trust.
The Data
In addition to the basic data required for the
HES dataset, the project is collecting a mass of wide-ranging
information on every aspect of midwifery care by independent midwives.
It is intended that the data collected will provide a basis for
study and analysis of a number of issues to do with low intervention
midwifery care.
The project was set up as a prospective data
collection. This has been achieved by asking each IM to keep a
list of all clients from date of booking, giving basic information
such as date of booking and estimated date of baby's birth. Each
participating IM returns a copy of this list to the project coordinator
three times a year. Details of clients who book late, and who
will have birthed before the next list copy is due, are notified
by e-mail, post or telephone, and are added to the co-ordinator's
list copy. In this way a complete list of each IM's clients is
built up during the year, and missing dataforms can easily be
detected and followed up.
In designing research databases it is normal
to aim at the minimum dataset that will satisfy the research requirements,
as these are expected to develop in ways initially unforeseen.
The IMA database tends to the other end of the spectrum, with
a wide range of data collected from the beginning. This is because
the number of independent midwives is limited, so that data accumulates
fairly slowly with time. It is expensive and poor research practice
to go back and try to resurrect additional details of reported
cases. The only other option is to gather a well-considered but
large dataset in the first place.
A result of this philosophy is that many detailed
studies will be possible on the existing dataset.
June 2003
| 
