Journal of Health Organization and Management

The creation of risk-related information: The UK General Medical Council's electronic

database
Sally Lloyd-Bostock

Article information:

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

To cite this document:

Sally Lloyd-Bostock, (2010),"The creation of risk-related information", Journal of Health Organization and
Management, Vol. 24 Iss 6 pp. 584 - 596
Permanent link to this document:
http://dx.doi.org/10.1108/14777261011088674
Downloaded on: 18 May 2015, At: 08:57 (PT)
References: this document contains references to 31 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 397 times since 2010*
Access to this document was granted through an Emerald subscription provided by 434496 []

For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.

About Emerald www.emeraldinsight.com

Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.

The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1477-7266.htm

JHOM
24,6

584

The creation of risk-related

information
The UK General Medical Councils electronic
database
Sally Lloyd-Bostock

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

ESRC Centre for Analysis of Risk and Regulation,

London School of Economics, London, UK
Abstract
Purpose This paper aims to clarify the potential to use data on doctors and fitness to practise (FTP)
cases held by the UK General Medical Council (GMC) for wider regulatory purposes, such as
identifying risk factors. The paper aims to concentrate on how data are shaped by the GMCs functions
and organisational concerns, and by the configuration and use of their electronic database.
Design/methodology/approach The GMC provided samples of their data, access to
documentation surrounding the configuration and use of the database, and meetings with staff able
to provide background on the database, GMC procedures, and the GMC as an organisation.
Findings The FTP database is designed to process cases within complex legal rules, and to provide
for accountability. The database and its use are adapted to these purposes. Attempts to use it for other
purposes are likely to find it difficult to use, the scope and quality of data uneven and some codes
unsuitable. The register data are very narrow in scope. While combining register and FTP data to
identify risk factors is by itself of limited value, the database can contribute to closer study of risks to
patient safety from poorly performing doctors.
Research limitations/implications The research was exploratory. It provides initial insights
and the basis for further research.
Practical implications The data have potential policy use for the GMC, but it is essential to
understand the limitations.
Originality/value The paper examines previously unanalysed influences on the GMCs data. It
also develops new angles on questions in the regulation literature about organisational risks and the
creation of risk data.
Keywords Risk management, Doctors, Medical information systems, Patients,
Performance management
Paper type Research paper

The changing context of medical and healthcare regulation, combined with new
approaches to promoting patient safety has produced an intense search for reliable
data on risks and patient safety incidents (see, e.g. Attride-Stirling et al., 2006; Vincent,
2006; Waring, 2009). Information about risks is fundamental to the risk-based
approaches that have spread across regulatory fields in the UK in the past decade (see,
e.g. Hutter, 2005), including the regulation of healthcare and regulation of the medical
Journal of Health Organization and
Management
Vol. 24 No. 6, 2010
pp. 584-596
q Emerald Group Publishing Limited
1477-7266
DOI 10.1108/14777261011088674

The research was supported by ESRC Award Number RES-153-25-0087 within its Public
Services Programme, and based at the ESRC Centre for Analysis of Risk and Regulation (CARR)
at the London School of Economics. The author is grateful to anonymous reviewers for their
helpful comments.

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

profession by the UK General Medical Council (GMC) (Lloyd-Bostock and Hutter,

2008). The approach emphasises systematic identification and assessment of risks,
predicting potential harm rather than reacting after the event, and targeting regulation
and inspection accordingly. The Care Quality Commission for example is creating
Quality and Risk Profiles of NHS Trusts, assembling what is known about a provider
to assess where risks lie and to prompt front line regulatory activity such as inspection
(Care Quality Commission, 2009). The approach can only be as good as the available
data. Gathering and integrating relevant risk data is resource intensive and difficult
(Lloyd-Bostock and Hutter, 2008). Use of existing sources of data is highly attractive,
but depends on the datas quality and relevance, and the context in which it was
created (Attride-Stirling et al., 2006; Vincent, 2006).
This article reports on an exploratory project to clarify the nature and potential uses
of risk-related data created by the GMC. In fulfilling its statutory functions the GMC
records basic information about every doctor on the medical register for the UK, and
further information about cases where a doctors fitness to practise (FTP) has been
questioned. Doctors subject to FTP procedures are a sub-set of all registered doctors
currently approximately 232,000 (GMC, 2010). According to GMC statistics, in 2008,
5,195 doctors were the subject of complaints or referrals to the GMC; 204 FTP panel
hearings were held; and 42 doctors had their names erased from the register (GMC,
2009). In March 2006 the GMC went live with a new electronic database (Siebel) that
now contains both the register and the FTP data. A combination of data on all doctors
registered and on FTP cases would seem to offer an opportunity to identify groups of
doctors more likely to present risks to patients, and the nature of those risks. The
possibility also arises of using GMC data in the wider patient safety context to help
identify patterns and sources of risk to patients.
The main premise underlying the project is that information in a database is always
a function of the context in which it is created, and that this affects the uses it can
validly serve. In general terms, this has long been recognised. Garfinkels classic paper
(Garfinkel, 1967) demonstrated the close link between recorded data and its purposes,
and the resulting difficulties facing attempts to use it for other purposes. Increasingly,
research on the regulation and management of risks to patients is examining in detail
the processes of creating and coding information about risks, and the methodological
limitations on using existing sources for purposes of healthcare regulation.
Attride-Stirling et al. (2006), for example, survey the quality and relevance of eleven
data-gathering tools. Waring (2009) traces through detailed ethnographic study how
concerns with blaming and professional boundaries, management assumptions, and an
organisations systems of risk management, construct and reconstruct narratives of
safety fed into the NHSs National Reporting and Learning System (NRLS). In addition,
the regulation literature is increasingly recognising the role of blame games and risks
to the organisation itself as influences on conceptualisations of risk and on regulatory
bodies (e.g. Hood, 2002; Hutter and Power, 2005). These literatures indicate possible
influences on the GMCs data and its wider usability. However, little was known
outside the GMC about the nature of its databases and the operation of its new
procedures. Earlier work relating GMC data on registration and fitness to practise
(Allen, 2000) predated the introduction of completely new procedures in 2004 (The

The creation of
risk-related
information
585

JHOM
24,6

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

586

Figure 1.
Factors affecting the flow
of information about risks
into the GMCs database

General Medical Council (Fitness to Practise) Rules 2004 (Fitness to Practise Rules
2004)) and the modern Siebel database.
The scope of the article is defined with reference to potential sources of influence, as
set out in Figure 1. The lower boxes represent an idealised flow from risk to risk
information. The upper row breaks down influences on the nature and flow of
information into Siebel. Each can influence both whether information is generated and
passed on (the first arrow); and if it is, the form it takes (the second arrow).
The main sources of FTP data are complaints from members of the public, and
referrals from NHS Trusts and other organisations including the police (GMC, 2009).
The literature on patient safety, incident reporting and complaints indicates numerous
likely influences at this stage (e.g. Lloyd-Bostock, 1999; Vincent, 2006; Walshe and
Boaden, 2006; Waring, 2009). The focus in this article, however, is the second and third
categories: the legal and organisational context (GMC factors); and the configuration
and use of the database (database factors). Within these broad areas, the focus is on
implications for wider use of the data. Interactions between the database and case
processing, although an essential aspect of the organisational context, are outside the
present scope. Central strands framing the analysis are thus: the GMCs legal
framework and disciplinary functions; risks to the GMC itself; and effects of the
configuration and use of the database:
.
Legal framework and functions. The GMC is a professional regulatory body with
legal powers, duties and procedural rules derived from the Medical Act 1983 (as
amended). The tight focus of its concerns is the fitness of individual doctors to be
registered and licensed in the UK and their continued fitness to practise. Where its
data do identify risks to patients, it will be within a particular framework that is
unlikely, for example, to be commensurate with systems- and human
factors-based approaches (e.g. DoH, 2000; Rasmussen, 1997; Reason, 1990).

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

Risks to the GMC. In the GMCs case, sources of organisational risk include its
high public profile and multiple stakeholders. Its legal remit is continually
changing and sometimes fuzzily defined, and it must work in close co-operation
with a network of organisations concerned with patient safety, medical
education and the medical profession (Lloyd-Bostock, 2009). Growing complexity
and continual change create room for ambiguities about responsibilities and for
failures in communication (Vaughan, 1999; West, 2006). The GMC depends on
other bodies to provide reliable information about employment environments,
doctors performance and conduct, and appropriate educational standards and
qualifications. Furthermore, as a professional body the GMC is especially
vulnerable to criticism for perceived regulatory failure (Lloyd-Bostock and
Hutter, 2008). Managing these various risks forms an integral part of the
organisational context (Hutter and Power, 2005).
Database configuration and use. As electronic databases proliferate, social
science research is accumulating on database configuration and use, and how
these affect the usefulness of a database within as well as beyond an
organisation (e.g. Agrawal, 2008; Bowker and Leigh Star, 1999). Research
includes ethnographic work on the setting up and use of modern databases
(Agrawal, 2008; Leslie, 2009). Once in place a database takes on a life of its own,
structuring and filtering information, and acting as the scaffolding of
knowledge (Bowker and Leigh Star, 1999). Data may be filtered by the wish to
present a particular version on official records or to achieve a particular outcome
(e.g. Agrawal and Silbey, 2008; Bossen, 2006; Leslie, 2009; Waring, 2009).

It was not known how these broadly defined influences might play out in practice in
the GMC context, nor what research data might be available and relevant. The research
was exploratory in methods and scope.
Methods
A confidentiality agreement was agreed with the GMC at the start of the project
covering information made available by the GMC. The project received approval from
the Research Ethics Committee at the London School of Economics.
Information was gathered from a combination of sources. The GMC provided
extensive assistance and support. The first stage was familiarisation with the GMC as
an organisation, its FTP and registration procedures, and its systems for collecting and
coding data. This included study of GMC guidance and internal documents; 13 hours of
interviews with GMC staff at policy, senior management and case-processing levels
within the Fitness to Practice and Registration Directorates, about case- and
information-processing systems and the history and purposes of the Siebel database;
and demonstration of the database by the information manager. Discussions were also
held with the Head of Analysis and Feedback at the NPSA (one hour), and the Head of
Information and Knowledge Management at NCAS (two hours).
The second stage was exploration of the database itself. Siebel incorporates an
Analytics function: selected coded data are saved daily in a data warehouse and
can be downloaded for analysis. The GMC provided anonymous Analytics data on all
complaints and referrals received by the Fitness to Practise Directorate 1/5/2006

The creation of
risk-related
information
587

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

JHOM
24,6

30/4/2008, and any resulting FTP cases. The GMC also provided data on all registered
doctors (excluding identifying information), derived from the register as it stood in
March 2009. In addition, arrangements were made for exploration of the database in
the GMCs offices.

588

Findings and interpretation

Siebel and its use within the GMC
The GMC uses a comprehensive Siebel customer relationship management (CRM)
application. The GMCs customised system and database are known within the GMC
as Siebel and the term is similarly used in this paper to refer to the GMCs system, not
to the application. Modifications to Siebel are continually being introduced within the
GMC, sometimes because of changes to legislative requirements governing the GMC.
The findings reported here relate to the time the study was conducted. Within the
Fitness to Practise Directorate Siebels primary function is the day-to-day processing
and management of cases, and handling and recording interactions with outside
bodies, replacing paper files. Documents are scanned in, decisions and case processing
information are coded, and some fields allow for free entry, e.g. of details of allegations.
Internal electronic communications are made in or accessible through Siebel and
appropriate parts of the database can be accessed from virtually every desk within in
the GMC.
The study was concerned with the registration and fitness to practise categories
of data. Findings from the interviews and study of the database are first described and
discussed separately for each category, with reference to the potential influences
identified above. The findings are then applied to consideration of the potential use of
the data to assess risks through combining the two categories.
Fitness to practise data: the legal context. The FTP database is based directly on the
GMCs FTP procedures, which embody the GMCs guidance on what is required of
doctors (GMC, 2006b). The procedures are governed by detailed rules with legal force
(the Fitness to Practise Rules, 2004). The Rules themselves, GMC terminology and FTP
procedures are explained in publicly available GMC documents (e.g. at: www.gmc-uk.
org/concerns/doctors_under_investigation/a_guide_for_referred_doctors.asp) and
Fitness to Practise Annual Statistics (GMC, 2009). In outline, complaints and
referrals to the GMC (enquiries) are triaged and either closed immediately, promoted
to Stream 1, or promoted to Stream 2. Stream 1 is for cases where the enquiry raises
concerns that could have implications for a doctors registration. These will be
investigated and a further decision then taken whether to refer the case to a fitness to
practise hearing. This may result in action on the doctors registration, including
erasure. An interim order panel may suspend the doctor pending the outcome of
investigation, and a range of other less common decisions and routes are possible
within the rules. Stream 2 is for enquiries where the allegations raised do not in
themselves have implications for the doctors registration, but which might require
GMC investigation if they form part of a pattern. In these cases the GMC contacts
employers or contractors to ask whether there are wider concerns. Once it has that
information it may close the case or promote it to Stream 1.
The process is designed to identify, direct and develop cases that might lead to
action on a doctors registration. Demonstrable compliance with legal requirements

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

and procedures is essential to the organisation. Siebel not only facilitates case
processing; it provides a record that procedures have been followed. Procedures are
framed from the outset by the possibility of an eventual FTP hearing: failure to follow
proper procedures can lead a case to fail. If an enquiry is promoted to Stream 1 and a
case is created, an outline investigation plan is set up within Siebel and seen by a
lawyer. As in many spheres, those processing cases frequently use language referring
to the legal rules governing what they do. For example, in internal communications,
staff might refer to a Rule 7 letter, referring to Rule 7 of the Fitness to Practise Rules
2004, applicable at a certain stage in the procedures at which a letter must be written to
a doctor with specified information including the allegations being made against
him/her. Some of Siebels pre-defined decision codes are expressed as the legal rule or
section that has been applied. For example, one of the most frequently used closure
reasons available for enquiries is Five Year Rule. The GMC cannot take action more
than five years after alleged events, unless exceptional circumstances can be shown.
Another is is R28 closure referring to Rule 28 of the Fitness to Practise Rules 2004
allowing for closure of a case after it has been referred to a panel hearing.
The legal framework governing the work of the GMC results in a complex and
specialised database configuration. Information is necessarily selected, lost and created
in the process of imposing codes on a complex world. Siebel not only copes with a
complex world, but also with complex procedures. The path from enquiry to hearing
involves several nested sub-paths. The focus of GMC decisions is variously the
enquiry, the case, the investigation, the doctor, allegations against a doctor, or
hearings. The relationships amongst these can be convoluted. One enquiry can give
rise to several triage decisions and thence to several cases. In the two-year data set, 28
per cent of enquiries involved triage decisions for two or more doctors; 8,397 enquiries
gave rise to 10,230 triage decisions; 101 gave rise to five or more. The circumstances
behind multiple triage decisions vary widely, and the complaint/referral may or may
not specifically refer to the doctors triaged. For example, one enquiry came from an
organisation processing prescriptions, raising questions about prescribing by a
number of named doctors. The enquiry was triaged in respect of each doctor. A
member of the public complaining about a relatives hospital care listed five doctors by
name, leading to triage in respect of each. In other cases doctors are not named but the
complaint relates to an episode of treatment involving several doctors, whom the
investigation team try to identify for triage. One NHS referral raised concerns about the
way a particular condition was being diagnosed and treated within a hospital. The
enquiry was triaged in respect of several doctors who were identified by the GMC team
as involved in treating the condition. Furthermore, if information arises during
investigation about matters that were not part of the original enquiry, a new enquiry
may be set up in Siebel to ensure that information is not shared inappropriately with
the source of the original enquiry.
The relationship between cases and doctors can also be complicated. Each case
concerns a single doctor, but one doctor may have several cases associated with
him/her, which may or may not arise from the same enquiry. Of the doctors concerned
in cases arising from enquiries in the two-year dataset, 7 per cent had more than one
case. Each case can involve several allegations. An investigation outcome is coded for
each allegation, which may be refer to hearing. The decision to refer a doctor to a

The creation of
risk-related
information
589

JHOM
24,6

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

590

hearing draws together cases and allegations. Allegations taken forward to a hearing
will be adjudicated separately and any action on registration will be related to proven
allegations. Furthermore, the database must accommodate departures from neat
progress. Cases can stop and start and sequences may be halted, sidetracked or looped
for a variety of reasons. Siebel captures these complicated intersections for practical
purposes through a configuration that takes forward decisions and investigations,
controlling the options available, requiring certain information to be completed, and
providing internal links to related processes and data. Progress of current cases can be
monitored and aggregated for management purposes. The system is not, however,
adapted to facilitate the production of summary statistics or data for wider purposes
such as risk assessment or patient safety research.
Siebel coding was often understandable only with reference to the GMCs
decision-making processes and legal functions. A particular example is the coding of
enquiries as a complaint or a referral. An investigation manager responsible for
triage explained that the choice of code relates to the need to seek complainants
consent to inform the doctors concerned. It is assumed that referrals (most often from
a Trust or the police) imply consent. On the other hand, codes requiring distinctions
that did not affect the channelling of a case could be used inconsistently, and
information concerning less serious or closed cases remains relatively thin and
undifferentiated.
Where identification of risks is concerned, the coding of allegations is crucial. The
allegation codes used in Siebel are designed, not to capture what is alleged, but rather
to define a potential case within the GMCs powers. The codes map directly onto the
GMCs Good Medical Practice, which sets out the standards required of a doctor
(GMC, 2006a). Allegations are coded at three levels. Allegation Category corresponds
to the seven general headings used in Good Medical Practice: clinical care; probity;
relationships with patients; working with colleagues; health, maintaining good medical
practice; and teaching/supervision, with an additional category compliance with a
GMC investigation, used when a doctor has not complied with a request such as to
undergo assessment as part of an FTP investigation. Allegation types correspond to
aspects of medical practice mentioned within each heading, and sub-types to
examples of those aspects. The category other is deliberately avoided, giving rise to
codes for allegation types and sub-types that are rarely used. The amount of detail
coded depends on how far the case progresses through FTP procedures. Allegations
are revised, added and amplified as investigation progresses towards stages requiring
specified allegations, notably a case examiners decision or a hearing. The resulting
hierarchy of codes, and the bunching of allegations within certain codes, limit the
usefulness of the allegation data in GMC format for purposes beyond progressing GMC
cases and monitoring by the GMC of its own activities. Other bodies that have
developed allegation codes in related areas (NCAS and the NRLS) said they found the
GMCs codes inappropriate for their purposes, and developed their own. Neither saw
value in attempting to integrate databases compiled for different purposes, though
there was value in sharing information, both in aggregate form and about particular
cases indicating new risks.
The new procedures implemented in 2004 cast a wide net, perhaps partly in reaction
to the Shipman case. Dame Janet Smith (2004) criticised the GMC for sending too many

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

complaints back for local resolution, and placing the onus on the complainant to come
back to the GMC. The triage process requires a computerised script to be completed
that records reasons for the decision from a pre-defined menu. There was some
suggestion in interviews within the GMC that it may have overdone the promotion of
matters that will never become cases of interest to the GMC. Around half of Stream 1
cases are closed after investigation (though the proportion is falling), and very few
Stream 2 cases are promoted to Stream 1 (GMC, 2009). In addition to handling the
information the procedures generate, Siebel serves to manage organisational risks by
enabling the GMC to demonstrate accountability. Stream 2 provides a route for cases
the GMC might be criticised for closing, enabling the GMC to pass responsibility for
closure to the doctors employers. Ambiguity over which risks various bodies ought to
regulate creates possibilities for blame transference and blame avoidance, increasing
the need to maintain records defensively. Siebel is configured to ensure that procedures
are tightly followed, and decisions (including decisions relying on other bodies) are
documented and justifiable.
Registration data. The registration data directly reflect the GMCs statutory
function of maintaining a register of doctors qualified and fit to practise in the UK
(Medical Act 1983 as amended). The GMC must establish certain facts in order to grant
registration, including details of qualifications and where they were obtained. The
information, and the identity of the doctor, is very carefully checked before registration
is granted. Beyond that little is recorded or recorded reliably on all doctors registered.
Date of first registration is recorded, but date of birth is not known with any accuracy.
Specialisation may be recorded either at the time of registration or when a doctor
notifies the GMC, but as doctors often move area of training the information is
unreliable. The register has a current address for the purposes of contacting the doctor,
but it does not indicate where the doctor is now employed nor indeed, whether s/he is
currently practising at all.
The scope and content of the register are constantly under review. With the
introduction of revalidation for all licensed doctors the GMC will hold fuller and more
up to date information on all doctors on its register. Some changes to the register are
already in place, for example the creation of the General Practitioner register. Ethnicity
data are now collected to enable the GMC to fulfil requirements to monitor for possible
discrimination in its activities. However, when asked about improving or extending
registration data, GMC policy staff said that resources cannot be devoted to collecting
and recording data without clear justification in terms of the GMCs functions and
governing legislation.
Scope for using GMC data. The nature of the register data clearly limits its value for
identifying risk factors. There is a danger that the information will loom larger than it
would if other information were also available (such as employment information); and
explanatory factors will be missing. Register data cannot be narrowed down to doctors
practising, let alone practising in a comparable field. Furthermore, Siebel is a living
database: data depend on the date they are extracted. Changes overwrite previous
data, limiting what can be analysed. Registration history can be traced for individual
doctors but it is not possible to reconstruct the register as it stood at a previous date in
order to make relevant comparisons.

The creation of
risk-related
information
591

JHOM
24,6

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

592

Identifying risk factors involves informed judgements about what risks should be
considered and how they are to be defined and measured. A starting point is the risk of
being the subject of a complaint or referral. The GMC presents some analyses on this
basis in its latest annual statistics, looking at the distribution of enquiries by the area
of the doctors primary medical qualification (GMC, 2009). However, enquiry data
include doctors who were not found to present any risk to patients. The further cases
progress towards a hearing case, the more concentrated the data become, but the
greater the influence of filters and the organisational context. The numbers rapidly
become small (2004 fitness to practise hearings were held in 2008 (GMC, 2009)), and
many risks to patients are likely to be filtered out in the process of specifying provable
allegations with implications for a doctors fitness to continue to practise.
The usability of a database depends on its technical usability as well as the content
and quality of data. The complex structure of the data and the continuous changes to a
living database create technical problems defining and relating data. Earlier codes,
such as allegation codes in FTP cases, are overwritten, obscuring the evolution of a
case. NCAS, faced with a similar situation established the practice of recording a
snapshot of its database each month so that it was possible to track changes to
allegations as a case proceeds. Analysis that requires tracing the flow of cases through
from enquiry to hearing entails attempting to match different parts of the database that
may not match perfectly. Samples filtered in to analyses using the Analytics function
can change each time the data warehouse is updated. As already outlined, the character
of Siebels configuration and codes make it difficult for an outsider to understand the
data or use it to produce statistics. The head of information and knowledge
management at NCAS similarly emphasised the need for caution in the use of NCAS
data, whether within or outside the organisation. Within the GMC, use of Siebel data
was described as a dark art, requiring specialist knowledge held by few GMC staff.
Conclusions
The clearest finding is the close fit of the GMCs database to its legal remit and the
demands of the legal framework within which it operates. These explicitly frame the
GMCs procedures, and through them the configuration of its registration and FTP
database. Also to emerge clearly are the effects of organisation-level concerns with
case- and data- management, which indeed determined the GMCs choice of a
comprehensive CRM application. These shape both the configuration and content of
the databases, and technical aspects of its usability. More tentatively, effects can be
discerned of organisational concerns with accountability, legitimacy and defensibility.
Here, it was suggested the database acts as a tool for managing organisational risk to
the GMC itself. These various effects combine to produce data that is very rich but
highly specialised for GMC purposes, of uneven quality, complex and technically
difficult to use. A combination of the register and FTP data cannot on its own reliably
tell us much about wider patterns of risks to patient safety from poorly performing
doctors, though the database does tell us about the GMC as an organisation, its
functions and priorities.
The study was exploratory and limited in scope, aiming to provide initial insights
and the basis for further and more detailed research. Study of Siebel illustrates how a
database structures and filters information through the categories and options made

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

available, which in turn are a function of the purposes of the database. As noted above,
where identification of risks to patients is concerned, the coding of allegations is
crucial. Waring (2009) shows how incident report forms shape knowledge about risks
to patient safety. White et al. (2008) describe a similar process in their study of the
impact of the Common Assessment Tool on professionals working with children,
writing of the descriptive tyranny of a tool which forces the writer to present
information within a certain structure and format. Much can be lost in the
transformation of story told by a patient into codes relating to individual doctors and
potentially provable allegations of departure from FTP standards. There are complex
interactions to be empirically disentangled here. For example, how do social and
cultural influences interact with formal legal and risk management frameworks and
tools, and with psychological processes of routine case processing?
In the regulation literature, organisational risks have been distinguished
conceptually from the societal risks that are the object of regulation such as harm
to patients. The distinction is unclear in practice in the GMC context. Concerns with
organisational and societal risk are not necessarily in conflict, nor even clearly
distinguishable. Risks to the GMCs reputation and powers might be thought of as
risks to the organisation rather than societal risks, and thus a distraction from its core
purposes. However, maintaining its reputation is also arguably part of the GMCs
remit. In addition to its broad duty to promote patient safety, the GMC has a duty to
maintain trust in the medical profession, as was made explicit in the Merrison Report
(Merrison, 1975) a duty that entails maintaining the publics trust and confidence in
the GMC itself. The risks associated with organisational complexity similarly have
several facets, and managing the risks is part of the GMCs functions. Thus, it has
explicitly confronted the need to integrate and confine its role within a network of
regulatory activities in a four layer model of medical regulation, which distinguishes
personal, team-based, and workplace from professional regulation where the
work of the GMC is concentrated (GMC, 2006b).
The study casts an unusual light on the effects of disasters on regulation. The
impact of the Shipman case is deeply embedded in GMC procedures, in GMC concerns
with accountability, and through these, in the database itself. The categories and
options available are partly determined by processes of selecting and negotiating codes
and dropdown lists when a database is designed, tending to crystallise concerns
relevant at the time (Agrawal, 2008). The 2004 Fitness to Practise Rules, which
underpin the FTP database, were devised in the aftermath of the Shipman case and
incorporate responses to concerns about its procedures raised by the subsequent
inquiry (see Smith, 2004, Ch. 25). The emphasis on recording legitimate reasons for
decisions and maintaining paper trails of interactions with other bodies may also owe
something to Shipman. Smith (2004) concluded that the GMC could not be held
responsible for Shipman continuing to practise when he was murdering patients, and
the value of the GMC targeting its efforts on preventing another Shipman is
questionable. The GMC does, however, seem anxious to avoid another Shipman in the
sense of avoiding the impact of such a case on the GMC. The GMC was sometimes in a
difficult position when giving evidence to the Inquiry and responding to its questions
and criticisms, for example, being unable to give reasons for GMC decisions because
they were not recorded.

The creation of
risk-related
information
593

JHOM
24,6

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

594

At the practical level, there is probably more potential value in focussed study
supplemented by other sources of data than in sweeping overviews of risk factors.
Because the FTP data are inherently concerned with the fitness of individual doctors to
practise, they do not map well onto contemporary approaches to patient safety, which
emphasise system-based and human factors approaches to medical error and patient
safety. However, the individual doctor and his or her individual failings remain an
essential aspect of risk to patient safety. Leape and Fromson (2006) estimated that in
the USA at least a third of physicians will, at some time in their career, have a condition
(most often a health or drug dependency problem) that impairs their ability to practice
medicine safely. Dealing with doctors whose health problems or clinical skills may
present a risk to patients as supportively as is compatible with protecting patients has
become a major part of the GMCs work. The GMC has a considerably larger fund of
information than is readily accessible in Siebel. Its data and expertise can contribute to
study of more closely defined risks, such as risks to patients arising from doctors
ill-health, or from different approaches to interaction with patients depending on where
a doctor was trained.
From the perspective of identifying risk factors, an important gap in the registration
data was information about employment and employment history. The GMCs
accumulated experience suggests that greater risks are associated with some work
environments, stages in a doctors career and specialties. More specifically, moving
employment frequently and working as a locum may be signs that a doctor is in
difficulty. The NHS holds much more detailed information about the registered doctors
it employs than does the GMC. For example published NHS statistics show numbers
practising in different specialties at different dates. Discussions with NCAS raised the
possibility that a combination of GMC and NHS data, even if only at an aggregate level,
would make fuller analysis possible. Optimising the quality and use of information in
Siebel requires resources, but special studies and modifications to the database could
make it possible to produce much more valuable information about risks to patients
from poorly performing doctors and where the risks are most likely to arise.
References
Agrawal, T. (2008), Fear and desire in systems design: negotiating database usefulness,
doctoral thesis, Massachusetts Institute of Technology, Cambridge, MA, available at:
http://dspace.mit.edu/handle/1721.1/42392?showfull (accessed 24 November, 2009).
Agrawal, T. and Silbey, S. (2008), Trust and surveillance in the technologically structured
workplace, paper presented at the Ecole Normale Superieure de Cachan (ENS Cachan),
Paris, October, available from the authors.
Allen, I. (2000), The Handling of Complaints by the GMC: A Study of Decision-making and
Outcomes, Policy Studies Institute, London.
Attride-Stirling, J., Humphrey, C., Tennison, B. and Cornwall, J. (2006), Gathering data for health
care regulation: learning from experience in England and Wales, Journal of Health
Services Research and Policy, Vol. 11 No. 4, pp. 202-10.
Bossen, C. (2006), Representations at work: a national standard for electronic health records,
Proceedings of the 20th Anniversary Conference on Computer-supported Co-operative
Work, November 4th-8th, Banff, Alberta, Canada, pp. 69-78.

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

Bowker, G. and Leigh Star, S. (1999), Sorting Things out: Classification and Its Consequences, The
MIT Press, Cambridge, MA.
Care Quality Commission (2009), Quality and risk profiles of NHS trusts in early 2010,
available at: www.cqc.org.uk/_db/_documents/Quality_and_Risk_Profile_v0_FINAL_
20091218.pdf
DoH (2000), DoH An Organisation with a Memory. Learning from Adverse Events in the NHS,
The Stationery Office, London.
Garfinkel, H. (1967), Good organizational reasons for bad clinic records, in Garfinkel, H. (Ed.),
Studies in Ethnomethodology, Prentice Hall, Englewood Cliffs, NJ, Ch. 6.
GMC (2006a), Good Medical Practice, The General Medical Council, London, available at: www.
gmc-uk.org/static/documents/content/GMC_GMP_0911.pdf (accessed 24 November 2009).
GMC (2006b), The GMCs Proposals on Healthcare Professional Regulation, The General Medical
Council, London.
GMC (2009), Fitness to Practise Annual Statistics 2008, GMC Council Paper 7 May, The General
Medical Council, London, available at: www.gmc-uk.org/about/council/2009_05.asp
GMC (2010), List of Registered Medical Practitioners Statistics, available at: www.gmc-uk.org/
doctors/register/search_stats.asp (accessed 9 February, 2010).
Hood, C. (2002), The risk game and the blame game, Government and Opposition, Vol. 37 No. 1,
pp. 15-37.
Hutter, B. (2005), The attractions of risk-based regulation: accounting for the emergence of risk
ideas in regulation, CARR Discussion Paper No. 33, available at: www.lse.ac.uk/
collections/CARR/pdf/Disspaper33.pdf
Hutter, B. and Power, M. (Eds.) (2005), Organisational Encounters with Risk, Cambridge
University Press, Cambridge.
Leape, L.L. and Fromson, J. (2006), Problem doctors: is there a system-level solution?,
The Annals of Internal Medicine, Vol. 144, pp. 1-8.
Leslie, M. (2009), Near misses: systems management in clinical medicine and public service,
paper presented at Close calls: Organizations, near misses, alarms, and early warnings,
ESRC Centre for Analysis of Risk and Regulation, London School of Economics, London,
March.
Lloyd-Bostock, S. (1999), Calling doctors and hospitals to account: complaining and claiming as
social processes, in Rosenthal, M., Mulcahy, L. and Lloyd-Bostock, S. (Eds), Medical
Mishaps: Pieces of the Puzzle, Open University Press, Buckingham, pp. 119-23.
Lloyd-Bostock, S. (2009), Risk-based approaches and professional regulation by the General
Medical Council, in Hood, C. and Miller, P. (Eds), Risk and Public Services. Special Report,
ESRC Centre for Analysis of Risk and Regulation, London School of Economics, London.
Lloyd-Bostock, S. and Hutter, B.M. (2008), Reforming regulation of the medical profession: the
risks of risk-based approaches, Health, Risk and Society, Vol. 10 No. 1, pp. 69-83.
Merrison, A. (1975), Report of the Committee of Inquiry into the Regulation of the Medical
Profession, The Merrison Report, Cmnd. 6018, HMSO, London.
Rasmussen, J. (1997), The concept of human error: is it useful for the design of safe systems?,
in Vincent, C. and de Mol, B. (Eds), Safety in Medicine, Elsevier Science, Oxford.
Reason, J. (1990), Human Error, Cambridge University Press, New York, NY.
Smith, J. (2004), Dame The Shipman Inquiry Fifth Report. Safeguarding Patients: Lessons from
the Past Proposals for the Future, The Stationery Office, London.

The creation of
risk-related
information
595

JHOM
24,6

Downloaded by Universiti Teknologi MARA At 08:57 18 May 2015 (PT)

596

Vaughan, D. (1999), The dark side of organizations: mistake, misconduct and disaster, Annual
Review of Sociology, Vol. 25, pp. 271-305.
Vincent, C. (2006), Patient Safety, Elsevier, London.
Walshe, K. and Boaden, R. (Eds.) (2006), Patient Safety: Research into Practice, Open University
Press, Maidenhead.
Waring, J. (2009), Constructing and re-constructing narratives of patient safety, Social Science
and Medicine, Vol. 69, pp. 1722-31.
West, E. (2006), Sociological contributions to patient safety, in Walshe, K. and Boaden, R. (Eds),
Patient Safety: Research into Practice, Open University Press, Maidenhead, pp. 19-30.
White, S., Hall, C. and Peckover, S. (2008), The descriptive tyranny of the Common Assessment
Framework: technologies of categorization and professional practice in child welfare,
British Journal of Social Work, Vol. 38 No. 3, pp. 1-21.
About the author
Sally Lloyd-Bostock is a Professorial Research Fellow at the LSEs Centre for Analysis of Risk
and Regulation. Her research concerns the relationship between psychology and law, and she has
particular interests in theoretical aspects of interdisciplinary work. Areas of her empirical
research have included the social psychology of negligence claims and formal complaints; health
and safety regulation; juries and courtroom decision making; and medical regulation by the
General Medical Council. She was previously Professor of Law and Psychology and Director of
the Institute for Judicial Administration at the University of Birmingham; and Senior Research
Fellow at the Centre for Socio-Legal Studies, University of Oxford. Sally Lloyd-Bostock can be
contacted at: s.lloyd-bostock@lse.ac.uk

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints