Key words
medical guidelines - evidence based medicine - clinical pathways - quality management
Schlüsselwörter
medizinische Leitlinien - evidenzbasierte Medizin - klinische Behandlungspfade - Qualitätsmanagement
Introduction
The changes in the structure of the German health system increasingly require stronger
interdisciplinary and specialised care. Doctors are presented with the increasingly
difficult task of finding a balance between optimising economics and meeting care
and quality standards. The inclusion of evidence and consensus-based recommendations
in routine clinical practice supports doctors in keeping up with these developments.
Medical guidelines have become established as the standard for synopsis of current literature and arriving
at consensus for clinical recommendations [1]. As scientifically based and systematically developed management tools for doctors
and patients they serve to support appropriate care for specific health problems [2]. Various studies have shown the positive effects of guideline-conform treatments.
As an example Wöckel et al. (2010) showed a direct association between adherence to
guidelines and both survival rates and recurrence-free interval in breast cancer patients
[3].
Clinical pathways support the immediate implementation of medical guidelines and the recommendations
they contain in clinical practice. They provide fixed documentation of the optimal
management for typical, defined disease patterns [4], [5]. They are interdisciplinary descriptions of treatment sequences for specific patient
types, adapted to local circumstances [6].
Despite the value of both tools being widely recognised [7], [8], German specialist associations state that guidelines are not yet used in routine
medical practice to the extent that those who compile them would hope [9], implying that there is optimisation potential. Barriers to their use include lack
of practice relevance, contradiction to other guidelines, difficulties with implementation
and poor availability [9], [10]. Likewise goal number 6 of the National Cancer Plan points out that it is not sufficient
to simply make medical guidelines available in the internet or medical journals. Rather,
in order to further develop guidelines in a timely manner, the development of a system
to continuously maintain and update the various guideline versions is necessary [11]. Although the methodological substance of guideline development, especially S3 guidelines,
is at a very high level, the information technological support and rigorous quality
assurance throughout the guideline life cycle[
1
] has proven difficult.
In addition it is apparent that the practical implementation of guidelines in general
and their adaption to specific clinical pathways have not been explicitly addressed
during guideline creation. Instead, the two instruments are often viewed as separate
entities [12], their synergistic potential remaining only partially exploited.
Doctors presumably still have reservations about the implementation and use of clinical
pathways and these have counterproductive effects. It can be concluded that doctors
are aware of medical guidelines [1], however it is not yet recognised that they have the potential to define medical
process standards. In the attempt to provide both guideline-conform and economically
sustainable medical care operative (executive) processes and medical guidelines will
have to become more strongly interlinked, especially in the hospital sector. The current
state of the practice of guideline compilation, distribution, implementation and revision
is summarised in [Table 1].
Table 1 Baseline situation – state of current practice in the guideline life cycle (derived
from [8]).
|
Planning, organisation, development and editing: Delay and high labour intensity in guideline development and revision due to inadequate
information technological support
|
-
Time and labour intensive processes (e.g. achieving consensus, coordination and communication between role-players in
the guideline commission, the systematic literature search and evidence evaluation)
[13], [14]
-
Use of Microsoft Office Solutions (MS Word, MS Excel)
-
Manual administration and verification of correlation between guideline components (e.g. fundamental information, supporting texts, recommendations,
statements, published evidence) and with other guidelines
|
|
Dissemination and implementation: construction of guideline-based clinical pathways impeded by lack of guidance for
adaption within the medical guidelines themselves
|
-
Passive dissemination of guidelines through their publication via specialist associationʼs publication
media and databases [15]
-
Supply in document form (prose, tables, algorithms)
-
Supply in various versions for different interest groups/purposes: full version, abridged version, patient guideline,
pocket or dia (software) versions [16], [17]
-
No process template or implementation tips for guideline users
|
|
Evaluation and update planning: No opportunity to monitor guideline implementation in practice or for documentation
and evaluation of divergence from guidelines (“guideline compliance”) due to a lack
of user-oriented quality control
|
-
Optional assistance for development and use of guideline quality indicators (QI) [18]
-
No generic method for comprehensive (all speciality groups) measurement and evaluation of the use of guideline recommendations in practice, i.e. no practical evaluation or knowledge/theory generation
-
Pioneering role of oncology guidelines: The standardised process of identifying and
implementing (result oriented) QI in medical guidelines by the Oncology Guidelines Program [19], and the uniform certification processes including standards for data communication
and outcome measurement [20]
-
Long updating cycles
|
Living Guidelines 2.0 – a Concept Proposal
Living Guidelines 2.0 – a Concept Proposal
The rigorous approach to active guidelines implementation affects almost all phases
of the guideline life cycle ([Table 1]). The concept of “Living Guidelines 2.0” is formulated here as a framework for a desired scenario. The aim is to provide guidance
for agreeing on and performing individual activities in the context of more stringent
guideline use.
Origins of the term
The term “living guidelines” is used in the literature to describe keeping medical
guidelines as up to date as possible with the best possible level of current medical
knowledge and recommendations[
2
]. This requires that guidelines be permanently updated, giving doctors timely access
to the latest evidence. Current research is focused increasingly on formalising the
if-then logic used in guidelines, to improve dealing with changes and to support computer-assisted
interpretation (e.g. decision support systems) [21], [22].
The term “Living Guidelines 2.0” goes further than simply staying up to date. It represents
a paradigm of care that has guideline-driven, quality-assured practice at its core.
This aim brings with it the challenge of a comprehensive interlinkage of guideline
compilation, practice implementation and maintenance (updating). The effect of guideline
recommendations on routine clinical practice, and the reflection of actual practice
in the guidelines would amount to direct integration of guidelines, thus strengthening
the relationship between guideline recommendations and clinical practice[
3
].
Consequence – Vision Living Guidelines 2.0
In the “Living Guidelines 2.0” concept, guidelines are no longer static, inert documentations
of current consensus opinion on treatment recommendations, but are applied rather
as dynamic, practice-rooted and therefore living tools. The concept expands the previous
focus of “updating guidelines” through integrating guideline recommendations into
the care process, implementing quality management adapted to the guideline, and feeding
back of practice-generated treatment outcomes into the further development of the
guideline and thus into the entire guideline life cycle ([Fig. 1]). In this way a more comprehensive, evidence-based standard of care should result,
achieving closer interlinkage of research/guideline compilation and clinical practice.
A description of the key fields of action follows.
Fig. 1 The “Living Guidelines 2.0” concept in the guideline life cycle (derived from [9]) – integration points and action recommendations.
In order to tackle deficits in current practice through the development and updating
of guidelines a guideline management system must be added to the existing MS Office
landscape. Such systems allow guideline fragments such as recommendations, proven
evidence, explanatory statements (background texts) and their shared properties to
be saved and maintained, semantically correct, in the guideline context. A clear modularisation
of guideline content is necessary to facilitate searching for individual guideline
elements, exchangeability and maintenance. The necessary improvement in content alignment
between different guidelines is also made possible [23]. Systematic guideline compilation is further supported by required acceptance and
testing procedures (e.g. consensus process) feeding directly into a workflow controlled
and documented by a guideline management system. Linking in to literature data bases
and the specification of continuous search requests would enable monitoring of the
latest, relevant publications and would thus further facilitate the updating process.
Guideline dissemination and implementation
A modularised system of guideline management on the basis of an electronic guideline
management system allows not only the circulation of guideline content in various
versions, and thus better adapted to the userʼs needs, but also supplies individual
modules relevant to specific user groups. Guideline content can be adapted flexibly
into diverse formats for various devices, and at the press of a button be transferred
to Apps for tablets or smartphones. Guideline users could also be allowed various
levels of right to access (e.g. reading, editing, administration). This would make
searching and finding content relevant to individual doctors and patients easier in
general and for specific treatment situations. On the basis of treatment relevant
information it would be possible to provide doctors with guideline-based recommendations
for specific patients during the treatment process. Doctors could simultaneously provide
feedback on recommendations in the decision-making phase, rating or commenting on
them. This would indeed make the guideline a living tool that brings together clinical
practice and research.
The current unsupported, laborious adaption of guideline recommendations to clinical
practice negatively affects the rapid application to routine clinical practice. For
these reasons a user-friendly, practical method of creating guideline-based clinical
pathways that can be integrated into existing structures needs to be developed and
evaluated. Existing methods can be used for this purpose (e.g. [24], [25]). The aim should be to provide guideline users with instructions (e.g. in handbook
form) to support the systematic development of quality assured clinical pathways based
on guideline recommendations. A possibility would be to provide the pathway developer
with a “pathway skeleton” or framework within the guideline that would need to be
adapted to hospital-specific circumstances. For this an expansion of the guideline
structure including an additional document would be necessary. Current medical guidelines
incorporate both explicit process descriptions in the form of clinical algorithms,
and implicit descriptions in background texts. The clinical algorithm, as the accepted
standard, should be utilised for describing and expanding the pathway framework. Implicit
and explicit process descriptions should be combined in such a way, with respect to
treatment scenarios, that the guideline user arrives at the fundamental medical course
of action intended by the guideline [24]. Such process descriptions should be compiled and published within the process of
guideline development. A first step could be to include the most important guideline
processes as an appendix, with descriptions in clinical flow diagrams that are already
incorporated.
Integrated quality control
If the transfer of guideline recommendations into clinical practice were methodically
and technologically better supported (with the aid of a guideline management system),
it is conceivable that data from guideline implementation in routine clinical practice
could be used for evaluation and for the generation of knowledge or theories for research
purposes. For this a standardised method of integration of guideline-based quality
indicators into clinical pathways is required (e.g. building on [26]). The expansion of guideline quality control systems into internal hospital process
standards improves procedural aspects of current outcome-based orientation. From the
perspective of processes, this would make a substantial contribution towards enabling
audit of guideline compliance, for which there is currently limited opportunity. The
interaction with clinical data intelligence methods may also have great potential.
These methods are suited to analysing general clinical practice data, organised in
various data bases, with a general framework overview and e.g. with respect to guideline
adherence. Available data can thus be used in a more targeted way for the evaluation
of medical and procedural issues (SmartData). It is clear that the prevailing IT infrastructure
as well as data protection organisation face great challenges.
The feedback of pathway-related QIs can provide important insights into the provision
of care, in turn allowing evaluation of guideline recommendations. The chain of information,
from first compilation through to final appraisal goes through the following steps:
compilation, annotation/commentary, conversion and appraisal. In this way e.g. repeated
departure from recommendations should be identified. Such application models are able
to highlight the possible need for revision of guideline recommendations. Research
hypotheses would continue to arise dependant on the analysis of specific care processes.
These functions have direct benefits for the compilation and updating processes since
they meet the requirements for problem-oriented prioritisation of guideline topics,
for tools demonstrating the effects of guideline implementation, and for the inclusion
of guideline users in guideline development [23].
Status Quo – Tools and Problem Solving Approaches
Status Quo – Tools and Problem Solving Approaches
The potential of support for guideline development and updating has been recognised
and addressed in various initiatives that have supplied specific IT products or methods.
Current approaches only function selectively and do not facilitate integrated, comprehensive
support as provided by “Living Guidelines 2.0” ([Fig. 2]). A number of typical solutions are discussed below. So far the guideline development
phase has experienced the most methodological and information technological support.
The AWMF Regelwerk (Engl.: handbook) [16], the German guideline appraisal tool DELBI (Deutsche Leitlinien-Bewertungsinstrument) and the GRADE method (Grading of Recommendations Assessment, Development and Evaluation) for evaluating
the quality of research results and medical evidence, and for the development and
compilation of reliable, trustworthy recommendations, are all established tools and
examples of methodological support for guideline development. Other guideline rating
instruments were analysed and compared by Semlitsch et al. (2015) [27]. They found that tools have increasingly been developed that provide an overview
of guideline quality using the systems of evidence analysis, the evidence supporting
recommendations, the clear formulation of aims and the naming of author teams as evaluation
criteria. The User Group – Med. Leitlinien e. V. (Engl.: Med. Guidelines) supplies
the guideline development portal to support collaboration between role-players in guideline compilation (e.g. preparation
of timescales, discussion forums, central data administration, tools for reaching
consensus and literature administration) [28]. The portal solution does not incorporate structural, semantic support for maintenance
and identification of guideline content. In contrast, the MagicApp developed within the EU-funded MAGIC project does cover this aspect. As a development
and publication platform it facilitates the compilation and revision processes as
well as the application of recommendations in practice. The XML export function creates
the basis for integration of guidelines into patient records, web portals or apps
for mobile devices. The toolʼs applicability to compilation ends, however, when content
deviates from suggested structure and in terms of predefined elements. The software
tool BridgeWiz
[29], which was developed with the GLIDES (Guidelines into Decision Support) project, also supports the compilation process
of individual recommendations using typical wh-questions (who, when, what, for whom,
how and why). The tool serves to formalise medical guidelines in order to translate
them into computer interpretable languages with the aim of supporting clinical decision-making.
It does not however permit the management of guidelines generated.
Fig. 2 Categorisation of existing solutions within the “Living Guidelines 2.0” concept.
The dissemination of guideline documents currently occurs passively via databases
such as provided by the AWMF. Distribution strategies have been compiled for various
interest groups (doctors, managers, patients, nursing staff, general public) as part
of the EU-funded DECIDE
[
4
] (Developing and Evaluating Communication Strategies to Support Informed Decisions
and Practice Based on Evidence) project and a corresponding toolkit has been developed.
The Cancer Council Australia wiki platform
[30] presents guideline content in the form of a wiki, enabling quick revision and linking
up of elements within and between guidelines. Though this fulfils the “Living Guidelines
2.0” requirement for dynamism, interlinkage with guideline implementation in clinical
practice is lacking. There are various other methods for this purpose that support
and guide the compilation of clinical pathways derived from medical guidelines [24], [25], [31]. These use completed guideline documents as their basis and have not yet been implemented
in the phase of guideline development.
Oncology Care in Obstetrics and Gynaecology as a National Benchmark
Oncology Care in Obstetrics and Gynaecology as a National Benchmark
The fundamental problem of interlinking of clinical practice and medical guidelines
has been recognised in the context of routine practice as well as by the previously
described approaches that mostly originate from scientific observations, and initial
solutions have been developed. The system of quality assurance among certified oncology
centres, which was essentially conceptualised by the Oncology Guideline Program (OL
program) and the German Cancer Association, is a good example. This system describes
the direct integration of QIs into medical guidelines and has also defined standardised
methods of documentation and recycling of QIs from clinical practice.
The process of identifying and implementing QIs in medical guidelines was developed
and standardised within the Oncology Guideline Program [19]. As an example, the current version of the interdisciplinary S3 guideline “Diagnosis,
Treatment and Follow-up of Breast Cancer” followed/applied the standard. The cornerstones
of care coordination are outlined in chapter 7 of the guideline emphasising clearly
that all measures must be transparent and quality controlled. The quality of the structure,
process and outcome quality must also be continuously appraised and the clinical cancer
register notified of relevant treatment data [32].
Furthermore, the agreed version of the guideline already contains a fixed, minimum
number of QIs derived on the basis of strong guideline recommendations and a corresponding
quality objective ([Table 2]).
Table 2 Integrated quality indicators for axillary lymphadenectomy in DCIS (guideline illustration)
[32].
|
QI no. 3 – Axillary lymphadenectomy in DCIS
|
|
Quality indicator
|
Content of recommendation (wording, LOE, EG); quality aim
|
Statements from the breast cancer guideline with respect to:
a. Impact on the health system
b. Source of statement from the 2012 guideline
c. Evidence base
|
|
Z:
Patients with axillary lymph node resection
N:
Patients with histology showing “DCIS” and completed operative treatment for primary
disease and breast conserving therapy
|
LOE 1b, recommendation level A
Quality target:
As few patients with primary axillary dissection or sentinel node biopsy (SLNE) in
DCIS with breast conserving therapy as possible
Target: < 5 %
|
a. Impact on the health system
Axillary dissection should not be performed in DCIS. Sentinel node biopsy should only
be performed when it is not possible to perform a secondary sentinel node biopsy for
technical reasons (e.g. after mastectomy)
b. Source of statement from the 2012 guideline
Statement Präinv-3b
c. Evidence base
Christiaens M et al. 2007; NZGG 2009
|
The QIs in turn form the basis of data to be incorporated into the tumour documentation
systems for analysis of structure, process and outcome quality. The standardised measurement
of results is assured via the “XML-OncoBox” whose central element is the uniform data
interface for submission of result data from tumour documentation systems and certified
cancer centre registers [20]. There is however still room for optimisation with respect to compatibility of gynaecological
and gynaecological oncology tumour registers to avoid double documentation.
For certification and re-certification centres must show their documentation and compliance
with key figures specified by the German Cancer Association (DKG). For certification
a key figure assessment form must be submitted, which is generated via the XML-OncoBox
and includes a result quality matrix.
The advantage of the XML-OncoBox is the support provided for calculation of the necessary
key figures required for certification. A single, uniform data interface has been
defined for this purpose that enables data transfer from the cancer centreʼs tumour
documentation system. Since the XML-OncoBox is directly linked to the tumour documentation
system data do not need to be exported externally but remain within the internal system
environment, which is a significant factor for data protection. For compatibility
with the multitude of tumour documentation systems the XML-OncoBox defines a unified
data schema. Data are standardised and the necessary key figures calculated using
standardised algorithms. In this way results from different centres can be compared.
When a documentation system is connected to it, the OncoBox can calculate all the
key figures and statistics required by the DKG, analyse trends, detect discrepancies
or derive hypotheses for further study.
The specification of QIs during guideline compilation and the requirement for documentation
and quality management systems based on these are fundamental premises for achieving
a closer correspondence between medical guidelines and routine clinical care.
This should be supported simultaneously by pragmatic measures such as standardisation
of graphic depictions of current guidelines. Even such simple measures have the potential
to boost guideline dissemination and implementation. As an example, a uniform guideline
template analogous to that of the OL program with additional features and functions
was developed within the DGGGʼs guideline program. This forms the basis of the recently
published guideline “Uterine Sarcoma”.
To conclude, [Fig. 3] illustrates how an integrated, quality system promotes the concept of living guidelines
(Living Guidelines 2.0). Further work is needed especially in the integration of clinical
pathway design recommendation and procedural orientation of hospital information systems.
Fig. 3 Guideline integration in practice – best practice taken from guideline-based cancer
treatment in Germany.
Conclusion
The provision of individual therapeutic measures always requires the consideration
of current scientific knowledge, the doctorʼs own expertise and the patientʼs consent.
Nevertheless medical guidelines offer the chance to further standardise medical treatments
and quality management and make them more comparable. The mode of action of guidelines
should be organized and determined systematically and not depend on the individual
doctorʼs desire to implement them. The realms of the medical guideline and routine
clinical practice should match up far better. Herein lies the chance of reliable tools
towards a quality-oriented, value-based remuneration system.
The existing quality management concept within certified cancer centres already shows
that the German health system is in a position to design systematic innovations for
large scale implementation although it is usually accused of being reluctant to change.
It is however clear that non-comprehensive measures do not take ground. An overall
strategy is required which allows current developments to be incorporated.
This article has provided an overview of the basic structure and existing framework
for guideline-conform treatment in Germany. The status of research in methodological
and information technological support of medical guideline development, dissemination
and implementation is analysed. The “Living Guidelines 2.0” concept forms the central
focus of discussion and individual aspects are illustrated. The interactions between
guideline developers and users is analysed in terms of the typical guideline life
cycle, and it is shown how clinical pathways and a coordinated quality system can
form the link between these two groups.
The analysis of the current situation shows that guideline compilation and (passive)
dissemination already enjoy good methodological and information technological support.
There is however very little support for implementation, monitoring at the hospital
level together with user and process-oriented guideline quality management, and recycling
of generated information back to the guideline group for updating. Currently existing
solutions with great potential for innovation are not yet comprehensively integrated,
functioning in isolation or fitting poorly together.
The major deficits at the interface between developers and users can only be overcome
if a large number of compatible methods and tools are developed and promoted in a
unified manner on the basis of a common vision for guideline production, implementation
and development. The “Living Guidelines 2.0” concept provides such a vision, which
now needs to be converted to a common plan of action. To this end measures have already
been undertaken by Prof. Beckmann (DGGG), the DKG e. V. and the TU Dresden. The increasing
number of activities towards integrated quality assurance shows that development is
in progress among both developers and users. The vision of “Living Guidelines 2.0”
will give this process a name.
