Occupational safety, hygiene, radiation protection, technical safety, pharmacy, transfusion medicine
Or copy link
Donabedian has written more than 100 articles and 7 books on quality assurance in medicine. His contributions go far beyond the concept of the triad of structure, process and outcome quality. He has written extensively on the epidemiology of patient needs, the importance of comprehensive insurance against the consequences of illness, the relationship between cost and quality, and the monitoring of service delivery.
Over the many years of his scientific career, Donabedian adapted his nomenclature to the focus of his work. Later on, he hardly ever spoke about the famous triad, but searched for irrefutable attributes of good medicine. He identified seven attributes, which he called "pillars", on which the quality of individual medicine and healthcare at a societal level should rest equally.
At the end of his life, Donabedian regretted that he had only become famous because of his "structure-process-outcome paradigm". He himself admitted that the triad did not fulfil all the needs of assessment (SHIP 2001).
In his classic publication DONABEDIAN In 1966, he examines the question of how quality should be understood. This is not a philosophical problem, but a prerequisite for objective evaluation. As long as an accumulation of value judgements on individual aspects, characteristics and contents of medical care is regarded as "quality", it remains no more than what each individual imagines it to be. For a scientific, empirical study, the multitude of possible dimensions and criteria must be narrowed down, their justification proven and their measurability analysed. Anyone who wants to "pay for performance", align state planning with indicators or carry out a ranking must achieve such objectivity - at the latest if the evaluation is to stand up in court.
Donabedian proposed to first determine what should be evaluated: 1. the results (outcome), which 2. result from the processes of the treatment and 3. the structure that is available for the processes.
He clearly recognises the limitations of evaluating the results of "good medicine" and soon comes to the conclusion that the results do not speak for themselves. Results must be viewed with great caution. He does not dismiss them as unsuitable, but sees them as an important indicator of the process characteristic "effectiveness". He was not yet familiar with the discussion about clinical studies as a tool for investigating the characteristic of "effectiveness", as we know it today as "evidence-based medicine".
The treatment process seems to him to be more important for the question of properly practised medicine. He distinguishes between the process itself, insofar as it is known to be "good", and the technical ability to carry it out (performance). The medical treatment can then be assessed on the basis of the characteristics appropriateness, completeness, redundancy, technical competence, coordination and continuity. At this point, he already adds the characteristic that he later reminds us of again and again: acceptability for the recipient of the service. Unfortunately, the concept of acceptability has so far been overlooked by many of his successors.
However, the treatment processes can only be carried out if the necessary resources are available: appropriate premises, equipment, qualified staff, an organisational process, an administrative structure and sufficient financial resources. Without appropriate structural requirements, there can be no good processes.
Donabedian fits the various evaluation methods of his time (1966) into this scheme. He weighed their performance against the extent to which they allowed well-founded judgements to be made. The result of his study is not particularly encouraging - but let's not forget that quality assurance in healthcare was still in its infancy.
There is only one thing he certainly did not want: he did not want to differentiate between three types of quality, each of which could be defined and assessed separately. His scepticism towards the enthusiasts of quality of results speaks against this.
To this day, many try to define quality in terms of "good quality". To do so, they list a number of features of medical treatment that they believe are part of this. Politicians and health insurance companies prioritise these differently to service providers and patients. Their descriptions are vague, such as "...must be orientated towards the well-being of the patient", "...must be holistic, human or empathetic" or "sufficient, appropriate and economical".
Others have drawn up a set of precise requirements for equipment, qualifications and scope of services that should be regarded as the standard for good medicine. Some consider the requirements to be insufficient, while others see them as excessive. Agreement on the catalogue is rarely reached. What is regarded as quality depends on the respective environment. All requirements must be differentiated according to the respective needs - there is no such thing as "one" quality. Quality is not an ideal state, but results from the requirements that are set.
Those who believe that they can somehow recognise what is good are attached to an intuitive understanding of quality. They perceive which characteristics belong to it and how they relate to each other as a whole. Many laypeople agree with this and spontaneously agree without realising that they mean something else.
A scientifically based assessment cannot be based on this. Quality is not a totality, but a set of characteristics that belong together but must be assessed using different methods. The characteristics influence each other, but are not equivalent.
The weighting of the characteristics is not the same in all situations. Sometimes effectiveness is considered extremely important, sometimes the demand for safety takes a back seat to the demand for better acceptability (in this case perhaps proximity to home). Different requirements are made depending on the situation and the person.
Today, quality is defined as the degree to which a set of characteristics of an object fulfils requirements (DIN EN ISO 9000:2015).
But what is the object of the analysis? Can structures, processes and results be considered separately? This is where the triad has not proved its worth. Worse: the confusion has only increased by adhering to this schematism.
The easiest way to explain why "structural quality" cannot be determined for itself is to say
What structural requirements could be agreed? For example, clear ideas have been developed for the structure of emergency centres (RIESSEN 2014). One of the requirements for the organisation of emergency centres is that they are closely linked to hospitals and the emergency medical service. Many details are very specifically required, e.g. that the initial assessment and care should be carried out by doctors and nurses trained in emergency medicine, who may consult appropriate specialists depending on the situation. They must have the diagnostic procedures relevant to emergency medicine (e.g. emergency laboratory, ECG, sonography/echocardiography, X-ray, computer tomography) available around the clock. They should be equipped with an emergency admission ward for short-stay patients, which would allow short-term inpatient observation without the need for significant further diagnostics and treatment. I will not list all the individual requirements, but I will add the desire for treatment rooms in which patients can be treated with discretion.
All of this would be roughly what is called "structural quality".
But what are the requirements based on? Emergency centres are assigned to three categories: Their equipment is based on the processes that take place in them. The "structural quality" required for the treatment processes is always demanded. The structure of an emergency centre in a district hospital is different from that of a university hospital and certainly different from that of a psychiatric clinic. There is no abstract "structural quality" that applies to everyone.
When looking at the treatment processes, "structural quality" appears as a "process resource": for each process, we can (and must) specify what we need for it. Rooms, equipment, materials, sufficient and qualified staff and monitoring of process control.
What is called "structural quality" is derived exclusively from the processes. Less leads to disruption of the process, more is decoration, luxury or waste.
The treatment processes require resources. Structural quality itself says very little about how good the medicine is.
Why we should not talk about the quality of results is more difficult to understand.
To the layperson, it seems so plausible: whether a treatment is good or bad is best judged by its results. If patients recover, that's good, if not, that's bad. Everyone can see for themselves from the results whether they have been treated well. He who heals is right. The whole world seems to be obsessed with this fallacy.
It is completely unclear what a result actually is.
In most cases, it is not even possible to agree on when the result occurred: the state of health with which you leave the hospital? Or that you are still sneezing 4 weeks later? It is often only possible to determine in the distant future whether the desired result has been achieved: has the patient "beaten" their cancer? How much longer did the treatment prolong their life? What is the decisive final result? What are merely interim results?
Antihypertensive therapy is supposed to protect against strokes - but protection cannot be measured in individual cases, only in terms of the lower probability of its occurrence in a larger population. Does the result mean "Reduction of stroke mortality in the population?
What matters about the result? If we have become so healthy, we may have suffered from side effects or complications that could have been avoided or must be accepted as unavoidable. Does the health benefit outweigh the damage we have suffered? In individual cases this may still be possible. However, if we look at all the patients treated, we eventually recognise considerable differences in the effectiveness and safety of the treatment procedures. As impressive as the desired results are, we have doubts as to whether the risks and opportunities are in reasonable proportion. We look at the positive effects and note the regrettable disadvantages.
Which of these is the result? We cannot simply add up the consequences to an overall result. But then what is "the" result? A mixture of positives and negatives. Acceptability falls to the back anyway.
Often the result cannot be read or measured at all. This is because most medical treatments are "special processes". The quickest way to understand what is meant by this is to look at the process of sterilisation. The result of sterilisation is sterility. But nobody can see or touch sterility (then it would no longer be sterile). All diagnostic processes are "special": you cannot tell from their results whether they are good or not. If the measurement procedure is correct and accurate, we trust the result - but we cannot judge the result.
A result is always what happens in the end. Naively, we see the cause of the result in the action that preceded it. In medicine, no one can say for sure in individual cases whether the desired event was caused by the treatment or whether it might not have materialised on its own. The cause cannot be seen in the result. Doctors and patients are easily deceived. They fall for a foil á deux.
Evidence-based medicine has made us aware of how sceptically we need to view results. In studies, events are counted - desirable and undesirable. If there are more desirable and fewer undesirable events in the group treated with A than in the group treated with B, then we say that A is more effective and safer than B.
So we don't even talk about results in clinical trials. That would be very naive. We test the quality characteristics of efficacy and safety of the treatment processes in carefully designed experiments. Some prove to be extremely effective and safe (e.g. anaesthetic procedures), others are effective but unsafe (e.g. certain surgical procedures or radiotherapy), some are very safe but not effective. You already know which ones I mean).
The rapture with which the quality of results is regarded as the stone of quality methods is responsible for many confusions in quality management. It takes a great deal of effort to distinguish between accidental and causal events. Large-scale clinical studies are usually indispensable. The characteristics of the treatment procedures must be verified and repeatedly validated. Only then are our decisions reasonably evidence-based.
Quality assurance of results is frustrating, costly and ineffective. It has been abandoned and replaced by mastery of the verified and repeatedly validated process.
In short: let's forget the quality of results. It is the quality characteristics of effectiveness and safety of treatment or accuracy and correctness of diagnostics that matter. The characteristics of the diagnostic and therapeutic processes can be tested and measured - quality of results cannot.
Donabedian writes in 1988 (DONABEDIAN 1988): "Because a variety of factors influence outcome, it is impossible to know with certainty the extent to which an observed outcome is attributable to prior treatment - even when extensive adjustments are made for differences between cases. What is needed is confirmation through a direct assessment of the process itself.
In the end, the process quality remains. The object of consideration in quality management are the production processes - in medical care, the treatment processes - diagnostic, therapeutic and nursing. For each process, a set of characteristics can be identified that can be checked and measured.
The characteristics of effectiveness and safety tell us something about the probability with which we can expect certain results. The configuration of the processes determines the essential resources. From the core processes, we can derive the need for support and management processes that ensure an undisturbed and efficient process. This is why the standard refers to a "process-orientated QM system" (DIN EN ISO 9001:2015). Everything revolves around the process.
Requirements are initially placed on each process. The process is designed in such a way that it fulfils the requirements to the highest possible degree. Proof of performance is provided during product development. Only then can the performance be reliably and effectively brought into routine use. Most processes, at least in medicine, are "special processes", i.e. we cannot directly read off the result. We rely on effectiveness and safety because we carry out a verified and validated process design under the conditions of process control.
Treatment processes are always designed for individual patients. In this respect, patient-centredness says everything and nothing. However, they are acceptable to patients in different ways. Today, we know the framework conditions for the acceptance of medicines, medical devices, medical and nursing services quite well. We know that the characteristics of acceptability are often enough the deciding factor - the pandemic has shown us this once again. The commitment to patient-centredness should motivate us to pay more attention to the process characteristic of acceptability that Donabedian cared so much about.
Donabedian was clear that the first step was to show how structure, process and outcome are actually connected. He hoped that organisational science, behavioural research and clinical research would contribute to this. His triad of structural, process and outcome quality successfully initiated quality assurance. Now it stands in the way of the modern concept of quality.
Author: Dr Ulrich Paschen QM elektronische post - Beiträge zur Guten Praxis in Medizin und Wissenschaft Broadcast 20 Fahrdorf, 22 May 2018 Reproduction is permitted provided the source is acknowledged and a specimen copy is supplied.
Save my name, email, and website in this browser for the next time I comment.
We firmly believe that the internet should be available and accessible to everyone and endeavour to provide a website that is accessible to the widest possible audience, regardless of circumstance or ability.
To achieve this, we adhere as strictly as possible to the Web Content Accessibility Guidelines 2.1 (WCAG 2.1) of the World Wide Web Consortium (W3C) at AA level. These guidelines explain how web content can be made accessible to people with a variety of disabilities. By adhering to these guidelines, we ensure that the website is accessible to all people: the blind, people with motor impairments, visual impairments, cognitive disabilities and more.
This website uses various technologies to make it as accessible as possible at all times. We use an accessible interface that allows people with certain disabilities to customise the user interface (UI) of the website and adapt it to their personal needs.
In addition, the website uses an AI-based application that runs in the background and constantly optimises accessibility. This application corrects the website's HTML code and adapts its functionality and behaviour to the screen readers used by blind users and to the keyboard functions used by people with motor impairments.
If you notice a malfunction or have suggestions for improvement, we look forward to hearing from you. You can contact the website operators by e-mail at upaschen@gutehospitalpraxis.de.
Our website uses Accessible Rich Internet Applications (ARIA) attribute technology and various behavioural modifications to ensure that blind users accessing the website with screen readers can read, understand and use the features of the website. Once a user with a screen reader accesses your website, they are immediately prompted to enter the screen reader profile so that they can browse and use your website effectively. Here's how our website fulfils some of the key requirements of screen readers, along with screenshots of sample code:
We want to support the widest possible range of browsers and supporting technologies so that our users can choose the tools that work best for them with as few restrictions as possible. That's why we've worked very hard to support all the major systems that make up over 95 % of the user market share, including Google Chrome, Mozilla Firefox, Apple Safari, Opera and Microsoft Edge, JAWS and NVDA (screen readers).
Despite our best efforts to make it possible for everyone to adapt the website to their needs, there may still be pages or sections that are not fully accessible, are in the process of being made accessible or for which there is no appropriate technical solution to make them accessible. Nevertheless, we are continuously improving our accessibility by adding, updating and improving options and features and developing and introducing new technologies. All of this is to achieve the optimum level of accessibility and to keep up with technological progress. If you have any questions, please contact upaschen@gutehospitalpraxis.de
