Evidence-Based Practice in Social Work and Nursing

PART 1 – From the single study to orientation: an overview of syntheses

Why not everyone reads all the studies themselves

In the earlier modules you learned to turn a practice question into a search question (Module 2), to classify study designs (Module 3), to search in a targeted way (Module 4) and to read a single study critically (Module 5). That is demanding and takes time. In everyday work nobody can start their own literature search for every decision, appraise studies and weigh the findings against each other. That is why there are documents that bundle knowledge and condense it into orientation. In this module we call them syntheses for practice: texts in which a group of experts collects and weighs findings according to a described procedure and turns them into recommendations, requirements or principles.

The decisive point is never just the question “What does it say?”, but: Who produced it, by what procedure, for whom and when? If you can read these four details, you can also classify a document you have never seen before. That is exactly what we practise here. The framework from Module 1 still applies: a synthesis is one building block of evidence – it replaces neither the expertise of the professional nor the question of what the person concerned wants and what their situation is.

The overview: five branches around classifying

Figure 1 shows the map of this module. At the top left are the AWMF guidelines, at the top right the DNQP national expert standards, at the bottom left the professional standards of social work, at the bottom right the law, and in the bottom centre the critical classification that links all the other branches. If you lose the thread later in the text, it is worth looking back at this figure.

Four terms you must keep apart

In everyday study life, “guideline”, “standard” and “directive” are often used as if they meant the same. For term papers and timed written papers that is a risk, because the documents differ in origin, procedure and binding force. We therefore work with four clearly separated terms.

Guideline (Leitlinie): an aid to orientation for decisions in a field of care, developed by a described procedure. In Germany the Association of the Scientific Medical Societies in Germany (AWMF) classifies guidelines by the degree of systematic method in the development process and distinguishes several classes for this purpose (AWMF, 2025). A guideline is a decision aid, not a law.

National expert standard (Expertenstandard): The German Network for Quality Development in Nursing (DNQP) describes national expert standards as evidence-based, monodisciplinary instruments that set out the contribution of nursing to key quality risks (DNQP, 2025b). “Monodisciplinary” means that the standard speaks from the perspective of one professional group, here nursing. Other professional groups in the team are not automatically included.

Professional standard (Fachstandard): In this course we use the term as a collective label for positions of professional and specialist associations that describe attitudes and professional requirements. These include the professional ethics of the German Professional Association for Social Work (DBSH, 2024) and the international definition of social work (IFSW, 2014). Both are positions of professional associations and are not law.

Law or statutes: binding legal rules. An example from nursing that you know from Module 1 is the professional code of the Nursing Chamber of North Rhine-Westphalia (Pflegekammer NRW, 2025), which, however, applies only to chamber members in NRW. You will find more in the legal basis box further down.

Figure 2 places the three professional document types side by side. Pay attention to the “Perspective” row: it often decides whether a document is intended for your field of work at all.

The following table adds law as a fourth type and makes the differences in binding force visible. The limit of your statement matters here: whether a professional document plays a role in a specific legal dispute, contract or administrative procedure is a legal question. This module does not answer it, and professionals do not give legal advice in individual cases.

A decision tree for everyday use

When you come across a document, four questions help, answered in order. Figure 3 shows the decision tree: first you check whether it has been officially published (then it is law), then whether it carries an AWMF class (then it is a guideline), then whether it comes from the DNQP (then it is a national expert standard), and finally whether it is the position of a professional association (then it is a professional standard). If a question stays open, you clarify publisher, procedure and version date and start again.

PART 2 – Guidelines: what the “S” tells you and how to read a guideline

The S in the AWMF classification

The AWMF rulebook for guidelines (AWMF-Regelwerk Leitlinien) is available in version 2.2 of 22 May 2025 (AWMF, 2025). It sets out how guidelines are developed and classified. The “S” stands for the degree of systematic method in the development process (AWMF, 2025). So it is a statement about the procedure – not about the correctness or importance of an individual recommendation. This is an obvious misunderstanding, and one sentence avoids it: the class describes how a guideline came about, not whether it fits your case.

Four classes, four procedures

According to AWMF (2025), the classes can be distinguished as follows. An S1 recommendation for action comes about through the informal consensus of a group of experts. An S2k guideline requires a representative guideline group and structured consensus-building under neutral moderation. An S2e guideline requires a systematic search for, selection and appraisal of the evidence. An S3 guideline combines all elements of systematic development. Figure 4 lays these descriptions over each other as a matrix: you can see at a glance which elements are required in which class and where there is only a “–”, meaning that no corresponding requirement is stated.

Two reading tips help you understand the matrix. First: S2k and S2e are not preliminary stages of each other, but set different emphases – here the breadth of participation and the structure of voting, there the systematic work with the body of studies. Second: the classes do not form an evidence pyramid in the sense of Module 3. There you saw that evidence hierarchies are a way of thinking and that the suitable study design depends on the question. The S-classes say something about the development procedure of the guideline, not about the type of study behind individual sources.

Guidelines for social work too? The example of the child protection guideline

The very name of the AWMF shows that it comes from medicine. The child protection guideline (Kinderschutzleitlinie) (AWMF register no. 027-069) shows that guideline work can open up across professional boundaries: its title expressly mentions the involvement of child and youth welfare services and education, and it is aimed at the health system, child and youth welfare and education (Kinderschutzleitlinie, 2022). 82 scientific societies, organisations, federal ministries and federal commissioners were involved in its development (Kinderschutzleitlinie, 2022). This breadth shows how many perspectives were brought together; it is expressly not, however, proof of the effect of individual measures. The content or recommendations of the guideline are not the subject of this module – here we are concerned with classifying.

The statusis also important: the short version was updated on 9 January 2022, because of the Child and Youth Strengthening Act (Kinder- und Jugendstärkungsgesetz). The document carries a note that it has not been updated for more than five years and is currently being revised (Kinderschutzleitlinie, 2022). This applies as of the retrieval date October 2026. We do not know of a date for a new version, and we do not speculate about one. For you this means: a guideline with a revision note is not worthless, but you state the note openly when you cite it.

Worked example: reading a guideline in six steps

Before you practise yourself, we work through an example. We take the child protection guideline and read out only what a profile sheet needs: title and register number, class, version date, validity note, target group and contributors. Figure 5 shows the six steps in the order in which you work through them on the document. Why this order? First the identity of the document, then its procedure, then its currency, then its audience – this stops you from taking over content from an outdated version without noticing.

Result as a profile sheet in three sentences: “The child protection guideline (AWMF register no. 027-069) expressly includes child and youth welfare services and education and was developed with the involvement of 82 scientific societies, organisations, federal ministries and federal commissioners. The short version dates from the update of 9 January 2022; the guideline carries a note that it is currently being revised (as of retrieval in October 2026). In my work I cite the version with its year and state the note.” Note what the profile sheet does not contain: not a single recommendation on content. You read those in the document itself and discuss them in the team or with your practice supervisor.

PART 3 – National expert standards: how the DNQP bundles knowledge in nursing

Definition and structure

National expert standards of the DNQP are, as defined above, evidence-based and monodisciplinary instruments. They consist of a preamble, a commentary and structure, process and outcome criteria for every step of the nursing process (DNQP, 2025b). The nursing process describes how nursing is planned, carried out and evaluated systematically. In a sense the standard lays a grid over this sequence and asks at each point: what is needed as a precondition, what should happen, and how do you recognise the result?

The three types of criteria can be understood like this: Structure criteria describe the preconditions, such as competences, resources and rules in the organisation. Process criteria describe the action itself. Outcome criteria describe what should be achieved. The commentary gives reasons for and explains the criteria; the preamble frames the standard. The following table turns each element into a question you can ask while reading. You find the answers in the standard itself; we deliberately do not reproduce any content of a standard here.

Four phases: from development to updating

A DNQP national expert standard goes through four phases (DNQP, 2025b). In the development phase a systematic literature analysis is carried out; an expert working group that includes at least one representative of those affected drafts the standard. In the consensus phase the draft is discussed at a public consensus conference. The model implementation takes place in around 25 organisations and comprises training, adaptation to local practice, introduction and audit. The audit is a structured review of how well the introduction went. Finally comes the updating: first after five years, then every seven years, with an eight-week online consultation.

Two points matter for you as a student. First, the involvement of a representative of those affected is built into the procedure – a foretaste of Module 7, which deals with participation. Second, a national expert standard is expressly not “finished”: it is revised at fixed intervals, and the year of the version therefore belongs in every source reference.

How evidence gets into the standard – and where the experts’ judgement counts

The DNQP searches hierarchically: first systematic reviews, then primary studies, among others in MEDLINE, CINAHL and the Cochrane Library (DNQP, 2025b). It documents the approach with a PRISMA flow diagram – you know the principle from Module 4. The studies are appraised with instruments such as AMSTAR-2 (for reviews), RoB 2 and AGREE II (for guidelines); the certainty of the evidence is presented with GRADE (DNQP, 2025b). You got to know RoB 2 in Module 5 as a tool for randomised trials. The fact that the DNQP first looks for reviews is an example of a pragmatic hierarchical approach, without this turning into a fixed evidence pyramid: which type of study suits the question remains the guiding question from Module 3.

One detail is particularly revealing for critical classification: where the evidence is weak, the judgement of the expert group carries particular weight (DNQP, 2025b). That is methodologically honest, but it also means that a standard visibly contains elements of professional judgement. That can make sense when studies are lacking; but it is something different from a finding based on several well-conducted studies, and you should make this distinction visible in your presentation and term paper.

The procedure itself is being developed further

It is not only the individual standards that are updated, but also the method by which they are produced. In November 2025 the DNQP published the sixth, revised version of its methods paper; earlier versions date from 2002, 2007, 2011, 2015 and 2019. The new version was developed in line with the GIN-McMaster Guideline Development Checklist (DNQP, 2025b). Figure 6 shows this sequence on a timeline. Look at the gaps: they are to scale, which is why the largest gap lies between the fifth and the sixth version. Do not confuse the timeline with the update intervals of the individual standards (after five years, then after seven): this is about the methods paper, not about a particular standard.

PART 4 – Professional standards of social work and the question of care settings

The professional ethics of the DBSH

The professional ethics of the DBSH have been available in a second edition since December 2024; it is a completely revised and expanded version of the first edition of 2014 (DBSH, 2024). Among its supreme principles are autonomy, solidarity and justice. It links ethics and professional practice and assumes that the practice competence of social work rests on scientifically sound knowledge (DBSH, 2024). For our topic this is an important link: the professional ethics anchor the claim that practice should be knowledge-based and at the same time preserve autonomy – exactly the tension we look at in Module 1 and Module 7.

The international definition of social work

The Global Definition of Social Work was adopted by the general assemblies of the IFSW and the IASSW in July 2014 (IFSW, 2014). Its commentary on knowledge describes social work as inter- and transdisciplinary. Research and theory are regarded in it as application-oriented and emancipatory and are often co-constructed together with service users; in addition, indigenous knowledge is recognised (IFSW, 2014). This understanding of knowledge is broad. A medically oriented guideline is one form of knowledge among several in it, and the voice of service users belongs to the production of knowledge itself. One possible interpretation (the definition itself does not say this): this explains why social work approaches guidelines with interest but also with critical questions.

Several perspectives on standards in social work

You know the debate about evidence-based practice in social work, with its supporting, critical and mediating positions, from Module 1. For standards this means: anyone who understands them as orientation sees in them an aid to justifying decisions. Anyone who understands them as a recipe meets exactly the objections of the critics. A reflective position lies in between: standards structure decisions, they do not take them away. In a term paper you convince not with one side, but with the ability to present both sides fairly.

Why standards matter for all care settings

To close this part, a sober look at figures. In December 2023, almost 5.7 million people in Germany were in need of care within the meaning of Book XI of the German Social Code (SGB XI). Of these, 86% or 4.9 million were cared for at home and 14% or 800,000 in full inpatient care in care homes (Destatis, 2024). Figure 7 shows the two groups as bars. The values: 4.9 million people at home, 0.8 million in full inpatient care, as of December 2023, Germany.

What follows from this for our topic? The great majority of people in need of care live at home. A view of standards that only looks at everyday life in care homes would overlook this majority; the reverse also applies. Social work meets people in need of care and their relatives in many fields of work, for example in counselling or in hospital social services. However, the figures say nothing about how and where an individual standard is used – that would be a different question. The increase of 730,000 (+15%) compared with December 2021 should also not be interpreted too quickly: according to Destatis, it is partly due to the broader definition of the need for care in force since 1 January 2017 (Destatis, 2024). Finding and interpretation stay separate.

PART 5 – Critical classification: limits, areas of tension and consequences

Currency: every synthesis has a date

Guidelines and standards are snapshots. The child protection guideline carries a revision note (Kinderschutzleitlinie, 2022), DNQP standards are updated at fixed intervals (DNQP, 2025b), and even the rulebooks change: the AWMF rulebook is at version 2.2 (AWMF, 2025), the DNQP methods paper in its sixth version. Consequence for your work: always state the version and the year, and point out notes. If you cite a document that is currently being revised, that is not a weakness in your argument but a sign of care.

Procedure is not fit

The class of a guideline and the phases of a national expert standard tell you how thoroughly a document was produced. They do not tell you whether it fits the person sitting in front of you. For that you need the framework from Module 1: evidence, condition and circumstances, preferences, and the professional expertise that brings everything together. Module 7 deepens this with the question of how life situation, wishes and resources feed into a reasoned decision. There are also limits to transferability: a document came about in a particular context; whether it fits your field of work, your organisation and your service users is a separate question. In social work, there is also the fact that for many social-pedagogical questions hardly any robust studies exist – a gap in research that you may name in a term paper without glossing over it.

Monodisciplinary in a multiprofessional team

The national expert standard is expressly designed to be monodisciplinary (DNQP, 2025b). That is a strength because it addresses nursing precisely, and at the same time a limit: other professional groups work in the team, and their tasks it does not describe. The child protection guideline, by contrast, brings the health system, child and youth welfare and education together (Kinderschutzleitlinie, 2022). For social work professionals this means: clarify interfaces in the team, name your own role, and do not assume that a document of another professional group describes your own task.

Open questions: standards and self-determination

One open question is how standards and self-determination fit together. The DBSH professional ethics put people’s autonomy first (DBSH, 2024); the NRW professional code commits chamber members to respecting the right to self-determination (Pflegekammer NRW, 2025). A standard describes what is professionally usual or what to aim for – and a person can decide differently. How such situations are to be resolved in an individual case depends on the case, the team and the procedures of the organisation; Module 7 prepares for this. Here the key sentence is enough: a standard is a starting point for the conversation, not a substitute for it.

How to deal with this in a term paper, presentation and timed written paper

In a term paper you name the document type (guideline, national expert standard, professional standard or law), briefly describe the procedure, state version and year, and discuss limits. In the APA 7 reference you list the issuing organisation as the author. In a presentation, a profile sheet with class or structure, version date, target group and contributors is often enough. In a timed written paper a typical task is the comparison: what distinguishes a guideline from a national expert standard, and what is neither of the two because it is law? With the table and decision tree from this module you can answer this in a structured way. Important: never present guidelines and standards as law.

For your studies

Self-reflection – 5 questions

Voluntary · not evaluated – your answers are not saved and nobody analyses them. Click an answer; the explanation helps you to judge your own understanding.