Evidence-Based Practice in Social Work and Nursing

PART 1 – Taking stock: what you bring with you from seven modules

From single tool to whole process

Evidence-based practice is not a single action but a sequence. The concept of evidence-based nursing describes steps for nursing that range from shared decision-making and formulating the problem through searching and critical appraisal to changing practice and evaluation. Exactly this logic is found – in modified form – in many study tasks: a term paper, a presentation or a timed-written-paper task often requires you to clarify a question, draw on referenced sources, classify them and in the end take a reasoned position. In the practical placement this becomes a team question to which you look for a comprehensible answer.

This module is therefore not a new block of content but a link. It shows how the building blocks of Modules 1 to 7 can be combined in three formats that really occur in studies and at the start of your career: the journal club, the literature report (a structured literature overview) and the critical review. Figure 1 gives you the overview: in the centre is evidence-based practice in your studies, around it the five building blocks that we will go through one after another.

Looking back: which module supplies which building block?

So that you do not learn anything twice, here is just the map. The definitions are in the respective modules; the table shows what you need them for now.

A why-question to think about: why is it not enough to know individual studies? Because, according to the understanding of the Institute for Quality and Efficiency in Health Care (IQWiG, 2025), a decision never follows from the evidence alone: it comes about when professional expertise brings the evidence together with a person’s condition, circumstances and preferences. A journal club therefore does not end at “What does the study say?” but at “What follows from it for these people in this situation?”

A working definition for the journal club

There is no uniform standard for the journal club. In this course we understand it as a one-off or regular meeting in which a group reads a chosen specialist text together, appraises it critically and relates it to its own practice. This is a working definition for this module, not a fixed technical term with a single source. In a seminar, in a study group and in a team the format often looks different; what matters is that selection, appraisal and consequence remain comprehensible.

PART 2 – Journal club: reading together, appraising critically, drawing consequences

The sequence in five steps

Figure 2 shows a practicable sequence for study groups and team discussions. It is deliberately kept lean so that it also works alongside lecture time and everyday practice.

Step 1 – Clarify the question: Before anyone looks for a text, it is settled which practice question lies behind it. A PICO formulation from Module 2 helps because it names population, intervention, comparison and outcomes (Higgins et al., 2024). Without this clarification you quickly end up discussing what is interesting rather than what fits.

Step 2 – Choose the text: The choice depends on the question, not on the title that promises the most. A question about effects needs a different design from a question about experience and meaning (Module 3). On access: you can reach many full texts through your university library; databases such as PubMed often supply only citations and abstracts, not full texts themselves (National Library of Medicine [NLM], n.d.).

Step 3 – Present: One person summarises question, method and finding in their own words. The separation matters: first you report what the text says, then you interpret what it means. If you mix the two, you lose the chance to question the interpretation later.

Step 4 – Appraise together: This is where the tools from Module 5 come in: where might systematic errors (bias) have distorted the results? Are effects reported in absolute numbers? How certain are the results, according to IQWiG (2025), with regard to internal validity, external validity and statistical precision? You do not have to answer every question – but you should be able to say which one remains open.

Step 5 – Draw the consequence: In the end there is a statement for practice that expressly takes the lifeworld into account (Module 7). Sometimes it is: “We change something.” Equally legitimate is: “The study does not fit our question” or “We need more information from the service users.” This consequence often gives rise to the next practice question – the dashed arrow in Figure 2.

Which texts are suitable – and what reporting guidelines make easier for you

Four types of text come into question for a journal club, all of which you know from the earlier modules: a primary study, a review, a guideline or a national expert standard. Each type brings its own reading aid.

For randomised trials this is the reporting guideline CONSORT 2025. It replaces CONSORT 2010, comprises 30 checklist items and contains seven new ones, including data sharing, conflicts of interest, patient and public involvement, the recording of harms and the handling of missing data; there is also a new section on “open science” (Hopewell et al., 2025). For systematic reviews PRISMA 2020 takes on this role, a checklist of 27 items in seven sections (Page et al., 2021). Both show what should be reported transparently. They help you when reading because you can look specifically for the information – for example, how missing data were handled. But they are not instructions for carrying out a study, and PRISMA 2020 is expressly not an instrument for quality appraisal (Page et al., 2021).

For guidelines and national expert standards you read other things first: for AWMF guidelines the class (S1, S2k, S2e or S3) and with it the degree of systematic method in the development process (Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen Fachgesellschaften [AWMF], 2025); for national expert standards of the German Network for Quality Development in Nursing (Deutsches Netzwerk für Qualitätsentwicklung in der Pflege, DNQP) the structure of preamble, commentary and criteria and the way the standard was developed (Deutsches Netzwerk für Qualitätsentwicklung in der Pflege [DNQP], 2025b). The content of individual recommendations is not the subject of this course; for that you read the original each time.

PART 3 – Literature report: structuring, delimiting, referencing

Literature report and systematic review – two genres with different ambitions

In your studies the term “Literaturbericht” (literature report) is often used for a structured literature overview: a reasoned selection of texts on a question that is summarised, evaluated and classified. A systematic review goes further methodologically: the question and the inclusion criteria are fixed in advance – in the Cochrane Handbook, PICO serves as a memory aid for this (Higgins et al., 2024) –, the search is comprehensive and documented, and the report follows a standard such as PRISMA 2020 (Page et al., 2021). Reviews of this kind are produced, for example, by Cochrane (Cochrane Deutschland, n.d.) and by the Campbell Collaboration, which, according to Welch (2021), provides systematic reviews and evidence and gap maps for social policy decisions.

The last point is more than quibbling over words. If you call your work a “systematic review” without having searched systematically, fixed criteria in advance and documented everything, you raise expectations that you cannot meet. It is more honest to describe it as a literature report with openly stated limits. This course does not qualify you to produce systematic reviews or guidelines; it helps you to find, read and classify them.

Worked example: outlining a literature report step by step

We take the team question from Module 2 and treat it expressly only as a question: does life-story work help young people in a residential group, compared with the usual support, on an outcome fixed in advance? Whether and how it helps, we do not know here – this example makes no statement about effectiveness but only shows the structure. Figure 3 walks through the six steps of the outline.

The most important step for the quality of your text is the fifth. Many drafts are “parades of authors”: study 1 is described, then study 2, then study 3. A synthesis instead organises by questions: what do several texts report in agreement? Where do they contradict each other, and what might the reason be – population, comparison or outcome? The distinction from Part 2 between finding and interpretation helps here: what a text reports, you write down factually; what you derive from it, you mark as your own classification. Figure 3 also shows that the limits come before the conclusion: they determine how far the conclusion may reach.

Referencing in APA 7: transparency in small things

The literature report stands or falls with clean references. In the text, APA 7 works with the author–year system. There are two forms: in the parenthetical form the reference stands in brackets at the end of the sentence, for example “(Page et al., 2021)”; in the narrative form the name is part of the sentence, for example “Page et al. (2021) describe …”, and with two authors, German-language texts use “und” instead of the ampersand, for example “Müller und Schmidt (2020)” (fictional example); in English-language texts this is “Müller and Schmidt (2020)”. In the parenthetical form the ampersand is used for two authors: (Müller & Schmidt, 2020). With longer author lists you shorten with “et al.”.

Three special features come up again and again in the research for this course. First: organisations are authors. The first time, you give the full name with its abbreviation; after that the abbreviation is enough, for example “IQWiG (2025)”. Second: if two texts by the same author appear in the same year, you tell them apart with the suffixes a and b – hence the “b” in DNQP (2025b). Third: if the year is missing, you write “n.d.” (no date), as with NLM (n.d.). Press releases, such as those of the Federal Statistical Office (Destatis), appear in the text with the year, with letters if there are several texts from the same year, here Destatis (2026b), and in the reference list with the date. Everything you cite in the text appears in the reference list; everything in the reference list is used in the text. Paraphrase in your own words and still give the source – ideas too must be referenced.

PART 4 – The framework in October 2026: current standards, professional law and data

Current developments at a glance

Anyone working in a journal club or a literature report should always cite the current version of a standard. Several of the documents you have met in this course were renewed recently. The overview shows what has changed.

Two observations connect these innovations. First, transparency and participation come more into focus: a separate IQWiG chapter on involving those affected, a CONSORT item on patient and public involvement. Second, methods papers are revised regularly; a source without a version is therefore harder to classify. That is a practical reason to always state version, date and retrieval date in your literature report.

Data: nursing training and nursing degree programmes

Figures belong in a critical review when they are reported soberly and not over-interpreted. The Federal Statistical Office (Destatis, 2026b) reports for 2025 around 63,900 new training contracts for registered nurses (Pflegefachfrau/Pflegefachmann), 8% more than in the previous year. Figure 4 shows the development of new training contracts from 2022 to 2025.

Summary of the values: In 2022, 52,100 new training contracts were concluded, in 2023 there were 54,400, in 2024 59,400 and in 2025 around 63,900. The number rose from year to year. In addition, the nursing degree programme under the Nursing Professions Act (Pflegeberufegesetz) has been remunerated and funded since 2024: just under 800 people began it in 2025 (about 8% more than in the previous year), and at the end of 2025 there were 1,800 students on the nursing degree programme (Destatis, 2026b).

What the figures do not say is part of the review too. They concern nursing, not social work. They say nothing about team sizes, quality of care or need, and they do not demonstrate any link between training numbers and quality. An interpretation – expressly as such – would be: where different training routes come together in one team, a shared language about evidence makes cooperation easier. That is a thought for the journal club, not a finding.

PART 5 – Critical review: opportunities, limits and what remains open

Opportunities of evidence-based practice

A critical review begins with what speaks in favour of working in an evidence-based way. First, transparency: if you base your decision on comprehensible sources, you can explain it and answer follow-up questions. This is the basic idea of evidence-based medicine according to Sackett et al. (1996), whose definition you know from Module 1: explicitly using the best available external evidence and combining it with professional expertise. Second, justifiability: IQWiG (2025) describes an evidence-based approach as a distinction from mere opinions, supported by scientific methods that are as objective as possible. Third, protection against routines without a reason: the sentence “We’ve always done it this way” is not an argument, but neither is it taboo – it is an occasion to ask for reasons. The professional ethics of the DBSH (2024) link ethics and professionalism and assume that the practice competence of social work rests on scientifically secured knowledge.

Limits and areas of tension

The debate about evidence-based practice in German-language social work is not closed. Positions in favour emphasise randomised studies of effectiveness. Critical positions point to a technology deficit of social interaction (Technologiedefizit) – meaning, roughly, that help between people cannot be planned like a technical process in which a particular action reliably produces a particular effect. They also name the limited transferability of causal models and concerns about professional autonomy. Mediating positions lie in between. We deliberately do not attribute individual arguments to individual people here. Figure 5 sets the three lines side by side.

There are also limits that you know from the earlier modules: context (a study carried out under other conditions cannot simply be transferred), values (what counts as a good outcome is not only a question of data), participation (the perspective of those affected belongs in the decision – and increasingly in the research itself; the DNQP involves at least one representative of the people affected in developing its national expert standards (DNQP, 2025b)) and research gaps (for many practice questions there are no suitable studies). Naming a gap is itself a result of a literature report – not an admission of failure.

The critical review in four sentences

A helpful structure for the end of a paper or presentation consists of four sentences: (1) finding – what the evaluated sources report, (2) limit – where their informative value ends, (3) weighing up – how you set opportunities and limits against each other, (4) consequence – what follows for practice, research or further enquiry. Watch out for two traps: a correlation in the data does not automatically imply a cause, and a survey such as the one in Figure 4 does not support any statement about quality.

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.