What is evidence-based practice?

Social workers need to understand and use research to provide effective support, but there are debates about how evidence should be used

Photo: aquarius83men/Fotolia

In a Community Care Inform Children guide, David Wilkins covers how to use evidence to inform your practice, and different ways to find relevant information and stay up-to-date with research. In the following excerpt from it, he goes through what evidence-based practice is, and some of the key ideas and controversies around it. Inform subscribers can read the full, in-depth guide.

Social workers need to understand and use research in practice if they are to provide effective help. The Professional Capabilities Framework (PCF) says that social workers should “make use of research to inform practice” and the Knowledge and Skills Statement for child and family practitioners says they should “make use of the best evidence from research to… support families and protect children”. This all sounds very simple – but evidence-based practice is more complicated than that.

Evidence-based practice (EBP)

Lindsay (2007) explains EBP in simple terms:

it involves using the best evidence you have about the most effective care of individuals, using it with the person’s best interests in mind, to the best of your ability and in such a way that it is clear to others that you are doing it.”

Put like this, it is hard to see how anyone could dispute the value of EBP – but there are many who do and for understandable reasons, even if one might in the end disagree. Most of us would feel uneasy if our GP suggested a treatment based not on the best available evidence but because of their ‘gut feeling’ or an anecdote they were told by a more senior colleague. It seems equally obvious that social workers should have a good working knowledge of the issues they encounter and an evidence-informed understanding of the best ways to help.

It is unlikely that anyone really believes otherwise. However, there are debates and controversies about the nature of evidence and how it should be used. Those critical of EBP often argue it is based on a medical model of evidence and that applying this approach in social work creates more problems than it solves. Put simply, medical experiments establish whether a medicine works for a specific illness. But are social work interventions comparable to medicines? And are the issues encountered by social workers comparable with illnesses? The answer to both these questions must surely be ‘no’.

Socially defined issues

Firstly, the issues social workers deal with tend to be socially defined. Deciding a child is being abused is not the same as saying they have measles. The latter can be determined via an objective blood test, which gives the same result irrespective of who administers the test. The former involves not only defining ‘abuse’ but determining what should be done about it and different results are not only possibly but likely depending on who make the judgement and when. A critical view of these issues is essential, and this makes interpreting ‘evidence’ – and deciding what counts as ‘evidence’ – more complicated.

For instance, both cholera and child abuse are more common in poor families; cholera, because of a lack of access to clean water and good hygiene among poorer people in poorer countries. The link between child abuse and poverty is much more complicated. It may be because the stress and difficulty of living in poverty makes a minority of parents more likely to abuse their children but it could also be definitional, because we are more likely to label the parenting practices of poor families as ‘child abuse’ than those of more affluent families. Accepting ‘common sense’ definitions of abuse and applying ‘evidence-informed’ interventions without critical thought is not good social work. EBP cannot provide simple solutions to complex problems for there are none (Pitts, 2001).

Social workers need to understand the contribution of social causes to many of the issues they deal with. Experimental evidence, on the other hand, tends to individualise problems. For instance, The Family Nurse Partnership is targeted at young pregnant woman, many of whom are living in poverty. The intervention appears helpful for many (although not in every case; see below). However, there remains a fundamental mismatch between the presenting issue and the response. Poverty is not an individual problem. Yet, gathering evidence about ‘what works’ in response to social problems is difficult using experimental approaches (although not impossible, as recent experiments involving ‘basic income’ may yet demonstrate). Experimental methods often shift the focus on to individuals and ignore the social causes of problems. This should be anathema to social work, where we aim not only to improve individual welfare but also for social justice.

Social, not medical, interventions

Social workers work in situations in which social, psychological and biological factors interact in complex ways. Social interventions are much more complicated than medical ones. Although the Family Nurse Partnership does seem to help pregnant women living in poverty in the USA, the same is apparently not true for pregnant women living in poverty in the UK. This might be because universal services in the UK are more readily available than the USA. Whatever the explanation, these different results for the same intervention in different locations highlights the importance of critically examining not simply ‘the intervention’ but the social and economic context in which it was administered and for whom (Pawson and Tilley, 1997).

EBP can risk giving the impression – or even creating a situation – in which social workers ‘apply’ interventions to people, rather than working with people. This would run counter to social work values. This does not mean that EBP is irrelevant for social work but it does indicate the need for caution, to ensure we still work in partnership with people, rather than assuming ‘we know best’. Using EBP requires highly professional individuals with the skills, knowledge and institutional support to use evidence appropriately.

Finally, a more pragmatic criticism of EBP can be made in relation to the very limited evidence base for social work. While we have a plethora of high-quality descriptive research, there is little strong evidence for the effectiveness of specific ways of working. As a result, social workers often need to ‘borrow’ evidence from other settings and translate it into their own context.


Lindsay, B (2007)
Understanding research and evidence-based practice
Exeter: Reflect Press

Pawson, R and Tilley, N (1997)
Realistic Evaluation
London: Sage

Pitts, J (2001)
Korrectional karaoke: New Labour and the zombification of youth justice
Youth Justice, Volume 1, Issue 2, pp3-16

More from Community Care

8 Responses to What is evidence-based practice?

  1. Winmor November 10, 2017 at 2:01 pm #

    Evidenced based research is important as ultimately there has to be some scrutiny of ideas and interventions that have some chance of relevance.

    True that social science is more messy that physical science, but to paraphrase, evidence based practice is the worst system aside from all the other options available

  2. Mark Wilberforce November 10, 2017 at 10:51 pm #

    Enjoyable article. I hope that one of the functions of EBP is that it can be used to help service users with decision-making about their support, with social workers able to share and discuss any knowledge or understanding that evidence can provide.

  3. Nora Duckett November 11, 2017 at 1:18 pm #

    Cogent and really helpful article. Will definitely use it in teaching to think more deeply about use of ‘evidence’. Stephen Webb’s (2001) article (https://academic.oup.com/bjsw/article-abstract/31/1/57/1654601?redirectedFrom=fulltext ) is really useful too. It helps see how complex processes turned into ‘facts’ can undermine professional judgement. Also a preoccupation with positivistic methods and determinism encourages social workers to become mechanistic eschewing critical reflection for technical rationality.

    • Pats November 15, 2017 at 7:28 pm #

      What! Plain language please.

  4. Romeo2001 November 12, 2017 at 2:30 pm #

    I understand the whole medicalisation of problems being anathema to social work but is the general tone of the article that the social worker is responsible for alleviating structural issues such as poverty either or fair? When was the last time any social worker achieved such ambitious aims?

    • Romeo2001 November 12, 2017 at 2:31 pm #

      *Either realistic or fair?

  5. Peter Endersby November 14, 2017 at 12:57 pm #

    As a teacher we are supposed to teach using ‘evidence’ but we don’t. You all weren’t when you were at school. The articles reference to medical is symptomatic of social works preoccupation with trying to be recognised. You guys have a serious chip on your shoulders. What about mental health? Simple?

    Teaching, Health and Social Work are dictated by policy and above all money. There are interventions that could should be employed across all but budgets and policy say no. Not evidence. To ignore this is folly or just wilful ignorance.

  6. Sid Fensome November 15, 2017 at 8:56 pm #

    Canada here. I find myself agreeing mostly with Winmor (above). I think the article’s author somewhat misunderstands the nature of evidence itself and what EBP is intended to do: focus our energies on approaches shown to have the most success with our client cohort (versus those with little success). Of course, one shouldn’t blindly apply an approach to client A (in the UK) merely because it worked with client B (in the U.S.) as having a clear understanding of the differences between the two (and the situational/cultural differences) is part of the evidence one is called to use to make good decisions. The same happens in good medicine: the physician who prescribes a common antibiotic to a patient without considering the possibility of allergic reaction or deadly side effect is also not using evidence in their practice. I think as Social Workers we have often not spent enough time learning how to recognize good research and discern where it can best be applied (and more importantly where it doesn’t apply).