Approaches to involvement
There is no evidence to show that any one way of involving people in a review is more or less impactful. Several different factors will influence the decision on the best approach for a specific systematic review. These factors may include the topic of the review, time available, resources available, and expertise of the review team.
Two different approaches to involvement have been used for other systematic reviews:
- Continuous involvement – people are involved ‘throughout’ the review process, perhaps as a member of the review author team or an advisory group.
- One-time involvement – people are involved at a specific stage in a review in order to complete a specific task or address a specific aim. For example, a group of people might be involved in discussing and reaching consensus on the question for a review, or people might be involved in order to contribute to the writing a plain language summary.
The aims of the PPI will help determine which approach might be best for a specific review. For example, if a key aim is to ensure that the outcomes included in the review reflect those that matter most to people affected by a particular health condition, then a one-time involvement approach may be more advantageous. This could enable a group of people to come together and reach consensus on the outcomes for the review. However, if the aim of involving people is to provide general oversight to the review process and ensure that all stages of the review process consider the views of patients and the public, then continuous involvement may be more advantageous. Some systematic reviews combine both approaches. For example, they may have PPI input on an advisory group throughout the review process, and then also plan 1 or more one-time events to get additional input into key stages of the review.
Levels of involvement
Involvement of people in a systematic review can be considered as a continuum, from more involvement and control, to less involvement and control. Pollock at al. (2019) explored the different tasks and activities in which patients and the public were involved in a range of systematic reviews. Using an iterative process, they developed a new taxonomy relating to the actions, responsibilities and tasks of those involved, called the ‘ACTIVE continuum of involvement’. It describes people as leading, controlling, influencing, contributing or receiving (see table 3).
Table 3 The ACTIVE continuum of involvement (from Pollock et al. 2019)
Level of involvement | Tasks |
Leading: Initiating the review; lead responsibility for carrying out and completion of review. | Tasks will include authorship of a review, and may include any activities associated with review completion, including key decisions relating to the methods and execution of the review. |
Controlling: Working in partnership with researchers, with varying degrees of control or influence over the review process. Making decisions, controlling, or both, 1 or more aspects of the review process, in collaboration with or under the guidance of the review authors. | Tasks may include defining outcomes of interest, inclusion criteria, key messages arising from review findings and writing a plain language summary. In completing tasks people have control over final decisions, such as application of inclusion criteria, categorisation of interventions, or recommendations for clinical practice. |
Influencing: Stating, commenting, advising, ranking, voting, prioritising, or reaching consensus. Providing data or information that should directly influence the review process, but without direct control over decisions or aspects of the review process. | Tasks may include assisting with review tasks, such as hand searching, screening, data extraction and assessment of risk of bias, possibly in a co-reviewer role. Tasks may include peer review, such as commenting on a protocol, systematic review or plain language summary. |
Contributing: Providing views, thoughts, feedback, opinions or experiences. Providing data or information that may indirectly influence the review process. People may be participants in a research study (for example, focus groups or interviews). | Tasks may include sharing views or opinions, for example, within a focus group or interview. May include ranking, voting or prioritising as participants in a research study (for example, in a Delphi study). |
Receiving: Receiving information about the systematic review, or results of the review. | Tasks may include attending events or reading or listening to information about the review. Although the results of a review may be discussed, these discussions do not influence the review process in any way. |
Although the level of involvement of people in a systematic review can be seen as a continuum, there is no evidence of a hierarchical association between level, impact, benefit or success of involvement. Indeed, current evidence and opinion suggests that different levels and methods of involvement may be useful at different stages in a systematic review. What is important is to consider the level of PPI involvement, and their level of control or influence over the process. Decisions about the level of control that people will have at various stages in the review process should be stated in advance, ideally within the systematic review protocol.
Format of involvement
Format of involvement means the ways in which people interact and communicate, such as through face-to-face meetings, events or workshops, individual or group telephone or video-calls, or email and written communication. The format of PPI in a systematic review will depend on several factors. These factors include (but are not limited to) the aim of involvement, the people who are being involved, and the resources and time available for this. In an exploration of the format of involvement adopted in a range of systematic reviews, Pollock et al. (2018) found that direct face-to-face interaction was the most common approach, and that this might comprise a small meeting, a larger workshop or public event, or a combination of these. In most cases, between 1 and 4 meetings or events were held throughout a review, although as many as 20 meetings had been held. Meetings varied in length from 1 hour to half a day. A small number of systematic reviews used electronic or remote methods to involve people. Most commonly, this was an electronic Delphi or survey method, usually involving 2 or 3 rounds of voting.
Research methods and processes
A range of different ways have been used when involving people in a systematic review. Often these methods and processes involve different ways of sharing thoughts and ideas, such as group discussions or written feedback. Several formal research methods have also been used when involving people. Adopting a formal research method or process can be useful when there is a clearly identified role, or aim, for the people involved. For example, the aim might be to reach consensus on the outcomes of relevance to the review, or to agree a way to synthesise the evidence so that it is accessible and understandable.
Formal research methods and processes that have been used when involving people in systematic reviews include:
- Participatory research approaches: Include ‘action research’ and ‘participatory action research’ and are usually considered as ‘approaches’ to research, rather than methods. These approaches integrate PPI with qualitative research, with a joint process of knowledge production by researchers and patients or the public. Participatory research approaches have key tenets: a democratic impulse; iterative data collection and analysis, and simultaneous contributions to science, improvement and change.
Box 1 Example of a participatory research approach
- Consensus decision-making techniques: Include using techniques for voting (that is, to make decisions about the review) and ranking (for example, to prioritise domains, such as outcomes, within a review). It also includes the nominal group technique, which involves a structured discussion and rounds of voting to reach consensus on a specific problem or issue, and the Delphi method, which involves several rounds of questionnaires or surveys to achieve consensus.
Box 2 Example of a consensus decision-making process
For an update of a Cochrane review relating to physiotherapy for people who had a stroke, Pollock et al. (2014, 2015) formed a stakeholder group comprising physiotherapists, stroke survivors and carers. During a series of 3 meetings, stakeholders made several decisions relating to the review. Decisions were made using the nominal group technique. In each case, the stakeholder group members first discussed a topic or statement for an agreed amount of time. Then each stakeholder group member individually ranked their agreement with that topic or statement and noted their reasons for this. The ‘voting’ sheets were anonymous, but were then collected and counted in front of the group members in order to see whether or not there was consensus on a topic. Further rounds of discussion and voting took place when needed. |
- Group process: Often the process of involving people within a systematic review entails a group meeting, which may be called a meeting, workshop or conference. These meetings commonly involve discussion and debate, perhaps supplemented with formal methods such as consensus decision-making techniques. The content and processes within these group meetings are often poorly reported. However, evidence suggests that these meetings do often combine careful planning and use of techniques known to enhance the group process. The planning and approach to running group meetings provides a way of addressing many of the general issues identified as important to involvement, such as effective communication, clarity, expectations, respect and trust.
Box 3 Resource on group process
- Qualitative research methods: These methods, such as interviews or focus groups, have been used to elicit views and opinions of patients and the public in relation to systematic reviews. The purpose has most commonly been to ‘contextualise’ the findings of a systematic review to a particular population or area. These data have then been analysed using methods for analysis of qualitative data, such as thematic analysis (Bunn et al. 2015, Martin et al. 2015). In such cases, the involvement of people has occurred after completion of the systematic review. However, it could be argued that the involvement relates to the final stages of a systematic review process (such as dissemination and translation of evidence into practice). The level of involvement is one of ‘contribution’ and, in these examples, the people involved could arguably be described as ‘participants’ in research.