2.1. Guideline development group

A WHO guideline steering group, led by the Department of Ageing and Life Course, was established with representatives from relevant WHO departments and programmes with an interest in the provision of scientific advice regarding older people. The guideline steering group provided overall supervision of the guideline development process. Two additional groups were formed: a guideline development group (GDG) and an external review group.

The GDG included a panel of academics and clinicians with multidisciplinary expertise on the conditions covered by the guidelines, plus geriatricians/specialist doctors in the care of older people. Consideration was given to the balance of gender and of geographically diverse representation (see Annex 1).

Potential members of the GDG were selected on the basis of their contribution to the area, as well as on the need for regional and area-of-expertise diversity. As a respected researcher in the field, the chair was selected for his extensive experience of guideline development methodology, and his participation in other guideline development groups. Each potential GDG member was asked to complete the WHO declaration-of-interest form. The personal statements were reviewed by the steering group.

2.2. Declarations of interest and management of conflicts of interest

All GDG members, peer reviewers and systematic review team members were requested to complete the declaration-of-interest form prior to the evidence-review process for guideline development. Invitations to participate in the GDG meeting were sent only after the declarations of interest had been reviewed and approved. These were reviewed by the responsible technical officer at WHO – in this case the director of the Department of Ageing and Life Course – and, when necessary, legal counsel. The group composition was finalized after this process. Annex 2 gives a summary of relevant declarations of interest.

The declarations were once more assessed for potential conflicts before the meeting in Geneva. The members who were involved in conducting either primary research or systematic reviews that would relate to the recommendations did not participate in the formulation of any recommendations themselves. The majority of the members had no major conflicts of interest. Minor conflicts of interest, of which there were two cases, were managed individually by restricting participation at relevant stages of the GDG meeting. All decisions were documented (see Annex 2).

2.3. Identifying, appraising and synthesizing available evidence

The scope of the guidelines and questions (Annex 3) were defined. A total of nine PICO (population, intervention, comparison group, outcomes) questions (23) were formulated by the GDG and steering group. Outcomes were rated by GDG members and external experts according to the importance of each outcome from the perspectives of older people and service providers, as not important (rated 1–3), important (4–6), or critical (7–9). Outcomes rated as critical were selected for inclusion into the PICO analysis. The GDG engaged in regular communications by email and discussions by teleconference.

When formulating the scoping questions and conducting the reviews, the focus was on evidence that applied specifically to older people who were frail or care-dependent or had priority conditions, and on interventions that could be used by non-specialist health workers in community settings or primary health care. The steps that were taken for evidence retrieval, assessment and synthesis are summarized in Annex 4. Further detail on the review methods and available evidence is summarized in the evidence profiles supporting these guidelines. The evidence profiles used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology (24) followed by the WHO guidelines handbook, and the profiles are available at the WHO web pages for ICOPE (http://www.who.int/ageing/health-systems/icope). The search strategy and methods of quality assessment and appraisal are included in each profile. This GRADE methodology for evidence-based medicine was also used to formulate the recommendations on the interventions, by providing a rating of the overall quality of evidence arising from each systematic review. All of the recommendations were based on direct evidence and analysis of quantitative data.

2.4. Consensus decision-making during the guideline development group meeting

The GDG met at the WHO headquarters in Geneva, Switzerland, 24–26 November 2015. The evidence reviews had been sent out in advance and were presented in a summarized version during the meeting. The GDG members discussed the evidence, clarified any points and interpreted the findings, to develop recommendations based on the draft prepared by the WHO Secretariat. The GDG then proceeded with deliberations and considered the relevance of the recommendations for older people based on:

• the balance of benefit and harm of each intervention;
• values and preferences of older people;
• costs and resource use;
• acceptability of the intervention to health care providers in low- and middle-income countries;
• feasibility of implementation;
• impact on equity and human rights.

To evaluate the values and preferences of older people and the acceptability of proposed interventions to health workers, no formal surveys were carried out; the discussion and assessment of these domains instead relied on the combined expertise and observations of the GDG members. Similarly, no formal cost-effectiveness studies were undertaken; again the GDG members informed the assessments of resource constraints based on their knowledge and experience.

Taking into account all of the above considerations, it was agreed that if a recommendation would be of general benefit, it would be rated as strong. If, however, there were caveats about its benefits in different contexts, and/or the quality of evidence was less robust, the recommendation would be rated as conditional. In the event of a disagreement, the chair and the methodologist would ascertain whether the dispute was related to the interpretation of the data or to the way that the recommendation was formulated. If a consensus agreement was not reached, the GDG members agreed to a simple majority vote (51%/49%), in which voting for this decision was by raised hands. GDG members reserved the right to have any objections recorded. Excluded from voting were any WHO staff members present at the meeting and any technical experts involved in the collection and review of the evidence.

The GDG reached a consensus agreement on the 13 recommendations and ratings given in this document. At the voting stage for recommendations on cognitive training and respite care, these further two were not supported due to insufficient evidence.

2.5. Document preparation and peer review

In addition to the GDG members, four peer reviewers provided expert input from specialized fields – psychiatry, nutrition, physical therapy and geriatrics. A preliminary version of these guidelines and the evidence profiles prepared by WHO staff and the GDG were circulated to the peer reviewers and the WHO steering group. All inputs and remarks from reviewers were discussed and agreed with the GDG by email. Additionally, peer reviewers were asked to rate the quality of the guidelines using a slightly modified version of the tool, Appraisal of Guidelines for Research and Evaluation (AGREE II). The original AGREE II tool lists 23 key items in the following domains: scope and purpose, stakeholder involvement, rigour of development, clarity of presentation, applicability, and editorial independence (25). The reviewers’ total AGREE II scores ranged from 22 to 154, and the average was 122.2.