Economic evaluations  of strategies targeting pre-diagnosis dementia  populations: Protocol for a systematic review

Registration

This systematic review protocol follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) guidelines¹². The PRISMA-P checklist is provided as part of the Extended Data¹³. The review was registered with the International Prospective Register of Systematic Reviews (PROSPERO CRD42024521521). The results will be reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines¹⁴.

Eligibility criteria

Population: The target population includes individuals at any stage prior to having a formal diagnosis of dementia, such as the general population, people at risk of dementia, those with mild cognitive impairment (MCI), caregivers, and patient-caregiver dyads.

Intervention: All types of strategies that target our pre-specified pre-diagnosis dementia population will be included. As a result, four categories of intervention will be included: (1) identification, (2) pharmacotherapy, (3) non-pharmacotherapy, and (4) management strategies at home, in the community, and in long-term care.

Articles will only be included if the strategies under investigation specifically target the aforementioned populations.

Comparator: The comparison group will contain any of the following: (1) usual care or standard of care or conventional interventions, (2) placebo, (3) doing nothing or without intervention, or (4) any treatment aimed at a pre-defined population.

Outcomes: Our primary outcomes will be cost-effectiveness assessments, such as incremental cost-effectiveness ratio (ICER), net monetary benefit (NMB), net health benefit (NHB), costs, incremental cost, effectiveness outcomes such as quality-adjusted life-year (QALY), and incremental effectiveness outcomes (e.g., incremental QALY). Additional outcomes will be included, such as clinical outcomes, which were used to indicate cost-effectiveness, and data related to resource use and utility.

Study design: Full EEs will be selected for this review. Full EEs were pre-defined according to Drummond’s classification scheme¹⁵. Accordingly, this review will include (1) Cost-Effectiveness Analysis (CEA), cost-utility analysis (CUA), and Cost-Benefit Analysis (CBA).

Settings: The review includes studies conducted in all countries globally.

Exclusion criteria

Articles will be excluded from the review if they (1) do not target pre-diagnosis dementia, or populations with unclear boundaries surrounding diagnosis (i.e., some pre- and some post-diagnosis) or individuals who may only exhibit MCI symptomatology but where neuropathology of dementia has been established (i.e., MCI due to Alzheimer’s Disease (AD)). This is because the criteria for MCI due to AD must be confirmed by biomarkers to determine the likelihood that the syndrome results from the underlying pathophysiology of AD¹⁶. The review will also exclude (2) do not include any comparator, (3) do not report economic outcomes, (4) partial EEs such as cost-of-illness analysis, efficacy effectiveness evaluation, cost-outcome description, (5) reviews, editorials, letters, commentaries, viewpoint/perspective, conference abstracts, (6) proceedings, protocols, case studies, and (7) are not reported in English.

Search strategy

Potentially relevant literature was identified by systematically searching six electronic databases: Embase, PubMed, Econlit, Web of Science, Cumulative Index to Nursing and Allied Health (CINAHL) and The NHS Economic Evaluation Database (NHS EED). The details of the search terms are provided in Extended Data¹³.

Study selection

References exported from the six databases will be transferred to EndNote, deduplicated, and then imported into the Covidence (https://www.covidence.org).

The study selection process will be conducted independently by two reviewers. Any disagreement will be resolved by a third reviewer. During the screening process, all titles and abstracts will be independently scanned for relevance, and the full texts of eligible articles will be assessed according to pre-defined inclusion and exclusion criteria.

Data extraction

The data will be extracted into a pre-defined template using Covidence. This process will be conducted independently by two reviewers, and any disagreements will be resolved by a third reviewer.

The extracted items will include:

Study details: Title, author, publication information, country, funding source, study design, population, intervention, comparator, and settings.

Methodology: Type of EEs, type of model (if applicable), time horizon, perspective, type of sensitivity analysis, currency, price, discount rate, method to measure and value costs, and method to measure and value outcomes.

Results: Total costs, incremental cost, total outcomes, incremental outcomes, ICER, NMB, and authors’ conclusions on the cost-effectiveness of the strategies.

Data synthesis

Pre-diagnosis dementia strategies will be categorized into four groups: identification, pharmacological and non-pharmacological interventions, and management.

Qualitative synthesis will report the number of EE studies in each subgroup. Where studies report both total costs and outcomes expressed in QALYs, results will be adjusted and will be visually presented on a cost-effectiveness plane to facilitate decision-making.

According to the available data and the variation in study quality, method, or design, narrative synthesis provides guidance on the interpretation and reliability of estimates to inform allocation decisions. The extracted data will be presented in a summary table of the key extracted items. Figures will be used to portray the production of studies over time and geographic location.

Quality appraisal

The Consolidated Health Economic Evaluation Reporting Standards (CHEERS) 2022 checklist was initially piloted to evaluate the reporting quality¹⁷. However, discrepancies arose during this process owing to varying interpretations by the two reviewers, leading to inconsistencies across items. To address these issues, several meetings were held with senior reviewers (DT and SS), resulting in a modified CHEERS 2022 framework comprising 67 sub-items, categorized as follows: title (two items), abstract (four items), introduction (three items), methods (38 items), results (11 items), discussion (six items), and other relevant information (two items). This adapted framework will be applied to assess reporting quality. Two reviewers will independently assess the reporting quality of the included articles using this framework, with a third reviewer resolving disagreements. All processes will be performed using Covidence.

Results of the reporting quality of included articles will be summarised using the traffic light system (i.e., green: fully satisfied, amber: partially satisfied, or red: where the standard was not achieved). Scoring of individual studies, in line with the framework, will be summarised in the summaries of each paper.