Article Text
Abstract
Introduction Late-life depression (LLD) is a global public health issue, often accompanied by cognitive impairments that can exacerbate the severity of depression and impair social functioning. Despite being a well-established treatment for LLD, the suitability of problem-solving therapy (PST) for individuals with LLD and varying degrees of cognitive impairments warrants further investigation. This paper presents the protocol for a systematic review and meta-analysis of randomised controlled trials (RCTs) aimed at evaluating the effectiveness and acceptability of PST for this specific demographic.
Methods/analysis Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we will conduct a systematic review to synthesise existing research on PST for individuals in this demographic. We will comprehensively search multiple databases and sources, including PubMed, EMBASE, the Cochrane Database and APA PsycNET from inception to October 2023, without language, publication year or type restrictions. Relevant studies will be manually screened from the references. Only RCTs involving PST for LLD will be included. The primary efficacy outcome will be the standardised mean difference in total scores on continuous depression severity scales across different comparison arms. Data extraction will be conducted independently by two reviewers (CH and J-JW), and methodological rigour will be assessed using the Cochrane Risk of Bias assessment tool. Subgroup and sensitivity analyses will be performed to investigate the impact of concomitant cognitive impairments and to evaluate the robustness of the findings.
Ethics and dissemination The meta-analysis project is expected to be ethically unproblematic and does not require approval from a research ethics committee. The results of this study will be shared through articles in scholarly peer-reviewed journals and presentations in various formats, both print and digital.
PROSPERO registration number CRD42023473782
- Depression & mood disorders
- Aged
- Delirium & cognitive disorders
- PSYCHIATRY
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
This meta-analysis presents the most comprehensive review to date of randomised controlled trials on problem-solving therapy (PST) for late-life depression with potential cognitive impairment.
We will conduct a thorough evaluation of the effectiveness and acceptability of PST, incorporating subgroup and sensitivity analyses to address pertinent clinical inquiries within this particular demographic.
The potential limitations include the variability in study designs and the quality of the included trials, which may impact the overall findings.
Introduction
Depression is a leading cause of global disability, particularly among the elderly, where the prevalence and incidence of severe depression increase twofold after the age of 70–85 years. Clinically significant depression in the elderly exacerbates various physical illnesses and contributes to the onset of disabilities. Research indicates that 9.3% to 23.0% of individuals with one or more chronic physical illnesses also experience depression, a significantly higher proportion compared with those without chronic diseases. When accounting for socioeconomic factors and overall health status, depression has a more pronounced negative impact on average health scores than other chronic diseases.
Late-life depression (LLD) refers to depressive episodes in individuals aged 60 years and above. The internationally accepted classifications include early-onset LLD (EOD), where depressive episodes first occur before the age of 60 years but continue or relapse in later life; late-onset depression (LOD), where depressive episodes first occur at the age of 60 years or older and very late-onset depression (VLOD), where depressive episodes first occur at the age of 75 years or older.1 With ageing, physical illnesses and specific psychological attributes prevalent in this group, increasing evidence suggests that LLD often accompanies cognitive impairment, a condition that may persist post-depression treatment.2–4 It is hypothesised that depression featuring prominent negative emotional symptoms is frequently associated with cognitive deficits, potentially due to dysfunction of the prefrontal cortex in emotional processing.5 An estimated one-third of depression patients experience some level of cognitive decline, which may worsen dramatically in the elderly.6 7 Consequently, the rising incidence of LLD has led to a current research focus on developing new therapies that can address both depression and cognitive comorbidities simultaneously.8 9
Antidepressants, specifically selective serotonin reuptake inhibitors,10 are pivotal in treating LLD. Despite their proven efficacy over placebos in LLD patients, these medications are consistently accompanied by intolerable side effects and a reduced response rate in the elderly.11 Additionally, cognitive complaints in LLD patients often indicate resistance to conventional antidepressant drugs. Conventional psychotherapies, including cognitive-behavioural therapy (CBT),12–14 interpersonal psychotherapy15 16 or behavioural activation treatments,17 18 have demonstrated promising outcomes for individuals with depression. Nonetheless, there is inconsistent evidence regarding their efficacy for patients with LLD. The successful implementation of psychotherapies is known to be significantly influenced by cultural factors and education level, particularly by the presence of comorbid cognitive impairments. Given the high prevalence of cognitive impairments in LLD, providing psychotherapy to these specific demographics is always challenging.
Problem-solving therapy (PST)19 is a practical approach focusing on resolving issues and taking action. It has demonstrated efficacy in treating LLD in the elderly, especially for patients with insight and normal cognitive functioning. However, there is a rising concern regarding the convergence of depression and cognitive impairment in older individuals, which is referred to as depression-executive dysfunction syndrome.20 21 Moreover, research has shown that PST not only significantly improves depressive symptoms but also produces positive therapeutic effects by enhancing skill developments, such as generating alternative solutions and making decisions.22–24 Therefore, as a practice-oriented approach, PST has the potential to improve depression and depression-related cognitive impairments, including executive function. Thus, further research is needed to examine the potential impact of PST and its interaction with cognition in the treatment of LLD.
The objective of this meta-analysis is to assess the efficacy and acceptability of PST for LLD, whether or not cognitive impairment is present. Subgroup and sensitive analyses will be performed to explore heterogeneity and identify factors contributing to variations in outcomes. The successful implementation of this meta-analysis will present the most comprehensive evidence and is expected to offer valuable insights to guide clinical decision-making regarding the use of PST in treating LLD with potential cognitive complaints.
Methods
Criteria for included studies
Types of studies
We will include pertinent randomised controlled trials (RCTs) and cluster-randomised trials. For studies employing a crossover design, only data from the initial randomisation phase will be analysed. However, quasi-randomised trials that allocate participants based on criteria like alternating days of the week25 will be excluded.
Types of participants
The original studies should encompass all subgroups of LLD participants, including those with early-onset LLD, late-onset LLD and very late-onset LLD of any gender. These participants must exhibit clinically significant depression meeting either of two criteria: (1) current diagnosis of major depression according to the International Classification of Diseases or the Diagnostic and Statistical Manual of Mental Disorders criteria or (2) clinically significant depression as indicated by depression rating scale threshold values. Participants with comorbid organic diseases are not excluded, while those diagnosed with schizophrenia or bipolar disorder are not included.
Types of interventions
The study will incorporate RCTs that compare interventions involving PST or its adaptively modified forms for LOD. The interventions will not exceed 12 weeks in duration and can be administered in person or online, irrespective of whether antidepressants are used as the primary treatment. The primary objective is to improve the patients’ problem-solving skills, enhance their ability to cope with stress and challenges and ultimately elevate their quality of life and decision-making capabilities.
Types of control groups
The eligible control group treatments encompass waiting lists, standard care, placebos and medication, as well as other psychotherapies like CBT, supportive therapy and reminiscence therapy.
Primary outcome
The primary efficacy outcome will be the standardised mean difference (SMD) in total scores on continuous depression severity scales across all comparison arms. This will first be assessed by the Hamilton Depression Rating Scale (HDRS),26 and if those data are not available, we will use the Beck Depression Inventory (BDI)27 or the Montgomery-Asberg Depression Rating Scale,28 followed by other depressive rating scales.
Secondary outcomes
The response rate refers to the percentage of patients showing significant improvement or a positive response to treatment, typically indicated by a 50% reduction in depression scores at the end of treatment.
The remission rate represents the percentage of patients whose symptoms either disappear or decrease to a clinically asymptomatic level, such as an HDRS score of 7 or less, a BDI score of 10 or less.
Improvement in cognitive function is related to the overall change in scores on continuous cognitive function assessment scales, such as the Mini-Mental State Examination29 or the WHO Disability Assessment Schedule II.30
The acceptability rate is defined by the all-cause discontinuation rate, indicating the percentage of patients discontinuing treatment for any reason and serving as an indicator of treatment acceptability.
Search strategy
Relevant studies will be systematically retrieved via the following electronic databases: PubMed, EMBASE, the Cochrane database and APA PsycNET. The deadline for the search is from the date of establishment to October 2023. There are no language limitations, and the study will encompass all published RCTs regarding PST treatment for LLD. Information regarding the search strategy is available in online supplemental etable 1 of the Supplementary materials. Additionally, we will reach out to corresponding authors to complete information for studies with incomplete data.
Supplemental material
Study selection
Two independent reviewers (CH and J-JW) will employ the designated search strategy to screen all retrieved studies according to predetermined inclusion criteria. Studies considered potentially suitable by reviewers will undergo full-text assessment and verification against the inclusion criteria. If there is a disagreement between the two reviewers on the selection of studies, a third reviewer (XL) will be consulted to resolve the disagreement.
Quality assessment
The included studies will be independently assessed by two reviewers using the Cochrane Risk of Bias assessment tool,31 categorising the risk of bias as ‘low’, ‘unclear’ or ‘high’. The assessment criteria encompass random sequence generation, allocation concealment, blinding of participants, treatment providers or outcome assessors, incomplete outcome data, selective reporting of results and other biases. Disagreements during this process will be resolved through discussion or consultation with a third reviewer if necessary. Study quality will be visualised using the risk of bias graphs generated by the RevMan software.32
Data management
The literature search results will be uploaded to EndNote, a reference management software that facilitates collaboration among reviewers during the study selection process. The team will screen and tag the literature based on inclusion and exclusion criteria. The EndNote’s duplicate detection feature will help identify and eliminate duplicate publications, reducing the risk of bias. Before the formal screening process, a calibration exercise will be conducted to pilot and refine the screening questions. Additionally, training will be provided to review team members who are not familiar with the EndNote software and the content area before the review begins.
Data extraction
Data will be extracted from each study by two reviewers using a predesigned standardised data extraction form. This includes basic information on the study, study design, participant characteristics (such as patient type, age, recruitment method and diagnostic criteria for depression), intervention measures (intervention course, duration and mode of intervention) and outcome measurements (changes in depression scale scores, cognitive measurement outcomes and follow-up results). To ensure accuracy and consistency, two researchers will independently perform data extraction and then compare their results to ascertain consistency. If there are discrepancies in the data extraction results of the two researchers, discussions will be held to resolve the differences, and if necessary the authors of the original study may be contacted to get the data clarified or obtain missing information.
Statistical analysis
Data analysis in this study will be executed using Review Manager (V.5.2).32 To evaluate the statistical heterogeneity among studies, I2 statistics will be computed. Depending on the I2 values, different models will be employed: a random-effects model for I2>50% and a fixed-effects model for I2<50%.33 The primary focus will be on the SMD for continuous outcomes across diverse comparison groups. In instances where studies report data from multiple control groups, we will integrate these datasets using established statistical methods. This integration approach will also be applied to multiple experimental groups compared to a single control group.34 The occurrence of positive effect sizes will indicate that the group treated with PST shows improved outcomes in LLD measures related to the control group. To assess publication bias, a funnel plot analysis will be conducted.35 Additionally, a meta-regression analysis will be performed to scrutinise the impact of factors like sponsorship or publication year on the outcome estimates.
Subgroup analysis
We anticipate conducting the following subgroup analyses: (1) participants subgroups (eg, early-onset LLD, LOD and VLOD); economic status of countries (patients from high-income and low-income countries); (2) method of recruitment; (3) form of treatment and (4) duration of treatment.
Sensitivity analyses
The following sensitivity analyses will be conducted: (1) restricting the study to those with higher quality; (2) limiting the research to studies employing standardised criteria for major depression and (3) excluding studies involving participants diagnosed with schizophrenia or bipolar affective disorder.
Meta-bias
To assess the presence of meta-biases such as publication bias and selective reporting within studies, we will implement several strategies. First, we will check whether the protocols of RCTs were published before the recruitment of patients commenced. For studies published after 1 July 2005, we will screen the Clinical Trial Register at the International Clinical Trials Registry Platform of the WHO (http://apps.who.int/trialssearch). We will evaluate the presence of outcome reporting bias by comparing outcomes reported in the trial protocols with those in the published reports. Sensitivity analyses will be conducted to evaluate the impact of potential selective reporting on the meta-analytic results. These analyses will help ensure that the findings of our systematic review are robust and not unduly influenced by reporting biases.
Levels of evidence
The Grading of Recommendation, Assessment, Development, and Evaluation (GRADE) instrument,36 as recommended by Cochrane, was used to evaluate the quality of evidence and the strength of recommendations in the studies included in the quantitative synthesis meta-analysis.37 38 This evaluation considered factors such as study design, consistency, directness, heterogeneity, precision, publication bias and other relevant aspects reported in the included studies of this systematic review. The quality of evidence was categorised as high, moderate, low or very low. The assessment of GRADE was conducted using tools available on the website: http://gradepro.org
Patient and public involvement statement
Research design: patients and the public provided valuable input on the research questions and objectives during the design phase, ensuring that the study addresses their needs and concerns.
Interpretation of results: patients and the public will participate in discussions to help researchers understand the clinical and practical implications of the findings, providing important perspectives.
Dissemination of results: we plan to communicate the findings to patients and the public through clear and concise reports and community meetings, ensuring that they fully understand the study’s discoveries and significance.
Amendments
When protocol amendments occur, the date of each amendment will be specified along with a description of the change and the reasoning behind it.
Ethics statements
Patient consent for publication
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
CH and YF are joint first authors.
CH and YF contributed equally.
Contributors XL is the guarantor. CH and JW made equal contributions during the meta-analysis protocol’s development, collaboratively shaping the study’s overarching framework and leading the drafting of the protocol. YF, FY, DC and XL were instrumental in the protocol design phase, offering crucial support and expert guidance to ensure that the research design was both comprehensive and practical. TW and ZZ, through their expertise, significantly influenced the formulation of the search strategy, guaranteeing the research’s extensive coverage of relevant literature. All authors involved meticulously reviewed and endorsed the final manuscript, upholding the study’s high standards of quality and accuracy.
Funding This study was funded by the National Key R&D Program of China (grant numbers 2023YFC3603200 and 2017YFC1310500); the STI2030-Major Projects (grant number 2022ZD0213100); the National Natural Science Foundation of China (grant number 82004284), the Collaborative Innovation Center for Clinical and Translational Science by the Ministry of Education & Shanghai (CCTS-202203), Sanming Project of Medicine in Shenzhen (Grant number. SZZYSM202111011), and Shenzhen Medical Research Fund (Grant number. A2403052).
Competing interests Our team affirms the absence of any conflicts of interest. This research protocol's drafting and execution were carried out autonomously, uninfluenced by any external peer review commissions or interventions.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.