Article Text
PDF
Abstract
Introduction ‘Task-shifting’ or ‘task-sharing’ is an effective strategy for delivering behavioural healthcare in lower resource communities. However, little is known regarding the actual steps (methods) in carrying out a task-shifting project. This paper presents a protocol for a systematic review that will identify steps in adapting an evidence-based psychological treatment for delivery by lay/non-licenced personnel.
Methods and analysis A systematic review of peer-reviewed, published studies involving a non-licenced, non-specialist (eg, community health worker, promotor/a, peer and lay person) delivering an evidence-based psychological treatment for adults will be conducted. Study design of selected articles must include a statistical comparison (eg, randomised controlled trials, quasiexperimental trials, pre–post designs and pragmatic trials). Study selection will follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Databases including PubMed, the Cochrane Library, Cochrane Central Register of Controlled Trials, SCOPUS, Cumulative Index to Nursing and Allied Health Literature, APA PsycInfo and Google Scholar will be searched from 2000 to 2020. Risk of bias will be assessed using the Cochrane Collaboration’s Risk of Bias (RoB 2) tool, and publication bias will be evaluated with the Cochrane GRADE approach. A narrative synthesis will be conducted for all included studies, and a summary table following Proctor’s framework for operationalising implementation strategies will be included. This protocol was developed following the 2015 guidelines of Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols.
Ethics and dissemination This review will analyse data from published studies only; thus, it will not require institutional board review. Findings will be presented at conferences, to the broader community via the Community Health Worker Translational Advisory Board and social media, and the final systematic review will be published in a peer-reviewed journal.
- mental health
- public health
- organisation of health services
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 protocol describes a planned systematic review to identify best practices for task-shifting evidence-based psychological treatments to non-licenced/lay health workers.
We will use established operationalised terms to identify and describe implementation strategies.
Studies will be identified via a thorough search strategy using independent data extraction techniques, with risk of bias mitigation strategies.
This protocol adheres to the 17-item checklist, Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols.
The review will only include studies in English and focuses on non-licenced/lay health workers (eg, not nurses or teachers) and identifies only studies examining delivery of psychological treatments (ie, not education or other programming).
Introduction
Mental health disorders are common worldwide,1 and although there are evidence-based psychological treatments (EBPTs) that improve health outcomes, most of the people who need treatment do not receive it.2 Mental health treatment is often provided by licenced mental health professionals in specialty settings.3 These specialty providers may not be available due to workforce shortages, cost of care and/or access difficulty, especially for lower income and economically challenged populations,4 who disproportionately suffer from physical and mental health concerns.5 6
Dissemination and implementation of EBPTs (eg, cognitive–behavioural therapy) to communities in need requires a multidimensional approach with innovative delivery methods.7 ‘Task-shifting’ or ‘task-sharing’ is one notable strategy that has emerged over the past two decades, largely in low-income and middle-income countries, as a method for delivering healthcare in lower resource communities. As described by the WHO, with task-shifting, ‘specific tasks are moved, where appropriate, from highly qualified health workers to health workers with shorter training and fewer qualifications in order to make more efficient use of the available human resources for health’.8
Task-shifting has great promise in improving access to EBPTs. A recent review of 27 trials in low-income and middle-income countries found that task-shifted EBPTs delivered by lay persons in primary care and community settings produced a pooled effect size of 0.49.9 Findings from this review indicate that EBPTs can be task-shifted and maintain effectiveness while delivered in nontraditional settings that improve scalability. Lay health workers, such as community health workers or promotor/as, are often trusted members of their communities and perform many important roles, such as delivering physical healthcare10 11 and mental healthcare12–14 and providing pandemic-related support.15 16Lay health workers help increase access to healthcare in lower resource areas around the world. Therefore, task-shifting EBPTs to lay personnel or paraprofessionals can help reduce disparities in access to evidence-based mental healthcare, thus improving health equity.8 17
Task-shifted psychological treatments
Many different EBPTs have been effectively task-shifted with cultural and contextual adaptations. For example, Patel and colleagues18 developed a task-shifted treatment programme for moderately severe to severe depression based on Behavioral Activation19, an established EBPT. Their rigorous randomised controlled trial (RCT) in India showed that the Healthy Activity Program (HAP), delivered by non-specialist lay counselors in primary care, significantly improved patients’ levels of depression (moderate effect size) and led to remission in almost two-thirds of patients treated.18 Likewise, the ‘Friendship Bench’ programme by Chibanda and colleagues20 was a task-shifting study conducted in Zimbabwe to address depression and common mental disorders. Their treatment programme was based on Problem Solving Therapy,21 an established EBPT22, and was delivered by lay health workers in a population with a high prevalence of people living with HIV. Results of their initial non-controlled clinical trial and a later RCT23 showed that this approach was efficacious in reducing psychological morbidity; a large cluster RCT is now underway.24 In both the HAP and the Friendship Bench studies, lay health workers were trained and supervised by licenced professional specialists (ie, clinical psychologists and/or psychiatrists).
While task-shifting is a recognised method for disseminating EBPTs, the best practice procedural steps for how to task-shift are unclear. A recognised problem in implementation research is that strategies are ‘often inconsistently labeled and poorly described … lack operational definitions … and are part of “packaged” approaches whose specific elements are poorly understood’ (p. 254).25 Efforts to scale up EBPTs are critical, yet there is no straightforward roadmap for how to implement task-shifting in new settings. This gap in the literature leaves interested stakeholders without clear guidance in deploying this promising implementation strategy. Therefore, we seek to operationalise such strategies used in task-shifting projects. Our research question is: what are the best practices in task-shifting an EBPT for delivery by lay/non-licenced personnel, including methods of (A) adapting the treatment for the new delivery context; (B) training lay personnel; (C) implementing the new treatment protocol and maintaining fidelity; and (D) sustaining the task-shifted programme over time?
Objectives
This protocol outlines our specific methods and planned analyses for a systematic review. This paper adheres to the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P).26 The systematic review seeks to identify specific task-shifting strategies using established definitions as described by Proctor and colleagues.25 27 In Proctor’s framework for operationalising implementation strategies, there are seven dimensions to consider: the actor, the action, action target, temporality, dose, the implementation outcome affected and the justification. Our review will focus on identifying each of these strategies employed in task-shifting studies, operationally defined in table 1. The most important outcome of this review is to identify best practices for conducting task-shifting implementation projects.
Dimensions and definitions of task-shifting strategies for the proposed systematic review based on Proctor et al.’s framework for operationalising implementation strategies
Methods and analysis
Eligibility criteria
Types of studies
Study inclusion criteria
(1) Studies must involve a non-licenced, non-specialist (eg, community health worker, promotor/a, peer and lay person) who is delivering the intervention. (2) Studies must address a ‘behavioural health’ problem, broadly defined as any psychological/mental health issue (eg., depression, eating disorders and substance use) and/or physical health concern (eg, chronic disease management, health behaviour change, lifestyle changes and adherence) using behavioural/psychological strategies. (3) The treatment components that have been shifted must be derived from an EBPT that has been found efficacious in at least one prior peer-reviewed RCT (eg, cognitive–behavioural therapy, motivational interviewing, behavioural activation and interpersonal psychotherapy). (4) The studies must include a statistical comparison of some kind. The comparator condition must be any of the following: baseline functioning of participants (as in pre–post design), or in an RCT, the control group must be attentional control, a waitlist control, a non-treated group, a treatment-as-usual group, a group receiving a different form of treatment or a group receiving treatment delivered by an expert provider (e.g, licenced psychologist). Eligible study designs also include RCTs, quasiexperimental trials, pre–post designs, pragmatic trials (eg, using stepped wedge or cluster RCT designs). (5) Eligible studies must report evidence of effectiveness of the task-shifting strategy (ie, clinical outcomes) by using a study design that statistically analyses outcomes using a comparator/control.
Study exclusion criteria
(1) Studies that deliver care solely using a licenced or specialist/non-lay person (eg, nurse and educator); (2) studies focused solely on task-shifting a primarily medical task (eg, HIV treatment and prenatal care); (3) studies reporting psychological/behavioural treatments that have not been previously proven effective as outlined above or not involving treatment (eg, screening only); (4) patient education studies with no behavioural intervention (eg, nutritional information only); (5) studies not involving a comparison; (6) descriptive studies, case reports or exclusively qualitative studies; (7) studies not published in a peer-reviewed journal (eg, dissertations, poster or paper presentations and newsletter articles); (8) books and book chapters; and (9) study protocol publications.
Types of participants
Participant inclusion criteria
(1) Study participants must have received a psychological/behavioural-based (ie, non-pharmacological) intervention for a ‘behavioural health’ problem, broadly defined as any psychological/mental health problem (eg, depression, eating disorder, parent–child behavioural issues and substance use) and/or physical health concern (eg, chronic disease management, lifestyle changes and adherence); (2) study participants must have received interventions delivered by non-licenced, non-specialists; and (3) study participants must be adults aged 18 years and older.
Participant exclusion criteria
(1) Patients treated using pharmacological, surgical or medical procedures as the primary intervention tested in the study; and (2) participants treated exclusively by licenced health professionals (eg, physicians and nurses).
Setting and timeframe
Inclusion criteria for setting and timeframe
(1) Task-shifting research studies conducted in high-income, low-income and middle-income countries; and (2) studies conducted in any setting (e.g, healthcare or community settings) or region (eg, urban and rural).
Exclusion criteria for ssetting and timeframe
Studies conducted prior to 2000 (ie, the approximate time when task-shifting was first reported).
Report characteristics
Information sources
The search strategy will be adapted for each of the following sources/databases: PubMed, the Cochrane Library, Cochrane Central Register of Controlled Trials, SCOPUS, Cumulative Index to Nursing and Allied Health Literature, APA PsycInfo and Google Scholar. The search will cover the time frame from January 2000 to July 2020. Peer-reviewed published literature will be sought, and posters, dissertations and presentations; descriptive or protocol articles; books and book chapters; and studies not published in English will be excluded. Unpublished studies will not be sought. The search will be rerun prior to the final analysis, and any further studies identified will be retrieved for inclusion.
Search strategy
The search strategy will be developed and overseen by a medical librarian in consultation with the primary researchers throughout the review (see online supplemental file 1 for PubMed search strategy example). Medical Subject Headings and free-text terms relating to lay health workers and the implementation of task-shifting/sharing will be included (see table 2).
Supplemental material
Supplemental material
Key word search terms
Study screening and selection
Using a coding guide and form, two reviewers will independently search titles and abstracts to remove publications not meeting inclusion criteria. Full texts will be retrieved. Multiple reports of the same study will be linked together (collated) per Cochrane guidelines (see Handbook sec 4.6.2)28 so that the unit of interest is each study, not each article. For example, if a single study was split into separate publications such as a protocol paper, report of the actual study, a qualitative analysis on acceptability and a follow-up, it will be counted as one study, and each of these articles will be searched for relevant data. We will examine prior reviews on the same topic and employ hand-searching of the reference lists for articles identified in the inclusion stage. We will iteratively refine our search strategy to refine the coding guide and form. We will add indexing terms as needed during the preliminary development of the inclusion article guide.
Data extraction
A data extraction chart has been developed by the team to aid in extracting the specific task-shifting steps from the included articles (see online supplemental file 2 for example of study extraction table). Data extracted will follow established definitions as described by Proctor and colleagues.25 27 Additional data we anticipate extracting include year of study publication, design, setting, participant demographics, geographic location of study, type of personnel delivering the intervention, demographics of personnel delivering intervention and reported effect sizes.
Two coauthors will independently conduct data extraction, and an additional author will review the data extracted for completion and accuracy. When necessary, consensus will be reached through discussions with an independent fourth author. Missing data will be sought out by contacting study investigators for unreported data and/or additional details. Data will be recorded in an Excel spreadsheet.
Assessing risk of bias
Reviewers will consider the quality of studies and risk of bias using the Cochrane Collaboration’s Risk of Bias (RoB 2) tool,29 incorporating considerations for evaluating psychotherapy outcome research.30 Although the RoB 2 is focused on RCTs, it is applicable to other types of study designs (eg, quasiexperimental and pre–post).29 All domains will be assessed, including bias related to the randomisation process, deviations from intended interventions, missing outcome data, measurement of the outcome, selection of reported result and overall bias. Two authors will review and independently rate risk of bias in each domain as ‘low risk of bias’, ‘some concerns’, or ‘high risk of bias’. When necessary, disagreements will be resolved by reaching consensus with a third reviewer.
Using the same procedures, we will evaluate publication bias by using the relevant section of the Cochrane GRADE tool31 and also mitigate publication bias by searching as extensively as possible using diverse databases, reviewing reference lists and any related systematic reviews. While many systematic reviews grade evidence of a particular treatment, the purpose of this review is to identify specific methods; therefore, we are restricting our search to published literature only.
Data synthesis and analysis
Data synthesis will primarily involve descriptive statistics with tables and graphs to visually communicate findings. Descriptive statistics will be employed to categorise and tally the different types of methodologies used in task-shifting studies, based on standardised language for operationalising implementation strategies,25 to include actors, actions, action targets, implementation outcome affected, temporality, dose and justification. Frequency counts and measures of central tendency will be included. We will report effect sizes for each study. Although we are not grading evidence (ie, incorporating all GRADE criteria,32 we have developed strict inclusion criteria for rigour and quality as outlined above.
We will consider different subgroups of studies in our review, such as design, population, personnel delivering intervention, location of studies and setting. Although this review is descriptive (no inferential statistical analyses are planned), different tables for each subgroup will be developed. Various subgroups are important to consider separately because differences are anticipated in methodologies depending on each subgroup, for example: design (randomised vs non-randomised trial), population (those with physical health vs mental health concerns), personnel delivering the intervention (community health worker vs other lay personnel), location of studies (low-income and middle-income countries vs high-income countries) and setting (community vs clinical or clinically affiliated).
Patient
Patients/public were not involved in choosing the methods or plans for dissemination of this protocol. However, we will seek feedback on dissemination plans for the forthcoming systematic review from our Community Health Worker Translational Advisory Board.
Ethics and dissemination
This review will analyse data from published studies only; thus, it will not require institutional board review. Any important protocol amendments will be documented in the methods section of the planned systematic review manuscript. Findings will be presented at conferences, to the broader community via the Community Health Worker Translational Advisory Board and social media, and the final systematic review will be published in a peer-reviewed journal.
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
Twitter @DrKatieK
Contributors KK: literature search, study design, writing, critical revision and referencing; LSK and JP: study design, writing and critical revision; LMG: study design and writing; CSG: search strategy and writing; ER: literature search and writing; ESL, YJ-E and JA: writing and critical revision; AGR and JT: study design and critical revision; EPF: study design, literature review, writing and critical revision.
Funding This work was supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health under Award Number K23DK123398 (PI: KK) and by National Institute on Aging of the National Institutes of Health grant number K76AG060003-01A1 (PI: LSK).
Disclaimer The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Competing interests None declared.
Patient consent for publication Not required.
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.