Chapter 16: Equity and specific populations

This is an archived version. For the current version, please go to training.cochrane.org/handbook/current

Vivian A Welch, Jennifer Petkovic, Janet Jull, Lisa Hartling, Terry Klassen, Elizabeth Kristjansson, Jordi Pardo Pardo, Mark Petticrew, David J Stott, Denise Thomson, Erin Ueffing, Katrina Williams, Camilla Young, Peter Tugwell

Key Points:
  • Health equity is the absence of avoidable and unfair differences in health.
  • Health inequity may be experienced across characteristics defined by PROGRESS-Plus (Place of residence, Race/ethnicity/culture/language, Occupation, Gender/sex, Religion, Education, Socio-economic status, Social capital and other characteristics (‘Plus’) such as sexual orientation, age and disability).
  • Cochrane Reviews can inform decision making by considering the distribution of effects in the population and implications for equity.
  • To address health equity in Cochrane Reviews, review authors may: consider health equity at the question formulation stage, possibly using a logic model; decide what methods will be used to identify and appraise evidence related to equity and specific populations; consider implications for ‘Summary of findings’ tables (e.g. separate tables for disadvantaged populations, separate rows for differences in risk of events); and interpret findings related to health equity in the discussion.

Cite this chapter as: Welch VA, Petkovic J, Jull J, Hartling L, Klassen T, Kristjansson E, Pardo Pardo J, Petticrew M, Stott DJ, Thomson D, Ueffing E, Williams K, Young C, Tugwell P. Chapter 16: Equity and specific populations. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.1 (updated September 2020). Cochrane, 2020. Available from www.training.cochrane.org/handbook.

16.1 Introduction to equity in systematic reviews

Health equity reflects a concern for social justice (Braveman 2006, Krieger 2008, Marmot et al 2008, Frieden 2011, Marmot et al 2012). When differences in health are avoidable, remediable and considered unjust and unfair, they are considered health inequalities (Whitehead 1992). Not all health differences are considered inequitable. For example, sickle cell disease is more common in some populations defined by ethnicity due to genetic differences and is not likely to be considered unfair. However, socio-economic differences in childhood asthma rates due to differential distribution of air pollutants would be considered an inequity. Reducing health inequities is considered an important public policy objective for social justice (i.e. moral grounds), social cohesion (for utilitarian reasons) and inter-generational solidarity (for sustainability).

We use the term ‘disadvantaged’ to denote disadvantage created by social, political and legal structures and processes (Welch et al 2015). Axes of potential disadvantage can be defined by the acronym PROGRESS-Plus (place of residence, race/ethnicity/culture/language, occupation, gender/sex, religion, education, socio-economic status and social capital) and ‘Plus’ refers to additional categories such as age, sexual orientation and disability which may influence opportunities for health of individuals and populations (O'Neill et al 2013). Other lists of characteristics may be helpful, depending on the intended audience of the review, such as the social determinants of health or SCRAP (sex, comorbidities, race, age and pathophysiology) (Dans et al 2008). The degree to which these factors are associated with disadvantage depends on time, place and interaction between the determinants (Lorenc et al 2013).

Review authors and decision makers increasingly recognize the importance of the impact of interventions on health equity. Some populations may not benefit from interventions to the same extent as others, which could lead to unintentional intervention-generated inequities (Lorenc et al 2013). Policy makers report that the lack of health equity considerations in systematic reviews limits their usefulness for decision making (Petticrew et al 2004).

Average results hide differences in effects between different populations. Therefore, review authors should consider not only what works on average, but also consider intervention impacts on health inequities. Systematic reviews may assess effects on health equity according to three types of interventions (Welch et al 2012):

  1. interventions aimed at the general population, where it is important to understand the distribution of effects across one or more PROGRESS-Plus characteristics;
  2. interventions focused on disadvantaged or at-risk populations in which there may not be equity outcomes but that may provide evidence about reducing inequities; and
  3. interventions aimed at reducing social gradients across populations or among subgroups of the population.

Trials often exclude populations that are disadvantaged or those above or below a certain age. The exclusion of these populations may influence the applicability of results beyond the trial settings. Review authors should report on the characteristics of the populations according to relevant PROGRESS-Plus factors as well as whether there are population subgroups with a higher risk of the condition or problem or if there are differences in factors that influence access to care. Such factors include values, preferences, affordability and feasibility from the patient/public perspective and conscious or unconscious bias by practitioners. Wait times for total joint arthroplasty provide an example of practitioner bias and gender differences in access to care (Pederson and Armstrong 2015). These factors may vary according to context.

It is usually not feasible to assess all PROGRESS-Plus characteristics. Thus, in choosing characteristics to assess, review authors should consider the perspective of the intended beneficiaries of the interventions and the intended users of the evidence.

16.2 Formulation of the review

Five issues are important for formulating the review question: (i) defining health equity; (ii) hypotheses related to equity and logic models; (iii) appropriate study designs; (iv) appropriate outcomes; and (v) context.

16.2.1 Defining health equity

As health equity implies a judgement about fairness, the first step for review authors is to define which populations experience health inequity with respect to the condition/problem or intervention being assessed. For example, in a Cochrane Review of school meals, socio-economic status, gender and rurality were considered important factors associated with health inequity, but proxy measures were also used: baseline nutritional status was used as a proxy measure for socio-economic disadvantage (Kristjansson et al 2007). Justification for the use of proxies should be given, their use should be transparent and their limitations should be clearly reported.

Review authors may need to consider specific populations separately, either within a broader review or in a focused review, depending on the question and the intended recipients of the intervention. For example, it may be important to consider a separate review for indigenous peoples such as a review on family-centred interventions for indigenous childhood well-being (McCalman et al 2017). For interventions delivered to diverse populations, review authors should assess the primary studies for transparent reporting of participant demographics. It is also important to assess the need for sensitivity or subgroup analyses to explore potential differences in effects. Equity reviews can consider these differences across populations defined by one or more PROGRESS-Plus factors (e.g. migrants, linguistic minorities, homeless); however, they likely cannot address all PROGRESS-Plus factors. Thus, at the question formulation stage, review authors should explicitly consider which factors are most important and how they will be addressed in the methods of the review. Box 16.2.a provides information related to considerations for deciding whether there may be differences in the relevance or appropriateness of an intervention based on whether it is being implemented in low- and middle- and/or high-income countries. Box 16.2.b provides resources that may be helpful when planning systematic reviews of studies including children and youth.

Moreover, rather than using one category to describe people’s experiences, intersectionality illuminates the complex ways a person experiences discrimination simultaneously – across ageism, sexism, racism, and other forms of institutionalized discrimination (Hankivsky 2014).

For example, a Cochrane Review of school feeding for improving the physical and psychosocial health of disadvantaged students reported: “children were classified as ‘predominantly disadvantaged’ by …the following criteria: 1) Living in a rural area or village; 2) Living in an urban area and described as socio-economically disadvantaged (e.g. poor or low-income) or from poor areas (e.g. slums); 3) if statistics were presented showing that 30% or more of the children in the sample were underweight, or stunted (nutritionist judgement) or that the average weight, height, and Body Mass Index (BMI) were low (nutritionist judgement) and 4) studies were implicitly or explicitly aimed at disadvantaged children, and indicators of disadvantage were provided in the paper.”(Kristjansson et al 2007)

Box 16.2.a Low- and middle-income countries (LMICs)

It is important to consider whether the functioning of an intervention or its relevance may differ among high-, middle- or low-income country settings and populations. For example, health systems may vary in financing, regulation, organization, and mechanisms of care delivery. There may also be differences in the wider context, e.g. economy and geography, and the relative importance of health issues. It may be appropriate to include only studies conducted in LMICs when:

  1. the intervention(s) that the review addresses is highly relevant in LMICs and of little or no relevance in high-income countries (HICs);
  2. there are compelling reasons to believe that the problem or the intervention(s) are different in LMICs;
  3. the outcomes of interest are different;
  4. the intervention(s) would be expected to function differently, so that the evidence would be unlikely to be transferrable between LMICs and HICs; or
  5. the researchers or review commissioners are particularly interested in evidence from LMICs.

Focusing solely on LMICs because the intervention is uncommon in HICs is not sufficient unless the problem or outcomes of interest are different in LMICs and HICs, and the intervention is expected to function differently.

For reviews that include studies from all countries, and where the topic is particularly important for LMICs but relevant for HICs, the Background of the review should address why the same intervention might have different absolute and/or relative effects in LMICs and HICs. Where appropriate, review authors should include subgroup analyses for LMICs and HICs and consider the applicability of the evidence for LMICs and HICs in the discussion.

For all reviews, review authors should consider (Oxman et al 2009):

  • if LMIC populations are likely to be disadvantaged by the intervention delivered;
  • whether there is evidence of differences in baseline conditions across LMIC and HICs, or for groups within these settings, which would result in differences in the absolute effectiveness of the intervention;
  • whether there is evidence of differences in access to care or the quality of care across LMIC and HICs; and
  • the implications of these differences for implementing the intervention to ensure that inequities are reduced if possible and they are not increased.

Box 16.2.b Systematic reviews including children and youth

Differences between children and adults, and amongst children and youth of disparate ages, mean that questions often arise around defining the population and planning subgroup analyses. Tools from the STAR Child initiative can be useful in planning a review (Sinha et al 2012, Williams et al 2012).

For reviews of conditions that are relevant to both children and adults, review authors should be aware of and document potential differences in:

  • the nature or course of the condition;
  • the intervention when delivered to adults and children;
  • the efficacy, effectiveness or safety profile of the intervention; and
  • important outcomes, measurement of outcomes, and clinically important differences (Sinha et al 2012, Williams et al 2012).

Note: Differences across sex/gender and other elements of PROGRESS-Plus may be relevant to consider.

16.2.2 Logic models and theories of change to articulate hypotheses about equity

Analytic frameworks such as logic models, causal chains and funnels of attrition are increasingly being used in systematic reviews to identify key questions across the population, intervention, comparison group and outcomes (PICO) of interest (Chapter 2, Section 2.4). Funnel-of-attrition or equity-effectiveness frameworks explain why effect sizes decrease along the causal chain and allow for identification of the various factors such as coverage and uptake that may impact the implementation of an intervention (Tugwell et al 2008, White 2014). Logic models, which show the relationships between inputs and results, can help identify the key questions that are relevant to assessing effects on health equity by predicting likely differences in response, differences in baseline risk, applicability and also factors that may mediate effects. These factors and differences can guide the methods of the review. They can help scope the review question, identify eligibility criteria, focus the search strategy, design a process evaluation and consider relevance to policy and/or practice (Anderson et al 2011, O’Connor et al 2011). For example, a Cochrane Review of food supplementation for improving the physical and psychosocial health of socio-economically disadvantaged children included a logic model showing how socio-economic factors and family structure might modify effectiveness of supplementary feeding (Kristjansson et al 2015).

Theories of change provide a comprehensive description and illustration of how and why a desired change is expected to happen in a particular context (Mackinnon et al 2006, Kneale et al 2015). Pathways to change may be uncovered in the process of doing the review, therefore, theories of change may need to be updated and revised during the review process to incorporate discoveries about the processes and barriers and facilitators to implementation.

16.2.3 Appropriate study designs to assess equity

Eligible study designs should be chosen according to their fitness for purpose (Tugwell et al 2010), and the rationale should be clearly explained (see Chapter 3). Review authors need to consider whether non-randomized studies may provide relevant and meaningful evidence about the impact of the intervention in populations and settings that they consider important (Tugwell et al 2010). These different study designs need different assessment of potential bias (see Chapter 24).

16.2.4 Appropriate outcomes for equity

Outcomes need to be selected based on the stakeholder/user groups. A framework may be helpful in defining the relevant groups. For example, these could include the 9 ‘P’s: patients, practitioners, the public, policy makers, press, product makers (e.g. drug and devices manufacturers), payers (e.g. medical insurers), purchasers (e.g. employers, governments) and principal investigators (Concannon et al 2012, Rader et al 2014). Outcomes to be considered include benefits and harms (and their trade-off): mortality (general/condition specific), impact (symptoms, physical/emotional/social/spiritual function, quality of life, utility, inconvenience, financial burden) and intermediate/surrogate outcomes/biomarkers (Boers et al 2014). See Box 16.2.c for specific considerations for outcomes of importance for children and youth and older adults.

The relative importance of health and social outcomes may differ for populations who experience health inequity. For example, maternal employment, family income and education are important outcomes in a Cochrane Review of day care for preschool children of disadvantaged mothers (Zoritch et al 2000). These outcomes may be less important for mothers with higher socio-economic status. A similar analysis of relative importance could be applied to older adults with pension or other forms of social security, in contrast to those without. The importance of outcomes for different settings and populations needs to be rated when selecting outcomes for ‘Summary of findings’ tables (Chapter 14, Section 14.1.2). Context should be considered in rating importance of outcomes (Section 16.2.5). Additionally, inconvenience, burden (e.g. out-of-pocket costs, travel time) and stigma need to be considered as potential outcomes even if they are not commonly reported in primary studies since they may be of utmost importance to the intended recipients of the intervention.

Box 16.2.c Outcomes for child health or ageing

There may be differences among children, adults, and older adults in disease pathogenesis, clinical features and natural history, physiological and psychological outcomes, and contrasting roles within the contexts of families and society in general. Across age groups, appropriate doses and likelihood of compliance will vary.

For children and youth:

  • developmental outcomes and growth will be important;
  • autonomy and independence may be important for youth; and
  • outcomes for parents and carers can have direct relevance for children.

For older adults:

  • appropriate outcomes should consider well-being, frailty, a continuum of abilities and disabilities, physical and cognitive decline, social participation and low mood;
  • outcomes are often measured in decades rather than years, in terms of trajectories over the life course; and
  • adverse effects are particularly common in later life, often presenting non-specifically, for example falls, immobility, cognitive problems (delirium and dementia) and incontinence. Other adverse events include loss of ability to live independently (e.g. requiring home care, community services or a move to residential care home) and impacts on informal carers (who may also be older adults), including caregiver stress and depression) (Jull 2010).

Note: Differences across sex/gender and other elements of PROGRESS-Plus may be relevant to consider.

16.2.5 Context and equity

Review authors should consider the social, cultural and political contexts in which interventions are planned and implemented (Marmot et al 2008). Primary research on health and social interventions is conducted within particular temporal, cultural, geographical, political and organizational settings (Pope et al 2007), and these may influence intervention effectiveness (Hawe et al 2004).

‘Taking context into account’ means understanding the important aspects of context, how these may influence the intervention (e.g. implementation), and describing, stratifying and exploring the extent to which they influence outcomes (Lewin et al 2017). For example, for reviews including older adults, multimorbidity without integrated care, and overall declines in capacities are an important contextual issue. One aspect can be assessed with the number of prescribed medicines and therefore review authors may wish to report this indicator. Some tools have been developed to collect and extract data on context, including the Context and Implementation of Complex Interventions (CICI) framework (Pfadenhauer et al 2017).

Review authors may wish to assess and document whether research procedures in included studies meet international ethical standards, since populations experiencing health inequities may be vulnerable in research and need additional protections (Welch et al 2017a). Systematic reviews can reinforce ethical practices by identifying ethical concerns in included studies.

Variations in context between studies can be assessed qualitatively and/or quantitatively. Context may be described in different sections of the primary studies or in accompanying papers, reports, policies or historical documents; finding these descriptions may need expert knowledge (Noyes et al 2013). Thus, the full team and advisory board (if the review has one) or other key stakeholders should be involved in interpretation to ensure that the review is useful, relevant and applicable. For example, a Cochrane Review on environmental interventions to reduce the consumption of sugar-sweetened beverages reported: “the context in which included studies were done can therefore be essential for assessing the transferability and applicability of their results… We will therefore extract contextual data, using the categories defined by the CICI (Context and Implementation of Complex Interventions) framework.” (von Philipsborn et al 2016)

16.3 Identification of evidence

Searches for equity-focused reviews should follow the general guidance (Chapter 4), but should ensure there is enough coverage of populations of interest. Searches related to health equity are likely to address perspectives beyond the biomedical lens. Thus, potentially relevant studies may be found in a wider range of literature sources and may be unreliably categorized. This may influence the databases and search terms chosen. A Cochrane Review of interventions for promoting reintegration and reducing harmful behaviour and lifestyles in street-connected children and young people searched a broad range of websites and grey literature sources (Coren et al 2016).

16.3.1 Databases to consider

Non-health databases may be relevant if the outcomes of interest include, for example, labour productivity or educational, economic or social outcomes. The information retrieval guidance of the Campbell Collaboration is an excellent resource for searches related to social outcomes (Kugley et al 2017), while the Norwegian Satellite of the Effective Practice and Organisation of Care (EPOC) Group maintains a list of databases relevant for low- and middle-income countries (EPOC 2013). For example, a Cochrane EPOC review of strategies to increase the ownership and use of insecticide-treated nets to prevent malaria searched multiple databases in addition to MEDLINE and EMBASE, including: CINAHL, Web of Science, Dissertations and Theses, African Index Medicus, LILACS and WHOLIS (Augustincic Polec et al 2015).

16.3.2 Term selection and use of search filters for equity

Using standard search filters (i.e. those available in the search interface of a database) for equity-related content carries significant risks, as many of the words describing PROGRESS-Plus categories are not indexed in the major databases (MEDLINE/Pubmed added a new MeSH term, ‘health equity’, in 2016). Paediatric studies are also often poorly indexed. Authors of studies on children-specific conditions may fail to use paediatric terms explicitly in the title, abstract, or even within the manuscript. Therefore, when searching electronic databases, we recommend using a paediatric search filter (a combination of the subject headings, age limits [if available], and free text terms) rather than indexing or age limits alone. Searching for studies related to older people may consider available search filters for relevance (van de Glind et al 2012). When validated filters are available, their use will save time in building the search and in reducing the number of articles to screen. For example, validated search filters have been developed for sex and gender specific outcome data (Lorenzetti and Lin 2017) and for equity-focused studies (Prady et al 2018) may be helpful in designing searches. Additional filters can be found on the ISSG search filters resource (ISSG 2018).

16.3.3 Practical advice

Appropriate retrieval strategies vary, depending on the research question and the specific populations and settings included. Practical suggestions include the following.

  • Use expert advice on planning and executing the search strategy, given the anticipated complexity of the searches (Chapter 4). Experts might know of unpublished, non-indexed or hard-to-locate evidence.
  • Identify validated filters, considering sensitivity and specificity, and trying to correct known limitations. If the filter is not validated, consider carefully the risk of missing vital information.
  • Look beyond traditional databases: small and specific databases addressing the research topic may be more relevant (Ogilvie et al 2005, Augustincic Polec et al 2015).
  • Develop logic models to make explicit the decisions on the search strategy.
  • Conduct iterative searches: language changes over time and varies by place.

16.4 Appraisal of evidence

For equity questions, baseline imbalance across PROGRESS-Plus factors may be important to assess by checking for poor randomization. Further, equity factors may be considered as potential confounders in non-randomized studies. Authors should document whether losses occurred differentially from specific populations defined by PROGRESS-Plus. Otherwise, the critical appraisal of evidence is similar to other reviews (discussed in Chapter 8 and Chapter 25).

16.5 Synthesis of evidence

Equity analysis involves three steps: first, identifying in the protocol which populations are likely to experience health inequity; second, assessing whether the intervention results in important improvement; and third, assessing whether the identified populations achieve the same improvement in both absolute and relative effects as other populations. Methods for assessing gradients of effects and gaps in absolute and relative effects are described by Evans et al (2001).

A Cochrane Review on culturally appropriate health education for type 2 diabetes mellitus in ethnic minority groups included equity considerations in the synthesis of the data: “we anticipated the need to stratify participants in age groups, as it can be an important effect modifier of outcomes; the effect of gender of participants, matched with gender of educators, were also analysed; … we tried to explore difference between different literacy subgroups, ability to speak language of the majority population and countries where the interventions take place; we stratified participants by ethnic groups.” Differences by age, gender and education were not explored because of insufficient data (Hawthorne et al 2008).

16.5.1 Subgroup analyses

For interventions provided to a broad population, equity may be considered through subgroup analyses across one or more PROGRESS-Plus factors, as pre-specified in the logic model and protocol (Chapter 2, Section 2.5.1).

Any subgroup analyses should be pre-specified and justified (Chapter 10, Section 10.11). In the process of doing the review, other important factors influencing outcomes may be uncovered. Authors should be open to this and all post-hoc decisions should be documented.

Meta-regression (Chapter 10, Section 10.11.4) may also be feasible to assess the role of explanatory variables such as population, context or process factors (Hollands et al 2015).

16.5.2 ‘Summary of findings’ tables

Authors may want to consider one of five methods to incorporate findings about health inequities in ‘Summary of findings’ tables (Welch et al 2017b):

  1. include health equity as an outcome;
  2. consider patient-important outcomes relevant to health equity;
  3. present separate tables for populations who experience health inequity to highlight important differences in relative effectiveness;
  4. create different rows within a single table to highlight differences in baseline risk for specific populations; and
  5. assess indirectness of evidence for populations that are predefined as important who experience health inequity.

16.6 Interpretation of evidence

Interpretation of evidence for specific populations defined across PROGRESS-Plus should focus on those populations identified at the protocol stage as important recipients of the intervention. Interpretation should consider the questions: Are findings likely to be applicable in those populations, even if they did not make up a large proportion of the participant populations in included studies? Why or why not? This section should be transparent and rely on details in the ‘Summary of findings’ table for specific populations. Any subgroup analyses should be interpreted with caution (Chapter 10, Section 10.11.6). See Box 16.6.a for specific examples of issues with interpretation for reviews including older adults.

Box 16.6.a Issues with interpretation for reviews including older adults

It is often difficult to determine applicability to all older people, including those who are frail and dependent. Frailty is an important concept, but it is of limited use as there are no widely adopted operational criteria. However, the following reported data can be useful:

  • type of residence, for example the proportion of patients living long-term in a care home (can be a proxy measure for those who are frail, disabled or have chronic cognitive impairment or dementia);
  • ability to perform basic activities of daily living (allows interpretation of whether results are applicable to older people living with disability); and
  • number and proportion of those with dementia, or whether dementia was a study exclusion criterion (allows consideration of whether results are generalizable to older people with major chronic cognitive impairment).

16.7 Concluding remarks

We recommend that review authors explicitly consider the relevance of health equity to their review at the title and protocol stages using tools such as the Equity Checklist (Ueffing et al 2009), then design their methods accordingly to assess effects on health equity and/or discuss generalizability and applicability. Checklists for review authors are listed in Box 16.7.a.

Box 16.7.a Checklists for review authors

Several published checklists can help review authors to work through and consider issues of equity:

  1. Cochrane and Campbell Equity Checklist for Systematic Review Authors for protocol planning (Ueffing et al 2009);
  2. PRISMA-Equity Checklist to report findings from equity-focused systematic reviews (Welch et al 2012);
  3. Sex and gender assessment tool (Doull et al 2010); and
  4. Sex/gender analysis briefing notes (Doull et al 2014)

For more information see the websites of the Campbell and Cochrane Equity Methods Group, the Sex/Gender Methods Group, Cochrane Child Health and Cochrane Global Ageing.

16.8 Chapter information

Authors: Vivian A Welch, Jennifer Petkovic, Janet Jull, Lisa Hartling, Terry Klassen, Elizabeth Kristjansson, Jordi Pardo Pardo, Mark Petticrew, David J Stott, Denise Thomson, Erin Ueffing, Katrina Williams, Camilla Young, Peter Tugwell

Acknowledgements: We would like to acknowledge Sari Tudiver and the Sex/Gender Methods Group whose work helped to inform this chapter and Ritu Sadana for useful comments. We acknowledge Simon Lewin for contributions to the equity concepts related to LMIC.

Funding: VAW holds an Early Researcher Award (2014–2019) from the Ontario Government. PT holds a Canada Research Chair in Health Equity (Tier 1), 2016–2024.

16.9 References

Anderson LM, Petticrew M, Rehfuess E, Armstrong R, Ueffing E, Baker P, Francis D, Tugwell P. Using logic models to capture complexity in systematic reviews. Research Synthesis Methods 2011; 2: 33-42.

Augustincic Polec L, Petkovic J, Welch V, Ueffing E, Tanjong Ghogomu E, Pardo Pardo J, Grabowsky M, Attaran A, Wells GA, Tugwell P. Strategies to increase the ownership and use of insecticide-treated bednets to prevent malaria. Cochrane Database of Systematic Reviews 2015; 3: CD009186.

Boers M, Kirwan JR, Wells G, Beaton D, Gossec L, d'Agostino MA, Conaghan PG, Bingham CO, 3rd, Brooks P, Landewe R, March L, Simon LS, Singh JA, Strand V, Tugwell P. Developing core outcome measurement sets for clinical trials: OMERACT filter 2.0. Journal of Clinical Epidemiology 2014; 67: 745-753.

Braveman P. Health disparities and health equity: concepts and measurement. Annual Review of Public Health 2006; 27: 167-194.

Concannon TW, Meissner P, Grunbaum JA, McElwee N, Guise J-M, Santa J, Conway PH, Daudelin D, Morrato EH, Leslie LK. A New Taxonomy for Stakeholder Engagement in Patient-Centered Outcomes Research. Journal of General Internal Medicine 2012; 27: 985-991.

Coren E, Hossain R, Pardo Pardo J, Bakker B. Interventions for promoting reintegration and reducing harmful behaviour and lifestyles in street-connected children and young people. Cochrane Database of Systematic Reviews 2016; 1: CD009823.

Dans A, Dans L, Guyatt G. Applying results to individual patients. In: Guyatt G, Rennie D, Meade M, Coon J, editors. Part B Therapy. New York (NY): McGraw-Hill Companies; 2008. p. 273–289.

Doull M, Runnels VE, Tudiver S, Boscoe M. Appraising the evidence: applying sex- and gender-based analysis (SGBA) to Cochrane systematic reviews on cardiovascular diseases. Journal of Women's Health 2010; 19: 997-1003.

Doull M, Welch V, Puil L, Runnels V, Coen SE, Shea B, O'Neill J, Borkhoff C, Tudiver S, Boscoe M. Development and evaluation of 'briefing notes' as a novel knowledge translation tool to aid the implementation of sex/gender analysis in systematic reviews: a pilot study. PloS One 2014; 9: e110786-e110786.

EPOC. LMIC Databases: Effective Practice and Organisation of Care; 2013. http://epoc.cochrane.org/lmic-databases.

Evans T, Whitehead M, Diderichsen F, Bhuya A, Wirth M. Challenging Inequities in Health: from Ethics to Action. New York (NY): Oxford University Press; 2001.

Frieden TR. Forward: CDC Health Disparities and Inequalities Report - United States, 2011. MMWR Supplements 2011; 60: 1-2.

Hankivsky O. Intersectionality 101 Vancouver (BC): Institute for Intersectionality Research & Policy, Simon Fraser University; 2014. https://docplayer.net/4773103-Intersectionality-101-olena-hankivsky-phd.html.

Hawe P, Shiell A, Riley T. Complex interventions: how "out of control" can a randomised controlled trial be? BMJ 2004; 328: 1561-1563.

Hawthorne K, Robles Y, Cannings-John R, Edwards AG. Culturally appropriate health education for type 2 diabetes mellitus in ethnic minority groups. Cochrane Database of Systematic Reviews 2008; 3: CD006424.

Hollands GJ, Shemilt I, Marteau TM, Jebb SA, Lewis HB, Wei Y, Higgins JPT, Ogilvie D. Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco. Cochrane Database of Systematic Reviews 2015; 9: CD011045.

ISSG. The InterTASC Information Specialists' Sub-Group Search Filter Resource 2018. https://sites.google.com/a/york.ac.uk/issg-search-filters-resource/home.

Jull J. Seniors caring for seniors: examining the literature on injuries and contributing factors affecting the health and well-being of older adult care-givers. Ottawa (ON): Public Health Agency of Canada; 2010.

Kneale D, Thomas J, Harris K. Developing and Optimising the Use of Logic Models in Systematic Reviews: Exploring Practice and Good Practice in the Use of Programme Theory in Reviews. PLoS ONE 2015; 10: e0142187.

Krieger N. Proximal, Distal, and the Politics of Causation: What’s Level Got to Do With It? American Journal of Public Health 2008; 98: 221-230.

Kristjansson E, Francis DK, Liberato S, Benkhalti Jandu M, Welch V, Batal M, Greenhalgh T, Rader T, Noonan E, Shea B, Janzen L, Wells GA, Petticrew M. Food supplementation for improving the physical and psychosocial health of socio-economically disadvantaged children aged three months to five years. Cochrane Database of Systematic Reviews 2015; 3: CD009924.

Kristjansson EA, Robinson V, Petticrew M, MacDonald B, Krasevec J, Janzen L, Greenhalgh T, Wells G, MacGowan J, Farmer A, Shea BJ, Mayhew A, Tugwell P. School feeding for improving the physical and psychosocial health of disadvantaged elementary school children. Cochrane Database of Systematic Reviews 2007; 1: CD004676.

Kugley S, Wade A, Thomas J, Mahood Q, Jørgensen AMK, Hammerstrom K, Sathe N. Searching for studies: a guide to information retrieval for Campbell systematic reviews. Campbell Collboration; 2017. Version 1.1 https://onlinelibrary.wiley.com/doi/10.4073/cmg.2016.1.

Lewin S, Hendry M, Chandler J, Oxman AD, Michie S, Shepperd S, Reeves BC, Tugwell P, Hannes K, Rehfuess EA, Welch V, McKenzie JE, Burford B, Petkovic J, Anderson LM, Harris J, Noyes J. Assessing the complexity of interventions within systematic reviews: development, content and use of a new tool (iCAT_SR). BMC Medical Research Methodology 2017; 17: 76.

Lorenc T, Petticrew M, Welch V, Tugwell P. What types of interventions generate inequalities? Evidence from systematic reviews. Journal of Epidemiology and Community Health 2013; 67: 190-193.

Lorenzetti DL, Lin Y. Locating sex- and gender-specific data in health promotion research: evaluating the sensitivity and precision of published filters. Journal of the Medical Library Association 2017; 105: 216-225.

Mackinnon A, Amott N, McGarvey C. Mapping change: Using a theory of change to guide planning and evaluation 2006. http://www.grantcraft.org/index.cfm.

Marmot M, Friel S, Bell R, Houweling T, Taylor S, Commission on Social Determinants of Health. Closing the gap in a generation: health equity through action on the social determinants of health. Lancet 2008; 372: 1661-1669.

Marmot M, Allen J, Bell R, Goldblatt P. Building of the global movement for health equity: from Santiago to Rio and beyond. Lancet 2012; 379: 181-188.

McCalman J, Heyeres M, Campbell S, Bainbridge R, Chamberlain C, Strobel N, Ruben A. Family-centred interventions by primary healthcare services for Indigenous early childhood wellbeing in Australia, Canada, New Zealand and the United States: a systematic scoping review. BMC Pregnancy and Childbirth 2017; 17: 71.

Noyes J, Gough D, Lewin S, Mayhew A, Michie S, Pantoja T, Petticrew M, Pottie K, Rehfuess E, Shemilt I, Shepperd S, Sowden A, Tugwell P, Welch V. A research and development agenda for systematic reviews that ask complex questions about complex interventions. Journal of Clinical Epidemiology 2013; 66: 1262-1270.

O'Neill J, Tabish H, Welch V, Petticrew M, Pottie K, Clarke M, Evans T, Pardo Pardo J, Waters E, White H, Tugwell P. Applying an Equity Lens to interventions: Using PROGRESS to ensure consideration of socially stratifying factors to illuminate inequities in health. Journal of Clinical Epidemiology 2013; 67: 56-64.

O’Connor D, Green S, Higgins JPT, editors. Chapter 5: Defining the review question and developing criteria for including studies. In: Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 (updated March 2011): The Cochrane Collaboration; 2011. Available from www.handbook.cochrane.org.

Ogilvie D, Hamilton V, Egan M, Petticrew M. Systematic reviews of health effects of social interventions: 1. Finding the evidence: how far should you go? Journal of Epidemiology and Community Health 2005; 59: 804-808.

Oxman AD, Lavis JN, Lewin S, Fretheim A. SUPPORT Tools for evidence-informed health Policymaking (STP) 10: Taking equity into consideration when assessing the findings of a systematic review. Health Research Policy and Systems 2009; 7 Suppl 1: S10.

Pederson A, Armstrong P. Sex, gender and systematic reviews: the example of wait times for hip and knee replacements. In: Armstrong P, Pederson A, editors. Women’s Health: Intersections of Policy, Research and Practice. Toronto: Women's Press; 2015. p. 56-72.

Petticrew M, Whitehead M, Macintyre SJ, Graham H, Egan M. Evidence for public health policy on inequalities: 1: The reality according to policymakers. Journal of Epidemiology and Community Health 2004; 58: 811.

Pfadenhauer LM, Gerhardus A, Mozygemba K, Lysdahl KB, Booth A, Hofmann B, Wahlster P, Polus S, Burns J, Brereton L, Rehfuess E. Making sense of complexity in context and implementation: the Context and Implementation of Complex Interventions (CICI) framework. Implementation Science 2017; 12: 21.

Pope C, Mays N, Popay J. Synthesising qualitative and quantitative health evidence: A guide to methods: A guide to methods: McGraw-Hill Education (UK); 2007.

Prady SL, Uphoff EP, Power M, Golder S. Development and validation of a search filter to identify equity-focused studies: reducing the number needed to screen. BMC Medical Research Methodology 2018; 18: 106.

Rader T, Pardo Pardo J, Stacey D, Ghogomu E, Maxwell LJ, Welch VA, Singh JA, Buchbinder R, Legare F, Santesso N, Toupin April K, O'Connor AM, Wells GA, Winzenberg TM, Johnston R, Tugwell P. Update of strategies to translate evidence from Cochrane Musculoskeletal Group systematic reviews for use by various audiences. Journal of Rheumatology 2014; 41: 206-215.

Sinha IP, Altman DG, Beresford MW, Boers M, Clarke M, Craig J, Alberighi OD, Fernandes RM, Hartling L, Johnston BC, Lux A, Plint A, Tugwell P, Turner M, van der Lee JH, Offringa M, Williamson PR, Smyth RL. Standard 5: selection, measurement, and reporting of outcomes in clinical trials in children. Pediatrics 2012; 129 Suppl 3: S146-152.

Tugwell P, Maxwell L, Welch V, Kristjansson E, Petticrew M, Wells G, Buchbinder R, Suarez-Almazor ME, Nowlan MA, Ueffing E, Khan M, Shea B, Tsikata S. Is health equity considered in systematic reviews of the Cochrane Musculoskeletal Group? Arthritis and Rheumatism 2008; 59: 1603-1610.

Tugwell P, Petticrew M, Kristjansson E, Welch V, Ueffing E, Waters E, Bonnefoy J, Morgan A, Doohan E, Kelly MP. Assessing equity in systematic reviews: realising the recommendations of the Commission on Social Determinants of Health. BMJ 2010; 341.

Ueffing E, Tugwell P, Welch V, Petticrew M, Kristjansson E. C1, C2 Equity Checklist for Systematic Review Authors. Version 2009-05-28. 2009. http://equity.cochrane.org/sites/equity.cochrane.org/files/uploads/equitychecklist.pdf

van de Glind EM, van Munster BC, Spijker R, Scholten RJ, Hooft L. Search filters to identify geriatric medicine in Medline. Journal of the American Medical Informatics Association 2012; 19: 468-472.

von Philipsborn P, Stratil JM, Burns J, Busert LK, Pfadenhauer LM, Polus S, Holzapfel C, Hauner H, Rehfuess E. Environmental interventions to reduce the consumption of sugar‐sweetened beverages and their effects on health. Cochrane Database of Systematic Reviews 2016; 7: CD012292.

Welch V, Petticrew M, Tugwell P, Moher D, O'Neill J, Waters E, White H. PRISMA-Equity 2012 extension: reporting guidelines for systematic reviews with a focus on health equity. PLoS Medicine 2012; 9: e1001333.

Welch V, Jull J, Petkovic J, Armstrong R, Boyer Y, Cuervo L, Edwards S, Lydiatt A, Gough D, Grimshaw J, Kristjansson E, Mbuagbaw L, McGowan J, Moher D, Pantoja T, Petticrew M, Pottie K, Rader T, Shea B, Taljaard M, Waters E, Weijer C, Wells G, White H, Whitehead M, Tugwell P. Protocol for the development of a CONSORT-equity guideline to improve reporting of health equity in randomized trials. Implementation Science 2015; 10: 146.

Welch VA, Norheim OF, Jull J, Cookson R, Sommerfelt H, Tugwell P. CONSORT-Equity 2017 extension and elaboration for better reporting of health equity in randomised trials. BMJ 2017a; 359: j5085.

Welch VA, Akl EA, Pottie K, Ansari MT, Briel M, Christensen R, Dans A, Dans L, Eslava-Schmalbach J, Guyatt G, Hultcrantz M, Jull J, Katikireddi SV, Lang E, Matovinovic E, Meerpohl JJ, Morton RL, Mosdol A, Murad MH, Petkovic J, Schünemann H, Sharaf R, Shea B, Singh JA, Sola I, Stanev R, Stein A, Thabaneii L, Tonia T, Tristan M, Vitols S, Watine J, Tugwell P. GRADE equity guidelines 3: considering health equity in GRADE guideline development: rating the certainty of synthesized evidence. Journal of Clinical Epidemiology 2017b; 90: 76-83.

White H. Current challenges in impact evaluation. The European Journal of Development Research 2014; 26: 18-30.

Whitehead M. The concepts and principles of equity and health. International Journal of Health Services 1992; 22: 429-445.

Williams K, Thomson D, Seto I, Contopoulos-Ioannidis DG, Ioannidis JP, Curtis S, Constantin E, Batmanabane G, Hartling L, Klassen T. Standard 6: age groups for pediatric trials. Pediatrics 2012; 129 Suppl 3: S153-160.

Zoritch B, Roberts I, Oakley A. Day care for pre-school children. Cochrane Database of Systematic Reviews 2000; 3: CD000564.