This is an excellent decision! Integrating quantitative and qualitative evidence provides a much richer, more comprehensive understanding of patient adherence, moving beyond "what works" to "how it works," "why it works (or doesn't)," and "what it means to patients."

Given your topic of "patient adherence to treatment," a mixed-methods systematic review is particularly powerful because:

• Quantitative studies can tell you about adherence rates, the effectiveness of interventions to improve adherence, and factors statistically associated with adherence.


• Qualitative studies can illuminate patients' experiences of adherence, their beliefs, motivations, barriers, facilitators, and the psychosocial context of their treatment, which often explains why the quantitative outcomes are observed.

Let's break down the integration approach and practical steps.

---

1. Deciding on the Integration Approach: Convergent Parallel Synthesis (Recommended)

For a systematic review integrating existing quantitative and qualitative studies, the most common and robust approach is Convergent Parallel Synthesis (sometimes called Concurrent Triangulation or Segregated Approach).

Why Convergent Parallel Synthesis?

• Nature of Data: You're dealing with two distinct bodies of evidence that have their own methodologies and epistemologies. This approach respects the integrity of each by analyzing them separately first.


• Review Context: Unlike a primary mixed-methods study where you collect data concurrently or sequentially, in a systematic review, you are synthesizing existing studies. A convergent approach allows you to synthesize each type of evidence using appropriate methods before bringing their findings together.


• Comprehensive Understanding: It aims to bring together different perspectives to provide a more complete picture, identify areas of convergence and divergence, and enrich understanding.


• Feasibility: It is generally more feasible for systematic reviews than sequential approaches (where one type of evidence directly informs the design or analysis of the other within the review, which is hard with pre-existing studies) or fully integrated approaches (which require a single overarching methodology across both).

How Convergent Parallel Synthesis Works:

• Phase 1: Separate Analysis: You conduct a quantitative synthesis (e.g., meta-analysis or narrative synthesis of quantitative findings) and a qualitative synthesis (e.g., thematic synthesis, meta-ethnography) independently.


• Phase 2: Integration/Comparison: You bring the findings from both separate syntheses together to compare, contrast, explain, and build a more integrated understanding in the discussion and synthesis phase of your review.

Other Approaches (and why they're less suitable for your review):

• Sequential Explanatory/Exploratory: One type of data informs the analysis or interpretation of the other. While powerful for primary studies, it's difficult to apply rigorously in a review of existing studies unless you're conducting a multi-stage review (e.g., a qualitative synthesis, then use its findings to frame questions for a new quantitative synthesis, which is a massive undertaking).


• Transformative: Focuses on social justice and advocacy, often driven by a specific theoretical lens. While adherence has social justice implications, this is more about the overall goal of the review rather than the methodological integration approach itself.


• Multifaceted/Parallel: Simply presenting quantitative and qualitative findings side-by-side without a deep, explicit integration of their themes and implications. This is the weakest form of integration and should be avoided in favor of truly synthesizing the findings.

Recommendation: Proceed with Convergent Parallel Synthesis. The key is to ensure the "integration" phase is robust and goes beyond mere juxtaposition.

---

2. Practical Steps for Integration (Convergent Parallel Synthesis)

Here’s a step-by-step guide on how to implement a convergent parallel synthesis for your systematic review on patient adherence:

A. Define Comprehensive Research Questions

Your review questions must be able to be addressed by both quantitative and qualitative evidence.

• Overarching Question: "What is known about patient adherence to [specific treatment/condition]?"


• Quantitative Sub-questions (e.g., PICO):


• "What are the rates of adherence to [specific treatment/condition]?"


• "What interventions are effective in improving patient adherence to [specific treatment/condition]?"


• "What patient-level, treatment-level, or system-level factors are statistically associated with patient adherence?"


• Qualitative Sub-questions (e.g., SPIDER or PICo):


• "What are patients' experiences and perceptions of adherence to [specific treatment/condition]?"


• "What do patients identify as barriers and facilitators to adherence to [specific treatment/condition]?"


• "How do patients make sense of and manage their adherence behaviors?"

B. Develop Separate Search Strategies

You'll need tailored search strategies for each type of evidence, run in parallel.

• Quantitative Search: Focus on keywords for adherence, interventions, outcomes, statistical associations, and relevant study designs (RCTs, observational studies, cohort studies). Use databases like PubMed, Embase, CINAHL, Cochrane Library.


• Qualitative Search: Focus on keywords for adherence, patient experiences, perceptions, beliefs, barriers, facilitators, and qualitative methodologies (interviews, focus groups, phenomenology, ethnography). Use databases like PsycINFO, CINAHL, Scopus, Web of Science, and consider searching for grey literature or specific qualitative journals.

C. Separate Screening and Data Extraction

Maintain separate but parallel processes.

• Screening: Screen titles, abstracts, and full texts against pre-defined inclusion/exclusion criteria for each evidence type.


• Data Extraction (Quantitative Studies):


• Study characteristics (authors, year, country, design).


• Participant characteristics (sample size, demographics, condition).


• Intervention details (type, duration, components).


• Adherence outcomes (measurement method, rates, effect sizes, statistical associations).


• Key findings relevant to your questions.


• Data Extraction (Qualitative Studies):


• Study characteristics (authors, year, country, design, methodology).


• Participant characteristics (sample size, demographics, condition).


• Phenomenon of interest (patient adherence).


• Rich descriptive data: Themes, sub-themes, conceptual categories, direct quotes (if possible/relevant).


• Authors' interpretations and conclusions.

D. Separate Quality Appraisal

Use appropriate and distinct critical appraisal tools for each type of study.

• Quantitative Studies:


• Risk of Bias 2 (RoB 2) for Randomized Controlled Trials (RCTs).


• ROBINS-I for Non-Randomized Studies of Interventions.


• JBI Critical Appraisal Checklists for various designs (cohort, cross-sectional, case-control).


• GRADE (Grading of Recommendations Assessment, Development and Evaluation) can be used to assess the certainty of evidence for quantitative findings.


• Qualitative Studies:


• CASP Qualitative Research Checklist.


• JBI Qualitative Assessment and Review Instrument (QARI).


• ENTREQ (Enhancing Transparency in Reporting the Synthesis of Qualitative Research) Statement can guide reporting of your qualitative synthesis.

E. Separate Synthesis Methods

Conduct the synthesis for each data type independently.

• Quantitative Synthesis:


• Meta-analysis: If studies are sufficiently homogenous regarding interventions, populations, and outcome measures (e.g., for specific adherence interventions).


• Narrative Synthesis: If heterogeneity prevents meta-analysis. This involves systematically describing, comparing, and synthesizing findings from multiple studies in text and tables (e.g., "vote counting" for factors, identifying trends in adherence rates).


• Reporting: Present adherence rates, effect sizes of interventions, and statistically significant factors related to adherence.


• Qualitative Synthesis:


• Thematic Synthesis: A common approach to identify overarching themes and concepts across qualitative studies regarding patient experiences, barriers, and facilitators of adherence.


• Meta-ethnography: More interpretive, aims to develop new conceptual understandings or "line-of-argument" from the findings of individual qualitative studies.


• Framework Synthesis: Applying a pre-existing theoretical framework (e.g., COM-B model, HBM) to categorize and analyze findings across studies.


• Reporting: Present emergent themes, sub-themes, and a conceptual model or framework describing patient experiences of adherence.

F. The Integration Phase (The Core of Convergent Synthesis)

This is where the magic happens. After you have the consolidated findings from your quantitative and qualitative syntheses, you bring them together.

• Map Findings: Create a table or matrix to visually display the key findings from both syntheses side-by-side.


• Columns: Quantitative Finding (e.g., "Intervention X significantly improves adherence by Y%"), Qualitative Finding (e.g., "Patients report Intervention X simplifies medication routine, reducing perceived burden").


• Rows: Corresponding themes or aspects of adherence (e.g., "Intervention effectiveness," "Patient perceptions of burden," "Social support").


• Compare and Contrast (Dialogue between findings):


• Convergence/Confirmation: Where do the findings align and strengthen each other? (e.g., "Quantitative studies show low adherence for complex regimens, and qualitative studies explain this via patient reports of confusion and frustration.")


• Divergence/Discrepancy: Where do the findings conflict or contradict each other? (e.g., "Quantitative studies show Intervention Z has no significant effect on adherence, but qualitative studies report patients find it highly supportive. Why? Perhaps the quantitative outcome measure didn't capture the subtle improvements, or patient perceptions of support didn't translate into measurable behavioral change.")


• Expansion/Elaboration: How do qualitative findings explain or add nuance to quantitative findings? (e.g., "Quantitative studies identify socioeconomic status as a barrier. Qualitative studies elaborate on how this manifests, e.g., through inability to afford transport to pharmacy, lack of health literacy, or competing life priorities.")


• Silence: What questions are raised by one set of findings that the other doesn't address? (e.g., "Quantitative studies show high adherence rates, but qualitative studies reveal significant distress or burden associated with maintaining that adherence.")


• Develop an Integrated Narrative and/or Conceptual Model:


• Weave the findings: In your discussion, don't just list findings; actively discuss them in relation to each other. Create a coherent narrative that draws from both quantitative and qualitative insights.


• Build a Conceptual Framework: If possible, synthesize the findings into a new or refined conceptual model that explains patient adherence more comprehensively, incorporating factors and pathways identified by both types of evidence. For example, a model illustrating the interplay between system factors (quant), intervention components (quant), patient beliefs (qual), and emotional responses (qual) influencing adherence behaviors.


• Answer Your Integrated Research Questions: Explicitly state how the combined evidence answers your overarching questions.

G. Reporting

• Structure: Follow PRISMA guidance (or PRISMA-ScR for scoping reviews) and explicitly report your mixed-methods approach.


• Transparency: Clearly describe each stage of your review, including search strategies, screening, data extraction, quality appraisal, and synthesis methods for both quantitative and qualitative evidence.


• Integration Section: Dedicate a specific section in your results and/or discussion to the integrated findings. Use headings that reflect your integration method (e.g., "Convergence of Evidence," "Divergent Findings," "Enriching Understanding of Adherence Barriers").


• Visual Aids: Use tables, matrices, and diagrams to illustrate the integrated findings and conceptual models.


• Balance: Ensure both quantitative and qualitative findings are given appropriate weight and space.


• Limitations: Discuss the limitations arising from integrating different types of evidence and the limitations of the included studies themselves.

---

Tips for Success:

• Interdisciplinary Team: Ideally, have researchers with expertise in both quantitative and qualitative methods on your review team.


• Protocol Registration: Register your systematic review protocol (e.g., on PROSPERO) and clearly outline your mixed-methods approach and integration strategy beforehand. This enhances rigor and transparency.


• Software: Use reference management software (e.g., EndNote, Zotero) and consider qualitative data analysis software (e.g., NVivo, ATLAS.ti) for your qualitative synthesis, and statistical software (e.g., R, Stata) for meta-analysis if applicable.


• Pilot Testing: Pilot your data extraction forms and critical appraisal tools on a few studies of each type to ensure consistency.


• Reflexivity: Be aware of potential biases and how your own perspective might influence the interpretation of findings, especially during the integration phase.

By following these steps, you'll be well-equipped to conduct a rigorous and insightful systematic review on patient adherence, leveraging the strengths of both quantitative and qualitative evidence. Good luck!