For a student with a draft literature review in a public health thesis, whether the review is the background to an empirical study or the review is the thesis itself. The page covers what the review has to show at population level, what each study design can and can't support, where equity and grey literature come in, how scoping and systematic reviews differ, and how far evidence from one country carries to another. Guides from Germany, Norway and Sweden and pages from the PRISMA statement, the Public Health Agency of Canada and EuroHealthNet are quoted. For other chapters, see thesis chapter feedback.
Esmond doesn't open your sources. From within your field, it points out where your use of them leaves a claim unsupported, where you cite a source in two conflicting ways, and other blind spots.
A public health review asks about groups and populations: how common a problem is, who carries more of it, what drives it, and which interventions or policies change it at scale. The University of Bergen's guide for the master's thesis in global health asks the introduction to “Describe the national and global public health relevance of the topic and support the text with relevant literature.”
The review then narrows to a question about a population, an exposure or intervention, an outcome and a setting. A reader checks whether each part of that question is covered by the studies you discuss, and whether the gap you name is a gap in that combination. For the gap in general terms, see research gap; for length, see how long a literature review should be.
A hypothetical example of a commentYour review concludes that school meal programs reduce childhood obesity, but four of the six studies you cite are cross-sectional. Which of them shows a change over time?
A first comment on one part of your own draft, right away and without an account.
Get a free previewThe label on the review sets what the method has to show. Karolinska Institutet's University Library addresses its systematic review page to researchers and tells students: “As a student you should rather do a structured literature review, including only certain aspects of the methodology described on this page.” At Stockholm University, two members of the Department of Public Health Sciences published guidelines for master's students that propose a systematized review, “which is based on the same principles as a systematic review (transparency and reproducibility) but is designed to ensure feasibility given the time and resources available to graduate students.”
Other programs allow the full review as the thesis. Bergen's global health program lists, as one of three approaches, a “Systematic or scoping literature review on a topic with a solid rationale and including a detailed evaluation of each paper’s methodological strengths and weaknesses.” For students who do such a review alone, the same guide states: “For those students who do a review alone, we will accept (as a minimum) that the screening and extraction is only done by one person, that the search is only done in one database, and that it is limited to publications in one language.”
| Review type | What it aims at | What the method part must show |
|---|---|---|
| Background review | Context and rationale for your own study | How sources were found and chosen, in a few sentences |
| Structured or systematized review | A focused question within a master's time frame | Databases, search strings, dates, eligibility criteria, screening, appraisal, and the limits of each |
| Scoping review | The extent, range and nature of research on a broad question | Protocol, search, charting form; appraisal is optional |
| Systematic review | An answer to a focused question from all available studies | Protocol, full search, screening by two people, risk of bias, synthesis |
The PRISMA statement's page on scoping reviews describes their purpose: “Scoping reviews serve to synthesize evidence and assess the scope of literature on a topic.” A scoping review maps what has been studied, where and how. A systematic review answers a question such as whether an intervention reduces an outcome, and weighs the studies by risk of bias. A reader checks that the conclusions match the type: a scoping review that ends with "the intervention is effective" claims something its method didn't test.
During a first test search, Karolinska Institutet's library advises that “it’s a good idea to see if there are any systematic reviews within your research area that have already been published”. If a recent review exists, your review needs a reason to repeat it: a newer period, another population, another setting.
Public health evidence comes from many designs, and the wording of your review has to follow the design. "Associated with" fits an observational finding; "reduces" or "causes" needs a design that supports it. Name the design next to the finding, with the population and country.
| Design | What it can show | What a reader checks when you cite it |
|---|---|---|
| Randomized trial, individual or cluster | The effect of an intervention under the trial's conditions | Who was randomized, adherence, loss to follow-up, how close the setting is to yours |
| Cohort study | Incidence, and the order of exposure and outcome | Which confounders were adjusted for; loss to follow-up |
| Case-control study | Suited to rare outcomes; reports an odds ratio | How controls were chosen; recall bias |
| Cross-sectional survey | Prevalence at one point in time | No time order, so no claim about cause or change on its own |
| Ecological study | Patterns across areas or countries | Conclusions about individuals drawn from group data |
| Natural experiment or interrupted time series | The effect of a policy introduced at a known time | Other changes in the same period; enough data points before and after |
Reviews are read for who is affected as well as for average effects. The Public Health Agency of Canada defines the term: “Health equity means that everyone has a fair chance to be as healthy as possible” EuroHealthNet's Health Inequalities Portal adds: “The term social gradient describes the phenomenon that for every step down the socio-economic ladder there is a corresponding decrease in health status.”
For a review, that means checking whether the included studies report results by socioeconomic position, sex, ethnicity, age or place, and saying so when they don't. An intervention that lowers the average can widen the gap between groups if it reaches the better-off first. The PROGRESS-Plus list (place of residence, race and ethnicity, occupation, gender, religion, education, socioeconomic status, social capital, plus factors such as age and disability) is one way to make that check systematic.
Part of the evidence in public health sits outside journals: WHO reports, surveillance data and reports from national public health institutes such as the Norwegian Institute of Public Health, the Public Health Agency of Sweden, the Robert Koch Institute or the Public Health Agency of Canada, as well as government evaluations and theses. Karolinska Institutet's library names dissertations and clinical trials as grey literature that is sometimes included in the search, and notes that government reports may have to be searched for on the organization's own website. Describe which sites you searched and when, and keep grey sources apart from peer-reviewed studies in the results.
For reviews done as a thesis, the Professorship of Public Health and Prevention at TUM asks for the method in detail: “In a systematic review/scoping review the search strategy should be clearly described, alongside eligibility criteria for studies and a description of the screening and data extraction process as well as intended approaches taken to quality appraisal and results presentation.” And for the results: “In a systematic review/scoping review, a flowchart of the retrieved studies should be presented and the main findings should be clearly described.” The PRISMA 2020 statement gives a checklist and flow diagram for systematic reviews, and PRISMA-ScR does the same for scoping reviews. A reader compares the numbers in the flow diagram with the text and the tables of included studies.
Both programs also have AI rules. The TUM guidelines state: “If AI tools are used in the work on the thesis, this must be clearly stated in the thesis.” The same TUM guidelines say that confidential information and personal data should never be shared with AI tools, so remove personal data from your draft before any upload. Bergen's global health guide lists the purposes for which AI tools can be used: general activities such as generating search strings, as long as AI content is not copied into the text, editing, and support for visual representations; it also states that breaking the university's rules on academic integrity is judged as cheating. Feedback on content is not in that list, so ask your supervisor before using Esmond for your thesis, and declare any use the way your program's rule asks. The can I use Esmond? page has more on this.
The studies in your review may come from countries other than the one your thesis is about, and an effect found in one setting depends on features of that setting: the health system, baseline prevalence, what goods and services cost, laws, how services are delivered and who uses them. A tax on sugar-sweetened beverages works through what drinks cost in shops and how people buy them; a screening program depends on the existing care pathway.
A reader asks two things. Which features of the original setting does the effect depend on, and are they present in yours? A column for country and setting in your table of included studies makes the first question easy to answer. Where a mismatch remains, the discussion is the place to say what it means for your conclusions; see results vs. discussion. If the review feeds a capstone, the MPH capstone project page covers that format.
These fragments are made up. None of them comes from a real thesis.
Esmond evaluates every assignment before you submit. It never writes it.
Esmond is an AI tool that doesn't write for you. If your course doesn't allow any AI tools, don't use Esmond for that work. If you're unsure, ask your instructor or advisor.
In a public health review, Esmond can flag, in the text you upload, causal wording on cross-sectional or ecological evidence, conclusions a scoping method can't carry, numbers that differ between the flow diagram and the text, conclusions about groups when the included studies report no subgroup results, and findings carried from one country to another without a word on the setting. See the sample report for the form of the comments, and how Esmond works for the steps.
Esmond reads what you upload. If the file holds the whole draft, it can set this chapter against the others; if it holds one chapter, the review stays within that chapter and may raise a point that another chapter already answers.
Esmond doesn't rerun your search, screen studies or appraise risk of bias; it reads what your review says about them. You decide what to change.
A review of the evidence on a health problem at population level: how common it is, who is affected, what determines it, and which interventions or policies change it. It can be the background to your own study, or the thesis itself as a structured, scoping or systematic review.
A scoping review maps the extent and nature of research on a broad question, and appraisal of study quality is optional. A systematic review answers a focused question from all available studies, with a protocol, screening and risk-of-bias assessment. Each comes with its own PRISMA reporting checklist.
Some programs allow it. Bergen's global health master's offers a systematic or scoping review as one of three approaches. Karolinska Institutet's library advises students to do a structured review that uses only parts of the systematic method. Check with your supervisor which form your program expects.
For a systematic or scoping review, TUM's Professorship of Public Health and Prevention expects a flow diagram and asks for the appropriate checklist with paragraph numbers; Bergen recommends reporting guidelines such as PRISMA when relevant. For a background review, ask your supervisor how much of the search to describe.
It comes from Judith Garrard's book Health Sciences Literature Review Made Easy. You build a table with one row per source and one column per item you extract, such as design, population, setting and main finding, and work from that table when you draft the review. It also makes gaps in the evidence visible.
It depends on the topic. Surveillance data, WHO reports and national institute reports can hold evidence that journals don't. Karolinska Institutet's library notes that grey literature such as dissertations or clinical trials is sometimes included in the search. State where and when you searched for them.
Read on November 8, 2026.
Get your paper evaluated before you submit