Free scoping read. A written assessment of your manuscript in two business days, at no cost.

Start one
Contents

Academy · Research Writing · Doctoral quality checklist

A Quality Checklist for Doctoral Students Writing Biomedical Research Articles

Much of what distinguishes a publishable first paper from a returned one is determined before drafting and can be checked against a list. This article provides that list, sets out what the evidence on writing training does and does not show, and establishes realistic expectations of the time to publication — at one institutional journal, a mean of 92 days against a target of under 30.

Pubrica AcademyWritten by the medical writing and biostatistics teams Reviewed byPubrica biostatistics team Published9 October 2026 Last checked9 October 2026 Reading time14 minutes Sources51 peer-reviewed

A first biomedical paper usually fails on structure rather than on sentences: a question never stated in one sentence, a methods section that cannot be evaluated, a results paragraph that interprets rather than reports, or a discussion that claims more than the design allows. All of these can be detected with a checklist applied before writing and again before submission. The checklist at the end of this page is the principal deliverable; the preceding sections explain the rationale for each item and what the evidence on writing training supports.

Learning objectives

  • What the evidence on scientific writing training shows — and why satisfaction and perceived gains are not the same as measured quality
  • Which five areas doctoral students report struggling with, from a review of 22 studies
  • The checks that must precede drafting, because they cannot be corrected afterwards
  • How to write a results paragraph that reports rather than interprets
  • Why citation and reference handling shows the smallest self-rated gain from training, and how it should be addressed
  • Realistic timelines: 92 days mean submission-to-publication against a 30-day target at one journal, and the implications for research schedules
  • How to use the checklist with a supervisor so that feedback addresses substance

Common misconceptions

  • “A writing course will make my paper publishable.” Fernandez and colleagues, in BMJ Open, studied 741 students across 27 editions of a 15-hour course and found high satisfaction (mean 4.6 of 5) and self-rated gains of 1.6 for knowledge, 1.3 for attitudes and 1.4 for skills on a 0–4 scale. Those are self-reported gains, not measured manuscript quality; this is a genuine finding, but a narrower one than it is usually presented as.
  • “References are the easy part.” In the same study’s subgroup of participants who later published, citing and writing references showed the smallest gain of all skills, at 1.0 (95% CI 0.8 to 1.2), against 1.5 (95% CI 1.3 to 1.7) for cover letters. It is the skill least improved by training.
  • “Submission to publication takes a few weeks.” Pasarica and colleagues, in Cureus, reported their journal averaging 92 days (SD 62) from submission to publication against an internal target of under 30. Authors should plan in months rather than weeks.
  • “A big meta-analytic effect settles the question.” Xiang and colleagues, in Frontiers in Medicine, pooled 13 studies and 1,396 participants and found that case-based learning raised critical-thinking disposition scores by a mean difference of 26.39 (95% CI 18.71 to 34.06), with heterogeneity of I² 99.03%, which renders the pooled estimate close to uninterpretable. Heterogeneity should be reported, and examined in the work of others.
  • “My supervisor will catch the problems.” Supervisors can identify only what they are shown. If a draft is submitted without a stated research question, the feedback will concern sentences, because sentences are what is visible.

Evidence on the effectiveness of writing training

This point merits candid treatment before a checklist is recommended, because the literature is more limited than the enthusiasm surrounding it. Fernandez and colleagues, in BMJ Open, reported a prospective longitudinal study of a 15-hour, two-day scientific writing course run in Spain across 27 editions from 2004 to 2013, with 741 students 51. Mean overall satisfaction was 4.6 out of 5, and self-rated gains were 1.6 points for knowledge, 1.3 for attitudes and 1.4 for skills on a 0–4 scale, all with P<0.001.

These are consistent findings over nearly a decade, and they measure what participants perceived to have changed. They do not measure whether the manuscripts improved, because the manuscripts were not assessed 51. The distinction has implications for how a course should be used: as a means of learning the conventions quickly, not as a substitute for the structural work that determines whether a paper is publishable.

Two other studies in this area are more encouraging but more selective. Bloothooft and colleagues, in BMC Medical Education, reported a hands-on publishing course in which students produced nine peer-reviewed reviews; among 31 respondents, 71% published at least one further paper and 55% said the project influenced their academic study or career path, and the nine manuscripts appeared in journals with a mean impact factor of 3.56 and were cited an average of 3.73 times per year 4. Pasarica and colleagues, in Cureus, described launching a university scientific journal, after which the share of graduating students reporting authorship of a peer-reviewed paper rose from 48.7% in 2015 to 70.5% in 2018 21. Both are single-institution studies with self-selected participants and no comparison group, so they should be read as demonstrations that structured practice can work, not as effect estimates 4,21.

741students across 27 editions of a 15-hour course, 2004 to 2013
4.6 / 5mean overall satisfaction
+1.6self-rated knowledge gain on a 0–4 scale (P<0.001)
+1.4self-rated skills gain (attitudes +1.3)

These are participants’ own ratings of satisfaction and perceived gain; manuscript quality was not assessed. Source: Fernandez E, Garcia AM, Seres E, Bosch F. Students’ satisfaction and perceived impact on knowledge, attitudes and skills after a 2-day course in scientific writing: a prospective longitudinal study in Spain. BMJ Open.

Common difficulties reported by doctoral students

Oktay and Shamsi, in the Arab World English Journal, reviewed the English-language literature on graduate nursing thesis writing and identified 22 studies, grouping their findings into five categories: challenges, feedback, support, needed skills and thesis writing itself 6. The grouping is more informative than it first appears, because three of the five categories concern the environment rather than the writing: feedback, support, and the challenges of the circumstances 6. A student who has stalled is therefore at least as likely to face a supervision or process problem as a skills problem, and the two require different remedies.

The skills evidence points to a specific and somewhat unexpected conclusion. In Fernandez and colleagues’ subgroup of 145 course participants who went on to publish a paper, the largest self-rated gain was in preparing cover letters, at 1.5 (95% CI 1.3 to 1.7), and the smallest was in citing and writing references, at 1.0 (95% CI 0.8 to 1.2) 51. Reference handling is the skill that training improves least, and it is also the one that most students assume requires no particular attention 51. This assumption is mistaken: reference errors are visible, easy for a reviewer to check, and are read as a sign of carelessness more generally.

Taken together, the two studies have a practical implication: the process should be addressed first (what is shown to a supervisor, and when), reference accuracy should be treated as a deliberate task with its own time allocation, and no single course should be assumed to substitute for either.

Checks required before drafting

These items cannot be repaired later, which is why they come first. There are five.

The question, in one sentence. It should be written out. If it requires two sentences or contains the word “and”, the manuscript probably contains two papers or no clear one. Everything else in the manuscript is judged against it.

The design, and whether it can answer that question. The reason should be stated in one sentence. A cross-sectional study cannot answer a question about change; a single-arm study cannot answer a comparative one. This mismatch produces most structural rejections, and no amount of editing addresses it.

The reporting guideline for the design, selected now. CONSORT 2010, described by Schulz, Altman and Moher in Trials, for parallel-group randomised trials; STROBE, described by von Elm and colleagues in BMJ, for observational studies, with 22 items of which 18 apply across cohort, case-control and cross-sectional designs and 4 are design-specific; PRISMA 2020, described by Page and colleagues in BMJ, for systematic reviews 6,18,32,36. Used before drafting, the checklist is an outline of what a reader expects; used afterwards, it is a list of items that can no longer be reported. Our reporting guidelines page covers choosing one when the design is a hybrid.

Authorship and contributions. This should be settled before writing, against the ICMJE criteria, all four of which must be met 1. Doctoral projects generate authorship disputes precisely because the contribution list is assembled at the end, when the order has become a matter of perception rather than record.

Target journal, chosen from its recent issues. Not from the aims-and-scope page, which describes aspiration. The last two issues should be read to identify a paper of comparable type and design. Our journal selection page covers how to do this systematically.

Methods and statistics under editorial scrutiny

Readers judge a methods section on whether the study could be repeated and whether each decision can be evaluated rather than taken on trust. The typical failure is not poor English but a paragraph that leaves every consequential choice unstated.

Four practices address most of the problem. The statistical test for each outcome should be named, rather than stating that appropriate tests were used; “appropriate” asks the reader to accept the choice on trust and is the most common warning sign in a student methods section. The analysis population should be stated explicitly, including who was excluded and why. The handling of missing data should be stated, because a complete-case analysis is a decision and must be declared as such. Finally, the primary outcome should be stated, in the singular, with its timepoint, and with an indication of where it was pre-specified 32.

Risk of bias merits a separate sentence where the design requires it. Sterne and colleagues’ RoB 2 tool, in BMJ, is the tool a reader will implicitly apply to a randomised design, and Guyatt and colleagues’ GRADE framework, in BMJ and the Journal of Clinical Epidemiology, is the framework a reader will apply to a body of evidence in a review 22,31,35. Reading the relevant tool before writing identifies the methodological details that are essential. If the analysis exceeds the author’s training, a statistician should be consulted before the data are collected, not afterwards; our biostatistics page covers what that involves.

Reporting rather than interpreting results

The most reliable marker of a first paper is a results section that explains what the findings mean. The reader is told the conclusion before being given the numbers, which makes both harder to evaluate. Interpretation belongs in the discussion, with one exception: where a competing explanation is intrinsic to the design, it is preferable to raise it alongside the finding rather than leave it to a reviewer.

Three specific rules apply. Effect sizes should be reported with intervals, not only P values; a P value indicates that the result is unlikely under the null hypothesis, not how large the effect is or whether it matters 32,36. A result reported in the text should never contradict the table; such discrepancies occur more often than might be expected and undermine confidence in the entire paper. The planned analysis should be reported before any additional analyses, and post-hoc analyses labelled as such, because a reader who discovers an unlabelled exploratory analysis is likely to reread the whole paper with suspicion.

One useful practice is to write the results section from the tables and figures outward: the tables are constructed first, and the text is then written to guide the reader through them. The resulting text is shorter, consistent with the numbers, and less prone to interpretation.

A results paragraph that interprets, and one that reports

Interprets too early

Patients in the intervention group showed a significant improvement in symptom scores compared with controls (P<0.05), clearly demonstrating the effectiveness of the intervention. Interestingly, a subgroup analysis revealed that younger patients benefited most, suggesting that age-targeted delivery may be the optimal strategy. Several other outcomes also trended towards improvement.

“Clearly demonstrating the effectiveness” is interpretation, and it appears before the reader has been given a single number to judge it against.

P<0.05 with no effect size. The reader cannot tell whether the difference is clinically meaningful or trivially small.

The subgroup finding is unlabelled. Presented as a result rather than as a post-hoc analysis, it invites a reviewer to assume the whole paper may be selectively reported.

“Suggesting that age-targeted delivery may be optimal” is a recommendation built on an exploratory subgroup, in the results section.

“Trended towards improvement” describes non-significant results without reporting them, which is reporting bias in miniature.

Reports, and lets the reader judge

Mean symptom score at 12 weeks was 14.2 (SD 5.1) in the intervention group and 17.8 (SD 5.6) in the control group; the adjusted between-group difference was −3.6 points (95% CI −5.4 to −1.8; P = 0.001), against a pre-specified minimal important difference of 3.0. Of the three secondary outcomes, one differed between groups (Table 3) and two did not; intervals are reported for all three. A post-hoc analysis stratified by age, not pre-specified, showed a larger difference in participants under 50 (−5.1 points, 95% CI −8.3 to −1.9) than in those aged 50 or over (−2.2 points, 95% CI −4.9 to 0.5); with 61 participants in the younger stratum this comparison is exploratory and the intervals overlap.

Group means, spread, adjusted difference and interval are all given, so the reader can judge both significance and magnitude.

The effect is anchored to a pre-specified minimal important difference, which answers the question a clinician actually has.

All three secondary outcomes are accounted for, including those that showed nothing — no outcome quietly disappears.

The subgroup analysis is labelled post-hoc, the stratum size is given, and the overlapping intervals are stated rather than glossed.

No interpretive verb appears anywhere; the reader reaches the discussion with the evidence already in hand.

The second version is longer and makes fewer explicit claims, and it is considerably harder to challenge. It depended on a pre-specified primary outcome, a stated minimal important difference and a complete list of secondary outcomes. These are protocol decisions, which is why the before-drafting half of the checklist matters more than the drafting half.

The discussion and substantive limitations

A discussion has four functions, in a straightforward order: to state what was found, in one paragraph, without new numbers; to place the finding against the existing literature, including studies that disagree; to state the limitations specifically; and to state the implications, scoped to the design.

Limitations are the weakest element of most doctoral drafts, usually because they are written as a defensive formality. “The sample size was small” and “further research is needed” are uninformative. A stronger version names the specific threat, its likely direction, and what the next study would do differently: which confounder was unmeasured and in which direction it would bias the estimate; what aspect of the sampling makes the result conservative or optimistic; which measurement lacks validation in the study population 36. Written in this way, the limitations paragraph demonstrates command of the material rather than apology, and it pre-empts the objections a reviewer would otherwise raise.

The corollary is avoiding overclaiming. Every verb in the discussion should match the design (associated with, predicted, caused), and the final sentence of the abstract should not claim more than the final paragraph of the discussion. Structural overreach of this kind underlies a large share of manuscript rejections, and it can be corrected at no cost before submission 20,25.

Commonly omitted mechanical checks

These checks are routine, yet they are the ones reviewers notice. Reference accuracy comes first, partly because it is the skill that, according to the evidence, training improves least: Fernandez and colleagues found that citing and writing references showed the smallest self-rated gain, at 1.0 (95% CI 0.8 to 1.2). Every reference should be checked against the original source rather than against a reference manager’s export, and each citation should be confirmed to support the specific claim to which it is attached. A reviewer who finds one misattributed citation will check several more.

The declarations follow: ethics approval, with the committee named and the reference number given; consent, including for case material; funding; conflicts of interest; and data availability 1. Where a trial is involved, registration in a registry meeting the ICMJE minimum 24-item data set, and, since 1 July 2018, a data sharing statement with any manuscript reporting clinical trial results.

Similarity checking should then be undertaken, before submission rather than after. Unintentional overlap in a methods section carried forward from the author’s own protocol or thesis chapter is common and is still flagged; running a plagiarism check in advance ensures that the author identifies it first. Finally, the journal’s own instructions should be checked: word limits, abstract structure, reference style and figure specifications 23,50. These are not quality signals, but failure to meet them generates a technical return before the science is read, which wastes weeks. A final proofreading pass also belongs here, last, after the structural work and not in place of it.

Time required for publication

Unrealistic timelines cause substantial damage to doctoral schedules, so planning should be anchored in published figures. Pasarica and colleagues, in Cureus, reported their institutional journal publishing 67 articles over two years with a 16% rejection rate, and averaging 92 days (SD 62) from submission to publication against an internal target of under 30 days 21. That journal was explicitly designed to be fast and student-friendly, with a low rejection rate, and it nevertheless took approximately three months on average, with wide variation 21.

Plans should be built on that basis. A submission in month one is not a publication in month two, and a thesis timeline that assumes otherwise is likely to fail. The structured-practice evidence also suggests that the effort is cumulative: in Bloothooft and colleagues’ course, 71% of the 31 respondents published at least one further paper afterwards and 55% said the project influenced their academic or career path, with the nine course-produced reviews appearing in journals averaging an impact factor of 3.56 and attracting 3.73 citations per year 4. Again, participants were self-selected and there was no control group, but the direction is consistent with the first paper being the most difficult.

In summary, the first paper typically takes longer than anticipated, mostly in revision and waiting, and the value of the checklist lies in removing avoidable rounds of revision.

What four studies of research-writing training and support report

StudyWhat was examinedReported resultWhat it does not establish
BMJ Open
Fernandez et al.
741 students, 27 editions of a 15-hour course, 2004–2013Satisfaction 4.6/5; self-rated gains 1.6 knowledge, 1.3 attitudes, 1.4 skills on 0–4 (all P<0.001)That manuscripts improved; no manuscript was assessed
BMC Medical Education
Bloothooft et al.
A hands-on publishing course producing nine peer-reviewed reviews; 31 respondents71% published at least one further paper; 55% reported career influence; mean impact factor 3.56; 3.73 citations per yearA comparison with students who did not take the course
Cureus
Pasarica et al.
Launching a university scientific journal; 67 articles over two yearsGraduating students reporting authorship rose from 48.7% (2015) to 70.5% (2018); mean 92 days (SD 62) to publication against a 30-day target; 16% rejection rateThat the journal caused the rise, or that these timelines apply elsewhere
Arab World English Journal
Oktay & Shamsi
Systematic literature review of graduate nursing thesis writing; 22 studiesFindings grouped into five categories: challenges, feedback, support, needed skills, thesis writingWhich interventions work; it maps the problem, not the solution

Figures as reported in each source. The four studies use different designs, populations and outcomes and are not comparable; they are placed together to show what this literature can and cannot support. Three of the four are single-institution studies without a comparison group.

Use of the checklist with a supervisor

Oktay and Shamsi’s five categories place feedback and support alongside skills for a reason: the quality of supervision a student receives is partly determined by what is submitted for review 6. A draft with no stated research question invites comments about sentences, because the structure is not sufficiently visible to comment on.

The checklist should therefore serve as the agenda for supervision rather than as a private audit. Part A, the before-drafting items, should be completed and sent to the supervisor as a single page before any writing begins. It requires about ten minutes to read and is where the supervisor’s expertise is most valuable, because those five items cannot be corrected later. The student then drafts, applies part B to the draft, and sends the draft with the completed part B attached, identifying the items that could not be ticked. This converts an open-ended review into a focused one, and it removes the most common frustration on both sides: a supervisor spending time on wording while a design problem remains unexamined.

The full checklist should be applied twice: once before writing and once before submission. The second pass identifies the items that have drifted, usually the final sentence of the abstract, references added late, and the declarations.

Key points

Key points

The items that determine whether a first paper is publishable are mostly settled before drafting: a question stated in one sentence, a design capable of answering it, the chosen reporting guideline, agreed authorship, and a target journal selected from its recent issues. The drafting checks that follow are genuine but secondary: naming the test, stating the analysis population and missing data, reporting effect sizes with intervals, labelling post-hoc analyses, and making the limitations specific enough to act on. The claims made for writing courses warrant scepticism: Fernandez and colleagues’ 741 students reported high satisfaction and clear perceived gains, but no manuscript was assessed. Publication should also be expected to take months: one student-friendly journal still averaged 92 days from submission to publication against a 30-day target.

Application to your study

Application to your study

  1. State the research question in one sentence and share it with the supervisor before drafting. If it requires two sentences or an “and”, resolve this first.
  2. Write one sentence explaining why the design can answer that question. A mismatch at this point is a problem that editing cannot correct.
  3. Obtain the reporting guideline for the study design at the outset and draft the methods against its items rather than mapping them afterwards.
  4. Agree authorship against the four ICMJE criteria before the first draft, and record each contribution while it is still clearly remembered.
  5. Replace every “appropriate statistical tests” with the named test per outcome, and state the analysis population and the missing-data approach.
  6. Construct the tables first, then write the results text from them. This prevents the text from contradicting the tables or drifting into interpretation.
  7. Label every post-hoc analysis as post-hoc and report every pre-specified outcome, including the ones that showed nothing.
  8. Rewrite each limitation to name the threat, its direction and what the next study would do differently. Delete “further research is needed”.
  9. Check every reference against the original source, not against the reference manager; this is the skill that training improves least.
  10. Run the checklist twice: once before writing, once before submission. The second pass catches the abstract’s last sentence, late references and the declarations.

Frequently asked questions

The evidence supports a narrower claim than promotional descriptions suggest. Fernandez and colleagues, in BMJ Open, followed 741 students across 27 editions of a 15-hour, two-day course from 2004 to 2013: mean satisfaction was 4.6 of 5 and self-rated gains were 1.6 for knowledge, 1.3 for attitudes and 1.4 for skills on a 0–4 scale, all with P<0.001. These are participants’ own ratings; no manuscript was assessed, and there was no comparison group. Courses are an efficient means of learning conventions. They do not address a question that was never clearly stated or a design that cannot answer it, and those factors determine publishability.

Five items, because they cannot be repaired afterwards. The research question, written in one sentence; if it requires two, the work may constitute two papers. The design, with one sentence explaining why it can answer that question. The reporting guideline for the design, selected at the outset so that drafting follows its items rather than mapping them afterwards. Authorship, agreed against the four ICMJE criteria, with contributions recorded while they are still clearly remembered. Finally, the target journal, selected by reading its last two issues rather than its aims-and-scope page. These five answers should be sent to the supervisor on one page before drafting, because that is where the supervisor’s time is most valuable.

Longer than most timelines assume. Pasarica and colleagues, in Cureus, described an institutional journal designed to be fast and student-friendly (67 articles in two years, with a 16% rejection rate), which nevertheless averaged 92 days (SD 62) from submission to publication against an internal target of under 30. That is approximately three months on average, with wide variation, at an unusually favourable venue, and it excludes the time spent writing and the time lost to rejection and resubmission elsewhere. Plans should therefore be made in months. The checklist derives its value mainly from removing avoidable revision rounds and technical returns, which account for most of the time.

Usually because wording is what the draft makes visible. Oktay and Shamsi’s review of 22 studies on graduate thesis writing grouped the findings into five categories (challenges, feedback, support, needed skills and thesis writing), and three of those five concern the environment rather than the writer. The feedback received can be changed by changing what is submitted: the five before-drafting items should be sent first on one page, followed by drafts with a completed checklist attached and the unticked items identified. This gives a supervisor something structural to respond to. An open-ended request for general comments reliably produces comments on sentences.

Each limitation should name the specific threat, its likely direction, and what the next study would do differently. “The sample size was small” and “further research is needed” convey nothing. By comparison: “we did not measure household income, which is associated with both exposure and outcome in this setting and would be expected to inflate the association we report”, or “the instrument has no published validation in adolescents, so scores may not be comparable with the adult literature we cite”. Each of these pre-empts an objection and informs a follow-up investigator what to change. Three specific limitations are more effective than eight generic ones, and they demonstrate command of the material rather than defensiveness.

Four, in order of how often reviewers raise them. Stating that “appropriate statistical tests” were used instead of naming the test for each outcome. Reporting P values without effect sizes and intervals, so that the reader cannot judge magnitude. Leaving the analysis population and the handling of missing data unstated, which makes a complete-case analysis an undeclared decision. Presenting post-hoc or subgroup analyses without labelling them, which also casts doubt on the pre-specified results. All four can be corrected within a day if the underlying analysis was sound. If it was not, a statistician should be consulted before data collection, not at the writing stage.

The heterogeneity should be checked before the pooled estimate is quoted. Xiang and colleagues, in Frontiers in Medicine, pooled 13 studies and 1,396 participants and reported that case-based learning raised critical-thinking disposition scores by a mean difference of 26.39 (95% CI 18.71 to 34.06), alongside heterogeneity of I² 99.03%. At that level of inconsistency the studies are not estimating a common effect, and the pooled figure is close to uninterpretable as a single quantity. The implication applies in both directions: the heterogeneity reported by others should be examined before their pooled estimates are cited, and authors conducting a review should report their own heterogeneity prominently rather than in a footnote.

Greater care than is usually expected, because it is the skill that training improves least. In Fernandez and colleagues’ subgroup of 145 participants who later published, citing and writing references showed the smallest self-rated gain of all skills at 1.0 (95% CI 0.8 to 1.2), against 1.5 (1.3 to 1.7) for cover letters. In practice, every reference should be checked against the original article rather than a reference-manager export, and each citation should be confirmed to support the specific claim attached to it rather than the general topic. Reviewers spot-check, and one misattributed citation prompts them to check several more. A similarity check should also be run before submission, since methods text carried forward from the author’s own protocol or thesis will be flagged.

References

  1. International Committee of Medical Journal Editors (ICMJE). Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. 2026. read 9 October 2026. icmje.org
  2. Adams SN, Kater KA, Khan A, et al. Development of a case-based learning framework for medical education: a scoping review. Medical Science Educator. 2026. doi:10.1007/s40670-025-02583-6 read 9 October 2026. doi.org/10.1007/s40670-025-02583-6
  3. Xiang Y, Liu D, Liu L, Liu IC, Wu L, Fan H. Impact of case-based learning on critical thinking dispositions in Chinese nursing education: a systematic review and meta-analysis. Frontiers in Medicine. 2025;12:1452051. doi:10.3389/fmed.2025.1452051 read 9 October 2026. doi.org/10.3389/fmed.2025.1452051
  4. Bloothooft M, Pennings HJM, van der Heyden MAG. The Utrecht University Honours Program review project: example based scientific publishing training aimed at bachelor medical students. BMC Medical Education. 2024;24(1):1-12. doi:10.1186/s12909-024-05098-7 read 9 October 2026. doi.org/10.1186/s12909-024-05098-7
  5. Karakus I. University students' cognitive flexibility and critical thinking dispositions. Frontiers in Psychology. 2024;15:1420272. read 9 October 2026. frontiersin.org
  6. Oktay YB, Shamsi AF. What English articles say about nursing postgraduate thesis writing: a systematic literature review. Arab World English Journal. 2024;15(4):85-98. doi:10.24093/awej/vol15no4.6 read 9 October 2026. doi.org/10.24093/awej/vol15no4.6
  7. Maia D, Andrade R, Afonso J, Costa P, Valente C, Espregueira-Mendes J. Academic performance and perceptions of undergraduate medical students in case-based learning compared to other teaching strategies: a systematic review with meta-analysis. Education Sciences. 2023;13(3):238. doi:10.3390/educsci13030238 read 9 October 2026. doi.org/10.3390/educsci13030238
  8. Yao J, et al. Case-based learning interventions for undergraduate nursing students in a theoretical course: a review of design, implementation, and outcomes. Journal of Professional Nursing. 2023;46:119-133. read 9 October 2026. frontiersin.org
  9. Arruzza E, Chau M, Kilgour A. Problem-based learning in medical radiation science education: a scoping review. Radiography. 2023;29:564-572. read 9 October 2026. frontiersin.org
  10. Showstark M, et al. Results and lessons learned from a virtual multi-institutional problem-based interprofessional learning approach: the VIPE program. Journal of Interprofessional Care. 2023;37:164-167. read 9 October 2026. frontiersin.org
  11. Borowiec BG. Ten simple rules for scientists engaging in science communication. PLOS Computational Biology. 2023;19(7):e1011251. doi:10.1371/journal.pcbi.1011251 read 9 October 2026. doi.org/10.1371/journal.pcbi.1011251
  12. Guraya SS, et al. Understanding medical professionalism using express team-based learning; a qualitative case-based study. Medical Education Online. 2023;28:2235793. read 9 October 2026. frontiersin.org
  13. Ma C, Zhou W. Effects of unfolding case-based learning on academic achievement, critical thinking, and self-confidence in undergraduate nursing students learning health assessment skills. Nurse Education Practice. 2022;60:103321. read 9 October 2026. frontiersin.org
  14. Wong FMF, Kan CWY. Online problem-based learning intervention on self-directed learning and problem-solving through group work: a waitlist controlled trial. International Journal of Environmental Research and Public Health. 2022;19:720. read 9 October 2026. frontiersin.org
  15. Garcia-Ponce AL, et al. A problem-/case-based learning approach as an useful tool for studying glycogen metabolism and its regulation. Biochemistry and Molecular Biology Education. 2021;49:236-241. read 9 October 2026. frontiersin.org
  16. Cen XY, Hua Y, Niu S, Yu T. Application of case-based learning in medical student education: a meta-analysis. European Review for Medical and Pharmacological Sciences. 2021;25(8):3173-3181. doi:10.26355/eurrev_202104_25726 read 9 October 2026. doi.org/10.26355/eurrev_202104_25726
  17. Yu Z, et al. Effects of blended versus offline case-centred learning on the academic performance and critical thinking ability of undergraduate nursing students: a cluster randomised controlled trial. Nurse Education Practice. 2021;53:103080. read 9 October 2026. frontiersin.org
  18. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021:n71. doi:10.1136/bmj.n71 read 9 October 2026. doi.org/10.1136/bmj.n71
  19. Zhao W, et al. The effectiveness of the combined problem-based learning (PBL) and case-based learning (CBL) teaching method in the clinical practical teaching of thyroid disease. BMC Medical Education. 2020;20:381. read 9 October 2026. frontiersin.org
  20. Jawaid SA, Jawaid M. Common reasons for not accepting manuscripts for further processing after editor's triage and initial screening. Pakistan Journal of Medical Sciences. 2019;35(1):1-3. doi:10.12669/pjms.35.1.28 read 9 October 2026. doi.org/10.12669/pjms.35.1.28
  21. Pasarica M, Bailey M, Cendan JC. Increasing students' publication productivity: could launching a university scientific journal be a catalyst? Cureus. 2019;11(1):e3953. doi:10.7759/cureus.3953 read 9 October 2026. doi.org/10.7759/cureus.3953
  22. Sterne JAC, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898. read 9 October 2026. frontiersin.org
  23. Fujita MSL, Agustin-Lacruz MdC, Terra AL. Journals' guidelines about title, abstract and keywords: an overview of Information Science and Communication Science areas. European Science Editing. 2018;44(4):76-79. read 9 October 2026. europeanscienceediting.org.uk
  24. Hong S, Yu P. Comparison of the effectiveness of two styles of case-based learning implemented in lectures for developing nursing students' critical thinking ability: a randomized controlled trial. International Journal of Nursing Studies. 2017;68:16-24. read 9 October 2026. frontiersin.org
  25. Bannas P, Reeder SB. How to write an original radiological research manuscript. European Radiology. 2017;27(11):4455-4460. doi:10.1007/s00330-017-4879-8 read 9 October 2026. doi.org/10.1007/s00330-017-4879-8
  26. Bodagh N, et al. Problem-based learning: a review. British Journal of Hospital Medicine. 2017;78:C167-C170. read 9 October 2026. frontiersin.org
  27. Kulak V, Newton G. An investigation of the pedagogical impact of using case-based learning in an undergraduate biochemistry course. International Journal of Higher Education. 2015;4:13-24. read 9 October 2026. frontiersin.org
  28. Wager E, Woolley K, Adshead V, Cairns A, Fullam J, Gonzalez J, Grant T, Tortell S. Awareness and enforcement of guidelines for publishing industry-sponsored medical research among publication professionals: the Global Publication Survey. BMJ Open. 2014;4(4):e004780. doi:10.1136/bmjopen-2013-004780 read 9 October 2026. doi.org/10.1136/bmjopen-2013-004780
  29. von Elm E, Altman D, Egger M, Pocock S, Gotzsche P, Vandenbroucke J. The STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement: guidelines for reporting observational studies. International Journal of Surgery. 2014. doi:10.1016/j.ijsu.2014.07.013 read 9 October 2026. doi.org/10.1016/j.ijsu.2014.07.013
  30. Kong LN, et al. The effectiveness of problem-based learning on development of nursing students' critical thinking: a systematic review and meta-analysis. International Journal of Nursing Studies. 2014;51:458-469. read 9 October 2026. frontiersin.org
  31. Guyatt G, Oxman AD, Akl EA, et al. GRADE guidelines: 1. Introduction - GRADE evidence profiles and summary of findings tables. Journal of Clinical Epidemiology. 2011;64(4):383-394. read 9 October 2026. link.springer.com
  32. Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials. Trials. 2010;11(1). doi:10.1186/1745-6215-11-32 read 9 October 2026. doi.org/10.1186/1745-6215-11-32
  33. Mokkink LB, Terwee CB, Patrick DL, et al. The COSMIN study reached international consensus on taxonomy, terminology, and definitions of measurement properties for health-related patient-reported outcomes. Journal of Clinical Epidemiology. 2010;63(7):737-745. doi:10.1016/j.jclinepi.2010.02.006 read 9 October 2026. doi.org/10.1016/j.jclinepi.2010.02.006
  34. Kuper A, Reeves S, Levinson W. An introduction to reading and appraising qualitative research. BMJ. 2008;337:a288. read 9 October 2026. link.springer.com
  35. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ. 2008;336(7650):924-926. read 9 October 2026. link.springer.com
  36. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, et al. Strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007;335(7624):806-808. doi:10.1136/bmj.39335.541782.AD read 9 October 2026. doi.org/10.1136/bmj.39335.541782.AD
  37. Williams B. Case based learning: a review of the literature. Emergency Medicine Journal. 2005;22:577-581. read 9 October 2026. frontiersin.org
  38. Kliewer MA. Writing it up: a step-by-step guide to publication for beginning investigators. American Journal of Roentgenology. 2005;185:591-596. read 9 October 2026. tech.snmjournals.org
  39. Hmelo-Silver CE. Problem-based learning: what and how do students learn? Educational Psychology Review. 2004;16:235-266. read 9 October 2026. frontiersin.org
  40. Sollaci LB, Pereira MG. The Introduction, Methods, Results, and Discussion (IMRAD) structure: a fifty-year survey. Journal of the Medical Library Association. 2004;92(3):364-371. read 9 October 2026. genreacrossborders.uwaterloo.ca
  41. Yeh ML. Assessing the reliability and validity of the Chinese version of the California Critical Thinking Disposition Inventory. International Journal of Nursing Studies. 2002;39:123-132. read 9 October 2026. frontiersin.org
  42. Simpson E, Courtney M. Critical thinking in nursing education: literature review. International Journal of Nursing Practice. 2002;8:89-98. read 9 October 2026. frontiersin.org
  43. International Committee of Medical Journal Editors. Uniform requirements for manuscripts submitted to biomedical journals. Annals of Internal Medicine. 1997;126(1):36-47. read 9 October 2026. stat.cmu.edu
  44. Facione NC, Facione PA, Sanchez CA. Critical thinking disposition as a measure of competent clinical judgment: the development of the California Critical Thinking Disposition Inventory. Journal of Nursing Education. 1994;33:345-350. read 9 October 2026. frontiersin.org
  45. Vickery B. The Royal Society Scientific Conference of 1948. Journal of Documentation. 1992;54(3):281-283. read 9 October 2026. stat.cmu.edu
  46. Atkinson D. The evolution of medical research writing from 1735 to 1985: the case of the Edinburgh Medical Journal. Applied Linguistics. 1992;13(4):337-374. read 9 October 2026. stat.cmu.edu
  47. Huth EJ. Structured abstracts for papers reporting clinical trials. Annals of Internal Medicine. 1987;106(4):626-627. read 9 October 2026. stat.cmu.edu
  48. Meadows AJ. The scientific paper as an archaeological artifact. Journal of Information Science. 1985;11(1):27-30. read 9 October 2026. stat.cmu.edu
  49. Bazerman C. Modern evolution of the experimental report in physics: spectroscopic articles in Physical Review, 1893-1980. Social Studies of Science. 1984;14:163-196. read 9 October 2026. stat.cmu.edu
  50. Radiology (Radiological Society of North America). Radiology scientific style guide and author instructions: scientific editing. Radiology. read 9 October 2026. pubs.rsna.org
  51. Fernandez E, Garcia AM, Seres E, Bosch F. Students' satisfaction and perceived impact on knowledge, attitudes and skills after a 2-day course in scientific writing: a prospective longitudinal study in Spain. BMJ Open. 8(1):e018657. read 9 October 2026. bmjopen.bmj.com

Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.

How to cite this article

Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.

Pubrica Academy. (2026, October 9). A Quality Checklist for Doctoral Students Writing Biomedical Research Articles. Pubrica. https://pubrica.com/academy/research/a-checklist-for-increased-quality-in-writing-biomedical-research-articles-for-doctoral-students/

This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.

The doctoral article quality checklist

Part A should be completed before writing begins and sent to the supervisor on its own. Parts B to E are applied to the draft, and part F last. The whole checklist should then be applied again before submission; the second pass is where the final sentence of the abstract, late references and the declarations are caught. Any item that cannot be ticked is either a task to complete or a limitation to report; it should not be left for a reviewer to find.

DOCTORAL ARTICLE QUALITY CHECKLIST
Manuscript: ______________________________  Design: ____________________
Target journal: ___________________________  Pass 1 date: ______________
Supervisor: _______________________________  Pass 2 date: ______________

================================================================
PART A - BEFORE YOU WRITE A WORD  (send this page on its own)
================================================================
These five cannot be fixed later. Do not start drafting until all five
are answered in writing.

A1. THE QUESTION, IN ONE SENTENCE
    ____________________________________________________________
    [ ] One sentence, no "and"
    [ ] Names population, exposure/intervention, comparator, outcome
    [ ] Could be falsified - it is not "to explore"

A2. THE DESIGN, AND WHY IT CAN ANSWER THAT QUESTION
    Design: ____________________________________________________
    Why it can answer A1, in one sentence: _____________________
    ____________________________________________________________
    [ ] If the question is about CHANGE, the design is longitudinal
    [ ] If the question is COMPARATIVE, there is a comparator
    [ ] If the question is CAUSAL, say what rules out confounding
        (usually: nothing - so A1 or the conclusion must change)

A3. REPORTING GUIDELINE, CHOSEN NOW
    Guideline: ______________________  (CONSORT / STROBE / PRISMA 2020 /
                                        CARE / other: ____________)
    [ ] Downloaded, and the methods will be drafted against its items
    [ ] For STROBE: the 4 design-specific items for my design identified
        (18 of the 22 apply to all three observational designs)
    [ ] Items I will NOT be able to report, listed now: ______________

A4. AUTHORSHIP AND CONTRIBUTIONS
    [ ] Every proposed author meets ALL FOUR ICMJE criteria
    [ ] Contributors who do not qualify are listed for acknowledgement
    [ ] Author order agreed IN WRITING, by everyone, now
    [ ] Corresponding author identified
    [ ] Contribution statement drafted while everyone remembers

A5. TARGET JOURNAL
    [ ] I have read the last TWO issues, not the scope page
    [ ] The journal has published this article type with this design
    [ ] Word limit: ______  Abstract format: ______  Ref style: ______
    [ ] Second-choice journal identified, with its differences noted

    >>> STOP. Send A1-A5 to your supervisor before drafting. <<<

================================================================
PART B - METHODS
================================================================
B1. [ ] Setting, dates and locations stated
B2. [ ] Eligibility criteria stated, and how participants were selected
B3. [ ] Sample size: justified, or stated as a convenience sample
B4. [ ] PRIMARY outcome - singular - named, with its timepoint
B5. [ ] Where the primary outcome was pre-specified (protocol /
        registration number): ______________________
B6. [ ] Every variable defined, including any diagnostic criteria and the
        edition or version used
B7. [ ] For each instrument: name, version, language, and the population
        its validation evidence comes from
B8. [ ] NAMED statistical test for each outcome
        ("appropriate statistical tests were used" does not appear)
B9. [ ] Analysis population stated, with exclusions and reasons
B10.[ ] Missing data: how much, and how handled - stated explicitly
B11.[ ] Confounders named, and the adjustment approach stated
B12.[ ] At least one sensitivity analysis planned or reported
B13.[ ] Software and version named
B14.[ ] Risk-of-bias or certainty tool named if the design calls for one

================================================================
PART C - RESULTS  (report; do not interpret)
================================================================
C1. [ ] Participant flow, numerically, from eligible to analysed
C2. [ ] Baseline characteristics by group
C3. [ ] Effect sizes WITH intervals for every outcome - not P values alone
C4. [ ] EVERY pre-specified outcome reported, including null ones
C5. [ ] No numbers in the text contradict the tables (check each one)
C6. [ ] Post-hoc and subgroup analyses explicitly labelled as such, with
        stratum sizes given
C7. [ ] No interpretive verbs: demonstrates, proves, confirms, clearly
        shows, suggests that we should
C8. [ ] "Trended towards" does not appear; non-significant results are
        reported with their intervals
C9. [ ] Tables and figures are self-explanatory with their captions alone
C10.[ ] Results text was written FROM the tables, not before them

================================================================
PART D - DISCUSSION AND LIMITATIONS
================================================================
D1. [ ] Opens with what was found, in one paragraph, with no new numbers
D2. [ ] Placed against the existing literature, INCLUDING studies that
        disagree
D3. [ ] Every causal verb checked against the design
        (associated with / predicted / caused)
D4. [ ] Limitations are SPECIFIC. For each one:
          threat: ______________  likely direction: ______________
          what the next study would do differently: ______________
D5. [ ] "Further research is needed" and "the sample size was small" as
        standalone limitations have been deleted
D6. [ ] Unmeasured confounders named individually
D7. [ ] Conclusion scoped to the design, and no stronger than D1
D8. [ ] LAST SENTENCE OF THE ABSTRACT is no stronger than the conclusion
        (check these two side by side - this is the commonest slip)

================================================================
PART E - REFERENCES AND DECLARATIONS
================================================================
E1. [ ] EVERY reference checked against the original source, not against
        a reference-manager export
E2. [ ] Each citation supports the SPECIFIC claim attached to it
E3. [ ] No citation of an aggregator or encyclopaedia
E4. [ ] Reference style matches the journal exactly
E5. [ ] Ethics approval: committee named, reference number given
E6. [ ] Consent statement, including for any identifiable case material
E7. [ ] Funding declared, including grant numbers
E8. [ ] Conflicts of interest declared for every author
E9. [ ] Data availability statement present
E10.[ ] Trial registration: registry and number (24-item minimum data set)
E11.[ ] Data sharing statement, if this reports clinical trial results
E12.[ ] Similarity check run BY ME, and any carried-forward methods text
        from my protocol or thesis rewritten

================================================================
PART F - SUBMISSION MECHANICS  (last, and quick)
================================================================
F1. [ ] Word count within limit, including or excluding what the journal says
F2. [ ] Abstract in the journal's required structure and length
F3. [ ] Figures at the required resolution and file format
F4. [ ] Title: specific, no colon-cliche, and no claim stronger than D7
F5. [ ] Keywords chosen from the journal's or index's controlled list
F6. [ ] Cover letter: states what is new in two sentences, and why this
        journal
F7. [ ] Suggested reviewers, if requested, with no conflicts
F8. [ ] A final read of the whole manuscript, aloud, in one sitting

================================================================
UNTICKED ITEMS - what each one becomes
================================================================
Item: __________  Becomes: task / limitation / design change
Item: __________  Becomes: task / limitation / design change
Item: __________  Becomes: task / limitation / design change
Item: __________  Becomes: task / limitation / design change

If any PART A item is untickable, stop and resolve it with your
supervisor before drafting further. Everything in parts B to F can be
fixed or written up honestly as a limitation. Part A cannot.

Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.

How Pubrica can help

Pubrica’s medical writing and biostatistics teams work with doctoral students at both ends of this process. Before drafting, the teams check that the question, the design and the chosen reporting guideline are aligned, and that authorship and the target journal are settled; after drafting, they read the manuscript as a desk editor would and identify the objections a reviewer is likely to raise while there is still time to address them. Editing cannot remedy a design problem, and the teams will state this if such a problem is found.

Find something else

Search the Pubrica Academy

Access is free and requires no account. Enter a topic, or open the full library to filter by category, sub-topic or resource type.

Pubrica can review a research question and draft and identify the checklist items that are not yet satisfied.

A scoping assessment is free and takes two business days. It provides a written assessment of what the work requires and which specialist would undertake it.

Get a free quote

Regulations

Check your project against this rule

  1. 1Your work
  2. 2Your details
  3. 3Send
What do you need?

Material you send is held under signed confidentiality, with access limited to the assigned team and encrypted transfer. It is not used to train anything and is not passed to third-party tools. Ask for the agreement before you send anything you would rather cover first.