Assessing authorship, study purpose and journals when evaluating clinical studies

This article is part of a comprehensive series exploring how to evaluate clinical studies when addressing information needs using a five-step process:

1. Identifying study type or design;
2. Appraising the journal, authors and study purpose;
3. Critiquing the methods used;
4. Understanding basic data or statistical analysis;
5. Analysing the study results, interpretation and conclusions.

This article explores the second step: appraising the journal, authors and study purpose.

There are thousands of medical journals published worldwide. MEDLINE — the National Library of Medicine’s comprehensive bibliographic database — indexed over 5,200 worldwide journals as of 2026. Meanwhile, Embase — the extensive medical literature database published by Elsevier — indexes over 8,500 journals, which include about 3,500 journals unique to Embase. These published journals and research studies can vary considerably in quality. Readers should consider several factors related to journals, authors/investigators and the purpose of the research when evaluating published studies. 

Editorial boards and peer review are two methods important for helping to ensure the overall quality of a journal and its studies. A study’s investigators might have potential conflicts of interest that could compromise the objectivity and quality of their work, and the journal should state these when publishing the study. As readers, after considering overall journal quality and potential conflicts of interest, we should examine a published study’s objectives and related hypotheses to determine if the design and methods used were sufficient to fulfil the study’s intended purpose. This article will review important considerations pertaining to journals, investigators and a study’s purpose.

Journal and investigator considerations

Journals should have an editorial board to help assure the quality of the studies they publish. An editorial board consists of individuals with expertise in a journal’s area(s) of focus. The board members assist in setting the direction of the journal and in making decisions about the acceptability of manuscripts (i.e. research or review papers) submitted by authors/investigators to that journal for publication consideration. The editor(s) will generally read a manuscript first to ensure it is consistent with the journal’s goals and readership. Manuscripts that do not ‘fit’ with the journal or are poorly written/designed would likely be rejected for publication by the editors. For manuscripts that appear reasonable to publish, ideally the editors’ decisions about publication will consider the comments and recommendations obtained from peer reviewers (see Box 1).

Peer review involves the journal sending a submitted manuscript to a small number of outside individuals (i.e. ‘peers’) who have expertise in that subject area. The peers provide expert review, critique, suggestions for revision and a recommendation about whether to publish. The peer reviewers serve valuable functions by identifying potential flaws in a research study, providing the editors with advice about the suitability of a manuscript for publication, and providing the editors, and authors, of the manuscript with suggested revisions that could improve the scientific quality of the research and the actual writing of the paper. Although the best approach to use for the peer review process (e.g. whether or not reviewers should be ‘blinded’ to the authors’ identities) has not been established, peer review is considered to be important for improving the quality of published studies. 

Potential conflicts of interest

The investigators should ideally conduct their study in a manner that is free from bias or other factors that might influence or affect their objective judgement. But sometimes investigators have relationships or other ties, referred to as potential ‘conflicts of interest’ or ‘competing interests’, that could alter their objectivity. Competing interests could influence the way the investigators conducted the study, reported results or interpreted findings. Potential conflicts of interest can be either personal or financial in nature, although financial conflicts (e.g. ties to a company funding the study) are easier to identify and thus tend to be focused upon (see Box 2).

Journals should ask authors of submitted manuscripts to disclose any potential conflicts of interest and should clearly state these disclosures when the study is published. Where can potential conflicts of interest be found when reading a study? Authors’ affiliations are usually found underneath their names at the top of the paper or near the bottom of the first page. Occasionally, they might be listed at the end of the paper. Financial ties of the authors, the funding source(s) for the study, honoraria or other grants received, consultant arrangements, etc. are generally found at the end of the paper — before or sometimes after the references — or occasionally on the first page. Readers should check for the presence of possible competing interests and, if present, consider whether they appeared to affect the study’s design, methods, or reporting or interpretation of the results. The presence of potential conflicts of interest does not automatically invalidate a study or indicate that the study has major problems — rather, it means that the reader should use extra care when analysing the report.

Questions for determining if a conflict of interest has led to bias

• Did the introduction appear overly positive or unduly focused on the potential benefits of therapy?
• Were the inclusion/exclusion criteria slanted so that patients enrolled would be more likely to respond favourably to the study drug?
• Was an active control used that would lead clinicians to choose the study drug even if both are shown to be equally efficacious? Examples include if the active control is more expensive, not routinely used in practice, or known to have more adverse effects than the study drug;
• Was standard error of the mean (SEM) provided for results when standard deviation (SD) would have been more appropriate to include? 
• Since SEM is always a smaller number than the SD, authors might include SEM with mean values simply to make the individual patient variability in their study appear less than it actually was;
• Did the authors discuss or interpret the findings in a way that was slanted towards or unduly emphasised positive (i.e. favourable) study findings? For example, did they downplay adverse effects or make non-significant differences appear more important than they actually were? At times, authors will make statements in their ‘Discussion’ section, such as ‘The adverse effects were as expected’. While that might be true for the class of drug involved, there could be a large number of adverse effects that should not be overlooked;
• Were statements made that were unsupported by the study’s results, or did the authors inappropriately extrapolate their findings to situations not warranted by the study design/ methods?

Study objectives and hypotheses

The rationale for performing a study should be clear from reading its introduction. The authors should have performed a thorough literature search to identify previous work in that area and the knowledge ‘gaps’ that their study will address. The literature cited should be balanced — to the extent possible — and include studies that found favourable, as well as unfavourable, findings about the study drug. The introduction should also provide an accurate assessment of the potential benefits versus risks from the therapy based on the articles cited.

At the end of the introduction, studies should clearly and concisely state their objective (i.e. purpose) for performing the study or a more definitive hypothesis that provides a statement of the results expected. The objective or hypothesis is a critical starting point for a study because the design and methods chosen must be appropriate to fulfil the study’s purpose.

Most studies usually report an objective instead of a hypothesis. An objective will indicate that a study is being conducted to determine if there is an effect of a therapy or whether one therapy is different than another or superior to another. But an objective does not provide a definitive statement of the results they expect to find in the study. In contrast, hypotheses provide a definitive statement of expected results. Hypotheses are tested statistically to meet the study objective. 

The ‘null hypothesis’ assumes there is no difference among treatments or comparisons or no relationship among variables. Remember the null hypothesis because it is the basis for statistical testing. ‘Alternative hypotheses’, such as one-tailed or two-tailed hypotheses, indicate that a difference is expected between therapies. A ‘one-tailed’ (i.e. one-sided) hypothesis states that the difference (i.e. change) will occur in only one direction, such as a studied drug will cause a greater effect, or a medication will cause less efficacy compared with another agent. A one-tailed hypothesis should be present to use one-tailed statistical tests appropriately for data analysis. A ‘two-tailed’ (i.e. two-sided) hypothesis states that there will be a difference between therapies but does not indicate a specific direction of change — rather the difference will occur in either direction, with one therapy being either more or less efficacious than another. 

Most clinical trials studying a new therapy or new use for an established therapy are interested in seeing if changes occur in either direction: more efficacious or less efficacious. Thus, study objectives are usually given that do not provide a definitive statement of what the investigators will find or a two-tailed hypothesis can be stated. Most statistical tests used for data analysis in a clinical study will therefore be two-tailed as well.

Worked examples

Worked example 1: the null hypothesis

Investigators wish to compare the efficacy of a new drug to treat hypertension with enalapril and state the objective of their study as: “The purpose of this study is to determine the efficacy of the new drug compared with enalapril for patients with mild-to-moderate hypertension.”

Does this study provide an objective or a hypothesis?

An objective is provided. A hypothesis that provides a clear statement of expected findings is not given.

Which of the following statements would best define the null hypothesis (i.e. statement of no difference)?

1. The new drug will be more efficacious than enalapril for the treatment of mild-to-moderate hypertension.
2. There will be no difference in efficacy between the new drug and enalapril for the treatment of mild-to-moderate hypertension.
3. The new drug will not be more efficacious than enalapril for the treatment of mild-to-moderate hypertension.

Answer

Statement 2 shows the null hypothesis. Statement 3 is not correct for this example since it implies that the new drug could in fact be less efficacious than enalapril. Statement 1 indicates a clear direction of effect and represents a one-sided/one-tailed hypothesis.

Worked example 2: conflicts of interest

The authors of a placebo-controlled study indicate their objective is to determine whether daily multivitamins have any effect on the development of common colds. The investigators received a grant to conduct this study from the manufacturer of ‘One a Day’ multivitamins. Each of the investigators were also listed as receiving prior grants from a few other pharmaceutical manufacturers who do not make the vitamin studied. One of the investigators was listed as a consultant to a manufacturer not involved with the vitamin studied. The manufacturer of ‘One a Day’ multivitamins supplied the vitamins and placebos. Subjects are enrolled into the study and assigned to receive either daily ‘One-a-Day’ multivitamins or an identically appearing placebo for 12 months.

Any potential conflicts of interest?

Yes. The manufacturer of the vitamins being studied provided the study funding (i.e. grant), which creates a possible competing interest for the investigators. The grant does not automatically mean the study is not strong, but rather the reader should carefully review this study when published to make sure there are no obvious biases present. It is not considered a competing interest for the investigators if they received past grant support from companies not involved with the current study or if they served as a consultant to such companies. It is also not considered a conflict of interest for the manufacturer to supply the drugs/placebos used in a study if that is their only involvement.

Statement of null hypothesis?

The null hypothesis would be: “There is no difference between placebo and multivitamins on the development of common colds.”

Worked example 3: type of hypothesis

It is unknown if certain statin drugs are more efficacious than others in preventing recurrences of atherosclerotic cardiovascular disease (ASCVD) in patients with stable disease. Therefore, a randomised, double-blind study was performed to determine if there were differences in atorvastatin, rosuvastatin, and pitavastatin therapy in decreasing the risk of ASCVD recurrences. It was hypothesised that there will be differences among the statins in their ability to reduce ASCVD recurrences.

What type of hypothesis is given?

A two-tailed hypothesis. It states that differences among the statins will occur, but no direction of the differences (i.e. one or two of the drugs studied are expected to be superior to the other[s] in decreasing the risk) is clearly stated. Note that the sentence that begins, “therefore, a … was performed to determine if there were differences in atorvastatin, rosuvastatin and pitavastatin therapy in decreasing the risk….“ is the study objective. It is not stating that a decreased risk will occur — rather, they plan to determine if there are differences in the risk among drugs.

Summary and key points

• Editorial boards and peer review are two methods that help ensure the quality of a journal and the studies they publish;
• There are several potential conflicts of interest (i.e. competing interests) that might affect the objectivity of a study’s investigators;
• Financial conflicts are the most straightforward to identify and are usually focused upon;
• Journals should ask authors to disclose all potential conflicts of interest and clearly state these in a published study;
• The introduction of a study should clearly provide the rationale for why it is needed, based on a thorough analysis of relevant literature, and clearly state an objective or hypothesis;
• The null hypothesis is very important and is the basis for most statistical testing.

How to apply to practice

• Be very cautious when reading studies published in journals that do not have editorial boards or that do not use peer review. The chance for bias might be much greater compared to studies from journals that use these methods;
• Always look for potential conflicts of interest for the authors/investigators of a published study and, if one or more exist, carefully consider the various ways in which bias might be present;
• Determine if a study is adequately designed to fulfil its stated objective or hypothesis.

One of the initial steps in reviewing a study is to consider where it was published (i.e. the journal), who the authors/investigators were (i.e. any potential conflicts of interest/competing interests) and why they conducted the study (i.e. the purpose/objective). This information can provide valuable insight into potential limitations of the study before a more detailed critique is conducted.

Self-assessment questions

Question 1

Investigators conducted a study comparing galantamine and placebo on quality of life in 60 patients aged 65–85 years with symptoms of mild Alzheimer’s disease.

The investigators state that this study is being conducted “to compare the efficacy of galantamine with placebo for quality of life in Alzheimer’s disease patients”. Two of the investigators are physicians employed by a university geriatrics centre and one of the investigators is employed by Ortho-McNeil Neurologics, which is the manufacturer of Razadyne brand of galantamine. Study funding came from a grant from Ortho-McNeil Neurologics. All three investigators own stock in Ortho-McNeil-Janssen Pharmaceuticals, and one investigator owns stock in Eli Lilly. Ortho-McNeil Neurologics provided identically appearing drug and placebo for the study. 

Are there any potential conflicts of interest for the investigators in this study? 

What would the null hypothesis be for this study?

Question 2

A study introduction describes how a shorter course of amoxicillin might be as effective as a longer duration of therapy. The investigators stated that the objective of their study was to compare the effectiveness of amoxicillin therapy for three days versus eight days in adults with hospital-acquired pneumonia. 

What would the null hypothesis be for this study?

A: Amoxicillin therapy given for three days is more effective than amoxicillin given for eight days for hospital-acquired pneumonia in adults.

B: Amoxicillin therapy given for three days is not as effective as amoxicillin given for eight days for hospital-acquired pneumonia in adults.

C: There is no difference in effectiveness between amoxicillin therapy given for three days and amoxicillin given for eight days for hospital-acquired pneumonia in adults.

D: There is no reason to give amoxicillin therapy for eight days for hospital-acquired pneumonia in adults.

E: There is no reason to give amoxicillin therapy for three days for hospital-acquired pneumonia in adults.

Question 3

You locate a study published in the Annals of Pharmacotherapy and find the following information on its website:

“The Annals is an independent, peer-reviewed medical journal that advances pharmacotherapy throughout the world by publishing evidence-based articles on practice and research.” 

Briefly describe what is meant by ‘peer-reviewed’ in this example.

Question 4 

The authors of a study stated the following: “We tested the hypothesis that melatonin could significantly improve the quality of sleep in patients with insomnia resulting from psychoactive drug therapy.” This hypothesis is:

A: One-tailed

B: Two-tailed

Question 5 

You are reading a published study comparing the efficacy of two drugs that states the following:

From the School of Medicine, Duke University, NC. Funded by a grant from the National Heart, Lung, and Blood Institute, NIH. Study medications were provided by Pfizer (manufacturer of both drugs). Manuscript received February 16, 2009. Accepted June 30, 2009.

Are there any apparent potential conflicts of interest for the study investigators based on the information provided?  

A: Yes

B: No

Question 6 

Given the information about the published study in Question 5, does it appear that peer review was used for this article?

A: Yes

B: No

Question 7 

You locate a study that has five authors listed: four of the authors are at the School of Medicine at the University of Washington and the fifth author is an employee of Abbott Labs, the manufacturer of the study drug. The study drug (Drug A) is being compared to another drug (Drug B) already marketed by another pharmaceutical company for treatment of the same condition as Drug A. You observe that Drug B is more expensive and not commonly used in clinical practice because of its cost and possible adverse effects. You also note that there are safe, less expensive, equally efficacious drugs available that could have been used for comparison with Drug A in this study. 

What was a likely reason for the selection of Drug B as the comparison drug in this study?

Answer guidance

Question 1

Yes, there are potential conflicts of interest for this study’s investigators. They include owning stock in the company (i.e. Ortho-NcNeil Neurologics) that manufactures galantamine, one of the investigators works for the manufacturer, and the funding to conduct the study came from the manufacturer. Supplying the study drug and placebo is not considered to be a conflict of interest. However, simply because potential competing interests exist does not mean that the study is of poor quality. It does mean that extra care should be taken when reading the study to identify any flaws or biases that might unnecessarily favour the galantamine. The null hypothesis for this study is: “There is no difference between galantamine and placebo on quality of life in Alzheimer’s disease patients.”

Question 2

C: The correct null hypothesis is: “There is no difference between amoxicillin therapy given for three days and amoxicillin given for eight days for hospital-acquired pneumonia in adults.”

Question 3

‘Peer-reviewed’ here is referring to the journal’s use of the peer review process for the articles it publishes. This means that, following review by editors, which is listed on the journal’s website, a manuscript that meets the journal’s goals would be sent to a small number of individuals with expertise in the subject area to review and critique the paper, make suggestions for change and offer a recommendation about whether it should be published. The peer reviewers’ comments are then reviewed by the editor(s), who make the final decision about publication.

Question 4

A: One-tailed. This hypothesis would be one-tailed since it is focused on studying change in only one direction, improving sleep quality. A two-tailed hypothesis would not give a direction of change, but would rather state that the researchers feel that melatonin will affect sleep quality. If stated as a study objective — not a hypothesis — the study would indicate that they are studying the effect of melatonin on sleep quality or to determine if melatonin will affect sleep quality.

Question 5

No: Potential conflicts of interest are not apparent based on the information provided. The investigators are from an academic institution and the grant was funded by a federal government agency that should be unbiased with regard to whether or not the involved drugs will be efficacious. No other competing interests are stated.

Question 6

Yes: It appears that peer review was likely from looking at the dates the manuscript was received and accepted. A four-month time span was involved, sufficient time for peer review to have been performed. This can always be verified by looking up information about the journal online or by checking print issues for peer review information.

Question 7

If you notice something unusual about a study, or a design feature that makes you question why the study was performed that way, always consider whether possible conflicts of interest exist that might have influenced the investigators’ objectivity. In this example, note that one of the authors works for Abbott Labs, the manufacturer of Drug A. Thus, the manufacturer has a vested interest in this study’s findings and this could be the reason for selecting comparison Drug B. From the manufacturer’s perspective, even if Drug A is found to be equally efficacious as Drug B, the problems with Drug B (e.g. cost, adverse effects) would likely result in clinicians favouring the use of Drug A.