Alexander- The doctor who was shot by his patient

How do you know if a scientific article is reliable?

Every topic that you search for has thousands of research articles. Although the internet had made it easy to search for every article we need, it also made it difficult by giving thousands of options for a single topic. The scientific reliability of the article becomes a question mark. We can’t review a thousand articles for one topic, it is practically impossible. Somehow we need to reduce it to 20-25 articles. These are some rules I realised in the long run from my reading experience to select the articles.

• Identifying the name of the journal:
  • First and the most important thing is identifying the name of the journal. Do not ever fall into the trap of predatory journals. Predatory journals are the journals which accept the publication with article processing fees and author charges and do not consider evaluating the content by peer-reviewing. Nowadays these journals also send requests personally on LinkedIn and g-mail for manuscript submissions. Hence a big NO to all articles of the predatory journals.
• Is it a peer-reviewed journal?
  • Peer reviewing is a very important part of publishing. This shows how well you can defend the research you have done and how well the scientific community apart from your research group accepts your ideas. Sometimes, having gone through peer reviewing adds unique perspectives to the manuscript and makes it more robust. Hence check the journal’s website, if it mentioned peer review.
• Impact factor:
  • Clarivate calculates the impact factor every year. It reflects the yearly mean number of citations that the journal receives for the last two years’ publications. Observing the trend of impact factors across years of a journal shows whether the quality of publications accepted by the particular journal is reducing or improving.
• Pub-med indexed:
  • If you are in a medical or paramedical field one easy way is to check if the journal is pub-med indexed. There are chances that if a publishing company is starting a new journal it might not have impact factors for the first few years since it takes some time to get noticed by the people. Pub-med indexed journals are more trustworthy than the non-indexed or no impact factor journals. But this principle applies only to the medical and paramedical subjects.
• Evidence-based pyramid:
  • Evidence based pyramid in medicine has the review articles and meta-analyses at the top with the highest valued content. As you go below the pyramid the scientific level of evidence becomes less relevant and unfiltered. Hence it is wise to start any literature review with the review articles.
• Check the author’s name:
  • Research the author’s experience in the field and see if they have multiple publications in a similar field. Experienced research authors or publications in collaboration with experienced research authors yield better scientific content.
• Publisher name:
  • Since there are some well-established publishers in the past few decades already, the articles from those publishers can be easily trusted without any doubt.
• Technicalities of the paper:
  • There are some technical details or factors that help you to understand the reliability of the publication. For example, Does the sample chosen for the study represent the population or is there any sampling bias? Is the methodology appropriate for the particular disease? Does the conclusion drawn from their results appropriate? Or the authors are biased toward any one side of the story? Do they have any financial interests? These thoughts will make you think about the credibility of the article.
• Number of citations:
  • Various people working in the same field often cite the articles of high significance . If the article has multiple citations, that represents the science in the article was useful to open up many answers and perspectives in the field. Any article that has been published recently will not have citations, since it has been in the journal for a very short period. Hence also correlate the year of publications along with the number of citations.
• Year of the publication:
  • Sometimes in the medical field, when you read about the treatment or diagnosis of a disease in a paper that was published 20 years ago and compare it with the treatment of current times, there would be a lot of difference. Hence it is always important that you keep the year of publication in mind to realise the importance of that particular evidence.

Once you select the articles for review, you will need to know how to read a research article?

http://optometryzone.com/2022/03/24/how-to-read-a-research-article/

All the above tips are not hard and fast rules. Take all these tips with a pinch of salt. They can be different in rare case scenarios. These only represent the maximum percentage of good publication characteristics. There is no single indicator that says that we can trust one journal or publisher or an author. I hope to keep in mind all the above tips you might get to some sort of conclusion, also I completely worked on medical and paramedical scientific research. Hence these tips are from a point of personal perspective.

Do you have more specific tips on how to narrow it down further? Let me know in the comments section.

How to read a research article?

Before starting to read a research article, decide what is the purpose for your read. Is it because you are interested in understanding a particular subject in-depth? or because you wanted to pursue the project in that field?

If you want an answer to a particular question, just select any recent review article and understand the conclusions well. If you want for literature search, so that you can pursue your future projects in that field, then follow the below steps.

Deciding a journal:

Make a list of journals in your subject with impact factors and whether they are PubMed indexed. Due to the rampant increase in the numbers of predatory journals, we need to be very cautious about the source. Do not read or trust any information or research article from predatory journals.

Start with review articles: 

For a person who is a newbie to research, I would suggest reading review articles, because they are the compilation summaries of multiple articles. Also, they contain very less mathematical numbers. Hence it will be easy for the brain to comprehend it as a story. Even if you are not a newbie, sometimes if you feel there’s any difficult concept, going back and searching for a review article is always a good option.

Highlighting/Making notes: 

Start highlighting with a marker (either on-screen or paper). Because, once you start reading a complex article, authors use multiple analogies, explanations to make complex concepts into simple concepts. But the important idea of the project lies mainly in one sentence that explains the purpose and another sentence that explains the methodology. If you do not underline/highlight there are high chances that you forget the main purpose before engrossing completely into the methodology. Additionally,  If you are patient enough, you can even make notes of the articles.

Non-understandable bits prevent us from following the main ideas: 

A variety of specialists are needed to pursue a single subject of research. For example, if you are studying about one part of the eye cornea, you might require engineers to build machines to image cornea, you might require doctors to assess the images, you might require data analysts to understand how to convert these images into valuable information. Once all these people collaborate and publish an article, it is not so easy for a doctor or an engineer to completely understand all three components of the research.

Hence it is okay to ignore non-understandable bits of the paper and concentrate on information that is of your interest. Any published article is a result of years of work in that particular field by a group of scientists. It is natural, of it takes multiple reads to understand in detail. Hence, take it easy on yourself and give time for multiple reads.

Contact the authors:

Once you really think there is something in the paper that you especially need to understand to do further research, you can always ask the author. You will be doing them a favour by letting them know that it’s complex for a person like you. They also feel happy to know someone is really interested in learning their area of interest. 

The authors tend to assume significant background knowledge from readers. It is not wrong on their part since, if a person is really interested in that topic, they will definitely learn all the basics and then come to research articles.  Also, since every journal has a word limit, authors have to explain their research concisely. Hence understanding an article depends on how much interest the reader has to learn that particle subject.

Our recent article on Best ways to keep you updated with eye and vision research (part 2)-Research alerts

Order of a research article:

The simplest way to understand an article is to read in the order it is presented by the journal. Most of the journals follow similar templates. Skimming the abstract should get you clear whether the article is of any use to your area of interest. Once you feel it is related, you start reading the introduction, methodology, results, discussion and conclusion. Sometimes if you don’t understand even the purpose of the article, it is always wise to start with a discussion. Authors use simple words in discussion with fewer numerical values and also extrapolate the specific results of the project into the practical applications and implications. Also, looking at the graphs and images gives a simple idea of the main results.

All the information above is from personal experience and from the below references. Open to any other simples ways of reading a research article. Comment your suggestions below.

References:

• https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3687192/
• https://towardsdatascience.com/guide-to-reading-academic-research-papers-c69c21619de6
• https://www.owlnet.rice.edu/~cainproj/courses/HowToReadSciArticle.pdf