Ideas are everywhere but they have to be converted to a valid RQ. The choice of RQ may be made from the evaluation of previous studies, one’s own experiences, from topic of interest or by the need of the time [1]. Most of the ideas for research come from one’s perceived gaps in the existing knowledge of a topic. These gaps could be the lack of clear conclusions or insufficient results from previous studies. Identification of such gaps could be taken up as a RQ which would then be used to build on the previous research. Ideas could also be developed from observations made on the previous work. Systematic analysis of this observation after brainstorming may also be one of the methods of finding the correct RQ. Unprecedented circumstances may also yield multiple RQs on a single topic.

As a rule of thumb, the broad ideas themselves do not form a RQ. They should be chiselled to yield one. This is achieved by an exhaustive and critical analysis of the broad idea by a thorough literature research. The brainstorming would include searching for studies conducted in past on same topic, identifying the lacunae in the existing knowledge or need to replicate the question in different settings, thinking of a hypothesis, and then generating a RQ [2]. Although multiple RQs may stem from a broad idea, it is best to choose only one primary RQ for a particular project (choosing the best RQ is described later). The RQ is the interrogative form of the solution which we are looking. Simply put, it is the purpose of study written in a question format [3]. It has to be specific, focused and clearly defined in terms of population targeted, planned intervention, outcome etc.

For example, while working in pediatric neurology ward, a student observed that there were a large number of children coming with breakthrough seizures. He brainstormed with peers to analyze the reasons for poor control of seizures in children with epilepsy. One of the RQ which was formed was ‘What are the causes of non-compliance of treatment among children with epilepsy at a tertiary-care center?’

The RQs are broadly classified into descriptive and inferential questions based upon the purpose, objective, and clinical context of the study [4]. Various types of RQs along with examples are given in Table I.

A RQ has multiple components. The most common approach to address these is referred to as PICO approach, the acronym standing for Population, Inter-vention/Exposure, Comparator and Outcome [5]. Some-times Timeframe, and Effect Size are also added to call it as PICOTES approach which makes it more compre-hensive. Although most commonly used in the inferential RQ, some of the elements (P and O) are also mandatory in the descriptive RQ. The salient features of PICO elements are explained in Box I.

Some of the framed RQs, although inviting and lucrative are not possible to do. A good RQ once framed, should be assessed using the FINER approach given by Hulley, et al. [6]. It should be feasible, interesting, novel, ethically sound, and relevant. It should be clearly stated and appropriately complex. This audit should be done before the start of the study to avoid wastage of manpower and resources. Doing a pilot study may also help in unearthing real time issues. These attributes are described in the Table II. The RQ should be chosen so that it addresses issues common to clinical setup of a particular area, builds upon previous gaps, ethically sound and is doable by the researcher in the given time limit.

The RQ should be written in one’s own wording, be appropriately complex and should not be very broad. It should be in question format and complete in itself [7]. One example of research question would be "How efficacious is intravenous phenobarbitone in comparison to intravenous levetiracetam, as a first line drug, in controlling neonatal seizures?" RQ should NOT be framed to provide answer in YES or NO. Binary outcome framing is not a correct approach as decisions in medicine are based on probabilities and cannot be absolute (0 or 100). Some examples of errors in framing RQ are shown below.

Incorrectly framed: In preterm neonates less than 30 weeks, does formula feeding predispose to necrotizing enterocolitis (NEC)?

Reason: RQ should not have a binary outcome.

Correct: What is the relationship of formula feeding with NEC in preterm neonate less than 30 weeks?

Incorrectly framed: What measures are being taken to prevent postoperative wound infection undergoing emergency surgeries?

Reason: Too simple and broad, should be appropriately complex.

Correct: What are the risk factors associated with deep surgical site infection following laparotomy for acute perforation peritonitis?

The writing of RQ should be followed by the framing of study outcomes and objectives. Outcomes are the measurable endpoints of the objectives which are monitored during the study and occurrence (or absence) of these indicate that result has been achieved. They should originate and be in alignment with the study objectives. It is of utmost importance to a priori define an outcome as well as a standard validated method to measure it as a particular entity may be definable or measurable by variable methods. The outcomes should be clearly written such that anyone can easily understand the nature of what is being measured and replicate the measurements at their research settings. The ideal outcomes should be reproducible under same conditions and have minimum inter observer difference, and valid i.e. should measure what they intend to measure. The primary outcome is the most important measurable endpoint and should correspond to the primary objective. The study design and sample size are also based upon the primary outcome of the study. The additional measurable endpoints pertaining to the secondary objectives are called as secondary outcomes. Usually outcomes are analyzed independently but ‘composite outcomes’ can be used when the individual occurrences are rare, correlate with each other or a combination is more informative as compared to an isolated outcome [12]. Mortality or bronchopulmonary dysplasia is a composite outcome used in RQs pertaining to respiratory support in neonates; death or presence of a severe neuro-development impair-ment (blindness, deafness, motor or cognitive disability) by two years of age is another example of composite outcome used in assessing long term impact of various interventions in extremely low birth weight neonates.

The RQ tells us that what we are planning to do. The objectives of the study give us accurate description of the steps about how we are going to achieve what we had thought of in the RQ. The objectives serve as milestones for the ultimate goal [8,9]. The objectives are classified as primary and secondary. The primary objective is the most important endpoint and should reflect the RQ. The additional endpoints which we want to study are termed as secondary objectives. It is better to have only one primary objective for a particular RQ. Secondary objec-tives can be multiple; however, should not exceed five.

Objectives are written in single infinitive sentence format starting with the word ‘To’ using SMART format. They should be specific, measurable, achievable, realistic and time defined [10]. The action verbs from Bloom’s measurable verb list e.g. determine, compare, verify, establish etc. should be used in the framing of the objective [11]. The objectives dictate the type of study design and help in developing the methodology section of the protocol. We should avoid writing too many objectives and should have a clear flow between the RQ and objectives as shown in Table III. The relationship between the individual objectives should have a synergistic impact [8,9].

The formulation of RQ is most critical and fundamental part of a study which should be done carefully and scientifically. Framing of objectives is a natural corollary after finalizing the research questions. Defining the outcome (mostly in the form of a numerical measurable expression) is a necessary intermediary between the two processes. A summary of the process of developing a research question and translating it into outcomes and objectives is exemplified (Table IV).

Contributors: SKD: literature research, drafted and revised the manuscript, approved final version of manuscript PG: conceived the idea, supervised the manuscript, reviewed the manuscript, finalised and approved the manuscript. Both authors have contributed to, designed and approved the manuscript.

Funding: None; Competing interest: None stated.

1. Jones R. Choosing a research question. Asia Pac Fam Med. 2003;2:42-4.

2. Garg R. Methodology for research I. Indian J Anaesth. 2016;60:640-5.

3. Lipowski EE. Developing great research questions. Am J Health Syst Pharm. 2008;65:1667-70.

4. Cañón M, Buitrago-Gómez Q. The research question in clinical practice: A guideline for its formulation. Rev Colomb Psiquiatr. 2018;47:193-200.

5. Richardson WS, Wilson MC, Nishikawa J, Hayward RS. The well-built clinical question: A key to evidence-based decisions. ACP J Club. 1995;123:A12-13.

6. Hulley SB. Designing clinical research, 3rd ed. Lippincott Williams & Wilkins; 2007.

7. Ratan SK, Anand T, Ratan J. Formulation of research question - Stepwise approach. J Indian Assoc Pediatr Surg. 2019;24:15-20.

8. Farrugia P, Petrisor BA, Farrokhyar F, Bhandari M. Research questions, hypotheses and objectives. Can J Surg. 2010;53:278-81.

9. Hanson BP. Designing, conducting and reporting clinical research. A step by step approach. Injury. 2006;37: 583-94.

10. Doran GT. There’s a S.M.A.R.T. way to write manage-ment’s goals and objectives. Manage Rev. 1981;70:35-36.

11. Adams NE. Bloom’s taxonomy of cognitive learning objectives. J Med Libr Assoc. 2015;103:152-3.

12. Velentgas P, Dreyer NA, Wu AW. Outcome definition and measurement [Internet]. Developing a protocol for observational comparative effectiveness research: A user’s guide. Agency for Healthcare Research and Quality (US); 2013. Accessed September 28, 2020. Available from: https://www.ncbi.nlm.nih.gov/books/NBK126186/