How to present an observational research publication

Guidance

a.       Stick to these headings TO SCORE your points. Mention the Headings to score your points.

b.       Be bold with your presentation – few people can match what you will be saying.

A.   Report the Paper

Title:

The title of my first paper is ……………………………….

Study design: Either of these statements

This is a prospective cohort study expected to produce level ….. evidence.

or

This is a retrospective cohort study expected to produce level ….. evidence.

or

This is a retrospective case control study expected to produce level ….. evidence.

Or

This is a retrospective cross-sectional study expected to produce level ….. evidence.

Or

The study design was unclear

Setting:

The study came of work of three hospitals in Sweden

 Or

The study was conducted in the community in Glasgow

Hypothesis testing:

This observational study was designed to explore the association between A and B.

Or

This observational study was designed to define the association between A and B

Or

This observational study was designed to generate a relationship between A and B

Or

This observational study was designed to quantify the relationship between A and B

Inclusion Criteria

The following were the inclusion criteria…………………………………………………

Exclusion Criteria

The following were the inclusion criteria…………………………………………………

Size of the Sample:

The study recruited ………….individuals. There was no requirement to calculate a sample size because a hypothesis was not being tested.

+

There were two [or three] groups. ] For example, Group 1 consisted of 7196 women with breast cancer, Group 2 consisted of 3200 women with colorectal cancer and Group 3 consisted of 1866 women with ovarian cancer].

Outcome Measures:

The main outcome measure was…………………………………

 Other outcomes included ……………………. 

Table 1

This shows an imbalance between the baseline characteristics.

[This is common with observational studies. It is expected that these will be minimised with techniques such as stratification, regression analysis o or some other statistical modelling technique]

Or

Table 1 shows a balance between baseline characteristics

Or

Table 1 shows a balance in the distribution of baseline characteristics or predictive factors or prognostic factors or confounding factors etc

Or, for example

Table 1 shows an imbalance in the distribution of stages of cancers of the breast, ovary and rectum.  

For example breast cancers without nodes constituted 53.2% of the breast cancers, while only 24.2% and 6.7% of ovarian and colorectal cancers were in the early stages.

The implication is that the outcome [mortality] might be affected when the three cancers are compared. [Note: later you need to say whether this was the case] 

Table 2

Pick up the most important findings – for example

Shows that the odds ratio of dying from breast cancer is 38% more than that of dying for ovarian cancer, after the woman has survived the first 5 years.

Table 3 and Table 4

Pick up the most important findings.

B.   What are the Strengths and Weaknesses

This information is usually in the methods and statistics sections.

There are four strengths in this study [List as many ‘strengths’ as you can find]

1.      The sample size was large

2.      The diagnosis was confirmed by review of histology or cross check of two databases

3.      The length of follow up was sufficiently long

4.      Follow up was complete.

There are 2 major weaknesses in this study [List as many ‘weaknesses’ as you can find]

1.      The study did not deal with the following biases:

a.       Recall bias

b.      Ascertainment bias

c.       Sampling bias

d.      Losses to follow up bias

[Note: the weaknesses in an observational study are what biases have not been mentioned or minimised.

C.   Did these weaknesses affect the results?

[This is an opinion, so you can say whatever you like]

The weaknesses did not significantly affect the results because.....................

The completeness of follow-up was a very important ‘strength’.

Or

The results were not affected by the imbalance of the distribution of the early stages of the cancers of the breast, ovary and colon.

D.   Summarise

[This is an opinion, so you can say whatever you like]

 The conclusions in this study are not justified because…….

Or

The conclusions in this study are well supported by the clear aims and strength of the study design. The authors made a huge effort to minimise the known biases of this study design. They acknowledge the biases that they could not overcome.