Safetxt for STIs

Safetxt for STIs

by | Sujit Rathod -
Number of replies: 19
Welcome to the 2022/23 edition of Epi in the News!

The first article of the year comes from the Daily Mail. I have colleagues who worked on this study, and so have been looking forward to these findings.

1) What is the study design?
2) What is the exposure? Outcome?
2b) What is the hypothesised mechanism of action (mediators)?

But researchers found 22.2% of those who received the Safetxts were reinfected with chlamydia or gonorrhoea. This compared to 20.3% in the group who did not receive the texts.

3) Are these figures prevalence or incidence figures? Who is in the numerator and denominator for these calculations?

4) Calculate and interpret a relative risk figure.

In reply to | Sujit Rathod

Re: Safetxt for STIs

by | NICOL REDZO -

1) What is the study design?
Two arm randomised controlled trial

2) What is the exposure? Outcome?
Exposure: previous chlamydia or gonorrhoea infection 
Outcome: Chlamydia or gonorrhoea re-infections

2b) What is the hypothesised mechanism of action (mediators)?

The group that receive safe sex text messages is less likely to be re-infected  with chlamydia or gonorrhoea

But researchers found 22.2% of those who received the Safetxts were reinfected with chlamydia or gonorrhoea. This compared to 20.3% in the group who did not receive the texts.

3) Are these figures prevalence or incidence figures? Who is in the numerator and denominator for these calculations?

These are incidence figures?

numerator= re-infections a

denominator=All youth in the study previously infected with chlamydia or gonorrhoea.

4) Calculate and interpret a relative risk figure.

RR=Risk of re-infection in exposed / risk of re-infection in unexposed group

22.2/20.3

=1.09
RR of 1.09 means the risk of chlamydia or gonorrhoea re-infection is the same for both groups.  

Regards
Nicol

In reply to | Sujit Rathod

Re: Safetxt for STIs

by | FATHIMA MINISHA -

Hi Sujit and everybody…

Just wrapped up all my modules and project for the MSc so now able to come back to this forum… thank you once again for the interesting read… 

1) its a prospective interventional study… the article does not mention the randomization part ( although i looked up bmj and found that it was randomized)

2) Exposure- receiving the Safe sex texts ( exposed are those who receive it and unexposed are those who dont)

Outcome- incidence of new sexually transmitted infections over the study period 

The mechanism of action expected would be that the text messages would encourage more of the participants to use safe sex techniques like condom use which in turn would reduce the reinfection… they were expecting the exposed group to have lesser incident risk.


3) these would be incidence figures… coz they are looking at the reinfection or new infections within that time period… numerator would be number of new infections in each group and denominator would be total number of participants in each group at the beginning of the study

4) the RR here would be 0.222/0.203= 1.1… so 10% higher risk of reinfection in the exposed group… of course we would need to calculate the CI to give more meaning to this number and see if this is due to chance or not… on calculating the 95% CI the values will be 0.99 to 1.20… p value 0.07… which means the true population RR can be anywhere between 0.99 to 1.20… which includes 1. This would translate as weak evidence against the null hypothesis of no difference… and that the difference noted is probably due to chance


Very interesting study… but I feel the population selected might not be the best for examining the effect of this intervention on the incidence of STI… people who have been infected once are generally more prone to get reinfected even with increased safety measures… maybe if they are able to recruit those who have not had an STI previously and then see if the texts help… that might show different results…

Fathima


In reply to | FATHIMA MINISHA

Re: Safetxt for STIs

by | Animesh Talukder -

Thank you, all, for sharing an interesting read and your thoughtful comments on it. I am Animesh, a new student in the MSc in Epidemiology course.

With regard to the selection issue, I wonder whether that would be a matter of concern at all since, I presume, similar subsets of the same sample were used across different treatment arms. Besides, the sample was randomised and that certainly attempted to make the groups comparable.

It would be wonderful to know your thoughts on this.

Cheers!

In reply to | Animesh Talukder

Re: Safetxt for STIs

by | Barun Kumar Singh -
Dear Animesh

thanks a lot for your thought, as this is behavioural kind of intervention, and behavioural change is often not straight forward and especially among young adults including adolescents. hence it would be interesting to know the age distribution of young and teens in both groups.

regards, barun
In reply to | FATHIMA MINISHA

Re: Safetxt for STIs

by | Sujit Rathod -
Welcome back Fathima!

You've made an interesting point about the eligibility criteria. I imagine those who already had an STI are at highest risk for a future STI, and so are higher priority for intervention.
In reply to | Sujit Rathod

Re: Safetxt for STIs

by | MAR ESTUPINAN FERNANDEZ DE MESA -

Hello All, 

Thanks for this very interesting exercise. 

I'm new to the forum and would appreciate some clarification of how this works:

- Do we have to answer questions only using the information included in the article? I couldn't find the reference to the BMJ paper and am not clear whether we should also check that to respond the questions. 

- I can see that students are submitting their responses to the forum, which is very helpful to see different views and reflections. However, will the correct answers be posted to help our learning?

Thanks, and best wishes, 

Mar

In reply to | MAR ESTUPINAN FERNANDEZ DE MESA

Re: Safetxt for STIs

by | Sujit Rathod -
Hi Mar -

Yes, my questions I ask are meant to be answered using the news article. That being said, sometimes I ask a question which is meant to reinforce how journalists gloss over important detail. (Fathima picked up on an important one, which is that the article doesn't specify that the exposure groups were randomly allocated). You are welcome and encouraged to check the source journal article.

You're welcome to post your own answers, to agree with a fellow student, or to provide your justification for an alternative answer.
In reply to | Sujit Rathod

Re: Safetxt for STIs

by | MAR ESTUPINAN FERNANDEZ DE MESA -
Thank you much, this is very helpful.

Some points from me:

1. Although the news article doesn't state the type of study, one can identify that researchers had two groups (a group who received the intervention, i.e. saftext, and a control group) and this was a prospective research (followed sample for 1 year). If the journalist wanted to show the strength/precision of the study, they'd have liked to say that this was a randomised study which is in higher position in the EBP hierarchy than cohort, case-control studies.

2. Exposure: receiving a safetxt
Outcome: incidence of chlamydia/gonorrhea reinfection at one year

2b. Hypothesis: safetxt would reduce the risk of chlamydia/gonorrhoea reinfection at one year

3. These are incidence figures (new reinfections at one year)
In each group:
Numerator: new infections
Denominator: total sample

4. The relative risk = RR (expose) / RR (non-expose) = 22.2%/20.3% = 0.222/.0203 = 1.09
This means that the exposed group (those who received the safetxt) were 1.09 (or 10% if we round the number) more likely to be reinfected than the control group (those who did not receive the safetxt).
I, however, would be cautious interpreting this figure just directly from the news article because it doesn't give you the precision/significance of this outcome and, therefore, it may be misleading. In this instance, I checked the paper and this shows OR=1.13, 95%CI 0.98 to 1.31, p=0.08. From these results, we can see that 95%CI includes 1 which means that the difference might be due to chance. Also important to highlight p > 0.05 level of statistical significance.

For me, it is also interesting to see how the journalist randomly picked another outcome, condom use, but we don't know why this was important, etc.

There isn't information about the characteristics of those who didn't participate, loss to follow up, effect of confounders, bias, the fact that women were over-represented, etc. This may be too much detail for this article; however, I'd say it's worth keeping in mind these points when we read these type of news to ensure we accurately interpret the results.

Thanks, and best wishes
In reply to | Sujit Rathod

Re: Safetxt for STIs

by | ANUSHA SREEJITH -
Hi Everyone,
This is a great exercise. I am new student in MSc. Epidemiology.

1) What is the study design?

Cohort study
2) What is the exposure? Outcome?
Exposure: Safe text

Outcome: chlamydia and gonorrhoea reinfection

2b) What is the hypothesised mechanism of action (mediators)?

It was hypothesised that Safetxt would reduce the risk of chlamydia and gonorrhoea reinfection.

3) Are these figures prevalence or incidence figures? Who is in the numerator and denominator for these calculations?

Incidence figures
RR= 22.2/20.3= 1.09

4) Calculate and interpret a relative risk figure.

People who receive safe text were 1.09 times more at risk to be reinfected than those who did not.
In reply to | Sujit Rathod

Re: Safetxt for STIs

by | SIH COLETTE -
Hello Sujit and everybody,

This was a very interesting read.

1) study design is a controlled trial with 2 arms: article does not say whether study was randomized or blinded to specific groups of persons

2) Exposure is a text messaging service promoting safe sex 
Outcome is new STIs infections (chlamydia, gonorrhoea, or “non-specific urethritis”)

2b) Hypothesized mediators were "encouraging participants to follow their STI treatment properly, including informing partners about their own infection, promoting condom use, and encouraging participants to seek STI testing before unprotected sex with a new partner."

But researchers found 22.2% of those who received the Safetxts were reinfected with chlamydia or gonorrhoea. This compared to 20.3% in the group who did not receive the texts.

3) The estimates presented are incidence risk figures. Numerator includes all new re-infections (chlamydia, gonorrhoea, or “non-specific urethritis”) and denominator is everyone at-risk of reinfection at he start of the study

4) Relative risk= incidence risk in exposed/incidence risk in unexposed = 22.2/20.3= 1.09
The Safetxt group had 1.09 times higher risk of STI reinfection compared to the non-Safetxt/control group

Kind regards,
Colette
In reply to | Sujit Rathod

Re: Safetxt for STIs

by | Harisd Phannarus -
Hello,
Let me try.

1) Prospective cohort study ... but if I have am original article, I would read the method section. It can be RCT when the author explained more.

2) Exposure: safetxt project which tailor to gender, sexual orientation. (dozens of texts in vary intervals)
Outcome: incidence of Chlamydia and Gonorrhea reinfection

2b) Hypothesis: Safetxt would reduce the risk of chlamydia and gonorrhoea reinfection in receiving safety group.

3) Incidence
Receiving safetxt group:
Numerator: case of reinfection in people who receiving safetxt
Denominator: total people who receiving safetxt at the start period
Non-receiving text group:
Numerator: case of reinfection in people who did not receiving safetxt
Denominator: total people who did not receiving safetxt at the start period

4) 222/203 = 1.094 = relative risk
Relative to people who did not receive safetxt, people who receive safetxt had 1.094 times the risk of chlamydia and gonorrhea reinfection through the follow up period

Thanks
Harisd
In reply to | Harisd Phannarus

Re: Safetxt for STIs

by | Sujit Rathod -
Welcome Harisd! And really well done for breaking down the numerators and denominators for the incidence figures.

I'd be curious to know if the journalist understood that this study was an RCT, and if so, why they chose to omit this detail.
In reply to | Sujit Rathod

Re: Safetxt for STIs

by | Harisd Phannarus -
Let me try again,

First, they have some bias to make their news look more interesting by omit scientific words or they try to make it sounds interesting by let some questionable hints in their news.

Second, RCT and cohort is not a general words to understand by lays person.

Third, other information in the news might mean RCT although it did not type in the news.

Thanks
In reply to | Sujit Rathod

Re: Safetxt for STIs

by | Barun Kumar Singh -
Based on the information provided in the news about the outcome of study without going through details in the published paper.

1. Study design is prospective cohort study with a kind of behavioural (health promotion) intervention through the text-customised according to the needs of target population.
2. Exposure in this study is Text messaging services promoting safe sex, and the outcome of study is reinfection (Sexually transmitted infection).
Hypothesised mechanism of action
Young adults and teen having previous history of STI after exposure to safe sex messages will be less likely to suffer from reinfection. However result came in opposite direction as those Young adults and teens not receiving any sort of text message developed relatively the reinfection at lower rate.
3. As the population have been followed after the intervention and outcome assessors are waiting for the specific outcome to happen, this is explicitly incidence figure (incidence risk). hence in numerator would be no. of young adults/teens (16-24 yrs) developed STI reinfection over the study period/total number of same age group population at the risk of reinfection.
4. relative risk=22.2/20.3=1.03
one group of young adult and teens (16-24 years) exposed to safe text message were 1.09 times more likely to suffer from STI-reinfection compared to other group receiving no intervention.
In reply to | Sujit Rathod

Re: Safetxt for STIs

by | RANMINI SUMUDITA KULARATNE -
1) What is the study design? Interventional study with an experimental arm and control arm. Not sure whether randomised, and how this was done.

2) What is the exposure? Outcome?
Exposure: educational test messaging around STI prevention and services.
Outcome: STI re-infections (gonorrhoea and chlamydia) or symptomatic urethritis during study period (how was this ascertained in those without symptoms; was there regular screening of asymptomatic participants? And at what intervals in the two groups?)

2b) What is the hypothesised mechanism of action (mediators)? Hypothesized mechanism of action: educational messaging would promote safe sex behaviour, and increased testing and treatment to prevent community spread of STI and re-infection.



3) Are these figures prevalence or incidence figures? Who is in the numerator and denominator for these calculations? Incidence of re-infections.
Numerator: number of re-infections - testing positive for STI pathogens (N. gonorrhoeae/ C. trachomatis) and those with symptomatic urethritis in each group during the study period. Were they screened at enrolment to ascertain that they were infection-free?

Denominator: total number of participants enrolled in each group ("at-risk persons").

4) Calculate and interpret a relative risk figure. Relative risk = 22.3/20.3 = 1.10 - 1.10 times higher risk in interventional arm. No confidence intervals or hypothesis test to ascertain evidence against null of no difference in arms.
In reply to | Sujit Rathod

Re: Safetxt for STIs

by | Mukovhe Phanguphangu -
1. This is an experimental study, likely a randomised control trial, as there is a cohort group that received intervention (texts) and a control group (did not receive the texts)

2. The exposure was receiving the texts (cohort group) and the outcome was whether they were re-infected with the STIs

2b. Receiving texts would lead to safer sexual practices and thus reduce the risk of reinfections

3. These are incidence figures - new "infections" during the study period/one year.

4. Relative Risk = 22.2/20.3 = 1.09 - those who received the text had 1.6 times the risk of reinfection compared to those who did not.
In reply to | Sujit Rathod

Re: Safetxt for STIs

by | Lilian Ngwongem Ngwana Ep Banmi -

1) study design - interventional study (Randomized controlled study). The interventional arm received Safetxt messages and control arm didn't. 

2)exposure - Safetxt messages

Outcome - reinfection by chlamydia, gonorrhoea 

2b)  Hypothesised mechanism of action : Safetxt messages will help youths (16-24) prevent re-infection through promoting safe sex practises like using condoms when having sex and informing one's sexual partner of any history of STIs.

3) incidence figures.

Numerator : all new re-infected cases in the study

Denominator : everyone at risk of re-infection at the start of the study.

4) Relative Risk (RR) = risk in exposed/risk in unexposed . = 22.2/20.3 = 1.094

Relative to those who didn't receive text messages, those who did had 1.09 times the risk of being re-infected with Chlamydia or gonorrhoea 


Accessibility
Reset All

Background Colour

Font Face

Font Size

1

Text Colour