NCCMT - URE - Effectiveness of Interventions - Understanding the Number Needed to Treat

Summary

The video script discusses the complexities of decision-making in public health, especially when selecting interventions with limited resources. It introduces the model of evidence-informed decision-making and emphasizes the importance of considering the magnitude of the problem, available resources, and societal preferences. The script also explains how to calculate and interpret the Number Needed to Treat (NNT), a measure that helps public health decision-makers understand the impact of interventions on a population level. It provides examples of calculating NNT for interventions like social media-based pregnancy prevention and car seat installation clinics, highlighting the balance between benefits and potential harms.

Takeaways

• 💡 Public health decision-making can be complex due to finite budgets and multiple effective interventions.
• 🔍 The National collaborating Center for methods and tools suggests using evidence-informed decision-making models in public health.
• 📊 Decision-makers should consider the magnitude of the problem, available resources, and societal preferences when choosing interventions.
• 🧮 Number Needed to Treat (NNT) is a useful measure to calculate how many people need an intervention for one to benefit, aiding in decision-making.
• 📈 NNT is calculated by dividing 1 by the Absolute Risk Reduction (ARR), which is the difference in outcome rates between control and intervention groups.
• 🤔 Calculating NNT helps interpret research results, showing how likely and impactful an intervention is on a population level.
• 🚫 NNTs are not always reported in studies, so public health professionals may need to calculate them using provided data.
• 👶 The script provides an example of calculating NNT for a social media-based teenage pregnancy prevention campaign.
• 🚗 Another example is given for a car seat installation clinic, demonstrating how NNT can be calculated and interpreted for different interventions.
• ⚠️ Number Needed to Harm (NNH) is also introduced to consider potential negative outcomes of interventions, balancing benefits and risks.
• 💉 Number Needed to Immunize (NNI) is mentioned as a related concept, important for immunization programs to prevent diseases.

Q & A

What is the primary challenge faced by health departments when choosing interventions?

The primary challenge is deciding which intervention to implement among many effective ones due to a finite budget.

How does the model of evidence-informed decision-making assist in public health decisions?

It assists by considering not only research evidence but also the magnitude of the problem, resources needed, and societal and political preferences.

What additional factors do decision-makers need to consider when making decisions across the spectrum of public health services?

They need to compare the benefits and resources required for different interventions to determine which delivers the greatest impact for the investment.

What are the three valuable concepts that can help in decision-making but do not tell you everything about an intervention's effectiveness?

Relative risks, odds ratios, and confidence intervals help estimate the likelihood and size of an effect but do not indicate who the intervention worked for or how many people experienced the desired outcome.

What is the Number Needed to Treat (NNT) and how is it calculated?

NNT is the number of people who need to receive an intervention for one person to experience a positive outcome. It is calculated as 1 divided by the absolute risk reduction (ARR).

Why is it important to calculate the NNT for public health interventions?

Calculating NNT provides a clear understanding of how many people need to be exposed to an intervention to achieve a single positive outcome, aiding in resource allocation and program decision-making.

In the example of teenage pregnancy prevention using social media, what was the calculated NNT?

The calculated NNT was 67, meaning 67 teenagers need to receive the intervention to prevent one pregnancy over a one-year period.

What is the significance of a smaller NNT in public health interventions?

A smaller NNT indicates that fewer people need to be exposed to the intervention for one positive outcome, suggesting a greater potential impact at a population level.

What is the Number Needed to Harm (NNH) and how does it differ from NNT?

NNH is the number of people who need to be exposed to an intervention for one person to experience an unintended negative outcome. It differs from NNT by focusing on harm rather than benefit.

How can the concept of NNT be applied to immunization programs?

The Number Needed to Immunize (NNI) is calculated similarly to NNT and tells us how many people need to be immunized to avoid one negative outcome, such as death from an infection.

Why are NNTs, NNHs, and NNIs considered powerful tools for public health decision-makers?

These measures provide clear, quantifiable information on the impact and potential harm of interventions, aiding decision-makers in making informed choices about public health programs.

Outlines

00:00

📊 Decision-Making in Public Health: Evidence-Informed Approach

This paragraph discusses the challenges faced by health departments in choosing which interventions to implement, given limited budgets. It introduces the model of evidence-informed decision-making in public health, which considers not only research evidence but also the magnitude of the problem, resources needed, and societal and political preferences. The paragraph emphasizes the complexity of decision-making, especially when comparing interventions across different public health services. It also introduces the concept of Number Needed to Treat (NNT), a statistical measure that helps determine the number of people who need to receive an intervention for one person to experience a positive outcome. The paragraph provides an example of how to calculate NNT using hypothetical data on teenage pregnancy rates and the effectiveness of social media-based pregnancy prevention messages.

05:03

📈 Calculating and Interpreting NNT in Public Health Interventions

This paragraph delves into the process of calculating the Number Needed to Treat (NNT) and its interpretation in the context of public health interventions. It explains the formula for calculating NNT, which is 1 divided by the Absolute Risk Reduction (ARR). The ARR is the difference in outcome rates between the intervention and control groups. Using the example of a social media campaign to reduce teenage pregnancy, the paragraph demonstrates how to calculate NNT and interpret its meaning. It also discusses the implications of a low NNT value, indicating a relatively effective intervention. The paragraph further explores other related measures such as Number Needed to Harm (NNH) and Number Needed to Immunize (NNI), which provide additional insights for decision-makers when considering the potential negative outcomes or benefits of interventions.

10:05

🔍 Utilizing NNT in Public Health Decision-Making

The final paragraph emphasizes the practical application of NNT in public health decision-making. It highlights that with practice, decision-makers can become proficient in calculating and interpreting NNT, which can inform decisions about public health interventions. The paragraph suggests that NNT, along with other measures like NNH and NNI, can provide valuable information to help balance the benefits and potential harms of interventions. It concludes by stating that these calculations are straightforward and can be performed using data typically available in published studies, thus empowering public health professionals to make informed choices.

Keywords

💡Evidence-Informed Decision-Making

Evidence-informed decision-making is a process that involves using the best available research evidence to inform decisions. In the context of the video, this concept is crucial for public health officials who must choose interventions with limited resources. The video suggests using a model from the National collaborating Center for methods and tools to consider not only research evidence but also the magnitude of the problem, available resources, and societal preferences.

💡Intervention

An intervention in public health refers to any action or program designed to improve health outcomes in a population. The video discusses how to choose among various interventions when a health department has a finite budget. Examples include a social media campaign to reduce teenage pregnancy or a clinic to promote correct car seat installation.

💡Magnitude of the Problem

The magnitude of the problem refers to the scale or severity of a health issue within a community. The video emphasizes that decision-makers should consider this when choosing interventions, as it affects the potential impact and resource allocation. For instance, addressing a higher teenage pregnancy rate in a community would be a priority.

💡Resources

Resources in the context of the video pertain to the financial and human capital required to implement and maintain public health interventions. The video highlights that decision-makers must evaluate the resources needed for each intervention, as this influences the feasibility and sustainability of the chosen strategy.

💡Societal and Political Preferences

Societal and political preferences refer to the values, beliefs, and priorities of a community and its leadership that can influence decision-making in public health. The video suggests that these preferences should be considered when selecting interventions, as they can affect the acceptance and effectiveness of the chosen strategy.

💡Relative Risks, Odds Ratios, and Confidence Intervals

These are statistical measures used in research to estimate the likelihood and effect size of outcomes. The video mentions that understanding these concepts is valuable for decision-makers to assess the statistical significance and the size of the effect of an intervention. However, these measures do not provide information on who the intervention worked for or the number needed to treat (NNT).

💡Number Needed to Treat (NNT)

The NNT is a measure defined as the number of people who need to receive an intervention for one person to experience a positive outcome. The video uses NNT to illustrate the practical application of calculating how many individuals need to be exposed to an intervention to achieve a single desired outcome, such as preventing a pregnancy or correctly installing a car seat.

💡Absolute Risk Reduction (ARR)

ARR is the difference in rates of an outcome between the control group and the intervention group. The video explains that ARR is used to calculate NNT, which helps in understanding the effectiveness of an intervention. It is calculated by subtracting the event rate in the intervention group from that of the control group.

💡Number Needed to Harm (NNH)

NNH is the counterpart to NNT, indicating the number of people who need to be exposed to an intervention for one person to experience an undesired negative outcome. The video suggests calculating NNH to weigh the benefits against potential harms of an intervention, such as feeling isolated after an anti-bullying campaign.

💡Number Needed to Immunize (NNI)

NNI is a measure similar to NNT and NNH, used to determine how many people need to be immunized to prevent one negative outcome, such as death from an infectious disease. The video mentions NNI as an example of how these measures can inform public health decisions by providing a clear numerical estimate of the impact of interventions.

Highlights

The challenge of selecting the most effective public health intervention given a finite budget.

The model of evidence-informed decision-making in public health.

Considering the magnitude of the problem, resources, and societal preferences in decision-making.

The complexity of choosing interventions across different public health programs.

Using relative risks, odds ratios, and confidence intervals in decision-making.

The importance of understanding who an intervention worked for and the desired outcomes.

Introduction to the Number Needed to Treat (NNT) as a decision-making tool.

Definition and calculation method of NNT in public health interventions.

Example of calculating NNT for a social media-based teenage pregnancy prevention program.

Interpretation of NNT results and their implications for public health interventions.

Calculating NNT for a car seat installation clinic intervention.

The significance of a smaller NNT in indicating a greater potential impact at a population level.

Introduction to the Number Needed to Harm (NNH) for assessing negative outcomes.

Example of calculating NNH for an anti-bullying campaign.

The concept of Number Needed to Immunize (NNI) in the context of vaccination programs.

Practical application of NNT, NNH, and NNI in public health decision-making.

The simplicity of calculating NNTs with data from published studies.

Building confidence in calculating and interpreting NNTs for public health interventions.

Transcripts

have you ever found yourself in a

situation where you have to decide which

intervention among many you want to

implement at your health department you

have looked at the research evidence and

even though a number of interventions

have been shown to be effective in

practice you have a finite budget and

can only choose one how do you decide

which one to implement you may have used

the model of evidence informed

decision-making in public health

suggested by the National collaborating

Center for methods and tools to assist

you in making your decision if you were

to use this model in addition to looking

at the research evidence you would also

look at the magnitude of the problem in

your community the resources needed to

implement and maintain the intervention

as well as the societal and political

preferences choosing the best

combination of interventions can be very

challenging even when the decision is

limited to one problem within one

program such as implementing a vaccine

promotion strategy used in social media

the decision-making process becomes even

more complicated when decisions across

the spectrum of public health services

are being made for example strategic

planning initiatives in these instances

decision makers may have to compare the

benefits and resources required to

maintain an environmental health

intervention such as monitoring swimming

conditions at public beaches to a

chronic disease prevention intervention

such as a healthy eating campaign given

limited resources decision-makers may

need to consider carefully across all

public health programs which

interventions deliver the greatest

impact for the required human and

financial investment please view the

other videos in this series to learn how

to use relative risks odds ratios and

confidence intervals in your

decision-making these valuable concepts

will give you an estimate of how likely

the result is whether it is

statistically significant and how large

the effect is what they don't tell you

is who an intervention worked for or how

many people had the desired outcome they

also don't tell you how many people need

to receive the intervention in order to

see one person with the desired outcome

to answer these questions we can use a

Mystikal measure called number needed to

treat or NNT it is defined as the number

of people who need to receive an

intervention in order for one person to

experience a positive outcome or to

avoid a negative one nnts can only be

calculated when there are only two

possible options for the outcome being

measured for example initiating breast

feeding or not initiating breast feeding

currently an antes are not usually

reported in the public health literature

but they are easy to calculate let's

take a look at some public health

examples to see just how easy it is to

calculate and interpret them remember

the data presented here are hypothetical

and are for demonstration purposes only

let's say data were presented at a

senior management meeting illustrating

that teenage pregnancy rates are higher

in your community than the provincial

average you've been asked to make

recommendations to address this issue

after searching health evidence org you

find a recently published systematic

review reporting that pregnancy

prevention messages disseminated using

social media lead to a statistically

significant reduction in teenage

pregnancy although a reduction in

teenage pregnancy is the outcome you

were hoping to see you wonder how many

teenagers need to receive the messages

in order for one pregnancy to be

prevented given there are only two

options for the outcome pregnant or not

pregnant you know that it is possible to

calculate an N NT unfortunately an N NT

is not reported in the systematic review

so you'll need to calculate it yourself

the formula for calculating and n n T is

1 divided by the absolute risk reduction

also referred to as an AR AR it sounds

complicated but it really isn't an AR R

is simply the difference in rates of the

outcome in the control group compared to

the intervention group let's do the math

say we have 1,000 teenagers in the

intervention group who receive weekly

messages over a one-year period about

pregnancy prevention and another 1,000

teenagers in a control group who do not

after one year we see how many teenagers

became pregnant in each

we find there were 15 pregnancies in the

intervention group and 30 pregnancies in

the control group the rate of pregnancy

in the intervention group is the number

of pregnancies divided by the total

number of teenagers in the intervention

group in this case 15 divided by 1,000

which equals 0.01 5 the rate of

pregnancy in the control group is 30

divided by 1000 or 0.030 we calculate

the absolute risk reduction by

subtracting 0.015 from 0.03 0 which

gives us an AR AR of 0.01 5 as mentioned

earlier to calculate the number needed

to treat we divide one by the 8 RR so 1

divided by 0.01 5 gives us an N NT of

66.7 because we can't have half a person

pregnant or half a pregnancy we round

the n NT up to the nearest whole number

so our n NT is 67 now we need to

interpret what an N NT of 67 means it

means we would need to expose 67

teenagers to the intervention in order

to prevent one pregnancy over a one-year

period this would constitute a

relatively effective intervention as few

teenagers need to receive the

intervention in order for one negative

outcome teenage pregnancy to be avoided

let's look at another example suppose

you are planning a strategy to increase

the number of properly installed car

seats a recently published meta-analysis

reported on the effectiveness of holding

clinics to promote the correct

installation of car seats the

intervention consisted of staff

assessing whether car seats were

installed correctly and teaching people

how to correctly install them the

control group received standard practice

where people could call and request

information on correct installation one

month after the intervention cars were

assessed to identify correctly installed

car seats following data were reported

in the meta-analysis among 500 people

exposed to the intervention 225

correctly installed their car seats

compared to 195 in the control group we

calculate the event rate in the

intervention group by dividing the

number of correctly installed car seats

by the total number of people in the

intervention group so 225 divided by 500

which equals 0.45 using the same formula

we divide 195 by 500 which gives us an

event rate in the control group of 0.39

to calculate the ARR we subtract 0.39

from 0.45 which gives us an AR r of 0.06

the N NT is calculated by dividing 1 by

0.06 which gives us an N NT of sixteen

point six seven we round that up to the

next whole number giving us an N NT of

seventeen which means that 17 people

need to attend the clinic in order for

one car seat to be correctly installed

one month later now that we have

determined how many people need to be

exposed to these two interventions in

order to either obtain one positive

outcome or avoid one negative outcome we

can identify both the financial and

human resources that will be required to

achieve a meaningful outcome for the

whole population while there are other

factors to consider in interpreting NN

teas such as the underlying risk of an

outcome in specific populations a

general understanding of how to

calculate and interpret them will

provide Public Health decision-makers

with powerful information upon which

they can make difficult program

decisions generally the smaller the NNT

the greater potential impact there will

be at a population level sometimes an

intervention can produce unanticipated

negative outcomes or harms which should

be considered when making a decision to

assess the magnitude of negative

outcomes we calculate the number needed

to harm or NNH the nnh tells us how many

people would need to be exposed to an

intervention in order for one unintended

negative outcome to be observed for

example let's say we are considering an

anti-bullying camp

the evidence indicates that the campaign

is effective in reducing the number of

bullying episodes however more children

exposed to the intervention report

feeling isolated compared to those not

exposed you can calculate the N and H to

determine how many students would need

to be exposed to the intervention for

one student to feel more isolated this

will help you weigh the benefits on an

intervention against the potential harms

one final use of this calculation is the

number needed to immunize or nni it is

calculated the same way as both the n n

h and n n t and tells us the number of

people who need to be immunized in order

to avoid one negative outcome such as

death as a result of infection you can

see how n n T's would be helpful in

adding information related to how many

people would need to be exposed to an

intervention in order for one person to

experience the outcome of interest

number needed to harm and number needed

to immunize also provide useful

information to decision makers along

with number needed to treat they are

simple to calculate and most published

studies provide sufficient data on

outcome measures so that you can easily

calculate them yourself with a little

practice you will not only become

confident in calculating n n T's but

also interpreting them and using them to

inform decisions about public health

interventions