Review Question - QID 217166

Action	Numeric Key	Letter Key	Function Key
Choose 1	1
Choose 2	2
Choose 3	3
Choose 4	4
Choose 5	5
Submit Response			Enter
Previous Question			Left Arrow
Next Question		N	Right Arrow
Open/Close Bookmode		C
Open Image			Spacebar

Measures of Association Number Needed to Treat and Number Needed to Harm Number needed to treat (NNT) QID: 217166 (M1.OMB.4770)

QID 217166 (Type "217166" in App Search)

A team of researchers conducts a double-blind, placebo-controlled randomized clinical trial of the efficacy of empagliflozin in patients with heart failure with reduced ejection fraction (HFrEF). The authors randomized 3,730 patients into either the treatment (n=1,863) or placebo (n=1,867) cohorts. The primary composite outcome was cardiovascular death or hospitalization for exacerbation of heart failure. At 16 months of follow-up, the composite outcome occurred in 361 patients in the treatment cohort and 462 patients in the placebo cohort. A clinician is designing a clinical care pathway for treating patients with HFrEF, in which empagliflozin will be added to prescription guidelines for all patients with HFrEF. A process quality measure is designed to measure adherence to these guidelines. According to the results of the study, failing to treat how many patients with HFrEF with empagliflozin would result in 1 additional composite outcome (cardiovascular death or hospitalization for HFrEF exacerbation) over 16 months?

Type & Select Correct Answer

Type in at least one full word to see suggestions list

1.3

0/0

4.0

0/0

5.0

0/0

19.0

0/0

28.0

0/0

Select Answer to see Preferred Response

Review TC In New Tab

The number needed to harm (NNH) denotes the number of patients that would need to be exposed to some risk factor (e.g., failing to treat with empagliflozin) to result in a single additional outcome. The NNH is calculated as 1/absolute risk increase, which is calculated in this scenario as the risk of the outcome in placebo-treated patients (462/1,867) less the risk of the outcome in empagliflozin-treated patients (361/1,863): NNH = 1/absolute risk increase = 1/(absolute risk of outcome in placebo-treated patients - absolute risk of outcome in empagliflozin-treated patients) = 1/(462/1,867 – 361/1,863) = 19.

The NNH is directly analogous to the number needed to treat (NNT), which is defined as the number of patients who need to be treated to prevent a single adverse outcome. While the NNT is calculated to determine the beneficial effects of treatment, the NNH is calculated to determine the harmful effects of exposure to some risk factor (or failure to treat when it is indicated). Therefore, while the NNT is defined mathematically as 1/absolute risk reduction, the NNH is defined as 1/absolute risk increase. Note that an absolute risk is simply the probability of an event occurring over a defined period of time.

Incorrect Answers:
Answer 1: 1.3 is incorrect because it is obtained by the ratio of the absolute risk of outcome with placebo to the absolute risk of outcome with empagliflozin (after rounding), which is also known as the unadjusted risk ratio (RR). RR = absolute risk of outcome with placebo / absolute risk of outcome with empagliflozin = (462/1,867) / (361/1,863) = 1.28.

Answer 2: 4.0 is incorrect because it is obtained by a simple inversion of the absolute risk of outcome with placebo: 1 / (462/1,867) = 4. This quantity has no epidemiological value.

Answer 3: 5.0 is incorrect because it is obtained by a simple inversion of the absolute risk of outcome with empagliflozin: 1 / (361/1,863) = 5. This quantity has no epidemiological value.

Answer 5: 28.0 is incorrect because this quantity, in percentage terms, is obtained by the absolute risk increase divided by the absolute risk of outcome in the treatment group, which denotes the relative risk increase (RRI): RRI = absolute risk increase / absolute risk of outcome in the treatment group = (462/1,867 - 361/1,863) / (361/1,863) * 100 = 28%.

Bullet Summary:
The number needed to harm is defined as the number of patients who need to be exposed to a risk factor to result in 1 additional adverse outcome and is calculated as 1/absolute risk increase.