Implementation of a nurse-initiated

pain protocol

Clinical Bottomline

A total of 13 studies evaluated the effects of implementing a nurse-initiated pain protocol. Of these studies, four had level of evidence 2b and nine had level of evidence 4.

In these studies several populations were evaluated:

- traumatic injuries (6 studies: 3009 patients)

- renal or biliary colic (1 study: 157 patients)

- abdominal pain (1 study: 200 patients)

- pain of any cause / mixed study populations (5 studies: 4296 patients)


As drug choices, 11 studies used opioids and 2 only used oral non-opioid analgesics. Furthermore, besides opioids, 1 pain protocol also permitted IV esketamine and 1 protocol utilized inhaled methoxyflurane. 

Taken together, it can be concluded* that...

1. a nurse-initiated pain protocol increases pain assessment

Level of recommendation C

The authors of 3 studies mentioned pain assessment and utilization of pain scores as endpoints. In 2 studies this increased from 2.9% to 31.8% and 19% to 81%, respectively. In the 3rd study, no numbers were reported, only the statement that pain assessment increased (as part of improved triage nurse performance).

2. a nurse-initiated pain protocol increases analgesic administration

Level of recommendation B

Administration of pain meds increased in all (6) studies in which this endpoint was evaluated. Baseline proportions varied between 10.3% - 46% and after implementation these numbers were 31.7% - 100%.

3. a nurse-initiated pain protocol leads to clinically relevant pain reduction

Level of recommendation C 

In 10 studies, pain reduction was measured using different methods, different pain scores and different time-points at which pain was assessed and reassessed. Absolute pain reduction was clinically relevant (at least 13 mm on a VAS scale) in 7 studies. A 3-point NRS reduction was reported in 53.2% of patients in 1 study and 1 study found a 30% NRS decrease in 77% of patients. Pain reduction was not clinically relevant (0.9 NRS points) in 1 study. 

4. a nurse-initiated pain protocol decreases time to analgesia

Level of recommendation C 

Time to analgesia decreased in 8 studies (compared to pre-implementation or to physicians' administration) with a mean decrease of 44.9 minutes. It increased (non-significantly) in 1 study (5 minutes) and was measured without any comparison in 1 study (mean 5.9 minutes). 

5. a nurse-initiated pain protocol is safe to use

- when IV fentanyl is administered

- leads to minor adverse events with inhaled methoxyflurane

- leads to minor adverse events with IV morphine

Level of recommendation C

Four studies reported the occurrence of adverse events, of which 2 used titrated IV morphine (adverse events in 4% and 12.9%, mainly nausea, vomiting, dizziness, hypotension and sedation). One study reported no adverse events using IV fentanyl and 1 study found an incidence of 64% of minor adverse events (mainly dizziness and drowsiness), utilizing inhaled methoxyflurane (and oral opioids). None required interventions.

6. a nurse-initiated pain protocol leads to no difference in Emergency Department length of stay

Level of recommendation B

This outcome was mentioned in 2 studies with level of evidence 2b. Both studies did not find statistically significant differences in total length of stay in the Emergency Department (5.2 versus 5.5 hours and 111 versus 105 minutes).

7. a nurse-initiated pain protocol increases patient satisfaction

Level of recommendation C

Two level 2b studies could not show statistically significant differences in patient satisfaction and 1 did (in favor of the after-implementation phase). Increased satisfaction was reported in a level 4 study and high satisfaction scores (>80 on a 100-VAS scale) were presented without comparison in 1 study.

* Disclaimer:

Included studies were different in several items, such as timing and methods of measurement of several endpoints; study population; baseline situation and pain medication. Due to these large differences in study methodology, results cannot be pooled and the level of recommendation should be regarded as an estimation of the true effect. The levels of recommendation were derived according to the OCEBM levels of evidence.