Effects of continued use of information interventions on opioid prescribing

How to cite

Ahomaki, Iiro; Saastamoinen, Leena; Böckerman, Petri; Pehkonen, Jaakko. 2021. "Effects of continued use of information interventions on opioid prescribing." AEA RCT Registry. AEARCTR-0008439. https://doi.org/10.1257/rct.8439-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Iiro Ahomaki* , Leena Saastamoinen , Petri Böckerman , Jaakko Pehkonen

* Primary investigator

Registry ID
AEARCTR-0008439
DOI
https://doi.org/10.1257/rct.8439-1.0
First registered
Last updated
Status
In development
Trial dates
2019-05-29 – 2022-12-31
Countries
Private
JEL codes
I10, I12, I19
Registry record
http://www.socialscienceregistry.org/trials/8439

Abstract

We study the effects of an information intervention on opioid prescribing and patients’ health outcomes. The intervention was a personal letter sent to physicians who had prescribed strong opioid painkillers (oxycontin or fentanyl) to a patient who had purchased these medicines for at least three months. These physicians were identified and randomized into treatment and control group. The letter was sent to physicians in the treatment group (n=3,689) on May 27 2019 and to physicians in the control group (n=3,679) on December 10 2019. The intervention letter contained personalized prescribing feedback and information about pain treatment with opioids. We also analyze the effects of the intervention separately for those physicians who received an information intervention in 2017 and/or 2018 as well. The study will inform policy makers on the usefulness of information letters as a long-term policy tool in reducing opioid prescribing and other potentially harmful clinical practices.

Intervention

From the group of physicians, who had prescribed oxycontin or fentanyl to a patient who had purchased at least three months’ supply of these substances in 2018, 3,689 randomly selected were sent a letter containing information about pain treatment with opioids and prescribing recommendations. The same letter was sent to 3 679 physicians in the control group six months later.

Experimental design

All physicians (N=7368) were identified who had prescribed strong opioids (oxycodone or fentanyl) to a patient (patients who were entitled to special reimbursement from national health insurance based on cancer diagnosis were excluded) who had purchased more than 3 months’ supply of strong opioids in 2018 in Finland. 3 689 of these physicians (treatment group) were randomly selected to receive a letter containing information on opioid use in pain treatment. The letters were sent in May 2019. The same letter was sent to other 3 679 of the identified physicians (control group) on December 10 2019.
We use nationwide register data on prescribing, health care utilization and diagnoses to analyze the effects of the intervention. We compare opioid prescribing and health outcomes in the treatment and control groups. For health outcomes, we compare the patients who have received a prescription from a physician in the treatment group to patients of the physicians in the control group. Furthermore, when studying effects on health, we exclude patients who have received a fentanyl or oxycodone prescription from physicians in both groups.

Design details

Randomization method
Randomization was done by the Social Insurance Institution of Finland in office by a computer.
Randomization unit
Physician
Clusters
3,689 physicians in the treatment group and 3,679 physicians in the control group.
Observations
3,689 physicians in the treatment group and 3,679 physicians in the control group.
Arms
3,689 physicians in the treatment group and 3,679 physicians in the control group.
Primary outcomes
1. Number of prescribed fentanyl (ATC: N01AH01) and oxycodone (ATC: N02AA05) Morphine Milligram Equivalences (MME) 2. Number of fentanyl and oxycodone prescriptions
Secondary outcomes
1. Number of prescribed opioid (ATC: N02A, N01AH, N07BC) MMEs 2. Number of opioid prescriptions Substitution to other drugs: 1. Prescribed Defined Daily Doses (DDDs) of other pain medicines (ATC: M01, N02BE01) 2. Number of other pain medicine prescriptions Patient-level outcomes: 1. Number of delivered fentanyl and oxycodone MMEs 2. Number of delivered opioid MMEs 3. Delivered DDDs of other pain medicines If we gain access to patient-level discharge data, we will also study the effects of the intervention on the following health outcomes: 1. Number of health care contacts 2. Number of opioid related health care contacts (ICD-10: F11, T36, and T40 and corresponding ICPC-2 codes). 3. Number of inpatient contacts 4. Number of opioid related inpatient contacts. 5. Number of days Hospitalized

Cited by

No citation data available for this registration yet.

Keywords