Power of Integrated Data Analytics Exploring Dentist Prescribing of Opioids.

Objectives: To develop and use analytics-based tools to examine integrated health information and deliver insight into prescribing patterns of network dental providers, outliers, demographics of patients treated with opioids, and relationship of opioid scripts to specific dental procedure codes.
Methods: Internal data sources were used to identify opioid prescriptions among Cigna Pharmacy and Dental bundled populations and by Dental Network providers for years Q1-2015 through Q2-2018. Information included:
1. Morphine Milligram Equivalence (MME),
2. Days Supplied (DS), and
3. Demographics of patient.
Claims data associated with both patient being prescribed the opioid and prescribing dentist analyzed for date of prescription and dental procedure codes by date of service. Dental service codes found in table scored and regression analyzed to represent likelihood of being clinically associated with a post-operative pain response. Analysis of population trends for MMEs and DS, outlier prescribing behavior, opioid use associated dental procedures using ANOVA test, and opioid use for different demographic groups conducted.
Results: 1. Opioid prescriptions written by Cigna network dentists demonstrated reduced prescribed MMEs by 18% and reduced DS by 10% overall from Q1-2015 to Q2- 2018.

2. General dentists in the Cigna network are less likely to write an opioid prescription than network specialists, but both groups reduced -MMEs and DS from Q1-2015 to Q2-2018 :
a. General Dentists: 12% MME & 9% DS;
b. Specialists: 19% MME & 10% DS.

3. Limited number of outliers could be identified in the provider data set, compared on MMEs and DS metrics against peers. Further examination and communication needed to determine reasonableness of opiate prescribing
.
4. The tool set showed ANOVA test agreement between opioid prescriptions supplied to patients and dental treatment codes clinically expected to be associated with a post-operative painful response
.
NOTE: Difference in Avg. MME per Rx for high vs. low pain causing procedures is 17.1 MME with a 95% CI of (15.14, 18.96). (2017 data).
Conclusions: Data analytics of integrated health information gains critical knowledge and performance data of groups of providers and their patients concerning use of opioids. Future development is aimed to enhance the user friendliness of this data to be actionable.