Demographic and Comorbidity Profile of Opioid-Related Hospitalizations in Mississippi: A Call for State Surveillance

Background: Opioid substance abuse is a fast growing and difficult to control epidemic, imposing an urgent need for building surveillance systems at a state level. As an integral part of such monitoring systems, secondary data sources can be utilized for the risk-identification and risk-stratification of patients and populations abusing opioid substances. The purpose of this study was to use hospital discharge data for identifying at-risk populations for opioid-related hospitalizations in Mississippi and for describing their demographics and comorbidity profiles. In particular, we evaluated the relation between opioid misuse and several clinical conditions known to be risk factors for opioid over-prescription and misuse.

Methods: Mississippi hospital discharge data for 2010-2011 were analyzed with SAS 9.3. We compared proportions of categorical variables with chi-square tests. The unit of analysis was a hospital discharge and we used primary and secondary diagnoses to identify discharges with opioid-related diagnostic codes.

Results: There were 9,781 opioid-related hospitalizations during 2010-11. The state’s two-year average rate was 15 opioid-related hospitalizations per 10,000 residents and the Southeast Public Health District had the highest hospitalization rate (27 per 10,000 persons). The proportion of patients with a diagnosis of opioid substance misuse was highest for Caucasians (83%), females (54%), rural residents (51%), and Medicare beneficiaries (31%). Compared to patients hospitalized for all other causes, however, Caucasians (83% vs. 59%), males (46% vs. 40%), urban residents (49% vs. 44%), and the uninsured (17% vs. 7%) were more likely to be hospitalized with a diagnosis indicating opioid misuse.

The bivariate analyses further revealed that 71% of all patients with an opioid-related diagnosis and 24% of all other patients had a co-existing mental health disorder. Compared to all other hospitalizations, patients hospitalized with a diagnosis of opioid abuse were more likely to have a co-existing diagnosis of chronic non-cancer-related pain (19% vs. 2%), lower back pain (12% vs. 2%), and myalgia (3% vs. 1%); however, these patients were less likely to have a co-existing cancer diagnosis (2% vs. 6%).  The above-mentioned differences were statistically significant at p = 0.001.

Conclusion: A disproportionate number of patients with an opioid-related diagnosis were Caucasians, lived in the state’s southeastern region and had mental disorders. These patients also had a high prevalence of non-cancer-related pain diagnoses. In addition to allocating resources to the populations at-risk, these findings can help public health programs and health care systems in their evaluation of the risk/benefit ratio of prescribing opioid substances.