Abstract
Objective
Determine if disparities exist for revisit complications after adult tonsillectomy.
Methods
Cases of adult tonsillectomy were extracted from the state ambulatory surgery databases and linked to the state emergency department databases and inpatient databases for California, Iowa, Florida and New York for 2010 and 2011. Revisits within 14 days for diagnoses of: post-tonsillectomy bleeding, acute pain and nausea/vomiting/dehydration were determined and analyzed for associations of these complications with age, sex, race, median household income and comorbidity score.
Results
Among 17,836 tonsillectomies (63.7% female; mean age, 29.0 years), revisit rates for post-tonsillectomy bleeding, acute pain and fever/dehydration were 5.1, 2.8 and 1.5%, respectively. On multivariate analysis, only female sex was associated with a lower post-tonsillectomy bleeding rate (odds, 0.48, p < 0.001). Decreasing household income, female sex, black and Hispanic race were associated with increased revisits for acute pain (odds, 1.21, 1.49, 2.03 and 1.32, p ≤ 0.002). Female sex was associated with an increased odds of a revisit for FNVD (odds, 1.94, p < 0.001).
Conclusions
Significant disparities with respect to income and race exist in the incidence of revisits and potentially avoidable complications after adult tonsillectomy.
1
Introduction
Healthcare disparities have been defined by the National Institutes of Health as differences in the incidence, prevalence, and burden of diseases and other adverse health conditions that exist among specific population groups in the United States. Population groups can be defined along racial, ethnic, gender, socioeconomic and/or behavioral lines. The past decade has been marked by a growing identification of healthcare disparities among disadvantaged patient populations. Recently, the Institute of Medicine outlined serious inequalities in the United States healthcare system leading to significant health care disparities .
Healthcare disparities have only recently been investigated in the otolaryngologic literature. We and others have quantified some of the healthcare disparities that exist with respect to racial, ethnic and insurance status for access to care and prescribing in pediatric acute otitis media, survival outcomes for head and neck cancer, pediatric sinusitis and pediatric sleep disordered breathing . While the study of inequalities in the epidemiology of disease and access to care has been at the forefront of disparity research, relatively fewer investigations have probed disparities that may exist in the outcomes after otolaryngologic surgical care.
Tonsillectomy is among the most commonly performed outpatient procedures in general and among otolaryngologic surgeries . Unfortunately, complications are relatively common after both pediatric and adult tonsillectomy with non-negligible rates of revisits and readmissions for postoperative problems such as post-tonsillectomy bleeding, fever/nausea/vomiting/dehydration (FNVD) and acute pain . Because such complications are to some degree “expected” after tonsillectomy, we sought to determine if healthcare disparities existed among population groups with respect to the incidence of these complications.
2
Methods
Cases of ambulatory adult (age ≥ 18.0 years) tonsillectomy or adenotonsillectomy were extracted from the State Ambulatory Surgery Databases for New York, Florida, Iowa and California for calendar years 2010 and 2011. These cases were linked to the corresponding State Emergency Department Databases and the State Inpatient Databases for visit encounters occurring after tonsillectomy, but within a 14-day postoperative window. Tonsillectomy cases performed for malignancy as the primary diagnosis were explicitly excluded. These de-identified databases are part of the Healthcare Cost and Utilization Project (HCUP) maintained by the Agency for Healthcare Research and Quality . This study was reviewed by our hospital’s committee on clinical investigations and deemed exempt from review. Standard demographic information was extracted and tabulated for the cases of adult tonsillectomy. Race/ethnicity was categorized as white, black, Hispanic or other (including Asian/Pacific Islander and Native American). Household income was assigned to one of the four state’s median household income quartiles to represent socioeconomic status. In order to control for comorbidities, the van Walraven point score modification of the Elixhauser comorbidity index was computed for each case .
Next, for each case, whether a revisit (including readmission) specifically for post-tonsillectomy bleeding, acute pain or fever/nausea/vomiting/dehydration (FNVD) after the tonsillectomy occurred was determined. The site of the revisit encounter was determined as return to the ambulatory surgery site, emergency department or inpatient admission. All diagnoses at the time of the revisit were examined to determine if any of the diagnoses encompassed post-tonsillectomy bleeding; acute pain and FNVD were considered if they were the primary revisit diagnosis. After univariate analysis, multivariate analysis was conducted with logistic regression to determine the associations between sex, race, household income and age with the occurrence of a revisit specifically for post-tonsillectomy bleeding, acute pain or FNVD.
2
Methods
Cases of ambulatory adult (age ≥ 18.0 years) tonsillectomy or adenotonsillectomy were extracted from the State Ambulatory Surgery Databases for New York, Florida, Iowa and California for calendar years 2010 and 2011. These cases were linked to the corresponding State Emergency Department Databases and the State Inpatient Databases for visit encounters occurring after tonsillectomy, but within a 14-day postoperative window. Tonsillectomy cases performed for malignancy as the primary diagnosis were explicitly excluded. These de-identified databases are part of the Healthcare Cost and Utilization Project (HCUP) maintained by the Agency for Healthcare Research and Quality . This study was reviewed by our hospital’s committee on clinical investigations and deemed exempt from review. Standard demographic information was extracted and tabulated for the cases of adult tonsillectomy. Race/ethnicity was categorized as white, black, Hispanic or other (including Asian/Pacific Islander and Native American). Household income was assigned to one of the four state’s median household income quartiles to represent socioeconomic status. In order to control for comorbidities, the van Walraven point score modification of the Elixhauser comorbidity index was computed for each case .
Next, for each case, whether a revisit (including readmission) specifically for post-tonsillectomy bleeding, acute pain or fever/nausea/vomiting/dehydration (FNVD) after the tonsillectomy occurred was determined. The site of the revisit encounter was determined as return to the ambulatory surgery site, emergency department or inpatient admission. All diagnoses at the time of the revisit were examined to determine if any of the diagnoses encompassed post-tonsillectomy bleeding; acute pain and FNVD were considered if they were the primary revisit diagnosis. After univariate analysis, multivariate analysis was conducted with logistic regression to determine the associations between sex, race, household income and age with the occurrence of a revisit specifically for post-tonsillectomy bleeding, acute pain or FNVD.
3
Results
A total of 17,836 adult tonsillectomies were examined across four states. There was a female preponderance (63.7%) with a mean age of 29.0 years. Figs. 1 and 2 depict the relative distribution of patients according to the states’ median household income quartiles and the distribution of race, respectively. Overall, the rate of revisit for post-tonsillectomy bleeding was 5.1%. Acute pain and FNVD as the primary revisit diagnoses were 2.8 and 1.5%, respectively.
Table 1 presents the univariate analysis for the associations between sex, race and household income with the revisit rates for post-tonsillectomy bleeding, acute pain and FNVD. Female sex was associated with a lower rate of revisits for post-tonsillectomy bleeding but slightly higher rates of revisit for acute pain and FNVD. Similarly, significant differences were noted for revisit rates according to race for acute pain and FNVD, although the latter differences were smaller in magnitude. Finally, lower household income quartile was associated with higher rates of acute pain revisits.
| Variable | Post-tonsillectomy bleeding (%) | p -Value | Acute pain (%) | p -Value | FNVD (%) | p -Value |
|---|---|---|---|---|---|---|
| Sex | ||||||
| Male | 8.01 | < 0.001 | 2.09 | < 0.001 | 0.82 | < 0.001 |
| Female | 3.84 | 3.17 | 1.67 | |||
| Race | ||||||
| White | < 0.001 | 0.466 | 2.39 | < 0.001 | 1.41 | < 0.001 |
| Black | 4.25 | 5.38 | 1.32 | |||
| Hispanic | 5.53 | 3.36 | 1.74 | |||
| Other | 5.33 | 2.82 | 1.05 | |||
| Household income | ||||||
| First quartile (poorest) | 5.22 | 0.692 | 4.04 | < 0.001 | 1.58 | 0.826 |
| Second quartile | 5.37 | 2.98 | 1.34 | |||
| Third quartile | 4.84 | 2.64 | 1.48 | |||
| Fourth quartile (wealthiest) | 5.03 | 1.98 | 1.54 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
