Abstract
This study examines the frequency of steroid use for acute respiratory tract infections in Louisiana and nationally, and factors associated with this clinical practice.
Clinical practice guidelines do not recommend systemic steroids in the treatment of acute respiratory tract infections (ARTIs).1 While some studies have shown earlier symptom resolution with steroids given for pharyngitis,2 clinical trials show no efficacy of systemic steroids for sinusitis3 and bronchitis.4 Adverse events can develop within 30 days of short-term steroid use, which raises concern about the safety of systemic steroids for ARTIs.5 We conducted the present study to examine the frequency of steroid use for ARTIs in Louisiana and nationally and to examine factors associated with this clinical practice.
Methods
Study Settings and Populations
We conducted a retrospective observational study of adults who had outpatient ambulatory care encounters that included an ARTI diagnosis (otitis, upper respiratory infection, sinusitis, pharyngitis, bronchitis, allergic rhinitis, influenza, and pneumonia) through Ochsner Health System primary care clinics in 2014 (“Health System”) and as reported in the National Ambulatory Medical Care Survey (NAMCS) in 2012 to 2013. Asthma and chronic obstructive pulmonary disease (COPD) encounter diagnoses were not considered inclusion criteria to minimize encounters for patients with exacerbations of these chronic diseases, which may be appropriately treated with steroids. Protocols for the health plan and health system analyses were reviewed by the Ochsner institutional review board and determined to be exempt based on federal regulations for human subjects research. Patient consent was deemed unnecessary for this retrospective observational study of adults.
Outcomes and Data Analysis
The main outcome for Health System analysis was steroid injection and for NAMCS analysis was steroid prescription. We chose steroid injection usage because we anecdotally observed that this was common practice in the Southeast United States. The NAMCS database does not query for intramuscular injections used in outpatient encounters; for this reason, we chose systemic oral steroid prescriptions. The covariates of interest included patient-, health care professional–, and encounter-specific factors. We conducted propensity score–adjusted, fixed-effects, and multivariable logistic regression analyses. Statistical significance was set at P < .05. All analyses were conducted using Stata statistical software (version 14.2; StataCorp).
Results
National Data
Almost 11% of adult outpatient encounters for ARTI included a steroid prescription. There was a significant regional variation in prescribing steroids for ARTI diagnoses, from a 13.6% prevalence in the South to a 8.3% prevalence in the Midwest (Table 1). In multivariate analysis, there were significantly higher odds for steroid prescriptions among patients’ with a medical history of COPD or asthma (odds ratio [OR], 2.62; 95% CI, 2.24-3.06), visit diagnosis of bronchitis (OR, 1.73; 95% CI, 1.22-2.46), and an encounter with a nurse practitioner (NP) (OR, 1.65; 95% CI, 0.79-3.42) or physician assistant (PA) (OR, 1.74; 95% CI, 0.98-3.06) (Table 1).
Table 1. Bivariable and Multivariable Regression Analysis of NAMCS ARTI Encounters by Steroid Prescriptions in 2012 and 2013.
| Characteristic | All Encounters (n = 107 462) | ARTI Encounters With Steroid Prescription | Steroid Prescriptiona | |||
|---|---|---|---|---|---|---|
| No (n = 5583) | Yes (n = 675) | P Value | OR (95% CI) | P Value | ||
| Patient age, mean (SD), y | 55.4 (18.4) | 49.8 (18.1) | 51.2 (17.2) | <.001 | 1.00 (0.99-1.01) | .56 |
| Female, % | 64 155 (59.7) | 3383 (60.6) | 412 (61.0) | .82 | ||
| Race, No. (%) | ||||||
| White | 85 862 (79.9) | 4483 (80.3) | 556 (82.4) | .57 | 1 [Reference] | |
| African American | 9134 (8.5) | 385 (6.9) | 39 (5.8) | 0.74 (0.41-1.36) | .34 | |
| Other | 12 466 (11.6) | 715 (12.8) | 80 (11.9) | 0.80 (0.55-1.17) | .26 | |
| Insurance, % | ||||||
| Commercial | 51 152 (47.6) | 3394 (60.8) | 379 (56.2) | .12 | 1 [Reference] | |
| Medicaid | 34 495 (32.1) | 1217 (21.8) | 161 (23.9) | 1.04 (0.59-1.85) | .89 | |
| Medicare | 6663 (6.2) | 329 (5.9) | 49 (7.3) | 0.87 (0.62-1.21) | .42 | |
| Other | 15 152 (14.1) | 642 (11.5) | 86 (12.7) | 1.07 (0.77-1.50) | .69 | |
| Chronic conditions, No. (%) | ||||||
| Diabetes | 14 400 (13.4) | 480 (8.6) | 46 (6.8) | .11 | 0.92 (0.58-1.48) | .76 |
| Asthma | 6770 (6.3) | 581 (10.4) | 143 (21.2) | <.001 | ||
| COPD | 5051 (4.7) | 681 (12.2) | 155 (23.0) | <.001 | ||
| COPD and asthma | 2.62 (2.24-3.06) | <.001 | ||||
| Osteoporosis | 2901 (2.7) | 123 (2.2) | 16 (2.4) | .73 | 1.42 (0.74-2.72) | .29 |
| Visit diagnoses, No. (%) | ||||||
| Sinusitis and otitis | 1290 (1.2) | 1100 (19.7) | 149 (22.1) | .15 | 1.28 (0.93-1.77) | .13 |
| Pharyngitis | 645 (0.6) | 581 (10.4) | 56 (8.3) | .09 | 0.75 (0.49-1.14) | .17 |
| Allergic rhinitis | 2042 (1.9) | 1803 (32.3) | 194 (28.7) | .06 | 0.90 (0.65-1.23) | .51 |
| URI-NOS | 1397 (1.3) | 648 (11.6) | 53 (7.8) | .002 | 0.68 (0.50-0.93) | .02 |
| Bronchitis | 1075 (1) | 871 (15.6) | 190 (28.2) | <.001 | 1.73 (1.22-2.46) | .002 |
| Pneumonia | 322 (0.3) | 274 (4.9) | 42 (6.2) | .15 | 1.23 (0.77-1.95) | .39 |
| Health care professional, No. (%) | ||||||
| Physician | 100 262 (93.3) | 5164 (92.4) | 596 (88.3) | <.001 | 1 [Reference] | |
| NP | 2579 (2.4) | 151 (2.7) | 31 (4.6) | 1.65 (0.79-3.42) | .18 | |
| PA | 4621 (4.3) | 268 (4.8) | 48 (7.1) | 1.74 (0.98-3.06) | .06 | |
| Either NP or PA | 1.76 (1.11-2.81) | .02 | ||||
| US region, No. (%) | ||||||
| Midwest | 13 970 (13.0) | 754 (13.5) | 68 (10.1) | <.001 | 1 [Reference] | |
| Northeast | 28 585 (26.6) | 1535 (27.5) | 133 (19.7) | 1.31 (0.81-2.13) | .27 | |
| South | 39 224 (36.5) | 2027 (36.3) | 318 (47.1) | 1.91 (1.36-2.70) | <.001 | |
| West | 25 683 (23.9) | 1267 (22.7) | 156 (23.1) | 1.18 (0.76-1.85) | .45 | |
Abbreviations: ARTI, acute respiratory tract infection; COPD, chronic obstructive pulmonary disease; NAMCS, National Ambulatory Medical Care Survey; NOS, not otherwise specified; NP, nurse practitioner; OR, odds ratio; PA, physician assistant; URI, upper respiratory infection.
Estimates from multivariable logistic regression models accounted for the complex survey design.
Health System
Twenty-three percent of adult primary care encounters for ARTI in the Health System included steroid injections (Table 2). In multivariate analysis, odds for steroid injection were significantly higher among patients with a medical history of COPD (OR, 1.47; 95% CI, 1.31-1.64); visit diagnoses of sinusitis or otitis (OR, 2.10; 95% CI, 1.89-2.33), allergic rhinitis (OR, 1.42; 95% CI, 1.30-1.56), upper respiratory infection (OR, 1.17; 95% CI, 1.05-1.30), or bronchitis (OR, 1.82; 95% CI, 1.67-1.99); and encounters with a NP (OR, 1.61; 95% CI, 1.53-1.71). Odds for steroid injection were lower among encounters with patients who were nonwhite (eg, African American [OR, 0.88; 95% CI, 0.83-0.93], or Medicaid or Medicare insured [OR, 0.80; 95% CI, 0.68-0.95 and 0.75; 95% CI, 0.69-0.81, respectively]), had medical history of diabetes (OR, 0.73; 95% CI, 0.67-0.79) and/or osteoporosis (OR, 0.88; 95% CI, 0.79-0.98), had been seen by a PA (OR, 0.78; 95% CI, 0.71-0.86), and with a visit diagnosis of pneumonia (OR, 0.55; 95% CI, 0.46-0.64) (Table 2). We observed substantial clinician variation with 17% never using steroid injections and 13% of clinicians using an injection more than 40% of the time (data not shown).
Table 2. Bivariable- and Propensity Score–Adjusted Fixed-Effects Multivariable Regression Analysis of Health System ARTI Encounters by Steroid Injections in 2014.
| Characteristic | All (n = 32 647) | ARTI Encounters With Steroid Injection | Steroid Injectiona | |||
|---|---|---|---|---|---|---|
| No (n = 24 987) | Yes (n = 7661) | P Value | OR (95% CI) | P Value | ||
| Patient age, mean (SD), y | 51.0 (17.8) | 51.5 (18.2) | 49.5 (16.5) | <.001 | 1.00 (1.00-1.00) | .53 |
| Female, No. (%) | 21 221 (65.0) | 16 366 (65.5) | 5041 (65.8) | .63 | ||
| Race/ethnicity, No. (%) | ||||||
| White | 14 789 (45.3) | 11 119 (44.5) | 3677 (48.0) | <.001 | 1 [Reference] | |
| African American | 16 487 (50.5) | 12 743 (51.0) | 3731 (48.7) | 0.88 (0.83-0.93) | <.001 | |
| Other | 1371 (4.2) | 1124 (4.5) | 253 (3.3) | 0.68 (0.56-0.83) | <.001 | |
| Insurance, No. (%) | ||||||
| Commercial | 22 461 (68.8) | 16 791 (67.2) | 5677 (74.1) | <.001 | 1 [Reference] | |
| Medicaid | 849 (2.6) | 650 (2.6) | 192 (2.5) | 0.80 (0.68-0.95) | .01 | |
| Medicare | 9043 (27.7) | 7321 (29.3) | 1716 (22.4) | 0.75 (0.69-0.81) | <.001 | |
| Other | 294 (0.9) | 225 (0.9) | 77 (1.0) | 1.35 (1.05-1.74) | .02 | |
| Chronic conditions, No. (%) | ||||||
| Diabetes | 4962 (15.2) | 4098 (16.4) | 873 (11.4) | <.001 | 0.73 (0.67-0.79) | <.001 |
| Asthma | 2579 (7.9) | 1874 (7.5) | 689 (9.0) | <.001 | 1.16 (1.06-1.27) | .001 |
| COPD | 1469 (4.5) | 1024 (4.1) | 460 (6.0) | <.001 | 1.47 (1.31-1.64) | <.001 |
| Osteoporosis | 2546 (7.8) | 2049 (8.2) | 506 (6.6) | <.001 | 0.88 (0.79-0.98) | .02 |
| Visit diagnoses, No. (%) | ||||||
| Sinusitis and otitis | 6627 (20.3) | 4223 (16.9) | 2421 (31.6) | <.001 | 2.10 (1.89-2.33) | <.001 |
| Pharyngitis | 6040 (18.5) | 4548 (18.2) | 1479 (19.3) | .03 | 0.99 (0.87-1.13) | .91 |
| Allergic rhinitis | 4832 (14.8) | 3648 (14.6) | 1180 (15.4) | .07 | 1.42 (1.3-1.56) | <.001 |
| URI-NOS | 12 732 (39.0) | 9770 (39.1) | 2972 (38.8) | .56 | 1.17 (1.05-1.3) | .004 |
| Bronchitis | 8423 (25.8) | 5947 (23.8) | 2475 (32.3) | <.001 | 1.82 (1.67-1.99) | <.001 |
| Pneumonia | 2318 (7.1) | 1999 (8.0) | 322 (4.2) | <.001 | 0.54 (0.46-0.64) | <.001 |
| Health care professional, No. (%) | ||||||
| Physician | 19 327 (59.2) | 15 542 (62.2) | 3785 (49.4) | <.001 | 1 [Reference] | |
| NP | 10 219 (31.3) | 6896 (27.6) | 3310 (43.2) | 1.62 (1.53-1.71) | <.001 | |
| PA | 3101 (9.5) | 2524 (10.1) | 575 (7.5) | 0.78 (0.71-0.86) | <.001 | |
| Urgent care, %b | 27 554 (84.4) | 20 339 (81.4) | 7232 (94.4) | <.001 | 2.68 (2.44-2.94) | <.001 |
| Weekend, No. (%) | 1273 (3.9) | 950 (3.8) | 322 (4.2) | .10 | ||
| Visit total, mean (SD) | 1.6 (1.6) | 1.4 (1.2) | 2.0 (2.4) | <.001 | ||
Abbreviations: ARTI, acute respiratory tract infection; COPD, chronic obstructive pulmonary disease; NOS, not otherwise specified; NP, nurse practitioner; OR, odds ratio; PA, physician assistant; URI, upper respiratory infection.
Estimates from triply robust propensity score-adjusted fixed effects multivariable logistic regression panel analysis.
Urgent care defined at clinic visit <48 hours from scheduling.
Conclusions
Adverse effects of systemic steroids, even for short-term use, are well documented. Future research is needed to further explore regional and national trends in use of corticosteroids for patients with ARTIs, as it likely represents high-cost, potentially harmful care. This study revealed high rates of systemic corticosteroid use among patients with ARTIs in Louisiana and nationally. Study limitations include possible inclusion of encounters for exacerbation of chronic respiratory illnesses such as asthma or COPD and discordance in corticosteroid outcomes measured (injection vs oral). This hypothesis-generating study highlights the need to further examine use of systemic corticosteroid for ARTIs and associated safety issues.
References
- 1.CDC Adult Treatment Recommendations https://www.cdc.gov/getsmart/community/for-hcp/outpatient-hcp/adult-treatment-rec.html. Accessed July 17, 2017.
- 2.Sadeghirad B, Siemieniuk RAC, Brignardello-Petersen R, et al. Corticosteroids for treatment of sore throat: systematic review and meta-analysis of randomised trials. BMJ. 2017;358:j3887. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Venekamp RP, Thompson MJ, Hayward G, et al. Systemic corticosteroids for acute sinusitis. Cochrane Database Syst Rev. 2014;(3):CD008115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hay AD, Little P, Harnden A, et al. Effect of oral prednisolone on symptom duration and severity in nonasthmatic adults with acute lower respiratory tract infection: a randomized clinical trial. JAMA. 2017;318(8):721-730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Waljee AK, Rogers MA, Lin P, et al. Short term use of oral corticosteroids and related harms among adults in the United States: population based cohort study. BMJ. 2017;357:j1415. [DOI] [PMC free article] [PubMed] [Google Scholar]