Orthopedics

Feature Article 

Hospital Stays and Medical Expenses Increase Nationwide Among Patients With Septic Arthritis of the Shoulder Who Inject Drugs

Alexander J. Toppo, BS; David H. W. Oh, BS; David J. Tybor, PhD, MS, MPH; Alysse G. Wurcel, MD, MS; Mariano E. Menendez, MD; Scott P. Ryan, MD; Matthew J. Salzler, MD

Abstract

The authors determined the proportion of patients nationwide with septic arthritis of the shoulder who inject drugs, evaluated differences in hospitalization outcomes and charges between patients with and without injection drug use (IDU), and quantified demographic trends among patients with IDU from 2000 to 2013. Nationally representative data of patients with a principal discharge diagnosis of shoulder septic arthritis were obtained from the Nationwide Inpatient Sample 2000–2013. Using published algorithms, the authors classified septic arthritis of the shoulder as related or unrelated to IDU. They compared length of stay, leaving against medical advice, number of procedures, and mortality rates between the 2 groups, using regression models to control for age, sex, and race. Fifteen percent (95% confidence interval [CI], 13.6%–16.5%) of septic arthritis cases were associated with IDU. From 2000 to 2013, shoulder septic arthritis associated with IDU increased 4-fold. After controlling for age, sex, and race, individuals who inject drugs stayed in the hospital for 3.7 more days (95% CI, 2.4–5.0), incurred an average of $13,250 more charges for medical care (95% CI, $2635–$23,866), and were 5.54 times more likely (95% CI, 3.22–9.55) to leave against medical advice than those without IDU. From 2000 to 2013, there was an increase in the proportion of patients with IDU-related septic arthritis of the shoulder between 35 and 54 years old and 55 and 64 years old, and an increase in the proportion who were white. Injection drug use–related shoulder septic arthritis is linked to suboptimal inpatient outcomes and greater resource use. [Orthopedics. 2020;43(4):e270–e277.]

Abstract

The authors determined the proportion of patients nationwide with septic arthritis of the shoulder who inject drugs, evaluated differences in hospitalization outcomes and charges between patients with and without injection drug use (IDU), and quantified demographic trends among patients with IDU from 2000 to 2013. Nationally representative data of patients with a principal discharge diagnosis of shoulder septic arthritis were obtained from the Nationwide Inpatient Sample 2000–2013. Using published algorithms, the authors classified septic arthritis of the shoulder as related or unrelated to IDU. They compared length of stay, leaving against medical advice, number of procedures, and mortality rates between the 2 groups, using regression models to control for age, sex, and race. Fifteen percent (95% confidence interval [CI], 13.6%–16.5%) of septic arthritis cases were associated with IDU. From 2000 to 2013, shoulder septic arthritis associated with IDU increased 4-fold. After controlling for age, sex, and race, individuals who inject drugs stayed in the hospital for 3.7 more days (95% CI, 2.4–5.0), incurred an average of $13,250 more charges for medical care (95% CI, $2635–$23,866), and were 5.54 times more likely (95% CI, 3.22–9.55) to leave against medical advice than those without IDU. From 2000 to 2013, there was an increase in the proportion of patients with IDU-related septic arthritis of the shoulder between 35 and 54 years old and 55 and 64 years old, and an increase in the proportion who were white. Injection drug use–related shoulder septic arthritis is linked to suboptimal inpatient outcomes and greater resource use. [Orthopedics. 2020;43(4):e270–e277.]

Septic arthritis is an urgent condition common to orthopedists of all subspecialties. It is associated with significant morbidity and mortality due to cartilage and bone degradation and potential systemic infection1,2 and had medical costs of $759 million in the United States in 2012 alone.3 Among those at high risk for septic arthritis are individuals with prosthetic joints, children, the elderly, and people who inject drugs (PWID).4,5 Injection drug use (IDU) can lead to septic arthritis through infection at the needle insertion site or the hematogenous seeding of joints.6–9

Given the current opioid epidemic, health care use among individuals with IDU-related septic arthritis is an orthopedic topic of special importance.10–12 Specifically, septic arthritis of the shoulder has a notoriously poor prognosis and has been associated with recurrent dislocation, subluxation, effusion, drainage, and other sequelae.13 Although differences in rates of complications, reoperation, leaving against medical advice, and other outcome measurements between PWID and people who do not inject drugs (non-PWID) have been documented for septic arthritis of the knee,14 these outcomes have not yet been investigated for septic arthritis of the shoulder. The most common area for injection drug use is the antecubital area, and proximity to the shoulder joint makes epidemiology of septic shoulders in PWID particularly relevant.

The authors sought to answer the following questions: (1) Among patients hospitalized for septic arthritis of the shoulder, what proportion inject drugs, and how has this proportion changed from year to year? (2) Are there any differences in outcome measures of length of stay, reoperations, and leaving against medical advice among patients diagnosed with septic arthritis of the shoulder who do and do not inject drugs? (3) How have patients with septic arthritis of the shoulder with and without IDU changed in terms of age and race trends from 2000 to 2013? The authors hypothesized that the proportion of septic arthritis of the shoulder patients with IDU would increase during the study time period, that PWID with septic arthritis of the shoulder would have worse outcome measures than non-PWID with septic arthritis of the shoulder, and that these 2 groups of patients would exhibit demographic changes from 2000 to 2013.

Materials and Methods

Database

In this cross-sectional study, the authors obtained data pertaining to patients treated for septic arthritis of the shoulder from 2000 to 2013 from the Healthcare Cost and Utilization Project (HCUP) Nationwide Inpatient Sample (NIS) database and conducted a comparative analysis. The NIS represents the largest US inpatient health care database available to the public and includes more than 8 million yearly inpatient observations.15 The database is a stratified, systematic random sample of approximately 20% of discharges from US nongovernmental community hospitals (including tertiary centers) with weights allowing for both statewide and national estimates. Diagnoses and procedural codes (International Classification of Diseases, 9th Revision, Clinical Modification [ICD-9-CM]), discharge status, patient demographics, and length of stay are included in the data. Given the variety of hospitalization indicators included in this nationally representative cohort, the database lends itself to several research topics, including medical practice patterns and treatment outcomes. In this study, the authors divided patients into 3 groups by age (15–34 years, 35–54 years, and 55–64 years, respectively), and 6 groups by race (white, black, Hispanic, Asian or Pacific Islander, Native American, and other) as was done in a prior publication studying IDU among infectious endocarditis patients.16 Those older than 64 years were omitted because this group is at higher risk of bacterial infection regardless of IDU status.17

Classification of Septic Arthritis of the Shoulder

Patients between 15 and 64 years old who were admitted to the hospital with a principal diagnosis of septic arthritis of the shoulder (ICD-9-CM 711.01) between January 1, 2000, and December 31, 2013, were included. Patients with a history of arthroplasty of the shoulder (ICD-9-CM V43.61) were excluded to remove this risk factor for septic arthritis from the analysis; patients with other types of shoulder surgery were included.

Classification of IDU-Related Septic Arthritis of the Shoulder

In the context of this study, IDU is defined as the injection of any nonprescription or illicit drugs, including cocaine, methamphetamines, and opioids. To the authors' knowledge, there is no direct method of identifying individuals who inject illicit or nonprescription drugs with ICD-9 codes. Although ICD-9-CM codes do not specify whether the drugs being used are injected or taken by other means (eg, smoked or snorted), a previous published proxy for the classification of IDU is ICD-9 codes for drugs that could be injected in addition to ICD-9 codes for hepatitis C virus (HCV).16 Because 80% of PWID have HCV, this disease is strongly associated with IDU.18 In the current study, patients with at least 1 ICD-9 code for either illicit drug use or HCV were classified as PWID.

Outcomes of Interest

The annual and overall proportions of IDU-related hospitalizations were calculated for those hospitalized with a principal diagnosis of septic arthritis of the shoulder and for all hospitalizations, respectively, captured by the NIS database from 2000 to 2013. The annual incidence of shoulder septic arthritis and IDU-related septic arthritis of the shoulder were determined using yearly population estimates from the US Census Bureau.19,20 In terms of hospitalization outcomes, the authors analyzed length of stay, mortality, proportion of patients who left against medical advice, and proportion of patients receiving repeat arthroscopy (ICD-9-CM codes 80.21, 80.41, 80.60, 80.76, and 80.81), arthrotomy (ICD-9-CM codes 80.01 and 80.11), and arthrocentesis (ICD-9-CM code 81.91); they also analyzed total charges and insurance status. The ICD-9-CM codes used to identify these procedures were consistent with those previously used for this purpose in the literature.1 The authors determined the distribution of patients with septic arthritis of the shoulder with and without IDU among the age and racial groups described here and conducted a trend analysis to ascertain the changes in the age and racial characteristics of these groups from 2000 to 2013.

Statistical Analysis

The authors used adjusted Wald tests to identify any differences in demographics and hospitalization outcomes between patients with septic arthritis of the shoulder with and without IDU. To determine annual trends in the demographic groups of IDU-related septic arthritis of the shoulder, the authors ran a linear regression using the year of hospitalization as the dependent variable. Logistic regression was employed to calculate the odds ratios (ORs) of leaving against medical advice, having surgical procedures, and mortality for septic arthritis of the shoulder in PWID relative to non-PWID. To calculate the ORs adjusted for age, race, and sex, the authors implemented a multivariable analysis of the linear and logistic regressions that controlled for these variables. The Hosmer-Lemeshow test was used to evaluate the goodness of fit for all logistic models (P<.05). Observations lacking a race categorization, which constituted 18% of the population, were excluded from the race analyses. The authors used Stata IC version 15.1 software (Stata Corp, College Station, Texas) to conduct all of the described analyses, with a 2-sided alpha level of 0.05. R 3.2.3 (R Development Core Team, Vienna, Austria) was used to produce Figure 1, and Excel version 16.15 for Mac (Microsoft, Redmond, Washington) was used to produce Figure 2. The appropriate discharge and trend weights were obtained from the HCUP to determine the national estimates and their respective standard errors.21 When the authors analyzed subsets of data pertaining to patients with IDU-related septic arthritis of the shoulder, they calculated appropriate standard errors using HCUP-NIS hospital data.22,23 Because this investigation did not involve human subjects, it was granted an exemption from the Tufts Health Sciences Institutional Review Board.

The proportion of people who inject drugs among patients with septic arthritis of the shoulder (Nationwide Inpatient Sample, 2000–2013). The error bars represent 95% confidence intervals.

Figure 1:

The proportion of people who inject drugs among patients with septic arthritis of the shoulder (Nationwide Inpatient Sample, 2000–2013). The error bars represent 95% confidence intervals.

The national hospital charges associated with septic arthritis of the shoulder patients who inject drugs and those who do not inject drugs (Nationwide Inpatient Sample, 2000–2013). The percentages represent the proportion of the total annual hospital charges for septic arthritis of the shoulder that are associated with injection drug use (IDU).

Figure 2:

The national hospital charges associated with septic arthritis of the shoulder patients who inject drugs and those who do not inject drugs (Nationwide Inpatient Sample, 2000–2013). The percentages represent the proportion of the total annual hospital charges for septic arthritis of the shoulder that are associated with injection drug use (IDU).

Results

Patient Population

The authors analyzed data from 2996 patients with a primary diagnosis of septic arthritis of the shoulder between 2000 and 2013. This corresponds to a national estimation of 14,193 (95% confidence interval [CI], 13,437–14,949) patients (Table 1).

Demographics for Patients With Septic Arthritis of the Shoulder in the United States, Stratified by Injection Drug Use: 2000–2013, Nationwide Inpatient Samplea

Table 1:

Demographics for Patients With Septic Arthritis of the Shoulder in the United States, Stratified by Injection Drug Use: 2000–2013, Nationwide Inpatient Sample

Patients with IDU-related septic arthritis of the shoulder were more frequently between 15 and 34 years old (IDU-SA: 16%, 95% CI, 12%–20%; non-IDU-SA: 11%, 95% CI, 10%–13%; P=.028) and 35 and 54 years old (IDU-SA: 61%, 95% CI, 56%–66%; non-IDU-SA: 50%, 95% CI, 48%–52%; P<.001) compared with patients with non–IDU-related septic arthritis of the shoulder, who were more frequently between 55 and 64 years old (IDU-SA: 23%, 95% CI, 19%–27%; non-IDU-SA: 39%, 95% CI, 37%–41%; P<.001). A greater percentage of PWID with septic arthritis of the shoulder were black or Hispanic compared with septic arthritis of the shoulder patients who did not use drugs (IDU-SA: 21%, 95% CI, 17%–26%; non-IDU-SA: 15%, 95% CI, 13%–17%; P=.011). Patients with IDU-related septic arthritis were more often insured by Medicaid (IDU-SA: 32%, 95% CI, 28%–37%; non-IDU-SA: 14%, 95% CI, 12%–15%; P<.001) or self-pay (IDU-SA: 20%, 95% CI, 16%–24%; non-IDU-SA: 8%, 95% CI, 7%–9%; P<.001); their non-IDU counterparts were more often insured by Medicare (IDU-SA: 16%, 95% CI, 13%–20%; non-IDU-SA: 21%, 95% CI, 19%–23%; P=.026). Only 19% of patients with IDU-related septic arthritis of the shoulder were privately insured, whereas 49% of patients with non–IDU-related septic arthritis of the shoulder were privately insured (Table 1).

Proportion of Patients With IDU-Related Septic Arthritis of the Shoulder

From 2000 to 2013, 15% (95% CI, 13.6%–16.5%) of the septic arthritis of the shoulder population had the IDU-related form of this condition (2128 patients; 95% CI, 1888–2367). The number of cases of septic arthritis of the shoulder in PWID increased by a factor of 4, from 67 (95% CI, 32–103) in 2000 to 265 (95% CI, 190–340) in 2013. During this time, the overall number of septic arthritis of the shoulder cases increased by a factor of 1.7, from 710 (95% CI, 571–850) in 2000 to 1205 (95% CI, 1038–1372) in 2013. From 2007 to 2013, the proportion of shoulder septic arthritis cases related to IDU increased from 11.6% (95% CI, 7.2%–16.0%) to 22.0% (95% CI, 16.6%–27.3%). The mean annual increase in the proportion of IDU among shoulder septic arthritis patients over this time period was 1.3 percentage points (95% CI, 0.37–2.2; P=.006) (Figure 1).

Hospitalization Outcomes

Among shoulder septic arthritis patients with and without IDU, 0.9% (95% CI, 0.3%–2.5%) and 0.5% (95% CI, 0.3%–0.9%) died while in the hospital, respectively. After adjusting for age, race, and sex, no significant difference in mortality during hospitalization was found between patients with IDU-related septic arthritis of the shoulder and those with non–IDU-related septic arthritis of the shoulder (adjusted OR [AOR], 1.89; 95% CI, 0.55–6.49; P=.313). The PWID with septic arthritis of the shoulder were less likely to receive arthroscopy than non-PWID (AOR, 0.76; 95% CI, 0.61–0.95; P=.017). There was no significant difference in arthrotomy (AOR, 1.21; 95% CI, 0.95–1.55; P=.126) or arthrocentesis (AOR, 1.05; 95% CI, 0.81–1.36; P=.737) between the 2 groups. The 2 groups did not differ in rates of repeat arthroscopy (AOR, 0.75; 95% CI, 0.53–1.06; P=.100), arthrotomy (AOR, 1.80; 95% CI, 0.88–3.70; P=.107) or reaspiration (AOR, 0.54; 95% CI, 0.24–1.19; P=.125). Compared with septic arthritis patients without IDU, septic arthritis patients with IDU were 5.54 times more likely to leave against medical advice (95% CI, 3.22–9.55; P<.001) (Table 2). On average, these PWID stayed in the hospital for 3.7 days longer (95% CI, 2.4–5.0 days; P<.001) than patients with non–IDU-related septic arthritis of the shoulder and received medical care that was $13,250 more expensive (adjusted for inflation, 95% CI, $2635–$23,866; P=.014) than that of septic arthritis patients who do not inject drugs (Table 3). The proportion of septic arthritis of the shoulder charges associated with IDU climbed from 11% (95% CI, 5.1%–13.9%) in 2000 to 26% (95% CI, 18.8%–30.3%) in 2013. During the course of the study period, hospital charges for shoulder septic arthritis grew by $2930 annually (adjusted for inflation, 95% CI, $1781–$4079; P<.001) (Figure 2).

Association of Injection Drug Use With Hospitalization Outcomes Among Patients With Septic Arthritis of the Shoulder, United States: 2000–2013, Nationwide Inpatient Samplea

Table 2:

Association of Injection Drug Use With Hospitalization Outcomes Among Patients With Septic Arthritis of the Shoulder, United States: 2000–2013, Nationwide Inpatient Sample

Hospitalization Outcomes for Patients With Septic Arthritis of the Shoulder in the United States, Stratified by Injection Drug Use: 2000–2013 Combined, Nationwide Inpatient Sample

Table 3:

Hospitalization Outcomes for Patients With Septic Arthritis of the Shoulder in the United States, Stratified by Injection Drug Use: 2000–2013 Combined, Nationwide Inpatient Sample

Age and Racial Trends

From 2000 to 2013, there was a decrease in the proportion of PWID with septic arthritis of the shoulder between 35 and 54 years old (P<.001) and an increase in the proportion of patients with IDU-related septic arthritis of the shoulder between 55 and 64 years old (P<.001). Among patients with IDU-related septic arthritis of the shoulder, the proportion of whites increased from 2000 to 2013 (P<.001).

Discussion

Based on the authors' results, patients with IDU-related septic arthritis of the shoulder were admitted to the hospital for longer periods of time, more frequently left against medical advice, and incurred higher hospital expenses than patients with non–IDU-related septic arthritis of the shoulder. Furthermore, the unadjusted annual increase in hospital expenses related to shoulder septic arthritis ($3610; 95% CI, $2502–$4717) far outpaced the unadjusted annual growth in national health expenditures per capita ($328 per year, unadjusted for inflation) during the study period.24 There were age and racial changes over time, with an increasing proportion of white PWID and an increasing proportion of 35- to 64-year-old PWID. Contrary to what the authors hypothesized, there was no statistically significant difference in rates of repeat procedures between these 2 groups. There was also no significant difference in mortality, although this may have been due to the small proportion of the authors' sample that died during hospitalization.

The increasing proportion of IDU among inpatients with septic arthritis of shoulder demonstrates another effect of the opioid epidemic on the field of orthopedic surgery. This is of greater importance given that orthopedic surgery has produced the third highest number of opioid prescriptions in the United States.25 The few guidelines regarding prescribing habits in orthopedic surgery recommend better determining who is at risk for opioid addiction, identifying who truly needs them, and helping patients set more accurate expectations for pain management and prescription administration.25 The described age demographic of patients with IDU-related septic arthritis along with the increasing number of PWID should inform screening practices among orthopedic surgeons both when treating septic arthritis and when prescribing opioids. These practices of secondary prevention and harm reduction align with existing guidelines for all clinicians.26,27

To the authors' knowledge, this investigation is the first to portray the burden of septic arthritis of the shoulder associated with IDU and adds to the existing body of literature showing that PWID have higher rates of postoperative complications and reinfection.28,29 If left untreated or undertreated, septic arthritis may lead to further bone destruction, systemic bacterial spread and sepsis, epidural abscesses, bacterial endocarditis, and other serious morbidities.

The fact that PWID did not have greater initial or repeat procedure rates than non-PWID and were less likely to have undergone arthroscopy may be related to delayed diagnosis of septic arthritis of the shoulder. Because this form of septic arthritis is less common than septic arthritis of the knee, presents without an effusion of the joint, and occurs less frequently than several other conditions that cause atraumatic shoulder pain, it may be initially overlooked and only diagnosed in the chronic phase. Additionally, patients with IDU may seek medical attention later than non-IDU patients. Given the paucity of guidelines for the clinical management of IDU-related septic arthritis, orthopedic surgeons must have a suspicion in patients with IDU for this disease so that they can treat them early. Further research can determine the effectiveness of total synovectomy as initial debridement for this disease in the aforementioned population. Additionally, because the peri- and post-hospital course of IDU-related septic arthritis may differ from that of the non–IDU-related form of the disease, the ultimate fate of patients with the former merits further investigation.

Orthopedic surgeons should accurately screen for IDU among those they are treating for septic arthritis and treat this condition accordingly. Finally, orthopedic surgeons should work in collaboration with addiction specialists and investigate other multidisciplinary approaches to managing the burgeoning number of PWID with septic arthritis.

This study had several strengths. It employed a nationally representative database to compare the hospital outcomes and expenses of PWID and non-PWID with septic arthritis of the shoulder; thus, its findings are generalizable to the entire US inpatient population. It is also the first of its kind to evaluate the differences in hospital outcomes and expenses between IDU-related and non–IDU-related septic arthritis of the shoulder.

This study also had several limitations. Because the authors' analysis relied heavily on proper hospital coding, coding misclassifications or omissions could have skewed their results. It is possible that these nonrandom errors biased the observed patient outcomes and demographic distributions; however, the use of weighted samples in this analysis mitigated any such effect. Additionally, it is not possible to follow individual patients through readmission, further health care use, and health outcomes with this database. Finally, the authors could not determine the proximity of IDU to the onset of septic arthritis, relative risk of PWID developing septic arthritis of the shoulder compared with those without IDU, or long-term outcomes of those diagnosed with septic arthritis. However, this study used the largest sample to date to investigate the hospitalization outcomes of shoulder septic arthritis. Further prospective cohort studies could help establish temporal causality and shed more light on these topics.

Although the authors used a previously published algorithm to identify individuals with IDU,16 it likely underestimated the number of individuals with IDU because this condition has traditionally been underdiagnosed. This methodology was also not immune to random misclassification of IDU status; however, this would only bias the results toward the null and mask any statistically significant relationship found. Additionally, the current study only included patients with a primary diagnosis of septic arthritis of the shoulder, omitting PWID who were given a secondary diagnosis of shoulder septic arthritis. For these reasons, this investigation provides a conservative estimate of the actual burden of hospitalizations linked to IDU-related septic arthritis.

The multivariable regressions used in this investigation have several limitations as well. The authors controlled for age, sex, race, and other key demographic factors; however, there may have been residual confounding from variables that could not be accurately obtained through the database. The NIS assigns each patient to a national income quartile based on his or her zip code's median household income, which only provides a crude estimate of socioeconomic status and does not account for homeless individuals. Furthermore, the frequent omission of homelessness and smoking status from billing records makes it difficult to account for these variables. To explore the effect of these covariates, further investigation with more comprehensive patient demographic information is indicated.

Conclusion

Injection drug use among patients with septic arthritis of the shoulder is increasing rapidly and is associated with longer hospital stays, higher medical charges, and increased rates of leaving against medical advice. Awareness of these trends may aid the orthopedic community in the early diagnosis and multidisciplinary treatment of septic arthritis of the shoulder in PWID.

References

  1. Jiang JJ, Piponov HI, Mass DP, Angeles JG, Shi LL. Septic arthritis of the shoulder: a comparison of treatment methods. J Am Acad Orthop Surg. 2017;25(8):e175–e184. doi:10.5435/JAAOS-D-16-00103 [CrossRef] PMID:28665804
  2. Mathews CJ, Weston VC, Jones A, Field M, Coakley G. Bacterial septic arthritis in adults. Lancet. 2010;375(9717):846–855. doi:10.1016/S0140-6736(09)61595-6 [CrossRef] PMID:20206778
  3. Singh JA, Yu S. The burden of septic arthritis on the US inpatient care: a national study. PLoS One. 2017;12(8):e0182577. doi:10.1371/journal.pone.0182577 [CrossRef]
  4. Gupta MN, Sturrock RD, Field M. A prospective 2-year study of 75 patients with adult-onset septic arthritis. Rheumatology (Oxford). 2001;40(1):24–30. http://www.ncbi.nlm.nih.gov/pubmed/11157138. doi:10.1093/rheumatology/40.1.24 [CrossRef]
  5. Sharp JT, Lidsky MD, Duffy J, Duncan MW. Infectious arthritis. Arch Intern Med. 1979;139(10):1125–1130. http://www.ncbi.nlm.nih.gov/pubmed/485744. PMID: doi:10.1001/archinte.1979.03630470037014 [CrossRef]485744
  6. Andersen K, Bennedbaek FN, Hansen BL. Septic arthritis in intravenous drug abuse. Ugeskr Laeger. 1994;156(26):3876–3880. http://www.ncbi.nlm.nih.gov/pubmed/8059469. PMID:8059469
  7. Larney S, Peacock A, Mathers BM, Hickman M, Degenhardt L. A systematic review of injecting-related injury and disease among people who inject drugs. Drug Alcohol Depend. 2017;171:39–49. doi:10.1016/j.drugalcdep.2016.11.029 [CrossRef] PMID:
  8. Peterson TC, Pearson C, Zekaj M, Hudson I, Fakhouri G, Vaidya R. Septic arthritis in intravenous drug abusers: a historical comparison of habits and pathogens. J Emerg Med. 2014;47(6):723–728. doi:10.1016/j.jemermed.2014.06.059 [CrossRef] PMID:25282119
  9. Roy S, Ninkovic J, Banerjee S, et al. Opioid drug abuse and modulation of immune function: consequences in the susceptibility to opportunistic infections. J Neuroimmune Pharmacol. 2011;6(4):442–465. doi:10.1007/s11481-011-9292-5 [CrossRef] PMID:21789507
  10. Jones CM, Christensen A, Gladden RM. Increases in prescription opioid injection abuse among treatment admissions in the United States, 2004–2013. Drug Alcohol Depend. 2017;176:89–95. doi:10.1016/j.drugalcdep.2017.03.011 [CrossRef] PMID:28531769
  11. Okie S. A flood of opioids, a rising tide of deaths. N Engl J Med. 2010;363(21):1981–1985. doi:10.1056/NEJMp1011512 [CrossRef] PMID:21083382
  12. Volkow ND, Frieden TR, Hyde PS, Cha SS. Medication-assisted therapies: tackling the opioid-overdose epidemic. N Engl J Med. 2014;370(22):2063–2066. doi:10.1056/NEJMp1402780 [CrossRef] PMID:24758595
  13. Gelberman RH, Menon J, Austerlitz MS, Weisman MH. Pyogenic arthritis of the shoulder in adults. J Bone Joint Surg Am.1980;62(4):550–553. http://www.ncbi.nlm.nih.gov/pubmed/7380854. PMID: doi:10.2106/00004623-198062040-00008 [CrossRef]
  14. Oh DHW, Wurcel AG, Tybor DJ, Burke D, Menendez ME, Salzler MJ. Increased mortality and reoperation rates after treatment for septic arthritis of the knee in people who inject drugs: Nationwide Inpatient Sample, 2000–2013. Clin Orthop Relat Res. 2018;476(8):1557–1565. doi:10.1097/01.blo.0000534682.68856.d8 [CrossRef] PMID:29762153
  15. Agency for Healthcare Research and Quality. Overview of the National (Nationwide) In-patient Sample (NIS). https://www.hcup-us.ahrq.gov/nisoverview.jsp. Accessed May 27, 2018.
  16. Wurcel AG, Anderson JE, Chui KKH, et al. Increasing infectious endocarditis admissions among young people who inject drugs. Open Forum Infect Dis. 2016;3(3):ofw157. doi:10.1093/ofid/ofw157 [CrossRef] PMID:27800528
  17. Cooper HLF, Brady JE, Ciccarone D, Tempalski B, Gostnell K, Friedman SR. Nationwide increase in the number of hospitalizations for illicit injection drug use-related infective endocarditis. Clin Infect Dis. 2007;45(9):1200–1203. doi:10.1086/522176 [CrossRef] PMID:17918083
  18. Zibbell JE, Asher AK, Patel RC, et al. Increases in acute hepatitis C virus infection related to a growing opioid epidemic and associated injection drug use, United States, 2004 to 2014. Am J Public Health. 2018;108(2):175–181. doi:10.2105/AJPH.2017.304132 [CrossRef] PMID:
  19. US Census Bureau. Annual estimates of the resident population for the United States, regions, states, and Puerto Rico: April 1, 2010 to July 1, 2017. https://www.census.gov/data/tables/2017/demo/popest/nation-total.html. Accessed May 27, 2018.
  20. US Census Bureau. Resident population by sex and age. Statistical abstract of the United States. https://www.census.gov/library/publications/2011/compendia/statab/131ed/population.html. Accessed June 21, 2018.
  21. Houchens R, Ross D, Elixhauser A. Using the HCUP national inpatient sample to estimate trends. HCUP methods series report # 2006–05 ONLINE; 2015. http://www.hcupus.ahrq.gov/reports/methods/methods.jsp. Accessed May 27, 2018.
  22. Houchens R, Elixhauser A. Final report on calculating Nationwide Inpatient Sample (NIS) variances for data years 2011 and earlier. http://www.hcup-us.ahrq.gov/reports/methods/methods.jsp. Accessed May 27, 2018.
  23. Houchens R, Ross D, Elixhauser A. final report on calculating National Inpatient Sample (NIS) variances for data years 2012 and later. 2015. http://www.hcup-us.ahrq.gov/reports/methods/methods.jsp. Accessed May 27, 2018.
  24. Centers for Medicare & Medicaid ServicesOffice of the Actuary, National Health Statistics GroupUS Department of CommerceBureau of Economic Analysis, US Bureau of the Census. National health expenditures; aggregate and per capita amounts. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData/NationalHealthAccountsHistorical.html. Accessed August 4, 2018.
  25. Morris BJ, Mir HR. The opioid epidemic: impact on orthopaedic surgery. J Am Acad Orthop Surg. 2015;23(5):267–271. doi:10.5435/JAAOS-D-14-00163 [CrossRef] PMID:25911660
  26. Alford DP, Zisblatt L, Ng P, et al. SCOPE of Pain: an evaluation of an opioid risk evaluation and mitigation strategy continuing education program. Pain Med. 2016;17(1):52–63. doi:10.1111/pme.12878 [CrossRef] PMID:26304703
  27. Zisblatt L, Hayes SM, Lazure P, Hardesty I, White JL, Alford DP. Safe and competent opioid prescribing education: increasing dissemination with a train-the-trainer program. Subst Abus. 2017;38(2):168–176. doi:10.1080/08897077.2016.1275927 [CrossRef] PMID:28418816
  28. Kim JB, Ejiofor JI, Yammine M, et al. Surgical outcomes of infective endocarditis among intravenous drug users. J Thorac Cardiovasc Surg. 2016;152(3):832–841.e1. doi:10.1016/j.jtcvs.2016.02.072 [CrossRef] PMID:27068439
  29. Menendez ME, van Dijk CN, Ring D. Who leaves the hospital against medical advice in the orthopaedic setting?Clin Orthop Relat Res.2015;473(3):1140–1149. doi:10.1007/s11999-014-3924-z [CrossRef] PMID:

Demographics for Patients With Septic Arthritis of the Shoulder in the United States, Stratified by Injection Drug Use: 2000–2013, Nationwide Inpatient Samplea

IndicatorNo. (%)Pb

All Septic ArthritisSeptic Arthritis Without Injection Drug UseSeptic Arthritis With Injection Drug Use
Patientsc14,193 (100)12,061 (85)2128 (15)
Age, y
  15–341685 (12)1352 (11)333 (16).028
  35–547306 (51)6000 (50)1306 (61)<.001
  55–645198 (37)4709 (39)489 (23)<.001
Sex
  Male9549 (67)8155 (68)1394 (66).417
  Female4635 (33)3901 (32)733 (34).417
Race
  White7644 (66)6612 (68)1032 (56)<.001
  Black1848 (16)1456 (15)393 (21).011
  Hispanic1470 (13)1138 (12)331 (18).005
  Asian or Pacific Islander165 (1)154 (2)11 (1).062
  Native American137 (1)116 (1)21 (1).943
  Other346 (2)288 (3)58 (3).820
Primary insurance
  Medicare2881 (20)2532 (21)349 (16).026
  Medicaid2351 (17)1663 (14)688 (32)<.001
  Private insurance6288 (44)5872 (49)411 (19)<.001
  Self-pay1405 (10)987 (8)419 (20)<.001
  No charge167 (1)111 (1)56 (3).070
  Other1077 (8)881 (7)195 (9).237

Association of Injection Drug Use With Hospitalization Outcomes Among Patients With Septic Arthritis of the Shoulder, United States: 2000–2013, Nationwide Inpatient Samplea

Hospital OutcomeAdjusted Odds Ratio (95% Confidence Interval)Pb
Left against medical advice5.54 (3.22–9.55)<.001
Underwent arthroscopy0.76 (0.61–0.95).017
Underwent repeat arthroscopy0.75 (0.53–1.06).100
Underwent arthrotomy1.23 (0.96–1.57).104
Underwent repeat arthrotomy1.80 (0.88–3.70).107
Underwent arthrocentesis1.05 (0.81–1.36).737
Underwent repeat arthrocentesis0.54 (0.24–1.19).125
Died1.89 (0.55–6.49).313

Hospitalization Outcomes for Patients With Septic Arthritis of the Shoulder in the United States, Stratified by Injection Drug Use: 2000–2013 Combined, Nationwide Inpatient Sample

Hospitalization OutcomePatients With Injection Drug UseaPatients Without Injection Drug UseaAdjusted Difference (Injection Drug Use-No Injection Drug Use)aPb
Length of stay, mean (range), d11.6 (10.5 to 12.7)7.8 (7.5 to 8.2)3.7 (2.4 to 5.0)<.001
Total charge unadjusted for inflation, mean (range), US $59,710 (50,869 to 68,552)46,575 (42,036 to 51,115)12,382 (2630 to 22,134).013
Total charge adjusted for inflation, mean (range), 2013 US $65,315 (55,554 to 75,076)51,004 (46,256 to 55,751)13,250 (2635 to 23,866).014
Left against medical advice (range)8.3% (5.9% to 1.1%)1.4% (1.0% to 2.0%)6.5% (3.5% to 9.4%)<.001
Underwent arthroscopy (range)34.0% (29.8% to 38.6%)43.9% (42.0% to 46.0%)−6.4% (−11.6% to −1.3%).014
Underwent repeat arthroscopyc (range)11.1% (8.3% to 14.5%)14.9% (13.5% to 16.4%)−3.2% (−6.8% to 0.3%).075
Underwent arthrotomy (range)27.4% (23.5% to 31.75%)25.1% (23.3% to 26.9%)3.9% (−0.9% to 8.8%).114
Underwent repeat arthrotomyc (range)3.6% (2.0% to 6.1%)2.0% (1.5% to 2.7%)1.5% (−0.7% to 3.7%).174
Underwent arthrocentesis (range)29.4% (25.0% to 34.1%)26.2% (24.4% to 28.0%)0.9% (−4.3% to 6.1%).738
Underwent repeat arthrocentesisc (range)1.6% (0.8% to 3.2%)2.8% (2.2% to 3.6%)−1.4% (−2.9% to 0.1%).064
Mortality (range)0.9% (0.3% to 2.5%)0.5% (0.3% to 0.9%)0.5% (−0.6% to 1.6%).364
Authors

The authors are from the Department of Public Health and Community Medicine (AGW), Tufts University School of Medicine (AJT, DHWO, DJT), and the Department of Geographic Medicine and Infectious Diseases (AGW) and the Department of Orthopaedics (MEM, SPR, MJS), Tufts Medical Center, Boston, Massachusetts.

The authors have no relevant financial relationships to disclose.

Correspondence should be addressed to: Matthew J. Salzler, MD, Department of Orthopaedics, Tufts Medical Center, Biewend Bldg, 7th Fl, 800 Washington St, Box 306, Boston, MA 02111 ( msalzler@tuftsmedicalcenter.org).

Received: November 08, 2018
Accepted: April 15, 2019
Posted Online: April 23, 2020

10.3928/01477447-20200415-03

Sign up to receive

Journal E-contents