Abstract
This cross-sectional study examines the rates of group A Streptococcus skin and soft tissue infections among persons experiencing homelessness.
Skin and soft tissue infections (SSTIs) are common among persons experiencing homelessness (PEH).1 While studies have investigated higher rates of methicillin-resistant Staphylococcus aureus (MRSA) SSTIs among PEH,2 group A Streptococcus (GAS) also causes SSTIs and requires different first-line antibiotic coverage.3 Given reports of invasive GAS outbreaks among PEH,4,5,6 we evaluated the association between homelessness and GAS SSTIs.
Methods
We performed a retrospective, cross-sectional analysis of inpatients seen by the dermatology consult services at University of California, San Francisco, Moffitt-Long Hospital or Zuckerberg San Francisco General Hospital between March 2018 and March 2020 for SSTI with test-positive bacterial skin cultures. We collected data through patient chart review. Housing status was dichotomized, with patients classified as homeless if they reported no primary address or an institutional primary address within the past 12 months. Monomicrobial and polymicrobial culture results were included. Racial and ethnic categorization (Asian, Black, White, other) was based on patient self-identification. Other racial and ethnic group information collected included patients self-identifying as American Indian, Alaska Native, Native Hawaiian, other Pacific Islander, or 2 or more races or ethnicities. The University of California, San Francisco, institutional review board approved this study, waiving patient informed consent for use of de-identified data.
To quantify referring clinicians’ therapeutic practices, rates of first-line antibiotic coverage against GAS (beta-lactam antibiotics) or MRSA (vancomycin, doxycycline, linezolid, daptomycin, trimethoprim-sulfamethoxazole) were compared between housing groups using χ2 testing. We determined associations between housing status (primary predictor) and both GAS SSTI (primary outcome) and MRSA SSTI (secondary outcome) in logistic regression after controlling for confounders identified through literature review, including age, gender, race, ethnicity, alcohol use disorder, and injection drug use. P < .05 was considered statistically significant.
Results
Among 239 patients with SSTI, 181 with microbiology data were included (age 51.9 [SD 19.8] years; 70 [39%] female patients; 52 [29%] PEH) (Table 1). Persons experiencing homelessness had significantly higher rates of ecthyma (21% [95% CI, 10%-32%] vs 3% [95% CI, 0.1%-6%]; P = .001) and ectoparasitic disease (8% [95% CI, 0.4%-15%] vs 0.8% [95% CI, 0%-2%]; P = .01) (Table 2). At the time of dermatology consultation, PEH had significantly higher rates of first-line MRSA coverage (91% [95% CI, 82%-99%] vs 70% [95% CI, 61%-80%]; P = .008) and nonsignificantly reduced rates of first-line GAS coverage (30% [95% CI, 16%-43%] vs 47% [95% CI, 37%-58%]; P = .05).
Table 1. Demographic Characteristics of Study Population Stratified by Housing Status.
| Characteristic | No. (%) | P value | |
|---|---|---|---|
| Stable Housing (n = 129) | Homeless (n = 52)a | ||
| Received care at SFGH | 31 (24) | 33 (63) | <.001b |
| Male | 74 (57) | 37 (71) | .09 |
| Non-Hispanic race and ethnicity | |||
| Asian | 23 (18) | 5 (10) | .28 |
| Black | 24 (19) | 9 (17) | |
| White | 58 (45) | 31 (60) | |
| Otherc | 24 (19) | 7 (13) | |
| Hispanic | 16 (12) | 5 (10) | .60 |
| Age, mean (SD) y | 53 (22) | 50 (14) | .41 |
| Health insurance coverage | <.001b | ||
| Medi-Cal or Healthy San Franciscod | 46 (36) | 37 (71) | |
| Medicare | 44 (34) | 7 (13) | |
| Private/other coverage | 36 (28) | 6 (12) | |
| Uninsured | 3 (2) | 2 (4) | |
| Injection drug use | 6 (5) | 14 (27) | <.001b |
| Alcohol use disorder | 7 (5) | 8 (15) | .03b |
| HIV infectione | 8 (10) | 11 (24) | .03b |
| Hepatitis C infectionf | 3 (5) | 12 (35) | <.001b |
| Proportion with primary SSTIg | 44 (34) | 29 (56) | .01b |
Abbreviations: SFGH, Zuckerberg San Francisco General Hospital and Trauma Center; SSTI, skin and soft tissue infection.
Homelessness was defined as a patient reporting no primary residence or listing a homeless shelter, hotel, place of worship, or hospital as their home address within the past 12 months.
Statistically significant (P < .05).
Self-identified as American Indian, Alaska Native, Native Hawaiian, Other Pacific Islander, or 2 or more races or ethnicities.
Healthy San Francisco enables uninsured residents of San Francisco who are not Medicare or Medi-Cal eligible to access medical services in San Francisco County.
Includes 83 patients with stable housing and 46 persons experiencing homelessness.
Includes 64 patients with stable housing and 34 persons experiencing homelessness.
Primary SSTI includes abscess, carbuncle, cellulitis, ecthyma, erysipelas, folliculitis, furuncle, or impetigo.
Table 2. Microbiologic Breakdown Stratified by Housing Status and Diagnostic Category.
| Characteristic | Stable housing (n = 129) | Homeless (n = 52) | P valuea | ||
|---|---|---|---|---|---|
| Proportion with GAS | Proportion with MRSA | Proportion with GAS | Proportion with MRSA | ||
| Primary SSTIb | |||||
| Abscess(es) | 1/6 | 4/6 | 1/2 | 1/2 | .81 |
| Carbuncle(s) or furuncle(s) | 0/4 | 4/4 | 0/1 | 1/1 | .66 |
| Cellulitis | 3/21 | 3/21 | 7/14 | 3/14 | .10 |
| Ecthyma | 0/4 | 2/4 | 7/11 | 7/11 | .001c |
| Erysipelas | 0/1 | 0/1 | 0/0 | 0/0 | .52 |
| Folliculitis | 0/3 | 0/3 | 0/0 | 0/0 | .27 |
| Impetigo | 1/5 | 1/5 | 0/1 | 0/1 | .51 |
| Secondary SSTId | |||||
| Acneiform disorder (eg, HS) | 0/4 | 0/4 | 0/0 | 0/0 | .20 |
| Bullous or vesicular disease | 1/9 | 2/9 | 0/0 | 0/0 | .05 |
| Burn, diabetic/traumatic ulcer | 1/3 | 2/3 | 1/1 | 0/1 | .87 |
| Cutaneous T-cell lymphoma | 0/3 | 1/3 | 0/0 | 0/0 | .27 |
| Drug reaction | 0/6 | 1/6 | 0/0 | 0/0 | .11 |
| Ectoparasitic disease (eg, lice) | 0/1 | 0/1 | 4/4 | 0/4 | .01c |
| Eczematous dermatitis | 3/15 | 4/15 | 0/4 | 1/4 | .43 |
| Fungal or mycobacterial skin infection (eg, tinea) | 1/1 | 0/1 | 0/0 | 0/0 | .52 |
| Neutrophilic dermatoses | 0/7 | 0/7 | 0/0 | 0/0 | .09 |
| Pruritus or excoriation disorders (eg, multifactorial pruritus, PN) | 3/6 | 0/6 | 1/1 | 0/1 | .39 |
| Psoriasis or seborrheic dermatitis | 0/3 | 1/3 | 1/3 | 1/3 | .24 |
| Rheumatologic disease | 1/1 | 0/1 | 0/0 | 0/0 | .52 |
| Syphilis infection | 0/3 | 0/3 | 0/0 | 0/0 | .27 |
| Vasculitis or vasculopathy | 0/3 | 1/3 | 0/1 | 1/1 | .87 |
| Venous stasis dermatitis/ulcer or lymphedema | 1/12 | 0/12 | 0/8 | 1/8 | .20 |
| Viral skin infection (eg, HSV) | 3/8 | 4/8 | 0/1 | 1/1 | .23 |
Abbreviations: GAS, group A Streptococcus; HS, hidradenitis suppurativa; HSV, herpes simplex virus; MRSA, methicillin-resistant Staphylococcus aureus; PN, prurigo nodularis; SSTI, skin and soft tissue infection.
Two-proportion z-test comparing rates of clinical diagnostic categories between housing groups.
Primary SSTI includes abscess, carbuncle, cellulitis, ecthyma, erysipelas, folliculitis, furuncle, or impetigo (44 patients with stable housing, 29 patients experiencing homelessness).
Statistically significant (P < .05).
Secondary SSTI includes underlying dermatologic disease with test-positive skin or tissue culture results requiring antibiotics (85 patients with stable housing, 23 patients experiencing homelessness).
Among PEH, 42% (95% CI, 29%-57%) had GAS-positive cultures and 33% (95% CI, 20%-47%) had MRSA-positive cultures. Among patients with stable housing, 15% (95% CI, 9%-22%) had GAS-positive cultures and 23% (95% CI, 16%-32%) had MRSA-positive cultures. Prior to adjustment, PEH had significantly higher odds of GAS SSTI (OR, 4.25 [95% CI, 2.04-8.85]; P < .001) and nonsignificantly higher odds of MRSA SSTI (OR, 1.60 [95% CI, 0.79-3.26]; P = .19) relative to patients with stable housing. After controlling for confounders, PEH had significantly higher odds of GAS SSTI (OR, 4.25 [95% CI, 1.86-9.70]; P = .001) and nonsignificantly higher odds of MRSA SSTI (OR, 1.49 [95% CI, 0.68-3.27]; P = .33).
Discussion
Our finding that PEH were more likely to have first-line MRSA coverage suggests that health care professionals may prioritize MRSA among PEH presenting with SSTI. Given that 42% of PEH had cultures that tested positive for GAS and PEH had 4.25 times higher odds of GAS SSTI, the present study results suggest that health care professionals may need to consider GAS more strongly among PEH presenting with SSTI.
This study has several limitations. First, we were unable to differentiate between purulent and non-purulent forms of SSTI and cannot confirm that purulent SSTIs were uniformly overrepresented. Second, GAS SSTIs excluded owing to lack of microbiologic data could be differential between the 2 groups. Third, our cohort likely had more severe skin disease relative to outpatients with SSTI or inpatients with SSTI managed without dermatology.
Conclusions
In summary, we found that homelessness was independently associated with GAS SSTI in our study population. Future studies are needed to demonstrate causation, establish a more precise effect size estimate, and evaluate outcomes among PEH receiving empirical GAS coverage for SSTI.
References
- 1.Adly M, Woo TE, Traboulsi D, Klassen D, Hardin J. Understanding dermatologic concerns among persons experiencing homelessness: a scoping review and discussion for improved delivery of care. J Cutan Med Surg. 2021;25(6):616-626. doi: 10.1177/12034754211004558 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Loewen K, Schreiber Y, Kirlew M, Bocking N, Kelly L. Community-associated methicillin-resistant Staphylococcus aureus infection: literature review and clinical update. Can Fam Physician. 2017;63(7):512-520. [PMC free article] [PubMed] [Google Scholar]
- 3.Stevens DL, Bisno AL, Chambers HF, et al. ; Infectious Diseases Society of America . Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014 update by the Infectious Diseases Society of America. Clin Infect Dis. 2014;59(2):e10-e52. doi: 10.1093/cid/ciu296 [DOI] [PubMed] [Google Scholar]
- 4.Cady A, Plainvert C, Donnio P-Y, et al. Clonal spread of Streptococcus pyogenes emm44 among homeless persons, Rennes, France. Emerg Infect Dis. 2011;17(2):315-317. doi: 10.3201/eid1702.101022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Dohoo C, Stuart R, Finkelstein M, Bradley K, Gournis E. Risk factors associated with group A Streptococcus acquisition in a large, urban homeless shelter outbreak. Can J Public Health. 2020;111(1):117-124. doi: 10.17269/s41997-019-00258-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Valenciano SJ, Onukwube J, Spiller MW, et al. Invasive group A Streptococcal infections among people who inject drugs and people experiencing homelessness in the United States, 2010-2017. Clin Infect Dis. 2021;73(11):e3718-e3726. doi: 10.1093/cid/ciaa787 [DOI] [PMC free article] [PubMed] [Google Scholar]