Overview

The review was guided by methodology outlined by the PRISMA statement (S1 Checklist). [7] A systematic search strategy was developed to identify articles that reported foot problems among homeless populations. A protocol of the review has not been previously published.

Eligibility Criteria

All original research articles that reported on foot conditions in homeless persons were included in the review. Articles that did not present data on clear foot health outcomes (e.g. reviews, case reports, and commentaries) were excluded.

Information Sources and Search Strategy

Articles were identified through a search of MEDLINE (1966–2016), EMBASE (1947–2016), and CINAHL (1982–2016) with no language exclusions. The search strategy was developed by one member of the research team (MJT) in consultation with a health sciences librarian and included terms related to homelessness that were cross-matched with foot health terms (Box 1). The yield from bibliographic databases was complemented by manual searches of reference lists from relevant research articles. The search concluded in July 2016.

Study Selection

Titles and abstracts of potentially eligible articles were screened by two independent reviewers (MJT, TDB). If it was unclear whether an article met the inclusion criteria, the full-text article was retrieved to determine eligibility. Articles that met the inclusion criteria were included in full-text analysis. Discrepancies in the screening process were resolved by consensus.

Data Extraction and Analysis

A standardized data extraction form was developed by the research team. Data were extracted by two independent reviewers (MJT, TDB) on the following variables: general study characteristics (e.g. primary author, year of publication, country of study, and source of funding); participants (age, sex, and sociodemographic data); study design; foot assessment methods; foot conditions and associated interventions; study findings. Quality of reports was independently assessed by two research team members (MJT, TDB) using the Downs and Black checklist. [8] The checklist has high internal consistency (Kruder-Richardson formula 20 = 0.89), test-retest reliability (r = 0.88), and inter-rater reliability (r = 0.75). The 27-item checklist assesses items related to reporting (clear description of objectives, participants, outcomes, findings, confounders, loss to follow-up, probability values), internal validity (biases, statistical analyses), external validity (representativeness of participants and treatment setting), and power. For the current review, 6 items related to randomized trials were excluded as no randomized interventions were identified. The checklist item that assessed adequacy of a study’s statistical power was simplified to a score of 0 or 1. The maximum quality score that could be attained was 22. Discrepancies were resolved by consensus. Data on all study variables were summarized. Quantitative variables were analyzed using Microsoft Excel.

Setting

Twelve studies included in the review were conducted in the United States, [9]-[20] two in the United Kingdom, [21], [22] two in France, [23], [24] and one in Italy. [25] The majority of studies included homeless shelter residents or patients who attended clinics dedicated to serving vulnerable populations. Seven (50%) studies included patients from clinics that served homeless individuals, [9], [10], [13]-[15], [21], [22] while eight (47%) studies recruited participants directly from homeless shelters. [11], [12], [16], [17], [19], [20], [23], [24] One study recruited homeless individuals from a volunteer health service outreach association, [25] and another study recruited homeless individuals from a community kitchen. [18] Ten (59%) studies included participants from more than one location (e.g. shelters, clinics). [10], [11], [13], [15], [16], [19], [29], [23]-[25]

Sample

Studies included a median number of 363 participants (range 95–930). The mean age of participants across studies was 40 years old (mean ages ranged from 32–48 years old). Five studies did not provide mean age of participants.[13], [19], [21]-[23] Although two studies included adolescents and one included children, no additional details were provided about these participants. [13],[14] All studies enrolled a majority of male participants, ranging from 51–100% of study participants. Definitions of homelessness varied across studies and were generally inadequately described. Three studies used duration of homelessness as a criterion for study enrollment.[10], [11], [15] While duration of homelessness was provided in a total of seven studies, there were no details about participants’ moves and housing transitions. Eleven (65%) studies included information about participants’ ethnic background or country of origin. Study samples were generally comprised of a higher number of individuals from minority populations (e.g. African-American). However, no studies assessed differences in outcomes by race. Only six (35%) studies provided information about participants’ level of education, employment, or income.[10], [11], [15]-[17], [19] Health insurance coverage rates were reported in three studies in the United States ranging from 14%-47%.[10], [15], [19]

Funding

Seven (41%) studies were supported by a government agency. Five studies reported a non-governmental source of funding and eight studies did not include a source of funding.

Study Design and Foot Assessment

The majority of studies were observational and descriptive in nature, and only one measured the effects of a foot care intervention. Most studies involved a participant questionnaire and/or a combination of a foot assessment and retrospective chart review of health clinics that served homeless individuals. For 13 (76%) studies, foot health of homeless individuals involved a clinical assessment. Clinical foot exams were often performed by a clinician or medical student. Four (24%) studies retrieved information about foot health solely through self-reported survey data. One larger case-control study involving 698 individuals compared residents of homeless shelters with people attending a pre-travel clinic. [24] Only two studies longitudinally assessed participants’ foot concerns over time. One study compared participants who remained homeless 18 months later with participants who obtained housing. [17] One interventional study of a mobile health service examined the effect of medical treatment on diabetic foot ulcers among homeless individuals during 17.6 ± 12 months. [19] No clinical trials or qualitative studies were identified in the systematic search.

Methodological Quality

The reported methodological quality in the majority of studies was generally moderate with a median score of 12 (range 8–20). The main study objectives were unclear in 2 (12%) studies. The main outcomes were inadequately described in 4 (24%) studies. Studies often lacked details on the sample such as mean age, sociodemographic data, recruitment procedures, and eligibility criteria. Data were often missing from reports without explanation and only 4 (24%) studies reported on participants lost to follow-up. Several studies appeared to have low internal and external validity and appeared to be insufficiently powered to detect clinically meaningful differences.

Foot Conditions

Study participants reported a wide range of foot pathologies. Prevalence of any foot problem ranged from 9% to 65% across study populations. Calluses and corns were among the most common concerns in homeless populations, which were identified in 7.7–57% of study participants. [12], [14], [16], [19] Nail pathologies such as ingrown toenails were common, ranging from 15%-65% across study samples. [14], [16], [19] Foot infections were also highly prevalent among study populations. Tinea pedis was identified as a presenting concern in six studies. [11], [12], [14], [16], [19], [24] Rates of tinea pedis ranged from 3.2% to 38%. In addition, one case-control study involving 698 participants found that homeless individuals were significantly more likely to have tinea pedis and scratching lesions of the socks compared to people presenting to a pre-travel clinic. [24] Upon further analysis, presence of scratching lesions of the socks were independently associated with homelessness. Prevalence of pitted keratolysis of the feet (20.4%) and toenail onychomycosis (15.5%) were reported in one study. [12] Cellulitis was also found in two studies. [16], [22] Prevalence of foot injuries ranged from 24–43% in studies. [14], [19] Foot deformities (14%), trauma (6%) and fractures (2.5%) were common. [16], [21] Bunions (14%), hammertoes (7%), gout (6%), plantar warts (2–4%), and foot ulcers (0.7–4%) were also identified. [12], [14], [19] Foot problems related to neurological disorders ranged from 37–43% in one study. [14] Acute medical problems such as deep vein thrombosis, frostbite, and gangrene were also found. [16] One study found the prevalence of frostbite among homeless participants was 2%. [19] In the same study, immersion (trench) foot was observed in 5% of participants. [19]

Foot pathologies related to diabetes were found in several studies. [13], [18], [19], [23] Prevalence of diabetes ranged from 6.2–23%. Arnaud et al. screened 488 homeless shelter residents for diabetes which identified 35 previously diagnosed and 2 newly diagnosed individuals with diabetes. [23] They found that 41% of homeless individuals with diabetes had difficulty walking, 42% had a loss of foot sensitivity, 43% had permanently reduced mobility, and 17% had experienced lower limb amputation. One in three homeless persons with diabetes had a high or very high podiatric risk as defined by international classification, warranting regular foot care. [23] Macnee et al. screened 58 homeless individuals for diabetes using capillary blood sugar values and found that 12% of patients who attended the screening clinic reported foot problems. [13] In one study of homeless adults who used foot care services, 56% reported a history of diabetes, hypertension, peripheral vascular disease or a combination of the three. [18] In one study of 299 homeless shelter residents, approximately 9% self-reported having diabetes with 5% having diabetes mellitus type 2 and 4% having diabetes mellitus type 1. [19] However, 16% of participants reported numbness, 21% had tingling in their feet, and 21% had swollen feet suggesting that some individuals may have had undiagnosed diabetes. [19] In this study, 3% of participants reported having peripheral artery disease. [19]

The only interventional study identified in the review was a prospective study of 930 homeless individuals which included 21(2.3%) individuals with diabetic foot ulcers and were treated with removal of necrotic tissue, incision and drainage of infected areas and abscesses, wound care, antibiotics and analgesics as needed weekly by a mobile health services team for a mean (SD) of 17.6 ± 12 months. [25] Participants requiring complex interventions such as surgery or revascularization were hospitalized. Protective footwear was provided to all participants. Eighteen (86%) participants had significant improvement: 13 had their condition completely resolved, and 5 showed partial improvement and no longer needed medication. [25] One patient required foot amputation and negative pressure wound therapy and later died of septic shock and kidney failure. Two patients required amputation due to worsening of ulcers. [25]

One study found no significant differences in the reporting of foot problems between homeless men and women. [20] One longitudinal, non-interventional study found that homeless participants’ podiatric concerns significantly decreased over time from 12.4% at baseline to 5.7% at 18 months. [17]

Limitations

This study was limited by a small body of literature related to foot conditions among homeless persons, with few studies specifically reporting on foot health as a primary study outcome. The review may not have been representative of the studies conducted on foot health among homeless individuals due to publication bias resulting from the small number of published articles on the topic. Low reported quality of several studies, which was evidenced by lack of details on participant characteristics and foot conditions, also limited the conclusions that could be drawn from them.

Our results are consistent with previous narrative reviews, case reports, and commentaries emphasizing the large burden of foot problems among homeless individuals, suggesting that more high quality data is needed. [3], [5], [6] Despite the prevalence of foot problems in this population, there are significant gaps between foot health needs and research studies to address these needs.

There is room for improvement in the quality of research reports given that a substantial proportion of studies did not include clear objectives and outcomes, provided limited data on participant samples, had low internal and external validity, and were insufficiently powered to detect clinically significant differences.

Future studies should assess foot health outcomes longitudinally and explore demographic and health status indicators that are associated with foot-related concerns. Comparative studies of homeless and housed individuals could yield more epidemiologic data and identify potential areas for intervention. Qualitative studies could further investigate facilitators and barriers to maintaining good foot health and accessing foot care services. Development of effective interventions to address the unmet foot health needs of this population should be prioritized. Interventional studies could explore multidisciplinary treatment approaches, provision of essential items such as clean socks and properly-fitting shoes, along with educational and outreach initiatives. Mobile interventions may be particularly effective in this population given the high burden of unmet medical and social needs.