Abstract
Background
Health disparities have important effects on orthopaedic patient populations. Socioeconomic factors and poor nutrition have been shown to be associated with an increased risk of complications such as infection in patients undergoing orthopaedic surgery. Currently, there are limited published data on how food insecurity is associated with medical and surgical complications.
Questions/purposes
We sought to (1) determine the percentage of patients who experience food insecurity in an orthopaedic trauma clinic at a large Level 1 trauma center, (2) identify demographic and clinical factors associated with food insecurity, and (3) identify whether there are differences in the risk of complications and reoperations between patients who experience food insecurity and patients who are food-secure.
Methods
This was a cross-sectional study using food insecurity screening surveys, which were obtained at an orthopaedic trauma clinic at our Level 1 trauma center. All patients 18 years and older who were seen for an initial evaluation or follow-up for fracture care between November 2022 and February 2023 were considered for inclusion in this study. For inclusion in this study, the patient had to have surgical treatment of their fracture and have completed at least one food insecurity screening survey. Ninety-eight percent (121 of 123) of patients completed the screening survey during the study period. Data for 21 patients were excluded because of nonoperative treatment of their fracture, nonfracture-related care, impending metastatic fracture care, and patients who had treatment at an outside facility and were transferring their care. This led to a study group of 100 patients with orthopaedic trauma. The mean age was 51 years, and 51% (51 of 100) were men. The mean length of follow-up available for patients in the study was 13 months from the initial clinic visit. Patient demographics, hospital admission data, and outcome data were collected from the electronic medical records. Patients were divided into two cohorts: food-secure versus food-insecure. Patients were propensity score matched for adjusted analysis.
Results
A total of 37% of the patients in this study (37 of 100) screened positive for food insecurity during the study period. Patients with food insecurity were more likely to have a higher BMI than patients with food security (32 kg/m2 compared with 28 kg/m2; p = 0.009), and they were more likely not to have healthcare insurance or to have Medicaid (62% [23 of 37] compared with 30% [19 of 63]; p = 0.003). After propensity matching for age, gender, ethnicity, current substance use, Charleston comorbidity index, employment status, open fracture, and length of stay, food insecurity was associated with a higher percentage of superficial infections (13% [4 of 31] compared with 0% [0 of 31]; p = 0.047). There were no differences between the groups in the risk of reoperation, deep infection, and nonunion.
Conclusion
Food insecurity is common among patients who have experienced orthopaedic trauma, and patients who have it may be at increased risk of superficial infections after surgery. Future research in this area should focus on defining these health disparities further and interventions that could address them.
Level of Evidence
Level III, therapeutic study.
Introduction
The effect of socioeconomic factors on disparities in medicine, such as unequal access to healthcare, healthcare use, and patient outcomes, has been well documented [19]. There have been reports on racial and ethnic disparities regarding surgical complications and mortality rates, with varying results [15]. It has also been reported that socioeconomic factors are associated with differences in the timing of operative treatment as well as use of surgery [13]. Although healthcare disparities have been widely examined in spine surgery and total joint arthroplasty, some recent investigations have included orthopaedic trauma populations [5, 20]. Overall, substantial healthcare disparities continue to exist in orthopaedic patient populations. The impact of malnutrition and food insecurity is one topic that is gaining interest [14].
However, the available evidence provides little information on the incidence and clinical impact of food insecurity in patients with musculoskeletal injuries. The United States Department of Agriculture defines food insecurity as “a household-level economic and social condition of limited or uncertain access to adequate food” [17]. Food insecurity is associated with malnutrition in older patients [16], which has important implications, because malnutrition has been linked to a higher risk of complications in patients with musculoskeletal conditions [14]. Recent reports have suggested that patients who have experienced orthopaedic trauma may be at increased risk of food insecurity compared with the national mean [7, 8, 18]. This may be in part because of the socioeconomic implications that sustaining a traumatic orthopaedic injury has on a patient, including temporary or permanent loss of employment and wages and decreased mobility. However, little has been published regarding food insecurity in the orthopaedic trauma population, and none in the outpatient setting [18].
We sought to (1) determine the percentage of patients who experience food insecurity in an orthopaedic trauma clinic at a large Level 1 trauma center, (2) identify demographic and clinical factors associated with food insecurity, and (3) identify whether there are differences in the risk of complications and reoperations between patients who experience food insecurity and patients who are food-secure.
Patients and Methods
Study Design, Setting
This was a cross-sectional study that evaluated patient data from an orthopaedic trauma clinic at a Level 1 academic trauma center in south-central Texas. The patients whose data were obtained for this study were from one out of the five orthopaedic trauma clinics at this trauma center, which has a large catchment area that encompasses West Texas and the southern border of Texas and Mexico. All patient data were retrieved from electronic medical records. At this clinic, a routine screening protocol for food insecurity during patient intake was implemented as part of clinical care. All patients whose data were included in this series were screened during their scheduled follow-up visits in the setting of a private clinical examination room.
Patient Inclusion and Exclusion Criteria
Data for patients who completed the clinic-implemented food insecurity survey between November 2022 and February 2023 were considered eligible for inclusion in this study. Data for patients 18 years and older who presented for their clinic visit after surgical fracture fixation were included. Patients undergoing nonoperative fracture treatment, surgical treatment at an outside facility and transferring care, nonfracture-related care, or prophylactic fixation of impending metastatic fractures were excluded from this study.
During the study period, 300 clinic patient encounters were conducted. Of these, approximately 59% were follow-up visits (177 of 300). This amounted to 123 unique patients seen and evaluated in the clinic during the study period. Ninety-eight percent (121 of 123) of patients completed the screening for food insecurity during the study period. After applying the inclusion and exclusion criteria outlined above, data for 100 patients were included in the analysis (Fig. 1). These patients had a mean length of follow-up of 13 months after the initial patient encounter in this cross-sectional study.
Fig. 1.
This figure demonstrates the inclusion and exclusion criteria and process that resulted in the 100 patients included in the final study analysis.
Descriptive Data
Of the 100 patients included in the analysis, 49% (49 of 100) were women, and the mean age was 51 years (Table 1). Most patients who had food insecurity were Hispanic (68% [25 of 37]), and most of the patients in the food-secure group were non-Hispanic White (52% [33 of 63]). In the food-insecure group, 51% (19 of 37) of the patients were insured, whereas 78% (49 of 63) patients in the food-secure group were insured. Sixty-five percent (24 of 37) of the patients with food insecurity were employed. In contrast, 40% (25 of 37) of the patients who were food-secure were employed.
Table 1.
Patient demographics
| Variable | Negative (n = 63) | Positive (n = 37) | p value |
| Age in years | 52 ± 19 | 49 ± 16 | 0.29 |
| Men | 49 (31) | 54 (20) | 0.79 |
| Marital status | |||
| Single | 30% (19) | 54% (20) | |
| Married | 49% (31) | 24% (9) | |
| Separated or widowed | 21% (13) | 22% (8) | |
| Race or ethnicitya | 0.08 | ||
| Black | 3% (2) | 3% (1) | |
| Hispanic | 44% (28) | 68% (25) | |
| White | 52% (33) | 30% (11) | |
| Insurance status | |||
| Private | 40% (25) | 22% (8) | |
| Medicare | 25% (16) | 14% (5) | |
| Medicaid | 8% (5) | 14% (5) | |
| Veterans Administration | 2% (1) | 0% (0) | |
| Workers Compensation | 3% (2) | 3% (1) | |
| None | 22% (14) | 49% (18) | |
| Insurance other than Medicaid | 78% (49) | 51% (19) | 0.01 |
| Medicaid or without insurance | 30% (19) | 62% (23) | 0.003 |
| Employed | 40% (25) | 65% (24) | 0.03 |
| Laborerb | 11% (7) | 19% (7) | 0.43 |
Data presented as mean ± standard deviation or % (n). Food insecurity screening was performed with the Hunger Vital Sign screening tool.
Ethnicity was self-reported by patient.
Laborer was defined as a patient who performs physical work for a living.
Screening Tool
The food insecurity screening survey used in this study was the well-known Hunger Vital Sign. This two-question survey is a subset of questions from the United States Department of Agriculture’s 18-item Household Food Security Survey, which has been validated to use as a screening tool to identify patients who are experiencing food insecurity. Answering Question 1 or 2 with an affirmative response (responding “often true” or “sometimes true” to either question) is reported to have a sensitivity of 93% or 82% and specificity of 85% or 95% in identifying patients with food insecurity, respectively [4]. The two-question screening tool was implemented in the same manner for this study.
The two questions in the screening tool are: “Within the past 12 months, the food we bought just didn’t last and we didn’t have money to get more,” and “Within the past 12 months, we worried whether our food would run out before we got money to buy more.” Patients were asked to select a response from the options of never true, sometimes true, and often true.
Patients who screened positive for food insecurity by selecting an affirmative response to either question were then provided with resources to help improve access to healthy food as part of clinical care. This included contact information and education by clinic staff on state-based and community-based food programs including Supplemental Nutrition Assistance Program, Catholic Charities of San Antonio, Christian Assistance Ministries, the Salvation Army of San Antonio, and the San Antonio Hope Center. In addition, patients screening positive for food insecurity were given a USD 5 food voucher.
Study Groups
In this cross-sectional study, the patient sample was divided into two groups: patients who screened positive (food-insecure) versus negative (food-secure) based on the survey provided in clinic.
Primary and Secondary Study Outcomes
The primary study goals were to determine the proportion of patients with food insecurity in an orthopaedic trauma clinic at a large Level 1 trauma center and identify demographic and clinical factors associated with food insecurity. The demographic information collected included age, gender, race and ethnicity, marital status, insurance status, and employment status. Gender, race or ethnicity, marital status, and employment status were all self-reported by patients. Clinical variables were collected including BMI, serum albumin levels, Charlson comorbidity index, and substance use. Hospital admission data included mechanism of injury, fracture type (open versus closed), fracture location, length of hospital stay, time to fixation, and discharge disposition. The Charlson comorbidity index considers several comorbidities, including diabetes mellitus, peripheral vascular disease, and congestive heart failure.
The secondary study goal was to identify whether there are differences in the risk of complications and reoperations in patients who experience food insecurity compared with patients who are food-secure. Outcome data including length of follow-up, complications, and reoperations were collected. Outcomes were defined as deep infections resulting in surgical debridement (confirmed by positive culture results or presence of pus, and wound breakdown or sinus with communication to the bone or hardware), superficial infection (surgical site infection without involvement of bone or hardware) resulting in the use of oral antibiotics or local wound care, reoperations, and nonunion. Reoperations included in this analysis were performed for a complication-related indication such as symptomatic hardware, infected hardware, or nonunion. None of the reoperations in the analysis were performed for planned, staged procedures.
Ethical Approval
Ethical approval for this study was obtained from the University of Texas Health San Antonio Institutional Review Board (protocol number: 20230635EX).
Statistical Analysis
Covariates that could confound or modify the outcome were identified (Fig. 2) [21]. Covariates selected included age, gender, ethnicity, current substance use, Charlson comorbidity index, employment as laborer, open fractures, and length of stay. Health determinants such as BMI and albumin level were not included because they likely act as mediating variables. Univariate logistic regression analysis confirmed that length of stay, and open fractures were effect modifiers on the outcomes of interest. Matching was performed to adjust for the covariates of age, gender, ethnicity, current substance use, Charlson comorbidity index, employment as laborer, open fractures, and length of stay. Patients in each cohort were matched 1:1 using a nearest-neighbor strategy without replacement and a caliper of 0.2. Propensity scores were estimated with logistic regression. Overall covariate balance was improved after adjustment for confounders (Supplemental Figs. 1 and 2; http://links.lww.com/CORR/B288).
Fig. 2.
This figure demonstrates the rationale for covariate selection. BMI and albumin level were excluded as covariates because they are mediating variables regarding surgical outcomes.
Absolute risk and risk ratios were calculated, along with their associated confidence intervals, for each outcome of interest. The Haldane continuity correction was performed to calculate risk ratios if a cohort provided zero contributions. Nonparametric data were analyzed using Wilcoxon rank sum tests. For categorical variables, we used Pearson chi-square tests, with Rao and Scott adjustments applied to allow for weight adjustments based on the propensity score–matched groups. A p value of 0.05 was considered statistically significant. The statistical analysis was performed using R 4.3.1 with the MatchIt (v4.5.4), cobalt (v4.5.1), and survey (v4.2-1) packages for propensity score analysis [3, 6, 9, 12].
Results
Proportion of Orthopaedic Trauma Patients With Food Insecurity
A total of 37% (37 of 100) of patients screened positive for food insecurity as defined above.
Demographic Factors Associated With Food Insecurity in Orthopaedic Trauma Patients
The groups (food-insecure versus food-secure) did not differ in terms of age, gender, ethnicity, and percentage of laborers (Table 1). A higher percentage of patients who screened positive for food insecurity were without healthcare insurance or had Medicaid insurance compared with those who were food-secure (62% [23 of 37] versus 30% [19 of 63]; p = 0.003). More patients who screened positive for food insecurity were employed (65% [24 of 37]) than those who screened negative (40% [25 of 63]) (p = 0.03). The mean BMI of the group with food insecurity (32 ± 7 kg/m2) was higher than in the group who was food-secure (28 ± 6 kg/m2) (p = 0.009), whereas other health metrics such as Charlson comorbidity index, serum albumin level on admission, and current substance use did not differ between the study groups (Table 2). There were no differences between cohorts regarding injury mechanism, fracture location, open fracture, periarticular injuries, length of stay, time to fixation, or length of follow-up (Table 3).
Table 2.
Patient health characteristics
| Characteristic | Negative (n = 63) | Positive (n = 37) | p value |
| BMI in kg/m2 | 28 ± 6 | 32 ± 7 | 0.009 |
| Charleston comorbidity index | 2.0 ± 2.2 | 1.5 ± 2.0 | 0.35 |
| Serum albumin level at admission | 3.3 ± 0.7 | 3.4 ± 0.7 | 0.66 |
| Current alcohol use | 40% (25) | 38% (14) | 0.86 |
| Current tobacco use | 18% (11) | 30% (11) | 0.16 |
| Current drug use | 11% (7) | 14% (5) | 0.73 |
Data presented as mean ± standard deviation or % (n). Food insecurity screening was performed with the Hunger Vital Sign screening tool.
Table 3.
Injury characteristics
| Characteristic | Negative (n = 63) | Positive (n = 37) | p value |
| Mechanism of injury | |||
| Fall from < 3 meters | 46% (29) | 46% (17) | |
| Motor vehicle collision | 27% (17) | 19% (7) | |
| Fall from > 3 meters | 14% (9) | 3% (1) | |
| Motorcycle collision | 5% (3) | 5% (2) | |
| Motor pedestrian collision | 2% (1) | 8% (3) | |
| Sports-related injury | 5% (3) | 0% (0) | |
| Crush injury | 0% (0) | 8% (3) | |
| Blunt trauma | 0% (0) | 5% (2) | |
| Ballistic trauma | 0% (0) | 3% (1) | |
| Dog bite | 2% (1) | 0% (0) | |
| Bicycle collision | 0% (0) | 3% (1) | |
| Open fracture | 25% (16) | 38% (14) | 0.28 |
| Extremity | |||
| Lower | 84% (53) | 81% (30) | |
| Upper | 10% (6) | 14% (5) | |
| Combined | 6% (4) | 5% (2) | |
| Periarticular injury | 67% (42) | 70% (26) | 0.88 |
| Time to definitive fixation in days | 3.4 ± 5.1 | 3.1 ± 4.2 | 0.98 |
| Length of stay in days | 7.1 ± 10.8 | 6.6 ± 11.3 | 0.58 |
| Discharge destination | |||
| Home | 65% (41) | 84% (31) | |
| Inpatient rehab | 19% (12) | 14% (5) | |
| Skilled nursing facility | 14% (9) | 3% (1) | |
| Length of follow-up in days | 341 ± 391 | 464 ± 527 | 0.21 |
Data presented as % (n) or mean ± standard deviation. Food insecurity screening was performed with the Hunger Vital Sign screening tool.
Is Food Insecurity Associated With a Higher Risk of Complications?
After controlling for potentially confounding factors of age, gender, ethnicity, current substance use, Charlson comorbidity index, employment as laborer, open fractures, and length of stay, there were no differences in the risk of reoperation, deep infection, and nonunion between the two groups (Table 4). There was a slightly higher association with superficial infection (13% [4 of 31] versus 0% [0 of 31], risk ratio 8.26 [95% CI 0.46 to 150]; p = 0.047) in the group with food insecurity.
Table 4.
Propensity-matched outcomes data
| Outcome | Food insecurity screening | Risk ratio (95% CI) | p value | |
| Negative (n = 31) | Positive (n = 31) | |||
| Reoperation | 19% (6) | 29% (9) | 1.50 (0.61-3.71) | 0.39 |
| Infection | ||||
| Superficial | 0% (0) | 13% (4) | 8.26 (0.46-150) | 0.047 |
| Deep | 3% (1) | 13% (4) | 4.00 (0.47-34) | 0.16 |
| Nonunion | 13% (4) | 10% (3) | 0.75 (0.18-3.08) | 0.67 |
Food insecurity screening was performed with the Hunger Vital Sign screening tool.
Discussion
The proportion of patients screening positive for food insecurity in the orthopaedic trauma population of this study was 37%, which is much higher than estimates of the national prevalence of 14% [2]. This finding suggests that a subset of orthopaedic trauma patients face severe disparities in access to food. Patients who experienced food insecurity during the study period were more likely to be less affluent, as demonstrated by the higher proportion of patients without insurance or patients receiving government assistance. There appeared to be an association between the patient experience of food insecurity and superficial infections, but this association was smaller than expected. There was no difference between patients with and without food insecurity in terms of deep infections, reoperations, and nonunion.
Limitations
Given the small sample and low incidence of complications in each of the study groups, this study is prone to statistical fragility. A few more or fewer outcome events in each arm of the study could have resulted in a change in a main finding. In addition, we found relatively few differences in outcomes between patients with and without food insecurity, but this may have been a function of insufficient statistical power. Further studies with larger samples and more statistical power are required to validate these results. Additionally, although the risk of superficial infection was higher in patients with food insecurity, the effect size may not be large as suggested by the risk ratio. Because no patients without food insecurity sustained a superficial infection, statistical continuity correction was required to avoid an infinite risk ratio. Moreover, the food insecurity screening was performed in the outpatient setting. As such, we cannot draw any conclusions about food insecurity among patients who are admitted to the hospital. Similarly, our study only included patients who presented for follow-up after surgical fracture care. Therefore, we cannot make any assumptions about patients undergoing nonsurgical treatment of musculoskeletal injuries.
Because this study relied on chart review for the study data, it is prone to misclassification bias. However, data were abstracted systematically using standardized forms by trained personnel who used consistent criteria during the data abstraction. Moreover, the study is subject to response bias because food insecurity screening was completed via a survey. Patients may not have wanted to answer the questions truthfully if they were concerned about the social or even clinical implications of their responses. Although the survey has been validated in low-income families with young children, demonstrating an overall sensitivity of 97% and specificity of 83% [4], this may be applied differently to the orthopaedic trauma population of our study. Lastly, the current study reflects the demographics of our specific patient population, which includes a city with a surrounding rural and hill country catchment area that is also near the United States–Mexico border. Patients in this study come from a variety of backgrounds including city dwellers, farmers and ranchers, and migrant workers. Therefore, care must be taken when extrapolating the current study experience to other groups of patients.
Proportion of Orthopaedic Trauma Patients With Food Insecurity
Other recent studies investigating food insecurity in the orthopaedic population have estimated proportions between approximately 13% and 25% [1, 8]. The recorded proportion in these patients was much higher, at 37%. There could be several reasons for this finding. The proportion of patients with food insecurity may vary substantially based on geographic differences across the United States. The demographics of this Level I urban trauma center near the Texas-Mexico border includes a relatively high representation of patients without healthcare insurance, as well as a high proportion of patients whose immigration status is undocumented, which could have contributed to this finding. Moreover, patients with food insecurity in this study were more likely to be employed, with a high representation of laborers, whereas many of the patients who were food-secure were retired. Many of the laborers worked in construction and had minimal to no insurance coverage. This highlights the issue of patients with working class employment not having access to healthy food, insurance coverage, and healthcare. This association should be further explored with social and health policy research that may identify methods to address this gap in healthcare and access to healthy food that much of the working population experiences. Additionally, the group of patients who were considered to have food insecurity included any patient who screened positive for food insecurity at any point during the study timeframe. This included patients who may not have initially screened positive during a clinic visit but screened positive at a subsequent visit. This relationship and its cause might be further explored in follow-up studies, but a possible explanation is the temporary or permanent loss of wages and decreased mobility that many orthopaedic trauma patients experience because of their injury. With these contributing factors, patients may later become food insecure as they go through the healing and recovery process.
Demographic Factors Associated With Food Insecurity in Orthopaedic Trauma Patients
The study groups (patients with food security versus patients with food insecurity) did not differ in terms of age, gender, ethnicity, percentage of laborers, alcohol use, or smoking status. This study demonstrated that patients with food insecurity were more likely to have a higher BMI than patients who were food-secure. To the best of our knowledge, this association has not yet been described for patients with orthopaedic trauma. However, it aligns with a report from the general United States population, which has shown that patients with food insecurity are at 32% increased odds of having obesity compared with people who are food-secure [10]. This finding has been related to the decreased access to healthy food that patients with food insecurity experience.
Is Food Insecurity Associated With a Higher Risk of Complications?
No differences in reoperations, deep infection, and nonunion were noted between the groups. A recently published study found no association between food insecurity and in-hospital complications [18]. However, that study was based on hospitalized patients, whereas this study included patients from an outpatient clinic with a mean of 1 year of follow-up data, accounting for longer-term complications. Patients with food insecurity in this study were more likely to experience superficial surgical site infections (13% versus 0%). However, this again should be interpreted with caution because the association was smaller than expected. Additionally, given the small study sample, the CI for this comparison was wide. Of note, previous research [16] has reviewed the relationship between food insecurity and malnutrition, and a recently published systematic review demonstrated an association between the two [11]. In this context, malnutrition has frequently been linked to an increased risk of complications in the orthopaedic trauma population, especially an increased risk of infection [14]. An association between food insecurity and the risk of infection was suggested in the current study, although the association was weaker than expected. This indicates an opportunity; healthcare professionals may wish to take steps to ensure that patients in this context have adequate access to food. Screening for food insecurity in the clinic setting can be done feasibly with this two-question survey, which can then open the conversation between the healthcare professional and patient regarding ways to increase access to healthy food. In our orthopaedic trauma clinic, patients who screened positive for food insecurity were provided education by clinic staff on local resources for food access and food vouchers.
Anecdotally, we observed that patients who screened positive for food insecurity and received food vouchers and referrals to food programs would occasionally report that they had not yet used these resources. Although a detailed analysis of this issue is beyond the scope of the current study, this observation was important to note. The exact reasons as to why patients did not use the resources provided remain unclear. Based on observations, several barriers might contribute to this, including cultural barriers, language barriers, health literacy, and limited access to technology. Further exploration of factors that play a role in patients using these types of resources is needed and will remain the subject of future research.
Conclusion
Food insecurity is common in patients sustaining orthopaedic trauma, especially in those with lower financial means. This study showed that food insecurity in orthopaedic trauma patients was not associated with differences in deep infections, reoperations, or nonunion compared with patients who were food-secure. There was a noted difference in superficial infections, although the difference was marginal. A suggestion for clinical practice would be to screen for food insecurity as a routine part of patient intake during clinic appointments, because it is feasible and can help healthcare professionals start connecting patients to resources and access to healthy food. Future research investigating the specific effects of food insecurity on overall nutritional status and postoperative outcomes would help us to continue to improve the way we care for patients with orthopaedic trauma.
Supplementary Material
Footnotes
Each author certifies that there are no funding or commercial associations (consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article related to the author or any immediate family members.
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request.
Ethical approval for this study was obtained from the University of Texas Health San Antonio Institutional Review Board (protocol number: 20230635EX).
Contributor Information
Loc-Uyen T. Vo, Email: vol@uthscsa.edu.
Luke Verlinsky, Email: verlinsky@uthscsa.edu.
Sohan Jakkaraju, Email: jakkaraju@livemail.uthscsa.edu.
Ana S. Guerra, Email: anaguerra0907@gmail.com.
References
- 1.Breslin MA, Bacharach A, Ho D, et al. Social determinants of health and patients with traumatic injuries: is there a relationship between social health and orthopaedic trauma? Clin Orthop Relat Res. 2023;481:901-908. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Coleman-Jensen A, Gregory C, Singh A. Household food security in the United States in 2013. Available at: https://papers.ssrn.com/abstract=2504067. Accessed September 26, 2023.
- 3.Greifer N. Cobalt: covariate balance tables and plots. Available at: https://CRAN.R-project.org/package=cobalt. Accessed September 1, 2023.
- 4.Hager ER, Quigg AM, Black MM, et al. Development and validity of a 2-item screen to identify families at risk for food insecurity. Pediatrics. 2010;126:e26-e32. [DOI] [PubMed] [Google Scholar]
- 5.Henstenburg JM, Lieber AM, Boniello AJ, et al. Higher complication rates after management of lower extremity fractures in lower socioeconomic classes: are risk adjustment models necessary? Trauma. 2022;24:131-137. [Google Scholar]
- 6.Ho DE, Imai K, King G, et al. MatchIt: nonparametric preprocessing for parametric causal inference. J Stat Sofw. 2011;42:1-28. [Google Scholar]
- 7.Jella TK, Cwalina TB, Schmidt JE, Wu VS, Yong TM, Vallier HA. Do patients reporting fractures experience food insecurity more frequently than the general population? Clin Orthop Relat Res. 2023;481:849-858. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Leary SM, Tully Z, Davison J, et al. Food insecurity is common in the orthopedic trauma population at a rural academic trauma center. Iowa Orthop J. 2023;43:137-144. [PMC free article] [PubMed] [Google Scholar]
- 9.Lumley T. Analysis of complex survey samples. J Stat Sofw. 2004;9:1-19. [Google Scholar]
- 10.Pan L, Sherry B, Njai R, Blanck HM. Food insecurity is associated with obesity among US adults in 12 states. J Acad Nutr Diet. 2012;112:1403-1409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Pereira MHQ, Pereira MLAS, Campos GC, Molina MCB. Food insecurity and nutritional status among older adults: a systematic review. Nutr Rev. 2022;80:631-644. [DOI] [PubMed] [Google Scholar]
- 12.R Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing. Available at: https://www.R-project.org/. Accessed September 1, 2023
- 13.Resad Ferati S, Parisien RL, Joslin P, Knapp B, Li X, Curry EJ. Socioeconomic status impacts access to orthopaedic specialty care. JBJS Reviews. 2022;10:e21.00139. [DOI] [PubMed] [Google Scholar]
- 14.Rong A, Franco-Garcia E, Zhou C, et al. Association of nutrition status and hospital-acquired infections in older adult orthopedic trauma patients. JPEN J Parenter Enteral Nutr. 2022;46:69-74. [DOI] [PubMed] [Google Scholar]
- 15.Schoenfeld AJ, Tipirneni R, Nelson JH, Carpenter JE, Iwashyna TJ. The influence of race and ethnicity on complications and mortality after orthopedic surgery: a systematic review of the literature. Med Care. 2014;52:842-851. [DOI] [PubMed] [Google Scholar]
- 16.Simsek H, Meseri R, Sahin S, et al. Prevalence of food insecurity and malnutrition, factors related to malnutrition in the elderly: a community-based, cross-sectional study from Turkey. Eur Geriatr Med. 2013;4:226-230. [Google Scholar]
- 17.US Department of Agriculture Economic Research Service. Definitions of food security. Available at: https://www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-u-s/definitions-of-food-security/. Accessed September 6, 2023.
- 18.Wetterhall M, Seilern und Aspang J, Hernandez-Irizarry R, et al. Food insecurity in orthopaedic trauma patients. J Am Coll Surg. 2022;235:S59. [Google Scholar]
- 19.Winkleby MA, Jatulis DE, Frank E, Fortmann SP. Socioeconomic status and health: how education, income, and occupation contribute to risk factors for cardiovascular disease. Am J Public Health. 1992;82:816-820. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Zelle BA, Morton-Gonzaba NA, Adcock CF, Lacci JV, Dang KH, Seifi A. Healthcare disparities among orthopedic trauma patients in the USA: socio-demographic factors influence the management of calcaneus fractures. J Orthop Surg Res. 2019;14:359. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Zhao Q-Y, Luo J-C, Su Y, Zhang YJ, Tu GW, Luo Z. Propensity score matching with R: conventional methods and new features. Ann Transl Med. 2021;9:812. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.