Abstract
Objective
This study aimed to identify phenotypes of nutritional needs of home-visited clients with low income, and compare overall changes in knowledge, behavior, and status of nutritional needs before and after home visits by identified phenotypes.
Materials and methods
Omaha System data collected by public health nurses from 2013 to 2018 were used in this secondary data analysis study. A total of 900 low-income clients were included in the analysis. Latent class analysis (LCA) was used to identify phenotypes of nutrition symptoms or signs. Score changes in knowledge, behavior, and status were compared by phenotype.
Results
The five subgroups included Unbalanced Diet, Overweight, Underweight, Hyperglycemia with Adherence, and Hyperglycemia without Adherence. Only the Unbalanced Diet and Underweight groups showed an increase in knowledge. No other changes in behavior and status were observed in any of the phenotypes.
Discussion and conclusions
This LCA using standardized Omaha System Public Health Nursing data allowed us to identify phenotypes of nutritional needs among home-visited clients with low income and prioritize nutrition areas that public health nurses may focus on as part of public health nursing interventions. The sub-optimal changes in knowledge, behavior, and status suggest a need to re-examine the intervention details by phenotype and develop strategies to tailor public health nursing interventions to effectively meet the diverse nutritional needs of home-visited clients.
Keywords: Omaha System, nutrition, standardized terminology, public health nurse, home visit
BACKGROUND AND SIGNIFICANCE
Proper nutrition is vital for health across the lifespan. However, many adults in the United States consume foods high in saturated fat and added sugars.1 Unhealthy diet increases risk for nutrition-related chronic diseases, including cardiovascular disease, type 2 diabetes, and obesity.2,3 These nutrition-related health issues affect low-income populations disproportionally.4,5 Approximately 10% of households in the United States experience food insecurity, and with 32% of low-income households identified as food insecure.6 People with food insecurity are more likely to have hypertension and hyperlipidemia.7 Adults living in poverty consume a lower quality diet8 and have a higher likelihood of developing type 2 diabetes and obesity.9,10 In contrast, adults and children from high-income families are less likely to be obese compared to those from low-income families.11,12 Disparities in nutrition and nutrition-related chronic diseases suggest a need to improve ongoing public health efforts.
One public health intervention available for low-income families is home visiting services offered by public health nurses. During a home visit, public health nurses identify the client’s unique needs and provide relevant interventions, including health education, chronic disease management, and care coordination.13 Public health nurses’ home visiting services have shown positive effects on children’s growth, maternal mental health, and the development of healthy habits.14 Public health nurses also provide interventions to address nutrition-related health needs.15 However, there is limited examination of the patterns or phenotypes of nutritional needs of home-visited clients in low-income households, despite the heterogeneity of nutritional needs across clients. It is also unclear whether intervention outcomes varied by phenotypes of nutritional needs. Identifying phenotypes is needed to develop more targeted and tailored public health nursing interventions.
An informatics approach utilizing an advanced statistical method such as latent class analysis (LCA) of standardized terminologies in existing electronic documentation provides an opportunity to address knowledge gaps. The use of standardized nursing terminology has been applied in other nursing studies examining phenotypes. For example, Koleck’s team identified five phenotypes of symptoms qualitatively documented in electronic health records using natural language processing to facilitate symptom management.16 Similarly, LCA can be used to identify phenotypes of nutritional needs that are quantitatively documented in standardized nursing terminologies.17 The identification of phenotypes will allow us to characterize unique combinations of diverse nutritional needs of home-visited clients which otherwise might have appeared homogenous. The phenotypes can then be utilized to compare home-visited clients’ responses to nutritional interventions. This effort will guide improvements for tailoring public health nursing interventions to better meet the diverse nutritional needs of clients and contribute to enhancing health equity.
OBJECTIVE
The purpose of this study was to identify phenotypes of nutritional needs of home-visited clients with low income, and compare overall changes in knowledge, behavior, and status of nutritional needs before and after home visits by the identified phenotypes using the Omaha System data. The Omaha System, a research-based, comprehensive, and standardized taxonomy is an electronic documentation system that public health nurses use to describe clients’ needs and care, including nutrition needs and interventions.18 The Omaha System consists of three components, including the Problem Classification Scheme, Intervention Scheme, and Problem Rating Scale for Outcomes. The Problem Classification Scheme assesses 42 problems across environmental, psychosocial, physiological, and health-behavior domains.18 The Intervention Scheme describes care plans and services provided to clients. The Problem Rating Scale for Outcomes has three, five-point Likert scales to assess knowledge, behavior, and status or condition of the client. For this study, the Problem Classification Scheme was mainly used to identify phenotypes of nutritional needs and the Problem Rating Scale for Outcome was used to compare overall knowledge, behavior, and status of nutritional needs by the identified phenotypes.
MATERIALS AND METHODS
Dataset
The secondary data analyses study used a subset of data from the Omaha System collected by public health nurses from 2013 to 2018. Clients with signs or symptoms reported in the Income and Nutrition problems in the Problem Classification Scheme (n = 900) were included in the analyses. Income problem is defined as “Money, wages, pensions, subsidies, interest, dividends, or other sources available for living and health care expenses” and included the following six signs or symptoms: low/no income, uninsured medical expenses, difficulty with money management, able to buy only necessities, difficulty buying necessities and other.18 This study’s protocol was exempted from review by the University of Minnesota Institutional Review Board as it was secondary research for which consent is not required (IRB ID Study00008512).
Study variables
Nutrition problem
The signs or symptoms of the Nutrition problem were used for analysis to identify patterns of the nutritional needs of the home-visited clients with an income problem. In the Omaha System, the nutrition problem was defined as “Select, consume, and use food and fluids for energy, maintenance, growth and health” and has the following 11 signs or symptoms: (1) overweight: adult body mass index (BMI) 25.0 or more or child BMI 95th percentile or more, (2) underweight: adult BMI 18.5 or less and child BMI 95th percentile or less, (3) lacks established standards for daily caloric/fluid intake, (4) exceeds established standards for daily caloric/fluid intake, (5) unbalanced diet, (6) improper feeding schedule for age, (7) does not follow recommended nutrition plan, (8) unexplained/progressive weight loss, (9) unable to obtain/prepare food, (10) hypoglycemia, and (11) hyperglycemia. Public health nurses recorded any of the nutrition signs or symptoms (yes versus no) based on their objective assessments.
Client characteristics
Race/ethnicity, marital status, and gender were recorded in the Omaha System and used to compare the characteristics of clients by group. Response options for race/ethnicity included (1) non-Hispanic white, (2) Latinx, (3) American Indian, and (4) Asian/Non-pacific Islander. Response options for marital status included (1) married, (2) single, (3) divorced, and (4) widowed. Gender was classified into female and male. Those who did not select any of the response options to race/ethnicity, marital status, and gender were classified as non-specified. In addition to the demographic characteristics, number of problems, number of total signs or symptoms, number of signs or symptoms in the nutrition problem, and number of interventions offered by the public health nurses were compared by group to describe the clients’ health needs.
Change in knowledge, behavior, and status
At the times of admission and discharge, public health nurses rated each client’s knowledge, behavior and status on a five-point Likert scale. Knowledge was defined as “ability of the client to remember and interpret information”. Clients’ knowledge was rated 1 if no knowledge and 5 if superior knowledge. Behavior referred to “observable responses, actions or activities of the client fitting the occasion or purpose”. Clients with no appropriate behavior were rated 1 and those with consistently appropriate behavior were rated 5. Status was defined as a “condition of the client in relation to objective and subjective defining characteristics”. Status of rating of 1 was used when client showed extreme signs or symptoms and 5 when there were no signs or symptoms.18
Statistical analysis
We used LCA to identify patterns of nutritional needs of home-visited clients with an income issue. A series of LCA models with two to six groups were conducted in SAS version 9.4. Model fit indices, including Bayesian Information Criterion (BIC) and Akaike Information Criterion (AIC) were compared across the models to identify models with best fit. We also reviewed probabilities of the indicators in each model to identify the ideal model with theoretically meaningful groups.19 Participant characteristics were compared by the identified groups using chi-square tests for categorical characteristics or analysis of variance (ANOVA) for continuous characteristics. The changes in knowledge, behavior and status from admission to discharge were compared by groups using ANOVA.
RESULTS
Of the 900 clients included in the analysis, 83% were female, 58% were non-Hispanic white, 24% were Latinx, American Indian, African American, or Asian/Non-pacific Islander, 19% were married, and 65% were single, divorced, or widowed. Mean age of the clients was 34.1 (SD 20.0).
Table 1 shows model fit indices from LCA. It supported models of three phenotypes and five phenotypes (Aim 1). Specifically, the smallest adjusted BIC and second smallest AIC supported the five-phenotype model. In comparison, the three-phenotype model had the smallest corrected AIC and BIC. The five-phenotype model was selected after reviewing theoretical meanings of phenotypes in each model.
Table 1.
Model-fit indices for latent class analyses of nutritional signs and symptoms among low-income home-visiting clients in the Omaha System data (n = 900)
| Number of Phenotypes |
|||||
|---|---|---|---|---|---|
| Index | 2 | 3 | 4 | 5 | 6 |
| Log-likelihood | −3967.37 | −3873.70 | −3843.11 | −3809.73 | −3792.29 |
| AIC | 934.65 | 771.31 | 734.12 | 691.36 | 680.49 |
| cAIC | 1068.10 | 974.39 | 1006.83 | 1033.71 | 1092.46 |
| BIC | 1045.10 | 939.39 | 959.83 | 974.71 | 1021.46 |
| aBIC | 972.06 | 828.24 | 810.57 | 787.33 | 795.97 |
| Entropy | 0.61 | 0.85 | 0.70 | 0.76 | 0.79 |
Abbreviations. AIC: Akaike Information Criterion; cAIC: Corrected AIC; BIC: Baysian Information Criterion; aBIC: Adjusted BIC.
Table 2 demonstrates the estimated probabilities of each phenotype in the selected five-phenotype model. Based on the distinct probabilities in each group, the phenotypes were named as Unbalanced Diet (35%), Overweight (33%), Underweight (19%), Hyperglycemia with Adherence (8%), and Hyperglycemia without Adherence (6%). As each phenotype name describes, the Unbalanced Diet group showed a distinctively high probability of having an unbalanced diet (probability = 0.704). All of the clients classified into the Overweight group reported an overweight problem (probability = 1.000). The third phenotype named Underweight only reported an underweight issue, which was a distinguished character of this phenotype, although the absolute value of the probability was not high (probability = 0.370). The remaining two phenotypes reported a hyperglycemia sign/symptom, but differed by the presence of an issue of non-adherence to a recommended nutrition plan. Therefore, the two phenotypes were named as Hyperglycemia with Adherence (probability of non-adherence = 0.209) and Hyperglycemia without Adherence (probability of non-adherence = 0.898).
Table 2.
Estimated probabilities by phenotype (n = 900)
| Unbalanced Diet | Overweight | Underweight | Hyperglycemia with Adherence | Hyperglycemia without Adherence | |
|---|---|---|---|---|---|
| n = 318 | n = 293 | n = 168 | n = 68 | n = 53 | |
| (35%) | (33%) | (19%) | (8%) | (6%) | |
| Unbalanced diet | 0.704 | 0.315 | 0.305 | 0.079 | 0.572 |
| Overweight | 0.212 | 1.000 | 0.000 | 0.042 | 0.655 |
| Underweight | 0.019 | 0.000 | 0.370 | 0.000 | 0.000 |
| Hyperglycemia | 0.000 | 0.078 | 0.027 | 0.942 | 1.000 |
| Non-adherence to recommended nutrition plan | 0.419 | 0.128 | 0.205 | 0.209 | 0.898 |
| Lack of established standards | 0.197 | 0.025 | 0.376 | 0.000 | 0.162 |
| Exceeding established standards | 0.104 | 0.190 | 0.024 | 0.000 | 0.277 |
| Improper feeding schedule | 0.078 | 0.000 | 0.047 | 0.000 | 0.073 |
| Unexplained weight loss | 0.015 | 0.021 | 0.226 | 0.032 | 0.066 |
| Unable to obtain food | 0.186 | 0.049 | 0.234 | 0.099 | 0.166 |
| Hypoglycemia | 0.009 | 0.000 | 0.028 | 0.341 | 0.361 |
Note. Distinguishable probabilities are bolded and highlighted in grey.
The five phenotypes had significant differences in demographic characteristics, health status assessed by the number of problems, signs/symptoms, as well as public health nursing interventions, received (see Table 3). For example, a high portion of American Indian was in the Hyperglycemia without adherence (15%, P < 0.001). Clients in this phenotype (Hyperglycemia without Adherence) reported the highest number of signs and symptoms in the problem specific to nutrition (F = 80.14, P < 0.001) as well as total problems in all domains (F = 39.19, P < 0.001) and received the highest number of interventions (F = 10.88, P < 0.001).
Table 3.
Client characteristics by phenotype
| Total | Unbalanced Diet | Overweight | Underweight | Hyperglycemia with Adherence | Hyperglycemia without Adherence | Chi-square | P-value | |
|---|---|---|---|---|---|---|---|---|
| Sample Size | 900 | 318 | 293 | 168 | 68 | 53 | ||
| Percent (%) | ||||||||
| Race/Ethnicity | 59.1 | 0.001 | ||||||
| Non-Hispanic white | 58% | 64% | 61% | 57% | 37% | 43% | ||
| Latinx | 16% | 17% | 12% | 20% | 26% | 9% | ||
| American Indian | 5% | 3% | 6% | 5% | 9% | 15% | ||
| African American | 2% | 2% | 2% | 1% | 1% | 2% | ||
| Asian/Non-pacific Islander | 1% | 0% | 1% | 4% | 3% | 0% | ||
| Not Specified | 17% | 14% | 18% | 15% | 24% | 30% | ||
| Marital Status | 76.0 | 0.001 | ||||||
| Married | 19% | 18% | 23% | 13% | 16% | 34% | ||
| Single | 50% | 58% | 51% | 55% | 26% | 17% | ||
| Divorced | 11% | 9% | 10% | 11% | 22% | 9% | ||
| Widowed | 4% | 3% | 3% | 8% | 10% | 8% | ||
| Not Specified | 16% | 14% | 14% | 14% | 25% | 32% | ||
| Gender | 71.5 | 0.001 | ||||||
| Female | 83% | 90% | 88% | 75% | 65% | 57% | ||
| Male | 15% | 8% | 11% | 21% | 34% | 42% | ||
| Not Specified | 2% | 2% | 1% | 4% | 1% | 2% | ||
| Mean ± Standard deviation | ||||||||
| Mean Age | 34.1 ± 20.0 | 27.2 ± 15.0 | 33.4 ± 16.6a | 35.4 ± 25.4a | 49.3 ± 21.5a,b,c | 55.0 ± 16.6a,b,c | 39.82 | 0.001 |
| No. of Problems | 10.1 ± 7.3 | 8.7 ± 5.4 | 9.1 ± 6.2 | 10.9 ± 8.3a | 14.7 ± 10.4a,b,c | 14.8 ± 7.8a,b,c | 19.29 | <0.0001 |
| No. of Total SS | 14.1 ± 13.0 | 12.3 ± 11.7 | 11.4 ± 10.4a | 15.0 ± 12.8b | 17.1 ± 13.0a,b | 32.8 ± 16.8a,b,c,d | 39.19 | <0.0001 |
| No. of Nutrition SS | 2.0 ± 1.2 | 2.0 ± 1.1 | 1.8 ± 0.8 | 1.9 ± 1.2 | 1.7 ± 0.7 | 4.4 ± 1.1a,b,c,d | 80.14 | <0.0001 |
| No. of Interventions | 312.7 ± 408.2 | 275.9 ± 348.2 | 273.4 ± 389.1 | 316.6 ± 398.1 | 390.6 ± 437.4 | 639.1 ± 634.2a,b,c,d | 10.88 | <0.0001 |
Abbreviations. No.: Number; SS: Signs or Symptoms.
P < 0.05 in Tukey’s Post-Hoc Test where the reference phenotype was Unbalanced Diet.
P < 0.05 in Tukey’s Post-Hoc Test where the reference phenotype was Overweight.
P < 0.05 in Tukey’s Post-Hoc Test where the reference phenotype was Underweight.
P < 0.05 in Tukey’s Post-Hoc Test where the reference phenotype was Hyperglycemia with Adherence.
Figure 1 displays changes in knowledge, behavior, and status scores assessed at admission and discharge from home visiting services by phenotype (Aim 2). Changes in knowledge scores were statistically different by phenotype (F = 6.60, P < 0.001) but not in behaviors (F = 1.82, P = 0.1221) and signs/symptoms scores (F = 1.93, P = 0.1041). Particularly, changes in knowledge among clients in the three phenotypes, including Overweight, Hyperglycemia with Adherence, and Hyperglycemia without Adherence were significantly smaller than those in the Unbalanced Diet and Underweight phenotypes.
Figure 1.
Public health nursing interventions and outcomes (knowledge, behavior, and status scores) by phenotype (n = 900).
Notes: Group 1 = Unbalanced Diet, Group 2 = Overweight, Group 3 = Underweight, Group 4 = Hyperglycemia with Adherence, Group 5 = Hyperglycemia without Adherence, Adm = assessment at admission, DC = Assessment at discharge. Only changes in the knowledge score were statistically significant at P < 0.05.
DISCUSSION
Conducting an advanced statistical method of LCA using a standardized nursing terminology dataset allowed us to discover five distinct phenotypes of nutritional needs among low-income clients receiving home visiting services from public health nurses, and compare the clients’ changes by the identified phenotypes. The three phenotypes (unbalanced diet, overweight, and underweight) mirror the existing nutritional signs/symptoms but the two remainder phenotypes highlighted the conditions of hyperglycemia with or without adherence in the low-income population targeted in this study. Hyperglycemia can cause life-threatening diabetic emergencies such as diabetic ketoacidosis and hyperosmolar hyperglycemic state.20 A recent study found that adults with low income or poor glycemic control have the greatest risks for diabetic emergencies, calling for interventions focusing on high risk groups.21 Consistently, our study findings emphasize a heightened need for hyperglycemia interventions among home visited clients with low-income, and an opportunity to advance health equity in this area through tailored public health nursing interventions.
The identified phenotypes are interesting in that it reinforces the three levels of prevention, primary, secondary, and tertiary22 that public health nurses can focus on when tailoring nutrition interventions for clients with low income. The areas included (1) nutrition education on balancing diet (primary prevention), (2) healthy weight management of those who are already overweight or underweight (secondary prevention), and (3) nutrition-related chronic disease management and adherence, particularly those with hyperglycemia (tertiary prevention). These areas are also well aligned with nutrition and healthy eating objectives in Healthy People 203023 and previous literature that demonstrated health disparities by income level.4,5,24 A replication of the analysis using other datasets is suggested to confirm the unique nutritional needs among low-income populations and determine whether these nutritional needs should be embedded into electronic health records systems for identification on admission, which may guide care planning and further our goals for more individualized patient-centered care.
In this study, we did not find significant changes in behavior and status/conditions related to nutritional needs among home-visited clients in each phenotype. Additionally, only the groups with seemingly lower risks (unbalanced diet and underweight groups), not the other groups with more complex risks, showed changes in knowledge. This is not surprising considering that home-visited clients with moderate levels of risks demonstrated the best outcomes in previous research.25 However, recent literature suggests home visiting services can be effective in providing nutrition interventions to prevent obesity, particularly for low-income populations.26–28 A comprehensive examination of the nutritional intervention details by the identified phenotypes is warranted to further inform and improve effectiveness of public health nurses’ nutrition interventions, allowing nurses to prioritize interventions on the areas that are most important for low-income clients. Additionally, we encourage future studies to account for the intervention content in addition to the number of intervention activities when evaluating changes in knowledge, behavior, and status.
This study has several limitations that are common to secondary data analysis. We were not able to fully describe the income level of the sample as no objective measure of income was available in the dataset. Any provider characteristics or details of the public health nursing interventions, such as the inter-rater reliability or the length of intervention that might have impacted the changes in knowledge, behavior, and status were not accounted for in the analysis. Additionally, we acknowledge that the age of the dataset is not recent, yet the same data schemes continue to be employed in the Omaha System, suggesting that this analytic approach can be used with more recent data. Lastly, the study only focused on nutrition signs/symptoms and relevant outcomes. While nutrition plays a critical role in the development and management of numerous chronic diseases and is an important intervention target for public health nurses’ home visiting services, it is likely that their physical (eg comorbidity), psychosocial (eg mental illness and family support) and environmental (eg access to healthful foods) conditions and resources were contributing to their nutritional needs. Future studies are warranted replicating this study with an extended data range, and further expanding the study by including other factors representing physical, social, and environmental factors that may affect home visiting clients’ nutritional needs.
The strengths of this study are important to note. This study is one of the first to analyze a standardized nursing terminology dataset using an advanced statistical method to guide advancement of public health nurses’ nutrition interventions for low-income populations. It identified 5 unique phenotypes, out of 11 nutritional signs and symptoms in the Omaha System Public Health Nursing data, that public health nurses may focus on these phenotypes to prioritize and tailor nutrition interventions and improve nutrition-related health equity for low-income populations. The study provided preliminary results on the effectiveness of public health nurses’ current nutrition interventions, and served as an example of how nursing informatics may be used to evaluate current interventions. Lastly, the study findings highlight the contributions of standardized nursing terminology to identify and prioritize various needs of home-visiting clients to develop tailored interventions to more effectively target clients’ needs.
Conclusion
Addressing nutritional needs, particularly for those with low income, is critical for public health nurses to promote well-being of home-visited clients and address nutrition disparity. Findings from this study using standardized nursing terminology in the Omaha System indicate home visited clients’ nutritional needs can be prioritized for nutrition education, healthy weight management and nutrition-related chronic disease management and adherence. While public health nurses are already providing some level of interventions to address these needs, the minimal change in knowledge, behaviors, and status shown in this study suggest a need for improving nutrition interventions through tailoring approaches to subsets of the population that may not be identified without using standardized nursing terminology and an informatics approach.
ACKNOWLEDGEMENTS
We would like to acknowledge the University of Minnesota, School of Nursing Center for Nursing Informatics and the Omaha System Partnership.
Contributor Information
Jiwoo Lee, School of Nursing, University of Minnesota, Minneapolis, Minnesota, USA.
Robin R Austin, School of Nursing, University of Minnesota, Minneapolis, Minnesota, USA.
Michelle A Mathiason, School of Nursing, University of Minnesota, Minneapolis, Minnesota, USA.
Karen A Monsen, School of Nursing, University of Minnesota, Minneapolis, Minnesota, USA.
FUNDING
This work was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health under award number 1K23HD107179-01. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
AUTHOR CONTRIBUTIONS
Dr. Lee conceptualized this secondary data analysis study, drafted, and revised the manuscript under the guidance of Drs. Austin and Monsen. Ms. Mathiason obtained de-identified datasets from Drs. Austin and Monsen, and conducted data analyses. All of the co-authors reviewed the data analysis results and the manuscript and provided critical feedback on intellectual content. All authors have approved the final manuscript.
CONFLICT OF INTEREST STATEMENT
None declared.
DATA AVAILABILITY
The data underlying this article will be shared on reasonable request to the corresponding author.
REFERENCES
- 1. Huth PJ, Fulgoni VL, Keast DR, Park K, Auestad N.. Major food sources of calories, added sugars, and saturated fat and their contribution to essential nutrient intakes in the U.S. diet: Data from the national health and nutrition examination survey (2003–2006). Nutr J 2013; 12: 116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Malik VS, Popkin BM, Bray GA, Despres JP, Hu FB.. Sugar sweetened beverages, obesity, type 2 diabetes and cardiovascular disease risk. Circulation 2010; 121 (11): 1356–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Reedy J, Krebs-Smith SM, Miller PE, et al. Higher diet quality is associated with decreased risk of all-cause, cardiovascular disease, and cancer mortality among older adults. J Nutr 2014; 144 (6): 881–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Hiza HAB, Casavale KO, Guenther PM, Davis CA.. Diet quality of Americans differs by age, sex, race/ethnicity, income, and education level. J Acad Nutr Diet 2013; 113 (2): 297–306. [DOI] [PubMed] [Google Scholar]
- 5. Thomson JL, Tussing-Humphreys LM, Goodman MH, Landry AS.. Diet quality in a nationally representative sample of American children by sociodemographic characteristics. Am J Clin Nutr 2019; 109 (1): 127–38. [DOI] [PubMed] [Google Scholar]
- 6. Coleman-Jensen A, Rabbitt MP, Gregory C, Singh A.. Household Food Security in the United States in 2021. U.S. Department of Agriculture, Economic Research Service; 2022. doi: 10.2139/ssrn.2504067. [DOI]
- 7. Rehm CD, Peñalvo JL, Afshin A, Mozaffarian D.. Dietary intake among US adults, 1999–2012. J Am Med Assoc 2016; 315 (23): 2542–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Seligman HK, Laraia BA, Kushel MB.. Food insecurity is associated with chronic disease among low-income NHANES participants. J Nutr 2010; 140 (2): 304–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Gaskin DJ, Thorpe RJ, McGinty EE, et al. Disparities in diabetes: the nexus of race, poverty, and place. Am J Public Health 2014; 104 (11): 2147–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Hsu CC, Lee CH, Wahlqvist ML, et al. Poverty increases type 2 diabetes incidence and inequality of care despite universal health coverage. Diabetes Care 2012; 35 (11): 2286–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Ogden CL, Carroll MD, Fakhouri TH, et al. Prevalence of obesity among youths by household income and education level of head of household — United States 2011–2014. MMWR Morb Mortal Wkly Rep 2018; 67 (6): 186–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ogden CL, Fakhouri TH, Carroll MD, et al. Prevalence of obesity among adults, by household income and education — United States, 2011–2014. MMWR Morb Mortal Wkly Rep 2017; 66 (50): 1369–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. American Public Health Association Public Health Nursing Section. The Definition and Practice of Public Health Nursing: A Statement of the Public Health Nursing Section. 2013. https://www.apha.org/~/media/files/pdf/membergroups/phn/nursingdefinition.ashx. Accessed February 21, 2023.
- 14. Ciliska D, Mastrilli P, Ploeg J, Hayward S, Brunton G, Underwood J.. The effectiveness of home visiting as a delivery strategy for public health nursing interventions to clients in the prenatal and postnatal period: a systematic review. Prim Heal Care Res Dev 2001; 2 (1): 41–54. [Google Scholar]
- 15. Horning ML, Olsen JM, Lell S, Thorson DR, Monsen KA.. Description of public health nursing nutrition assessment and interventions for home-visited women. Public Health Nurs 2018; 35 (4): 317–26. [DOI] [PubMed] [Google Scholar]
- 16. Koleck TA, Tatonetti NP, Bakken S, et al. Identifying symptom information in clinical notes using natural language processing. Nurs Res 2021; 70 (3): 173–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Mori M, Krumholz HM, Allore HG.. Using latent class analysis to identify hidden clinical phenotypes. J Am Med Assoc 2020; 324 (7): 700–1. [DOI] [PubMed] [Google Scholar]
- 18. Martin KS. The Omaha System: A Key to Practice, Documentation, and Information Management. 2nd ed. Omaha, NE: Health Connections Press; 2005. [Google Scholar]
- 19. Lanza ST, Cooper BR.. Latent class analysis for developmental research. Child Dev Perspect 2016; 10 (1): 59–64. doi: 10.1111/cdep.12163 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Maletkovic J, Drexler A.. Diabetic ketoacidosis and hyperglycemic hyperosmolar state. Endocrinol Metab Clin North Am 2013; 42 (4): 677–95. doi: 10.1016/j.ecl.2013.07.001 [DOI] [PubMed] [Google Scholar]
- 21. McCoy RG, Galindo RJ, Swarna KS, et al. Sociodemographic, clinical, and treatment-related factors associated with hyperglycemic crises among adults with type 1 or type 2 diabetes in the US from 2014 to 2020. JAMA Netw Open 2021; 4 (9): e2123471. doi: 10.1001/jamanetworkopen.2021.23471 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Simeonsson RJ. Primary, secondary, and tertiary prevention in early intervention. J Early Interv 1991; 15 (2): 124–34. doi: 10.1177/105381519101500202 [DOI] [Google Scholar]
- 23. US Department of Health and Human Services and U.S. Department of Agriculture. Nutrition and Healthy Eating. Healthy People 2030. https://health.gov/healthypeople/objectives-and-data/browse-objectives/nutrition-and-healthy-eating. Accessed February 21, 2023.
- 24. Lorson BA, Melgar-Quinonez HR, Taylor CA.. Correlates of fruit and vegetable intakes in US children. J Am Diet Assoc 2009; 109 (3): 474–8. doi: 10.1016/j.jada.2008.11.022 [DOI] [PubMed] [Google Scholar]
- 25. Gomby DS. Understanding evaluations of home visitation programs. Futur Child 1999; 9 (1): 27–43. [PubMed] [Google Scholar]
- 26. Ordway MR, Sadler LS, Holland ML, Slade A, Close N, Mayes LC.. A home visiting parenting program and child obesity: a randomized trial. Pediatrics 2018; 141 (2): e20171076. doi: 10.1542/peds.2017-1076 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Rosenstock S, Ingalls A, Foy Cuddy R, et al. Effect of a home-visiting intervention to reduce early childhood obesity among Native American children: a randomized clinical trial. JAMA Pediatr 2021; 175 (2): 133–42. doi: 10.1001/jamapediatrics.2020.3557 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Salvy SJ, de la Haye K, Galama T, Goran MI.. Home visitation programs: an untapped opportunity for the delivery of early childhood obesity prevention. Obes Rev 2017; 18 (2): 149–63. doi: 10.1111/obr.12482 [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.
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.