Accuracy of Electronic Health Record Food Insecurity, Housing Instability, and Financial Strain Screening in Adult Primary Care

Harle, Christopher A.; Wu, Wei; Vest, Joshua R.

doi:10.1001/jama.2022.23631

Cited by 7 publications

(6 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our findings contribute to the growing literature linking housing instability and racial disparities in health . Our findings can also inform emerging research on screening for social determinants of health including housing insecurity and efforts to target health care outreach to individuals in unstable living arrangements …”

Section: Introductionmentioning

confidence: 51%

Examining Excess Mortality Associated With the COVID-19 Pandemic for Renters Threatened With Eviction

Graetz,

Hepburn,

Gershenson

et al. 2024

JAMA

View full text Add to dashboard Cite

ImportanceResidential evictions may have increased excess mortality associated with the COVID-19 pandemic.ObjectiveTo estimate excess mortality associated with the COVID-19 pandemic for renters who received eviction filings (threatened renters).Design, Setting, and ParticipantsThis retrospective cohort study used an excess mortality framework. Mortality based on linked eviction and death records from 2020 through 2021 was compared with projected mortality estimated from similar records from 2010 through 2016. Data from court records between January 1, 2020, and August 31, 2021, were collected via the Eviction Lab’s Eviction Tracking System. Similar data from court records between January 1, 2010, and December 31, 2016, also collected by the Eviction Lab, were used to estimate projected mortality during the pandemic. We also constructed 2 comparison groups: all individuals living in the study area and a subsample of those individuals living in high-poverty, high-filing tracts.ExposuresEviction filing.Main Outcomes and MeasuresAll-cause mortality in a given month. The difference between observed mortality and projected mortality was used as a measure of excess mortality associated with the pandemic.ResultsThe cohort of threatened renters during the pandemic period consisted of 282 000 individuals (median age, 36 years [IQR, 28-47]). Eviction filings were 44.7% lower than expected during the study period. The composition of threatened renters by race, ethnicity, sex, and socioeconomic characteristics during the pandemic was comparable with the prepandemic composition. Expected cumulative age-standardized mortality among threatened renters during this 20-month period of the pandemic was 116.5 (95% CI, 104.0-130.3) per 100 000 person-months, and observed mortality was 238.6 (95% CI, 230.8-246.3) per 100 000 person-months or 106% higher than expected. In contrast, expected mortality for the population living in similar neighborhoods was 114.6 (95% CI, 112.1-116.8) per 100 000 person-months, and observed mortality was 142.8 (95% CI, 140.2-145.3) per 100 000 person-months or 25% higher than expected. In the general population across the study area, expected mortality was 83.5 (95% CI, 83.3-83.8) per 100 000 person-months, and observed mortality was 91.6 (95% CI, 91.4-91.8) per 100 000 person-months or 9% higher than expected. The pandemic produced positive excess mortality ratios across all age groups among threatened renters.Conclusions and RelevanceRenters who received eviction filings experienced substantial excess mortality associated with the COVID-19 pandemic.

show abstract

Section: Introductionmentioning

confidence: 51%

Examining Excess Mortality Associated With the COVID-19 Pandemic for Renters Threatened With Eviction

Graetz,

Hepburn,

Gershenson

et al. 2024

JAMA

View full text Add to dashboard Cite

show abstract

“…The 18-item version was used for households with children and the 10-item version for households without children. The 6-item short form was used in six cases [ 30 , 31 , 32 , 48 , 49 , 50 ]. Two of these were with households with children [ 31 , 32 ].…”

Section: Resultsmentioning

confidence: 99%

Food Insecurity Screening in High-Income Countries, Tool Validity, and Implementation: A Scoping Review

Baker,

Gallegos,

Rebuli

et al. 2024

Nutrients

View full text Add to dashboard Cite

Household food insecurity has significant negative implications across the lifespan. While routine screening is recommended, particularly in healthcare, guidelines are lacking on selection of screening tools and best-practice implementation across different contexts in non-stigmatizing ways. The objective of this scoping review was to synthesize evidence on household food insecurity screening tools, including psychometrics, implementation in a range of settings, and experiences of carrying out screening or being screened. Four electronic databases were searched for studies in English published from 1990 until June 2023. A total of 58 papers were included, 21 of which focused on tool development and validation, and 37 papers described implementation and perceptions of screening. Most papers were from the USA and described screening in healthcare settings. There was a lack of evidence regarding screening in settings utilized by Indigenous people. The two-item Hunger Vital Sign emerged as the most used and most valid tool across settings. While there is minimal discomfort associated with screening, screening rates in practice are still low. Barriers and facilitators of screening were identified at the setting, system, provider, and recipient level and were mapped onto the COM-B model of behavior change. This review identifies practical strategies to optimize screening and disclosure.

show abstract

“…As reported by others, this observation illustrates the importance of combining the information provided by ICD-9 codes and other structured data (e.g., questionnaires) with unstructured data in the EHR to obtain a more representative assessment of the SBDH prevalence in a population. 7,[10][11][12]32 On the other hand, implementing SDoH screening tools across a large health system is impractical and potentially biased.…”

Section: Discussionmentioning

confidence: 99%

Large-scale identification of social and behavioral determinants of health from clinical notes: Comparison of Latent Semantic Indexing and Generative Pretrained Transformer (GPT) models

Roy,

Morrell,

Zhao

et al. 2024

Preprint

View full text Add to dashboard Cite

Background: Social and behavioral determinants of health (SBDH) are associated with a variety of health and utilization outcomes, yet these factors are not routinely documented in the structured fields of electronic health records (EHR). The objective of this study was to evaluate different machine learning approaches for detection of SBDH from the unstructured clinical notes in the EHR. Methods: Latent Semantic Indexing (LSI) was applied to 2,083,180 clinical notes corresponding to 46,146 patients in the MIMIC-III dataset. Using LSI, patients were ranked based on conceptual relevance to a set of keywords (lexicons) pertaining to 15 different SBDH categories. For Generative Pretrained Transformer (GPT) models, API requests were made with a Python script to connect to the OpenAI services in Azure, using gpt-3.5-turbo-1106 and gpt-4-1106-preview models. Prediction of SBDH categories were performed using logistic regression model that included age, gender race and SBDH ICD-9 codes with a natural cubic spline of 2 degrees of freedom for age. Results: LSI retrieved patients according to 15 SBDH domains, with an overall average PPV ≥ 83%. Using manually curated gold standard (GS) sets for nine SBDH categories, the macro-F1 score of LSI (0.74) was better than ICD-9 (0.71) and GPT-3.5 (0.54), but lower than GPT-4 (0.80). Due to document size limitations, only a subset of the GS cases could be processed by GPT-3.5 (55.8%) and GPT-4 (94.2%), compared to LSI (100%). Using common GS subsets for nine different SBDH categories, the macro-F1 of ICD-9 combined with either LSI (mean 0.88, 95% CI 0.82-0.93), GPT-3.5 (0.86, 0.82-0.91) or GPT-4 (0.88, 0.83-0.94) was not significantly different. After including age, gender, race and ICD-9 in a logistic regression model, the AUC for prediction of six out of the nine SBDH categories was higher for LSI compared to GPT-4.0. Conclusions: These results demonstrate that the LSI approach performs comparable to more recent large language models, such as GPT-3.5 and GPT-4.0, when using the same set of documents. Importantly, LSI is robust, deterministic, and does not have document-size limitations or cost implications, which make it more amenable to real-world applications in health systems.

show abstract

Accuracy of Electronic Health Record Food Insecurity, Housing Instability, and Financial Strain Screening in Adult Primary Care

Abstract: This study assesses the accuracy of electronic health record–based screening questionnaires about social risk factors using external single-domain questionnaires as a comparator.

Cited by 7 publications

References 3 publications

Examining Excess Mortality Associated With the COVID-19 Pandemic for Renters Threatened With Eviction

Examining Excess Mortality Associated With the COVID-19 Pandemic for Renters Threatened With Eviction

Food Insecurity Screening in High-Income Countries, Tool Validity, and Implementation: A Scoping Review

Large-scale identification of social and behavioral determinants of health from clinical notes: Comparison of Latent Semantic Indexing and Generative Pretrained Transformer (GPT) models

Contact Info

Product

Resources

About