HomeADScreen: Developing Alzheimer's disease and related dementia risk identification model in home healthcare.
Alzheimer's disease and related dementia (ADRD)
Clinical notes
Home healthcare (HHC)
Machine learning
Natural language processing
Journal
International journal of medical informatics
ISSN: 1872-8243
Titre abrégé: Int J Med Inform
Pays: Ireland
ID NLM: 9711057
Informations de publication
Date de publication:
09 2023
09 2023
Historique:
received:
27
03
2023
revised:
22
06
2023
accepted:
07
07
2023
pmc-release:
01
09
2024
medline:
14
8
2023
pubmed:
17
7
2023
entrez:
16
7
2023
Statut:
ppublish
Résumé
More than 50 % of patients with Alzheimer's disease and related dementia (ADRD) remain undiagnosed. This is specifically the case for home healthcare (HHC) patients. This study aimed at developing HomeADScreen, an ADRD risk screening model built on the combination of HHC patients' structured data and information extracted from HHC clinical notes. The study's sample included 15,973 HHC patients with no diagnosis of ADRD and 8,901 patients diagnosed with ADRD across four follow-up time windows. First, we applied two natural language processing methods, Word2Vec and topic modeling methods, to extract ADRD risk factors from clinical notes. Next, we built the risk identification model on the combination of the Outcome and Assessment Information Set (OASIS-structured data collected in the HHC setting) and clinical notes-risk factors across the four-time windows. The top-performing machine learning algorithm attained an Area under the Curve = 0.76 for a four-year risk prediction time window. After optimizing the cut-off value for screening patients with ADRD (cut-off-value = 0.31), we achieved sensitivity = 0.75 and an F1-score = 0.63. For the first-year time window, adding clinical note-derived risk factors to OASIS data improved the overall performance of the risk identification model by 60 %. We observed a similar trend of increasing the model's overall performance across other time windows. Variables associated with increased risk of ADRD were "hearing impairment" and "impaired patient ability in the use of telephone." On the other hand, being "non-Hispanic White" and the "absence of impairment with prior daily functioning" were associated with a lower risk of ADRD. HomeADScreen has a strong potential to be translated into clinical practice and assist HHC clinicians in assessing patients' cognitive function and referring them for further neurological assessment.
Sections du résumé
BACKGROUND
More than 50 % of patients with Alzheimer's disease and related dementia (ADRD) remain undiagnosed. This is specifically the case for home healthcare (HHC) patients.
OBJECTIVES
This study aimed at developing HomeADScreen, an ADRD risk screening model built on the combination of HHC patients' structured data and information extracted from HHC clinical notes.
METHODS
The study's sample included 15,973 HHC patients with no diagnosis of ADRD and 8,901 patients diagnosed with ADRD across four follow-up time windows. First, we applied two natural language processing methods, Word2Vec and topic modeling methods, to extract ADRD risk factors from clinical notes. Next, we built the risk identification model on the combination of the Outcome and Assessment Information Set (OASIS-structured data collected in the HHC setting) and clinical notes-risk factors across the four-time windows.
RESULTS
The top-performing machine learning algorithm attained an Area under the Curve = 0.76 for a four-year risk prediction time window. After optimizing the cut-off value for screening patients with ADRD (cut-off-value = 0.31), we achieved sensitivity = 0.75 and an F1-score = 0.63. For the first-year time window, adding clinical note-derived risk factors to OASIS data improved the overall performance of the risk identification model by 60 %. We observed a similar trend of increasing the model's overall performance across other time windows. Variables associated with increased risk of ADRD were "hearing impairment" and "impaired patient ability in the use of telephone." On the other hand, being "non-Hispanic White" and the "absence of impairment with prior daily functioning" were associated with a lower risk of ADRD.
CONCLUSION
HomeADScreen has a strong potential to be translated into clinical practice and assist HHC clinicians in assessing patients' cognitive function and referring them for further neurological assessment.
Identifiants
pubmed: 37454558
pii: S1386-5056(23)00164-8
doi: 10.1016/j.ijmedinf.2023.105146
pmc: PMC10529395
mid: NIHMS1918585
pii:
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Langues
eng
Sous-ensembles de citation
IM
Pagination
105146Subventions
Organisme : NIA NIH HHS
ID : R21 AG065753
Pays : United States
Informations de copyright
Copyright © 2023 Elsevier B.V. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Références
Pain Manag Nurs. 2021 Feb;22(1):36-43
pubmed: 32680825
Alzheimers Dement. 2013 Mar;9(2):208-45
pubmed: 23507120
Alzheimers Dement (N Y). 2018 Dec 07;4:661-668
pubmed: 30560199
Alzheimers Dement (N Y). 2019 Dec 10;5:918-925
pubmed: 31879701
Expert Rev Pharmacoecon Outcomes Res. 2017 Apr;17(2):189-202
pubmed: 28351177
Alzheimers Dement (N Y). 2022 Jul 26;8(1):e12341
pubmed: 35910670
Methods Mol Biol. 2010;609:223-39
pubmed: 20221922
Front Aging Neurosci. 2019 Jan 15;10:437
pubmed: 30697157
Lancet Neurol. 2006 Sep;5(9):735-41
pubmed: 16914401
Curr Pharm Des. 2016;22(7):812-8
pubmed: 26648470
Comput Inform Nurs. 2019 Nov;37(11):583-590
pubmed: 31478922
Physiother Can. 2017;69(2):161-170
pubmed: 28539696
Arq Neuropsiquiatr. 2010 Dec;68(6):855-61
pubmed: 21243241
Front Aging Neurosci. 2019 Dec 20;11:363
pubmed: 31920631
Gerontol Geriatr Med. 2020 Sep 24;6:2333721420959861
pubmed: 33029550
Alzheimers Dement. 2014 Sep;10(5):562-70
pubmed: 24035147
J Am Geriatr Soc. 2017 May;65(5):931-936
pubmed: 28295134
Int J Environ Res Public Health. 2015 Jul 17;12(7):8281-94
pubmed: 26193295
PLoS One. 2019 Jul 5;14(7):e0203246
pubmed: 31276468
Aging Ment Health. 2020 Jun;24(6):947-955
pubmed: 30789028
Laryngoscope Investig Otolaryngol. 2017 Mar 16;2(2):69-79
pubmed: 28894825
Intern Med. 2005 Jul;44(7):717-21
pubmed: 16093593
J Neurol Neurosurg Psychiatry. 2019 Apr;90(4):373-379
pubmed: 29954871
Am J Epidemiol. 2002 Jun 15;155(12):1081-7
pubmed: 12048221
BMC Neurol. 2014 May 07;14:101
pubmed: 24885474
JMIR Med Inform. 2020 Jun 3;8(6):e17819
pubmed: 32490841
J Gerontol A Biol Sci Med Sci. 2022 Jun 1;77(6):1261-1271
pubmed: 34919686
Am J Alzheimers Dis Other Demen. 2018 Dec;33(8):500-507
pubmed: 30068225
Neurology. 2009 Jul 21;73(3):173-9
pubmed: 19439724
Am J Alzheimers Dis Other Demen. 2022 Jan-Dec;37:15333175221129384
pubmed: 36242532
BMC Med Inform Decis Mak. 2022 May 12;22(1):129
pubmed: 35549702
Lancet. 2020 Aug 8;396(10248):413-446
pubmed: 32738937
Patient Prefer Adherence. 2021 Dec 14;15:2793-2807
pubmed: 34934309
BMJ Open. 2017 Feb 3;7(2):e011146
pubmed: 28159845
Int J Alzheimers Dis. 2019 May 21;2019:3637954
pubmed: 31263595
Neuron. 2020 Jun 17;106(6):881-883
pubmed: 32553204