Use of multidimensional item response theory methods for dementia prevalence prediction : an example using the Health and Retirement Survey and the Aging, Demographics, and Memory Study

Emma Nichols; Foad Abd-Allah; Amir Abdoli; Ahmed Abualhasan; Eman Abu-Gharbieh; Ashkan Afshin; Rufus Olusola Akinyemi; Fahad Mashhour Alanezi; Vahid Alipour; Amir Almasi-Hashiani; Jalal Arabloo; Amir Ashraf-Ganjouei; Getinet Ayano; Jose L. Ayuso-Mateos; Atif Amin Baig; Maciej Banach; Miguel A. Barboza; Suzanne Lyn Barker-Collo; Bernhard T. Baune; Akshaya Srikanth Bhagavathula; Krittika Bhattacharyya; Ali Bijani; Atanu Biswas; Archith Boloor; Carol Brayne; Hermann Brenner; Katrin Burkart; Sharath Burugina Nagaraja; Felix Carvalho; Luis F. S. Castro-de-Araujo; Ferran Catala-Lopez; Ester Cerin; Nicolas Cherbuin; Dinh-Toi Chu; Xiaochen Dai; Antonio Reis de Sa-Junior; Shirin Djalalinia; Abdel Douiri; David Edvardsson; Shaimaa I. El-Jaafary; Sharareh Eskandarieh; Andre Faro; Farshad Farzadfar; Valery L. Feigin; Seyed-Mohammad Fereshtehnejad; Eduarda Fernandes; Pietro Ferrara; Irina Filip; Florian Fischer; Shilpa Gaidhane; Lucia Galluzzo; Gebreamlak Gebremedhn Gebremeskel; Ahmad Ghashghaee; Alessandro Gialluisi; Elena V. Gnedovskaya; Mahaveer Golechha; Rajeev Gupta; Vladimir Hachinski; Mohammad Rifat Haider; Teklehaimanot Gereziher Haile; Mohammad Hamiduzzaman; Graeme J. Hankey; Simon I. Hay; Golnaz Heidari; Reza Heidari-Soureshjani; Hung Chak Ho; Mowafa Househ; Bing-Fang Hwang; Licia Iacoviello; Olayinka Stephen Ilesanmi; Irena M. Ilic; Milena D. Ilic; Seyed Sina Naghibi Irvani; Masao Iwagami; Ihoghosa Osamuyi Iyamu; Ravi Prakash Jha; Rizwan Kalani; Andre Karch; Ayele Semachew Kasa; Yousef Saleh Khader; Ejaz Ahmad Khan; Mahalaqua Nazli Khatib; Yun Jin Kim; Sezer Kisa; Adnan Kisa; Mika Kivimaki; Ai Koyanagi; Manasi Kumar; Ivan Landires; Savita Lasrado; Bingyu Li; Stephen S. Lim; Xuefeng Liu; Shilpashree Madhava Kunjathur; Azeem Majeed; Preeti Malik; Man Mohan Mehndiratta; Ritesh G. Menezes; Yousef Mohammad; Salahuddin Mohammed; Ali H. Mokdad; Mohammad Ali Moni; Gabriele Nagel; Muhammad Naveed; Vinod C. Nayak; Cuong Tat Nguyen; Huong Lan Thi Nguyen; Virginia Nunez-Samudio; Andrew T. Olagunju; Samuel M. Ostroff; Nikita Otstavnov; Mayowa O. Owolabi; Fatemeh Pashazadeh Kan; Urvish K. Patel; Michael R. Phillips; Michael A. Piradov; Constance Dimity Pond; Faheem Hyder Pottoo; Sergio I. Prada; Amir Radfar; Fakher Rahim; Juwel Rana; Vahid Rashedi; Salman Rawaf; David Laith Rawaf; Nickolas Reinig; Andre M. N. Renzaho; null et al.

doi:10.1186/s12911-021-01590-y

Use of multidimensional item response theory methods for dementia prevalence prediction : an example using the Health and Retirement Survey and the Aging, Demographics, and Memory Study

Emma Nichols, Foad Abd-Allah, Amir Abdoli, Ahmed Abualhasan, Eman Abu-Gharbieh, Ashkan Afshin, Rufus Olusola Akinyemi, Fahad Mashhour Alanezi, Vahid Alipour, Amir Almasi-Hashiani, Jalal Arabloo, Amir Ashraf-Ganjouei, Getinet Ayano, Jose L. Ayuso-Mateos, Atif Amin Baig, Maciej Banach, Miguel A. Barboza, Suzanne Lyn Barker-Collo, Bernhard T. Baune, Akshaya Srikanth BhagavathulaKrittika Bhattacharyya, Ali Bijani, Atanu Biswas, Archith Boloor, Carol Brayne, Hermann Brenner, Katrin Burkart, Sharath Burugina Nagaraja, Felix Carvalho, Luis F. S. Castro-de-Araujo, Ferran Catala-Lopez, Ester Cerin, Nicolas Cherbuin, Dinh-Toi Chu, Xiaochen Dai, Antonio Reis de Sa-Junior, Shirin Djalalinia, Abdel Douiri, David Edvardsson, Shaimaa I. El-Jaafary, Sharareh Eskandarieh, Andre Faro, Farshad Farzadfar, Valery L. Feigin, Seyed-Mohammad Fereshtehnejad, Eduarda Fernandes, Pietro Ferrara, Irina Filip, Florian Fischer, Shilpa Gaidhane, Lucia Galluzzo, Gebreamlak Gebremedhn Gebremeskel, Ahmad Ghashghaee, Alessandro Gialluisi, Elena V. Gnedovskaya, Mahaveer Golechha, Rajeev Gupta, Vladimir Hachinski, Mohammad Rifat Haider, Teklehaimanot Gereziher Haile, Mohammad Hamiduzzaman, Graeme J. Hankey, Simon I. Hay, Golnaz Heidari, Reza Heidari-Soureshjani, Hung Chak Ho, Mowafa Househ, Bing-Fang Hwang, Licia Iacoviello, Olayinka Stephen Ilesanmi, Irena M. Ilic, Milena D. Ilic, Seyed Sina Naghibi Irvani, Masao Iwagami, Ihoghosa Osamuyi Iyamu, Ravi Prakash Jha, Rizwan Kalani, Andre Karch, Ayele Semachew Kasa, Yousef Saleh Khader, Ejaz Ahmad Khan, Mahalaqua Nazli Khatib, Yun Jin Kim, Sezer Kisa, Adnan Kisa, Mika Kivimaki, Ai Koyanagi, Manasi Kumar, Ivan Landires, Savita Lasrado, Bingyu Li, Stephen S. Lim, Xuefeng Liu, Shilpashree Madhava Kunjathur, Azeem Majeed, Preeti Malik, Man Mohan Mehndiratta, Ritesh G. Menezes, Yousef Mohammad, Salahuddin Mohammed, Ali H. Mokdad, Mohammad Ali Moni, Gabriele Nagel, Muhammad Naveed, Vinod C. Nayak, Cuong Tat Nguyen, Huong Lan Thi Nguyen, Virginia Nunez-Samudio, Andrew T. Olagunju, Samuel M. Ostroff, Nikita Otstavnov, Mayowa O. Owolabi, Fatemeh Pashazadeh Kan, Urvish K. Patel, Michael R. Phillips, Michael A. Piradov, Constance Dimity Pond, Faheem Hyder Pottoo, Sergio I. Prada, Amir Radfar, Fakher Rahim, Juwel Rana, Vahid Rashedi, Salman Rawaf, David Laith Rawaf, Nickolas Reinig, Andre M. N. Renzaho, et al.

Western Sydney University

Research output: Contribution to journal › Article › peer-review

4 Citations (Scopus)

Abstract

Background Data sparsity is a major limitation to estimating national and global dementia burden. Surveys with full diagnostic evaluations of dementia prevalence are prohibitively resource-intensive in many settings. However, validation samples from nationally representative surveys allow for the development of algorithms for the prediction of dementia prevalence nationally. Methods Using cognitive testing data and data on functional limitations from Wave A (2001-2003) of the ADAMS study (n = 744) and the 2000 wave of the HRS study (n = 6358) we estimated a two-dimensional item response theory model to calculate cognition and function scores for all individuals over 70. Based on diagnostic information from the formal clinical adjudication in ADAMS, we fit a logistic regression model for the classification of dementia status using cognition and function scores and applied this algorithm to the full HRS sample to calculate dementia prevalence by age and sex. Results Our algorithm had a cross-validated predictive accuracy of 88% (86-90), and an area under the curve of 0.97 (0.97-0.98) in ADAMS. Prevalence was higher in females than males and increased over age, with a prevalence of 4% (3-4) in individuals 70-79, 11% (9-12) in individuals 80-89 years old, and 28% (22-35) in those 90 and older. Conclusions Our model had similar or better accuracy as compared to previously reviewed algorithms for the prediction of dementia prevalence in HRS, while utilizing more flexible methods. These methods could be more easily generalized and utilized to estimate dementia prevalence in other national surveys.

Original language	English
Article number	241
Number of pages	10
Journal	BMC Medical Informatics and Decision Making
Volume	21
Issue number	1
DOIs	https://doi.org/10.1186/s12911-021-01590-y
Publication status	Published - 2021

Open Access - Access Right Statement

Â© The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Access to Document

10.1186/s12911-021-01590-y

Cite this

Nichols, E., Abd-Allah, F., Abdoli, A., Abualhasan, A., Abu-Gharbieh, E., Afshin, A., Akinyemi, R. O., Alanezi, F. M., Alipour, V., Almasi-Hashiani, A., Arabloo, J., Ashraf-Ganjouei, A., Ayano, G., Ayuso-Mateos, J. L., Baig, A. A., Banach, M., Barboza, M. A., Barker-Collo, S. L., Baune, B. T., ... et al. (2021). Use of multidimensional item response theory methods for dementia prevalence prediction : an example using the Health and Retirement Survey and the Aging, Demographics, and Memory Study. BMC Medical Informatics and Decision Making, 21(1), Article 241. https://doi.org/10.1186/s12911-021-01590-y

@article{847acb8cc77c4db79602872fdab74988,

title = "Use of multidimensional item response theory methods for dementia prevalence prediction : an example using the Health and Retirement Survey and the Aging, Demographics, and Memory Study",

abstract = "Background Data sparsity is a major limitation to estimating national and global dementia burden. Surveys with full diagnostic evaluations of dementia prevalence are prohibitively resource-intensive in many settings. However, validation samples from nationally representative surveys allow for the development of algorithms for the prediction of dementia prevalence nationally. Methods Using cognitive testing data and data on functional limitations from Wave A (2001-2003) of the ADAMS study (n = 744) and the 2000 wave of the HRS study (n = 6358) we estimated a two-dimensional item response theory model to calculate cognition and function scores for all individuals over 70. Based on diagnostic information from the formal clinical adjudication in ADAMS, we fit a logistic regression model for the classification of dementia status using cognition and function scores and applied this algorithm to the full HRS sample to calculate dementia prevalence by age and sex. Results Our algorithm had a cross-validated predictive accuracy of 88% (86-90), and an area under the curve of 0.97 (0.97-0.98) in ADAMS. Prevalence was higher in females than males and increased over age, with a prevalence of 4% (3-4) in individuals 70-79, 11% (9-12) in individuals 80-89 years old, and 28% (22-35) in those 90 and older. Conclusions Our model had similar or better accuracy as compared to previously reviewed algorithms for the prediction of dementia prevalence in HRS, while utilizing more flexible methods. These methods could be more easily generalized and utilized to estimate dementia prevalence in other national surveys.",

author = "Emma Nichols and Foad Abd-Allah and Amir Abdoli and Ahmed Abualhasan and Eman Abu-Gharbieh and Ashkan Afshin and Akinyemi, {Rufus Olusola} and Alanezi, {Fahad Mashhour} and Vahid Alipour and Amir Almasi-Hashiani and Jalal Arabloo and Amir Ashraf-Ganjouei and Getinet Ayano and Ayuso-Mateos, {Jose L.} and Baig, {Atif Amin} and Maciej Banach and Barboza, {Miguel A.} and Barker-Collo, {Suzanne Lyn} and Baune, {Bernhard T.} and Bhagavathula, {Akshaya Srikanth} and Krittika Bhattacharyya and Ali Bijani and Atanu Biswas and Archith Boloor and Carol Brayne and Hermann Brenner and Katrin Burkart and Nagaraja, {Sharath Burugina} and Felix Carvalho and Castro-de-Araujo, {Luis F. S.} and Ferran Catala-Lopez and Ester Cerin and Nicolas Cherbuin and Dinh-Toi Chu and Xiaochen Dai and Sa-Junior, {Antonio Reis de} and Shirin Djalalinia and Abdel Douiri and David Edvardsson and El-Jaafary, {Shaimaa I.} and Sharareh Eskandarieh and Andre Faro and Farshad Farzadfar and Feigin, {Valery L.} and Seyed-Mohammad Fereshtehnejad and Eduarda Fernandes and Pietro Ferrara and Irina Filip and Florian Fischer and Shilpa Gaidhane and Lucia Galluzzo and Gebremeskel, {Gebreamlak Gebremedhn} and Ahmad Ghashghaee and Alessandro Gialluisi and Gnedovskaya, {Elena V.} and Mahaveer Golechha and Rajeev Gupta and Vladimir Hachinski and Haider, {Mohammad Rifat} and Haile, {Teklehaimanot Gereziher} and Mohammad Hamiduzzaman and Hankey, {Graeme J.} and Hay, {Simon I.} and Golnaz Heidari and Reza Heidari-Soureshjani and Ho, {Hung Chak} and Mowafa Househ and Bing-Fang Hwang and Licia Iacoviello and Ilesanmi, {Olayinka Stephen} and Ilic, {Irena M.} and Ilic, {Milena D.} and Irvani, {Seyed Sina Naghibi} and Masao Iwagami and Iyamu, {Ihoghosa Osamuyi} and Jha, {Ravi Prakash} and Rizwan Kalani and Andre Karch and Kasa, {Ayele Semachew} and Khader, {Yousef Saleh} and Khan, {Ejaz Ahmad} and Khatib, {Mahalaqua Nazli} and Kim, {Yun Jin} and Sezer Kisa and Adnan Kisa and Mika Kivimaki and Ai Koyanagi and Manasi Kumar and Ivan Landires and Savita Lasrado and Bingyu Li and Lim, {Stephen S.} and Xuefeng Liu and Kunjathur, {Shilpashree Madhava} and Azeem Majeed and Preeti Malik and Mehndiratta, {Man Mohan} and Menezes, {Ritesh G.} and Yousef Mohammad and Salahuddin Mohammed and Mokdad, {Ali H.} and Moni, {Mohammad Ali} and Gabriele Nagel and Muhammad Naveed and Nayak, {Vinod C.} and Nguyen, {Cuong Tat} and Nguyen, {Huong Lan Thi} and Virginia Nunez-Samudio and Olagunju, {Andrew T.} and Ostroff, {Samuel M.} and Nikita Otstavnov and Owolabi, {Mayowa O.} and Kan, {Fatemeh Pashazadeh} and Patel, {Urvish K.} and Phillips, {Michael R.} and Piradov, {Michael A.} and Pond, {Constance Dimity} and Pottoo, {Faheem Hyder} and Prada, {Sergio I.} and Amir Radfar and Fakher Rahim and Juwel Rana and Vahid Rashedi and Salman Rawaf and Rawaf, {David Laith} and Nickolas Reinig and Renzaho, {Andre M. N.} and {et al.}",

year = "2021",

doi = "10.1186/s12911-021-01590-y",

language = "English",

volume = "21",

journal = "BMC Medical Informatics and Decision Making",

issn = "1472-6947",

publisher = "BioMed Central Ltd.",

number = "1",

}

Nichols, E, Abd-Allah, F, Abdoli, A, Abualhasan, A, Abu-Gharbieh, E, Afshin, A, Akinyemi, RO, Alanezi, FM, Alipour, V, Almasi-Hashiani, A, Arabloo, J, Ashraf-Ganjouei, A, Ayano, G, Ayuso-Mateos, JL, Baig, AA, Banach, M, Barboza, MA, Barker-Collo, SL, Baune, BT, Bhagavathula, AS, Bhattacharyya, K, Bijani, A, Biswas, A, Boloor, A, Brayne, C, Brenner, H, Burkart, K, Nagaraja, SB, Carvalho, F, Castro-de-Araujo, LFS, Catala-Lopez, F, Cerin, E, Cherbuin, N, Chu, D-T, Dai, X, Sa-Junior, ARD, Djalalinia, S, Douiri, A, Edvardsson, D, El-Jaafary, SI, Eskandarieh, S, Faro, A, Farzadfar, F, Feigin, VL, Fereshtehnejad, S-M, Fernandes, E, Ferrara, P, Filip, I, Fischer, F, Gaidhane, S, Galluzzo, L, Gebremeskel, GG, Ghashghaee, A, Gialluisi, A, Gnedovskaya, EV, Golechha, M, Gupta, R, Hachinski, V, Haider, MR, Haile, TG, Hamiduzzaman, M, Hankey, GJ, Hay, SI, Heidari, G, Heidari-Soureshjani, R, Ho, HC, Househ, M, Hwang, B-F, Iacoviello, L, Ilesanmi, OS, Ilic, IM, Ilic, MD, Irvani, SSN, Iwagami, M, Iyamu, IO, Jha, RP, Kalani, R, Karch, A, Kasa, AS, Khader, YS, Khan, EA, Khatib, MN, Kim, YJ, Kisa, S, Kisa, A, Kivimaki, M, Koyanagi, A, Kumar, M, Landires, I, Lasrado, S, Li, B, Lim, SS, Liu, X, Kunjathur, SM, Majeed, A, Malik, P, Mehndiratta, MM, Menezes, RG, Mohammad, Y, Mohammed, S, Mokdad, AH, Moni, MA, Nagel, G, Naveed, M, Nayak, VC, Nguyen, CT, Nguyen, HLT, Nunez-Samudio, V, Olagunju, AT, Ostroff, SM, Otstavnov, N, Owolabi, MO, Kan, FP, Patel, UK, Phillips, MR, Piradov, MA, Pond, CD, Pottoo, FH, Prada, SI, Radfar, A, Rahim, F, Rana, J, Rashedi, V, Rawaf, S, Rawaf, DL, Reinig, N, Renzaho, AMN & et al. 2021, 'Use of multidimensional item response theory methods for dementia prevalence prediction : an example using the Health and Retirement Survey and the Aging, Demographics, and Memory Study', BMC Medical Informatics and Decision Making, vol. 21, no. 1, 241. https://doi.org/10.1186/s12911-021-01590-y

Use of multidimensional item response theory methods for dementia prevalence prediction : an example using the Health and Retirement Survey and the Aging, Demographics, and Memory Study. / Nichols, Emma; Abd-Allah, Foad; Abdoli, Amir et al.
In: BMC Medical Informatics and Decision Making, Vol. 21, No. 1, 241, 2021.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - Use of multidimensional item response theory methods for dementia prevalence prediction : an example using the Health and Retirement Survey and the Aging, Demographics, and Memory Study

AU - Nichols, Emma

AU - Abd-Allah, Foad

AU - Abdoli, Amir

AU - Abualhasan, Ahmed

AU - Abu-Gharbieh, Eman

AU - Afshin, Ashkan

AU - Akinyemi, Rufus Olusola

AU - Alanezi, Fahad Mashhour

AU - Alipour, Vahid

AU - Almasi-Hashiani, Amir

AU - Arabloo, Jalal

AU - Ashraf-Ganjouei, Amir

AU - Ayano, Getinet

AU - Ayuso-Mateos, Jose L.

AU - Baig, Atif Amin

AU - Banach, Maciej

AU - Barboza, Miguel A.

AU - Barker-Collo, Suzanne Lyn

AU - Baune, Bernhard T.

AU - Bhagavathula, Akshaya Srikanth

AU - Bhattacharyya, Krittika

AU - Bijani, Ali

AU - Biswas, Atanu

AU - Boloor, Archith

AU - Brayne, Carol

AU - Brenner, Hermann

AU - Burkart, Katrin

AU - Nagaraja, Sharath Burugina

AU - Carvalho, Felix

AU - Castro-de-Araujo, Luis F. S.

AU - Catala-Lopez, Ferran

AU - Cerin, Ester

AU - Cherbuin, Nicolas

AU - Chu, Dinh-Toi

AU - Dai, Xiaochen

AU - Sa-Junior, Antonio Reis de

AU - Djalalinia, Shirin

AU - Douiri, Abdel

AU - Edvardsson, David

AU - El-Jaafary, Shaimaa I.

AU - Eskandarieh, Sharareh

AU - Faro, Andre

AU - Farzadfar, Farshad

AU - Feigin, Valery L.

AU - Fereshtehnejad, Seyed-Mohammad

AU - Fernandes, Eduarda

AU - Ferrara, Pietro

AU - Filip, Irina

AU - Fischer, Florian

AU - Gaidhane, Shilpa

AU - Galluzzo, Lucia

AU - Gebremeskel, Gebreamlak Gebremedhn

AU - Ghashghaee, Ahmad

AU - Gialluisi, Alessandro

AU - Gnedovskaya, Elena V.

AU - Golechha, Mahaveer

AU - Gupta, Rajeev

AU - Hachinski, Vladimir

AU - Haider, Mohammad Rifat

AU - Haile, Teklehaimanot Gereziher

AU - Hamiduzzaman, Mohammad

AU - Hankey, Graeme J.

AU - Hay, Simon I.

AU - Heidari, Golnaz

AU - Heidari-Soureshjani, Reza

AU - Ho, Hung Chak

AU - Househ, Mowafa

AU - Hwang, Bing-Fang

AU - Iacoviello, Licia

AU - Ilesanmi, Olayinka Stephen

AU - Ilic, Irena M.

AU - Ilic, Milena D.

AU - Irvani, Seyed Sina Naghibi

AU - Iwagami, Masao

AU - Iyamu, Ihoghosa Osamuyi

AU - Jha, Ravi Prakash

AU - Kalani, Rizwan

AU - Karch, Andre

AU - Kasa, Ayele Semachew

AU - Khader, Yousef Saleh

AU - Khan, Ejaz Ahmad

AU - Khatib, Mahalaqua Nazli

AU - Kim, Yun Jin

AU - Kisa, Sezer

AU - Kisa, Adnan

AU - Kivimaki, Mika

AU - Koyanagi, Ai

AU - Kumar, Manasi

AU - Landires, Ivan

AU - Lasrado, Savita

AU - Li, Bingyu

AU - Lim, Stephen S.

AU - Liu, Xuefeng

AU - Kunjathur, Shilpashree Madhava

AU - Majeed, Azeem

AU - Malik, Preeti

AU - Mehndiratta, Man Mohan

AU - Menezes, Ritesh G.

AU - Mohammad, Yousef

AU - Mohammed, Salahuddin

AU - Mokdad, Ali H.

AU - Moni, Mohammad Ali

AU - Nagel, Gabriele

AU - Naveed, Muhammad

AU - Nayak, Vinod C.

AU - Nguyen, Cuong Tat

AU - Nguyen, Huong Lan Thi

AU - Nunez-Samudio, Virginia

AU - Olagunju, Andrew T.

AU - Ostroff, Samuel M.

AU - Otstavnov, Nikita

AU - Owolabi, Mayowa O.

AU - Kan, Fatemeh Pashazadeh

AU - Patel, Urvish K.

AU - Phillips, Michael R.

AU - Piradov, Michael A.

AU - Pond, Constance Dimity

AU - Pottoo, Faheem Hyder

AU - Prada, Sergio I.

AU - Radfar, Amir

AU - Rahim, Fakher

AU - Rana, Juwel

AU - Rashedi, Vahid

AU - Rawaf, Salman

AU - Rawaf, David Laith

AU - Reinig, Nickolas

AU - Renzaho, Andre M. N.

AU - et al., null

PY - 2021

Y1 - 2021

N2 - Background Data sparsity is a major limitation to estimating national and global dementia burden. Surveys with full diagnostic evaluations of dementia prevalence are prohibitively resource-intensive in many settings. However, validation samples from nationally representative surveys allow for the development of algorithms for the prediction of dementia prevalence nationally. Methods Using cognitive testing data and data on functional limitations from Wave A (2001-2003) of the ADAMS study (n = 744) and the 2000 wave of the HRS study (n = 6358) we estimated a two-dimensional item response theory model to calculate cognition and function scores for all individuals over 70. Based on diagnostic information from the formal clinical adjudication in ADAMS, we fit a logistic regression model for the classification of dementia status using cognition and function scores and applied this algorithm to the full HRS sample to calculate dementia prevalence by age and sex. Results Our algorithm had a cross-validated predictive accuracy of 88% (86-90), and an area under the curve of 0.97 (0.97-0.98) in ADAMS. Prevalence was higher in females than males and increased over age, with a prevalence of 4% (3-4) in individuals 70-79, 11% (9-12) in individuals 80-89 years old, and 28% (22-35) in those 90 and older. Conclusions Our model had similar or better accuracy as compared to previously reviewed algorithms for the prediction of dementia prevalence in HRS, while utilizing more flexible methods. These methods could be more easily generalized and utilized to estimate dementia prevalence in other national surveys.

AB - Background Data sparsity is a major limitation to estimating national and global dementia burden. Surveys with full diagnostic evaluations of dementia prevalence are prohibitively resource-intensive in many settings. However, validation samples from nationally representative surveys allow for the development of algorithms for the prediction of dementia prevalence nationally. Methods Using cognitive testing data and data on functional limitations from Wave A (2001-2003) of the ADAMS study (n = 744) and the 2000 wave of the HRS study (n = 6358) we estimated a two-dimensional item response theory model to calculate cognition and function scores for all individuals over 70. Based on diagnostic information from the formal clinical adjudication in ADAMS, we fit a logistic regression model for the classification of dementia status using cognition and function scores and applied this algorithm to the full HRS sample to calculate dementia prevalence by age and sex. Results Our algorithm had a cross-validated predictive accuracy of 88% (86-90), and an area under the curve of 0.97 (0.97-0.98) in ADAMS. Prevalence was higher in females than males and increased over age, with a prevalence of 4% (3-4) in individuals 70-79, 11% (9-12) in individuals 80-89 years old, and 28% (22-35) in those 90 and older. Conclusions Our model had similar or better accuracy as compared to previously reviewed algorithms for the prediction of dementia prevalence in HRS, while utilizing more flexible methods. These methods could be more easily generalized and utilized to estimate dementia prevalence in other national surveys.

UR - http://hdl.handle.net/1959.7/uws:65509

U2 - 10.1186/s12911-021-01590-y

DO - 10.1186/s12911-021-01590-y

M3 - Article

SN - 1472-6947

VL - 21

JO - BMC Medical Informatics and Decision Making

JF - BMC Medical Informatics and Decision Making

IS - 1

M1 - 241

ER -

Use of multidimensional item response theory methods for dementia prevalence prediction : an example using the Health and Retirement Survey and the Aging, Demographics, and Memory Study

Abstract

Open Access - Access Right Statement

Access to Document

Other files and links

Fingerprint

Cite this