A comparison of prediction approaches for identifying prodromal Parkinson disease.
Identifying people with Parkinson disease during the prodromal period, including via algorithms in administrative claims data, is an important research and clinical priority. We sought to improve upon an existing penalized logistic regression model, based on diagnosis and procedure codes, by adding...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2021-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0256592 |
| _version_ | 1831715445673033728 |
|---|---|
| author | Mark N Warden Susan Searles Nielsen Alejandra Camacho-Soto Roman Garnett Brad A Racette |
| author_facet | Mark N Warden Susan Searles Nielsen Alejandra Camacho-Soto Roman Garnett Brad A Racette |
| author_sort | Mark N Warden |
| collection | DOAJ |
| description | Identifying people with Parkinson disease during the prodromal period, including via algorithms in administrative claims data, is an important research and clinical priority. We sought to improve upon an existing penalized logistic regression model, based on diagnosis and procedure codes, by adding prescription medication data or using machine learning. Using Medicare Part D beneficiaries age 66-90 from a population-based case-control study of incident Parkinson disease, we fit a penalized logistic regression both with and without Part D data. We also built a predictive algorithm using a random forest classifier for comparison. In a combined approach, we introduced the probability of Parkinson disease from the random forest, as a predictor in the penalized regression model. We calculated the receiver operator characteristic area under the curve (AUC) for each model. All models performed well, with AUCs ranging from 0.824 (simplest model) to 0.835 (combined approach). We conclude that medication data and random forests improve Parkinson disease prediction, but are not essential. |
| first_indexed | 2024-12-21T00:18:48Z |
| format | Article |
| id | doaj.art-4f17074530b341eaa642d598bcba33fe |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-12-21T00:18:48Z |
| publishDate | 2021-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-4f17074530b341eaa642d598bcba33fe2022-12-21T19:22:09ZengPublic Library of Science (PLoS)PLoS ONE1932-62032021-01-01168e025659210.1371/journal.pone.0256592A comparison of prediction approaches for identifying prodromal Parkinson disease.Mark N WardenSusan Searles NielsenAlejandra Camacho-SotoRoman GarnettBrad A RacetteIdentifying people with Parkinson disease during the prodromal period, including via algorithms in administrative claims data, is an important research and clinical priority. We sought to improve upon an existing penalized logistic regression model, based on diagnosis and procedure codes, by adding prescription medication data or using machine learning. Using Medicare Part D beneficiaries age 66-90 from a population-based case-control study of incident Parkinson disease, we fit a penalized logistic regression both with and without Part D data. We also built a predictive algorithm using a random forest classifier for comparison. In a combined approach, we introduced the probability of Parkinson disease from the random forest, as a predictor in the penalized regression model. We calculated the receiver operator characteristic area under the curve (AUC) for each model. All models performed well, with AUCs ranging from 0.824 (simplest model) to 0.835 (combined approach). We conclude that medication data and random forests improve Parkinson disease prediction, but are not essential.https://doi.org/10.1371/journal.pone.0256592 |
| spellingShingle | Mark N Warden Susan Searles Nielsen Alejandra Camacho-Soto Roman Garnett Brad A Racette A comparison of prediction approaches for identifying prodromal Parkinson disease. PLoS ONE |
| title | A comparison of prediction approaches for identifying prodromal Parkinson disease. |
| title_full | A comparison of prediction approaches for identifying prodromal Parkinson disease. |
| title_fullStr | A comparison of prediction approaches for identifying prodromal Parkinson disease. |
| title_full_unstemmed | A comparison of prediction approaches for identifying prodromal Parkinson disease. |
| title_short | A comparison of prediction approaches for identifying prodromal Parkinson disease. |
| title_sort | comparison of prediction approaches for identifying prodromal parkinson disease |
| url | https://doi.org/10.1371/journal.pone.0256592 |
| work_keys_str_mv | AT marknwarden acomparisonofpredictionapproachesforidentifyingprodromalparkinsondisease AT susansearlesnielsen acomparisonofpredictionapproachesforidentifyingprodromalparkinsondisease AT alejandracamachosoto acomparisonofpredictionapproachesforidentifyingprodromalparkinsondisease AT romangarnett acomparisonofpredictionapproachesforidentifyingprodromalparkinsondisease AT bradaracette acomparisonofpredictionapproachesforidentifyingprodromalparkinsondisease AT marknwarden comparisonofpredictionapproachesforidentifyingprodromalparkinsondisease AT susansearlesnielsen comparisonofpredictionapproachesforidentifyingprodromalparkinsondisease AT alejandracamachosoto comparisonofpredictionapproachesforidentifyingprodromalparkinsondisease AT romangarnett comparisonofpredictionapproachesforidentifyingprodromalparkinsondisease AT bradaracette comparisonofpredictionapproachesforidentifyingprodromalparkinsondisease |