John Ko
John Ko has a total of 73 co-authors. Here are the top co-authors by number of shared publications:
- Stefan Varga
WorldWide Medical Immunology, Bristol Myers Squibb, Princeton, New Jersey, USA
4 shared publications Vivian Herrera
Novartis Pharmceuticals Corp. East Hanover NJ United States
4 shared publications- Bozena Katic
Communicable Disease Service, New Jersey Department of Health , Trenton, New Jersey
3 shared publications - Debbie Moreton
PatientsLikeMe Cambridge, MA United States
3 shared publications - Kristina Simacek
PatientsLikeMe, Cambridge, MA, USA
3 shared publications - Daniel Kantor
Univ Penn, Ctr Global Hlth, Perelman Sch Med, Philadelphia, PA USA
2 shared publications
Latest Publications
- Open Access
- 2 July 2020
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 15 (7) , e0235574
Detecting rare diseases in electronic health records using machine learning and knowledge engineering: Case study of acute hepatic porphyria
With the growing adoption of the electronic health record (EHR) worldwide over the last decade, new opportunities exist for leveraging EHR data for detection of rare diseases. Rare diseases are often not diagnosed or delayed in diagnosis by clinicians who encounter them infrequently. One such rare disease that may be amenable to EHR-based detection is acute hepatic porphyria (AHP). AHP consists of a family of rare, metabolic diseases characterized by potentially life-threatening acute attacks and chronic debilitating symptoms. The goal of this study was to apply machine learning and knowledge engineering to a large extract of EHR data to determine whether they could be effective in identifying patients not previously tested for AHP who should receive a proper diagnostic workup for AHP. We used an extract of the complete EHR data of 200,000 patients from an academic medical center and enriched it with records from an additional 5,571 patients containing any mention of porphyria in the record. After manually reviewing the records of all 47 unique patients with the ICD-10-CM code E80.21 (Acute intermittent [hepatic] porphyria), we identified 30 patients who were positive cases for our machine learning models, with the rest of the patients used as negative cases. We parsed the record into features, which were scored by frequency of appearance and filtered using univariate feature analysis. We manually choose features not directly tied to provider attributes or suspicion of the patient having AHP. We trained on the full dataset, with the best cross-validation performance coming from support vector machine (SVM) algorithm using a radial basis function (RBF) kernel. The trained model was applied back to the full data set and patients were ranked by margin distance. The top 100 ranked negative cases were manually reviewed for symptom complexes similar to AHP, finding four patients where AHP diagnostic testing was likely indicated and 18 patients where AHP diagnostic testing was possibly indicated. From the top 100 ranked cases of patients with mention of porphyria in their record, we identified four patients for whom AHP diagnostic testing was possibly indicated and had not been previously performed. Based solely on the reported prevalence of AHP, we would have expected only 0.002 cases out of the 200 patients manually reviewed. The application of machine learning and knowledge engineering to EHR data may facilitate the diagnosis of rare diseases such as AHP. Further work will recommend clinical investigation to identified patients’ clinicians, evaluate more patients, assess additional feature selection and machine learning algorithms, and apply this methodology to other rare diseases. This work provides strong evidence that population-level informatics can be applied to rare diseases, greatly improving our ability to identify undiagnosed patients, and in the future improve the care of these patients and our ability study these diseases. The next step is to learn how best to apply these EHR-based machine learning approaches to benefit individual patients with a clinical study that provides diagnostic testing and clinical follow up for those identified as possibly having undiagnosed AHP.
- 2 July 2020
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 15 (7) , e0235574
Open AccessDetecting rare diseases in electronic health records using machine learning and knowledge engineering: Case study of acute hepatic porphyria
With the growing adoption of the electronic health record (EHR) worldwide over the last decade, new opportunities exist for leveraging EHR data for detection of rare diseases. Rare diseases are often not diagnosed or delayed in diagnosis by clinicians who encounter them infrequently. One such rare disease that may be amenable to EHR-based detection is acute hepatic porphyria (AHP). AHP consists of a family of rare, metabolic diseases characterized by potentially life-threatening acute attacks and chronic debilitating symptoms. The goal of this study was to apply machine learning and knowledge engineering to a large extract of EHR data to determine whether they could be effective in identifying patients not previously tested for AHP who should receive a proper diagnostic workup for AHP. We used an extract of the complete EHR data of 200,000 patients from an academic medical center and enriched it with records from an additional 5,571 patients containing any mention of porphyria in the record. After manually reviewing the records of all 47 unique patients with the ICD-10-CM code E80.21 (Acute intermittent [hepatic] porphyria), we identified 30 patients who were positive cases for our machine learning models, with the rest of the patients used as negative cases. We parsed the record into features, which were scored by frequency of appearance and filtered using univariate feature analysis. We manually choose features not directly tied to provider attributes or suspicion of the patient having AHP. We trained on the full dataset, with the best cross-validation performance coming from support vector machine (SVM) algorithm using a radial basis function (RBF) kernel. The trained model was applied back to the full data set and patients were ranked by margin distance. The top 100 ranked negative cases were manually reviewed for symptom complexes similar to AHP, finding four patients where AHP diagnostic testing was likely indicated and 18 patients where AHP diagnostic testing was possibly indicated. From the top 100 ranked cases of patients with mention of porphyria in their record, we identified four patients for whom AHP diagnostic testing was possibly indicated and had not been previously performed. Based solely on the reported prevalence of AHP, we would have expected only 0.002 cases out of the 200 patients manually reviewed. The application of machine learning and knowledge engineering to EHR data may facilitate the diagnosis of rare diseases such as AHP. Further work will recommend clinical investigation to identified patients’ clinicians, evaluate more patients, assess additional feature selection and machine learning algorithms, and apply this methodology to other rare diseases. This work provides strong evidence that population-level informatics can be applied to rare diseases, greatly improving our ability to identify undiagnosed patients, and in the future improve the care of these patients and our ability study these diseases. The next step is to learn how best to apply these EHR-based machine learning approaches to benefit individual patients with a clinical study that provides diagnostic testing and clinical follow up for those identified as possibly having undiagnosed AHP.
- Top Cited PapersOpen Access
- 12 September 2019
- Published by Wolters Kluwer Health in Hepatology
- Vol. 71 (5) , 1546-1558
EXPLORE: A Prospective, Multinational, Natural History Study of Patients with Acute Hepatic Porphyria with Recurrent Attacks
Background and Aims: Acute hepatic porphyria comprises a group of rare genetic diseases caused by mutations in genes involved in heme biosynthesis. Patients can experience acute neurovisceral attacks, debilitating chronic symptoms, and long‐term complications. There is a lack of multinational, prospective data characterizing the disease and current treatment practices in severely affected patients. Approach and Results: EXPLORE is a prospective, multinational, natural history study characterizing disease activity and clinical management in patients with acute hepatic porphyria who experience recurrent attacks. Eligible patients had a confirmed acute hepatic porphyria diagnosis and had experienced ≥3 attacks in the prior 12 months or were receiving prophylactic treatment. A total of 112 patients were enrolled and followed for at least 6 months. In the 12 months before the study, patients reported a median (range) of 6 (0‐52) acute attacks, with 52 (46%) patients receiving hemin prophylaxis. Chronic symptoms were reported by 73 (65%) patients, with 52 (46%) patients experiencing these daily. During the study, 98 (88%) patients experienced a total of 483 attacks, 77% of which required treatment at a health care facility and/or hemin administration (median [range] annualized attack rate 2.0 [0.0‐37.0]). Elevated levels of hepatic δ‐aminolevulinic acid synthase 1 messenger ribonucleic acid levels, δ‐aminolevulinic acid, and porphobilinogen compared with the upper limit of normal in healthy individuals were observed at baseline and increased further during attacks. Patients had impaired quality of life and increased health care utilization. Conclusions: Patients experienced attacks often requiring treatment in a health care facility and/or with hemin, as well as chronic symptoms that adversely influenced day‐to‐day functioning. In this patient group, the high disease burden and diminished quality of life highlight the need for novel therapies.
- 12 September 2019
- Published by Wolters Kluwer Health in Hepatology
- Vol. 71 (5) , 1546-1558
Top Cited PapersOpen AccessEXPLORE: A Prospective, Multinational, Natural History Study of Patients with Acute Hepatic Porphyria with Recurrent Attacks
Background and Aims: Acute hepatic porphyria comprises a group of rare genetic diseases caused by mutations in genes involved in heme biosynthesis. Patients can experience acute neurovisceral attacks, debilitating chronic symptoms, and long‐term complications. There is a lack of multinational, prospective data characterizing the disease and current treatment practices in severely affected patients. Approach and Results: EXPLORE is a prospective, multinational, natural history study characterizing disease activity and clinical management in patients with acute hepatic porphyria who experience recurrent attacks. Eligible patients had a confirmed acute hepatic porphyria diagnosis and had experienced ≥3 attacks in the prior 12 months or were receiving prophylactic treatment. A total of 112 patients were enrolled and followed for at least 6 months. In the 12 months before the study, patients reported a median (range) of 6 (0‐52) acute attacks, with 52 (46%) patients receiving hemin prophylaxis. Chronic symptoms were reported by 73 (65%) patients, with 52 (46%) patients experiencing these daily. During the study, 98 (88%) patients experienced a total of 483 attacks, 77% of which required treatment at a health care facility and/or hemin administration (median [range] annualized attack rate 2.0 [0.0‐37.0]). Elevated levels of hepatic δ‐aminolevulinic acid synthase 1 messenger ribonucleic acid levels, δ‐aminolevulinic acid, and porphobilinogen compared with the upper limit of normal in healthy individuals were observed at baseline and increased further during attacks. Patients had impaired quality of life and increased health care utilization. Conclusions: Patients experienced attacks often requiring treatment in a health care facility and/or with hemin, as well as chronic symptoms that adversely influenced day‐to‐day functioning. In this patient group, the high disease burden and diminished quality of life highlight the need for novel therapies.
- Open Access
- 30 October 2018
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 20 (10) , e11168
The Impact of Disease-Modifying Therapy Access Barriers on People With Multiple Sclerosis: Mixed-Methods Study
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: In the United States, relapsing-remitting multiple sclerosis (RRMS) patients can face difficulty accessing disease-modifying therapies (DMTs) due to insurance, pharmacy, or provider policies. These barriers have been associated with poor adherence and negative health outcomes. Objective: The goals of the current study were to describe the overall occurrence of difficulties and delays associated with gaining access to DMTs among patients with RRMS, to assess DMT adherence during periods of reduced access, and to contextualize the patient journey from receipt of a prescription for DMT to obtaining and taking their medication when faced with access barriers. Methods: U.S.-based adult patients self-reporting RRMS were recruited from an online health data-sharing social network, PatientsLikeMe. Patients were invited to complete a web-based survey if they reported a diagnosis of RRMS and were prescribed a DMT for MS. Follow-up phone interviews were conducted with 10 respondents who reported experiencing a MS-related relapse during the time they had experienced challenges accessing DMTs. Results: Among 507 survey completers, nearly half were either currently experiencing an issue related to DMT assess or had difficulty accessing a DMT in the past (233/507, 46.0%). The most frequently reported reasons related to access difficulty were authorization requirements by insurance (78/182, 42.9% for past issues and 9/42, 21% for current issues) and high out-of-pocket costs (54/182, 29.7% for past issues and 13/42, 31% for current issues). About half (51%) of patients with current access issues and over a third (38%) of those with past issues went without their medication until they could access their prescribed DMT. Relapses were reported during periods of reduced DMT access for nearly a third (56/118, 29.8%) of those with past issues and nearly half (22/45, 49%) of those with current issues. Resolving access issues involved multiple stakeholder agents often coordinated in a patient-led effort. Among those who had resolved issues, about half (57/119, 47.9%) reported that doctors/office staff were involved, under half (48/119, 40.3%) were involved themselves, and about a third (39/119, 32.8%) reported the drug manufacturer was involved in resolving the issue. Follow-up interviews revealed that the financial burden associated with obtaining a prescribed DMT led to non-adherence. Additionally, patients felt that DMT treatment delays and stress associated with obtaining the DMT triggered relapses or worsened their MS. Conclusions: This study expands current research by using a patient-centered, mixed-methods approach to describe the patient perspective on barriers to MS treatment, process to resolve barriers, and the perceived impact of treatment barriers on outcomes. Issues related to DMT access occur frequently, with patients often serving as their own agents when navigating access difficulties to obtain their medication(s). Patient support for resolution of DMT access is needed to prevent undue stress and nonadherence.
- 30 October 2018
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 20 (10) , e11168
Open AccessThe Impact of Disease-Modifying Therapy Access Barriers on People With Multiple Sclerosis: Mixed-Methods Study
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: In the United States, relapsing-remitting multiple sclerosis (RRMS) patients can face difficulty accessing disease-modifying therapies (DMTs) due to insurance, pharmacy, or provider policies. These barriers have been associated with poor adherence and negative health outcomes. Objective: The goals of the current study were to describe the overall occurrence of difficulties and delays associated with gaining access to DMTs among patients with RRMS, to assess DMT adherence during periods of reduced access, and to contextualize the patient journey from receipt of a prescription for DMT to obtaining and taking their medication when faced with access barriers. Methods: U.S.-based adult patients self-reporting RRMS were recruited from an online health data-sharing social network, PatientsLikeMe. Patients were invited to complete a web-based survey if they reported a diagnosis of RRMS and were prescribed a DMT for MS. Follow-up phone interviews were conducted with 10 respondents who reported experiencing a MS-related relapse during the time they had experienced challenges accessing DMTs. Results: Among 507 survey completers, nearly half were either currently experiencing an issue related to DMT assess or had difficulty accessing a DMT in the past (233/507, 46.0%). The most frequently reported reasons related to access difficulty were authorization requirements by insurance (78/182, 42.9% for past issues and 9/42, 21% for current issues) and high out-of-pocket costs (54/182, 29.7% for past issues and 13/42, 31% for current issues). About half (51%) of patients with current access issues and over a third (38%) of those with past issues went without their medication until they could access their prescribed DMT. Relapses were reported during periods of reduced DMT access for nearly a third (56/118, 29.8%) of those with past issues and nearly half (22/45, 49%) of those with current issues. Resolving access issues involved multiple stakeholder agents often coordinated in a patient-led effort. Among those who had resolved issues, about half (57/119, 47.9%) reported that doctors/office staff were involved, under half (48/119, 40.3%) were involved themselves, and about a third (39/119, 32.8%) reported the drug manufacturer was involved in resolving the issue. Follow-up interviews revealed that the financial burden associated with obtaining a prescribed DMT led to non-adherence. Additionally, patients felt that DMT treatment delays and stress associated with obtaining the DMT triggered relapses or worsened their MS. Conclusions: This study expands current research by using a patient-centered, mixed-methods approach to describe the patient perspective on barriers to MS treatment, process to resolve barriers, and the perceived impact of treatment barriers on outcomes. Issues related to DMT access occur frequently, with patients often serving as their own agents when navigating access difficulties to obtain their medication(s). Patient support for resolution of DMT access is needed to prevent undue stress and nonadherence.
- Open Access
- 29 May 2018
- Published by JMIR Publications Inc. in JMIR Preprints
The Impact of Disease-Modifying Therapy Access Barriers on People With Multiple Sclerosis: Mixed-Methods Study (Preprint)
BACKGROUND In the United States, people with relapsing-remitting multiple sclerosis (RRMS) can face difficulty accessing disease-modifying therapies (DMTs) because of insurance, pharmacy, or provider policies. These barriers have been associated with poor adherence and negative health outcomes. OBJECTIVE The goals of this study were to describe the overall occurrence of difficulties and delays associated with gaining access to DMTs among people with RRMS, to assess DMT adherence during periods of reduced access, and to contextualize the patients’ journey from receipt of a prescription for DMT to obtaining and taking their medication when faced with access barriers. METHODS We recruited US-based adults self-reporting RRMS from a Web-based health data-sharing social network, PatientsLikeMe. Individuals were invited to complete a Web-based survey if they reported a diagnosis of RRMS and were prescribed a DMT for MS. Follow-up phone interviews were conducted with 10 respondents who reported experiencing an MS-related relapse during the time they had experienced challenges accessing DMTs. RESULTS Among 507 survey completers, nearly half were either currently experiencing an issue related to DMT assess or had difficulty accessing a DMT in the past (233/507, 46.0%). The most frequently reported reasons for access difficulty were authorization requirements by insurance companies (past issues: 78/182, 42.9%; current issues: 9/42, 21%) and high out-of-pocket costs (past issues: 54/182, 29.7%; current issues: 13/42, 31%). About half (20/39, 51%) of participants with current access issues and over a third (68/165, 41.2%) of those with past issues went without their medication until they could access their prescribed DMT. Relapses were reported during periods of reduced DMT access for almost half (56/118, 47.5%) of those with past issues and nearly half (22/45, 49%) of those with current issues. Resolving access issues involved multiple stakeholder agents often coordinated in a patient-led effort. Among those who had resolved issues, about half (57/119, 47.9%) reported that doctors or office staff were involved, under half (48/119, 40.3%) were involved themselves, and about a third (39/119, 32.8%) reported the drug manufacturer was involved in resolving the issue. Follow-up interviews revealed that the financial burden associated with obtaining a prescribed DMT led to nonadherence. Additionally, participants felt that DMT treatment delays and stress associated with obtaining the DMT triggered relapses or worsened their MS. CONCLUSIONS This study expands current research by using a patient-centered, mixed-methods approach to describe barriers to MS treatment, the process to resolve barriers, and the perceived impact of treatment barriers on outcomes. Issues related to DMT access occur frequently, with individuals often serving as their own agents when navigating access difficulties to obtain their medication(s). Support for resolution of DMT access is needed to prevent undue stress and nonadherence.
- 29 May 2018
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessThe Impact of Disease-Modifying Therapy Access Barriers on People With Multiple Sclerosis: Mixed-Methods Study (Preprint)
BACKGROUND In the United States, people with relapsing-remitting multiple sclerosis (RRMS) can face difficulty accessing disease-modifying therapies (DMTs) because of insurance, pharmacy, or provider policies. These barriers have been associated with poor adherence and negative health outcomes. OBJECTIVE The goals of this study were to describe the overall occurrence of difficulties and delays associated with gaining access to DMTs among people with RRMS, to assess DMT adherence during periods of reduced access, and to contextualize the patients’ journey from receipt of a prescription for DMT to obtaining and taking their medication when faced with access barriers. METHODS We recruited US-based adults self-reporting RRMS from a Web-based health data-sharing social network, PatientsLikeMe. Individuals were invited to complete a Web-based survey if they reported a diagnosis of RRMS and were prescribed a DMT for MS. Follow-up phone interviews were conducted with 10 respondents who reported experiencing an MS-related relapse during the time they had experienced challenges accessing DMTs. RESULTS Among 507 survey completers, nearly half were either currently experiencing an issue related to DMT assess or had difficulty accessing a DMT in the past (233/507, 46.0%). The most frequently reported reasons for access difficulty were authorization requirements by insurance companies (past issues: 78/182, 42.9%; current issues: 9/42, 21%) and high out-of-pocket costs (past issues: 54/182, 29.7%; current issues: 13/42, 31%). About half (20/39, 51%) of participants with current access issues and over a third (68/165, 41.2%) of those with past issues went without their medication until they could access their prescribed DMT. Relapses were reported during periods of reduced DMT access for almost half (56/118, 47.5%) of those with past issues and nearly half (22/45, 49%) of those with current issues. Resolving access issues involved multiple stakeholder agents often coordinated in a patient-led effort. Among those who had resolved issues, about half (57/119, 47.9%) reported that doctors or office staff were involved, under half (48/119, 40.3%) were involved themselves, and about a third (39/119, 32.8%) reported the drug manufacturer was involved in resolving the issue. Follow-up interviews revealed that the financial burden associated with obtaining a prescribed DMT led to nonadherence. Additionally, participants felt that DMT treatment delays and stress associated with obtaining the DMT triggered relapses or worsened their MS. CONCLUSIONS This study expands current research by using a patient-centered, mixed-methods approach to describe barriers to MS treatment, the process to resolve barriers, and the perceived impact of treatment barriers on outcomes. Issues related to DMT access occur frequently, with individuals often serving as their own agents when navigating access difficulties to obtain their medication(s). Support for resolution of DMT access is needed to prevent undue stress and nonadherence.
- 18 April 2017
- Published by Wolters Kluwer Health in Neurology
- Vol. 88 (16_supplem)
Do multiple sclerosis cost-effectiveness studies follow best practice guidelines? A review of 5 case studies (P3.399)
- 18 April 2017
- Published by Wolters Kluwer Health in Neurology
- Vol. 88 (16_supplem)
Do multiple sclerosis cost-effectiveness studies follow best practice guidelines? A review of 5 case studies (P3.399)
- 18 April 2017
- Published by Wolters Kluwer Health in Neurology
- Vol. 88 (16_supplem)
Multiple sclerosis patient experiences with access to disease-modifying therapies: A qualitative analysis on the patient impact of medication access barriers (P3.333)
- 18 April 2017
- Published by Wolters Kluwer Health in Neurology
- Vol. 88 (16_supplem)
Multiple sclerosis patient experiences with access to disease-modifying therapies: A qualitative analysis on the patient impact of medication access barriers (P3.333)
- 5 April 2016
- Published by Wolters Kluwer Health in Neurology
- Vol. 86 (16_supplem)
Evaluation of the Impact of Managed Care Restrictions on Clinical Outcomes, Adherence/Persistence, Financial Outcomes, and Healthcare Resource Utilization in Neurological Conditions: A Systematic Literature Review (P1.366)
- 5 April 2016
- Published by Wolters Kluwer Health in Neurology
- Vol. 86 (16_supplem)
Evaluation of the Impact of Managed Care Restrictions on Clinical Outcomes, Adherence/Persistence, Financial Outcomes, and Healthcare Resource Utilization in Neurological Conditions: A Systematic Literature Review (P1.366)
- 5 April 2016
- Published by Wolters Kluwer Health in Neurology
- Vol. 86 (16_supplem)
Treatment Preference in Relapsing-Remitting Multiple Sclerosis: An Online Choice Study in Patients and Neurologists (P3.102)
- 5 April 2016
- Published by Wolters Kluwer Health in Neurology
- Vol. 86 (16_supplem)
Treatment Preference in Relapsing-Remitting Multiple Sclerosis: An Online Choice Study in Patients and Neurologists (P3.102)