EHR data and AI to predict response to antidepressant treatment

Antidepressants are frequently prescribed for adults with depression, a common and often disabling psychiatric condition. However, identifying the most effective treatment for a particular patient is often a trial-and-error process that can result in prolonged morbidity, disability, and exposure to adverse effects, as well as substantial healthcare costs. Precision psychiatry aims to optimise treatment matching using patient-specific profiles, but there are few evidence-based predictors available to clinicians initiating antidepressant treatment.

Although average response rates are similar across different antidepressant classes, individual responses can vary widely in clinical practice. Therefore, accurately and scalably guiding antidepressant selection presents specific challenges. The gold standard for characterising antidepressant response from electronic health records (EHRs) remains expert chart review, which is labor- and time-intensive.

However, advances in machine learning (ML) and the growing availability of large-scale health data, such as EHRs, offer new opportunities for developing clinical decision-support tools that may address this challenge. In a recent study published in the peer-reviewed open-access medical journal Nature Partner Journals (npc) Digital Medicine, researchers used machine learning models to accurately predict differential treatment response probabilities for patients and between antidepressant classes based on real-world EHR data. The pipeline incorporated AI and non-AI features, as well as unstructured data (i.e. clinical notes) to maximize the use of information contained in EHRs.

The study included 17,556 patients who received a new antidepressant prescription from non-psychiatrists, and data were obtained from 20 years of EHRs spanning from January 1990 to August 2018. The patients had at least one International Classification of Diseases (ICD) code for depression and at least one ICD code for non-recurrent depression during their history.

ICD codes from EHR data were obtained for adult patients (age ≥ 18 years) with at least one visit (the first visit with an antidepressant prescription is defined as the “index visit” for each patient) with a diagnostic ICD code for a depressive disorder (defined as ICD-9-CM: 296.20–6, 296.30–6, and 311; ICD-10-CM: F32.0–9, F33.0–9) co-occurring with an antidepressant prescription, and at least one ICD code for non-recurrent depression (ICD-9-CM: 296.20–6 and 311; ICD-10-CM: F32.0–9) any time during their history.

The resulting models achieved good accuracy, discrimination, and positive predictive value, which could be valuable for further efforts aiming to provide clinical decision support for prescribers. However, the researchers noted several limitations, including missing data in EHRs(e.g. patients who may receive some of their care outside of the healthcare system), and secular trends in clinician prescribing or documentation practices that may have affected model performance.

In summary, the study presents a novel computational pipeline based on real-world EHR data for predicting differential responses to commonly used classes of antidepressants. The approach demonstrated here could be adapted to a wide variety of other clinical applications for optimising and individualising treatment selection.

REFERENCES:

  1. Sheu, Yh., Magdamo, C., Miller, M. et al. AI-assisted prediction of differential response to antidepressant classes using electronic health records. npj Digit. Med. 6, 73 (2023). https://doi.org/10.1038/s41746-023-00817-8


Future of EHR/EMR trends in 2023: what to expect

The global EHR market is projected to grow to $38.5 billion by 2030, with patient engagement, integration, big data implementation, and standardisation is the most significant areas for improvement.

However, physician practices still face concerns with documentation issues like risk-based management codes and the use of modifiers.

It’s crucial to keep up with EHR trends as they are now a common feature in the healthcare industry and have largely replaced paper-based record-keeping.

The COVID-19 pandemic has accelerated the adoption of telehealth, which has become essential for non-emergency visits during lockdowns and restrictions.

Tech trends will continue to shape the future of healthcare facilities.

According to a report by Grand View Research, the global EHR market was worth $28.1 billion in 2022 and is expected to grow at a CAGR of 4%. This growth will likely bring about new and recurring trends in the EHR industry.

As the market continues to expand, we can expect the emergence of new and recurring trends.

What specific trends can we anticipate? The following infographic illustrates new and recurring trends, as envisaged by Austin, Texas, USA-based start-up SelectHub which is a combination of a software selection platform and research firm.

References:
Kashish, A. (2023, February 28). Future of EHR/EMR: Experts Predict Trends in 2023. SelectHub. Selecthub: EHR & EMR Trends 2023

Goodbye to manuscript in medical notes

The readability and speed of information retrieval have improved dramatically with electronic patient encounters.

For quick data retrieval and trend analysis, almost every industry is now automated and digitalised. Take a look at the stock market or organisations like Federal Express or Walmart.

Why not the medical field?

EMR in the Twelfth Plan

As the healthcare industry transitions to digitisation, the Electronic Medical Records / Electronic Health Records (EMR / EHR) systems that store patients’ health information in a digital format are becoming increasingly popular.

Many governments encourage doctors to use EMR / EHR systems and invest in the training of healthcare information technology professionals. An April 2021 report by Grand View Research, Inc. reports that with Government initiatives to encourage healthcare Internet Technology (IT) usage, the global EHR market was valued at USD 26.8 billion in 2020 and is expected to witness a compound annual growth rate (CAGR) of 3.7% from 2021. According to Bloomberg, the CAGR rate is expected to peak at 3.30% and be worth $ 33.69 billion globally by 2028.

For instance, the Government of Malaysia joins this increasing zeal to adopt digital platforms for the continued use of digital technologies in the healthcare field. Therefore, it aims to implement a uniform system for maintaining EMR / EHR by public hospitals and healthcare providers.

The Twelfth Malaysia Plan 2021-2025 (aka Twelfth Plan or 12MP or Malaysia 5 Year Plan), like all previous Malaysia 5 Year Plans starting from the first plan introduced in 1965, is Malaysia’s version of a comprehensive outline of government development policies and strategies. The Twelfth Plan period beginning the year 2021 up to 2025 was tabled in parliament on the 27th of September 2021 for implementation.

The sub-section Leveraging Technology and the sub-section Digitalising Healthcare Services under section Priority Area B, Strategy B3 as a part of the entire Twelfth Plan document gives special mention to EMR implementation. These sections highlight that the government will continue all efforts in the healthcare subsector to digitalise healthcare services to ensure seamless utilisation of information to solve the issue of fragmented health information systems, thereby reducing costs and unnecessary procedures.

One aspect of this digitalisation plan initiative is to leverage an emerging technology like the EMR. This initiative will be rolled out in phases to ensure the seamless flow of information among healthcare facilities and create a lifetime health record.

However, following an attempt to implement an EMR system nationwide for the 146 hospitals in Malaysia, the then Minister of Health first mooted, spoke to the media in November 2018. He said that an EMR system for all government hospitals and clinics in Malaysia could be realised within three years and completed over the next five years.

Mid-2020, the new administration in charge after a change of government, the new Minister of Health, revealed in a written reply to a Parliament session that 25% of all government hospitals and 9% of all 1090 public health clinics were already using an EMR system. Furthermore, his replies added that ten hospitals and a public health clinic were also sharing digital records through a health information exchange platform called MyHix. The Minister also gave details in his replies about an EMR pilot implementation in Negeri Sembilan. This pilot trialed an EMR system that covered online patient registration & appointment, electronic payments, and virtual consultation involved seven hospitals, 44 public health clinics, and 12 dental clinics under Phase One of the National EMR Project.

To quote Benjamin Franklin, Founding Father of the United States, when he said, “By failing to prepare, you are preparing to fail.” So let us hope this time around Malaysia will genuinely realise its dream of a successful EMR implementation by knowing what it wants and needs and knowing what to expect. If any, few people would contest the long-term benefits of a quality EMR system.

References:

  1. EMR implementation for public hospitals and clinics to cost RM1.5B, says Malaysian Health Minister, HIMSS, retrieved Oct 16, 2021. Available online: https://www.healthcareitnews.com/news/asia/emr-implementation-public-hospitals-and-clinics-cost-rm15b-says-malaysian-health-minister
  2. Electronic medical record system to cost Putrajaya up to RM1.5b, The Edge Markets, retrieved Oct 16, 2021. Available online: https://www.theedgemarkets.com/article/electronic-medical-record-system-cost-putrajaya-rm15b
  3. Electronic Health Record Market worth $ 33.69 Billion, Globally, by 2028 at 3.30% CAGR: Verified Market Research, Business, Bloomberg, retrieved Oct 16, 2021. Available online: https://www.bloomberg.com/press-releases/2021-07-15/electronic-health-record-market-worth-33-69-billion-globally-by-2028-at-3-30-cagr-verified-market-research
  4. Electronic Health Records Market Size, Share & Trends Analysis Report By Type (Post-acute, Acute), By End-use (Ambulatory Care, Hospitals), By Product (Web-, Client-server-based), By Business Models, And Segment Forecasts, 2021 – 2028, Report Overview, retrieved Oct 16, 2021. Available online: https://www.grandviewresearch.com/industry-analysis/electronic-health-records-ehr-market
  5. Twelfth Malaysia Plan 2021 – 2025, retrieved Oct 16, 2021. Available online: https://rmke12.epu.gov.my/en

Patient Access and Amendment to Medical Records

Hello readers. I am not a writer but I just maintain and write blog posts for this website / blog as its rightful owner.

Like most writers, I too have become an inveterate procrastinator. In the course of writing for this website / blog, I had actually dumped my gold standard and writing policy to make sure that I have at least written frequently something that someone would actually want to read over the past months.

As 2015 is fading away, here I have now this new post and hope to endeavour with more posts into 2016 and also take a moment to breathe and probably as psyched as you may have watched Kylo Ren in Star Wars: The Force Awakens who has the following dialogue, “Nothing will stand in our way. I will finish what you started.”, and now say to myself and tell you readers that “Nothing will stand in my way. I will finish what I started.”, by finishing some unfinished blog posts in the course of 2015 and continue to write in MRPALSMY.

Paper-based medical record practitioners, doctors and healthcare management have been influenced for example after attending healthcare conferences and lured by vendors of the promise that Electronic Medical Records (EMRs) were supposed to improve patient care and make doctors’ lives easier, do away with paper documents, and provide greater governance and stewardship for medical records practitioners.

Unfortunately, the promise of how EMRs are supposed to be in theory, haven’t worked out as well in practice as they were to be.

A detailed report in the Chicago Tribune, laments how doctors in the U.S. are even looking forward to retiring because these doctors are crying out in pain over an increasing  “burnout“ to the demands of clicking through page after page of records.

Their predicament was compounded from the findings of a new 2015 survey by Accenture PLC which found that fewer U.S. doctors believe that the EMR has improved treatment decisions, reduced medical errors or improved health outcomes. This is compared to a similar study conducted in 2012.

Headquartered in Dublin, Ireland¸ Accenture PLC is the world’s largest consulting firm and as the world’s most admired Information Technology Services Company – providing multinational management consulting, technology services, and outsourcing.

Accenture PLC had commissioned a six-country online survey of 2,619 doctors to assess their adoption and attitudes toward electronic health records and healthcare IT. The survey conducted by Nielsen between December 2014 and January 2015 included doctors across six countries, which included neighboring Singapore (200 respondents) and the U.S. (601 respondents).

Another recent study by Mayo Clinic researchers, working with the American Medical Association, further strengthened the case against the use of EMRs which found that more than half of physicians felt emotionally exhausted. Heavier workloads and “increased clerical responsibilities.” were among the chief complains.

But all is not bad for EMRs.

There is evidence that EMRs are helping patients to get more access to their medical information.

In Malaysia, a patient’s medical record on which the paper it was printed on belongs to the medical practitioner and the healthcare facility and its services. Thus, the medical practitioner and the healthcare facility and its services hold all rights associated with ownership of the physical medical record.

Nonetheless, the contents of a medical record jointly belong morally and ethically between the practitioner and the patient, simply because the practitioner who wrote the medical record holds the intellectual property right over the medical record while the patient who confided with the practitioner considers his or her “confidential” information therein contained in the medical record as “private” in observance of the on-going ethical doctor-patient relationship.

This longstanding Malaysian, almost a decade old guideline on medical records by the Malaysian Medical Council concurs well with the universal and traditionally accepted view that the information contained within the health / medical record belonged to the individual patient, and the paper it was printed on belonged to the healthcare facility.

This guideline further acknowledges and asserts that since the patient views that all the information contained in his/her medical records (i) is about him/her, (ii) that he/she should have access to records containing information about his/her medical condition for legitimate purpose and in good faith between the practitioner and him/her, (iii) he/she has a right to know what personal information is recorded, (iv) rightfully expects the records are accurate, and (v) also knows who has access to his/her personal information.

The patient’s views that all the information contained in his/her medical records (i) is about him/her, (ii) that he/she should have access to records containing information about his/her medical condition for legitimate purpose and in good faith between the practitioner and him/her, is supported by the findings from the 2015 Accenture study which shows there is evidence in the U.S. in particular, that increased access to online medical records in particular has provided patients (55 percent) with better understanding of their illness as well as having a positive impact on patient-doctor relationship.

While patients have right of such access to their medical records, the Malaysian guideline does not say if patients may be permitted to make their own changes to the mostly paper-based medical record systems available in Malaysia, and stops at proclaiming that the Malaysian patient rightfully expects the records are accurate.

The 2015 Accenture study also does not report if patients are allowed to alter medical records in the U.S., but reports that patients there do monitor their medical records and inform the practitioner of any factual errors in their personal patient information and seek to change any entries made by the practitioner in the course of consultation, diagnosis and management, thus increasing the accuracy of their medical records (60 percent).

I rest the case for EMRs that promises better patient care, which is accentuated by findings from the Accenture survey that a high percentage (82 percent of respondents) of U.S. patients when allowed by their doctors to update their own medical records, it increases their engagement in their own health as well as improves patient satisfaction, boosts understanding of their health conditions, increases patient and physician communication.

Readers can view an infographic below which summarises the 2015 Accenture study.

2015 Healthcare IT Check-Up Shows Progress (And Some Pain)

Infographic credit: Accenture PLC

In Malaysia, since we do not have specific laws or regulations that address how amendments should be processed in medical records, I think it is about time healthcare organisations in Malaysia structure their practices to comply with the greater awareness and requirements of patients’ rights and the promise of better health care in Malaysia through both paper-based and electronic medical records.

References:

  1. Accenture, 2015, Accenture doctors survey 2015: Healthcare IT pain and progress, Accenture PLC, [https://www.accenture.com/us-en/insight-accenture-doctors-survey-2015-healthcare-it-pain-progress.aspx]
  2. John, R 2015, Beleaguered by electronic record mandates, some doctors burning out, Chicago Tribune, [http://www.chicagotribune.com/business/ct-doctors-hate-records-mandate-1213-biz-20151211-story.html#]
  3. Malaysian Medical Council , 2006, Medical Records And Medical Reports, Guideline Of The Malaysian Medical Council, http://mmc.gov.my/v1/docs/Medical%20Records%20&%20Medical%20Reports.pdf]
  4. Patricia, C 2011, Patient Access and Amendment to Health Records (Updated), American Health Information Management Association (AHIMA), [http://library.ahima.org/xpedio/groups/public/documents/ahima/bok1_048587.hcsp?dDocName=bok1_048587\]