Short Review Vol. 2 No. 2 (2018): International J Health Systems and Implementation Research
Role Of Laboratory Diagnostic Data In Real – World Evidence Generation In India
Downloads
Topics
Keywords:
- Lab data
- Real world evidence
- Data analytics
- Disease surveillance
- Public health Risk prediction
Abstract
The review article provides an overview of how the analysis of the diagnostic laboratory data can support better understanding of the diseases of public health importance. This article encompasses potential applications of laboratory data to different healthcare domains such as, disease surveillance, targeted interventions of health programs, patient profiling, and predictive analytics.
Full text
This web version was prepared from the published PDF, which is the version of record. Please cite and quote from the PDF.
Background
Diagnostic laboratories are a major source of health care data. The laboratory tests have become an integral part of the patient management protocol across various healthcare settings such as inpatient, outpatient or emergency department. It is estimated that about 70% of the medical decision making is influenced by results of laboratory tests.[1]
2. Magnitude of the Diagnostic Laboratory Data in India
Current market estimate suggests Indian diagnostic industry to be around $5 billion and growing at a compound annual growth rate (CAGR) of 15-16% annually. India boasts over 100,000 diagnostic laboratories though 88-90% of them fall into an unorganized sector. Nearly, 70% of the diagnostic industry is represented by pathology laboratories.[2] Increase in disease burden, rising coverage under medical insurance and launch of technologically advanced procedures are fuelling the growth of the Indian diagnostic laboratories market.[3] In recent decades, the Indian population is witnessing epidemiological transition characterized by the double burden of infectious as well as lifestyle diseases. Though infectious diseases have declined in recent decades they are still a significant cause of morbidity and mortality in India. Non-communicable diseases now contribute to 53% of overall disease burden in India with everyone in five people is suffering from a chronic disease. [4] To aid in diagnosis and monitoring of growing number of patient population, millions of laboratory tests take place every day and in turn generate an enormous amount of healthcare data that resides in the diagnostic laboratories. This review provides an overview of the laboratory diagnostic data and its applicability in generating real-world evidence in the Indian context.
3. Laboratory Diagnostic Data and Real-World Evidence Research
Any healthcare data collected outside the clinical trial setting can be considered as ‘real-world data (RWD)’. Sources of RWD include but are not limited to inpatient records, outpatient records, clinical histories, laboratory tests, prescription data and patient reported outcomes. In developed countries, widespread adoption of electronic medical records (EMR), financial incentives by the government, and requirements by payers (insurance agencies, government, etc) for reimbursing healthcare costs have led to the generation of large administrative datasets that collect information about a patient from various sources. EMR adoption on large scale in India has been dismal and restricted to few large, corporate hospitals and individual physician clinics. Additionally, there are no provisions for health information exchange between hospitals. Both these factors lead to restriction of healthcare data in patient files or in the hospital system where it was collected.
Laboratory diagnostic data have characteristics that make it useful for real-world evidence (RWE) generation. Since most laboratory tests are conducted using computerized machines, the biomarker values are electronically coded and stored. This process of reporting of test results is automated and hence leads to easy storage of data on the servers that can be retrieved using information technology services. Thus analysing, reporting, and storing of data is free of manual intervention and assures data quality. The stored lab data contains information on patient demographics such as name, age, gender, contact information, address, referring doctor/hospital, name of the tests ordered, biomarker values, reference ranges, indicators for whether values are within/out-of-range and conclusion (wherever applicable).
4. Application of Lab Diagnostic Data
Where available, lab data should be used along with patient’s clinical data to generate holistic insights such as clinical details like complaints, findings of examination, outcomes can be correlated with diagnostic lab variables. However, in the absence of such combined data repository, standalone lab diagnostic data can still be harnessed to generate important real-world health insights. Moreover, such data when aggregated over time and across geographical areas from multiple laboratories can help in ascertaining and comparing the health status of the population, generate alerts, and help in prioritization of disease control measures. Some applications of standalone lab diagnostic data are mentioned below:
4.1 Infectious disease surveillance
Since lab tests are ordered by physicians based on signs and symptoms of the patients either to diagnose or monitor the disease, infectious disease surveillance forms an important application of lab diagnostic data. Demographics such as age and gender of suspected and positive cases and seasonal trend can be easily identified by analysing diagnostic data without the need of clinical information. Aggregated analysis of such data can be used by various stakeholders to easily pick up changes in demographics, identify trends as compared to historical data and can be used to for public health preparedness, physician education and awareness or to alert public about heightened risk during a given time by plotting heat maps. Influenza Surveillance Laboratory Network is one such example of laboratory surveillance in India, where clinically suspected influenza-like illness (ILI)/severe acute respiratory infections (SARI) patients are sent for laboratory confirmation.[5] This network currently consists of 12 regional laboratories with the Microbiology Division at the National Centre for Disease Control (NCDC) as the National Reference Laboratory. In the absence of clinical history available with the private laboratories, data from private laboratories without information on symptoms (e.g. fever, runny nose, etc) can still be used to supplement such existing surveillance system to enhance the coverage under the assumption that patients will be sent for laboratory tests only in presence of symptoms.
4.2 Targeted Interventions of Health Programs
Especially in the Indian context, several targeted interventions, under various health programs can be delivered more efficiently by using big data analytics, as high-risk population and their spatiotemporal distribution can be ascertained, and resources can be focussed. In maternal and child health programs, lab and clinic data can provide information about the high-risk pregnancies and children, who are at high-risk of malnourishment. In a similar way, the Revised National Tuberculosis Control Program (RNTCP) can be supported by identifying potential hot spots of drug resistance by analysing distribution of multidrug resistant tuberculosis (MDR TB) in different populations. For example, in the National Programme for Prevention and Control of Cancer, Diabetes, Cardiovascular Diseases and Stroke (NPCDCS) , population estimates can be generated to identify pre-diabetic population and preventive interventions and resources can be directed towards them.
4.3 Patient Profiling
Individuals who get their testing done at the same laboratory diagnostic chain are provided a unique identification number and hence can be grouped by their biomarker values into high and low risk[6] and can be followed up in time to see how they progress. Such longitudinal data can be used to formulate different hypothesis and can be confirmed by clinical data. For example, based on vitamin D levels, individuals can be first classified into deficient, insufficient, normal, or toxic categories. These groups can then be further classified according to age, gender, or region. Individuals with available lipid values can then be classified in these various subgroups and can be tracked to assess changes in lipid values over time.
4.4 Risk Prediction by Using Predictive Analytics
Predictive analytics is an important tool for risk prediction and timely decision making.[7] Risk prediction can be of great help in preventive health for personal engagement of patients in terms of chronic diseases. Late detection of chronic kidney diseases (CKD) is a common problem in India. The Kidney disease improving global outcomes (KDIGO), 2012 algorithm[8] can be applied on laboratory-based values of albuminuria and serum creatinine to find high-risk population. This is particularly important given that during early stages of CKD, most diabetic patients consult primary health physicians, and do not get the consultation from nephrologists.
5. Data Issues
5.1 Data Quality
Indian diagnostic industry is composed of largely unorganized laboratories. The industry also faces challenges such as lack of stringent regulations and accreditation requirements raising questions about the validity and accuracy of their results. Any conclusion from research will be as good as the quality of data being used and hence it is prudent to use data from laboratories that have at least an the College of American Pathologists (CAP) or National Accreditation Board of testing and calibration laboratories (NABL) accreditation.
5.2 Patient Confidentiality
Though medical records have been the source of research by physicians since a long time, it is of prime importance to ensure that no protected health information (PHI) that can identify an individual patient is used at any point. According to the Ministry of Health and Family Welfare’s guidelines on the use of electronic health records (EHRs), it is possible to use an individual’s data without his/her authorization if it is totally anonymized by removing the following patient identifying information.[9]
6. Way Forward
No dataset is perfect and even the gold-standard clinical trials have limitations. With the Government of India, Ministry of Health and Family Welfare directive on use of anonymized electronic healthcare data, it is possible to make use of such datasets wherever available ensuring patient confidentiality.[7] To generate RWE in countries where widespread EMR adoption is still a far-fetched goal, researchers need to make use of the different retrospective datasets available to them and design studies to generate or to validate hypotheses considering the inherent limitations of the datasets. Standalone laboratory diagnostic data has definite applications in generating RWE data and when supplemented with additional clinical information will expand its utility by leaps and bounds.
7. Contributors
Dr. Pankdeep Chhabra (while employed at THB) and Dr. Manmohan Singh conceived the whole study and wrote the manuscript; Akansh Khurana and Sukanya Ghildiyal provided the intellectual support to the overall manuscript.
8. Conflict of Interest
No conflict of interest is declared.
9. References
- Westgard, JO, Darcy T. The truth about quality: Medical usefulness and analytical reliability of laboratory tests. Clin Chim Acta. 2004,2;346(1):3-11. [doi:10.1016/j.cccn.2003.12.034]
- Dey S. Lack of regulation haunts India's test labs. [updated 2014 April 20; cited 2018 Jun 28]. Available from http://www.business-standard.com/article/companies/lack-of-regulation-haunts-india-s-test-labs-114041400536_1.html
- Growing Prevalence of Diseases, Coverage under Medical Insurance and Launch of Technologically Advanced Procedures to Drive India Diagnostic Laboratories Market. KenResearch [Internet]. 2017 [cited 28 June 2018]. Available from:https://www.kenresearch.com/blog/2017/03/india-diagnostic-laboratories-market/
- NCDs account for 53% of the disease burden in India: Study. The Hindu – Business line [Internet]. 2016 [cited 2018 June 28]. Available from: https://www.thehindubusinessline.com/news/science/ncds-account-for-53-of-the-disease-burden-in-india-study/article9187723.ece
- Influenza surveillance lab network [Internet]. Integrated Disease Surveillance Program (IDSP). 2009 [cited 28 June 2018]. Available from: http://idsp.nic.in/index1.php?lang=1&level=1&sublinkid=5789&lid=3722
- Sbrana F, Cocci F, Papa A, Landic P, Sampietro T, Rossi G, et al. Routine laboratory tests to risk-stratify patients with chronic coronary artery disease. J Cardiol. 2013;61(2):132-7. [doi:10.1016/j.jjcc.2012.09.005]
- Srinivas TR, Taber DJ, Su Z, Zhang J, Mour G, Northrup D, et al. Big data, predictive analytics and quality improvement in kidney transplantation - A proof of concept. Am J Transplant. 2017;17(3):671-81. [Google Scholar]
- Clinical Practice Guideline Update for the Diagnosis, Evaluation, Prevention, and Treatment of Chronic Kidney Disease–Mineral and Bone Disorder (CKD-MBD). KDIGO [Internet]. 2017 [cited 28 June 2018]. Available from; https://kdigo.org/wpcontent/uploads/2017/02/2017-KDIGO-CKD-MBD-GL-Update.pdf
- Electronic Healthcare Records (EHR) Standards for India. Ministry of Health and Family Welfare [Internet]. 2016 [cited 28 June 2018]. Available from: https://mohfw.gov.in/sites/default/files/17739294021483341357.pdf
Published
Article metrics
- Views
- 1,498
- Downloads
- 1,074
Based on this site's usage statistics.
Issue
Section
How to Cite
More in this topic: Health information & digital health
View Point · 2024
Digitalization of ASHA Work: Challenges and Way Forward