Real World Resource

Introduction to real‑world evidence studies

Deepa Chodankar
Clinical Study Unit, Sanofi Synthelabo (India) Limited, Mumbai, Maharashtra, India

Address for correspondence: Dr. Deepa Chodankar,

Sanofi Synthelabo (India) Limited, Sanofi House, CTS No. 117‑B, L&T Business Park, Saki Vihar Road, Powai, Mumbai ‑ 400 072, Maharashtra, India.
E‑mail: [email protected]

Real‑world data (RWD) are data relating to patient health investigational product. RCTs are designed to focus
status and/or the delivery of health care routinely collected on internal validity (capability of a clinical study to
from a variety of sources. provide reliable results which are actually true and not
due to an error), which may sometimes compromise
Real‑world evidence (RWE) is the clinical evidence about generalizability to general population. [2] RCTs are
the usage and potential benefits or risks of a medical conducted on very selective populations, so patients
product derived from analysis of RWD.”[1] with comorbidities may be excluded. Furthermore,
these studies are conducted in very controlled settings.
RWD can be generated from: Most of the clinical treatment guidelines are formulated
• Electronic health records (EHRs)
based on the RCT results. However, these results do
• Medical claims, billing data, and insurance data
not truly represent the actual entire population, since
• Data from product and disease registries
these RCTs have many inclusion and exclusion criteria.
• Patient‑generated data, including from in‑home‑use
Hence, these results from RCT require support from
settings
diverse situations that would be present in a real‑world
• Data gathered from other sources that can inform on
clinical scenario [Table 1].[3]
health status, such as mobile devices.[1,2]

Findings from clinical trials cannot be generalized to Safety evaluation is another aspect that can be influenced by
population at large due to the stringent eligibility criteria. the type of study design. In RCTs, small sample size, strict
RWE studies complement clinical trials by generalizing patient eligibility criteria, and short‑term follow‑up do not
the findings from clinical trial to general population. always allow the measurement of rare adverse events (AEs).
Furthermore, RWE can provide information on other areas, Hence, AE rates recorded during RCTs do not correctly
such as natural history and course of disease, effectiveness reflect the incidence of AEs in real‑life settings. The short
studies, outcome studies, and safety surveillance.[2] duration of RCTs may not allow AEs to be detected in
patients who would have developed the AE after a longer
DIFFERENCE REAL‑WORLD EVIDENCE STUDIES period of drug exposure; the selection criteria might have
AND RANDOMIZED CLINICAL TRIALS excluded older patients and patients with comorbidities,
so the frequency of AE reported from RCTs might have
Randomized clinical trials (RCTs) are initial studies been lower.[4]
conducted to establish the safety and efficacy of an
This is an open access journal, and articles are distributed under the terms of the Creative Commons
Attribution-NonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and
Access this article online build upon the work non-commercially, as long as appropriate credit is given and the new creations
Quick Response Code: are licensed under the identical terms.
Website:
www.picronline.org
How to cite this article: Chodankar D. Introduction to real-world evidence
studies. Perspect Clin Res 2021;12:171-4.

DOI: © 2021 Perspectives in Clinical Research | Published by Wolters Kluwer - Medknow

10.4103/picr.picr_62_21
Received: 23‑03‑21, Accepted: 24-03-21, Published: 07-07-21.

© 2021 Perspectives in Clinical Research | Published by Wolters Kluwer - Medknow 171

 Real World Resource

Table 1: Randomized clinical trials versus real‑world evidence[2,3]

RCT RWE
Setting Experimental or interventional setting Real‑world setting or observational or noninterventional setting
Study conduct Protocol‑based, GCP compliant Real‑life clinical practice
Treatment Fixed pattern Variable pattern
Participant population Strict and many inclusion and exclusion criteria Very few inclusion and exclusion criteria
Attending physician Investigator Practitioner
Comparator Placebo/selective alternative interventions Either no control arm or standard treatment or care
Outcome Efficacy Effectiveness
Randomization and blinding Yes No
GCP=Good clinical practice, RCT=Randomized clinical trial, RWE=Real‑world evidence

ADVANTAGES OF REAL‑WORLD EVIDENCE iii. State Medicaid

COMPARED WITH RANDOMIZED CLINICAL iv. Insurance companies.
TRIALS[2,3]
2. Clinical studies
i. Clinical trial data such as pragmatic study
1. As compared to RCT, depending on the type of RWE,
ii. Product registry
it may take much less time, less resources, and less cost
iii. Disease registry.
2. In RWE studies, data can be accessed rapidly and data
3. Clinical setting
can be retrieved easily
i. EHR from clinicians/hospitals
3. RWE studies can be used to evaluate the natural history
of disease, prevalence, incidence, unmet medical need, ii. Laboratory test
current treatment patterns, and standard of care iii. Billing data.
4. RWE studies can be used to support patient outcomes 4. Pharmacy
and health economics i. Sales data
5. RWE studies can be used to understand current ii. Prescription data.
health‑care services 5. Patient powered
6. Research which is not possible with RCT can be done i. Social media
with RWE, for example, studies on high‑risk groups ii. Patient advocacy groups/patient communities.
7. Side effects which are less frequently seen can be
EXAMPLE OF REAL‑WORLD EVIDENCE STUDIES
studied better with a RWE study as compared to RCT CONDUCTED IN INDIA
since RCT is conducted in a smaller population and
with a shorter duration. Few examples of different types of RWE studies conducted
are given below:
SOURCES OF REAL‑WORLD DATA
1. Retrospective study: This is a retrospective study done
Health records can be paper based or electronic. RWD to evaluate to what extent type 1 diabetic guidelines
can be leveraged if it is digitally recorded so that it can be are followed in clinical practice in Sweden. These
aggregated and analyzed appropriately for research. Over guidelines recommend quarterly or more frequent
the last 2 decades, significant progress has been made in hemoglobin A1c (HbA1c) assessments in patients with
digitalization of health records.[5] Electronic health‑care uncontrolled type 1 diabetes mellitus. 5989 patients
records (EHRs) can be defined as an organized set of were recruited from 10 outpatient diabetes clinics
health‑care data, which can be accessed electronically. in Sweden. Diab‑Base electronic medical record
They contain a diversity of data, the most frequent being database was used for data collection. Data on patient
medical records from general practitioners, specialists or characteristics, including treatment, general risk
hospitals, pharmacies, prescription data, and sometimes factors for diabetic complications, and frequency of
lifestyle‑related information.[6] HbA1c measurements, were retrieved for all patients.
This study provided important insight into HbA1c
RWD can be accessed from various sources as mentioned measurement in routine clinical practice in Sweden. It
below:[7,8] was found that the measurements were done less than
1. Claims that recommended by guidelines recommend[9]
i. Medical claims 2. Prospective study: LANDMARC study is a prospective,
ii. Prescription drug claims multicenter study evaluating a large cohort of people

172 Perspectives in Clinical Research | Volume 12 | Issue 3 | July-September 2021

 Real World Resource

with type 2 diabetes mellitus across India over a • Registry for orphan/rare disease, such as thalassemia
period of 3 years. This study will reveal the trends • Health economics and outcomes research in
in complications associated with diabetes; treatment government as compared to private hospitals
strategies used by physicians; and correlation among • Identification of unmet medical needs which would
treatment, control, and complications of diabetes[10] be relevant to Indian population
3. Cross‑sectional study: This multicenter study was • Comparative effectiveness research of branded
designed to determine the control of dyslipidemia products versus generic and biologic versus
in the Indian diabetic population treated with biosimilar
lipid‑lowering drugs. The study was conducted in • Comparative effectiveness research of Indian prescribed
178 sites in India. This study found that dyslipidemia dose as compared to prescribing information.
control in Indian type 2 diabetes mellitus patients is
very poor with almost half of them not reaching their In order that RWE studies achieve its true objectives
low‑density lipoprotein‑cholesterol goal.[11] and are designed, conducted, analyzed, and reported
appropriately in India, the following needs to be done:
CURRENT SCENARIO OF REAL‑WORLD • Different stakeholders such as government,
EVIDENCE IN INDIA academicians, investigators, pharmaceutical industry,
and patient support group would have to come
Recording patient data in the hospital needs time and
together and encourage the conduct of RWE studies
effort. There is no legislative framework or guidelines in which are relevant to Indian health care
India, which would enable collating data from hospitals • India‑specific RWE guidelines are required to be drafted
on standard indicators of quality of patient care. This • Support in terms of data management, statisticians,
could be one of the reasons for the poor quality of and technical support for public and private hospitals
medical records. Many of the hospitals do not even so as to help them in developing structured format for
record patient history in detail and information about collecting data and also to analyze and report RWE
treatment provided is sometimes very less. The private studies appropriately.
sector hospitals have better resources and funds as
compared to public hospitals. However, in the absence Financial support and sponsorship
of any government regulations, there is no requirement Nil.
to collect these data. Furthermore, public hospitals are
overloaded with patients. For RWD, well‑defined formats Disclosure
whether in the physician’s clinic or in the operation The insights stated in this article are author’s personal
theater are required to yield data which can be analyzed opinion and does not reflect those of the current and
appropriately. Some prestigious medical institutes in the previous employers.
country have developed a simple format to get structured
data.[12] Funding
This initiative was supported by Sanofi.
WHAT CAN BE DONE NEXT
Conflict of interest
To generate RWD, it is crucial to develop brief and easy to Dr Deepa Chodankar is an employee of Sanofi India and
fill computer‑readable formats for collecting information might be shareholders of Sanofi.
at the clinic or at the hospital.[12]
REFERENCES
India with its diverse vast population does have an
1. Framework for FDA’s Real‑World Evidence Program. U.S: Food and
advantage for conducting RWE studies, especially in areas Drugs Administration; 2018.
as listed below:[2] 2. Bhatt A. Conducting real‑world evidence studies in India. Perspect
• Natural history and long‑term complications of Clin Res 2019;10:51‑6.
3. Kim HS, Lee S, Kim JH. Real‑world evidence versus randomized
communicable diseases, such as drug‑resistant
controlled trial: Clinical research based on electronic medical records.
tuberculosis, and noncommunicable diseases, such as J Korean Med Sci 2018;33:e213.
diabetes, hypertension, and heart failure 4. Monti S, Grosso V, Todoerti M, Caporali R. Randomized controlled

Perspectives in Clinical Research | Volume 12 | Issue 3 | July-September 2021 173

 Real World Resource

trials and real‑world data: Differences and similarities to untangle is real‑world data? A review of definitions based on literature and
literature data. Rheumatology 2018;57:vii54‑8. stakeholder interviews. Value Health 2017;20:858‑65.
5. National Academies of Sciences. Engineering, and Medicine; Health 9. Matuleviciene V, Attvall S, Ekelund M, Clements M, Dahlqvist S,
and Medicine Division; Board on Health Sciences Policy; Forum on Fahlén M, et al. A retrospective study in 5,989 patients with type 1
Drug Discovery, Development, and Translation. Real‑World Evidence diabetes in 10 outpatient diabetes clinics in Sweden of the frequency
Generation and Evaluation of Therapeutics: Proceedings of a of measuring HbA1c in clinical practice. J Diabetes Metab 2014;5:5.
Workshop. 3, Opportunities for Real‑World Data. Washington (DC): 10. Das AK, Mithal A, Kumar KM, Unnikrishnan AG, Kalra S, Thacker H,
National Academies Press (US); February 15, 2017. Available from: et al. Rationale, study design and methodology of the LANDMARC
https://www.ncbi.nlm.nih.gov/books/NBK441694/. [Last accessed trial: A 3‑year, pan‑India, prospective, longitudinal study to assess
on 2021 Jan 12]. management and real‑world outcomes of diabetes mellitus. Diabet
6. European Medicines Agency. Observational Data (Real World Data). Med 2020;37:885‑92.
European Medicines Agency; 2017. 11. Mithal A, Majhi D, Shunmugavelu M, Talwarkar PG, Vasnawala H,
7. US Food and Drug’s Administration. Real World Evidence. Raza AS. Prevalence of dyslipidemia in adult Indian diabetic patients:
Available from: https://www.fda.g ov/science‑research/ A cross sectional study (SOLID). Indian J Endocrinol Metab
science‑and‑research‑special‑topics/real‑world‑evidence. [Last 2014;18:642‑7.
accessed on 2021 Jan 20]. 12. Rajivlochan M. Clinical audits and the state of record keeping in India.
8. Makady A, de Boer A, Hillege H, Klungel O, Goettsch W. What Ann Neurosci 2015;22:197‑8.

174 Perspectives in Clinical Research | Volume 12 | Issue 3 | July-September 2021