The document discusses the steps for developing clinical prediction models. It begins by outlining the key questions to consider such as the target outcome, population, and users. Step 2 involves selecting appropriate datasets, preferably longitudinal cohorts that reflect the target population. Step 3 covers handling variables, including variable selection, categorization, and dealing with missing data. Step 4 describes generating the model through techniques like logistic regression. The final step involves evaluating the model through measures of discrimination and calibration in both development and validation datasets. Validation in external populations is ideal to test generalizability. The overall goal is to create an accurate and useful model to aid clinical decision-making.
3. Managing Customer Relationships: A Strategic Framework,
Third Edition, Don Peppers and Martha Rogers
3
3
Dialogue Requirements
Parties at both ends have been clearly identified
All parties in the dialogue must be able to participate in it
All parties to a dialogue must want to participate in it
Dialogues can be controlled by anyone in the exchange
A dialogue with an individual customer will change an
enterprise’s behavior toward that individual (and vice versa)
A dialogue should pick up where it last left off
Managing Customer Relationships: A Strategic Framework,
Third Edition, Don Peppers and Martha Rogers
4
4
Implicit and Explicit Bargains
Implicit bargains
Expressed through nonaddressable (one-way) media
Example: early television commercials - “Watch our ad and see
the show for free.”
Explicit bargains
Made possible through addressable (two-way) media
Example: website operators that offer free e-mail to individuals
willing to view targeted advertisements
Dialogue and interaction are so valuable that enterprises offer
4. free services or discounts to customers willing to participate
Managing Customer Relationships: A Strategic Framework,
Third Edition, Don Peppers and Martha Rogers
5
5
Two-Way, Addressable Media
World Wide Web
Social media
Wireless technology
Voicemail
E-mail
Texting
Fax
Digital video recorders
Interactive voice response
Managing Customer Relationships: A Strategic
Framework, Third Edition, Don Peppers and Martha Rogers
6
6
Integration of Interaction across the Enterprise
Laterally coordinated across media types
Longitudinally coordinated over the individual customer’s
relationship with the firm
Results in customer’s conversation with firm picking up exactly
where it left off – no matter what channel was used last time
5. Managing Customer Relationships: A Strategic Framework,
Third Edition, Don Peppers and Martha Rogers
7
7
Touchpoint Mapping
Dimensions of customer experience:
Physical – location based
Emotional – related to culture of customer-facing employees
and their behaviors
Logical – information flow within enterprise; includes
touchpoint mapping
Touchmap: a graphical depiction of the interactions a company
has with each customer segment across each available channel –
from the customer’s perspective (outside in)
Managing Customer Relationships: A Strategic Framework,
Third Edition, Don Peppers and Martha Rogers
8
8
Uses of a Touchmap
Tracks technology implementation
Tracks employee induction and training
Tracks employee performance
Can help secure executive sponsorship
Identifies redundancies and other opportunities for efficiency
Reveals data gaps and silos
6. Managing Customer Relationships: A Strategic Framework,
Third Edition, Don Peppers and Martha Rogers
9
9
Customer Dialogue
Most valuable dialogue information:
Customer needs
Potential value
Dialogue methods
Drip irrigation dialogue: learning incrementally about a
customer through each interaction
Golden Questions: designed to reveal most important customer
information while requiring least possible customer effort
(focused on needs, not product)
Managing Customer Relationships: A Strategic Framework,
Third Edition, Don Peppers and Martha Rogers
10
10
Earning Trust in Customer Dialogue
Use a flexible opt-in policy
Make an explicit bargain
Make your privacy policy clear and simple
7. Create a culture based on customer trust
Remember you’re responsible for your partners, too
Managing Customer Relationships: A Strategic Framework,
Third Edition, Don Peppers and Martha Rogers
11
11
Complaining Customers,
Hidden Assets
Complaints provide the following opportunities:
To fix the relationship
To expand scope of knowledge about the customer
To provide data points about the enterprise’s products and
services
Studies show complaining customers may have highest potential
value
The most loyal customers are the ones who take time to
complain in the first place
Customers who have a problem resolved are more loyal than
those who never had a problem
Managing Customer Relationships: A Strategic Framework,
Third Edition, Don Peppers and Martha Rogers
12
8. 12
Complaining Customers,
Hidden Assets
Seek out complaints, don’t avoid them
Complaint discovery:
Is there anything more we can do for you?
Is there anything we can do better?
Managing Customer Relationships: A Strategic Framework,
Third Edition, Don Peppers and Martha Rogers
13
13
image1.png
image2.jpeg
image3.jpeg
image4.png
image5.png
38 www.e-enm.org
Endocrinol Metab 2016;31:38-44
http://dx.doi.org/10.3803/EnM.2016.31.1.38
pISSN 2093-596X · eISSN 2093-5978
9. Review
Article
How to Establish Clinical Prediction Models
Yong-ho Lee1, Heejung Bang2, Dae Jung Kim3
1Department of Internal Medicine, Yonsei University College of
Medicine, Seoul, Korea; 2Division of Biostatistics, Department
of Public Health Sciences, University of California Davis
School of Medicine, Davis, CA, USA; 3Department of
Endocrinology
and Metabolism, Ajou University School of Medicine, Suwon,
Korea
A clinical prediction model can be applied to several
challenging clinical scenarios: screening high-risk individuals
for asymp-
tomatic disease, predicting future events such as disease or
death, and assisting medical decision-making and health
education.
Despite the impact of clinical prediction models on practice,
prediction modeling is a complex process requiring careful
statisti-
cal analyses and sound clinical judgement. Although there is no
definite consensus on the best methodology for model develop-
ment and validation, a few recommendations and checklists
have been proposed. In this review, we summarize five steps for
de-
veloping and validating a clinical prediction model: preparation
for establishing clinical prediction models; dataset selection;
handling variables; model generation; and model evaluation and
validation. We also review several studies that detail methods
for developing clinical prediction models with comparable
examples from real practice. After model development and
vigorous
10. validation in relevant settings, possibly with evaluation of
utility/usability and fine-tuning, good models can be ready for
the use
in practice. We anticipate that this framework will revitalize the
use of predictive or prognostic research in endocrinology,
leading
to active applications in real clinical practice.
Keywords: Clinical prediction model; Development; Validation;
Clinical usefulness
INTRODUCTION
Hippocrates emphasized prognosis as a principal component of
medicine [1]. Nevertheless, current medical investigation
mostly focuses on etiological and therapeutic research, rather
than prognostic methods such as the development of clinical
prediction models. Numerous studies have investigated wheth-
er a single variable (e.g., biomarkers or novel clinicobiochemi-
cal parameters) can predict or is associated with certain out-
comes, whereas establishing clinical prediction models by in-
corporating multiple variables is rather complicated, as it re-
quires a multi-step and multivariable/multifactorial approach to
design and analysis [1].
Clinical prediction models can inform patients and their
physicians or other healthcare providers of the patient’s proba-
bility of having or developing a certain disease and help them
with associated decision-making (e.g., facilitating patient-doc-
tor communication based on more objective information). Ap-
Received: 9 January 2016, Revised: 14 January 2016,
Accepted: 27 January 2016
Corresponding authors: Dae Jung Kim
Department of Endocrinology and Metabolism, Ajou University
School of
12. tions. Furthermore, in some instances, certain models can pre-
dict the possibility of having future disease or provide a prog-
nosis for disease (e.g., complication or mortality). This review
will concisely describe how to establish clinical prediction
models, including the principles and processes for conducting
multivariable prognostic studies and developing and validating
clinical prediction models.
CONCEPT OF CLINICAL PREDICTION
MODELS
In the era of personalized medicine, prediction of prevalent or
incident diseases (diagnosis) or outcomes for future disease
course (prognosis) became more important for patient manage-
ment by health-care personnel. Clinical prediction models are
used to investigate the relationship between future or unknown
outcomes (endpoints) and baseline health states (starting point)
among people with specific conditions [2]. They generally
combine multiple parameters to provide insight into the relative
impacts of individual predictors in the model. Evidence-based
medicine requires the strongest scientific evidence, including
findings from randomized controlled trials, meta-analyses, and
systematic reviews [3]. Although clinical prediction models are
partly based on evidence-based medicine, the user must also
adopt practicality and an artistic approach to establish clinically
relevant and meaningful models for targeted users.
Models should predict specific events accurately and be rela-
tively simple and easy to use. If a prediction model provides
inaccurate estimates of future-event occurrences, it will mislead
healthcare professionals to provide insufficient management of
patients or resources. On the other hand, if a model has high
predictability power but is difficult to apply (e.g., with compli-
cated calculation or unfamiliar question/item or unit), time con-
suming, costly [4] or less relevant (e.g., European model for
Koreans, event too far away), it will not be commonly used.
For example, a diabetes prediction model developed by Lim et
13. al. [5] has a relatively high area under the receiver operating
curve (AUC, 0.77), while blood tests that measure hemoglobin
A1c, high density lipoprotein cholesterol, and triglyceride are
included in the risk score, which would generally require clini-
cian’s involvement so could be a major barrier for use in com-
munity settings. When prediction models consist of complicat-
ed mathematical equations [6,7], a web-based application can
enhance implementation (e.g., calculating 10-year and lifetime
risk for atherosclerotic cardiovascular disease [CVD] is avail-
able at http://tools.acc.org/ASCVD-Risk-Estimator/). There-
fore, achieving a balance between predictability and simplicity
is a key to a good clinical prediction model.
STEPS TO DEVELOPING CLINICAL
PREDICTION MODELS
There are several reports [1,8-13] and a textbook [14] that de-
tail methods to develop clinical prediction models. Although
there is currently no consensus on the ideal construction meth-
od for prediction models, the Prognosis Research Strategy
(PROGRESS) group has proposed a number of methods to im-
prove the quality and impact of model development [2,15]. Re-
cently, investigators on the Transparent Reporting of a multi-
variable prediction model for Individual Prognosis Or Diagno-
sis (TRIPOD) study have established a checklist of recommen-
dations for reporting on prediction or prognostic models [16].
This review will summarize the analytic process for developing
clinical prediction models into five stages.
Stage 1: preparation for establishing clinical prediction
models
The aim of prediction modeling is to develop an accurate and
useful clinical prediction model with multiple variables using
comprehensive datasets. First, we have to articulate several im-
portant research questions that affect database selection and the
15. may have relatively low power for predicting future incidence
of disease when different risk factors come into play. Accord-
ingly, longitudinal or prospective cohort datasets should be
used for prediction models for future events (Table 1). Models
for prevalent events are useful for predicting asymptomatic
diseases, such as diabetes or chronic kidney disease, by screen-
ing undiagnosed cases, whereas models for incident events are
useful for predicting the incidence of relatively severe diseases,
such as CVD, stroke, and cancer.
A universal clinical prediction model for disease does not
exist; thus, separate specific models that can individually as-
sess the role of ethnicity, nationality, sex, or age on disease risk
are warranted. For example, the Framingham coronary heart
disease (CHD) risk score is generated by one of the most com-
monly used clinical prediction models; however, it tends to
overestimate CHD risk by approximately 5-fold in Asian popu-
lations [17,18]. This indicates that models derived from one
ethnicity sample may not be directly applied to populations of
other ethnicities. Other specific characteristics of study popula-
tions beside ethnicity (e.g., obesity- or culture-related vari-
ables) could be important.
There is no absolute consensus on the minimal requirement
for dataset sample size. Generally, large representative, contem-
porary datasets that closely reflect the characteristics of their
target population are ideal for modeling and can enhance the
relevance, reproducibility, and generalizability of the model.
Moreover, two types of datasets are generally needed: a devel-
opment dataset and a validation dataset. A clinical prediction
model is first derived from analyses of the development dataset
and its predictive performance should be assessed in different
populations based on the validation dataset. It is highly recom-
mended to use validation datasets from external study popula-
tions or cohorts, whenever available [19,20]; however, if it is
not possible to find appropriate external datasets, an internal
validation dataset can be formed by randomly splitting the orig-
16. inal cohort into two datasets (if sample size is large) or statisti-
cal techniques such as jackknife or bootstrap resampling (if not)
[21]. The splitting ratio can vary depending on the researchers’
particular goals, but generally, more subjects should be allocat-
ed to the development dataset than to the validation dataset.
Stage 3: handling variables
Since cohort datasets contain more variables than can reason-
ably be used in a prediction model, evaluation and selection of
the most predictive and sensible predictors should be done.
Generally, inclusion of more than 10 variables/questions may
decrease the efficiency, feasibility and convenience of predic-
tion models, but expert’s judgment that could be somewhat
subjective is required to assess the need for each situation. Pre-
dictors that were previously found to be significant should nor-
mally be considered as candidate variables (e.g., family history
of diabetes in diabetes risk score). It should be noted that not
all significant predictors need to be included in the final model
(e.g., P<0.05); predictor selection must be always guided by
clinical relevance/judgement to prevent nonsensical or less rel-
evant or user-unfriendly variables (e.g., socioeconomic status-
related) or possible false-positive associations. Additionally,
Table 1. Characteristics of Different Clinical Prediction Models
according to Their Purpose
Characteristic Prevalent/concurrent events Incident/future
events
Data type Cross-sectional data Longitudinal/prospective cohort
data
Application Useful for asymptomatic diseases for screening
undiagnosed cases (e.g., diabetes, CKD)
Useful for predicting the incidence of diseases
18. small and variable unification does not alter the statistical sig-
nificance of the model materially. Although continuous param-
eters are usually included in a regression model, assuming lin-
earity, researchers should consider the possibility of non-linear
associations such as J- or U-shaped distributions [24]. Further-
more, the relative effect of a continuous variable is determined
by the measurement scale used in the model [8]. For example,
the impact of fasting glucose levels on the risk of CVD may be
interpreted as having a stronger influence when scaled per 10
mg/dL than per 1 mg/dL.
Researchers often emphasize the importance of not dichoto-
mizing continuous variables in the initial stage of model devel-
opment because valuable predictive information can be lost
during categorization [24]. However, prediction models—is
not the same thing as regression models—with continuous pa-
rameters may be complex and hard to use or be understood by
laypersons, because they have to calculate their risk scores by
themselves. A web or computer-based platform is usually re-
quired for the implementation of these models. Otherwise, in a
later phase, researchers may transform the model into a user-
friendly format by categorizing some predictors, if the predic-
tive capacity of the model is retained [8,19,25].
Finally, missing data is a chronic problem in most data anal-
yses. Missing data can occur various reasons, including uncol-
lected (e.g., by design), not available or not applicable, refusal
by respondent, dropout, or “don’t know.” To handle this issue,
researchers may consider imputation technique, dichotomizing
the answer into yes versus others, or allow “unknown” as a
separate category as in http://www.cancer.gov/bcrisktool/.
Stage 4: model generation
Although there are no consensus guidelines for choosing vari-
ables and determining structures to develop the final prediction
model, various strategies with statistical tools are available
[8,9]. Regression analyses, including linear, logistic, and Cox
models are widely used depending on the model and its intend-
19. ed purpose. First, the full model approach is to include all the
candidate variables in the model; the benefit of this approach is
to avoid overfitting and selection bias [9]. However, it can be
impractical to pre-specify all predictors and previously signifi-
cant predictors may not be in a new population/sample. Sec-
ond, a backward elimination approach or stepwise selection
method can be applied to remove a number of insignificant
candidate variables. To check for overfitting of the model,
Akaike information criterion (AIC) [26], an index of model fit-
ting that charges a penalty against larger models, may be useful
[19]. Lower AIC values indicate a better model fit. Some inter-
pret that AIC addresses explanation and Bayesian information
criterion (BIC) addresses prediction, where BIC may be con-
sidered a Bayesian counterpart [27].
If researchers prefer algorithm modeling culture instead of
data modeling culture, e.g., formula-based regression [28], a
classification and regression tree analysis or recursive parti-
tioning could be considered [28-30].
With regard to determining scores for each predictor in the
generation of simplified models, researchers using expert judg-
ment may create a weighted scoring system by converting β
coefficients [19] or odds ratios [20] from the final model to in-
teger values, while preserving monotonicity and simplicity. For
example, from the logistic regression model built by Lee et al.
[19], β coefficients <0.6, 0.7 to 1.3, 1.4 to 2.0, and >2.1 were
assigned scores of 1, 2, 3, and 4, respectively.
Stage 5: model evaluation and validation (internal/
external)
After model generation, researchers should evaluate the predic-
tive power of their proposed model using an independent datas-
et, where truly external dataset is preferred whenever available.
There are several standard performance measures that capture
different aspects: two key components are calibration and dis-
crimination [8,9,31]. Calibration can be assessed by plotting the
observed proportions of events against the predicted probabili-
21. sification index and integrated discrimination improvement are
known to lead to non-proper scoring and vulnerable to miscali-
brated or overfit problems [33]. AUC and R2 are often hard to
increase by a new predictor, even with large odds ratio. Despite
similar names, AIC and BIC address slightly different issues
and
information in BIC can be decreased with sample size increases.
The Hosmer-Lemeshow test is highly sensitive when sample
size is large, which is not an ideal property as a goodness-fit
sta-
tistic. Calibration plot can easily provide a high correlation
coef-
ficient (>0.9), simply because they are computed for predicted
versus observed values on grouped data (without random vari-
ability). Finally, AUC also needs caution: a high value (e.g.,
>0.9) may mean excellent discrimination but it can also reflect
the situation where prediction is not so relevant: (1) the task is
closer to diagnostic or early onset rather than prediction; (2)
cas-
es vs. non-cases are fundamentally different with minimal over-
lap; or (3) predictors and endpoints are virtually the same things
(e.g., current blood pressure vs. future blood pressure).
Despite the long list provided above, we do not think this is
a discouraging news to researchers. We may tell us no method
is perfect and “one size does not fit all” is also true to statistical
methods; thus blinded or automated application can be danger-
ous.
It is crucial to separate internal and external validation and
to conduct the previously mentioned analyses on both datasets
to finalize the research findings (see the following for example
reports [19,20,34]). Internal validation can be done using a ran-
dom subsample or different years from the development dataset
or by conducting bootstrap resampling [22]. This approach can
particularly assess the stability of selected predictors, as well as
prediction quality. Subsequently, external validation should be
22. performed on an independent dataset from that which was pre-
viously used to develop the model. For example, datasets can
be obtained from populations from other hospitals or centers
(see geographic validation [19]) or a more recently collected
cohort population (temporal validation [34]). This process is
often considered to be a more powerful test for prediction mod-
els than internal validation because it evaluates transportability,
generalizability and true replication, rather than reproducibility
[8]. Poor model performance may occur after use of an external
dataset due to differences in healthcare systems, measurement
methods/definitions of predictors and/or endpoint, subject
characteristics or context (e.g., high vs. low risk).
CONCLUSIONS
For patient-centered perspectives, clinical prediction models
are useful for several purposes: to screen high-risk individuals
Table 2. Statistical Measures for Model Evaluation
Sensitivity and specificity
Discrimination (ROC/AUC)
Predictive values: positive, negative
Likelihood ratio: positive, negative
Accuracy: Youden index, Brier score
Number needed to treat or screen
Calibration: Calibration plot, Hosmer-Lemeshow test
Model determination: R2
24. clinical prediction models to the public. For example, Korean
risk models for diabetes, fatty liver, CVD, and osteoporosis are
readily available at http://cmerc.yuhs.ac/mobileweb/. Simple
model may be translated into a one page checklist for patient’s
self-assessment (e.g., equipped in waiting room in clinic). We
anticipate that the framework that we provide/summarize,
along with additional assistance from related references or text-
books, will help predictive or prognostic research in endocri-
nology; this will lead to active application of these practices in
real world settings. In light of the personalized- and precision-
medicine era, further research is needed to attain individual-
level predictions, where genetic or novel biomarkers can play
bigger roles, as well as simple generalized predictions which
can further help patient-centered care.
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was re-
ported.
ACKNOWLEDGMENTS
This study was supported by a grant from the Korea Healthcare
Technology R&D Project, Ministry of Health and Welfare, Re-
public of Korea (No. HI14C2476). H.B. was partly supported
by the National Center for Advancing Translational Sciences,
National Institutes of Health, through grant UL1 TR 000002.
D.K. was partly supported by a grant of the Korean Health
Technology R&D Project, Ministry of Health and Welfare, Re-
public of Korea (HI13C0715).
ORCID
Yong-ho Lee http://orcid.org/0000-0002-6219-4942
Dae Jung Kim http://orcid.org/0000-0003-1025-2044
25. REFERENCES
1. Moons KG, Royston P, Vergouwe Y, Grobbee DE, Altman
DG. Prognosis and prognostic research: what, why, and how?
BMJ 2009;338:b375.
2. Hemingway H, Croft P, Perel P, Hayden JA, Abrams K,
Timmis A, et al. Prognosis research strategy (PROGRESS)
1: a framework for researching clinical outcomes. BMJ
2013;346:e5595.
3. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Rich-
ardson WS. Evidence based medicine: what it is and what it
isn’t. BMJ 1996;312:71-2.
4. Greenland S. The need for reorientation toward cost-effective
prediction: comments on ‘Evaluating the added predictive
ability of a new marker. From area under the ROC curve to re-
classification and beyond’ by M. J. Pencina et al., Statistics in
Medicine (DOI: 10.1002/sim.2929). Stat Med 2008;27:199-
206.
5. Lim NK, Park SH, Choi SJ, Lee KS, Park HY. A risk score
for predicting the incidence of type 2 diabetes in a middle-
aged Korean cohort: the Korean genome and epidemiology
study. Circ J 2012;76:1904-10.
6. Griffin SJ, Little PS, Hales CN, Kinmonth AL, Wareham NJ.
Diabetes risk score: towards earlier detection of type 2 diabe-
tes in general practice. Diabetes Metab Res Rev 2000;16:164-
71.
7. Goff DC Jr, Lloyd-Jones DM, Bennett G, Coady S,
D’Agostino RB, Gibbons R, et al. 2013 ACC/AHA guideline
on the assessment of cardiovascular risk: a report of the
27. proach to development, validation, and updating. New
York: Springer; 2009.
15. Steyerberg EW, Moons KG, van der Windt DA, Hayden
JA, Perel P, Schroter S, et al. Prognosis Research Strategy
(PROGRESS) 3: prognostic model research. PLoS Med
2013;10:e1001381.
16. Collins GS, Reitsma JB, Altman DG, Moons KG. Transpar-
ent Reporting of a multivariable prediction model for Indi-
vidual Prognosis or Diagnosis (TRIPOD): the TRIPOD
statement. Ann Intern Med 2015;162:55-63.
17. Liu J, Hong Y, D’Agostino RB Sr, Wu Z, Wang W, Sun J, et
al. Predictive value for the Chinese population of the Fram-
ingham CHD risk assessment tool compared with the Chi-
nese Multi-Provincial Cohort Study. JAMA 2004;291:2591-
9.
18. Jee SH, Jang Y, Oh DJ, Oh BH, Lee SH, Park SW, et al. A
coronary heart disease prediction model: the Korean Heart
Study. BMJ Open 2014;4:e005025.
19. Lee YH, Bang H, Park YM, Bae JC, Lee BW, Kang ES, et
al.
Non-laboratory-based self-assessment screening score for
non-alcoholic fatty liver disease: development, validation and
comparison with other scores. PLoS One 2014;9:e107584.
20. Bang H, Edwards AM, Bomback AS, Ballantyne CM, Bril-
lon D, Callahan MA, et al. Development and validation of a
patient self-assessment score for diabetes risk. Ann Intern
Med 2009;151:775-83.
21. Kotronen A, Peltonen M, Hakkarainen A, Sevastianova K,
Bergholm R, Johansson LM, et al. Prediction of non-alco-
28. holic fatty liver disease and liver fat using metabolic and
genetic factors. Gastroenterology 2009;137:865-72.
22. Harrell FE Jr. Regression modeling strategies: with applica-
tions to linear models, logistic regression, and survival
analysis. New York: Springer; 2001.
23. Sun GW, Shook TL, Kay GL. Inappropriate use of bivari-
able analysis to screen risk factors for use in multivariable
analysis. J Clin Epidemiol 1996;49:907-16.
24. Royston P, Altman DG, Sauerbrei W. Dichotomizing con-
tinuous predictors in multiple regression: a bad idea. Stat
Med 2006;25:127-41.
25. Boersma E, Poldermans D, Bax JJ, Steyerberg EW, Thom-
son IR, Banga JD, et al. Predictors of cardiac events after
major vascular surgery: role of clinical characteristics, dobu-
tamine echocardiography, and beta-blocker therapy. JAMA
2001;285:1865-73.
26. Sauerbrei W. The use of resampling methods to simplify re-
gression models in medical statistics. J R Stat Soc Ser C
Appl Stat 1999;48:313-29.
27. Shmueli G. To explain or to predict? Stat Sci 2010:289-310.
28. Heikes KE, Eddy DM, Arondekar B, Schlessinger L. Diabe-
tes risk calculator: a simple tool for detecting undiagnosed
diabetes and pre-diabetes. Diabetes Care 2008;31:1040-5.
29. Breiman L, Friedman J, Stone CJ, Olshen RA. Classifica-
tion and regression trees. Belmont: Wadsworth Internation-
al Group; 1984.
29. 30. Breiman L. Statistical modeling: the two cultures (with
com-
ments and a rejoinder by the author). Statist Sci 2001;16:199-
231.
31. Steyerberg EW, Vickers AJ, Cook NR, Gerds T, Gonen M,
Obuchowski N, et al. Assessing the performance of predic-
tion models: a framework for traditional and novel mea-
sures. Epidemiology 2010;21:128-38.
32. Meffert PJ, Baumeister SE, Lerch MM, Mayerle J, Kratzer
W, Volzke H. Development, external validation, and com-
parative assessment of a new diagnostic score for hepatic
steatosis. Am J Gastroenterol 2014;109:1404-14.
33. Hilden J. Commentary: on NRI, IDI, and “good-looking”
sta-
tistics with nothing underneath. Epidemiology 2014;25:265-
7.
34. Lee YH, Bang H, Kim HC, Kim HM, Park SW, Kim DJ. A
simple screening score for diabetes for the Korean popula-
tion: development, validation, and comparison with other
scores. Diabetes Care 2012;35:1723-30.
35. Wyatt JC, Altman DG. Commentary: Prognostic models:
clinically useful or quickly forgotten? BMJ 1995;311:1539.
Review
www.thelancet.com Vol 395 May 16, 2020 1579
Artificial intelligence and the future of global health
30. Nina Schwalbe*, Brian Wahl*
Concurrent advances in information technology infrastructure
and mobile computing power in many low and
middle-income countries (LMICs) have raised hopes that
artificial intelligence (AI) might help to address challenges
unique to the field of global health and accelerate achievement
of the health-related sustainable development goals. A
series of fundamental questions have been raised about AI-
driven health interventions, and whether the tools,
methods, and protections traditionally used to make ethical and
evidence-based decisions about new technologies can
be applied to AI. Deployment of AI has already begun for a
broad range of health issues common to LMICs, with
interventions focused primarily on communicable diseases,
including tuberculosis and malaria. Types of AI vary, but
most use some form of machine learning or signal processing.
Several types of machine learning methods are
frequently used together, as is machine learning with other
approaches, most often signal processing. AI-driven
health interventions fit into four categories relevant to global
health researchers: (1) diagnosis, (2) patient morbidity
or mortality risk assessment, (3) disease outbreak prediction and
surveillance, and (4) health policy and planning.
However, much of the AI-driven intervention research in global
health does not describe ethical, regulatory, or
practical considerations required for widespread use or
deployment at scale. Despite the field remaining nascent,
AI-driven health interventions could lead to improved health
outcomes in LMICs. Although some challenges of
developing and deploying these interventions might not be
unique to these settings, the global health community will
need to work quickly to establish guidelines for development,
testing, and use, and develop a user-driven research
agenda to facilitate equitable and ethical use.
31. Introduction
AI is changing how health services are delivered in many
high-income settings, particularly in specialty care
(eg, radiology and pathology).1–3 This development has
been facilitated by the growing availability of large
datasets and novel analytical methods that rely on such
datasets. Concurrent advances in information technology
(IT) infrastructure and mobile computing power have
raised hopes that AI might also provide opportunities to
address health challenges in LMICs.4 These challenges,
including acute health workforce shortages and weak
public health surveillance systems, undermine global
progress towards achieving the health-related sustainable
development goals (SDGs).5,6 Although not unique to
such countries, these challenges are particularly relevant
given their contribution to morbidity and mortality.7,8
AI-driven health technologies could be used to address
many of these and other system-related challenges.4
For example, in some settings, AI-driven interventions
have supplemented clinical decision making towards
reducing the workload of health workers.9 New dev-
elopments in AI have also helped to identify disease
outbreaks earlier than traditional approaches, thereby
supporting more timely programme planning and
policy making.10 Although these interventions provide
promise, there remain several ethical, regulatory, and
practical issues that require guidance before scale-up
or widespread deployment in low and middle-income
settings.4
The global health community, including several large
donor agencies, has increasingly recognised the urgency
of addressing these issues towards ensuring that
populations in low and middle-income settings benefit
from developments in digital health and AI.11 Several
32. global meetings have taken place since 2015.12–14 For
example, in May, 2018, the World Health Assembly
adopted a resolution on digital technologies for universal
health coverage.15 In 2019, the United Nations Secretary
General’s High-Level Panel on Digital Cooperation
recommended that “by 2030, every adult should have
affordable access to digital networks, as well as digitally-
enabled financial and health services, as a means to
make a substantial contribution to achieving the SDGs”.16
Lancet 2020; 395: 1579–86
*Joint first authors
Heilbrunn Department of
Population and Family Health,
Columbia Mailman School of
Public Health, New York, NY,
USA (N Schwalbe MPH); Spark
Street Advisors, New York, NY,
USA (N Schwalbe, B Wahl PhD);
and Department of
International Health, Johns
Hopkins Bloomberg School of
Public Health, Baltimore, MD,
USA (B Wahl)
Correspondence to:
Nina Schwalbe, Columbia
Mailman School of Public Health,
New York, NY 10032, USA
[email protected]
Search strategy and selection criteria
We reviewed PubMed, MEDLINE, and Google Scholar.
33. This Review included peer-reviewed research articles
published in English between Jan 1, 2010, and Dec 31, 2019.
Relevant articles were identified using search terms that
included low and middle-income country names (appendix
pp 2–7) and “artificial intelligence”, “augmented intelligence”,
“computational intelligence”, and “machine learning”.
The titles and abstracts of identified articles were initially
reviewed by a study reviewer to assess whether the study was
done in a low-income or middle-income country, according
to the World Bank Atlas country classification method, and
focused on health or health system challenges that could be
addressed with artificial intelligence (AI) interventions.
We synthesised key themes and trends, using a previously
described classification for AI-driven health interventions
(ie, expert systems, machine learning, natural language
processing, automated planning and scheduling, and image
and signal processing) and broad categories of health
interventions (ie, diagnosis, risk assessment, disease outbreak
prediction and surveillance, and health policy and planning).
We excluded studies done in LMICs where AI might have been
used to develop a drug or diagnostic, but was not a central
component of the final health tool being studied.
http://crossmark.crossref.org/dialog/?doi=10.1016/S0140-
6736(20)30226-9&domain=pdf
Review
1580 www.thelancet.com Vol 395 May 16, 2020
In October, 2019, The Lancet and Financial Times
inaugurated a joint Commission focused on the
convergence of digital health, AI, and universal health
coverage.17 A report from this Commission is expected
in 2021.
34. In the context of these efforts to achieve the health-
related SDGs and ensure universal health coverage, we
aim to assess current AI research related to health in
LMICs. We identified the types of health issues being
addressed by AI, types of AI used in these interventions
(eg, machine learning, natural language processing,
signal processing), and whether there is sufficient
evidence that such interventions could improve health
outcomes in LMICs. In this Review we aim to highlight
additional research requirements, inform national and
global policy discussions, and support efforts to develop
a research and implementation agenda for AI in global
low-income and middle-income countries.
Current research on AI in LMICs
A full list of studies included in this narrative Review is
provided in the appendix (pp 8–11). AI interventions focus
on a broad range of health issues common to LMICs.
Most AI studies focused on communicable diseases,
including tuberculosis, malaria, dengue, and other
infectious diseases. Other AI studies focused on non-
infectious diseases in children and infants, preterm birth
complications, and malnutrition. Some interventions
aimed to address non-communicable diseases, including
cervical cancer. AI studies in LMICs addressed public
health from a broader perspective, particularly, health
policy and management. These studies include AI
research aimed at improving the performance of health
facilities, improving resource allocation from a systems
perspective, reducing traffic-related injuries, and other
health system issues.
The types of AI deployed in health research in LMICs
are described in the table. Most AI-driven health
interventions used some form of machine leaning or
35. signal processing, or both. Studies often evaluated
the use of machine learning together with other AI
approaches, most often with signal processing. In
addition, several types of machine learning methods
were frequently used together. For example, a common
approach used in machine learning and signal processing
was the use of convolutional neural networks for feature
extraction, and support-vector machines for classifi-
cation. A few research studies assessed interventions
based on natural language processing, data mining,
expert systems, or advanced planning.
AI-driven interventions for health
AI-driven health interventions broadly fit into four
categories described in the table. The automation or
support of diagnosis for communicable and non-com-
municable diseases emerged from studies as one of the
main uses of AI. Signal processing methods are often
used together with machine learning to automate the
diagnosis of communicable diseases. Signal processing
interventions focused specifically on the use of radiological
data for tuberculosis18,23 and drug-resistant tuberculosis,19
ultrasound data for pneumonia,24 micro scopy data for
malaria,25–27 and other biological sources of data for
tuberculosis.28–30 Most diagnostic interventions using AI in
LMICs reported either high sensitivity, specificity, or high
accuracy (>85% for all), or non-inferiority to comparator
diagnostic tools. Machine learning aids clinicians in
diagnosing tuberculosis,31 and expert systems are used
for diagnosing tuberculosis32 and malaria.27 Studies
mostly reported high diagnostic sensitivity, specificity, and
accuracy; however, at least one study reported low accuracy
when attempting to identify asymptomatic cases of
malaria.27
36. AI-driven interventions also focused on the diagnosis
of non-communicable diseases in LMICs, primarily
using signal processing methods for disease detection,
including cervical cancer and pre-cervical cancer using
microscopy,33–36 or data from photos of the cervix called
cervigrams.37 The accuracy has been reported to be
greater than 90%. One study aimed to evaluate a low-
cost, point-of-care oral cancer screening tool using cloud-
based signal processing and reported high sensitivity and
specificity relative to that of an onsite specialist.38
Morbidity and mortality risk assessment is another
area for which AI driven interventions have been
assessed in the global health context. These interventions
are based largely on machine learning classification tools
and typically compare multiple machine learning
approaches with the aim of identifying the optimal
approach to characterise risk. This approach has also
been used at health facilities to predict disease severity in
patients with dengue fever20 and malaria,39 and children
with acute infections.40 Researchers have used this
approach to quantify the risk of tuberculosis treatment
failure41 and assess the risk of cognitive sequelae after
malaria infection in children.42
See Online for appendix
Types of AI* Example
Diagnosis Expert system; machine learning;
natural language processing;
signal processing
Researchers applied machine learning and signal
processing methods to digital chest radiographs
to identify tuberculosis cases18 and drug-resistant
37. tuberculosis cases19
Mortality and
morbidity risk
assessment
Data mining; machine learning;
signal processing
To quantify the risk of dengue fever severity,
researchers applied machine learning algorithms
to administrative datasets from a large tertiary
care hospital in Thailand20
Disease outbreak
prediction and
surveillance
Data mining; machine learning;
natural language processing;
signal processing
Remote sensing data and machine learning
algorithms were used to characterise and predict
the transmission patterns of Zika virus globally21
Health policy and
planning
Expert planning; machine
learning
Machine learning models were applied to
administrative data from South Africa to predict
length of stay among health-care workers in
underserved communities22
38. AI=artificial intelligence. *Many types AI were implemented
together.
Table: Public health functions and associated types of AI
Review
www.thelancet.com Vol 395 May 16, 2020 1581
Machine learning classification tools were also used
to estimate the risk of non-infectious disease health
outcomes. For example, studies have focused on esti-
mating anaemia risk in children using standardised
household survey data,43 identifying children with the
greatest risk of missing immunisation sessions,44 and
detecting high-risk births using cardiotocography
data.45 A study from Brazil aimed to assess the
behavioural risk classification of sexually active teen-
agers.46 The reported accuracy of these tools ranged
from moderate (approximately 65%) to high (almost
99%).
Signal processing and machine learning have also been
used to estimate perinatal risk factors—eg, to automat-
ically estimate gestational age using data from ultrasound
images and other patient variables.47–49 Studies reported
high accuracy (>85%) relative to trained experts and other
standard gestational age estimation techniques.
Researchers are using AI for public health surveil-
lance to predict disease outbreak and evaluate disease
surveillance tools. Researchers have evaluated prediction
models using machine learning algorithms and remote
39. (ie, data collected by satellite or aircraft sensors) or local
(ie, data measured on site such as rainfall) sensing data to
estimate outbreaks of dengue virus. Although one study
reported high sensitivity and specificity for identifying
dengue outbreaks using a data-driven epidemiological
prediction method,50 other researchers51 found that
machine learning approaches for predicting dengue
outbreaks outperformed approaches based on linear
regression. Researchers have also used remote sensing
data and machine learning methods to predict malaria52,53
and Zika virus21 outbreaks with accuracy greater than
85%.
Another common approach to disease prediction and
surveillance is the use of machine learning and data
mining, together with data from online social media
networks and search engines. One study used this
approach to predict dengue outbreaks54 and other studies
to track and predict influenza outbreaks.55,56 All studies
reported high accuracy compared with observed data.
Social media data and machine learning using artificial
neural networks were also used to improve surveillance
of HIV in China.57
AI-driven health interventions can also be used to
support programme policy and planning. One such
study used data from a health facility in Brazil and an
agent-based simulation model to compare programme
options aimed at increasing the overall efficiency of the
health workforce.58 In another study, researchers used
several government datasets—including health system,
environmental, and financial data—together with
machine learning (ie, artificial neural networks) to
optimise the allocation of health system resources by
geography based on an array of prevalent health
challenges.59 Expert planning methods and household
40. survey data to optimise community health-worker visit
schedules were reported in the literature; however, no
results have yet been published.60
Additionally, AI methods aimed at informing pro-
gramme planning efforts within facilities have been
evaluated in low and middle-income settings. Some
examples include forecasting the number of outpatient
visits at an urban hospital61 and the length of health
-worker retention,22 using machine learning methods
and large administrative datasets from health facilities.
In another example, researchers used expert systems
and administrative data to design a system for measuring
the performance of hospital managers.62
Researchers are also using machine learning and data
mining methods to improve road safety in LMICs. In one
study, researchers used street imagery available online
and machine learning to estimate helmet use prev-
alence.63 In another study, a large government dataset of
road injuries and data mining techniques were used to
predict road injury severity.64
Accelerating access to AI
Numerous data are available to show how AI is being
tested to address health challenges relevant to the
achievement of SDGs. Such interventions include disease-
specific applications and those aimed at strengthening
health systems. Many AI health interventions have shown
promising preliminary results, and could soon be used to
augment existing strategies for delivering health services
in LMICs. Especially in disease diagnosis, where AI-
powered interventions could be used in countries with
insufficient numbers of health providers, and in risk
assessment, where tools based largely on machine
41. learning could help to supplement clinical knowledge.9
Although the research identified in this Review
indicates that AI-driven health interventions can help to
address several existing and emerging health challenges,
many issues are not sufficiently described in these
studies and warrant further exploration. These issues
relate to the development of AI-driven health inter-
ventions; how efficacy and effectiveness are assessed
and reported; planning for deployment at scale; and
the ethical, regulatory, and economic standards and
guidelines that will help to protect the interests of
communities in LMICs. Although these issues have
been described elsewhere,4,11,65–67 they have not been
systematically or explicitly addressed in research
published to date. We highlight these areas and suggest a
framework for consideration in future development,
testing, and deployment.
From development to deployment
One of the most important challenges facing AI in
LMICs relates to appropriate development and design.
Although none of the articles we reviewed here have
explained the impetus for project development, there are
most likely multiple reasons that explain why particular
health challenges in LMICs have been targeted by AI
Review
1582 www.thelancet.com Vol 395 May 16, 2020
developers. Communicable diseases—including malaria
and tuberculosis—continue to account for a pronounced
burden of disease in LMICs5 and attract substantial
42. donor funding.68 In addition, the characteristics of some
common health challenges in LMICs are able to be
addressed by AI—eg, the use of ultrasound data to
diagnose respiratory diseases and identify preterm birth
risk factors. The availability and portability of digital
ultrasound units and large datasets that can be used to
train AI algorithms (including in high-income settings),
have contributed to the development and testing of such
interventions in LMICs.
Although interventions such as those identified in this
Review might be beneficial, it is important that the
research agenda and development of interventions is
driven by local needs, health system constraints, and
disease burden rather than availability of data and
funding. A global research agenda for AI interventions
relevant to LMICs would help to ensure that new tools are
developed to respond to population needs. Step should
also be taken during the development of AI applications
to avoid ethnic, socioeconomic, and gender biases found
in some AI applications.
Another major challenge relates to comparative
performance of algorithms—including benchmarking
against any current standard care—and for continuously
assessing performance after deployment. Although
processes to enable benchmarking and assessment have
begun, including a collaboration between WHO and the
UN International Telecommunications Union (ITU),12,69
this type of testing will require adequate and representative
datasets from observational and surveillance studies,
electronic medical records, and social media platforms.
Open access to diverse datasets representing different
populations is particularly important, considering that
most AI-driven health interventions from the research
literature we identified are based on machine learning.
43. Enabling access across borders will require new types of
data sharing protocols and standards on inter-operability
and data labelling. This global movement could be
facilitated by an international collaboration so that data
are rapidly and equitably available for the development
and testing of AI-driven health interventions. Such
collaborations are already being developed in the UK by
initiatives such as the Health Data Research Alliance70
and the Confederation of Laboratories for Artificial
Intelligence Research in Europe.71
Reporting and methodological standards are also
required for AI health interventions in LMICs, particu-
larly those used for diagnostic tools. Although the
epidemiological and statistical methods used in studies
that we identified seem largely appropriate for the
research questions addressed, results were not reported
consistently. For example, some studies assessing diag-
nostic tools provide estimates of sensitivity, specificity,
and overall accuracy—ie, the probability of an individual
being correctly identified by a diagnostic test, which is
mathematically equivalent to a weighted average of the
sensitivity and specificity of the test. However, other
studies provided only a subset of these measurements.
The use of comparators was also inconsistently reported.
The Standards for Reporting of Diagnostic Accuracy
Studies72 provide guidelines for diagnostic assessments
and could be a starting place for standardising of
research in AI diagnostics.
None of the reviewed studies described whether
health technology assessments for an AI-driven health
intervention had been done. Standardised methods for
these assessments, including the extent to which these
interventions add value over current standards of care,
44. are urgently needed. Such methods should show how
well AI tools work outside study settings and highlight
related health system costs, including unintended
clinical, psychological, and social consequences. The
costs associated with false positive and false negative
results are also important to assess.
Although many studies reviewed here used statistical
methods that follow classic epidemiology methods,
basing their hypotheses on plausible models of causality,
some new AI-driven health interventions—particularly
those applying machine learning algorithms—identify
disease patterns and associations without a priori
hypotheses. Such approaches hold promise because they
are not necessarily affected by developer-introduced bias.
However, there remains a threat that false associations
could be identified and integrated into new AI-driven
health interventions.
The successful deployment of many AI-driven health
interventions will require investment to strengthen the
underlying health system. In addition to ethical concerns
related to diagnosing disease when treatment is not
available, the effectiveness of new diagnostic tools will
be limited if access to treatment is not expanded for all
patients. Similarly, tools that aim to predict outbreaks
and supplement surveillance would need to be supported
and complemented by robust surveillance systems to
guide an adequate public health emergency response if
an outbreak is accurately predicted.
Recommendations
Given the nascent stage of research on AI health
interventions in LMICs, global standards and guidelines
are needed to inform the development and evaluate
performance of tools in these settings. To support such
45. efforts, we provide several recommendations for research
and development of AI-driven health interventions in
low and middle-income settings using the AI application
value chain (figure).
Throughout the development and deployment phases,
we propose that researchers consider the principles for
digital development (panel).13 These principles provide
guidance on the best practice for development of digital
health technologies. Although none of the studies
reviewed here explicitly acknowledge digital principles,
Review
www.thelancet.com Vol 395 May 16, 2020 1583
we believe that they are helpful for development of
AI-driven health technologies. However, the digital
principles alone are insufficient. Institutional structures
also have an important role to play in the development
and deployment of new health technologies. Such
structures include appropriate regulatory and ethical
frameworks, benchmarking standards, pre-qualification
mechanisms, guidance on clinical and cost-effective
approaches, and frameworks for issues related to data
protection, in particular for children and youth, many of
whom now have a digital presence from birth. The
impact of AI tools on gender issues is another important
consideration and an area in which global guidance is
currently lacking.
AI does not need to be held to a higher standard of
research; however, its unique complexities, including the
requisite use of large datasets and the opaque nature of
46. some AI algorithms, will require approaches specifically
tailored to interventions and consideration of how efficacy
and effectiveness are assessed. Guidelines, such as those
from the EQUATOR network including the Transparent
Reporting of a Multivariable Prediction Model for
Individual Prognosis or Diagnosis—statement specific to
Machine Learning (TRIPOD-ML), Standard Protocol
Items: Recommendations for Interventional Trials
(SPIRIT)-AI, and Consolidated Standards of Reporting
Trials (CONSORT)-AI, that aim to harmonise termi-
nologies and reporting standards in prediction research,66
might help to guide researchers as they design and assess
AI interventions. Agencies in high-income countries,
including the US Food and Drug Administration, have
begun to develop separate regulatory pathways for
AI-driven health intereventions.67 In addition to the UN
ITU benchmarking initiative, WHO has recently created a
new digital health department and released new guidelines
on digital health.73 These efforts can help to provide
valuable insight for LMICs.
Current AI research highlights additional areas for
strengthening standards and guidelines for AI research
in LMICs. Although most AI investigators report neces-
sary approvals by institutional review boards, indicating
that the studies were all done ethically, only a few
described how the research teams addressed issues of
informed consent or ethical research design in tools that
used large datasets and electronic health records.
Reporting on ethical considerations would help future
researchers to address these complex yet essential issues.
Similarly, only a few studies reported on the usability
or acceptability of AI tools from the provider or patients’
perspective, despite acknowledging that usability is
an important factor for AI interventions, particularly
47. in LMICs. Human-centred design, an approach to
programme and product development frequently cited in
technology literature, considers human factors to ensure
that interactive systems are more usable. Human-
centered design is acknowledged as an important factor
for the development of new technologies in LMICs.65
There was also an absence of randomised clinical
trials (RCTs) identified in the literature. Clinical trials
help to establish clinical efficacy in LMICs. Given the
challenges associated with conducting RCTs for new
health technologies,74 new approaches such as the Idea,
Development, Exploration, Assessment, and Long Term
(IDEAL) follow-up framework75 recommended for the
evaluation of novel surgical practices, could serve to
provide relevant learning. This framework provides
guidance on clinical assessment for surgical inter-
ventions, in the context of challenges that make clinical
trials difficult, including variation in setting, disparities
in quality, and subjective interpretation.
There were only a few references to any type of
implementation research to assess questions related to
adoption or deployment at scale. Assessing implemen-
tation-related factors could help to identify potential
Figure: Recommendations for development of artificial
intelligence driven health applications in low and
middle-income countries
Research and development
• Incorporate human centred
design principles into
application development
48. • Ensure equitable access to
representative datasets
Assessment
• Standardise reporting of efficacy
and effectiveness
• Build consensus around
appropriate statistical and
epidemiological methods and
reporting
• Assess relative benefits over
current standard of care
Deployment
• Develop standards for health
technology assessments
• Encourage cost-effectiveness
and cost–benefit evaluations
• Conduct implementation and
systems-related research
• Do continuous assessments of
efficacy and effectiveness
User-driven research agenda aligned with digital principles
Statistical, ethical, and regulatory standards
Panel: Digital principles for artificial intelligence driven
interventions in global health
49. • User-centred design starts with getting to know the people you
are designing for by
conversation, observation, and co-creation
• Well designed initiatives and digital tools consider the
particular structures and needs
that exist in each country, region, and community
• Achieving a larger scale requires adoption beyond a pilot
population and often
necessitates securing funding or partners that take the initiative
to new communities
and regions
• Building sustainable programmes, platforms, and digital tools
is essential to maintain
user and stakeholder support, and to maximise long-term effect
• When an initiative is data driven, quality information is
available to the right people
when they need it, and those people will use data to act
• An open approach to digital development can help to increase
collaboration in the
digital development community and avoid duplicating work that
has already been done
• Reusing and improving is about taking the work of the global
development
community further than any organisation or programme can do
alone
• Addressing privacy and security in digital development
involves careful
consideration of which data are collected and how data are
50. acquired, used, stored,
and shared
• Being collaborative means sharing information, insights,
strategies, and resources
across projects, organisations, and sectors, leading to increased
efficiency and effect
Review
1584 www.thelancet.com Vol 395 May 16, 2020
unintended consequences at an individual and system
level of AI interventions. Further, there was no
description of the costs related to patients, providers, or
systems. A thorough assessment of these costs is crucial
to inform cost-effectiveness analyses and the potential
for scalability.
Limitations and conclusions
First, relevant articles might have been published before
2010. However, The field of AI, particularly in global
health, is rapidly evolving and any articles that were not
included as a result of being published before 2010 are
unlikely to be representative of this field as it is today. In
addition, our Review included only English-language
articles. Given the prominence of AI research around the
world, excluding articles published in languages other
than English could be a limitation.
As with all reviews, publication bias is another potential
limitation. There are two probable sources of this bias in
AI research. First, studies with null results are less likely
to be published.76 For that reason, AI-driven health
51. interventions that have not shown statistically significant
results might be under-represented in our literature
Review. Furthermore, investments in AI and health were
forecasted to have reached US$1∙7 billion in 2018,77 and
are increasingly dominated by private equity firms78 and
driven by so-called big tech companies such as Google
and Baidu ventures.79 Given that many interventions are
developed in the private sector for commercial use, some
AI developers might not place a high priority on
publishing the results in academic literature.80
AI is already being developed to address health issues in
LMICs. Current research is addressing a range of health
issues and using various AI-driven health interventions.
The breadth and promising results of these interventions
emphasise the urgency for the global community to act
and create guidance to facilitate deployment of effective
interventions. This point is particularly crucial given the
rapid deployment of AI-driven health interventions
which are being rolled out at scale as part of the severe
acute respiratory syndrome coronavirus 2 (SARS-CoV-2)
pandemic response. In many cases this roll-out is being
carried out without adequate evidence or appropriate
safeguards.
In accordance with our recommendations, the global
health community will need to work quickly to: incorporate
aspects of human-centred design into the development
process, including starting from a needs-based rather
than a tool-based approach; ensure rapid and equitable
access to representative datasets; establish global systems
for assessing and reporting efficacy and effectiveness of
AI-driven interventions in global health; develop a
research agenda that includes implementation and
system related questions on the deployment of new
AI-driven interventions; and develop and implement
52. global regulatory, economic, and ethical standards and
guidelines that safeguard the interests of LMICs. These
recommendations will ensure that AI helps to improve
health in low and middle-income settings and contributes
to the achievement of the SDGs, universal health
coverage, and to the coronavirus disease 2019 (COVID-19)
response.
Contributors
NS and BW are joint first authors. NS and BW reviewed the
literature
and wrote the manuscript.
Declaration of interests
We declare no competing interests.
Acknowledgments
Fondation Botnar funded the data collection and supported an
initial
synthesis of the literature which provided the basis for this
Review.
The funder had no role in study design, data collection, data
analysis,
data interpretation, writing of the report, or the decision to
submit for
publication. All authors had full access to all the data used in
the study
and the corresponding author had final responsibility for the
decision to
submit for publication.
References
1 Hosny A, Parmar C, Quackenbush J, Schwartz LH, Aerts
HJWL.
Artificial intelligence in radiology. Nat Rev Cancer 2018;
53. 18: 500–10.
2 Chang HY, Jung CK, Woo JI, et al. Artificial intelligence in
pathology. J Pathol Transl Med 2019; 53: 1–12.
3 Jha S, Topol EJ. Adapting to artificial intelligence:
radiologists and
pathologists as information specialists. JAMA 2016; 316: 2353–
54.
4 Wahl B, Cossy-Gantner A, Germann S, Schwalbe NR.
Artificial
intelligence (AI) and global health: how can AI contribute to
health in resource-poor settings? BMJ Glob Health 2018;
3: e000798.
5 Lozano R, Fullman N, Abate D, et al. Measuring progress
from 1990
to 2017 and projecting attainment to 2030 of the health-related
Sustainable Development Goals for 195 countries and
territories:
a systematic analysis for the Global Burden of Disease Study
2017.
Lancet 2018; 392: 2091–138.
6 Bartolomeos KK. The case for investing in public health
surveillance
in low- and middle-income countries. Afr J Emerg Med 2018; 8:
127–28.
7 Alkire BC, Peters AW, Shrime MG, Meara JG. The economic
consequences of mortality amenable to high-quality health care
in
low- and middle-income countries. Health Aff (Millwood) 2018;
37: 988–96.
54. 8 Kruk ME, Gage AD, Joseph NT, Danaei G, García-Saisó S,
Salomon JA. Mortality due to low-quality health systems in the
universal health coverage era: a systematic analysis of amenable
deaths in 137 countries. Lancet 2018; 392: 2203–12.
9 Guo J, Li B. The application of medical artificial intelligence
technology in rural areas of developing countries. Health Equity
2018; 2: 174–81.
10 Lake IR, Colón-González FJ, Barker GC, Morbey RA, Smith
GE,
Elliot AJ. Machine learning to refine decision making within a
syndromic surveillance service. BMC Public Health 2019; 19:
559.
11 USAID Center for Innovation and Impact. Artificial
intelligence in
global health: defining a collective path forward. 2019.
https://www.
usaid.gov/sites/default/files/documents/1864/AI-in-Global-
Health_
webFinal_508.pdf (accessed May 26, 2019).
12 International Telecommunication Union. AI for good global
summit. 2017. https://www.itu.int/en/ITU-T/AI/Pages/201706-
default.aspx (accessed May 28, 2019).
13 Digital Principles. Principles for digital development. 2017.
https://digitalprinciples.org/ (accessed May 26, 2019).
14 Digital Investment Principles. The principles of donor
alignment
for digital health. 2018. https://digitalinvestmentprinciples.org/
(accessed June 3, 2019).
15 WHO. Seventy-first world health assembly. Digital health.
55. 2018.
https://apps.who.int/gb/ebwha/pdf_files/WHA71/A71_R7-en.pdf
(accessed Nov 19, 2019).
16 UN Secretary-General’s High-level Panel on Digital
Cooperation.
The age of digital interdependence. 2018.
https://www.un.org/en/
pdfs/DigitalCooperation-report-for%20web.pdf (accessed
Nov 19, 2019).
Review
www.thelancet.com Vol 395 May 16, 2020 1585
17 Kickbusch I, Agrawal A, Jack A, Lee N, Horton R.
Governing health
futures 2030: growing up in a digital world-a joint The Lancet
and
Financial Times Commission. Lancet 2019; 394: 1309.
18 Lopes UK, Valiati JF. Pre-trained convolutional neural
networks as
feature extractors for tuberculosis detection. Comput Biol Med
2017;
89: 135–43.
19 Jaeger S, Juarez-Espinosa OH, Candemir S, et al. Detecting
drug-
resistant tuberculosis in chest radiographs. Int J CARS 2018;
13: 1915–25.
20 Phakhounthong K, Chaovalit P, Jittamala P, et al. Predicting
the
56. severity of dengue fever in children on admission based on
clinical
features and laboratory indicators: application of classification
tree
analysis. BMC Pediatr 2018; 18: 109.
21 Jiang D, Hao M, Ding F, Fu J, Li M. Mapping the
transmission risk
of Zika virus using machine learning models. Acta Trop 2018;
185: 391–99.
22 Moyo S, Doan TN, Yun JA, Tshuma N. Application of
machine
learning models in predicting length of stay among healthcare
workers in underserved communities in South Africa.
Hum Resour Health 2018; 16: 68.
23 Aguiar FS, Torres RC, Pinto JV, Kritski AL, Seixas JM,
Mello FC.
Development of two artificial neural network models to support
the
diagnosis of pulmonary tuberculosis in hospitalized patients in
Rio de Janeiro, Brazil. Med Biol Eng Comput 2016; 54: 1751–
59.
24 Correa M, Zimic M, Barrientos F, et al. Automatic
classification of
pediatric pneumonia based on lung ultrasound pattern
recognition.
PLoS One 2018; 13: e0206410.
25 Go T, Kim JH, Byeon H, Lee SJ. Machine learning-based in-
line
holographic sensing of unstained malaria-infected red blood
cells.
J Biophotonics 2018; 11: e201800101.
57. 26 Torres K, Bachman CM, Delahunt CB, et al. Automated
microscopy
for routine malaria diagnosis: a field comparison on Giemsa-
stained blood films in Peru. Malar J 2018; 17: 339.
27 Andrade BB, Reis-Filho A, Barros AM, et al. Towards a
precise test
for malaria diagnosis in the Brazilian Amazon: comparison
among
field microscopy, a rapid diagnostic test, nested PCR, and a
computational expert system based on artificial neural
networks.
Malar J 2010; 9: 117.
28 Khan S, Ullah R, Shahzad S, Anbreen N, Bilal M, Khan A.
Analysis of tuberculosis disease through Raman spectroscopy
and
machine learning. Photodiagn Photodyn Ther 2018; 24: 286–91.
29 Mohamed EI, Mohamed MA, Moustafa MH, et al. Qualitative
analysis of biological tuberculosis samples by an electronic
nose-
based artificial neural network. Int J Tuberc Lung Dis 2017; 21:
810–17.
30 Kuok CP, Horng MH, Liao YM, Chow NH, Sun YN. An
effective and
accurate identification system of Mycobacterium tuberculosis
using
convolution neural networks. Microsc Res Tech 2019; 82: 709–
19.
31 Elveren E, Yumuşak N. Tuberculosis disease diagnosis using
artificial neural network trained with genetic algorithm. J Med
Syst
58. 2011; 35: 329–32.
32 Osamor VC, Azeta AA, Ajulo OO. Tuberculosis-Diagnostic
Expert
System: an architecture for translating patients information
from
the web for use in tuberculosis diagnosis. Health Informatics J
2014;
20: 275–87.
33 Zhao M, Wu A, Song J, Sun X, Dong N. Automatic screening
of
cervical cells using block image processing. Biomed Eng Online
2016; 15: 14.
34 Chankong T, Theera-Umpon N, Auephanwiriyakul S.
Automatic cervical cell segmentation and classification in pap
smears. Comput Methods Programs Biomed 2014; 113: 539–56.
35 Kumar R, Srivastava R, Srivastava S. Detection and
classification of
cancer from microscopic biopsy images using clinically
significant
and biologically interpretable features. J Med Eng 2015;
2015: 457906.
36 Su J, Xu X, He Y, Song J. Automatic detection of cervical
cancer
cells by a two-level cascade classification system.
Anal Cell Pathol (Amst) 2016; 2016: 9535027.
37 Hu L, Bell D, Antani S, et al. An observational study of deep
learning and automated evaluation of cervical images for cancer
screening. J Natl Cancer Inst 2019; 111: 923–32.
38 Uthoff RD, Song B, Sunny S, et al. Point-of-care,
59. smartphone-based,
dual-modality, dual-view, oral cancer screening device with
neural
network classification for low-resource communities. PLoS One
2018; 13: e0207493.
39 Johnston IG, Hoffmann T, Greenbury SF, et al. Precision
identification of high-risk phenotypes and progression pathways
in
severe malaria without requiring longitudinal data. NPJ Digit
Med
2019; 2: 63.
40 Kwizera A, Kissoon N, Musa N, et al. A machine learning-
based
triage tool for children with acute infection in a low resource
setting. Pediatr Crit Care Med 2019; 20: e524–30.
41 Hussain OA, Junejo KN. Predicting treatment outcome of
drug-
susceptible tuberculosis patients using machine-learning
models.
Inform Health Soc Care 2019; 44: 135–51.
42 Veretennikova MA, Sikorskii A, Boivin MJ. Parameters of
stochastic
models for electroencephalogram data as biomarkers for child’s
neurodevelopment after cerebral malaria. J Stat Distrib Appl
2018;
5: 8.
43 Meena K, Tayal DK, Gupta V, Fatima A. Using classification
techniques for statistical analysis of Anemia. Artif Intell Med
2019;
94: 138–52.
60. 44 Chandir S, Siddiqi DA, Hussain OA, et al. Using predictive
analytics to identify children at high risk of defaulting from a
routine immunization program: feasibility study.
JMIR Public Health Surveill 2018; 4: e63.
45 Hoodbhoy Z, Noman M, Shafique A, Nasim A, Chowdhury
D,
Hasan B. Use of machine learning algorithms for prediction of
fetal
risk using cardiotocographic data. Int J Appl Basic Med Res
2019;
9: 226–30.
46 Waleska Simões P, Cesconetto S, Toniazzo de Abreu LL, et
al.
A data mining approach to identify sexuality patterns in a
Brazilian
university population. Stud Health Technol Inform 2015; 216:
1074.
47 van den Heuvel TLA, Petros H, Santini S, de Korte CL,
van Ginneken B. Automated fetal head detection and
circumference
estimation from free-hand ultrasound sweeps using deep
learning
in resource-limited countries. Ultrasound Med Biol 2019; 45:
773–85.
48 Rittenhouse KJ, Vwalika B, Keil A, et al. Improving preterm
newborn identification in low-resource settings with machine
learning. PLoS One 2019; 14: e0198919.
49 Papageorghiou AT, Kemp B, Stones W, et al. Ultrasound-
based
gestational-age estimation in late pregnancy.
Ultrasound Obstet Gynecol 2016; 48: 719–26.
61. 50 Buczak AL, Koshute PT, Babin SM, Feighner BH, Lewis SH.
A data-driven epidemiological prediction method for dengue
outbreaks using local and remote sensing data.
BMC Med Inform Decis Mak 2012; 12: 124.
51 Scavuzzo JM, Trucco F, Espinosa M, et al. Modeling Dengue
vector
population using remotely sensed data and machine learning.
Acta Trop 2018; 185: 167–75.
52 Haddawy P, Hasan AHMI, Kasantikul R, et al.
Spatiotemporal
Bayesian networks for malaria prediction. Artif Intell Med
2018;
84: 127–38.
53 Kabaria CW, Molteni F, Mandike R, et al. Mapping intra-
urban
malaria risk using high resolution satellite imagery: a case
study of
Dar es Salaam. Int J Health Geogr 2016; 15: 26.
54 Althouse BM, Ng YY, Cummings DA. Prediction of dengue
incidence using search query surveillance. PLoS Negl Trop Dis
2011;
5: e1258.
55 Xu Q, Gel YR, Ramirez Ramirez LL, Nezafati K, Zhang Q,
Tsui KL.
Forecasting influenza in Hong Kong with Google search queries
and statistical model fusion. PLoS One 2017; 12: e0176690.
56 Clemente L, Lu F, Santillana M. Improved real-time
influenza
surveillance: using internet search data in eight Latin American
62. countries. JMIR Public Health Surveill 2019; 5: e12214.
57 Nan Y, Gao Y. A machine learning method to monitor
China’s AIDS
epidemics with data from Baidu trends. PLoS One 2018;
13: e0199697.
58 Yousefi M, Yousefi M, Ferreira RPM, Kim JH, Fogliatto FS.
Chaotic genetic algorithm and Adaboost ensemble
metamodeling
approach for optimum resource planning in emergency
departments. Artif Intell Med 2018; 84: 23–33.
59 Rosas MA, Bezerra AF, Duarte-Neto PJ. Use of artificial
neural
networks in applying methodology for allocating health
resources.
Rev Saude Publica 2013; 47: 128–36.
60 Brunskill E, Lesh N. Routing for rural health: optimizing
community health worker visit schedules. Association for the
Advancement of Artificial Intelligence. 2010.
https://www.cs.cmu.
edu/~ebrun/brunskilllesh.pdf (accessed June 3, 2019).
Review
1586 www.thelancet.com Vol 395 May 16, 2020
61 Huang D, Wu Z. Forecasting outpatient visits using
empirical mode
decomposition coupled with back-propagation artificial neural
networks optimized by particle swarm optimization. PLoS One
2017;
63. 12: e0172539.
62 Shafii M, Hosseini SM, Arab M, Asgharizadeh E,
Farzianpour F.
Performance analysis of hospital managers using fuzzy AHP and
fuzzy TOPSIS: Iranian experience. Glob J Health Sci 2015; 8:
137–55.
63 Merali HS, Lin LY, Li Q, Bhalla K. Using street imagery and
crowdsourcing internet marketplaces to measure motorcycle
helmet use in Bangkok, Thailand. Inj Prev 2019; published
online
March 4. DOI:10.1136/injuryprev-2018-043061.
64 Beshah T, Hill S. Mining road traffic accident data to
improve
safety: role of road-related factors on accident severity in
Ethiopia.
2010. AAAI Spring Symposium Series.
65 The Lancet. Artificial intelligence in global health: a brave
new
world. Lancet 2019; 393: 1478.
66 Collins GS, Moons KGM. Reporting of artificial intelligence
prediction models. Lancet 2019; 393: 1577–79.
67 He J, Baxter SL, Xu J, Xu J, Zhou X, Zhang K. The practical
implementation of artificial intelligence technologies in
medicine.
Nat Med 2019; 25: 30–36.
68 Chapman N, Doubell A, Oversteegen L, et al. Neglected
disease
research and development: reaching new heights. G-FINDER
Report. 2018.
64. https://www.vaccineresources.org/details.php?i=2597
(accessed June 3, 2019).
69 Wiegand T, Krishnamurthy R, Kuglitsch M, et al. WHO and
ITU
establish benchmarking process for artificial intelligence in
health.
Lancet 2019; 394: 9–11.
70 Health Data Research UK. UK health data research alliance.
2019.
https://www.hdruk.ac.uk/infrastructure/uk-health-data-research-
alliance/ (accessed May 28, 2018).
71 Confederation of Laboratories for Artificial Intelligence
Research in
Europe. A European vision for AI. https://claire-
ai.org/?lang=de.
(accessed May 30, 2019).
72 Cohen JF, Korevaar DA, Altman DG, et al. STARD 2015
guidelines
for reporting diagnostic accuracy studies: explanation and
elaboration. BMJ Open 2016; 6: e012799.
73 WHO. WHO guideline recommendations on digital
interventions
for health system strengthening. 2019. https://apps.who.int/iris/
bitstream/handle/10665/311941/9789241550505-eng.pdf?ua=1
(accessed Dec 2, 2019).
74 Lorenzo C, Garrafa V, Solbakk JH, Vidal S. Hidden risks
associated
with clinical trials in developing countries. J Med Ethics 2010;
36: 111–15.
65. 75 McCulloch P, Altman DG, Campbell WB, et al. No surgical
innovation without evaluation: the IDEAL recommendations.
Lancet
2009; 374: 1105–12.
76 Easterbrook PJ, Berlin JA, Gopalan R, Matthews DR.
Publication
bias in clinical research. Lancet 1991; 337: 867–72.
77 Franklin Templeton. Artificial intelligence: real opportunity.
2019.
https://www.franklintempleton.com/investor/
article?contentPath=html/ftthinks/en-us-
retail/featured/artificial-
intelligence-real-opportunity.html (accessed June 3, 2019)
78 OECD. Private Equity Investment in Artificial Intelligence.
2018.
https://www.oecd.org/going-digital/ai/private-equity-
investment-in-
artificial-intelligence.pdf (accessed June 3, 2019).
79 Hui M, Lahiri T. This Chinese tech giant was 2018’s most
active
corporate investor in AI startups. Quartz. 2019. https://qz.com/
1555973/baidu-ventures-was-2018s-most-active-investor-in-ai/
(accessed June 2, 2019).
80 Breschi S, Lassébie J, Menon C. A portrait of innovative
start-ups
across countries. OECD. 2018. https://www.oecd-ilibrary.org/
docserver/f9ff02f4-
en.f?expires=1580483659&id=id&accname=guest&
checksum=3BD05CBBFA0446D064FFFA10A059C23D
(accessed
June 5, 2019).
68. .
CHAPTER 2
Health Determinants,
Measurements, and the Status
of Health Globally
LEARNING OBJECTIVES
By the end of this chapter, the reader will be
able to do the
following:
■ Describe the determinants of health
■ Define the most important health
indicators and key terms
related to measuringhealth status and the burden
of
disease
■ Discuss the status of health
globally and how it varies by
country income group, region, and age group
Skolnik, Richard. Global Health 101, Jones & Bartlett Learning,
LLC, 2019. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/indianatech-
ebooks/detail.action?docID=5894023.
Created from indianatech-ebooks on 2022-09-10 00:46:19.
C
op
70. ht
s
re
se
rv
ed
.
M
▶ Vignettes
aria is a poor woman who lives in the highlands of
Peru.
She is from an ethnic group called Quechua. In
Peru, poor
people tend to live in the mountainsand be
indigenous, be less
educated, and have worse health status than other
people. In
Eastern Europe, the same issues occur among
ethnic groups that
are of lower socioeconomic status, such as the
Roma people. In
the United States, thereare also enormous health
disparities, as
seen in the health status of African Americans
and Native
Americans, compared to white Americans. If
we want to
understand and address differences in health
71. status among
different groups, how do we measure health status?
Do we
measure it by age? By gender? By socioeconomic
status? By level
of education? By ethnicity? By location?
Yevgeny is a 56-year-old Russian male. Life
expectancy in Russia
in 1985 was about 64 years for males and 74
years for females. It
then fell to about 59 years for males and 72
years for females in
2001, before rising again to 67 for males
and 77 for females in
2016. What does life expectancy at birth
measure? What are the
factors contributing to the earlier decline in
life expectancy at birth
in Russia? What has happened to trends in
life expectancy in
othercountries?Which countries have the longest and
shortest life
expectancies, and why?
Sarah is a 27-year-old woman in northern
Nigeria. While women in
high-income countries very rarely die of pregnancy-
related causes
and have a maternal mortality ratio of about 10
per 100,000 live
births, the maternal mortality ratio for women in
low-income
countries like Sarah is about 500 per 100,000
live births. This is
50 times higher than that in the best-off country
78. rv
ed
.
▶ The Determinants and Social
Determinants of Health
Why are somepeople healthy and somepeople not
healthy?
When asked this question, many of us will
respond that good
health depends on access to health services.
Yet, as you will
learn, whether or not people are healthy
depends on a large
number of factors, many of which are
interconnected, and most of
which go considerably beyond access to health
services.
The World Health Organization (WHO) defines
the determinants
of health as the “range of personal, social,
economic and
environmental factors which determine the health
status of
individuals or populations.” WHO defines the
social determinants
of health as the “conditions in which people
are born, grow, live,
work and age.”
There has been considerable writing about the
79. determinants and
social determinants of health, which different
organizations depict
in a range of ways. The next section builds
on the work of a
number of actors and agencies. It briefly
discusses the
determinants and social determinants of health
and how they
influence health. It is essential to understand
theseconcepts if one
wants to understand why people are healthy or
not and what can
be done to address different health conditions in
different settings.
FIGURE 2-1 shows one way of depicting the
determinants of
health.
6
7
Skolnik, Richard. Global Health 101, Jones & Bartlett Learning,
LLC, 2019. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/indianatech-
ebooks/detail.action?docID=5894023.
Created from indianatech-ebooks on 2022-09-10 00:46:19.
C
op
yr
ig
ht
81. se
rv
ed
.
FIGURE 2-1 The Determinants of Health
Reproduced from Dahlgren, G., & Whitehead, M.
(1991). Policies and strategies to
promote social equity in health. Stockholm,
Sweden: Institute for Futures Studies.
Retrieved from
http://www.iffs.se/media/1326/20080109110739filmZ8UVQv2w
QFShMRF6cuT.pdf
The first group of factors that helps to
determine health relates to
the personal and inborn features of individuals.
These include
genetic makeup, sex, and age. Our genetic makeup
contributes to
what diseases we get and how healthy we are. One
can inherit, for
example, a genetic marker for a particular disease,
such as
Huntington’s disease, which is a neurological
disorder. One can
also inherit the genetic component of a disease
that has multiple
84. breast cancers. For
similar reasons, age is also an important determinant
of health.
Young children in low- and middle-income countries
oftendie of
diarrheal disease, whereas olderpeople are much
more likely to
die of heartdisease, to cite one of many
examples of the
relationship between health and age.
Individuallifestyle factors, including people’s own health
practices
and behaviors, are also important determinants of
health. Being
able to identify when you or a family member
is ill and needs
health care can be critical to good health. One’s
health also
depends greatly on how one eats, or if one smokes
tobacco,
drinks too much alcohol, or drives safely.
We also know that being
active physicallyand getting exercise regularly is
better for one’s
health than is being sedentary.
The extent to which people receive social
support from family,
friends, and community also has an important link
with health.
The stronger the social networks and the stronger
the support that
people get from those networks, the healthier people
will be. Of
course, culture is also an extremely important
85. determinant of
health.
Living and working conditions also exertan
enormous influence on
health. These include, for example, housing, access
to safe water
and sanitation, access to nutritious food, and
access to health
services. Crowded housing, for example, is a risk
factor for the
transmission of tuberculosis. The lack of safe water
and sanitation,
coupled with poor hygiene in many settings, is
one of the major
risk factors for the diarrheal disease that is
associatedwith so
much illness and death in young children.
Nutrition is central to
health, beginning at conception, and families have to
be able to
8
9
Skolnik, Richard. Global Health 101, Jones & Bartlett Learning,
LLC, 2019. ProQuest Ebook Central,
http://ebookcentral.proquest.com/lib/indianatech-
ebooks/detail.action?docID=5894023.
Created from indianatech-ebooks on 2022-09-10 00:46:19.
C
op
yr