Health Policy Brief

March 2016

Prediabetes in California: Nearly Half

of California Adults on Path to Diabetes

Susan H. Babey, Joelle Wolstein, Allison L. Diamant, Harold Goldstein

SUMMARY: In California, more than 13 million

adults (46 percent of all adults in the state) are

estimated to have prediabetes or undiagnosed

diabetes. An additional 2.5 million adults have

diagnosed diabetes. Altogether, 15.5 million

adults (55 percent of all California adults) have

prediabetes or diabetes. Although rates of

prediabetes increase with age, rates are also high

among young adults, with one-third of those ages

18-39 having prediabetes. In addition, rates of

prediabetes are disproportionately high among

young adults of color, with more than one-third

of Latino, Paciﬁc Islander, American Indian,

African-American, and multiracial Californians

ages 18-39 estimated to have prediabetes. Policy

efforts should focus on reducing the burden of

prediabetes and diabetes through support for

prevention and treatment.

iabetes, particularly type 2 diabetes,

is a signiﬁcant and growing health

problem that affects both adults and children

and can cause a number of serious complications,

including blindness, kidney disease,

cardiovascular disease, amputation, and

premature death. Nationally, the prevalence

of diabetes among adults has nearly tripled

over the past 30 years.

In 2014, 29.1 million

people in the U.S., or 9.3 percent of the

population, had diabetes (including 8.1 million

with undiagnosed diabetes).

In California,

the prevalence of diabetes among adults

increased by 35 percent between 2001

and 2012.

Prediabetes, also referred to as impaired

glucose tolerance or impaired fasting glucose,

is a condition in which blood glucose

levels are higher than normal but not high

enough for a diagnosis of diabetes. People

with prediabetes have a much higher risk

of developing type 2 diabetes, as well as

an increased risk for cardiovascular disease.

Results from the Diabetes Prevention

Program (DPP) clinical trial indicated that

among those with prediabetes, increased

physical activity, improvements in diet, and

weight loss can prevent or delay the onset

of diabetes signiﬁcantly more than placebo

or medication.

Results also indicated that

medication, while effective, is not as effective

as lifestyle changes.

Nationally, more than one in three adults

is estimated to have prediabetes, and 90

percent of these individuals are not aware

that they have the condition.

Between 1999

and 2010, the prevalence of prediabetes

among adults in the U.S. increased from 29

percent to 36 percent.

Moreover, between

1999 and 2008, the prevalence of diabetes

and prediabetes among adolescents in the

U.S. rose dramatically, from 9 percent to

23 percent.

Without intervention efforts,

up to 30 percent of people with prediabetes

will develop type 2 diabetes within ﬁve

years, and up to 70 percent will develop

diabetes within their lifetime.

There are very

effective interventions available, including

lifestyle modiﬁcation programs recognized

by the CDC’s National Diabetes Prevention

This policy brief was developed

in partnership with the California

Center for Public Health Advocacy

with funding from the California

Health Care Foundation and

The California Endowment

‘‘

More than

13 million

California adults

—nearly half of

the state’s adult

population—are

estimated to have

prediabetes.

’’

UCLA CENTER FOR HEALTH POLICY RESEARCH

Program, that can prevent or delay the

progression from prediabetes to diabetes.

The current trends in diabetes and prediabetes

are troubling because of the associated

human and ﬁnancial costs. Not only does

diabetes increase the risk of serious medical

complications, but it is also extremely costly

to families, businesses, health care plans,

states, and the nation. Nationally, diabetes

was estimated to cost $245 billion in 2012,

including $176 billion in direct medical

costs and $69 billion in lost productivity.

In California, the total cost of diabetes was

estimated to be more than $27 billion, with

$19 billion of that spent on direct medical

care for diabetes and $8 billion on the indirect

costs associated with the disease.

In addition,

undiagnosed diabetes is estimated to cost

California $2.8 billion and prediabetes $5.3

billion in direct medical care.

This study used data from the 2013-14

California Health Interview Survey (CHIS)

and the National Health and Nutrition

Examination Survey (NHANES) to estimate

the prevalence of prediabetes in California.

NHANES 2009-2012 data were used to

build and test a statistical model predicting

prediabetes, deﬁned by hemoglobin A1c and

fasting plasma glucose (blood tests commonly

used to diagnose diabetes and prediabetes).

This predictive model was then applied to

CHIS data to produce California-speciﬁc

estimates of the prevalence of prediabetes

and undiagnosed diabetes (herein referred

to as prediabetes when reporting California

estimates). The percentage of California adults

with undiagnosed diabetes is expected to

comprise a relatively small proportion of the

prediabetes estimates presented. Nationally,

less than 4 percent of adults have undiagnosed

diabetes. This policy brief describes the

estimated prevalence of prediabetes, including

undiagnosed diabetes, statewide as well as by

age, race and ethnicity, and county.

Prediabetes in California

One-Third of Young Adults in California

Have Prediabetes

In California, more than half of adults (55

percent) have either prediabetes or diabetes.

This includes 2.5 million adults, or 9 percent of

the state’s adult population, who have diagnosed

diabetes. In addition, nearly half of adults (46

percent) are estimated to have prediabetes.

This represents more than 13 million California

adults. Prediabetes prevalence increases with

age, rising from 33 percent among adults ages

18-39 to 49 percent among those ages 40-

59 (Exhibit 1). Prevalence then levels off at

approximately 60 percent among adults 55

and older.

Percent of Adults Diagnosed with Diabetes and Estimated to Have Prediabetes by

Age Group, California, 2013-14

Exhibit 1

Diabetes Prediabetes

Age % %

18-39 2% 33%

40-54 9% 49%

55-69 16% 60%

70+ 20% 59%

All California Adults 9% 46%

Source: 2013-14 California Health Interview Survey

Note: Estimates of prediabetes are based on predictive models

developed using 2009-2012 NHANES data and applied to

CHIS 2013-14 data. Prediabetes estimates include adults

with undiagnosed diabetes. Nationally, approximately 3.9 percent of

adults have undiagnosed diabetes. Conﬁdence intervals for estimates

presented in this table are available here: http://healthpolicy.ucla.edu/

publications/search/pages/detail.aspx?PubID=1472.

‘‘

Not only does

diabetes increase

the risk of

serious medical

complications,

but it is also

extremely costly

to families,

businesses, health

care plans, states,

and the nation.

’’

UCLA CENTER FOR HEALTH POLICY RESEARCH

Prediabetes Higher Among Adults of Color

Prediabetes disproportionately affects certain

racial and ethnic groups. In California, at

least half of Paciﬁc Islanders (55 percent),

American Indians (51 percent), and African-

Americans (50 percent) are estimated to

have prediabetes (Exhibit 2). Among young

adults, more than one-third of Latinos

(36 percent), Paciﬁc Islanders (43 percent),

American Indians (38 percent), African-

Americans (38 percent), and those of

multiple races (37 percent) are estimated to

have prediabetes.

Prediabetes Varies by County

The prevalence of prediabetes varies from

county to county among California adults.

Because age is a particularly strong risk

factor for diabetes and prediabetes, Exhibit 3

displays estimates of county-level prediabetes

prevalence broken out by age group. High

rates among young adults are particularly

concerning, because the risk of complications

from diabetes increases signiﬁcantly the

longer one has the condition. Among adults

ages 18-39, the prevalence of prediabetes

ranged from 26 percent in Lake County to 40

percent in both Kings and Imperial counties

(Exhibit 3). Among this younger age group,

ﬁve counties had rates below 30 percent

(Lake, San Benito, Butte, San Francisco, and

San Luis Obispo), and ﬁve had rates over 37

percent (Tulare, Merced, San Joaquin, Kings,

and Imperial). Among all adults, rates ranged

from 43 percent in Sutter and Butte counties

to 54 percent in Nevada County and the

combined counties of Tuolumne, Calaveras,

Amador, Inyo, Mariposa, Mono, and Alpine.

This regional variation is likely due to a

number of factors, including differences

in demographic, social, economic, and

environmental characteristics.

Percent of Adults Estimated to Have Prediabetes by Race or Ethnicity and Age Group,

California, 2013-14

Exhibit 2

Source: 2013-14 California Health Interview Survey

Note: Estimates of prediabetes are based on predictive models

developed using 2009-2012 NHANES data and applied to

CHIS 2013-14 data. Prediabetes estimates include adults

Age Group

Race and Ethnicity 18-39 40-54 55-69 70+ All Adults

Latino 36% 51% 55% 51% 44%

Paciﬁc Islander 43% 54% 76% 53% 55%

American Indian 38% 52% 65% 70% 51%

Asian 31% 45% 53% 58% 42%

African-American 38% 56% 61% 57% 50%

White 29% 49% 63% 61% 48%

Multiracial 37% 51% 58% 52% 45%

California 33% 49% 60% 59% 46%

with undiagnosed diabetes (approximately 3.9 percent of

adults nationally). Conﬁdence intervals for estimates presented

in this table are available here: http://healthpolicy.ucla.edu/

publications/search/pages/detail.aspx?PubID=1472.

‘‘

High rates

among young

adults are

particularly

concerning,

because the risk

of complications

from diabetes

increases

signiﬁcantly the

longer one has

the condition.

’’

UCLA CENTER FOR HEALTH POLICY RESEARCH

Percent of Adults Estimated to Have Prediabetes by County or County Group and Age,

California, 2013-14

Exhibit 2

Source: 2013-14 California Health Interview Survey

Note: Estimates of prediabetes are based on predictive models

developed using 2009-2012 NHANES data and applied

to CHIS 2013-14 data. Prediabetes estimates include

Age Group

County or County Group 18-39 40-54 55-69 70+ All Adults

Northern and Sierra Counties 31% 50% 61% 60% 48%

Butte 28% 52% 62% 53% 43%

Shasta 30% 52% 62% 54% 50%

Humboldt 32% 47% 67% 67% 48%

Del Norte, Siskiyou, Lassen, Trinity, Modoc, Plumas, Sierra 32% 48% 64% 63% 49%

Mendocino 30% 44% 66% 65% 48%

Lake 26% 43% 58% 58% 46%

Tehama, Glenn, Colusa 34% 58% 47% 59% 46%

Sutter 32% 51% 48% 58% 43%

Yuba 33% 55% 58% 57% 48%

Nevada 33% 46% 66% 71% 54%

Tuolumne, Calaveras, Amador, Inyo, Mariposa, Mono, Alpine 36% 53% 64% 60% 54%

Greater Bay Area 32% 48% 62% 62% 47%

Santa Clara 32% 43% 62% 65% 46%

Alameda 34% 51% 58% 64% 47%

Contra Costa 33% 44% 61% 62% 47%

San Francisco 28% 51% 66% 55% 44%

San Mateo 31% 48% 67% 65% 47%

Sonoma 33% 53% 61% 60% 49%

Solano 32% 48% 61% 50% 45%

Marin 31% 48% 61% 67% 50%

Napa 33% 48% 66% 65% 48%

Sacramento Area 31% 50% 63% 60% 47%

Sacramento 31% 51% 63% 58% 46%

Placer 31% 47% 61% 63% 47%

Yolo 32% 51% 59% 57% 44%

El Dorado 32% 49% 67% 62% 50%

San Joaquin Valley 36% 50% 60% 57% 47%

Fresno 37% 45% 68% 65% 49%

Kern 34% 58% 51% 49% 45%

San Joaquin 39% 46% 67% 58% 48%

Stanislaus 34% 54% 58% 52% 45%

Tulare 38% 41% 56% 56% 44%

Merced 38% 55% 51% 55% 46%

Kings 40% 49% 58% 60% 48%

Madera 32% 55% 63% 49% 45%

Central Coast 33% 51% 61% 58% 46%

Ventura 32% 53% 59% 61% 47%

Santa Barbara 33% 50% 64% 56% 47%

Santa Cruz 30% 45% 66% 61% 45%

San Luis Obispo 29% 52% 63% 57% 46%

Monterey 37% 48% 54% 50% 45%

San Benito 27% 53% 58% 62% 47%

Los Angeles 33% 48% 57% 56% 44%

Other Southern California 33% 51% 60% 61% 46%

Orange 31% 49% 62% 61% 46%

San Diego 32% 50% 62% 59% 46%

San Bernardino 35% 51% 52% 64% 45%

Riverside 34% 54% 63% 62% 48%

Imperial 40% 53% 43% 41% 44%

California 33% 49% 60% 59% 46%

adults with undiagnosed diabetes (approximately 3.9 percent of

adults nationally). Conﬁdence intervals for estimates presented in

this table are available here: http://healthpolicy.ucla.edu/publications/

search/pages/detail.aspx?PubID=1472.

UCLA CENTER FOR HEALTH POLICY RESEARCH

Conclusions and Recommendations

More than 13 million California adults —

nearly half of the state’s adult population

—are estimated to have prediabetes. This

suggests that more effort is needed to address

the prevention of diabetes and the detection

of and intervention for prediabetes statewide.

Health promotion and disease prevention

efforts such as maintaining a healthy weight,

consuming healthy foods and beverages,

limiting intake of sugar and other simple

carbohydrates, and being more physically

active all reduce the risk of developing type 2

diabetes. To aid in the prevention of diabetes,

particularly among those with prediabetes,

policymakers should consider the following:

• Support diabetes prevention efforts. Most

people with prediabetes do not know they

have the condition. Providing coverage for

and ensuring the regular medical practice

of appropriate screening can identify

people with prediabetes while it is still

possible to prevent the onset of type 2

diabetes. In addition, insurance coverage

for and referral to recognized diabetes

prevention programs can remove critical

barriers to education and care for people

with prediabetes and can facilitate lifestyle

changes that can prevent diabetes.

• Promote community and workplace

environments that support healthy

eating. Local and state policy initiatives

can improve the food and beverage

environment by increasing access to fruits

and vegetables, decreasing marketing

of unhealthy options, encouraging large

institutions such as hospitals to follow

healthy food procurement guidelines,

developing educational strategies to assist

consumers in making more informed food

and beverage choices, and ensuring the

availability of safe and low-cost drinking

water.

• Promote built environments that

encourage regular physical activity.

Lack of physical activity is a signiﬁcant

risk factor for diabetes, and further

policies should be developed to facilitate

active living—for example, creating safe

environments for walking and biking,

providing access to safe parks and

other places for recreation and physical

activity, and offering worksite programs

to facilitate regular physical activity for

adults of all ages.

• Support adequate access to quality

primary and specialty care. At-risk

individuals need to have adequate and

sufﬁcient access to quality health care

services. Lack of continuous health

insurance coverage and insufﬁcient beneﬁts

packages create signiﬁcant ﬁnancial

barriers to accessing primary and specialty

care services. In addition, increased access

to recognized diabetes-prevention lifestyle

modiﬁcation programs has been shown to

be particularly beneﬁcial for adults with

prediabetes.

‘‘

Most people with

prediabetes do not

know they have

the condition.

’’

UCLA CENTER FOR HEALTH POLICY RESEARCH

This publication contains

data from the California

Health Interview Survey

(CHIS), the nation’s largest

state health survey.

Conducted by the UCLA

Center for Health Policy

Research, CHIS data give

a detailed picture of the

health and health care

needs of California’s large

and diverse population.

Learn more at:

www.chis.ucla.edu

Data Sources and Methods

The ﬁndings in this brief are based on data from the

2013-14 California Health Interview Survey (CHIS).

CHIS 2013-14 completed interviews with more

than 40,000 households that included 40,240 adults,

drawn from every county in the state. Interviews

were conducted in English, Spanish, Chinese (both

Mandarin and Cantonese), Korean, Vietnamese, and

Tagalog. California estimates of diabetes prevalence

are based on self-report. Adults were asked whether

they had ever been diagnosed with diabetes by a

doctor. Those who responded “yes” were classiﬁed as

having diabetes.

Estimates of prediabetes are statistically modeled.

Data from the 2009-2012 National Health and

Nutrition Examination Survey (NHANES)

were used to build and test predictive models of

blood glucose levels above cutoffs associated with

prediabetes. NHANES is a cross-sectional survey

that provides a nationally representative sample

of the noninstitutionalized population. NHANES

participants completed a household interview as

well as a physical examination that included a blood

sample. Predictive models were developed for the

adult population (18 and older) using data from the

NHANES fasting subsample. Cutoffs associated

with prediabetes were applied to hemoglobin A1c

(HbA1c) and fasting plasma glucose (FPG) values in

NHANES: HbA1c of 5.7 percent or above, or FPG

of 100 or above. People who reported having been

diagnosed with diabetes were classiﬁed as having

diabetes.

The predictive model was developed using

Generalized Boosted Regression Models (GBM)

implemented in R.

This iterative, machine-

learning algorithm increases in complexity until it

minimizes out of training-sample predictive error,

which was assessed using tenfold cross-validation.

The NHANES predictive model displayed good

predictive ability: Pseudo R-squared = 0.304 and

Coefﬁcient of Discrimination = 0.301. These metrics

are taken from the cross-validation and represent the

prediction for cases not used in the training. Models

predicted blood glucose levels above prediabetes

cutoffs. As a result, estimates of prediabetes include

adults with undiagnosed diabetes. However, those

with undiagnosed diabetes are expected to represent

a relatively small proportion of the prediabetes

estimates presented here. Variance was estimated

using multiple imputation. Conﬁdence intervals for

estimates presented in this publication are available

here: http://healthpolicy.ucla.edu/publications/search/pages/

detail.aspx?PubID=1472.

For consistency with earlier estimates, the National

Center for Health Statistics applies regression

equations to fasting glucose values collected

after 2005. The current analysis does not involve

comparison with earlier estimates. Therefore, fasting

glucose values are based on the current laboratory

measurement methods and have not been adjusted

to be comparable to values collected in previous

NHANES cycles. Based on our analysis of 2009-2012

NHANES data using HbA1c and FPG values not

adjusted for comparability with earlier NHANES

cycles, approximately 42 percent of U.S. adults 18

and over have prediabetes, and an additional 3.9

percent have undiagnosed diabetes. The predictive

model developed in NHANES was applied to CHIS

2013-14 data to produce California-speciﬁc estimates

of the prevalence of prediabetes (which include

undiagnosed diabetes). Although the California

prediabetes estimates include undiagnosed diabetes,

the proportion with undiagnosed diabetes is expected

to be relatively small, given that nationally less than

4 percent of adults have undiagnosed diabetes.

The California Health Interview Survey is a

collaboration of the UCLA Center for Health Policy

Research, the California Department of Public

Health, the California Department of Health Care

Services, and the Public Health Institute. For funders

and other information on CHIS, visit www.chis.ucla.edu.

UCLA CENTER FOR HEALTH POLICY RESEARCH

Author Information

Susan H. Babey, PhD, is a senior research scientist at

the UCLA Center for Health Policy Research. Joelle

Wolstein, PhD, MPP, is a research scientist at the

UCLA Center for Health Policy Research. Allison L.

Diamant, MD, MSHS, is a professor in the Division

of General Internal Medicine and Health Services

Research at the David Geffen School of Medicine

at UCLA. Harold Goldstein, DrPH, is executive

director of the California Center for Public Health

Advocacy.

Acknowledgments

The authors wish to thank Hongjian Yu, PhD;

Pan Wang, PhD; Akbar Esfahani, MIS; Carl Ganz;

Terry Schell, PhD; Gwen Driscoll; Venetia Lai; and

Celeste Maglan Peralta for their assistance. The

authors would also like to thank the following

individuals for their helpful comments: Xiao Chen,

PhD, associate director of the Health Economics

and Evaluation Program, UCLA Center for Health

Policy Research; Francine R. Kaufman, MD,

Distinguished Professor Emerita of Pediatrics at

USC, the Center for Endocrinology, Diabetes and

Metabolism, Children’s Hospital Los Angeles;

Paul Simon, MD, MPH, director of the Division of

Chronic Disease and Injury Prevention, Los Angeles

County Department of Public Health.

Suggested Citation

Babey SH, Wolstein J, Diamant AL, Goldstein H.

Prediabetes in California: Nearly Half of California

Adults on Path to Diabetes. Los Angeles, CA: UCLA

Center for Health Policy Research and California

Center for Public Health Advocacy, 2016.

Endnotes

1 Centers for Disease Control and Prevention, National

Center for Health Statistics, Division of Health Interview

Statistics, data from the National Health Interview

Survey. Statistical analysis by the Centers for Disease

Control and Prevention, National Center for Chronic

Disease Prevention and Health Promotion, Division of

Diabetes Translation. Accessed October 12, 2015. http://

www.cdc.gov/diabetes/statistics/prev/national/ﬁgageadult.htm

2 Centers for Disease Control and Prevention. National

Diabetes Statistics Report: Estimates of Diabetes and Its Burden

in the United States. 2014. Atlanta, GA: U.S. Department

of Health and Human Services.

3 Meng YY, Pickett MC, Babey SH, Davis AC, Goldstein

H. Diabetes Tied to a Third of California Hospital Stays,

Driving Health Care Costs Higher. 2014. Los Angeles, CA:

UCLA Center for Health Policy Research and California

Center for Public Health Advocacy.

4 Knowler WC, Barrett-Conner E, Fowler SE, et al.

Diabetes Prevention Program Research Group.

Reduction in the Incidence of Type 2 Diabetes with

Lifestyle Intervention or Metformin. 2002. N Engl J Med

(346): 393–403; Tuomilehto J, Lindstrom J, Eriksson J,

et al. Finnish Diabetes Prevention Study Group. 2001.

Prevention of Type 2 Diabetes Mellitus by Changes

in Lifestyle Among Subjects with Impaired Glucose

Tolerance. N Engl J Med (344): 1343–1350.

5 Bullard KM, Saydah SH, Imperatore G, Cowie CC,

Gregg EW, Geiss LS, Cheng YJ, Rolka DB, Williams

DE, Caspersen CJ. Secular Changes in U.S. Prediabetes

Prevalence Deﬁned by Hemoglobin A1c and Fasting

Plasma Glucose. National Health and Nutrition

Examination Surveys, 1999–2010. 2013. Diabetes Care

36(8): 2286-93.

6 May AL, Kuklina EV, Yoon PW. Prevalence of

Cardiovascular Disease Risk Factors Among U.S.

Adolescents, 1999−2008. 2012. Pediatrics 129(6):

1035-1041.

7 Centers for Disease Control and Prevention. Prediabetes.

http://www.cdc.gov/diabetes/basics/prediabetes.html. Published

October 21, 2014. Accessed January 27, 2016; Tabák

AG, Herder C, Rathmann W, Brunner EJ, and Kivimäki

M. Prediabetes: A High-Risk State for Diabetes

Development. 2012. The Lancet 379 (9833): 2279–90.

8 American Diabetes Association. 2013. Economic Costs

of Diabetes in the U.S. in 2012. Diabetes Care

36(4):1033-46.

9 Dall TM, Yang W, Halder P, Pang B, Massoudi M,

Wintfeld N, Semilla AP, Franz J, Hogan PF. The

Economic Burden of Elevated Blood Glucose Levels in

2012: Diagnosed and Undiagnosed Diabetes, Gestational

Diabetes Mellitus, and Prediabetes. 2014. Diabetes Care

37(12):3172-9.

10 Ridgeway G. Generalized Boosted Models: A

Guide to the GBM Package. May 23, 2012. http://

gradientboostedmodels.googlecode.com/git/gbm/inst/doc/gbm.pdf.

Accessed November 20, 2015; Friedman JH. Greedy

Function Approximation: A Gradient Boosting Machine.

2001. Annals of Statistics 129(5):1189-1232.

UCLA CENTER FOR HEALTH POLICY RESEARCH

10960 Wilshire Blvd., Suite 1550

Los Angeles, California 90024

The UCLA Center

for Health Policy Research

is part of the

UCLA Fielding School of Public Health.

The analyses, interpretations, conclusions,

and views expressed in this policy brief are

those of the authors and do not necessarily

represent the UCLA Center for Health Policy

Research, the Regents of the University

of California, or collaborating

organizations or funders.

PB2016-1

Editor-in-Chief: Gerald F. Kominski, PhD

Phone: 310-794-0909

Fax: 310-794-2686

Email: [email protected]

www.healthpolicy.ucla.edu

Read this publication online