An Innovative Approach to Enhancing the Surveillance ...

VOLUME 2: NO. 3 JULY 2005

An Innovative Approach to Enhancing theSurveillance Capacity of State-based

Diabetes Prevention and Control Programs:The Diabetes Indicators and Data Sources

Internet Tool (DIDIT)

TOOLS & TECHNIQUES

Suggested citation for this article: Mukhtar Q, Brody ER,Mehta P, Camponeschi J, Clark CK, Desai J, et al. Aninnovative approach to enhancing the surveillance capaci-ty of state-based diabetes prevention and control pro-grams: the Diabetes Indicators and Data Sources InternetTool (DIDIT). Prev Chronic Dis [serial online] 2005 Jul[date cited]. Available from: URL: http://www.cdc.gov/pcd/issues/2005/jul/04_0126.htm.

PEER REVIEWED

Abstract

The Diabetes Indicators and Data Sources Internet Tool(DIDIT) is an interactive Web-based resource with infor-mation on 38 diabetes indicators (e.g., diabetes-associatedcomplications, care, lifestyle) and 12 associated datasources frequently used by state diabetes prevention andcontrol programs. This tool is designed to strengthen theability of states to conduct diabetes surveillance and topromote consistency in defining and tracking indicatorsacross states. In this way, the DIDIT supports one of the10 essential public health services: the timely and accurateassessment of public health.

In addition to serving as a central repository of informa-tion on diabetes surveillance, the DIDIT also allows usersto share experiences of using these indicators and data

sources in their diabetes surveillance activities, dataanalysis, and tracking of diabetes-related objectives statedby Healthy People 2010. The DIDIT is an innovativeapproach to enhancing public health surveillance at thestate and national levels.

Introduction

Diabetes is a complex, chronic disease consisting of agroup of conditions resulting from the impaired productionor effectiveness of insulin (1). It is increasingly importantto conduct surveillance to accurately assess the prevalenceof diabetes and its complications. The global prevalence ofdiabetes is nearly 200 million cases, and there is a senseof urgency in the public health field about understandingand defining the burden of diabetes and its associated riskfactors (2). Traditionally, diabetes surveillance has includ-ed monitoring indicators such as incidence, prevalence,complications, mortality, quality of care, and preventive-care practices (3). Reflecting recent developments in theidentification of risk factors for type 2 diabetes (4,5), sur-veillance has expanded to include lifestyle variables suchas diet and physical activity (6), socioeconomic status (7),quality of life (8), and a condition called prediabetes.People who have blood glucose levels that are higher thannormal, but not high enough to be classified as diabetes,have prediabetes (4).

The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services,the Public Health Service, the Centers for Disease Control and Prevention, or the authors’ affiliated institutions. Use of trade names is for identification only

and does not imply endorsement by any of the groups named above.

www.cdc.gov/pcd/issues/2005/jul/04_0126.htm • Centers for Disease Control and Prevention 1

Qaiser Mukhtar, PhD, Erica R. Brody, MPH, Prachi Mehta, DrPH, Jenny Camponeschi, MS, Cynthia K. Clark,MA, Jay Desai, MPH, Michael Friedrichs, MS, Angela M. Kemple, MS, Heidi R. Krapfl, MS, Brenda Ralls,

PhD, Jackson P. Sekhobo, PhD

VOLUME 2: NO. 3JULY 2005

The Centers for Disease Control and Prevention’s (CDC’s) NationalDiabetes Prevention and ControlProgram (NDPCP) supports a DiabetesPrevention and Control Program(DPCP) in each state, each U.S. terri-tory, and the District of Columbia. TheNDPCP and the 59 DPCPs are chargedwith defining and monitoring the bur-den of diabetes and demonstrating suc-cess in accomplishing NDPCP nationalobjectives, which were established in1999 to reduce preventable morbidityand mortality associated with diabetes.These objectives incorporate goals toincrease the proportion of people withdiabetes who receive annual foot exam-inations, dilated eye examinations,hemoglobin A1c (HbA1c) tests, and rec-ommended vaccinations againstinfluenza and pneumococcal infection.The national objectives also addressreducing health disparities for popula-tions at high risk for diabetes andestablishing links with other programsthat promote wellness among people with diabetes (10).

Additionally, the timely and accurate assessment of pub-lic health is one of the 10 essential public health servicesand thus a fundamental function of public health agencies,such as DPCPs, that are responsible for population-basedhealth (11). The DPCPs also track and strive to achieveHealthy People 2010 objectives, 17 of which directlyaddress diabetes (12).

Surveillance is essential to effective program plan-ning, advocacy, and evaluation, and surveillance datacontribute significantly to developing and carrying outpolicies at the national and state levels. Surveillance isthe continuous monitoring or routine collection of dataon various factors (e.g., behaviors, complications,deaths) over a regular interval of time, and surveillancedata can be useful for program evaluation. In contrast,evaluation provides tailored information to answer specific questions about a program. Data collection forevaluation is more flexible than for surveillance andmay allow specific program areas to be assessed ingreater depth (13).

The NDPCP requires each DPCP toestablish a diabetes surveillance sys-tem to monitor the burden of diabeteswithin its state and systematicallyevaluate the impact of DPCP pro-grams on the morbidity and mortalityassociated with diabetes. The 59DPCPs differ in their resources, fund-ing levels, expertise, priorities,staffing, and partnerships. These factors affect a program’s ability to col-lect, analyze, and interpret data and toconduct surveillance and evaluate programs. A diabetes-specific central-ized resource providing comprehensivebackground information on definitionsof indicators, data sources, methodolo-gy, and issues relevant to analysis isimportant to achieving these surveil-lance and evaluation goals.

In the fall of 2001, representativesfrom NDPCP and five of the 59 DPCPsformed a work group that began devel-oping a tool to assist the DPCPs with

surveillance and program evaluation. The result was theDiabetes Indicators and Data Sources Internet Tool(DIDIT), which was developed to strengthen the DPCPs’ability to conduct surveillance and program evaluation.

The DIDIT is an interactive resource that provides com-prehensive information on 38 indicators of diabetes andtheir data sources. Although most of the materials found in the DIDIT are available elsewhere (e.g.,www.cdc.gov/diabetes/statistics/index.htm; www.cdc.gov/brfss), the intent of the DIDIT is to offer, in a centralizedlocation, the most recent information on diabetes surveil-lance. Another goal is to present clear, concise, and easy-to-read information by organizing it into a flexible and aninteractive format. In addition to providing information on38 diabetes indicators (Figure 1) and other associated datasources (if available) for each indicator, the DIDIT pro-vides information such as lists of related articles, Web sitelinks, and recommended standards of care. Updated on anongoing basis, the DIDIT facilitates communication in thediabetes community by allowing for user feedback andcomments about indicators and data sources. In addition,users have the opportunity to share their general experi-ences with using the tool. In this article, we describe the

2 Centers for Disease Control and Prevention • www.cdc.gov/pcd/issues/2005/jul/04_0126.htm



Figure 1. List of all indicators in the DiabetesIndicators and Data Sources Internet Tool(DIDIT) by type.

DIDIT’s content, provide examples ofhow its information is presented, andhighlight its public health applications.We provide an overview of the processused to develop the DIDIT; a compre-hensive description of the DIDIT devel-opment process will be published in acompanion article.

Developing the DIDIT

The DIDIT was developed by themembers of a work group representingDPCPs and the NDPCP. The workgroup members have extensive expert-ise in diabetes surveillance and evalua-tion. User input has also been obtainedfrom additional DPCPs through 1)focus groups, 2) usability testing, and3) pilot testing, which the NDPCP con-ducted with volunteer DPCPs beforethe DIDIT’s release. Content and Website development are supported by twoindependent contractors. To facilitatecontent development, the work groupmembers developed a list of 55 diabetesindicators and, using two rounds ofDelphi process, ranked them based onimportance, mutability, and availabili-ty of state-level data. In this way, theDIDIT’s 38 indicators were selected.The national- and state-level datasources described in the DIDIT wereselected because they are publiclyavailable, reliable, and accurate as aresult of quality assurance proceduresin place at the various agencies thatmaintain them (e.g., the CDC, theAgency for Healthcare Research andQuality, and the Centers for Medicareand Medicaid.) The first version of theDIDIT was released to DPCPs inOctober 2003. In April 2004, the DPCPstaff and the NDPCP project officerswere trained to use the DIDIT using NetMeeting(Microsoft Corp, Redmond, Wash), which allowed for atwo-way interaction between stakeholders in various geographical areas.

The DIDIT is still a work inprogress, and several enhancementsare underway. The DPCPs and CDCpartners have been kept informed ofthe DIDIT development and enhance-ment on an ongoing basis throughpresentations, e-mail updates, andconference calls.

Contents of the DIDIT

Diabetes indicators

The DIDIT provides informationabout each of 38 diabetes indicators(Figure 2), including its relevance todiabetes surveillance, NDPCP nation-al objectives, and Healthy People 2010objectives (12). Additionally, it pro-vides information about the agenciesrecommending the indicator, relatedarticles and Web sites, and a list ofassociated data sources. Many of theindicators are related to health carepractices (e.g., annual foot examina-tions) or recommendations on lifestyleissues for people with diabetes (e.g.,physical activity). If relevant, informa-tion on recommended standards ofcare is provided. For example, theAmerican Diabetes Association’s posi-tion statement on preventive carepractices is linked to many of the indicators.

Data sources

The DIDIT includes 12 frequentlyused state and national data sourcesfor diabetes surveillance (Figure 3).The data sources are cross-referencedwith information about the indicators.For each indicator, the DIDIT providesmethodologies for computing the

numerator and denominator using the associated datasources. In addition, DPCPs can enter and share informa-tion about a DPCP-specific data source to inform others ofspecial methods for collecting local data. As of April 2005,





Figure 2. Information in the Diabetes Indicatorsand Data Sources Internet Tool (DIDIT) on eachof 38 diabetes indicators.

Figure 3. List of Diabetes Prevention andControl Program-specific data sources in theDiabetes Indicators and Data Sources InternetTool (DIDIT).

Figure 4. General information about the EndStage Renal Disease Networks data source thatis included in the Diabetes Indicators and DataSources Internet Tool (DIDIT).


California, Utah, and Ohio had enteredinformation about DPCP-specific datasources they have used for diabetessurveillance (Figure 3). The DIDITdata sources include surveys (e.g.,Behavioral Risk Factor SurveillanceSystem [BRFSS], National HealthInterview Survey [NHIS]), administra-tive records (e.g., health insuranceclaims), and vital records (e.g., deathand birth certificates). These data arecollected by various methods, such astelephone interviews, in-person inter-views, and laboratory testing. Weshould note that the DIDIT does notprovide data but instead gives compre-hensive information about data,including where and how to access it.For each data source, the DIDIT pro-vides information on the populationrepresented by the data, the methodand purpose of data collection, accessissues, privacy issues, and general ana-lytic issues, in addition to providing alist of diabetes indicators that can betracked using that data source (Figure4). The DIDIT is designed to provideusers with the information needed touse a data source effectively. Thus, theDIDIT serves as a reference guide forlearning about and using key datasources for diabetes surveillance. TheDIDIT also has a search feature, allow-ing users to search for specific datasources, indicators, or both.

Specification of numerators anddenominators

Standardization in computing andreporting estimates across states isessential to high-quality diabetes sur-veillance. Any differences in calculat-ing the numerator or denominator of an indicator at thestate level can cause variations, making it difficult to com-pare data over time or across states.

The DIDIT provides instructions for constructingnumerators and denominators of indicators (Figure 5) for

each data source linked to each indica-tor. For example, state-level data onHbA1c testing is available fromsources such as the Health PlanEmployer Data and Information Set(HEDIS), BRFSS, Medicare, andMedicaid. The DIDIT provides infor-mation about computing the numera-tor and denominator for HbA1c testingusing each of these four data sources.It provides information on which casesshould be included in the numeratorand denominator in addition to rele-vant diagnosis and procedure codeswhere appropriate (e.g., for adminis-trative and inpatient hospital dis-charge data). For survey data, relevantsurvey questions are provided (Figure5), allowing users to analyze andreport diabetes indicators consistentlywith other programs. All of this infor-mation allows users to compare differ-ent potential data sources and decidewhich are best suited for their pro-gram needs and how to correctlydefine the numerator and denomina-tor for data analysis.

Technical issues in analysis

Data analysis can be complex, andmost people learn about its pitfallsthrough trial and error. A unique fea-ture of the DIDIT is the technical notesfield, which provides backgroundinformation and caveats for analyzingdata for an indicator and a data source(Figure 6).

Furthermore, the DIDIT encouragesusers to share experiences in analyz-ing indicators and data sources. Byusing the comments view/add feature

(Figure 4) and posting their comments on the DIDIT, userscan inform their peers about technical issues related todiabetes data analysis. This interactive feature of theDIDIT sets up an avenue for sharing, documenting, andlearning. Currently, all DPCPs can enter this informationabout their own or their partners’ experiences.




Figure 5. General and diabetes indicator-spe-cific information provided by the Behavioral RiskFactor Surveillance System (BRFSS) datasource that is included in the DiabetesIndicators and Data Sources Internet Tool(DIDIT).

Figure 6. Information to construct the inci-dence of end-stage renal disease attributed todiabetes indicators using the End Stage RenalDisease Networks data source.

At a Glance

The At a Glance function in theDIDIT provides a snapshot of the rela-tionship between indicators and theirdata sources. For example, in Figure 7,the 38 indicators are presented in thefirst column, with their national- andstate-level data sources presented inthe second and third columns. This lay-out allows users to quickly assess thenumber and names of data sourcesavailable in the DIDIT for each indica-tor. Similarly, the DIDIT presentsanother table with the 12 data sourceslisted in the first column and the indi-cators that can be measured with eachdata source in the second column(Figure 8).

Upcoming Enhancementsand Activities

In its current capacity, the DIDIT isa powerful tool, but it continues toevolve in response to the needs of theNDPCP and DPCPs. Thus, the DIDITis a work in progress. While we workon populating the current fields,efforts are underway to enhance thetool by adding a glossary of epidemiol-ogy and surveillance terms, links touseful references, and a section calledResources that lists background mate-rials frequently used by epidemiolo-gists. We plan to update the DIDITcontent as needed and develop an online training curriculum.Additionally, we plan to evaluate theDIDIT to ensure that it is meeting theneeds of DPCPs and their partners.

Using the DIDIT for ProgramEvaluation Activities

An additional function of the DIDITis to assist in evaluating public health

programs. The information provided inthe DIDIT can address steps 3–5 of theCDC’s 6-step framework for programevaluation (14). For example, if a DPCPis planning to evaluate the increase inthe percentage of people with diabeteswho receive the recommended eyeexamination, they might use the DIDITto help them and their community part-ners understand the reason it is impor-tant to track eye examinations amongpeople with diabetes and how this indi-cator is defined. Next, using informa-tion on available data sources, theywould identify data sources for trackingeye examinations (e.g., BRFSS,HEDIS) and use the descriptive infor-mation about each data source to selectthe best data source for their purpose.Using information in the indicator-spe-cific section, they would construct theirnumerator and denominator for eyeexaminations in a way that is consis-tent with methods used by otherDPCPs. The information in technicalnotes would help them analyze the dataand interpret their conclusions.

Implications for PublicHealth Practice

Effective public health surveillancerequires access to timely, accurate, andreliable information from a wide vari-ety of sources (3). The DIDIT is a valu-able resource for helping public healthpractitioners and clinical staff to assessthe burden, complications, and risk fac-tors of diabetes.

The DIDIT is the only reference toolof its kind that contains diabetes-spe-cific information required for surveil-lance and, to some extent, for programevaluation activities. This comprehen-sive resource saves time becauseDIDIT users do not have to search for information on several Web sites





Figure 7. List of diabetes indicators and theirassociated national-, state-, and DiabetesPrevention and Control Program (DPCP)-specificdata sources.

Figure 8. List of Diabetes Indicators and DataSources Internet Tool (DIDIT) data sources bytype and related indicators.


or read lengthy descriptions to find information on diabetes.

The information about indicators has numerous poten-tial uses. People new to diabetes surveillance can use theinformation to become familiar with multiple aspects ofeach indicator. The information also assists experiencedpublic health practitioners by helping them work with newindicators or stay current on the most recent diabetesinformation, such as the latest position statement of theAmerican Diabetes Association. Having a list of relatedpeer-reviewed articles, summaries, or Web sites for eachindicator saves users the effort of searching the literature.Informally, some users have reported using informationand resources from the description pages on various indi-cators to prepare presentations, write articles, and locateand develop educational materials.

The DIDIT provides a common platform and languagefor epidemiologists, state program coordinators, the CDC’sDivision of Diabetes Translation, and project officers tocommunicate about issues on diabetes surveillance andanalysis. In addition, it is designed to promote nationaluniformity in collecting and defining data and computingdiabetes indicators. Finally, it provides a mechanism forDPCPs to share their unique state and local data sources,in addition to their own experiences with using and ana-lyzing diabetes indicators and data sources.

The concepts and participatory processes used for theDIDIT’s development were designed to be user friendlyand can be applied to many areas of public health practice.They also may be used to develop information technologysolutions for other diseases, public health settings, or pro-gram areas. The DIDIT serves as a model approach forother local, state, and national public health programsinterested in enhancing their and their partners’ technicaland scientific ability by emphasizing collective experienceand information technology. Additionally, the DIDIT pro-vides a template that may be applied to other chronic dis-ease programs at the CDC.

Access to the DIDIT

The DIDIT is currently accessible to all NDPCP andDPCP staff through the Diabetes ManagementInformation System. To ensure data integrity and securi-ty, this system is password protected on a secure site.

Individuals in the diabetes community can request accessto the DIDIT by visiting the CDC’s Diabetes Public HealthResource Web site (www.cdc.gov/diabetes/statistics/index.htm) and filling out a request form. Since its releasein October 2003, more than 30 external partners haverequested and been granted access to the DIDIT.

Author Information

Corresponding Author: Qaiser Mukhtar, PhD, Divisionof Diabetes Translation, Centers for Disease Control andPrevention (CDC), 4770 Buford Hwy, Mail Stop K-10,Atlanta, GA 30341. Telephone: 770-488-5505. E-mail:[email protected].

Author Affiliations: Erica R. Brody, MPH, RTIInternational, Research Triangle Park, NC; Prachi Mehta,DrPH, Northrop Grumman CITS Contract, Atlanta, Ga;Jenny Camponeschi, MS, Wisconsin Department ofHealth and Family Services, Division of Public Health,Diabetes Prevention and Control Program (DPCP),Madison, Wis; Cynthia K. Clark, MA, CDC, Atlanta, Ga;Jay Desai, MPH, Minnesota Diabetes Program, MinnesotaDepartment of Health, St. Paul, Minn; Michael Friedrichs,MS, Utah Bureau of Health Promotion, Department ofHealth, Salt Lake City, Utah; Angela M. Kemple, MS,Oregon Department of Human Services, DPCP, Portland,Ore; Heidi R. Krapfl, MS, New Mexico Diabetes ControlProgram, Department of Health, Santa Fe, NM; BrendaRalls, PhD, Utah DPCP, Department of Health, Salt LakeCity, Utah; Jackson P. Sekhobo, PhD, DiabetesSurveillance and Evaluation, Bureau of Chronic DiseaseEpidemiology and Surveillance, New York StateDepartment of Health, New York, NY.

References

1. National Institute of Diabetes & Digestive & KidneyDiseases. National diabetes statistics fact sheet: gen-eral information and national estimates on diabetes inthe United States. Revised edition. Bethesda (MD):U.S. Department of Health and Human Services,National Institutes of Health; 2004.

2. Boyle JP, Honeycutt AA, Narayan KM, Hoerger TJ,Geiss LS, Chen H, et al. Projection of diabetes burdenthrough 2050: impact of changing demography anddisease prevalence in the U.S. Diabetes Care




2001;24(11):1936-40.3. Desai J, Geiss L, Mukhtar Q, Harwell T, Benjamin S,

Bell R, et al. Public health surveillance of diabetes inthe United States. J Public Health Manag Pract Nov2003;Suppl:S44-51.

4. Williamson DF, Vinicor F, Bowman BA; Centers forDisease Control and Prevention Primary PreventionWorking Group. Primary prevention of type 2 diabetesmellitus by lifestyle intervention: implications forhealth policy. Ann Intern Med 2004;140(11):951-7.

5. Knowler WC, Barrett-Connor E, Fowler SE, HammanRF, Lachin JM, Walker EA, et al; Diabetes PreventionProgram Research Group. Reduction in the incidenceof type 2 diabetes with lifestyle intervention or met-formin. N Engl J Med 2002;346(6):393-403.

6. Hu G, Lindstrom J, Valle TT, Eriksson JG, JousilahtiP, Silventoinen K, et al. Physical activity, body massindex, and risk of type 2 diabetes in patients with nor-mal or impaired glucose regulation. Arch Intern Med2004;164(8):892-6.

7. Brown AF, Ettner SL, Piette J, Weinberger M, GreggE, Shapiro MF, et al. Socioeconomic position andhealth among persons with diabetes mellitus: a con-ceptual framework and review of the literature.Epidemiol Rev 2004;26:63-77.

8. Brown DW, Balluz LS, Giles WH, Beckles GL,Moriarty DG, Ford ES, et al. Diabetes mellitus andhealth-related quality of life among older adults.Findings from the behavioral risk factor surveillancesystem (BRFSS). Diabetes Res Clin Pract2004;65:105-15.

9. Centers for Disease Control and Prevention.Prevalence of diabetes and impaired fasting glucose inadults, United States, 1999-2000. MMWR MorbMortal Wkly Rep 2003;52(35):833-7.

10. Safran MA, Mukhtar Q, Murphy DL. Implementingprogram evaluation and accountability for populationhealth: progress of a national diabetes control effort. JPublic Health Manag Pract 2003;9(1):58-65.

11. Institute of Medicine Committee on Assuring theHealth of the Public in the 21st Century. The future ofthe public’s health in the 21st century. Washington(DC): National Academies Press; 2002.

12. Land GH. Measuring 2010 national objectives andleading indicators at the state and local level. J PublicHealth Manag Pract 2002;8(4):9-13.

13. Centers for Disease Control and Prevention.Introduction to Program Evaluation forComprehensive Tobacco and Control Programs.

Atlanta (GA): Centers for Disease Control andPrevention; 2001 Nov: p. 6-9.

14. Centers for Disease Control and Prevention.Evaluation Working Group [Homepage on theInternet]. Atlanta (GA): Centers for Disease Controland Prevention; 2004 [updated 2004 Aug 4; cited 2004 Sep 27]. Available from: URL:http://www.cdc.gov/eval/framework.htm.





An Innovative Approach to Enhancing the Surveillance ...

Documents

Transcript of An Innovative Approach to Enhancing the Surveillance ...