Oral Health Surveillance

	October 25, 2004
	Based on reading:
	Eugenio D. Beltran-Aguilar, Dolores M. Malvitz, Stuart A. Lockwood, Gary Rozier, Scott L. Tomar

The problem and the need

	Since 1930’s, important changes in the prevalence and severity of dental caries
		Also a need to monitor other oral conditions or risk factors
	Our surveillance efforts, however, have changed very little in scope or format
		Local and state programs face challenges in an increasingly competitive environment for public resources
		require development and implementation of alternative surveillance tools
	Development of alternative surveillance tools needed

"Alternative surveillance efforts on state..."

	Alternative surveillance efforts on state and local level needed in order to:
		Assess oral health needs
		Monitor oral health status, including disparities among population groups
		Plan intervention programs at state and local levels
		Establish sound health policies
		Evaluate progress toward state health objectives

Past/current* efforts - Data collection on national level

	National Health and Nutrition Examination Surveys (NHANES)
		NHANES I (1971-74)
		NHANES II (1982-84): Hispanic health and nutrition examination survey
		NHANES III (1988-94)
		NHANES IV (1999-present)*

Past efforts - Data collection on national level

	National Institutes of Dental Research (now the NIDCR)
		1979-80 and 1986-87
			surveyed children 5-17 yrs
		1985-86
			surveyed employed adults and seniors

Past efforts to collect data on national level

	Systematic collection of data from representative samples
		mostly at the national level
	One-time or sporadic experiences
		especially for data at state or local level
	Use of visual-tactile protocols
		implemented at the tooth-surface or tooth-site level for data collection
	Focus mainly on dental caries and periodontal diseases
	Lengthy time from data collection to publication of results

Past efforts to collect data on national level

	Lowest level of statistical representation =  US geographic regions
		Data collected on representative sample of US population
	Demand high level of human and material resources
		In case of dental caries, use of visual-tactile assessments
		of all teeth/surfaces
		by trained, standardized dentists

Other forms of data collection at national level – besides clinical surveys

	Questionnaires have been used to collect oral health data at national level
	National Health Interview Survey (NHIS)
		Annual self-reported data
		Representative sample of US population
		Face-to-face interviews
		Core and supplemental modules
		Since 1983 – Track oral health topics: dental visits, use of fluoride supplements, toothbrushing practices, dental insurance status, screening for oral cancer
			have been part of basic, periodic, and topical sections

Past efforts – Data collection on state level

	States needed state-specific data:
	State and local government agencies tried to  implement visual-tactile exams
		Often tried to use same protocol and dx criteria used by federal agencies
	Usually sporadic experiences
		with limited ability to evaluate trends/needs at state level on a regular basis

Limitations of past efforts on state level

	Only provided snapshots of oral health status
		Could not be used to evaluate trends, particularly at the state and local levels
			Limited use for policy makers
	Required high levels of both human and material resources

Key characteristics of public health surveillance systems

	Key characteristics demanded of data systems for surveillance:
		Integrated
		Ongoing
		Cost efficient
		Translatable into public health interventions

Key characteristics of public health surveillance systems

	Monitoring activities identified by the term “epidemiologic surveillance”
	Defined as:
		ongoing
		systematic collection, analysis, and interpretation
		of outcome-specific data
		for use in planning, implementation, and evaluation of public health practice

Key characteristics of public health surveillance systems

	Surveillance activities grouped into systems:
	By one or more conditions
		Oral disease, cardiovascular disease, etc.
	By different aggregates of the population
		Local, state, national, international

Key characteristics of public health surveillance systems

	Two critical elements in this definition:
	Ongoing
		(regularly occurring, not episodic in nature) use of collected data
	Demands system of efficient analysis and dissemination
		For public health purposes

Key characteristics of public health surveillance systems

	Centers for Disease Control and Prevention (CDC)
		Federal agency responsible for monitoring diseases, conditions, and risk factors
		Provides data to policy makers and decision makers
		Data to serve as basis for implementing public health interventions

Key characteristics of public health surveillance systems

	National public health surveillance systems monitored by:
		the CDC Epidemiology Program
		the Council of State and Territorial Epidemiologists (CSTE)
		CDC and CSTE have established a set of standards
	State surveillance systems vary in the number and scope of conditions monitored
	Currently CDC monitors approximately 102 surveillance systems

Key characteristics of public health surveillance systems

	To be effective surveillance system needs:
	A functional structure that allows collection, processing and dissemination of information
	Participation of many within health service system, including clinicians and policy makers
	Data obtained from variety of sources
		Vital statistics
		Registries
		Sample surveys
		Administrative data systems
		Sentinel surveillance

Limitation of oral surveillance systems

	In oral health, have monitored disease primarily via conduct of clinical surveys
		on dental caries
		at both national & state levels
	If basic, desired characteristics are:
			integrated, ongoing, cost efficient, and
			translatable into public health interventions
	Infrequent collection of oral health data, using visual-tactile exam, does not make an oral health surveillance system

Limitations of using visual-tactile exams in oral health surveillance

	Rely heavily on primary data collection from calibrated dental professions
		Rationale is that only dental professionals, calibrated to a standard, can make valid diagnoses of oral diseases and conditions
	Very few public health surveillance systems rely so heavily on primary data
		Almost all surveillance systems accept a certain level of error as a consequence of misdiagnosis, misclassification, or incompleteness of data
	Example – mortality at national level measured with death certificates
		Filled out by a variety of individuals- physicians, physician assts, midwives, medical coroners – few have received standardization training on how to complete a death certificate

Limitations of using visual-tactile exams (cont.)

	Protocol developed primarily to measure dental caries
		Dental caries continues to be the most prevalent of all oral conditions
		However its prevalence and severity has declined dramatically during the past 30 years
			No indication will return to the previous levels of disease
	Because dental caries no longer so universal
		need surveillance tools to identify, at pop level, those still affected or at risk of dental caries
		and tools to measure other oral conditions and their risk factors

Limitations of using visual-tactile exams (cont.)

	Currently lack appropriate surveillance tool to measure periodontal disease
	Have a large number of indices:
		Measure soft and hard deposits in the supra- and sub-gingival areas
		Indices of gingivitis
		Indices of periodontal involvement alone – loss of attachment (LOA) or pocket depth
			Or combined with measures of gingivitis
		Also digital radiography and enzymatic tests to detect specific microorganisms
	None of these measures appropriate to collect surveillance data
		Issues of validity, reliability, and cost

Limitations of using visual-tactile exams (cont.)

	Consumes large amount of human and financial resources to conduct clinical examination surveys:
	Need to recruit, train, and standardize examiners
	Need resources to get & transport portable equipment, instruments, and infection control supplies
	State and local departments need to fund consultation on sampling and data analysis
	Too resource intensive for state and local policy making

Limitations of using visual-tactile exams (cont.)

	Difficult to secure participants in oral health surveys
	Risk of response bias
		If responders differ from nonresponders, risk of response bias
	Nonresponse rates high, especially among older children and adolescents
	Various approaches used to reduce this problem
			Negative consent – if parent does not return signed consent form, implies implicit consent for child to be examined
			Monetary incentives

Limitations of using visual-tactile exams (cont.)

	Most protocols collect information at tooth or surface levels
		Tooth/surface levels for dental caries
		LOA/pocket depth for periodontal disease
	Changes in prevalence and severity for most subjects
		Most of 32 teeth or 148 surfaces diagnosed and coded as sound
		Similar situation observed in site-specific assessment of LOA and pocket depth
	Most oral health objectives use person as unit of measurement
	Therefore may be unnecessary to collect surveillance information on dental caries at tooth or surface level

Limitations of using visual-tactile exams (cont.)

	In assessing dental caries, we measure both past and “present” episodes of the disease
	Surveillance generally does not measure past events
		such as past episodes of influenza or active TB
		or how many of these infections occurred in the lifespan of the individual
	Most clinical presentations of dental caries represent the past
	Restorations and missing teeth not always direct consequence of dental caries
		Therefore may be invalid in identifying populations at risk for oral diseases

 Limitations of using visual-tactile exams (cont.)

	Late reporting due to complicated planning and initiation procedures
	Often results reported years after initiation
	Inadequate for timely implementation of public policies and evaluation of outcomes

Weaknesses in the current oral health surveillance system

	Visual-tactile examination has been regarded as gold standard –
	Ongoing belief that oral health surveillance data requires same level of rigor and precision as research related to clinical treatment
		Virtually no public health surveillance systems conduct primary collection of data with same rigor as researchers conducting randomized clinical trials

Weaknesses in the current oral health surveillance system

	DMF and LOA too resource intensive to be used as primary oral health surveillance tools at state and local level
		Put programs in undue disadvantage against other public health programs
	Cross-sectional prevalence data have not been used successfully for program planning
		these efforts often end up with publication of data long after a planning decision should have been made

Alternative methods utilized in oral health surveillance

	Seven-step model for needs assessment
		Developed in 1995-96 by Health Resources and Services Administration (HRSA)
		Designed to meet needs to collect oral disease data at state & local levels with limited resources
		Step-by-step model assumed different levels of resources and proposed data collection methods for each level
		Methods range from most simple (expert opinion, focus groups) to complex (screenings, surveys, secondary data)
		Model emphasizes need to start with simpler techniques and to move to more complex ones after initial data generated, and there is justification to move to more complex level

Alternative methods: seven-step model for needs assessment (cont.)

	Model tested in Louisiana and Nebraska
		Not used extensively
	But helped some public health officials to consider techniques other than visual-tactile exams for obtaining oral disease data

Alternative methods: Use of visual-only screening models

	Defined as intraoral assessment and reporting of status at the person level
		Used to collect data among school children and preschool children in OR
		Later in a state-wide screening of school children in WA
	CDC tested visual-only screening protocol that used person-based assessment of oral status
		Assessment included dental caries, presence of sealants, urgency of treatment needs, enamel fluorosis, and injuries
	Protocol designed to require minimal instruction of examiners
		(a dental hygienist and a registered nurse)
	Take little time to conduct
	Require no sophisticated equipment or instruments
	Later used in LA, MA on school and pre-school children, Special Olympics population

Alternative methods: Basic screening survey (BSS) – visual-only exam

	Visual-only model
		to provide timely data with sufficient validity and reliability
	Less resource intensive
		less demanding training process and lower time requirements
	Standardized protocol using a video to train screeners
		Used in 1999 to assess oral health status of approx 21,000 children in grades 1-3 in Ohio at the county level
		Expected to help identify people at risk as data are collected and analyzed more frequently

Alternative methods: Programmatic and administrative data

	Forms completed by dentists and hygienists in 20 local health agencies in Michigan
			number of decayed and filled teeth, presence of early childhood caries, presence of sealants, root caries, and presence of two or more teeth in adults
	Medicaid claims data
			number of children affected by early childhood caries  and resultant treatment costs
			Medicaid data also used in Iowa and NC
	Insurance claims data
			Michigan to examine trends in dental tx provided to enrollees in a private dental insurance plan

Alternative methods: Self-reports, established surveys

	Self- or parent reports
		Use of global descriptors (i.e., excellent to fair and poor)
		Show strong correlation with clinical health status
		Preliminary results from sample of children in WA with prevalence of ECC of 15%
			show mothers can accurately assess oral health status of their children’s teeth
	Face to face and telephone interviews
		Optional and core modules included in existing surveys
		(see following examples)

 Alternative methods:
 Self-reports, established surveys (cont.)

	The Behavioral Risk Factor Surveillance System (BRFSS)
		State health departments
		Telephone surveys, using random-digit selection
		Generates prevalence estimates of chronic disease risk factors
		Core set of questions and number of optional modules
		States allowed to add their own questions
	In 1995 optional module of oral health-related questions was introduced
		Over a 4-year period, 48 states used the module
	In 1999 three oral health-related questions were included on the BRFSS core questionnaire
		Thus data available for every state for that year and included again in 2002

 Alternative methods:
 Self-reports, established surveys (cont.)

	Pregnancy Risk Assessment Monitoring System (PRAMS)
		Conducted by CDC
		Collects state-specific information
		On health-related behaviors and experiences
		Representative sample of mothers who delivered live infants
			About one-third of mothers reported dental visit during pregnancy
			Of those who said that they needed to see a dentist for a problem:
				only about half had dental visit during their pregnancy

Alternative methods:
Self-reports, established surveys (cont.)

	Youth Risk Behavior Surveillance System (YRBSS)
		Has potential for tracking oral health information in youth
		School-based survey – CDC system
		Administered biennially through state Depts. of Education
		Assesses prevalence of health risk behaviors among high school students
		Data on tobacco use and other behaviors and trends obtained

Alternative methods: Important characteristics - summary

	These procedures have important characteristics for use in surveillance of oral diseases, conditions, and risk behaviors:
		1) integrated into existing data collection mechanisms
		2) data collection is frequent and systematic, thus providing timely data
		3) data do not rely on visual-tactile examinations
		4) when clinical data needed, secondary data sources or visual screenings used

Other alternative methods

	Geographic Information System (GIS)
		To identify geographic areas, administrative areas, counties, or census tracts with populations at higher risk of disease
		Used to examine distribution of dentists by aggregate measures of factors associated with caries
			percentage of schoolchild population eligible to receive free and reduced-cost lunch

Other alternative methods

	Sentinel surveillance
		Collects data from sites such as hospitals, clinics and health centers, and schools where patients at risk receive care
			Routinely used to monitor multiple conditions including influenza, HIV, and cancer
		Due to low national prevalence and association with poverty, Early Childhood Caries (ECC) could be monitored by sentinel surveillance
			Most children with ECC are treated at pediatric dental offices, dental schools

National Oral Health Surveillance System (NOHSS)

	First step in development of comprehensive state and local surveillance systems
	Includes exiting surveillance data from
		BRFSS, NHIS, and WFRS
	Calls for use of BSS to collect person-based data for set of oral health disease indicators
	Is expected that more state programs will be able to generate their own data for action
		technical support would be available from ASTDD (Assoc of State and Territorial Dental Directors)

National Oral Health Surveillance System (NOHSS)

	Developed by the ASTDD and the CDC’s Division of Oral Health
	User-friendly, resource-sparing, and integrated oral health surveillance system
	Includes eight basic oral health indicators obtained from existing surveillance systems, BSS, CDC Water Fluoridation Reports, and cancer registries
		Adult dentals visits - Child caries experience
		Adult tooth cleaning - Child untreated caries
		Adult tooth loss - Child dental sealants
		Fluoridation status - Oral and pharyngeal cancer
	Also provides data on state demographics, oral health program infrastructure, administration, and activities at the state level
	www.cdc.gov/nohss

Utility of surveillance data

	Most state, local, national agencies recognize need for timely data
		Useful for generating support for their dental public health programs
	States have been able to survive administrative reorganizations and increase funding after documenting statewide needs

Challenges for the future for oral health surveillance

	Testing the validity of self-reporting and visual assessment in seniors
	Developing a screening protocol for periodontal diseases
	Implementing standardized codes for treatment claims data
	Process of exchange with research community on validation of new surveillance tools
	NIDCR/CDC Dental, Oral, and Craniofacial Data Resource Center has been developed to promote access to surveillance data and collaboration among researchers
	Envision system that will generate data from the local level to the state, regional, national levels