Modeling the Dynamics of Dental Health in Older Adults
Sara S. Metcalf
Department of Geography
The State University of New Y ork at Buffalo
105 Wilkeson Quad, Ellicott Complex, North Campus
Buffalo, NY 14261
Phone: (716) 645-0479
Fax: (716) 645-2329
smetcalf@ buffalo.edu
Michael J. Widener
Department of Geography
The State University of New Y ork at Buffalo
105 Wilkeson Quad, Ellicott Complex, North Campus
Buffalo, NY 14261
mjwidene@ buffalo.edu
Mary E. Northridge
Department of Epidemiology and Health Promotion
New Y ork University College of Dentistry
250 Park Ave South
New Y ork, NY 10003
men6@nyu.edu
Mark Paich
Principal, Decisio Consulting
P.O. Box 1466, Portage, MI 49081
m.paich@ att.net
Stephen Marshall
ElderSmile Program Director
Columbia University College of Dental Medicine
P&S Box 20, 650 W 168th St, New York, NY 10032
sm15@columbia.edu
Ira B. Lamster
Dean, Columbia University College of Dental Medicine
630 W. 168" Streeet, New Y ork, NY 10032
ibl1@ columbia.edu
Acknowledgment: Work on this paper is part of a National Institutes of Health (NIH) project
(grant 1R21DE021187-01) titled, “Leveraging Opportunities to Improve Oral Health in Older
Adults: A System Dynamics Model for Developing and Prioritizing Interventions, Programs, and
Policies.” Within NIH, the project is funded by the National Institute for Dental and Craniofacial
Research (NIDCR) and the Office of Behavioral and Social Science Research (OBSSR).
1
ABSTRACT
As the population of older adults in the United States expands with the aging of the Baby Boom
generation, the advantages of developing coordinated and cost effective health policies have
become clearer. Co-morbidities between medical and dental health conditions become more
prevalent with age. For example, periodontal disease is a risk factor for other chronic illnesses,
notably diabetes and cardiovascular disease. Despite this link, medical and dental disorders are
rarely screened for and treated as related conditions. Additionally, access to dental care for
older adults may be adversely affected by lack of insurance coverage and complex social,
cognitive, and physical factors that result in missed opportunities for intervention and prevention
of more serious conditions. Developing interventions to improve dental health among older
adults is a particular challenge, given the complex set of causal pathways and time delays over
the life course that are involved. In this paper, we draw upon the experience of Columbia
University's ElderSmile outreach program to inform construction of a dynamic model as part of
a larger research endeavor that explores the intra- and inter-personal relationships relevant to
oral health in older adults.
1, INTRODUCTION
Oral health has been recognized as a matter of public health concern because it affects a large
proportion of the population and is linked with general health status (IOM, 2002). The landmark
publication, Oral Health in America: A Report of the Surgeon General, highlighted a lack of
awareness of the importance of oral health among the general public, and found a significant
disparity between racial and socioeconomic groups with regard to oral health and ensuing overall
health issues (US DHHS, 2000). According to the Surgeon General’s report, the mouth is the
gateway of the body, in that it senses and responds to the external world and reflects what is
happening deep inside the body. The mouth signals nutritional deficiencies and serves as an early
waming system for diseases such as HIV/AIDS, other immune system problems, general
infections, and stress. Poor oral health is associated with diabetes, heart disease, and stroke
(Allukian, 2006). A comprehensive exam in the mouth is a novel way to study the body as an
integrated system, since it is an accessible entry point for the medically complex patient (Glick,
1999).
As the Baby Boom demographic enters retirement, the need for regular and proactive dental care
among older adults is becoming ever more acute. The oral health of older adults is influenced by
complex medical and social factors including multiple medications for chronic illness, cognitive
impairment, social isolation, and/or physical disabilities that interfere both with oral hygiene
activity and access to dental care. Early intervention strategies to address oral health problems
can prevent illness, diagnose serious conditions early, and maintain optimum overall health.
Patient-centered oral health promotion, such as the preventive screening (PS) and referral
services offered through Columbia University’s ElderSmile program in upper Manhattan,
provides a way to identify and address conditions that compound oral health problems for this
underserved population.
2, CONCEPTUAL FRAMEWORK
Because a systems perspective cannot be gained by studying component parts in isolation, a
consideration of cross-scalar relationships is necessary to develop and implement oral health
interventions for older adults. Factors at the neighborhood scale (e.g., community access),
interpersonal scale (e.g., oral health promotion), and individual scale (e.g., nutrition and chronic
illness) are particularly important in influencing dental health outcomes such as tooth retention,
dental caries, and periodontal disease. This paper employs a systems perspective to frame oral
health in older adults as due to the lifelong accumulation of advantageous and disadvantageous
experiences at multiple scales, from the micro-scale of the mouth to the societal scale that
involves U.S. federal policy, including lack of routine dental care coverage under Medicare.
Methodological developments in system dynamics, geographic information systems (GIS),
agent-based modeling, and social network analysis have enabled a growing body of research on
the effects of multiple scales (macro, meso, and micro) and environmental dimensions
(geographic, institutional, and social) on health behaviors and outcomes. Intersections of scale
and environmental factors on health in general, and dental health in particular, are outlined in
Table 1.
Table 1. Multi-level framework for environmental influences on health
Macro Meso Micro
Geographic F Individual Activity
Efvironnignt Urban Context Neighborhood Space
Institutional Health Care Outreach
Environment System Programs Self Care
Social Socio-economic Community Personal
Environment Structures Relationships Network
The multiple scales and environmental dimensions outlined in Table 1 represent a multi-level
framework for modeling health, where the level of the social environment encompasses a diverse
set of mechanisms operating among and within social structures existing at different levels
(Northridge et al, 2003; Bachrach & Abeles, 2004). At the macro level are structures and
processes that involve and affect populations broadly: goverment, media, economic systems,
social stratification, political processes and policymaking, and commonly held cultural values
and practices. Some of these processes also operate at the meso level, that is, in communities,
neighborhoods, and institutions such as workplaces. Processes contributing to social cohesion,
social support, social control, social and cultural conflict, and the development and enforcement
of social and cultural norms also play a significant role. In families and small groups,
interpersonal processes such as conflict and support, socialization, and sharing of resources play
a dominant role, at both the meso level and micro level. Characteristics of the individual and
biological mechanisms (micro level) fill out the multi-level conceptualization for this research.
2.1 CAUSAL DYNAMICS
Modeling with system dynamics enables policymakers to assess the impact of different health
interventions to identify ones that yield the greatest leverage in both the short term and the longer
3
term (Homer and Hirsch, 2006). The difficulty of selecting effective programs and policies that
provide the greatest impact may be mitigated by outlining the complex set of causal pathways
that links risk factors to dental health. A causal map of the system dynamics is developed to
illustrate leverage points for potential health interventions that would reduce the burden of dental
health in older adults. Through the process of articulating causal relationships, the research team
interacts with dental practitioners and policy scholars in a participatory process of knowledge
sharing.
Adequacy of dental health in older adults requires attention to factors affecting tooth retention,
such as periodontal disease and tooth decay. Figure 1 maps the causal dynamics involved in
dental health. The reinforcing nature of the feedback loops in Figure 1 reveals their capacity to
destabilize the system into one or more vicious cycles. Because health issues tend to compound
with age, the reinforcing feedback loops outlined in Figure 1 can trigger vicious cycles of dental
health decline among older adults. Our conceptualization of these causal dynamics continues to
evolve through an ongoing collaborative and iterative group modeling process (Richardson and
Andersen, 1995; van den Belt, 2004).
range of food
choice
ability to chew
healthy food
choice <age> yy
4 f « preventive
cognitive chronic (a) a4 screenings
function ~~ illness
: v. health ‘2
oppete a a knowledge
prion Le community
oral ypiene access
metabolism activity social oral health
engagement promotion
(d)
physical social support
ability a
O_O ct
Figure 1. Reinforcing dynamics of dental health in older adults.
The left side of Figure 1 features individual-level factors that influence dental health, whereas
the right side reflects the level of dental health care available to the individual. Solid arrows
indicate proportional relationships, whereas dotted arrows indicate inverse relationships.
Specifically, the central reinforcing loop (a) indicates that chronic illness is often implicated in
dental disease, furthering the utility of preventive screenings for both medical and dental health
purposes. Chronic illness is more likely with increasing age, and frequently affects both physical
ability and cognitive function. Resulting impacts may make it harder for individuals to maintain
effective levels of oral hygiene activity, worsening dental health. Moreover, the variety of
medications used to manage chronic illness may manifest in the oral cavity, such as a dry mouth.
As a key contributor to both physical and mental health, nutrition (b) plays an important role in
the system dynamics of oral health in older adults. Older adults are particularly prone to have
missing teeth, which reduces their ability to chew foods thoroughly. Structural decline of teeth
with age also hampers mastication, or ability to chew, limiting the range of foods available for
consumption, and thus the capacity to make healthy food choices. Both the structure and content
of raw, unprocessed fruits and vegetables are particularly important for the maintenance of dental
health. Adequate nutrition enables both cognitive function and physical ability. If nutrition is
compromised, a frailty loop is triggered as metabolism slows from reduced physical activity,
decreasing appetite and therefore nutrition. At the intra-personal scale, improved physical ability
may accompany more diligent oral hygiene activity, self care that (like nutrition) enhances dental
health over time through retention of natural teeth and prevention of dental disease. Just as self
care improves dental health, certain harmful behaviors such as the use of tobacco and alcohol
have been shown to be risk factors for oral pathology.
As implied by the name of the ElderSmile program, an incentive to maintain healthy teeth is to
sustain the propensity to smile and interact with others. Dental problems such as missing teeth
and bad breath can inhibit social behavior. As such, one consequence of healthy tooth retention,
disease prevention, and improved overall dental health is self-confidence, which gives rise to an
increased propensity for social behavior, the social engagement term depicted in Figure 1. The
hatched arrow from dental health to social engagement indicates a time lag for shifts in social
behavior to result from the increased confidence that accompanies improved health.
Social support combines with individual ability to make transportation easier for older adults to
access preventive screenings and treatment centers, via a factor termed community access in
Figure 2. Community access expands the range of food choices available to older adults. Oral
health promotion (c), such as that made available through Columbia University’s ElderSmile
program, includes activities such as preventive screenings for the population. The social support
(d) that results from increased social engagement then provides a mechanism for spreading
awareness of oral health promotion programs such as ElderSmile. Such social support enables
opportunities for transportation of older adults who have physical limitations. Social support for
older adults is abruptly severed when a partner passes, for example. A loss of social support in
the home setting can make it hard to remember effective oral hygiene habits and further restricts
community access to nutritious food, especially if cognitive function has been weakened with
age, as in the common case of dementia. The latter condition can complicate otherwise routine
matters of denture care.
Consistent with the need for patient-centered (dental and medical) health homes (Glick, 2009),
oral health promotion involves dissemination of dental practitioner knowledge regarding the
unique considerations of older adults, such as how to treat patients with chronic illness (Lamster,
2004). Importantly, as word of mouth spreads about preventive screening and referral
opportunities, community knowledge grows and reinforces social support. The impact of oral
health promotion on dental health is mediated through timely treatment of tooth decay.
Depending upon the procedures necessary, the affordability of treatment may render it
prohibitive for individuals lacking coverage, a significant portion of the overall population of
older adults.
2.2 A STRUCTURAL VIEW OF DENTAL HEALTH DECLINE
For many adults, age induces a decreased sensitivity to tooth pain, resulting in longer delays
before symptoms of decay compel professional treatment. Figure 2 outlines the process by which
healthy teeth become decayed, filled, missing teeth and/or prosthetic teeth. Stocks (boxes) reflect
the prevalence of conditions, and flows (arrows with valves) reflect the incidence rates among
older adults.
Decayed set
Healthy |_sz_gy| Decayed Teeth with Missing
al >|
Teeth 4 Symptoms t
Filled Teeth lag Brosinete
Figure 2. A structural view of dental health decline among older adults.
The stock and flow structures mapped in Figure 2 are appropriate for formalization at the scale of
the individual, at the aggregate scale, or at the scale of a population subgroup (e.g., by
neighborhood or socioeconomic group). One-way arrows indicate irreversible flows: athough the
process of decay can be mitigated with fillings, it cannot be undone. Once decay has begun, teeth
cannot return to their original state of health. Moreover, filled teeth are prone to complications
and further decay, as indicated by the two-way flow between decayed and filled teeth. A
majority of older adults have teeth that are decayed, filled, or missing. Once teeth are missing,
prosthetic teeth may be sought, depending upon affordability of treatment as well as cultural
attitudes about the importance of prosthetic teeth. A key indicator of oral health among older
adults is therefore the retention of natural teeth.
Stock and flow structures inform the modeling at various scales. In addition to those implicated
in the diagnosis of dental health decline, stocks and flows are appropriate to simulate age cohorts
and anticipate demographic shifts in the population, and to distinguish between levels of risk
among subpopulations (Hirsch, 1975). An important factor affecting dental disease progression
and tooth loss is frequency and quality of dental care. Adults receiving regular dental care are
more likely to have decayed teeth diagnosed and filled before symptoms of pain develop, but
underinsured adults are likely to have teeth that progress into the symptomatic state. Consistent
with our multi-level conceptual framework, modeling participation in preventive screening and
social processes provides a contrast with biophysical processes of tooth decay, enabling
intersection of dynamics relevant at different scales of the system.
3, PREVENTIVE SCREENING INTERVENTION
The ElderSmile prevention centers are located at senior centers and other locations in which older
adults gather in Harlem and Washington Heights/Inwood. The prevention centers host a
combination of services, including: general presentations and discussions in both English and
Spanish of oral health promotion in later life (e.g., potential oral health problems, how to choose
oral health care products, and access to oral health care, including transportation issues);
6
demonstrations of brushing and flossing techniques and care of prosthetic devices; and oral
cancer and oral health examinations for seniors who elect to participate. Services are provided by
Columbia University’s College of Dental Medicine (CDM) faculty dentists and dental students.
Outreach sessions are conducted at community-based prevention centers by ElderSmile staff, two
of whom speak fluent Spanish. A bout five ElderSmile team members participate in each outreach
session, including a program coordinator, a health educator, dentists, and CDM dental students.
Numbers vary because of the availability of dental students who volunteer their time and services.
Older adults are enrolled in the ElderSmile program after participating in the oral health
promotion activities and being examined by one of the ElderSmile dentists.
Figure 3 outlines the reference mode for participation in ElderSmile preventive screening
activities, illustrating monthly participation rates over a 2-year period from 2006-2008.
80
60
Participants per Month
ee a ee ae
SPSS SERS SSS SLPF ES RES ERS SETESES
Figure 3. Participants in the ElderSmile preventive screening (PS) outreach program.
The oscillatory dynamic in Figure 3 may be analogous to a problem posed by Arthur (1999)
about the El] Farol bar in Santa Fe, New Mexico. The bar has a weekly Irish night. If the
potential Irish bar-goers (such as Arthur) wish to avoid a crowd and anticipate its likelihood
based upon recent experiences, oscillatory behavior results from commonly held expectations
about the system. However, the capacity for the El Farol dynamic is fixed, based upon the size
of the bar and the constant (weekly) frequency of the Irish night. For the ElderSmile program,
workshops are not evenly spread throughout the year, and they are furthermore distributed
among the 27 participating senior centers in upper Manhattan. Figure 4 illustrates the
relationship between monthly participation and workshop availability between 2006 and 2008.
Variation in workshop utilization indicates that resources may be fully utilized or underutilized.
Z
Further analysis will consider the location-specific variation, accounting for differences among
senior centers participating in the ElderSmile program.
90
80
70
Participants per Month
20
10
oe
oe
y =11.776x +3.2331
R? =0.4523
Figure 4. Monthly utilization versus availability of ElderSmile preventive screenings (PS).
3
Workshops per Month
The descriptive statistics in Table 2 for average DMFT (Decayed Missing Filled Teeth) among
participant subgroups reveal disparities by age, gender, ethnicity, education, and smoking history.
Differences between missing (MT) and filled (FT) teeth reflect accumulations of advantages and
disadvantages over the life course, with filled teeth reflecting some degree of access to care.
Table 2. Means and Standard Errors (SE) of Decayed Filled Missing Teeth (DMFT), Decayed Teeth (DT), Missing Teeth (MT),
and Filled Teeth (FT) by Age, Gender, Race/ Ethnicity, Education Level, and Smoking History of Older Adults Aged 65 Years and
Older (N = 662) Who Participated in Community-Based Oral Health Examinations Conducted by Dentists: The ElderSmile
Program, New York, NY, 2006-2009
Characteristic DMFT (SE) | p-value | DT(SE) | p-value | MT(SE) | p-value | FT(SE) | p-value
Age in years
65-74 20.6 (0.3) 0.19 0.9 (0.1) 0.66 14.2 (0.5) 0.05 5.5 (0.3) 0.04
75+ 21.2 (0.3) 1.0 (0.1) 15.6 (0.5) 4.7 (0.3)
Gender
Men 20.5 (0.4) 0.14 1.4 (0.2) <0.01 13.9 (0.6) 0.04 5.2 (0.4) 0.66
Women 21.2 (0.3) 0.7 (0.1) 15.4 (0.5) 5.0 (0.3)
Race/ ethnicity
Non-Hispanic White 20.4 (0.5) 0.09 1.2 (0.3) 0.13 8.9 (0.9) <0.01 10.4 (0.7) <0.01
Non-Hispanic Black 21.2 (0.4) 1.0 (0.1) 16.4 (0.6) 3.9 (0.3)
Non-Hispanic Other 18.4 (1.2) 1.3 (0.4) 10.4 (1.5) 6.8 (0.8)
Hispanic 21.1 (0.4) 0.7 (0.1) 16.1 (0.6) 4.2 (0.3)
Education level
Less than high school 21.7 (0.5) 0.02 1.0 (0.2) 0.89 17.4 (0.7) <0.01 3.4 (0.3) <0.01
High school 21.0 (0.4) 0.9 (0.1) 15.5 (0.6) 4.6 (0.3)
More than high school 20.0 (0.4) 0.9 (0.1) 11.4 (0.7) 7.7 (0.4)
Smoking history
Current smoker 23.0 (0.7) <0.01 1.8 (0.5) <0.01 18.2 (1.2) <0.01 3.0 (0.6) <0.01
Former smoker 22.0 (0.4) 1.0 (0.2) 15.8 (0.7) 5.2 (0.5)
Never smoked 20.0 (0.3) 0.7 (0.1) 13.5 (0.5) 5.7 (0.3)
Total 20.9 (0.2) 0.9 (0.1) 14.9 (0.4) 5.1 (0.2)
Racial disparities are manifest in the increased prevalence of decayed and missing teeth among
African American and Hispanic participants in the ElderSmile screenings, at a rate nearly double
that of their white counterparts, whereas the latter had more teeth with fillings (Table 2). These
disparities reflect accumulations of advantages and disadvantages over the life course, such as
insurance coverage and access to quality care. Nearly all of the participants of the ElderSmile
program required dental treatment at neighborhood sites, and the vast majority (88%) of those
referred have followed up with treatment, demonstrating the effectiveness of the program as well
as the widespread need for such services (Northridge et al, 2011).
The utilization of preventive care and treatment is driven by the accessibility and affordability of
services. The region of northern Manhattan served by ElderSmile is a U.S. federally designated
manpower shortage area. Disparities in oral health and health care have been documented for
seniors in the study region as compared to the U.S. senior population overall. Only 23% of
seniors in Harlem and Washington Heights/Inwood visit a dentist annually, compared to 50% of
seniors nationally. Among the U.S. population at large, 70% visit the dentist at least once a year,
significantly more than those who visit a doctor within the past year. Among the older adults in
upper Manhattan participating in the ElderSmile program, this relationship is reversed: only 45%
visited the dentist during the past year, whereas doctor visits are more frequent. The utilization of
dental care tracks strongly with the availability of dental insurance, as only 48% of ElderSmile
participants had dental insurance (Marshall et al, 2009).
Many adults develop edentulism, or a complete absence of permanent teeth in the mouth. In
northem Manhattan, 45% of seniors have no teeth, compared with 25% of seniors nationally. For
older adults with missing teeth, prosthetic replacements may not be sought for reasons of
affordability. While the Medicaid program in New Y ork State offers dental coverage, only 20%
of enrolled ElderSmile participants access dental services through Medicaid, and only 10% of
private dentists provide over $10,000 in Medicaid dental services (CDM, 2006). The relationship
between age and tooth loss is apparent in data from the ElderSmile program: 17% of participants
aged 65-74 years were edentulous, whereas 22% of adults over 75 years had the condition
(Northridge et al, 2011). Among social networks in which the absence of teeth is a norm, the
need for prosthetic teeth may be questioned as an unnecessary expense in later life. While the
logic that ‘you can eat without teeth’ is prevalent among many, including some budget-minded
policymakers who view dental health as a luxury, a systems perspective reveals the
compounding impacts of neglecting oral care.
3.1 MODELING WORD OF MOUTH
Because this project is designed to inform oral health interventions, the model centers on the role
of preventive screening (PS) outreach programs in assisting seniors with securing dental
treatment. The stocks in Figure 5 distinguish broad subgroups of the population by whether or
not older adults have participated in the preventive screenings and treatment opportunities
offered by the ElderSmile program. In this representation, a reinforcing word-of-mouth
mechanism (R1) enables knowledge diffusion about preventive screening events. Although
desired PS participation grows rapidly, it is tempered by perceived availability (B1) when
screening sessions are at capacity. Although PS capacity constrains participation (B2), it is
adjusted (R2) to accommodate increasing interest.
perception delay
total
= population
ld
perceived PS
availability marketing effect
peer contact rate
PS availability WOM efect
¢ B1 ——
4 an desired PS :
indicated PS
ultlzation @)
~
participation
_wPS capacity
Communicators
visits wow decay
Gap
wom day fraction ier
time
Screened PS Treated PS
participation | Participants treatment, Participants
intial number of PS. ~
participants
Jno treatment treatment
MS feetsion ~~
Capaciiydeley Potential PS
Participants
PS capacity
utilization
( ia
\
utilization
function
treatmentdelay treatment
cumulative : ecien
surveys
Figure 5. Structural dynamics of preventive screening (PS) intervention.
The model structure in Figure 5 accommodates the El Farol dynamic described above through
the use of delayed adjustment of capacity and perceived availability via B1 and B2. Desired
participation is diminished with perception of unavailability, as with the crowd deterrent on
attendance at El Farol’s Irish night. Balancing feedback with delays can induce such oscillation.
Figure 6 illustrates a range of weekly participation outcomes induced by variation in the
perception and capacity delay times relative to the base run, which has a capacity delay of 24
weeks and a perception delay of 4 weeks. Using a minimum 2-week delay for both, capacity
adjustment time is varied to 48 weeks (doubling the base run delay), and perception adjustment
time is varied to 16 weeks (quadrupling the base run delay). These delays induce a range of
transient behavior before weekly participation rates approach a constant equilibrium.
50% 75% (I 95% [I 100% (a
participation
20
17
0 39 78 117 156
Time (week)
Figure 6. Effects of varying capacity and perception delays on weekly participation.
10
In Figure 7, the same sensitivity analysis is performed as in Figure 6, with the inclusion of a
marketing boost after one year (52 weeks). Doubling the marketing effect ultimately results in a
doubling of participation rate (note the shift in scale on the y-axis). The shape of the plot
indicates that the marketing effect thereby crosses a bifurcation threshold, enabling an approach
toward a higher equilibrium rate of participation in preventive screenings.
50% 75% (NN 95% I 100% (a
participation
40
30
0 39 78 117 156
Time (week)
Figure 7. Participation effect of a marketing boost at one year while varying delays.
Figure 8 reveals a range of outcomes from varying contact rate, and therefore the amplitude of
the word of mouth effect, from zero to four times the base run setting (0.04/week).
50% 75% NIN 95% I 100%
participation
40
30
20
10
0 39 78 117 156
Time (week)
Figure 8. Effects of varying contact rate on PS participation.
11
Significantly greater contact rates induce participation beyond the rates implied to be feasible via
analysis of ElderSmile workshop offerings and utilization (as per Figure 4 above). Because the
range of simulation outcomes indicated in Figure 8 varies significantly with contact rate, a
consideration of heterogeneity among social networks of older adults is warranted.
Social networks may be empirically specified or simulated to test alternative mechanisms for
developing social ties and communicating. Agent-based representation is appropriate in cases
where a particular social network structure deviates from the random mixing implied by a
continuous stock-flow structure (Rahmandad and Sterman 2008). Figure 9 maps a simulated
proximity-based social network across residential locations of the ElderSmile participants, using
AnyLogic software to integrate GIS data within an agent-based model.
Thm
Figure 9. Simulated agent-based network among ElderSmile participants.
The network structure in Figure 9 is used to explore the relationship between social and physical
distance among participating older adults. By virtue of their presence at senior centers,
ElderSmile PS participants exhibit a greater physical and social mobility than that of the broader
population of older adults in upper Manhattan. Such activity may explain why the dental health
of participants is better than national averages in terms of tooth retention, as indicated by lower
levels of edentulism (Northridge et al, 2011). Questions about social mobility and geographic
distance include: Do participants who come from farther away have greater influence on others?
Do they reflect the outcome of a strong word of mouth dynamic? Is the El Farol capacity and
perception dynamic actually experienced in the ElderSmile program? We continue to explore
these questions using the process of participatory systems modeling. Extensions of this research
will explore the variability of social dynamics in both agent-based and stock-flow formulations,
and will examine the utility of individual-level representations of dental health decline.
12
4, CONCLUSION
Ongoing efforts to refine and reform systems of health care for older adults may be meaningfully
abetted by modeling that explicitly links dental health to chronic illness, community access, oral
health promotion, and nutrition. Education and outreach programs help promote effective oral
hygiene practices and routine care, so as to retain natural teeth as long as they are functional, and
to keep them healthy. Oral health promotion also enables health education and early
identification of tooth decay and co-morbid conditions. Once the progression of dental decay has
started, it can be mitigated or reinforced, but not reversed. This intrinsic irreversibility
underscores the importance of regular preventive dental care throughout the life course.
Coordinated intervention efforts minimize the impact of oral diseases and conditions prevalent
among older adults so as to improve their quality of life. A systems perspective for oral health in
older adults helps to identify interventions as leverage points that are effective and cost-saving.
Use of the system dynamics methodology has enabled the research team and project stakeholders
to explicitly recognize, discuss and modify feedback relationships relevant to treatment of oral
health among older adults. A number of reinforcing dynamics that deteriorate oral health are
induced with age, and yet inadequate insurance coverage, cultural beliefs and daily routines often
preclude preventive community care.
Outreach programs such as evaluative preventive screenings generate useful opportuities to share
critical information between participants and public health providers seeking to mitigate
deterioration of dental health. Modeling to address dental health should consider educational
incentives to enhance resource capacity for dental hygienists and dentists to improve quality of
life for older adults as a growing and underserved population. With the benefit of system
dynamics, policymakers and program managers sharpen their sense of how to allocate resources
and time interventions for improving oral health and health care for older adults. Our ongoing
analysis centers on preventive screening interventions that provide a range of services including
transportation, education and instruments for effective oral hygiene, and referrals for full dental
treatment.
13
REFERENCES
Allukian M, Horowitz AM. Oral Health. In Levy BS, Sidel VW (eds.) Social Injustice and
Public Health. New Y ork: Oxford University Press, 2006, pp. 357-377.
Arthur WB. Complexity and the economy. Science. 1999;284: 107-109.
Bachrach CA, Abeles RP. Social science and health research: growth at the National Institutes of
Health. AmJ Public Health. 2004;94: 22-28.
Columbia University College of Dental Medicine (CDM). ElderSmile, 2006.
Glick M. A home away from home: the patient-centered health home. J Am Dent Assoc.
2009;140: 140-142.
Hirsch G. Examining Alternatives for Improving the Nation’s Oral Health: A System Dynamics
Model of the Dental Care Delivery System. Cambridge, MA: Pugh-Roberts Associates,
1975.
Homer J, Hirsch G. System dynamics modeling for public health: background and opportunities.
AmJ Public Health. 2006;96: 452-458.
Institute of Medicine (IOM). The Future of the Public’s Health in the 21% Century. Washington,
DC: The National Academies Press, 2002.
Lamster IB. Oral health care services for older adults: a looming crisis. Am J Public Health.
2004;94:699-702.
Lamster IB, Northridge ME (eds.) Improving Oral Health for the Elderly: An Interdisciplinary
Approach. New Y ork, NY : Springer, 2008.
Marshall S, De La Cruz L, Northridge ME, Vaughan RD, O’Neil-Dunne J, Lamster IB.
ElderSmile: a comprehensive approach to improving oral health for older adults. Am J
Public Health. 2009;99:595-599.
Northridge ME, Sclar E, Biswas P. Sorting out the connections between the built environment
and health: a conceptual framework for navigating pathways and planning healthy cities.
J Urban Health. 2003;80:556-568.
Northridge ME, Ue F, Borrell LN, De La Cruz L, Chakraborty B, Bodnar S, Marshall S, Lamster
IB. Tooth loss and dental caries in community-dwelling older adults in northem
Manhattan. Gerodontology. 2011; in press.
Rahmandad H, Sterman J. Heterogeneity and network structure in the dynamics of diffusion:
comparing agent-based and differential equation models. Manag Sci. 2008;54:998-1014.
14
Richardson GP, Andersen DF. Teamwork in group model building. System Dynamics Review.
1995;11(2): 113-137.
US Department of Health and Human Services (US DHHS). Oral Health in America: A Report
of the Surgeon General. Rockville, MD: National Institute of Dental and Craniofacial
Research, 2000.
van den Belt, M. Mediated Modeling: A System Dynamics Approach to Environmental
Consensus Building. Washington, DC: Island Press, 2004.
15
