Examining Household Asthma Management Behavior Through a Microeconomic Framework

Limits of the Classic Health Behavior Model

Key elements of asthma management, particularly proper pharmacotherapy and avoidance of triggers, are largely under control of patients and their family members (Clark, Jones, Keller, & Vermeire, 1999). As a result, the primary strategy to improve asthma management has been through educational interventions, based on the premise that knowledge of asthma management is the necessary component to reduce asthma morbidity (Guevara, Wolf, Grum, & Clark, 2003). Asthma educational programs have ranged from clinic based to community based (R. Evans et al., 1999; Garrett et al., 1994; Greineder, Loane, & Parks, 1999; Krieger, Takaro, Song, & Weaver, 2005; Lewis, Rachelefsky, Lewis, de la Sota, & Kaplan, 1984; Madge, McColl, & Paton, 1997). As detailed in the excellent overview by Clark and Valerio (2003), there are myriad behavioral models that serve as the basis for asthma management programs (Clark & Valerio, 2003), including social cognitive theory (Clark et al., 1986; D. Evans et al., 1987), self-efficacy (Toelle et al., 1993), social support and networks (Shah et al., 2001), and self-monitoring (Kelly et al., 2000; Madge et al., 1997). However, as evidenced by the persistent and increasing inequality in asthma-related morbidity in groups within urban communities (Akinbami, LaFleur, & Schoendorf, 2002; Akinbami & Schoendorf, 2002; Shapiro, & Stout, 2002), these behavioral models have not demonstrated meaningful improvements in asthma management. We believe this is due to two reasons.

First, for those interventions that have focused on increasing knowledge of asthma management in affected populations, it is widely recognized that “knowledge is not necessarily followed by the appropriate action” (Blessing-Moore, 1996). Several recent systematic reviews, meta-analyses, and interventions have suggested that school-based asthma education programs are related consistently to positive changes in knowledge, self-efficacy, and self-management, but have been inconsistently associated with health endpoints, such as respiratory symptoms, and indicators of morbidity, such as school absence (Coffman, Cabana, Halpin, & Yelin, 2008; Coffman, Cabana, & Yelin, 2009; Wolf, Guevara, Grum, & Clark, 2003). It is likely that educational interventions often fail to result in improvements in asthma-related morbidity because these programs affect one element in a complex system characterized by multiple determinants of behavior.

Second, although behavioral interventions have been implemented based on the theoretical existence of certain causal pathways, and have demonstrated value in measuring intention, there has been little empirical evidence of the utility or validity of these models with respect to outcome evaluation (Plotnikoff, Costigan, Karunamuni, & Lubans, 2013). The aforementioned theories based on cognitive perspectives assume that attitudes, beliefs, expectation of events and outcomes are major determinants of health-related behaviors (Munro, Lewin, Swart, & Volmink, 2007) but have demonstrated limited association with intervention effectiveness (Plotnikoff et al., 2013; Prestwich et al., 2013; Ramirez, Hodges Kulinna, & Cothran, 2012). As Mermelstein and Revenson (2013) state, “the health literature is replete with studies where strong correlational findings . . . fail to translate to effective interventions and more nuanced theory.”

In addition, there is no one unifying theory that provides quantitative descriptions or forecasts as to how people will behave vis-à-vis asthma management. It is unclear whether existing health behavior models can (1) explain heterogeneity in behavior; (2) provide valid, quantitative predictions of future behavior; or (3) formulate empirical, testable, hypotheses. Moreover, no single theory or conceptual framework provides a dominant paradigm for health behavior research or practice (Glanz & Bishop, 2010; Glanz, Rimer, & Lewis, 1997). Therefore, the mechanisms by which current interventions operate—especially those focused on behavioral change—exist largely in a black box. This issue is not disease specific and pervades nearly all of public health practice. Successful implementation of public health interventions requires understanding of the operational link by which provision of intervention resources translate into positive health outcomes.

Introduction to Economic Theory

The main premise of microeconomic theory is that individuals act to maximize their welfare (i.e., happiness, well-being), given constraints on income and time. Grossman (1972) first extended this concept to examine how individuals choose to invest in their health (Grossman, 1972). In short, individuals’ behaviors or choices produce a level of health; the relation between the investments and the level of health outcome can be thought of as a health production function.

In the household health production function, individuals produce the commodity “health” by combining their own time and effort with purchased goods, such as medications, health care services, healthful foods, and so on (Clemmer, Kenkel, Ohsfeldt, & Webb, 1994). For a household to achieve a desired goal, a family may evaluate its current levels of consumption and shift its available resources (i.e., time, income) in order to achieve that goal. This theory relies on the stipulation that health, rather than considered a predetermined state of being or an outcome imposed on the individual, is affected by the individual’s behavior. Therefore, an individual can make trade-offs between goods to maintain or improve their health and goods for other purposes. Observation of the trade-offs that individuals make as they choose among various combinations of health and other goods indicates the relative value they place on health (Cropper & Freeman, 1991) or their priority for health compared with other goods available. For instance, a caretaker might choose to allocate resources to a “symptom-free day” for a child by using a HEPA (high-efficiency particulate air) vacuum and cleaning supplies along with time and effort to clear a home environment from asthma triggers. These investments require a trade-off between current happiness or pleasure and other choices for improvements in future well-being (Cutler & Glaeser, 2005).

Budget Constraints

Although it would be optimal to consume all goods that provide utility, there are practical limits to an individual’s utility function, namely, that people do not have infinite funds or time to obtain all of their preferred goods and services. Economists incorporate a pragmatic element into this optimization framework—a budget constraint. Given individual preferences and a limited pool of funds and a limited amount of time, individuals can substitute, or trade off, goods or actions to maximize utility. Trade-offs may exist between goods or between actions, and these investments can change over time based on an individual’s preference. Given an equal price of a good, or an equal investment in time, an individual would choose a good or action based solely on the ranking of his/her preferences. However, in situations where the price or time investment for a desired good is beyond the limits of the budget, the complexity of an individual’s system of trade-offs increases: Goods and time investments would need to shift to accommodate preferences (Figure 1).

OPEN IN VIEWER

Figure 1.

(A) Basic and universal monetary and temporal aspects of the budget constraint and how a family might shift resources, given asthma control scenarios. (B) More frequent use of a HEPA (high-efficiency particulate air) vacuum to reduce indoor allergens would increase the time devoted to household responsibilities and reduce leisure time. (C) A household decision to purchase controller medication may reduce budget devoted to clothing and also increase child care time for daily medication administration. (D) A preventive care visit to a primary care provider may result in a caregiver missing work; in the case of hourly employees, this could result in an overall reduction in the monthly monetary budget. However, several costs (e.g., housing) are fixed and would comprise a larger part of the budget, reducing money for other basic needs.

This concept can be applied to interventions that seek to improve indoor environmental quality for children with asthma. Case managers or community health educators visit with families to identify potential triggers for asthma and teach practical mitigation methods. If households enrolled in the program do not currently follow the program recommendations, families would either have to trade off time allocated to another activity to accomplish the environmental remediation tasks or trade off money allocated to other goods and services to have resources to pay for these mitigation services. Although many environmental remediation programs provide equipment to the families (e.g., HEPA vacuums, mattress, and pillow covers), these programs do not provide increases in the time allotment or monetary budgets for families, or maintenance, or replacement when needed. Furthermore, there are no studies that reveal how a household may shift resources to accomplish these activities without discussion of what other activities, goods, or services will be forgone. As both financial and temporal resources are limited in low-income families, knowledge generated through participation in environmental remediation programs may not be actualized if trade-offs are not feasible. For example, recommended wash instructions for bedding would be more practicable if families did not have to travel to a laundromat and could complete tasks in home. Given increases in income and/or time, a household’s options for engagement in asthma management activities would increase.

Economists have examined the influence of budget constraints on recommended physical activity levels, hypothesizing that time costs may be an important barrier to regular exercise. Using data from the National Health and Nutrition Examination Surveys, Meltzer and Jena (2010) found that individuals in the highest income group exercised an average of 7 more hours per month than individuals in the lower income group, and that exercise intensity (measured in metabolic equivalents) increased with income levels. These findings suggest that increasing exercise intensity has the capacity to reduce the time costs of the investment in health. From the microeconomic perspective, time investments in healthy behaviors can be reduced by substitution of exercise intensity with amount of time committed to exercise. With regard to asthma, the time costs necessitated by disease management strategies may be reduced with efficient use of care resources. For example, the development of school-based health centers in communities with high asthma prevalence may reduce both time and budget commitments for health care encounters, in addition to reducing missed school days and creating educational opportunities for asthma management (Webber et al., 2003).

Welfare Maximization: An Example

Application of the optimization framework can be illustrated by examining asthma-related ED visits, a common indicator of mismanaged asthma. The underlying premise of many community-based asthma interventions is that reliance on emergency care in urban populations interferes with development of partnerships between clinicians and caretakers and limits opportunities for asthma education (Halfon, Newacheck, Wood, & St Peter, 1996; Lozano, Connell, & Koepsell, 1995; Warman, Silver, McCourt, & Stein, 1999). Children enrolled in public insurance programs disproportionately use the ED for asthma care compared with children with private insurance (Flores et al., 2009; Jones, Lin, Munsie, Radigan, & Hwang, 2008; Liu & Pearlman, 2009; Smith, Wakefield, & Cloutier, 2007). The lack of utilization of primary care can result in improper medication use among urban children (Celano et al., 2010), absence of written self-management plans (Findley et al., 2003; Haby, Powell, Oberklaid, Waters, & Robertson, 2002), and inadequate information about asthma triggers and management (Haby et al., 2002). As such, reduction in the number of ED visits is a primary outcome of asthma management interventions (Boyd et al., 2009). From a public health perspective, the optimal model of asthma management is the establishment of an ongoing partnership between a patient’s family and a primary care physician. Once informed about proper management techniques, families would decrease their reliance on the ED.

However, when examining this problem through an optimization framework, treatment of asthma in an outpatient emergency setting may be rational and efficient for these parents. Asthma management is time and resource intensive; low-income families may view the incremental daily cost of these activities as inefficient. Furthermore, the lag between starting inhaled corticosteroids and experiencing their potential benefits may lead some families to underestimate the future benefit of consistent use. Stated from an economic perspective, while multiple ED visits are often viewed as a red flag of mismanaged asthma, parents may view utilizing the ED as maximizing efficiency. In the ED, children receive expeditious care in a setting where acute asthma exacerbations are frequently seen and effectively treated. In contrast, the use of inhaled corticosteroid is a costly investment (requiring both payment and time) with uncertain future benefits. Coupled with evidence that primary care services are often difficult to access in underserved areas (Baren et al., 2001; Scarfone, Zorc, & Capraro, 2001; Zorc et al., 2003), and managing chronic disease through the ED may be a community norm (Farber, Johnson, & Beckerman, 1998), efforts to educate families and promote prevention behavior may not coincide with families’ notion of optimization. Empirical evidence supports this hypothesis. Low-income parents have reported the ED as the preferred source of care because of difficulties in attainment of urgent advice or appointments with the primary care physician; parents also reported lack of access to transportation, child care, and inflexible work schedules that preclude primary care physician visits for asthma treatment (Fredrickson, Molgaard, Dismuke, Schukman, & Walling, 2004). We may also infer that the monetary costs associated with ED utilization for asthma exacerbations are largely hidden to the household, whereas the costs associated with primary care treatment (e.g., eligibility and availability of affordable insurance, ability to meet copay requirements) may exceed a household’s budget constraints. We include an additional example of asthma management behaviors and the economic constructs described in Figure 2. This schematic emphasizes not only the household-level factors necessary for decision making (i.e., preferences, budget constraints, and time/risk preferences), but the milieu in which these decisions are made, which is a critical component of models of interpersonal behavior.

OPEN IN VIEWER

Figure 2.

A welfare optimization approach that would lead to a decision to engage in an asthma management behavior (e.g., decision to purchase a controller medication). The system includes household-level preferences that are influenced by household-level factors (budget constraints, risk tolerance, and discount rate), which influence and are influenced by macro-level factors (health care system) as well as personal beliefs about treatment. The components of the welfare optimization approach are similar to the multiple factors included in the ecological model(s) of health behavior.

Joint Ventures

Despite its flexibility and intuitive appeal, the economic model is not a panacea for amelioration of the current level of asthma-related morbidity. Though the concepts of trade-offs and individual choice are central to classic economic theory and the household productive function, contemporary economic research has not explored the choices available to households in pursuit of good health, particularly in urban and low-income communities. For example, two individuals may be given the same information on the use of an asthma inhaler, a purchased good. However, if one individual has a greater amount of education, or greater access to pharmaceutical advice, the medication would be more likely to be effective in that individual compared with the other. Though the cost of an educational program might be the same for individuals, their ability to use personal skills and resources to produce health may be vastly different. Health behavior theories recognize the inherent psychological and behavioral heterogeneity among people: An individual may have a different reserve of health efficacy compared with another, may be at a stage of change closer to making a behavioral modification, or hold health beliefs that make them more likely to not only engage in a post–asthma prevention or management behavior, but more efficiently produce a positive health outcome. We are encouraged by examples in the literature that reflect collaborative efforts among clinicians, epidemiologists, and economists to implement economic models to address behavioral change. For example, in a randomized controlled trial to promote smoking cessation, Volpp et al. (2009) provided $100 for participants who completed a smoking-cessation program, an additional $250 if cessation continued 6 month after program completion, and an additional $400 if cessation continued after 6 additional months. Compared with the usual care group, who only received education, the intervention group participants were significantly more likely to participate in a smoking-cessation program and continue cessation in the two time periods after program completion (Volpp et al., 2009). Given long-term health care costs of smoking (Manning, Keeler, Newhouse, Sloss, & Wasserman, 1989), the upfront costs of financial incentives are relatively negligible. This intervention model is currently being tested by Medicaid programs in several states across the country (National Association of Chronic Disease Directors, 2011). These types of interdisciplinary approaches have also been developed for the infectious disease literature, where incorporation of an economic risk-response framework into an epidemiological model for infectious disease dynamics improved model accuracy (Klein, Laxminarayan, Smith, & Gilligan, 2007). Furthermore, economic models of time and risk preference have recently been mapped onto more traditional health behavior frameworks, such as the transtheoretical model (Leonard et al., 2013).

Behavioral economics represents a transdisciplinary approach that implements developments in cognitive psychology to refinement of the microeconomic framework, particularly regarding the way individuals use information and consider alternatives. ² Prospect theory and loss aversion are examples in which the economics of incentives is combined with insight from psychology on how people behave in real-world situations (Rice, 2013). In addition, the theory of utility maximization assumes that an individual will consider all information available in decision making; behavioral economics suggests that individuals use general “rules of thumb” or heuristics to avoid information overload that comes with an abundance of information. The use of heuristics may not result in welfare maximization, but a decision that will be sufficient for current needs. Though behavioral economics has been lauded as a potential solution for a wide array of public health issues (Thaler & Sunstein, 2008), several economists suggest that behavioral economics should be considered a complement to, rather than a replacement for, broad-based public health policies (Lowewenstein & Ubel, 2010; Oliver, 2011; Thaler & Sunstein, 2008).

Whether an investigator’s theoretical training falls under the health behavior models or the microeconomic framework, there is overall consensus that context is a key constituent for health behaviors. Complex health problems are largely embedded within systems and cultures with dynamic processes that both affect and are affected by individual behavior (Mermelstein & Revenson, 2013). Ignoring social factors and exposures lead to an overemphasis of individually focused interventions that may not result in sustainable improvements in population health (Link & Phelan, 1995). Coupled with limited applicability of carefully constructed health behavior trials in community settings,(Cohen, Scribner, & Farley, 2000; Susser, 1995), the ecological perspective has reemerged as a guiding principle for behavioral interventions (Glanz & Bishop, 2010; Richard, Gauvin, & Raine, 2011). The complex social dynamics that have resulted in the current burden of asthma morbidity requires the application of a broad array of social and behavioral sciences that result in feasible and relevant solutions within communities (Livingood et al., 2011). As such, contemporary models in health promotion have incorporated concepts of social ecology in the extant health behavior frameworks (Best et al., 2003; Burke, Joseph, Pasick, & Barker, 2009; Samuel, Commodore-Mensah, & Dennison Himmelfarb, 2013).

Conclusion

Asthma morbidity and mortality are, in theory, largely preventable when patients and families are educated about the disease and have access to high-quality health care (National Heart Lung and Blood Institute, 1997). However, despite the evidence that asthma control is an achievable outcome, poor disease management for children living in low-income urban communities persists (Global Intiative for Asthma, 2002; National Heart Lung and Blood Institute, 1997).

The causal pathway(s) that underlie asthma-related morbidity are complex and involve choices and behaviors in multiple domains and on multiple levels, including individuals, households, schools, health plans, and local, state, and federal governments. Implementation of policies in the United States (U.S. Department of Health and Human Services & Centers for Medicare and Medicaid Services, 2009; U.S. Department of Labor, 1993) and the state level, such as statutes that permit students to carry inhalers in schools (Jones & Wheeler, 2004) have improved the range of options for asthma patients to engage in positive management behavior. Furthermore, community-based interventions that have included a diverse set of stakeholders and focused on policy and systems change have demonstrated decreases in asthma-related morbidity among target populations (Clark et al., 2013) as well as societal cost savings (Bhaumik et al., 2013). However, strategies for successful intervention efforts that require individuals to be active and constantly involved have not been well defined. Much like other current chronic diseases that are burdens in terms of morbidity and cost, asthma requires day-to-day management in an unpredictable world. This unpredictability makes it difficult to communicate effective management strategies to families and to design paternalistic-style interventions that would have a population-level morbidity impact.

The microeconomic perspective offers an empirically testable framework in which to understand the burden of asthma-related morbidity in urban communities. The concepts of preferences, trade-offs, and the production of health by a combination of inputs are not only intuitively appealing, but can be expressed quantitatively and continuously refined. This perspective assumes that families seek to minimize asthma-related morbidity for their children, but the production of good health competes with other needs of the family for both time and income. Health, as an input to overall welfare or happiness, competes with other needs and preferences of a household for both monetary and time investments. This model may better reflect a household’s perspective of realistic components to maximize welfare. If asthma symptoms are not present, urban households may be able to focus on other needs, such as housing, schooling, safety, food security, as well as leisure. These competing needs are infrequently addressed or included in the public health perspective, particularly in asthma management intervention evaluations. Intervention studies have often failed to recognize and operationalize the concept that socially disadvantaged families living in urban settings have a restricted choice set, that is, the trade-offs that families can make to maximize welfare are constrained. Furthermore, despite a desire to improve the health status of a child with asthma, intervention studies are not designed to respond to the competing needs of a household. Welfare maximization may relate to meeting basic needs (e.g., food or shelter) as opposed to improving the health status of a child with asthma. Identification of these choices and priorities made by families should form the basic design of interventions to address asthma morbidity, especially in low-income urban communities. The consideration of health as an input to overall welfare, rather than equal to welfare, may be an opportunity to better understand the restricted choice sets urban households encounter and promote programs or policies to either reduce time and investment on other household needs or improve the efficiency of health production. This concept is readily illustrated by data analysis techniques employed in economics. Rather than analyzing a health state as the dependent variable, in economic analysis, health is a right-hand side variable to predict general overall happiness as health competes with other needs of the family from both a monetary perspective and a time perspective.

As the preponderance of asthma epidemiology is currently conducted through questionnaires, incorporation of the economic framework into asthma surveillance is a logical place to establish joint research endeavors. Asthma educational programs targeted at adolescents may focus on understanding what outcome participants would like to maximize, for example, participation in athletics or control of their life and disease, as well as identification of competing preferences. Questionnaires such as the National Asthma Survey can readily incorporate elements, such as preference ranking through discrete choice experiments to better understand the barriers to effective asthma management and control. DaVanzo and Gertler’s (1990) illustrative list of types of data to collect to define the health production function in the developing world could be readily adapted by economists and epidemiologists to address questions of resource uptake for household asthma management behavior (Table 1). In summary, the public health community must realign our understanding of asthma-related morbidity to the communities most affected by this disease. We must also recognize that health is one component of overall welfare of a household. Reductions in asthma-related morbidity may not derive from asthma-specific interventions, but an overall reduction in other stressors and challenges a household may face, in order to focus more resources on the production of health. The consideration of economic perspectives to understand household-level decisions related to asthma management and care may offer a new pathway to understand the underlying causal behavioral mechanisms related to asthma management and control.

OPEN IN VIEWER

Table 1.

Select Economic Questions for Asthma Surveys From DaVanzo and Gertler (1990).

Health status of all household members, including,

• Morbidity

• Chronic conditions

• Nutritional status

• Days lost from work or normal activities

Health behaviors and inputs, including,

• Utilization of and expenditures (money and time)

• Health habits (washing hands, smoking, using drugs, and alcohol)

• Diet of household members

Influences on health behaviors, including,

• Amount and sources of income and household resources and control over their use

• Access to credit and to other resources outside the household

• Value of household members’ time (i.e., wage rates for workers in the formal sector)

• Household composition

House characteristics

• Community-level data on the availability, prices, and type of health services offered

• Community characteristics (such as proximity to major transportation)