Eating Habits and Developing High

¶ … Eating Habits and Developing High Cholesterol Levels

Coronary heart disease (CHD) remains the leading single cause of death in the United States today, and elevated serum cholesterol is widely recognized as being the risk factor responsible for myocardial infarction and CHD death; furthermore, a growing body of research supports the acceptance of hypercholesterolemia being as a causal and treatable agent in coronary artery arthersclerosis (Baum, Jennings, Manuck & Rabin, 2001).. According to statistics released by the National Institutes of Health (NIH), approximately half a million Americans die each year from coronary heart disease (Adams & Jennings, 1993). To date, researchers have determined that controllable risk factors such as the level of physical inactivity, smoking, overweight or obesity, high blood pressure, high blood cholesterol, and diabetes are all major influences on the development and severity of heart disease (Meadows, 2003). While coronary heart disease (CHD) is a complex, multifaceted health problem, there has been increased attention focused on dietary cholesterol and saturated fats as factors that also contribute to elevated blood cholesterol levels; in this regard, elevated blood cholesterol, specifically LDL (low density lipoprotein, the so-called "bad cholesterol") cholesterol, can lead to arteriosclerosis (a narrowing of the arteries that slows or blocks the flow of blood) and greatly increases the risk of heart attack (Adams & Jennings, 1993; Ulrich, 2002).

High blood pressure and elevated lipids represent a lethal combination; in fact, it has been estimated that fully 80% of those with hypertension also have high cholesterol levels (or hypercholesterolemia) and that 50% of these individuals will require some type of medication in order to lower their cholesterol to safe levels (Griffith & Wood, 1997). According to these authors, "For the most part, high blood pressure and high cholesterol are connected by a common denominator: poor lifestyle choices. Both diseases are more prevalent in sedentary and obese persons, and tobacco users" (Griffith & Wood, p. 240). High cholesterol levels are particularly dangerous for people with high blood pressure because excess cholesterol in the blood can become trapped in the crevices of arteries that have been damaged by chronic hypertension. As time passes, cholesterol accumulates along with other materials and constricts the opening of the affected arteries; if this blockage takes place in the coronary arteries that supply blood to the heart muscle, the result is a heart attack while blockages in the arteries that provide the brain with blood can result in a stroke (Griffith & Wood, 1997).

In a number of cases, blood cholesterol levels have been shown to be able to be lowered through diet and exercise, and the risk of CHD can thereby be reduced (Adams & Jennings, 1993). Cholesterol, though, represents just one causal factor that underlies CHD, and other risk factors for CHD include genetic predisposition, obesity, high blood pressure, diabetes, and smoking; however, just one factor, cholesterol, has been shown to be an important factor in this medley. In fact, an individual with a blood cholesterol level of 240 mg/dL (milligrams per deciliter) has a risk of CHD more than double that of an individual whose cholesterol is 200 mg/dL (NIH, 1990 cited in Adams & Jennings, p. 146).

The public awareness of the health risks typically associated with high cholesterol levels has also grown significantly; for example, a recent study found the percentage of Americans who were informed that their blood cholesterol level was too high increased from 11 to 20% between 1987 and 1990, and the number of people tested during the same period increased by 25% (Adams & Jennings, 1993). In this regard, there have been three dietary habits that have been commonly shown to contribute to elevated blood cholesterol levels:

1. Diets with high levels of saturated fatty acids which can elevate LDL cholesterol levels; these are found in both animal fats (e.g., butterfat) and plant oils; a number of tropical oils (e.g., palm, palm kernel, and coconut oils) contain particularly high concentrations of saturated fatty acids. "Fortunately, tropical oils are a minor component of the U.S. food supply" (Adams & Jennings, 1993, p. 146).

2. Diets with relatively high levels of cholesterol; dietary cholesterol is present only in foods and byproducts of animal origin (Adams & Jennings, 1993).

3. High calorie diets that exceed normal body requirements and contribute to obesity. Accordingly, health experts have provided relatively consistent dietary guidelines for reducing blood cholesterol levels. Generally, the specific recommendations include lowering the intake of saturated fats and dietary cholesterol, increasing the relative proportion of foods high in complex carbohydrates, and reducing total caloric intake for overweight persons (Adams & Jennings, 1993).

Although many Americans have responded to the threat represented by elevated blood cholesterol levels, significant problems continue to persist; for example, the average blood cholesterol level for adult Americans is approximately 210 mg/dL, and about 55% of the adult population have cholesterol levels of 255 mg/dL or higher (Adams & Jennings, 1993). In 1986, the National Center for Health Statistics estimated that 27.4 million adult Americans have cholesterol levels that placed them at high risk of CHD, and another 19.6 million adults could be placed in the "moderate risk" category; in addition, there has been an increasing amount of concern expressed over elevated cholesterol levels in children. According to Kagawa-Singer, Katz, Taylor, and Vanderryn (1996), Asian and Latino children have health problems comparable to those previously identified among white and African-American children. These authors report that, "Many were overweight and had low cardiovascular endurance and high cholesterol levels. In this Los Angeles sample, 38% of the Asian and Latino children had above-normal cholesterol levels for children, and 13% had cholesterol levels above normal for adults" (Kagawa-Singer et al., p. 149). By today's health standards, these rates would be even more alarming, with 40% of the boys and 45% of the girls being categorized as moderately to severely obese. Indeed, the authors predicted that these poor health trends would likely continue until better health education and physical education programs were provided for at-risk children and their parents (Kagawa-Singer et al., 1996). Likewise, a study conducted in New York found that approximately 80% of the 9-year-old children surveyed were consuming too much saturated fat and 60% were ingesting excessive amounts of dietary cholesterol (Adams & Jennings, 1993). Today, a number of researchers believe that high cholesterol levels in childhood can also contribute to an increased risk of heart disease and hypertension in adulthood (Adams & Jennings, 1993).

B. Rationale. Contemporary lifestyles frequently prevent consumers from enjoying balanced nutritional intake. Increasing pressures associated both with work and a decreasing amount of time being spent in the home has forced many Americans to "eat on the run"; further, despite increased public awareness, there is a clear trend today that Americans are becoming more obese; an article in American Demographics pointed out that "Americans claim to be concerned about nutrition, but demanding lifestyles and hunger pains are more likely to determine the foods they eat" (Are Americans Eating Better?, 1989, p. 30). Both high blood pressure and high lipids levels, though, are routinely treated with lifestyle modification regimens, including changes to diet and exercise. High blood pressure is a very common disease, but it is frequently symptomless; however, the condition can be successfully self-managed given the proper information, motivation, and medical guidance (Griffith & Wood, 1997). According to Griffith and Wood, if an individual has high cholesterol and high blood pressure that are related to obesity, a low-fat diet will simultaneously treat the hypertension and high cholesterol. Furthermore, a regular aerobic exercise program has been shown to lower blood pressure, reduce total cholesterol, and increase HDL (or so-called "good" cholesterol) levels (Griffith & Wood, 1997).

In their book, Behavior, Health and Aging, Baum et al. (2000) note that young and middle-aged men with relative hypercholesterolemia are known to have an increased risk of heart disease; in addition, the relationship between cholesterol and CHD risk is graded, and high cholesterol remains a risk factor in men with manifest CHD and in countries where the mean cholesterol concentration is relatively low. Furthermore, elevated serum cholesterol has been linked with preclinical atherosclerotic plaques in the carotid, femoral and coronary arteries (Baum et al., 2000).

Recent innovations have permitted the separation of total cholesterol into lipoprotein fractions; this has allowed researchers to better identify the relationship between serum lipids and atherosclerotic disease (Baum et al., 2000). In this regard, while it is closely correlated with total cholesterol, low density lipoprotein (LDL) cholesterol appears to represent a more accurate predictor of CHD; further, there is a growing body of evidence that points to an independent, inverse relationship between high density lipoprotein (HDL) cholesterol and CHD, whereas the role of elevated triglyceride levels in coronary atherosclerosis remains a point of debate (Baum et al., 2000). Recent studies have shown, though, that HDL and triglycerides are more important factors for women than men and, conversely, that elevated LDL cholesterol is an inconsistent predictor of heart disease in women (Baum et al., 2000).

C. Hypotheses and/or Objectives. The general hypothesis of this study is that eating foods that are high in cholesterol is not the only reason why people develop problems with cholesterol; the research will be guided by the following research questions.

D. Research questions. This study will be guided by the following three research questions:

1. Can high cholesterol levels be genetically related?

2. Can high cholesterol levels be anatomically induced?

3. Do high cholesterol levels always result from poor eating choices?

E. Assumptions and Limitations. For the purposes of this study, it will be assumed that a chi-square analysis represents a superior methodology for the investigation of the above-stated general hypothesis.

F. Definition of terms.

1. Coronary heart disease (CHD).

1. High-density lipoprotein (HDL). This is the so-called "good" cholesterol (Griffith & Wood, 1997).

2. Hypercholesterolemia.

This term refers to an elevation of cholesterol in blood plasma (Albertine, 2001).

3. Low-density lipoprotein (LDL). This is the so-called "bad" cholesterol; these are the particles that contribute to atherosclerosis (e.g., the hardening and narrowing of the arteries) (Ulrich, 2002).

Chapter 2: Review of the Literature

Overview. As noted above, there are two types of lipoproteins in the blood; their relative quantities in the blood are main factors in heart disease risk (Henkel, 1999).

1. Low-density lipoprotein (LDL). This is the so-called "bad" cholesterol; LDLs are the form in which cholesterol is communicated into the blood and represent the primary cause of harmful fatty buildup in arteries. Henkel notes that the higher the LDL cholesterol level in the blood, the higher the risk of contracting heart disease.

2. High-density lipoprotein (HDL). By contrast, HDLs are the so-called "good" form of cholesterol, HDLs carry blood cholesterol back to the liver, where it can be eliminated. According to Henkel, HDL serves to prevent cholesterol accumulations in blood vessels; therefore, low HDL levels tend to increase the associated risk of heart disease.

One of the basic ways LDL cholesterol levels can reach dangerous levels is through eating too much of two nutrients: 1) saturated fat (commonly found mostly in animal products, and 2) cholesterol (found only in animal products); saturated fat increases LDL levels more than anything else in the diet. A wide range of other factors, though, also affect blood cholesterol levels including:

1. Genetic. According to Henkel, high cholesterol levels can frequently be found in families. "Even though specific genetic causes have been identified in only a minority of cases, genes still play a role in influencing blood cholesterol levels," he says (p. 23).

2. Overweight and Obesity. The author reports that excess weight tends to increase blood cholesterol levels; therefore, reducing weight may help to lower levels (Henkel, 1999).

3. Sedentary or Active Lifestyles. Henkel points out that regular physical activity may not only lower LDL cholesterol, but it may increase the levels of desirable HDL.

4. Age and Gender. Prior to experiencing menopause, Henkel points out that women tend to have total cholesterol levels that are lower than men of the same age; however, as shown in Figure 1 below, cholesterol levels naturally increase as both men and women age. According to Henkel, menopause has been associated with increases in LDL cholesterol in women in a number of studies.

5. Stress. Finally, although clinical studies have not demonstrated any direct connection between cholesterol levels and stress, clinicians advise that because people sometimes eat fatty foods to console themselves when under stress, this can result in higher blood cholesterol (Henkel, 1999).

While high total and LDL cholesterol levels, together with low HDL cholesterol, can tend to increase heart disease risk, these are just some of a wider range of other risk factors that include cigarette smoking, high blood pressure, diabetes, obesity, and physical inactivity. "If any of these is present in addition to high blood cholesterol, the risk of heart disease is even greater" (Henkel, p. 23).

Etiology. Total serum cholesterol levels gradually rise from childhood through adulthood. Cross-sectional studies indicate that cholesterol tends to peak in both sexes at around age 60 years; in fact, after a certain age, cholesterol levels tend to decline (see Figure 1 below) (Baum et al., 2000). According to Baum and his colleagues, "Prospective data corroborate these age and gender patterns, and the data further reveal that body mass index changes in parallel with cholesterol levels. Therefore, the rise in cholesterol during adulthood and the fall after age 60 may be attributable, in part, to weight change" (p. 227). While there is some evidence that the incidence of hypocholesterolemia (cholesterol < mg/dl) tends to increase in prevalence between the ages of 65 and 85 years, Baum et al. point out that other evidence suggests that individuals who manage to reach age 80 years in otherwise good health tend to enjoy "normal" cholesterol levels, without an overrepresentation of either hypo- or hypercholesterolemia (Baum et al., 2000).

Figure 1. Mean serum cholesterol levels by sex and age.

Source: Baum et al., 2000 p. 227.

Review of Selected Studies to Date. A public health demonstration project by Greenblatt-Ives, Kuller, & Traven (1993) evaluated the efficacy of community-based cholesterol-lowering interventions in elderly individuals. In this study, approximately 1,200 ambulatory, noninstitutionalized men and women 65 to 79 years of age suffering from serum cholesterol levels over 240 mg/dl were randomly assigned to health screenings and promotion services at local hospitals and clinics, or to a control group. According to Baum and his colleagues, the intervention developed by Greenblatt et al. was provided through scheduled appointments that were free of charge; approximately half of the intervention group subjects completed their appointments. The results revealed of this study showed that after 2 to 3 years, neither of these groups (the intervention group as a whole nor those actually attending the intervention sessions) experienced reduced cholesterol levels that were lower than those of the control group. In fact, only active drug treatment significantly was shown to reduce serum cholesterol concentrations in this study. The authors concluded that aggressive treatment (i.e., with drugs) is therefore required to in order to achieve any substantive reduction in cholesterol among elderly, hypercholesterolemic individuals (Baum et al., 2000).

In their study, "Baseline Assessment of the Health Status and Health Behaviors of African-Americans Participating in the Activities-for-Life Program: A Community-Based Health Intervention Program," Paschal, Lewis, Martin, Dennis-Shipp, and Simpson (2004) report that, "Obesity is a major contributor to the high rates of hypertension and diabetes among Americans, particularly African-Americans. For instance, African-Americans have a higher prevalence of being overweight than their Caucasian counterparts" (p. 305). Furthermore, the authors emphasize that today, 30% of African-Americans are categorized as being obese compared to 28% of Caucasians.

Further exacerbating the already high rate of obesity and related conditions among African-Americans is the widespread problem of poor nutrition and inactivity. Although national figures suggest that 23% of the adult population has a sedentary lifestyle, a disproportionate number of African-Americans seem to have such, with 55-75% of African-American women rarely exercising, and 30 to 66% of men not exercising (Bronner, 2001). In addition, general nutrition and eating habits are poor for this population. For instance, approximately 76% of the African-American population do not currently meet minimum recommendations for daily fruit servings, and less than half (42%) currently meet the minimum daily requirements for vegetables per day (Bronner, 2001).

Purpose and Goal of Study. The purpose of the study by Dennis-Shipp and his colleagues was to determine the baseline results from a 9-month "Activities-for-Life" program; this regimen provided financial incentives, nutrition education, and physical fitness activities for African-American men and women. According to the authors, "This was a culturally relevant health education and physical fitness program that was community-based and easily accessible to its predominantly African-American community" (p. 306). The goal of the intervention program was to reduce those risk factors commonly associated with obesity and related health problems such as diabetes and hypertension in the sample population (Dennis-Shipp et al., 2004).

Methods and Participants. The Activities-for-Life initiative was implemented in 2002, by the Center for Health & Wellness, a primary health care and prevention and wellness center located in Wichita, Kansas; the participants in the program were provided formal education about life-threatening conditions (obesity, poor nutrition, hypertension, and diabetes) and were required to participate in physical fitness activities each week at the Center for Health & Wellness for a period of nine months (Dennis-Shipp et al., 2004).

The study sample was comprised of 134 African-Americans who resided in northeast Wichita, Kansas; 94 (or 70%) of the participants were females and 40 (30%) were males. The authors report that the age of the participants ranged from 20-74 years of age with the majority (52%) being between the ages of 35 and 54 years; in addition, 55% of the subjects were employed or self-employed, and 15% were unemployed (not including those that were retired or students) and 40% of the subjects were married. The authors also report that 83% of the subjects had some type of medical or health insurance. The amount of formal education achieved by the participants ranged from one year of schooling (or first grade) to the completion of graduate school work, with the majority (65%) having completed somewhere between 12 and 14 years of formal education; approximately 55% had household incomes of $30,000 or less (Dennis-Shipp et al., 2004).

The researchers recruited their subjects from a low-income, urban area of Wichita (financial incentives of $150 paid over the course of the 9-month intervention were shown to be the most effective of those used). The initiative was promoted through a combination of press release, urban radio announcements, radio talk shows, outreach referrals to predominately black churches, individual referrals, and the distribution of a program brochure in the targeted community (Dennis-Shipp et al., 2004). The initiative was promoted in these advertisements as being a health education program that would focus on nutrition and other health issues -- it was not promoted as a program for weight loss or obesity. Upon recruitment, the potential participants were told that the program would emphasize life-style changes that would improve their long-term health. Participation in the program was strictly voluntary. Participants were asked to complete consent forms and a behavioral contract stipulating their intent to attend education prevention classes and to attend at least two physical fitness classes provided by the center every week. After program orientation the participants were administered the health survey. The health screenings were carried out the following week at the Center for Health & Wellness.

Questionnaire and Other Instrumentation Used. The health survey used by the researchers consisted of 63 questions and was comprised of constructs from the following three sources:

1. Thirty-five (35) questions from a health survey used in a community-based prevention intervention program that provided formal education about health and nutrition;

2. Twenty (20) questions from a health survey used to assess African-American church attendees' attitudes and beliefs about health and testing; and, 3. Eight (8) questions from a standard SF-8 Health Survey form that is frequently used in clinical health settings (Dennis-Shipp et al., 2004).

The items on the current survey included basic demographic questions, items targeting the respondent's health, specific health behavior questions, and health attitude and belief inquiries. Besides the data collected from the health survey, the researchers also obtained data from the health screenings that each of the participants was required to complete; those health screenings involved blood pressure, blood sugar, and cholesterol checks, as well as height and weight assessments to determine body mass indices (Dennis-Shipp et al., 2004).

Statistical Analysis. The researchers used SPSS Version 11.0 to perform the statistical analysis of the sample data; simple frequencies and means were calculated and used to summarize the data collected from the health surveys including the demographic characteristics of the sample, and the health behavior and attitude questions. Frequencies and means were also used to summarize the data collected from the health screenings (Dennis-Shipp et al., 2004).

Results.

The statistical analysis of the health screening data described a sample that was basically overweight, with participants' weights ranging from 130 to 345 pounds; however, 212 pounds was the mean weight for males and 209 pounds was the mean weight for females. The authors suggest that based on the analysis of the body mass indices (BMI's) for each participant, 30% of the overall sample was overweight and approximately 60% were obese. Just one percent of the sample was found to be underweight, and only 9% were determined to have ideal BMI's with the remaining 90% being regarded as overweight or obese. There were some profound gender differences identified in the analysis as well; for example, the results showed that 96% of the females in the sample were overweight and obese (27% and 69% respectively). While the male subjects experienced a higher percentage (38%) of overweight individual compared to their female counterparts, they showed a lower percentage that were obese (41% among males versus 69% among females) (Dennis-Shipp et al., 2004).

The findings from the self-report data showed that a sedentary lifestyle, either voluntary or otherwise, was a major problem with this sample. According to the authors, "The majority of the participants (79%) stated that they exercised less than three times per week for at least 20 minutes. Nearly half of the sample (49%) reported that they ate fewer than two servings of fruits and vegetables per day" (emphasis added) (p. 307). The authors also report that 38% of the subject indicated that they perceived their overall physical health to be only fair to poor when they were requested to rate their overall physical health (on a scale with the following responses: excellent, very good, good, fair, poor) (Dennis-Shipp et al., 2004).

The results of the health screenings found a high incidence of hypertension among this sample, with almost two-thirds (62%) being diagnosed with this condition, and the health survey data indicated that more than half (56%) of the sample reported being diagnosed with hypertension by a physician before participating in the current program. According to the health screening results, 17% more males than females were more likely to have hypertension (74% compared to 57%, respectively). In addition, the health screenings showed that 17% of the participants experienced high blood sugar levels; however, 14% responded with this factor on the surveys. The health screenings results indicated that 17% of the female subjects had high blood sugar compared to 9% of the men (Dennis-Shipp et al., 2004).

Not surprisingly, the authors also report that the sample as a whole had high cholesterol levels; approximately 37% of the participants were identified with high cholesterol levels based upon recommended ranges. These percentages were comparable to the 36% of participants that indicated high levels by a physician prior to their health screenings for this program. The females in the study were more likely to have high cholesterol, with 45% of the 94 women in the study having this condition. Eighteen percent of the men had high cholesterol (Dennis-Shipp et al., 2004).

One of the most important contributions this study presents to the literature on obesity is the health information it provides about African-American males participating in a community-based health intervention program. Participation from adult males has been extremely low in similar programs. Moreover, few prior studies utilized stipends or financial incentives, so this program may have been more successful in recruiting male participants because it was able to provide this. It is widely known that it is difficult to get racial/ethnic minorities to be involved in research. Incentives might be a way to get a cross-section of the African-American population to participate. Financial incentives or stipends could be instrumental in recruiting low-income minority participants who are oftentimes short on extra time for program commitment and limited in income. Therefore, interventions such as the Activities-for-Life Program may help recruit more participants by including stipends to help reimburse the participants for their time as well as representing an investment in the overall outcome of project by providing more eligible subjects for analysis (Dennis-Shipp et al., 2004).

Furthermore, this study provided timely and valuable information concerning the status of health of African-Americans in low-income areas; the findings of this study emphasized the glaring need for programs to address the alarming overweight and obesity rates typically reported among this segment of the population. Therefore, this study might have implications for reducing health disparities across the country. The data collected from the community-based program indicates a growing need for community-based interventions to modify diet and activity levels in order to prevent obesity, hypertension and other health diseases in at-risk African-American communities; these types of interventions may serve to reduce the rate of excess weight gain and improve overall physical fitness (Dennis-Shipp et al., 2004). The authors, though, caution that there were a number of limitations with this study, including: 1) It is difficult to generalize these results to all African-Americans in Wichita, Kansas. "Even in Wichita, African-Americans are not a monolithic group and different subgroups and cultures exist within the African-American community. However, given the demographics regarding income, age and marital status, there appears to be a cross-section of African-Americans from which we might be able to draw conclusions" (p. 307); 2) The results could have been skewed by the composition of the sample itself, which was comprised of a larger percentage of women than men; whenever possible, though, the authors point out that their data was analyzed by gender. Finally, the report that while this study had more women than men, African-American men are rarely studied (Dennis-Shipp et al., 2004).

The authors recommend that any future studies of this nature should focus on collecting community- and region-specific data because of cultural diversity that is evinced among this population segment; furthermore, they recommend that there is a crying need for such obesity prevention initiatives through culturally appropriate community-based behavioral interventions. Furthermore, Dennis-Shipp and his colleagues determined that any such future intervention programs would benefit from the recruitment of larger samples and attempts to recruit more male subjects. In this regard, the authors report that financial incentives may improve the recruitment of males, a feature that has been identified in other studies as well. Finally, the authors state that future studies might also assess associated barriers to not exercising and eating right (Dennis-Shipp et al., 2004).

Taken together, the results of this study suggest that incentives to compensate for their limited time and commitment, low-income African-Americans, especially African-American men, represent a highly effective tool to increase participating of this population segment in health promotion interventions. Based on the national efforts to provide for more equitable distribution of healthcare services, this type of program reflects the importance of timely disseminating of crucial health status and behavior information for minorities; it also stresses the need for the implementation of such programs in inner-city and rural areas of the country. In the final analysis, the authors conclude that, "In order to screen and determine the baseline levels of behaviors of minority populations, health disparities will likely not be reduced until we determine what are the health behaviors and status of minority populations. More research conducted by community-based agencies may be our greatest asset" (Dennis-Shipp et al., 2004, p. 309).

Chapter 3: Methodology

A. Background. Beyond the paucity of timely research into the relationship between the suspected causative factors and high cholesterol levels, there remains a profound need to identify an improved method of assessing these potential relationships. In this regard, social researchers frequently use a chi-square analysis to help identify any associations between two or more variables by amplifying and describing their respective relationships (Neuman, 2003). According to Mirkin (2001), the chi-squared coefficient was first proposed by K. Pearson (1900) as an application of his concepts for testing observed frequencies against expected values with the chi-square distribution that he invented for this specific purpose. "In this context," Mirkin notes, "the coefficient was to be used only for testing independence in a bivariate distribution, not for evaluating association. However, the coefficient came to be used as an association measure, notably by A.A. Tschuprow and H. Cramer" (p. 111).

B. Assumptions and Limitations Associated with Chi-Square Analysis. According to Neuman (2003), chi-squared actually has two different uses: "It can be used as a measure of association in descriptive statistics, or in inferential statistics. As a measure of association, chi-squared can be used for nominal and ordinal data. It has an upper limit of infinity and a lower limit of zero, meaning no association" (p. 350).