No. 1, January 2009
Health-Related Quality of
Life in Adults From 17 Family Practice Clinics in North Carolina
Leigh F. Callahan, PhD, Jack Shreffler, PhD, Thelma J. Mielenz, PT, PhD, Jay S. Kaufman, PhD, Britta Schoster, MPH, Randy Randolph, MRP, Philip Sloane, MD, MPH, Robert DeVellis, PhD, Morris Weinberger, PhD
Suggested citation for this article: Callahan LF, Shreffler J, Mielenz TJ, Kaufman JS, Schoster B, Randolph R, et al. Health-related quality of life in adults from 17 family practice clinics in North Carolina. Prev Chronic Dis 2009;6(1):A05.
jan/07_0215.htm. Accessed [date].
We examined health-related quality of life (HRQOL) in white and African American
patients based on their own and their community’s socioeconomic status.
Participants were 4,565 adults recruited from 17 family physician practices in urban and rural areas of North Carolina. Education
was used as a proxy for individual socioeconomic status, and the census block-group poverty level was used as a proxy for community socioeconomic status. HRQOL measures were the 12-Item Short Form Survey Instrument,
physical component summary (PCS) and mental component summary (MCS), and 3 Centers for Disease Control and Prevention HRQOL healthy days measures. Multilevel analyses
examined independent associations of individual and community poverty level with HRQOL, adjusting for demographics and clustering by family practice. Analyses were stratified by race and
were conducted on subgroups of arthritis and cardiovascular disease patients.
Among whites, all 5 HRQOL measures were significantly associated with the lowest individual socioeconomic status, and 4 HRQOL measures were associated with the lowest community socioeconomic status (MCS being the exception).
Among African Americans, 4 HRQOL measures were significantly associated with the lowest individual socioeconomic status and the lowest community socioeconomic status (PCS being the exception).
Arthritis and cardiovascular disease subgroup analyses showed generally analogous findings.
Better HRQOL measures generally were associated with low levels of community poverty and high levels of education, emphasizing the need for further exploration of
factors that influence health.
Back to top
Despite the growing ability of societies to extend life and prevent disease,
health disparities persist (1). As researchers and policy makers have tried to answer the question of why some societies are healthier than others, attention has focused on individual-level factors such as biology (eg, genomics), psychology (eg, coping strategies), community (eg, place of residence, work environment), and society (eg, social and economic policies).
Although a strong association is well-established between lower levels of individual socioeconomic status (SES)
and poor health outcomes from many diseases throughout the developed world (1), studies increasingly suggest that community social determinants
(ie, the socioeconomic environment of a person’s neighborhood) may influence health (2-7), independent of the
person’s SES (4,8-11). Regardless of personal socioeconomic position (12,13), characteristics of the
physical, social, and service environments of neighborhoods and communities may influence the lives of
people who live there (14), including their health (3).
Many studies that have examined the association of both individual and community social determinants with
illness and death have been conducted primarily in large urban areas (5,15). The purpose of this study was to examine the relationship between individual and community SES with health-related quality of life (HRQOL) by race in a geographically diverse sample of community-dwelling non-Hispanic whites and African Americans in a
southeastern state. Study participants were recruited from a representative cohort of adult primary care patients followed in 17 practices from rural and urban areas of North Carolina. Analyses were conducted on the entire sample and on subgroups of
people who self-reported the 2 most common chronic conditions and leading causes of disability, arthritis and cardiovascular disease (CVD) (16).
Back to top
In 2001, the North Carolina Family Medicine Research Network established a practice-based cohort for primary care research (17). The practices were selected to represent the geographic and racial/ethnic diversity of North Carolina. In each of the 17 participating practices, all consecutive patients at least 18 years of age who had a scheduled appointment during a 1-month period were asked by study staff to enroll in this study (N = 7,687). Of the consecutive patients approached, 4,876
(63.4%) enrolled in our study. Participants were asked to complete a self-report questionnaire that included items on demographics, chronic conditions, health behaviors, and health status. This study focused on the 4,565 participants whose self-designated race was either white (n = 3,612) or African American (n = 953). Participants were equally distributed between urban and rural practices.
In addition to examining the whole group,
we also performed analyses on subgroups of
patients who reported the 2 most common chronic conditions, arthritis (white, n = 969; African American, n = 275), and CVD/hypertension (white, n = 1,271; African American, n = 469). Participants were classified as self-reporting arthritis if they reported they had osteoarthritis, rheumatoid arthritis, or fibromyalgia. They were classified as self-reporting CVD/hypertension if they reported they had heart disease, CVD, or hypertension. In the subgroup analyses, individuals
appeared in both subgroup analyses if they self-reported both arthritis and CVD/hypertension. All components of the study were approved by the medical institutional review board of the University of North Carolina at Chapel Hill.
Two self-administered instruments were used to measure HRQOL: the Medical Outcomes Study’s 12-Item Short Form Survey Instrument (SF-12v2) and the Centers for Disease Control and Prevention Health-Related Quality of Life (CDC HRQOL) healthy days measure. The SF-12v2 yields 2 summary scores: physical component summary (PCS) and mental component summary (MCS). The SF-12v2 is strongly correlated with the SF-36 and is reliable in general populations (18). Higher scores on the PCS and MCS
indicate better health, and scores range from 0 to 100. The healthy days measure assesses physical and mental HRQOL (19). In this study, we used responses to the following questions as single-item indicators of HRQOL: 1) “Now thinking about your physical health, which includes physical illness and injury, for how many days during the past 30 days was your physical health not good?" 2)
“Now thinking about your mental
health, which includes stress, depression, and problems with emotions, for how many days during the past 30 days was your mental health not good?” and 3) “During the past 30 days, for about how many days did poor physical or mental health keep you from doing your usual activities, such as self-care, work, or recreation?” CDC HRQOL questions have been validated by statistically correlating the responses with measures from more comprehensive or established instruments such as the SF-36v2
(20). The CDC HRQOL measures have good construct validity, acceptable criteria, and known groups validity, and they have been suggested for use in surveillance and research (20).
Individual SES was defined by education level, which was categorized as low (less than high school
diploma), middle (high school diploma), or high (more than high school diploma). The number of years of formal education
is the socioeconomic variable most closely associated with health (21) and is commonly used in epidemiologic studies (22). Community SES was defined by the block-group poverty level (percentage of the population in households with income below the
federal poverty threshold), derived by matching each participant’s home address to his or her census block group, a geographic entity containing an average of
approximately 1,000 residents, obtained from the 2000 US Census (23,24) with MapMarker Plus 7.2 (Empower Geographics, Des Plaines, Illinois). Only results with precise geography were used.
Some studies have suggested that block-group characteristics are better indicators of the immediate SES environment than
are census tract measures (4).
A poverty-level category was assigned as either low, medium, or high for both racial groups, with cutpoints designed to divide each racial group into tertiles. The cutpoints for whites were poverty levels of 6.9% and 13.8%. The corresponding cutpoints for African Americans were 11.8% and 21.3%. Therefore, in absolute terms, the highest poverty category for the African Americans was more severe than the highest category for whites, and the categories established hardship in relative
terms within the communities, which were largely segregated.
We decided to stratify by race because it is a strong correlate of education and poverty level (25). We also evaluated for effect measure modification for all social variables in all 5 of the HRQOL outcome models. For every model, a different effect was found for whites and African Americans for 1 or more of the covariates, confirming our decision to stratify.
All data were analyzed by using Stata 9.0 (StataCorp LP, College Station, Texas). Descriptive statistics were computed to describe the sample, and t tests and χ2 tests were performed to evaluate statistical differences between white and African American groups and between
people with arthritis or CVD/hypertension and those without arthritis or CVD/hypertension.
Because data were collected at 17 family practice sites across the state, some correlation within sites was possible. Multilevel analyses were performed by using multiple linear regressions, specifically analysis of covariance, which examined the independent associations of education and poverty with the 5 HRQOL outcomes. These models were adjusted for age, sex, and body mass index (BMI) and
for clustering by family practice site by adjusting the estimated standard errors for intra-site
correlation. The adjusted mean outcomes are computed from the estimated model equation as follows: the nonreferent indicators for poverty levels (middle, high) are pov2 and pov3, respectively. The nonreferent indicators for education (middle, low) are ed2 and ed3, respectively.
HRQOL = b0 + b1 × pov2 + b2 × pov3 + b3 × ed2
+ b4 × ed3 + b5 × age + b6 × BMI + b7
In many cases, study participants in a given block group attend different family practices, obviating the possibility of building a 3-level random-effects model (block groups are not nested within practice sites). Tests of formal 2-level models (people with practice sites) did not show distinct advantage over the population average-effects model.
In the models, poverty and education appeared as indicator variables, and the most beneficial categories (high education and low poverty level) served as the references. For example, the test for significance of an outcome in the low-education group evaluated the difference from the outcome in the high-education group. Models were run for the total group and then on subgroups of
people who self-reported arthritis and CVD.
Back to top
Characteristics of study participants are displayed in
Table 1. In the total sample, African Americans were significantly more likely to be younger, be female, and have a higher BMI, less education, and worse HRQOL (except for CDC HRQOL poor mental days) compared with whites. The block-group poverty-level tertiles were calculated separately for whites and African Americans and reflected higher poverty levels in African Americans. Patients within each of the disease-specific subgroups
were significantly more likely to be older, have a higher BMI, and be less educated than the patients not in the disease subgroup. The disease-specific subgroups had significantly lower HRQOL than the whole group (with the exception of MCS for the CVD/hypertension group). The mental HRQOL showed an overall less negative effect of disease than
did physical HRQOL.
Table 2 presents the adjusted means for the 5 HRQOL outcomes by education and poverty level for the white total group and the arthritis and CVD/hypertension subgroups. The highest level of education and the lowest poverty level are the referent categories. The adjusted means varied more
within categories of education than within categories of poverty. For example, in the white total group, SF-12v2 PCS increased with education from 38.1 to 46.6
whereas PCS increased with decreasing poverty level
from 43.8 to 46.0.
For the PCS in the total white group, groups with both the middle and low
levels of education had significantly lower mean scores (poorer outcomes) than
the group with the highest level of education, after adjusting for poverty
level, age, BMI, sex, and family practice site. Also, for the PCS, both the
middle and high poverty-level groups had significantly lower mean scores than
the lowest poverty-level group, adjusting for education level and the other
covariates (Table 2). In the
whole population, mean SF-12v2 MCS scores were significantly poorer in
the middle- and low-education groups, but we found no significant
differences among the poverty-level groups.
All 3 of the CDC HRQOL healthy days measures showed significantly higher mean scores (poorer outcomes) in both the lowest education-level groups and the highest poverty-level groups, adjusting for each other and age, BMI, sex, and family practice setting (Table 2) in whites. Similar findings were identified in the subgroup analyses for participants who self-reported arthritis and
those who self-reported CVD/hypertension (Table 2).
Table 3 shows the adjusted means for the 5 health status measures by education and poverty level for the African American total group and the
2 disease subgroups (arthritis and CVD/hypertension). The smaller sample size, particularly for African Americans in the arthritis subgroup, resulted in fewer findings of significant differences, but the general trend of the findings follows that seen for whites. Low individual levels of education and high poverty levels were significantly associated
with poorer status in the SF-12v2, MCS, more poor physical and mental health days, and more limited activity days in the total group.
Back to top
In this community-dwelling sample of patients living in rural and urban areas of North Carolina, both individual (education) and community (poverty level) SES measures were associated with physical and mental measures of HRQOL. A 1-day difference is considered meaningful for the CDC HRQOL healthy days scores at the individual level (26). All 3
days scores had a difference of at least 1 day for both the individual and the community SES measures. The minimum clinically important
difference for the SF-12v2 PCS and SF-12v2 MCS in a chronic disease has been reported as 1.26 and 2.28, respectively (27). In our study, the differences for the PCS and MCS exceeded these thresholds for both education and community poverty level. Thus, all differences reported here appear to be meaningful.
Overall, the magnitude of difference in HRQOL was greater for education compared with community poverty level.
For example, for whites in any group, changes in the SF-12 PCS scores, going
from low to high SES, ranged from 8.1 to 9.6 for education and from 1.7 to 2.2
for community poverty level. Although the independent effects of community socioeconomic context may be relatively small, the overall importance of community socioeconomic context to individual health may be more substantial, both because it affects all
people in a
community and because community context shapes the person’s sense of control and all
The association between lower levels of individual SES and poorer health outcomes has been documented repeatedly in various parts of the developed world (28-32). Individual-level SES can be examined according to a number of variables, including formal education level, which was used in this study, income, occupation, home ownership, race, and marital status. Formal education level is, in part, a marker for behavioral variables (22,33,34), such as self-management,
problem-solving abilities, efficiency in use of medical services, capacity to cope with stress, social skills, psychological status, and economic skills, which singly or together enable
people to more effectively prevent, overcome, or cope with adversity (35).
Although the body of public health research relating community factors to patterns of health and disease is well-established (36), the underlying hypothesis of this prior work proposes that factors operating at the level of the communities may affect the health outcomes of individuals. Studies in recent years have suggested that area-level or community-level variables may provide information that is not captured by individual-level variables (2,37,38). In our study, using census-based
block-group poverty levels as a community-level SES indicator, we found that the community indicator was significantly associated with HRQOL after controlling for the individual indicator.
Our study has several limitations. Our data are all self-reported, and we had only
education level as our measure of individual SES. Although we lacked data on income and occupation as individual SES measures, many studies have shown education to be a strong marker of individual SES. The effects of community socioeconomic level on health may be underestimated in our study because we used a crude measure of community SES, block-group poverty level from the census. Future research should begin
to include information about self-defined communities or at least purposefully delineate community boundaries to more closely match theoretical constructs.
The US health care research agenda places priority on reducing disparities in health outcomes among people from different socioeconomic and ethnic groups by examining the mechanisms believed to
affect health. Now, more than ever,
the interest in a more explicit investigation of the complex issues about health disparities
is increasing (39). However, the statement in Healthy People 2010 recognizing that communities, states, and national
organizations will need to take a multidisciplinary approach to achieving health equity is often overlooked (40). The development and communication of effective actions to reduce health disparities
depends on clarifying relationships between community variables (social context), individual variables (social position), and health outcomes.
Back to top
This project was supported by a grant from the National Institute of Arthritis and Musculoskeletal and Skin Diseases Multidisciplinary Clinical Research Center, Rheumatic Diseases: P60-AR49465-01.
We thank Jennifer Milan Polinski and Shannon Currey for assistance in the establishment of the cohort. We thank the following participating family practices in the North Carolina Family Medicine Research Network for their assistance: Black River Health Services, Burgaw; Bladen Medical Associates, Elizabethtown; Blair Family Medicine, Wallace; Cabarrus Family Medicine, Concord; Cabarrus Family Medicine, Harrisburg; Cabarrus Family Medicine, Kannapolis; Cabarrus Family Medicine, Mt. Pleasant;
Chatham Primary Care, Siler City; CMC-Biddle Point, Charlotte; CMC-North Park, Charlotte; Community Family Practice, Asheville; Cornerstone Medical Center, Burlington; Crissman Family Practice, Graham; Dayspring Family Medicine, Eden; Family Practice of Summerfield, Summerfield; Goldsboro Family Physicians, Goldsboro; Henderson Family Medicine Clinic, Henderson; Lumber River Family Practice, Lumberton; Moncure Community Health Center-Piedmont Family Health Services, Moncure; Orange Family
Medical Group, Hillsborough; Person Family Medical Center, Roxboro; Pittsboro Family Medicine, Pittsboro; Prospect Hill Community Health Center, Prospect Hill; Robbins Family Practice, Robbins; and Village Family Medicine, Chapel Hill.
Back to top
Corresponding Author: Leigh F. Callahan, PhD, University of North Carolina at Chapel Hill, Thurston Arthritis Research Center, 3300 Thurston Bldg, CB#7280, Chapel Hill, NC 27599-7280. Telephone: 919-966-0564. E-mail: email@example.com.
Author Affiliations: Jack Shreffler, Thelma J. Mielenz, Jay S. Kaufman,
Britta Schoster, Randy Randolph, Philip Sloane, Robert DeVellis, University of North Carolina at Chapel Hill; Morris Weinberger, University of North Carolina at Chapel Hill,
Durham Veterans Affairs Medical Center, Durham, North Carolina.
Back to top
- Marmot M, Wilkinson RG. Social determinants of health. Oxford (UK): Oxford University Press; 1999.
- Diez-Roux AV, Nieto FJ, Muntaner C, Tyroler HA, Comstock GW, Shahar E, et al.
Neighborhood environments and coronary heart disease: a multilevel analysis. Am J Epidemiol 1997;146:48-63.
- Robert SA. Socioeconomic position and health: the independent contribution of community socioeconomic context. Annu Rev Sociol 1999;25:489-516.
- Diez-Roux AV, Merkin SS, Arnett D, Chambless L, Massing M, Nieto FJ, et al.
Neighborhood of residence and incidence of coronary heart disease. N Engl J Med 2001;345(2):99-106.
- Yen IH, Kaplan GA.
Neighborhood social environment and risk of death: multilevel evidence from the Alameda county study. Am J Epidemiol 1999;149:898-907.
- Bosma H, van de Mheen H, Borsboom G, Mackenbach JP.
Neighborhood socioeconomic status and all-cause mortality. Am J Epidemiol 2001;153(4):363-71.
- Lochner K, Pamuk E, Makuc D, Kennedy BP, Kawachi I.
State-level income inequality and individual mortality risk: a prospective, multilevel study. Am J Public Health 2001;91(3):385-91.
- Kobetz E, Daniel M, Earp JA.
Neighborhood poverty and self-reported health among low-income, rural women, 50 years and older. Health Place 2003;9(3):263-71.
- Nordstrom CK, Diez-Roux AV, Jackson SA, Gardin JM.
The association of personal and neighborhood socioeconomic indicators with subclinical cardiovascular disease in an elderly cohort. The cardiovascular health study. Soc Sci Med 2004;59(10):2139-47.
- Sundquist K, Winkleby M, Ahlen H, Johansson SE.
Neighborhood socioeconomic environment and incidence of coronary heart disease: a follow-up study of 25,319 women and men in Sweden. Am J Epidemiol 2004;159(7):655-62.
- Winkleby M, Cubbin C, Ahn D.
Effect of cross-level interaction between individual and neighborhood socioeconomic status on adult mortality rates. Am J Public Health 2006;96(12):2145-53.
- Morland K, Wing S, Diez-Roux AV, Poole C.
Neighborhood characteristics associated with the location of food stores and food service places. Am J Prev Med 2002;22(1):23-9.
- Inagami S, Cohen DA, Finch BK, Asch SM. You are where you shop: grocery store locations, weight, and neighborhoods. Am J Prev Med 2006;31(1):10-7.
- Taylor CB, Ahn D, Winkleby MA.
Neighborhood and individual socioeconomic determinants of hospitalization. Am J Prev Med 2006;31(2):127-34.
- Yen IH, Yelin EH, Katz P, Eisner MD, Blanc PD.
Perceived neighborhood problems and quality of life, physical functioning, and depressive symptoms among adults with asthma. Am J Public Health 2006;96(5):873-9.
- Brownson RC, Remington P, Davis JR. Chronic disease epidemiology and control. Washington (DC): American Public Health Association; 1998.
- Sloane PD, Callahan L, Kahwati L, Mitchell CM.
Development of a practice-based patient cohort for primary care research. Fam Med 2006;38(1):50-7.
- Ware J Jr, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996;34:220-33.
- Measuring healthy days: population assessment of health-related quality of life. Atlanta
(GA): Centers for Disease Control and Prevention; 2000.
- Mielenz T, Jackson E, Currey S, DeVellis R, Callahan LF.
Psychometric properties of the Centers for Disease Control and Prevention Health-Related Quality of Life (CDC HRQOL) items in adults with arthritis. Health Qual Life Outcomes 2006;4:66.
- Hinkle LE Jr, Whitney LH, Lehman EW, Dunn J, Benjamin B, King R, et al.
Occupation, education, and coronary heart disease. Risk is influenced more by education and background than by occupational experiences, in the Bell System. Science 1968;161(838):238-46.
- Liberatos P, Link BG, Kelsey JL.
The measurement of social class in epidemiology. Epidemiol Rev 1988;10:87-121.
- Krieger N.
Overcoming the absence of socioeconomic data in medical records: validation and application of a census-based methodology. Am J Public Health 1992;82(5):703-10.
- Krieger N.
Women and social class: a methodological study comparing individual, household, and census measures as predictors of black/white differences in reproductive history. J Epidemiol Community Health 1991;45:35-42.
- Kaufman JS, Cooper RS.
considerations for use of racial/ethnic classification in etiologic research. Am J Epidemiol 2001;154(4):291-8.
- Moriarty DG, Zack MM, Kobau R.
The Centers for Disease Control and Prevention’s Healthy Days Measures
population tracking of perceived physical and mental health over time. Health Qual Life Outcomes 2003;1(1):37.
- Bennett SJ, Oldridge NB, Eckert GJ, Embree JL, Browning S, Hou N, et al.
Comparison of quality of life measures in heart failure. Nurs Res 2003;52(4):207-16.
- Weinblatt E, Ruberman W, Goldberg JD, Frank CW, Shapiro S, Chaudhary BS.
Relation of education to sudden death after myocardial infarction. N Engl J Med 1978;299(2):60-5.
- Callahan LF, Pincus T.
Formal education level as a significant marker of clinical status in rheumatoid arthritis. Arthritis
- Lynch JW, Kaplan GA, Shema SJ.
Cumulative impact of sustained economic hardship on physical, cognitive, psychological, and social functioning. N Engl J Med 1997;337:1889-95.
- Marmot M, Shipley M, Brunner E, Hemingway H.
contribution of early life and adult socioeconomic factors to adult
morbidity in the Whitehall II study. J Epidemiol Community Health 2001;55:301-7.
- Turrell G, Lynch JW, Kaplan GA, Everson SA, Helkala E, Kauhanen J, et al.
position across the lifecourse and cognitive function in late middle age. J Gerontol 2002;57B(1):S43-51.
- Callahan LF, Cordray DS, Wells G, Pincus T.
Formal education and five-year mortality in rheumatoid arthritis: mediation by helplessness scale score. Arthritis Care Res 1996;9(6):463-72.
- Winkleby MA, Jatulis DE, Frank E, Fortmann SP.
Socioeconomic status and health: how education, income, and occupation contribute to risk factors for cardiovascular disease. Am J Public Health 1992;82(6):816-20.
- Sagan LA. The health of nations: true causes of sickness and well-being. New York (NY): Basic Books; 1987.
- Tyroler HA, Cassel J.
Health consequences of culture change II: the effect of urbanization on coronary heart mortality in rural residents. J Chronic Dis 1964;17:167-77.
- Susser M.
The logic in ecological: I.
The logic of analysis. Am J Public Health 1994;84:825-9.
- Krieger N, Rowley DL, Herman AA, Avery B, Phillips MT.
Racism, sexism, and social class: implications for the studies of health, disease, and well-being. Am J Prev Med 1993;9(Suppl 6):82-122.
- Evans T, Whitehead M, Diderichsen F, Bhuiya A, Wirth M. Challenging inequities in health: from ethics to action. Oxford (UK): Oxford University Press; 2001.
- Metzler M. Social determinants of health: what, how, why, and now. Prev Chronic Dis 2007;4(4):85.
Back to top