Joanne M. Garrett

The Outcomes and Costs of Care for Acute Low Back Pain among Patients Seen by Primary Care Practitioners, Chiropractors, and Orthopedic Surgeons

BACKGROUND Patients with back pain receive quite different care from different types of health care practitioners. We performed a prospective observational study to determine whether the outcomes of and charges for care differ among primary care practitioners, chiropractors, and orthopedic surgeons. METHODS Two hundred eight practitioners in North Carolina were randomly selected from six strata: urban primary care physicians (n = 39), rural primary care physicians (n = 48), urban chiropractors (n = 32), rural chiropractors (n = 32), orthopedic surgeons (n = 29), and primary care providers at a group-model health maintenance organization (HMO) (n = 28). The practitioners enrolled consecutive patients with acute low back pain. The patients were contacted by telephone periodically for up to 24 weeks to assess functional status, work status, use of health care services, and satisfaction with the care received. RESULTS The status at six months was ascertained for 1555 of the 1633 patients enrolled in the study (95 percent). The times to functional recovery, return to work, and complete recovery from low back pain were similar among patients seen by all six groups of practitioners, but there were marked differences in the use of health care services. The mean total estimated outpatient charges were highest for the patients seen by orthopedic surgeons and chiropractors and were lowest for the patients seen by HMO and primary care providers. Satisfaction was greatest among the patients who went to the chiropractors. CONCLUSIONS Among patients with acute low back pain, the outcomes are similar whether they receive care from primary care practitioners, chiropractors, or orthopedic surgeons. Primary care practitioners provide the least expensive care for acute low back pain.

What Is Wrong With End‐of‐Life Care? Opinions of Bereaved Family Members

OBJECTIVE: To describe family perceptions of care at the end of life.

A Prospective Study of Advance Directives for Life-Sustaining Care

BACKGROUND The use of advance directives is recommended so that people can determine the medical care they will receive when they are no longer competent, but the effectiveness of such directives is not clear. METHODS In a prospective study conducted over a two-year period, 126 competent residents of a nursing home and 49 family members of incompetent patients were interviewed to determine their preferences with respect to hospitalization, intensive care, cardiopulmonary resuscitation, artificial ventilation, surgery, and tube feeding in the event of critical illness, terminal illness, or permanent unconsciousness. Advance directives, consisting of signed statements of treatment preferences, were placed in the medical record to assist in care in the nursing home and to be forwarded to the hospital if necessary. RESULTS In an analysis of 96 outcome events (hospitalization or death in the nursing home), care was consistent with previously expressed wishes 75 percent of the time; however, the presence of the written advance directive in the medical record did not facilitate consistency. Among the 24 events in which inconsistencies occurred, care was provided more aggressively than had been requested in 6 cases, largely because of unanticipated surgery or artificial ventilation, and less aggressively than requested in 18, largely because hospitalization or cardiopulmonary resuscitation was withheld. Inconsistencies were more likely in the nursing home than in the hospital. CONCLUSIONS. The effectiveness of written advance directives is limited by inattention to them and by decisions to place priority on considerations other than the patients autonomy. Since our study was performed in only one nursing home and one hospital, other studies are necessary to determine the generalizability of our findings.

The Natural History of the Development of Obesity in a Cohort of Young U.S. Adults between 1981 and 1998

Context Studies of the natural history of obesity have had limitations, such as cross-sectional design, short follow-up, or narrowly defined populations. Contribution Using data from a national study of Americans born between 1957 and 1964 and followed for nearly two decades, these investigators found that more than 25% were obese by age 35 years. Obesity developed most quickly in black women, with moderate rapidity in Hispanic women, and most slowly in white women. Hispanic men developed obesity more quickly than other men. More recent birth cohorts became obese faster than earlier cohorts. Implications Understanding ethnic differences in the age of onset and rate of progression of obesity may help efforts to prevent obesity. The Editors The proportion of Americans who are overweight or obese is increasing at an alarming rate. From 1960 to 1999, the prevalence of excess weight (body mass index [BMI] 25 kg/m2) increased from 44% to 61% of the adult population (persons aged 20 to 74 years) and the prevalence of obesity (BMI 30 kg/m2) doubled from 13% to 27% (1, 2). Obesity has been linked to increased mortality and chronic morbidity from such diverse causes as hypertension, diabetes, sleep apnea, certain types of cancer, and depression. In the context of this rising prevalence, obesity is increasingly being recognized as an important medical and public health problem (3-6). Treatment options are available for obesity. However, they can be expensive, often involve difficult lifestyle modifications, and pose adherence problems for many patients. In addition, weight loss, once achieved, is difficult to sustain. Because the health consequences are often long term, the costbenefit ratio of treatment may be more favorable if people who have recently become obese are targeted. Similarly, prevention should focus on those at immediate risk for becoming obese or overweight. Such approaches require a clear understanding of who is at highest risk for obesity, when this risk is greatest, and how excess weight develops over time. Although inactivity and high caloric intake are the primary risk factors for obesity, other risk factors have been established. Three of the most influential are age, sex, and race or ethnicity (1, 7-9). To better identify people at risk for obesity and those who might benefit from early intervention, we assessed the relationship between weight development and these influential factors. Despite the morbidity and mortality burden of obesity, its natural history has not been well characterized. Current knowledge is based largely on cross-sectional data. Although several large studies have examined weight development in adults, they have important limitations. Only a few tracked weight long enough to detect age-based trends. The first National Health and Nutrition Examination Survey (NHANES I) Epidemiologic Follow-up Study examined a representative group of U.S. residents but sampled only two time points, spaced 10 years apart, and had insufficient follow-up to track weight patterns in most minority groups (10). The Coronary Artery Risk Development in Young Adults study followed weight development in a large cohort of young black and white Americans but did not provide data for other populations and was geographically restricted (8). Other studies have been performed outside of the United States, but the results may not apply here (10, 11). We designed our study to address the limitations described earlier. We examined data from the National Longitudinal Survey of Youth 1979 (NLSY79). This large, prospective cohort study allowed longitudinal examination of weight trends over a time frame sufficient to capture the dynamics of such a slowly changing problem and of an age range during which obesity becomes prevalent. Because not all weight gain has been documented to be harmful, our analysis focused on obesity onset (defined as the time at which a person reaches obesity) rather than weight gain. We used the clinical definition of obesity (BMI 30 kg/m2) as our end point because it reflects a point at which mortality risk increases substantially (12). Understanding how obesity develops over time, who is at greatest risk, and when that risk is highest may be essential in reversing the population trend toward excess weight. Methods Study Group and Obesity Measurement We analyzed data from the NLSY79, which was initiated by the U.S. Department of Labor, Bureau of Labor Statistics. This survey was designed to gather information about the labor market activities and other life events of young U.S. citizens. It has been used extensively in economic and sociologic analyses, and the sample design is described elsewhere (13-16). The cohort of men and women, born between 1957 and 1964, has been followed for more than two decades. It consists of a nationally representative group, with supplemental samples overrepresenting minority racial and ethnic groups and poor persons. We excluded a sample of military personnel because this population was removed from the study in 1985 and because physical activity requirements for these individuals could lead to atypical weight patterns. We included only persons who reported that the ethnicity with which they most closely identified was Hispanic, black, or white (or European origin). Height and weight data were self-reported; height was reported in 1981 and 1985, and weight was reported 12 times between 1981 and 1998. Examination of Age-Specific Body Mass Index Patterns Age-specific BMI was calculated for all participants. Height was assumed to remain constant after age 21 years. Standard clinical definitions were used for overweight (BMI 25 to 29.9 kg/m2) and obesity (BMI 30 kg/m2). Trends in age-specific BMI were examined by race or ethnicity and birth year. Cohort Differences We examined whether rates of obesity onset differed for people maturing at different times in our sample. Birth cohorts were grouped by 2-year increments. To standardize relative rates to age rather than calendar year, time was expressed as the number of years until a person reached obesity, starting from a standard age. In the cohort analysis, obesity onset was evaluated from age 24 years (the baseline age of the oldest participants). The outcome of interest was obesity, and the time variable was years from age 24 years to obesity. People who were already obese at age 24 years were excluded. KaplanMeier plots visually demonstrated differences in rates of obesity onset. Cox proportional-hazards modeling estimated the ratio of rates (hazard ratio) between groups. For this method to be correct, the ratio of rates must be constant over time (proportional). To assess this assumption, we used two methods: a statistical test of the scaled Schoenfeld residuals, with a P value greater than 0.05 implying that the rates were proportional and a visual method with parallel lines implying proportional rates. In all our analyses, results of proportional hazards testing were reported whenever the assumption was violated. Timing of Obesity Onset by Race or Ethnicity and Sex We evaluated relative rates of obesity onset in 17- and 18-year-old persons of different race or ethnicity who were not obese at the start of the study; different models were used for men and women. Time was expressed as the number of years from age 17 years until reaching obesity. To control for birth cohort differences, models were adjusted for birth year. As before, KaplanMeier plots were examined and hazard ratios were estimated by using Cox proportional-hazards modeling. To determine relative importance of weight gain during childhood compared with adulthood in terms of the timing of obesity onset, we adjusted the models for baseline BMI at 17 or 18 years of age and examined the effect on hazard ratio estimation. Prevalence and Incidence of Obesity in Early Adulthood For this analysis, we defined early adulthood as the period from 20 to 22 years of age to 35 to 37 years of age. Persons born between 1959 and 1963 were queried about their weight one to two times during each of these age intervals. We calculated the prevalence and incidence of obesity across the early adulthood range. Logistic regression modeling predicted obesity status at age 35 to 37 years from an individuals race or ethnicity, sex, and BMI at age 20 to 22 years. The model was adjusted in 2-year increments for year of birth. Its predictive ability was quantified by using receiver-operating characteristic (ROC) curves. This study was evaluated and exempted from review by the Institutional Review Board of the University of North Carolina at Chapel Hill. Role of the Funding Sources Primary funding was through the Robert Wood Johnson Clinical Scholars Program, with additional support from the National Institute of Child Health and Human Development. Neither entity had a role in the design, conduct, or reporting of this study. Results Description of the Study Sample The total sample included 9179 people: 19% Hispanic, 31% black, and 50% white (Table). People who did not report ethnicity (n = 106) were not included. Age (sample year minus birth year) ranged from 17 to 24 years in 1981 to 34 to 41 years in 1998, and participants were equally distributed across 8 years of birth. Baseline weight was 31.7 to 136.1 kg, and baseline BMI was 11.3 to 51.5 kg/m2. Table. Sample Characteristics in 1981 Relationship of Body Mass Index to Time Mean age-specific BMI increased with time. This relationship between weight and time has two components, illustrated with data from the oldest and youngest birth cohorts of our sample (Figure 1). The first is the natural history of body weight: Average BMI increased with age throughout early adulthood. The second is a secular weight trend: Persons born in later calendar years tended to have larger age-specific BMI values. In every case in which the same age was sampled for both birth cohorts, the group born in 1957 had a lower mean BMI than the group born i

Differences in end-of-life decision making among black and white ambulatory cancer patients

OBJECTIVE: African-American (black) and white individuals have been shown to differ in their desire for life-sustaining treatments and their use of living wills for end-of-life care, but the reasons for these differences are unclear. This study-was designed to test the hypothesis that these ethnic differences exist because black patients trust the health care system less, fear inadequate medical treatment more, and feel less confident that living wills can give them more control over their terminal care.DESIGN: Cross-sectional, in-person survey conducted from November 1993 to June 1994.SETTING: Two medical oncology clinics with 40% to 50% black patient representation.PARTICIPANTS: Ambulatory cancer patients, 92 black and 114 white, who were awaiting their physician visits and agreed to participate (76% of those eligible). Patients were excluded if they were under age 40 or if they had nonmelanoma skin cancer only.MEASUREMENTS AND MAIN RESULTS: Black ambulatory cancer patients wanted more life-sustaining treatments (odds ratio [OR] 2.8; 95% confidence interval [CI] 1.4–5.3), and were less likely to want to complete a living will at some time in the future (OR 0.36; 95% CI 0.17–0.75) than were white patients after controlling for socioeconomic variables. However, these differences were not related to lack of trust or fear of inadequate medical treatment in this study population. Both groups of patients trusted the health care system and felt that physicians treated patients equally well. Neither group feared inadequate or excessive medical care, and the majority of both groups agreed that living wills would help them keep control over their terminal care.CONCLUSIONS: Black and white cancer patients make different end-of-life choices, even after adjusting for likely explanatory variables. The other factors that influence decision making remain unclear and need to be further explored if physicians are to understand and help their patients make choices for end-of-life care.

Life-sustaining treatments during terminal illness - Who wants what?

Objective: To determine patient characteristics associated with the desire for life-sustaining treatments in the event of terminal illness.Design: In-person survey from October 1986 to June 1988.Setting: 13 internal medicine and family practices in North Carolina.Patients: 2,536 patients (46% of those eligible) aged 65 years and older who were continuing care patients of participating practices, enrolled in Medicare. The patients were slightly older than the 65+ general population, 61% female, and 69% white, and most had one or more chronic illnesses.Measurements and main results: The authors asked the patients whether they would want each of six different treatments (hospitalization, intensive care, cardiopulmonary resuscitation, surgery, artificial ventilation, or tube feeding) if they were to have a terminal illness. The authors combined responses into three categories ranging from the desire for more treatment to the desire for less treatment. After adjustment for other factors, 53% of women chose less treatment compared with 43% of men; 35%ofblacksvs 15% of whites and 23% of the less well educated vs 15% of the better educated expressed the desire for more treatment. High depression scores also were associated with the desire for more treatment (26% for depressed vs 18% for others).Conclusion: Patients’ choices for care in the event of terminal illness relate to an intricate set of demographic, educational, and cultural factors. These results should not be used as a shortcut to determine patient preferences for care, but may provide new insights into the basis for patients’ preferences. In discussing choices for future life-sustaining care, physicians need to explore with each individual the basis for his or her choices.

Beyond the good prognosis: Examination of an inception cohort of patients with chronic low back pain

Study Design. A 22-month prospective cohort study. Objectives. To describe the course of an inception cohort of patients with chronic low back pain. Summary of Background Data. Chronic low back pain is a debilitating condition with great medical and social cost. Methods. A cohort of 1246 patients with acute low back pain who sought treatment from 208 North Carolina providers was observed. Patients who developed chronic low back pain were identified. Entry criteria were back pain of less than 10 weeks’ duration, no previous care for this episode of low back pain, no previous spine surgery, not pregnant, no nonskin malignancy, and access to a telephone. The providers were of four types: primary care medical doctors, doctors of Chiropractic, orthopedic surgeons, and health maintenance organization-based primary care providers. Patients were contacted by telephone shortly after enrollment and at 2, 4, 8, 12, and 24 weeks, with a final interview at 22 months. Patient functional status, care-seeking, and satisfaction were evaluated. Results. Ninety-six patients had chronic, continuous symptoms for 3 months, forming the inception cohort of chronic low back pain. A valid, reliable measure of back-specific functional disability also was used. Predictors of the development of chronicity were poor baseline functional status and sciatica. A more powerful predictor of chronicity was poor functional status at 4 weeks. Two thirds of patients with chronic low back pain at 3 months had functionally disabling symptoms at 22 months, and a majority of these were employed. Satisfaction with care was low. Forty-six patients (2.6% of the entire cohort) underwent surgery, with no statistically significant difference in surgical rates among initial provider strata. Patients who underwent surgery after 3 months had a Roland disability score at 22 months of 10 (7.7, 12.3). Forty-one percent of patients with chronic low back pain see an orthopedic or neurologic surgeon. Chronic low back pain occurs in 7.7% of patients who seek care for acute low back pain, with unremitting pain for 22 months in 4.7%. Conclusion. Once established, chronic low back pain is persistent. Most patients with chronic low back pain seek little care, and a majority are employed. Future research should emphasize maintenance of employment and function.

Patient Preference for Physician Discussion and Practice of Spirituality

OBJECTIVE: To determine patient preferences for addressing religion and spirituality in the medical encounter.DESIGN: Multicenter survey verbally administered by trained research assistants. Survey items included questions on demographics, health status, health care utilization, functional status, spiritual well-being, and patient preference for religious/spiritual involvement in their own medical encounters and in hypothetical medical situations.SETTING: Primary care clinics of 6 academic medical centers in 3 states (NC, Fla, Vt).PATIENTS/PARTICIPANTS: Patients 18 years of age and older who were systematically selected from the waiting rooms of their primary care physicians.MEASUREMENTS AND MAIN RESULTS: Four hundred fifty-six patients participated in the study. One third of patients wanted to be asked about their religious beliefs during a routine office visit. Two thirds felt that physicians should be aware of their religious or spiritual beliefs. Patient agreement with physician spiritual interaction increased strongly with the severity of the illness setting, with 19% patient agreement with physician prayer in a routine office visit, 29% agreement in a hospitalized setting, and 50% agreement in a near-death scenario (P<.001). Patient interest in religious or spiritual interaction decreased when the intensity of the interaction moved from a simple discussion of spiritual issues (33% agree) to physician silent prayer (28% agree) to physician prayer with a patient (19% agree; P<.001). Ten percent of patients were willing to give up time spent on medical issues in an office visit setting to discuss religious/spiritual issues with their physician. After controlling for age, gender, marital status, education, spirituality score, and health care utilization, African-American subjects were more likely to accept this time trade-off (odds ratio, 4.9; confidence interval, 2.1 to 11.7).CONCLUSION: Physicians should be aware that a substantial minority of patients desire spiritual interaction in routine office visits. When asked about specific prayer behaviors across a range of clinical scenarios, patient desire for spiritual interaction increased with increasing severity of illness setting and decreased when referring to more-intense spiritual interactions. For most patients, the routine office visit may not be the optimal setting for a physician-patient spiritual dialog.

Recurrence and care seeking after acute back pain: Results of a long-term follow-up study

OBJECTIVE To explore the relationship between type of initial care as well as the likelihood of recurrence and consequent care seeking behavior. RESEARCH DESIGN Prospective observational cohort recruited from 208 randomly selected North Carolina practices. Cohort study examined the recurrence of low back pain among patients free of back pain 3 months after their index visit to a practitioner for that problem. The following four practitioner strata were examined: primary care providers, chiropractors, orthopedic surgeons, and practitioners in a group model HMO. Patients were interviewed by telephone at 6 and 22 months after the initial visit. MAIN OUTCOME MEASURES Rates of disabling and non-disabling low back pain; functional status using the Roland back disability scale; and care seeking. RESULTS Rates of recurrence were substantial; functionally disabling recurrence rates varied between 8% and 14% between 3 to 6 months, and 20% to 35% between 6 to 22 months. Differences in rates among practitioner strata were statistically significant only between 6 to 22 months with higher recurrence rates for HMO patients. Functional status, number of bed days, and time off work were very similar among the practitioner strata. Care seeking, however, was greater among those patients who had initially seen a chiropractor for their back pain. Patients with recurrence saw the same practitioner type they had seen for the index episode 88% of the time. Satisfaction was slightly greater for patients who saw chiropractors when compared with patients seeing allopathic physicians. CONCLUSIONS The recurrence of low back pain is common. Severe disability is rare. Patients who had sought care from chiropractors are more likely to return for recurrences than patients who had initially sought care from MDs.

A prospective study of the impact of patient preferences on life-sustaining treatment and hospital cost.

OBJECTIVES Ethicists advise that life-sustaining treatment decisions should be made in keeping with patient preferences. Until recently, there has been little systematic study of the impact of patient preferences on the use of various life-sustaining treatments or the consequent cost of hospital care. This prospective study was designed to answer the following questions: a) Do patient treatment preferences about the use of life-sustaining treatment influence the treatments they receive? and b) Do patient treatment preferences influence the total cost of their hospitalization? DESIGN A prospective, cohort study. SETTING A university teaching hospital. PATIENTS Hospitalized patients, at least 50 yrs of age, with short life expectancy due to end-stage heart, lung, or liver disease, metastatic cancer, or lymphoma. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Patients were interviewed to determine their desire for life-sustaining treatment and other characteristics and then were followed for 6 months to determine life-sustaining treatment use and costs during hospitalization. Two hundred forty-four patients were interviewed. Fifty-eight percent of patients expressed a desire for life-sustaining treatments to prolong life for 1 wk. During 245 subsequent hospitalizations, there were 20 episodes of mechanical ventilation, 63 episodes of intensive care, and 66 cancer treatments given. Bivariate and multivariate analyses showed no significant association between patient desire to receive treatment to prolong life and either life-sustaining treatment use (p = .59) or hospital costs (p = .20). CONCLUSION In a university teaching hospital setting, there is no systematic evidence that patient preferences determine life-sustaining treatment use or hospital costs.