Characteristics of adult primary care patients as predictors of future health services charges.

BACKGROUND: Utilization risk assessment is potentially useful for allocation of health care resources, but precise measurement is difficult. OBJECTIVE: Test the hypotheses that health-related quality of life (HRQOL), severity of illness, and diagnoses at a single primary care visit are comparable case-mix predictors of future 1-year charges in all clinical settings within a large health system, and that these predictors are more accurate in combination than alone. RESEARCH DESIGN: Longitudinal observational study in which subjects' characteristics were measured at baseline, and their outpatient clinic visits and charges and their inpatient hospital days and charges were tracked for 1 year. SUBJECTS: Adult primary care patients. MEASURES: Duke Health Profile for HRQOL, Duke Severity of Illness Checklist for severity of illness, and Johns Hopkins Ambulatory Care Groups for diagnostic groups classification. RESULTS: Of 1,202 patients, 84.4% had follow up in the primary care clinic, 63.2% in subspecialty clinics, 14.8% in the emergency room, and 9.6% in the hospital. Of $6,290,775 total charges, $779,037 (12.2%) was for follow-up primary care. The highest accuracy was found for predicting primary care charges, where R2 for predictors ranged from 0.083 for medical record auditor-reported severity of illness to 0.107 for HRQOL. When predictors were combined, the highest R2 of 0.125 was found for the combination of HRQOL and diagnostic groups. CONCLUSIONS: Baseline HRQOL, severity of illness, and diagnoses were comparable predictors of 1-year health services charges in all clinical sites but most predictive for primary care charges, and were more accurate in combination than alone.

A telephone needs assessment for potential high utilizers.

BACKGROUND AND OBJECTIVES: We conducted a telephone needs assessment of patients at risk for high utilization of health services to determine the type of intervention that might be most responsive to their needs. METHODS: Family practice patients who were classified as at high risk based on self-reported health-related quality of life (HRQOL), age, and gender received a structured telephone interview to ask about their health problems, difficulties with their health care, and types of help programs that might be useful to them. RESULTS: Of 867 adult patients randomly selected as eligible for the study, 352 completed the Duke Health Profile to measure HRQOL, 122 were classified at risk for high utilization, and 104 were interviewed by telephone. Patient-reported health problems were similar to provider-reported problems. High-risk patients reported difficulties receiving help from the clinic on issues of communication and logistics (43.9%), limitations of clinic facilities and services (26.3%), and financial concerns (26.3%). Of the highest risk patients, 45.8% expressed willingness to participate in help programs conducted over the telephone. CONCLUSIONS: Many high-risk patients might benefit from participation in a telephone intervention program. A future study is needed to measure the effectiveness of such a program to improve utilization of health services.

Health-related quality of life predictors of survival and hospital utilization.

The health-related quality of life (HRQOL) of 103 end-stage renal disease (ESRD) patients on hemodialysis was studied for prediction of 1-year survival and hospital days in the context of other predictors. Higher HRQOL physical functioning, higher provider-reported functional performance, fewer private religious activities, living with family, black race, and having a diagnosis of hypertension predicted survival. Lower HRQOL energy, higher pain, and not living with family predicted more hospital days. Patients living with family reported more social support and better HRQOL general health, emotional well-being, social health, and quality of social interactions than other patients.

An international comparison of the reliability and responsiveness of the Duke Health Profile for measuring health-related quality of life of patients treated with alprostadil for erectile dysfunction.

OBJECTIVES: It is important that health measures are both reliable and responsive to clinical change. The aim of this study was to assess the reliability and responsiveness of the physical, mental, and social health scales of the Duke Health Profile (DUKE). METHODS: Impotent males self-administered the Duke Health Profile before and during treatment with alprostadil for erectile dysfunction during a 19-month period. Subjects were 490 patients in the United States and 583 patients in 12 other countries. Each of the three basic Duke Health Profile scales has only five items, and each is heterogeneous because each measures more than one health concept. RESULTS: Cronbach's alpha reliability estimates were: physical health, 0.68 for United States and 0.64 for other countries; mental health, 0.62 and 0.52, respectively; and social health, 0.53 and 0.47, respectively. Alprostadil was expected to improve mental health primarily, and results of the study were consistent with this hypothesis. For example, at approximately 14 months from therapy onset, mental health improved for patients both in the United States (standardized response mean, SRM, = 0.17) and other countries (mean SRM = 0.30), whereas physical health worsened in the United States and was unchanged in other countries, and social health was unchanged in the United States and improved in other countries. Maximum responsiveness was shown for mental health in the other countries, where the mean standardized response means at four follow-ups during a 19-month period were 0.11, 0.21, 0.30, and 0.36. CONCLUSIONS: This study provides support for the responsiveness of the Duke Health Profile mental health scale.

[The Duke Severity of Illness Checklist]. / Die Duke-Checkliste für den Krankheitsschweregrad (DUSOI).

Developed at the Duke University Medical Center of Durham, North Carolina, U.S.A., the Duke Severity of Illness Checklist (DUSOI) is a tool for measuring a person's illness severity. The instrument comprises four parameters of a health problem, e.g. in a given diagnosis, namely: symptoms, complications, 6-months prognosis without treatment, treatment potential. Using the numerical ratings (from 0 to 4) for each parameter of every diagnosis, the following three types of severity score (from 0 [lowest degree of severity] to 100 [highest degree of severity]) can be calculated: (1) the DUSOI diagnosis score for each diagnosis stated, (2) the DUSOI overall score for the set of all health problems stated for a patient, and (3) the DUSOI comorbidity score, i.e., all problems except for any one problem of principal interest. The DUSOI is suitable for patients from the entire chain of medical and rehabilitative care, although it had initially been developed for the ambulatory sector. Completing the DUSOI form is very economical in terms of time needed, and is recommended to be done by the physician in charge immediately following the patient's visit. Alternatively, the form could also be filled in by a reviewer on the basis of the patient's medical record. A certain level of judgement is required on the part of the physician or reviewer carrying out the DUSOI assessments.

Quality of life effects of alprostadil therapy for erectile dysfunction: results of a trial in Europe and South Africa.

OBJECTIVES: Quality of life (QOL) data were used to evaluate the effects of self-administered intracavernosal injection of alprostadil for erectile dysfunction, when used for up to 18 months during a 13 country Phase III clinical trial. METHODS: The Duke Health Profile was used to measure patients' physical and psychosocial QOL at baseline, 3, 6, 12 and 18 months. Changes from baseline were measured using paired t-tests, with additional analyses by cause of dysfunction, starting dosage, and prior treatment. RESULTS: Patients displayed significant improvements in mental and social health and self-esteem at six months (P < 0.01, n = 570), with greater improvements at 12 and 18 months. Anxiety and depression measures also improved significantly at 12 and 18 months, as did the summary general health score. Worse pain scores were observed in the first year but not at 18 months. Those with a starting dosage of 10-20 micrograms, those with psychogenic causes of dysfunction, and those with no prior treatment for erectile dysfunction generally showed the greatest improvements. CONCLUSION: In this study, the clinical improvements in erectile function due to intracavernosal alprostadil therapy were complemented by QOL improvements, particularly in the mental health, of many patients.

Safety of autologous, ex vivo-expanded human immunodeficiency virus (HIV)-specific cytotoxic T-lymphocyte infusion in HIV-infected patients.

We infused six human immunodeficiency virus (HIV)-seropositive subjects with autologous CD8+ cytotoxic T cells (CTLs) enriched for HIV-specific cytotoxicity targeted against a diversity of HIV epitopes in gp120, gag p17 and p24, and nef. There was no toxicity and no subject deteriorated clinically. In the first 2 weeks, CD4 counts increased for all subjects and plasma viremia decreased in five of six subjects. Twenty-four weeks later, the mean values of all measures of viral burden and surrogate markers of HIV infection were either unchanged or improved, but none of the changes was statistically significant. Two subjects continued to have decreased cell-associated viral burden and another subject had more than doubled CD4 cell count. HIV-specific CTL activity increased in most subjects. The increase in CD4 T-cell counts in the first weeks after the infusion suggests that antiviral CTLs of diverse specificities do not play a significant role in CD4 T-cell decline. The lack of any acute toxicity or adverse effect on viral burden suggests that therapy with antiviral CTLs deserves further study.

Predictors of functional health status of end stage renal disease patients.

Potential predictors of the functional health status of 125 end stage renal disease (ESRD) patients were studied cross-sectionally. When health status was assessed by the physician with the Karnofsky Index, younger patient age, lower ESRD severity of illness, lower comorbidity severity, and higher albumin levels were predictors of better health [R-square = 0.48]. When patients self-reported their health status with the Duke Health Profile, African-American race, higher family support, lower family stress, and lower ESRD severity were positive predictors [R-square = 0.23]. The importance of measuring functional status, severity of illness, and social support and stress of ESRD patients is supported by these findings.

Screening for anxiety and depression in primary care with the Duke Anxiety-Depression Scale.

BACKGROUND AND OBJECTIVES: Anxiety and depression are highly prevalent and underdiagnosed in primary care. This study tested the seven-item Duke Anxiety-Depression Scale (DUKE-AD) in primary care adult patients as a screener for anxiety and depression as defined by the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R). METHODS: Receiver operating characteristic curves (ROC) and odds ratios were used to test screener accuracy, and sensitivities and specificities were used to test screener efficiency in patients with anxiety and/ or depression. RESULTS: In 481 patients, the ROC area for patients with major anxiety disorders (panic disorder, agoraphobia, or generalized anxiety) was 72.3%. The ROC area for major depressive disorders (major depressive disorder and/or dysthymia) was 78.3%, and the ROC area for both major anxiety and/or depressive disorders was 76.2%. Odds ratios for these same groups after controlling for sociodemographic factors were 1.043, 1.057, and 1.053, respectively. Sensitivities and specificities for these groups at the DUKE-AD score cutoff point of > 30 on a 0-100 scale were 71.4% and 59.2%, 81.8% and 63.6%, and 73.9% and 66.1%, respectively. CONCLUSIONS: The DUKE-AD is a brief, easily scored questionnaire that serves as a valid screener for DSM-III-R anxiety and depression in the primary care setting.

Recognition of a small number of diverse epitopes dominates the cytotoxic T lymphocytes response to HIV type 1 in an infected individual.

Mitogen-activated T cell lines may be reproducibly used to identify relatively conserved HIV-1 epitopes that dominate CTL recognition of HIV-infected cells. Using a combination of nested truncations of HIV-vaccinia recombinants encoding HIV-1LAI Env and overlapping peptides that span the coding regions of the HIV-1 SF2 subclone of env, gag, nef, rev, and tat, we have mapped the immunodominant, relatively conserved CTL epitopes recognized by 25 HIV-seropositive individuals with CD4 counts between 100 and 500/mm3 and no history of AIDS opportunistic infection. We could characterize at least 1 peptide CTL epitope recognized by the T cell lines of 18 of 25 of the subjects; the T cell lines from 2 additional subjects recognized HIV-vaccinia presenting targets, but no dominant peptide epitope was identified. CTL epitopes were most frequently encoded by gag (recognized by 16 of 25 patient T cell lines), followed by nef and env (11 of 25 each), and the RT region of pol (9 of 25). Tat and Rev were rarely the sites of CTL epitopes. The identified epitopes occurred predominantly in relatively conserved regions of HIV-1. The mean number of HIV peptides identified at a single time for each cell line was 2.7 +/- 1.7. Although no single peptide dominated CTL recognition in more than four individuals, clusters of epitopes were found in the N-terminal region of gp160 and in two central regions of Nef. The dominant HIV-1 CTL epitopes in infected patients were not predictable on the basis of MHC expression and varied widely in an MHC-diverse population.

Perceived mental health and disablement of primary care and end-stage renal disease patients.

OBJECTIVE: To compare perceived current mental health and disablement between primary care and end-stage renal disease (ESRD) patients, and to study social support and stress and severity of illness as possible determinants of mental health and disablement. METHOD: Observational cross-sectional analysis of 414 primary care patients in a rural community health center and 125 ESRD patients requiring hemodialysis in two community dialysis units. The Duke Health Profile (DUKE) anxiety-depression scale was used to assess mental health; the DUKE disability scale, to indicate disablement; the Duke Social Support and Stress Scale, to measure support and stress; and the Duke Severity of Illness Scale, to rate severity of illness. RESULTS: Perceived current mental health in terms of anxiety and depression symptoms was worse for primary care than for ESRD patients, and perceived current disablement was no different for the two groups. Patients' perception of their health status and of stress from family members were more closely associated with their level of anxiety and depression symptoms than were their diagnostic profiles or overall severity of illness. In turn, their level of anxiety and depression symptoms was the principal correlate of their disablement. CONCLUSIONS: The demonstration of strong relationships among anxiety and depression symptoms, disablement, and family stress in these two very different patient populations should stimulate further research and motivate clinicians to evaluate all three parameters as part of routine patient care.

Duke Case-Mix System (DUMIX) for ambulatory health care.

The Duke Case-Mix System (DUMIX), which combines age, gender, patient-reported perceived and physical health status, and provider-reported or auditor-reported severity of illness to classify patients by their risk of high future utilization, explained 17.1% of the variance in future clinic charges and 16.6% of the variance in return visits. When a random half of 413 ambulatory adults were classified into four risk classes by predictive regression coefficients from the other half, there was a stepwise increase in actual future utilization by risk class. The most accurate classification was for Class 4 (highest risk) patients, with a sensitivity of 40.8%, specificity of 82.1%, and likelihood ratio of 2.3. These 23.7% of patients accounted for 44.2% of charges for all patients. When predictive coefficients from this population were used to classify a different group of 206 ambulatory adults, past utilization also increased in stepwise order by case-mix class.

Classification of severity of health problems in family/general practice: an international field trial.

BACKGROUND: A methodology is needed for classification of health problems by severity. OBJECTIVES: We aimed to test the Duke Severity of Illness Checklist (DUSOI) for feasibility and usefulness. METHOD: The DUSOI was field tested internationally by 22 family/general practitioners in 9 countries. RESULTS: The DUSOI was found to be feasible for rating severity of illness of health problems in family/general practice. The measure was shown to be clinically useful in older patients and those with chronic and more severe health problems. Variability of severity ratings was less within the same rater than between different raters (i.e. higher intrarater than interrater reliability). Clinical face validity was supported by the finding that DUSOI ratings classified patients with the same diagnosis and those with different diagnoses according to the severity differences that would be expected clinically. CONCLUSIONS: Although research is needed to improve reliability and to test validity further, the DUSOI was shown in the present study to be a methodology that is reasonable for consideration as an international classification of health problems by their severity in primary care patients.

Anxiety and depressive symptom identification using the Duke Health Profile.

Duke Health Profile (DUKE) subscales were compared for their ability to identify anxiety and depressive symptoms as measured by the State Anxiety Inventory (SAI) and the Center for Epidemiologic Studies Depression Scale (CES-D) in 413 primary care patients. The seven-item Duke Anxiety-Depression Scale (DUKE-AD) was the best symptom identifier, with sensitivities and specificities greater than 70% for high scores on both the SAI and CES-D. Also, baseline DUKE-AD scores predicted five clinical outcomes during an 18-month follow-up period, with receiver operating characteristic (ROC) curve areas ranging from 57.1 to 58.7%. Patients shown by DUKE-AD scores to be at high risk (>30, scale 0-100) for symptoms of anxiety and/or depression were more often women, less well-educated, not working, and with lower socioeconomic status. The severity of illness was higher than that of low-risk patients. Although the providers did not know which patients were at high risk, they made a clinical diagnosis of anxiety or depression more often in high-risk patients.

Perceived family stress as a predictor of health-related outcomes.

OBJECTIVE: To measure the predictive effect of patient-perceived family stress for health-related outcomes. DESIGN: Prospective study of patients whose social stress and support were measured by self-report at baseline with the Duke Social Support and Stress Scale and used as predictors of outcomes during an 18-month follow-up period. SETTING: Rural primary care community health clinic. PATIENTS: Convenience sample of ambulatory adults. INTERVENTION: None. MAIN OUTCOME MEASURES: Follow-up (one or more follow-up visits), frequent follow-up (more than six visits), referral and/or hospitalization (one or more), high follow-up severity of illness (upper-tertile mean Duke Severity of Illness Checklist scores), and high follow-up total charges (> or = $268). RESULTS: There were 413 patients with a mean age of 40.4 years. Of these, 58.6% were women; 47.2%, African American; 52.8%, white; 56.7%, married; 77.2%, wage earners or housekeepers; and 52.3% had more than one health problem. At baseline, patients with high self-reported family stress (upper-tertile Duke Social Support and Stress Scale scores) had lower quality of life, functional health, and social support scores and higher dysfunctional health and social stress scores than other patients. High baseline family stress scores (scale of 0 to 100) predicted follow-up (odds ratio [OR] = 1.014), frequent follow-up (OR = 1.021), referral and/or hospitalization (OR = 1.018), high severity of illness at follow-up (OR = 1.016), and high follow-up charges (OR = 1.018) after controlling for the effects of social support, age, gender, and race. Family stress scores were stronger predictors of these outcomes than the other social stress and support variables. CONCLUSION: The finding of patient-perceived family stress as a risk factor for unfavorable health-related outcomes suggests the need for early detection and treatment of family stress by family physicians.

Health status and severity of illness as predictors of outcomes in primary care.

Two measures of health status and severity of illness were tested as indicators of patient case-mix to predict health-related outcomes in a rural primary care community health clinic, using a convenience sample of 413 ambulatory adults (mean age = 40.4 years: 58.6% women, and 47.2% black). At baseline; patients completed the Duke Health Profile, and providers completed the Duke Severity of Illness Checklist. During the 18-month follow-up study, patients experienced the following outcomes: at least one follow-up visit (74.3%), more than six visits (20.6%), at least one referral or hospital admission (17.3%), upper tertile severity scores (24.9%), and upper tertile office charges (24.9%). Baseline physical health, perceived health, and severity scores were statistically significantly predictive of all five outcomes. Predictive accuracy (i.e., area under the receiver operating characteristic curves) for outcome probabilities estimated from a case-mix model of physical health, severity, age, gender, and race was 72.3% for follow-up, 69.7% for frequent follow-up, 70.5% for referral and/or hospital stay, 65.7% for high follow-up severity of illness, and 67.6% for high follow-up charges. These data support health status and severity of illness as case-mix indicators and outcome predictors of follow-up utilization, severity of illness, and cost in the primary care setting.

Feasibility and potential clinical usefulness of a computerized severity of illness measure.

OBJECTIVE: To assess the feasibility and potential clinical usefulness of the computerized Duke Severity of Illness Checklist (DUSOI). DESIGN: Cross-sectional study of patients whose severity of illness was measured with the DUSOI. Providers assessed the clinical usefulness of the DUSOI and recorded the length of time required for rating severity. Auditors rated severity using progress note information. Demographic and financial data from clinic records were also obtained. SETTING: University-based family practice clinic with 64,621 annual visits. PATIENTS: Convenience sample of ambulatory patients. MAIN OUTCOME MEASURES: Clinical usefulness and time required to rate severity. RESULTS: For 117 patients (63.3% female; mean age, 46.3 years), the mean charge was $105.38, the mean number of health problems was 2.0, the mean overall provider DUSOI score was 33.7, and the mean auditor DUSOI score was 34.0 (scale = 0 to 100). There was excellent agreement between provider and auditor DUSOI scores (intraclass correlation coefficient, .77). Providers required 1.1 minutes to record severity; the principal auditor required 1.6 minutes. Providers found the DUSOI potentially useful in 30.3% of patients. Usefulness was greater in women (38.2% of women vs 18.2% of men), older patients (mean age, 54.5 years in useful group vs 41.9 in nonuseful group), and sicker patients (mean DUSOI score, 55.1 vs 25.9). The DUSOI was more clinically useful in patients with health problems such as type II diabetes mellitus (75.0%) than in those with problems such as tobacco use (25.0%). Higher charges correlated with a higher number of health problems and with female gender but not with severity scores. CONCLUSIONS: The computerized DUSOI is feasible for all patients and is potentially useful for women, older, and sicker patients.

Disease-specific versus generic measurement of health-related quality of life in insulin-dependent diabetic patients.

The health-related quality of life of 170 adult insulin-dependent diabetic patients was measured cross-sectionally to compare a disease-specific instrument, Diabetes Quality of Life (DQOL) questionnaire, and two generic instruments, the Duke Health Profile (DUKE) and the General Health Perceptions Questionnaire (GHP). The generic measures provided as much or more information about health-related quality of life as the disease-specific instrument. This was demonstrated both by comparison of the DQOL with the DUKE and GHP and by comparison of the disease-specific with the generic components of a modified version of the DQOL. Patients with the diabetic complication of nephropathy had increased worry over their health and lower general health perceptions. Neither the duration of diabetes nor the intensity of insulin therapy, however, was found to have a statistically significant effect on any of the health-related quality of life scores. Nondiabetic factors, such as the comorbidity, nondiabetic medications, marital status, social relationships, and family arguments were found to be predictors of health-related quality of life more often than the diabetic factors duration of diabetes, complications, and intensity of insulin therapy. These analyses suggest the clinical value of using generic questionnaires to measure health-related quality of life and psychosocial factors to identify nondiabetic problems that might respond to intervention, thereby potentially enhancing the effect of diabetes-specific therapy.