Original Article
Agreement of self-reported comorbid conditions with medical and physician reports varied by disease among end-stage renal disease patients

https://doi.org/10.1016/j.jclinepi.2006.09.003Get rights and content

Abstract

Objective

To compare self-report of eight diseases with review of medical records and physician reports.

Study Design and Setting

In a cohort of 965 incident end-stage renal disease (ESRD) patients (Choices for Healthy Outcomes in Caring for End-stage renal disease study), data on existing medical conditions were obtained from medical record abstraction, physician report (CMS Form 2728), and self-report in a baseline questionnaire. We evaluated agreement with kappa statistics (k) and sensitivity of self-report. Regression models were used to examine characteristics associated with agreement.

Results

The results showed excellent or substantial agreement between self-report and the medical record for diabetes (k = 0.93) and coronary artery intervention (k = 0.79), and poor agreement for chronic obstructive pulmonary disease (k = 0.20). Physician-reported prevalence for all diseases was equal or lower than that by self-report. Male patients were more likely to inaccurately report hypertension. Compared to white patients, African American patients were more likely to inaccurately report cardiovascular diseases.

Conclusion

In ESRD patients, self-report agreement with the medical record varies with the specific disease. Awareness of diseases of the cardiovascular system appears to be low. African American and male ESRD patients are at risk of low awareness of disease and educational interventions are needed in this high-risk population.

Introduction

Self-reported medical history of chronic diseases, such as hypertension and diabetes mellitus, is used in research studies to determine an individual participant's risk status and also the prevalence of specific diseases in a study population. The reported accuracy of self-reported health information is inconsistent in the literature. Most reliability or validation studies of self-reported chronic diseases compare participant self-report with the medical record; however, comparisons with a clinical examination [1], [2], [3], [4], [5], [6], [7] and also administrative data have been performed [8], [9], [10]. When comparing the medical record with patient self-report, some studies found good agreement for diabetes, hypertension, myocardial infarction (MI), and cerebrovascular disease [1], [11], [12], [13], [14], [15], [16], while others found lower agreement for some of these same conditions [5], [7], [9], [10], [14], [17], [18], [19].

Agreement between self-report and the medical record varies, depending upon the specific disease that is being evaluated. Chronic diseases that are diagnostically complex and require a clinical judgment in addition to medical testing may be more vulnerable to misclassification from self-reports [1], [13], [17]. Agreement also likely varies in different influenced populations, given their respective demographic, educational, and clinical characteristics. Previous studies have been performed in cohorts of study volunteers [11], [20], men of the Veterans Administration system in the United States [15], participants in the Women's Health Initiative [21], elderly populations [12], [14], and other cohorts that are more representative of the general population [1], [2], [7], [12], [17], [22], [23].

Patients with end-stage renal disease (ESRD) often have a very complex medical history with many comorbid diseases. The validity of self-report for disease classification in research studies of this population is unknown. Evaluating the agreement between self-report of medical history of ESRD patients and the medical record is necessary to interpret the use of self-report for risk factor classification in research and also potentially for clinical use. Identification of patients who are not in agreement with the medical record may represent patients who are unaware of their clinical diagnoses and may be targets for educational interventions.

In the United States, the Centers for Medicare & Medicaid Services (CMS) requires the completion of the Form 2728 for all incident ESRD patients [24]. This form has to be completed by the attending nephrologist and includes 20 questions about prevalent comorbid conditions. The Form 2728 represents an individual physician report for each patient. The agreement between physician report and the medical record in ESRD has been described [25], but the agreement between self-reported comorbid chronic diseases and physician report (Form 2728) has not been described.

The objective of this study was to examine the agreement between patient self-report and medical record in a cohort of ESRD patients with regard to eight chronic comorbid conditions, including congestive heart failure (CHF), MI, cerebrovascular disease, angioplasty or coronary artery bypass graft surgery (CABG), hypertension, diabetes, chronic obstructive pulmonary disease (COPD), and cancer. In addition, we compared the agreement between patient self-report and physician report, represented by the Form 2728, for each of these diseases. We further examined the effects of patient characteristics, including race, age, gender, or education on the positive agreement of patient self-reports with the medical record.

Section snippets

Study design and population

We conducted a cross-sectional analysis of baseline data from the Choices for Healthy Outcomes in Caring for End-stage renal disease (CHOICE) study, a national prospective cohort study of ESRD patients undergoing dialysis therapy [26]. At the time of enrollment, participants in this study were at least 18 years old, English or Spanish speaking, and had been diagnosed with kidney failure requiring initiation of outpatient dialysis. From October 1995 to June 1998, 1,041 patients were recruited

Characteristics of the study population

Of the 1,041 patients enrolled in the CHOICE cohort study, 965 (93%) patients completed ≥85% of the baseline questionnaire, including the self-report of comorbid diseases, and were included in this study. The average age of the study population was 58 years and 54% were male; 67% were white, and 28% were African American. At baseline, 75% of the patients received hemodialysis and 25% received peritoneal dialysis. The average number of existing comorbid conditions was three, and average days to

Discussion

This cross-sectional study indicates that in a population of incident ESRD patients, self-reported prevalence of comorbid diseases agreed variably with the medical record depending upon the specific disease. Self-report of diabetes had excellent agreement with the medical record. Self-report of CHF, MI, cerebrovascular disease/stroke, and cancer had only moderate agreement with medical records, while hypertension and COPD showed poor agreement. However, self-report of diabetes demonstrated

References (42)

  • S.A. Hunt

    ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure)

    J Am Coll Cardiol

    (2005)
  • S.L. Norris et al.

    Increasing diabetes self-management education in community settings. A systematic review

    Am J Prev Med

    (2002)
  • F.C. Lampe et al.

    Validity of a self-reported history of doctor-diagnosed angina

    J Clin Epidemiol

    (1999)
  • A.R. Feinstein et al.

    High agreement but low kappa: I. The problems of two paradoxes

    J Clin Epidemiol

    (1990)
  • E.S. Ford et al.

    Test characteristics of self-reported hypertension among the Hispanic population: findings from the Hispanic Health and Nutrition Examination Survey

    J Clin Epidemiol

    (1990)
  • W.H. Giles et al.

    The validity of self-reported hypertension and correlates of hypertension awareness among blacks and whites within the stroke belt

    Am J Prev Med

    (1995)
  • Y. Tsubono et al.

    Accuracy of self-report for stomach cancer screening

    J Clin Epidemiol

    (1994)
  • J. Luck et al.

    How well does chart abstraction measure quality? A prospective comparison of standardized patients with the medical record

    Am J Med

    (2000)
  • N. Haapanen et al.

    Agreement between questionnaire data and medical records of chronic diseases in middle-aged and elderly Finnish men and women

    Am J Epidemiol

    (1997)
  • S. Tretli et al.

    Reliability of questionnaire information on cardiovascular disease and diabetes: cardiovascular disease study in Finnmark county

    J Epidemiol Community Health

    (1982)
  • M.F. Weaver et al.

    HCV prevalence in methadone maintenance: self-report versus serum test

    Am J Health Behav

    (2005)
  • Cited by (89)

    • Evaluating agreement between clinic- and patient-reported outcomes for weight and co-morbidities at 1 year after bariatric surgery

      2023, Surgery for Obesity and Related Diseases
      Citation Excerpt :

      SCRs abstract data that are available in the medical records and provider notes according to specific criteria, and the PROs platform asks qualitative assessment questions directly to patients without accompanying definitions. Diabetes is a tightly controlled and closely monitored disease, so defining disease status may be more straightforward, resulting in strong agreement between patient self-report and clinical records for diabetes, as has been reported elsewhere [6,8,9]. Conversely, no singular diagnostic criteria exist for GERD, and the risk of misclassification is increased on both the clinician and the patient ends [17,18].

    View all citing articles on Scopus
    View full text