Article Figures & Data
Tables
Characteristic n % Female 530 55.2 White race 787 81.9 Low socioeconomic status 129 13.4 Depressive symptoms (PHQ-9 >4)a 342 35.6 Anxiety symptoms (GAD-7 >4)a 201 20.9 Any inpatient admissiona 209 21.8 Any emergency department visitb 222 23.1 Death within 10 months of survey 24 2.5 Mean No. (SD) Median (5%, 95%) Age at survey, y 961 75.6 (5.7) 75 (68, 86) Years enrolled before survey 961 12.4 (4.4) 14 (4, 17) Self-reported disease count 961 7.9 (2.7) 8 (4, 12.5) Self-reported disease burden score 960 16.8 (10.0) 15 (5, 36) CCI year before survey 961 1.9 (1.9) 1 (0, 6) Inclusion conditionsc 961 3.6 (0.8) 3 (3, 5) Outcomes General health scored 961 58.9 (21.4) 62 (20, 87) Physical component scaled 958 36.4 (11.4) 36 (19, 54) Mental component scaled 958 54.8 (9.0) 57 (37, 66) Financial constraints scored 958 77.7 (26.2) 92 (25, 100) Overwhelmed scored 959 78.5 (22.4) 83 (33, 100) General self-efficacy scored 959 80.7 (16.2) 81 (50, 100) Inpatient admissionsb (range = 0–7) 961 0.32 (0.72) 0 (0, 2) Emergency department visitsb (range=0–14) 961 0.34 (0.80) 0 (0, 2) Office visitsb (range = 0–62) 961 6.20 (4.97) 5 (3, 16) -
CCI = Charlson comorbidity index; PHQ-9 = 9-item Patient Health Questionnaire; GAD-7 = Generalized Anxiety Disorder 7-item scale.
-
a Score of >4 indicative of symptoms of each condition.
-
↵b During the 10 months after the survey.
-
↵c From original inclusion criteria for cohort of having 3 or more of a list of 10 chronic conditions.
-
↵d Scored on a range from 1 to 100; higher scores indicate better outcomes (eg, better health, fewer financial constraints, greater self-efficacy).
-
- Table 2
Summary of Significant Associations Between Morbidity Measures and Patient-Reported Outcomes
Patient-Reported Outcomesa Morbidity Measure General Health Status β (CI)b Physical Well-being β (CI)b Emotional Well-being β (CI)b Fewer Financial Constraints Odds Ratio (CI)c Less Overwhelmed Odds Ratio (CI)c Self-efficacy Odds Ratio (CI)c Quan adaptation of CCI (ICD-9)15 (range=0–12) −1.91 (−2.50 to −1.33)d −0.68 (−0.99 to −0.37d 0.06 (−0.19 to 0.31) 0.82 (0.73 to 0.91)d 0.87 (0.78 to 0.96)e 0.96 (0.86 to 1.08) Self-reported disease burden26 (range=1–89) −0.71 (−0.84 to −0.59)d −0.49 (−0.56 to −0.42)d −0.11 (−0.16 to −.05)d 0.96 (0.94 to 0.99)f 0.95 (0.93 to 0.97)d 0.96 (0.94 to 0.99)f Anxiety symptoms (GAD-7)39 (range=0–21) −2.75 (−5.88 to 0.37) 1.41 (−0.24 to 3.06) −5.90 (−7.25 to −4.56)d 0.64 (0.37 to 1.09) 0.57 (0.35 to 0.93)e 0.79 (0.48 to 1.31) Depressive symptoms (PHQ-9)38 (range=0–27) −12.01 (−14.75 to −9.27)d −4.99 (−6.44 to −3.54)d −5.87 (−7.05 to −4.69)d 0.66 (0.39 to 1.13) 0.30 (0.18 to 0.50)d 0.18 (0.10 to 0.33)d -
CCI = Charlson comorbidity index; GAD-7 = Generalized Anxiety Disorder 7-item scale; ICD- 9=International Classification of Disease, 9th edition; PHQ-9=9-item Patient Health Questionnaire.
-
Note: Associations expressed as point estimates with confidence intervals within separate models for each outcome.
-
↵a For all outcomes, higher outcome values represent a better state (eg, better physical functioning, fewer financial constraints). All models adjusted for morbidity measures above and age, sex, race, socioeconomic status, and length of enrollment.
-
↵b Linear regression: β estimates; nonsignificant confidence intervals cross zero.
-
↵c Logistic regression: odds ratios; nonsignificant confidence intervals cross 1.0.
-
↵d P value <.001.
-
↵e P value <.05.
-
↵f P value <.01.
-
- Table 3
Summary of Significant Associations Between Morbidity Measures and Utilization Outcomes
Utilization Outcomesa Morbidity Measure Outpatient Utilization Negative Binomial Regression Rate Ratio (CI)b Inpatient Admission Logistic Regression Odds Ratio (CI)c Emergency Department Admission Logistic Regression Odds Ratio (CI)c Quan adaptation of CCI (ICD-9)15 (range=0–12) 1.05 (1.02–1.09)d 1.17 (1.08–1.26)d 1.12 (1.04–1.22)d Self-reported disease burden26 (range=1–89) 1.02 (1.01–1.02)e 1.03 (1.01–1.04)d 1.01 (0.99–1.03)f Anxiety symptoms (GAD-7)39 (range=0–21) 1.23 (1.03–1.47)f 1.01 (0.65–1.58) 0.94 (0.62–1.43) Depressive symptoms (PHQ-9)38 (range=0–27) 1.00 (0.86–1.16) 0.81 (0.55–1.20) 1.72 (1.19–2.49)d -
CCI = Charlson comorbidity index; GAD-7 = Generalized Anxiety Disorder 7-item scale; ICD-9=International Classification of Diseases, Ninth Edition; PHQ-9=9-item Patient Health Questionnaire.
-
Note: Associations expressed as rate or odds ratios with confidence intervals within separate models for each outcome.
-
↵a All models adjusted for other morbidity measures and age, sex, race, socioeconomic status, follow-up time, and length of enrollment.
-
↵b Negative binomial regression; nonsignificant confidence intervals cross 1.0.
-
↵c Logistic regression; nonsignificant confidence intervals cross 1.0.
-
↵d P value <.01.
-
↵e P value <.001.
-
↵f P value <.05.
-
Additional Files
The Article in Brief
Association of Patient-Centered Outcomes With Patient-Reported and ICD-9-Based Morbidity Measures
Elizabeth A. Bayliss , and colleagues
Background Evaluating patient-centered care for complex patients requires the ability to approriately measure morbidity (illness) for a variety of clinical outcomes. This study compares the contributions of self-reported morbidity and morbidity measured using administrative diagnosis data for both patient-reported outcomes and utilization outcomes.
What This Study Found A comprehensive assessment of a patient�s morbidity requires both subjective and objective measurement of diseases and disease burden, as well as an assessment of emotional symptoms. Comparing two different approaches to gauging morbidity - (1) objective measurement using ICD-9 diagnosis codes and (2) subjective measurement using patient-reported disease burden and emotional symptoms - researchers conclude both are needed. In data on 961 older adults with three or more medical conditions, morbidity measured by diagnosis code is more strongly associated with higher utilization, whereas self-reported disease burden and emotional symptoms are more strongly associated with patient-reported outcomes.
Implications
- Accurate measurement strategies to account for morbidity burden will become increasingly important in developing new methods for evaluating patient-centered care delivery for complex patients.