Clinical multimorbidity and physical function in older adults: a record and health status linkage study in general practice

Abstract
Background. Multiple chronic conditions occurring in the same individual are associated with adverse health outcomes. In family practice, individuals are seen who, over time, may experience many different symptoms, illnesses and chronic diseases. Measures for defining multimorbidity, which incorporate the diverse range of health problems seen in population-based family practice, remain to be developed. We have investigated whether routinely collected consultation data could be used as the basis for a simple classification of multimorbidity that reflects an individual's overall health status. Methods. Morbidity consultation data for 9439 English patients aged 50 years and over in an 18-month time period were linked to their self-reported physical health status measured by Short-Form 12 at the end point. Associations between physical function and all-cause multimorbidity counts were estimated relative to single morbidity only, and between physical function and morbidity severity (185 morbidities categorized on four ordinal scales of severity) relative to persons who had not consulted about any of the 185. Results. In the 18-month period, 19% had consulted for a single morbidity and 23% for six or more (a high multimorbidity count). An estimated 24% of poor physical function in the family practice consulting population may be attributable to high multimorbidity. There was an increasing strength of association between poor physical function and increasing severity of multimorbidity on all four severity scales. Estimated associations (adjusted odds ratios) of the most severe morbidity categories with poor physical function were, for each of the four scales, respectively, 5.6 for chronicity [95% confidence interval (CI) 4.4–7.1], 7.0 for time course (4.5–10.6) and 3.6 for health care use (2.0–6.6) and for patient impact (6.7; 5.2–8.8). Conclusions. Multimorbidity defined by using routinely collected family practice consultation data and classified by count and by severity was associated with poorer physical function. This approach offers the potential for systematic use of routine records to classify multimorbidity and to identify groups with high likelihood of poor physical status for needs assessment and targeted intervention.