Co-morbidity: we need a guideline for each patient not a guideline for each disease

Imagine you are a GP seeing a 79-year-old woman who has osteoporosis, type 2 diabetes mellitus, hypertension and chronic obstructive pulmonary disease, all of moderate severity. She is complaining of worsening pain in her left knee when walking, which you think might be osteoarthritis and start to investigate. While doing that you are wondering if you are following all the recommendations of the five individual disease guidelines. This issue of a patient with multi-morbidity each addressed by single-disease guidelines is a common scenario that faces most primary care physicians. The simultaneous presence of multiple diseases is now the normal pattern seen in populations of patients.1 In Holland, multi-morbidity of four or more chronic diseases occurs in 7% of 45- to 64-year olds, increasing to 30% of 65- to 74-year olds and to 55% of the population aged ≥75.2 In Québec, the prevalence of patients with two or more medical conditions in the 18- to 44-year-old, 45- to 64-year-old and ≥65 years age groups is 68%, 95% and 99% among women and 72%, 89% and 97% among men, respectively.3 Over half of the patients with cancer from the 2004–05 Australian National Health Survey report arthritis as a co-morbidity followed by mental health problems (16.1%), asthma (15.1%), cardiovascular disease (14.6%) and diabetes (9.0%).4 Guidelines describing care of patients largely remain single disease-oriented documents. Primary care health professionals trying to implement the evidence from these guidelines find dealing with multi-morbidity challenging5; only 14% of physicians in primary care report consulting a guideline weekly.6 With 2550 individual summaries on the National Guideline Clearinghouse, it is not surprising that the three main reasons for the low rate of use are lack of awareness, lack of familiarity and lack of agreement between the guidelines.7,8 Following advice from multiple guidelines may actually have unanticipated and undesirable effects.9 Using the hypothetical 79-year-old woman described above and reviewing five clinical practice guidelines, each addressing one of these conditions, resulted in 12 separate medications or 19 doses per day. This combination of drugs may cause drug adverse effects, interactions between drugs and adverse effects of drugs used for one condition on other conditions. There has been a gradual move in guideline development from advice delivered by expert panels towards consensus statements from clinicians representing differing contexts of care. The process includes epidemiologists providing expertise in appraisal of the evidence using a standardized approach.10,11 This standardization provides an opportunity to share this information across different guideline groups reducing duplication of effort. This also allows information from the evidence to be used in a novel manner. It is possible to create from this appraised evidence a database of screening, diagnostic and therapeutic benefits and harms for patients with multi-morbidity. Appropriately considered by clinicians and epidemiologists, working with patients with multi-morbidity, this information would allow a much more coherent guidance for the most common disease groupings. For example in patients with chronic diseases in whom life expectancy is shortened, guidelines regarding screening may needed to be adjusted. In a 60-year-old man with diabetes, congestive heart failure, lung disease, stroke and substantial frailty, the payoff time for colorectal cancer screening (minimum time until benefits exceed harms) is 7.3 years, whereas life expectancy is 3.7 years, making screening more questionable.12 Tables of risk benefit for all the evaluated interventions ranked by greatest benefit based on the individual patient's age, gender and morbidities would give the clinician and patient specific guidance. As time is so limited during primary care consultations, this is likely be very helpful in choosing which two or three high impact interventions to discuss during the consultation with the patient. Such a system would work well within an electronic medical record where the morbidities are already available and so display of risk benefit would be automatic. This multi-morbidity-based guidance, helping patients and clinicians see clearly the effectiveness of interventions, might effectively address limitations of current guidelines and reduce the care gap that exists between what we know and what we do. Funding: None. Ethical approval: none. Conflict of interest: none.