Chronic care management in Danish general practice - a cross‒sectional study of workload and multimorbidity

We included persons aged 40 years or more. Patients were categorised into those with none, one, two, three or more chronic conditions.

We included consultations, home visits, email contacts, and telephone consultations when counting the total number of contacts; the total number of contacts was used as the denominator when calculating the frequency of contacts. In the analysis of the characteristics of the contacts, we limited the contacts to symptom‒ and disease‒related face‒to‒face contacts (excl. telephone, email contacts and consultations on prophylactic or administrative issues).

We dichotomized the number of additional psychosocial problems in the contacts (yes, no). The GPs were asked to assess the experienced burden of the contacts (1–10, 10 being the heaviest), and afterwards we categorized theses values into three groups (1‒4 = very easy/easy, 5‒6 = medium, 7‒10 = heavy/very heavy). Referral to specialist care was divided into three groups (outpatient clinics, primary care specialists or admission to hospital).

We identified the ten most frequently registered chronic conditions and calculated the proportion of contacts for each of them. In a subpopulation of contacts with patients having only one of the ten most frequent chronic conditions, we analysed how the GP weighted the chronic disease with regard to additional psychosocial problems raised during the contact, the time consumption and experienced workload of the contacts and their possible substitution by nurses.

Chi‒square tests were used to examine the difference between groups having no, one, two, three or more chronic conditions. An exploratory factor analysis of the 25 most frequently registered chronic conditions was performed with an orthogonal varimax rotation of the factors to identify occurring clusters of chronic conditions, expressed as eigenvalues greater than one. The isolated burden of each of the ten most common chronic conditions was compared using Chi‒square tests, and a Poisson regression was used to estimate the prevalence ratio of the experienced burden of these ten conditions adjusted for additional psychosocial problems and time consumption. Stata 11.1 was used to perform the statistical analyses.

The project was approved by the Danish Data Protection Agency and by the National Board of Health. The participating GPs received a fee that was partly dependent on the number of registered contacts.

In general practice, two out of three 40+ year‒old patients presented with at least one chronic disease. Patients with chronic conditions were more likely to raise additional psychosocial problems during the consultation than those without chronic conditions, and the GPs reported that the time consumption and the burden of these contacts tended to increase with the number of chronic conditions. Particularly demanding were contacts with patients suffering from depression and dementia. Accounting for one fifth of all registered chronic conditions, hypertension was the most commonly presented chronic condition. GPs estimated that nurses could substitute them in 32% of these contacts, but in only 7% of contacts with depressive patients.

COPD patients accounted for only 5% of the total number of contacts. This may be rather few compared with the number of contacts with hypertensive patients, because even though the prevalence of COPD is lower than the prevalence of hypertension [11, 12] COPD may, indeed, be a disabling condition with a strong impact on quality of life and mortality and with a potential higher need for health services. It is therefore possible that the consumption of general practice services by patients with chronic conditions to some degree is due to some patients being aware of the importance of prevention (e.g. of hypertension) whereas others are more reticent concerning presenting at their GP. Such a difference may also be due to sociodemographic factors as COPD is shown to be associated with a low socio‒economic status [13]. Moreover, Danish GPs are paid by a capitation fee as well as fee‒for‒service and the latter makes it profitable to do many consultation checks. It cannot be ruled out that this plays a role, as hypertension checks may be easy consultations, often performed by the clinical staff. Contrary to this consultations with COPD patients according to our findings are more complex, indicating that the frequency of hypertension checks is partly defined by other factors than by a genuine need for physical care delivered by the GP.

We may expect that the rise in primary sector consultations for chronic conditions will translate into a growing workload in general practice in the future. At the same time, clinical staff is increasingly involved in substituting the GPs. However, as the staff primarily eases the workload of the GPs by handling less complicated problems, the GPs’ work with patients with complicated problems can be expected to intensify.

The present study included a large number of GPs, nurses and patient encounters and we obtained information not otherwise available about the contents of these encounters and how the GPs experienced these contacts. The study included a representative share of the GPs in the region, corresponding to 13% of all GPs in Denmark. Information about the contacts was based on the GPs’ direct experience as the questionnaires were filled in immediately after the contacts and recall bias was thereby minimised.

We were able to offer a detailed characterization of these encounters as we invited the GPs to comment on several aspects of the their experience. The number of chronic conditions may be underestimated, as the registration by the GPs may have been incomplete. The extent of this possible misclassification is unknown. Moreover, we did not present the GPs with a definition of chronic condition or disease, as we aimed in this pragmatic study to picture the activities and the daily work of the GPs as they perceived it. It may be argued that for example uncomplicated hypertension and overweight are risk factors rather than chronic conditions and should as such not be included as chronic conditions as this would imply a risk of overestimating the prevalence of chronic conditions in general practice.

Encounters in general practice may be difficult to clearly mark out with a single reason for encounter, as these contacts often are characterized by a history of GP‒patients relationship, which makes the GP take psychological factors of the patient into consideration as well as the actual reason for encounter. However, as this study was based on the GPs’ stating chronic diseases, the potential uncertainty of the character of the reasons for encounter does not hamper the results.

We have identified no other studies exploring how GPs perceive the workload of managing patients with chronic conditions. However, several studies have demonstrated an increase in the prevalence of psychiatric disorders in primary care. In an earlier study of our data, we found that a larger share of patients gave depression as the reason for their encounter in 2009 than in a previous study conducted 16 years earlier [10]. This rise in the prevalence of depression in general practice is supported by a study from 2008 in Dutch general practice, which showed a significant increase in the prevalence of depression and anxiety disorders [14]. Likewise, an increase in the prevalence of depression from 1998 to 2008 was found in Australia [15]. Even though these studies focused on the general morbidity in general practice and not on registration of contacts, including chronic diseases, as the present study, the results are likely to show the same pattern of an increase in the prevalence of depression in primary care.

Even though severe somatic illnesses as such as COPD will continue to be a marked challenge for the GPs it is worth being aware that GPs are likely to be facing mounting pressures in their daily work in the future due to the general rise in the prevalence of depression, stress and other anxiety disorders and complaints. Moreover, patients with depression and anxiety disorders have been documented to consult their GPs more frequently about minor ailments than patients without these psychiatric disorders, which emphasizes the complexity of the needs of these patients [16]. Further research is needed to identify the needs of these patients and how best to meet them, and studies on changes in populations’ threshold for seeking help would be relevant to shed light on whether there is a growing tendency to present the GP for minor concerns compared with earlier.