Methods

Setting

Danish healthcare is mainly tax-financed and includes free-of-charge access to services11. General practices are typically independent, physician-owned clinics, and nearly all Danes are listed with a specific general practice clinic. General practitioners (GPs) are remunerated through a mix of capitation and fee-for-services based on a national agreement between the Danish Regions and the Organisation of GPs. In Denmark, GPs are responsible for most prescriptions and chronic care management12.
The current study was a part of a larger quality improvement project focusing on polypharmacy and communication inspired by the World Health Organisations global initiative “Medication without harm”13. The project was conducted in a close collaboration between the Centre for Health and Care in the Municipality of Frederikshavn, Denmark, and a large GP clinic in Frederikshavn (hereafter GPF).
The GPF is a large clinic with a strategic focus on older patients and patients with chronic diseases. The GPF has a close collaboration with the municipal and regional health services. The GPF has an affiliated population of approximately 8,900 patients, of which more than 2,300 citizens are older than 65 years. The GPF employs eight GPs, ten nurses, ten medical students, or GP trainees, a social and health assistant, a pharmaconomist, and a physiotherapist.
In the Municipality of Frederikshavn, the Centre for Health and Care runs 12 care homes, of which eleven are covered by a specific GP practice. The GPF is affiliated ”care home doctor” for four of the care homes in the municipality.

Ethics

The project was approved by the Management in the Municipality of Frederikshavn. According to Danish legislation, no formal permission from the national or regional Committee on Health Research Ethics was required for this type of study, as patients were not treated inferior to usual care and no biological material was collected. It was conducted as a quality improvement project and informed consent was not required for the specific data collected. The study was conducted in accordance with the Basic & Clinical Pharmacology & Toxicology policy for experimental and clinical studies14. The study is in compliance with the General Data Protection Regulation15 and a part of North Denmark Region’s record of processing activities (K2023-008). The study is registered in ClinicalTrials.gov (registered January 31, 2023, awaiting ClinicalTrials.gov ID).

Study design and population

The study was conducted in the 2-years period from the January 2020 to December 2021 using an uncontrolled before-and-after design. The study included care home residents living in selected care homes, in which the GPF was associated “care home doctors”, and community-dwelling patients with chronic disease listed with the GPF.

The Chronic Care Model

In Denmark, chronic care consultations are provided to patients with one or more chronic conditions. The organisation of these consultations varies across GP clinics, depending on e.g. the size of the clinic and the competencies in the staff group16. In connection with this project, a new, local Chronic Care Model was drawn up for patients with one or more chronic conditions such as diabetes, chronic obstructive pulmonary disease, hypertension, heart failure, or atrial fibrillation. The overall aim of the model was to obtain sufficient depth and breadth in the chronic care consultations over a one-year period.
The new Chronic Care Model is illustrated in figure 1.
In addition to the Chronic Care Model, a new cross-sectoral communication model was established. This included regular contact between the care home nurses and the GPF (weekly by telephone, e-mail, visit, and/or online conference); support opportunity from a pharmacist employed in Frederikshavn Municipality; and support opportunity from a specialized geriatric department at the hospital every second week.
————————————————–figure 1 near here——————————————————————-

Care home residents

The four care homes affiliated with the GPF accommodate 190 residents. Of these, 128 were patients in the GPF (the remaining residents kept their family doctor when moving into the care home) (figure 2). In the period from March 24, 2020, to June 16, 2021, the intervention was offered to new residents and residents that had not yet attended a consultation focusing on pharmacological treatment in The Chronic Care Model.

Community-dwelling patients with chronic disease

The GPF had 1,800 community-dwelling patients with chronic diseases listed in the period 2020-2021 (figure 2). From June 3, 2020, to November 16, 2021, patients were invited for the consultation focusing on pharmacological treatment in the month of their birthday and, thereby, included randomly and consecutively throughout the study period.
—————————————————figure 2 near here————————————————————–

Intervention

In this study, the focus was placed on the annual consultation focusing on pharmacological treatment in the Chronic Care Model. This specific consultation constituted the “intervention”. It included a structured review of the patient’s health state, in addition to a structured medication review with a focus on appropriate medication and deprescribing. Medication changes were registered as deprescribing (dose reduction or stopping/pausing of medications), new prescription, and other medication changes (e.g., dose increase or change in dosing interval). Additionally, issues such as treatment plans for addictive drugs, dose dispensing, resuscitation, life-prolonging treatment, and terminal care were discussed when relevant. A selected group of providers (two doctors and three nurses) from GPF were responsible for conduction of the intervention in the present study. The GPs performed the medication reviews. In care home residents, the GPs also carried out the related consultation. In community-dwelling patients, the nurses were responsible for the consultations with the GPs as close support.

Data collection and outcomes

Before and 3-4 months after the consultation focusing on pharmacological treatment, information regarding medication changes and health-related outcomes were collected during consultations.
Health-related outcomes were collected by a nurse or the patient’s contact person together with the patient and, if possible, also relatives. The primary outcome was changes from baseline to 3-4 months follow-up in 1) self-reported health status (on a scale from 1 to 10). Secondary outcomes were 2) general condition (rated on a 5-point Likert Scale as “much below average”, “below average”, “average”, “above average” and “much above average”); and 3) functional level (rated on a 5-point Likert Scale as “independent”, “frail”, “mild disability”, “disability” and “severe disability”). General condition and functional level were determined by clinical evaluation. The outcomes were developed with inspiration from Garfinkel17.

Statistical analysis

Descriptive and non-parametric data were summarized and displayed by medians [inter quartile range (IQR)] for continuous data and as proportions (percentages) for categorical data. Parametric data was displayed by means and standard deviation (SD). Paired t-test was used to compare means of self-reported health status at baseline and follow-up. McNemars test was used when comparing paired proportions for categorical variables. General condition and functional level were dichotomised and analysed as proportion of patients with general condition rated as “average or above” (defined as: “average”, “above average” or “much above average”). The proportion of patients with functional level rated as “without any disability” was defined as the categories: “independent” or “frail”. Statistical analyses were performed in STATA 17. Statistical significance was indicated by a two-tailed p value of 0.05.