Multidimensional analysis of treatment adherence in patients with multiple chronic conditions. A cross-sectional study in a tertiary hospital

Abstract

Objective

Determine treatment adherence in patients with multiple chronic conditions (MCC).

Methods

A random patient sample ≥15 years, discharged from hospital with ≥1 chronic conditions (CC) was interviewed after 6–12 months. Analysis included variables in 5 dimensions (WHO): socio-demographics, disease, treatment, patient and health system characteristics. Morisky–Green adherence questionnaire was used. High chronic treatment complexity was defined as: >3 pills/day, >6 inhalations/day, >1 injection/day, pharmacological treatment plus diet or self-monitoring techniques.

Results

301 patients were interviewed (62 ± 15 years, 59% males). Despite good treatment information perception (79%), only 3% followed the patient education programme. Poor adherents (82%) were older (64 ± 14 years vs. 55 ± 16 years), had more CC (3.25 ± 2.02 vs. 2.62 ± 2.72), a higher frequency of hypertension (44% vs. 15%), ischaemic heart diseases: (21% vs. 4%), hyperlipidaemia (19% vs. 6%), more pills/day (5.78 ± 4.14 vs. 3.20 ± 4.70) and more complex treatments (95% vs. 70%) (p < 0.05). On multivariate analysis number of CC [3.68 (0.75–18.15)], pills/day [2.23 (1.02–4.84)], treatment complexity [4.00 (1.45–11.04)], and hypertension [2.57 (1.06–6.25)] were predictive of poor adherence (OR 95% CI p < 0.05).

Conclusion

The WHO conceptual framework allows the construction of poor adherence risk profiles in patients with MCC after hospital discharge.

Practice implications

Predictive variables of poor adherence could help clinicians detect patients with MCC most likely to present poor adherence.

Introduction

As countries experience epidemiological changes, chronic conditions (CC) are rapidly becoming more important determinants of national disease burden [1], [2] and a cause of overcrowding in tertiary hospitals [3].

Effective, evidence-based, treatments are now available. Their effectiveness however, is limited by many factors including inequities in drug prescription [4], lack of affordability in low income countries and lack of adherence to treatment.

Adherence in patients with CC is disappointingly low, ranging from 20% to 50% [5], [6] It is associated with increased morbidity and mortality [7] and drops dramatically within the first 6 months of therapy [8], despite effective interventions such as technical (pill boxes, plypills), behavioural (memory aids and reminders), educational or multifaceted approaches [8], [9]. As many as 200 factors have been hypothesised to influence adherence and these factors can also be classified as either intentional (rational decision-making, secondary effects) or non-intentional (treatment complexity, depression) [10], [11], [12].

Literature concerning adherence to treatment has grown impressively during the last decades [5], [6], [8], [9], [13], [14], [15]. Nonetheless, despite these efforts, lack of adherence remains an unresolved problem. As pointed out by several authors [9] and the World Health Organization (WHO) [16], the common belief that patients are solely responsible for taking their treatment is misleading, often reflects a misunderstanding of how other factors affect people's behaviour and avoids the development of effective interventions.

Therefore, the WHO has proposed the use of a conceptual framework which groups all factors into five sets named “dimensions”: social and economics, health care team and system characteristics, chronic conditions, as well as therapy- and patient-related factors. The “five dimensions approach” was developed after an in depth review of the literature by an expert committee, which considered that in order to improve adherence, it was necessary to work on several levels simultaneously. Therefore, these dimensions should be taken into account when studying adherence and designing interventions aimed at improving it. Unfortunately, the WHO approach has not yet been adequately tested.

Moreover, most studies focus on adherence in a single disease [5], [6]; which ignores the complex mix of conditions and treatments that characterise patients in the clinical setting, especially those over 65 years of age, who often present multiple chronic conditions (MCC) [17]. In fact, only a few studies in this area are available [18], and the relationship between treatment complexity and poor adherence has not been analysed in depth.

Finally, measurement of adherence in clinical practice is not systematically undergone, at least in part because there is not a widely accepted gold standard [9]. Therefore, the clinical consequences of the lack of adherence are often overlooked, while medication adherence has been called the “next frontier in quality improvement” [15].

We hypothesised that the analysis of adherence in the five dimensions of factors focusing on chronic patients who often receive complex treatments proposed by the WHO would allow the construction of a risk profile of poor adherence patients. In addition, this better knowledge might result in more effective strategies to improve adherence.

Therefore, the main objective of this study was to identify factors determining poor adherence to medication in patients with CC using the WHO conceptual framework [16].