Concordance between elderly patients’ understanding of and their primary healthcare physician's diagnosis of heart failure

Abstract

Objective. The objective was to study primary healthcare patients’ understanding of their diagnosis of heart failure (HF), using patients treated for diabetes mellitus (DM) as a comparative group. Design. A cross-sectional community based study. Setting. Karlskrona community situated on the Swedish south-east coast with 60 600 inhabitants. Subjects. A total of 1402 subjects, aged 60–96 years in 10 age cohorts, selected randomly from the national population registry participating in the Swedish National study on Ageing and Care – Blekinge. Main outcome measures. Understanding of diagnosis of HF or DM in primary healthcare. Prevalence of cognitive impairment. Results. In all, 39.8% of patients with a diagnosis of HF treated in primary healthcare and 97.1% of patients with DM had an understanding of their respective diagnosis. Cognitive impairment was significantly more prevalent in the groups of patients treated for HF (OR 1.9, 95% CI 1.2 to 3.0) and DM (OR 1.8, 95% CI 1.1 to 3.1), when compared with those not treated for either HF or DM. The odds ratio for understanding of diagnosis was 0.013 (95% CI 0.003 to 0.052, p<0.001) in patients treated for HF, compared with patients treated for DM, when adjusted for the subject's age, sex, and cognitive function. Conclusion. In this study it was shown that patients’ understanding of their diagnosis was highly dependent on diagnosis, independently of age, sex, or cognitive function. The results suggest that there is room for improvement of care in primary healthcare, to increase HF patients’ understanding of their diagnosis.

• Cognitive impairment
• diabetes mellitus
• heart failure
• primary healthcare

Heart failure (HF) is a major and growing health problem in the Western world, due to decreased mortality in acute myocardial infarction, improved pharmacological treatment of HF, and an ageing population [1–3]. Despite improvement in pharmacological treatment of HF the prognosis remains poor, with a 5-year survival of 25% in men and 38% in women and a 10-year survival of only 11% and 21% in men and women respectively [4]. Prevalence of HF rises from approximately 1% in persons 50 years of age to over 10% in persons older than 80 years [1], [2].

Heart failure is in contrast to diabetes mellitus a difficult diagnosis in primary health care. Concordance between patient and physician is important for treatment results.

• Understanding of diagnosis was found to be significantly lower in patients treated for heart failure than patients treated for diabetes mellitus.

• Differences in understanding could not be explained by the patients’ age, sex, or cognitive function.

In Sweden a major proportion of elderly patients with HF are treated in primary healthcare, but at most healthcare centres no special patient education is currently provided. In primary healthcare diagnosis of HF is difficult [5], [6] and is still often based on clinical examination, electrocardiogram (ECG), and chest X-ray [7]. Treatment of patients with HF is also often complex, requiring several medications, lifestyle modifications, and symptom monitoring. Studies have shown that compliance with pharmacological therapy and dietary restrictions among patients with HF is low and that it is a major cause for hospitalizations [8]. Patient education is an important part of HF management to improve compliance with pharmacological treatment [9], decrease the number of hospital admissions [10], and improve quality of life [11]. In contrast to HF, diagnosis of diabetes mellitus (DM) today is easy and patients learn how to self-monitor their disease and to modify treatment when so needed. This, however, is a result of a major effort to increase patients’ understanding of their disease, using patient education and treatment support. Whereas nurse-led care for patients with DM has been around for more than 10 years in Swedish primary healthcare, the equivalent for patient education of HF patients only exist at a few healthcare centres.

The aim of this study was to describe patients’ understanding of their diagnosis of HF, treated in primary healthcare, using subjects with a diagnosis of DM as a comparative group and to study factors associated with differences in understanding.

Material and methods

The Swedish National study on Ageing and Care [12] is a population-based, multi-centre cohort study, which started enrolment of subjects in 2001. The study has four participating centres in Sweden. One of the centres is the Swedish National study on Ageing and Care – Blekinge, which encompasses the Karlskrona community with 60 600 inhabitants. Since 1991 nurse-led care for patients with DM has been practised at all five healthcare centres in the community, but has not been available for any patients with HF.

Briefly the study sample was selected randomly in age cohorts of 60, 66, 72, and 78 years old. In the age cohorts of 81, 84, 87, 90, 93, and 96 years old all inhabitants were selected. The participants were invited by mail to take part in a medical, psychological, dental, and questionnaire examination by research staff in two sessions of three hours. If the subjects agreed to participate but were not able to come to the research centre the investigations were performed in their homes.

Informed consent was obtained from all participants and they were asked to sign a release form for their medical records. After the invitations had been sent out all those who had not responded were called and invited again. If they decided not to participate the reason for this was registered.

The present study was carried out as a cross-sectional study on all the 1402 enrolled participants in the baseline examination of the Swedish National study on Ageing and Care – Blekinge. As a part of the medical examination the subjects were asked if they were under any treatment by a physician for hypertension, ischaemic heart disease (angina pectoris, previous myocardial infarction), DM, or HF. If the subject could not respond the closest relative or principal care-giver was asked. As part of the psychological test battery, subjects underwent testing with the Mini Mental State Examination [13]. Electrocardiogram was done and investigated for signs of arrhythmia or ischemia. The computerized medical records in primary healthcare for all subjects with at least one positive answer to the above questions or ECG abnormalities (positive screen) were reviewed to see if they were diagnosed with HF or DM by their primary healthcare physician. The patient was thus categorized as having HF or DM based on the primary healthcare physician's diagnosis, as stated in the medical records, regardless of which investigations and tests the diagnosis was based on. The study was approved by the Research Ethics Committee at Lund University.

Statistical analysis

An independent samples t-test was used to compare differences in mean levels of continuous variables (age) in two groups. Categorical data (understanding of diagnosis of HF/DM between age groups) were compared with chi-square analysis. The Mini Mental State Examination score was dichotomized and a subject with a Mini Mental State Examination score <24 was characterized as having cognitive impairment [14].

Logistic regression was used to fit a model with cognitive impairment (CI) as the dependent variable and diagnosis as the independent variable. The model was adjusted for sex and age. Logistic regression was also used to fit a model, with understanding of diagnosis as the dependent variable and diagnosis according to primary healthcare medical records as the independent variable. The model was adjusted for sex, age, and CI. All analyses were done using the statistical package Stata version 8.0 (Stata Corporation,Texas, USA).

Results

This study comprised all 1402 participants in the baseline examination of the Swedish National study on Ageing and Care – Blekinge (Table I). Of these, 40 who were treated by private care physicians were excluded as we were not able to gain access to their medical records and also because their care was organized differently. A further 25 had missing results on the Mini Mental State Examination. The overall participation rate was 60.5%. The participation rate was 75.2%, 66.3%, 53.7%, and 47.0% in the age groups 60–69, 70–79, 80–89, and 90-years. Those that did not participate were older and often stated that they were too ill to participate.

Table I.  The number of individuals, with percentages in parentheses, in different subgroups of participants in the baseline investigation of the Swedish National study on Ageing and Care – Blekinge.

Group of individuals	Number	(%)
Negative screen	566	(39.6)
Positive screen	771^a	(55.0)
Treated privately^b	40	(2.8)
Missing MMSE^c	25	(1.8)
Total	1402

^aSubjects treated for hypertenson, ischaemic heart disease, diabetes mellitus, or heart failure. ^bPatients treated by private primary care physicians. ^cMini Mental State Examination.

Among the 771/1337 (58.0%) participants who were screened positive, 401/771 (52.0%), 361/771 (46.8%), and 130/771 (16.9%) stated that they were treated for hypertension, ischemic heart disease, and diabetes mellitus respectively and 370/771 (48%) had ECG abnormalities. According to primary healthcare medical records 133/771 (17.2%) were treated for HF and 130/771 (16.9%) for DM. Twenty-five participants were treated for both HF and DM and were excluded from the analyses below.

Patients treated for HF were significantly older (mean age 83.6 years, 95% CI 82.4 to 84.9 years), than patients treated for DM (mean age 75.3 years, 95% CI 73.6 to 77.1 years), p<0.001.

Of the 108 subjects treated for HF according to primary healthcare medical records, this was known to 43 (39.8%) and of the 105 subjects treated for DM this was known to 102 (97.1%). Understanding of treatment was 1/4 (25.0%), 6/17 (35.3%), 25/65 (38.5%), and 11/22 (50.0%) in subjects treated for HF and 29/30 (96.7%), 31/32 (96.9%), 37/38 (97.4%), and 5/5 (100%) in those treated for DM in the age groups 60–69, 70–79, 80–89, and 90– respectively (Figure 1). There was no significant difference in understanding between age groups with the same diagnosis.

Figure 1.  Understanding of diagnosis of heart failure and diabetes mellitus in different age groups.

Cognitive impairment was present in 48/108 (44.4%) of subjects treated for HF compared with 25/105 (23.8%) in those treated for DM and 193/1099 (17.6%) of those not treated for HF or DM. Cognitive impairment was significantly more prevalent in the groups of patients treated for HF (OR 1.9, 95% CI 1.2 to 3.0, p<0.05) and DM (OR 1.8 95%, CI 1.1 to 3.1, p<0.05), when compared with those not treated for either HF or DM. There was no significant difference in prevalence of CI between patients treated for HF or DM, when adjusted for age and sex.

In a univariate logistic model the odds ratio for understanding of diagnosis was 0.019 (95% CI 0.006 to 0.065, p<0.001) in patients treated for HF compared with patients treated for DM. When the model was adjusted for age, sex, and CI the odds ratio was 0.013 (95% CI 0.003 to 0.052, p<0.001). The effect of the independent variables age, sex, and CI was not statistically significant.

Discussion

In this study we found that patients treated for HF in primary healthcare had a much lower understanding of their diagnosis than patients treated for DM. Age, sex, and cognitive function did not have a significant effect on the HF patients’ understanding of their diagnosis. The number of patients with HF could be underestimated, as those who had not participated in the Swedish National study on Ageing and Care – Blekinge were more likely to be from the older age groups and often stated that they were too ill to participate.

There are several different factors that may explain the difference in understanding seen between patients treated for HF and DM. HF is a difficult diagnosis in primary care [5], [6] in contrast to DM. The use of echocardiography, which is essential for a correct diagnosis, is still seldom used in primary healthcare [7]. Brain natriuretic peptide, which recently has been introduced as a specific marker for HF, is often not used in routine care [17], [18]. Studies have shown that optimal treatment, according to guidelines, is seldom used, which might not be surprising when the diagnosis is often mainly based on clinical signs [19]. The diagnosis is thus not always clear for the primary healthcare physician and this may reflect on both his/her choice of medical treatment and the information given to the patient.

Much effort has been put into care and education of patients with DM in Swedish primary healthcare since the early 1990s, in contrast to patients with HF. With regular visits to a specially trained nurse for advice on exercise, nutrition, self-monitoring etc. and telephone contact as needed, combined with at least yearly visits to a primary healthcare physician, the programme has been successful and has increased the patient's ability to understand and to deal with his/her disease [15].

The presence of comorbidities in the elderly is high with symptoms similar to those of HF [16]. The high prevalence of comorbidities in this elderly group of patients also makes HF a difficult disease, both for the primary healthcare physician to diagnose and for the patients to understand.

A large part of the elderly population is cognitively impaired. Both HF and DM have been shown to be independently associated with poorer cognitive function [20], [21]. Cognitive impairment is important to identify as it has in been shown to increase mortality independently of diagnosis [22]. In this study CI was more prevalent in both HF and DM patients than in those not treated for either HF or DM, but had no significance for understanding of diagnosis. Compliance in treatment of HF has been shown to differ regarding different aspects of overall compliance [23]. Cognitive function has in another study been shown to be important for participation in outpatient treatment programmes [24].

Heart failure is a major health problem and contributes to a large part of the healthcare costs. The cost for hospitalization account for 50–75% of the total cost for treatment of HF [25]. A reason for this is the high risk of readmission, with a rate of 40–50% within six months [26]. The main factors behind this, which are potentially preventable with patient education, are poor compliance with medications, non-adherence to lifestyle modification, and failure to seek medical advice when symptoms become worse [27], [28].

Conclusions

The findings in this study show that a large proportion of patients treated for HF in primary healthcare have no understanding of their diagnosis, in contrast with patients treated for DM. Our results argue that there is a great need for improvement of care for patients treated for heart failure in primary healthcare. Further studies are needed to elucidate the effect of patient education and treatment support in this group of elderly and often cognitively impaired patients.

The authors would like to thank the staff and participants of the Swedish National study on Ageing and Care – Blekinge for their help and interest in the study. They are grateful to the Swedish Ministry of Health and Social Affairs, Blekinge Institute of Technology, Kristianstad University, Blekinge County Council and Blekinge Institute for Research and Development for funding this study.