Many elderly are treated with several different drugs. With an increase in the number of medicines, there is a greater risk of interactions and adverse effects. There are however also risks associated with the underuse of recommended treatment. For several diseases, it has been shown that it is more common for elderly than for younger patients to receive suboptimal treatment with the recommended drugs [
The recommendation for
For the diagnosis
For patients with
Aspirin is also the best documented secondary prevention of
This study was performed as a quality improvement project on a middle-sized family health care centre. We wanted to describe how well general practitioners (GPs) adhere to current guidelines when they investigate and treat elderly patients with cardiovascular disease. We also wanted to evaluate feedback and local continuing education as an act for improvement of this treatment.
The study was performed in 2006–2008 in a family health care center in Eslöv, a municipality in southern Sweden with a population of approximately 30 000. The center has a registered population of 10500. All nine GPs at the center participated in the intervention. Data was collected from the patient’s medical records on patients who met the following inclusion criteria: ≥65 years, visited the health care center between April (July for hypertension) and December 2006, on this visit, received one or more of the following diagnoses: hypertension, heart failure (HF), ischemic heart disease (IHD), chronic atrial fibrillation (CAF), or prior stroke (inclusive TIA).
For the diagnosis hypertension, patients who received the diagnosis less than 6 months ago were excluded.
The medical records were systematically reviewed by a doctor and the following variables were investigated, as recommended in the guidelines of 2006.
For Target blood pressure reached (130/80 for diabetes and 140/90 for others)? If not, number of antihypertensive drugs? ECG during the last 2 years? Blood lipids tested during the last 5 years?
For Diagnosis based on cardiac stress test/myocardial scintigraphy/prior verified acute myocardial infarction? Treated with betablocker? Treated with aspirin (or clopidogrel/warfarin)? Blood lipids tested during the last 5 years? Target level for cholesterol reached (LDL ≤ 2.5 mmol/L and total cholesterol ≤ 4.5 mmol/L)?
For Diagnostics performed: Echocardiography performed ever? NT-proBNP tested ever? Pulmonary X-ray performed initially? Treated with ACEI or, if intolerant, angiotensin receptor blocker (ARB)? Target dose used for ACEI/ARB? Treated with betablocker with indication for HF? Treated with NSAID regularly?
For Treated with warfarin? If no, treated with aspirin 320 mg? Which CHADS2-score does the patient reach?
For Target blood pressure 140/90 reached? If ischemic stroke: Treated with recommended anticoagulation (aspirin 160 mg, warfarin or clopidogrel)?
Any information not easily accessible in the medical records was regarded as missing. Only a very small number of patients (<10) were excluded due to insufficient quality of data. Patients with hypertension were divided into one group with patients aged 65–84 and one group aged 85 or more. This was due to the recent research finding that the oldest elderly perhaps do not gain from a blood pressure treated to the same low target as younger adults [
If current recommendations were not followed, feed-back was given to the patient’s doctor with suggestions for possible improvements. Local education was organized in 2007 for all clinicians at the health center, focusing on cases and the regional guidelines. The education was repeated and followed up after three months. Repeated measurements were performed for the patients who visited the health care center between April and December 2008.
Power calculations for changes in proportions of adherence to guidelines were performed before the study started. For each diagnosis one variable was considered as the main variable. These were for hypertension the proportion who reached target blood pressure, for IHD treatment with beta blocker, for heart failure treatment with ACEI/ARB, for CAF treatment with warfarin, and for stroke reaching target blood pressure. The proportion of the main variable was analyzed in a smaller sample. The power calculation was performed on clinically reasonable improvements in the variables. For example, the proportion who reached target blood pressure in the sample was 35% and a clinically relevant proportion in the hypertension population should be 60%. The observation period was set up in order to get sufficient number of patients. It was enough to investigate a random sample of every fifth patient with hypertension aged 65–84. For IHD, HF, and CAF adequate numbers of patients were available during the observation period. Enough stroke patients were not found at the center to expect significant changes. Observed proportions of adherence and their exact confidence limits were calculated. The proportions from the studied periods were compared using two-sample test for equality of proportions. Computer software R version 2.6.0 was used for all statistical analyses (R Foundation for Statistical Computing, Vienna, Austria).
The study did not require an ethical approval according to a protocol by the ethical committee of Lund University no. 228/2007. The study was performed in accordance with Swedish law and the declaration of Helsinki.
Characteristics of the patients with hypertension are presented in Table
Patients with hypertension who reach target blood pressure and are investigated with ECG and blood lipids.
Percentage (95% CI) | Differencea (95% CI) | ||||
---|---|---|---|---|---|
Patients 65–84 years | |||||
Total number of patients | 2006 | 39 | |||
2008 | 54 | ||||
Target blood pressure level | 2006 | 11 | 28.2 (15.0, 44.9) | 7.0 (−14.2, 28.2) | 0.63 |
2008 | 19 | 35.2 (22.7, 49.4) | |||
ECG taken last two years | 2006 | 17 | 43.6 (27.8, 60.4) | −4.7 (−27.2, 17.8) | 0.81 |
2008 | 21 | 38.9 (25.9, 53.1) | |||
Lipids assessed | 2006 | 29 | 74.4 (57.9, 87.0) | 1.6 (−17.8, 21.0) | 1 |
2008 | 41 | 75.9 (62.4, 86.5) | |||
Patients ≥85 years | |||||
Total number of patients | 2006 | 24 | |||
2008 | 41 | ||||
Target blood pressure level | 2006 | 9 | 37.5 (18.8, 59.4) | −5.8 (−33.1, 21.6) | 0.31 |
2008 | 13 | 31.7 (18.1, 48.1) | |||
ECG taken last two years | 2006 | 14 | 58.3 (36.6, 77.9) | −7.1 (−35.4, 21.2) | 0.77 |
2008 | 21 | 51.2 (35.1, 67.1) | |||
Lipids assessed | 2006 | 3 | 12.5 (2.7, 32.4) | 26.5 (3.3, 49.8) | 0.047 |
2008 | 16 | 39.0 (24.2, 55.5) |
aDifference between 2008 and 2006.
The measurements of 2006 and 2008 of the patients with ischemic heart disease are presented in Table
Patients with ischemic heart disease who are pharmacologically treated and investigated according to the guidelines.
Percentage (95% CI) | Differencea (95% CI) | ||||
---|---|---|---|---|---|
Number of patients | 2006 | 113 | |||
2008 | 105 | ||||
Investigation performed | 2006 | 105 | 92.9 (86.5, 96.9) | 0.54 (−8.0, 6.9) | 1 |
2008 | 97 | 92.4 (85.5, 96.7) | |||
Treated with beta blocker | 2006 | 70 | 61.9 (52.3, 70.9) | 10.5 (−2.9,23.7) | 0.14 |
2008 | 76 | 72.4 (62.8, 80.7) | |||
Treated with aspirin | 2006 | 100 | 88.5 (81.1, 93.7) | −1.8 (−11.5, 7.9) | 0.84 |
2008 | 91 | 86.7 (78.6, 92.5) | |||
Treated with aspirin/warfarin | 2006 | 107 | 94.7 (88.8, 98.0) | −1.4 (−8.6, 5.9) | 0.89 |
2008 | 98 | 93.3 (86.7, 97.3) | |||
Lipids assessed | 2006 | 84 | 74.3 (65.3, 82.1) | 6.6 (−5.3, 18.5) | 0.31 |
2008 | 85 | 81.0 (72.1, 88.0) | |||
Target lipid level | 2006 | 32/84 | 38.1 (27.7, 49.3) | 9.0 (−7.1, 25.0) | 0.31 |
2008 | 40/85 | 47.1 (36.1, 58.2) |
a Difference between 2008 and 2006.
For patients with HF, treatment with beta blocker increased. Characteristics of HF patients are shown in Table
Patients with chronic heart failure who are pharmacologically treated and investigated according to the guidelines.
Percentage (95% CI) | Differencea (95% CI) | ||||
---|---|---|---|---|---|
Number of patients | 2006 | 75 | |||
2008 | 94 | ||||
Investigation with ECO | 2006 | 42 | 56.0 (44.1, 67.5) | 10.0 (−6.0, 25.9) | 0.24 |
2008 | 62 | 66.0 (55.5, 75.4) | |||
Investigation with plasma NT-proBNP | 2006 | 32 | 42.7 (31.3, 54.6) | 46.7 (32.7, 60.7) | <10−4 |
2008 | 84 | 89.4 (81.3, 94.8) | |||
Treated with ACEI or ARB | 2006 | 45 | 60.0 (48.0, 71.1) | 11.3 (−4.30, 26.9) | 0.17 |
2008 | 67 | 71.3 (61.0, 80.1) | |||
ACEI recommended dose | 2006 | 16/45 | 35.6 (21.9, 51.2) | −1.3 (−20.4, 18.0) | 1 |
2008 | 23/67 | 34.3 (23.2, 46.9) | |||
Treated with beta-blocker | 2006 | 22 | 29.3 (19.4, 41.0) | 19.6 (4.0, 35.2) | 0.015 |
2008 | 46 | 48.9 (38.5, 59.5) |
aDifference between 2008 and 2006.
The measurements in 2006 and in 2008 of the patients with chronic atrial fibrillation or prior stroke are presented in Table
Patients with chronic atrial fibrillation or prior stroke who are treated according to the guidelines.
Percentage (95% CI) | Differencea (95% CI) | ||||
---|---|---|---|---|---|
Chronic atrial fibrillation | |||||
Treated withwarfarin | 2006 | 34/77 | 44.2 (32.8, 55.9) | 6.5 (−10.8, 23.8) | 0.52 |
2008 | 37/73 | 50.7 (38.7, 62.6) | |||
Treated with aspirin 320 mg if not warfarin | 2006 | 5/43 | 11.6 (3.9, 25.1) | 27.3 (6.1, 48.4) | 0.010 |
2008 | 14/36 | 38.9 (23.1, 56.5) | |||
Stroke | |||||
Target blood pressure level | 2006 | 31/50b | 62.0 (47.2,75.3) | ||
2008 | 40/59b | 67.8 (54.4,79.4) | |||
Treated with recommended anticoagulation | 2006 | 35/48b | 72.9 (58.2,84.7) | ||
2008 | 45/57b | 78.9 (66.1,88.6) |
aDifference between 2008 and 2006.
bTwo patients had a bleeding stroke.
CHADS2 score of patients with CAF, related to anticoagulation. The table shows the number of patients who are treated with Warfarin, Aspirin, or no recommended anticoagulation, respectively.
CHADS2 score 2006 | N | Warfarin | Aspirin 320 mg | No anticoagulation |
---|---|---|---|---|
0 | 5 | 2 | 0 | 3 |
1 | 20 | 13 | 0 | 7 |
2 | 22 | 11 | 2 | 9 |
3 | 12 | 1 | 2 | 9 |
4 | 13 | 4 | 1 | 8 |
5 | 5 | 3 | 0 | 2 |
6 | 0 | 0 | 0 | 0 |
Total number | 77 | 34 | 5 | 38 |
CHADS2 score 2008 | ||||
0 | 5 | 4 | 0 | 1 |
1 | 17 | 11 | 0 | 6 |
2 | 25 | 9 | 8 | 8 |
3 | 18 | 7 | 6 | 5 |
4 | 6 | 5 | 0 | 1 |
5 | 2 | 1 | 0 | 1 |
6 | 0 | 0 | 0 | 0 |
Total number | 73 | 37 | 14 | 22 |
As previously described in other studies in primary care, the adherence to guidelines for the treatment of cardiovascular disease was low in this study. In addition, the study revealed that educational efforts may have a positive impact on the adherence and therefore should be encouraged.
The study was performed as a quality improvement project on a family health care station. We applied the regional guidelines of 2006. These are built on evidence-based data and widely known guidelines of international societies.
In 2006, the recommendation for the treatment of hypertension was a target BP of 130/80 for patients with diabetes and 140/90 for others. For patients aged between 65 and 80, we found a trend of more patients reaching target level after the education intervention (28% in 2006 versus 35% in 2008) but this was not significant (
The Hypertension in the Very Elderly Trial (HYVET) revealed that using antihypertensive drug therapy decreases the risk of new cardiovascular events as well as total mortality in the elderly (over 80) [
Treating blood pressure to target is important in secondary prevention of stroke. The HYVET study showed a 30% reduction in strokes in the treatment group [
In the patient group over 80 with hypertension, we found a significant improvement from 2006 to 2008 in testing of the patient’s lipid level. This is not the most important part of the intervention but it reveals that the educational program has an effect.
The positive trend of increased lipid level testing, as well as treatment to lipid target level, was also seen among the patients with IHD. For these patients, we also found a positive trend for the treatment with beta-blocker, from 62% in 2006 to 72% in 2008 (not significant). This is more than was seen in Finland 2003, where 51% of patients with IHD aged 75 and above were treated with beta-blocker [
Remme et al. found that despite the widespread availability of evidence-based guidelines, there are differences between physicians and countries in the management of HF [
Only 43% of European GPs would always, and 51% often, prescribe an ACEI when treating an HF patient. Correspondingly, 59% of the Swedish GPs often, and 34% always, prescribe ACEI. This is similar to the results from our study where 60% of the HF patients were treated with ACEI in 2006. The proportion was 71% in 2008. The increase is not significant (
The use of beta blockade for HF treatment has increased significantly from 29% to 49% (
Almost no patient with HF was prescribed NSAID regularly, which also indicates an awareness of the recommendations.
Chronic atrial fibrillation is common among the elderly in primary health care, and about half of these patients are treated with warfarin, according to a Swedish study from 2004 [
There is still much to improve regarding compliance to guidelines. More local education, as in this project, could be one way. GPs attitudes towards guidelines are yet another factor to deal with. A Swedish study about attitudes revealed that the degree of reliance on research data varied among GPs. Some were convinced of an actual and predictable risk for the individual; others strongly doubted it. Some were relying firmly on protection from disease by pharmaceutical treatment; others were strongly questioning its effectiveness in individual cases [
Hence, since we can show an effect of education programs like the present one, we should encourage a more widespread use of them. Still we cannot be definitely sure of the effect due to the intervention, since some of the positive results could be explained by an increasing awareness in the medical community about investigations or some of the treatments.
More research is needed about optimal treatment, doses, and targets for these elderly, in order to make the guidelines more specific for this population. This would presumably also increase GPs’ adherence.
This study has some limitations. Firstly we did not have a control group that was not exposed to the intervention. This means that we cannot eliminate other influences on the results, such as better availability of laboratory and other tests or wishes of the GPs to conform to local quality indicators. However, the project is not first and foremost an intervention study, but a description of how it is possible to work with quality improvement in the clinics and on small units. Second, we identified the patients through diagnoses in the medical records. There might be patients who were not captured, but there is no reason to believe that they differ from the investigated patients.
By showing that adherence to guidelines about the treatment of cardiovascular disease is low in primary care, we confirm the results of previous studies. For example, only approximately one-third of the patients with hypertension reach target blood pressure. One half of the patients with CAF are treated with Warfarin, although more than two-thirds have a CHADS2 score of 2 or more. Educational efforts appear to increase the adherence and therefore should be encouraged. For example, the use of beta blockade to heart failure patients and the assessment of NT-proBNP in these patients increased significantly after the intervention.
The work was supported by the Governmental Funding of Clinical Research within the NHS (ALF) and the National Board of Health and Welfare.