Fifteen years with clinical guidelines in the treatment of hypertension--still discrepancies between intentions and practice

Sex Differences in the Assessment of Cardiovascular Risk in Primary Health Care: A Systematic Review

OBJECTIVE

To determine whether sex differences exist in the assessment of cardiovascular disease (CVD) risk scores/risk factors in primary health care.

DESIGN/METHODS

PubMed and EMBASE were systematically searched on 31 January 2017. Clinical trials and observational studies were included if they reported on the assessment of CVD risk score, blood pressure (BP), cholesterol or smoking status in primary health care, stratified by sex. Meta-analyses were performed, using random effects models, to determine differences between sexes, separately for adjusted and unadjusted data.

RESULTS

Of 14,928 studies found in the search, 22 studies (including 4,754,782 patients) were included in the systematic review with the meta-analysis for quantitative assessment. Overall, the assessment rates of CVD risk score and risk factors were similar in women and men (CVD risk score: 30.7% vs. 35.2% [difference (95% CI): -4.5 (-5.1, -3.9)]; BP: 91.3% vs. 88.5% [2.8 (2.5, 3.0)]; cholesterol: 69.9% vs. 71.0% [-1.1 (-1.5, -0.8)]; and smoking: 85.9% vs. 86.7% [-0.8 (-1.1, -0.5)]). The pooled, adjusted likelihood of having the risk score, BP and cholesterol assessments were comparable between women and men: OR (95% CI): 0.87 (0.70, 1.07); 1.41 (0.89, 2.25); and 1.15 (0.82, 1.60), respectively. However, women were 32% less likely to be assessed for smoking (0.68 [0.47, 1.00]). There was substantial heterogeneity between studies and the risk of publication bias was moderate.

CONCLUSION

Despite the guideline recommendations, assessment of CVD risk score in primary health care was low in both sexes. Further, women were less likely to be assessed for their smoking status than men, whereas no sex discrepancies were found for BP and cholesterol assessments.

Long-term trends in the prevalence of chronic kidney disease and the influence of cardiovascular risk factors in Norway

Surveillance of chronic kidney disease (CKD) prevalence over time and information on how changing risk factors influence this trend are needed to evaluate the effects of general practice and public health interventions. Because very few studies addressed this, we studied the total adult population of a demographically stable county representative of Norway using cross-sectional studies 10 years apart (Nord-Trøndelag Health Study (HUNT)2 and Nord-Trøndelag Health Study (HUNT)3, 65,237 and 50,586 participants, respectively). Thorough quality-control procedures and comparisons of methods over time excluded analytical drift, and multiple imputations of missing data combined with nonattendance weights contributed to unbiased estimates. CKD prevalence remained stable in Norway from 1995 through 1997 (11.3%) to 2006 through 2008 (11.1%). The association of survey period with CKD prevalence was modified by a strong decrease in blood pressure, more physical activity, and lower cholesterol levels. Without these improvements, a 2.8, 0.7, and 0.6 percentage points higher CKD prevalence could have been expected, respectively. In contrast, the prevalence of diabetes and obesity increased moderately, but the proportion of diabetic patients with CKD decreased significantly (from 33.4% to 28.6%). A CKD prevalence of 1 percentage point lower would have been expected without these changes. Thus, CKD prevalence remained stable in Norway for more than a decade in association with marked improvements in blood pressure, lipid levels, and physical activity and despite modest increases in diabetes and obesity.

Is there an association between disease ignorance and self-rated health? The HUNT Study, a cross-sectional survey

OBJECTIVE

To explore whether awareness versus unawareness of thyroid dysfunction, diabetes mellitus or hypertension is associated with self-rated health.

DESIGN

Large-scale, cross-sectional population-based study. The association between thyroid function, diabetes mellitus and blood pressure and self-rated health was explored by multiple logistic regression analysis.

SETTING

The second survey of the Nord-Trøndelag Health Study, HUNT2, 1995-1997.

PARTICIPANTS

33 734 persons aged 40-70 years.

PRIMARY OUTCOME MEASURES

Logistic regression was used to estimate ORs for good self-rated health as a function of thyroid status, diabetes mellitus status and blood pressure status.

RESULTS

Persons aware of their hypothyroidism, diabetes mellitus or hypertension reported poorer self-rated health than individuals without such conditions. Women with unknown and subclinical hypothyroidism reported better self-rated health than women with normal thyroid status. In women and men, unknown and probable diabetes as well as unknown mild/moderate hypertension was not associated with poorer health. Furthermore, persons with unknown severe hypertension reported better health than normotensive persons.

CONCLUSIONS

People with undiagnosed but prevalent hypothyroidism, diabetes mellitus and hypertension often have good self-rated health, while when aware of their diagnoses, they report reduced self-rated health. Use of screening, more sensitive tests and widened diagnostic criteria might have a negative effect on perceived health in the population.

[Patterns in the prescription of antihypertensive drugs in Norway, 1975-2010]

BACKGROUND

A complete overview of drugs used for hypertension is not available from official data sources. The aim of this study was to investigate the use of blood pressure medication over the years to identify trends in consumption patterns.

MATERIAL AND METHOD

Data were collected from the Norwegian drug wholesaler statistics for the years 1975-2010 and from the Norwegian prescription database 2004-2010. Defined daily doses (DDD) per 1000 inhabitants per day, number of users and prevalence by gender and age were analysed.

RESULTS

In the period 1975 to 2010 the use of antihypertensives increased from 55 to 248 DDD/1000 inhabitants/day (Norwegian drug wholesaler statistics). Data from the Norwegian prescription database show that in 2010 there were a total of 754 909 users of antihypertensives, of whom 638 830 had received the diagnosis hypertension. Thiazides and angiotensin II antagonists had the greatest increase in number of users from 2004 to 2010. In all ages up to 73 years, a higher proportion of men were prescribed drugs for hypertension in 2010, whereas in the over-73 s these drugs were most widely used by women.

INTERPRETATION

The use of antihypertensives is steadily increasing. Angiotensin II antagonists and thiazides are the most commonly used drugs. The authorities'' decision in 2004 that thiazides should be regarded as first-line therapy for hypertension probably contributed to increased use of thiazides.

The role of Decision Support System (DSS) in prevention of cardiovascular disease: a systematic review and meta-analysis

Background

The potential role of DSS in CVD prevention remains unclear as only a few studies report on patient outcomes for cardiovascular disease.

Methods and Results

A systematic review and meta-analysis of randomised controlled trials and observational studies was done using Medline, Embase, Cochrane Library, PubMed, Amed, CINAHL, Web of Science, Scopus databases; reference lists of relevant studies to 30 July 2011; and email contact with experts. The primary outcome was prevention of cardiovascular disorders (myocardial infarction, stroke, coronary heart disease, peripheral vascular disorders and heart failure) and management of hypertension owing to decision support systems, clinical decision supports systems, computerized decision support systems, clinical decision making tools and medical decision making (interventions). From 4116 references ten studies met our inclusion criteria (including 16,312 participants). Five papers reported outcomes on blood pressure management, one paper on heart failure, two papers each on stroke, and coronary heart disease. The pooled estimate for CDSS versus control group differences in SBP (mm of Hg) was - 0.99 (95% CI −3.02 to 1.04 mm of Hg; I² = 0; p = 0.851).

Conclusions

DSS show an insignificant benefit in the management and control of hypertension (insignificant reduction of SBP). The paucity of well-designed studies on patient related outcomes is a major hindrance that restricts interpretation for evaluating the role of DSS in secondary prevention. Future studies on DSS should (1) evaluate both physician performance and patient outcome measures (2) integrate into the routine clinical workflow with a provision for decision support at the point of care.

Evaluation of the agreement between guidelines and initial antihypertensive drug treatment using a national health care reimbursement database

RATIONALE, AIMS AND OBJECTIVES

To test the agreement between guidelines for the management of hypertension and medical practices while avoiding frequent limitations such as the use of non-representative samples of practitioners and self-reporting of their practices over a short period of time.

METHODS

The characteristics of initial antihypertensive drug treatment in a large representative sample of the French population aged 50-80 (n = 17 855) were collected from a national health care reimbursement database and compared with national guidelines over a 5-year period.

RESULTS

Major discrepancies are observed including the use of non-recommended drug classes such as loop and potassium sparing diuretics alone or in association and the absence of distinction between patients according to their age. More minor discrepancies are the preferential use of mono-therapies over drug combinations and of some bi-therapies among those recommended. Some degree of concordance with the guidelines is also observed including the specific characteristics of the treatment of diabetics compared with other categories of patients and the preferential use of long-acting dihydropyridine calcium antagonists and of low-dose thiazide diuretics when these drug classes are chosen. Several of these discrepancies or concordances, which mainly reflect general practitioner (GP) activity, show time trends over the entire follow-up period with no significant effect of the guideline released during this period.

CONCLUSIONS

At the French national level, the agreement between initial antihypertensive drug treatment and guidelines varies considerably depending on the characteristics of the treatment that are considered. The GPs who delivered the treatment do not seem to have been influenced by the guidelines released over the last decade.

Use and usefulness of guidelines for sickness certification: results from a national survey of all general practitioners in Sweden

Objectives Diagnoses-specific sickness certification guidelines were recently introduced in Sweden. The aim of this study was to investigate to which extent general practitioners (GPs) used these guidelines and how useful they found them, 1 year after introduction. Design A cross-sectional questionnaire study. A comprehensive questionnaire about sickness certification practices in 2008 was sent to all physicians living and working in Sweden (n=36 898, response rate 60.6%). In all, 19.7% (n=4394) of the responders worked as GPs. Setting Primary healthcare in all Sweden. Participants The participating GPs who had consultations concerning sickness certification at least a few times a year (n=4278, 97%). Main outcome measures Descriptive statistics and prevalence ratios for the 11 questionnaire items about the use and usefulness of the sickness certification guidelines. Results A majority (76.2%) of the GPs reported that they used the guidelines. In addition, 65.4% and 43.5% of those GPs reported that the guidelines had facilitated their contacts with patients and social insurance officers, respectively. The guidelines also helped nearly one-third (31.5%) of the GPs to develop their competence and improve the quality of their management of sickness certification consultations (33.5%). About half experienced some problems when using the guidelines and 43.7% wanted better competence in using them. A larger proportion of non-specialists and of GPs with fewer sickness certification consultations had benefitted from the guidelines. Conclusions The national sickness certification guidelines implemented in Sweden were widely used by GPs already a year after introduction. Also, the GPs consider the guidelines useful in several respects, for example, in patient contacts and for competence development.

GPs' thoughts on prescribing medication and evidence-based knowledge: the benefit aspect is a strong motivator. A descriptive focus group study

OBJECTIVE

To describe GPs' thoughts of prescribing medication and evidence-based knowledge (EBM) concerning drug therapy.

DESIGN

Tape-recorded focus-group interviews transcribed verbatim and analysed using qualitative methods.

SETTING

GPs from the south-eastern part of Västra Götaland, Sweden.

SUBJECTS

A total of 16 GPs out of 178 from the south-eastern part of the region strategically chosen to represent urban and rural, male and female, long and short GP experience.

METHODS

Transcripts were analysed using a descriptive qualitative method.

RESULTS

The categories were: benefits, time and space, and expert knowledge. The benefit was a merge of positive elements, all aspects of the GPs' tasks. Time and space were limitations for GPs' tasks. EBM as a constituent of expert knowledge should be more customer adjusted to be able to be used in practice. Benefit was the most important category, existing in every decision-making situation for the GP. The core category was prompt and pragmatic benefit, which was the utmost benefit.

CONCLUSION

GPs' thoughts on evidence-based medicine and prescribing medication were highly related to reflecting on benefit and results. The interviews indicated that prompt and pragmatic benefit is important for comprehending their thoughts.

[The quality of first line management of arterial hypertension in the sanitary region of Sousse]

The aim of the study is to value determinants of the quality of management of hypertension in structures of primary health care, a medical audit has been achieved on a representative sample of 456 hypertensive patients followed in the sanitary region of Sousse during the year 2002. It takes out again this work that the global quality of management of hypertension in primary health care have been considered satisfactory at only 28,7% of the hypertensive patients. It was statistically differential according to surroundings (farming: 40,5%, urban: 24,9%) and categories of the seniority of follow-up in primary health care (< or = five years: 34,6%, > five years: 23,9%). A survey multi varied by logistical regression controlling the other factors of confusion (kind, seniority of the illness, geographical and financial accessibility) kept these two factors: the farming middle (ORa: 1,97; P = 0,003) and the lower seniority to five years (ORa: 1,64; P = 0,023). So, the hypertensive patients followed in the urban health centres since more that five years should constitute the population targets a program of improvement of the quality of health care dispensed to hypertensive patients in extra hospital structures of health.

Posting Guidelines: A Practical and Effective Way to Promote Appropriate Hypertension Treatment

BACKGROUND

Despite publication and periodic updating of treatment guidelines, hypertension remains undertreated in the United States, and physicians underuse recommended drugs.

METHODS

Hypertension treatment guidelines were summarized and posted in five places in a hospital-based primary care clinic staffed by internists and internal medicine residents. Costs and recommended doses of five commonly used antihypertensive drugs were included. The charts of all 253 patients seen during a four-month period with a diagnosis of hypertension were analyzed. Blood pressures and physician prescribing habits were compared at baseline and at 8, 12, and 16 months after posting the guidelines.

RESULTS

The number of patients with blood pressures < 140/90 mm Hg increased from 41% to 58%, p = .001. Median (IQR) systolic pressure fell from 143 (119-167) to 137 (116-158) mm Hg, p < .0001 and diastolic pressure from 78 (65-91) to 77 (64-90) mm Hg, p = .0002. Physicians prescribed more recommended drugs, more total antihypertensive drugs, larger doses of hydrochlorothiazide and lisinopril, and more inexpensive drugs. The total cost of antihypertensive drugs per patient increased slightly.

CONCLUSION

Regular exposure to clinical guidelines, presented in a practical and simple way, can change physician behavior and improve patient care.

Metabolic disorders associated with uncontrolled hypertension

AIM

To examine the prevalence and characteristics of uncontrolled hypertension (HT).

METHODS

A cross-sectional community-based study (1992-93) was carried out in Skara, Sweden, including 894 patients who consecutively underwent an annual follow-up at the hypertension outpatient clinic in primary care. Controlled HT was defined as diastolic blood pressure (DBP) < or =90 mmHg and systolic blood pressure (SBP) < or =160 mmHg and was used as reference. Uncontrolled DBP was defined as DBP >90 mmHg regardless of SBP level, and isolated uncontrolled SBP was defined as SBP >160 mmHg and DBP < or =90 mmHg. Proportions were age-standardized using the Skara population as reference.

RESULTS

The prevalence of uncontrolled HT was 43% (isolated uncontrolled SBP 18% and uncontrolled DBP 25%). Both men and women with isolated uncontrolled SBP were older (73 years, CI: 70-75; and 73 years; CI: 72-75) than patients with controlled HT (64 years, CI: 63-66; and 65 years, CI: 64-66). Men and women with known cardiovascular disease (CVD) less often had isolated uncontrolled SBP (OR: 0.4, CI: 0.2-0.9; and OR: 0.5, CI: 0.3-0.9), whereas men and women with known diabetes more often had uncontrolled DBP (OR: 2.3, CI: 1.3-4.1; and OR: 3.3, CI: 1.9-5.7). Men with known CVD less often had uncontrolled DBP (OR: 0.5, CI: 0.3-1.0, p = 0.04), and men with fasting blood glucose >5.5 mmol/l more often had isolated uncontrolled SBP (OR: 1.9, CI: 1.0-3.5, p = 0.04). In women, the following high risk factor levels were associated with uncontrolled DBP: fasting blood glucose >5.5 mmol/l (OR: 1.4, CI: 1.1-1.8), fasting triglycerides > or =1.7 mmol/l (OR: 1.4, CI: 1.1-1.8), body mass index (BMI) >30 kg/m2 (OR: 1.5, CI: 1.1-1.9), waist/hip ratio (WHR) >0.85 cm/cm (OR: 1.7, CI: 1.3-2.2), insulin resistance (homeostasis model assessment (HOMA) >third quartile) (OR: 1.4, CI: 1.1-1.9) and microalbuminuria (OR: 3.2, CI: 1.7-6.2).

CONCLUSION

Uncontrolled DBP is in both sexes related to type 2 diabetes, whereas isolated uncontrolled SBP is related to older age. In women, uncontrolled DBP, furthermore, is related to several other CVD risk factors of the metabolic syndrome. Patients with uncontrolled DBP should be carefully evaluated for metabolic disorders.

1999 WHO/ISH Guidelines applied to a 1999 MONICA sample from northern Sweden

BACKGROUND

Treating hypertension with drugs is so far the most cost-effective way to reduce this important risk factor for cardiovascular disease (CVD). It is, however, important to determine absolute risk, and thereby estimate indication for drug treatment, in order to maintain a cost-effective drug treatment. WHO/ISH Hypertension Guidelines from 1999 propose a risk stratification for estimating absolute risk for CVD based on blood pressure and additional risk factors, target organ damage (TOD) and CVD.

OBJECTIVES

We studied the consequences of applying the recent WHO/ISH risk stratification scheme to a MONICA sample of 6000 subjects from a geographically defined population in northern Sweden, regarding indications for treatment, target blood pressure and risk distribution.

METHODS

We have risk-classified each of these patients using a computer program, according to the WHO/ISH scheme. Data on TOD were not available.

RESULTS

In all, 917 (15%) had drug-treated hypertension. Three-quarters (n = 737) were inadequately treated, with blood pressure levels at or above 140 or 90 mmHg. 1773 (30% of 5997) untreated subjects had a blood pressure of 140/90 or above; 16% in the low-, 62% in the medium-, 8% in the high-, and 14% in the very-high-risk group. The corresponding risk-group pattern for the inadequately treated hypertensives (n = 737) was 5.5, 48.3, 11.1 and 35.2%, respectively. If we shifted the target blood pressure from below 140/90 to below 130/85 for drug-treated subjects under 60 (n = 278) the number of inadequately treated subjects increased by 34 (12.2% of 278); 14 in the low-risk group, 15 in the medium-risk group, and only five in the high- or very-high-risk groups.

CONCLUSIONS

Only one-fifth of the drug-treated hypertensives were well controlled. Moreover, the incidence of newly detected blood pressure elevation was high. The majority of younger subjects with high blood pressure had low risk, but in those aged 45-54 this had already risen to a medium risk. Changing the target blood pressure to below 130/85, for subjects aged below 60, as recommended by WHO/ISH, affects predominantly low- and medium-risk groups.

Hypertension guidelines and their limitations--the impact of physicians' compliance as evaluated by guideline awareness

OBJECTIVE

The initial step of an optimal therapeutic strategy for patients with arterial hypertension is the recognition and acceptance of new recommendations by the physicians themselves. This guideline awareness of the physicians has never been evaluated in detail.

DESIGN

The awareness of content of current recommendations in hypertension diagnosis, treatment and treatment control was therefore assessed in primary care physicians using a questionnaire. The guidelines of the German Hypertension Society were used as the reference standard.

PARTICIPANTS

A total of 24 899 German physicians, including all internists, all cardiologists and 22% of general practitioners were contacted in a nationwide survey.

MAIN OUTCOME MEASURES

The number of answers in agreement with the guideline was used as a measure of guideline awareness. Adequate awareness of content of guideline recommendations was defined as the correct answer to five out of eight questions; the correct answers had to include the appropriate definition of hypertension.

RESULTS

The analysis was based on 11 547 returned questionnaires (47.1%). An adequate guideline awareness was found in 23.7% of the total study population, especially in 37.1% of cardiologists, in 25.6% of internists and in 18.8% of general practitioners. While the guideline awareness was significantly influenced by the duration of private practice, regional and municipal factors had only minor influence on the results.

CONCLUSION

The impact of hypertension guidelines on actual medical knowledge is modest. Thus, the information strategies about current treatment guidelines must be improved and tailored to the needs of physicians in clinical practice to ultimately improve patient care.

Assessment of coronary risk in general practice in relation to the use of guidelines: a survey in Denmark

BACKGROUND

There is a considerable gap between recommended and actually conducted preventive cardiology in general practice. The effect of guidelines is not fully evaluated.

METHODS

A questionnaire containing 10 questions on preventive cardiology, including the use of clinical guidelines, together with four case stories for cardiovascular risk estimation was mailed to 205 general practitioners (GPs).

RESULTS

Response rate was 81%. Twenty-five percent of the GPs had consultations in preventive cardiology at least once a day and 60% of the GPs thought lifestyle intervention had significant effect on cardiovascular risk. Approximately two-thirds of the GPs were regular users of national guidelines on prevention of cardiovascular disease. While the majority of GPs correctly assigned a patient with multiple risk factors to the high-risk category there was a much larger variation in risk estimations if fewer risk factors were present. GPs who reported use of guidelines overestimated coronary risk twice as frequently as nonusers of guidelines.

CONCLUSION

Preventive cardiology in general practice is common and the effect of lifestyle intervention is well accepted. Poor discrimination between high- and low-risk patients may, however, lead to suboptimal preventive care. The use of guidelines does not seem to improve risk estimation and further dissemination of better tools for risk estimation is needed.

Cardiovascular disease and risk factors in persons with type 2 diabetes diagnosed in a large population screening: the Nord-Trøndelag Diabetes Study, Norway

OBJECTIVE

To study cardiovascular status and risk factors in persons with newly diagnosed type 2 diabetes and controls in a large population.

DESIGN

Case-control study.

SETTING

Population screening.

SUBJECTS

The screening of 74 499 individuals (88.1%), aged 20 years and older, in Nord-Trøndelag County, Norway, during 1984-86 detected 428 persons with undiagnosed diabetes according to the 1980 WHO criteria, of whom 205 attended a clinical follow-up examination assessing cardiovascular status and risk factors.

METHODS

For each of 205 cases, one control person matched by age and sex underwent the same clinical examination. Lipids, body mass index, waist/hip ratio, blood pressure, pulse rate, blood pressure medication, kidney function, cardiovascular disease, family history and lifestyle were recorded.

RESULTS

At the screening prior to the diagnosis of diabetes, those with diabetes reported poorer general health, less physical activity, more siblings with diabetes and more frequent use of antihypertensive medication. They had higher body mass index, systolic and diastolic blood pressure and pulse rate compared with controls. At the clinical evaluation, diabetics had higher urine albumin levels, increased waist/hip ratio, and higher total cholesterol/HDL cholesterol ratios than the controls. They also reported a greater incidence of angina pectoris and had more ECG changes.

CONCLUSIONS

Diabetics presented with more cardiovascular risk factors, angina pectoris and ECG changes than the controls, and they had an established metabolic syndrome more often than controls. These results suggest that prevention of cardiovascular disease in diabetics requires earlier diagnosis of the diabetes.

Cardiovascular risk factors and testing of relatives amongst patients with familial hyperlipidaemia one decade after a clinical trial

OBJECTIVES

We examined the cardiovascular disease risk factor status of men and women with familial hyperlipidaemia (FH) 10-11 years after a clinical trial and asked whether first-degree relatives had undergone lipid testing.

DESIGN SETTING AND SUBJECTS

Subjects started lipid-lowering drugs in 1987-88. Of 60 subjects, 12 had died, one emigrated and 35 men and 12 women took part in a follow-up clinical examination in 1998.

RESULTS

Total cholesterol level was reduced by 41% and high-density lipoprotein (HDL) cholesterol level increased by 13% compared with baseline (diet alone). Low-density lipoprotein (LDL) cholesterol level was lower at the end of the trial than at follow-up (3.6 +/- 1.5 vs. 4.6 +/- 2.2 mmol L-1; P = 0.01) and was higher in the group taking a low dose of a statin alone compared with other drug groups. Thus, two-thirds of the subjects required adjustment of lipid-lowering drugs to reach target lipid levels. One-fifth consumed at least two food groups rich in saturated fat regularly. Body mass index (BMI) increased from 25.6 +/- 2.9 to 26.8 +/- 3.3 kg m-2 (P < 0.001). Five subjects compared with one at baseline had type II diabetes or glucose intolerance; 12 compared with four at baseline had a blood pressure of >/= 160 mmHg systolic or >/= 95 mmHg diastolic. Plasma total homocysteine was higher in subjects with coronary artery disease than in subjects without disease (11.7 +/- 3.9 vs. 9.0 +/- 2.3 micromol L-1; P = 0.01). Barriers to testing for lipids amongst children or siblings included family feuds, fear of increased insurance and psychiatric disease.

CONCLUSION

The majority of subjects were undertreated. Increases in BMI, blood pressure and glucose levels and the diet posed challenges to risk reduction. Plasma homocysteine levels should be considered in this group. Testing of all first-degree relatives may not be achievable because of psychological barriers.

Development and maintenance of guideline-based decision support for pharmacological treatment of hypertension

The objective was to build a computer-based decision support system (DSS), which could apply the formal rules embedded in guidelines regarding pharmacological treatment of hypertension. The aim was also to test VISUAL BASIC as a development tool for DSS's in health care. From the Swedish guidelines for treatment of hypertension, the most widely accepted and scientifically best proved treatment strategies were chosen and implemented as rules. A DSS that is capable of applying the evidence-based rules extracted from guidelines regarding drug treatment of hypertension, to any patient's medical profile, was constructed. The output consists of a recommendation regarding preferred generic drug class and also a written report, reflecting decision steps provided by the rule-base and inference engine. We also provide methods for formalising an implementable language of guidelines. A mainstream programming language like VISUAL BASIC can be an alternative when building complicated decision support systems. A logic formal notation can facilitate communication between the expert and the programmer. The program is a stand-alone product independent of computerized medical records and thereby easy to install and maintain.

Evaluation of a computer-based decision support system for treatment of hypertension with drugs: retrospective, nonintervention testing of cost and guideline adherence

OBJECTIVE

To evaluate a computerized decision support system (DSS) for drug treatment of hypertension, regarding quality, safety, and cost compared to actual antihypertensive drug treatment.

DESIGN

The medical profiles of 338 hypertensive patients treated with drugs against hypertension were processed by the DSS. The drug treatment proposed by the system was then compared to actual treatment given by their physician.

SETTING

Four health centres in the county of Västerbotten, in Sweden.

SUBJECTS

A list of hypertensive patients was extracted from the computerized medical records of each health centre and every fifth patient's medical profile was assessed by the system.

INTERVENTIONS

None.

MAIN OUTCOME MEASURES

Drug used, drug used in relation to certain major diseases such as diabetes mellitus, asthma, ischaemic heart disease (IHD), and previous myocardial infarction. Adherence to hypertension guidelines, safety, and cost.

RESULTS

The DSS suggested significantly more thiazides and significantly fewer calcium antagonists than the physicians had prescribed, with a total cost reduction of 33-40%, depending on doses chosen. The DSS drug profile was more adherent to guidelines in patients with major complicating diseases, suggesting an improvement in treatment quality for these patients by the DSS.

CONCLUSION

The DSS which fully implements current guidelines may improve the quality of antihypertensive treatment, concurrently leading to a considerable reduction in drug costs.

Treatment of diabetes mellitus--physicians' adherence to clinical guidelines in Norway

OBJECTIVE

To assess general practitioners' current adherence to the Norwegian clinical guidelines for the treatment of diabetes mellitus.

DESIGN

Descriptive, retrospective registration of information from patient records of one year; 7 November 1993 to 7 November 1994. In addition, mailed questionnaire to examine the representativeness of the participating doctors.

SETTING

General practice in Sør- and Nord-Trøndelag counties in Norway, 380,000 inhabitants.

PARTICIPANTS

In one year 1119 patients were registered with the diagnosis of diabetes mellitus in the records of 56 general practitioners. The patients were 53% women and 47% men; 51% were 70 years or older.

MAIN OUTCOME MEASURES

Levels of HbA1c and blood pressure in accordance with the recommendations of the Norwegian clinical guidelines for diabetes mellitus. Fractions of patients with a measured HbA1c, blood pressure, and serum cholesterol in one year.

RESULTS

At least one HbA1c was recorded in 77% of the diabetic patients during the specified year. In patients under 70 years of age, 56% had HbA1c above the recommended treatment level of 7.5%. In patients 70 years of age or older, 36% were above the recommended limit of 8.5%. At least one blood pressure was recorded in 79% of the patients during the specified year. In patients under 70 years of age, 53% had a systolic blood pressure above the recommendations of 140 mmHg, and 22% had a diastolic blood pressure above the recommendations of 90 mmHg. Serum cholesterol was not recorded during the specified year in 75% of the patients.

CONCLUSION

There are still major discrepancies between current practice and the intentions laid down in the Norwegian clinical guidelines. A discussion of alternative methods for implementation and evaluation of the efficacy of clinical guidelines is needed.