Blood pressure (BP) control remains unsatisfactory worldwide. Better knowledge of BP management in clinical practice is needed to develop more effective improving strategies. Using a large Italian primary care database, we selected the subjects diagnosed with hypertension, and extracted the diagnosis of myocardial infarction, angina pectoris/coronary disease, stroke/transitory ischemic attack (TIA), heart failure, atrial fibrillation, peripheral arterial disease, diabetes mellitus, the serum total cholesterol, HDL cholesterol, triglycerides, creatinine, BP, electrocardiogram, weight, height and the prescription of cardiovascular (CV) drugs. Hypertension was recorded in 119.065 individuals (prevalence 19.3%), 19.134 (16%) had no ambulatory visit and 33.183 (27.8%) had no BP value recorded. Overall, 14.594 (21.9%) had at least one recorded diagnosis showing high CV risk. BP was controlled (mean of BP values <140/90 mm Hg) in 28.918 patients (16.690 women, 12 189 men and 40 gender not recorded), that is, 43.23% of the subjects with recorded BP. Among the non-controlled patients, 21.866 (57.8%) were non-high risk grade 1 (mean BP 142.5/84.5 mm Hg; s.d. 13.1/8.2) and 7.123 (18.8%) high-risk grade 1 hypertensives (mean BP 150/83 mm Hg; s.d. 6.2/7.2). Less than three drugs were prescribed in 29.919 (79.1%) of non-controlled patients. Low attendance rate, BP under-recording and suboptimal use of politherapy are major obstacles to hypertension control. Most uncontrolled individuals are low-CV risk, grade 1 hypertensive patients, for whom the personal benefit of adding another drug is modest. Aiming at the recommended BP target in uncontrolled grade 2–3 hypertensive/high-CV risk patients would probably require two additional drugs.
It is well known that hypertension is one of the most important modifiable risk factor for atherothrombotic cardiovascular (CV) disease;1 unfortunately blood pressure (BP) control is unsatisfactory worldwide and particularly in Europe.2 The barriers for controlling hypertension are well known and include patient factors, such as non-adherence to lifestyle advice or drug treatment, and healthcare provider factors, including the organisation or environment where care is delivered. As far as the factors pertaining to physicians are concerned, suboptimal drug prescription, and particularly therapeutic inertia, is of paramount importance. According to the guidelines,3 decisions about the treatment intensity should consider the patient's risk level and the corresponding health benefit expected from the therapy. Surprisingly, the relationship between risk level/expected benefit and prescribing attitude is not considered in recent reviews on improvement of BP control.4, 5
For these reasons, the usual evaluation of clinical practice based on the percentage of ‘controlled’ or ‘non-controlled’ patients, (defined as such according to a BP threshold) appear to be of limited help. Furthermore, because of the continuous relationship between BP levels and CV risk, the absolute BP level is more relevant than simply being above or below the recommended BP target. On the patient's side, non-attendance obviously impairs all the physician's efforts to reach the optimal treatment levels for hypertension. This problem is hardly considered when the level of hypertension control is evaluated.2 Information about real practice problems in hypertension management are crucial to develop new and more effective improvement strategies. The vast majority of hypertensive patients are managed in primary care, and almost all the drug prescriptions are filled by General Practitioners (GPs) in Italy. We examined a large primary care database to analyse the BP management in Italy.
We conducted a retrospective study using information obtained from the Health Search Database, set up by the Italian College of GPs in 1998. The quality of the information gathered has been assessed in earlier studies.6, 7
In Italy, all the citizens are registered with a GP or with a Paediatrician, depending on their age. All the ‘life saving’ drugs are reimbursed by the National Health Service and, as far as hypertension is concerned, almost all the prescription are filled by GPs. The Health Search Database currently contains data from over 550 GPs, representing more than 40 000 Italian GPs, with a patient population of more than 800 000 individuals. Following the training in the optimal use of the clinical software, all participating GPs have been recording all patients’ data during their normal daily clinical practice, that is, the complete recording of all patients’ data, according to the guidelines, is not a requirement, and therefore, only the spontaneous clinical activity is recorded. For this reason, the reported prevalence is not the true prevalence of the diseases, but the prevalence of detected and recorded diagnoses. This may be an important limitation for pure epidemiological studies, but a strength when an accurate and generalizable picture of the real practice is needed. All the complete, original and anonymous clinical records are periodically sent to the database. All the diagnoses are coded using the ninth revision of the International Classification of Diseases (ICD-9).8 Prescription records are coded according to the Anatomical Therapeutic Chemical classification system.9 A unique patient code links demographic, prescription information, clinical events and diagnoses, hospital admissions and cause of death. Data are periodically subjected to quality control, exploring four realms: (a) recorded basic data (smoking habits, weight and height), (b) prescriptions associated with a clinical problem, (c) codified diagnoses and (d) prescribed test with associated recorded results.10 Any variations within the agreed ranges are investigated and submitted to each participating GP. Physicians who fail to meet the quality standard criteria required are not considered for epidemiological studies.11 At the time of the initiation of this study, 400 GPs were considered up to standard and only their patients were the subjects of this study. We selected all the living subjects on 31 December 2005 with the diagnosis of hypertension (ICD-9 codes 401, 402, 403, 404 and 437.2).
Presence of myocardial infarction (ICD-9 410 and 412), angina pectoris/coronary disease ICD-9 411.1, 411.89, 413, V45.81 and V45.82), stroke/transitory ischemic attack (TIA) (ICD-9 434.9, 435), heart failure (ICD-9 428), atrial fibrillation (ICD-9 427.3), peripheral arterial disease (ICD-9 440.2, 440.20, 440.21, 440.22, 440.23, 440.24, 440.29 and 443.9) and diabetes mellitus (ICD-9 250) were determined. We extracted the serum total cholesterol, HDL cholesterol, triglycerides, creatinine (last available data recoded in 2004 and 2005), systolic and diastolic BP (mean of the recorded values in 2005), electrocardiogram, weight and hight. We calculated the glomerular filtration rate (GFR) with the MDRD2 formula.12 Prescriptions (at least one in the last 4 months) of the CV drugs were also examined.
We used a standard descriptive analysis (means and confidence intervals), the χ2-test (to examine differences in the drug pattern prescription between controlled and non-controlled patients) and the correlation coefficient (to examine the correlation between the number of prescribed drugs and the level of CV risk).
In this study, we have only reported the results concerning BP control and drug prescription. First, we identified the patients with evaluable BP level (that is, subjects with recorded BP values in 2005) and those with controlled (<140/90 mm Hg) or non-controlled (⩾140/90 mm Hg) BP; then, we examined the prescription pattern and the mean BP values according to the risk level (High risk vs non-high risk) and the hypertension classification (grades according to the guidelines),7 both in controlled and non-controlled patients.
The 400 GPs cared for 615 373 subjects. Hypertension diagnosis was recorded in 119 065 individuals (19.3%), 66 017 women, 52 788 men and 264 gender not recorded; 19 134 (16% of all the hypertensive population) had no ambulatory visit and 33 183 (27.8%) had no BP value recorded during 2005. Among patients with recorded BP values, a diagnosis of diabetes mellitus, myocardial infarction, coronary artery disease (without myocardial infarction), stroke, renal failure and left ventricular hypertrophy were recorded, respectively in 11.114 (16.6%), 1927 (2.9%), 1048 (1.6%), 1769 (2.6%) and 2281 (3.4%) cases. Overall, 14 594 (21.9%) hypertensive patients had at least one of the above-mentioned diagnosis, and were therefore classified as high-CV risk individuals.
Blood pressure was controlled (BP values <140/90 mm Hg)7 in 28 918 patients (16 690 women, 12 189 men and 40 gender not recorded), that is, 43.23% of the subjects with recorded BP and 24.28% of the whole group of patients with recorded diagnosis of hypertension. The mean systolic and diastolic BP in the controlled group were 128.4 mm Hg (s.d. 7.7) and 78.37 mm Hg (s.d. 5.83), respectively.
We classified the patients with non-controlled BP according to the European Guidelines:3 28 989 (79%) had grade 1 hypertension (140–159/90–99 mm Hg), 7.506 (20%) grade 2 hypertension (160–179/100–109 mm Hg) and 1.473 (1%) grade 3 hypertension (⩾180/110 mm Hg). The mean systolic and diastolic BP in grades 1, 2 and 3 hypertensive subjects were, respectively, 145.2 mm Hg (s.d. 6.9 mm Hg) and 84.37 mm Hg (s.d. 6.9 mm Hg); 161.9 mm Hg (s.d. 8.9 mm Hg) and 90.69 mm Hg (s.d. 9.3 mm Hg); and 183.4 mm Hg (s.d. 12.8 mm Hg), 97.4 mm Hg (s.d. 12.6 mm Hg). It should be noticed that 26.4% of hypertensive patients are included in the grade 1 group because of either systolic BP=140 mm Hg (6.530) or diastolic BP=90 mm Hg (1.132).
The mean BP of non-high-risk patients with grade 1 hypertension was 142.5/84.5 mm Hg (s.d. 13.1/82); the mean BP of grade 1 high-risk hypertensives was 150/83 mm Hg (s.d. 6.2/7.2 mm Hg).
The different classes of antihypertensive drugs prescribed in controlled and non-controlled patients are reported in Table 1.
The number of prescribed drug is significantly higher in high- vs low-risk risk patients for each hypertension grade (Table 2).
More interestingly, there is a highly significant (R2=0.9426) linear relationship between the level of CV risk and the mean number of prescribed antihypertensive drugs (Figure 1).
Among non-controlled patients, 20.130 (53.02%) had the serum creatinine value recorded, and 4.461 (22.16% of those with recorded serum creatinine) had GFR value <60 ml−1 min−1 per 1.73 m2, indicative of renal failure (1.317–29.5% men; 3.144 −70.5% women). In these patients, diuretics were prescribed in 2.733 (61.3%): thiazidic diuretic in 2.011(73.6%) and loop diuretics in 722 (26.4%).
In the group of subjects with GFR <60 ml−1 min−1 per 1.73 m2, renal failure was a recorded diagnosis in only 510 patients (11.4%).
International guidelines on hypertension treatment are regularly issued every few years, not to mention thousands of congresses and educational meetings. Despite this huge effort, millions of patients do not reach the recommended BP targets, especially in Europe,2 and BP suboptimal control is still causing thousands of potentially avoidable CV events every year. The hypertensive subjects are usually classified as ‘controlled’ or ‘not controlled’. In our opinion, this is not an adequate method to describe the problem, thus impairing the development of effective improvement strategies. Knowledge of the relevant variables, such as coexisting diseases, the level of CV risk, the degree of BP control and the drugs prescribed, are important to better understand the reasons of suboptimal treatment. This information should be obtained in the setting where hypertensive patients are usually cared for, that is, in general practice. We, therefore, used a large primary care database to investigate how Italian GPs treat their hypertensive patients.
Surprisingly, the most important problems in hypertension management seem to be the low attendance rate of hypertensive patients and BP under-recording by GPs. Even if many of the subjects without BP, a recorded BP had this taken at some point during their treatment, the lack of an up to date record greatly impairs the effective management of hypertension. The magnitude of these problems does not change even when some of the non-attending subjects were hause-bound and possibly well controlled at home. It is clear that any strategy aimed at improving BP control cannot work if 44% of the hypertensive patients have unknown BP values. We think that the approach to this huge shortcoming should involve both the GPs and the Health Authorities. A pay-for-performance contract for GPs13 can significantly increase the BP recording rate,14 and, on the other hand, solo or group practices can improve their performances adopting simple an unexpensive self auditing procedures.15 Both approaches could increase the number of patients with recorded BP values and possibly reduce the number of non-attainders, focusing GPs on the need to better motivate and involve their hypertensive patients.
Among the subjects whose BP was recorded, 43% had uncontrolled BP. The prescription of antihypertensive drugs in non-controlled patients is suboptimal. Even if the number of drugs raises significantly with the increases in BP levels and in CV risk, the magnitude of under-prescription is enormous. This is a well-known problem. A recent survey in the United States16 showed that although the treatment rate of patients with diabetes mellitus, coronary artery disease, stroke and chronic kidney disease is higher than that of non-high-risk subjects, lack of therapy remains unacceptable—from 10.6 to 34.1% depending on the associated disease, a percentage very close to ours.
Our data clearly showed that the problem of the inadequate control of hypertension should be viewed under two different perspectives: that of the single doctor/patient and that of the public health management.
Considering the clinical practice of the GPs, therapeutic inertia—the failure to start new drugs or increasing the dose in patients with an abnormal clinical measurement—17 seems to be playing a pivotal role; this is a well-known phenomenon. Observational studies in the United States have found that therapeutic inertia is common in hypertension, diabetes and hypercholesterolaemia. Therapeutic inertia can even persist when BP is quite high; in a recent large US study, one-third of patients with persistent BP ⩾160/100 mm Hg had no changes in their treatment or had a spontaneous return to lower BP over 6 months (although the study did not examine whether this reflected prescribing decisions or patient adherence).18 This trend also was confirmed in Europe. In a British study, the treatment was not intensified in nearly half (45%) of consultations in which the patients had a single suboptimal BP reading. Similarly, no intensification occurred in 36% of consultations after two successive suboptimal BP readings, and in 27% of those taking fewer than three drugs.19 In our study, only 21% of non-controlled patients were prescribed with ⩾3 drugs, showing that therapeutic inertia is a major problem also in Italian primary care. Of most concern also is the under-prescription among grade 3 hypertensive and high-risk patients.
On the other hand, it must be noticed that over three quarter of the non-controlled patients have grade I hypertension, most of them with BP values <150/90 mm Hg; this BP level is considered satisfactory according to the standards of the British GPs. Furthermore, most of our grade 1 hypertensive patients are non-high risk individuals with a mean BP very close to that recommended by the guidelines. For these patients, the individual odds of benefit from small BP decreases are low. Furthermore, to reach the current targets, many patients will require up to four drugs, with many also taking aspirin and a statin (five or six drugs in total), but in terms of lowering CV risk, which is the purpose of treatment, the benefits from adding a fifth, or even a sixth drug, are scant.20 Additional drugs have diminishing benefit, but an equal or greater chance of side effects and interactions. It is not, therefore, surprising that both doctor and patients are not eager to increase the number of daily tablets. On the other hand, such an effort could be justified for high-CV risk patients, for whom a small BP decrease offers the greater benefit. This study shows that the number of prescribed drugs is, in fact, higher for high-risk and for grade 2–3 hypertensive patients. Despite this, according to our data, ⩾2 additional antihypertensive drugs could be necessary to aim to the target BP recommended for high-CV risk patients;3 this is a relevant effort for patients, GPs, and National Health Services. Computer-based strategies21, 22 could be a useful tool both to identify the high-risk patients and to remind the GP to increase the antihypertensive therapy (and to check patient compliance).
According to the Cochrane Collaboration,23 antihypertensive drug therapy should be vigorously increased when patients do not reach target BP levels. In our opinion, this therapeutic approach would not suit the cases of the great number of low-risk, grade 1 hypertensive patients, for whom the individual benefit of a further BP reduction is small. Should we be happy with a <150/90 mm Hg BP, accepting real practice and British GPs contract standards, or should we struggle to reach guidelines’ targets? We think that only the Governments can answer this question, as the advantages of such a strategy are relevant when the whole population is considered. The availability of low-cost, well-tolerated and safe drugs make this option attractive and cost effective. It is clear that such a population strategy should be agreed with doctors and patients, and adequately supported by the National Health Service.
Our study has many limits. We examined the data form GPs who volunteered to participate to a research database; it is, therefore, probable that the recording rate of BP and CV risk factors is higher than that of the other Italian GPs; on the other hand, the prescription pattern of these GPs does not differ from that of the other Italian GPs.
We must also remember that <140/90 mm Hg is not the target for high-risk patients, according to the international guidelines.3 The rate of BP control in these patients is thus overestimated. On the other hand, in clinical practice, some patients have their BP measured in non-ideal condition; therefore, there is the possibility of overestimating the rate of non-control.
We did not examine the therapeutic adherence. Successful BP control is possible with a combination of intensification and adherence, suggesting that the therapy intensification must be coupled with interventions to enhance medication adherence.24 Unfortunately, both persistence25 and adherence26 are low among Italian hypertensive patients; it is, therefore, possible that these problems play a relevant role in BP under-control.
It must be also noticed that even with politherapy and good therapeutic adherence, current targets are low enough to be unachievable for most patients. Even in clinical trials, with protocol driven prescribing and willing participants, most fail to achieve systolic BP below 140 mm Hg.27 Even if 100% control is not possible, our data about the prevalence of renal failure and diuretic use in non-controlled patients clearly show that the identification of low GFR and a more rationale use of loop diuretics can be a simple and easy improvement strategy.
Low attendance rate of hypertensive patients, and BP under-recording are major obstacles to hypertension control, and should be immediately addressed both by the Health Authorities and GPs. The use of politherapy increases on accordance with the risk level, but remains largely suboptimal, both among low- and high-CV risk patients, thus representing a valuable target for improvement. Most uncontrolled individuals are non-high-CV risk, grade 1 hypertensive patients, for whom the single personal benefit of adding another drug is modest; however, increasing the therapy in such a great number of people can lead to a relevant nationwide decrease of CV events. Different improving strategies should therefore be developed for low-risk subjects and for grade 2–3 hypertensive/high-CV risk patients.
Canto JG, Iskandrian AE . Major risk factors for cardiovascular disease: debunking the ‘only 50%’ myth. JAMA 2003; 290: 947–949.
Wolf-Maier K, Cooper RS, Banegas JR, Giampaoli S, Hense HW, Joffres M et al. Hypertension prevalence and blood pressure levels in 6 European countries, Canada, and the United States. JAMA 2003; 289: 2363–2369.
Mancia G, de Backer G, Dominiczak A, Cifkova R, Fagard R, Germano G et al. 2007 Guidelines for the Management of Arterial Hypertension The Task Force for the Management of Arterial Hypertension of the European Society of Hypertension (ESH) and of the EuropeanSociety of Cardiology (ESC). J Hypertens 2007; 25: 1105–1187.
Düsing R . Overcoming barriers to effective blood pressure control in patients with hypertension. Curr Med Res Opin 2006; 22: 1545–1553.
Coca A, Aranda P, Bertomeu V, Bonet A, Esmatjes E, Guillén F et al. [Strategies for effective control of arterial hypertension in Spain. Consensus document]. Rev Clin Esp 2006; 206: 510–514.
Filippi A, Bignamini AA, Sessa E, Samani F, Mazzaglia G . Secondary prevention of stroke in Italy: a cross-sectional survey in family practice. Stroke 2003; 34: 1010–10149.
Cricelli C, Mazzaglia G, Samani F, Marchi M, Sabatini A, Nardi R et al. Prevalence estimates of chronic diseases in Italy: exploring the differences between selfreport and primary care databases. J Public Health Med 2003; 25: 254–257.
World Health Organisation. International Classification of Diseases. Manual of the international statistical classification of diseases: injuries and causes of deaths, 9th revision. World Health Organisation: Geneva, 1977.
World Health Organisation Collaborating Centre for Drug Statistics Methodology. ATC Index with DDDs. World Health Organisation: Oslo, 2001.
Sessa E, Samani F, Niccolai C, Pecchioli S, Ventriglia G, Mazzaglia G . The creation of a validated sample of doctors in general medicine for database Health Search. J Italian Soc Gen Pract (SIMG) 2004; 3: 10–14.
Lawrenson R, Williams T, Farmer R . Clinical information for research: the use of general practice databases. J Pub Health Med 1999; 21: 299–304.
Brosius FC, Hostetter TH, Kelepouris E, Mitsnefes MM, Moe SM, Moore MA et al. Detection of chronic kidney disease in patients with or at increased risk of cardiovascular disease. A Science Advisory From the American Heart Association Kidney and Cardiovascular Disease Council; the Councils on High Blood Pressure Research, Cardiovascular Disease in the Young, and Epidemiology and Prevention; and the Quality of Care and Outcomes Research Interdisciplinary Working Group Circulation 2006; 114: 1083–1087.
Roland M . Linking physician pay to quality of care: a major experiment in the UK. N Engl J Med 2004; 351: 1488–1454.
Doran T, Fullwood C, Gravelle H, Reeves D, Kontopantelis E, Hiroeh U et al. Pay-for-performance programs in family practices in the United Kingdom. NEJM 2006; 355: 375–384.
Jamtvedt G, Young JM, Kristoffersen DT, O’Brien MA, Oxman AD . Audit and feedback: effects on professional practice and health care outcomes. Cochrane Database Syst Rev 2006; 19: CD000259.
Wong ND, Lopez VA, L’Italien G, Chen R, Kline SEJ, Franklin SS . Inadequate control of hypertension in US adults with cardiovascular disease comorbidities in 2003–2004. Arch Intern Med 2007; 167: 2431–2436.
Phillips LS, Branch WT, Cook CB, Doyle JP, El Kebbi IM, Gallina DL et al. Clinical inertia. Ann Intern Med 2001; 135: 825–834.
Rodondi N, Peng T, Karter AJ, Bauer DC, Vittinghoff E, Tang S et al. Therapy modifications in response to poorly controlled hypertension, dyslipidaemia and diabetes mellitus. Ann Intern Med 2006; 144: 475–484.
Inkster ME, Montgomery A, Donnan P, MacDonald TM, Sullivan F, Fahey T . Organisational factors in relation to control of blood pressure: an observational study. Br J Gen Pract 2006; 55: 931–937.
Marshall T . Coronary heart disease prevention: insights from modelling incremental cost effectiveness. BMJ 2003; 327: 1264–1267.
Garg AX, Adhikari NKJ, McDonald H, Rosas-Arellano MP, Devereaux PJ, Beyene J . Effects of computerized clinical decision support systems on practitioner performance and patient outcomes. JAMA 2005; 293: 1223–1238.
Filippi A, Sabatini A, Badioli L, Samani F, Mazzaglia G, Catapano A et al. Effects of an automated electronic reminder in changing the antiplatelet drug-prescribing behavior among Italian general practitioners in diabetic patients: an intervention trial. Diabetes Care 2003; 26: 1497–1500.
Fahey T, Schroeder K, Ebrahim S . Interventions used to improve control of blood pressure in patients with hypertension. Cochrane Database Syst Rev 2006; 18: CD005182.
Ho PM, Magid DJ, Shetterly SM, Olson KL, Peterson PN, Masoudi FA et al. Importance of therapy intensification and medication nonadherence for blood pressure control in patients with coronary disease. Arch Intern Med 2008; 168: 271–276.
Mazzaglia G, Mantovanic LG, Sturkenboome MC, Filippi A, Trifiró G, Cricelli C et al. Patterns of persistence with antihypertensive medications in newly diagnosed hypertensive patients in Italy: a retrospective cohort study in primary care. J Hypertens 2005, 2093–2100.
Poluzzi E, Strahinja P, Vaccheri A, Vargiu A, Silvani MC, Motola D et al. Adherence to chronic cardiovascular therapies: persistence over the years and dose coverage. Br J Clin Pharmacol 2007; 63: 346–355.
Mancia G, Grassi G . Systolic and diastolic blood pressure control in antihypertensive drug trials. J Hypertens 2002; 20: 1461–1464.
About this article
Cite this article
Filippi, A., Paolini, I., Innocenti, F. et al. Blood pressure control and drug therapy in patients with diagnosed hypertension: a survey in Italian general practice. J Hum Hypertens 23, 758–763 (2009). https://doi.org/10.1038/jhh.2009.14
The role of baseline and post-treatment frontal QRS-T angle for detecting arterial blood pressure control
Clinical and Experimental Hypertension (2021)
Blood pressure control in Australian general practice: analysis using general practice records of 1.2 million patients from the MedicineInsight database
Journal of Hypertension (2021)
Current prescription status of antihypertensive drugs in Japanese patients with hypertension: Analysis by type of comorbidities
Clinical and Experimental Hypertension (2019)
Renal Failure (2018)
Hypertension Status and Associations with Self-Rated Health and General Practitioner Health Seeking in a Rural Australian Cohort
Journal of Cardiovascular Development and Disease (2018)