- Top of page
Recently there has been an enormous increase in interest in measuring the quality of medical care.1 Improving quality of care is the first reason for measuring it.1 Another reason is accountability of the medical profession. Measurement and monitoring of quality can be achieved by developing quality indicators (QIs) or quality measures, derived directly from clinical guidelines or other appropriate sources.2 The criteria for developing QIs and the principles of evidence-based medicine should be in mind when assessing quality.3,4
To whom information on quality of care should be available is another issue for debate. Emphasis should also be given to the principles and models of quality improvement, which may be defined as a process of planned activities that have the aim of improving care. It involves the specification of desired performance, the review of actual performance, and the implementation of changes in care if necessary, followed by further review of performance to check the impact of the changes. The process is frequently depicted as a cycle.5 However, the first thing that is needed is baseline data from which to draw comparisons in the future. The process of measuring and improving quality is regarded as a priority by policy makers, patients, and professionals of healthcare systems and remains an important issue for further investigation.6
Hypertension (HT) is among the most common outpatient diagnoses in the United States.7 High blood pressure (BP) is a common medical condition affecting 1 in 4 Americans.8 More than 90% of Americans will be affected by it at some point in their lives.9 About half of Americans age 45 years or older have high BP.10
Hypertension (HT) is a significant risk factor for cardiovascular disease that increases with age.11 Patients with HT are at risk for stroke, ischemic heart disease, and other cardiovascular diseases. Vulnerable populations such as the elderly and those with comorbidities (e.g., diabetes, kidney disease) are at even greater risk. Hypertension (HT) doubles the lifetime risk of stroke.12 High BP was listed as a primary or contributing cause of death in approximately 278 000 deaths in the United States in 2003.13 Nearly one-third of adults with high BP are unaware of their condition, increasing the risk of complications and death.10
Despite available effective treatment options, 2 in 3 people with HT are untreated or undertreated.11 Antihypertensive therapies have been associated with a 35% to 40% reduction in stroke incidence, a 20% to 25% reduction in ischemic heart disease, and a more than 50% reduction in heart failure.14
In 2007, the estimated direct and indirect costs associated with high BP in the United States totalled $66.4 billion.13
- Top of page
Table 1 shows the percentages of patients who reached the target for every QI at first and last visit. Blood pressure (BP) was the only QI measured carefully. In more than 70% there were no measurements for LVH, and 60% had no measurements for LDLc. Large percentages of missing values are depicted for the rest of the QIs. Forty percent had achieved the target for systolic and diastolic BP, but recommended care was not delivered for the other QIs, although for some of them there was a statistically significant difference. QI improvement was obtained for BP, TChol, LDLc, HDLc (Figure 1). At last visit, fewer patients showed K and Na abnormalities. The following QIs showed no improvement: TGs, Hematocrit (Hct), Glu, UA, CrCl, BMI, ECG Sokolow-Lyon index, and LVH (Figure 2). Despite antihypertensive treatment and better BP control, renal dysfunction and LVH did not show any improvement.
Figure 1. Quality indicators with improvement. Abbreviations: BP, blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TChol, total cholesterol.
Download figure to PowerPoint
Figure 2. Quality indicators without improvement. Abbreviations: BMI, body mass index; CrCl, creatinine clearance; ECG, electrocardiogram Sokolow-Lyon <35 mm; Glu, glucose; LVWT, left ventricular wall thickness; IVS, Interventricular Septum; PW, Posterior Wall <11 mm; TGs, triglycerides; UA, uric acid.
Download figure to PowerPoint
Table 1. Quality Indicators
| ||First Visit||Last Visit|
| ||Present, N (%)||Missing, N1 (%)||Present, N (%)||Missing, N1 (%)|
|BP <140/90 mm Hg||133 (11)a||0 (0)||481 (40)a||4 (0.3)|
|Systolic BP <140 mm Hg||212 (17.6)a||0 (0)||610 (50.7)a||4 (0.3)|
|Diastolic BP <90 mm Hg||348 (28.9)a||0 (0)||789 (65.6)a||3 (0.2)|
|Total cholesterol <200 mg/dL||174 (17.7)a||224 (18.6)||203 (26)a||424 (35.2)|
|LDLc <160 mg/dL||278 (57.7)a||724 (60)||357 (68.1)a||682 (56.6)|
|LDLc <130 mg/dL||120 (24.9)a||724 (60)||174 (33.2)a||682 (56.6)|
|HDLc >40 mg/dL male, >50 mg/dL female||445 (69)a||563 (46.7)||453 (72)a||577 (47.8)|
|Triglycerides <150 mg/dL||550 (69)a||409 (33.9)||473 (65.3)a||482 (40)|
|Hct >37% female, >40% male||889 (91)a||230 (19.1)||653 (86)a||447 (37.1)|
|Glucose ≤125 mg/dL||895 (89.1)a||202 (16.7)||656 (83.2)a||418 (34.7)|
|Uric acid ≤6.5 mg/dL||704 (76.3)a||283 (23.5)||519 (73.6)a||501 (41.5)|
|Potassium 3.5–5.5 mmol/L||654 (93.6)b||507 (42)||545 (94.8)b||631 (52.3)|
|Na 135–145 mmol/L||581 (83.8)c||513 (42.5)||494 (87)c||638 (52.9)|
|Creatinine clearance ≥50 mL/min||665 (92.4)a||486 (40.3)||551 (84.8)a||556 (46.1)|
|BMI ≤25||240 (24.9)a||243 (20)||214 (23)a||277 (23)|
|Sokolow-Lyon <35 mm||767 (94.1)a||391 (32.4)||590 (93.7)a||576 (47.8)|
|Left ventricular wall thickness||181 (65.8)a||931 (77.2)||126 (47.9)a||943 (78.2)|
Table 2 shows the variation of BP according to stratification. There was a significant decrease in the number of patients with grade 2 and grade 3 HT at first visit, from 39.1% to 16.1% and 22.1% to 3.5% at last visit, respectively. But there was an increase in the number of patients with grade 1 HT from 27.8% at first visit to 40.4% at last one. Further analysis showed that from 471 patients with grade 2 HT at first visit, 215 (45.7%) had grade 1 HT at last visit and only 161 (34.3%) reached the target for BP. From 267 patients with grade 3 HT at first visit, 96 (36.2%) had grade 1 HT at last visit and only 83 (31.1%) reached the target for BP. From 335 patients with grade 1 HT at first visit, 132 (39.5%) did not improve, 154 (46.1%) had BP <140/90 mm Hg at last visit, and 48 (14.4%) had grade 3 HT. Of 133 patients with BP <140/90 mm Hg at first visit, 83 (62.4%) remained in target, 43 (32.3%) had grade 1 HT, and 7 (5.3%) had grade 2 HT at last visit.
Table 2. Variation According to Stratification of Blood Pressure
| ||First Visit||Last Visit|
|Blood Pressure, mm Hg||N||%||N||%|
|Normal and high normal, ≤139/89||133||11||481||40a|
|Grade 1 HT, 140/90 to 159/99||335||27.8||486||40.4a|
|Grade 2 HT, 160/100 to 179/109||471||39.1||193||16.1a|
|Grade 3 HT, ≥180/110||267||22.1||42||3.5a|
Only 18 (11.3%) patients with diabetes mellitus had BP <130/80 mm Hg at last visit vs 2 (1.6%) at first visit. There were 122 patients (98.4%) who had BP ≥130/80 mm Hg at first visit, and 141 patients (88.7%) needed better BP control, as they had BP ≥130/80 mm Hg at last visit.
Table 3 gives some important information for the use of antihypertensive, hypolipidemic, and antidiabetic treatment. Sixty percent of patients were taking diuretics at last visit, which may affect Glu and CrCl control. Betaβ-Blockers, calcium channel blockers, and ACE inhibitors were more frequently used (44%, 43%, and 39%, respectively) than ARBs (14%). Although 68% of patients had LDLc >160 g/dL, only 15.7% were treated with statin at last visit. Indeed the results for diabetes were disappointing; 17% of patients had Glu >125 mg/dL, and 7% were treated at last visit.
Table 3. Medication
| ||First Visit||Last Visit|
The number of cardiovascular events, estimated by the outcome indicators, is shown on Table 4. Coronary artery disease (CAD) affected 19.4% of the population at last visit vs 9.7% at first visit. There was a smaller but statistically significant increase in MI (2% vs 1.6%), stroke, or TIA (4.3% vs 3.8%). Three patients died of stroke. The percentages would probably be higher if the medical records were better updated.
Table 4. Outcome Indicators
| ||First Visit||Last Visit|| |
| ||Present, N (%)||Missing, N1 (%)||Present, N (%)||Missing, N1 (%)||P|
|CAD||45 (9.7)||743 (61.6)||70 (19.4)||845 (70.0)||<0.0001|
|MI||19 (1.6)||44 (3.6)||23 (2.0)||61 (5.1)||<0.0001|
|Cerebrovascular events (stroke, TIA)||45 (3.8)||31 (2.6)||50 (4.3)||42 (3.5)||<0.0001|
There is no doubt that the majority of patients were satisfied with the quality of care provided. Table 5 shows that 54.4% of patients were satisfied with their health, and a large proportion of them were very satisfied with the structure of the HT unit; 94% to 98.4% were satisfied with the appointment system, duration of the appointment, health services, equipment, and cleanliness; and 48.6% knew that they needed little salt, and 12.2% needed no salt. Their opinion in regard to the need for medical treatment to function in their daily life ranged broadly: 14.8% said they need it not at all, 26.8% a little, 31% a moderate amount, 18.4% very much, and 9% an extreme amount. The percentage of smokers was 34.4% at the beginning of the study, and at the end it was 23.7%. Our patients were well informed and well trained.
Table 5. Patient Satisfaction
| ||Dissatisfied, N (%)||Neither Dissatisfied, nor Satisfied, N (%)||Satisfied, N (%)|
|How satisfied are you with your health?||54 (10.8)||174 (34.8)||272 (54.4)|
|How satisfied are you with your sleep?||91 (18.2)||120 (24.0)||289 (57.8)|
|How satisfied are you with the quality of care that you get from the hypertension unit?||0 (0)||16 (3.2)||484 (96.8)|
|How satisfied are you with the time-duration of every appointment?||3 (0.6)||5 (1.0)||492 (98.4)|
|How satisfied are you with your access to health services and appointment system?||2 (0.4)||28 (5.6)||470 (94.0)|
|How satisfied are you with the room of the hypertension unit, equipment, cleaning?||2 (0.4)||25 (5.0)||473 (94.6)|
Finally, 76% believed their response to antihypertensive treatment was good. The patients were extremely well informed about target levels in BP, Chol, and Glu; 97.4% believed that BP should be less than 140/90 mm Hg, and only 0.2% did not know the answer; 84.4% believed that Chol should be <200 mg/dL, and 1.4% did not know the answer; and 96.1% believed that Glu should be <110 mg/dL, and 11.6% did not answer.
- Top of page
Previous reports in the literature have indicated suboptimal BP control rates.18–22 Oliveria et al have examined barriers to primary-care physicians' willingness to increase the intensity of treatment among patients with uncontrolled HT.19 In that study, the most frequently cited reason for not addressing uncontrolled HT related to satisfaction with the current BP level, even if it was above the threshold level for treatment. On average, physicians reported that 150 mm Hg was the lowest systolic BP at which they would recommend pharmacologic therapy to a patient without comorbidities. In another study, intensifications in the antihypertensive regimen were significantly more likely to occur only when the systolic BP level reached 160 mm Hg or greater. They also found that patients who had prior medication increases were more likely to have intensification of their therapy. Prior efforts to modify the medication regimen may reduce any subsequent reluctance to making additional changes.18 Hyman and Pavlik23 found that a significant proportion of primary-care physicians do not seek the treatment goal of an systolic BP of less than 140 mm Hg and a diastolic BP of less than 90 mm Hg recommended by the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC) guidelines. Furthermore, the response patterns to the clinical scenarios provided indicated that physicians are more likely to intensify treatment for a mildly elevated diastolic BP than for a mildly elevated systolic BP and that higher BPs were tolerated in older patients. Forty percent of the respondents reported limited familiarity with the consensus guidelines in effect at the time (JNC-V). Physicians who reported familiarity with the JNC guidelines were consistently more likely to report lower BP thresholds for treatment.
Two studies emphasize the opportunities that exist to improve the care of patients with HT in the HT unit. The development of successful strategies to improve the quality of HT management hinges on an enhanced understanding of the process of how physicians make decisions about initiating and modifying pharmacologic treatment.14,21
However, in developed countries, maintaining good health increasingly involves management of problems such as HT, dyslipidemia, and diabetes, which often have no symptoms. Limitations in managing such problems are often due to clinical-inertia failure of healthcare providers to initiate or intensify therapy when indicated. Clinical inertia is due to at least 3 problems: overestimation of care provided; use of “soft” reasons to avoid intensification of therapy; and lack of education, training, and practice organization aimed at achieving therapeutic goals. Physicians will need to build into their practice a system of reminders and performance feedback to ensure necessary care. Special efforts are clearly required to reduce therapeutic inertia,24 particularly in regard to modestly elevated systolic BP levels, as the benefits of BP reduction can be substantial.25
A large study estimated that an intervention reducing BP by 5 mm Hg reduced death due to stroke by 14%, death due to coronary heart disease by 9%, and death from all causes by 7%.14 The decrease in life expectancy for individuals with high BP is 5.1 years in men and 4.9 years in women.26 Data from the most recent National Health and Nutrition Examination Survey (NHANES 1999–2000) indicate that HT prevalence is increasing in the United States, and HT control rates are unacceptably low.4 In 1999 to 2000, 28.7% of NHANES participants had HT, a 3.7% increase compared with 1988 to 1991.27 Control of high BP has been included as a Health Plan Employer Data and Information Set measure by the National Committee for Quality Assurance as part of its program to compare the performances of managed healthcare plans.27 Measure definition given from the US National Committee For Quality Assurance for BP is as follows: “This measure estimates the percentage of hypertensive adults 18 to 85 whose blood pressure was controlled. Adequate control is defined as a blood pressure reading less than 140/90 mm Hg during the past year. Both systolic and diastolic pressure must be at or under this threshold for blood pressure to be considered controlled.” Alexander et al previously demonstrated that it is feasible to assess BP control as a quality measure through review of medical records in a health maintenance organization population.20 In another study,19 they evaluated the quality of HT management in patients followed in a managed-care setting to assess the opportunities that exist for improvement. In this study, 11% were at target BP for all visits, 38% were at target BP for at least 50% of visits, and 33% were not at target BP for any visit during 1999. Only 12% had their antihypertensive therapeutic regimen intensified (dose increase or switch to a new agent).
In our study, the documentation of drug-prescribing decisions concerning BP and BP measurements were complete. Medical records were updated for every visit with BP measurements and information about whether patients were given antihypertensive treatment. This task was better focused in every visit from the responsible doctor of the hypertensive unit. We examined our patients every 3 months only if their BP was well controlled. If necessary, the next appointment could be earlier, even in a week, depending on each patient's profile. Follow-up was absolutely individualized, and adjustment of treatment took place at the right time. A special effort was made to achieve the BP target (<140/90 mm Hg) for every patient, in every visit. Although there was a team of 6 physicians, they all referred to the director of the hypertensive unit, so there was a common line in decision making for every patient. The need for change in antihypertensive treatment was well discussed. These are probably the reasons we were able to achieve better control of BP than has been shown in previous studies.18–22
Because in our study we succeeded in controlling BP in 40% of patients, efforts are still needed for better improvement. Godley et al showed that a quality improvement program in a group-model managed-care organization increased BP control from 37% to 49%.28 Fifteen percent of patients were at target BP for all visits, and 30% were not at target for any visit. We did not succeed in controlling grade 3 and grade 2 HT but did improve the grade of HT. The number of patients with grade 1 HT increased at last visit to 40.4% from 27.8% at first visit. A great percentage of patients of every grade of HT at first visit had grade 1 HT at last visit. It is obvious that we did not manage to control BP levels between 140/90 and 159/99 mm Hg.
In our unit we treat a wide range of hypertensive patients, as it is a reference center. Patients with resistant, secondary, and other types of HT are referred to us from other doctors. This could be a reason for not achieving optimal BP control. Adaptation of a quality improvement program is the next step for further improvement.
We did not manage to improve lipid profile and to provide the desirable quality of care for other QIs, such as Glu, BMI, CrCl, UA, and LVH. There was a slight improvement of TChol, LDLc, and HDLc but no improvement of TGs level. Incomplete documentation and lack of protocols for the follow-up of these QIs could explain these disappointing results. Although we informed and advised our patients of the desirable targets, we did not prescribe and more often did not adjust hypolipidemic and antidiabetic treatment. The treatment of patients with dyslipidemia and diabetes was suboptimal. An explanation could be that we did not focus enough on the cardiovascular risk stratification of every patient with HT, as guidelines require.15 Although we focused on therapeutic lifestyle changes, we did not manage to have a follow-up for lipid profile or diabetes at least every 6 months, as we had for HT. This could be a reason for suboptimal therapy.
Renal function, as it is represented by CrCl, and left ventricle wall thickness are 2 important indicators of target organ damage in the hypertensive population according to the guidelines ESC/ESH. Our focus on CrCl, LVH, Glu levels, and BMI was not enough, although we are aware of their prognostic importance for HT. That could be a good explanation for the increase in cardiovascular events assessed by outcome indicators. The percentages probably would be higher if the medical records were more informative in the results of those indicators. According to a previously published report on more than 400 indicators of healthcare quality, recommended care was not delivered to about a third of patients with HT,29 a finding that is similar to our study.
One limitation of this study was the use of medical records, as it is a retrospective study. Incomplete documentation of diagnoses or laboratory test results may have led to under-ascertainment of comorbidities. Satisfaction of our patients was unexpectedly high. A reason for that is the near-optimal structure of our unit. Every patient has his own card where we check the next appointment. There is also a specific telephone number for the patient's contact with us. The equipment and the adequate number of physicians reduce the delay time in each visit. The common line in decision making increases the confidence of patients. A very interesting observation is that more than 90% of patients know the desirable targets for BP, Chol, and Glu, and more than 50% are very careful with the use of salt. A statistically significant number of patients stopped smoking during the follow-up period; 34.4% were smokers at first visit, and 23.7% were smokers at the last visit. Well-informed and trained patients were another outcome of good structure and continuous follow-up and discussion with them. There were 38.8% who answered that their diet was very satisfactory and 52.6% that it was not.
The paradox of this population is that they did not manage to control their weight, as BMI was worse at last visit. Plasma Glu and lipid levels also were not improved. Balancing the diet and losing weight as part of a lifestyle change are goals that need a lot of work. Maybe physician and patient efforts could be enhanced by including other specialists, such as dieticians and psychologists, in the team of the HT unit.
The main purpose of this study was not to compare its findings with previous studies but to measure the provided quality in our hypertensive unit to enhance the efforts for improvement. We constructed a big database, we assessed the QIs, and we are working on the necessary changes in the structure and the process. We also realized the importance of good documentation, and so we created new forms targeting an easier, timeless selection of adequate information for every patient to serve the need to treat the whole patient and not only his BP. We educated and convinced our doctors to complete the forms correctly and to emphasize not only BP but every factor involving cardiovascular risk stratification. We are also cooperating with the Diabetology Centre and the Obesity Outpatients Unit to treat our patients more effectively. A quality improvement program focused on the overall profile of the hypertensive population has already started. Good structure could be the base for an improved process and an optimal outcome.