SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References

J ClinHypertens (Greenwich). 2010;12:502–507. © 2010 Wiley Periodicals, Inc.

The purpose of this study was to evaluate the reasons physicians provided when the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) guidelines recommending a thiazide diuretic as a first line treatment for hypertension were not followed. A subsample of patients from a randomized controlled study who had uncontrolled blood pressure at an index visit and were not prescribed a thiazide were evaluated. Differences in groups that received any medication change or therapeutic lifestyle changes counseling and those that did not were compared. Differences in treatment were also compared for patients who received educational materials with or without telephone calls and financial incentive with a control group. The authors examined whether patients achieved blood pressure control in 12 months. The results show providers are not aggressive enough with getting blood pressure to goal and patients who are more educated about hypertension may be less likely to experience clinical inertia.

Uncontrolled hypertension (HTN) persists in the population, regardless of age, race, gender, and frequency of health care visits. Actual control rates vary between those that are treated and untreated and those with comorbid conditions, but are as low as 33% of adult hypertensive patients in the United States.1 According to current Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) guidelines, a thiazide diuretic is a first line therapy alone or in combination with other antihypertensive agents for most patients.2 Many hypertensive patients are not being treated to recommended goal blood pressures (BPs)3–6 even though most providers are aware of and agree with JNC 7 guidelines.7 Many patients with “resistant” HTN can frequently achieve controlled BP if a thiazide diuretic is added.3,4,7,8

Many reasons are given for failing to intensify therapy when a patient has hypertensive readings. Even though guidelines clearly state the threshold at which patients are hypertensive, some providers extend those thresholds to higher levels9,10 and justify their approach because the patient’s BP is “controlled.”10 Additionally, some physicians do not attempt to change treatment during a clinic visit because the visit is not focused on HTN or they prefer to continue monitoring BP before making changes.9 These and other issues frequently result in a patient’s BP not being adequately controlled BP for many months or even years.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References

This study evaluated secondary data from a parent study from a randomized, controlled trial that evaluated the implementation of guidelines and changing the behavior of providers by encouraging patients to discuss thiazide diuretics with their providers.11 This parent study recruited 958 veterans from the Veterans Affairs (VA) Medical Centers in Iowa City, Iowa and Minneapolis, Minnesota and at an associated VA Community Based Outpatient Clinic. Inclusion and exclusion criteria are listed in Table I.

Table I.   Inclusion/Exclusion Criteria
Inclusion criteria
 Patients seen in primary care clinics at the study sites within the last 13 months and who did not have an active prescription for a thiazide diuretic; and
 Prior diagnosis of hypertension, as documented in Veterans Affairs (VA) outpatient administrative files; and
 Less than 80 years of age; and
 Blood pressure above treatment goal at the 2 most recent VA outpatient clinic visits; or
 Blood pressure at goal (as defined above) during one of the last 2 outpatient clinic visits, but the patient is receiving a prescription for a calcium channel blocker (CCB)
Exclusion criteria
 Less than 2 or more visits to a primary care clinic in the prior 12 months
 Active prescription for a thiazide or loop diuretic or lithium
 Documented allergy to thiazides or to sulfa agents
 Previously documented intolerance or adverse drug reaction to thiazide diuretics
 Renal insufficiency, defined by a glomerular filtration rate <30 mL/min
 No serum creatinine lab value collected in the past year
 Prior history of hypokalemia or serum potassium <3.5 meq/L in the prior year
 Diagnosis of gout or active prescription for allopurinol
 Congestive heart failure due to systolic dysfunction with a documented left ventricular ejection fraction <35% by echocardiography, nuclear medicine study, or ventriculography
 Residence in a long-term care facility
 No telephone for follow-up calls
 Life expectancy <6 months
 Inability to give informed consent or impaired cognitive function (defined as >4 errors on the 10-item Pfeiffer Portable Mental Status Questionnaire, administered during study intake)

Physicians were randomized into 2 groups (Figure). Patients were either placed in the Pure Control Group (n=60) or the Intervention Group (n=511) by virtue of their physician’s assignment. The Intervention Group was further randomized into Control Group 2 (n=127), Intervention Group A (n=140), Intervention Group B (n=119), or Intervention Group C (n=125). Patients in the Pure Control Group had a provider who had no patients in any of the intervention groups. Patients in Intervention Group A received an intervention letter with the patient’s latest BP, Framingham score, and the benefits of a thiazide diuretic and a postcard to take to their primary care physician (see below). Patients in Intervention Group B received the same intervention letter, postcard and a $20 incentive for discussing the letter with their providers as well as 6 months reimbursement for their medication copayment. Patients in Intervention Group C received the letter, postcard, the financial incentive, 6 months reimbursement for their medication copayment, and a reminder call from the research coordinator to remind them of their appointment and of the letter and ask if they had any questions prior to their index visit. The patients were asked to attend their regularly scheduled visit with a primary care provider. BP was measured at each subsequent primary care visit.

image

Figure Figure.  Study design. BP indicates blood pressure; CBOC, community-based outpatient clinic.

Download figure to PowerPoint

A postcard that was given to all 3 intervention groups was designed for the patient to take the postcard to their index appointment and then return it to the research team after the appointment to receive the financial incentive(s). The postcard required the provider’s signature confirming a discussion regarding thiazides had taken place. The postcard also provided a space for the provider to explain why a thiazide was not prescribed.

Only patients from the parent study noted above who had uncontrolled BP and who were not receiving a thiazide diuretic, were included in the present analyses including those with diabetes or renal insufficiency with systolic BP (SBP) ≥130 mm Hg or diastolic BP (DBP) ≥80 mm Hg at the index visit and were not prescribed a thiazide diuretic. Patients with uncomplicated HTN were eligible if they had SBP ≥140 mm Hg or DBP ≥90 mm Hg as defined by JNC 7 guidelines and were not prescribed a thiazide diuretic. Reasons for not prescribing a thiazide diuretic were compiled using either the information from the postcard and/or from the electronic medical record. Documentation of a discussion of a thiazide with the provider from the postcard or electronic medical record was noted. BP measurements from the 12 month visit were used to ascertain if the patient met their BP goal.

The data were analyzed using SAS 9.1.3, service pack 2 (SAS Institute Inc, Cary, North Carolina). Chi-square tests (or Fisher exact tests where appropriate) were used to assess statistical significance.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References

There were 269 patients with uncontrolled HTN who met the study criteria with 112 patients in the 2 control groups and these were combined since there were no significant differences between the 2 control groups. Likewise, there were 157 in the 3 intervention groups and these were combined because there were no significant differences in the intervention groups. The majority of the patients were men (n=264, 98.1%), the mean age was 64.3 years (8.5 standard deviation), 54.7% had comorbid diabetes (n=140), and 17.5% had renal insufficiency (n=47). None of these patients were receiving a thiazide diuretic at the initial index visit, even though there were no known contraindications. Table II lists the reasons that providers gave for not initiating thiazide therapy for their patient.

Table II.   Reasons Given by Physicians for Not Prescribing a Thiazide
  1. aControl stated by physician even though the patient’s blood pressure was not controlled at the clinic visit.

Other medication change was made following the intervention letter, n=98
 Increase dose of current medication, n=56, 57.1%
 Add new nonthiazide medication, n=21, 21.4%
 Patient noncompliance—restart previous medication,  n=11, 11.2%
 Therapeutic lifestyle change counseling (low-salt diet, exercise, etc), n=8, 8.2%
 Change current medication (stop current medication and start new medication), n=2, 2%
Reasons no medication change was made following the intervention letter, n=171
 Blood pressure controlled,a n=105, 61.4%
  Controlled—lower home blood pressures,   n=44, 25.7%
  Controlled—provider indicated that patient’s  hypertension was controlled at primary care visit,  even though the blood pressure was above goal,  n=61, 35.6%
 Watch and wait, n=30, 17.5%
 Not addressed, n=29, 17%
 Previous adverse drug reaction or contraindication to a thiazide, n=2, 1.2%
 Patient comanaged by a provider outside the VA, n=2, 1.2%

The majority of study patients, (n=171, 63.6%), received no medication change or therapeutic lifestyle counseling (TLC). Table III shows the difference in medication changes or TLC for different situations or patient types. Patients with isolated diastolic HTN, patients with lower BP goals (eg, diabetes), and within 10 points of SBP and/or DBP goal were significantly less likely to receive any change in therapy. Patients who received educational materials about thiazide diuretics and HTN, those who had a discussion about thiazide diuretics with their provider, and patients with only elevated SBP had no statistically significant difference in rates of medication changes or TLC from their provider than patients without these characteristics.

Table III.   Percent of Patients Receiving Medication Changes or TLC
Patient TypeComparator GroupP Value
  1. Abbreviations: BP, blood pressure; DBP, diastolic blood pressure; SBP, systolic blood pressure; TLC, therapeutic lifestyle changes. aThe 3 intervention groups were combined. bThe 2 control groups were combined.

Intervention groups,a 36.9%Control group,b 35.7%.84
Had a lower goal BP, <130/80 mm Hg, 29.4%Had a higher therapeutic goal, <140/90 mm Hg, 48.5%.0017
Elevated DBP only, 12.9%Elevated SBP or SBP and DBP, 39.5%.0038
Within 10 points of goal, 26.1%BP above 10 points of goal, 47.3%.0003
Discussion of thiazide with provider, 34.2%No discussion of thiazide with provider, 42.1%.22
SBP only, 40%Elevated DBP or SBP and DBP, 33.3%.26

Patients in the 2 control groups (n=112) and patients who received any intervention (n=157) were compared to see what differences existed in their provider’s choice of treatment and why they were not prescribed a thiazide. Two treatment differences were significant between the control groups compared to the intervention groups (Table IV). Providers in the combined intervention groups were significantly less likely to state they did not make a medication change because they wanted to “watch and wait” (P=.03), but they were significantly more likely to not make a change because BP was lower at home (P=.035). The control groups were more likely to be reported as controlled in the physician’s opinion or not have HTN addressed in the medical record note or postcard that was returned to the research team than the interventional material patients, but these results did not achieve statistical significance.

Table IV.   Treatment Differences Between Patients in the Control Group and Patients in the Intervention Groups
Type of ActionControl Group, n=112, %Intervention Group, n=157,*%P Value
  1. *Intervention groups A, B, and C combined.

Add new medication 6.310.8.19
Increase dose18.822.2.48
None—controlled33.942.7.15
None—not addressed12.5 9.6.44
None—watch and wait16.1 7.6.03
Lower home blood pressures10.720.4.03

Table V shows the differences between the groups who did and did not receive an intervention and rates of BP control. There was no significant difference between patients who did or did not receive a medication change or TLC on achieving BP goal at 12 months after the index visit.

Table V.   Medication Changes or TLC and Target Blood Pressure at 12 Months
 % at BP Goal% Not at BP Goal
  1. Abbreviations: BP, blood pressure; TLC, therapeutic lifestyle changes. aP=0.76.

No medication change or TLC, n=18735.864.2
Medication change/TLC,a n=8237.862.2

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References

This study identified reasons patients with uncontrolled HTN were not prescribed a thiazide diuretic in spite of 3 possible interventions for providers. Interestingly, the most common reason given by the physicians was that the patient had controlled BP (22.7% of all reasons) even though it was clearly not controlled, and 16.4% said that BP was controlled at home. It is not known why the physicians would indicate BP was controlled, when it clearly was not. The problem with home BP is that the physician seemed to accept these values without acknowledging that the goal for home BP is at least 5 mm Hg lower than clinic BP or that the veracity of home BP is often in question unless the BP device stores the values.2 Another common reason (11%) for not changing therapy was “watch and wait.” Many of these reasons seem to indicate typical patterns of “clinical inertia” or failure to act when the BP is not controlled which contributes to overall poor BP control.

Patients who initiated a discussion about thiazides with their physician were no more likely to receive a medication change (not involving a thiazide) or TLC than patients who did not initiate a discussion regarding thiazides. Perhaps targeting the discussion to a thiazide drug instead of overall HTN control drew the focus of the conversation away from a candid discussion between the provider and the patient about other treatment options if the provider did not choose to prescribe a thiazide.

Patients in this study who did not receive any medication change or TLC at the index visit had no significant differences in the ability to reach their BP goals at 12 months.

Patients with diabetes or renal insufficiency that have a lower BP goal were less likely to have changes made to their medication regimen or be given TLC than those with a standard goal of <140/90 mm Hg. Patients with isolated DBP were less likely to have medication changes or TLC than those with elevated SBP and DBP or isolated SBP. Patients who were within 10 points of their BP goal were also less likely to have medication changes made or receive TLC compared with those that were above 10 points of their BP goal. When comparing the control groups to the intervention groups, the control groups were more likely to have no changes to medication therapy or TLC with the provider choosing to monitor future readings. These patients were also less likely to have lower home BP readings noted as the reason for no change in the chart note or postcard that was returned to the research team. This finding could be explained by the fact that patients who received intervention educational materials were more educated about their HTN and wanted to be more proactive about their treatment. The physician may have been more willing to note home BP while having a discussion about thiazide diuretics with their patients.

Reasons for No Medication Change or TLC

Patients who had a lower BP goal as defined in JNC 7 were less likely to receive a medication change or TLC than patients with the goal of <140/90 mm Hg. Of the 269 patients in this study, 170 patients had a BP goal of <130/80 mm Hg due to a diagnosis of diabetes or renal insufficiency. Over one-fourth, (n=50, 29.4%), of these patients had medication changes made or were given TLC. Studies have found that only 24%–37.5% of patients who were currently receiving HTN treatment and had diabetes are reported as controlled compared to 50%–64% of patients without diabetes.1 However, a recent clinical trial found that HTN was controlled in 82% of patients with diabetes following an intensive team-based intervention.7 Our findings suggest physicians were not aggressive enough for patients with diabetes or renal insufficiency.

Patients with only elevated diastolic values were less likely to receive a medication change or TLC than patients with either isolated systolic HTN or patients with elevations in both SBP and DBP. Interestingly, other studies found that physicians were more aggressive with an elevated DBP than an elevated SBP.5,9 Most cases of diastolic elevation occur in younger populations and this younger population is more likely to have controlled BP.6,10 In this study, rates of medication changes or TLC did not differ when a patient only had isolated systolic HTN.

Patients within 10 points of their goal BP were less likely to receive a medication change or TLC. The study by Oliveria and coworkers9 found that physicians reported the lowest BP they would treat was 150 mm Hg systolic and 91 mm Hg diastolic. One-third of physicians did not recommend treating patients with SBP from 140–160 mm Hg and DBP from 90–100 mm Hg.10 Our findings are consistent with these studies and suggest many physicians do not intensify therapy when BP is close to goal. Even so, a 10 mm Hg reduction in SBP can result in a 40% reduction in stroke and 30% reduction in ischemic heart disease.12

Patient Perceptions

In a qualitative evaluation of the parent clinical trial used for the present study, patients were interviewed shortly after their index visit and asked why they felt they were not prescribed a thiazide diuretic.11 Most patients gave the reason that their BP was controlled and additional medication was not needed. Unfortunately, in many cases these patients’ BP was not controlled to current recommended goals. The next most common reason for not prescribing a thiazide was the presence of comorbidities, specifically benign prostatic hyperplasia and diabetes. Diabetes is not a contraindication to starting a thiazide diuretic and JNC 7 guidelines recommend a thiazide even with this compelling indication.2 Patients who had diabetes and received a thiazide had even greater reductions in risk than those not prescribed a thiazide.13

Limitations

The BPs in this study were routine clinic measurements rather than validated measurements performed by trained research personnel. The readings were rarely repeated. The study population was mainly white, elderly, and male. Our data were obtained from either the patient chart or the postcard that the patient returned to determine the reason for not starting a thiazide. We did not observe the discussion between the patient and provider to know the thought process behind their choice. Although the patient was scheduled to have a visit with their primary care provider, we could not control if the visit was focused on HTN.

Conclusions

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References

Patients with only elevated DBP, a lower BP goal, and that were within 10 points of their BP goal were less likely to receive an intervention regardless of guidelines and a discussion initiated by the patient about thiazides. Patients who received intervention educational materials were less likely to receive any medication changes or TLC and more likely to be reported as controlled due to lower home BP. These patients were also less likely to have providers choose to watch future BP and wait to make a decision on further treatment. There were many cases in which medication was not intensified or thiazide added, even though BP was clearly not well controlled.

Acknowledgments and disclosures:  The authors would like to thank John Holman, MA, for his contribution to this paper by performing the statistical analyses. The research reported here was supported by the Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development (HSR&D) Service Merit Review Grant (IMV 04-066-1) and through the Center for Research in the Implementation of Innovative Strategies in Practice (CRIISP) (HFP 04-149). Dr Carter is also supported by the Agency for Healthcare Research and Quality (AHRQ) Centers for Education and Research on Therapeutics Cooperative Agreement #5U18HSO16094 and the National Heart Lung and Blood Institute, (HL091841). The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs. The Veterans Administration Project to Implement Diuretics (VAPID) was registered at Clinicaltrials.gov: NCT00265538.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Conclusions
  7. References