PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Clin Hypertens (Greenwich). Author manuscript; available in PMC Apr 1, 2009.
Published in final edited form as:
PMCID: PMC2453045
NIHMSID: NIHMS53282
A Cluster-Randomized Trial to Evaluate Physician/Pharmacist Collaboration to Improve Blood Pressure Control
Barry L. Carter, PharmD,1,2 George R. Bergus, MD,2 Jeffrey D. Dawson, ScD,3 Karen B. Farris, PhD,1 William R. Doucette, PhD,1 Elizabeth A. Chrischilles, PhD,4 and Arthur J. Hartz, MD, PhD.2
1Division of Clinical and Administrative Pharmacy, College of Pharmacy, University of Iowa, Iowa City, IA
2Department of Family Medicine, Roy J. and Lucille A. Carver College of Medicine, University of Iowa, Iowa City, IA
3Department of Biostatistics, College of Public Health, University of Iowa, Iowa City, IA
4Department of Epidemiology, College of Public Health, University of Iowa, Iowa City, IA
Correspondence: Barry L. Carter, Pharm.D., FCCP, FAHA, Division of Clinical and Administrative Pharmacy, Room 527, College of Pharmacy, University of Iowa, Iowa City, Iowa, 52242, Phone: 319−335−8456, Fax: 319−353−5646 Email: barry.carter/at/uiowa.edu
The purpose of the present study was to evaluate a physician/pharmacist collaborative model to improve BP control. The study was a prospective, cluster-randomized controlled clinical trial that enrolled 179 patients with uncontrolled primary hypertension aged 21−85 years (mean 61 years). At 9 months, the mean adjusted difference in SBP was 8.7 (95% CI: 4.4, 12.9) mm Hg, while the difference in DBP was 5.4 (CI: 2.8, 8.0) mm Hg. The 24-hour BP levels showed similar effects with mean SBP 8.8 (CI: 5.0, 12.6) mm Hg and DBP 4.6 (CI: 2.4, 6.8) mm Hg lower in the intervention group. BP was controlled in 89.1% of patients in the intervention group and 52.9% in the control group (adjusted odds ratio 8.9; CI: 3.8, 20.7; p<0.001). Physician/pharmacist collaboration achieved significantly better mean BP and overall BP control rates primarily by intensification of medication therapy and improving patient adherence.
Keywords: hypertension management, clinical trial, pharmacist management, blood pressure control
Hypertension is a serious problem throughout the world affecting more than one billion people.1, 2 Uncontrolled BP is thought to contribute to 7 million deaths worldwide each year.2 Controlling BP can reduce heart failure by over 50%, strokes by 35−40% and myocardial infarctions by 20−25%.3 The reasons for poor control are multifactorial and include patient, physician and structural factors.4-8 Reports suggest that physicians do not adhere to hypertension guidelines.8-12 One common finding is that medications are frequently not used optimally when BP remains uncontrolled.4-8 These problems contribute to the findings that hypertension is only controlled in 37% of the 65 million Americans with this condition.13
Studies have evaluated numerous approaches to improve BP control but have generally found that education is an insufficient strategy.13-17 Various analyses have found that changing the organization of office practice including adding pharmacists can improve outcomes for chronic conditions including hypertension.18-22 An international review of best practices to affect change in improving practice suggested that expanding pharmacists' roles led to better prescribing behaviour.18 An analysis of clinical trials conducted by the Stanford-UCSF Evidence-based Practice Center for the Agency for Health Care Research and Quality (AHRQ) found that the most effective strategies include interdisciplinary management of hypertension.13, 15 These analyses found that using clinical pharmacists in case management resulted in some of the greatest improvements in BP control.22-29 Many previous studies, however, were small single site studies, involved only one intervention pharmacist, did not control for many patient, physician or clinic variables and/or did not use an unbiased BP measurement technique. There was also a suggestion of publication bias since most studies were small.13, 15
The purpose of the present efficacy study was to evaluate the ability of a physician\pharmacist collaborative model to improve BP control while maintaining high internal validity. We hypothesized that patients cared for using this model would achieve lower mean BP values and higher rates of BP control as defined by the seventh Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure guidelines (JNC-7).3
Study sites
This study was a prospective, cluster-randomized controlled efficacy trial involving five clinics operated by one university. Clinics were stratified based on whether the clinic initially had a clinical pharmacist on staff and then randomized to control (n=3) or intervention clinics (n=2). Randomization of clinics was performed using a table of random numbers. Randomization at the clinic level was used to minimize contamination at the physician level. The trial was designed as an efficacy study in order to have high internal validity.
General internists or family physicians staffed all clinics. One intervention site and one control site had clinical pharmacists on staff prior to the study. The three clinical pharmacists in the control site abstained from making recommendations for any patients in the control group but they continued to answer general treatment questions from physicians in their clinic.
Patients, Physicians and Data Collection
Educational lectures were provided to physicians in all five clinics (control and intervention sites) by one investigator (BLC) before patients were enrolled. The majority of participating physicians attended these training sessions. Handouts, slides and the JNC-7 express version were supplied to all physicians including those who were unable to attend these sessions.
To reduce variation due to physician training, only patients cared for by faculty physicians were included in the study. Due to financial pressures to increase efficiency and clinic revenues, these physicians scheduled all their patients much like private practice so typical visits were brief.
Lists of patients with diagnostic codes for hypertension were obtained and the research nurses screened medical records of all patients with hypertension. Patients who met the study criteria were sent a post card and asked to contact the research nurse. In addition, research nurses periodically screened daily schedules for patients with hypertension. Finally, physicians could refer patients to the study.
Males or females aged 21 to 85 years with a diagnosis of hypertension were eligible if they did not have diabetes and their clinic BP was between 145−179 mm Hg systolic BP or 95−109 mm Hg diastolic BP. Patients with diabetes with a clinic BP between 135−179 mm Hg systolic BP or 85−109 mm Hg diastolic BP were eligible. Exclusion criteria included: BP medication or dose change within four weeks of the baseline visit, enrolment in the 24-hour BP monitoring consult service within the previous 6 months, stage 3 hypertension (BPs ≥ 180/110 mm Hg), evidence of hypertensive urgency or emergency, recent myocardial infarction or stroke (6 months prior to enrolment), New York Heart Association Class III or IV CHF, unstable angina, serious renal or hepatic disease, pregnancy, poor prognosis (life expectancy less than 3 years), dementia or cognitive impairment.
Prior to this study, there were no clinical trials of this model that randomized by clinic. Thus, several fixed and random effects that affect power were unknown a priori, such as within- and between-patient variability, between-physician variability, and between-clinic variability. Therefore, we used several techniques to estimate power and sample size. First, we powered the study by letting sigma denote the population standard deviation of the change scores averaged across physicians within each clinic for a given quantitative outcome (e.g., mean BP) and assumed an alpha=0.05 (two-sided). The power for detecting a 3.4-sigma difference between 2 intervention clinics and 3 control clinics would be 80%, and a 3.9-sigma difference will be detected with 90% certainty. Even with this uncertainty, the clinical effect was known from previous studies to be 10−12 mm Hg SBP.13, 15 We assumed a typical two-sample comparison of normal data and a two-tailed test with α and β of 0.05 and 0.95, respectively. The estimated sample size was 47 patients per group. As noted above, because this was a longitudinal study with several fixed and random effects, we inflated the sample size to 90 patients per group (180 total) in consideration of this unknown variability.
The study was approved by the University of Iowa Institutional Review Board and all patients signed informed consent. Patients and physicians were aware that there were control and intervention groups as part of the consent process but this fact was not emphasized by the investigators and research nurses. Nonetheless, most physicians and patients likely recognized their group assignment. However, both physicians and patients were blind to the 24-hour BP results so the 24-hour results were the most robust for analysis purposes. Two different research nurses were dedicated to patients in either control sites or intervention sites to minimize contamination. The research nurses were employed by the investigators specifically for this study. The research nurse collected the following data at the baseline visit: patient age, gender, race, educational degree, insurance status, household income, marital status, smoking status, alcohol intake and history of co-existing conditions. They measured the patient's height and weight, calculated a BMI, recorded all antihypertensive medications, doses and dates of last refills and performed a pill count of BP medications. The nurse personally administered questions on adverse reactions.30 The adverse reaction questionnaire was developed for another study and included 47 questions of typical medication side effects.30 For each potential reaction the patient was asked: “in the past 4 weeks how much have you been bothered by....” The patient could rate the potential reaction: zero (not at all), 1 (a little bit), 2 (somewhat), 3 (quite a bit) or 4 (very much). The responses for each patient were summed (potential range from 0−188).
Research nurses were specially trained to measure BP using American Heart Association guidelines and the process used in the AASK trial.31, 32 Specifically, the nurses measured the subjects' BP three times at each data collection visit using a mercury sphygmomanometer using standardized techniques from BP clinical trials.31, 32 The second and third values were averaged and used as the clinic BP. The nurses were certified quarterly in their ability to accurately position patients and measure BP to ensure consistent and valid readings. The clinic BP values were provided to the physician and/or clinical pharmacist for patients in both the control and intervention groups. The clinical pharmacist then interviewed patients in intervention sites (see intervention below). Patients in both the control and intervention groups saw their physicians at the baseline visit.
Next, the research nurse placed a 24-hour BP monitor set to measure the BP every 20 minutes during the day and every 30 minutes during sleep (SpaceLabs 90217-A, SpaceLabs Medical, Redmond, Washington).33 The 24-hour results were used as a blinded objective outcome and were not made available to either the patient's physician or clinical pharmacist until the patient completed the trial. Finally, patients in both groups were given written information on hypertension from NHLBI. The research nurses encouraged all patients (control and intervention) to follow the lifestyle modifications (diet, exercise, stopping smoking) as described in these resources. Patients were also provided with their goal BP.
Patients returned at 2, 4, 6 and 8 months for follow-up data collection visits with the research nurses where clinic BP measurements, adverse reaction and pill counts were repeated. At the 9-month visit, the nurses performed all of the same procedures as performed at the baseline visit, including repeating the 24-hour BP. Patients received $100 if they completed both 24-hour BP measurements to reimburse them for the inconvenience of wearing the 24-hour monitors and the extra time required to return the monitors. Patients were telephoned prior to clinic visits to encourage adherence with study visits.
Intervention physicians and pharmacists underwent teambuilding exercises conducted by two investigators (KBF, WRD) using previous strategies.34 The sessions explored strategies to investigate sub-optimal treatment, poor medication adherence, potential adverse reactions, drug interactions or other barriers to success. If there was disagreement, the physician made the final decision and these instances were recorded to determine the degree of acceptance of the pharmacists' recommendations.
There were five intervention clinical pharmacists, four of whom were faculty or clinical pharmacy residents in the university family medicine intervention site. The fifth was placed into the community-based intervention clinic that had never had a clinical pharmacist on staff prior to this study. We hoped to increase the generalizability of the study by including five different clinical pharmacists and one site that had not previously had a clinical pharmacist. The pharmacists were well versed in hypertension treatment. However, two initial 90-minute training sessions were conducted by one investigator (BLC) to ensure that intervention pharmacists provided a consistent intervention. These training sessions included the JNC-7 guidelines, strategies to improve BP control, methods to optimize therapy and strategies to improve medication adherence. Follow-up discussions were held at least quarterly with the pharmacists to ensure fidelity to the intervention.
The intervention protocol specified a patient interview at baseline by the clinical pharmacist. The pharmacist assessed the patient's regimen, suggested a goal BP and provided recommendations to improve BP control. BP control was defined as an office BP <130/80 mm Hg for patients with diabetes or chronic kidney disease and <140/90 mm Hg for all other patients.3 The protocol specified that pharmacists should recommend therapies consistent with JNC-7 and educate the physician by providing background information if necessary.3, 35 The primary focus of the pharmacists was to address suboptimal medication regimens. For instance pharmacists suggested to add thiazide diuretics if not in the regimen, increase medication doses to at least moderate levels, utilize appropriate combination regimens based on pharmacology and utilize agents for co-existing conditions when appropriate (e.g. angiotensin converting enzyme inhibitors for patients with diabetes). The second major area of the intervention protocol was for patients with poor medication adherence. The pharmacist recommended adherence aids if poor adherence was unintentional. If poor adherence appeared to be intentional, the pharmacist tried to negotiate a strategy to improve adherence. The pharmacists educated all patients with poor medication adherence using written information from NHLBI, and/or taught them to perform home monitoring. All study visits with intervention pharmacists occurred in the medical office clinic. Pharmacists were encouraged to attend each clinic visit (2, 4, 6, and 8 months) and they were encouraged to initiate additional visits or telephone contact if BP remained uncontrolled. The results of these interviews served as the basis for patient-specific recommendations and feedback to the physician. Pharmacists could not independently prescribe therapy so all changes were approved by the physician. Most recommendations to the physician were performed face-to-face during the patient visit but some physicians provided the authority for pharmacists to make dosage changes and then inform them immediately after the visit. Every encounter with the pharmacist was recorded on a case report form that included all recommendations made by the pharmacist. We confirmed how the physician reacted by reconciling the recommendation with the medication list and dose the research nurse collected at each study visit. In the control group, we examined the baseline medication list and dosages and examined changes at each study visit to determine changes by physicians in the control group.
All patient data were entered into case report forms by the research nurses. Individual data elements were double-entered into an Access® database by a blinded data management team that included data technicians, the data manager and the biostatistician (JDD).
Descriptive statistics (means, standard deviations, and percentages) of patient demographic and health-related variables were calculated at baseline for each group. Medication adherence was calculated from the pill counts as the percent of predicted doses measured at each study visit. Baseline comparisons between the groups were made using Student's t-test and Fisher's Exact test. Preliminary analysis revealed that the response variables were correlated within-subject, but no significant clustering due to clinics or physicians was observed. For continuous responses (SBP and DBP), likelihood-based mixed models with random patient effects were fit in SAS Proc Mixed to incorporate all available data from baseline through 9 months in an intention-to-treat analysis. For BP control, a Generalized Estimating Equation (GEE) model using the binomial distribution and the logit link was fit in SAS Proc Genmod, accommodating the correlations across patients. For both of these types of models (mixed and GEE), contrasts were estimated in order to test for the treatment effect at 2, 4, 6, 8, and 9 months post-baseline. Also in these models, we adjusted for baseline BP level, age, gender, race, education, insurance status, household income, marital status, smoking status, alcohol intake, body mass index, number of co-existing conditions at baseline, number of baseline antihypertensive medications, baseline medication adherence, and total number of visits.
Patient recruitment began in January 2004 and patients were assigned to the control or intervention group by virtue of the clinics' randomization. The last patient completed the trial in October 2006. We enrolled 179 patients and 160 (89.4%) subjects had data at both the baseline and 9-month visit (Figure 1) (p=0.47 between groups). Most patients (87%) completed all six study visits (5.7 ± 1.0 intervention group vs 5.5 ± 1.3 control group; p=0.43). There were optional visits with the pharmacists in the intervention group, resulting in 6.8 ± 1.6 total visits in that group which we controlled for in the analyses. When adjusted for the intervention effect, the within-clinic interclass correlation coefficient (ICC) for SBP at 9 months was 0.0084 (within clinic variance 139.0, between clinic variance 1.2) (not significant). When adjusting for all relevant baseline covariates, the ICC went from 0.0084 down to 0.0010. Similarly, the physician effects appeared to be very small, as within-physician ICC was 0.0097, within-physician variance was 138.4 and the between-physician variance was 1.4 (not significant). When adjusting for the baseline covariates, the between-physician ICC went from 0.0097 down to 0.0005. These results demonstrate there was no clustering of effect by clinic or physician.
Figure 1
Figure 1
Flow of patients through the study protocol
Baseline characteristics of the subjects are shown in Table 1. The baseline number of antihypertensives were not different between the intervention (1.5 ± 1.0) and control groups (1.4 ± 1.0). There was no difference in the percentage of patients who were prescribed medication at baseline between the control (76%) and intervention (84%) groups (p=0.185).
Table 1
Table 1
Patient Demographics at Baseline —
Table 2 displays the BP results at each pre-specified study visit. After adjustment for the covariates, the mean difference (control group minus the intervention group) in 9-month SBP was 8.7 (95% CI: 4.4, 12.9) mm Hg (Figure 2), while the adjusted mean difference in 9-month DBP was 5.4 (CI: 2.8, 8.0) mm Hg (data not shown). The 24-hour BP effect size was nearly identical with a mean difference of 8.8 (CI: 5.0, 12.6) mm Hg for SBP and 4.6 (CI: 2.4, 6.8) mm Hg for DBP.
Table 2
Table 2
Clinic BP, 24-hour blood pressure and BP control
Figure 2
Figure 2
Clinic measured systolic BP
BP was controlled in 89.1% of patients in the intervention group and 52.9% in the control group (adjusted odds ratio 8.9; CI: 3.8, 20.7; p<0.001, Figure 3). BP was controlled in 62.8% of non-diabetics in the control group and 91.4% in the intervention group (adjusted odds ratio of 10.2; CI: 3.4, 29.9; p<0.001). For patients with diabetes, BP was controlled in 23.5% of patients in the control group and 81.8% in the intervention group (adjusted odds ratio of 40.1; CI: 4.1, 394.7; p=0.002). Table 3 illustrates that adjusted between-group differences were similar to the crude effects.
Figure 3
Figure 3
Blood pressure control based on clinic blood pressure based on <140/90 mm Hg for uncomplicated hypertension and <130/80 mm Hg for patients with diabetes or chronic kidney disease
Table 3
Table 3
Unadjusted and adjusted effects of intervention vs control at 9 months.
We performed a sensitivity analysis to check the robustness of our findings in the presence of informative dropout. First, we reran our analysis under a scenario that all 19 subjects who dropped out had uncontrolled BP at the end of the study and found that the intervention and control BP control rates would be 81.2% and 46.2%, respectively, (adjusted OR 6.1; CI: 2.1, 17.7, p<0.001). More pessimistically, we considered the scenario where all dropouts in the intervention group had uncontrolled BP and all dropouts in the control group had controlled BP. In this situation, the respective BP control rates would be 81.2% and 59.0%, (adjusted OR 5.3 CI: 1.9, 14.2; p=0.001).
By the end of the study, the mean number of antihypertensives, was significantly higher (p=0.003) in the intervention group (2.4 ± 0.9) when compared to the control group (1.9 ± 1.0). At baseline, medication adherence was significantly better in the control group (89%) compared to the intervention group (71%) (p<0.001). There was no apparent reason for this baseline difference. By the 9 month visit there was no difference in medication adherence (92% in the control group vs 94% in the intervention group p=0.369).
There was no difference in the side effect score at baseline (mean 26.5 control group vs. 28.8 intervention group, p=0.397). In spite of the increase in medications in both groups, side effects scores declined at 9 months to 18.3 in the control group (p=0.003 vs baseline) and 22.2 in the intervention group (p = 0.014 vs baseline). There was no difference in side effect scores between groups at 9-months (p=0.135).
The clinical pharmacists made 267 recommendations (2.6 per patient) to change BP medications and physicians accepted 256 (95.9%). Of all the drug-therapy (n=256) changes made by physicians on the recommendations of the pharmacists, the majority were to increase the dose (34%), add another non-diuretic antihypertensive (30%) and add a thiazide diuretic (17%). Other recommendations included to switch within class (5%), decrease a dose (4%), discontinue a drug (7%) or change dose frequency (3%). Most of the recommendations (60%) occurred at or before the two-month visit. Physicians in the control group changed medications 100 times (1.28 per patient, p <0.001 compared to the intervention group).
The clinical pharmacists made 441 recommendations concerning lifestyle modifications to patients and most involved increasing activity (45%), reducing weight (27%) or initiating the DASH diet (22%). Many of these recommendations were multiple suggestions to the same patients at several clinic visits. There were only 17 (4%) recommendations to improve medication adherence which suggests that nonadherence was rarely a problem. Other than medication adherence, we did not systematically evaluate the degree to which patients changed their exercise level or diet.
We found very high BP control rates (89%) and mean BP reductions which confirms reviews that suggest interdisciplinary management of BP is a highly effective approach.15, 22 Most studies involving pharmacists reviewed for AHRQ found control rates of 45−70% and a difference of approximately 14 mm Hg in systolic BP.13 Our controlled efficacy trial found a BP control rate of 89% and a difference in SBP of 8.7 mm Hg for research-measured BP and 8.8 mm Hg with 24-hour BP.
There are several explanations for the good BP results in the control group including: the research nurse reinforced the goal BP, adherence and lifestyle modifications and provided written material on hypertension which are known to reduce BP. The protocol required all patients see their physicians at the baseline visit and this increased surveillance probably caused physicians in the control group to provide medical care once they were alerted to the lack of BP control. There was also the possibility of a Hawthorne effect. Thus, the control group cannot be considered usual care. Even so, our intervention was still much more effective at achieving BP control. The effect in the control group appeared to peak at 6 months and then wane at the 9-month period. In contrast, BP control was continuing to increase in the intervention group. It is not known if longer follow-up would increase the differences between groups or if the effect in the intervention group might also wane with time.
Controlling BP within six months reduces cardiovascular risks.36 Most of the pharmacists' recommendations were made in the first two months and BP was controlled in 73% of patients at 6 months and 89% at 9 months. Importantly, BP was controlled to current standards in 81.8% of patients with diabetes. BP began to diverge at the 2-month visit and became significant at 4−6 months that was likely due to the pharmacodynamic delay in antihypertensive response.37
A recent study provided a clinical pharmacist educational intervention along with unit dose blister packages, both intended to improve medication adherence in patients with hypertension.16 Following a six month intervention, systolic BP decreased from 133.2 ± 14.9 mm Hg to 129.9 ± 16.0 mm Hg (p=0.02). Patients (n=159) were then randomized to continued clinical pharmacist intervention with unit dose medications or usual care with standard prescription bottles for another six months. This study did not attempt to intensify medication doses, but focused on medication adherence. SBP continued to decline in the intervention group to 124.4 ± 14.0 mm Hg but deteriorated slightly in the usual care group 133.3 ± 21.5 mm Hg (p=0.005). It is not possible to determine, however, if the improvement was due to the clinical pharmacists, the unit dose packaging or both. It is interesting that SBP after 12 months was similar to our study at 9 months; 133.3 vs 133.0 mm Hg in the control group, respectively, and 124.4 vs 124.2 mm Hg in the intervention group respectively. However, our baseline systolic BP values were much higher (151 vs 134 mm Hg) and most of our patients had good medication adherence.
The usual strategy to overcome suboptimal therapy or poor guideline adherence has been to provide educational lectures and information on the guidelines. These approaches in our study, along with increased surveillance noted above achieved BP control in 53% of patients with previous uncontrolled BP in our control group. The present study suggests that educational approaches will not be optimal. Our findings suggest that poor BP control was due to sub-optimal medication regimens and support our previous cross sectional study in the same group of physicians that found no relationship between knowledge and BP control.17 The majority of recommendations in our study involved adding medications or increasing dosages and doing this early in the intervention. In most cases decisions were made collaboratively by the physician and pharmacist. Physicians agreed with the pharmacists' recommendations 95% of the time. We did not capture the instances when physicians actually turned care over to the pharmacists. However, since pharmacists in our state cannot prescribe medications independently, any new prescriptions had to be at least signed by the physician.
It may appear that the interventions to change drug therapy were simple and it might be questioned why physicians did not make these changes on their own. However, Berlowitz found that physicians frequently did not increase medication dosages even though patients continued to have uncontrolled BP in spite of up to six physician visits per year.5 Oliveria and coworkers found that the primary barrier for poor BP control was related to physicians who were satisfied with poorly controlled BPs.6 Garg et al evaluated the causes of “resistant” hypertension referred to a specialty hypertension clinic.4 The most common reasons for “resistant” hypertension were: drug-related causes (61% including sub-optimal regimens), patient nonadherence (13%), secondary hypertension (7%) or other (18%). These findings are consistent with our study, which found that the lack of BP control was primarily due to suboptimal regimens. Primary care physicians are required to deal with multiple competing priorities during short office visits. Our intervention was probably effective because pharmacists could focus on achieving therapeutic goals for patients taking medications for hypertension.38
Importantly, the greater use of antihypertensives in the intervention group did not lead to higher side effect scores, which may seem counterintuitive. Other studies have demonstrated increased quality of life and reduced adverse symptoms when BP becomes controlled following the use of medications including thiazides.28, 39, 40
Patient nonadherence was a less common cause of poor BP control in our study which is also consistent with the studies by Garg and Oliveria.4, 6 Nevertheless, medication adherence based on pill counts improved from 71% to 94% in the intervention group. It is not known if this improvement was due to the counseling by the pharmacists or by the increased attention in the study or both. It may also seem unusual that medication adherence was so high at both the baseline and 9-month periods, especially in the control group. Caro et al found medication adherence was 97% after one year of follow-up and 82% after 4.5 years of follow-up in patients with established hypertension.41 Our patients and those of Caro et al, continued to receive care and are likely to be more adherent than patients who have dropped out of care. In addition, studies have also found that patients taking more medications may adhere at a higher rate possibly explained by the Health Belief Model.42
This study had several strengths including the use of standardized clinic BP measurements,31 24-hour BP monitoring, intention-to-treat analyses and control of numerous co-variables. The importance of standardized BP measurements is reflected by the fact that usual office BP measurements are often performed incorrectly.43 It might be assumed that the data on pharmacist-assisted BP management were already conclusive. A review of the international literature identified 63 quality improvement strategies in hypertension and 12 involved pharmacy interventions.13 These and other authors have suggested that modifying the role of pharmacists to include disease management may improve outcomes of care.18, 21, 44 However, the studies of pharmacist-assisted management of hypertension had many limitations. Few studies used research nurses to measure BP,23 none controlled for baseline co-variates or used an intention-to treat analysis. In addition, our study is the first to use 24-hour BP monitoring. Another strength of this study is that we randomized by clinic, which avoided contamination that would have occurred if we had randomized by patient or by physician. Finally, we studied five clinics and used five intervention pharmacists, which are more than nearly all previous studies of physician\pharmacist collaboration in hypertension.
While the study was designed to have high internal validity, there are some limitations of the present study. First, we used a cluster design that randomized a small number of clinics. However, this study used the largest numbers of clinics to date and our ICC for clinics and physicians were extremely low and not statistically significant so uneven clinic or physician characteristics did not likely influence these results. Our design was stronger than randomization at the patient level where physicians would have had patients in both the control and intervention group, potentially leading to serious contamination. Our design resulted in uneven enrolment between the two groups but this did not negatively impact either the ICC or the overall results. This efficacy study can only be generalized to clinics that principally utilize faculty physicians. Our design was used to simulate other efficacy clinical trials in hypertension to achieve high internal validity.31, 45 It is likely that BP control rates would be different in other practice settings and when patients were not required to attend specific research study visits.
While the intervention may be considered unusual or costly, current guidelines for the U.S. Department of Veterans Administration/Department of Defense state: “If BP continues to be elevated, clinicians should consider...management by a pharmacist in the follow-up and adjustment of medications to improve BP goal.”46 Expanded roles for pharmacists in managing medications for chronic conditions have been proposed in the US and supported by reviews of the literature by authors from the Netherlands, Canada and the UK.18, 21, 44 Future studies of this model should include more clinics with greater geographic, ethnic and socioeconomic diversity since these populations are likely to respond differently to the intervention. Finally, our study only lasted 9 months and future studies should determine how long such an intervention is effective.
An intervention involving physician\pharmacist collaboration that focused on optimizing and intensifying medications was associated with significant reductions in BP and improvements in BP control. This study was the first to include 24-hour BP monitoring to objectively confirm clinic pressures. These improvements were correlated with increased intensity of medication use, which suggests the model had an effect to overcome sub-optimal medication regimens. The intervention also improved medication adherence in the small number of patients with poor adherence without increasing side effects. This study suggests that for clinics or health systems that have clinical pharmacists, their reallocation to provide more direct patient management can significantly improve BP control.
Acknowledgements
The authors acknowledge Alan Zillich, PharmD, Janyce Stewart RN and Gail Ardery PhD (project managers); Karen Kluesner, RN and Sheryl Eastin RN (research nurses); Jessica Milchak, PharmD, Michael Ernst, PharmD, Cynthia Weber, PharmD, Jennifer Steffensmeier, PharmD, Michael Kelly, PharmD (intervention pharmacists); Yinghui Xu, MS (data management); Carrie Franciscus (database development); Paul James, MD, Christopher Goerdt, MD and David Katz, MD (data and safety monitoring board).
Supported by the National Heart, Lung, and Blood Institute, HL069801. Dr Carter is also supported by the Center for Research in Implementation in Innovative Strategies in Practice (CRIISP), Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development Service (HFP 04-149). 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.
Funding for this project was supported by the National Heart, Lung, and Blood Institute, HL069801. Dr Carter is also supported by the Center for Research in Implementation in Innovative Strategies in Practice (CRIISP), Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development Service (HFP 04-149). 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.
Footnotes
Presented, in part, at the 22nd Annual Scientific Meeting and Exposition of the American Society of Hypertension, May 19, 2007, Chicago, Illinois.
1. 2003 European Society of Hypertension-European Society of Cardiology guidelines for the management of arterial hypertension. J Hypertens. 2003;21:1011–1053. [PubMed]
2. World Health Report 2002: Reducing risks, promoting healthy life. World Health Organization.; Geneva, Switzerland: 2002. http://www.who.int/whr/2002/ [PubMed]
3. Chobanian AV, Bakris GL, Black HR, et al. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension. 2003;42:1206–1252. [PubMed]
4. Garg JP, Elliott WJ, Folker A, Izhar M, Black HR. Resistant hypertension revisited: a comparison of two university-based cohorts. Am J Hypertens. 2005;18:619–626. [PubMed]
5. Berlowitz DR, Ash AS, Hickey EC, et al. Inadequate management of blood pressure in a hypertensive population. N Engl J Med. 1998;339:1957–1963. [PubMed]
6. Oliveria SA, Lapuerta P, McCarthy BD, L'Italien GJ, Berlowitz DR, Asch SM. Physician-related barriers to the effective management of uncontrolled hypertension. Arch Intern Med. 2002;162:413–420. [PubMed]
7. Hyman DJ, Pavlik VN. Characteristics of patients with uncontrolled hypertension in the United States. N Engl J Med. 2001;345:479–486. [PubMed]
8. Milchak JL, Carter BL, James PA, Ardery G. Measuring adherence to practice guidelines for the management of hypertension: an evaluation of the literature. Hypertension. 2004;44:602–608. [PMC free article] [PubMed]
9. Carter BL, Malone DC, Ellis SL, Dombrowski RC. Antihypertensive Drug Utilization in Hypertensive Veterans With Complex Medication Profiles. J Clin Hypertens (Greenwich) 2000;2:172–180. [PubMed]
10. Hill MN, Levine DM, Whelton PK. Awareness, use, and impact of the 1984 Joint National Committee consensus report on high blood pressure. American Journal of Public Health. 1988;78:1190–1194. [PubMed]
11. Naiman DJ, Barker LR. Adherence to the JNC VI guidelines for the treatment of hypertension in the resident clinic. Maryland Medical Journal. 1999;48:307–311. [PubMed]
12. Nelson CR, Knapp DA. Trends in antihypertensive drug therapy of ambulatory patients by US office-based physicians. Hypertension. 2000;36:600–603. [PubMed]
13. Ong KL, Cheung BM, Man YB, Lau CP, Lam KS. Prevalence, awareness, treatment, and control of hypertension among United States adults 1999−2004. Hypertension. 2007;49:69–75. [PubMed]
14. Goldstein MK, Lavori P, Coleman R, Advani A, Hoffman BB. Improving adherence to guidelines for hypertension drug prescribing: cluster-randomized controlled trial of general versus patient-specific recommendations. Am J Manag Care. 2005;11:677–685. [PubMed]
15. Walsh JM, McDonald KM, Shojania KG, et al. Quality improvement strategies for hypertension management: a systematic review. Med Care. 2006;44:646–657. [PubMed]
16. Lee JK, Grace KA, Taylor AJ. Effect of a Pharmacy Care Program on Medication Adherence and Persistence, Blood Pressure, and Low-Density Lipoprotein Cholesterol: A Randomized Controlled Trial. JAMA. 2006:2563–71. [PubMed]
17. Carter BL, Hartz A, Bergus G, et al. Relationship between physician knowledge of hypertension and blood pressure control. J Clin Hypertens (Greenwich) 2006;8:481–486. [PubMed]
18. Grol R, Grimshaw J. From best evidence to best practice: effective implementation of change in patients' care. Lancet. 2003;362:1225–1230. [PubMed]
19. Battersby MW. Health reform through coordinated care: SA HealthPlus. BMJ. 2005;330:662–665. [PMC free article] [PubMed]
20. Wilson T, Buck D, Ham C. Rising to the challenge: will the NHS support people with long term conditions? BMJ. 2005;330:657–661. [PMC free article] [PubMed]
21. Avery AJ, Pringle M. Extended prescribing by UK nurses and pharmacists. BMJ. 2005;331:1154–1155. [PMC free article] [PubMed]
22. Carter BL, Zillich AJ, Elliott WJ. How pharmacists can assist physicians with controlling blood pressure. J Clin Hypertens. 2003;5:31–37. [PubMed]
23. Bogden PE, Abbott RD, Williamson P, Onopa JK, Koontz LM. Comparing standard care with a physician and pharmacist team approach for uncontrolled hypertension. J Gen Intern Med. 1998;13:740–745. [PMC free article] [PubMed]
24. Bond CA, Monson R. Sustained improvement in drug documentation, compliance, and disease control. A four-year analysis of an ambulatory care model. Arch Intern Med. 1984;144:1159–1162. [PubMed]
25. Borenstein JE, Graber G, Saltiel E, et al. Physician-pharmacist comanagement of hypertension: a randomized, comparative trial. Pharmacotherapy. 2003;23:209–216. [PubMed]
26. Bogden PE, Koontz LM, Williamson P, Abbott RD. The physician and pharmacist team. An effective approach to cholesterol reduction. J Gen Intern Med. 1997;12:158–164. [PMC free article] [PubMed]
27. Carter BL, Barnette DJ, Chrischilles E, Mazzotti GJ, Asali ZJ. Evaluation of hypertensive patients after care provided by community pharmacists in a rural setting. Pharmacotherapy. 1997;17:1274–1285. [PubMed]
28. Erickson SR, Slaughter R, Halapy H. Pharmacists' ability to influence outcomes of hypertension therapy. Pharmacotherapy. 1997;17:140–147. [PubMed]
29. Hawkins DW, Fiedler FP, Douglas HL, Eschbach RC. Evaluation of a clinical pharmacist in caring for hypertensive and diabetic patients. Am J Hosp Pharm. 1979;36:1321–1325. [PubMed]
30. Kaboli PJ, Hoth A, Carter BL, et al. The VA Enhanced Pharmacy Outpatient Clinic (EPOC) Study: A randomized-controlled pharmacist-physician intervention trial. J Gen Intern Med. 2004;19:227. (abstract)
31. Wright JT, Jr., Bakris G, Greene T, et al. Effect of blood pressure lowering and antihypertensive drug class on progression of hypertensive kidney disease: results from the AASK trial. JAMA. 2002;288:2421–2431. [PubMed]
32. Pickering TG, Hall JE, Appel LJ, et al. Recommendations for blood pressure measurement in humans and experimental animals: Part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Hypertension. 2005;45:142–161. [PubMed]
33. O'Brien E, Coats A, Owens P, et al. Use and interpretation of ambulatory blood pressure monitoring: recommendations of the British hypertension society. BMJ. 2000;320:1128–1134. [PMC free article] [PubMed]
34. Farris KB, Cote I, Feeny D, et al. Enhancing primary care for complex patients. Demonstration project using multidisciplinary teams. Can Fam Physician. 2004;50:998–1003. [PMC free article] [PubMed]
35. Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA. 2003;289:2560–2572. [PubMed]
36. Julius S, Kjeldsen SE, Weber M, et al. Outcomes in hypertensive patients at high cardiovascular risk treated with regimens based on valsartan or amlodipine: the VALUE randomised trial. Lancet. 2004;363:2022–2031. [PubMed]
37. Carter BL, Frohlich ED, Elliott WJ, Moore MA, Mann RJ, Roberts RW. Selected factors that influence responses to antihypertensives. Choosing therapy for the uncomplicated patient. Arch Fam Med. 1994;3:528–536. [PubMed]
38. Singer GM, Izhar M, Black HR. Goal-oriented hypertension management: translating clinical trials to practice. Hypertension. 2002;40:464–469. [PubMed]
39. Neaton JD, Grimm RH, Jr., Prineas RJ, et al. Treatment of Mild Hypertension Study. Final results. Treatment of Mild Hypertension Study Research Group. JAMA. 1993;270:713–724. [PubMed]
40. Materson BJ, Reda DJ, Cushman WC, et al. Single-drug therapy for hypertension in men. A comparison of six antihypertensive agents with placebo. The Department of Veterans Affairs Cooperative Study Group on Antihypertensive Agents. N Engl J Med. 1993;328:914–921. [PubMed]
41. Caro JJ, Salas M, Speckman JL, Raggio G, Jackson JD. Persistence with treatment for hypertension in actual practice. CMAJ. 1999;160:31–37. [PMC free article] [PubMed]
42. Billups SJ, Malone DC, Carter BL. The relationship between drug therapy noncompliance and patient characteristics, health-related quality of life, and health care costs. Pharmacotherapy. 2000;20:941–949. [PubMed]
43. Villegas I, Arias IC, Botero A, Escobar A. Evaluation of the technique used by health-care workers for taking blood pressure. Hypertension. 1995;26:1204–1206. [PubMed]
44. Rothman R, Weinberger M. The role of pharmacists in clinical care: where do we go from here? Eff Clin Pract. 2002;5:91–94. [PubMed]
45. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA. 2002;288:2981–2997. [PubMed]
46. VA/DoD Clinical practice guideline for diagnosis and management of hypertension in the primary care setting. Department of Veterans Administration, Department of Defense. 2004. Revision July 2005.