Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Association between encounter frequency and time to blood pressure control among patients with newly diagnosed hypertension: a retrospective cohort study

Association between encounter frequency and time to blood pressure control among patients with... INTRODUCTIONHypertension is the most common condition seen in primary care today. An estimated 80 million adults older than 20 years currently have a diagnosis of hypertension. This represents 38.9 million physician office visits and 3.6 million outpatient department visits annually. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) guidelines recommend that once antihypertensive medications are started, patients should return for follow‐up at monthly intervals until blood pressure (BP) goals are reached and then every 3 to 6 months once stable. The 2014 Evidence‐Based Guideline for the Management of High BP in Adults from the panel members of the Eighth Joint National Committee (JNC 8) does not provide follow‐up intervals, but does recommend that if BP goals are not reached within a month of starting treatment, an increase in the initial medication should be made or a second medication from one of the recommended classes should be started.Kaiser Permanente (KP) Southern California is a large, integrated healthcare system that delivers care to approximately 4.1 million members. Since 2004, KP has implemented a series of changes to improve the treatment and identification of hypertension, including: (1) creation of a hypertension registry, (2) BP measurement standardization, (3) the use of an evidence‐based and easy‐to‐remember treatment algorithm, and (4) a multidisciplinary approach to care, which includes medical assistants, nurses, and pharmacists with shared responsibilities in BP management. According to the Healthcare Effectiveness Data and Information Set (HEDIS) from 2004 to 2012 assessing BP control at KP, overall hypertension control increased from 54% to 86%.The main objective of this study was to examine and describe patients with newly diagnosed hypertension and determine whether the frequency and intervals between outpatient encounters were associated with achieving BP control. In addition, this study examined the potential risk factors associated with the time to BP control in the first year since hypertension diagnosis.METHODOLOGYIn this retrospective, observational, cohort study, we identified patients with newly diagnosed hypertension at KP Southern California between January 1, 2008, and December 31, 2014. We identified patients using electronic health records that capture sociodemographic, medical encounters, and pharmacy information for all patient care provided by KP. KP's prescription drug database was used as the source of all prescription data. This data set provided detailed information regarding the patients’ prescriptions including date written, dispensed date, refill dates, quantities dispensed, and quantity of the prescription written.Patients included in the study were aged at least 18 years with an International Classification of Diseases—Ninth Revision diagnosis code for hypertension and an antihypertensive medication filled within the 1‐year period after that diagnosis. The study inclusion criteria relative to that date were: continuous membership with the health plan for at least 1 year before diagnosis with no more than a 45‐day gap in coverage allowed, and a standard pharmacy benefit. Patients were excluded if they did not have a valid BP measurement within 1 year after hypertension diagnosis, were pregnant at any point during the study period, or if a BP medication was dispensed before the hypertension diagnosis date. The index date (ie, first day of follow‐up) was the first recorded elevated BP date on or after the date of hypertension diagnosis.Clinical outcomesBP measurements from outpatient encounters were examined to define a “hypertensive period”: the period of time between the index date and the first subsequent date at which BP was at the JNC 7 or JNC 8 targets. Essentially, the hypertensive period is the same as the follow‐up period, meaning that follow‐up stopped when patients achieved BP control. Clinical outcomes were measured during the hypertensive period, which was limited to 1 year follow‐up after the index date. Within that 1 year of follow‐up, participants were considered to have BP control if they reached their BP targets per guidelines. Patients were censored if their BP remained uncontrolled (ie, if their BP remained above goal), died, disenrolled from the health plan, or reached the end of the follow‐up period of December 31, 2014, whichever occurred first. See appendix A for number of patients censored by category.The JNC 7 BP targets were used for the study period from January 1, 2008, through December 31, 2013: BP goal was <140/90 mm Hg unless patients had chronic kidney disease or diabetes mellitus in which the BP goal was <130/80 mm Hg. The JNC 8 BP targets were used for the study period on and after January 1, 2014: if patients were 60 years and older, the BP goal was <150/90 mm Hg and if patients were younger than 60 years, had chronic kidney disease, or diabetes mellitus, then BP goal was <140/90 mm Hg.HEDIS criteria was applied if patients had multiple BPs taken on the same day, where the lowest systolic and lowest diastolic BPs were recorded and need not be from the same measurement set. Only outpatient office‐based BP measurements subsequent to the hypertension diagnosis date (index date) were used in our data.Although patient‐reported BPs during a virtual encounter do get recorded into the KP electronic medical record system, we excluded these BPs from the analysis because of risk of subjectivity. Only BPs that were recorded during office visits were used in this study. All face‐to‐face encounters during the follow‐up period were included in the analysis, whether or not BP measurements were taken on that date. We excluded any healthcare specialist visits or visits to the emergency department, urgent care, inpatient admissions, or procedures.Statistical analysisOutcomes analyzed included age, sex, race/ethnicity, body mass index, median neighborhood household income and education derived from US Census data, preferred spoken language (English, non‐English), baseline systolic and diastolic BPs, and Quan's adaptation of Charlson Comorbidity Index (CCI).Comparisons between patients who reached BP control and those who did not were made using standardized differences that estimate the difference between the mean or proportions in each group divided by the standard error. The standardized difference is a useful metric to identify potential risk factors between the two groups and an imbalance in the covariate distribution between the two groups is defined by an absolute value of the standardized difference being greater than or equal to 0.10, indicating at least a small effect size of the covariate. Crude and multivariable‐adjusted Cox proportional hazards models were developed to assess the effects of each significant covariate on BP control outcome status. Additionally, time to control was examined by developing crude and multivariable linear regression models among patients with controlled BP, which evaluated days of hypertensive period as the dependent variable. All models adjusted for the number of encounters in the hypertensive period, age at diagnosis, sex, race/ethnicity, median neighborhood household income, CCI, and the days between encounters.RESULTSA total of 338 756 patients with a new diagnosis of hypertension were identified in the medical record database during the study period. Of these, 242 799 did not meet eligibility criteria. The study cohort totaled 95 957, with 83 358 cases with controlled BP and 12 599 with uncontrolled BP (Figure ). A comparison of patient baseline characteristics for controlled and uncontrolled BP is presented in Table . Using an absolute value of the standardized difference cutoff of 0.10, patient's age at diagnosis, sex, race/ethnicity, neighborhood median household income, CCI, and baseline body mass index were considered significantly associated with BP control.Flow diagram of the study cohort population.Baseline demographic and clinical characteristics by BP control status in a cohort of patients with newly diagnosed hypertension—Kaiser Permanente Southern California 2008–2014Controlled BPStandardized differenceNo n = 12,599 (13.1%)Yes n = 83,358 (86.9%)Age at hypertension diagnosis, yMean (SD)49.3 (12.16)52.4 (12.36)0.2618–<402684 (21.3)11 896 (14.3)−0.1840–<607517 (59.7)48 219 (57.8)−0.0460–<802249 (17.9)21 764 (26.1)0.2080 y149 (1.2)1479 (1.8)0.05SexFemale5209 (41.3)40 972 (49.2)0.16Male7390 (58.7)42 386 (50.8)−0.16Race/ethnicityWhite, non‐Hispanic4017 (31.9)32 076 (38.5)0.14Black, non‐Hispanic1673 (13.3)9317 (11.2)−0.06Hispanic4521 (35.9)27 587 (33.1)−0.06Asian, non‐Hispanic1333 (10.6)9857 (11.8)0.04Other, non‐Hispanic1055 (8.4)4521 (5.4)−0.12Neighborhood household income, thousands US$Median (IQR)59.6 (43.9–78.9)62.2 (45.7–83.1)0.11Preferred spoken languageEnglish10 543 (83.7)69 864 (83.8)<0.01Non‐English2056 (16.3)13 494 (16.2)<0.01Charlson Comorbidity Index08974 (71.3)62 189 (74.7)0.081 or 23495 (27.8)20 091 (24.1)−0.083116 (0.9)1008 (1.2)0.03Baseline BP, mm HgSystolic, mean (SD)145.8 (12.60)145.0 (10.87)−0.06Diastolic, mean (SD)87.2 (10.09)85.6 (9.96)−0.16Baseline body mass index, kg/m2No.11 79179 755Mean (SD)32.0 (7.17)30.7 (6.75)−0.19Median (IQR)30.9 (27.1–35.7)29.6 (26.1–34.1)Prevalence of chronic kidney disease29 (0.2)248 (0.3)−0.01Prevalence of diabetes mellitus3845 (30.5)14 876 (17.8)−0.30Values are expressed as number (percentage) unless otherwise indicated. BP, blood pressure; IQR, interquartile range; SD, standard deviation.Using JNC 7 and JNC 8 criteria for values measured before and after January 1, 2014, respectively.Standardized difference is the difference in means or proportions divided by standard error; an imbalance is defined as the absolute value of the standardized difference ≥0.10 (indicating a small effect size).Follow‐up characteristics were also significantly associated with BP control status (Table ). On average, follow‐up was significantly longer in the uncontrolled patients at 292.9 (standard deviation 105.72) days vs 232.2 (standard deviation 137.35) days in those with controlled BP (standardized difference –0.5). The controlled BP group had significantly more encounters on average compared with the uncontrolled BP group (4.1 vs 3.1, standardized difference 0.33) during the hypertensive period. The median number of days between encounters was significantly lower for the controlled BP group compared with the uncontrolled BP group, respectively (29 vs 63, standardized difference −0.44). The first prescribed hypertensive medications did not differ between groups.Follow‐up characteristics by BP control status in a cohort of patients with newly diagnosed hypertension—Kaiser Permanente Southern California 2008–2014Controlled BPStandardized differenceNo( n = 12,599 (13.1%)Yes )n = 83,358 (86.9%)Days of follow‐upMean (SD)292.9 (105.72)232.2 (137.35)−0.501–<1484 (0.7)1848 (2.2)0.1314–<31265 (2.1)6055 (7.3)0.2531–<60426 (3.4)7610 (9.1)0.2460–<90349 (2.8)5348 (6.4)0.1790–<1801075 (8.5)11 726 (14.1)0.18180 d10 400 (82.5)50 771 (60.9)−0.50No. of encountersMean (SD)3.1 (2.41)4.1 (3.07)0.331 or 26228 (49.4)27 648 (33.2)−0.333 or 43908 (31)30 214 (36.2)0.115 or 61468 (11.7)13 984 (16.8)0.157995 (7.9)11 512 (13.8)0.19Average days between encountersMean (SD)101.2 (99.17)62.5 (75.28)−0.44Median (IQR)63.0 (19.0–162.0)29.0 (13.0–86.0)1–<142123 (16.9)22 208 (26.6)0.2414–<312264 (18)20 736 (24.9)0.1731–<601748 (13.9)13 029 (15.6)0.0560–<901201 (9.5)7226 (8.7)−0.0390–<1802468 (19.6)11 819 (14.2)−0.14180 d2795 (22.2)8340 (10)−0.34BP at control, mm HgSystolic, mean (SD)N/A126.9 (9.13)—Diastolic, mean (SD)N/A76.1 (8.22)—First prescribed hypertension medicationβ‐Blocker970 (7.7)7193 (8.6)<0.01Calcium channel blockers860 (6.8)6016 (7.2)0.02ACEIs/angiotensin II receptor antagonists2944 (23.4)16 563 (19.9)<0.01Andrenolytics93 (0.7)584 (0.7)<0.01Vasodilators1 (0)23 (0)0.01Diuretics2349 (18.6)16 470 (19.8)0.03Antihypertensive Combinations5382 (42.7)36 509 (43.8)0.02Values are expressed as number (percentage) unless otherwise indicated. ACEIs, angiotensin‐converting enzyme inhibitors; IQR, interquartile range; SD, standard deviation.Using Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) and Eighth Joint National Committee (JNC 8) criteria for values measured before and after January 1, 2014, respectively.Standardized difference is the difference in means or proportions divided by standard error; an imbalance is defined as the absolute value of the standardized difference >0.10 (indicating a small effect size).Follow‐up was measured from hypertension diagnosis through the earliest of: loss of Kaiser Permanente membership, death, blood pressure (BP) control, or 1 year from diagnosis.Table  presents crude and multivariable‐adjusted Cox proportional hazards ratios (HRs) with 95% confidence interval (CI) limits for BP control as the outcome. The results demonstrate a significant linear trend between encounter interval and achieving BP goals. As the number of days increased between encounters from the1 to  < 14 days (reference), there was a consistently lower likelihood of achieving BP control within 1 year of follow‐up. When the encounter interval was ≥180 days, it was associated with the lowest likelihood of achieving BP control compared with 1 to  < 14 days (HR, 0.35; 95% CI, 0.34–0.36). Figure  shows a Kaplan‐Meier curve of patient time in the hypertensive period. All patients began in the hypertensive period and dropped off as they achieved BP control. The graph showed that time to control rose progressively as the encounter interval increased. The curve also clearly demonstrated that this effect was consistent throughout the follow‐up period.Crude and multivariable‐adjusted hazard ratios with 95% confidence interval limits for blood pressure controlCrudeMultivariableNo. of encounters1 or 2ReferenceReference3 or 41.40 (1.38–1.43)1.24 (1.22–1.27)5 or 61.28 (1.26–1.31)1.02 (1.00–1.05)71.11 (1.08–1.13)0.78 (0.76–0.80)Days between encounters1–<14ReferenceReference14–<310.83 (0.81–0.84)0.82 (0.80–0.84)31–<600.67 (0.66–0.69)0.64 (0.63–0.66)60–<900.58 (0.57–0.60)0.55 (0.53–0.56)90–<1800.49 (0.48–0.50)0.45 (0.44–0.46)180 d0.38 (0.37–0.39)0.35 (0.34–0.36)SociodemographicsAge at diagnosis (5‐y increment)1.02 (1.02–1.02)1.02 (1.01–1.02)Male vs female sex0.94 (0.92–0.95)0.96 (0.94–0.97)Race/ethnicityWhite, non‐HispanicReferenceReferenceBlack, non‐Hispanic1.01 (0.99–1.04)1.01 (0.99–1.04)Hispanic0.98 (0.97–1.00)0.98 (0.96–1.00)Asian, non‐Hispanic0.93 (0.91–0.95)0.95 (0.92–0.97)Other, non‐Hispanic0.96 (0.93–0.99)0.98 (0.94–1.01)Neighborhood household income, thousands US$ (10,000 US$ increment)1.00 (0.99–1.00)1.00 (0.99–1.00)Charlson Comorbidity Index0ReferenceReference1 or 21.10 (1.08–1.11)1.10 (1.08–1.12)3 or 41.50 (1.41–1.60)1.47 (1.37–1.57)Baseline body mass index (per 5‐unit increment)1.00 (1.00–1.01)1.00 (1.00–1.01)Kaplan‐Meier curve of time to blood pressure control by mean encounter interval length. Shown is a Kaplan‐Meier curve with 95% Hall‐Wellner bands comparing different encounter interval length (mean days between encounters) groups from the first hypertension diagnosis to the first blood pressure measurement at goal. Encounter interval length (d) groups are: 0 to <14 (green), 14 to <31 (blue), 31 to <60 (purple), 60 to <90 (yellow), 90 to <180 (black), and ≥180 (red). Each incremental encounter interval group achieved less blood pressure control at a slower rate consistently throughout the study follow‐up period.Compared with those with only one or two encounters (reference) during the hypertensive period, patients with three or four encounters (HR, 1.24; 95% CI, 1.22–1.27) were significantly more likely to reach BP control. Patients with five or six encounters were not significantly different compared with the reference. Patients with seven or more encounters were significantly less likely to reach control (HR, 0.78; 95% CI, 0.76–0.80).Men were less likely than women to reach BP control during follow‐up (HR, 0.96; 95% CI, 0.94–0.97). Compared with whites (reference), Asians were found to have a significantly lower likelihood of achieving BP control within 1 year (HR, 0.95; 95% CI, 0.92–0.97). All other race associations were found not to be statistically significantly. In addition, as the CCI increased from zero as the reference, there was a significantly higher likelihood of BP control within 1 year.Table  presents crude and multivariable linear regression model estimates and standard errors for days of the hypertensive period as an outcome among patients with controlled BP only. The results demonstrate a significant linear trend between encounter interval and time to control. As the number of days increased between encounters from 1 to  < 14 days (reference), it took patients more days to achieve BP control. When the encounter interval was ≥180 days, the hypertensive period lasted 172 days longer than the reference 1 to  < 14 days (P < .001).Crude and multivariable linear estimates and SE for time to BP controlCrudeMultivariableEstimate (SE)P valueEstimate (SE)P valueNo. of encounters1 or 2ReferenceReference3 or 4–25.25 (1.13)<.0001−6.69 (1.11)<.00015 or 66.08 (1.41)<.000134.56 (1.37)<.0001742.47 (1.5)<.000186.04 (1.48)<.0001Days between encounters1–<14ReferenceReference14–<3129.25 (1.23)<.000130.4 (1.26)<.000131–<6065.92 (1.4)<.000169.48 (1.43)<.000160–<9091.75 (1.72)<.000196.52 (1.75)<.000190–<180120.48 (1.45)<.0001129.78 (1.48)<.0001180 d154.08 (1.63)<.0001171.8 (1.72)<.0001SociodemographicsAge at diagnosis (5‐y increment)2.07 (83 286).05−0.33 (0.04)<.0001Male vs female sex2.88 (0.95)<.00012.41 (0.91).0082Race/ethnicityWhite, non‐HispanicReferenceReferenceBlack, non‐Hispanic–8.66 (1.62)<.0001−4.56 (1.55).0033Hispanic–1.63 (1.13).152.15 (1.12).0555Asian, non‐Hispanic12.79 (1.58)<.00018.99 (1.54)<.0001Other, non‐Hispanic–5 (2.18).02−1.55 (2.12).4646Neighborhood household income, thousands US$ (10,000 US$ increment)0.02 (77 051).050.1 (0.02)<.0001Charlson Comorbidity Index0ReferenceReference1 or 2–18.53 (1.11)<.0001−19.22 (1.06)<.00013 or 4–59.5 (4.35)<.0001−53.96 (4.09)<.0001Baseline body mass index (per 5‐unit increment)0.07 (79 683).05−0.47 (0.07)<.0001BP, blood pressure; SE, standard error.Compared with those with only one or two encounters (reference) during the hypertensive period, patients with three or four encounters had seven fewer hypertensive period days (ie, faster time to control). Patients with five or six encounters had 35 additional hypertensive period days. Patients with seven or more encounters had 86 additional hypertensive period days (all P values <.001).DISCUSSIONThis study identified that patients with newly diagnosed hypertension who had their BP controlled within the first year of diagnosis had an average of about four encounters. Of those achieving BP goals, the mean days between encounters was 63 and the median was 29. This study found that an encounter interval of less than 2 weeks was associated with a higher likelihood of having BP control. Additionally, fewer days between encounters was also found to be associated with significantly faster time to BP control.The findings of this study are consistent with previously published studies evaluating the number and frequency of encounters needed to control BP. A study by Turchin and colleagues examined whether hypertension encounter intervals between primary care physician visits led to faster normalization of BP in patients with diabetes mellitus. It found that time to normalization was significantly shorter (0.7 months vs 1.9 months) when the interval was ≤2 weeks vs between 2 weeks and 1 month, respectively. Morrison and colleagues found that median time to goal BP of 130/85 mm Hg was 1.3 months vs 13.9 months when encounters with physicians were between 1 and 2 weeks vs 3 to 6 months in patients with diabetes mellitus, respectively. A recently published retrospective nested case‐control study by Low and colleagues that sought to identify risk factors associated with patients with continuously uncontrolled hypertension found that patients who had a greater average number of walk‐in BP visits per year and greater average number of primary care physician visits per year were associated with controlled hypertension. This study added to the understanding by examining all outpatient visits associated with hypertension.Two previous studies conducted with patients with diabetes mellitus and hypertension found that similar encounter interval lengths were associated with faster time to BP control. Interestingly, even though our study only allowed participants to contribute their very first hypertensive period after diagnosis for analysis, it had similar results as the aforementioned studies that allowed participants to contribute multiple hypertensive periods over the course of the study period. Although current hypertension guidelines recommend monthly follow‐up until BP is controlled, based on the results from this study and previous studies, patients should be seen at least every 2 weeks as it was associated with higher likelihood of BP control and faster time to control.The reasons shorter encounter intervals led to a higher likelihood of achieving BP targets in a shorter amount of time was likely to be multifactorial. One likely explanation is that as the interval between encounters increases, clinical inertia may become more apparent. Clinical inertia is defined as failure by healthcare providers to appropriately initiate or intensify therapy during visits with patients. Intuitively, the more often a patient is seen over a shorter period of time, the more opportunities a provider has to appropriately adjust therapies and order appropriate labs. Another possibility is that as encounters become more spread out, especially for patients with uncontrolled BP on multiple medications, medication adherence may become of bigger concern. According to previous studies looking at risk factors for uncontrolled hypertension, medication adherence was found to be associated with BP control. In addition, increased encounter frequency has been linked to improved treatment adherence. When encounters are spaced more closely together, the healthcare professional has more opportunities and better follow‐up to assess medication adherence, potential side effects, and disease state understanding. It is also important to consider that most antihypertensive medications, except for thiazides, achieve most of their effect in less than 2 weeks. Having a patient come in biweekly allows providers to assess whether the current antihypertensive medication dose is appropriate and can adjust as needed until BP is controlled.There were a number of factors that were associated with a greater likelihood of BP control. As CCI increased, there was a significantly higher likelihood of BP control. Patients with a higher CCI have a higher rate of mortality within 1 year depending on the comorbid condition. These individuals may indirectly require closer follow‐up with tight medication management and adherence. Additionally, being a woman and older at hypertension diagnosis was associated with a higher likelihood of BP control.There were a few interesting findings that were associated with a lower likelihood of achieving BP control. Previous studies indicated that there was a significantly higher risk of uncontrolled hypertension in blacks compared with whites. The current study found the same finding; however, this was not shown to be statistically significant. Interestingly, the linear regression shows that blacks who did achieve BP control did so in a faster manner compared with whites. Additionally, Asians were found to be less likely to achieve BP control and those who did achieve BP control did so in a slower manner than whites. Further studies need to be conducted to examine relationships between race and the likelihood and time to achieve BP control.Compared with one or two encounters as reference, the likelihood of achieving BP control was significantly higher among patients who had three or four encounters but was significantly lower for patients who had seven or more encounters. Furthermore, patients who had three or four encounters were significantly found to reach BP control faster compared with patients with one or two encounters (reference). Consistently, patients who had five or more encounters were found to reach BP control significantly slower compared with those with one or two encounters. These findings suggest that there may be an optimal number of encounters to benefit patients with hypertension. Some previous studies found that increased frequency of encounters was associated with better control of chronic diseases.Study strengthsThis study had a number of strengths, including a large patient population that reflected communities within Southern California. We derived our data using a prospectively collected electronic medical database including pharmacy prescription data. Moreover, a large number of variables were investigated to explore multiple associations with the number of encounters to BP control. Although the type of hypertensive medication did not differ between groups, an exploratory analysis was conducted to add the hypertensive medication class used as an adjustment variable. Adjusting for medication class had negligible effects on all of the results. This study also used accepted national measures via the HEDIS criteria to assess BP control. Additionally, this study used JNC 7 and JNC 8 BP targets in order to accurately reflect real‐world clinical practice and goals at time of care.Study limitationsThis study's limitations are those characteristic of retrospective observational designs, including the inability to control for confounding variables that are unknown or unmeasurable. This study was dependent on International Classification of Diseases––Ninth Revision codes for the diagnosis of hypertension, which may lead to misclassification bias. In addition, because of the retrospective nature of this study, it was not possible to determine causation, and information and selection bias are also possibilities. The use of HEDIS criteria to define BP control using clinic BP recordings from a single day may lead to an inflated rate of control since SBP and DBP could come from different BP sets from the same day. Future studies exploring a more stringent definition for control and examining time in control may be valuable. Also, there is potential for patients to have multiple hypertensive periods, but this study focused solely on patients with newly controlled hypertension. Another limitation of this study is that we did not explore medication adherence as this study used hypertension medications for the purpose of identifying patients with hypertension, descriptive clinical characteristics, and adjustment in multivariate analysis. Last, this study was limited to insured patients with 1 year of continuous coverage, which may limit generalizability.CONCLUSIONSThis study found that shorter encounter intervals lead to higher likelihood of achieving BP control in a shorter amount of time in patients with newly diagnosed hypertension who received pharmacological therapy. The findings of this study suggest that consideration should be placed on updating current hypertension guidelines to have patients with newly diagnosed hypertension followed up at least biweekly once a BP medication is initiated until BP is controlled. Additional studies are needed to better understand how patient characteristics and specific therapies may be benefited by frequency and type of encounters.DISCLOSURES AND CONFLICT OF INTERESTMB is a former Kaiser employee. All other authors report no conflicts of interest and have nothing to disclose.REFERENCESJames PA, Oparil S, Carter BL. 2014 Evidence‐based guideline for the management of high BP in adults. Report from panel members appointed to the eighth joint national committee (JNC 8). JAMA. 2014;311:507‐520.Elperin DT, Pelter MA, Deamer RL, et al. A large cohort study evaluating risk factors associated with uncontrolled hypertension. J Clin Hypertens (Greenwich). 2014;16:149‐154.Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics—2015 update: a report from the American Heart Association. Circulation. 2015;131:e29‐e322.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.Shaw KM, Handler J, Wall HK. Improving BP control in a large multiethnic California population through changes in health care delivery 2004–2012. Prev Chronic Dis. 2014;11:140173.Low KJ, Pelter MA, Deamer RL, et al. Identification and evaluation of risk factors in patients with continuously uncontrolled hypertension. J Clin Hypertens (Greenwich). 2015;17:281‐289.Sim JJ, Handler J, Jacobsen SJ, et al. Systematic implementation strategies to improve hypertension: the Kaiser Permanente Southern California experience. Can J Cardiol. 2014;30:544‐552.National Committee for Quality Assurance. HEDIS & Performance Measurement. Washington, DC: National Committee for Quality Assurance; 2014. http://www.ncqa.org/hedisqualitymeasurement.aspx. Accessed Septeber 12, 2015.Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD‐9‐CM and ICD‐10 administrative data. Med Care. 2005;43:1130‐1139.Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373‐383.Yang D, Dalton JE. A unified approach to measuring the effect size between two groups using SAS®. http://support.sas.com/resources/papers/proceedings12/335-2012.pdf. Accessed January 23, 2018.Turchin A, Goldberg SI, Shubia M, et al. Encounter frequency and BP in hypertensive patients with diabetes mellitus. Hypertension. 2010;56:68‐74.Morrison A, Shubina M, Turchin A. Encounter frequency and serum glucose level, BP, and cholesterol level control in patients with diabetes mellitus. Arch Intern Med. 2011;171:1542‐1550.Guthmann R, Davis N, Brown M, et al. Visit frequency and hypertension. J Clin Hypertens (Greenwich). 2005;7:327‐332.Lebeau JP, Cadwallader JS, Aubin‐Auger I, et al. The concept and definition of therapeutic inertia in hypertension in primary care: a qualitative systematic review. BMC Family Practice. 2014;12:130.Patel NC, Crismon ML, Miller AL, et al. Drug adherence: effects of decreased visit frequency on adherence to clozapine therapy. Pharmacotherapy. 2005;25:1242‐1247.Plantinga LC, Fink NE, Sadler JH, et al. Frequency of patient‐physician contact and patient outcomes in hemodialysis care. J Am Soc Nephrol. 2004;15:210‐218.Asao K, McEwen LN, Crosson JC, et al. Revisit frequency and its association with quality of care among diabetic patients: translating research into action for diabetes (TRIAD). J Diabetes Complications. 2014;28:811‐818.Redfern J, Menzies M, Briffa T, et al. Impact of medical consultation frequency on modifiable risk factors and medications at 12 months after acute coronary syndrome in the CHOICE randomized controlled trial. Int J Cardiol. 2010;145:481‐486.Koebnick C, Langer‐Gould A, Gould M, et al. Sociodemographic characteristics of members of a large, integrated health care system: comparison with US Census Bureau data. Perm J. 2012;16:37‐41.AAPPENDIXNumber of patients censored by categoryCensorFrequencyPercentCumulative frequencyCumulative percent1_year_since_incident_date602747.84602747.84Death270.21605448.05Hypertensive_period_end562444.6411 67892.69Membership end9217.3112 599100 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Clinical Hypertension Wiley

Association between encounter frequency and time to blood pressure control among patients with newly diagnosed hypertension: a retrospective cohort study

Download PDF
9 pages

Loading next page...
1
 
/lp/wiley/association-between-encounter-frequency-and-time-to-blood-pressure-ctSPJ8jxYS

References (24)

Publisher
Wiley
Copyright
Copyright © 2018 Wiley Periodicals, Inc.
ISSN
1524-6175
eISSN
1751-7176
DOI
10.1111/jch.13223
Publisher site
See Article on Publisher Site

Abstract

INTRODUCTIONHypertension is the most common condition seen in primary care today. An estimated 80 million adults older than 20 years currently have a diagnosis of hypertension. This represents 38.9 million physician office visits and 3.6 million outpatient department visits annually. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) guidelines recommend that once antihypertensive medications are started, patients should return for follow‐up at monthly intervals until blood pressure (BP) goals are reached and then every 3 to 6 months once stable. The 2014 Evidence‐Based Guideline for the Management of High BP in Adults from the panel members of the Eighth Joint National Committee (JNC 8) does not provide follow‐up intervals, but does recommend that if BP goals are not reached within a month of starting treatment, an increase in the initial medication should be made or a second medication from one of the recommended classes should be started.Kaiser Permanente (KP) Southern California is a large, integrated healthcare system that delivers care to approximately 4.1 million members. Since 2004, KP has implemented a series of changes to improve the treatment and identification of hypertension, including: (1) creation of a hypertension registry, (2) BP measurement standardization, (3) the use of an evidence‐based and easy‐to‐remember treatment algorithm, and (4) a multidisciplinary approach to care, which includes medical assistants, nurses, and pharmacists with shared responsibilities in BP management. According to the Healthcare Effectiveness Data and Information Set (HEDIS) from 2004 to 2012 assessing BP control at KP, overall hypertension control increased from 54% to 86%.The main objective of this study was to examine and describe patients with newly diagnosed hypertension and determine whether the frequency and intervals between outpatient encounters were associated with achieving BP control. In addition, this study examined the potential risk factors associated with the time to BP control in the first year since hypertension diagnosis.METHODOLOGYIn this retrospective, observational, cohort study, we identified patients with newly diagnosed hypertension at KP Southern California between January 1, 2008, and December 31, 2014. We identified patients using electronic health records that capture sociodemographic, medical encounters, and pharmacy information for all patient care provided by KP. KP's prescription drug database was used as the source of all prescription data. This data set provided detailed information regarding the patients’ prescriptions including date written, dispensed date, refill dates, quantities dispensed, and quantity of the prescription written.Patients included in the study were aged at least 18 years with an International Classification of Diseases—Ninth Revision diagnosis code for hypertension and an antihypertensive medication filled within the 1‐year period after that diagnosis. The study inclusion criteria relative to that date were: continuous membership with the health plan for at least 1 year before diagnosis with no more than a 45‐day gap in coverage allowed, and a standard pharmacy benefit. Patients were excluded if they did not have a valid BP measurement within 1 year after hypertension diagnosis, were pregnant at any point during the study period, or if a BP medication was dispensed before the hypertension diagnosis date. The index date (ie, first day of follow‐up) was the first recorded elevated BP date on or after the date of hypertension diagnosis.Clinical outcomesBP measurements from outpatient encounters were examined to define a “hypertensive period”: the period of time between the index date and the first subsequent date at which BP was at the JNC 7 or JNC 8 targets. Essentially, the hypertensive period is the same as the follow‐up period, meaning that follow‐up stopped when patients achieved BP control. Clinical outcomes were measured during the hypertensive period, which was limited to 1 year follow‐up after the index date. Within that 1 year of follow‐up, participants were considered to have BP control if they reached their BP targets per guidelines. Patients were censored if their BP remained uncontrolled (ie, if their BP remained above goal), died, disenrolled from the health plan, or reached the end of the follow‐up period of December 31, 2014, whichever occurred first. See appendix A for number of patients censored by category.The JNC 7 BP targets were used for the study period from January 1, 2008, through December 31, 2013: BP goal was <140/90 mm Hg unless patients had chronic kidney disease or diabetes mellitus in which the BP goal was <130/80 mm Hg. The JNC 8 BP targets were used for the study period on and after January 1, 2014: if patients were 60 years and older, the BP goal was <150/90 mm Hg and if patients were younger than 60 years, had chronic kidney disease, or diabetes mellitus, then BP goal was <140/90 mm Hg.HEDIS criteria was applied if patients had multiple BPs taken on the same day, where the lowest systolic and lowest diastolic BPs were recorded and need not be from the same measurement set. Only outpatient office‐based BP measurements subsequent to the hypertension diagnosis date (index date) were used in our data.Although patient‐reported BPs during a virtual encounter do get recorded into the KP electronic medical record system, we excluded these BPs from the analysis because of risk of subjectivity. Only BPs that were recorded during office visits were used in this study. All face‐to‐face encounters during the follow‐up period were included in the analysis, whether or not BP measurements were taken on that date. We excluded any healthcare specialist visits or visits to the emergency department, urgent care, inpatient admissions, or procedures.Statistical analysisOutcomes analyzed included age, sex, race/ethnicity, body mass index, median neighborhood household income and education derived from US Census data, preferred spoken language (English, non‐English), baseline systolic and diastolic BPs, and Quan's adaptation of Charlson Comorbidity Index (CCI).Comparisons between patients who reached BP control and those who did not were made using standardized differences that estimate the difference between the mean or proportions in each group divided by the standard error. The standardized difference is a useful metric to identify potential risk factors between the two groups and an imbalance in the covariate distribution between the two groups is defined by an absolute value of the standardized difference being greater than or equal to 0.10, indicating at least a small effect size of the covariate. Crude and multivariable‐adjusted Cox proportional hazards models were developed to assess the effects of each significant covariate on BP control outcome status. Additionally, time to control was examined by developing crude and multivariable linear regression models among patients with controlled BP, which evaluated days of hypertensive period as the dependent variable. All models adjusted for the number of encounters in the hypertensive period, age at diagnosis, sex, race/ethnicity, median neighborhood household income, CCI, and the days between encounters.RESULTSA total of 338 756 patients with a new diagnosis of hypertension were identified in the medical record database during the study period. Of these, 242 799 did not meet eligibility criteria. The study cohort totaled 95 957, with 83 358 cases with controlled BP and 12 599 with uncontrolled BP (Figure ). A comparison of patient baseline characteristics for controlled and uncontrolled BP is presented in Table . Using an absolute value of the standardized difference cutoff of 0.10, patient's age at diagnosis, sex, race/ethnicity, neighborhood median household income, CCI, and baseline body mass index were considered significantly associated with BP control.Flow diagram of the study cohort population.Baseline demographic and clinical characteristics by BP control status in a cohort of patients with newly diagnosed hypertension—Kaiser Permanente Southern California 2008–2014Controlled BPStandardized differenceNo n = 12,599 (13.1%)Yes n = 83,358 (86.9%)Age at hypertension diagnosis, yMean (SD)49.3 (12.16)52.4 (12.36)0.2618–<402684 (21.3)11 896 (14.3)−0.1840–<607517 (59.7)48 219 (57.8)−0.0460–<802249 (17.9)21 764 (26.1)0.2080 y149 (1.2)1479 (1.8)0.05SexFemale5209 (41.3)40 972 (49.2)0.16Male7390 (58.7)42 386 (50.8)−0.16Race/ethnicityWhite, non‐Hispanic4017 (31.9)32 076 (38.5)0.14Black, non‐Hispanic1673 (13.3)9317 (11.2)−0.06Hispanic4521 (35.9)27 587 (33.1)−0.06Asian, non‐Hispanic1333 (10.6)9857 (11.8)0.04Other, non‐Hispanic1055 (8.4)4521 (5.4)−0.12Neighborhood household income, thousands US$Median (IQR)59.6 (43.9–78.9)62.2 (45.7–83.1)0.11Preferred spoken languageEnglish10 543 (83.7)69 864 (83.8)<0.01Non‐English2056 (16.3)13 494 (16.2)<0.01Charlson Comorbidity Index08974 (71.3)62 189 (74.7)0.081 or 23495 (27.8)20 091 (24.1)−0.083116 (0.9)1008 (1.2)0.03Baseline BP, mm HgSystolic, mean (SD)145.8 (12.60)145.0 (10.87)−0.06Diastolic, mean (SD)87.2 (10.09)85.6 (9.96)−0.16Baseline body mass index, kg/m2No.11 79179 755Mean (SD)32.0 (7.17)30.7 (6.75)−0.19Median (IQR)30.9 (27.1–35.7)29.6 (26.1–34.1)Prevalence of chronic kidney disease29 (0.2)248 (0.3)−0.01Prevalence of diabetes mellitus3845 (30.5)14 876 (17.8)−0.30Values are expressed as number (percentage) unless otherwise indicated. BP, blood pressure; IQR, interquartile range; SD, standard deviation.Using JNC 7 and JNC 8 criteria for values measured before and after January 1, 2014, respectively.Standardized difference is the difference in means or proportions divided by standard error; an imbalance is defined as the absolute value of the standardized difference ≥0.10 (indicating a small effect size).Follow‐up characteristics were also significantly associated with BP control status (Table ). On average, follow‐up was significantly longer in the uncontrolled patients at 292.9 (standard deviation 105.72) days vs 232.2 (standard deviation 137.35) days in those with controlled BP (standardized difference –0.5). The controlled BP group had significantly more encounters on average compared with the uncontrolled BP group (4.1 vs 3.1, standardized difference 0.33) during the hypertensive period. The median number of days between encounters was significantly lower for the controlled BP group compared with the uncontrolled BP group, respectively (29 vs 63, standardized difference −0.44). The first prescribed hypertensive medications did not differ between groups.Follow‐up characteristics by BP control status in a cohort of patients with newly diagnosed hypertension—Kaiser Permanente Southern California 2008–2014Controlled BPStandardized differenceNo( n = 12,599 (13.1%)Yes )n = 83,358 (86.9%)Days of follow‐upMean (SD)292.9 (105.72)232.2 (137.35)−0.501–<1484 (0.7)1848 (2.2)0.1314–<31265 (2.1)6055 (7.3)0.2531–<60426 (3.4)7610 (9.1)0.2460–<90349 (2.8)5348 (6.4)0.1790–<1801075 (8.5)11 726 (14.1)0.18180 d10 400 (82.5)50 771 (60.9)−0.50No. of encountersMean (SD)3.1 (2.41)4.1 (3.07)0.331 or 26228 (49.4)27 648 (33.2)−0.333 or 43908 (31)30 214 (36.2)0.115 or 61468 (11.7)13 984 (16.8)0.157995 (7.9)11 512 (13.8)0.19Average days between encountersMean (SD)101.2 (99.17)62.5 (75.28)−0.44Median (IQR)63.0 (19.0–162.0)29.0 (13.0–86.0)1–<142123 (16.9)22 208 (26.6)0.2414–<312264 (18)20 736 (24.9)0.1731–<601748 (13.9)13 029 (15.6)0.0560–<901201 (9.5)7226 (8.7)−0.0390–<1802468 (19.6)11 819 (14.2)−0.14180 d2795 (22.2)8340 (10)−0.34BP at control, mm HgSystolic, mean (SD)N/A126.9 (9.13)—Diastolic, mean (SD)N/A76.1 (8.22)—First prescribed hypertension medicationβ‐Blocker970 (7.7)7193 (8.6)<0.01Calcium channel blockers860 (6.8)6016 (7.2)0.02ACEIs/angiotensin II receptor antagonists2944 (23.4)16 563 (19.9)<0.01Andrenolytics93 (0.7)584 (0.7)<0.01Vasodilators1 (0)23 (0)0.01Diuretics2349 (18.6)16 470 (19.8)0.03Antihypertensive Combinations5382 (42.7)36 509 (43.8)0.02Values are expressed as number (percentage) unless otherwise indicated. ACEIs, angiotensin‐converting enzyme inhibitors; IQR, interquartile range; SD, standard deviation.Using Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) and Eighth Joint National Committee (JNC 8) criteria for values measured before and after January 1, 2014, respectively.Standardized difference is the difference in means or proportions divided by standard error; an imbalance is defined as the absolute value of the standardized difference >0.10 (indicating a small effect size).Follow‐up was measured from hypertension diagnosis through the earliest of: loss of Kaiser Permanente membership, death, blood pressure (BP) control, or 1 year from diagnosis.Table  presents crude and multivariable‐adjusted Cox proportional hazards ratios (HRs) with 95% confidence interval (CI) limits for BP control as the outcome. The results demonstrate a significant linear trend between encounter interval and achieving BP goals. As the number of days increased between encounters from the1 to  < 14 days (reference), there was a consistently lower likelihood of achieving BP control within 1 year of follow‐up. When the encounter interval was ≥180 days, it was associated with the lowest likelihood of achieving BP control compared with 1 to  < 14 days (HR, 0.35; 95% CI, 0.34–0.36). Figure  shows a Kaplan‐Meier curve of patient time in the hypertensive period. All patients began in the hypertensive period and dropped off as they achieved BP control. The graph showed that time to control rose progressively as the encounter interval increased. The curve also clearly demonstrated that this effect was consistent throughout the follow‐up period.Crude and multivariable‐adjusted hazard ratios with 95% confidence interval limits for blood pressure controlCrudeMultivariableNo. of encounters1 or 2ReferenceReference3 or 41.40 (1.38–1.43)1.24 (1.22–1.27)5 or 61.28 (1.26–1.31)1.02 (1.00–1.05)71.11 (1.08–1.13)0.78 (0.76–0.80)Days between encounters1–<14ReferenceReference14–<310.83 (0.81–0.84)0.82 (0.80–0.84)31–<600.67 (0.66–0.69)0.64 (0.63–0.66)60–<900.58 (0.57–0.60)0.55 (0.53–0.56)90–<1800.49 (0.48–0.50)0.45 (0.44–0.46)180 d0.38 (0.37–0.39)0.35 (0.34–0.36)SociodemographicsAge at diagnosis (5‐y increment)1.02 (1.02–1.02)1.02 (1.01–1.02)Male vs female sex0.94 (0.92–0.95)0.96 (0.94–0.97)Race/ethnicityWhite, non‐HispanicReferenceReferenceBlack, non‐Hispanic1.01 (0.99–1.04)1.01 (0.99–1.04)Hispanic0.98 (0.97–1.00)0.98 (0.96–1.00)Asian, non‐Hispanic0.93 (0.91–0.95)0.95 (0.92–0.97)Other, non‐Hispanic0.96 (0.93–0.99)0.98 (0.94–1.01)Neighborhood household income, thousands US$ (10,000 US$ increment)1.00 (0.99–1.00)1.00 (0.99–1.00)Charlson Comorbidity Index0ReferenceReference1 or 21.10 (1.08–1.11)1.10 (1.08–1.12)3 or 41.50 (1.41–1.60)1.47 (1.37–1.57)Baseline body mass index (per 5‐unit increment)1.00 (1.00–1.01)1.00 (1.00–1.01)Kaplan‐Meier curve of time to blood pressure control by mean encounter interval length. Shown is a Kaplan‐Meier curve with 95% Hall‐Wellner bands comparing different encounter interval length (mean days between encounters) groups from the first hypertension diagnosis to the first blood pressure measurement at goal. Encounter interval length (d) groups are: 0 to <14 (green), 14 to <31 (blue), 31 to <60 (purple), 60 to <90 (yellow), 90 to <180 (black), and ≥180 (red). Each incremental encounter interval group achieved less blood pressure control at a slower rate consistently throughout the study follow‐up period.Compared with those with only one or two encounters (reference) during the hypertensive period, patients with three or four encounters (HR, 1.24; 95% CI, 1.22–1.27) were significantly more likely to reach BP control. Patients with five or six encounters were not significantly different compared with the reference. Patients with seven or more encounters were significantly less likely to reach control (HR, 0.78; 95% CI, 0.76–0.80).Men were less likely than women to reach BP control during follow‐up (HR, 0.96; 95% CI, 0.94–0.97). Compared with whites (reference), Asians were found to have a significantly lower likelihood of achieving BP control within 1 year (HR, 0.95; 95% CI, 0.92–0.97). All other race associations were found not to be statistically significantly. In addition, as the CCI increased from zero as the reference, there was a significantly higher likelihood of BP control within 1 year.Table  presents crude and multivariable linear regression model estimates and standard errors for days of the hypertensive period as an outcome among patients with controlled BP only. The results demonstrate a significant linear trend between encounter interval and time to control. As the number of days increased between encounters from 1 to  < 14 days (reference), it took patients more days to achieve BP control. When the encounter interval was ≥180 days, the hypertensive period lasted 172 days longer than the reference 1 to  < 14 days (P < .001).Crude and multivariable linear estimates and SE for time to BP controlCrudeMultivariableEstimate (SE)P valueEstimate (SE)P valueNo. of encounters1 or 2ReferenceReference3 or 4–25.25 (1.13)<.0001−6.69 (1.11)<.00015 or 66.08 (1.41)<.000134.56 (1.37)<.0001742.47 (1.5)<.000186.04 (1.48)<.0001Days between encounters1–<14ReferenceReference14–<3129.25 (1.23)<.000130.4 (1.26)<.000131–<6065.92 (1.4)<.000169.48 (1.43)<.000160–<9091.75 (1.72)<.000196.52 (1.75)<.000190–<180120.48 (1.45)<.0001129.78 (1.48)<.0001180 d154.08 (1.63)<.0001171.8 (1.72)<.0001SociodemographicsAge at diagnosis (5‐y increment)2.07 (83 286).05−0.33 (0.04)<.0001Male vs female sex2.88 (0.95)<.00012.41 (0.91).0082Race/ethnicityWhite, non‐HispanicReferenceReferenceBlack, non‐Hispanic–8.66 (1.62)<.0001−4.56 (1.55).0033Hispanic–1.63 (1.13).152.15 (1.12).0555Asian, non‐Hispanic12.79 (1.58)<.00018.99 (1.54)<.0001Other, non‐Hispanic–5 (2.18).02−1.55 (2.12).4646Neighborhood household income, thousands US$ (10,000 US$ increment)0.02 (77 051).050.1 (0.02)<.0001Charlson Comorbidity Index0ReferenceReference1 or 2–18.53 (1.11)<.0001−19.22 (1.06)<.00013 or 4–59.5 (4.35)<.0001−53.96 (4.09)<.0001Baseline body mass index (per 5‐unit increment)0.07 (79 683).05−0.47 (0.07)<.0001BP, blood pressure; SE, standard error.Compared with those with only one or two encounters (reference) during the hypertensive period, patients with three or four encounters had seven fewer hypertensive period days (ie, faster time to control). Patients with five or six encounters had 35 additional hypertensive period days. Patients with seven or more encounters had 86 additional hypertensive period days (all P values <.001).DISCUSSIONThis study identified that patients with newly diagnosed hypertension who had their BP controlled within the first year of diagnosis had an average of about four encounters. Of those achieving BP goals, the mean days between encounters was 63 and the median was 29. This study found that an encounter interval of less than 2 weeks was associated with a higher likelihood of having BP control. Additionally, fewer days between encounters was also found to be associated with significantly faster time to BP control.The findings of this study are consistent with previously published studies evaluating the number and frequency of encounters needed to control BP. A study by Turchin and colleagues examined whether hypertension encounter intervals between primary care physician visits led to faster normalization of BP in patients with diabetes mellitus. It found that time to normalization was significantly shorter (0.7 months vs 1.9 months) when the interval was ≤2 weeks vs between 2 weeks and 1 month, respectively. Morrison and colleagues found that median time to goal BP of 130/85 mm Hg was 1.3 months vs 13.9 months when encounters with physicians were between 1 and 2 weeks vs 3 to 6 months in patients with diabetes mellitus, respectively. A recently published retrospective nested case‐control study by Low and colleagues that sought to identify risk factors associated with patients with continuously uncontrolled hypertension found that patients who had a greater average number of walk‐in BP visits per year and greater average number of primary care physician visits per year were associated with controlled hypertension. This study added to the understanding by examining all outpatient visits associated with hypertension.Two previous studies conducted with patients with diabetes mellitus and hypertension found that similar encounter interval lengths were associated with faster time to BP control. Interestingly, even though our study only allowed participants to contribute their very first hypertensive period after diagnosis for analysis, it had similar results as the aforementioned studies that allowed participants to contribute multiple hypertensive periods over the course of the study period. Although current hypertension guidelines recommend monthly follow‐up until BP is controlled, based on the results from this study and previous studies, patients should be seen at least every 2 weeks as it was associated with higher likelihood of BP control and faster time to control.The reasons shorter encounter intervals led to a higher likelihood of achieving BP targets in a shorter amount of time was likely to be multifactorial. One likely explanation is that as the interval between encounters increases, clinical inertia may become more apparent. Clinical inertia is defined as failure by healthcare providers to appropriately initiate or intensify therapy during visits with patients. Intuitively, the more often a patient is seen over a shorter period of time, the more opportunities a provider has to appropriately adjust therapies and order appropriate labs. Another possibility is that as encounters become more spread out, especially for patients with uncontrolled BP on multiple medications, medication adherence may become of bigger concern. According to previous studies looking at risk factors for uncontrolled hypertension, medication adherence was found to be associated with BP control. In addition, increased encounter frequency has been linked to improved treatment adherence. When encounters are spaced more closely together, the healthcare professional has more opportunities and better follow‐up to assess medication adherence, potential side effects, and disease state understanding. It is also important to consider that most antihypertensive medications, except for thiazides, achieve most of their effect in less than 2 weeks. Having a patient come in biweekly allows providers to assess whether the current antihypertensive medication dose is appropriate and can adjust as needed until BP is controlled.There were a number of factors that were associated with a greater likelihood of BP control. As CCI increased, there was a significantly higher likelihood of BP control. Patients with a higher CCI have a higher rate of mortality within 1 year depending on the comorbid condition. These individuals may indirectly require closer follow‐up with tight medication management and adherence. Additionally, being a woman and older at hypertension diagnosis was associated with a higher likelihood of BP control.There were a few interesting findings that were associated with a lower likelihood of achieving BP control. Previous studies indicated that there was a significantly higher risk of uncontrolled hypertension in blacks compared with whites. The current study found the same finding; however, this was not shown to be statistically significant. Interestingly, the linear regression shows that blacks who did achieve BP control did so in a faster manner compared with whites. Additionally, Asians were found to be less likely to achieve BP control and those who did achieve BP control did so in a slower manner than whites. Further studies need to be conducted to examine relationships between race and the likelihood and time to achieve BP control.Compared with one or two encounters as reference, the likelihood of achieving BP control was significantly higher among patients who had three or four encounters but was significantly lower for patients who had seven or more encounters. Furthermore, patients who had three or four encounters were significantly found to reach BP control faster compared with patients with one or two encounters (reference). Consistently, patients who had five or more encounters were found to reach BP control significantly slower compared with those with one or two encounters. These findings suggest that there may be an optimal number of encounters to benefit patients with hypertension. Some previous studies found that increased frequency of encounters was associated with better control of chronic diseases.Study strengthsThis study had a number of strengths, including a large patient population that reflected communities within Southern California. We derived our data using a prospectively collected electronic medical database including pharmacy prescription data. Moreover, a large number of variables were investigated to explore multiple associations with the number of encounters to BP control. Although the type of hypertensive medication did not differ between groups, an exploratory analysis was conducted to add the hypertensive medication class used as an adjustment variable. Adjusting for medication class had negligible effects on all of the results. This study also used accepted national measures via the HEDIS criteria to assess BP control. Additionally, this study used JNC 7 and JNC 8 BP targets in order to accurately reflect real‐world clinical practice and goals at time of care.Study limitationsThis study's limitations are those characteristic of retrospective observational designs, including the inability to control for confounding variables that are unknown or unmeasurable. This study was dependent on International Classification of Diseases––Ninth Revision codes for the diagnosis of hypertension, which may lead to misclassification bias. In addition, because of the retrospective nature of this study, it was not possible to determine causation, and information and selection bias are also possibilities. The use of HEDIS criteria to define BP control using clinic BP recordings from a single day may lead to an inflated rate of control since SBP and DBP could come from different BP sets from the same day. Future studies exploring a more stringent definition for control and examining time in control may be valuable. Also, there is potential for patients to have multiple hypertensive periods, but this study focused solely on patients with newly controlled hypertension. Another limitation of this study is that we did not explore medication adherence as this study used hypertension medications for the purpose of identifying patients with hypertension, descriptive clinical characteristics, and adjustment in multivariate analysis. Last, this study was limited to insured patients with 1 year of continuous coverage, which may limit generalizability.CONCLUSIONSThis study found that shorter encounter intervals lead to higher likelihood of achieving BP control in a shorter amount of time in patients with newly diagnosed hypertension who received pharmacological therapy. The findings of this study suggest that consideration should be placed on updating current hypertension guidelines to have patients with newly diagnosed hypertension followed up at least biweekly once a BP medication is initiated until BP is controlled. Additional studies are needed to better understand how patient characteristics and specific therapies may be benefited by frequency and type of encounters.DISCLOSURES AND CONFLICT OF INTERESTMB is a former Kaiser employee. All other authors report no conflicts of interest and have nothing to disclose.REFERENCESJames PA, Oparil S, Carter BL. 2014 Evidence‐based guideline for the management of high BP in adults. Report from panel members appointed to the eighth joint national committee (JNC 8). JAMA. 2014;311:507‐520.Elperin DT, Pelter MA, Deamer RL, et al. A large cohort study evaluating risk factors associated with uncontrolled hypertension. J Clin Hypertens (Greenwich). 2014;16:149‐154.Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics—2015 update: a report from the American Heart Association. Circulation. 2015;131:e29‐e322.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.Shaw KM, Handler J, Wall HK. Improving BP control in a large multiethnic California population through changes in health care delivery 2004–2012. Prev Chronic Dis. 2014;11:140173.Low KJ, Pelter MA, Deamer RL, et al. Identification and evaluation of risk factors in patients with continuously uncontrolled hypertension. J Clin Hypertens (Greenwich). 2015;17:281‐289.Sim JJ, Handler J, Jacobsen SJ, et al. Systematic implementation strategies to improve hypertension: the Kaiser Permanente Southern California experience. Can J Cardiol. 2014;30:544‐552.National Committee for Quality Assurance. HEDIS & Performance Measurement. Washington, DC: National Committee for Quality Assurance; 2014. http://www.ncqa.org/hedisqualitymeasurement.aspx. Accessed Septeber 12, 2015.Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD‐9‐CM and ICD‐10 administrative data. Med Care. 2005;43:1130‐1139.Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373‐383.Yang D, Dalton JE. A unified approach to measuring the effect size between two groups using SAS®. http://support.sas.com/resources/papers/proceedings12/335-2012.pdf. Accessed January 23, 2018.Turchin A, Goldberg SI, Shubia M, et al. Encounter frequency and BP in hypertensive patients with diabetes mellitus. Hypertension. 2010;56:68‐74.Morrison A, Shubina M, Turchin A. Encounter frequency and serum glucose level, BP, and cholesterol level control in patients with diabetes mellitus. Arch Intern Med. 2011;171:1542‐1550.Guthmann R, Davis N, Brown M, et al. Visit frequency and hypertension. J Clin Hypertens (Greenwich). 2005;7:327‐332.Lebeau JP, Cadwallader JS, Aubin‐Auger I, et al. The concept and definition of therapeutic inertia in hypertension in primary care: a qualitative systematic review. BMC Family Practice. 2014;12:130.Patel NC, Crismon ML, Miller AL, et al. Drug adherence: effects of decreased visit frequency on adherence to clozapine therapy. Pharmacotherapy. 2005;25:1242‐1247.Plantinga LC, Fink NE, Sadler JH, et al. Frequency of patient‐physician contact and patient outcomes in hemodialysis care. J Am Soc Nephrol. 2004;15:210‐218.Asao K, McEwen LN, Crosson JC, et al. Revisit frequency and its association with quality of care among diabetic patients: translating research into action for diabetes (TRIAD). J Diabetes Complications. 2014;28:811‐818.Redfern J, Menzies M, Briffa T, et al. Impact of medical consultation frequency on modifiable risk factors and medications at 12 months after acute coronary syndrome in the CHOICE randomized controlled trial. Int J Cardiol. 2010;145:481‐486.Koebnick C, Langer‐Gould A, Gould M, et al. Sociodemographic characteristics of members of a large, integrated health care system: comparison with US Census Bureau data. Perm J. 2012;16:37‐41.AAPPENDIXNumber of patients censored by categoryCensorFrequencyPercentCumulative frequencyCumulative percent1_year_since_incident_date602747.84602747.84Death270.21605448.05Hypertensive_period_end562444.6411 67892.69Membership end9217.3112 599100

Journal

Journal of Clinical Hypertension – Wiley

Published: Jan 1, 2018

There are no references for this article.