- Email: [email protected]
Subclinical Mood and Cognition Impairments and Blood Pressure Control in a Large Cohort of Elderly Hypertensives
JAMDA 17 (2016) 864.e17e864.e22
JAMDA journal homepage: www.jamda.com
Original Study
Subclinical Mood and Cognition Impairments and Blood Pressure Control in a Large Cohort of Elderly Hypertensives Karolina Piotrowicz MD, PhD a, Aleksander Prejbisz MD, PhD b, Marek Klocek MD, PhD c, Roman Topór-Ma˛ dry MD, PhD d, Paulina Szczepaniak MSc e, Kalina Kawecka-Jaszcz MD, PhD c, Krzysztof Narkiewicz MD, PhD f, Tomasz Grodzicki MD, PhD a, Andrzej Januszewicz MD, PhD b, Jerzy Ga˛ sowski MD, PhD a, * a
Department of Internal Medicine and Gerontology, Jagiellonian University Medical College, Kraków, Poland Department of Hypertension, Institute of Cardiology, Warsaw, Poland c 1st Department of Cardiology, Interventional Electrocardiology and Hypertension, Jagiellonian University Medical College, Kraków, Poland d Faculty of Health Sciences, Jagiellonian University Medical College, Kraków, Poland e Medycyna Praktyczna, Liszki, Poland f sk, Poland Department of Hypertension and Diabetology, Medical University of Gdansk, Gdan b
a b s t r a c t Keywords: Cognitive impairment mood disturbances subclinical BP control elderly survey
Background: Blood pressure (BP) control in the elderly is often limited by poor compliance with prescribed regimen. Both can be influenced by clinical cognitive or mood impairments; however, the impact of subclinical alterations of cognition or mood remains unknown. Objectives: To assess the relation between cognition, mood, and BP control in treated older hypertensive patients. Design: Cross-sectional association study. Setting: Predefined substudy to the POLFOKUS nationwide survey investigating the correlates of poor BP control in patients randomly drawn from primary and specialist practices across Poland. Participants: 1988 outpatients 65 years of age treated for hypertension for at least 1 year. Measurements: BP was mean of at least 2 office measurements. We assessed adherence to antihypertensive medications using a questionnaire and performed screening tests for cognitive deficits [Abbreviated Mental Test Score (AMTS)] and mood disorders [Geriatric Depression Scale (GDS)]. In all patients, we used a unified (BP <140/90 mm Hg) and in 80 years old a unified or age-specific (<150 mm Hg systolic BP) definition of BP control. We fitted logistic regression models to assess the probability of poor BP control in association with cognitive and mood disturbances. Results: The mean [standard deviation (SD)] age of 1988 (65.6% women) patients was 73.9 (6.0) years (19.3% 80 years old). Cognitive and mood impairments were observed in 8.0% and 37.2%, respectively. Mean systolic and diastolic BP were 141.8 (16.4) and 83.6 (9.5) mm Hg, respectively. According to agestratified and unified definition of proper BP control, goal BP was achieved in 65.4% and 38.5% patients 80 years of age, respectively. In younger patients, the control reached 46%. Globally, 66% patients adhered to antihypertensive medications. Poor compliance was related to cognitive and mood impairments. When unified goal was applied, there was a 15.0% higher risk of finding poor BP control per 1 score lost in AMTS and an 8.0% increase per 1 score gained in GDS (all P < .001). Conclusion: Poorer BP control is related to subclinical worsening of cognition and mood, which supports widespread use of the Comprehensive Geriatric Assessment even in apparently self-dependent older patients with hypertension. Ó 2016 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
A.J. and J.G. are both senior authors. The POLFOKUS was funded by an unrestricted educational grant by KRKA Poland. The authors of this study were in no way remunerated by the sponsor for their involvement, and performed all tasks within the scope of their respective regular employments. The funder had no role in study design, the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the article for http://dx.doi.org/10.1016/j.jamda.2016.06.021 1525-8610/Ó 2016 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
publication. All the researchers were independent from the funder. The authors declare no conflicts of interest. * Address correspondence to Jerzy Ga˛ sowski, MD, PhD, Department of Internal Medicine and Gerontology, Jagiellonian University, Medical College, 10 Sniadeckich Street, 31-531 Kraków, Poland. E-mail address: [email protected] (J. Ga˛ sowski).
864.e18
K. Piotrowicz et al. / JAMDA 17 (2016) 864.e17e864.e22
The prevalence of hypertension increases with advancing age of populations, making the elderly hypertensives the largest, and swiftly increasing, subgroup of patients.1 Antihypertensive treatment is beneficial across a wide spectrum of age. Current evidence supports that antihypertensive therapy be instituted in older patients 65 years old, and this extends onto the population of functionally independent hypertensive patients aged 80 years or older.2,3 Despite the vast advances in therapeutic approaches to hypertension, at best 50% to 60% of individuals who have been prescribed antihypertensive medications reach blood pressure (BP) targets, a number far from satisfactory.4,5 This is especially important in view of some recently published data indicating that the BP target should be set at about 120 mm Hg systolic.6 Disputable as these results are, we are not close enough to achieving a 140e or 150emm Hg goal in very many a patient. The most important factor implicated in poor BP control is patients’ noncompliance with the prescribed regimen.7 Many factors, such as socioeconomic status, education, age, and gender, in turn influence the compliance itself.7 In the context of advancing age, level of cognition and mood may assume special importance.8e10 Memory impairments, both clinically manifest and especially at a subclinical level, increase with ageing.11 Despite lower overall prevalence in older individuals when compared with younger adults, depression both persists into older age and increases across the geriatric age strata.12 In the past, it has been demonstrated that uncontrolled midlife hypertension increases the risk of late-life cognitive decline and dementia, and that such risk diminished with increasing duration of antihypertensive treatment.13e15 Of note, patients who developed dementia tend to have lower BPs than their healthy peers, although when their BPs are tracked to the predementia period, they are significantly higher than in individuals who would not have developed marked cognitive decline.15 Mood disorders were linked to both too high and too low BP values.16,17 However, little is known about the way and strength of the possible association between changes in cognition and mood, especially at the subclinical level, that is, when an intervention could still make a difference, and compliance and thus the on-treatment BP in elderly hypertensives.
7 days, (4) omitted on 8 to 10 days, (5) omitted on 11 to 14 days, (6) took medications occasionally, and (7) did not adhere to the regimen at all. Measurements of Mood and Cognitive Impairments For the assessment of mood, we used a shortened 15-item Geriatric Depression Scale (GDS).20 For the assessment of cognitive impairment, we used a 10-item Abbreviated Mental Test Score (AMTS) and the Clock Drawing Test (CDT).21e23 Both assessment areas are an integral part of the so-called Comprehensive Geriatric Assessment. Cognitive impairment was suspected if the subject was scored at 6 or less points in AMTS test21; mood disorders were diagnosed when the patient achieved 6 and more points in GDS scale.20 We implemented a version of CDT using three preprinted dials. Patients were asked to put in the numbers, and on the second and on the third preprinted clock face to indicate 3:00 and 11:10 hours, respectively. The results were scanned and digitalized. For each participant, we assessed three clocks. First, we randomly selected 10% of all clocks, and 2 researchers (K.P. and J.G.) read them independently according to the Sunderland and modified Shulman criteria.22,23 Cases with discordant results were discussed, and a consensus approach was forged, to be implemented in similar cases. After that, one researcher (K.P.) read the CDTs for all patients. Statistical Analysis Means were compared using a standard normal Z test, and proportions, with a chi-square test. The correlation analysis was used to identify factors associated with levels of BP. Further, to study the association between level of cognition or mood and level of BP, we used the linear regression modeling with adjustment for possibly important confounders. Finally, we used the multiple logistic regression modeling to study the independence of association between cognitive function or mood and BP control. In this analysis, we used a unified (BP <140/90 mm Hg for all patients) and age-specific (<150 mm Hg SBP for octogenarians and older) definition of BP control.19 Theory
Methods Study Design, Participants The POLFOKUS study was a cross-sectional assessment of factors associated with poor BP control in patients with hypertension in Poland. The study was endorsed by the Ethics Committee of the Jagiellonian University, Kraków, Poland (KBET/196/B/2010). The design of the study has been described in detail elsewhere.18 Briefly, in fall 2010, we randomly drew 1500 general practitioners, 500 cardiologists or hypertension specialists from the registry of medical practices in Poland. The sample was to be representative for Poland. Further, each of the general practitioners had to include 6 to 8 patients and a specialist to include 8 to 10 patients. During each day of enrollment, the second and third hypertensive patients with prior appointment were included until the prespecified quota was reached. The current report is based on the data acquired in the framework of the predefined geriatric segment of the study. To be eligible to enter the geriatric segment, the patient had had to be treated for hypertension for at least 12 months and be 65 years of age at the time of the visit. We gathered information on patient’s medications, history of cardiovascular and metabolic diseases, education, and socioeconomic status. We measured sitting systolic (SBP) and diastolic (DBP) BPs twice during the visit, in accordance with current guidelines.19 The drug adherence was classified into the following: (1) patient took medications regularly as prescribed, (2) omitted medications on 1 to 3 days per month, (3) omitted on 4 to
Although several studies showed that the clinically overt depression or dementia can be traced as reasons for poor compliance with medications and thus poor BP control, no study thus far addressed whether subclinical deviations in assessment inventories could in a dose-dependent fashion translate into poor compliance with medications and poor BP control. Calculation For the first time we found that poor compliance is related to subclinical cognitive and mood impairments. When the below 140/ 90emm Hg goal was applied, there was a 15.0% higher risk of finding poor BP control per 1 score lost in AMTS and an 8.0% increase per 1 score gained in GDS (all P < .001). The observation that the subclinical worsening of cognition and mood assessed with the screening tools are related to poorer BP control lends support to the widespread use of the Comprehensive Geriatric Assessment even in apparently selfdependent oldest patients with hypertension. Results General Characteristics The mean [standard deviation (SD)] age of the 1988 patients included was 73.9 (6.0) years. Overall, 384 (19.3%) patients were at least 80 years of age. Overall, 1304 (65.6%) women were older than
K. Piotrowicz et al. / JAMDA 17 (2016) 864.e17e864.e22
men [74.2 (6.1) years vs 73.5 (5.7) years, P ¼ .02, women vs men, respectively]. The systolic and diastolic BP averaged 141.8 (16.4)/83.6 (9.5) mm Hg. The average duration of hypertension amounted to 12.4 (7.7) years. On average, a patient was prescribed three antihypertensive medications, 191 patients were prescribed one, 593 two, and 1189 three or more medications. Antihypertensive Medications and Compliance The structure of medication intake is presented in Table 1. The angiotensin-converting enzyme inhibitors (ACE-I) predominated, followed by beta-blockers and by thiazide and thiazide-like diuretics (Table 1). Overall, patients’ compliance with the prescribed regimen was as follows: 65.9% took medications regularly as prescribed, 21.3%
Table 1 Characteristics of the Study Population Characteristic
65e79 Years 80þ Years P for All Difference
Age, years Sex, n (%) Women Education, n (%) Primary Vocational Secondary school College and higher Medications, n (%) ACE inhibitors Beta-blockers Thiazide diuretics Dihydropyridine CCB ARB Loop diuretics Aldosterone antagonists Alpha-blockers CCB, other Central Compliance, n (%) Regular use (as prescribed) Nonadherent 1e3 d/mo Nonadherent 4e7 d/mo Nonadherent 8e10 d/mo Nonadherent 11e14 d/mo Drug use from time to time (a few/mo) Not at all Previous complications, n (%) Cardiovascular disease Arrhythmias Heart failure Stroke TIA CKD BPH* COPD/asthma OSA Diabetes mellitus, type 2 Impaired glucose fasting and impaired glucose tolerance Elevated TC and/or LDL cholesterol Elevated TG and/or decreased HDL cholesterol Need to use NSAIDs regularly
71.7 4.1
83.1 3.1 .001
73.9 6.0
1035 (64.5)
269 (70.1) .041
1304 (65.6)
363 444 500 243
(23.4) (28.7) (32.3) (15.7)
129 92 93 59
(34.6) (24.7) (24.9) (15.8)
.0001 .124 .006 .947
492 536 593 302
(25.6) (27.9) (30.8) (15.7)
1163 919 844 575 394 224 187 52 38 24
(73.0) (57.7) (53.0) (36.1) (24.7) (14.1) (11.7) (3.3) (2.4) (1.5)
286 220 196 150 93 108 77 14 14 0
(74.9) (57.6) (51.3) (39.3) (24.4) (28.3) (20.2) (3.7) (3.7) (0)
.449 .982 .564 .245 .88 .0001 .0001 .694 .16 .016
1449 1139 1040 725 487 332 264 66 52 24
(73.3) (57.6) (52.6) (36.7) (24.6) (16.8) (13.4) (3.3) (2.6) (1.2)
1045 340 129 30 7 25
(66.1) (21.5) (8.2) (1.9) (0.4) (1.6)
245 78 35 9 2 7
(65.2) (20.7) (9.3) (2.4) (0.5) (1.9)
.742 .751 .468 .535 .818 .699
1290 418 164 39 9 32
(65.9) (21.3) (8.4) (2.0) (0.5) (1.6)
6 (0.4) 851 577 353 94 174 97 285 147 93 421 397
(55.6) (38.3) (23.7) (6.3) (11.7) (6.5) (54.4) (10.0) (6.3) (26.7) (26.9)
0 (0) 245 185 178 40 86 52 64 43 9 102 87
(66.6) (51.3) (48.9) (11.0) (23.8) (14.7) (58.7) (12.1) (2.6) (27.0) (24.4)
.232 .0001 .0001 .0001 .002 .0001 .0001 .41 .253 .006 .899 .348
6 (0.3) 1096 762 531 134 260 149 349 190 102 523 484
(57.7) (40.8) (28.7) (7.2) (14.0) (8.1) (55.1) (10.4) (5.6) (26.7) (26.4)
965 (65.3)
217 (60.6) .097
1182 (64.4)
671 (46.0)
137 (38.9) .017
808 (44.6)
472 (31.5)
139 (38.8) .008
611 (32.9)
ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blockers; BPH, benign prostatic hypertrophy; CCB, calcium channel blocker; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; HDL, high-density lipoprotein; LDL, low-density lipoprotein; NSAIDs, nonsteroidal anti-inflammatory drugs; OSA, obstructive sleep apnea; TC, total cholesterol; TG, triglycerides; TIA, transient ischemic attack. *Men only.
864.e19
omitted medications on 1 to 3 days per month, 8.4% omitted 4 to 7 days, 2.0% omitted 8 to 10 days, 0.5% omitted 11 to 14 days, 1.6% took medications occasionally, and 0.3% did not adhere to the regimen at all. BP and BP Control in Younger and Older Patients In the younger group (65e79 years of age), mean (SD) systolic and diastolic BPs were 141.4 (16.5) and 83.5 (9.4) mm Hg, respectively. BPs in patients aged 80 years averaged 143.2 (16.3) and 83.9 (9.6) mm Hg SBP and DBP, respectively, and were not different from those observed in the younger group (all P > .06). To analyze the extent to which BP is controlled, we employed the then accepted definitions of BP target values. In the younger group, BP was considered to be under control if values of SBP/DBP were less than 140/90 mm Hg. According to this definition, 46.0% patients had adequate BP control. When we applied the same criteria to patients 80 years of age the control rate was lower (38.5%, P ¼ .008). Overall, when the above-mentioned cut-off was applied to the entire group, BP control was 44.6%. When we applied stratified definition, which included <140/90 mm Hg target for younger, and <150 mm Hg SBP for older, patients, the control rates were as follows: 49.7% in the entire group and 65.4% control among the older patients. Cognitive Function and Mood Cognitive function impairment was diagnosed in 8.0% of respondents. In 7.2% of them mild and in 0.8% severe dementia was suspected. Cognitive impairment was diagnosed significantly more often among patients aged 80 years than in younger ones, 13.2% and 6.8%, respectively (P < .001). Depressed mood was diagnosed in 37.2% of patients. In 29.4% mild and in 7.8% severe depression was suspected. Similarly, there was a difference in frequency of depressed mood according to age as it was found in 45.5% of the older and 35.2% of the younger patients (P < .001). The response rates for AMTS and GDS were 91.8% and 86.7%, respectively. The frequency distributions of cognitive and mood measure scores were shown in Figure 1. BP and BP Control in Relation to Measures of Impaired Cognition and Mood There was a linear relation between worsening of compliance and worsening of both AMTS and GDS scoring (Figure 2). There was a negative correlation between BP values and AMTS results (separately, for both SBP and DBP, r ¼ 0.13, P < .0001). In the patients with cognitive impairment, mean (SD) SBP and DBP were 146.7 (16.6) and 85.3 (9.6) mm Hg, respectively. BPs in patients without suspicion of dementia were significantly lower than in those with cognitive impairment, and averaged 141.3 (16.5) and 83.3 (9.5) mm Hg, systolic and diastolic, respectively (all P < .02). Figure 3 shows the relation between BP control and decreasing AMTS score. Adequate BP control according to unified definition was achieved by 29.3% and 46.1% of the patients with and without suspicion of dementia, respectively (P < .001). Applying the logistic regression models to the entire group, with adjustment for sex, age, educational level, and then additionally to previous cardiovascular complications (myocardial infarction, stroke, heart failure, arrhythmia), diabetes mellitus, and antihypertensive medications, we showed that cognitive decline was associated with a 15% higher risk of lack of BP control per each score below 10 in the AMTS scale, when both unified and stratified definitions of BP goal were applied (all P .004; Table 2). Concordant results were found for BP and the mood disorders measures. A positive correlations between BP values and GDS results were seen (for SBP, r ¼ 0.17, P < .0001; and for DBP, r ¼ 0.12, P < .0001). In the patients with mood disorders, mean (SD) SBP and DBP were
864.e20
K. Piotrowicz et al. / JAMDA 17 (2016) 864.e17e864.e22
Fig. 1. The frequency distributions of the AMTS and GDS scores. NDA, no data available.
144.2 (16.3) and 84.3 (9.5) mm Hg, respectively. BPs in those without depression were significantly lower than in those with mood disorders, and averaged 140.3 (16.6) and 83.0 (9.4) mm Hg, systolic and diastolic, respectively (all P < .01). Figure 3 shows the relation between BP control and increasing GDS score. The goal BP values according to the unified definition were reached by 37.4% and 49.1% of the patients with and without depression suspicion, respectively (P < .0001). When we applied the logistic regression models to the entire group, with adjustment for sex, age, educational level, and then additionally to previous cardiovascular complications (myocardial infarction, stroke, heart failure, arrhythmia), diabetes mellitus, and antihypertensive medications, we showed that depressed mood was associated with 8.0% higher risk of lack of BP control per each point above 1 in the GDS scale, for both BP goal definitions (all P < .0001, Table 2). Discussion Based on data from a Polish nationwide cohort of hypertensives 65 years of age, we found that 44.6% of them reach BP targets when treated for hypertension for at least 1 year. When we subdivided the cohort according to age, the 65 to 79eyears cohort had 46.0% control and the 80eyears cohort had a control rate of 38.5%. The latter increased to 65.4% when instead of the goal value of 140/90 mm Hg, we employed the <150emm Hg SBP value in the oldest olds. We found that poorer compliance with the regimen and poor BP control rates were associated with greater GDS scores and lower AMTS scores. On average, a difference of 1 score in GDS was associated with 8% difference in probability of lack of BP control. The corresponding difference in probability of lack of BP control associated with 1 score difference in AMTS was 15%. In our study, the majority of patients did not attain GDS or AMTS scores suggestive of either depression or dementia. This means that already at the subclinical level, mood or cognition disturbances are an
Fig. 2. The compliance with regimen and the AMTS and GDS scores.
important factor in BP control. Populations worldwide are ageing, and hypertension is one of most common chronic conditions in the elderly.1 Deleterious consequences of partial adherence to antihypertensive treatment or medication errors and therefore inadequate BP control include increased total and cardiovascular mortality or cardiovascular complications.7 Another important consequence is an increase in the prevalence of old-age-onset dementia.11 We showed control rate of hypertension in Polish old adults to be comparable to those from some of the European countries.5,24e26 Our results show improvement when compared with earlier estimates for the general population of Poland (NATPOL PLUS, 2002: 12% of treated and untreated hypertensives combined aged 18e93 achieved target values of BP,27 implying approximately 25% control rate in treated individuals). However, the overall control rate that we report is still lower than those published for Canada [The Canadian Health Measures Survey (CHMS), 66%] or the United States [The National Health and Nutrition Examination Survey (NHANES), 53%].4 The relationships between impaired cognition or mood disorders and poor hypertension control were hypothesized and demonstrated in our study. We postulated that poorer mood and cognition, possibly acting via disorganized daily activities, adversely influence compliance with medications, which in turn translates to higher BP values. This is consistent with previous research underlining the association between dementia and depression, and antihypertensive drugs nonadherence.28e32 Vinyoles et al showed in the COGNIPRES crosssectional study of 1579 Spanish old hypertensives that those with cognitive impairment had 47% less probability for therapeutic compliance and were 40% less likely to achieve adequate BP control.28 In the Rotterdam Study, Salas et al29 determined limited cognitive capacity being an independent factor for antihypertensive drugs
K. Piotrowicz et al. / JAMDA 17 (2016) 864.e17e864.e22
864.e21
Table 2 Odds Ratios of Worse Concomitant BP Control Associated With 1 Score Greater AMTS or GDS Model
AMTS (OR, 95% CI) P Value GDS (OR, 95% CI) P Value
Entire group, stratified approach 1 Entire group, stratified approach 2 Entire group, unified approach 1 Entire group, unified approach 2 80þ, stratified approach 1 80þ, stratified approach 2 80þ, unified approach 1 80þ, unified approach 2 65e79, stratified approach 1 65e79, stratified approach 2
0.87 (0.80e0.94)
.001
1.08 (1.05e1.12) <.0001
0.87 (0.79e0.96)
.004
1.08 (1.03e1.12)
0.86 (0.79e0.94)
.0008
1.08 (1.05e1.12) <.0001
0.87 (0.79e0.95)
.004
1.07 (1.03e1.12)
.0004
0.91 (0.80e1.04)
.15
1.06 (1.00e1.13)
.07
0.86 (0.60e1.23)
.39
1.11 (0.93e1.32)
.26
0.88 (0.77e1.01)
.07
1.04 (0.98e1.11)
.22
0.83 (0.70e0.97)
.02
1.07 (1.00e1.16)
.06
0.86 (0.79e0.93)
.0003
1.09 (1.05e1.12) <.0001
0.87 (0.79e0.95)
.002
1.07 (1.04e1.11)
.0003
.0001
CI, Confidence interval; OR, odds ratio. Unified BP control: <140/90 mm Hg; age-stratified BP control: <140/90 mm Hg for those aged 65e79 years, SBP <150 mm Hg for those 80þ years old. In the 65e79 years stratum, the stratified and unified BP control are of definition identical. Model 1: adjusted for sex, age, education; model 2: adjusted for sex, age, education, previous complications (myocardial infarction, stroke, heart failure, arrhythmias, diabetes mellitus type 2); and classes of hypertensive medications.
Fig. 3. The percentage of blood pressure control rates according to the AMTS and GDS scores, in the entire (ALL) and the 80 years (80þ) groups.
nonadherence. In a prospective study of 1573 hypertensive patients 55 years old, they demonstrated that cognitively impaired participants had 2 times greater risk of lack of compliance, with even higher risk for those who lived alone.29 In the Cohort Study of Medication Adherence among Older Adults (CoSMO), Krousel-Wood et al30 identified depressive symptoms as an independent, modifiable risk factor associated with 84% greater probability of decline in adherence in 2-year follow-up in 1965 hypertensive patients. In the same cohort, the decline in antihypertensive medication adherence was related to a 68% higher risk of a lack of BP control.30 In meta-analysis of risk indicators for nonadherence to antihypertensive medications provided by Lemstra and Alsabbagh,31 depression or using antidepressants were mentioned among nine factors that were related to poor compliance.31 Interestingly, in the study conducted by Siegel et al32 in 40,492 old hypertensive patients who were included in the databases of the US Department of Veterans Affairs, depression but not dementia was related to 14% higher risk of drug noncompliance.32 However, to our knowledge, our study is the first one that was designed specifically to address the issue of the relation between simple measures of mood and cognition on one hand, and patient’s compliance and BP control on the other, using simple, easy-toperform, standardized measures of cognition and mood in a representative sample of elderly hypertensives. An especially innovative aspect in our study, it is the finding that a subclinical mental and mood decline, expressed by the loss of 1 score in AMTS and GDS, translates into the risk of detecting uncontrolled hypertension. Likewise, none of the above-mentioned studies showed a clear-cut relation between subclinical changes in mood and cognition and poor BP control. Despite the fact that the clinical assessment tools we used were not designed to pick up the minute, subclinical changes in cognition and
mood, they showed better performance than the CDT, which at the time of study design was thought to be more sensitive in this regard (data not shown).33 However, the results published in 2011 are in line with our conclusion of poor performance of the CDT in asymptomatic patients.34,35 Our findings are potentially of high impact to the health of old hypertensive patients, as the geriatric assessment, including the simple and easy-to-perform mood and cognition assessment tools, may help selecting individuals at an increased risk of poor compliance and thus help curbing the poor BP control rates in the ageing populations. Not to mention the fact of the value of screening for cognitive and mood disturbances themselves. We showed that less than half of elderly hypertensive outpatients had satisfactory BP control that, at least partially, may be attributed to some undetected or untreated mental deficits. Such groups of patients should be followed regularly with assessment of mental health. The caregivers or families may need to be advised to oversee the antihypertensive therapy, or in fact all pharmacologic therapy in case of even subtle cognitive deficits and mood disturbances of their proxies. Our study needs to be interpreted in the context of its limitations. First, the study was a cross-sectional assessment and, although probable, the causative relation cannot be strictly asserted. There may have been 2 sources of possible bias in the selection of our patients. First, as the sampling unit was an outpatient practice (whether general or specialist), and not the patient, the enrollees were not strictly randomly drawn. On the other hand, we achieved representativeness of our sample of practices at the national level and ensured strict adherence to the protocol that predefined close to random inclusion of patients. The second source of bias is associated with the fact that the patients had to be fit enough to make it to the clinics; thus, our sample probably is not representative of the frail and severely demented, depressed, or disabled old individuals. However, this may have at best diluted the strength of the associations we show. Some bias in the reporting of the adherence to medications may have been introduced by inclusion of a small group affected with more substantial cognitive impairment.
864.e22
K. Piotrowicz et al. / JAMDA 17 (2016) 864.e17e864.e22
The strength of our study rests on its having been specifically designed to address the question of relation between BP control and cognitive and mood disturbances in the community-dwelling old hypertensive patients, and all shown analyses had been prespecified. In conclusion, poor BP control is related to the subclinical worsening of cognition and mood, which supports widespread use of the Comprehensive Geriatric Assessment even in apparently selfdependent older patients with hypertension. Conclusions With ageing of populations the subclinical and clinical cognitive and mood disturbances are increasing in frequency. We assessed BP control in a representative sample of Polish hypertensive outpatients at or above the age of 65 as a function of cognition and mood. Even at the subclinical level, worse results of screening tests for cognitive and mood disorders were in a quantifiable way associated with poor compliance, and further with greater probability of concomitant lack of adequate BP control. References 1. Global status report on non-communicable diseases 2014. Available at: http:// www.who.int/nmh/publications/ncd-status-report-2014/en/. Accessed May 8, 2016. 2. Staessen JA, Gasowski J, Wang JG, et al. Risks of untreated and treated isolated systolic hypertension in the elderly: Meta-analysis of outcome trials. Lancet 2000;355:865e872. 3. Benetos A, Rossignol P, Cherubini A, et al. Polypharmacy in the aging patient: Management of hypertension in octogenarians. JAMA 2015;314:170e180. 4. Joffres M, Falaschetti E, Gillespie C, et al. Hypertension prevalence, awareness, treatment and control in national surveys from England, the USA and Canada, and correlation with stroke and ischaemic heart disease mortality: A crosssectional study. BMJ Open 2013;3:e003423. 5. Pereira M, Lunet N, Azevedo A, Barros H. Differences in prevalence, awareness, treatment and control of hypertension between developing and developed countries. J Hypertens 2009;27:963e975. 6. SPRINT Research Group, Wright JT Jr, Williamson JD, Whelton PK, et al. A randomized trial of intensive versus standard blood-pressure control. N Engl J Med 2015;373:2103e2116. 7. Sabaté E, editor. Adherence to Long-Term Therapies: Evidence for Action. Geneva, Switzerland: World Health Organization. Available at: http://www. who.int/chp/knowledge/publications/adherence_introduction.pdf; 2003. Accessed May 16, 2016. 8. Khatib R, Schwalm JD, Yusuf S, et al. Patient and healthcare provider barriers to hypertension awareness, treatment and follow up: A systematic review and meta-analysis of qualitative and quantitative studies. PLoS One 2014;9:e84238. 9. Campbell NL, Boustani MA, Skopelja EN, et al. Medication adherence in older adults with cognitive impairment: A systematic evidence-based review. Am J Geriatr Pharmacother 2012;10:165e177. 10. Krousel-Wood M, Joyce C, Holt E, et al. Association of depression with antihypertensive medication adherence in older adults: Cross-sectional and longitudinal findings from CoSMO. Ann Behav Med 2010;40:248e257. 11. The World Alzheimer Report 2014, Dementia and Risk Reduction: An analysis of protective and modifiable factors. Available at: http://www.alz.co.uk/ research/world-report-2014. Accessed May 8, 2016. 12. Aziz R, Steffens DC. What are the causes of late-life depression? Psychiatr Clin North Am 2013;36:497e516. 13. Gottesman RF, Schneider AL, Albert M, et al. Midlife hypertension and 20-year cognitive change: The Atherosclerosis Risk in Communities neurocognitive study. JAMA Neurol 2014;71:1218e1227.
14. Launer LJ, Hughes T, Yu B, et al. Lowering midlife levels of systolic blood pressure as a public health strategy to reduce late-life dementia: Perspective from the Honolulu Heart Program/Honolulu Asia Aging Study. Hypertension 2010;55:1352e1359. 15. Stewart R, Xue QL, Masaki K, et al. Change in blood pressure and incident dementia: A 32-year prospective study. Hypertension 2009;54:233e240. 16. Meng L, Chen D, Yang Y, et al. Depression increases the risk of hypertension incidence: A meta-analysis of prospective cohort studies. J Hypertens 2012;30: 842e851. 17. Hildrum B, Romild U, Holmen J. Anxiety and depression lowers blood pressure: 22-year follow-up of the population based HUNT study, Norway. BMC Public Health 2011;11:601. 18. Prejbisz A, Klocek M, Ga˛ sowski J, et al. Factors associated with resistant hypertension in a large cohort of hypertensive patients: Pol-Fokus study. Pol Arch Med Wewn 2015;125:249e259. 19. Mancia G, Fagard R, Narkiewicz K, et al. 2013 ESH/ESC Guidelines for the management of arterial hypertension: The Task Force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens 2013;31: 1281e1357. 20. Sheikh JI, Yesavage JA. Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. In: Brink TL, editor. Clinical Gerontology: A Guide to Assessment and Intervention. New York: Haworth Press; 1986. 21. Hodkinson HM. Evaluation of a mental test score for assessment of mental impairment in the elderly. Age Ageing 1972;1:233e238. 22. Shulman KI. Clock-drawing: Is it the ideal cognitive screening test? Int J Geriatr Psychiatry 2000;15:548e561. 23. Sunderland T, Hill J, Mellow A, et al. Clock drawing in Alzheimer’s disease. A novel measure of dementia severity. J Am Geriatr Soc 1989;37:725e729. 24. Grassi G, Cifkova R, Laurent S, et al. Blood pressure control and cardiovascular risk profile in hypertensive patients from central and eastern European countries: Results of the BP-CARE study. Eur Heart J 2011;32: 218e225. 25. Llisterri JL, Rodriguez-Roca GC, Escobar C, et al, Working Group of Arterial Hypertension of the Spanish Society of Primary Care Physicians Group HTASEMERGEN, PRESCAP 2010 Investigators. Treatment and blood pressure control in Spain during 2002-2010. J Hypertens 2012;30:2425e2431. 26. Van der Niepen P, Dupont AG. Improved blood pressure control in elderly hypertensive patients: Results of the PAPY-65 Survey. Drugs Aging 2010;27: 573e588. 27. Zdrojewski T, Szpakowski P, Bandosz P, et al. Arterial hypertension in Poland in 2002. J Hum Hypertens 2004;18:557e562. 28. Vinyoles E, De la Figuera M, Gonzalez-Segura D. Cognitive function and blood pressure control in hypertensive patients over 60 years of age: COGNIPRES study. Curr Med Res Opin 2008;24:3331e3339. 29. Salas M, In’t Veld BA, van der Linden PD, et al. Impaired cognitive function and compliance with antihypertensive drugs in elderly: The Rotterdam Study. Clin Pharmacol Ther 2001;70:561e566. 30. Krousel-Wood M, Joyce C, Holt E, et al. Predictors of decline in medication adherence: Results from the cohort study of medication adherence among older adults. Hypertension 2011;58:804e810. 31. Lemstra M, Alsabbagh MW. Proportion and risk indicators of nonadherence to antihypertensive therapy: A meta-analysis. Patient Prefer Adherence 2014;8: 211e218. 32. Siegel D, Lopez J, Meier J. Antihypertensive medication adherence in the Department of Veterans Affairs. Am J Med 2007;120:26e32. 33. Yamamoto S, Mogi N, Umegaki H, et al. The clock drawing test as a valid screening method for mild cognitive impairment. Dement Geriatr Cogn Disord 2004;18:172e179. 34. Parsey CM, Schmitter-Edgecombe M. Quantitative and qualitative analyses of the clock drawing test in mild cognitive impairment and Alzheimer disease: Evaluation of a modified scoring system. J Geriatr Psychiatry Neurol 2011;24: 108e118. 35. Ehreke L, Luck T, Luppa M, et al. Clock drawing testdScreening utility for mild cognitive impairment according to different scoring systems: Results of the Leipzig Longitudinal Study of the Aged (LEILA 75þ). Int Psychogeriatr 2011;23: 1592e1601.
JAMDA journal homepage: www.jamda.com
Original Study
Subclinical Mood and Cognition Impairments and Blood Pressure Control in a Large Cohort of Elderly Hypertensives Karolina Piotrowicz MD, PhD a, Aleksander Prejbisz MD, PhD b, Marek Klocek MD, PhD c, Roman Topór-Ma˛ dry MD, PhD d, Paulina Szczepaniak MSc e, Kalina Kawecka-Jaszcz MD, PhD c, Krzysztof Narkiewicz MD, PhD f, Tomasz Grodzicki MD, PhD a, Andrzej Januszewicz MD, PhD b, Jerzy Ga˛ sowski MD, PhD a, * a
Department of Internal Medicine and Gerontology, Jagiellonian University Medical College, Kraków, Poland Department of Hypertension, Institute of Cardiology, Warsaw, Poland c 1st Department of Cardiology, Interventional Electrocardiology and Hypertension, Jagiellonian University Medical College, Kraków, Poland d Faculty of Health Sciences, Jagiellonian University Medical College, Kraków, Poland e Medycyna Praktyczna, Liszki, Poland f sk, Poland Department of Hypertension and Diabetology, Medical University of Gdansk, Gdan b
a b s t r a c t Keywords: Cognitive impairment mood disturbances subclinical BP control elderly survey
Background: Blood pressure (BP) control in the elderly is often limited by poor compliance with prescribed regimen. Both can be influenced by clinical cognitive or mood impairments; however, the impact of subclinical alterations of cognition or mood remains unknown. Objectives: To assess the relation between cognition, mood, and BP control in treated older hypertensive patients. Design: Cross-sectional association study. Setting: Predefined substudy to the POLFOKUS nationwide survey investigating the correlates of poor BP control in patients randomly drawn from primary and specialist practices across Poland. Participants: 1988 outpatients 65 years of age treated for hypertension for at least 1 year. Measurements: BP was mean of at least 2 office measurements. We assessed adherence to antihypertensive medications using a questionnaire and performed screening tests for cognitive deficits [Abbreviated Mental Test Score (AMTS)] and mood disorders [Geriatric Depression Scale (GDS)]. In all patients, we used a unified (BP <140/90 mm Hg) and in 80 years old a unified or age-specific (<150 mm Hg systolic BP) definition of BP control. We fitted logistic regression models to assess the probability of poor BP control in association with cognitive and mood disturbances. Results: The mean [standard deviation (SD)] age of 1988 (65.6% women) patients was 73.9 (6.0) years (19.3% 80 years old). Cognitive and mood impairments were observed in 8.0% and 37.2%, respectively. Mean systolic and diastolic BP were 141.8 (16.4) and 83.6 (9.5) mm Hg, respectively. According to agestratified and unified definition of proper BP control, goal BP was achieved in 65.4% and 38.5% patients 80 years of age, respectively. In younger patients, the control reached 46%. Globally, 66% patients adhered to antihypertensive medications. Poor compliance was related to cognitive and mood impairments. When unified goal was applied, there was a 15.0% higher risk of finding poor BP control per 1 score lost in AMTS and an 8.0% increase per 1 score gained in GDS (all P < .001). Conclusion: Poorer BP control is related to subclinical worsening of cognition and mood, which supports widespread use of the Comprehensive Geriatric Assessment even in apparently self-dependent older patients with hypertension. Ó 2016 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
A.J. and J.G. are both senior authors. The POLFOKUS was funded by an unrestricted educational grant by KRKA Poland. The authors of this study were in no way remunerated by the sponsor for their involvement, and performed all tasks within the scope of their respective regular employments. The funder had no role in study design, the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the article for http://dx.doi.org/10.1016/j.jamda.2016.06.021 1525-8610/Ó 2016 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
publication. All the researchers were independent from the funder. The authors declare no conflicts of interest. * Address correspondence to Jerzy Ga˛ sowski, MD, PhD, Department of Internal Medicine and Gerontology, Jagiellonian University, Medical College, 10 Sniadeckich Street, 31-531 Kraków, Poland. E-mail address: [email protected] (J. Ga˛ sowski).
864.e18
K. Piotrowicz et al. / JAMDA 17 (2016) 864.e17e864.e22
The prevalence of hypertension increases with advancing age of populations, making the elderly hypertensives the largest, and swiftly increasing, subgroup of patients.1 Antihypertensive treatment is beneficial across a wide spectrum of age. Current evidence supports that antihypertensive therapy be instituted in older patients 65 years old, and this extends onto the population of functionally independent hypertensive patients aged 80 years or older.2,3 Despite the vast advances in therapeutic approaches to hypertension, at best 50% to 60% of individuals who have been prescribed antihypertensive medications reach blood pressure (BP) targets, a number far from satisfactory.4,5 This is especially important in view of some recently published data indicating that the BP target should be set at about 120 mm Hg systolic.6 Disputable as these results are, we are not close enough to achieving a 140e or 150emm Hg goal in very many a patient. The most important factor implicated in poor BP control is patients’ noncompliance with the prescribed regimen.7 Many factors, such as socioeconomic status, education, age, and gender, in turn influence the compliance itself.7 In the context of advancing age, level of cognition and mood may assume special importance.8e10 Memory impairments, both clinically manifest and especially at a subclinical level, increase with ageing.11 Despite lower overall prevalence in older individuals when compared with younger adults, depression both persists into older age and increases across the geriatric age strata.12 In the past, it has been demonstrated that uncontrolled midlife hypertension increases the risk of late-life cognitive decline and dementia, and that such risk diminished with increasing duration of antihypertensive treatment.13e15 Of note, patients who developed dementia tend to have lower BPs than their healthy peers, although when their BPs are tracked to the predementia period, they are significantly higher than in individuals who would not have developed marked cognitive decline.15 Mood disorders were linked to both too high and too low BP values.16,17 However, little is known about the way and strength of the possible association between changes in cognition and mood, especially at the subclinical level, that is, when an intervention could still make a difference, and compliance and thus the on-treatment BP in elderly hypertensives.
7 days, (4) omitted on 8 to 10 days, (5) omitted on 11 to 14 days, (6) took medications occasionally, and (7) did not adhere to the regimen at all. Measurements of Mood and Cognitive Impairments For the assessment of mood, we used a shortened 15-item Geriatric Depression Scale (GDS).20 For the assessment of cognitive impairment, we used a 10-item Abbreviated Mental Test Score (AMTS) and the Clock Drawing Test (CDT).21e23 Both assessment areas are an integral part of the so-called Comprehensive Geriatric Assessment. Cognitive impairment was suspected if the subject was scored at 6 or less points in AMTS test21; mood disorders were diagnosed when the patient achieved 6 and more points in GDS scale.20 We implemented a version of CDT using three preprinted dials. Patients were asked to put in the numbers, and on the second and on the third preprinted clock face to indicate 3:00 and 11:10 hours, respectively. The results were scanned and digitalized. For each participant, we assessed three clocks. First, we randomly selected 10% of all clocks, and 2 researchers (K.P. and J.G.) read them independently according to the Sunderland and modified Shulman criteria.22,23 Cases with discordant results were discussed, and a consensus approach was forged, to be implemented in similar cases. After that, one researcher (K.P.) read the CDTs for all patients. Statistical Analysis Means were compared using a standard normal Z test, and proportions, with a chi-square test. The correlation analysis was used to identify factors associated with levels of BP. Further, to study the association between level of cognition or mood and level of BP, we used the linear regression modeling with adjustment for possibly important confounders. Finally, we used the multiple logistic regression modeling to study the independence of association between cognitive function or mood and BP control. In this analysis, we used a unified (BP <140/90 mm Hg for all patients) and age-specific (<150 mm Hg SBP for octogenarians and older) definition of BP control.19 Theory
Methods Study Design, Participants The POLFOKUS study was a cross-sectional assessment of factors associated with poor BP control in patients with hypertension in Poland. The study was endorsed by the Ethics Committee of the Jagiellonian University, Kraków, Poland (KBET/196/B/2010). The design of the study has been described in detail elsewhere.18 Briefly, in fall 2010, we randomly drew 1500 general practitioners, 500 cardiologists or hypertension specialists from the registry of medical practices in Poland. The sample was to be representative for Poland. Further, each of the general practitioners had to include 6 to 8 patients and a specialist to include 8 to 10 patients. During each day of enrollment, the second and third hypertensive patients with prior appointment were included until the prespecified quota was reached. The current report is based on the data acquired in the framework of the predefined geriatric segment of the study. To be eligible to enter the geriatric segment, the patient had had to be treated for hypertension for at least 12 months and be 65 years of age at the time of the visit. We gathered information on patient’s medications, history of cardiovascular and metabolic diseases, education, and socioeconomic status. We measured sitting systolic (SBP) and diastolic (DBP) BPs twice during the visit, in accordance with current guidelines.19 The drug adherence was classified into the following: (1) patient took medications regularly as prescribed, (2) omitted medications on 1 to 3 days per month, (3) omitted on 4 to
Although several studies showed that the clinically overt depression or dementia can be traced as reasons for poor compliance with medications and thus poor BP control, no study thus far addressed whether subclinical deviations in assessment inventories could in a dose-dependent fashion translate into poor compliance with medications and poor BP control. Calculation For the first time we found that poor compliance is related to subclinical cognitive and mood impairments. When the below 140/ 90emm Hg goal was applied, there was a 15.0% higher risk of finding poor BP control per 1 score lost in AMTS and an 8.0% increase per 1 score gained in GDS (all P < .001). The observation that the subclinical worsening of cognition and mood assessed with the screening tools are related to poorer BP control lends support to the widespread use of the Comprehensive Geriatric Assessment even in apparently selfdependent oldest patients with hypertension. Results General Characteristics The mean [standard deviation (SD)] age of the 1988 patients included was 73.9 (6.0) years. Overall, 384 (19.3%) patients were at least 80 years of age. Overall, 1304 (65.6%) women were older than
K. Piotrowicz et al. / JAMDA 17 (2016) 864.e17e864.e22
men [74.2 (6.1) years vs 73.5 (5.7) years, P ¼ .02, women vs men, respectively]. The systolic and diastolic BP averaged 141.8 (16.4)/83.6 (9.5) mm Hg. The average duration of hypertension amounted to 12.4 (7.7) years. On average, a patient was prescribed three antihypertensive medications, 191 patients were prescribed one, 593 two, and 1189 three or more medications. Antihypertensive Medications and Compliance The structure of medication intake is presented in Table 1. The angiotensin-converting enzyme inhibitors (ACE-I) predominated, followed by beta-blockers and by thiazide and thiazide-like diuretics (Table 1). Overall, patients’ compliance with the prescribed regimen was as follows: 65.9% took medications regularly as prescribed, 21.3%
Table 1 Characteristics of the Study Population Characteristic
65e79 Years 80þ Years P for All Difference
Age, years Sex, n (%) Women Education, n (%) Primary Vocational Secondary school College and higher Medications, n (%) ACE inhibitors Beta-blockers Thiazide diuretics Dihydropyridine CCB ARB Loop diuretics Aldosterone antagonists Alpha-blockers CCB, other Central Compliance, n (%) Regular use (as prescribed) Nonadherent 1e3 d/mo Nonadherent 4e7 d/mo Nonadherent 8e10 d/mo Nonadherent 11e14 d/mo Drug use from time to time (a few/mo) Not at all Previous complications, n (%) Cardiovascular disease Arrhythmias Heart failure Stroke TIA CKD BPH* COPD/asthma OSA Diabetes mellitus, type 2 Impaired glucose fasting and impaired glucose tolerance Elevated TC and/or LDL cholesterol Elevated TG and/or decreased HDL cholesterol Need to use NSAIDs regularly
71.7 4.1
83.1 3.1 .001
73.9 6.0
1035 (64.5)
269 (70.1) .041
1304 (65.6)
363 444 500 243
(23.4) (28.7) (32.3) (15.7)
129 92 93 59
(34.6) (24.7) (24.9) (15.8)
.0001 .124 .006 .947
492 536 593 302
(25.6) (27.9) (30.8) (15.7)
1163 919 844 575 394 224 187 52 38 24
(73.0) (57.7) (53.0) (36.1) (24.7) (14.1) (11.7) (3.3) (2.4) (1.5)
286 220 196 150 93 108 77 14 14 0
(74.9) (57.6) (51.3) (39.3) (24.4) (28.3) (20.2) (3.7) (3.7) (0)
.449 .982 .564 .245 .88 .0001 .0001 .694 .16 .016
1449 1139 1040 725 487 332 264 66 52 24
(73.3) (57.6) (52.6) (36.7) (24.6) (16.8) (13.4) (3.3) (2.6) (1.2)
1045 340 129 30 7 25
(66.1) (21.5) (8.2) (1.9) (0.4) (1.6)
245 78 35 9 2 7
(65.2) (20.7) (9.3) (2.4) (0.5) (1.9)
.742 .751 .468 .535 .818 .699
1290 418 164 39 9 32
(65.9) (21.3) (8.4) (2.0) (0.5) (1.6)
6 (0.4) 851 577 353 94 174 97 285 147 93 421 397
(55.6) (38.3) (23.7) (6.3) (11.7) (6.5) (54.4) (10.0) (6.3) (26.7) (26.9)
0 (0) 245 185 178 40 86 52 64 43 9 102 87
(66.6) (51.3) (48.9) (11.0) (23.8) (14.7) (58.7) (12.1) (2.6) (27.0) (24.4)
.232 .0001 .0001 .0001 .002 .0001 .0001 .41 .253 .006 .899 .348
6 (0.3) 1096 762 531 134 260 149 349 190 102 523 484
(57.7) (40.8) (28.7) (7.2) (14.0) (8.1) (55.1) (10.4) (5.6) (26.7) (26.4)
965 (65.3)
217 (60.6) .097
1182 (64.4)
671 (46.0)
137 (38.9) .017
808 (44.6)
472 (31.5)
139 (38.8) .008
611 (32.9)
ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blockers; BPH, benign prostatic hypertrophy; CCB, calcium channel blocker; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; HDL, high-density lipoprotein; LDL, low-density lipoprotein; NSAIDs, nonsteroidal anti-inflammatory drugs; OSA, obstructive sleep apnea; TC, total cholesterol; TG, triglycerides; TIA, transient ischemic attack. *Men only.
864.e19
omitted medications on 1 to 3 days per month, 8.4% omitted 4 to 7 days, 2.0% omitted 8 to 10 days, 0.5% omitted 11 to 14 days, 1.6% took medications occasionally, and 0.3% did not adhere to the regimen at all. BP and BP Control in Younger and Older Patients In the younger group (65e79 years of age), mean (SD) systolic and diastolic BPs were 141.4 (16.5) and 83.5 (9.4) mm Hg, respectively. BPs in patients aged 80 years averaged 143.2 (16.3) and 83.9 (9.6) mm Hg SBP and DBP, respectively, and were not different from those observed in the younger group (all P > .06). To analyze the extent to which BP is controlled, we employed the then accepted definitions of BP target values. In the younger group, BP was considered to be under control if values of SBP/DBP were less than 140/90 mm Hg. According to this definition, 46.0% patients had adequate BP control. When we applied the same criteria to patients 80 years of age the control rate was lower (38.5%, P ¼ .008). Overall, when the above-mentioned cut-off was applied to the entire group, BP control was 44.6%. When we applied stratified definition, which included <140/90 mm Hg target for younger, and <150 mm Hg SBP for older, patients, the control rates were as follows: 49.7% in the entire group and 65.4% control among the older patients. Cognitive Function and Mood Cognitive function impairment was diagnosed in 8.0% of respondents. In 7.2% of them mild and in 0.8% severe dementia was suspected. Cognitive impairment was diagnosed significantly more often among patients aged 80 years than in younger ones, 13.2% and 6.8%, respectively (P < .001). Depressed mood was diagnosed in 37.2% of patients. In 29.4% mild and in 7.8% severe depression was suspected. Similarly, there was a difference in frequency of depressed mood according to age as it was found in 45.5% of the older and 35.2% of the younger patients (P < .001). The response rates for AMTS and GDS were 91.8% and 86.7%, respectively. The frequency distributions of cognitive and mood measure scores were shown in Figure 1. BP and BP Control in Relation to Measures of Impaired Cognition and Mood There was a linear relation between worsening of compliance and worsening of both AMTS and GDS scoring (Figure 2). There was a negative correlation between BP values and AMTS results (separately, for both SBP and DBP, r ¼ 0.13, P < .0001). In the patients with cognitive impairment, mean (SD) SBP and DBP were 146.7 (16.6) and 85.3 (9.6) mm Hg, respectively. BPs in patients without suspicion of dementia were significantly lower than in those with cognitive impairment, and averaged 141.3 (16.5) and 83.3 (9.5) mm Hg, systolic and diastolic, respectively (all P < .02). Figure 3 shows the relation between BP control and decreasing AMTS score. Adequate BP control according to unified definition was achieved by 29.3% and 46.1% of the patients with and without suspicion of dementia, respectively (P < .001). Applying the logistic regression models to the entire group, with adjustment for sex, age, educational level, and then additionally to previous cardiovascular complications (myocardial infarction, stroke, heart failure, arrhythmia), diabetes mellitus, and antihypertensive medications, we showed that cognitive decline was associated with a 15% higher risk of lack of BP control per each score below 10 in the AMTS scale, when both unified and stratified definitions of BP goal were applied (all P .004; Table 2). Concordant results were found for BP and the mood disorders measures. A positive correlations between BP values and GDS results were seen (for SBP, r ¼ 0.17, P < .0001; and for DBP, r ¼ 0.12, P < .0001). In the patients with mood disorders, mean (SD) SBP and DBP were
864.e20
K. Piotrowicz et al. / JAMDA 17 (2016) 864.e17e864.e22
Fig. 1. The frequency distributions of the AMTS and GDS scores. NDA, no data available.
144.2 (16.3) and 84.3 (9.5) mm Hg, respectively. BPs in those without depression were significantly lower than in those with mood disorders, and averaged 140.3 (16.6) and 83.0 (9.4) mm Hg, systolic and diastolic, respectively (all P < .01). Figure 3 shows the relation between BP control and increasing GDS score. The goal BP values according to the unified definition were reached by 37.4% and 49.1% of the patients with and without depression suspicion, respectively (P < .0001). When we applied the logistic regression models to the entire group, with adjustment for sex, age, educational level, and then additionally to previous cardiovascular complications (myocardial infarction, stroke, heart failure, arrhythmia), diabetes mellitus, and antihypertensive medications, we showed that depressed mood was associated with 8.0% higher risk of lack of BP control per each point above 1 in the GDS scale, for both BP goal definitions (all P < .0001, Table 2). Discussion Based on data from a Polish nationwide cohort of hypertensives 65 years of age, we found that 44.6% of them reach BP targets when treated for hypertension for at least 1 year. When we subdivided the cohort according to age, the 65 to 79eyears cohort had 46.0% control and the 80eyears cohort had a control rate of 38.5%. The latter increased to 65.4% when instead of the goal value of 140/90 mm Hg, we employed the <150emm Hg SBP value in the oldest olds. We found that poorer compliance with the regimen and poor BP control rates were associated with greater GDS scores and lower AMTS scores. On average, a difference of 1 score in GDS was associated with 8% difference in probability of lack of BP control. The corresponding difference in probability of lack of BP control associated with 1 score difference in AMTS was 15%. In our study, the majority of patients did not attain GDS or AMTS scores suggestive of either depression or dementia. This means that already at the subclinical level, mood or cognition disturbances are an
Fig. 2. The compliance with regimen and the AMTS and GDS scores.
important factor in BP control. Populations worldwide are ageing, and hypertension is one of most common chronic conditions in the elderly.1 Deleterious consequences of partial adherence to antihypertensive treatment or medication errors and therefore inadequate BP control include increased total and cardiovascular mortality or cardiovascular complications.7 Another important consequence is an increase in the prevalence of old-age-onset dementia.11 We showed control rate of hypertension in Polish old adults to be comparable to those from some of the European countries.5,24e26 Our results show improvement when compared with earlier estimates for the general population of Poland (NATPOL PLUS, 2002: 12% of treated and untreated hypertensives combined aged 18e93 achieved target values of BP,27 implying approximately 25% control rate in treated individuals). However, the overall control rate that we report is still lower than those published for Canada [The Canadian Health Measures Survey (CHMS), 66%] or the United States [The National Health and Nutrition Examination Survey (NHANES), 53%].4 The relationships between impaired cognition or mood disorders and poor hypertension control were hypothesized and demonstrated in our study. We postulated that poorer mood and cognition, possibly acting via disorganized daily activities, adversely influence compliance with medications, which in turn translates to higher BP values. This is consistent with previous research underlining the association between dementia and depression, and antihypertensive drugs nonadherence.28e32 Vinyoles et al showed in the COGNIPRES crosssectional study of 1579 Spanish old hypertensives that those with cognitive impairment had 47% less probability for therapeutic compliance and were 40% less likely to achieve adequate BP control.28 In the Rotterdam Study, Salas et al29 determined limited cognitive capacity being an independent factor for antihypertensive drugs
K. Piotrowicz et al. / JAMDA 17 (2016) 864.e17e864.e22
864.e21
Table 2 Odds Ratios of Worse Concomitant BP Control Associated With 1 Score Greater AMTS or GDS Model
AMTS (OR, 95% CI) P Value GDS (OR, 95% CI) P Value
Entire group, stratified approach 1 Entire group, stratified approach 2 Entire group, unified approach 1 Entire group, unified approach 2 80þ, stratified approach 1 80þ, stratified approach 2 80þ, unified approach 1 80þ, unified approach 2 65e79, stratified approach 1 65e79, stratified approach 2
0.87 (0.80e0.94)
.001
1.08 (1.05e1.12) <.0001
0.87 (0.79e0.96)
.004
1.08 (1.03e1.12)
0.86 (0.79e0.94)
.0008
1.08 (1.05e1.12) <.0001
0.87 (0.79e0.95)
.004
1.07 (1.03e1.12)
.0004
0.91 (0.80e1.04)
.15
1.06 (1.00e1.13)
.07
0.86 (0.60e1.23)
.39
1.11 (0.93e1.32)
.26
0.88 (0.77e1.01)
.07
1.04 (0.98e1.11)
.22
0.83 (0.70e0.97)
.02
1.07 (1.00e1.16)
.06
0.86 (0.79e0.93)
.0003
1.09 (1.05e1.12) <.0001
0.87 (0.79e0.95)
.002
1.07 (1.04e1.11)
.0003
.0001
CI, Confidence interval; OR, odds ratio. Unified BP control: <140/90 mm Hg; age-stratified BP control: <140/90 mm Hg for those aged 65e79 years, SBP <150 mm Hg for those 80þ years old. In the 65e79 years stratum, the stratified and unified BP control are of definition identical. Model 1: adjusted for sex, age, education; model 2: adjusted for sex, age, education, previous complications (myocardial infarction, stroke, heart failure, arrhythmias, diabetes mellitus type 2); and classes of hypertensive medications.
Fig. 3. The percentage of blood pressure control rates according to the AMTS and GDS scores, in the entire (ALL) and the 80 years (80þ) groups.
nonadherence. In a prospective study of 1573 hypertensive patients 55 years old, they demonstrated that cognitively impaired participants had 2 times greater risk of lack of compliance, with even higher risk for those who lived alone.29 In the Cohort Study of Medication Adherence among Older Adults (CoSMO), Krousel-Wood et al30 identified depressive symptoms as an independent, modifiable risk factor associated with 84% greater probability of decline in adherence in 2-year follow-up in 1965 hypertensive patients. In the same cohort, the decline in antihypertensive medication adherence was related to a 68% higher risk of a lack of BP control.30 In meta-analysis of risk indicators for nonadherence to antihypertensive medications provided by Lemstra and Alsabbagh,31 depression or using antidepressants were mentioned among nine factors that were related to poor compliance.31 Interestingly, in the study conducted by Siegel et al32 in 40,492 old hypertensive patients who were included in the databases of the US Department of Veterans Affairs, depression but not dementia was related to 14% higher risk of drug noncompliance.32 However, to our knowledge, our study is the first one that was designed specifically to address the issue of the relation between simple measures of mood and cognition on one hand, and patient’s compliance and BP control on the other, using simple, easy-toperform, standardized measures of cognition and mood in a representative sample of elderly hypertensives. An especially innovative aspect in our study, it is the finding that a subclinical mental and mood decline, expressed by the loss of 1 score in AMTS and GDS, translates into the risk of detecting uncontrolled hypertension. Likewise, none of the above-mentioned studies showed a clear-cut relation between subclinical changes in mood and cognition and poor BP control. Despite the fact that the clinical assessment tools we used were not designed to pick up the minute, subclinical changes in cognition and
mood, they showed better performance than the CDT, which at the time of study design was thought to be more sensitive in this regard (data not shown).33 However, the results published in 2011 are in line with our conclusion of poor performance of the CDT in asymptomatic patients.34,35 Our findings are potentially of high impact to the health of old hypertensive patients, as the geriatric assessment, including the simple and easy-to-perform mood and cognition assessment tools, may help selecting individuals at an increased risk of poor compliance and thus help curbing the poor BP control rates in the ageing populations. Not to mention the fact of the value of screening for cognitive and mood disturbances themselves. We showed that less than half of elderly hypertensive outpatients had satisfactory BP control that, at least partially, may be attributed to some undetected or untreated mental deficits. Such groups of patients should be followed regularly with assessment of mental health. The caregivers or families may need to be advised to oversee the antihypertensive therapy, or in fact all pharmacologic therapy in case of even subtle cognitive deficits and mood disturbances of their proxies. Our study needs to be interpreted in the context of its limitations. First, the study was a cross-sectional assessment and, although probable, the causative relation cannot be strictly asserted. There may have been 2 sources of possible bias in the selection of our patients. First, as the sampling unit was an outpatient practice (whether general or specialist), and not the patient, the enrollees were not strictly randomly drawn. On the other hand, we achieved representativeness of our sample of practices at the national level and ensured strict adherence to the protocol that predefined close to random inclusion of patients. The second source of bias is associated with the fact that the patients had to be fit enough to make it to the clinics; thus, our sample probably is not representative of the frail and severely demented, depressed, or disabled old individuals. However, this may have at best diluted the strength of the associations we show. Some bias in the reporting of the adherence to medications may have been introduced by inclusion of a small group affected with more substantial cognitive impairment.
864.e22
K. Piotrowicz et al. / JAMDA 17 (2016) 864.e17e864.e22
The strength of our study rests on its having been specifically designed to address the question of relation between BP control and cognitive and mood disturbances in the community-dwelling old hypertensive patients, and all shown analyses had been prespecified. In conclusion, poor BP control is related to the subclinical worsening of cognition and mood, which supports widespread use of the Comprehensive Geriatric Assessment even in apparently selfdependent older patients with hypertension. Conclusions With ageing of populations the subclinical and clinical cognitive and mood disturbances are increasing in frequency. We assessed BP control in a representative sample of Polish hypertensive outpatients at or above the age of 65 as a function of cognition and mood. Even at the subclinical level, worse results of screening tests for cognitive and mood disorders were in a quantifiable way associated with poor compliance, and further with greater probability of concomitant lack of adequate BP control. References 1. Global status report on non-communicable diseases 2014. Available at: http:// www.who.int/nmh/publications/ncd-status-report-2014/en/. Accessed May 8, 2016. 2. Staessen JA, Gasowski J, Wang JG, et al. Risks of untreated and treated isolated systolic hypertension in the elderly: Meta-analysis of outcome trials. Lancet 2000;355:865e872. 3. Benetos A, Rossignol P, Cherubini A, et al. Polypharmacy in the aging patient: Management of hypertension in octogenarians. JAMA 2015;314:170e180. 4. Joffres M, Falaschetti E, Gillespie C, et al. Hypertension prevalence, awareness, treatment and control in national surveys from England, the USA and Canada, and correlation with stroke and ischaemic heart disease mortality: A crosssectional study. BMJ Open 2013;3:e003423. 5. Pereira M, Lunet N, Azevedo A, Barros H. Differences in prevalence, awareness, treatment and control of hypertension between developing and developed countries. J Hypertens 2009;27:963e975. 6. SPRINT Research Group, Wright JT Jr, Williamson JD, Whelton PK, et al. A randomized trial of intensive versus standard blood-pressure control. N Engl J Med 2015;373:2103e2116. 7. Sabaté E, editor. Adherence to Long-Term Therapies: Evidence for Action. Geneva, Switzerland: World Health Organization. Available at: http://www. who.int/chp/knowledge/publications/adherence_introduction.pdf; 2003. Accessed May 16, 2016. 8. Khatib R, Schwalm JD, Yusuf S, et al. Patient and healthcare provider barriers to hypertension awareness, treatment and follow up: A systematic review and meta-analysis of qualitative and quantitative studies. PLoS One 2014;9:e84238. 9. Campbell NL, Boustani MA, Skopelja EN, et al. Medication adherence in older adults with cognitive impairment: A systematic evidence-based review. Am J Geriatr Pharmacother 2012;10:165e177. 10. Krousel-Wood M, Joyce C, Holt E, et al. Association of depression with antihypertensive medication adherence in older adults: Cross-sectional and longitudinal findings from CoSMO. Ann Behav Med 2010;40:248e257. 11. The World Alzheimer Report 2014, Dementia and Risk Reduction: An analysis of protective and modifiable factors. Available at: http://www.alz.co.uk/ research/world-report-2014. Accessed May 8, 2016. 12. Aziz R, Steffens DC. What are the causes of late-life depression? Psychiatr Clin North Am 2013;36:497e516. 13. Gottesman RF, Schneider AL, Albert M, et al. Midlife hypertension and 20-year cognitive change: The Atherosclerosis Risk in Communities neurocognitive study. JAMA Neurol 2014;71:1218e1227.
14. Launer LJ, Hughes T, Yu B, et al. Lowering midlife levels of systolic blood pressure as a public health strategy to reduce late-life dementia: Perspective from the Honolulu Heart Program/Honolulu Asia Aging Study. Hypertension 2010;55:1352e1359. 15. Stewart R, Xue QL, Masaki K, et al. Change in blood pressure and incident dementia: A 32-year prospective study. Hypertension 2009;54:233e240. 16. Meng L, Chen D, Yang Y, et al. Depression increases the risk of hypertension incidence: A meta-analysis of prospective cohort studies. J Hypertens 2012;30: 842e851. 17. Hildrum B, Romild U, Holmen J. Anxiety and depression lowers blood pressure: 22-year follow-up of the population based HUNT study, Norway. BMC Public Health 2011;11:601. 18. Prejbisz A, Klocek M, Ga˛ sowski J, et al. Factors associated with resistant hypertension in a large cohort of hypertensive patients: Pol-Fokus study. Pol Arch Med Wewn 2015;125:249e259. 19. Mancia G, Fagard R, Narkiewicz K, et al. 2013 ESH/ESC Guidelines for the management of arterial hypertension: The Task Force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens 2013;31: 1281e1357. 20. Sheikh JI, Yesavage JA. Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. In: Brink TL, editor. Clinical Gerontology: A Guide to Assessment and Intervention. New York: Haworth Press; 1986. 21. Hodkinson HM. Evaluation of a mental test score for assessment of mental impairment in the elderly. Age Ageing 1972;1:233e238. 22. Shulman KI. Clock-drawing: Is it the ideal cognitive screening test? Int J Geriatr Psychiatry 2000;15:548e561. 23. Sunderland T, Hill J, Mellow A, et al. Clock drawing in Alzheimer’s disease. A novel measure of dementia severity. J Am Geriatr Soc 1989;37:725e729. 24. Grassi G, Cifkova R, Laurent S, et al. Blood pressure control and cardiovascular risk profile in hypertensive patients from central and eastern European countries: Results of the BP-CARE study. Eur Heart J 2011;32: 218e225. 25. Llisterri JL, Rodriguez-Roca GC, Escobar C, et al, Working Group of Arterial Hypertension of the Spanish Society of Primary Care Physicians Group HTASEMERGEN, PRESCAP 2010 Investigators. Treatment and blood pressure control in Spain during 2002-2010. J Hypertens 2012;30:2425e2431. 26. Van der Niepen P, Dupont AG. Improved blood pressure control in elderly hypertensive patients: Results of the PAPY-65 Survey. Drugs Aging 2010;27: 573e588. 27. Zdrojewski T, Szpakowski P, Bandosz P, et al. Arterial hypertension in Poland in 2002. J Hum Hypertens 2004;18:557e562. 28. Vinyoles E, De la Figuera M, Gonzalez-Segura D. Cognitive function and blood pressure control in hypertensive patients over 60 years of age: COGNIPRES study. Curr Med Res Opin 2008;24:3331e3339. 29. Salas M, In’t Veld BA, van der Linden PD, et al. Impaired cognitive function and compliance with antihypertensive drugs in elderly: The Rotterdam Study. Clin Pharmacol Ther 2001;70:561e566. 30. Krousel-Wood M, Joyce C, Holt E, et al. Predictors of decline in medication adherence: Results from the cohort study of medication adherence among older adults. Hypertension 2011;58:804e810. 31. Lemstra M, Alsabbagh MW. Proportion and risk indicators of nonadherence to antihypertensive therapy: A meta-analysis. Patient Prefer Adherence 2014;8: 211e218. 32. Siegel D, Lopez J, Meier J. Antihypertensive medication adherence in the Department of Veterans Affairs. Am J Med 2007;120:26e32. 33. Yamamoto S, Mogi N, Umegaki H, et al. The clock drawing test as a valid screening method for mild cognitive impairment. Dement Geriatr Cogn Disord 2004;18:172e179. 34. Parsey CM, Schmitter-Edgecombe M. Quantitative and qualitative analyses of the clock drawing test in mild cognitive impairment and Alzheimer disease: Evaluation of a modified scoring system. J Geriatr Psychiatry Neurol 2011;24: 108e118. 35. Ehreke L, Luck T, Luppa M, et al. Clock drawing testdScreening utility for mild cognitive impairment according to different scoring systems: Results of the Leipzig Longitudinal Study of the Aged (LEILA 75þ). Int Psychogeriatr 2011;23: 1592e1601.