Frailty in Chinese older adults with hypertension: Prevalence, associated factors, and prediction for long‐term mortality

Abstract

Hypertension and frailty are associated and often coexist in older adults. Few studies have examined the association between hypertension and frailty in Chinese population. We explored the prevalence of and the factors associated with frailty as well as whether frailty could identify patients at risk of adverse outcomes among older adults with hypertension. Data were from the Beijing Longitudinal Study of Aging. A total of 1111 hypertensive participants aged ≥60 years old who completed the comprehensive geriatrics assessment were included. All participants were followed up for 8 years. The total number of deaths was 604. Frailty was assessed by the 68‐item frailty index. Stepwise forward logistic regression was used to explore the association between the associated factors and frailty in hypertensive participants. The prediction for mortality was assessed using the adjusted Cox proportional hazards model. Two hundred and eighteen older adults were determined as frail (prevalence rate: 19.6%). Frail older adults with hypertension had worse physical performance, worse psychological, and social function, as well as worse lifestyle habits, compared to nonfrail older adults with hypertension. Chair stand test failure, balance test failure, fracture, disability, depression, and physical frailty measured with modified frailty phenotype were independently associated with frailty. Frailty was associated with a higher 8‐year mortality, hazard ratio (HR) = 3.40, adjusted for age and sex, HR = 2.61. Frailty is associated with poorer physical function and higher mortality in community‐dwelling hypertensive older adults in China. These findings emphasize the importance and need for frailty intervention and prevention in older adults with hypertension.

1. INTRODUCTION

With the population aging and the increasing life expectancy, the increasing prevalence of chronic diseases and functional impairment poses big problems for society and the health care system. One of the main health challenges among older adults is hypertension. As a major risk factor for both cardiovascular and cerebrovascular diseases, hypertension is a pivotal factor for overall mortality risk in the older population. However, an inverse association between blood pressure and cardiovascular risk was found in the oldest age groups among the elderly.1, 2, 3, 4 Considering that multiple comorbidities accompany hypertension, blood pressure control poses a significant challenge to physicians. The biological age at which individuals are referred to as frail5 is often more important than the chronological age and thus should be considered in the hypertension treatment in older adults.6

Frailty is a physiological state of increased vulnerability to inner and external stressors resulting from multi‐systemic dysregulation with aging and is associated with a higher risk of physical functional decline, institutionalization, disability, dependence, long‐term care use, as well as a reduction in life expectancy among older adults.7, 8 Hence, it is considered as another pressing health concern for older adults. The prevalence of frailty in older adults ranges from 7% to 28%, is higher in women, and increases with age.9, 10 Current literature on the association between hypertension and frailty, especially the beneficial effect of antihypertensive treatment in frail, older adults with hypertension, is open to debate, since the evidence is still scarce and based on studies with different criteria for frailty.11, 12, 13, 14, 15 Frailty assessments are clinically useful for addressing the heterogeneity of health status among older adults. The risk for frailty was associated with the presence of hypertension.11, 14, 16, 17 Moreover, hypertension treatment seems to positively affect adverse outcomes among older adults with frailty.12 Yet, there is scarce information about the prevalence of frailty among hypertensive older participants in Chinese population. Given that better control of frailty risk factors could improve prognosis, we aimed to explore the prevalence and associated factors of frailty in older adults with hypertension and whether frailty could identify patients who are at risk of increased mortality.

2. METHODS

2.1. Study population

We obtained the data for the analyses from the Beijing Longitudinal Study of Aging (BLSA), 2004‐2012. In 2004, we conducted a cross‐sectional survey based on sample data from the fourth census of Beijing using well‐established statistical sampling techniques, which included clustering, stratification, and random selection. A total of 1865 adults aged ≥60 years from a city district (Xuanwu), an outskirt (Daxing), and an exurban area (Huairou) were selected as baseline participants. A total of 1111 older adults with hypertension who completed the comprehensive geriatrics assessment (CGA) were included, 519 were male (46.7%) and 592 were female (53.3%), 602 were rural (54.2%) and 509 were urban (45.8%), with the mean age 74.68 (60‐100) years. Details of the sampling scheme have been previously published.7, 18 The study was approved by the Xuanwu Hospital's Committee on Ethics of Human Experiments. All study participants provided written informed consent prior to enrollment.

2.2. Data collection

Trained staff completed the assessments using a standard, well‐designed, and structured questionnaire. In the person‐to‐person interviews, data were collected on the following aspects: socioeconomic and demographic characteristics (including age, sex, residential area, ethnic group, occupation, educational level, marital status, monthly income, and family relationships); physical health (self‐reported history of chronic disease and clinical syndromes); physical function (activities of daily living [ADL], instrumental activities of Daily Living [IADL], and short physical performance battery); life behavior and social function (smoking, drinking, physical exercise, eating habits, sleep, and social activities); neuropsychological health (Center for Epidemiological Studies Depression [CES‐D] and Mini–Mental State Examination [MMSE]); and medical condition. Residential areas were divided into urban (Xuanwu) and rural (Daxing and Huairou). The education level of the participants was recorded as illiterate or literate. Hypertension was defined as current systolic blood pressure ≥ 140 mm Hg, diastolic blood pressure ≥ 90 mm Hg, self‐reported hypertension, or receiving antihypertensive medication.

2.3. Follow‐up and outcome ascertainment

The primary outcome of this study was all‐cause mortality. Separate follow‐up studies were conducted in 2007, 2009, and 2012. Death certificates were obtained from the provincial vital statistics offices, village committees, hospital records, medical insurance data, and interviews with family members or neighborhood. Any participants still alive were recorded as censored. The 8‐year follow‐up mortality was virtually complete. The total number of deaths in 2012 was 604.

2.4. Frailty assessment

As detailed in a previous publication, frailty was measured by CGA‐frailty index (CGA‐FI) with 68 parameters from six variables: demographic characteristics; physical health; physical function; living behavior and social function; mental health; and cognitive function.7, 10 Frailty was defined as an FI ≥ 0.25 .19, 20 The Frailty Screening Questionnaire (FSQ) with four self‐reported components (slowness, weakness, inactivity, and exhaustion) based on modified Fried criteria was used to measure the frailty phenotype .21 A score of 0 was considered robust; 1‐2 was considered prefrail; and ≥3 indicated frail.

2.5. Physical function

Physical function was assessed by balance test, chair stand test, ADL, and IADL. The balance test comprises four parts: standing unsupported for 10 seconds with the feet together; a semi‐tandem stand; full‐tandem stand; and standing unsupported for 10 seconds with the feet together and eyes closed. We considered that subjects who were unable to complete the four tests had failed the balance test. The chair stand test22 was performed with the subject seated in a chair, with the feet flat on the floor and arms held flat against their side with the elbows at 90°. We considered that subjects who were unable to stand five times stand from the chair had failed the test. ADL and IADL were each assessed in terms of 14 items; the performance of each activity was rated as being performed independently (score of 1), partially dependently (score of 2), or completely dependently (score of 3). ADL consisted of eating, bathing, dressing, toileting, indoor walking, and getting in and out of bed. IADL consisted of shopping, phone calls, housekeeping, laundry, money management, medication, and transportation. We defined fractures as spontaneous fractures over the previous 2 years. Falls were defined as occurred twice in the past year.

2.6. Psychological function

Cognitive function was measured using MMSE scale, with a total score of 0‐30. The thresholds for participants who were illiterate or who were educated up to elementary school, middle school and high school, or college or above were <17, 17‐20, 21‐22, and 23‐24, respectively.23 Participants who scored below the threshold value for their education group were regarded as cognitively impaired. Depression was measured using the CES‐D scale, with a total score of 0‐60. Scores ≥ 16 are generally consistent with clinical depression.24

2.7. Social function

Social frailty (SF) was assessed by the HALFT scale with the following 5 items: inability to help others, limited social participation, loneliness, financial difficultly, and not having anyone to talk to. Total score ranges from 0‐5 points: a score of 0 was considered non‐SF; 1‐2 was considered pre‐SF; and a score of ≥3 indicated SF.25 Literacy status was categorized as illiterate and literate. Marital status was divided into married or not married; the not married category included individuals who were divorced, widowed, and never married.

2.8. Lifestyle

Inactivity was assessed by BLSA physical activity questionnaire (BLSA‐PAQ) including the following variables: walking, gardening, low‐level exercise, and high‐level exercise. BLSA‐PAQ index (BLSA‐PAQ total score) = Walking score +Outdoor chores score + 2 × (low‐intensity exercise score) + 3 × (moderate‐intensity exercise score). Inactivity was defined as the lowest quartile of the BLSA‐PAQ index.26 Smoking was defined as a positive response to the question “do you usually smoke.” Meat intake was measured by asking participants whether they usually eat meat ≥2 times every week. Tea intake was defined by answering yes to the question “do you usually drink tea”.

2.9. Statistical methods

We performed all statistical analyses using SPSS software. Chi‐square tests were conducted for discrete variables and analysis of variance, and Student's t tests were used to compare means of the groups for continuous variables with Tukey post hoc tests. Stepwise forward logistic regression was used to explore the association between the risk factors (age, sex, living area, hypertension course, systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate, IADL disability, cognitive impairment, social frailty, marriage status, education, inactivity, smoking, meat intake, tea intake, fall, chair stand test, balance test, fracture, ADL disability, depression, and FSQ physical frailty) and frailty in participants with hypertension. Hazard ratios (HRs) for all‐cause mortality associated with frailty were calculated. Cox proportional hazards regression analysis was used to evaluate the effect of frailty on mortality. Two different Cox models were used: model 1 was not adjusted; model 2 was adjusted for age and sex. We considered P < 0.05 (two‐tailed) statistically significant.

3. RESULTS

The distribution of FI was skewed with mean values 0.17 (standard deviation [SD], 0.10) and range 0.02 to 0.59 (Figure 1A). Median FI was higher (P = 0.03, Mann‐Whitney rank‐sum test) in women (0.15; IQR, 0.10‐0.22) compared to men (0.13; IQR, 0.10‐0.21) (Figure 1B).

Figure 1.

The distribution of frailty index (FI) in older adults with hypertension. A, Histogram showing the distribution of FI among 1111 older participants with hypertension. B, The distribution of FI among older participants with hypertension in male and female

Among the 1111 older community‐dwelling adults, 218 were determined to be frail according to CGA‐FI criteria (prevalence rate: 19.6%). The prevalence of frailty increased with age and was higher in patients who lived in rural areas and who were illiterate and unmarried (Table 1). Frail old hypertensive residents exhibited worse performance in chair stand and balance test and had increased disability, falls, and fractures compared to those old hypertensive participants without frailty (Table 1). Old residents with cognitive impairment and depression had a higher prevalence of frailty. Both physically frail (defined by FSQ) and socially frail (defined by HALFT) residents were prone to have frailty. Participants with less physical activity and less meat and tea intake had a higher prevalence of frailty (Table 1).

Table 1.

Characteristics of the nonfrail and frail participants with hypertension

	n		Nonfrail	Frail	P value
Epidemiology
Age (y)	1111		73.78 ± 6.66	78.36 ± 7.01	<0.001
Sex	1111	Male	425 (81.9)	94 (18.1)	0.235
		Female	468 (79.1)	124 (20.9)
Area	1111	Rural	455 (75.6)	147 (24.4)	<0.001
		Urban	438 (86.1)	71 (13.9)
Hypertension
Course (ys)	613		3.20 ± 0.97	3.37 ± 0.97	0.065
SBP (mmHg)	1095		148.79 ± 18.47	152.15 ± 21.49	0.037
DBP (mmHg)	1095		83.00 ± 10.68	85.05 ± 12.68	0.031
HR (/min)	1111		72.91 ± 20.27	73.69 ± 9.91	0.587
Physical
Chair stand failure	1070	No	733 (95.6)	34 (4.4)	<0.001
		Yes	147 (48.5)	156 (51.5)
Balance test failure	970	No	664 (96.9)	21 (3.1)	<0.001
		Yes	91 (86.7)	14 (13.3)
ADL disability	1111	No	866 (90.0)	96 (10.0)	<0.001
		Yes	27 (18.1)	122 (81.9)
IADL disability	1109	No	615 (97.9)	13 (2.1)	<0.001
		Yes	277 (57.6)	204 (42.4)
FSQ frailty	1056		0.60 ± 0.86	3.03 ± 1.28	<0.001
		No	561 (97.6)	14 (2.4)	<0.001
		Prefrail	286 (82.9)	59 (17.1)
		Frail	32 (23.5)	104 (76.5)
Psychological
CI	1067	No	775 (88.4)	102 (11.6)	<0.001
		Yes	107 (56.3)	83 (43.7)
Depression	1047	No	792 (89.2)	96 (10.8)	<0.001
		Yes	81 (50.9)	78 (49.1)
Social
SF	1039		0.75 ± 0.93	1.90 ± 1.16	<0.001
		No	447 (96.1)	18 (3.9)	<0.001
		Pre‐SF	384 (78.7)	104 (21.3)
		SF	41 (47.7)	45 (52.3)
Married	1111	Yes	597 (83.7)	116 (16.3)	<0.001
		Not married	296 (74.4)	102 (25.6)
Education	1111	Illiterate	365 (74.3)	126 (25.7)	<0.001
		Unilliterate	528 (85.2)	92 (14.8)
Lifestyle
Inactivity	1107	No	763 (89.1)	93 (10.9)	<0.001
		Yes	129 (51.4)	122 (48.6)
Smoking	1111	No	569 (81.1)	133 (18.9)	0.457
		Yes	324 (79.2)	85 (20.8)
Meat intake	1110	Yes	610 (83.3)	122 (16.7)	0.001
		No	282 (74.6)	96 (25.4)
Tea	1108	Yes	474 (87.5)	68 (12.5)	<0.001
		No	418 (73.9)	148 (26.1)
Outcome
Fracture	1111	No	871 (81.3)	201 (18.8)	<0.001
		Yes	22 (56.4)	17 (43.6)
Fall	1105	No	813 (84.0)	155 (16.0)	<0.001
		Yes	75 (54.7)	62 (45.3)

ADL, activities of daily living; BMI, body mass index; CI, cognitive impairment; DBP, diastolic blood pressure; FSQ, Frailty Screening Questionnaire; HR, heart rate; IADL, instrumental activities of daily living; SBP, systolic blood pressure; SF, social frailty.

Forward stepwise logistic regression showed the independent risk factors for frailty in old, community‐dwelling adults with hypertension, adjusted for age, sex, living area, hypertension course, SBP, DBP, heart rate, IADL disability, cognitive impairment, social frailty, marriage status, education, inactivity, smoking, meat intake, tea intake, and fall. These factors are chair stand test failure (odds ratio OR [95%CI], 4.27 (1.53‐11.88]), balance test failure (OR [95%CI], 2.96 [1.13‐7.77]), fractures (OR [95%CI], 7.89 [1.22‐51.84]), ADL disability (OR [95%CI], 7.34 [2.14‐25.15]), depression (OR [95%CI], 17.43 [6.57‐46.26]), and FSQ frailty (OR [95%CI], 9.74 [3.280‐28.95]) (Table 2).

Table 2.

Forward stepwise logistic regression of factors associated with frailty in older participants with hypertension

Variables	Coefficient	Standard error	P value	95% CI
Chair stand test	4.27	0.52	0.005	1.53‐11.88
Balance test	2.96	0.49	0.027	1.13‐7.77
Fracture	7.89	0.96	0.031	1.20‐51.84
ADL disability	7.34	0.63	0.002	2.14‐25.15
Depression	17.43	0.50	<0.001	6.57‐46.26
FSQ frailty	9.74	0.56	<0.001	3.28‐28.95

ADL, activities of daily living; CI, confidence interval; FSQ, Frailty Screening Questionnaire.

The variables not in the equation were age, sex, living area, hypertension course, SBP, DBP, heart rate, IADL disability, cognitive impairment, social frailty, marriage status, education, inactivity, smoking, meat intake, tea intake, and fall.

Estimates of the frailty HR for mortality were conducted by Cox proportional hazards model and adjusted for age and sex. Frailty was associated with a higher 8‐year mortality in patients with hypertension, HR = 3.40 (95% CI: 2.77‐4.17). Adjusting for age and sex did not alter these findings, HR = 2.61 (95% CI: 2.11‐3.23) (Figure 2).

Figure 2.

Survival curve for hypertension and frailty in older adults. A, not adjusted; B, Adjusted for age and sex. Hazard ratios and associated 95% CI from Cox regression models showing the effect of adjusting for frailty on the estimate of treatment effect in those for whom a frailty index was calculated n = 1111

4. DISCUSSION

Our study shows that the mean FI for this subset of subjects was about 0.17, similar to that of HYVET study (0.16‐0.17).12 Older individuals with the same chronological age vary largely in health and functional ability. Frailty, characterized by decreased physiologic reserve, is closely associated with biological age, morbidity, and decreased survival in older adults,7, 27 indicating the need for an individualized approach.28 Older adults have an increasing burden of comorbidity, especially those frail older adults, leading to higher risk of polypharmacy. A U‐shaped relationship between blood pressure and mortality exists in both cognitively and functionally impaired patients.29 Polypharmacy contributes to drug interactions and increases the risk of adverse effects of antihypertensive medication. Accordingly, those who are frail, with comorbidities, and polypharmacy should be treated differently.30 However, no evidence of an interaction between treatment effect and frailty was shown in HYVET study.12 The Systolic Blood Pressure Intervention Trial (SPRINT) study revealed frail older patients could also benefit from intensive blood pressure lowering,31 very recently, and it was criticized as the included patients were very lightly frail, therefore, frailty degree is important for the harms‐benefits balance of antihypertensive medication in patients with frailty.32

The present study found that the prevalence of frailty in hypertensive older adults was high in rural areas, which was in accordance with our previous studies.10 Participants in rural areas in China have a poor financial situation, poor nutrition, and poor hypertension control,18 these factors could lead to worse physical function. We have reported the prevalence of frailty in Chinese community‐dwelling population to be 7.1% and 9.9% in our previous studies,10, 21 moreover, the prevalence of frailty in older hypertensive adults was much higher (19.6%) in this study. A study found that 80% of hypertensive outpatients were frail; furthermore, frail patients showed lower blood pressure values and might easily be over‐treated.33 Although the above study has a small sample size (56 hypertensive old outpatients), the results of both the above study and our study indicate that frail patients with hypertension need more geriatric attention.

This study shows that frailty is associated with chair stand test failure, balance test failure, fracture, disability, depression, and physical frailty in older, community‐dwelling participants with hypertension. Previous studies found that frailty was associated with limited physical activity and ADL difficulty in hypertensive patients.11 However, decreased ambulation as well as higher blood pressure levels were associated with a lower risk of death in very old patients.34 An excessive reduction in blood pressure in frail subjects is dangerous, and positive outcomes are associated with higher blood pressure values indicate the need for a person‐tailored interventions.33 Since the prevalence of frailty increases with age,35 a systematic screening for frailty was proposed for very old people by the Working Group for the Management of Hypertension36 to improve the therapeutic effect and promote better personalization of interventions.

Our results showed that frailty is associated with 8‐year mortality in older adults with hypertension. Population‐based studies have consistently found an association between higher blood pressure and lower rates of mortality in older people.37, 38 Higher blood pressure was associated with a lower risk of stroke in subjects aged 85 years and older with comorbidities.39 Frailty was independently and additively associated with left ventricular hypertrophy and reduced global longitudinal strain; greater left atrial volume index and abnormalities of cardiac structure and function together show the greatest association with frailty.40 Although we did not observe the cardiovascular outcomes in the present study, in a study conducted in Korean older population, frailty was associated with the adverse cardiovascular outcomes.41 Considering all the above, more attention should be paid to the blood pressure control in the frail older adults with hypertension.

As FI is a multidimensional indicator, some variables associated with frailty were included in CGA‐FI; furthermore, variables related to cardiovascular diseases which could be a major driver of mortality were also included in CGA‐FI in the present study. Thus, further studies on the association between frailty and hypertension using frailty phenotype criteria are needed. Another limitation of this study is that polypharmacy and ambulatory blood pressure measurement were not included in the data analysis. In spite of these limitations, one strength of our study is the long‐term follow‐up and the representative of BLSA as the Chinese population, which provided the most convincing results of the prevalence of frailty and its predictive power for mortality. The other strength is that psychological and social functions of the patients were assessed through logistic regression, which showed that most of the independent risk factors of frailty are due to physical function, which highlights the importance of intervention in the physical domain in older adults with hypertension.

5. PERSPECTIVES

Frailty is associated with hypertension in old Chinese adults. Poorer physical function and higher mortality were found in community‐dwelling hypertensive older adults with frailty, which emphasizes the need for frailty intervention and prevention.

CONFLICT OF INTEREST

The authors have no conflict of interests and no financial associations to disclosure.

AUTHOR CONTRIBUTIONS

LM and ZT contributed to the design of the work. LM drafted the manuscript and wrote it together with YL and ZT. LZ and FS contributed to the analysis and interpretation of data. All the authors contributed to writing the paper and revising it critically and gave final approval of this version.

Ma L, Zhang L, Sun F, Li Y, Tang Z. Frailty in Chinese older adults with hypertension: Prevalence, associated factors, and prediction for long‐term mortality. J Clin Hypertens. 2018;20:1595–1602. 10.1111/jch.13405