Lifestyle Changes and Blood Pressure Control: A Community‐Based Cross‐Sectional Survey (2006 Ontario Survey on the Prevalence and Control of Hypertension)

Abstract

To evaluate lifestyle changes and their impact on hypertension control in a sample of hypertensive respondents in Ontario, Canada, diet, physical activity, and other nonpharmacologic measures were recorded using a structured questionnaire during the 2006 Ontario Survey on the Prevalence and Control of Hypertension. Responses were weighted to the total adult population of 7,996,653 in Ontario. The prevalence of hypertension was 21%; 42% of hypertensive persons received therapy with antihypertensive drugs and lifestyle changes, and 41% received therapy with drugs only. Blood pressure was controlled in 85% of respondents who used only drugs and in 78% of those who stated that they received therapy with combined drug treatment and lifestyle changes. Fewer than half of hypertensive respondents practiced lifestyle changes (in combination with drug treatment) for blood pressure control. Lifestyle measures in addition to medication use did not result in better control of hypertension compared to only medication use.

Lifestyle modification is recommended as the first step in the management of hypertension by Canadian and international guidelines. Salient features of lifestyle changes include reduction of excessive body weight, a Dietary Approaches to Stop Hypertension (DASH)–like diet and sodium reduction, moderate physical activity, limiting alcohol consumption, and stress management. The beneficial effect of weight reduction in lowering blood pressure (BP) is well established. ¹ The DASH diet ² and others have provided evidence for the efficacy of a low‐sodium diet as well as a diet with fruits and vegetables and low‐fat dairy products in reducing BP among individuals with and without hypertension. Likewise, regular aerobic exercise may lead to BP lowering in hypertensive patients. ³ With respect to the intake of alcohol, excess alcohol consumption increases the risk of hypertension, ⁴ while prolonged abstinence or reduction of an excess to a moderate intake reduces elevated BP levels. ⁵ Last, individualized stress management strategies may significantly reduce BP. ⁶ Smoking cessation, which is an integral part of lifestyle counseling aimed to prevent cardiovascular diseases, lacks consensus regarding its long‐term effects on BP.

While the efficacy of lifestyle interventions in the management of hypertension is generally acknowledged, there is a dearth of information about the extent of adoption of these nonpharmacologic measures in the general population of hypertensive patients. The Canadian Heart Health Survey (CHHS), ⁷ which was conducted from 1989 to 1991, found that 8% of respondents reported that lifestyle measures were used exclusively to control hypertension, while 14% combined nonpharmacologic treatment with the use of antihypertensive medications.

Recently, Neutel and Campbell ⁸ reported changes in lifestyle habits in patients with newly diagnosed hypertension over a period of 8 years in a cohort of 9538 respondents who participated in the National Population Health Survey in Canada. During this time, a total of 1281 respondents received new diagnoses of hypertension. The main and most significant lifestyle modification adopted among participants with newly diagnosed hypertension was smoking cessation (27.0% to 22.1%; P<.001); also, an increase was reported in physical activity (56.1% to 60.9%; P<.01). Paradoxically, respondents with newly diagnosed hypertension showed a significant increase in obesity after the diagnosis. In this survey, the impact of lifestyle measures on BP was not studied.

As part of the 2006 Ontario Survey on the Prevalence and Control of Hypertension (ON‐BP), we evaluated lifestyle changes as practiced by hypertensive respondents in this Province. ⁹ ON‐BP is a community‐based cross‐sectional study that used direct observations to assess the current prevalence of hypertension and its management in this most populous province of Canada. The survey included questions regarding lifestyle measures adopted by hypertensive individuals. Thus, we were able to assess the extent of nonpharmacologic measures as reported in the hypertensive population and evaluate the impact of these interventions on BP control.

Methods

ON‐BP, conducted from March to October 2006, consisted of a random and representative sample of the adult population (20–79 years) in the province of Ontario. The aim was to recruit 2500 respondents. Specific details regarding survey sampling design and methodology were recently described elsewhere. ⁹ , ¹⁰ Survey participants were interviewed in their homes by an experienced interviewer using a structured questionnaire. Participants were then invited to attend a clinic visit in which BP measurements were taken and anthropometric data, such as height, weight, and waist/hip circumferences, were collected by a trained nurse. During the in‐home interview, participants were asked a specific set of questions pertaining to BP history, including diagnosis and management of hypertension. Participants were also asked whether they were using any nonpharmacologic and/or lifestyle treatment(s) for BP control. The possible list of answers were (1) diet (dietary modifications such as decreasing salt and fat intake), (2) physical activity, (3) medications, (4) herbal remedies, (5) reduced alcohol consumption, (6) other, and (7) no treatment. Participants were free to choose all the options that applied.

A standardized protocol was followed to ensure proper BP measurement technique. In preparation for their clinic visit, participants were instructed to avoid eating, drinking caffeinated beverages, and smoking 30 minutes before attending the clinic visit. Participants were also asked not to engage in vigorous physical activity 2 hours before their visit and to abstain from consuming any alcohol for 8 hours. During the clinic visit, participants (1) were instructed to empty their bladder, (2) had their left arm circumference measured to determine proper BP cuff size, (3) were seated in a comfortable chair with both feet flat on the ground and the left arm resting on a pillow at the level of the heart, and (4) were left alone to rest quietly for a period of 5 minutes prior to having their BP measured. BP was measured using the automated BpTRU (BpTRU Medical Devices Ltd., Coquitlam, British Columbia, Canada) instrument, which was validated by Wright and associates. ¹¹ In every 10th participant, BP was measured using both the BpTRU and mercury sphygmomanometer, alternating the device used first between participants. When measuring BP with the BpTRU, the first reading, which was observed to ensure proper functioning of the device, was discarded. Five readings, taken at 1‐minute intervals while the participant was left alone, were then averaged. BpTRU values were adjusted to be consistent with mercury sphygmomanometer readings in order to compare findings with those of other studies utilizing sphygmomanometer measurements exclusively. Thus, a linear regression equation derived from the subsample of participants who had their BP measured with both instruments was used to adjust BpTRU measurements recorded for this survey. ¹²

As in the main publication of the survey findings, ⁹ hypertension was defined as a systolic BP ≥140 mm Hg and/or a diastolic BP ≥90 mm Hg or treatment with antihypertensive drugs. “Treatment” was defined as using antihypertensive drugs or antihypertensive drugs and lifestyle changes. Of the 2992 respondents who agreed to participate in ON‐BP, 2551 attended the study clinic and had their BP measured. According to the definition of hypertension, 733 respondents were classified as being hypertensive.

Given the complexity of the sampling plan and of the adjustments to the estimation weights, sampling errors were estimated using a replication method known as “jackknife.” ¹³ , ¹⁴ Replication weights were created using the WesVar 4.2 software (Westat, Rockville, MD); the same software was used to create all cross‐tables. The chi‐square analysis was used to test for differences in proportions by age, ethnicity, sex, treatment, and control. BP values are presented as weighted means (and standard errors) and t‐test was used to test for differences in BP values. The coefficient of variation is calculated as the standard deviation of an estimate divided by its mean. Estimates with coefficients of variation >0.33 or cells with a sample size <10 were considered unreliable ¹⁵ and marked with a superscript “c” (^c) in the tables.

Results

The overall prevalence of hypertension was 21%. In this hypertensive population, 83% reported that they were receiving treatment for their elevated BP.

Forty‐one percent reported using only antihypertensive medications for their BP, while 42% of hypertensive participants reported that they also adopted lifestyle changes in addition to using medications as part of their treatment (Table I). The remaining hypertensive participants (17%) did not report any type of treatment. No sex‐ or ethnicity‐related differences with respect to treatment modality were found (Table I). Among the middle‐aged group (40–59 years), a combination of medications and lifestyle changes tended to be more frequent compared with treatment with medications only (P=.053).

Table I.

 Distribution of Treatment Modality in the Hypertensive Population by Age, Sex, and Ethnicity^a

	Drugs Only	Drugs and Lifestyleb	No Treatment
Overall	41 (3.0)	42 (3.2)	17 (2.1)
Age, y
20–39	59 (21)c	8 (5.1)c	33 (17.8)c
40–59	33 (4.5)	46 (4.5)	21 (3.5)
60–79	45 (5.2)	42 (5.1)	12 (2.6)
Sex
Male	40 (4.1)	44 (3.9)	17 (2.9)
Female	42 (4.7)	40 (4.1)	18 (3.7)
Ethnicity
White	40 (3.8)	42 (3.5)	17 (2.4)
Black	38 (9.6)	52 (10.1)	10 (4.0)c
East Asian	53 (10.2)	26 (8.2)	21 (8.9)c
South Asian	30 (8.8)	55 (11.7)	15 (8.5)c

Values are percentages (SE). ^aAll estimates are weighted to represent the Ontario hypertensive population (7,996,653). ^bLifestyle indicates diet, exercise, and other (eg, herbal, reduced alcohol consumption). ^cEstimate with coefficient of variation >0.33. Treatment by age groups: P=.053.

Table II depicts the level of BP control by treatment modality. No significant difference regarding BP control and BP values was found between hypertensive persons who were relying only on medications for their treatment and those who had added lifestyle changes to their BP medications. The average number of antihypertensive medications taken by these 2 groups was practically identical (1.7 [0.1] vs 1.6 [0.1], respectively; P=.623).

Table II.

 Treatment Modality and Control of Hypertension^a

	Controlled	Uncontrolled
Drugs only	85 (3.3)	15 (3.3)
SBP	118 (1.6)	151 (2.0)
DBP	72 (1.2)	84 (1.5)
BMI	32 (1.2)	33.8 (2.2)
Drugs and lifestyle	78 (4.4)	22 (4.4)
SBP	120 (1.5)	145 (4.0)
DBP	75 (1.3)	87 (3.1)
BMI	31 (0.5)	34b (1.1)

Values are percentages (SE). Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; SBP, systolic blood pressure. ^aAll estimates are weighted to represent the Ontario hypertensive population (7,996,653). ^b P=.057 vs the corresponding value for the controlled group.

Regardless of the BP treatment modality, all groups of hypertensive patients had a high prevalence of obesity (body mass index [BMI] ≥30 kg/m²) (Table II). In both groups with uncontrolled BP, BMI was similarly higher compared with the groups of patients in whom BP was controlled. Furthermore, patients with uncontrolled hypertension who were receiving antihypertensive therapy and who practiced lifestyle changes tended to have higher BMIs than those in the same group in whom hypertension was controlled (34 [1.1] vs 31 [0.5], respectively; P=.057).

The most frequently adopted lifestyle change, as reported, was diet (86%). Other lifestyle changes included some physical activity (60%) and, in 13% of cases, other nonpharmacologic measures.

Discussion

The efficacy of lifestyle changes has been established in controlled trials. However, when comparison is made between drug treatment and lifestyle changes, the outcomes of lifestyle interventions in “real life” are disappointing. Nicolson and colleagues ¹⁶ carried out a systematic review of 5 randomized trials comparing treatment with drugs and lifestyle changes and reported that the results regarding the effect of lifestyle changes on BP control were inconclusive. This analysis confirmed similar findings of an earlier study by Ebrahim and Smith. ¹⁷

The ON‐BP study ascertained that 42% of hypertensive respondents in Ontario reported that they used various lifestyle maneuvers to control their BP. This number represents a significant change in patients’ attitudes compared with 20 years ago, when 22% of hypertensive persons participating in the CHHS reported changes in their lifestyle. ⁷

In the ON‐BP hypertensive population, the most frequently utilized lifestyle modification was diet change, followed by increased physical activity. However, in this population, it was shown that practicing lifestyle changes did not add a beneficial effect to BP control compared with antihypertensive treatment with medications only. This finding is also reflected by similar levels of BP and an equal number of prescribed antihypertensive drugs in the 2 groups of hypertensive participants.

When speculating as to why the combination of lifestyle changes and antihypertensive medications did not result in better BP control compared with drugs alone, several possible reasons may be considered:

• • Hypertensive individuals who adopted lifestyle measures may have been prescribed less medication compared with those who did not adopt lifestyle changes as part of their treatment. For the ON‐BP study population, however, we ascertained that this was not the case, as both groups of hypertensive respondents had the same number of drugs prescribed.
• • Another possibility is that individuals using both drugs and lifestyle measures are less compliant in taking the prescribed medication and rely more heavily upon lifestyle changes to control their hypertension.
• • Perhaps some physicians are reluctant to increase the number and dosage of antihypertensive medications when facing a patient who is treated with drugs but has poorly controlled hypertension. In this case, they may attempt to complement the therapeutic effects of medication with lifestyle modification.
• • Finally, it is likely that the lifestyle measures as practiced by the respondents of the survey did not reach the standards set by efficacy trials and guideline recommendations and, thus, had only minimal impact on BP.

These speculations may explain why the 2 groups of hypertensive patients also had similarly high BMIs. In this population, practicing lifestyle changes such as diet change and increased physical activity did not result in a lower BMI compared with hypertensive persons who relied strictly on medications for BP treatment.

The main limitation of the present study is that the assessment of treatment for BP, including adoption of lifestyle changes and medication use, was based on self‐reported information. In this respect, as with any study of a similar nature (ie, a cross‐sectional questionnaire‐based survey), the findings should be interpreted with caution.

When reflecting on clinical guidelines that recommend lifestyle measures for BP management, the fact that nondrug interventions are neither cheap nor necessarily harmless is ignored. ¹⁸ In the PREMIER randomized trial, ¹⁹ 2 multicomponent lifestyle interventions (one of them combined with the DASH diet) were studied for their effect on BP levels among 810 adults with above‐optimal BP and stage 1 hypertension. In the PREMIER study, the participants had to manage their diet themselves, unlike in the original efficacy study of the DASH diet project ² in which respondents were provided with the experimental food free of charge. As part of the behavioral interventions, participants had to attend 18 counseling sessions in the first 6 months (14 group meetings and 4 individual sessions). During the 12‐month follow‐up, patients had an additional 12 group meetings and 3 individual sessions. Despite the costly effort and an initial significant reduction in the prevalence of hypertension at 6 months, the prevalence of hypertension increased again (differences between intervention and nonintervention groups were nonsignificant) at an 18‐month follow‐up of these participants. The DASH diet was successful as long as food was provided to the study paticipants. As soon as the respondents had to take care of their diet themselves (as in the PREMIER study ¹⁹ ; ie, a real‐life situation), the beneficial effects of this diet diminished or disappeared entirely.

Relying on the effectiveness of lifestyle changes can also lead to the postponement of efficacious drug treatment with serious consequences. As Moser ²⁰ pointed out, “Prolonging lifestyle interventions in the hope that some benefit may accrue may, in view of available data on the effects of prolonged BP elevation, be a poor strategy.” We share this opinion.

Conclusions

ON‐BP, a community‐based cross‐sectional survey, found that nearly one‐half of hypertensive individuals state that they practice some form of lifestyle strategy to control their BP. Hypertensive respondents who combined drug treatment with lifestyle changes did not have better BP levels or body weight control compared with those who were solely receiving drug treatment. We conclude that lifestyle changes, as practiced in the community, appear to be ineffective in controlling hypertension in the general, free‐living population of Ontario.