Prevalence of postural hypotension in primary care, community and institutional care settings: systematic review and meta-analysis

Problem

Postural hypotension (PH), the reduction in blood pressure (BP) when rising from sitting or lying to standing, is a major risk factor for falls, cognitive decline and mortality. Prevalence estimates for PH vary greatly, depending on the definition, population, care setting and BP measurement method. PH is often asymptomatic and overlooked in primary care. Awareness of prevalence of PH in the community, and in specific populations, could contribute to improved detection and management in primary care.

Aim

To estimate the prevalence of PH for primary care, community and care home populations across different disease subgroups.

Approach

Systematic review and meta-analysis. We searched Medline and Embase to July 2017 for cross-sectional and cohort studies representative of primary care, community or care home settings which reported prevalence of PH. Data on prevalence, definition of PH used, measurement method and study level demographics were extracted to a standard proforma by two reviewers. Pooled estimates for mean prevalence of PH were calculated and compared between care settings and disease subgroups using meta-analyses in random effects models. Statistical heterogeneity of pooled prevalence estimates was quantified using I2 statistics and explored using sensitivity analyses and meta-regression of study level variables.

Findings

1497 unique citations reporting PH were screened, 302 full texts reviewed and 83 contributed to the review. We pooled data from studies that met the consensus definition for PH (decrease in systolic BP by ≥20 mmHg or diastolic BP by ≥10 mmHg within 3 min of standing). Pooled prevalences of PH were 15% (95%CI 13 to 18; I2 =99%) within community samples, 18% (13 to 24; I2 =98%) in primary care and 26% (18 to 35; I2 =93%) in care home settings (P=0.04 between groups). For subgroups according to chronic disease, pooled prevalences of PH were 19% (16 to 23; I2 =98%) in hypertension, 21% (16 to 26; I2 =92%) in diabetes, 27% (15 to 40; I2 =93%) in Parkinson’s disease and 28% (24 to 32; I2 =0%) in dementia, compared with 13% (11 to 15; I2 =99%) for those without these conditions (P<0.01 between groups). Univariable meta-regression identified female gender, increasing age and supine systolic BP as predictors of higher prevalences of PH.

Consequences

Our results demonstrate substantial background prevalences of PH in community and primary care settings, which are greater for persons residing in institutions and those with chronic diseases. Care setting and disease specific prevalences of PH should be taken into account when assessing patients. We previously reported low rates of testing for PH in primary care; these data suggest that asymptomatic PH could frequently go undetected, risking oversight of an avoidable cause of morbidity. These figures should help to inform future strategies for detection and management of PH.