The importance of different frailty domains in a population based sample in England

To assess frailty, we used the 1990’s Strawbridge questionnaire [8] which was based on an early model of frailty, where the underpinning concept was, vulnerability to environmental challenge based upon complex underlying problems. Building on former studies [9‐11], the authors combined impairments across four domains of body function into a single outcome and created a multi-dimensional frailty questionnaire. The Strawbridge questionnaire is a relatively simple and user friendly instrument, and although its reliability has not been reported [4], it has established validity for postal administration and self-completion in older community-dwelling adults [4, 8, 12, 13].

The Strawbridge questionnaire [8] includes 16 items that assess frailty across four domains (physical, nutritional, cognitive, and sensory). Four items represent the physical domain (sudden loss of balance, weakness in arms, weakness in legs, dizzy when stand up quickly), two items represent the nutritive domain (loss of appetite, unexplained weight loss) and four items represent the cognitive domain (difficulty paying attention, trouble finding the right word, difficulty remembering things, forgetting where put things). The final six items represent difficulties in the sensory domain (reading newspapers, recognizing a friend across the street, reading signs at night, hearing over the phone, hearing a normal conversation, hearing a conversation in a noisy room). Our mode of administration, scoring and the final Strawbridge frailty classification was according to the original instructions [8]. Thus, for each of the 16 items, participants self-reported if they had experienced problems over the past 12 months and responses were scored: 1 (rarely or never), 2 (sometimes), 3 (often) and 4 (very often). Participants scoring ≥3 (often or very often) on at least one item in any domain were considered to have impairment/frailty within that domain. Participants were classified as frail (overall frailty) if they reported impairments in two or more domains. Missing values were not replaced, and the scoring rules are such that impairment and frailty can be assessed despite some missing values for individual items.

We collected self-report data on sex, age, ethnic group, marital status, living arrangements, age leaving full time education (years), height (feet and inches or metres), weight (stones and pounds or kilograms), cognition (clock draw test), self-rated health (SRH), chronic diseases and health conditions (angina/heart trouble; anxiety/depression; cancer; arthritis; chronic lung disease; dementia; diabetes; osteoporosis; Parkinson’s disease; urinary incontinence; stroke (right/left side)). Body Mass Index (BMI kg/m²) was calculated using weight and height converted to metric units if appropriate. We used a clock-drawing test [14] to assess global cognition (zero to six point scale whereby higher scores equate to better cognitive ability). SRH was scored on a five point scale (excellent, very good, good, fair, poor) from the 12-Item Short-Form Health Survey (SF-12) [15]. Responses for SRH were collapsed into three categories, 1 = excellent or very good; 2 = good; 3 = fair or poor.

We asked seven questions about adverse health states. Falls were defined using an internationally agreed definition, by recall over the previous 12 months [16]. Participants were asked “are you able to get out and about on foot outside the house” and “on average how many hours a day do you spend walking?” We classified people as having poor outdoor mobility if they were unable to get out and about on foot outside the house unaided. Those walking less than 1 h per day were considered to have low physical activity. We used the three-level version of the EuroQol five-dimensional questionnaire (EQ-5D-3L) [17] to identify participants with limitation in self-care (i.e. some problem or inability to wash or dress themselves), and restricted participation was defined as having some problems or being unable to perform usual activities.

Finally, we used the following two questions from the Short Form 12 Health Survey (SF-12) instrument [15]. We asked, “how much of the time during the past four weeks did you have a lot of energy?” and classed those with no or little energy as having low energy. We asked, “during the past four weeks, have you accomplished less than you would like as a result of your physical health?” and used responses to identify participants who accomplished less than they liked, all or most of the time.