Materials and Methods
Sample
The data are from wave 6 (2012/2013) of ELSA, a nationally representative panel survey of community-dwelling men and women aged 50 years and older in England. Full details on the study design and methods have been described previously. Data collection consisted of a face-to-face interview and self-completion questionnaires. A total of 10 601 individuals participated in wave 6, with 7079 (67%) completing and returning the paper-based SRA-Q. This analysis included only men with complete data on erectile function and PDE5i usage, and was additionally restricted to the age range covered by the PDE5i users (51–87 years), leaving a total sample of 2612. ELSA wave 6 received ethical approval from the NRES Committee South Central—Berkshire, and all participants were provided with a letter and leaflet to allow them to give informed consent.
Sexual Relationships and Activities
The ELSA SRA-Q includes questions on attitudes to sex, sexual activities and behaviours, problems and concerns, and details about current sexual partnerships. Items included in the SRA-Q were taken from validated instruments, with minor modification to ensure gender specificity, and chosen to maximise harmonisation with both the US National Social Life, Health, and Aging Study (NSHAP) and the third UK National Survey of Sexual Attitudes and Lifestyles (Natsal-3). Face validity of the ELSA SRA-Q was assessed by cognitive interviews among a sample (n=45) of the ELSA pilot/dress rehearsal panel. The cognitive testing was carried out by the NatCen Social Research (NatCen) and interviews were conducted by specialist cognitive interviewers. The full range of sexuality measures assessed in the SRA-Q, and cross-sectional associations with demographic, lifestyle and health factors have been described previously.Table 1 summarizes the items from the SRA-Q presented in this analysis. Men were also asked if they had used any oral medications (Viagra, Cialis, Levitra) to enhance their erections over the past 3 months. Those answering yes were additionally asked to rate the effect of these medications on their sex life (5-point Likert scale: response set 'very positive', 'positive', 'neither positive nor negative', 'negative', 'very negative'). The SRA-Q was completed in private and the questionnaire sealed in an envelope upon completion. The male and female versions of the SRA-Q are freely available from the ELSA website (http://www.elsa-project.ac.uk/).
Other Assessments
Participants were asked about their current living arrangements, general health and lifestyle factors during the face-to-face interview. Specifically, they were asked whether a doctor had ever told them they had any of several common conditions, including hypertension, arthritis, cardiovascular diseases, diabetes and asthma. Self-rated health was ranked on a five-point scale (excellent, very good, good, poor or fair), smoking status was recorded as current or non-smoker, and typical frequency of alcohol consumption over the past year as never or rarely (never–once or twice), regularly (once every 2 months–twice a week) or frequently (3 days a week–almost every day). Depressive symptoms were assessed using the eight-item version of the Center for Epidemiologic Studies Depression scale, with a score of four or more indicative of risk for depression.
Statistical Analysis
All analyses were conducted using STATA SE v13.1 (StataCorp, College Station, TX, USA). Weights were used to correct for sampling probabilities and differential non-response, including to the SRA-Q, and to calibrate back to the 2011 Census population distributions for sex and age. Specifically, these weights accounted for (i) the differential probability of being included in the wave 6 sample and (ii) for non-response to the SRA-Q instrument (full details available from http://www.elsa-project.ac.uk/). Multiple logistic regression was used to examine the association of PDE5 inhibitor use with sexual activities, function and problems among men, with results expressed as odds ratios (ORs) and 95% confidence intervals (CIs).