Some experiments with factors that might affect the response of mothers to a postal questionnaire

The effects of a number of controllable factors on the response rate of mothers to a postal questionnaire were assessed by a series of experiments. In one a factorial design was used to look at seven factors with a random sample of 1600 mothers. The main findings were that the responses to questionnaires of 8, 16 or 24 pages were similar, but that rather more mothers replied when the questionnaires contained only factual questions than when they covered both facts and attitudes. Another experiment, based on a similar sample, compared the response rates when the questionnaires were sent out by a government organization or by the Institute for Social Studies in Medical Care, and found no difference. The final experiment looked at the effect of precoding the answers or asking respondents to tick boxes beside their replies and again found no difference in response rates. The overall response rate was 79 per cent, but this varied between 66 and 87 per cent in the ten randomly selected study areas--a much greater variation than that found with any of the experimental factors.     

Dietary behaviours of volunteers for a nutrition education program,compared with a population sample

Abstract: The dietary behaviours of and recent dietary change by volunteers for a nutrition education program were compared with those of a more population-representative sample. The population sample was randomly selected from the electoral rolls of three Australian cities. Those selected received questionnaires which were to be completed and returned by mail. The sample of volunteers was recruited from the electoral rolls of suburbs of either high or low social status in one of these cities. Volunteers were posted a questionnaire, to be returned in person. A quantified food frequency questionnaire was used to estimate relative intake of 19 nutrients, plus energy intake. Occupation, age, sex, reported recent dietary change and diet-related beliefs were also assessed. The population sample was weighted to the age and occupational distribution of the education program sample. Compared to the 874 respondents in the population sample (70.4 per cent response rate), the 487 volunteers (24.2 per cent response rate) for the nutrition education program had healthier nutrient intakes and reported more dietary behaviour changes. Recruitment in the education program was greater in areas of higher social status (32 per cent) than in areas of lower social status (20 per cent). The potential effect of such a program on the whole population was demonstrated by the proportions that volunteered and the characteristics of these volunteers. The need to provide a range of opportunities for changing dietary behaviour, according to the health-related behaviours and beliefs of target groups and their socioeconomic circumstances, was highlighted.     

The accuracy of general practitioner records of smoking and alcohol use: comparison with patient questionnaires

BACKGROUND: General practitioner (GP) records are increasingly being used as sources of information on potential confounders such as smoking use and alcohol intake in epidemiological studies. The aim of this study was to assess the accuracy of GP records on smoking use and alcohol intake compared with data from patient questionnaires. METHODS: Patients registered with 42 practices in Oxfordshire that agreed to take part in a post-marketing surveillance study of omeprazole were sent a postal questionnaire that included questions about alcohol and tobacco use. Two years later, data on these aspects of lifestyle were abstracted from the GP records. RESULTS: A total of 892 patients agreed to take part in the study; 804 (90 per cent) completed the postal questionnaire, and the records of 856 (96 per cent) were reviewed. Information on smoking and alcohol use was present in 74 per cent and 63 per cent of GP records, respectively. Agreement between the two data sources was moderate for both smoking (kappa = 0.50) and alcohol use (kappa = 0.52). With regard to smoking, the main discrepancy between the two data sources was that 46 per cent (94/206) of patients who reported themselves as exsmokers were recorded as being never smokers in the GP record. With regard to alcohol, there were no systematic differences between the two data sources. CONCLUSION: Data from GP records on smoking status and alcohol use are incomplete and subject to some misclassification. This is a source of potential failed adjustment for confounding, which should be considered in epidemiological studies that make use of these records.     

A new, brief questionnaire (PEQ) developed in primary health care for measuring patients' experience of interaction, emotion and consultation outcome

BACKGROUND: A deepened understanding of patients' perspectives is essential in order to improve medical communication. By changing focus from patient satisfaction to patient experiences, more immediate, personal and affective responses may be captured. OBJECTIVE: Our aim was to develop a new consultation-specific questionnaire on patient experiences. METHODS: The questionnaire was developed in Norwegian primary care in three main phases. Phase 1: focus groups with patients in order to identify important aspects of patients' experiences, and their words and language when describing such experiences. Phase 2: a questionnaire survey with 110 items including 660 patients. Extensive testing resulted in a reduction to 25 items on six dimensions. Phase 3: a questionnaire survey with 25 items including 1092 patients. Psychometric analyses and feedback from patients and physicians involved dimensionality and tests of validity and reliability. RESULTS: A final questionnaire was produced with 18 items on five dimensions: communication; emotions; short-term outcome; barriers; and relations with the auxiliary staff. The validity and reliability estimates were highly satisfactory. Three scales were skewed while two were more equally distributed. Forty-eight per cent of the patients described less than optimal communication experiences; some communication barriers were detected in 70% of the visits and less helpful experiences with the staff were reported in 55% of the visits. Twenty-four per cent of patients left with no positive feelings, and 48% scored low on the outcome scale (knowledge, perceived result). CONCLUSIONS: The patient experience questionnaire (PEQ) emphasizes what patients value the most, i.e. interaction, emotions and outcome, and may represent a valuable tool for doctors who want feedback from their patients on the function of their doctor-patient relationships.     

An assessment of the construct validity of the SF-12 summary scores across ethnic groups

BACKGROUND: The aim of the study was to determine the construct validity of the 12-item Short Form health survey questionnaire (SF-12) across ethnic groups in a large community sample of the United Kingdom. METHODS: A postal survey was carried out in English using a questionnaire booklet, containing the SF-12 and a number of other items relating to experiences of chronic illness and utilization of health care services. The dataset was the National Survey of NHS Patients. The sample consisted of 1000 residents within each Health Authority in England who were randomly selected from the electoral registers, giving an initial sample of 100000. RESULTS: A total of 61 426 (61.4 per cent) questionnaires were returned; 94.3 per cent of respondents classified themselves as white and 5.7 per cent classified themselves as members of other ethnic groups. Construct validity of the SF-12 was assessed by comparing results from the two summary scores (the Mental Health Component Summary (MCS) score and the Physical Health Component Summary (PCS) score) with overall self-assessed health and limiting longstanding illness. Although there were generally consistent patterns of association between overall self-assessed health or limiting longstanding illness and the MCS and PCS scores in all the ethnic groups, there were significant differences between the MCS and PCS scores of Indians, Pakistanis and Bangladeshis who understood English fluently and those who did not. Furthermore, there were differences in the completion rates of the SF-12 between ethnic groups and a reversal of the general pattern of increasing MCS scores with increasing age in Bangladeshis. CONCLUSION: The results indicate that the use of the SF-12 to measure the health of ethnic minorities seems acceptable in most instances, but may prove problematic in those instances where respondents complete the questionnaire via an untrained translator, such as a friend or family member. The systematic differences in MCS and PCS scores between ethnic minorities who understood English fluently and those who did not suggest that the meaning of specific SF-12 items may change when informally translated. Future research using the SF-12 to measure the health status of ethnic minorities in the United Kingdom via postal surveys must include questions on whether respondents completed the questionnaires via informal translations. In general, those wishing to measure the health of members of ethnic groups who are unable to read English might consider using different techniques to gain the information from these groups.     

Assessment of self-report in HIV surveillance: a pilot study

Abstract: The basis of HIV exposure category classification was investigated among selected cases of newly diagnosed HIV infection. Questionnaires seeking specific information on patient-reported exposure to HIV were forwarded to doctors who had requested the HIV antibody test for patients who met the study sample criteria. The cases of interest were those newly diagnosed between 1 January and 31 October 1991 and notified to state and territory health authorities as having been attributed to exposures to HIV other than male homosexual contact or receipt of blood, blood products or tissue. A total of 158 questionnaires was forwarded and 59 per cent were returned. Among the returned questionnaires included in the study sample, exposure to HIV on the original notification to the health authority was given as injecting drug use (8 per cent, 3 of 37), heterosexual contact (46 per cent, 17 of 37), or unavailable (46 per cent, 17 of 37). A clear basis for HIV exposure category classification was provided on the questionnaires for 70 per cent (7 of 10) of cases among women, whereas among men whose infection was attributed to heterosexual contact, a basis for exposure category classification was specified for only 43 per cent (10 of 23) of cases. Although the study was limited by the low response rate, use of the questionnaire provided a relatively simple means for assessing self-reported HIV exposure history.     

Managing patients with identical names in the same ward

PURPOSE: To review the experience of managing two patients with identical names in the same ward during a five-month period. DESIGN/METHODOLOGY/APPROACH: The records of the patients were reviewed to look for incorrect entries, errors in specimens sampling, administration of blood products and chemotherapy, and misplacement of clinical notes. Doctors and nurses involved were also invited to complete a questionnaire study to comment on the usefulness of the measures implemented for correct patient identification. A random sample of 60 patients was also selected to see if their full names were shared with other patients attending the same hospital. FINDINGS: Among the 1442 sheets of hospital records from the two patients, no errors pertaining to the clinical activities were found. However, 13 (0.9 per cent) sheets of the hospital records were misplaced. The 21 doctors and nurses participating in the questionnaire study gave positive support to all the additional measures implemented for safeguarding patient identification, of which the automated alerting feature in the electronic clinical management system received the highest scores. A total of 32 (53 per cent) of the 60 sampled patients shared a common full name with one to 101 other patients attending the same hospital. ORIGINALITY/VALUE: Patients with identical names staying in the same ward present a unique challenge to acute health-care settings. The situation is especially relevant in communities where most people's names are not unique. Specific guidelines and measures are needed to prevent patient misidentification. Errors in filing of patient notes and laboratory reports to the hospital record deserve further attention.     

Incidence of neurodermatitis (atopic dermatitis) in students starting school in Ostalbkreis in Baden-Württemberg

The prevalence of atopic eczematous dermatitis differs considerably between countries. In a former study with preschool children the prevalence in Baden-Württemberg according to data from parent questionnaires was about 10 per cent. Taking into account the expected false positives and false negatives led to the approximated unbiased prevalence of 7 to 8 percent. To evaluate this estimate, the Health Department Ostalbkreis used the same questionnaire in a second study of the same age group. The questionnaires were self-administered by the parents during the medical examination before school entrance. The parents of positive administered cases (11.2 per cent) were asked for permission to contact the family doctor to have the diagnosis confirmed (or not). 75 per cent of the positive identified cases by questionnaires were confirmed by family doctors. Therefore, the prevalence of atopic eczematous dermatitis in preschool children aged six to seven years in the area of Ostalbkreis is about 8.5 per cent.     

Provider expectations and consumer perceptions of the importance and value of dental care.

Seventy-eight inner city mothers of third and fourth grade children in three racial groups--white, black, and American Indian--known to need dental treatment for disease on permanent teeth, were interviewed at home by a trained community resident interviewer. Sixty-two per cent of the mothers were on public assistance. Information relating to the mothers' perceptions of the importance and value of dental care both for herself and for her children were collected. Utilization data were obtained via two dental examinations conducted one year apart. Data were also collected from a sample of provider dentists via mail questionnaire. A yield of 315 usable questionnaires was obtained, a return rate of 53 per cent. Provider-dentists felt that low socioeconomic consumers do not value dental services, as compared to other types of consumer goods and services, and that they do not believe dental care is important. Low income mothers in the same city reported that they did value dental care and believed it important. Expectations of and orientations toward the importance of dental care were found discongruent between the two groups of respondents. These discongruities on the dentist-patient relationship are discussed as a barrier to utilization. Although financial resources were available to many of the study families, only 49 per cent of these children received the needed care. It is suggested that the psychological cost to a patient of seeking care in inhospitable settings may act as a barrier to utilization.     

Solar protection behaviours: a study of beachgoers

Abstract: This study explored the prevalence and predictors of solar protection behaviour in a community sample of beachgoers. A total of 670 participants was randomly selected from six beaches in the Newcastle district. The solar protection behaviour of each participant was assessed by direct observation and interview. A subsample was also asked to complete a written questionnaire to assess attitudes to solar protection use, knowledge of skin cancer and awareness of recommended solar protection behaviours. Forty-five per cent of the beachgoers in this sample were using a high level of solar protection, and a substantial proportion (16 per cent) of the sample was not using any kind of solar protection. Sunscreen with SPF 15+ was applied to at least one body region by 69 per cent of the sample, 17 per cent of the sample were wearing a recommended hat, 15.1 per cent were using shade, and 3.4 per cent were wearing a recommended style of shirt at the time of observation. Chi-square analyses of the data for under-15-year-olds indicated that a significantly greater proportion of 0- to 9-year-olds were well protected compared to 11- to 14-year-olds, but there was no difference in use of protection by boys and girls under 15 years of age. Stepwise regression analysis of the adult sample (15 years and over) showed that the predictors of overall level of solar protection were marital status and frequency of skin self-examination in the past 12 months. The practical implications of these findings for future community-based skin cancer prevention programs are discussed.     

Women's satisfaction with medical termination

One hundred women aged between 14 and 43 completed a questionnaire regarding their experience and satisfaction with medical termination of pregnancy at Bedford Hospital. A visual analogue score from 0-10 was used. Age, occupation, marital status, parity, previous terminations and awareness of the medical method were recorded. Additional comments were also invited. Overall satisfaction with the method was good, with a mean score of 8.19 and women were very appreciative of the care given by nursing staff. Counselling and choice of method offered scored less well and need improvement. Eighty five per cent of this sample had already had a previous termination of pregnancy, with 30 per cent having had two or more, indicating a need for improved contraceptive counselling both before and after the termination.     

Patient and nurse assessment of quality of care in postoperative pain management

Objective: To describe and compare patient and nurse assessments of the quality of care in postoperative pain management, to investigate differences between subgroups of patients, and to compare patient assessments in different departments.

Design: Patient and nurse questionnaires.

Setting: Five surgical wards in general surgery, orthopaedics, and gynaecology in a central county hospital in Sweden.

Sample: Two hundred and nine inpatients and 64 registered nurses. The response rates were 96% for the patients and 99% for the nurses; there were 196 paired patient-nurse assessments.

Method: The Strategic and Clinical Quality Indicators in Postoperative Pain Management patient questionnaire was used which comprises14 items in four subscales (communication, action, trust, and environment). The items were scored on a 5 point scale with higher values indicating a higher quality of care. Five complementary questions on levels of pain intensity and overall satisfaction with pain relief were scored on an 11 point scale. Twelve of the 14 items in the patient questionnaire and two of the complementary questions were adjusted for use in the nurse questionnaire.

Results: The patients' mean (SD) score on the total scale (scale range 14–70) was 58.6 (8.9) and the nurses' mean (SD) score (scale range 12–60) was 48.1 (6.2). The percentage of patients who scored 1 or 2 for an individual item (disagreement) ranged from 0.5% to 52.0%, while for nurses the percentage ranged from 0.0% to 34.8%. Forty two patients (24%) reported more pain than they expected; these patients assessed the quality of care lower. There were differences between patient and nurse assessments concerning the environment subscale, the question on overall satisfaction, and patients' experience of worst possible pain intensity.

Conclusion: The results provided valuable baseline data and identified important areas for quality improvement in postoperative pain management.

     

Do women read poster displays on breast cancer in waiting rooms?

BACKGROUND: This study reports data from a survey carried out in the Iranian Center for Breast Cancer (ICBC) to examine whether women read poster displays in waiting rooms and whether they would have any suggestions to improve posters and thus meet their informational needs. METHODS: Five specially designed posters were displayed in waiting rooms in the ICBC. The content of posters was related to risk factors, early detection, signs and symptoms, and prevention of breast cancer. During a 3 month period a self-administered questionnaire was distributed to all attendees and they were asked to complete the questionnaire. RESULTS: Overall, 850 women were given the questionnaire and 777 completed questionnaires (91 per cent) were returned. The mean age of the women was 37.0 years (SD = 10.7) and they mostly had secondary education (47 per cent). In all, 691 women (86 per cent) reported that they had seen the posters and 620 (80 per cent) said that they had read the displays. The vast majority of the women reported that posters were readable (89 per cent) and understandable (80 per cent). However, 25 per cent of the respondents indicated that materials on the displays created more questions rather than answering their questions and some reported that they became upset (26 per cent) or felt anxiety (42 per cent) while reading the posters. Finally, 218 women (28 per cent) had suggestions to improve posters, of whom 110 (50 per cent) believed that the posters should be simpler. CONCLUSION: The study findings suggest that despite the limitations of posters as a means of health communication, their use in public places may be useful but consideration should be given to the content of the poster displays to prevent anxiety.     

Hearing conservation in the mining industry. Evaluation of a risk factor questionnaire

The objectives of this study were to assess the validity ofa self-administered hearing questionnaire in the assessmentof risk to hearing, to amend the questionnaire design as necessary,and to give general guidance on the use of audiometric questionnairesin hearing conservation programmes. The study was carried outat a colliery in Staffordshire. The odds of deafness, accordingto the Health and Safety Executive definition, were analysedby exposure to the hearing risk variables in the questionnaire.The results of the hearing questionnaire were linked to a cross-sectionalaudiometric survey. There were 189 questionnaires eligible forinclusion. Noise exposure at work was reasonably well predictedby having to shout at work (OR 1.42, 95 per cent Cl 0.58–3.54)and being exposed to noisy tools and machinery (OR 2.12, 95per cent Cl 0.72–6.72). Risk to hearing from disease waspredicted by having ‘abnormal’ hearing or dischargingears, but not by any of the other exposure variables. Risk tohearing from acoustic trauma was demonstrated for fire-arms(OR 1.23, 95 per cent Cl 0.5–2.98). None of the hobbiesincluded in the questionnaire usefully predicted hearing loss.The conclusions are that, contrary to Health and Safety Executiveadvice, audiometric questionnaires may reasonably be administeredto all employees, as they are useful in the documentation ofhearing loss risk factors. They should not, however, be usedto detect ear disease. A revised questionnaire was drafted asthe result of these findings, and will be piloted in a largerpopulation.     

Caries experience among children in fluoridated Townsville and unfluoridated Brisbane

Abstract: Fluoridation of community water supplies constitutes the main public health strategy for prevention of dental caries in Australia. In recent years questions have been raised about the effectiveness of water fluoridation. The aim of this paper was to examine differences in caries experience of children aged 5 to 12 years who were lifetime residents either of Brisbane (the unfluoridated Queensland capital) or Townsville (fluoridated since 1965). Children from each city were sampled from patients of the school dental service. Dental therapists and dentists from the school dental service recorded data describing dental caries experience and parents were asked to complete a questionnaire about their children's residential history and exposure to other fluorides. Of the 18 348 children sampled, 10 195 (55.6 per cent) provided completed questionnaires, and 4588 were lifetime residents of their respective cities. Caries rates were significantly lower ( P < 0.01) among children in Townsville than in Brisbane, both in the deciduous dentition (according to age, 32 to 55 per cent fewer tooth surfaces affected) and permanent dentition (20 to 65 per cent fewer tooth surfaces affected). Significantly lower rates in Townsville persisted ( P < 0.01) in multivariate analyses that controlled for oral hygiene practices, exposure to fluoride supplements and household income. Water fluoridation appears to provide a substantial public health benefit for children in Townsville.     

Knowledge and attitudes of hospital staff to occupational exposure to bloodborne viruses

In order to assess awareness of occupational risk of exposure to bloodborne viruses a questionnaire was sent to 245 health care workers, representing a 10% sample of employees with patient contact in a large teaching hospital in Scotland, stratified by occupational group. One hundred and eight questionnaires (44%) were returned. Seventy per cent of respondents in laboratory and clinical areas described themselves as having sufficient knowledge for their own area of practice, but many gave incorrect answers or expressed uncertainty about the infectivity of HIV and hepatitis B and hepatitis C viruses. Ninety-four respondents were unaware that a regimen containing more than one antiretroviral drug is now recommended for post exposure prophylaxis of HIV infection, 37 thought they had been at risk of bloodborne viral infection and had contacted the occupational health department for advice, and 68 respondents disagreed with guidelines from the United Kingdom's General Medical Council on testing of patients for bloodborne viruses. The results indicate a need for educational initiatives for new and existing staff.     

Questionnaire to assess patient satisfaction with pharmaceutical care in Spanish language.

OBJECTIVE: To develop and validate a questionnaire, in Spanish, for assessing patient satisfaction with pharmaceutical care received in community pharmacies. DESIGN: Selection and translation of questionnaire's items; definition of response scale and demographic questions. Evaluation of face and content validity, feasibility, factor structure, reliability and construct validity. SETTING: Forty-one community pharmacies of the province of Santa Fe. Argentina. PARTICIPANTS: Questionnaire administered to patients receiving pharmaceutical care or traditional pharmacy services. MAIN OUTCOME MEASURE: Pilot test to assess feasibility. Factor analysis used principal components and varimax rotation. Reliability established using internal consistency with Cronbach's alpha. Construct validity determined with extreme group method. RESULTS: A self-administered questionnaire with 27 items, 5-point Likert response scale and demographic questions was designed considering multidimensional structure of patient satisfaction. Questionnaire evaluates cumulative experience of patients with comprehensive pharmaceutical care practice in community pharmacies. Two hundred and seventy-four complete questionnaires were obtained. Factor analysis resulted in three factors: Managing therapy, Interpersonal relationship and General satisfaction, with a cumulative variance of 62.51%. Cronbach's alpha for the whole questionnaire was 0.96, and 0.95, 0.88 and 0.76 for the three factors, respectively. Mann-Whitney test for construct validity did not showed significant differences between pharmacies that provide pharmaceutical care and those that do not, however, 23 items showed significant differences between the two groups of pharmacies. CONCLUSION: The questionnaire developed can be a reliable and valid instrument to assess patient satisfaction with pharmaceutical care in community pharmacies in Spanish. Further research is needed to deepen the validation process.     

Results and cost of meeting the National Service Framework for Coronary Heart Disease requirement for 12 month follow-up after acute coronary events

BACKGROUND: The National Service Framework (NSF) for Coronary Heart Disease (CHD) set standards, targets and milestones. In the case of acute myocardial infarction (AMI) or coronary revascularization, Milestone 3 of Standard 12 requires a 12 month audit of exercise and smoking habit and of body mass index (BMI) for patients who have attended cardiac rehabilitation (CR). The targets are that 50 per cent of patients should be exercising regularly, not smoking and have a BMI of <30 kg/m(2). The purpose of this study was to find out whether the targets are realistic and to measure the cost of retrieving the data. METHODS: A postal questionnaire was used to follow up all the patients who attended our CR programme over a 12 month period. The project was costed. RESULTS: Four hundred and three CHD patients who had attended the programme between April 2001 and March 2002 were sent questionnaires 12 months after their index event. Their diagnoses were AMI in 147 (36.5 per cent), coronary artery surgery in 157 (39 per cent) and angioplasty in 99 (24.5 per cent). Completed questionnaires were received from 358 (89 per cent). Of the responders, 69 per cent were exercising regularly, 91.6 per cent were not smoking (73 per cent had been non-smokers before their index cardiac event) and 79 per cent had a BMI of <30 kg/m(2)(the figure at the start of rehabilitation had been 79 per cent). The cost of performing the audit was pounds sterling 1204. CONCLUSION: This audit is inexpensive. The targets for smoking and BMI set by the NSF were achieved by a very large margin before either the index cardiac event or starting CR.     

A self–administered questionnaire for detection of unrecognised coronary heart disease

Abstract: On an individual and a population basis, an increased incidence of coronary heart disease is associated with classical cardiovascular risk factors, but many cases occur in people not identified as at high risk. Conversely, many people at high statistical risk do not develop coronary disease. We used a questionnaire to identify unrecognised coronary heart disease in people attending large–scale health survey centres. Participants were required to report the presence and characteristics of any chest pain. Those returning responses consistent with myocardial ischaemia were offered treadmill exercise ECG tests. Over 18 months, 4070 questionnaires were returned. Of 475 respondents offered testing, 229 (198 male, 131 female) accepted. Thirty–two subjects (15 male, 17 female: a detection rate of 13.9 per cent of those assessed as likely on questionnaire, or 0.8 per cent of all respondents) had results consistent with significant coronary heart disease. Follow–up was available in 30 cases. There was no difference in classical risk–factor distribution (including multivariate risk percentiles: 42.4 (male) and 46.7 (female)) between those newly diagnosed with coronary heart disease and their community counterparts. More women than men were identified as suffering from unrecognised coronary heart disease, with a preponderance of younger women. Cost per case identified was A$1220. Screening by self–administered questionnaire is a useful and relatively cost–effective means of identifying unrecognised coronary heart disease. ( Aust N Z J Public Health 1997; 21: 545–7)     

Women in management. Why women in the NHS don't get a share of the action

The survey was designed to give a representative picture of staff attitudes across the South East Thames region. A key intention of the survey was to collect information from all staff groups and locations. Questionnaires were sent out during September and October 1992 to a random sample of women and men. By the close of the survey, 1,886 usable questionnaires had been returned from members of the survey sample, representing a response rate of 43 per cent. In addition, 512 questionnaires had been received from volunteers. The analysis presented in the report is based on 2,398 survey respondents, 18 per cent of whom were men.