The levels of use of over-the-counter drugs, particularly those with potential side effects, indicate that further studies are needed to examine patterns of use in more detail and that health education about self-medication is appropriate among young people aged 11-12 years and above.

Study objectives -These were as follows:to study incompleteness of data, herein called item non-response, generated by a self completion questionnaire; to identify the characteristics ofitem non-responders and the types of questions liable to high item non-response rates; and to discuss possible reasons for item non-response. Design -Item non-response patterns in 12 307 responders (62%) to a representative postal survey based on a stratified sample drawn from family health services authorities' (FHSA) registers were investigated. Main outcome measures -Data were analysed for item non-response in three groups depending on when the questionnaire was returned (wave analysis). The overall completion rate of the questionnaire was examined and the natural logarithm of the proportion of completed questions was used as an outcome variable in multiple regression analysis. Item In postal studies, bias may be introduced through respondents failing to complete all the questions, referred to here as "item nonresponse".' This bias is additional to any bias caused by total non-response,2 such as the under representation of young men34 and smokers.56 Therefore, a better understanding of item non-response should help in the interpretation ofdata generated by postal surveys. Reasons for item non-response vary and include poor question design, questionnaire layout, and the inability or unwillingness of the respondent to supply information.7A previous study has focused on item nonresponse in a series of three postal surveys of patients who had recently undergone three different surgical procedures (each involving around 300 patients).8 In all three surveys, more than half of all respondents missed one or more items (56%, 63%, and 60%). The amount of variance in item non-response attributable to respondents' characteristics, such as gender, age, education, or health perception, in regression modelling was fairly low (R2 = 7%, 12%, and 9%), suggesting that patient characteristics do not relate strongly to item non-response. In all three surveys questions at the end of the questionnaire were no more likely to be omitted than those at the beginning and the two most frequently omitted questions were on family income (13%, 14%, and 23%) and employment status (6%, 12%, and 17%).

Objective -To examine the quality ofsocial support among unemployed residents in Trent, England. Design -Secondary analysis of data generated by those ofworking age drawn from a postal lifestyle survey of the adult population of Trent region. Subjects and setting -Subjects were 6987 individuals (males 16-64 years and females 16-59 years of age), of whom 9.90% (6891 6987) were unemployed. Main outcome measures -Responses about the quality of social support obtained from three key questions. Results -Generally, the unemployed reported poorer quality of social support than employed persons ( p<0.0001 ) on all three key elements examined: 31% v 17% respectively had no practical support; 19% v 10% had no help with solving problems, and 21% v 10% had no emotional support. Only 57% of the unemployed had all three of these elements compared with 75% of the employed. Unemployment and lack of social support had independent and deleterious effects on perceptions of general health and mental health. Relationships remained after allowing for the possible confounding effects of age, gender, and household composition. Conclusions -There is a relationship between unemployment and poorer quality ofsocial support which may help to explain some ofthe increased morbidity and mortality experienced by this group, especially that related to mental health.

Study Objective-The study aimed to consider the impact of two different types of reminder on response rates and costs in a postal survey. Design-The study was a cross sectional survey. A self-completion lifestyle questionnaire was used. Those who did not respond after the initial mailing were randomly allocated to receive either a postcard or questionnaire as a first reminder. All outstanding nonresponders received a questionnaire as a second reminder. Subjects-A representative sample of 698 adults aged 16-70 was used, drawn from a family health services authority register. Main results-Postcard reminders were as effective as questionnaire reminders in increasing response whether one or two reminders are sent. The costs per response were calculated. Two questionnaires as reminders were found to be 1-7 times more expensive than a postcard plus questionnaire. Including the initial mailing, the cost per response using all questionnaires was 1-3 times the cost when a postcard was used for the first reminder. Conclusions-To increase the response to a postal survey effectively and economically, two reminders should be sent-first a postcard and then a questionnaire. 7 ipidemiol CommunitY Health 1993; 47: 334-335 questionnaire and letter were sent.7 In most studies the only or final reminder is a further copy of the questionnaire. The cost of the various types of reminder do not seem to have been reported in detail.We report findings from a pilot survey carried out in the autumn of 1991 as part of the Trent Health Lifestyle Survey,8 and address the practical issue of what type of reminders should be sent in a postal survey. MethodThe sample for the pilot study was drawn from the family health services authority patient register of a district with a population of 205 000. The register was stratified by gender and age. From each stratum a systematic sample was drawn, with random start and fixed sampling fraction, designed to provide a total sample of 700.The study was carried out using a selfcompleted questionnaire, delivered by post and returned by freepost. After three weeks, first reminders were sent out. The non-respondents were randomly assigned to be sent only a simple postcard reminder or a follow up letter together with another copy of the questionnaire and a freepost envelope. After a further three weeks, all remaining non-respondents were sent a follow up letter plus a copy of the questionnaire and a freepost envelope. The initial mailing was on 1 November, a first reminder was sent of 21 November, and a second reminder on 2 December 1991. December 1992 Throughout the 1 980s and into the 1 990s increasing emphasis in health promotion has been placed on the need for information by which to inform policy and monitor the impact of activity on populations.' Routine information about health related behaviour, knowledge, beliefs, and attitudes is unavailable from existing sources.2 An increasingly popular method of collecting population lifestyle data in Great Britain is a postal survey using a self-com...