Secondary Logo

Journal Logo

Proceedings: Sex Differences and Analgesics; London, UK, 11 November 2001: Conference Paper

Epidemiology of analgesic use: a gender perspective

Isacson, D.; Bingefors, K.

Author Information
European Journal of Anaesthesiology: 2002 - Volume 19 - Issue - p 5-15



Analgesics are the most commonly used non-prescription medications and some of the most commonly used prescription medications. The extent of analgesic use can be described from the viewpoints of unnecessary use, dependence and side-effects [1-3]. Problems with concurrent use of multiple analgesic products have also been discussed [4]. In Sweden, in 1999, sales of different types of analgesics as measured in Defined Daily Doses per 1000 inhabitants and day are sufficient to provide for approximately 10% of the Swedish population to use a full dose of analgesics every day [5]. Epidemiological studies relying on self-reported use consistently show a gender difference, with women reporting significantly more frequent analgesic use [6-11]. An American study analysing the difference in use of abusable prescription drugs, i.e. anxiolytics and narcotic analgesics, between women and men showed that women were 48% more likely to use such drugs as compared to men controlling for demographics, health status, economic status, and diagnosis [12].

In spite of the common use of analgesics and the potential risks involved, relatively few epidemiological studies on factors influencing analgesic use have been conducted. Some factors have been found to be of importance in some studies but not in others. Relief from pain, headache, and discomfort are major reasons for using prescription analgesics [4,6-8,13]. However, these factors are not found to be of the same importance for the use of non-prescription analgesics [7,14]. Mental health problems have also been associated with the use of analgesics and several explanations have been discussed [6,8,15,16]. One is that these problems cause or are expressed as pain [17]. Another is that pain causes mental health problems and a third is that psychiatric disorders, such as depression, may share a common pathogenic mechanism with pain [18,19]. Further, non-prescription analgesics may be used as a substitute for hypnotics and tranquillizers due to their availability [14]. Social structure, social status, marital status, educational level, economic situation, work and working conditions have been found to be associated with use of analgesics in some studies but not in others [4,6-8,14,20].

Studies have also consistently shown a higher health care utilization among women as compared to men. The reasons for these differences have been discussed by several authors [21-25]. The discussions have been summarized by Verbrugge into five categories: biological risks, acquired risks, psychosocial aspects of symptoms and care, health reporting behaviour, and prior health care and caretakers as causes [25]. The first category 'biological risks' includes gender differences due to genetics, sex hormones, reproductive events, and sex-specific pathology. 'Acquired risks' includes gender differences due to differences in work, leisure, lifestyle, health habits, psychological stress, and social environment. Differences in perception and evaluation of symptoms as well as action taken are categorized as 'psychosocial aspects of symptoms and care'. Further, it has been suggested that women report symptoms more willingly, that they recall health problems to a greater extent than men and that men and women use different language when reporting illness, as included in the category 'health reporting behaviour'. Finally, Verbrugge has the category 'prior health care and caretakers as causes' to explain gender differences. This category includes explanations such that women may recover more rapidly from illness due to more active care; that women are diagnosed earlier due to more frequent health care contacts, that women have different health perceptions, knowledge, and attitudes due to greater prior experience with the sick role; and that a physician gender bias may exist [25].

The aim of our study was to analyse the influence of medical and non-medical factors on the difference in use of analgesics between women and men from a population perspective.

Study population and methods

This investigation is based on a postal questionnaire sent to a random sample of 8000 persons aged 20-84 years from the population registry in the county of Uppsala, Sweden, in October-December 1995. The aim of the survey was to study different aspects of health, health care utilisation, and the use of drugs. Uppsala County is comprised of a university city, smaller towns, and agricultural areas. It had a population of approximately 290 000 inhabitants in 1995. In all 5404 persons answered the questionnaire, i.e. 68% of those included in the survey. The majority of non-respondents were those who gave no reason for not responding (n = 2275) and those who stated that they did not want to participate (n = 132). A comparison between the study population and the total population of Uppsala County showed that the distributions of gender, age, marital status, and educational level were similar [26].

Comparability with other surveys was one important factor in determining which questions to include; validity was another. In deciding questions on drug use, health care utilisation, diseases and medical complaints, recall periods, sociodemographic variables, etc., the national Swedish survey, 'The Swedish Survey of Living Conditions', was considered. This survey is a series of cross-sectional surveys conducted annually since 1975 [27,28]. Several methodological studies have been conducted on the survey [27,29]. The influence of education, marital status, economic difficulties, knowledge about and attitudes towards medication, self-care orientation, alcohol use, smoking and body mass index ((BMI) men (women): <20.1 (18.7) = underweight, 20.1-24.9 (18.7-23.7) = normal, > 25.0 (23.8) = overweight), contacts with primary and hospital care were determined using questions described in detail elsewhere [26,30](Appendix 1).

Users of analgesic drugs were identified by a question on medication use, phrased 'Have you during the last two weeks used any of the following medicines' followed by a list of prescription drugs and non-prescription drugs. The respondent had the opportunity to add drugs that were not covered by the list. The importance of pain and ache as well as other psychiatric and somatic complaints was also studied using self-report. Further, analyses were carried out on health related quality of life (HRQoL) and the difference in use between men and women. Finally, the importance of menstrual problems and menopausal problems was investigated.

The medical outcomes study (MOS) short form 36 (SF-36) was used to measure HRQoL (Appendix 2). SF-36 covers eight domains of health: physical function (PF), role limitation because of physical health (RP), bodily pain (BP), general health perceptions (GH), vitality (VT), social functioning (SF), role limitation because of emotional health problem (RE) and mental health (MH) [31,32]. Only a small number of respondents did not complete the questions included in SF-36. For each dimension those 25% (or closest to 25%) who scored lowest were identified (PF = 85.0, RP = 75.0, BP = 54.0, GH = 60.0, VT = 27.3, SF = 75.0, RE = 66.7 and MH = 66.7 were used).

Statistical analysis

In order to explain differences in the frequency of analgesic use between women and men, changes in the odds ratio (OR) for gender difference in analgesic use were calculated using consecutive sets of independent variables shown to differ between genders in the descriptive analyses studied. In a first step (Model A) age, education, marital and economic status, knowledge about and attitudes towards medication, self-care orientation, alcohol use, smoking and body mass index were included. Variables statistically associated with the use of analgesics were kept in the model. In a second step (Model B), variables on pain and ache, i.e. backache, ache in arms and legs, shoulder ache, and headache were added to the variables in the first model. In a third step (Model C) other somatic and psychiatric complaints were added. In order to analyse the importance of medical complaints which are possible for women only, we also tested the influence on the OR of excluding women with menopausal or menstrual problems from the analyses (Model D). The eight domains of SF-36 were added in the fourth step (Model E). Finally, the importance of contacts with health care was studied (Model F). The statistical analyses were carried out using the SAS statistics program [33]. In the logistic regression analyses the LOGISTIC procedure was used.


As shown in Table 1, 34.8% of the women and 21.4% of the men had used one or more type of analgesic - both prescription and non-prescription - during the two week study period. Descriptive statistics on use by age, marital and economic status, education, selfcare orientation, medication knowledge, attitudes towards drugs, body mass index, alcohol use, smoking and low HRQoL is also presented in Table 1. The prevalences of various medical problems and complaints in relation to the use of analgesics are shown in Table 2. Backache was the most common problem followed by shoulder ache and sleeping problems. Use of analgesics was most prevalent among individuals suffering from headache.

Table 1
Table 1:
Number in the population and use of analgesics in the county by marital status, education, economic status, self-care orientation, medication knowledge, attitudes towards drugs, body mass index, smoking, alcohol use and low health related quality of life, in Uppsala County, Sweden.
Table 2
Table 2:
Prevalence of various medical complaints (self-reported) and use of analgesics among persons with these complaints, in Uppsala County, Sweden.

Changes in the OR for women compared to men when introducing the consecutive sets of variables stepwise into the model are shown in Table 3. The uncorrected OR for analgesics use was 1.96 (CI (95%) 1.73 to 2.21) for women compared with men in the population. Table 3 shows the changes in the OR when introducing the different sets of explanatory variables.

Table 3
Table 3:
Factors associated with the use of analgesics in Uppsala County.

In Model A economic problems were a significant risk factor for analgesic use as was good medication knowledge and a positive view of self care. Individuals with a negative attitude towards medical drugs used analgesics to a lesser extent than others. Furthermore, obesity was associated with analgesic use (Table 3). However, the inclusion of these variables only had a minor effect on the OR for women versus men. It decreased from 1.96 to 1.89 (CI (95%) 1.66 to 2.15).

In Model B the pain variables were added to Model A. Ache and pain were strongly correlated with the use of analgesics (Table 3). The inclusion of these variables also had a strong impact on the OR for women vs men which decreased from 1.89 to 1.56 (CI (95%) 1.36 to 1.79). Besides a significant influence of economic difficulties, good medication knowledge, a positive view of self care, a negative attitude to drugs and being overweight, it was found that age and education still were significantly associated with the use of analgesics in this model. Those having the least education used analgesics to a lesser extent than others in the population.

Of the variables relating to medical problems presented in Table 2, colds or influenza, sleeping problems and anxiety were found to be significant when added in Model C (Table 3). Having had a cold or influenza was strongly correlated while sleeping problems and anxiety measures were of less importance. The inclusion of these variables in Model C only had a marginal effect on the OR for women versus men.

With respect to medical problems specific to women, we analysed the importance of menstrual and menopausal problems by excluding women reporting these problems. Menopausal problems were found to be of minor importance in the use of analgesics and women with these problems therefore remained in the model. On the other hand, menstrual problems were found to strongly influence the use of analgesics among women. A separate logistic regression analysis including only women aged 20-52 years showed that menstrual problems were the second most important factor for the use of analgesics in this group (OR = 2.37, CI (95%) 1.71 to 3.30), headache being most important factor (OR = 2.63, CI (95%) 2.00 to 3.45). The OR for women versus men decreased somewhat - from 1.51 to 1.44 (CI (95%) 1.25 to 1.66) - when women with menstrual problems were excluded from the analyses.

In Model E, variables on low HRQoL were added. It was found that those scoring low on the bodily pain dimension and those scoring low on the vitality dimension used analgesics to a greater extent than those scoring higher (Table 3). The other dimensions were not found to be statistically associated with the use of analgesics. When the two dimensions - bodily pain and vitality - were included the sleeping problems and anxiety variables ceased to be statistically significant. Once again there was a slight decrease in the OR for women versus men. Model E, shows headache as the most influential factor explaining analgesic use (OR = 2.87, CI (95%) 2.32 to 3.55). The second most important factor was a negative attitude towards drugs (OR = 0.42, CI (95%) 0.23 to 0.75). Severe bodily pain was in third place (OR = 2.16, CI (95%) 1.79 to 2.60) followed by having influenza or a cold (OR = 1.49, CI (95%) 1.22 to 1.81). The OR for women vs men was 1.39 (CI (95%) 1.20 to 1.60).

In a final step - Model F - contacts with health care, i.e. physician or district nurse visits in primary care, or utilization of hospital care, were added to the analysis. Neither of these variables was found to be of statistical significance when added to Model E. Neither, did these variables turn out to be of importance for the difference in analgesic use between women and men.


In this study we found that 34.8% of the women and 21.4% of the men had used analgesics during the two week period studied. The gender difference in analgesic use remained statistically significant after controlling for factors differentially distributed between women and men and known to affect use.

When interpreting the results one should note that the participation rate in the survey was 68% and women participated more commonly than men did. It is reasonable to assume that there are several factors, e.g. health and quality of life, that are of importance for the willingness to participate in a survey like this one. Missing values can occur completely at random, or for a definite reason [34]. Consequently our missing values cannot be considered completely random. It should also be noted that in a survey relying on self-report there may well be problems with recall bias and lay perception errors.

This study is based on self-report of health conditions and medication use. Some of the gender differences may be caused by different reporting behaviours shown to differ substantially between men and women [35]. In most studies, women report higher frequencies of somatic symptoms, more bodily distress, including pain, and more medication use [35,36].

Pain and ache had the most significant effect on the gender difference in analgesic use in the study. Thus, the prevalence of different types of self-reported pain was much higher among women than among men. For instance, 16.1% of the women but only 6.4% of the men reported suffering from headache, 26.5% of the women but only 15.0% of the men had problems with shoulder ache. We also found a higher proportion of women among those who scored low on the SF-36 dimension bodily pain. Pain or a perception of pain should logically be the main cause behind a decision to take an analgesic drug. However, the pain variables do not explain all analgesic use.

It has been suggested that there are differences in the perception of pain between men and women. Women have lower pain threshold and tolerance in experimental studies, but it is not clear whether this is due to a difference in reporting behaviour [37]. Women experience more pain caused by biological difference, thus the prevalence of menstrual problems was high in our study. Gender differences in the pharmacokinetics and pharmacodynamics of analgesics, possibly the consequence of modulation by sex hormones, have been found in some studies [38]. It is, however, not clear if these differences are of clinical importance and how they may influence analgesic use.

Differences in 'acquired risks', i.e. gender differences due to differences in work, leisure, lifestyle, health habits, psychological stress, and social environment, was one of the five categories used by Verbrugge [25] for explaining differences in health care utilization.

In our study, as well as in other studies, the importance of factors such as social structure, social status, marital status, educational level, economic situation and lifestyle, were of a more marginal importance for the use of analgesics [4,6-8,14,20]. We found that education, economic problems and obesity were associated while marital status, smoking and alcohol use were not associated with use of analgesics. Personal factors such as attitudes and knowledge have been included in various models including the health belief model [39-41], the theory of reasoned action [39,42,43] and in the behavioural model developed by Andersen [44]. We found that self-care orientation, attitudes to drugs and medication knowledge were strongly associated with the use of analgesics. However, somewhat surprisingly the inclusion of variables that would represent differences in 'acquired risks' was of little importance for the difference in use between men and women.

Unfortunately, the survey did not include detailed information on working conditions which may differ significantly between men and women and which have a significant influence on work-related pain conditions. Headache, the most common cause of pain in our study, has been shown to be strongly associated with physical and mental work stress [45]. A study in Sweden showed that, compared with the seventies, women in the nineties are more often than men employed in monotonous jobs with a high physical work load [46]. Also, in the same study, physical work load was not related to educational level among women while men with higher education experienced less physical workload [46]. Musculoskeletal pain conditions are strongly related to working conditions and, in women, the added burden of a greater responsibility for the home. In a longitudinal study, risk factors for neck and shoulder pain, with the exception of repetitive motion, were found to differ between genders; among women psychosocial factors such as skill development, opportunities of influencing job characteristics and perceived total workload (including home responsibilities) were important while among men physically demanding work, particularly in combination with unfavourable social conditions at work, were important [47]. Women also suffer from fibromyalgia, believed to be a stress related disease, to a much greater extent than men [48].

A recent government report on the alarmingly rapid increase in stress and work-related sick-leave, particularly among women, states that the gender differences in illness can be explained by women having a considerably weaker position in the labour market including worse psychosocial work environments, worse physical work environments and fewer rehabilitation programmes [49]. Vocational rehabilitation has also been shown to be less successful for women [50,51].

Severe stress at work and/or at home is often expressed as depression. Somewhat surprisingly, there was no effect of depression or of the mental health dimension of SF-36 on analgesic use or on the gender difference. However, there was an effect of low scores on the vitality dimension on analgesic use. Depression and other mental health conditions, usually more frequent among women, have been consistently shown to be strongly associated with pain [17,18] and depressed patients are frequent users of analgesics [52]. Men seem to have a higher tendency toward 'self-medication' for the somatic symptoms of depression while not admitting to the socially stigmatized disease of depression [53,54].

The reported use of analgesics can be considered to be a decision to treat by the individual. Women may differ from men in their propensity for initiating medication use. The appraisal of pain, found to differ between genders, may have important implications on coping strategies [36,55]. Attitudes and beliefs about the social acceptability of different pain behaviours for men and women may be important in the decision to use medication [56]. Behaviours learned from older generations in the family have also been shown to influence pain appraisal and coping strategies [56]. Medication use for health concerns is typically higher for women [25]. Women with headache, particularly tension-type headache, used medications to a greater extent than did men with the same condition in one study [57]. However, other studies have not shown a gender difference in prescription medication use for headache [36]. It has been suggested that men use non-prescription analgesics as a substitute for health care while women use non-prescription analgesics as a supplement for health care [26].

Even after controlling for several sets of potentially explanatory factors we found a significant gender difference in the use of analgesics. This remaining difference may possibly be explained by biological and psychosocial factors not measured in the study. The gender difference in analgesic use that remains after controlling for a number of factors, including pain, should be investigated further. There are, as yet, no satisfactory explanations in the literature, but presumably several causes, biological as well as environmental and psychological, are involved. The most important issue is whether the difference is explained by adequate drug use or whether women are unnecessarily exposed to a much greater risk for adverse effects and dependency through their frequent use of analgesics.


In the population, women use analgesics much more frequently than men. Consequently women may be at greater risk for adverse effects and dependency. Some of the gender difference is explained by the greater frequency of pain conditions among women, but a significant difference in use still remains to be explained.


1. Silberstein SD, Young WB. Analgesic rebound headache. How great is the problem and what can be done? Drug Saf 1996; 13: 133-144.
2. Rantucci M, Segal HJ. Hazardous non-prescription analgesic use by the elderly. J Soc Adm Pharm 1991; 8: 108-120.
3. Guess HA, West R, Strand LM, et al. Fatal upper gastrointestinal hemorrhage or perforation among users and nonusers of nonsteroidal anti-inflammatory drugs in Saskatchewan, Canada 1983. J Clin Epidemiol 1988; 41: 35-45.
4. Chrichilles EA, Lemke JH, Wallace RB, Drube GA. Prevalence and characteristics of multiple analgesic drug use in an elderly study group. J Am Geriatr Soc 1990; 38: 979-984.
5. Apoteksbolager AB (The National Corporation of Swedish Pharmacies). Svensk läkemedelsstatistik 1995 (Swedish drug statistics 1999). Stockholm: Apoteksbolaget AB, 1999.
6. Antonov K, Isacson D. Use of analgesics in Sweden - the importance of sociodemographic factors, physical fitness, health and health related factors, and working conditions. Soc Sci Med 1996; 42: 1473-1481.
7. Antonov K, Isacson D. Prescription and non-prescription analgesic use in Sweden. Ann Pharmacother 1998; 32: 485-494.
8. Eggen AE. The Tromsø study: Frequency and predicting factors of analgesic drug use in a free-living population (12-56 years). J Clin Epidemiol 1993; 46: 1297-1304.
9. Del Rio M, Prada C, Alvarez F. The use of medication by the Spanish population. Pharmacoepidemiol Drug Saf 1997; 6: 41-48.
10. Svarstad BL. Stress and the use of non-prescription drugs: an epidemiological study. Res Comm Ment Health 1983; 3: 233-254.
11. Kung F, Gibson S, Helme R. Factors associated with analgesic and psychotropic medications use by community-dwelling older people with chronic pain. Aust NZ J Publ Health 1999; 23: 471-474.
12. Simoni-Wastila L. The use of abusable prescription drugs: the role of gender. J Women Health Gend Med 2000; 9: 289-297.
13. Antonov K, Isacson D. Headache and analgesic use in Sweden. Headache 1998; 38: 97-104.
14. Ahonen R, Enlund H, Klaukka T, Vohlonen I. Use of analgesics in a rural Finnish population. J Pharmacoepidemiol 1991; 2: 3-17.
15. Hanlon JT, Fillenbaum GG, Studenski SA, Ziqubu-Page T, Wall WE Jr. Factors associated with suboptimal analgesic use in community-dwelling elderly. Ann Pharmacother 1996; 30: 739-744.
16. Jacobsen BK, Hansen V. Mental problems and frequent use of analgesics. Lancet 1989; 333: 273.
17. von Korff M, Dworkin SF, le Resche L, Kruger A. An epidemiologic comparison of pain complaints. Pain 1988; 32: 173-183.
18. von Knorring L, Ekselius L. Idiopathic pain and depression. Qual Life Res 1994; 3: S57-S68.
19. Breslau N, Davies G, Schultz L, Peterson E. Migraine and major depression: a longitudinal study. Headache 1994; 34: 387-393.
20. Jylhä M. Ten-year change in the use of medical drugs among the elderly - a longitudinal study and cohort comparison. J Clin Epidemiol 1994; 47: 69-79.
21. Cleary P, Mechanic D, Greenley J. Sex differences in medical care utilization: an empirical investigation. J Health Soc Behav 1982; 23: 106-119.
22. Isacson D. Pharmacological patterns among heavy users of prescription drugs. J Soc Adm Pharm 1987; 5: 12-20.
23. Meininger J. Sex differences in factors associated with use of medical care and alternative illness behaviors. Soc Sci Med 1986; 22: 285-292.
24. Svarstad B, Cleary P. Gender differences in the acquisition of prescribed drugs: an epidemiological study. Med Care 1987; 25: 1089-1098.
25. Verbrugge LM. Gender and health: an update on hypotheses and evidence. J Health Soc Beh 1985; 26: 156-182.
26. Antonov K. Pharmacoepidemiological studies on the use of analgesics in Sweden. PhD: Uppsala University, 1997.
27. Statistics Sweden. Living Conditions. Appendix 13. Technical report for 1984-85, 1986-87 and 1988-89 Surveys of Living Conditions. Örebro: SCB-Tryck, 1991.
28. Statistics Sweden. Living Conditions. Report no 76. Health and medical care 1980-1989. Örebro: SCB-Tryck, 1992.
29. Thorslund M, Wärneryd B. Methodological research in the Swedish surveys of living conditions. Problems of measurements and data collection. Soc Ind Res 1985; 16: 77-95.
30. Isacson D, Bingefors K. Attitudes towards drugs. A survey in the general population. Pharmacy World and Science 2002; in press.
31. Ware J, Sherbourne C. The MOS 36 item short form health survey (SF-36). Med Care 1992; 30: 473-483.
32. Ware J, Snow K, Kosinski M, Gandek B. SF-36 Health Survey Manual and Interpretation Guide. Boston: New England Medical Center, 1993.
33. SAS Institute Inc. The SAS system for Windows, Release 6.12. Cary, NC, USA, 1996.
34. Little RJA, Rubin DB. Statistical Analysis with Missing Data. New York: John Wiley, 1987.
35. Barsky A, Peekna H, Borus J. Somatic symptom reporting in women and men. J Gen Int Med 2001; 16: 266-275.
36. Unruh A. Gender variations in clinical pain experience. Pain 1996; 65: 123-167.
37. Fillingim R, Maixner W. Gender differences in the response to noxious stimuli. Pain Forum 1995; 4: 209-221.
38. Beierle I, Meibohm B, Derendorf H. Gender differences in pharmacokinetics and pharmacodynamics. Int J Clin Pharmacol Ther 1999; 37: 529-547.
39. DiMatteo MR, DiNicola DD. Achieving patient compliance. The Psychology of the Medical Practitioner's Role. New York: Pergamon Press, 1982.
40. Becker MH, Maiman LA, Kirscht JP, Haefner DP, Drachman RH, Taylor DW. Patient perception and compliance: recent studies of the Health Belief Model. In: Haynes RB, Taylor DW, Sackett DL, eds. Compliance in Health Care. Baltimore and London: The Johns Hopkins University Press, 1979: 78-109.
41. Strecher VJ, Rosenstock IM. The Health Belief Model. In: Glanz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education, 2nd edn. San Francisco: Jossey-Bass Publishers, 1996: 41-59.
42. Fishbein M, Ajzen I. Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Reading, Massachusetts: Addison-Weseley, 1975.
43. Montano DE, Kaspryzyk D, Taplin HT. The theory of reasoned action and the theory of planned behavior. In: Glanz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education. San Francisco: Jossey-Bass Publishers, 1996: 85-112.
44. Andersen RM. Revisiting the Behavioral Model and access to medical care: does it matter? J Health Soc Behav 1995; 36: 1-10.
45. Antonov K, Isacson D. Headache in Sweden - the importance of working conditions. Headache 1997; 37: 228-234.
46. Torgén M, Kilbom Å. Physical work load between 1970 and 1993 - did it change? Scand J Work Environ Health 2000; 26: 161-168.
47. Fredriksson K, Alfredsson L, Bildt Thorbjörnsson C, et al. Risk factors for neck and shoulder disorders: a nested case-control study covering a 24-year period. Am J Ind Med 2000; 38: 516-528.
48. Anderberg UM. Fibromyalgia syndrome in women - a stress disorder? PhD: Uppsala University, 1999.
49. Riksförsäkringsverket. RFV rapporterar 2001: 5. Stockholm: Riksförsäkringsverket (National Social Insurance Board), 2001.
50. Marnetoft S-U, Selander J, Bergroth A, Ekholm J. Factors associated with successful vocational rehabilitation in a Swedish rural area. J Rehab Med 2001; 33: 71-78.
51. Ahlgren C, Hammarström A. Has increased focus on vocational rehabilitation led to an increase in young employee's return to work after work-related disorders? Scand J Publ Health 1999; 27: 220-227.
52. Bingefors K, Isacson D, von Knorring L, Smedby B. Prescription drug use and health care utilization among patients treated with antidepressants in a Swedish community. Ann Pharmacother 1995; 29: 566-572.
53. Bingefors K. Antidepressant-treated patients. Longitudinal, population-based studies. PhD: Uppsala University, 1996.
54. Bingefors K, Isacson D, von Knorring L, Smedby B, Ekselius L, Kupper L. Antidepressant-treated patients in ambulatory care. Long-term use of non-psychotropic and psychotropic drugs. Br J Psychiatry 1996; 168: 292-298.
55. Unruh A, Ritchie J, Merskey H. Does gender affect appraisal of pain and pain coping strategies? Clin J Pain 1999; 15: 31-40.
56. Fillingim R. Sex, gender, and pain: women and men really are different. Curr Rev Pain 2000; 4: 24-30.
57. Rasmussen B. Migraine and tension-type headache in a general population. Int J Epidemiol 1992; 21: 1138-1143.
58. Lisper L, Isacson D, Sjödén P, Bingefors K. Medicated hypertensive patients' view and experience of information and communication concerning antihypertensive drugs. Patient Educ Counc 1997; 32: 147-155.

    Appendix 1.

    Questions used in the Uppsala County survey on health, health care utilization, and health related quality of life as reported inTable 3.

    Sociodemographic, attitudes and knowledge, lifestyle


    What education have you received? Only give one answer and report the highest level attained.

    Nine-year basic school or less

    At least 2 years of high school

    At least 3-4 years of high school


    Other education, which? ...

    Economic standard:

    During the past 12 months, have you at any time had problems with meeting expenses for food, rent or other necessary bills?

    Yes, once

    Yes, several times


    Medication knowledge:

    Mark each of the following statements either 'correct', 'wrong' or 'don't know':

    Penicillin strengthens the immune system

    Using nasal sprays continously for more than ten days can cause a blocked nose

    Some drugs can be absorbed into the blood stream through the skin

    If the prescription label says '1 tablet twice a day', it means that the tablets should be taken 8 hours apart.

    Treo and Magnecyl (well known aspirin brand names) contain the same active drug

    On some drug packages there is a red triangle, which means that one should not drive a car after taking the drug

    Attitude towards medicines:

    Which of the following alternatives do you think best describes your view of medicines?

    Medicines are something positive, a help.

    Medicines are a necessary evil

    Medicines are something negative, a danger

    (Based on findings from qualitative studies in Sweden [58])

    Self-care orientation:

    Below you will find a list of various symptoms. Mark for each one what you would first do if you suffered from that symptom. Would you: contact a doctor; treat yourself (rest, home remedies, over-the-counter drugs); wait and expect the symptoms to disappear; or ignore the symptom (one answer per symptom).

    Breathlessness while resting

    Diarrhoea or constipation

    Skin rash


    Persistent cough

    Unexpected weight loss

    Sore throat

    Back pain


    Sleeping problems

    Feeling tired and worn out

    Common cold


    (From surveys at the Kaiser Permanente Center for Health Research, Oregon, USA: Johnson R. Personal communication)

    Health problems

    Are you at present suffering from one or more of the following diseases or symptoms?



    Leg ulcers





    Skin problems

    High blood pressure

    Menopausal (Climacteric) problems


    Gastrointestinal problems

    Sight problems

    Hearing problems

    Back or hip pains, sciatica

    Urinary tract problems

    Ache in legs and/or arms

    Persistent menstrual problems


    Have you, during the past 2 weeks, experienced one or more of the following symptoms?

    Sleeping problems

    Heavy cold or influenza

    Heartburn or gastritis

    Gynaecological problems

    Sprained limb

    Recurring headaches or migraines


    Have you, at any time in your life, suffered from any of the following diseases'?




    Myocardial infarction

    Stomach ulcer

    (From the Swedish Survey on Living Conditions [27-29])

    Appendix 2.

    Health related quality of life instrument, SF-36, used in the investigation. TABLE

    SF-36 dimensions with lowest and highest possible score definitions.

    ANALGESICS; SEX; pharmacoepidemiology

    © 2002 European Society of Anaesthesiology