1

Perception or real illness? How chronic conditions contribute to gender

inequalities in self-rated health.

European Journal of Public Health, 2012, 22(6), 781-6.

Unformatted, pre-print manuscript. Article available at http://dx.doi.org/10.1093/eurpub/ckr184

Davide Malmusi 1,2,3, Carme Borrell 1,2,4,5, Lucía Artazcoz 1,2,4, Joan Benach4,6.

1 Agència de Salut Pública de Barcelona

2 Institut d’Investigació Biomèdica (IIB-Sant Pau)

3 Unitat Docent de Medicina Preventiva i Salut Pública IMAS-UPF-ASPB

4 CIBER Epidemiología y Salud Pública (CIBERESP), Spain

5 Universitat Pompeu Fabra (UPF)

6 Health Inequalities Research Group (GREDS), Employment Conditions Network (EMCONET), UPF.

Address for correspondence:

Davide Malmusi

Agència de Salut Pública de Barcelona

Pl. Lesseps, 1

08023 Barcelona, Spain

Phone: +34 932027798

Fax: +34 933686943

dmalmusi@aspb.cat

2

ABSTRACT

Background: In Spain, like in many countries, women report poorer general health and more

daily activity limitations due to health reasons when compared to men. This study aims to

examine whether these poorer indicators are due to a greater prevalence of health

problems, and to identify the types of problems that contribute most to gender inequalities.

Methods: Cross-sectional study on the population aged over 15 and residing in Spain, with

data from the 2006 National Health Survey (n=29,139). The sex prevalence ratios (PR) of

poor self-rated health and chronic limitation of activity are sequentially adjusted by age and

the presence of 27 chronic conditions by means of robust Poisson regression.

Results: At equal number of disorders, women reported equal or even better health than

men. The excess of poor health in women (age-adjusted PR and 95% CI: self-rated health =

1.36, 1.29-1.41; chronic limitation = 1.25, 1.18-1.32) disappeared when adjusting for the

number of chronic diseases (self-rated health = 1.00, 0.96-1.04; chronic limitation = 0.90,

0.85-0.96). Musculoskeletal, mental and other pain disorders accounted for most of the

association. The results were consistent in different strata of age, social class and type of

country of birth.

Conclusion: These results suggest that the poorer self-rated health of women is a reflection

of the higher burden of disease they suffer. A health system responsive to gender

inequalities should increase its efforts in addressing and resolving musculoskeletal, mental

and other pain disorders, usually less considered in favour of disorders with greater impact

on mortality.

KEYWORDS

Gender inequalities; Health Status; Chronic Limitation of Activity; Musculoskeletal Diseases.

3

Introduction

The health of women and men is different and inequitable. The paradox is described that in

most countries of the world women have longer life expectancy but poorer health status1,2,

although this latter finding is more consistent for indicators of mental health, chronic illness

or disability than in the case of self-rated health3-5. Most of these differences are unfair and

avoidable inequalities derived from the unequal hazards of social positions, roles and

expectations of men and women in society because of the gender system6,7. While men’s

shorter life expectancy can be attributed to both biological disadvantages and gendered

patterns of health-related behaviours and risk-taking8, the poorer health status of women

has been interpreted as a result of patriarchy, the systematic domination of women by men,

which restricts women's access to social privileges, paid work and economic resources, while

allocating them the majority of domestic and family responsibilities9-11.

It has been argued that differences in symptoms perception, evaluation and reporting may

account for sex differences in illness reports12. This view transcends into practice where

studies conducted in the US found that women are more likely than men to be told by

physicians that their problems are “in their heads” and to have their symptoms attributed to

“over-anxiousness”, even when results indicated an organic disorder13. However, it has been

shown that women are not more prone to report illness14,15, and that their higher

expressiveness doesn’t account for their increased distress16. In some studies it is possible to

appreciate that, once adjusted for other measures of morbidity or functional capacity,

gender differences in self-rated health are eliminated or reversed17-19. Nevertheless, no

studies have been conducted focusing on this specific issue and particularly exploring the

individual and joint contribution of different chronic conditions to the disadvantage of

women in general health status. We hypothesise that inequalities in general indicators of

health status, quality of life or activity limitation, shall reflect the differences in the burden

of disease that women and men experience. This information could be useful to introduce

gender equity as a criterion for prioritising health problems in the health system as well as

for implementing policies related to the social determinants of these conditions.

Therefore, the objectives of this study are; 1) to assess whether, within the Spanish

population, poorer general health outcomes in women compared to men are attributable to

4

a higher prevalence of chronic conditions and 2) to identify the type of conditions that

contribute most to these inequalities.

Methods

Study population and data source

The study population consisted of all persons living in Spain in 2006, aged 16 years and

above and not institutionalised. The data source is the 2006 Spanish National Health Survey,

representative of the whole population living in households. Data were collected through

face-to-face interviews at home between June 2006 and June 2007. A stratified multistage

sampling was applied, the units being census tracts at first stage, main family dwellings at

second stage, and individuals at third stage20. The total number of people interviewed was

29,476. Analyses were performed on subjects with valid answers to all general health and

chronic conditions variables (N=29,139).

Variables

Dependent variables. The following health indicators were used as the main dependent

variables:

Self-rated general health status. It was derived from the single question “Over the last

twelve months, would you say your overall health has been very good, good, fair, poor,

very poor?”. This widely used indicator has predictive power on mortality, morbidity and

use of health services, and reflects a global evaluation of diseases, symptoms, functional

abilities and overall well-being21. A dichotomous outcome variable was created (very

good, good = good health; fair, poor, very poor = poor health)22.

Chronic limitation of activity. It was determined from the question "During the last six

months, to what extent have you been limited due to a health problem in the activities

people usually do?". Answers "severely limited" and "limited but not seriously" were

considered as having a chronic limitation.

Self-rated health was the main study variable, while activity limitation was used to test the

robustness of the findings.

5

Sex. The sex of the respondent (man or woman) was used as the main stratification variable

or independent variable depending on the analysis.

Specific chronic conditions. These were used as dependent, stratification and adjustment

variables depending on the analysis. The 27 conditions included in the survey were used,

which ranged from risk factors (high cholesterol) to chronic pain (e.g. back pain or headache)

or diseases with high mortality (e.g. stroke): for each, respondents stated whether they have

ever suffered from the condition; during the last twelve months; a physician has diagnosed

them; and have taken medication for it. The following variables were used:

o Number of prevalent conditions (suffered in the last 12 months);

o Number of diagnosed conditions (suffered in the past 12 months and diagnosed by

a physician at any time);

o Each prevalent condition individually;

o Groups of conditions. Starting from those conditions that individually reduced the

association between sex and self-rated health more than 10% (see Table 2), two

groups were created:

musculoskeletal disorders: a) arthrosis, arthritis or rheumatism; b) neck pain; c)

back pain; and d) osteoporosis;

mental disorders: a) reporting the disorders themselves, identified by a single

question (depression, anxiety or other); or b) having poor mental health (score

higher than two on the 12-item version of General Health Questionnaire, GHQ-

1223);

These variables were included in the models separately.

Stratification variables. As a way to test the consistency of results in different population

groups, some analyses were repeated in the following strata:

Age: 16 to 44 years, 45 to 64, and 65 years or older (representing respectively the

younger active age, adults in active age but who experience increasing health limitations,

and the legal inactive age in Spain).

Social class was assigned according to the current or last occupation of the respondent,

or of the household reference person for those never engaged in paid work, and

classified following the recommendations of the Spanish Society of Epidemiology24. The

five original classes were collapsed into two groups: non-manual (I-III) and manual (IV-V).

6

Type of country of birth was categorized as “Spanish-born”, “born in another high

income country” (countries with human development indexes greater than 0.9 in 2006,

year of the survey 25), and “born in middle or low income countries” (all other countries).

Due to the small sample size in some categories, the stratified analysis of social class and

type of country of birth was simplified in three categories: non-manual social classes (any

origin), Spanish-born manual social classes, or manual class born in middle or low income

countries (excluding from this analysis people of manual social classes from other high-

income countries).

Statistical analysis

All the analyses included weights derived from the complex sample design. The sample

distribution is described, as well as crude prevalence by sex of the general health indicators

and the 27 chronic conditions. Sex differences are tested with Pearson’s chi-square.

To analyse the contribution of specific chronic conditions to gender inequalities in general

health indicators (self-rated health and limitation of activity), we first described these

indicators by sex and number of chronic conditions. Then we calculated the prevalence

ratios (PR), with their confidence intervals at 95%, of the health indicators by sex by means

of robust Poisson regression models26, first entering age as the only adjustment variable, and

then adding to the model one or more variables of specific chronic conditions. An approach

to the contribution size was obtained by calculating the percentage reduction in the strength

of association between sex and the health indicator while adjustment variables were added

with the formula 100*(PRmodel 1-PRmodel 2)/ (PRmodel 1-1)27.

The regression analysis with and without adjustment for the number of chronic conditions

was repeated with stratification by age group, social class and type of country of birth.

Results

A total of 14,249 men and 14,746 women had valid answers for general health indicators

and chronic conditions. Of the participants, 52.4% were aged 16 to 44, 58.5% were in

manual social class, and 87.6% were born in Spain.

7

Inequalities in health indicators and in chronic conditions

Prevalences by sex of poor self-rated health, chronic activity limitation and the 27 chronic

conditions included, are shown in Table 1. Women reported poorer health (38.8% vs. 27%,

p<0.001) and more activity limitation (25.7% vs. 19.3%, p<0.001) than men. Women also had

higher prevalence for most specific chronic conditions. The prevalence was higher in women

within the 11 most common conditions (all p<0.001), and in 6 of these conditions prevalence

in women was more than double that of men. These results were consistent when the

criterion was restricted to conditions diagnosed by a physician.

The contribution of chronic conditions to gender inequalities in general health

Figure 1 describes the distribution of the number of chronic conditions, with more men than

women reporting no condition (37% vs. 22%) or one condition (23% vs. 17%) and more

women reporting three or more (46% vs. 25%). Furthermore, it shows the proportion of men

and women reporting poor health or chronic limitation for each category of the number of

conditions. It is quite evident that in this case, women did not report poorer health than

men with the same number of conditions; on the contrary, men appeared to report poorer

health and more chronic limitations than women with the same number of conditions,

especially from three conditions on.

Table 2 shows the same phenomenon in terms of multivariate analysis. Adjusted only by

age, women reported significantly poorer health (PR = 1.36, 95% CI= 1.29-1.41) and more

activity limitation (PR = 1.24, 1.18-1.32). Once further adjusted by the number of conditions,

there were no differences between women and men in the likelihood to report poor self-

rated health (PR adjusted by age and number of prevalent conditions = 1.00, 0.96-1.04),

while women had a lower risk of reporting activity limitation compared to men (PR adjusted

by age and number of prevalent conditions = 0.90, 0.85-0.96).

Table 2 also displays the conditions that reduced 10% or more, on their own, the strength of

the association between sex and self-rated health. Five disorders (arthritis, mental disorders,

neck pain, headache and back pain) individually reduced the association by 30% or more.

With limitation of activity as the dependent variable, the order of importance was consistent

8

but reductions were more emphasised. The same table shows the result of adjusting

simultaneously for groups of variables that accounted for similar health problems:

musculoskeletal disorders alone accounted for most of the association (71% with self-rated

health, 102% with chronic limitation), and so did the available mental disorder variables

(58% and 91%).

Stratification by age, social class and type of country of birth

Figure 2 shows that the findings of Table 2 are substantially similar in all subgroups: that is,

the probability of poorer health of women versus men disappears when adjusting for the

number of prevalent chronic conditions. For self-rated health, this was the case for younger

than 45 and the three groups by social class and type of country of birth; for those aged 45

to 64, the association was even reversed (PR 0.89, 95% CI 0.83-0.95), while for those over

64, it was reduced but still minimally significant (PR 1.07, 1.02-1.13). In the case of activity

limitation, the association was significantly reversed, as in the whole population, in the 16-

44 and 45-64 age groups and in the manual Spanish-born; disappeared in non-manual; and

was reduced while loosing significance in those over 64 and in manual from low-income

countries.

Discussion

In Spain, women reported poorer general health status and more chronic limitations of

activity than men. This article shows that these reports reflect or even underestimate the

fact that women suffer more often from a wide range of chronic conditions, where women

reported equal health and less activity limitation compared to men with the same number of

conditions. Musculoskeletal and other pain disorders, as well as mental health, account for

most of the excess of poor health of women.

Major strengths of the study include the use of a large population sample, and the

consistency of results with more than one health indicator, and in the various population

subgroups analyzed. In all the subgroups, sex differences disappeared when adjusting for the

number of reported conditions, except in the oldest group where a significant but very slight

9

excess persisted. This latter finding deserves replication, though it might be attributed to

residual confounding such as the severity of conditions.

Musculoskeletal, mental and other pain disorders were the primary causes of the excess of

poor health of women. Many other studies have shown that these groups of conditions are

also the ones with the strongest impact on well-being, disability and burden of ill health of

the entire population 28-32. In spite of their relevance, and the existence of effective

interventions for their management in primary care 33,34, they have been largely overlooked

in favour of other chronic diseases or cardiovascular risk factors 35,36. Previous studies have

pointed out the strict relation between physical and mental disorders37, and argued that

gender inequalities in mental illness may contribute to the gap in chronic physical conditions

15,38, although both may also share a common cause, and women’s higher rates of

depression themselves reflect real health differences rooted in disadvantages rather than

artifacts39.

The main finding of this study – gender differences in self-rated health disappear when

adjusting for number of chronic conditions – has also been observed in other studies in

Southern European cities such as Florence 17 and Barcelona 19. However, results could be

different in countries such as the UK and Finland where several studies have shown higher

morbidity in women but equal or poorer self-rated health in men 3,5,17,18. These contradictory

findings point out to the importance of examining social determinants of health status in

different contexts in order to tailor policies to reduce gender inequalities in health, which

are rooted in large inequities in socialisation, power, and exposures to determinants such as

employment status and working conditions, income, or the burden of unpaid work2,10.

The study has some potential limitations. One may arise from the use of self-reporting for

both the general health and chronic conditions variables, that may therefore share a

common variance. Nevertheless, checklists of conditions have been shown to encourage

reporting by the genuinely ill and give an accurate reflection of health needs29. Results were

also consistent when using the number of physician diagnosed disorders: although the same

diagnosis is self-reported, and often dependent on having posed the problem to a health

professional, we found no sex differences in the ratio between diagnosed and prevalent

disorders. Finally, other studies using different approaches have ruled out reporting bias as a

cause of gender differences in health indicators 14, 16.

10

The results can also be influenced by the use of a concrete checklist of conditions, which

despite being large, does not include all possible conditions. For example, externalising

mental disorders, such as those related with substance abuse, which are usually more

frequent in men, were not included. However, their impact on quality of life (not mortality)

at the population level is generally lower than that of internalising disorders28.

Because of the high level of correlation between different chronic conditions, the

percentages of contribution to gender inequalities of single conditions could be

overestimated. However, despite that, they are useful to approximate the relative

importance of one disorder over the other. For this same reason, the joint contribution of

chronic conditions was examined with summary variables such as the total number of

conditions. The alternative option of including each condition separately produces nearly

identical results (data not shown).

This study adds to the literature demonstrating that reporting bias is not an explanation for

gender differences in self-rated general health, whereas musculoskeletal and mental

disorders emerge as major contributors, a finding that also has new implications for both

health policy and research. A health care system responsive to gender inequalities should

increase its efforts in addressing and resolving these groups of disorders, which despite their

limited impact on mortality, have a strong impact on everyday health and well-being of not

only women but the entire population28, and are usually insufficiently addressed by health

systems vis-à-vis potentially fatal conditions with similar impacts13,36. At the same time,

research and policy on gender equity in health care have mainly focused so far on

inequalities in the quality of care provided for certain conditions, such as coronary heart

disease, or on women's specific issues such as reproductive health. However, these policies

should go beyond to include the prioritisation of health problems according to their impact

on gender inequalities, and the study of dimensions such as somatic pain or mental distress

from a gender perspective.

11

ACKNOWLEDGMENTS

This work was undertaken as part of Davide Malmusi’s doctoral dissertation at the

Universitat Pompeu Fabra.

FUNDING

This work was partially funded by Observatorio de Salud de las Mujeres, Dirección General

de la Agencia de Calidad, Ministerio de Sanidad, Política Social e Igualdad.

Davide Malmusi was partially supported by the IV grant for young epidemiologists “Enrique

Nájera” awarded by the Sociedad Española de Epidemiología and sponsored by the Escuela

Nacional de Sanidad.

Conflicts of interest: None declared.

KEYPOINTS

Women’s disadvantage in general health reports can be explained by their higher

prevalence of chronic conditions.

Musculoskeletal, mental and other pain disorders account for most of this

disadvantage.

A health care system responsive to gender inequalities should increase its efforts in

addressing and resolving these groups of disorders while considering their social

determinants.

12

References

1. Case A, Paxson C. Sex differences in morbidity and mortality. Demography 2005;42:189-214.

2. Cherepanov D, Palta M, Fryback DG, Robert SA. Gender differences in health-related

quality-of-life are partly explained by sociodemographic and socioeconomic variation

between adult men and women in the US: evidence from four US nationally representative

data sets. Qual Life Res 2010;19(8):1115-24.

3. Lahelma E, Martikainen P, Rahkonen O, Silventoinen K. Gender differences in illhealth in

Finland: patterns, magnitude and change. Soc Sci Med 1999; 1;48(1):7-19.

4. Rueda S, Artazcoz L, Navarro V. Health inequalities among the elderly in western Europe. J

Epidemiol Community Health 2008;62(6):492-8.

5. Bambra C, Pope D, Swami V, et al. Gender, health inequalities and welfare state regimes: a

cross-national study of 13 European countries. J Epidemiol Community Health 2009;63:38-

44.

6. Verbrugge LM. Gender and health: an update on hypotheses and evidence. J Health Soc

Behav 1985;26(3):156-82.

7. Krieger N. Genders, sexes, and health: what are the connections—and why does it matter?.

Int J Epidemiol 2003;32(4):652-7.

8. Bird CE, Rieker PP. Gender matters: an integrated model for understanding men's and

women's health. Soc Sci Med 1999;48(6):745-55.

9. Bird CE, Fremont AM. Gender, time use, and health. J Health Soc Behav 1991;32(2):114-29.

10. Ross CE, Bird CE. Sex stratification and health lifestyle: consequences for men's and

women's perceived health. J Health Soc Behav 1994;35(2):161-78.

11. Doyal L. What makes women sick: gender and the political economy of health. London:

MacMillan; 1995.

12. Hibbard JH, Pope CR. Another look at sex differences in the use of medical care: illness

orientation and the types of morbidities for which services are used. Women Health 1986;

11(2):21-36.

13. Tannenbaum C, Mayo N. Women's health priorities and perceptions of care: a survey to

identify opportunities for improving preventative health care delivery for older women. Age

Ageing 2003;32(6):626-35.

13

14. Macintyre S, Ford G, Hunt K. Do women ‘over-report' morbidity? Men's and women's

responses to structured prompting on a standard question on long standing illness. Soc Sci

Med 1999;48(1):89-98.

15. Gijsbers van Wijk CM, Huisman H, Kolk AM. Gender differences in physical symptoms and

illness behavior. A health diary study. Soc Sci Med 1999;49(8):1061-74.

16. Mirowsky J, Ross CE. Sex differences in distress: Real or artifact?. Am Sociol Rev

1995;60(3):449-68.

17. Jylhä M, Guralnik JM, Ferrucci L, Jokela J, Heikkinen E. Is self-rated health comparable across

cultures and genders?. J Gerontol B Psychol Sci Soc Sci 1998;53(3):S144-52.

18. Arber S, Cooper H. Gender differences in health in later life: the new paradox?. Soc Sci Med

1999; 48(1):61-76.

19. Orfila F, Ferrer M, Lamarca R, Tebe C, Domingo-Salvany A, Alonso J. Gender differences in

health-related quality of life among the elderly: The role of objective functional capacity

and chronic conditions. Soc Sci Med 2006;63(9):2367-80.

20. Ministerio de Sanidad y Consumo. Encuesta Nacional de Salud de 2006, Metodología

detallada.

www.msc.es/estadEstudios/estadisticas/encuestaNacional/encuestaNac2006/metodENS20

06.pdf Accessed March 2011.

21. Quesnel–Vallée A. Self-rated health: caught in the crossfire of the quest for ‘true’health?.

Int J Epidemiol 2007;36(6):1161-4.

22. Manor O, Matthews S, Power C. Dichotomous or categorical response? Analysing self-rated

health and lifetime social class. Int J Epidemiol 2000;29(1):149-57.

23. Goldberg D. Manual of the general health questionnaire. Windsor, England: NFER

Publishing; 1978.

24. Domingo-Salvany A, Regidor E, Alonso J, Alvarez-Dardet C. Una propuesta de medida de la

clase social. Aten Primaria 2000;25:350-63.

25. United Nations Development Programme. Human Development Report 2006: beyond

scarcity: power, poverty and the global water crisis. New York: United Nations Development

Programme; 2006.

26. Barros AJD, Hirakata VN. Alternatives for logistic regression in cross-sectional studies: an

empirical comparison of models that directly estimate the prevalence ratio. BMC Medical

Research Methodology 2003;3(1):21.

14

27. Szklo M, Nieto FJ. Epidemiology: beyond the basics. Gaithersburg, MD: Aspen; 2000.

28. Saarni SI, Härkänen T, Sintonen H, et al. The impact of 29 chronic conditions on health-

related quality of life: a general population survey in Finland using 15D and EQ-5D. Quality

of Life Research 2006;15(8):1403-14.

29. Knight M, Stewart-Brown S, Fletcher L. Estimating health needs: the impact of a checklist of

conditions and quality of life measurement on health information derived from community

surveys. J Public Health Med 2001; 23(3):179-86.

30. Moussavi S, Chatterji S, Verdes E, Tandon A, Patel V, Ustun B. Depression, chronic diseases,

and decrements in health: results from the World Health Surveys. Lancet 2007; 370:851-8.

31. Centers for Disease Control and Prevention (CDC). Prevalence and most common causes of

disability among adults--United States, 2005. MMWR Morb Mortal Wkly Rep 2009;

58(16):421-6.

32. Pueyo MJ, Surís X, Larrosa M, et al. [Importance of chronic musculoskeletal problems in the

population of Catalonia (Spain): prevalence and effect on self-perceived health, activity

restriction and use of health services.]. Gac Sanit 2011. doi:10.1016/j.gaceta.2011.03.008 .

Spanish.

33. Rost K, Nutting P, Smith JL, Elliott CE, Dickinson M. Managing depression as a chronic

disease: a randomised trial of ongoing treatment in primary care. BMJ 2002; 325(7370):934.

34. Dobscha SK, Corson K, Perrin NA, et al. Collaborative Care for Chronic Pain in Primary Care.

JAMA 2009; 301(12):1242-52.

35. Eisenberg L. Treating depression and anxiety in primary care. N Engl J Med 1992;326:1080-

4.

36. Verbrugge LM, Patrick DL. Seven chronic conditions: their impact on US adults' activity

levels and use of medical services. Am J Public Health 1995; Feb;85(2):173-82.

37. Gureje O, Von Korff M, Simon GE, Gater R. Persistent Pain and Well-being. JAMA 1998; July

08;280(2):147-51.

38. Needham B, Hill TD. Do gender differences in mental health contribute to gender

differences in physical health?. Soc Sci Med 2010; 71(8):1472-9.

39. Piccinelli M, Wilkinson G. Gender differences in depression. Critical review. Br J Psychiatry

2000; 177:486-92.

15

Table 1 Presence of chronic conditions and poor health indicators by sex (%).

Prevalent a Diagnosed a

Men

(n=14,249)

Women

(n=14,746)

Men

(n=14,249)

Women

(n=14,746)

Chronic disorders b

Chronic back pain 18.9 28.6** 15.1 24.5**

Chronic neck pain 15.5 31.3** 12.3 26.5**

Arthrosis. arthritis or rheumatism 13.9 28.8** 12.5 27.0**

Hypertension 17.1 20.4** 16.5 20.1**

Varicose veins in the legs 7.0 26.1** 4.5 18.2**

Migraine or frequent headaches 8.2 20.4** 5.9 16.0**

High cholesterol level 13.8 14.5** 13.6 14.3**

Depression. anxiety or other mental disorders 8.2 18.9** 6.6 16.2**

Chronic allergies 10.0 13.1** 8.9 12.1**

Haemorrhoids 6.0 10.2** 4.3 7.4**

Chronic constipation 3.1 11.1** 2.5 8.2**

Diabetes 6.2 5.7 6.2 5.7

Chronic skin problems 5.4 6.3** 5.0 6.0**

Prostate problems 5.8 --- 5.6 ---

Menopausal period problems --- 5.6 --- 4.9

Cataracts 3.8 6.2** 3.5 6.1**

Osteoporosis 1.3 8.2** 1.2 7.9**

Anaemia 1.3 7.7** 1.2 7.6**

Asthma 4.1 5.0** 3.9 4.9**

Urinary Incontinence 3.0 5.6** 2.6 4.4**

Heart diseases other than infarction 3.6 4.8** 3.5 4.7**

Chronic bronchitis 4.3** 3.6 4.2** 3.5

Stomach or duodenal ulcer 3.6 3.5 3.3 3.4

Thyroid problems 0.7 5.8** 0.7 5.7**

Malignant neoplasms 1.2 1.2 1.2 1.2

Myocardial infarction 1.4** 0.6 1.4** 0.6

Stroke 0.7 0.7 0.7 0.7

Mean number (SD) 1.7 (2.1) 2.9** (2.8) 1.5 (2) 2.6** (2.7)

General health indicators

Poor general self-rated health 27.0 38.8**

Chronic limitation of activity 19.3 25.7**

Population aged 16 or more. Source: Spanish National Health Survey. 2006. a Prevalent: Having had the disorder in the last 12 months. Diagnosed: Having had the disorder in the last 12

month. and having had it diagnosed by a physician at any time. b Conditions are ordered by prevalence in the whole study population.

** Pearson’s chi-square p-value < 0.001.

16

Table 2. Association between sex and general health indicators. adjusted sequentially for age,

number of diagnosed and prevalent chronic conditions, and single chronic conditions.

Poor self-rated health Activity limitation

PR

(Women

vs. Men)

% change PR

(Women

vs. Men)

% change

Adjusting by age and... 1.36** 1.24**

Number of chronic conditions (diagnosed) 1.04 89% 0.93* 130%

Number of chronic conditions (prevalent) 1.00 99% 0.90** 139%

Single chronic conditions a

Arthrosis, arthritis or rheumatism 1.18** 48% 1.08* 66%

Depression, anxiety or other mental disorders 1.20** 45% 1.08* 66%

Chronic neck pain 1.20** 44% 1.10* 58%

Migraine or frequent headaches 1.25** 31% 1.14** 43%

Chronic back pain 1.25** 30% 1.14** 42%

Varicose veins in the legs 1.27** 25% 1.15** 39%

Osteoporosis 1.29** 18% 1.16** 35%

Anaemia 1.30** 16% 1.18** 26%

Chronic constipation 1.30** 15% 1.19** 22%

Groups of conditions

Musculoskeletal disorders b 1.10** 71% 1.00 102%

Mental disorders c 1.15** 58% 1.02 91%

Population aged 16 or more. Source: Spanish National Health Survey. 2006.

PR: Prevalence ratio calculated by means of robust Poisson regression.

% change: Percent reduction in the strength of association between the variable sex and the health indicator,

compared to age-adjusted model only. a Ordered from higher to lower reduction in the strength of association. Only those conditions reducing the

association more than 10% are shown. b Musculoskeletal disorders: Arthritis, back pain, neck pain, osteoporosis. c Mental disorders: depression, anxiety or other mental disorders (self-reported); mental health according to GHQ-

12 (good, poor, non-response).

* p-value of the sex coefficient < 0.01 ** p-value of the sex coefficient < 0.001

17

Figure 1 Prevalence of poor health and activity limitation by sex and number of prevalent

chronic conditions.

Population aged 16 years or more. Source: Spanish National Health Survey, 2006

18

Figure 2 Association between sex and self-rated general health, adjusted sequentially for age

and number of prevalent chronic conditions. Different strata of age, social class and type of

country of birth.

PR, prevalence ratio of poor self-rated general health between women and men; CI, confidence interval; Model 1,

adjusted by age; Model 2, adjusted by age and number of prevalent chronic conditions. MLIC: Born in a middle- or

low-income country. Source: Spanish National Health Survey, 2006