Social and Economic Inequalities in the Use of Contraception Among Women in Spain

Abstract

Background:

Socioeconomic inequalities affect different areas of health, including sexual and reproductive health. The aim of the present study was to analyze inequalities in the use of contraception among women resident in Spain in 2006.

Methods:

This is a cross-sectional study of women aged 15–49 years and resident in Spain in 2006 that analyzes the use of contraception during their first experience of sexual intercourse and during the 4 weeks before the interview (n = 5,141). Socioeconomic inequalities are measured with indicators of socioeconomic position, such as level of education, social class, and country of origin, and such characteristics as age, religion, age at first intercourse, living with partner, and number of children.

Results:

Contraception was used by 70.4% of the women during their first experience of sexual intercourse and by 78.1% during the previous 4 weeks. The women who used contraception most during their first experience of sexual intercourse were nonreligious younger women from developed countries who had a higher level of education and who had their first experience after the age of 18. The women who used contraception most often during sex in the previous 4 weeks were younger women with a higher level of education who did not live with a partner, who had children, and who had used contraception during their first experience.

Conclusions:

There are socioeconomic inequalities in the use of contraception among women in Spain. The use of contraception during the first experience of sexual intercourse was associated with more frequent use of contraception during sex in the 4 weeks before the interview.

Introduction

It is widely accepted that contraception is the most effective means of preventing unintended pregnancy and sexually transmitted infections (STI).^1,2 In developed countries, however, where access to contraception is unlimited, at least for the general population, rates of unintended pregnancy and STI remain nonnegligible.^3,4 In these countries, the main reasons for inequality seem to be related to differences in the behavior of different social classes (in terms of the quantity and quality of contraception used), as opposed to differences in access.^5,6

Several studies have been carried out in higher-income countries⁷ with the aim of discovering the factors that influence a woman's decision to use contraception. They have indicated differences in availability and accessibility and also in social and cultural attitudes toward fertility control, sexuality, and the roles of women in society.^8,9 Women in the lowest socioeconomic strata usually use the least effective forms of contraception and are, therefore, at highest risk of unintended pregnancy.^10,11

Since contraception was legalized in Spain in 1978, there has been a gradual increase in its use (33% of women in 1985 used some kind of contraception, 56% in 1997, and 72% in 2003).¹² However, the rate of use of some of the most effective methods, such as the pill, is still lower than in other European countries,¹³ and unintended pregnancy rates remain high, especially at the limits of childbearing age.^1,9,10,14 Few studies have examined the use of contraception or unintended pregnancy in the general population in Spain, and still fewer have explored the socioeconomic characteristics related to lower rates of use. In fact, only one study analyzed socioeconomic factors and contraception use, using data for the period 1985–1995, when the population of Spain was socially quite different from today.¹⁵

There are various indicators of socioeconomic position that can be used to study socioeconomic inequalities in health. None of them can be considered as the best one because each of them reflects a different part of the total existing variation.¹⁶ As socioeconomic characteristics also change during the life course, a combination of indicators can be used to try to capture this great variation.¹⁷ Level of education and social class based on occupation are two appropriate indicators to study socioeconomic inequalities in contraception use for two main reasons: first, they are related to different times in the life course of women, and second, they complement each other because level of education can be collected from all women, not only from those who are in the labor market.¹⁸ The country of origin also can be an indicator of socioeconomic position in Spain because people who come from developing countries usually work in jobs below their educational level, often experiencing job insecurity and social vulnerability.¹⁹

The aim of the present study was to describe the extent of the use of contraceptives in Spain in 2006 and to explore socioeconomic inequalities that could exist in their use. Such inequalities could be discussed and dealt with in the future so that women from different socioeconomic strata can have the same opportunities in planning their pregnancies with regard to contraception use. To achieve this aim, rates of contraceptive use were evaluated during two different periods: during women's first experience of sexual intercourse and during the 4 weeks before the interview.

Materials and Methods

Design, setting, and patients

This is a population-based cross-sectional study of noninstitutionalized sexually active women aged 15–49 years residing in Spain in 2006. The information source used was the 2006 fecundity interview developed by the Center for Sociological Research (CIS).²⁰ This survey consisted of a face-to-face interview at women's homes that included sociodemographic data and information about the sexual attitudes and reproductive life of women aged ≥15 years residing in Spain. A fixed number of 400 interviews were carried out in each of the autonomous communities (the 17 regions into which Spain is divided), and the remaining interviews needed to achieve the intended number of 10,000 interviews were assigned in proportion to the size of the study population (interviews were conducted for 9,737 women, corresponding to a response rate of 97.4%). Weighting coefficients were applied to restore proportionality to the sample. A random multistage sample selection strategy was used, with stratification by autonomous communities. Primary (municipality) and secondary (census section) sampling units were randomly selected in proportion to the population of the autonomous community. Women from these units were selected using random route procedures, with quotas calculated on the basis of female activity rates (activity as defined by the labor force survey of Spain [EPA]) and age. Considering a 95.5% confidence level (CI) (two sigma), the real error of the overall sample was ± 1.1%, and the maximum error of the autonomous communities was ± 4.99%.

From the initial sample, 5,141 sexually active women of childbearing age (i.e., between 15 and 49 years) who resided in Spain in 2006 were selected. This sample is representative at both national and autonomous community levels.

Measurements and variables

Two dependent variables were considered. The first considered the use of contraception during the woman's first experience of sexual intercourse. After ascertaining the age at which the participant had had her first complete heterosexual intercourse, she was asked: During your first experience of sexual intercourse, did you or the other person take any kind of precaution in order to avoid pregnancy? The response was coded as a dichotomous variable.

The second dependent variable concerned the use of contraception during sex in the 4 weeks before the interview. This was a dichotomous variable that recorded the participant's answer to the question: Have you and/or your partner used any kind of contraceptive method or a combination of methods when having sex during the last 4 weeks? This question was asked only of women who confirmed that they had had sexual intercourse during the previous 4 weeks.

The independent variables explored to identify possible associations with the use of contraception during first intercourse were selected on the basis of our conceptual framework and a review of the literature and were as follows: age (15–19, 20–24, 25–34, and 35–49 years), level of education (illiterate or incomplete primary education; primary education, up to age 14–16; secondary education, usually up to age 17–18; and university education), country of origin (developed countries, including Switzerland, United States, Bahamas, Japan, and the European Union prior to May 2004 [note that 98% of the women from developed countries were Spanish], or developing countries, including all other countries), being religious (yes, including women of any faith, or no, including women who report being nonreligious or nonpracticing Catholics), age at first intercourse (<16, 16–17, 18–19, and >19 years), and the year of first intercourse, calculated using the woman's age at interview and her age at first intercourse (<1986, 1986–1995, and >1995).

In the analysis of the use of contraception during the previous 4 weeks, in addition to the independent variables noted, we also included cohabitation status (living with a partner or not), number of children (0, 1, or >1), contraception use at first intercourse (yes or no), having a religious partner (yes or no), employment (yes or no), main activity (employed, unemployed, housewife, student), individual social class (I–II, III, IV–V), and household social class (taken as that of the woman or of her partner, if higher: I–II, III, IV–V), following the classification system proposed by the Spanish Society of Epidemiology.²¹

Respondents who said they had used contraceptives during their first experience of sexual intercourse or during the previous 4 weeks were asked if they used any of the methods on the following list: condom, pill, withdrawal, injection, intrauterine device (IUD), emergency contraception, periodic abstinence, male sterilization, female sterilization, cream/foam spermicides or sponge, diaphragm, any other method. The rates of missing data were 3.6% and 6.1% for the first and second dependent variables, respectively.

Statistical analysis

We performed a descriptive univariate analysis and a bivariate analysis using the chi-square test. To quantify the relationships between the independent and dependent variables, bivariate and multivariate logistic regression models were fitted, and crude and adjusted odds ratios (OR and aOR, respectively) and the corresponding 95% confidence intervals (95% CI) were calculated. We found no evidence for interaction between the independent variables, so interaction terms were not included in the adjusted models. Although we did not observe a statistical interaction between the variables for use of contraception at first intercourse and during the previous 4 weeks, we performed an analysis of contraceptive use during the previous 4 weeks stratified by contraceptive use at first intercourse because the latter was found to be very relevant in multivariate models.

All statistical analyses were performed using STATA, version 10.1 (Stata Corp., College Station, TX). Missing values were excluded from the analysis.

Results

Table 1 shows rates of use of contraception during the first experience of sexual intercourse and during the previous 4 weeks among women of childbearing age in Spain. The most commonly used method of contraception was the condom (53.7% and 32.9%, respectively), followed by the pill (10.8% and 20.7%, respectively). It should be noted that a nonnegligible percentage of women reported having used ineffective methods, such as withdrawal, periodic abstinence, and spermicides or sponges as the sole method of contraception (withdrawal was used by 1.8% of women at their first intercourse and by 1.8% of women during the 4 weeks before the interview).

Table 1.

Contraception Methods Used During First Intercourse and During Previous 4 Weeks in Women Aged 15–49, Spain, 2006

	First intercourse		Last 4 weeks
Contraception method	n	%	n	%
None	1,467	29.74	770	22.00
Condom	2,651	53.74	1,153	32.94
Pill	532	10.79	725	20.70
Intrauterine device	15	0.29	218	6.23
Male sterilization	3	0.06	251	7.16
Female sterilization	2	0.04	178	5.09
Withdrawal	90	1.83	62	1.78
Injections	32	0.65	7	0.21
Emergency postcoital contraception	9	0.18	1	0.03
Periodic abstinence	6	0.12	7	0.21
Cream/foam spermicides, sponge	2	0.04	2	0.07
Diaphragm	0	0.00	8	0.23
Others	11	0.22	59	1.69
Condom + pill	88	1.78	27	0.78
Condom + other	11	0.21	19	0.55
Other combinations	16	0.33	12	0.34
Total	4,933	100.00	3,500	100.00

Contraception use during first complete heterosexual intercourse

In Spain, 70.4% of women of childbearing age used some kind of contraceptive method during their first heterosexual intercourse (Table 2). Women aged 15–19 years used contraception most during first intercourse (aOR 1.91, 95% CI 1.01-3.63, compared to women aged 20–24 years), and those aged 35–49 years used it least (aOR 0.13, 95% CI 0.09-0.19). Moreover, women with a higher level of education were more likely to have used contraception (aOR 5.88, 95% CI 4.33-7.99 for women with a university education compared to women with less than a primary education). Women from developing countries used contraception less than those from developed countries (aOR 0.22, 95% CI 0.16-0.29), and religious women also used contraception less than nonreligious women (aOR 0.73, 95% CI 0.61-0.88). Women who started having intercourse before age 16 years or between 16 and 17 years used contraception less often than those who started between 18 and 19 years (aOR 0.39, 95% CI 0.27-0.55 and aOR 0.71, 95% CI 0.57-0.90, respectively) (Table 2).

Table 2.

Bivariate (Number of Women and Prevalence of Contraception Use) and Multivariate Association Between Use of Contraception During First Intercourse and Individual Characteristics, Spain, 2006

	n	%	p value chi-square	OR	95% CI	aOR	95% CI
Age, years
15–19	281	89.52		2.47	1.47-4.14	1.91	1.01-3.63
20–24	566	87.65		1		1
25–34	1,622	80.83		0.51	0.37-0.69	0.47	0.33-0.66
35–49	2,487	57.49		0.16	0.12-0.21	0.13	0.09-0.19
Total	4,956	70.39	<0.001
Level of education
Less than primary	524	42.33		1		1
Primary	1,887	66.89		2.26	1.84-2.78	2.26	1.73-2.94
Secondary	1,354	75.87		3.19	2.54-4.01	3.61	2.72-4.79
University	1,129	83.31		4.83	3.75-6.22	5.88	4.33-7.99
Total	4,894	70.53	<0.001
Origin
Developed countries	4,416	73.31		1		1
Developing countries	523	46.15		0.32	0.26-0.39	0.22	0.16-0.29
Total	4,939	70.43	<0.001
Religious
Yes	1,306	60.48		0.56	0.48-0.65	0.73	0.61-0.88
No	3,606	74		1		1
Total	4,912	70.41	<0.001
Age at first intercourse
<16	355	61.59		0.56	0.43-0.75	0.39	0.27-0.55
16–17	1,150	72.33		0.92	0.76-1.12	0.71	0.57-0.90
18–19	1,517	73.93		1		1
>19	1,671	66.82		0.71	0.60-0.85	0.90	0.74-1.10
Total	4,693	70.08	<0.001
Year of first intercourse
<1986	1,473	51.24		1
1986–1995	1,698	73.38		2.62	2.21-3.11
>1995	1,491	85.78		5.74	4.68-7.03
Total	4,662	70.35	<0.001

Totals of different variables are not the same because of missing values.

Totals can differ from the sum of categories because of the sample weights.

aOR, odds ratio (OR) adjusted for all variables included in last column; CI, confidence interval.

Regression models were not adjusted for year of first intercourse, as this variable did not improve the fit of the final model. This is likely because most of the variability explained by this variable was already accounted for by the variables age and age at first intercourse. Year of first intercourse was only relevant in women aged 35–49 years, where starting to have intercourse after 1986 was associated with increased use of contraception (aOR 1.95, 95% CI 1.55-2.45 in women who had their first experience of sexual intercourse between 1986 and 1995 compared to those who had it before 1986). The only area where missing could potentially affect the results was in the analysis of education level, where women with university studies had the most missing data (5.7%, p < 0.01).

Contraception during 4 weeks before interview

In this sample, 80.5% of women of childbearing age had had sexual relationships during the 4 weeks before the interview, a percentage that increased with age, and 78.1% of these used some form of contraception (Table 3).

Table 3.

Bivariate (Number of Women and Prevalence of Contraception Use) and Multivariate Association Between Use of Contraception During Previous 4 Weeks and Individual Characteristics of Women, Spain, 2006

	n	%	p value chi-square	OR	95% CI	aOR	95% CI
Age, years
15–19	173	94.67		2.47	1.47-4.14	1.10	0.44-2.77
20–24	360	91.51		1		1
25–34	1,161	80.70		0.51	0.37-0.69	0.41	0.24-0.68
35–49	1,822	72.20		0.16	0.12-0.21	0.26	0.15-0.45
Total	3,515	78.09	<0.001
Level of education
Less than primary	352	66.70		1		1
Primary	1,377	76.66		2.26	1.84-2.78	1.35	0.96-1.91
Secondary	962	81.29		3.19	2.54-4.01	1.60	1.10-2.34
University	783	81.73		4.83	3.75-6.22	1.86	1.23-2.79
Total	3,473	78.08	<0.001
Origin
Developed countries	3,166	78.79		1		1
Developing countries	338	71.31		0.32	0.26-0.39	0.84	0.59-1.20
Total	3,504	78.07	<0.01
Religious
Yes	904	71.84		0.56	0.48-0.65	0.81	0.64-1.02
No	2,583	80.10		1		1
Total	3,487	77.96	<0.001
Age at first intercourse
<16	263	83.73		0.56	0.43-0.75	1.16	0.71-1.89
16–17	811	80.22		0.92	0.76-1.12	0.76	0.57-1.03
18–19	1,089	81.19		1		1
>19	1,162	73.14		0.71	0.60-0.85	0.77	0.60-0.99
Total	3,325	78.34	<0.001
Year of first intercourse
<1986	1,067	69.90		1
1986–1995	1,282	77.27		2.62	2.21-3.11
>1995	964	89.30		5.74	4.68-7.03
Total	3,313	78.39	<0.001
Living with partner
Yes	2,689	74.12		1		1
No	827	90.99		3.52	2.61-4.75	2.94	2.00-4.31
Total	3,515	78.09	<0.001
Children
0	1,137	83.54		1		1
1	780	70.70		0.48	0.37-0.61	1.37	0.96-1.95
>1	1,554	78.28		0.71	0.57-0.89	3.13	2.19-4.49
Total	3,472	78.30	<0.001
Use at first intercourse
Yes	2,454	84.29		3.11	2.56-3.77	2.76	2.18-3.48
No	1,022	63.30		1		1
Total	3,476	78.12	<0.001
Religious partner
Yes	398	68.53		0.71	0.56-0.90
No	1,561	75.35		1
Total	1,959	73.86	0.004
Employment
Yes	1,392	78.71		1.08	0.90-1.30
No	1,323	77.38		1
Total	2,715	78.06	0.412
Main activity
Employed	1,537	78.65		1.29	1.04-1.60
Unemployed	291	76.13		1.12	0.81-1.54
Housewife	644	74.07		1
Student	241	96.49		9.62	4.61-20.09
Total	2,713	78.51	<0.001
Women's social class
I–II	344	81.13		1
III	295	79.06		0.88	0.58-1.32
IV–V	716	77.34		0.79	0.57-1.11
Total	1,356	78.65	0.388
Household social class
I–II	629	78.82		1
III	430	75.97		0.85	0.63-1.14
IV–V	1,224	76.15		0.86	0.68-1.09
Total	2,282	76.83	0.405

Totals of different variables are not the same because of missing values.

Totals can differ from the sum of categories because of the sample weights.

aOR, odds ratio (OR) adjusted for all variables included in last column; CI, confidence interval.

The results presented in Table 3 were not adjusted for the last five independent variables or for the year of first intercourse, as none of these variables were found to be relevant to the model. For example, having a religious partner did not provide extra information when women's religious beliefs were taken into account; the association with women's main activity observed in the crude model was not maintained in the adjusted model because the majority of women in the youngest age group are students; the associations with employment and social class were not associated with use of contraception in any case, and most of the variability in year of first intercourse was explained by women's age and age at first intercourse. Again, year of first intercourse was relevant only for women aged 35–49 years, among whom starting to have intercourse after 1986 was associated with increased use of contraception (aOR 1.95, 95% CI 1.55-2.45 in women who had their first experience of sexual intercourse between 1986 and 1995 compared to those who had it before 1986). Differences between groups in terms of country of origin, religiousness, and age at first intercourse lose their significance when the variables living with a partner, number of children, and use of contraception at first intercourse are included in the model. The other independent variables considered in Table 2, women's age and level of education, continued to be associated, although the magnitude of association decreased.

For contraception use during the previous 4 weeks, the most important variables were having children (aOR 3.13, 95% CI 2.19-4.49 two or more compared to none), not living with a partner (aOR 2.94, 95% CI 2.00-4.31), and having used contraception at first intercourse (aOR 2.76, 95% CI 2.18-3.48). The change in results for the variable corresponding to the number of children between the crude and adjusted models was mainly because of introduction of the variables for age and cohabitation status into the model (Table 3).

We observed significant differences between the results according to whether or not contraception was used at first intercourse (Table 4). For women who used contraception during their first intercourse, such variables as age, education level (aOR 2.47, 95% CI 1.37-4.47 in women with university studies), not living with a partner (aOR 4.09, 95% CI 2.41-6.96), and having children were associated with the use of contraception during the previous 4 weeks in the same direction as for the nonstratified adjusted analysis. For women who did not use contraception during their first intercourse, the use of contraception during the previous 4 weeks was clearly associated only with having more than two children (aOR 2.65, 95% CI 1.53-4.60 compared to not having children), also in the same direction as in the nonstratified adjusted analysis. The only areas where missingness could potentially affect the results was in the analysis of age group, where women aged 20–24 years had the most missing values (9.9%, p = 0.04), and in the analysis of use of contraception at first intercourse, where women who used contraception had more missing values in questions related to use during the last 4 weeks (6.6%, p = 0.03).

Table 4.

Adjusted Regression Models for Use of Contraception During Previous 4 Weeks Stratified by Having or Not Having Used Contraception During First Intercourse, Spain, 2006

	Used contraception at first intercourse		Not used contraception at first intercourse
	aOR ^a	95% CI	aOR ^a	95% CI
Age, years
15–19	1.94	0.43-8.72	0.49	0.11-2.28
20–24	1		1
25–34	0.30	0.15-0.61	0.79	0.33-1.89
35–49	0.18	0.08-0.38	0.53	0.22-1.26
Level of education
Less than primary	1		1
Primary	1.73	1.01-2.98	1.15	0.75-1.76
Secondary	1.95	1.12-3.41	1.46	0.88-2.44
University	2.47	1.37-4.47	1.48	0.82-2.69
Origin
Developed countries	1		1
Developing countries	1.04	0.58-1.88	0.79	0.48-1.31
Religious
Yes	0.88	0.64-1.21	0.72	0.50-1.03
No	1		1
Age at first intercourse
<16	0.79	0.38-1.67	1.61	0.86-3.02
16–17	0.59	0.40-0.87	1.23	0.78-1.94
18–19	1		1
>19	0.70	0.50-0.98	0.90	0.62-1.33
Living with partner
Yes	1		1
No	4.09	2.41-6.96	1.68	0.95-2.95
Children
0	1		1
1	1.53	0.98-2.39	1.33	0.75-2.36
>1	3.65	2.29-5.82	2.65	1.53-4.60

aOR refers to OR adjusted for the rest of the variables.

Discussion

In Spain, 70% of sexually active women aged 15–49 years used contraception during their first intercourse, which is similar to the results found in other countries.^22,23 Factors that influence use of contraception include age, level of education, age at first intercourse, year of first intercourse, country of origin, and religion, also consistent with other studies.²³ The importance of age and the year in which a woman started having intercourse is undoubtedly related to the cohort effect.²⁴ Older women who started having sexual intercourse longer ago used contraceptive methods less because they were not as readily available as in more recent times. In this study, initiating intercourse at an early age was associated with lower rates of use of contraception, which is in agreement with the results of various studies that highlight the relationship between early initiation of sexual activity and sexual risk behaviors.² Country of origin and religion make up women's cultural background, which can influence their behavior with respect to contraception in various ways, especially when women are young and are having their first intercourse.^2,25,26 Women with a lower level of education and lower socioeconomic position were found to use contraception less during their first intercourse.²³

In our study, 78% of sexually active women aged 15–49 years used contraception during the 4 weeks before the interview, a percentage similar to those of other European countries and higher than that in the United States.^9,27 In this group, the influence of level of education is less marked than at first intercourse, but women with a higher level of education continue to use contraception more than others, indicating socioeconomic inequalities.^27
–29 In other European studies, single women, those with a higher level of education, those with children, and those who had previously had an induced abortion generally used contraception more.⁸ Despite the importance of culture and religion in influencing the contraceptive behavior of practicing women, when faced with the realities of family life, they are inclined to attend to priorities other than those of strict doctrines, with birth control being more relevant than religious teachings.³⁰

We found that the use of contraception during the previous 4 weeks was largely associated with living with a partner and having children. It seems clear that living with a partner can affect women's contraceptive use, not only by predicting sexual intercourse but also because these women can share contraceptive decision making with their partners, which may not be the case for those who have sporadic relationships.^28,31 We observed a similar pattern among women who have children, and according to the number of children. Women who already have the number of children they want may, therefore, use more contraception, whereas those who do not yet have children or do not have the number they want may be trying to get pregnant. Moreover, level of education can have an important influence on patterns of childbearing; thus, educational level, number of children, and use of contraception are all related. It is necessary to highlight how much the use of contraception at first intercourse affects its use during the 4 weeks before the interview. Remarkably, socioeconomic factors influence the use of contraception by women who had used it during their first intercourse but do not seem to be related to women who had not used it. These results support previous evidence on how the use of contraception at first intercourse reinforces the habit of using it for the rest of a woman's fertile life.^32,33

The contraceptive method used most by women in Spain at first intercourse is the condom, followed by the pill, as is the case in other studies.^23,31,34 The same is true of contraception use during the previous 4 weeks. It has been noted that women in Italy and Spain use condoms more frequently than do those from Northern European countries or from the United States, who tend to use the pill in a higher proportion.^8,9,27

There was a significant use of ineffective contraceptive methods not only at first intercourse but also during sexual intercourse in the previous 4 weeks. Such ineffective methods included withdrawal, periodic abstinence, and spermicides or sponges as the sole contraceptive method used.³⁵ In previous studies, various reasons were given for not using contraception or for using ineffective methods, such as the perception of a low risk of pregnancy, having unplanned sex, fear that parents will find out they are having sex (in adolescents), problems with accessing contraception, their partner's preference for nonuse, or other reasons.⁵

The reliability of reported sexual behavior is difficult to verify. There may be a bias toward reporting behaviors seen as more socially desirable and avoiding reporting those that can cause emotional distress. However, if the survey is structured, as is the case in this study, so that participants can trust in its legitimacy and confidentiality and if the interviewer takes an appropriate professional approach, this can facilitate accurate disclosure of sensitive information.³⁶ It has also been found that the reliability of self-reported sexual behavior differs according to a variety of factors, including the time frame for recall. It has also been reported that incidence reports (such as first intercourse) are generally more reliably reported than frequency reports, where reliability decreases with longer recall periods and more frequent behaviors.³⁷

The main limitations of this study are related to survey design. Because this survey was designed to collect data other than those related to socioeconomic position, a number of important socioeconomic variables are not included. Unlike other studies, social class did not appear to be associated with contraception use in this study,¹¹ and although we were able to make a very close approximation, we cannot claim that social class does not influence women's contraceptive behavior in Spain because the occupational categories registered cannot be directly translated into social class groups. If we assume our approximation of social class is good enough, these results can reflect that when one aims to study the existing variability in the use of contraception in Spain, level of education would be a better indicator of socioeconomic position than social class derived from the occupation. Probably the latter indicator is not good enough for women who are out of the labor market,¹⁸ even if we take into account their partners' occupation. Moreover, data regarding parents' socioeconomic position were not available, which would have allowed us to perform a more detailed study of contraception use in adolescent women.³⁸

Another point is that although the participants' country of origin was reported, the sample size was not sufficient to stratify the analyses further. Thus, it was not possible to divide the group of developing countries into more categories, which would allow us to study differences between women of different origins. Nevertheless, considering the fact that most of these women do not have a high level of education and tend to be of social classes IV–V, the results shown according to country of origin are also a good approximation of socioeconomic inequalities in contraception use. Within the group of developed countries, most of the sample is from Spain, but this country was not separated into an isolated group because its behavior was the same as that of other developed countries.

An important limitation of this study is that the question that asks if some form of contraception was used during the previous 4 weeks is likely to overestimate its use during this time. This question does not discriminate between effective and ineffective methods; this information is only ascertained by a second question that asks about the method used. Moreover, answers indicating that an effective form of contraception was used during the previous 4 weeks do not guarantee that it was used during all instances of sexual intercourse during this period. This problem can be solved only by collecting data about how often contraception was used and the way in which it was used during this period; these data were not available in most of the studies conducted. Finally, it is important to note that this survey did not take into account the fact that some women may have been trying to become pregnant; this percentage may be significant, especially in some age groups.⁹

This study has two important strengths. First, members of the population who do not have sexual intercourse were excluded, which is important because almost 40% of women who report not having used contraception recently are not sexually active.⁹ Second, this is the first study that has analyzed socioeconomic inequalities in the use of contraception among the general population in Spain and where it has been possible to compare the use of contraception during the first intercourse to that during recent sexual activity within the same sample of participants. Accepting the limitations mentioned, our study demonstrates the presence of inequalities according to socioeconomic position in the use of contraception among women in Spain. This issue has not be extensively explored in higher-income countries,⁷ particularly in Spain, where only one previous study has been reported that analyzed data for the period 1985–1995¹⁵ and obtained results similar to ours.

The inequalities treated in this study are individual-level inequalities. It is highly possible that socioeconomic characteristics of the context of women's place of residence will also affect their use of contraception. Incorporating contextual characteristics would be a necessary step for understanding how socioeconomic factors can influence the use of contraception among women living in Spain on both individual and contextual levels.

In order to plan effective intervention strategies to promote the use of contraception, it is necessary to be aware of the current situation in a country.¹³ Many more studies of this issue are required in Spain, where interventions in populations at greatest risk are needed. One such intervention could involve working with adolescents who have not yet had their first sexual intercourse, as this could have important repercussions on their sexual habits in the future.³²

Footnotes

Acknowledgments

This study was funded by the Health-Care Technology Assessment Agency of Spain (PI07/90050). We thank Birgit Ferran, Dave Macfarlane, Gavin Lucas, and Mariona Casals for their help in correcting the English version of this article. This article forms part of the doctoral dissertation of Dolores Ruiz-Muñoz at the Pompeu Fabra University (UPF) of Barcelona.

Disclosure Statement

No competing financial interests exist.

References

Lete

, Bermejo

, Parrilla

et al. Use of contraceptive methods and risk of unwanted pregnancy in Spanish women aged 40–50 years: Results of a survey conducted in Spain. Eur J Contracept Reprod Health Care, 2007; 12:46–50.

Raine

, Minnis

, Padian

. Determinants of contraceptive method among young women at risk for unintended pregnancy and sexually transmitted infections. Contraception, 2003; 68:19–25.

Sable

, Libbus

. Beliefs concerning contraceptive acquisition and use among low-income women. J Health Care Poor Underserved, 1998; 9:262–275.

Olsen

. Sexual and reproductive health “choice”: Women living with contraception. Int J Public Health, 2007; 52:195–196.

Jones

, Darroch

, Henshaw

. Contraceptive use among U.S. women having abortions in 2000–2001. Perspect Sex Reprod Health, 2002; 34:294–303.

Ranjit

, Bankole

, Darroch

, Singh

. Contraceptive failure in the first two years of use: Differences across socioeconomic subgroups. Fam Plann Perspect, 2001; 33:19–27.

Bentley

, Kavanagh

, Smith

. Area disadvantage, socioeconomic position and women's contraception use: A multilevel study in the U.K. J Fam Plann Reprod Health Care, 2009; 35:221–226.

Spinelli

, Talamanca

, Lauria

. Patterns of contraceptive use in 5 European countries. European Study Group on Infertility and Subfecundity. Am J Public Health, 2000; 90:1403–1408.

Skouby

. Contraceptive use and behavior in the 21st century: A comprehensive study across five European countries. Eur J Contracept Reprod Health Care, 2004; 9:57–68.

10.

Font-Ribera

, Perez

, Salvador

, Borrell

. Socioeconomic inequalities in unintended pregnancy and abortion decision. J Urban Health, 2008; 85:125–135.

11.

Layte

, McGee

, Rundle

, Leigh

. Does ambivalence about becoming pregnant explain social class differentials in use of contraception? Eur J Public Health, 2007; 17:477–482.

12.

Lete

, Martinez-Etayo

. [Reproductive health: Some data and reflections.] Gac Sanit, 2004; 18,Suppl 1:170–174.

13.

Lete

, Duenas

, Martinez-Salmean

et al. Contraceptive practices and trends in Spain: 1997–2003. Eur J Obstet Gynecol Reprod Biol, 2007; 135:73–75.

14.

Lete

, Bermejo

, Coll

et al. Use of contraceptive methods in Spain: Results of a national survey. Contraception, 2001; 63:235–238.

15.

Ruiz

, Cabré

, Castro

, Solsona

. Factores socio-demográficos que influyen en el uso de anticonceptivos y en la selección de métodos. Anticoncepción y salud reproductiva en España: Crónica de una (r)evolución. Consejo Superior de Investigaciones Cientificas. Instituto de Economia y Geografia: Madrid, 2005; 183–203.

16.

Galobardes

, Lynch

, Smith

. Measuring socioeconomic position in health research. Br Med Bull, 2007; 81–82:21–37.

17.

Galobardes

, Shaw

, Lawlor

, Lynch

, Davey

. Indicators of socioeconomic position (part 2) J Epidemiol Community Health, 2006; 60:95–101.

18.

Galobardes

, Shaw

, Lawlor

, Lynch

, Davey

. Indicators of socioeconomic position (part 1) J Epidemiol Community Health, 2006; 60:7–12.

19.

Agudelo-Suarez

, Ronda-Perez

, Gil-Gonzalez

et al.

[The migratory process, working conditions and health in immigrant workers in Spain (the ITSAL project).]

Gac Sanit, 2009; 23,Suppl 1:115–121.

20.

Centro de Investigaciones Sociológicas (CIS). Fecundidad y valores en la España del siglo XXI. Ficha técnica, 2009. Estudio CIS N^a 2639.

21.

Domingo-Salvany

, Regidor

, Alonso

, Alvarez-Dardet

. [Proposal for a social class measure. Working Group of the Spanish Society of Epidemiology and the Spanish Society of Family and Community Medicine.] Aten Primaria, 2000; 25:350–363.

22.

Manning

, Longmore

, Giordano

. The relationship context of contraceptive use at first intercourse. Fam Plann Perspect, 2000; 32:104–110.

23.

Svare

, Kjaer

, Thomsen

, Bock

. Determinants for non-use of contraception at first intercourse: A study of 10841 young Danish women from the general population. Contraception, 2002; 66:345–350.

24.

Wellings

, Nanchahal

, Macdowall

et al. Sexual behaviour in Britain: Early heterosexual experience. Lancet, 2001; 358:1843–1850.

25.

Galazios

, Tsikouras

, Liberis

et al. Attitudes towards contraception in three different populations. Clin Exp Obstet Gynecol, 2008; 35:22–26.

26.

Jones

, Darroch

, Singh

. Religious differentials in the sexual and reproductive behaviors of young women in the United States. J Adolesc Health, 2005; 36:279–288.

27.

Frost

, Darroch

, Remez

. Improving contraceptive use in the United States. Issues brief. New York: Alan Guttmacher Institute, 2008; 1–8.

28.

Martin

. Contraceptive use patterns among Spanish single youth. Eur J Contracept Reprod Health Care, 2005; 10:219–228.

29.

Singh

, Darroch

, Frost

. Socioeconomic disadvantage and adolescent women's sexual and reproductive behavior: The case of five developed countries. Fam Plann Perspect, 2001; 33:251–258.

30.

Srikanthan

, Reid

. Religious and cultural influences on contraception. J Obstet Gynaecol Can, 2008; 30:129–137.

31.

Gomez

, Sola

, Cortes

, Mira

. Sexual behaviour and contraception in people under the age of 20 in Alicante, Spain. Eur J Contracept Reprod Health Care, 2007; 12:125–130.

32.

Shafii

, Stovel

, Davis

, Holmes

. Is condom use habit forming? Condom use at sexual debut and subsequent condom use. Sex Transm Dis, 2004; 31:366–372.

33.

Shafii

, Stovel

, Holmes

. Association between condom use at sexual debut and subsequent sexual trajectories: A longitudinal study using biomarkers. Am J Public Health, 2007; 97:1090–1095.

34.

Darroch

, Singh

, Frost

. Differences in teenage pregnancy rates among five developed countries: The roles of sexual activity and contraceptive use. Fam Plann Perspect, 2001; 33:244–250.

35.

Lete

, Bermejo

, Coll

et al. Spanish population at risk of unwanted pregnancy: Results of a national survey. Eur J Contracept Reprod Health Care, 2003; 8:75–79.

36.

Mitchell

, Wellings

, Elam

, Erens

, Fenton

, Johnson

. How can we facilitate reliable reporting in surveys of sexual behaviour? Evidence from qualitative research. Cult Health Sex, 2007; 9:519–531.

37.

Fenton

, Johnson

, McManus

, Erens

. Measuring sexual behaviour: Methodological challenges in survey research. Sex Transm Infect, 2001; 77:84–92.

38.

Hassani

, Kosunen

, Rimpela

. The use of oral contraceptives among Finnish teenagers from 1981 to 2003. J Adolesc Health, 2006; 39:649–655.