Knowledge about the intrauterine device and interest in using it among women users of primary care services*

Objective: to analyze the level of knowledge about the intrauterine device, the interest in using it and the relationship between these events among women in reproductive age. Method: cross-sectional study conducted with 1858 women between 18 and 49 years old, attending Primary Health Care Facilities. Data were obtained in face-to-face interviews. The level of knowledge was evaluated by items with answers options “agree”, “disagree” and “I don’t know”. Knowledge was categorized as below/equal and above the median. Chi-square and multiple logistic regression were used in Stata 14.2 (95% confidence level). Results: intrauterine device current use was not frequent (1.7%; n=32) and the level of knowledge was higher among women between 25 and 34 years old, white, living in Aracaju (Sergipe), who were more educated, and who were currently using or had already used the intrauterine device. Interest in using the intrauterine device (38.0%; n=634) was higher among younger women, single, more educated, had health insurance, no children and with higher level of knowledge about the intrauterine device. Conclusion: the level of knowledge about the intrauterine device was associated with the interest in using it.

Thus, investigating the knowledge about the IUD and the interest in using it can provide insights to explain the low percentage of use in the country and to find strategies to promote and expand its use. A review of studies on the knowledge of health professionals and of women/couples about the IUD emphasized that in lowand middle-income countries, such as Brazil and other Latin American countries, there is a lack of information on the perspective of these groups regarding IUDs, which in turn, limits the knowledge about what women consider when choosing or rejecting a contraceptive (19) .
The objective of this study was to analyze the level of knowledge on the IUD, the interest in using it and the relationship between these events among women in reproductive age who were users of Primary Health Care Facilities (PHCF) in the three cities in Brazil.

Method
We conducted a quantitative, cross-sectional study with women ranging 18 to 49 years of age, who were users of PHCF from three Brazilian cities: São Paulo (São Paulo state), Aracaju (Sergipe state) and Cuiabá (Mato Grosso state).
The parameters to determine the sample size in each city were: proportion of women estimated equal to 50%; 95% confidence level; relative sampling error of 5%; and design effect (deff) of 2 (20) . Sample size calculation indicated the need for 768 women, 800 in São Paulo and 385 in Aracaju and 368 in Cuiabá.
Considering that the estimates should be obtained from women using contraceptive methods, which have an estimated proportion of 80% (4) , the number of women who should be interviewed was 1000 in São Paulo and 482 in Aracaju and in Cuiabá. Then, considering the percentage of 25% of women who would not answer the questionnaire (refusal or loss due to interviewer's problems) and 33% of women between 18 and 49 years old who would not be eligible for the interview (the exclusion criteria were never having had sexual intercourse, using a permanent contraceptive method and reporting not being aware of IUD) (4) . Thus, 1993 women should be selected in the city of São Paulo, in order to obtain 1000 valid interviews, and 963 women in Aracaju and 963 in Cuiabá, in order to obtain at least 482 valid interviews in each of these capitals.
The sampling plan was conducted in two-stage cluster sampling. The first-stage (primary sampling unit) consisted of the PHCF, drawn with probability proportional to size, measured by the number of Borges ALV, Araújo KS, Santos AO, Gonçalves RFS, Fujimori E, Divino EA. In order to elaborate the variable "level of knowledge about the IUD", 12 statements about the mechanisms of action, indications for use and side effects of the IUD were elaborated, based on studies that also measured the level of knowledge about the IUD (10)(11)(21)(22)(23)(24) .
Women who knew or had heard about the IUD responded to each statement with the following response options: "agree", "disagree" or "I don't know". Because knowledge is a latent variable, that is, one that cannot be directly measured or observed, its analysis considered the subjectivity and the difficulty of "measuring" women's knowledge about the IUD (25) . This was done through three strategies: 1) Item Response Theory (IRT); 2) Factor analysis, with an eigenvalue greater than one, according to the Kaiser criterion; 3) Cronbach's Alpha. The intention was to choose, based on these strategies, which statements should compose the construct "knowledge level about the IUD".
The IRT showed that the following items presented moderate, high or very high discrimination and therefore were considered in the next step (factor analysis): item 4 -The man feels the IUD during sexual intercourse (correct option=disagree, discrimination index 1.59); item 7 -The IUD is inserted through surgery (correct option=disagree, discrimination index 1.94); item 8 -The IUD is abortive (correct option=disagree, discrimination index 1.58); item 9 -After removing the IUD, women have difficulty getting pregnant (correct option=disagree, discrimination index 1.80); item 10 - The IUD increases the risk of uterine cancer (correct option=disagree, discrimination index 1.99); and item 12 -The IUD has many unpleasant side effects (correct option=disagree, discrimination index 1.45).
The factor analysis confirmed that these six items were part of a single factor (Eigenvalue equal to 1.665), that is, they compound a single latent construct/variable, which is IUD knowledge. Also, all items showed factor loadings greater than 0.30 (item 4=0.540; item 7=0.390; item 8=0.541; item 9=0.571; item 10=0.597; item 12=0.494). Kaiser-Meyer-Olkin (KMO) was 0.798, which shows that the data set was adequate to proceed with factor analysis (26) . Finally, Cronbach's alpha coefficient was 0.706, indicating that this latent variable had adequate reliability.
Correct answers were coded as 1, while incorrect and "I don't know" answers were coded as zero. Each woman had her score added up, obtaining a final regression analysis, the variable use of contraceptive methods also considered women who were pregnant, so it had one more category, "is currently pregnant". The main covariate was IUD use (previous and/or current).
For the analysis of interest in using the IUD, women with no female sterilization and with partners without vasectomy, who did not use IUD and had never used it, were asked the reasons why they had never used it and if they would like to use it in the future. Women who did not know if they would like to use IUD or who said that they would not want to use it expressed their reasons. Aspects associated with future interest in using an IUD were analyzed using the chi-square test. Subsequently, multiple logistic regression was conducted, with simultaneous entry of the independent variables already mentioned. In this model, the main covariate was the level of knowledge on the IUD. The dependent variables were the level of knowledge on the IUD (dichotomous: below the median and equal to or greater than the median) and the interest in using IUD (also dichotomous: no and yes). All data were analyzed in Stata 14.2. A confidence level of 95% was considered and the Hosmer-Lemeshow test was applied to check models goodness of fit. were obtained before beginning data collection.

Results
As the purpose of this study was to analyze the relationship between knowledge on the IUD and interest  (Table 1).
Borges ALV, Araújo KS, Santos AO, Gonçalves RFS, Fujimori E, Divino EA.   Women who had never used the IUD (n=1759) were asked if they would like to use it, and 58.7% said they would not. The most cited reasons for never having used the IUD was that they were not interested in the method because they were satisfied with their current method (42.1%), followed by the fact that they had no information about the IUD and that it has never been offered to them (26.7%). The most cited reasons for not having interest in using the IUD in the future were lack of interest in the method and satisfaction with the current method (25.1%), followed by fear of the insertion procedure (13.4%) and desire for an irreversible method (12.3%). Rev. Latino-Am. Enfermagem 2020;28:e3232.   Rev. Latino-Am. Enfermagem 2020;28:e3232.

Discussion
Our study on the knowledge on the IUD and interest in using it among women attending PHCF in three Brazilian state capitals showed that a reasonable proportion of women had a level of knowledge below the median, that a little more than third were interested in using an IUD in the future and, finally, that the level of knowledge on the IUD was associated with current and previous use of IUD and with the interest in using it.
These results are consistent with findings from other contexts, in which the level of knowledge on the IUD is considered unsatisfactory (10)(11)(12)21,23) and is related to socio-demographic variables such as age, education and race/skin color (27) . Thus, younger women, self-identified as non-white and with lower level of education had the lowest level of knowledge on the IUD. Studies on reproductive health analyzing the use of contraceptive methods, preconception preparation, pregnancy planning and prenatal care found that groups with this profile face more difficulties to access this type of care and to engage in these practices (15,28) .
The items with high proportions of incorrect answers were "The IUD has many unpleasant side effects" and "The IUD increases the risk of uterine cancer". Similar results were obtained elsewhere (21) , from data from a national survey in the United States and verified that women associated IUD with infertility and cancer. Likewise, other authors (22) found that fear of side effects, of pain and of the infection´s possibility were reasons for not using the IUD. In fact, there are two side effects commonly associated with the copper IUD, that is the only LARC available at PHCF in the SUS: increased menstrual flow and worse menstrual cramps.
Other side effects are described in more specific cases and are not reported by most users of copper IUD. Very rare complications related to copper IUD are pelvic inflammatory disease and perforation of the uterine wall.
No type of cancer has been associated with its use (16) .
In addition to the fear of side effects and risk of cancer, the literature also shows that women do not choose the IUD as a contraceptive method because they are afraid that the IUD might move around the body; because they are afraid of the insertion pain; because they depend on a health professional for insertion and removal of the device; because they think that there is an increased risk of ectopic pregnancy and infections; because they think that it is less effective than the pill; among other reasons (10)(11)(21)(22)(23) .
The scenario described, in which many users do not know whether the IUD is associated with uterine cancer and what are its side effects, or mistakenly believe that it is inserted through surgery, seems to reveal that women in these three capitals would not choose a method surrounded by so many uncertainties, including doubts regarding its effectiveness and effects on the body. This is mainly because knowledge about the method might be related to its use (29) , even though this was not observed in other methods users such as emergency contraception (30) .
In our study, we did not analyze the sources of information on the IUD, but it can be assumed that younger women access the internet to better understand the method and solve their doubts. This, on the one hand, is positive, because of the easy access to information; however, it should be noted that not all websites on contraception are reliable. A US study found that the quality of information on the IUD available on specialized websites is inconsistent and that about half of these websites provided misleading information, which could contribute to the method underutilization (31) .
On the other hand, there is evidence that educational interventions increase the proportion of women with positive attitude in relation to the IUD and that women with prior knowledge around the method are more interested in using it (22,24) .
Despite the low percentage of current and/or previous use of the IUD, the proportion found among the women studied was higher than the one found in the 2006 DHS (4) . However, this comparison should be carefully taken, given that women who used irreversible methods were excluded from our sample, as were those who reported not being aware of IUD, which may have overestimated the percentage.
As observed in our findings, the level of knowledge on IUD is strongly associated with its use and interest in using it, which may be related to the fact that it is a stigmatized method, both among women and health professionals. A study (11) conducted with US adolescents found that most did not consider the IUD as a suitable method for them. That is, adolescents do not include the IUD in the list of contraceptive methods indicated for their age group. However, the WHO considers the IUD a safe method for most women, including nulliparous women and adolescents (16) . Thus, this result may be It is interesting to note that young women, between 18 and 24 years old, showed more interest in using IUD compared to older women. It is possible that the use of a long-term method is in line with their reproductive intention to delay pregnancy. However, in the population studied, reproductive intention was not associated with interest in using IUD, but parity was. Women with children were more likely to be interested in using IUD when compared to women without children. Women in stable unions were less likely to be interested in using the method. These associations seem complex, but the profile of women who would be interested in using an IUD is clear: young, single, and with children. These women do not yet meet the requirements for adopting an irreversible method but may also not be willing to use a method that requires self-discipline and daily use, such as the contraceptive pill, making the IUD a very convenient option.
On the "CHOICE" project (38) , which offered counseling and contraceptive supplies to women at a clinic in the United States, with the advantage that the methods were available on the same day at no cost, the researchers noted that adolescents and young women,