Strategies for notifying sexual partners of people with sexually transmitted infections: a randomized clinical trial*

2 Universidade Estadual do Ceará, Programa de Pós-Graduação Cuidados Clínicos em Enfermagem e Saúde, Fortaleza, CE, Brazil. ABSTRACT Objective: To compare the effectiveness of verbal communication and communication by card in getting sexual partners of people with sexually transmitted infections to attend a health service and the factors associated with the success of these types of communication. Method: Clinical, controlled, and randomized study, whose intervention was offering a reporting card for index patients to hand to their sexual partners. Results: The sample was 189 index patients, 94 of whom were in the control group, and verbally invited their sexual partners to receive care, and 95 were allocated to the intervention group, and took their partner’s reporting card to their partners as a way to invite them to receive care. The percentage of partners invited by card who came to the service was 52.6%, in contrast with 43.6% among partners who were invited verbally, but no significant statistical difference was found (p=0.215). The factors associated with failure to convince partners to come to the service were: not living with the partner (p=0.0001); not having a steady partner (p=0.0001); having casual partners (p=0.028); and using condoms with a steady partner (p=0.045). The infection type did not influence the studied partners’ visits to the service. Conclusion: Given the failure to achieve effectiveness when applying the reporting by card, the authors recommend another card model containing information for partners to be used in combination with other methods. Brazilian Clinical Trials Registry: RBR-7jp5mr.


INTRODUCTION
Partner reporting (PR) is an action that benefits sexual partners of index patients (for instance, people with sexually transmitted infections or STIs) by making them aware that they need to visit a health unit to be informed about their exposure, do diagnostic tests, and have access to treatment and/or prophylaxis when necessary (1) . It is one of the main prevention strategies for STIs in developed countries in North America and Europe. Efforts have been made to incorporate this practice into services, together with a continuous commitment to using different reporting methods to facilitate the access of sexual partners of index patients to health services (2)(3) .
A systematic review showed the relevance of PR, taking into account especially the high frequency of human immunodeficiency virus (HIV ) diagnosis in partners of index patients infected with this virus (4) . Another study reinforced its importance, mostly in cases with serodiscordant partners, because offering prophylaxis to seronegative partners is an effective strategy to prevent the transmission of the disease (5) .
Usually, three types of PR are used. The first is known as PR by the index patient, who is encouraged to tell their partner(s) about their diagnosis. The second method is reporting by contract. In this situation, the healthcare professional drafts an agreement under which the index patient must, in a short period, inform their sexual partner(s) and refer them to a counseling service. If this scheme is unsuccessful, the professional takes action, keeping the information confidential. In the third method, PR is carried out by a professional, who directly contacts the sexual partner(s), maintaining the index patient's anonymity (3,(6)(7) .
Structural barriers, deficient counseling, professionals' insufficient knowledge, and lack of minimum training to handle STI cases stand out among the difficulties in implementing PR, in addition to personal barriers resulting from fear of the reaction of partner(s) and the stigma associated with STIs (8)(9)(10) .
In Brazil, the lack of visibility of PR occurs because of the evident insecurity professionals have in carrying out this strategy, especially as a consequence of lack of the institutional support necessary to work the cases and of details about how this action can be developed by health professionals (8)(9) .
When the method of making the index patient responsible for notifying their sexual partners is chosen, it is recommended, in Brazil, that the professional instruct the patient to do it in two ways. In the first, the index patient him/herself can go to the unit accompanied by their sexual partners. In the second, the professional gives the index patient a PR card to be handed to the sexual partners, who can visit the health unit, bringing the card, without the need for the presence of the index patient. If the partners fail to come to the unit after 15 days, the method in which the professional communicates with the partners via correspondence or other means of communication that guarantee confidentiality (phone, email, or social media) is applied (6) .
There is a lack of details in the Brazilian literature about how PR is carried out, especially by using PR cards (8)(9) . These cards were devised to be a more feasible means of reporting, because they cause less embarrassment when compared to verbal reporting. Data and/or passwords contained on the card allow professionals to know the reasons that originated the reporting without the need to ask further questions (6) . Additionally, the card allows professionals to identify the partners of the index patient so they can receive care immediately, respecting secrecy and the confidentiality of the information provided by the person who originated the reporting.
Studies focused on this subject may provide resources for researchers to know the best way to carry out PR, aiming to help professionals and facilitating access of sexual partners of people with STIs to the healthcare network. It must be emphasized that most studies on PR are limited to the reality of other countries (5) . The objective of the present study was to compare the effectiveness of verbal communication and communication by card in getting sexual partners of people with STIs to come in to health services and factors associated with the success of these types of communication.

Study type
This was a randomized controlled clinical trial.

Setting
The study was carried out from August 2016 to July 2017 in an STI reference center in Fortaleza, state of Ceará, Brazil. In addition to providing care for cases referred from primary care, the service offers medications necessary for treatment and develops a set of complementary actions such as: providing counseling; offering rapid tests to diagnose HIV, syphilis, and hepatitis B and C; collecting material to perform venereal disease laboratory testing in cases that showed a positive result in rapid tests for syphilis; and providing condoms, vaccination against hepatitis B, instructions, communication to sexual partners, and follow-up for test results and healing control (6) . To provide this type of care, the unit has a team made up of four nurses, two doctors, one psychologist, two biochemists, one pharmacist, two laboratory technicians, three nursing aides, and one social worker.

Study population
All patients diagnosed with STIs (clinically or by serological or laboratory tests) during 11 consecutive months were included in the study. This period was the time necessary to complete the sample calculation for each group. Clinical diagnosis of STIs included people with: 1) anogenital ulcer syndrome; 2) warts; 3) urethral or cervical discharge; 4) pelvic inflammatory disease (PID) syndrome; and 5) laboratory diagnosis of PID, syphilis, trichomoniasis, HIV, and hepatitis B and C, as determined by the Clinical Protocol and Therapeutic Guidelines for comprehensive care of people with STIs (6) .

Selection criteria
The selection criteria were: 1) people 18 years old or older, with a serological or laboratory diagnosis of STI, hepatitis B or C, and/or a syndrome associated with an STI (genital ulcers, urethral discharge, genital warts, PID) diagnosed when receiving care in the health unit, the socalled index patients, who had sexual partners to be notified (6) regardless of the emotional or sexual bond and who showed a desire to communicate with their partners about their condition; and 2) sexual partners who were notified by index patients and came to health units to receive care. A steady relationship was defined as one with an emotional bond, regardless of marital status, and a casual relationship was considered to be sexual contact with a person involving no proximity or affectivity bond and no intention of having other sexual encounters.
The exclusion criteria were: 1) index patients who did not agree to notify their sexual partners or did not agree to have them referred during one month after the STI diagnosis; and 2) patients who had previous knowledge of the diagnosis. Women diagnosed with vaginal discharge syndrome were also excluded, because of the possible existence of alterations caused by factors not associated with STIs (11) .

intervention
The study examined two groups, defined as: a) the intervention group, made up of people with STIs who took the reporting cards to their sexual partner(s) as a way to invite them to receive care; and b) the control group, made up of people with STIs who verbally invited their sexual partners to receive care. The patients allocated in the control group received only verbal information to invite/encourage their sexual partners to visit the health service.
Sealed envelopes handed out by the research assistant were used to randomize the patients in both groups. The envelopes contained written content informed the subjects as to which intervention type they would receive. The study outcome was whether sexual partners came to the health service after the use of the reporting card. The maximum period for waiting/evaluation of whether sexual partners came in for both groups was one month.
The PR card proposed by the Brazilian Health Ministry was used (6) . The first part of the card (part A) shows the name of the health unit and is used only by the professional that carried out the reporting. The second part (part B) is intended for the patient to hand to their sexual partners. Both parts have the same number and are put together when the partners visit the health unit. The numeric sequence works as a "password" to be used confidentially by the professional to allow access to the information available in the medical record of the index patient, an alternative designed to ensure total anonymity for the person who generated the reporting (6) . When there is more than one partner, each receives a card, and all contain the same index patient number. This procedure allows for knowing afterwards which partners came to the health unit.

data collection
There were no requirements for advance scheduling of the day and time for the partners to visit the unit, because priority care in the unit was guaranteed to the users of the cards. Consequently, PR was considered concluded when the service was sought by the notified people (without registering the time when they did it).
Data collection was carried out by using a form designed by one of the researchers to obtain information about index patients. It had the following sections: I) reporting type: verbal or by card; II) sociodemographic characterization: gender, age, skin color, level of education, whether the person lived with the partner, religion, per capita income, sexual orientation, professional activity, origin; III) health habits: alcohol consumption, illicit drug consumption, whether the person smoked when the form was applied; IV) sexual behavior: sexual orientation, number of sexual partners in the past 12 months, relationship type (steady or casual), whether the person had a casual partner to bring to the service, frequency of use of condoms in steady and/or casual relationships; V) clinical data: type of STI 1 (main STI of the index patient, which originated their visit to the unit), diagnosis form for STI 1, type of STI 2 (second type of STI identified in the same index patient), diagnosis form for STI 2, type of STI 3 (third type of STI identified in the same patient).
The outcome occurrence proportion was defined based on previous studies that reported a mean PR acceptance rate of 30% (13)(14) . The following values were adopted when applying the formula: Zα=1.96, Zβ=0.84, p=30%, d=20%. The formula indicated the need to include 82 people with STI in each group, totaling 164 participants. After adding a safety percentage equal to 10% to account for potential losses, a sample of 90 people in each group was suggested. The final sample after recruitment was 189 participants, with 94 allocated to the control group and 95 to the intervention group ( Figure 1).

data analySiS and treatment
Quantitative data were entered into an Excel software spreadsheet and transferred to SPSS version 20.0, license number 10101131007, with which they were processed. Descriptive data analysis was performed by obtaining absolute and relative frequencies. The answer to the question "Did the patient's partner come in?" was considered the dependent variable for statistical associations, and the other variables were classified as independent. The association between categorical variables was evaluated by applying Pearson's chi-squared test, and the odds ratio (OR) and the 95% confidence interval (CI) were calculated. Analyses with p<0.05 were considered statistically significant.

RESULTS
Regarding the reporting type, 94 index patients carried out verbal reporting (control group) and 95 patients carried out reporting by card (intervention group). In the latter group, the percentage of sexual partners who came to the health service was 52.6% (n=50), whereas the percentage in the group of patients who informed their partners verbally was 43.6% (n=41). However, there was no statistically significant difference between the two reporting types (verbal and by card) regarding success in getting partners to come to the health service (p=0.215) ( Table 1).  Considering the 189 index patients who participated in the study, regardless of the PR types adopted, the analysis of correlation with other factors that could be associated with success or failure in getting their sexual partners to come to the service resulted in the identification of some important aspects, which can be described as follows.
The only item among the sociodemographic characteristics and health habits of index patients that showed a statistically significant association with failure to get their partners to come to the service was not living with the partner (p=0.0001; 95% CI=1.90-6.76), which was associated with a 3.58-fold increase in the chances of index patients not getting partners to come in (Table 2).
Concerning the association of sexual behavior data and partners coming to the service, the statistically significant factors were not having a steady/stable partner (p=0.001; 95% CI=1.95-18.04), having a casual relationship (p=0.028; 95% CI=1.08-5.14), and using condoms with the steady partner (p=0.045; 95% CI=1.01-3.86). The probability of not getting the partner to come in among the index patients who did not have a steady partner was 5.93 times higher when compared with those who had a steady partner. The chances of index patients who had casual partners not getting them to come to the health unit were 2.35 higher in comparison with those who did not have this type of relationship. Additionally, the probability of index patients who used condoms with a steady partner always/sometimes getting them to come to the service was 1.97 higher when compared with those who did not use condoms (Table 3). Regarding the STI diagnosis in index patients, a higher prevalence of syphilis (49.2%) in the group of main STIs in index patients that got their partners to visit (STI 1) and a higher frequency of diagnosis by using the rapid or laboratory test (52.0%) were identified. Regarding the second type of STI detected in the same index patient (STI 2), HIV infection (41.0%) and rapid or laboratory test as a method of diagnosis (70.6%) were predominant. However, bivariate analysis taking into account the STIs and the way the index patient was diagnosed did not show a statistically significant association with getting partners to come to the health unit (Table 4).
Regarding the STI diagnosis of index patients and getting partners to come to the health unit, there was a higher percentage of visits by partners when STI 1 was syphilis (50.5%) and STI 2 was trichomoniasis (100%), followed by HIV infection (85.7%). It must be stressed that partner visits to the health unit did not occur in the cases in which STI 1 or STI 2 of the index patients was hepatitis B.
Getting partners to come to the health unit after diagnosis in index patients by rapid or laboratory test prevailed in all the cases, for both STI 1 (50.0%) and STI 2 (66.7%). Despite this result, it was found that STI type and diagnosis method did not show a statistically significant association with partner visits to the health unit (Table 4).
There was only one index case with a diagnosed third infection (STI 3), which was classified as an anogenital wart associated with human papillomavirus. However, this infection was not confirmed in the partner who came to the health unit.

DISCUSSION
The present study did not show a significant difference in success in getting partners to come to the health service when verbal reporting and reporting by card were compared, which corroborates the results of other international investigations (5,15) . In the face of this finding, it is necessary to emphasize the importance for STI programs to introduce a card model containing informative content that could be used as a tool to support other types of PR and counseling (5) .
Data collected in the present study indicated the occurrence of flaws in the execution of actions in both situations, given that less than half of the partners came to the health unit, a result similar to that found in other studies (15)(16)(17) .
Even with PR by card being a feasible modality, additional alternatives can be used to increase the proportion of exposed contacts and overcome the communication barriers (15) . Some of these options are: providing treatment for patients to give to their partners (in the cases involving curable STIs) (5,13) ; offering further support to index patients (active search by professionals by phone or home visits) (18) ; introducing new communication modalities based on internet insertion strategies or mobile devices and apps for PR (12,19) ; and using electronic reporting cards, which can replace traditional ones (5) .
In Brazil, although the use of new technological tools to notify partners has been recommended since 2015 (6) , efforts to put them into practice are necessary because of frequent reports of lack of other PR alternatives to be applied by professionals, other than those focused on the need for patients themselves to commit to informing their sexual partners (8)(9) .
However, increasing the application of these new technologies (email, social media, cell phone apps) in Brazilian health services is challenging, because of weaknesses reported by professionals that pose a barrier to their implementation.
These result from lack of technical training in the STI area and of specific knowledge about handling sexual partners, in combination with insufficient time, lack of clarity regarding the correct way to carry out PR, and lack of institutional support for the management of these cases, especially when new PR modalities are inserted (8)(9) . Consequently, investments involving actions to improve the management process and the infrastructure of the services are urgent to achieve progress in PR, with technical training of health professionals in comprehensive care of people with STIs.
Additionally, it implies the need to create guidelines that include these new PR strategies as an important public health tool, with clear instructions about how they must be used, respecting the right to anonymity and confidentiality of information for people with STIs and, most importantly, the patient's decision (20) .
Regarding the correlation between index patients and whether their partner(s) came to the health unit, the only item in the list of other sociodemographic characteristics of index patients associated with higher chances of not getting sexual partners to come in was not living with these contacts. This finding reinforces the evidence that implementing this action depends especially on the type of relationship between index patients and their sexual partners, mainly when there is a higher possibility of doing it in a direct and face-to-face way, given that it involves a private subject that is inherent to STI diagnosis (21) . The approximation resulting from living together facilitates open communication and trust in relationships, which are considered to be factors that influence the decision to notify a partner (22) .
The present study showed higher chances of failure to get partners to come in for patients who did not have steady relationships when compared with those who said that they had this type of relationship. Consequently, the data pointed *Pearson's chi-squared test **Genital ulcer, urethral discharge, associated pelvic pain (PID), anogenital wart …continuation to a higher probability of not getting casual sexual partners to come to the health unit, in comparison with that of noncasual ones. Similar results were described in a study on PR carried out in the United States, according to which patients who said that they had casual sexual partners were less likely to notify them (23) , confirming that success in this action is more common in cases involving steady relationships (24) . Analysis identified a higher probability of partners not coming to the health unit in the group of index patients who said that they used condoms in comparison with those who reported not using them at all, which corroborated a study that showed the existence of higher chances of notifying partners in patients who said that they had not used condoms in their last intercourse events when compared to those who said that they had used this protection alternative (16) .
Regarding the correlation between success in getting sexual partners to come to the health unit and STI types (regardless of having more than one infection) or the way the diagnosis was obtained (syndromic, laboratory, or rapid tests), analysis showed no association with any of these variables, a fact that had been verified in another study (24) . However, these factors were pointed out as being able to influence the patient's decision regarding PR, especially if the case involves the presence of HIV, because of the chronic nature of the viral infection and the stigma associated with it (22) .
These divergent results corroborate a systematic review on the effectiveness of PR methods in people with STIs, which indicated the nonexistence, among the evaluated methods, of one that was considered to be highly effective in recruiting partners for any type of STI (1) .
Although the present study did not find differences regarding the way the diagnosis of index patients was obtained and the success of the actions, especially in symptomatic cases diagnosed by using the syndromic approach, the authors emphasize that recruiting partners of patients with symptoms associated with STIs is convenient, because the evidence that the presence of these symptoms itself favors involuntary disclosure of diagnosis of these infections and the search for information related to sexual partners is strong (1,21) .
The limitation of the present study was the small sample size, which affects the power to detect statistical associations.

CONCLUSION
Failure of sexual partners to come to the health unit when using both the verbal approach and the provision of the PR card reinforces the need to introduce other PR methods to increase the success of this action. Given the lack of evidence of higher effectiveness in the approach that resorted to the use of cards, the authors propose evaluating this strategy as a tool to be combined with other reporting methods and recommend that STI programs develop a card model focusing on informative messages intended for sexual partners.
In cases involving casual encounters, the authors suggest the incorporation of new technologies to recruit partners by using the internet or text messages sent through cell phones. Implementing these technologies involves greater participation by professionals and greater efforts in health services to make resources available for these alternatives to be put into practice.