Disrespectful care in family planning services among youth and adult simulated clients in public sector facilities in Malawi

Background: Provision of high-quality family planning (FP) services improves access to contraceptives. Negative experiences in maternal health have been documented worldwide and likely occur in other services including FP. This study aims to quantify disrespectful care for adult and adolescent women accessing FP in Malawi. Methods: We used simulated clients (SCs) to measure disrespectful care in a census of public facilities in six districts of Malawi in 2018. SCs visited one provider in each of the 112 facilities: two SCs visits (one adult and one adolescent case scenario) or 224 SC visits total. We measured disrespectful care using a quantitative tool and field notes and report the prevalence and 95% confidence intervals for the indicators and by SC case scenarios contextualized with quotes from the field notes. Results: Some SCs (12%) were refused care mostly because they did not agree to receive a HIV test or vaccination, or less commonly because the clinic was closed during operating hours. Over half (59%) of the visits did not have privacy. The SCs were not asked their contraceptive preference in 57% of the visits, 28% reported they were not greeted respectfully, and 20% reported interruptions. In 18% of the visits the SCs reported humiliation such as verbal abuse. Adults SCs received poorer counseling compared to the adolescent SCs with no other differences found. Conclusions: We documented instances of refusal of care, lack of privacy, poor client centered care and humiliating treatment by providers. We recommend continued effort to improve quality of care with an emphasis on client treatment, regular quality assessments that include measurement of disrespectful care, and more research on practices to reduce it. This sub-study was part of a larger evaluation of the national family planning program in Malawi [28]. The aim of the evaluation was to determine whether district-level differences in fertility and contraceptive use were associated with quality of family planning services. The evaluation measured quality of care using direct observation, client exit interviews, knowledge assessments, and simulated clients (SCs). SCs are trained to act as clients seeking services in order to evaluate care without the provider knowing that they were being assessed [29]. This sub-study used the SC protocol and expanded it to collect information related to disrespectful care for facility-based family planning providers. For the larger evaluation, we selected six of the 28 administrative districts in Malawi: one group of three low FP outcome districts and another group of three FP high outcome districts. We developed a district database of FP outcomes from the 2015–2016 DHS: changes in total fertility rate, modern contraceptive prevalence, unmet need, demand for FP services satisfied, and adolescent pregnancy. We purposefully selected Chitipa, Dedza and Salima as the low outcome group based on the FP outcomes relative to the other districts, and to maximize variation in geographic spread and religious representation (i.e. at least one district had a significant Muslim population). We matched this district group to high outcome districts (Machinga, Mangochi and Nkhata Bay) based on theorized confounders including proportion of rural households, women’s education, religion, poverty, and facilities per population by district using coarsened exact matching – method of matching by categories of values rather than exact values [30]. This study is a census of all public sector facilities in the six districts. We did not include private for-profit facilities, facilities managed by non-government organization (e.g. Banja La Mtsogolo), or religious facilities that do not offer contraceptives. In 2016, Harris et al. published a specific framework of disrespectful care and abuse for family planning by applying a framework previously developed for intrapartum care [31]. They defined respectful family planning care as support for women’s contraceptive method choice free from coercion, and creation of a positive, client-centered environment [31]. Adapting this framework, we defined four domains of disrespectful care: poor client-centered care, non-private consultations, refusal of care, and non-dignified care [31]. We conducted a desk review of existing quality of care indicators and assessment protocols that measured provider-delivered family planning care [26, 32–38].1 For each domain of the framework, we identified measurable indicators given the study design, used the questions from existing tools where they existed and developed new questions if needed (Table 1). For instance, for the poor client-center care domain, one indicator we measured was the proportion of visits where the provider did not ask the client preferred method. For privacy, we measured the proportion of visits where there was not auditory or visual privacy. We developed the non-dignified care questions based on reported provider behaviors from a qualitative study on adolescent perceptions of family planning in Malawi [39]. Questions used to develop indicators of disrespectful care • Were you prescribed, given or referred for a method during this visit? • If no, referred to field notes for reason. • Did the provider talk to you about your family planning methods in a group or by yourself? • If individual counseling, was the consultation conducted in an area where no one could see you and provider? • If individual counseling, was the consultation conducted in an area where no one could hear your conversation? • Did the provider talk to you (or the group) about any contraceptive methods? • If yes, what contraceptive method(s) did the provider talk with you about? • Did the provider talk to you (or the group) about any contraceptive methods? • If yes, do you feel the provider advocated a specific method for you during the consultation? • Did the provider talk to you (or the group) about any contraceptive methods? • If yes, did the provider ask you about your preference in contraceptive methods? • Did the provider interrupt you while you were speaking? • Did the provider interrupt the consultation to conduct other business? • Did provider raise their voice or yell at you? • Did provider use a disparaging term to describe you? • Did provider do anything else considered disrespectful or abusive? • If yes, what did the provider say or do? • Did the provider make any critical or judgmental comments about: • The number of children you have? Or do not have? • Your plans for whether you want to have more children and when? • Your partner/marital status? • The involvement of your partner in your family planning? • Your sexual activity? • The involvement of your parents? • Your age in regards to accessing family planning? • Your preferred method of contraceptives? • Your physical appearance? Judgement comments by staff • At any point, did you feel unwelcome by other health facility staff? • At any point, did other health facility staff make disrespectful or judgmental remarks to you or about you to others where you could overhear? • If yes, what did the staff say or do? • Did provider ask for additional money (informal payment)? • Did any other health facility staff ask you for additional money (informal payment)? We then created a quantitative tool with pre-coded responses and translated it into Chichewa and Tumbuka, two languages spoken in Malawi. During the early stages of data collection, it was apparent the details of the SC encounters with the health system could not be adequately captured in a pre-coded, quantitative tool. To capture this, we added an ad hoc qualitative field notes tool or a notebook for the SCs to describe their encounters at the facility and with the providers. The case scenarios were adapted from Malawi-specific family planning training materials, pretested with non-study clinicians in Malawi, and reviewed by a SC training consultant working with a Malawian organization for clinical and cultural accuracy (Additional file 1) [40–42]. One scenario was a married, adult, “method-switcher” who wants to change from hormonal injectable contraceptives to hormonal contraceptive pills. The other was an adolescent, unmarried woman who has just become sexually active and is a “first-time user” of contraceptives. To elicit comparable care across the providers, the details of each scenario were standardized including medical history, parity, age, and method preference among other factors (Additional file 1). The study team hired twelve data collectors with previous survey experience for the simulated clients: six assigned to an adult case scenario and six assigned to an adolescent case scenario. The SCs participated in a two-week training for the main study, including a one-day training focused entirely on client simulation, and a pilot to practice their scenario at a non-study facility. The survey coordination team assigned six data collectors to the adolescent SC based on their resemblance to younger woman; all data collectors were over 18 years of age. For the medical safety of the data collectors, the SCs could only accept pills or condoms and were trained to deploy standardized exit strategies to avoid injectables, implants, IUDs, or other invasive procedures (e.g. needle stick or cervical exam). To avoid invasive procedures, the SCs were trained to tell the provider they would come back at a later date for the procedure and ask for pills or condoms to use in the meantime. We selected hormonal contraceptive pills as the method preference since the counseling is more complex compared to condoms and measuring counseling quality was one of the aims of the overall study. SCs were trained to provide the information from their assigned scenario when asked by the provider to mimic an actual consultation. Data collection was carried out from January through March 2018. There were six teams, one team assigned to a district and two SCs (adult and adolescent SC) per team. SCs visited the first facility in each district approximately 1 to 2 days after all the listed providers in that district were consented, and the last SC visit occurred 61 days later. Each facility was visited once by two SCs. Both the “Adult” and “Adolescent” SCs arrived at a facility on the same morning, traveling separately. All SCs presented as clients accessing the facility for the first time. When the consultation was complete, the SCs returned to the field vehicle parked out sight of the facility and were immediately interviewed by their supervisor in the vehicle (unobserved by community members) using the quantitative tool. The supervisor then returned to interview the providers who consulted the SCs with a standardized instrument that solicited their background characteristics, education, and information about their position. Later that evening, the SCs recorded their encounter in the field notes. All providers gave their consent to participate in the study. One team inadvertently deleted a SC form from the data collection device before transfer to the server. In this study, we included data from 111 facilities (> 99% response rate) or 222 SC consultations (111 adult and 111 adolescent consultations). We created a listing of public sector facilities that offer family planning services in the six districts by updating a 2017 census of family planning facilities through discussions with the District Health Office [28]. Study teams contacted each of the facility administrators to create a listing of all family planning providers working at that facility. From these listings, all facility-based providers in the six districts were called prior to data collection for a verbal consent using a standardized form. The consent form stated the providers may be visited by masked, simulated clients sometime in the next 3 months. The provider offering family planning services the day the SCs visited the facility were included in the study. If two providers worked as a team the more senior provider was enrolled. During the provider interviews conducted after the SC consultations, the supervisors confirmed whether the provider had given consent to participation in the study. If not, the study team supervisor read the consent form to the provider and they were given the opportunity to be removed from the study and their data deleted. Reasons for not initially being consented by mobile phone include poor network connectivity, new hire or transfer from another facility, or the provider was missed when the facility in-charge listed the family planning providers. We entered the checklist responses into Open Data Kit on Android tablets and used R and Stata 14.2 software for analysis [43–45]. .We reported the prevalence of disrespectful actions and associated 95% confidence intervals for the six districts aggregated, assuming a binomial distribution with no survey design effect. We conducted a stratified analysis to determine whether the indicators differed by district, case scenario or phase of data collection. Since many of the indicators were subjective, we aimed to test whether SC reporting changed from the first half (< 30 days) versus the second half (> 30 days) of data collection. To compare the levels of care for these stratified analyses, we reported the proportions and compared 95% confidence intervals among the groups. We digitized the field notes from the SCs for analysis in Microsoft Excel and they were coded and organized by theme using a framework analysis as described by Ritchie and Lewis [46, 47]. If we found an event that fit within the four domains of disrespectful care, it was coded with a binary score and triangulated with the quantitative data. We selected excerpts from the translated field notes to provide descriptive context to the quantitative data in each domain.