i Strengthening Family Planning Project تعزيز تنظيم األسرة Lessons from a Contraceptive Health Insurance Coverage Pilot Submitted to: Dr. Nagham Abu Shaqra Agreement Officer’s Representative Population & Family Health Section USAID/Jordan Submitted by: Timothy Irgens Chief of Party Prepared by: Josef Tayag August 2015 Strengthening Health Outcomes through the Private Sector (SHOPS) Associate Cooperative Agreement No. 278-A-00-10-00434-00 The information contained in this document is considered CONFIDENTIAL and is intended for the recipient and their authorized representatives only. Any unauthorized distribution is strictly prohibited without the prior written consent of submitter. i TABLE OF CONTENTS Table of Contents.........................................................................................................................i List of Tables and Figures...........................................................................................................ii Introduction................................................................................................................................ 1 Overview of the Contraceptive Coverage Pilot ........................................................................... 3 Assessment Methodology .......................................................................................................... 5 Beneficiary Surveys................................................................................................................ 5 Cost Analysis.......................................................................................................................... 5 Process Evaluation................................................................................................................. 6 Data Limitations...................................................................................................................... 6 Characteristics of the Beneficiaries ............................................................................................ 8 Pilot Implementation .................................................................................................................14 Challenges Related to Disseminating Information .................................................................14 Challenges Related to Measuring Unmet Demand ................................................................16 Conclusions and Lessons Learned ...........................................................................................20 Bibliography ..............................................................................................................................22 ii LIST OF TABLES AND FIGURES Table 1. Characteristics of respondents to Ta’ziz beneficiary survey.......................................... 8 Table 2. Summary comparison of cost and utilization statistics of maternity claims...................16 Table 3. Cost and frequency of modern family planning claims by type reported to Ta’ziz from RSS beneficiaries .....................................................................................................................18 Figure 1. Hypothesized result chain for the contraceptive coverage pilot ................................... 4 Figure 2. Reasons for not using modern family planning............................................................ 9 Figure 3. Percent of MWRA using modern family planning against the estimated 57 percent required modern contraceptive use condition for break-even ....................................................10 Figure 4. Among respondents using a family planning, distribution of the method used............11 Figure 5. Among users of family planning, the type of provider used.........................................12 Figure 6. Among respondents reporting willingness to use contraceptive coverage, distribution of method type ..........................................................................................................................13 Figure 7. Chain of communications to reach MWRA beneficiaries for the contraceptive coverage pilot...........................................................................................................................................15 1 INTRODUCTION Research studies show that ensuring a population’s access to modern family planning leads to improved population health outcomes. Between 1990 and 2010, access to modern family planning helped cut worldwide maternal deaths in half (Osotimehin, 2012). Access to modern family planning is estimated to prevent as many as one in every 11 child deaths (Osotimehin, 2012). Globally, researchers have estimated that addressing unmet need for family planning would reduce unwanted pregnancies by two-thirds, resulting in 21 million fewer unplanned births and 26 million fewer induced abortions (Singh & Darroch, 2012). Studies also show that access to private health insurance positively influences the use of family planning in a many developing countries (Comfort, Peterson, & Hatt, 2013). It is estimated that every one dollar (JD 0.71) spent on family planning saves USD 1.40 (JD 0.99) in maternal and newborn health care costs (Singh & Darroch, 2012). In theory, because health insurance providers carry the monetary risk of financing health care for their beneficiaries, such savings may also benefit private health insurance funds. In Jordan, the contraceptive prevalence rate (CPR) has been relatively stagnant over the past 10 years (Department of Statistics, MOH of Jordan and ICF International, 2013). Currently, an estimated 42 percent of married women of reproductive age (MWRA) are using modern family planning methods, and 19 percent are using traditional methods (Department of Statistics, MOH of Jordan and ICF International, 2013). 1 In addition, 12 percent of MWRA have an unmet need for family planning services (Department of Statistics, MOH of Jordan and ICF International, 2013). Jordan has become more reliant on the private sector as a source for contraceptives, with 56 percent of all contraceptives being procured from the private sector.2 Given the reliance upon the private sector, private sector strategies may help address stagnant CPR and unmet need for family planning in Jordan. However, working with the private sector requires a business case: evidence that demonstrates that implementing a strategy is financially sustainable or profitable for the company undertaking the strategy. In Jordan, an estimated 70 percent of the population is covered by health insurance (World Health Organization). Thus, mobilizing priority services such as modern family planning holds great potential for growing the CPR. Employers—health insurance companies’ main customers—have expressed their interest in expanding insurance coverage for their employees 1 The intrauterine device (IUD) is the most widely adopted modern method (21 percent), followed by the pill and male condom (8 percent each), female sterilization (2 percent), and long acting methods (LAM) and injectables (1 percent each). Withdrawal (14 percent) and rhythm (4 percent) are the most common traditional methods (Department of Statistics, MOH of Jordan and ICF International, 2013). In the past 22 years, there has been a decline in the use of long-term methods such as female sterilization and IUD and an increase in the use of short-term methods like condoms (Department of Statistics, MOH of Jordan and ICF International, 2013). 2 The majority of pills (52 percent), IUDs (62 percent), and male condoms (58 percent) are procured through the private sector (Department of Statistics, MOH of Jordan and ICF International, 2013). 2 to include modern family planning. In a study conducted by the Higher Population Council (HPC), 87 of the 100 private- and public-sector employers interviewed by HPC reported a willingness to pay up to a two percent additional premium3 to expand coverage to include modern family planning services (Tarawneh, Tarawneh, & Ayyoubeen, 2011). In the same study, HPC estimates that this would only require an increase of 1.3 to 1.6 percent in overall health insurance premium (Tarawneh, Tarawneh, & Ayyoubeen, 2011). This means that employers are willing to pay more than what would be required hypothetically to offer family planning benefits to their employees. Despite interest from customers, very few health insurance providers in Jordan offer family planning benefits. At the start of this study, only one health insurance company offered modern family planning benefits to the employees of the Embassy of the United States. In recent years, health insurance has been limited to low profitability attributed to intense competition and low barriers to market entry (Yaghmour, Masri, & Kamhieh, 2011). These factors created pressure for health insurance providers to keep the price of their premiums low while the costs of medication and medical services increase. The loss ratio for the health insurance industry stood at 91 percent in 2009, up from 83 percent in the previous year; this means that 91 percent of all revenues collected for health insurance were used for medical care, leaving only a small margin of profit for the industry (Yaghmour, Masri, & Kamhieh, 2011). This is evidence to support that the health insurance industry as a whole is in need of strategies to control its costs. Offering contraceptive coverage may help alleviate this pressure. The Strengthening Family Planning Project in Jordan (in Arabic, Ta’ziz Tanzim al-Usra, or Ta’ziz for short) implemented a pilot program to test the cost-benefit of offering coverage of modern family planning methods through private health insurance schemes in Jordan. Funded by the United States Agency for International Development (USAID), Ta’ziz seeks to expand the availability, quality, and use of modern family planning services and products through partnership with the private sector. Through this pilot, Ta’ziz sought to investigate whether offering coverage of modern family planning methods and services through private health insurance companies would increase contraceptive use among MWRA while at the same time decrease maternity-related costs related to unplanned pregnancies. This report describes the pilot program implemented by Ta’ziz and the methodology used to conduct the assessment. It also presents information on pilot beneficiaries, discusses the pilot implementation and derives lessons learned. 3 Premium refers to the money required to be paid to health insurance providers in exchange for coverage. In Jordan, premiums are usually paid by employers in whole or in part, with the remainder paid by the employee receiving coverage. 3 OVERVIEW OF THE CONTRACEPTIVE COVERAGE PILOT Ta’ziz estimated that health insurance providers could profit from covering modern family planning services if: 4  Overall utilization of modern family planning methods among beneficiaries reached at least 57 percent  Beneficiaries opted to use the most cost-effective methods (such as IUDs)5  Beneficiaries reduced their reliance on less cost-effective methods such as condoms that are high-cost to the insurer, but relatively available to beneficiaries outside of health facilities (Tayag, 2013). Thus, Ta’ziz piloted contraceptive coverage to assess if these conditions were feasible in a real￾world scenario and if these conditions would allow health insurance companies to, in fact, break-even financially (see Figure 1).6 4 Prior to the start of the pilot, Ta’ziz used cost data and utilization rates from HPC’s 2011 study on health insurance and family planning to estimate the factors that would lead to a health insurance company to break-even in costs. The model used data from the 2009 Jordan Population Family Health Survey (JPFHS) for the business case model: a general fertility rate of 127 births per 1,000; average modern family planning method effectiveness of 96 percent; 71 percent of births as normal deliveries, 17 percent of births as caesarian sections, and 12 percent of births as miscarriages or others; and 2.6 percent of MWRA as having been sterilized women (Department of Statistics [Jordan], 2010). Ta’ziz also assumed a MWRA population of 2,000 (as estimated by health insurers prior to the start of the pilot); total increase in modern family utilization of 16 percent; and a modern family planning utilization of 35 percent using IUDs, two percent using injectables, nine percent combined pill, one percent implants, three percent opting for female sterilization surgery, and seven percent using male and female condoms. See Annex 1 for the complete Excel-based model. 5 The cost-effectiveness of modern methods is different for each type of method. IUDs and implants are the most cost-effective methods; to prevent a single pregnancy it costs JD 8.60 annually using IUD and JD 10.10 using implants. Injectables are less effective at JD 29.7 annually; while female sterilization, birth control pills, and condoms are the least effective at JD 73.50, JD 87.40, and JD 96.10, respectively (Tarawneh, Tarawneh, & Ayyoubeen, 2011). 6 Result chains represent a program or intervention as a linear process with activities at the front and long-term outcomes at the end. Such a visual display helps to explain how intermediate results relate to longer term intended results and identify steps to be monitored or assessed by program implementers (Funnel & Rogers, 2011). 4 Figure 1. Hypothesized result chain for the contraceptive coverage pilot Ta’ziz approached eight health insurance companies and only two companies agreed to participate in the pilot. Mediterranean and Gulf Insurance Co. P.L.C. Jordan (MedGulf) and Al Nisr Al Arabi Co. Ltd. (Al Nisr) both signed memoranda of understanding (MOU). Ta’ziz committed to cover modern family planning services. The companies committed to grant Ta’ziz access to needed de-identified claims data and inform their network doctors of the coverage. Ta’ziz would deliver educational materials to help promote the contraceptive coverage. The pilot was designed such that health insurance companies would cover modern contraceptives for existing employer clients. The health insurance companies identified two current clients, the Royal Scientific Society (RSS) and Manaseer Group, to participate in the pilot. The employers committed to providing contact lists for MWRA beneficiaries. Also, the employers committed to allow Ta’ziz to distribute announcement cards and educational materials about family planning and conduct educational lectures for employees’ families to promote the coverage.7 At the start of the pilot, the insurers estimated that there would be more than 2,000 MWRA beneficiaries; however, after collecting the beneficiary lists, only 891 MWRAs (545 from Manaseer Group and 346 from RSS) were employees or wives of employees and were beneficiaries under the employees’ insurance. 7 Ta’ziz signed non-disclosure agreements with each of the stakeholders noted above. Ethical approval for this study was obtained from the Abt Associates Inc.’s Institutional Review Board (IRB). Ta’ziz facilitates the pilot design and undertakes business case analysis Ta’ziz delivers educational materials to raise awareness of contraceptive coverage Health insurance companies include family planning into their benefit packages Employees are aware of contraceptive coverage Utilization of modern family planning increases; unplanned pregnancies decrease; insurers save money Employers distribute information 5 ASSESSMENT METHODOLOGY This section describes the approach used to assess the pilot implementation. Ta’ziz conducted beneficiary surveys to estimate demand for and use of contraceptives among pilot beneficiaries, a cost analysis to estimate changes in maternity- and family planning-related costs to insurance companies, and key informant interviews to evaluate the pilot implementation process. BENEFICIARY SURVEYS Ta’ziz conducted a baseline beneficiary survey to identify patterns of contraceptive use prior to the coverage. The beneficiary survey was conducted via telephone with 891 MWRA beneficiaries. Ta’ziz used a structured questionnaire designed to collect information associated with use of modern family planning, provider preferences for family planning, and interest in modern family planning. The questionnaire also probed perceptions of whether insurance coverage would affect contraceptive method use, method choice, and provider choice. There were 444 total responses (279 from Manaseer Group and 165 from RSS) for an overall response rate of 50 percent. Nearly a year after the start of the coverage, Ta’ziz conducted a follow-up telephone survey. The follow-up survey consisted of a structured questionnaire designed to collect information on awareness of the coverage, use of the benefit, reasons for not using the benefit, and interest in learning more about the benefit. A total of 515 MRWA beneficiaries were contacted, but only 283 MWRA participated in the survey for a response rate of 55 percent.8 All interviewees were asked to provide verbal consent before participating in the study and were free to decline to participate. COST ANALYSIS At the start of the pilot, Ta’ziz analyzed historical claims data related to maternity care (consisting of pregnancy, maternal, or post-natal care) from health insurance companies’ pilot participants. Ta’ziz compared this data to the data on incidence of claims incurred under the contraceptive coverage pilot. Ta’ziz received de-identified historical data from maternity care claims made by Manaseer Group beneficiaries insured by MedGulf (July 2011 to May 2013) and from RSS beneficiaries insured by Al Nisr (January 2010 to August 2013). Ta’ziz analyzed the data for the number of MWRA who made claims, total annualized cost of claims, and per capita cost of claims. Furthermore, Ta’ziz conducted a more detailed analysis into the frequency and cost per completed term of pregnancy using this dataset (Annex 2 and 3). 8 At the start of the pilot, health insurance companies identified 891 MWRA beneficiaries. During the follow-up survey, health insurers identified only 515 beneficiaries. This decrease could be attributed to changes in the number of policyholders registered as well as the removal of inactive beneficiaries from insurers’ membership lists. 6 At the end of the pilot, Ta’ziz collected claims data related to maternity care and family planning from beneficiaries of both the Manaseer Group and RSS for a period of 15 months and 18 months, respectively (Manaseer Group’s coverage began three months after RSS’ coverage started). This data was analyzed for the frequency of unique MWRA making claims, total annualized cost of claims, and per capita cost of claims; in addition, Ta’ziz estimated the cost and frequency of family planning services utilized under the pilot for RSS claims. Ta’ziz was unable to analyze Manaseer Group claims for family planning because MedGulf, the insurance company covering the employer, did not identify family planning services claims as originally agreed at the start of the pilot. PROCESS EVALUATION Ta’ziz conducted a series of semi-structured key informant interviews to evaluate the implementation process and better understand reasons behind the use or non-use of the contraceptive coverage. The evaluation used the Consolidated Framework for Implementation Research (CFIR) (Damschroder, Aron, Keith, Kirsh, Alexander, & Lowery, 2009). CFIR is composed of five domains of implementation: 1. Characteristics of the intervention 2. External influences such as policies or programs that are outside of the intervention that may affect implementation 3. The culture surrounding the implementation of the intervention 4. Characteristics of the individuals implementing the pilot 5. The process of implementation. Ta’ziz interviewed two pilot program managers, six human resources managers from Manaseer Group, one senior executive from Al Nisr, and two medical managers from MedGulf. Though RSS officially agreed to participate in the process evaluation, it provided no representatives to be interviewed. All interviewees were asked to provide verbal consent before participating in the study and were free to decline to participate. The interview guides were designed to gather information pertaining to each of the five domains of implementation outlined in the CFIR. The information obtained was contextualized with gray literature, the beneficiary surveys, and the claims data provided by the health insurance companies. DATA LIMITATIONS There are many limitations associated with the data available for the pilot assessment. Firstly, delays from partners shortened the timeline to observe changes in pregnancy patterns. Originally the pilot coverage was supposed to run for more than 24 months; however, coverage was only available to Manaseer Group for 15 months and to RSS for 18 months. Secondly, when signing the MOUs with Ta’ziz, insurers estimated the number of MWRAs as beneficiaries to be greater than 2,000, yet the pilot started with only 891 MWRAs. Claims data were for a relatively small number of full pregnancies observable within the claims data (113 pregnancies observed with Manaseer Group and 191 pregnancies observed RSS beneficiaries) and longer 7 term trends were not available. Further, Ta’ziz was unable to distinguish family planning claims for Manaseer Group which prevented a comprehensive analysis of cost. Due to turnover or unavailability, Ta’ziz was unable to interview six of the anticipated 16 key informant interviewees that were planned for the process evaluation. Ta’ziz, despite these issues, was able to derive important lessons that will inform the implementation of future similar programs in Jordan. 8 CHARACTERISTICS OF THE BENEFICIARIES As presented in Table 1, the mean age of MWRA was slightly higher for respondents from RSS at 34.2 years old as compared to respondents from Manaseer Group who were on average 31.2 years old. Respondents from RSS also had more children on average with 2.9 children as compared to 2.4 for Manaseer Group respondents. Among RSS respondents, more women reported that they were the primary policyholders for their health insurance policy (as opposed to being a dependent on the insurance policy). RSS respondents also tended to have more years of education compared to respondents of Manaseer Group. This may be related to the types of industries in which the companies are involved. RSS is a research and technical support service provider, whereas companies in Manaseer Group focus on labor-oriented industries such as manufacturing and infrastructure. Table 1. Characteristics of respondents to Ta’ziz baseline beneficiary survey Manaseer Group (n=279) RSS (n=165) Combined (n=444) Mean age 31.2 34.2 32.3 Average number of children 2.4 2.9 2.6 Percent of policyholders 13.6% 23.6% 17.3% Average number of years of education 12.9 14.3 13.4 Percent currently pregnant 12.5% 13.3% 12.8% Before the pilot began, 62 percent of Manaseer Group and 56 percent of RSS respondents reported not using a family planning method. The most common reasons for not using a family planning method included: being currently pregnant, being medically unfit or having health concerns, preferring a traditional method, planning to get pregnant, wanting rest from the method, and fear of side effects (Figure 2). Thirteen percent of Manaseer Group and nine perecent of RSS respondents reported preferring traditional methods. Seven percent of Manseer Group and 13 percent of RSS respondents reported fear of side effects for not using a family planning method. These proportions suggest that MWRAs who were not currently using modern contraceptive may have benefited from additional information about the advantages of modern family planning versus traditional methods and about concerns related to side-effects. Furthermore, a significant proportion of employees (79 percent of Manaseer Group and 81 percent of RSS respondents) expressed an interest in learning more about contraceptive benefits. 9 Figure 2. Reasons for not using modern family planning At baseline, neither company had the 57 percent modern contraceptive prevalence that was estimated as one of the conditions for insurance companies to break-even. Thirty percent of Manaseer Group and 44 percent of RSS respondents were using a modern family planning method at the time of the baseline beneficiary survey. This means that both Manaseer Group and RSS beneficiaries would have to increase their modern contraceptive use by 19 and 13 percentage points, respectively, to reach the estimated level at which contraceptive coverage would be profitable for the insurance companies (Figure 3). 20% 16% 13% 11% 10% 7% 10% 10% 1% 0% 1% 0% 0% 1% 23% 10% 9% 11% 9% 13% 6% 5% 3% 4% 2% 1% 1% 3% 0% 5% 10% 15% 20% 25% Currently pregnant Medically unfit/health concerns Prefers traditional method Planning to get pregnant Wanted rest from method Fear of side effects Postpartum Difficult to get pregnant Respondent opposed Spouse absent/no sex Husband opposed Costs too much Other Don't know RSS Manaseer Group 10 Figure 3. Percent of MWRA using modern family planning against the estimated 57 percent break-even condition The most popular modern methods used by respondents in both the Manaseer Group and RSS were IUDs (35 percent and 49 percent, respectively) and oral contraceptives (19 percent and 26 percent, respectively). Twenty-eight percent of Manaseer Group and 17 percent of RSS respondents preferred to use traditional methods (Figure 4). Hypothetically, if these users of traditional methods converted to modern methods, then the groups of beneficiaries would have surpassed the hypothesized 57 percent CPR condition estimated for breaking even, financially. Furthermore, as figure 3 presents, the percent of non-users who reported being interested in using a modern through the insurance also suggests a likelihood of attaining the 57 percent contraceptive prevalence. 38% 44% 33% 21% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Manaseer Group RSS Non-users interested in using a modern method with insurance Currently using modern family planning 57 percent break-even threshold 11 Figure 4. Distribution of methods among respondents using any form of family planning Among those currently using modern family planning, 86 percent of Manaseer Group and 82 percent of RSS respondents reported being satisfied with their current method of choice. Among those that reported being dissatisfied with their current method, the main reasons for dissatisfaction were side effects and inconvenience. Only seven percent of Manaseer Group respondents and no RSS respondents reported being dissatisfied with their method choice because of cost reasons. Among users of modern family planning, 30 percent of Manaseer Group respondents and 19 percent of RSS respondents prefer to switch their method if family planning was covered by health insurance. Of these, 45 percent of Manaseer Group respondents and 50 percent of RSS respondents would switch to IUDs; 16 percent of Manaseer Group and 14 percent of RSS respondents would switch to oral contraceptives; and six percent of Manaseer Group and seven percent of RSS respondents would switch to injectables. The most common source of modern family planning services was a private doctor, hospital, or clinic (29 percent among Manaseer Group and 43 percent among RSS respondents), followed by pharmacies (25 percent among Manaseer Group and 19 percent among RSS respondents). Government health centers and maternal child health (MCH) clinics were also widely chosen for modern family planning (34 percent among Manaseer Group and 24 percent of among RSS respondents) (Figure 5). 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Manaseer Group RSS Combined Injectable Implants Female sterilization Condoms Pills/Oral contraceptives IUD Traditional methods 12 Figure 5. Type of provider used among modern family planning users Thirty-three percent of Manaseer Group and 32 percent of RSS group respondents preferred to switch provider types if family planning was covered by health insurance. Among these, 97 percent of Manaseer Group and 74 percent RSS respondents preferred to switch to private providers. Among respondents who were not currently using modern family planning, 54 percent of Manaseer Group and 38 percent of RSS respondents reported that they would use a modern family planning method if family planning was covered by their insurance. Among these, 50 percent of Manaseer Group and 60 percent of RSS respondents preferred to use their coverage to finance their IUDs. Nineteen percent of Manaseer Group and 14 percent of RSS respondents preferred to use their coverage to finance oral contraceptives. Twenty-one percent of Manaseer Group and 17 percent of RSS respondents reported that they do not know what type of method on which they preferred to use their health insurance coverage (Figure 6). 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Manaseer Group RSS Combined Others Jordan Association for Family Planning and Protection Government clinic or hospital Pharmacy Private doctor, hospital, or clinic 13 Figure 6. Distribution of methods among respondents reporting willingness to use contraceptive coverage Prior to the start of coverage, fairly large proportions of respondents reported willingness to use their contraceptive coverage, with most wanting to use it for IUDs. These results supported Ta’ziz’s hypothesis that there may be unmet demand for modern family planning that may be met through health insurance coverage. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Manaseer Group RSS Combined Condoms Implants Injectable Pills/Oral contraceptives Don’t know IUD 14 PILOT IMPLEMENTATION Two conditions were needed to see an increase in modern family planning use among beneficiaries in the pilot program: (1) MWRA beneficiaries being aware of their modern family planning insurance coverage and (2) contraceptive coverage meeting an otherwise unmet demand for modern family planning. As this section describes, meeting these two conditions proved to be challenging. CHALLENGES RELATED TO DISSEMINATING INFORMATION Ta’ziz planned to conduct an interpersonal campaign incorporating posters, marketing collateral, educational lectures, and family-oriented events. However, companies eventually prevented Ta’ziz from conducting the interpersonal activities and limited Ta’ziz to distributing brochures and educational information through employer human resource representatives. Ta’ziz provided human resources representatives with brochures, a small gift, and cards welcoming members to the coverage. Human resource representatives reported that they sent news about the coverage through employees’ supervisors. Supervisors were then held responsible for disseminating the news to employees. However, according to Ta’ziz’s beneficiary survey, only 17 percent of MWRA respondents were policyholders; most (83 percent) of MWRA respondents were dependents to policyholders. Getting messages to MWRA beneficiaries about the contraceptive coverage required a lengthy chain of communication (Figure 7). News about the coverage had to travel from health insurance companies to employers’ human resources representatives, to employee supervisors, to employees, and finally to dependents if the MWRA was not the primary policyholder (only 17 percent of MWRA respondents were primary policyholders). Furthermore, health insurance companies needed to communicate to their provider networks to inform them that the coverage was now available to their patients. 15 Figure 7. Chain of communications for news on the contraceptive coverage to reach MWRA beneficiaries Unfortunately, Ta’ziz discovered that employers had not communicated with the MWRA beneficiaries directly. Nearly a year after Ta’ziz delivered the gift announcing the new benefits, Ta’ziz called beneficiaries and discovered that 89 percent of Manaseer Group and 87 percent of RSS respondents recalled receiving the gift.9 Two percent of Manaseer Group and 10 percent of RSS respondents reported being informed by someone about the benefit through an interpersonal conversation. One company, RSS, distributed new member identification cards that included a sticker on the cards announcing the coverage to ensure that policyholders were informed. However, key informant interviews revealed that this approach did not directly reach MWRA beneficiaries who were married to the main policyholder because only the main policyholder was responsible for picking up the new cards. Key informant interviews revealed that human resource representatives viewed health insurance coverage as essential to recruiting and retaining their employees. However, human resource representatives did not believe that the family planning benefit was important for retaining their employees. Human resources representatives from Manaseer companies 9 As noted in the limitations section, the response rate for this telephone survey was fairly low at 55 percent. Thus, these proportions are not fully generalizable to the whole population of beneficiaries. Furthermore, the beneficiaries were surveyed nearly a year after the gifts and announcements were distributed which likely influences how the respondents recalled the announcements. It is unclear if the small sample size and recall bias influenced responses positively or negatively. Health insurers Employer HR Reps Supervisors Employees/ Policyholders Married to policyholder 17 % 83 % Percent of MWRA beneficiaries in pilot Communicated to provider network to inform them of coverage 16 reported that most of their employees were male laborers and had health risks related to injury that would have been of greater financial concern to the company than family planning. Employers also reported that while they actively promoted the coverage of pregnancy-related services, family planning was not seen as a priority because lower proportions of their employees were MWRAs. Further, representatives from Manaseer Group and RSS reported feeling that family planning was a private matter; they felt the work place was not the proper venue to have family planning conversations. This likely explains why Ta’ziz was prevented from conducting educational lectures at workplaces and why such small proportions of MWRA reported having conversations related to their coverage. These communication challenges prevented potentially useful information from reaching MWRA beneficiaries. As of one year after the start of the benefit, five percent of Manaseer Group and 22 percent of RSS respondents reported using the modern family planning benefit. CHALLENGES RELATED TO MEASURING UNMET DEMAND Claims utilization and cost data were to reveal if a rise in the utilization of contraceptive coverage could be associated with a reduction in the number of maternity claims. However, Ta’ziz faced challenges related to obtaining a sufficient sample of claims structured in sufficient detail. Health insurance companies originally estimated a population of nearly 2,000 MWRA beneficiaries to participate in the pilot; however, only 891 were eventually included in the pilot due to the lack of interest from employers. Also, health insurance companies were not able to provide beneficiary and detailed encounter data (as originally prearranged in the MOUs) which limited Ta’ziz’s ability to draw conclusions from analyzing claims. Ta’ziz had originally planned to analyze claims frequency for a minimum of 24 months but encountered frequent delays in getting administrative approvals from health insurance companies and participating employers. Ta’ziz analyzed claims utilization and cost for a period of 15 months with employees of Manaseer Group and 18 months with employees of RSS. These delays limited the range of claims available for analysis. At the start of the pilot, Manaseer Group recorded 545 and RSS recorded 346 MWRA in their beneficiary lists (Table 2). The health insurance companies did not provide Ta’ziz with the total number of MWRAs at the end of the pilot despite several requests and follow-ups. These challenges prevented Ta’ziz from fully conducting the intended analysis. However, the information related to annualized maternity claims cost and the unit cost of providing family planning services may be useful for planning future executions of contraceptive coverage, which is why the information is presented here. Manaseer Group and RSS each had 112 MWRAs make maternity claims per year based on historical trends; thus, RSS had higher proportions of MWRAs making maternity claims compared to Manaseer Group. By the end of the pilot, 180 of Manaseer Group and 125 of RSS beneficiaries made maternity claims per year. The increase in maternity claims may be due to changes in the number of MWRA beneficiaries in both companies. However, Ta’ziz is unable to assess this without the total number MWRAs at the end of the pilot from both companies. Manaseer Group and RSS saw an increase in the total annualized cost of all maternity claims between the start of the pilot and the end of the pilot. Without the total number of MWRAs at the 17 end of the pilot, Ta’ziz is unable to determine if this increased cost of maternity claims is related to an increase in the number of MWRA beneficiaries or if MWRAs were utilizing their health insurance benefits more frequently. Table 2. Summary comparison of cost and utilization statistics of maternity claims Start of pilot End of pilot Number of MWRAs in membership lists provided by insurance companies Manaseer Group 545 N/A Royal Scientific Society 346 N/A Number of unique MWRAs making maternity claims per year Manaseer Group 112 180 Royal Scientific Society 112 125 Annualized cost of all maternity-related claims (in JD) Manaseer Group 57,109.00 64,736.96 Royal Scientific Society 46,875.75 50,684.78 Per capita annualized cost of all maternity-related claims (in JD) Manaseer Group 509.90 359.65 Royal Scientific Society 418.53 405.48 Both Manaseer Group and RSS saw a decrease in per capita annualized cost for maternity claims during the pilot. Annualized per capita cost of Manaseer Group claims decreased from JD 509.90 to JD 359.65, and RSS annualized per capita claims cost decreased from JD 418.53 to JD 405.48. Theoretically, access to modern family planning may reduce annualized per capita cost of maternity claims for a population because individuals who have a higher risk of birth complications (e.g., MWRAs who have had multiple births and older MWRA) would have access to family planning for limiting or spacing. However, without more detailed information about beneficiaries, claim encounters, and a larger sample, Ta’ziz is unable to comment if the pilot contributed to the observed decrease per capita cost of maternity claims. At the end of the pilot, Ta’ziz analyzed the family planning claims from RSS beneficiaries. Ta’ziz was not able to analyze utilization of the benefit among Manaseer Group beneficiaries because MedGulf, the insurer administering the benefit was not able to disaggregate family planning claims from its maternity claims as originally agreed upon and documented in the MOU. A total of 26 MWRA RSS beneficiaries used the benefit in the period between March 2014 and February 2015. This population generated a total of 58 individual claims. Regarding cost of claims, the cost of addressing complications such as abdominal pain, vaginal bleeding, or vertigo associated with a modern family planning service was on average JD 7.00. The average cost of contraceptive management was JD 14.14, JD 9.67 for a family planning consultation, JD 46.65 for an IUD insertion, and JD 5.31 for an oral contraceptive (Table 3). The small sample of 18 modern family planning claims available poses limitations for the applicability of this data in other settings.10 Table 3. Cost and frequency of modern family planning claims by type reported to Ta’ziz from RSS beneficiaries Type of family planning service Min. Avg. Max. Count Complication 2.57 7.00 10.00 6 Abdominal pain 5.17 6.01 6.86 2 Vaginal bleeding 2.57 5.29 8.00 2 Vaginitis or vulvovaginitis 10.00 10.00 10.00 1 Vertigo 9.38 9.38 9.38 1 Contraceptive management11 1.92 14.14 66.85 23 Family planning consultation 8.00 9.67 13.00 3 IUD Insertion 8.00 46.65 110.02 12 Family planning consultation 8.00 8.00 8.00 1 IUD insertion 10.73 43.17 69.00 7 Mirena 104.10 107.06 110.02 2 Nova T 16.20 17.78 19.35 2 Oral contraceptive 2.04 5.31 12.00 14 Belara 3.34 3.34 3.34 1 Cerazette 3.54 5.99 12.00 4 Marvelon 2.04 5.62 12.00 3 Yasmin 4.00 5.04 6.13 6 All family planning services 1.92 17.76 110.02 58 In sum, the pilot reveals that future executions of contraceptive coverage may encounter numerous challenges related to data management practices of health insurers. Health insurance companies in Jordan find it difficult to provide detailed claims encounter data, restructure coding systems to disaggregate medical services like family planning, and maintain updated beneficiary lists. These data elements would be required to fully assess the profitability of contraceptive coverage. 10 Presenting this data on claims may help to contextualize HPC’s calculations of annual cost per users based on service provider interviews. The claims analysis also informs the estimated averted cost of pregnancies that HPC used for its 2011 study. The HPC study found the expected cost per pregnancy to be JD 809, which assumed claiming for all pregnancy-related visits and events and used the cost ceilings reported by health insurance companies. However, Ta’ziz’s claims cost analysis revealed that the average amount paid by the insurance company in this pilot is about JD 653 for a Caesarian Section and JD 269 for a normal delivery, as opposed to the insurance ceiling JD 816 and JD 656, respectively, used in the HPC report. This may be because many women do not make claims at the end of a pregnancy event. They may be giving birth at a non-network facility or abroad or may choose to instead pay-out-of￾pocket for reasons that were outside the bounds of this study. 11 Without detailed encounter data, Ta’ziz is unable to determine what is included in the service category, “contraceptive management.” 19 Challenges related to the limited data have been compounded by the perspective shared among health insurance representatives that the pilot was low-priority. On one hand, health insurance companies saw the social value in modern family planning coverage. On the other hand, these same individuals were not convinced of the cost-benefit of offering family planning coverage and therefore held contraceptive coverage as lower priority compared to other forms of coverage that are in great demand from their clients (such as maternity benefits). As the Chief Executive Officer of Al Nisr stated, “[modern family planning] is just not in our best interest, financially.” 20 CONCLUSIONS AND LESSONS LEARNED At the start of the pilot, Ta’ziz hypothesized that there was an unmet need for modern family planning services among MWRA beneficiaries of private health insurance programs in light of DHS survey findings. Ta’ziz also hypothesized that these needs could be met through education about modern family planning as well as by covering the cost of contraception through health insurance. From the perspective of health insurance providers, contraceptive coverage may save insurers the cost of financing unwanted pregnancies. In Jordan, the execution of contraceptive coverage had never before been piloted to assess its cost-effectiveness for private health insurers. During its execution, the Ta’ziz contraceptive coverage pilot identified numerous challenges between the theoretical cost-benefit of offering coverage and the practical realities of ensuring informed choice to modern family planning. One lesson learned was that contraceptive costs were not the primary barrier to family planning use and that attitudinal or knowledge barriers may be more significant. Ta’ziz’s beneficiary surveys revealed that large proportions of MWRA beneficiaries were still interested in using contraceptive coverage and learning more about modern family planning methods. Common reasons for not using modern family planning included fear of side-effects as well as preference for traditional methods of contraception. These attitudinal reasons may be better addressed through education and information sharing. Employers and health insurance providers needed to prioritize contraceptive coverage in order to effectively use their communications channels to promote the benefits. Regarding communication, strategies for disseminating culturally-sensitive information about contraception have long been developed. The challenge lies in gaining buy-in throughout the chain of communications required to reach beneficiaries. Employers have the most influence in raising awareness of coverage and ensuring that beneficiaries are informed. However, cultural stigma and misunderstanding around modern contraception must be addressed. Future executions of contraceptive coverage could benefit from gaining buy-in from human resource officers representing employers. While Ta’ziz first sought buy-in from executive level management at the employer companies, getting buy-in from individual human resource managers about the financial value of contraceptive coverage is equally important. This would facilitate better communication, problem-solving, and identify alternative ways of reaching MWRA beneficiaries. Despite the challenges encountered in the pilot implementation, health insurance companies decided to continue providing contraceptive coverage beyond the life of the pilot. Representatives from health insurance companies explained that this was a social benefit they would like to continue promoting. This may incentivize other competing health insurance companies to offer similar coverage in the future. Future executions of contraceptive coverage could be strengthened by assessing the profitability of such strategies to reinforce the social 21 rationale. To measure profitability and better assess contraceptive coverage, future executions will need to build the capacity of health insurance providers to manage their claims and membership data. Data on active beneficiaries, detailed data on claims encounters, disaggregate coding for family planning services, as well as larger samples over longer time periods will strengthen the evidence for the profitability of contraceptive coverage. Future executions of contraceptive coverage should use the lessons documented in this study to inform communication plans, assessment timelines, sampling, and activities related to updating health insurer claims and membership management databases. With an estimated 70 percent of the population covered by health insurance, mobilizing priority services such as modern family planning holds great potential for growing the CPR (World Health Organization). Furthermore, executing strategies such as contraceptive coverage may assist the health insurance industry in controlling its overall costs. Realizing these benefits, however, hinges on overcoming the challenges documented in this study. 22 BIBLIOGRAPHY Comfort, A., Peterson, L., & Hatt, L. (2013). Effect of Health Insurance on the Use and Provision of Maternal Health Services and Maternal and Neonatal Health Outcomes: A Systematic Review. Journal of Health, Population, and Nutrition(31), 81-105. Damschroder, L. J., Aron, C. D., Keith, E. R., Kirsh, R. S., Alexander, A. J., & Lowery, C. J. (2009). Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implementation Science. Department of Statistics [Jordan]. (2010). Jordan Population and Family Health Survey 2009. Calverton, Maryland: Department of Statistics and ICF Macro. Department of Statistics, MOH of Jordan and ICF International. (2013). Jordan Population and Family Health Survey 2012. Calverton, MD: Department of Statistics and ICF International. Funnel, S., & Rogers, P. (2011). Purposeful Program Theory: Effective Use of Theories of Change and Logic Models . San Franciso: John Wiley & Sons. Osotimehin, B. (2012). Family Planning Saves Lives, Yet Investments Falter. Lancet, 82-83. Singh, S., & Darroch, J. (2012). Adding It Up: Costs and Benefits of Contraceptive Services￾Estimates for 2012. New York: Guttmacher Institute. Tarawneh, M., Tarawneh, B., & Ayyoubeen, Y. (2011). Feasibility of Family Planning Services Inclusion within Public and Private Employers Health Insurance Plans. Amman: Higher Population Council. Tayag, J. (2013, April). Jordan Modern Family Planning Coverage Insurance Business Case Scenarios. Amman, Amman, Jordan. World Health Organization. (n.d.). Jordanian experience: Keeping out-of-pocket spending under control (Jamal Abu Saif). Retrieved August 6, 2015, from http://www.emro.who.int/about￾who/universal-health-coverage/jordanian-experience-keeping-out-of-pocket-spending￾under-control.html 23 Annexes Annex 1: Health Insurance Business Case Model Annex 2: Manaseer Group Claims Data: Baseline Cost Summary Annex 3: Al Nasr Al Arabi Group Claims Data: Baseline Cost Summary