Syphilis control in pregnancy: decentralization of screening facilities to primary care level, a demonstration project in Nairobi, Kenya

1

ei

G:ECOLOGY

& OBSTETRICS

International Journal of Gynecology & Obstetrics 48 Suppl. (1995) S121-S128

Syphilis control in pregnancy: decentralization of screening facilities to primary care level, a demonstration project in Nairobi, Kenya F. Jenniskens*a"b,E. Obwakaa, S. Kirisuaha, S. Moses",', F. Mohamedali Yusufalie, J.O. Ndinya Acholaa, L. Fransen', M. Lagab, M. Tenimerman', b,` "University of Nairobi, Nairobi, Kenya binstitute of Tropical Medicine, Antwerp, Belgium

`University of Ghent, Brussels, Belgium 'University of Manitoba, Winnipeg, Canada `Nairobi City Council, Nairobi, Kenya 'European Commission, Health & AIDS Unit, Brussels, Belgium

Abstract A decentralized syphilis control program in pregnant women was implemented in nine Nairobi City Council for between July 1992 August 1993, were women screened syphilis, treated pregnant and whereby antenatal clinics before leaving the clinic if RPR seroreactive, and counselled on the importance of partner treatment and sexual abstinence during treatment in order to protect their unborn babies from getting congenital syphilis. A total of 13 131 500/ for 87.3% (RPR treated test), on site and of seroreactive women were syphilis screened pregnant women were of partners returned to the clinic and were treated. The prevalence of RPR reactivity was 6.5%. Based on other data the program could theoretically have prevented 413 cases of congenital syphilis at a cost of approximately 50 USD per prevented case. This demonstration project shows that decentralized prevention of congenital syphilis in antenatal in health is feasible inexpensive in by allocation reproductive and and resource nurses should receive priority clinics and child survival programs. Keywords: Syphilis; Congenital syphilis; Perinatal mortality

1. Introduction Maternal syphilis infection during pregnancy is a serious condition with a dramatic effect on least If 60% of at untreated, pregnancy outcome.

* Corresponding author, Institute of Tropical Medicine, Antwerp, Tel: + 32 3 2476530; Fax: + 32 3 2476333.

pregnant syphilitic women will experience adverse pregnancy outcomes. A recent population based 26% in Malawi that of stillbirths, showed study 11% of neonatal deaths, 5% of post neonatal deaths, and 8% of infant deaths were attributable to active maternal syphilis infection [I]. Moreover recent reports confirm the increased risk of abortion, stillbirth, prematurity and perinatal death in syphilis seroreactive women [2-5].

0020-7292f95!509.50 © 1995 International Federation of Gynecology and Obstetrics SSD10020-7292(95)02326-T

S122

F. Jenniskens et al. / International Journal of Gynecolo~:v & Obstetrics 48 Suppl. (1995) S121 $128

Several studies in Africa have reported high prevalence rates of syphilis among pregnant women as well as in the general population, ranging from 3 to 15% [1 11]. Most African countries advocate a policy of syphilis screening in pregnancy, but in reality only a few countries achieve implementation of these policies due to managerial, budgetary and logistic constraints. In Zambia, Hira et al. have demonstrated the success of a decentralized syphilis screening program in primary health care clinics, with a two-thirds reduction in syphilis associated adverse pregnancy outcome [9]. In Kenya, stable syphilis seroreactivity rates of approximately 3% have been reported in urban settings for many years [4-11]. Recent observations however, suggest a rise in syphilis seroreactivity among pregnant women [12]. In 1989, the Government of Kenya set up a National STD Control Program within the Ministry of Health, with syphilis control in pregnancy as one of its highest priorities. Prevention of congenital syphilis through screening and treating pregnant women has been a policy in Nairobi City Council primary health care clinics for many years. A centralized system was in place where women were bled at the first antenatal visit. The specimen were then transported to a central laboratory (Pumwani Maternity Hospital) to be tested with a RPR (Rapid Plasma Reagin) test and the results sent back to the clinic. The turn-around time for giving a woman her result, was between 2 and 4 weeks. Subsequently, women found to be reactive, were referred to the Nairobi Special Treatment Clinic, the main STD referral center in the city. In 1989, an evaluation of this system revealed low effectiveness, even with additional logistic support [12]. Not more than 60% of pregnant women were bled at the first visit, of these, 87% had their results registered on their antenatal clinic card, but only 9.1% of RPR reactive women received adequate treatment. In view of these observations, the health authorities adopted a plan for a pilot congenital syphilis control program, through RPR testing and treatment of seroreactive pregnant women at antenatal clinics by nurses. This paper will discuss the feasibility and operational aspects of this de-

centralized program among pregnant women in nine health centers in Nairobi. 2. Methods

A demonstration project was developed, targeted at prevention of congenital syphilis in Nairobi. In this paper congenital syphilis is defined as a broad term for all different adverse pregnancy outcomes related to a maternal syphilis infection, including abortion, prematurity, stillbirth, perinatal death and children born with or developing signs or symptoms of congenital syphilis. Nairobi and its peri-urban area have approximately 2.7 million inhabitants, of whom the larger part live in peri-urban slums. The population served by this project generally live in single rented rooms with their families. Dwellings are characterized by mud, wood or iron sheet construction. The Public Health Department of the Nairobi City Council (NCC) is the major health provider in the city of Nairobi, operating 54 of the 154 registered health units within the city. All 54 health units provide maternal and child health (MCH) and family planning (FP) services and 30 of them also provide curative services. The centers are distributed evenly in the peri-urban and urban areas and the catchment population per clinic is estimated at 80 000. The main strategy of the project was the implementation of a decentralized, clinic-based model for effective on-site diagnosis and treatment of maternal/partner syphilis. Components of this program included: (1) providing laboratory support; (2) providing supplies and drugs to the M C H clinics; (3) training of antenatal clinic nurses in performing a rapid screening test, in treating seroreactive women before leaving the clinic, and in counselling patients on safer sexual behaviors, stressing the importance of their sexual partners being treated; and (4) supervision and monitoring. Out of the 30 NCC health centers with both M C H / F P and curative components, nine were selected on the basis of geographical distribution and population served (largest lower income populations, situated in urban and peri-urban areas) to have screening programs set up.

F. Jenniskens et al. / International Journal of Gynecology & Obstetrics 48 Suppl. (1995) S121-$128

S123

Table 1 Indicators used to monitor the program I. 2. 3. 4. 5. 6. 7. 8. 9.

Number of new ANC attenders per month Number RPR tests done/Number of new ANC attenders Number of women RPR reactive/Number of RPR tests done Number of women RPR reactive treated correctly/Number of women RPR reactive Number of women given partner notification card/Number of women RPR reactive Number partners spontaneous return/Number of women RPR reactive Number partners assisted return/Number of women RPR reactive Number of partners treated/Number of women RPR reactive Number false negative and false positive tests at the MCH clinic as compared with the gold standard at the UON.

Abbreviations: RPR, rapid plasma reagin; UON, Universityof Nairobi, Department of Medical Microbiology:and ANC, antenatal clinic. Syphilis serology was performed by antenatal clinic nurses, using the RPR card test (RPR, Wellcome, London, UK). The project did not opt for a micro-hemagglutination essay for antibodies to Treponema Pallidum at clinic level for the sake of simplicity and cost reduction. Quality control (Macro-value, R P R card test, Becton Dickinson, Cockeysville, USA) was performed by a reference laboratory at the University of Nairobi (UON). Each M C H clinic stored all serum samples collected during 1 day each week for retesting at the U O N laboratory, and the results were discussed with the clinic staff on a monthly basis. R P R seroreactive women were treated with a single dose of 2.4 million units of intramuscular benzathine penicillin. To facilitate partner referral a standard form for partner notification was developed wherein the partner was invited to come to the clinic for reasons related to the pregnancy. Each seroreactive woman was asked to bring her partner(s) to the clinic for treatment. The program was initiated in June 1992 and the counselling component began in October 1992 in close collaboration with the clinic staff\ Formative research, using focus group discussions and in depth interviews, was carried out among pregnant women with syphilis, their partners and nurses at antenatal clinics. The results of the formative research will be reported elsewhere. A manual was developed for training antenatal clinic nurses on the public health importance of maternal syphilis, screening and diagnosis of syphilis in pregnancy, antenatal syphilis control

programs: components of patient care, R P R testing techniques, promoting behavior change and counselling women with syphilis and their partners using syphilis counselling materials. Between June 1992 and January 1993, 69 antenatal clinic nursing staff in the nine selected health centers were trained in the principles and practice of syphilis control in pregnancy. At the same time screening units were set up in the health centers' M C H clinics and provided with minimal laboratory equipment and supplies, drugs and counselling facilities. Laboratory equipment consisted of a centrifuge and a RPR card shaker. Refresher workshops were conducted in February 1993. During these workshops the acceptance of the program by the clinic staff was discussed and the counselling session of the training program was tested. The project team supervised the clinics on a monthly bas'is. A standardized form was used, which stated the objective of the visit, the topics discussed and a checklist to evaluate the laboratory, stock management, clinic staff on duty, record keeping, quality control and partner tracing activities. The cost-effectiveness of the decentralized approach was estimated using existing estimates of adverse pregnancy outcomes related to maternal syphilis. 2. 1. Process indicators used to m o n i t o r the program

Table 1 illustrates the process indicators which are used to monitor and evaluate the implementation of the program.

S124

F. Jenniskens et at. /International Journal of Gynecology & Obstetrics 48 Suppl. (1995) S 1 2 1 - S 1 2 8

percentages 10 9.0

8.8

8 7.2 S.0

5./;5.61

6 4,8

4

2

o 1

2

3

4

6

7

8

9

Clinic Name

Fig, 1. Prevalence of RPR reactivity in the nine intervention clinics (July 1992 August 1993): I = Ngong Road; 2 = Baba dogo; 3 = Dandora; 4 = N g a r a ; 5 - Mathare North: 6 = Langata; 7 = Ritura; 8 = Kangemi; 9 = Kariobangi.

3. Results

3.1. Screening and treatment Between July 1992 and August 1993, a total of 13 131 pregnant women were tested for syphilis. Overall, 860 (6.5%) were seroreactive. Fig. 1 shows the prevalence of R P R seroreactivity

in the nine clinics, which ranged from 2.7% in Ngara to 9.0% in Babadogo. N g a r a is a middle income area in the center of Nairobi, whereas Babadogo is on the edge of an industrial area where commercial sex is common. The other clinics are all situated near slum areas. Table 2 shows the principle results of the intervention during the study period. Virtually 100% of new antenatal clinic attenders were screened for syphilis on their first clinic visit. The mean gestational age at first visit varied from 23.3 to 28.1 weeks. Overall, 87.3% of seroreactive pregnant women were treated at their first visit according to the guidelines, ranging from 74.6% in Kangemi to 100% in Ngara. Reasons for non-treatment were refusal of some clinics to treat a w o m a n before bringing her partner to the clinic. Moreover some women did not have time to wait for their results.

3.2. Partner notification and treatment After the introduction of the partner notification form, an average of 86% of all seroreactive pregnant women were counselled by the clinic nurses about the importance of their partners being treated and were given a partner notification form. It is unknown how m a n y of the women actually notified their partners, but 428 (49.8%) of partners were treated at the same clinic as the index case. It is possible that some partners

Table 2 Proportion of women screened, women and partners treated and the gestational age at booking Clinic name

Kariobangi Kangemi Riruta Langata Mathare North Ngara Dandora Baba Dogo Ngong Road Total

Women screened (%) 2648 1063 2218 2166 904 442 1587 1080 1023

Mean GA at first visit (weeks)

% 1:8 confirmed by a Micro-hemagglutination assay for antibodies to Treponema Pallidum [1]. In our setting confirmatory tests were not performed for operational reasons. A study, currently underway, to assess the impact of the decentralized program on adverse pregnancy outcomes, will further address the issue of low RPR titres and their associated risk of adverse pregnancy outcomes. Only 1.1% of the tests resulted in excess treatment of pregnant women. The consequences of over-treatment with penicillin are minimal, but the implications for violence against women incorrectly diagnosed as having syphilis following partner notification may be serious. The syphilis prevalence among pregnant women in Nairobi is still on the rise. Temmerman et al. reported an RPR seroreactivity rate of 5.3% in 1991 [12], while we found an overall RPR seroreactivity of 6.5'70 in 1993. This trend shows that it is still important to invest in syphilis control programs in pregnancy, particularly in urban centers. Several recent reports have described costeffectiveness analyses of antenatal syphilis screening programs even in low seroprevalence settings [16-19]. The cost per averted case of congenital syphilis (fetal loss, perinatal death, congenitally infected infant) was estimated at 10 USD and 70 USD for syphilis seroprevalence

S127

levels of 15% and 1%, respectively. The cost of detecting one case of syphilis in the former centralized screening system in Nairobi was 66 USD, and the cost of preventing one case of congenital syphilis was at least 730 USD due to the considerable delay in communicating the results, problems in follow-up and low treatment compliance [18]. Decentralizing diagnostic and treatment services and integrating them at the primary health care level reduced the cost per averted case more than 10-fold. Our calculation of costs was higher than the estimates by other authors [20], as we included the costs of personnel, supervision and quality control. This project has addressed the operational aspects of the implementation of prevention and control activities of congenital syphilis, and did not include a follow-up phase to measure directly the effect of the program on preventing adverse pregnancy outcomes. More accurate data on the cost-effectiveness of the program can be calculated once its effect on adverse pregnancy outcome has been examined. Further research is needed on the effectiveness of the counselling program in reducing re-infection rates, and an educational campaign needs to be designed in order to promote earlier antenatal clinic attendance. Our experience clearly demonstrates that syphilis screening and control by nurses at antenatal clinic level is feasible. The proportion of women adequately treated was increased almost 10-fold and half of the partners were treated. Moreover, the decentralized program was clearly less expensive than the former centralized program. Similar programs should receive priority in resource allocation to reproductive and child health programs.

Acknowledgements This work was supported by grants from MotherCare/John Snow, Inc. (Arlington, USA); the Commission of the European Communities, Directorate General VIII for Development (Brussels, Belgium) and by the subprogram Science and Technology for Development, Commission of the European Communities, Directorate General XII for Research (Brussels, Belgium).

S128

F. Jenniskens et al. / International Journal o f Gynecology & Obstetrics 48 Suppl. (1995) $121 $128

References [l] McDermott, Steketee R, Larsen S, Wirima J. Syphilis-associated perinatal and infant mortality in rural Malawi. Bull World Health Organ 1993; 71(6): 773 780. [2] Watts TE, Larsen SA, Brown ST. A case-control study of stillbirths at a teaching hospital in Zambia, 1979-1980: serological investigations for selected infectious agents. Bull World Health Organ 1984; 62:803 808. [3] Larsson Y, Larsson V. Congenital syphilis in Addis Ababa. Ethiop Med J 1970; 8:163 165. [4] Temmerman M, Lopita M, Sinei S, Plummer FA, Nagelkerke N, Piot P. Sexually transmitted infections as risk factors for spontaneous abortion. Int J STD AIDS 1992; 3: 418-422. [5] Temmerman M, Plummer FA, Mirza NB et al. Infection with Human Immunodeficiency virus (HIV) as a risk factor for adverse obstetrical outcome. AIDS 1990; 4: 1087-1093. [6] De Clercq A. Problemes en obstetrique et gynecologic. In: Meheus A, Butera S e t al. editors. Sante et maladies au Rwanda. Bruxelles: Administration Generale de la Cooperation au Developpement 1982; 627-656. [7] Vuylsteke B, Bastos R, Barreto R et al. High prevalence of sexually transmitted diseases in a rural area in Mozambique. Genitourin Med 1993; 69:427 430. [8] Friedman PS, Wright DJM. Observations on syphilis in Addis Ababa. 2. Prevalence and natural history. Br J Vener Dis 1977; 53:276 280. [9] Hira SK, Bhat G J, Chikamata DM et al. Syphilis intervention in pregnancy: Zambian demonstration project. Genitourin Med 1990; 66:159 164. [10] Liljestrant J, Bergstrom S, Nieuwenhuis F, Hederstedt B. Syphilis in pregnant women in Mozambique. Genit Med 1985; 61: 355-358. [11] Mati JKG, Aggarwal VP, Lucas S, Sanghvi HCG, Corkhill R. The Nairobi birth survey, I the study design,

[12]

[13]

[14]

[15]

[16]

[17] [18]

[19] [20]

the population and outline results. J Obstet Gynaecol East Centr Afr 1981; 1:132 139. Temmerman M, Mohamed Ali F, Ndinya-Achola JO, Moses S, Plummer FA, Piot P. Rapid spread of both HIV-1 and syphilis among antenatal women in Nairobi, Kenya. AIDS 1992; 6: 1181. Wasserheit JN, Holmes KK. Reproductive tract infections: challenges for international health policy, programs and research. In: Germaine A e t al. editors. Reproductive Tract Infections, Global Impact and Priorities for Womens' Reproductive Health. New York: Plenum Press, 1992: 16. Pido D, Jenniskens F. MotherCare maternal/congenital syphilis control project qualitative research report. MotherCare Working Paper No. 18. Arlington: MotherCare/ John Snow, Inc., 1993. van Dyck E, van de Velden L, Ndoye I e t al. Evaluation of the rapid plasma reagin "teardrop" card test for screening of syphilis in field conditions. Sex Transm Dis 1993; 20: 194-197. Stray-Pedersen B. Economic evaluation of maternal screening to prevent congenital syphilis. Sex Transm Dis 1983; 10: 167-172. Garland S, Kelly N. Is antenatal screening for syphilis worthwile? Med J Austral 1989; 151: 368-372. Temmerman M, Mohamed Ali F, Fransen L. Syphilis prevention in pregnancy: an opportunity to improve reproductive and child health in Kenya. Health Policy Plann 1993; 8(2): 122 127. Hira SK. Guidelines for prevention of adverse outcomes of pregnancy due to syphilis. WHO/VDT/91.455. Schulz KF, Schulte JM, Berman SM. Maternal health and child survival: opportunities to protect both women and children from the adverse consequences of reproductive tract infections. In: Germaine A e t al., editors. Reproductive Tract Infections: Global Priorities for Womens' Reproductive Health. New York: Plenum Press, 1992: 148.