Early discontinuation of intrauterine device in Nepal – a retrospective study

WHO South-East Asia Journal of Public Health 2012;1(3):309-319 309 Original research Early discontinuation of intrauterine device in Nepal – a retrospective study Subash Thapaa Background: The Intrauterine Device (IUD) is the least popular method of contraception for family planning in Nepal. In addition, discontinuation in the early days after insertion is very high and information related to factors affecting early discontinuation is very limited. Methods: A retrospective study was carried out over a 5-year period to determine the associated factors for early discontinuation of the IUD. Insertion, consultation and removal records of IUD users of Kshetrapati Family Welfare Centre, Kathmandu, were reviewed. Removal within 12 months of insertion is defined as early discontinuation. The study participants were selected randomly. Uni-variate and logistic regression analysis were used to analyse predictors of early discontinuation of IUD. Results: The results showed that one fifth of the 230 women using IUD discontinued within the first year of insertion. Side-effects were the main reason for early discontinuation followed by expulsion. Woman’s occupational status, husband’s occupational status, husband’s educational status, experience of side-effects and follow-up practice were associated with early discontinuation of IUD. Menstrual disorder and abnormal vaginal discharge were also associated with early discontinuation. Conclusion and recommendation: Risks of IUD discontinuation were found to be very high during the early days of insertion and side-effects were identified as the major cause. So, proper management of side-effects would be very important to reduce early discontinuation of IUD. Key Words: Intrauterine device, discontinuation, follow-up, side-effect, reproductive intention. a Department of Public Health, Nepal Institute of Health Sciences, Kathmandu, Nepal. Correspondence to Subash Thapa (e-mail: soobesh@gmail.com) Introduction The intrauterine device (IUD) is the least popular method of contraception for family planning in Nepal. Only 0.4% of women use the IUD and among them, about 13% remove it within the first year and 20% remove it within the second year.1 This behaviour is mainly governed by women’s reproductive intentions, reproductive status, the combined impact of age, duration of marriage and number of surviving children.2,3 L ike l ihood o f d i scont inuat ion o f contraception is higher among women of low parity and those who had not achieved their desired family size at the start of the segment of use.4–6 The family size preference is apparently Book 1.indb 309 12-Dec-2012 9:02:29 AM WHO South-East Asia Journal of Public Health 2012;1(3):309-319310 Early discontinuation of intrauterine contraceptive device in Nepal – a retrospective study Subash Thapa more important than preference about sex of children in determining contraception use.7,8 However, women who want to space births are twice as likely to discontinue using the IUD within the first year as compared with those who want to limit childbirth.9 Studies had noted that women of lower age, along with low education and those who have tried other contraceptive methods are more likely to discontinue IUD, although education does not always emerge as a key determinant.3,10,11 A study in Southern Jordan reported that planned pregnancy was the main reason for discontinuation of all contraceptive devices.8 Especially for IUDs, side-effects and health concerns also play an important role in discontinuation.7,12 Irregular bleeding, lower abdominal pain, and vaginal discharge appear to be other key reasons for discontinuation.10 In the context of Nepal, side-effects are the most frequently cited reason for stopping use. Only a few discontinue use because they want more children.1,13 Another important point is that 40% of IUD users do not practice follow- up visits even when they were told by the health workers. The main reason is that many women experience no pain or problems with the method. Similarly, the women do not feel any need for the follow-up and it takes a long time to reach to the health facility.1 Low use and high discontinuation is a public health concern. The study of discontinuation and its causes clearly remains important for programme planning and guidance. In Nepal, very few studies on IUD have been documented; and studies that have been conducted lack appropriate design and power of analysis. In the current study, an attempt has been made to determine the factors associated with the early discontinuation of IUD. The study, at first, examined the socio- demographic characteristics of IUD users. Then, the study identified the reasons for early discontinuation of IUD and examined the association between those factors. Material and methods The study was a retrospective one in which women inserting IUD from January to December, 2004 were followed over a period of five years (2004–2009), based on the data of the IUD users of the Kshetrapati Family Welfare Centre (KFWC) in Kathmandu. During the period, some women were found to discontinue using IUD whereas some were continuing the use. Within a period of one year, 540 women opted for IUD (TCu 380A) at the family welfare centre and simple random sampling was done to cover the required sample size of 230 women. The sample size was estimated by means of power calculation based on the discontinuation rate of 13%1 in the first year with a level of significance of 0.05 and power of 95%.14,15 In KFWC, women going in for IUD were at first registered in an insertion record, then their follow-up visits and complaints were recorded in a consultation record and finally, the women discontinuing use were recorded in a removal record. These three different records were reviewed for the study purpose. Consent for data review was taken from the board of directors of KFWC, Kathmandu. The institutional ethical review board of B. P. Koirala Institute of Health Sciences (BPKIHS), Dharan, Nepal approved the study protocol and consent process. Variables studied were age during insertion, place of residence, educational status, occupational status, husband’s educational status, husband’s occupational status, total number of children, sex of last child, reproductive intention, duration of use of IUD, side-effects and follow-up practice. During routine clinic hours, information from the IUD users regarding side-effects such as Book 1.indb 310 12-Dec-2012 9:02:29 AM WHO South-East Asia Journal of Public Health 2012;1(3):309-319 311 Early discontinuation of intrauterine contraceptive device in Nepal – a retrospective study Subash Thapa pain in abdomen, and excessive menstrual bleeding were all taken through interview by the IUD trained nurses in the family planning centre and recorded in the appropriate record. Similarly, early discontinuation of IUD has been defined as removing IUD within the first 12 months after insertion and the women in this category were known as “early discontinuers”.1,2,10, Meanwhile, women removing the IUD after 12 months as well as women who were still continuing use during the study period were known as “women who did not discontinue early”. All analyses were carried out using PASW statistic version 18.0. Univariate statistics were used to test association and to calculate the crude odds ratio and confidence intervals. A logistic regression analysis was then carried out to identify the selected variables that were the best independent predictors of early discontinuation of IUD, after adjusting for the husband’s education status, husband’s occupation status and women’s occupation status. Husband’s education status, husband’s occupation status and women’s occupation status were entered in the model at the first step and then, selected variables were entered into the model. Similarly, Kaplan–Meier analysis was used to identify the relationship between side-effects and continuation of IUD. Results Characteristics of the IUD users and reasons for early discontinuation In order to determine the associated factors for early discontinuation of IUD, insertion, consultation and removal records of 230 IUD users were reviewed for a five-year period. During the period of study, all the women were using TCu 380A IUD. No significant difference was found between users whose records were reviewed and those whose records were not reviewed in terms of sociodemographic characteristics. Table 1 shows a comparison of different characteristics between the women who discontinued early with those who did not. Ove ra l l (N=230) , B rahm in and Kshetri women (35.7%) were found to be proportionately higher than other ethnicity. Similarly, a great majority of the women were Hindus (74.8%) followed by Buddhist (16.5%) and Christian (4.8%). IUD was used by women of almost all reproductive age groups with the highest number in the age group of 24–34 years. IUD was mostly used by suburban women (79.5%) and early discontinuation was also very high among them. By education, IUD was mostly used by women who can only read and write (37.8%). A majority of the women were housewives by occupation and early discontinuation was also found to be high among them. Similarly, the husbands of the women were more involved in business, followed by agriculture. The women whose husbands were farmers had higher rates of early discontinuation, whereas most of the women whose husbands were businessmen did not discontinue early. Similarly, early discontinuation was high among women whose husbands had secondary or higher education. Women having no child were not found using IUD and a majority of women having two children were found to be using IUD. Early discontinuation was more among women who have two children. A great majority of women had an IUD inserted while the age of their last child was less than four and early discontinuation was higher among these women. However, equal proportions of women were found having a male or female child at last pregnancy, but early discontinuation was higher among women having a female child at last pregnancy. Health workers and Book 1.indb 311 12-Dec-2012 9:02:29 AM WHO South-East Asia Journal of Public Health 2012;1(3):309-319312 Early discontinuation of intrauterine contraceptive device in Nepal – a retrospective study Subash Thapa Table 1: Characteristics of IUD users, and comparison between women who discontinued early and who did not discontinue earlyi Characteristics Early discontinuers (N=49) Who did not discontinue early (N=181) Total (N=230) P-Value ii Ethnicity iii Brahmin and Kshetri 13 (26.5%) 69 (38.1%) 82 (35.7%) 0.12 Mangolian 17 (34.7%) 35 (19.3%) 52 (22.6%) Newar 13 (26.5%) 49 (27.1%) 62 (27.0%) Terai caste and Dalit 6 (12.2%) 28 (15.5%) 34 (14.7%) Religion Hindu 36 (73.5%) 136 (74.5%) 172 (74.8%) 0.59 Buddhist 9 (18.3%) 29 (15.3%) 38 (16.5%) Christian and others 4 (3.2%) 16 (5.8%) 20 (8.7%) Age Less than 24 years 16 (32.7%) 49 (27.1%) 65 (28.3%) 0.21 24 to 34 years 23 (46.6%) 97 (53.6%) 120 (52.2%) More than 35 years 10 (20.4%) 35(19.3%) 45(19.5) Place of residence iv Urban 9 (18.4%) 61 (33.7%) 70 (30.4%) 0.89 Rural 10 (20.4%) 37 (20.4%) 47 (20.4%) Suburban 30 (61.2%) 83 (45.9%) 113 (49.2%) Educational status Illiterate 13 (26.5%) 54 (29.8%) 67 (29.1%) 0.62 Can only read and write 16 (32.7%) 71 (39.2%) 87 (37.8%) Up to secondary level 15 (30.6%) 41 (22.7%) 56 (24.4%) Higher 5 (10.2%) 15 (8.3%) 20 (8.7%) Occupational status Agriculture 7 (14.3%) 14 (7.7%) 21 9.1%) 0.00* Service 9 (18.4%) 28 (15.5%) 37 (16.1%) Labour 7 (14.3%) 5 (2.8%) 12 (5.2%) Business 4 (8.2%) 20 (11.0%) 24 (10.4%) Housewife 22 (44.9%) 114 (63.0%) 136 (59.2%) Husband’s occupational Status Agriculture 21 (42.9%) 46 (25.4%) 67 (29.1%) 0.00* Service 4 (8.2%) 50 (27.6%) 54 (23.5%) Labour 3 (6.1%) 16 (8.8%) 19 (8.3%) Business 16 (32.7%) 63 (34.8%) 79 (34.3%) Foreign employment 5 (10.2%) 6 (3.3%) 11 (4.8%) Book 1.indb 312 12-Dec-2012 9:02:29 AM WHO South-East Asia Journal of Public Health 2012;1(3):309-319 313 Early discontinuation of intrauterine contraceptive device in Nepal – a retrospective study Subash Thapa Characteristics Early discontinuers (N=49) Who did not discontinue early (N=181) Total (N=230) P-Value ii Husband’s educational status Illiterate 9 (18.4%) 20 (11.0%) 29 (12.6%) 0.00* Can only read and write 8 (16.3%) 79 (43.6%) 87 (37.8%) Up to secondary level 16 (32.7%) 59 (32.6%) 75 (32.6%) Higher 16 (32.7%) 23 (12.7%) 39 (17.0%) Total number of children during insertion One 10 (20.4%) 33 (18.2%) 43 (18.7%) 0.94 Two 25 (51.0%) 100 (55.2%) 125 (54.2%) Three 10 (20.4%) 36 (19.9%) 46 (20.0%) Four and more 4 (8.2%) 12 (6.7%) 160 (7.1%) Age of last child Less than or equal to four years 37 (75.5%) 130 (71.8%) 167 (72.6%) 0.60 More than four years 12 (24.5%) 51 (28.2%) 63 (27.4%) Sex of last child Male 19 (38.8%) 97 (53.6%) 116 (50.4%) 0.06 Female 30 (61.2%) 84 (46.4%) 114 (49.6%) Source of information about IUD Friends 16 (32.7%) 36 (19.9%) 52 (22.6%) 0.13 Husband 11 (22.4%) 31 (17.1%) 42 (18.3%) Health worker 16 (32.7%) 83 (45.9%) 99 (43.0%) Radio/TV 6 (12.2%) 31 (17.1%) 37 (16.1%) Reproductive intention Birth spacing 24 (49.0%) 86 (47.5%) 110 (47.8%) 0.09 Limiting childbirth 14 (28.6%) 74 (40.9%) 88 (38.3%) Don’t know 11 (22.4%) 21 (11.6%) 32 (13.9%) i The dependent variable is dichotomous (who discontinued early/ who did not discontinue early). Early discontinuation was taken for the period of removal of IUD within the first 12 months of insertion. ii P-value is calculated from chi-squared statistics. Association is significant when P value is less than 0.05 and significant association is denoted by.* iii Brahmin and Khhetri are the higher ethnic groups in Nepal. Terai caste is used to represent the indigenous group living in the plains of Nepal. Dalit is used to indicate lower ethnic group. iv Suburban are the periphery villages nearby a municipality having larger population. friends were found to be the major sources of information about IUD. Similarly, early discontinuation was higher among women who intended to space birth than women who intended to limit birth by using IUD. Book 1.indb 313 12-Dec-2012 9:02:29 AM WHO South-East Asia Journal of Public Health 2012;1(3):309-319314 Early discontinuation of intrauterine contraceptive device in Nepal – a retrospective study Subash Thapa Table 2 shows the reasons for early discontinuation of IUD and association of different side-effects with early discontinuation. Side-effect was the main reason mentioned by the women for early discontinuation of IUD, followed by expulsion. Of the total users (N=230), 47.4% complained of having side-effects after insertion of IUD and early Table 2: Reasons for early discontinuation of IUD and its association with side-effects Characteristics Early Discontinuers (N=49) Who did not discontinue early (N=181) Total (N=230) P-Value Reasons for early discontinuation + Side-effects and health concern 17 (34.7%) 27 (14.5%) 44 (19.1%) NA Expulsion 13 (26.5%) 11 (6.1%) 24 (10.4%) No need≠ 10 (20.4%) 15 (8.3%) 25 (10.9%) Dissatisfied with IUD§ 9 (18.4%) 13 (7.2%) 22 (9.6%) Side-effects Yes 30 (61.2%) 79 (43.6%) 109 (47.4%) 0.02* No 19 (38.8%) 102 (56.4%) 121 (52.6%) Bleeding Yes 13 (26.5%) 30 (16.6%) 43 (18.7%) 0.11 No 36 (73.5%) 151 (83.4%) 187 (81.3%) Menstrual disorder Yes 7 (14.3%) 7 (3.9%0 14 (6.1%) 0.00* No 42 (85.7%) 174 (96.1%) 216 (93.9%) Abdominal pain Yes 10 (20.4%) 37 (20.4%) 47 (20.4%) 0.99 No 39 (79.6%) 144 (79.6%) 183 (79.6%) Abnormal vaginal discharge Yes 11 (22.4%) 16 (8.8%) 27 (11.7%) 0.00* No 38 (77.6%) 165 (91.2%) 203 (88.3%) Itching and other complaints Yes 5 (10.2%) 18 (9.9%) 23 (10.0%) 0.95 No 44 (89.8%) 163 (90.1%) 207 (90.0%) Follow-up Yes 25 (51.0%) 126 (69.6%) 151 (65.7%) 0.01* No 24 (49.0%) 55 (30.4%) 79 (34.3%) + The total number of early discontinuers is 49 and late discontinuers is 66 in this analysis, because till the period of study, only 115 discontinued (early or not early) and the remaining 115 were continuing to use of IUD. ≠ No need of family planning includes need of a child and husband migrated to foreign country. § Dissatisfied with the device includes shifting of the choice of device, husband’s refusal and failure of the device. Book 1.indb 314 12-Dec-2012 9:02:29 AM WHO South-East Asia Journal of Public Health 2012;1(3):309-319 315 Early discontinuation of intrauterine contraceptive device in Nepal – a retrospective study Subash Thapa discontinuation was also found higher among them. The major complaints were bleeding, menstrual disorder, abnormal vaginal discharge, itching and other complains. Menstrual disorder and abnormal vaginal discharge were associated with early discontinuation of IUD. Of the total users (N=230), 65.7% visited for follow-up which includes both, follow-up for complaining side-effects (40.9%) and follow- up for routine checkup without side-effects (24.8%). Of the women visiting for follow- up, 39.1% made follow-up visits twice and 33.1% made a follow-up visit only once. Early discontinuation was higher among women who do not practice follow-up. One fifth of the total users (21.3%) discontinued IUD within the first year of insertion. The discontinuation rate reached 31.3% in the second year, 40.4% in the third year, 46.5% in the fourth year and 50% in the fifth year of insertion. A great majority of the women (N=163) had also used family planning methods in the past. Most of them used depo-provera (49.1%). The proportion of women who repeated IUD for the second time was 9.6%. The other devices used in the past were Norplant (5.7%), pills (5.7%) and condom (0.9%). Among the women removing IUD (N=115), 22.6% used nothing and 14% wanted childbirth after discontinuation of IUD. However, 23.4% of them used depo-provera and 13 % went for permanent sterilization. Table 3: Logistic regression analysis of selected variables and risk of early discontinuation of IUD+ Characteristics Crude OR (95% CI) Adjusted OR≠ (95% CI) Place of residence Urban 1.00 (Ref.) 1.00 (Ref.) Rural 2.450(1.084 - 5.534) 1.604 (0.611–4.206) Suburban 1.337(0.593 - 3.018) 1.176 (0.467-2.961) Reproductive Intention Birth spacing 1.00 (Ref.) 1.00 (Ref.) Limiting childbirth 1.877(0.796 - 4.429) 2.036 (0.704-5.892) Don’t know 2.769(1.096 - 6.993) 1.945 (0.634-5.965) Follow-up Yes 1.00 (Ref.) 1.00 (Ref.) No 2.199 (1.156 – 4.185) 1.556(0.731-3.313) Side-effects No 1.00 (Ref.) 1.00 (Ref.) Yes 2.039 (1.069 – 3.887) 2.623 (1.182-5.822) + The dependent variable is dichotomous. Early discontinuation was taken for the period of removal of IUD within the first 12 months of insertion. ≠ Adjusted odds ratios were calculated by logistic regression analysis between the early discontinuation and selected factors, after adjusting for the women’s occupational status, husband’s occupational status and husband’s educational status. Book 1.indb 315 12-Dec-2012 9:02:29 AM WHO South-East Asia Journal of Public Health 2012;1(3):309-319316 Early discontinuation of intrauterine contraceptive device in Nepal – a retrospective study Subash Thapa Association of selected factors with early discontinuation of IUD Univariate analysis identified association of woman’s occupational status, husband’s occupational status, husband’s educational status, side-effects and follow-up with early discontinuation of IUD. Results based on logistic regression analyses are shown in Table 3. In these models, place of residence, reproductive intention, side-effects, and follow- up exposure variables were used. Significant association was detected only between side-effect and early discontinuation of IUD after adjusting with woman’s occupational status, husband’s occupational status, and husband’s educational status. Similarly, Figure 1 illustrates the relationship between side-effects and continuation of IUD. It has been shown from the diagram that women experiencing side-effects have low probability of continuing use of IUD as compared with those not experiencing any side-effects. Discussion This study was carried out to examine the factors associated with early discontinuation of IUD in Nepal. The results are consistent with previous studies suggesting that experiencing side- effects is associated with early discontinuation of IUD. This study found that side-effect is the main reason for IUD discontinuation which was well supported by the evidence shown by some previous studies.1,13,16 Sociodemographic variables are not the major constraints for discontinuation of the IUD.17 This study also supported the same finding that women’s literacy and age were not associated with early discontinuation, while some studies had found women with low education; lower parity and age are more likely to discontinue use of the IUD.10,11 Studies in Nepal suggest that the IUD is popular among urban and suburban women, which was also well documented in this study.1 Residing in a rural area is associated with increased risk Figure 1: Kaplan-Meier analysis showing relationship between side-effect and continuation of IUD 1.0 0.8 0.6 0.4 0.2 0.0 0.00 20.00 Duration of use in months side-effects occured C u m u la ti ve S u rv iv al Pr ob ab ili ty yes no yes-censored no-censored 40.00 60.00 80.00 100.00 Book 1.indb 316 12-Dec-2012 9:02:30 AM WHO South-East Asia Journal of Public Health 2012;1(3):309-319 317 Early discontinuation of intrauterine contraceptive device in Nepal – a retrospective study Subash Thapa of discontinuation.18 However, in this study, place of residence was not associated with early discontinuation of IUD, but women’s occupational status was associated with early discontinuation. Studies documented that spousal approval has a direct impact on women’s reproductive behaviour.19 Similarly, experience of side- effects and their impact on continuation were significantly affected by spousal factors.19 This study also identified early discontinuation of IUD associated with husband’s educational and occupational status. Muslim women appear to be more likely to discontinue use of IUD than women of other religions, although in some Muslim countries, such as Turkey and Iran, discontinuation rates are fairly low.9,10,20 A great majority of the women in this study were Hindu, and no association was found between religion and early discontinuation. The family size preference is apparently more important than sex of the last child in determining reproductive intention and contraception behaviour.4,7 Women’s reproductive intentions govern their choice of method and guide them for making a conscious decision for discontinuation.2,9 However, no association was found between total number of children and sex of last child with early discontinuation behaviour. Similarly, no association was found between reproductive intention and early discontinuation of IUD. The rate of discontinuation of IUD in the first year in the study was higher as compared with a cohort study done among Jordanian women.21 Some studies concluded planned pregnancy as the most common reason for discontinuation of all methods of contraception.7,8 This study found that side-effect was the major reason for IUD discontinuation. The risk of early discontinuation was nearly three times higher among women experiencing side- effects, even after adjusting with women’s occupational status, husband’s occupational status, and husband’s educational status. This finding was consistent with previous studies elsewhere.1,13,20 This study identified abdominal pain as the most common side- effect while a study done among Jordanian women found bleeding as the most common side-effect experienced.21 The study identified menstrual disorder and abnormal vaginal discharge as the most significant side-effects for early discontinuation of IUD. In Nepal, although the majority of users report that they were told when to return to the facility for a follow-up visit, only about 60% of IUD users practice follow-up.1 An association between early discontinuation of IUD and follow-up visit was found in this study. Different methods of contraception used by women after discontinuation of IUD had been identified by the study. Only one fourth of the women discontinuing IUD switched to a permanent method and a great majority switched to less effective methods. On the other hand, 22.7% of women did not use any devices after removing IUD, which signifies the high risk of unwanted pregnancy. This study has several programmatic implications.5 First, the high rate of early discontinuation due to side-effects and expulsion suggests the need of special pre- insertion screening and counselling services. Second, issues related to side-effects need to be more openly discussed and properly addressed. Third, there might be some possibilities that encouraging for follow-up service within a few months after insertion will minimize and prevent early discontinuation. The current study has several methodological strengths. Women using IUD were recruited from a large set of 540 IUD users. A detailed analysis of the variables was carried out over a long period of time. However, the findings of Book 1.indb 317 12-Dec-2012 9:02:30 AM WHO South-East Asia Journal of Public Health 2012;1(3):309-319318 Early discontinuation of intrauterine contraceptive device in Nepal – a retrospective study Subash Thapa the current study need to be interpreted in the context of its methodological shortcomings. Due to the retrospective method, it could be argued that the observed associations are due to information and selection bias. Small sample size is an important limitation of this study. In addition, the study was carried out among the users of a family planning centre; the findings may have limited generalizability. Health service factors, familial factors as well as sociocultural factors were also shown to be associated with the discontinuation of contraception in previous studies.9,10,11 Unfortunately, it was beyond the scope of this study to collect information on these factors. Conclusion and recommendation This study ident i f ied that the ear ly discontinuation of IUD was very high in Nepal and side-effects were found to be the only major reason for early discontinuation. So, it was understood that proper management of side-effect in early days would reduce early discontinuation. This IUD discontinuation study, along with future research in this area, can help policy-makers and programme managers track family planning progress and help them to meet the goal of reproductive health for all. Thus, further research in this area is required. References USAID, Engender Health. 1. Contraceptive use and discontinuation patterns in Nepal. 2003. http:// www.engenderhealth.org/files/pubs/family-planning/ nepal_report_on_contraceptive_use.pdf - accessed 18 July 2012. Ali Mohamed, John Cleland. Contraceptive 2. discontinuation in six developing countries: a cause-specific analysis. International Family Planning Perspectives. 1995; 21(3):92-97. Zlidar VM. 3. New survey findings: the reproductive revolution continues. Baltimore: Johns Hopkins Bloomberg School of Public Health, 2003. Chowdhury AI, Fauveau V, Aziz KM. Effect of child 4. survival on contraception use in Bangladesh. Journal of Biosocial Science. 1992; 24(4):427–432. Mahdy NH, El-Zeiny NA. Probability of contraceptive 5. continuation and its determinants. Eastern Mediterranean Health Journal. 1999; 5(3):526– 539. Rehan N, Inayatullah A., Chaudhary I. Efficacy 6. and continuation rates of Norplant in Pakistan. Contraception. 1999; 60(1):39–43. Asari VG. Determinants of contraceptive use in 7. Kerala: the case of son/daughter preference. Journal of Family Welfare. 1994; 40(3):19–25. Youssef RM. Contraception use and probability of 8. continuation: community-based survey of women in southern Jordan. Eastern Mediterranean Health Journal. 2005; 11(4):545-558. Bhat PN, Halli SS. Factors influencing continuation of 9. IUD use in South India: evidence from a multivariate analysis. Journal of Biosocial Science. 1998; 30(3):297–319. Rivera R, Chen-Mok M, McMullen S. Analysis of user 10. characteristics that may affect early discontinuation of the TCu-380A IUD. Contraception. 1999; 60(3):155-160. Petta CA, Amatya R, Farr G, Chi I. An analysis of 11. the personal reasons for discontinuing IUD use. Contraception. 1994; 50(4):339-347. Salhan S, Tripathi V. Factors influencing discontinuation 12. of intrauterine contraceptive devices: an assessment in the Indian context. European Journal of Contraception and Reproductive Health Care. 2004; 9(4):245–259. Ministry of Health, University Research Corporation. 13. Developing strategies to increase IUD use in urban areas. Kathmandu, 1993. Lemeshow S, Hosmer DW, Klar J, Lwanga SK. 14. Adequacy of sample size in health studies. Statistics in Medicine. 1990; 9(11):1382. Machin D, Campbell MJ, Fayers MP, Pinal APY. 15. Sample size tables for clinical studies. 2nd edn. London, Blackwell Science, 1997. Book 1.indb 318 12-Dec-2012 9:02:30 AM WHO South-East Asia Journal of Public Health 2012;1(3):309-319 319 Early discontinuation of intrauterine contraceptive device in Nepal – a retrospective study Subash Thapa Mahdy NH, El-Zeiny NA. Probability of contraceptive 16. continuation and its determinants. Eastern Mediterranean Health Journal. 1999; 5(3):526- 539. Marie Stopes International Nepal. 17. Intra uterine contraception in Nepal - a cohort study. 2006. Petta CA, Amatya R, Farr G, Ching I-cheng. An 18. analysis of the personal reasons for discontinuing IUD use. Contraception. 1994; 50(4):339-347. Mahboob El-Alam, Searing H, Bradley J, Shabnam 19. F. IUD use and discontinuation in Bangladesh. New York; The Acquire Project, 2007. Jenabi E., Alizade SM., Baga RI. Continuation rates 20. and reasons for discontinuing TCu380A IUD use in Tabriz, Iran. Contraception. 2006; 74(6):483–486. Khader YS, El-Qaderi S, Khader AM. Intrauterine 21. contraceptive device discontinuation among Jordanian women: rate, causes and determinants. Journal of Family Planning and Reproductive Health Care. 2006; 32:161-164. Book 1.indb 319 12-Dec-2012 9:02:30 AM