FACTORS ASSOCIATED WITH THE PERCEPTION OF SIDE-EFFECTS RELATING TO THE USE OF CONTRACEPTIVE METHODS

Die doel met die ondersoek was om gebruikers van moderne kliniese kontraseptiewe (orale pil, hormonale inspuiting en IUT) se persepsies van newe-effekte wat met gebruik gepaard gaan te bepaal, en om faktore wat met hierdie persepsies verband hou te identifiseer. Die studie is onder Swart vroue in ’n woonbuurt van Benoni gedoen. ’n Steekproef van tweehonderd gebruikers van elke metode is op ewekansige wyse uit die rekords van die plaaslike gesondheidskliniek getrek. Daar is bevind dat ’n aansienlike persentasie gebruikers van elke metode te kenne gegee het dat hulle voorheen of tans newe-effekte ervaar het wat hulle me* die besondere kontraseptief in verband gebring het. Probleme is ondervind deur 39,5 % van die inspuitinggebruikers, ietwat minder IUT-gebruikers (35,0 %) en ’n aansienlike laer persentasie pilgebruikers (21,5 %). Die meervoudige stapsgewyse regressie-analises het getoon dat uiteenlopende verklarende veranderlikes ’n statistiese beduidende verband met die persepsie van newe-effekte by die gebruikers van die onderskeie metodes gehad het. By gebruikers van die pil was dit die veranderlikes ekonomiese aktiwiteit, ouderdom en aantal metodes wat reeds gebruik is; in die geval van die inspuiting was dit kindertal en tydperk wat die metode gebruik is, terwyl slegs een veranderlike, naamlik aantal metodes wat reeds gebruik is by IUT-gebruikers uitgesonder is. Ofskoon daar wel veranderlikes behoort te wees wat met die voorkoms van newe-effekte verband hou en nie by die onder­ soek ingesluit is nie, kan dit uit die huidige ondersoek afgelei word dat die persepsie van newe-effekte in groter mate die gevolg van psigologiese as fisiologiese faktore kan wees.


R E S E A R C H DESIGN
The study was conducted in Daveyton, a Black residential township near B enoni.T h ree sam ples of 200 respondents for each o f the three contraceptive m ethods w ere drawn from the records o f a health clinic situated in the tow nship.A s the study was n o t focused on the influ ence of side-effects on continuation of a specific m eth o d , a low er limit of one year of continuous use of a specific m ethod was set in o rd er to exclude w om en w ho w ould be unable to give reliable responses due to a short period of use.This condition also excluded wom en who m ight be experim ent ing with a m ethod; who w ere not strongly m otivated to prev en t a pregnancy; who w ere unhappy with their m ethod and w ere contem plat ing changing to an o th er m ethod; and for whom one o r m ore of the m ethods w ere contra-indicated b e cause they had recently given birth or were breastfeeding.T he field work was undertaken during 1980.In te rv ie w s w e re c o n d u c te d by trained fieldw orkers in the language of the respondent.

INCIDENCE OF SIDE-EFFECTS
A list of side-effects o r problem s as sociated with the use of each of the three m ethods was com piled after consultation with m edical and nurs ing personnel attached to the national Family Planning P ro gramm e. T he list contained side-effects referred to in scientific litera ture and problem s and com plaints associated with th e use of co ntra ceptives that are frequently heard from clients at family planning clin ics.
D uring the interview s the respon dents were asked w hether they ex perienced any of the listed side-effects or problem s from the contra ceptive they w ere using, and if so, to elaborate on these.N otes w ere m ade of problem s that respondents m entioned which w ere not included in the schedule.
The percentage users of the three m ethods who indicated th at they had or were experiencing side-effects which they associated with the use of the contraceptive varied sub stantially.W hereas 21,5 % of pill users reported side-effects, the inci dence am ongst users of the IU D and injection w ere considerably higher, 35,0 % and 39,5 % respec tively.

F A C T O R S R E L A T E D TO TH E P ER CEPTIO N O F S ID E-EFFEC TS
As relatively little research has been done in South A frica concern ing factors related to perceived side-effects of contraceptive use, it was decided to include explanatory variables which are usually used in KAP surveys in the analyses.
T hree dem ographic variablesage, parity and desire for m ore chil dren -w ere selected.E ducational level and econom ic activity w ere in cluded as socio-econom ic variables.V ariables related to contraceptive use that were included were: parity at which a clinical m ethod was first accepted; reason for first accep tance and present use being for spacing or for term ination of fertil ity; the num ber of clinical m ethods ever used; w hether contraceptive use began on own initiative o r was m o tiv a te d by another p e r s o n ; period since the last birth; and du ration which the present m ethod was being used.
E xploratory m ultivariate analy ses showed th at four of the explana tory variables, desire for m ore chil dren, period since last birth, parity at which a clinical m ethod was first accepted, and w hether the reason for first acceptance was to space or term inate fertility, did not have a statistically significant relationship (p > 0,05) with the d ependent varia ble (side-effects) am ong all three of the samples of m ethod users.
The relationships betw een the dependent variables and the re maining eight explanatory variables were subsequently analysed by means of m ultivariate stepwise re gression analyses.In the case of the oral pill users, three of the eight ex planatory variables (age (p < 0,025), econom ic activity (p < 0,005) and num ber of m ethods used (p < 0 ,0 0 1 )) had a statistically sig nificant relationship with the d e pendent variable (side-effects).The analyses regarding IU D users re sulted in only one explanatory vari able (num ber of m ethods used (p < 0,025)), and in the case of the injec tion two variables (parity (p < 0,025) and period of m ethod use (p < 0,025) being statistically signifi cant.
The percentages of w om en re porting side-effects are shown only for explanatory variables that w ere statistically significant.T he p er centages w ere adjusted in o rd e r to elim inate the influence of all o ther explanatory variables included in the original regression models.
In the case of pill users percep tions of side-effects were far m ore prevalent am ong older w om en, wom en who w ere economically active, and w om en who had used m ethods in addition to the pill.In the case of injection users, the p er centages of wom en reporting sideeffects in the parity categories 3 -4 and 5 + , w ere m ore than double than that for w om en with parity 0 -2.M ore than half of the wom en who had used the injection for the shortest period rep o rted side-effects as against slightly m ore than one q uarter for the categories with longer use.A s with pill users, sideeffects w ere m ore prevalent among women who had used m ethods in addition to the IU D .

DISCUSSION
A lthough it is not possible from this research to conclude w hether the perceived side-effects had physio logical or psychological origins, it seems that m ost of the com plaints could be related to the latter rather than the form er. O lder wom en and wom en with higher parity m ay be m ore anxious to prevent p regnan cies, m ore concerned ab out the ef fectiveness of the m ethod, and thus more method-conscious than younger and low er parity wom en.T he sam e arg u m en t m ay hold for econom ically active users for whom an unplanned pregnancy may hold considerable financial im plications.W ith an increase in th e period of m ethod use, confidence in the m ethod m ay im prove and few er sym ptom s perceived as related to the m ethod p er se.

Pill
Lim ited research has been done inSouth A frica regarding the inci dence of side-effects and p ercep tions of these am ong users of con traceptives.N um erous know ledge, attitu d e and practice (K A P) surveys have b een co nducted by th e H um an Sciences R esearch C ouncil (E ras mus 1981a; E rasm us 1981b; E ra s m u s 1 9 8 2 ; Groenewald 1 9 7 8 ; L ó tter, V an T o n d er 1976; M ostert, M alherbe 1974; M o stert 1974; Strydom 1981; V an d er M erw e 1982) and in som e of these studies ques tions concerning side-effects w ere included to d eterm in e reasons for discontinuation of specific contra ceptive m ethods.T he literature concerning research on side-effects related to contraceptive use usually pertains eith er to clinical studies Gerda Erasmus, BA Hons, Senior Researcher, Institute for Sociological and Demographic Research W.P. Mostert DPhil, Senior Chief Researcher, Institute for Sociological and Demographic Research aim ed at determ ining types o f sideeffects associated w ith specific m ethods or to the influence o f these effects on the discontinuation of contraceptive use.The present study was conducted in order to d eterm ine (a) th e inci dence of perceived side-effects am ongst users o f th e intra-uterine device, the horm onal pill and the injection, and (b) factors associated with these perceptions.
W om en w ho have used m ore than one m eth o d could be regarded as m ore p ro n e o r hypersensitive to the som atic effects o f m edicines and changed m ethods for th at reason.H ow ever, th e fact th at these trends are n o t sim ilar for each o f th e three m ethods, suggests th at the p ercep tion of side-effects should n o t only be related to psychological factors.T h e fin d in g s i n d ic a te , furtherm ore, th at the presence and perception of side-effects relating to m o d e rn c o n tr a c e p tio n a re n o t clearly understood.It is suggested th at in-depth m ultidisciplinary re search be u n d ertaken in ord er to understand why w om en, speci fically B lacks, react to contracep tive use th e way they do.T he exis tence and perceptions of side-ef fects am ong considerable percent ages of w om en m ust be im portant factors in th e discontinuation of contraceptive use, leading to un w anted pregnancies and high fertil ity.