Attitudes and knowledge of nurse practitioners towards traditional healing , faith healing and complementary medicine in the Northern Province of South Africa

O bjective: To investigate the attitudes and knowledge of nurses towards traditional healing, faith healing and com ple­ mentary therapies in the Northern Province in South Africa. Design: Survey of nurses. Setting: Registered professional nurses at health centres and clinics. Participants: 84 registered professional nurses Results: N urse’s perceptions were basically positive toward ethnomedical therapy (traditional healing, faith healing and complementary medicine); this also included their integration into the primary health care system. M ean ratings for referral to a faith healer was 2.7, followed by com plem entary medicine (2.6) and traditional healing (2.2). Although low rates o f referrals to ethnomedical therapists were practised, it was done so mainly in the patient’s interest and not as a last resort for chronic or term inal illness. Most did not discuss with a patient benefits o f traditional healing but 71% discussed the possible harmful effects. However, the majority discussed the benefits rather than harmful effects o f faith healing. With respect to mean ratings on knowledge, faith healing was considered the most important (4.3), followed by com ple­ mentary medicine (4.2), and traditional healing (4.1). C onclusion : Faith healing was considered as more important than com plem entary medicine and traditional healing. Implications are relevant for nursing health care and policy.


Introduction
There is som e resurgence o f interest in traditional healing m ethods as practised by indigenous South A fricans (Free m an & M otsei 1992: 1183).A lthough traditional and w est ern health care system s have operated side-by-side in South A frica since the advent o f the E uropeans, w estern healing has en jo y ed g re a te r fo rm a l (i.e.o ffic ia l) ac ce p ta n ce by su ccessive p revious govern m en ts because it w as seen to be based on scientific and rational know ledge.In contrast, traditional and faith healing has been officially frow ned upon and m arginalised because it w as perceived to be based on m ystical and m agical religious beliefs (Freem an & M otsei 1992(Freem an & M otsei : 1183)).B efore the A pril 1994 elections the A frican N atio n al C ongress (A N C ) pro p o sed the follow ing in its N ational H ealth Plan: T raditional healing w ill becom e an integral and recognised part o f health care in South A frica.C onsum ers w ill be al low ed to ch o o se w h o m to co n su lt fo r th eir h ealth care, and legislation will be changed to facilitate controlled use o f traditional practitioners.Soon after the elections, the new Government of National Unity form alised this policy in its Reconstruction and D evelopm ent Plan (RDP) (African National Congress 1994b: 3-5.).There are deep divisions, fueled by mutual suspicion and lack o f com m unication, betw een traditional and other com plem en tary healers and m edical and social workers.This is not in the interest o f people who use all types o f healers.The R econ struction and D evelopm ent Program m e (RDP) m ust aim to im prove com m unication, understanding and cooperation between different types o f healers (African N ational Congress 1994b: 3 5.).Hopa, S im bayeand d u T o it (1998: 8) have investigated the perceptions on integration o f traditional and w estern healing in the new South Africa.D ifferent stakeholders (psychiatrists, m edical doctors, psychologists, traditional healers and con sum ers) favoured the form al cooperation option for the inte gration o f the tw o health systems.A s noted in South Africa traditional and faith healers play an im portant role in health care (Peltzer 1998: 191, Peltzer 1999: 387, Pretorius 1989: 101, Shai-M ahoko 1996: 31).This m ay be expressed by the fact that for m any South A fricans the first choice o f treatm ent is the traditional healer (Farrand 1984: 779, Louw & Pretorius 1995:52f., Mabunda 1999:34-36, Swartz 1986: 280).M abunda (1999:10) surveyed hospital patients and staff, church m em bers and university students in the N orthern P rov ince in South A frica and found that -acco rd in g to the partici pants-a num ber o f diseases can be best prevented or cured by traditional healing, e.g.w itchcraft (like Sejeso and Sefolane) and ancestor (badim o) related problem s, "traditional" diseases like H logwana (pulsating fontanelle; litt.'littlehead'), M akgoma (assortm ent o f ailm ents w hich follows the breach o f particular taboos), infertility, sexually transm itted diseases, asthma, m en tal disorders, epilepsy, and diarrhea w hile biom edicine was considered successful in diseases such as tuberculosis, chicken

Curationis May 2002
pox, A ID S, hypertension, diabetes, m alaria, m easles, cancer, anem ia, m ental retardation, ulcers.Freem an andM otsei (1992: 1184) stated that there are broadly three types o f traditional healers available to South A frican consum ers.First the traditional doctor or inyanga.This is gen erally a m ale w ho uses herbal and other m edicinal preparations for treating disease.Second the dingaka (Sotho).This is u su ally a w om an who operates w ithin a traditional religious super natural context and acts as a m edium with the ancestral shades.T hird the faith healer who integrates Christian ritual and tradi tional practices.In addition, Fenn (1998: 3-4) gives a guide to com plem entary therapies in South A frica including hy p n o therapy, regression therapy, rebirthing, past life regression, psychoneuroim m unology (PNI) therapy, astrological counsel ling, tarot counselling, m etaphysical counselling, etc. N em ec (1980: 2) states that tw o-thirds o f the people in the w orld today depend on the healing m ethods used by their ancestors, and in som e areas it is the only form o f care avail able and w ith w hich people are com fortable.In every culture, illness, the responses to it, individuals experiencing it and treat ing it, and the social institutions relating to it are all system ati cally interconnected.The totality o f these interrelationships is the health care system .Included here are patterns o f belief about the causes o f illness, the norm s governing the choice and evaluation o f treatm ent, and socially legitim ated roles, pow er relationships, interacting setting and institutions.P a tients and healers are basic com ponents o f such system s and thus are em bedded in specific configurations o f cultural m ean ings and social relationships (Kleinm an 1980: 24).Original A f rican concepts o f disease causation incorporate b elief in natu ral or G od-given illness, and in supernatural forces, including sorcery, w itchcraft, spirit disturbance, and in breaching o f ta boos or breaking kinship rules or religious obligations.The causal fram ew ork em erges from a fundamentally Unitarian con ception o f nature w hich encom passes the physical w orld and a sociological environm ent that expresses continuity betw een the living and the dead, together with the m ystical forces o f the universe (W H O 1976:10-12).M ost traditional medical theories have a social and religious character and em phasize p reven tion and holistic features.Traditional m edical practices are usu ally characterized by the h ea le r's personal involvem ent, by secrecy and a rew ard system .B iom edical theory and practice show an alm ost opposite picture: asocial, irreligious, curative and organ-directed, and professional detachm ent (Van der Geest 1997:903).
T he role o f traditional and faith healing and com plem entary m edicine in the future o f health care in South A frica largely depend on its acceptance by the established m edical com m u nity.D efinitions used here are: traditional healers are herbal ists o r diviners, fa ith healers are prophets, priests, pastors w ho heal, and com plem entary m edicine includes therapies such as acupuncture, hypnosis, hom eopathy, chiropractic, m assage, touch therapy, vitam in therapy, etc. C om plem entary or alternative m edicine is here to cover W estern practices al though traditional or faith healing could also be understood as com plem entary medicine (Fenn 1998:11).Several developing countries in Africa, Asia and Latin America have experim ented with the integration o f traditional and W est ern health care system s, but the case o f traditional m edicine in C hina is perhaps the best m odel for the successful integration o f the two medical systems (Lesley in Pretorius 1999:4).W HO has strongly advocated prom otion o f cooperation between tra ditional and m odem health practitioners.In Botswana 'United Health Com mittees have been established aim ing at creating a dialogue betw een the different types of health professionals (Akerele 1987: 16).There is a general lack o f studies in South A frica investigating the attitude and know ledge o f nurse practitioners towards non biomedical or ethnomedical therapies.Mahape and Peltzer (1998: 39) studied the attitudes o f psychiatric nurses tow ards tradi tional healers as providers o f mental health in South Africa.Findings show ed that the nurses agree that it is possible to develop a w orking partnership w ith tradi tional healers in m ental health care, provided form al policy guidelines structuring the prac tice o f traditional healing are specified.
To assess attitudes and know ledge with respect to traditional healing, faith healing and com plem entary m edicine a survey was conducted am ong nurse practitioners in the N orthern Prov ince in South Africa.The N orthern Province is the m ost im poverished o f the nine provinces in South A frica with a real G G P per capita o f R 1712 Rand in 1994.The area is largely rural with 91% o f the inhabit ants living in non-urban areas w here accesses to health care facilities are inadequate (e.g.2.4 hospital beds per 1000 people in 1992 and a physician: population ratio o f less than 1: 20000).H ealth indicators are low (an infant m ortality rate o f 57:1000) (Developm ent Bank of Southern A frica 1998: 10).

Method Sample and procedure
The sam ple consisted o f nurse practitioners w orking at the randomly selected 9 health centres and 14 clinics o f the Lowveld and N orthern regions o f the N orthern Province.The tw o re gions com prised o f 18 health centres o f w hich every second, alphabetically chosen, was selected.T he study only utilised clinics found in the L ow veld because o f their geographic ac cessibility.E very sixth o f a total o f 83 clinics in the Low veld w as selected for the study.At all these 9 health centres and 13 clinics nurse practitioners were chosen by convenience.The final sample included 84 nurse practitioners; 20 w ere discarded because their questionnaires w ere not com plete.The nurse practitioners w ere given a self-adm inistered questionnaire af ter inform ed consent w as taken from them and anonym ity and confidentiality was assured.The nurses were 78 female (92.9%) and 6 m ale (7.1%) in the age range o f 23 to 58 years (M age 38.0 yr., SD=8.0).All participants w ere A frican (black).The num ber o f years o f health service ranged from 3 to 27 years, with a m ean o f 13.0 years (SD=6.9).Perm ission to conduct the study w as given by the U niversity o f the N orth Ethics C om m ittee and the N orthern Province D e partm ent o f H ealth and Welfare.

Measuring instrument
The questionnaire was a m odified and expanded version of w hat Boucher and Lenz (1992: 60-61) had used with American physicians.Section 1 included dem ographic data: sex, age, num ber o f years o f nursing practice, and ethnicity.Section 2 31 Curationis May 2002 had 8 item s on the perceptions about traditional healing (such as 'Is traditional healing a threat or benefit to the public', which was rated from l=extrem e threat to 5= extreme benefit), 8 items on faith healing and 8 item s on com plem entary m edicine.Sec tion 3 asked for referral inform ation: 3 items on referrals to traditional healer (such as 'H ave you ever referred or recom m ended a patient to a traditional healer?' w ith the response options 'Y es', 'No o f ca se s' and 'N o '), 3 items on faith healer and 3 item s on com plem entary m edicine practitioner; 6 items for each ethnom edical m ethod for possible reasons for referral (such as patient interest, as a last resort (chronic or term inal illness), know healer or com plem entary practitioner with good reputation, personally found it beneficial, for nonspecific com plaints and other reasons); and 8 item s on discussing benefits and harm ful effects for each ethnom edical m ethod (such as 'Have your ever discussed with a patient the possible benefits o f faith healing?').Section 4 identified the know ledge about various ethnom edical nesses regarding the appropriate referral agency (psychiatrist, m edical practitioner, clinical psychologist, traditional healer, faith healer, and com plem entary medicine practitioner).Finally, open questions asked for case descriptions of patients who had consulted a traditional healer, faith h ealer or com plem en tary m edicine practitioner before consulting the nurse and case descriptions o f patients he or she had referred or recom m ended to a traditional healer, faith healer or com plem entary m edicine practitioner.A pilo t study was conducted on 20 nurses to test the reliability o f the questionnaire.T he same nurses (from the pilot study) w ere asked to respond to the questionnaire after 3 w eeks.A test-retest reliability o f .82 was found.

Data analysis
R eported case studies w ere content analysed, and descriptive statistics were calculated using SPSS version 8.0.treatm ent m ethods (such as acupuncture and herbal therapy, w hich was anchored from l= unfam iliar to 3=understanding of proposed uses).Section 5 rated on a 5-point scale 9 types of evidence for the evaluation of ethnom edical m ethods (such as 'prospective clinical random ized controlled trials and case re ports in com plem entary' or 'alternative m edicine jo u rn als', w hich was rated from 1= this type o f evidence w ould have no impact on my thoughts about ethnom edical therapies to 5=this type o f evidence w ould convince me that the treatm ent us highly effective).Section 6 asked to rate 14 problem s or ill-

Perceptions
O ne the one hand so-called ethnom edical m ethods were seen as quackery by 41 % for faith healing, 4 6 % traditional healing and 49% for com plem entary healing, and on the other hand 69% felt that traditional healing held prom ise for the treatm ent o f sym ptom s, conditions, and/or diseases, 65% for faith heal ing and 64% for com plem entary m edicine, respectively.M ore than one third (betw een 33% and 42% ) felt that any o f the

Curationis May 2002
ethnom edical m ethods had no true im pact on the treatm ent of sym ptom s, conditions, and/or diseases.It appeared that it d e p ended m uch on the type and expertise o f the p articu lar ethnom edical m ethod w hether it was considered prom ising or quackery.Similarly, the overw helm ing m ajority (surprisingly) o f the nurse practitioners felt that the integration o f faith heal ing (98%), traditional healing (93%) and com plem entary m edi cine (77% ) into the national health care system had som e posi tive im pact on patient satisfaction.However, many of the nurses also perceived that the integration o f ethnom edical m edicine (traditional healing: 76% , faith healing: 77% , com plem entary m edicine: 49% ) into the national health care system had a m a jo r negative im pact on patient satisfaction (see Table 1).
Regarding the potential threat or benefit of ethnomedical m eth ods to the public, the m ajority (77% to 87% ) felt that they had a m oderate benefit.However, on the other hand betw een 68% and 75% felt that they were a m oderate threat, even 63% felt faith healing to be an extrem e threat, traditional healing 52% and faith healing 52% respectively (see Table 2).
Table 2 : Threat or benefit of traditional healing, faith healing, and complementary medicine to the public in percent N urse practitioners would more likely refer a patient to a faith healer, second to a com plem entary m edicine practitioner and last to a traditional healer (see Table 4).
A list o f 5 possible reasons for referral w as provided.Partici pants w ere asked to check each reason that applied.An 'other' category was also added on the list so participants could write in their ow n responses.The m ost com m on reason for referral w as 'patient interest', follow ed by 'personally found it benefi cial', 'as a last resort' and 'acquaintance w ith a good therapist' (see Table 5).
Table 6 indicates "discussing benefits and harm ful effects of referring a patient" .
A lm ost h alf o f the nurse practitioners (43% ) discussed the benefits of ethnomedical treatment with their patients.The nurse initiated the discussion about the benefits of faith healing (52%, as opposed to 56% the patient) and com plem entary m edicine (40% , as opposed to 39% the patient) and the patient initiated T hree questions were asked about rates of referral.The first inquired w hether nurse practitioners had ever referred or rec om m ended a patient to an ethnom edical practitioner, the sec ond asked w hether it was likely they would do so in the future, and third asked about rates o f referral or recom m endation o f a patient from a ethnom edical practitioner.D epending on the ethnom edical m ethod only betw een 14% and 26% o f the nurse practitioners had referred a patient.38% would likely or very likely refer a patient to a faith healer in the future as opposed to 21% to a traditional healer.M ost referrals were received from traditional healers (55%), 33% from faith healers and 20% from com plem entary practitioners (see Table 3).
the discussion about the benefits o f traditional healing (30% as opposed to 43% the patient).Clearly, the m ajority o f the nurse practitioners (71% ) discussed the harm ful effects o f re ferral to a traditional healer, 58% said that they initiated the discussion and 51 % said the patient initiated the discussion.
A lm ost h alf of the nurses (42% ) also discussed the harm ful effects o f faith healing, and 38% said that they discussed the harm ful effects o f com plem entary m edicine.A m ong the nurse practitioners w ho said that they discussed the possible harm ful effects o f ethnom edical treatment, the discussion was m ore often initiated by the nurses than by the patient.

Curationis May 2002
Overall, for most problems the medical practitioner seem to be perceived as the most appropriate for referral, followed by the clinical psychologist, psychiatrist, faith healer, complementary medicine practitioner and last the traditional healer.The five most frequently rated problems for referral for the medical prac titioner were: (1) asthma, (2) allergies, (3) HIV/ADDS, (4) cancer, and (5) terminal illness; for the clinical psychologist: (1) sexual referred or recommended to a ethnomedical therapist are sum marized below: Table 5 From traditional healer: To traditional healer:  On the basis of the case analysis and also as earlier stated where most referrals were received from traditional healers (55%) it was found that patients had often consulted traditional heal ers before coming to the clinic and that nurses frequently re ferred or recommended patients to a faith healer.

Referral case descriptions
From the case descriptions from a healer to the nurse and to a healer from the nurse selected examples are described below: (1) "A lady of 35years old experienced infertility.She consulted a traditional healer.She was given some herbs to drink for to conceive a baby.After a day's dosage, she was brought to hospital in a critical state.She died of over dosage." (2) "A lady was suffering from hypertension.A nurse referred 35 Curationis May 2002 her to a faith healer.I quote 'God can heal any illness, no matter what.The blood of Jesus was shed to wash our sickness and sins away.God can heal hypertension and diabetes if and only you believe and have faith in him'."Survey participants were also queried about their knowledge of various ethnomedical therapies.Specific questions were asked concerning acupuncture, traditional healing, homeopa thy, etc. Blank spaces were left for participants to write in other therapies.Participants were asked to rank their knowledge of (3) "A mother on psychiatric treatment of depression was re ferred by a nurse to attend a church conference.She was prayed for and was healed." (4) "A patient refused mastectomy as it was still early for op eration.A nurse referred her to a traditional healer for treat ment.She took treatment for 3 months.Thereafter the mass disappeared, and the patient was healed." (5) "A subordinate who experienced work related stress that affected his productivity.His supervisor recommended mas sage therapy and a gym for stress relieving exercises."

Knowledge
When asked to evaluate the statement "Nurses should have knowledge about the most prominent ethnomedical treatments", the highest rating was given for faith healing (4.3), followed by complementary therapies (4.2), and traditional healing (4.1) (see Table 8).
About one fifth (22%) of the participants had attended lec tures or workshops or received training in the use of ethnomedical therapy.Workshop topics included herbal medi cine and acupuncture (8%).
The calculated means indicated that faith healing was best understood and acupuncture was least understood (2.1 and 1.3, respectively).

Treatment outcomes and scientific evidence
When asked whether the use of ethnomedical therapy would result in favourable or unfavourable treatment outcomes, more than 80% of the nurses thought that there would be 'some' positive outcome.However, more than 68% thought there would be 'some' negative outcome.There was also a high rate of 'don't know' answers (66% to 75%) (see Table 10).
Using the following 5-point Likert scale, participants were asked to evaluate 9 types of evidence in ethnomedical treatments and the impact of each on their thinking: (1)  (5) This type of evidence would convince me that the treat ment is highly effective.Nurse practitioners indicated that personal experience would provide the most convincing evidence of the efficacy of a ethnomedical modality (M=2.87),followed by recommendation of a specialist (M=2.83), and case reports in complementary medicine journals (M=2.81).The break down for each category is presented in Table 11.

Discussion
Generally, nurse practitioners ranked faith healing as most im portant, followed by complementary medicine practitioner and traditional healing in that order.This may be explained by the African cultural and religious background of the nurses.There was both a high positive and negative attitude towards any of the three ethnomedical therapies, which seems to indicate that the quality differs.This needs to be further investigated.Surely, the role of ethnomedical therapies in rural health care has been identified and a positive attitude towards their integration into the national health care system has been expressed (se also Osei 1994:39, Ryan 1998:209).In line with other studies (e.g.Mohape & Peltzer 1998: 39) the majority of the psychiatric nurses favoured the integration of faith healing (98%), traditional healing (93%), and complemen tary medicine (77%) into the national health care system.These psychiatric nurses were all African South Africans; most were Protestants (60%), followed by Roman Catholics (19%) and other religious denominations such as Zion Christian Church and Jehovah's Witnesses.However, low rates of referral to a faith healer, complementary medicine practitioner and traditional healers were practised, and the majority discussed rather the harmful effects than the benefits of ethnomedical therapies.This may indicate some opposition in the collaboration of the nurses with ethnomedical therapies.Nurses in Botswana were also found to be opposed medical health system (Barbee 1986: 75).Among nurses in Malawi, a number of cultural conflicts were identified: (1) con flict with traditional healing and medicine, (2) conflict with tra ditional diet and feeding practices, (3) conflict with family/gen der taboos, (4) conflict with concepts of illness, (5) conflict with religion, (6) conflict with sexual practices, (7) conflict with family planning, and (8) conflict with "death".Further it was found that the nurses generally would ask the patient about the belief underlying her/his cultural practice.Thereafter the nurse will try to convince the patient about a biomedical con cept of health and health care (Peltzer 1997: 159).This study found that patients had often consulted traditional healers before coming to the clinic, in particular for conditions such as tuberculosis, HIV/AIDS, cancer, renal failure, epilepsy and sexually transmitted diseases.Generally, there seemed to have been low referral rates by the nurse clinicians to any of the ethnomedical practitioners (14% to traditional healers, 22% to faith healers and 26% to comple mentary medical practitioners).For example, in the UK Perkin, Pearcy and Fraser (1994: 523) found that 70% of hospital doc tors and 93% of general practitioners had, on at least one occa sion, suggested a referral for complementary treatment.Patients normally initiated discussing the benefits and nurses normally initiated the harmful effects of traditional and faith healing.This means that patients and nurses may have differ ing views on the value of traditional and faith healing.On the other hand, in regard to the treatment outcome of ethnomedical therapies most nurses acknowledged positive effects, which was also found in other studies (e.g., Farrand 1984: 779).

Recommendations
The study provided evidence that there is a two way referral process of clients/patients from a healer to a nurse and from a nurse to a healer.Therefore, it is important that healers and primary health care nurses have forums to share experiences concerning their beliefs and treatment modalities of health prob- to the collaboration with spiritual and traditional healers.They ^ems Prevalent >n the communities.were perceived to be caught in a dialectical tension between Further research 1S needed t0 exPlore the cultural behefs of their traditional beliefs and their acquired beliefs in the bio-Pnmary health care nurses about health and lllness' and how these beliefs have an impact on their practice as nurses.

Curationis May 2002
A nurse has a role to assist a patient to choose among different ethnomedical therapies for the sake of the patient's well-being.Therefore, there is a need that nurses understand and respect the patient's value system with regard to his or her choice about maintaining good health.In order to improve the quality of traditional, faith healing and complementary medicine, the Health and Welfare, Northern Province, for giving permission to conduct the study.

References
AFRICAN NATIONAL CONGRESS 1994A: A national health plan fo r South Africa.Johannesburg: Author.sectors in which they can be most effective should be identi fied such as psycho-social and mental instability seem to be some of them (Peltzer 2001:10).

number o f X ethnomedical methods hold promise 58 (69) 55 (65) 54 (64) Has the integration o f X ethnomedical method into the national health care system some (major) positive impact on patient satisfaction? 78 (93) 82 (98) 65 (77) Has the integration o f X healing/medicine into the national health care system some (major) negative impact on patient satisfaction? 64 (76) 65 (77) 41 (49)
*X stands for each respective treatm ent m odality (traditional healing, faith healing and com plem entary m edicine)

Table 7
indicates 14 different problem types in relation to 6 different referral options.

lik ely w o u ld you refer/recom m end a patient to a faith healer in the future 2 .7 4 1.44 H ow lik ely w o u ld y o u refer/recom m end a patient to an com p lem entary m ed icin e practitioner in the future 2.61 1.46 H ow lik e ly w o u ld y o u refer/recom m end a patient to a traditional healer in the future
Nurses' case descriptions of patients who had consulted a traditional healer, faith healer or complementary medicine prac titioner before consulting and case descriptions of patients