Violence is the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community that either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment or deprivation (UNO 1993; WHO 1996). Violence can be self-directed, interpersonal, or collectively carried out. Intimate partner violence (IPV), wife battering, in particular, is the most common form of interpersonal violence.

The issue of intimate partner violence has reached such a level globally that it now constitutes a matter of international public health priority. It was estimated that in the year 2,000 alone, about 520,000 persons were killed as a result of IPV, a rate of 8.8 per 100,000. Possibly, many more cases of such deaths could have escaped detection having been concealed as accidents, natural or death from unknown causes. For every mortality that occurred through violence, many more people are physically injured or psychologically damaged (WHO 2002). Violence against women (VAW) occurs in all countries, in every culture, and in all strata of society, although it may be common in lower socioeconomic groups. Population-based studies in many countries have reported 10% to 69% of women as having been abused at one time or the other in their lives (Ellsberg et al. 1999; Heise et al. 1999; Martin et al. 1999). Prevalence rates of 21% to 50% have been reported in Africa (Butchart and Brown 1991; Odujinrin 1993; Van der Straten et al. 1995). Violent tendencies in men are rooted in many complex factors including biological, social, cultural, economic, and religious ones (Ilika and Ilika 2005; Oyediran and Isiugo-Abanibe 2005). However, the common risk factors are growing up in a broken home or where wife battering is common, substance abuse, social isolation, income inequality, poverty, alcoholism, and low self-esteem. Events that trigger abusive behaviour are also fairly consistent in many developing countries: disobeying or arguing with the husband, query about money or extramarital affairs, food not being ready on time, suspicion of the wife having an extramarital relationship, and refusal to submit to sexual intercourse (Fawole et al. 2005; Schuler et al. 1996).

The burdens of both fatal and non-fatal physical and psychological injuries resulting from IPV are mostly left for the health sector to bear. Far more victims experiencing IPV are more likely to report to the health facilities rather than the police stations in Nigeria (Ilika et al. 2002).

The increasing recognition that IPV is a public health issue at international level has fortunately helped in sensitizing the health sector to the role it needs to play. IPV is a problem to be managed and prevented using the same tools that have been employed for campaigns against other priority public health issues, including HIV/AIDS, tuberculosis, and malaria (Butchart et al. 2004). The role of health professionals in the management of IPV cannot be over-emphasized, and goes beyond simple clinical interventions and drug prescriptions. They need the capacity to suspect, recognize, diagnose, and treat IPV. Beyond this, they should be involved with other stakeholders in the care, support, and rehabilitation of victims, and campaign against IPV. In other words, health professionals should be able to collect information on the cases, and identify risk factors and the consequences of IPV for them to be able to contribute directly to its prevention at local and national levels, and also be able to render quality care directly to the persons affected. Unfortunately, despite the global enthusiasm in the management of IPV, there is the widely observed lack of competence on the part of many health professionals. They may lack adequate capacity in detecting many of the cases and when they succeed in identifying cases, they may fail to get involved adequately as to manage the situation appropriately (Chambliss et al. 1995; Noel and Yam 1992; Turmen 1997; Webster et al. 1996). Other constraints to recognition, diagnosis, and treatment of IPV include fear of offending the person affected or her spouse, powerlessness on what to do next, lack of comfort in treating such cases, not wanting to get involved in police cases, and time constraints (Sugg and Inui 1992). Educational intervention program for health professionals have been found to be useful in improving the detection rate of cases (Berenson et al. 1991; Berrios and Grady 1991). Our earlier research in this environment has shown that the competence in detection of indicators of IPV is inadequate among the health professionals (Adeyemi et al. 2005). These indicators are the physical and psychosomatic symptoms manifested by patients, which mask the actual physical and psychological abuse and violation.

Our objective in the present study is to determine the knowledge base of health professionals about what constitutes IPV or Violence Against Women (VAW) and assess their attitudes, beliefs, and actions towards women experiencing spousal violence. Findings from this study will assist us to determine the training needs of the health workers on the required knowledge, change of attitude, and skills to live up to the challenges of the global pandemic of violence against women.

Methods

This was a cross-sectional study carried out in the Obafemi Awolowo University Teaching Hospitals Complex (OAUTHC), Ile-Ife, Nigeria. OAUTHC is one of the tertiary health institutions in the south-western part of Nigeria and serves at least five out of the 36 states of Nigeria. The Research and Ethical Committee of OAUTHC approved of the study. A trained facilitator was available to guide respondents to fill the questionnaire. It took 15 min to complete the entire questionnaire. Medical doctors and nursing staff received VAW questionnaire instruction in separate groups.

Medical doctors (n = 58) and nurses (n = 76) working in OAUTHC were included in the study via simple random sampling, and written informed consents were obtained from them.

Data were collected using the VAW questionnaire which had been pilot tested for face and content validity among the staff of a General Hospital within the location of the study and by consensual agreement among members of the study group (Domestic Violence Study Group). The questionnaire, which had three sections, was designed in line with the relevant literature (Asher et al. 2001; Easteal and Easteal 1992; Forgaty and Brown 2002). The first section was designed to obtain socio-demographic characteristics of the respondents. The second section, consisting of 18 items, assessed knowledge of the respondents about what constitutes VAW. Participants rated the 18 items on a five-point Likert scale—strongly disagree (0), disagree (2.5), uncertain (5.0), agree (7.5) and strongly agree (10). The third section, consisting of 20 items, assessed the attitudes/practices of the respondents towards women experiencing violence. The rating was on a five-point Likert scale as well. However, the scoring of attitudes/practices supporting VAW (last eight items) was reversed i.e., strongly disagree (10) to strongly agree (0).

The data were analyzed using the Statistical Package for Social Sciences (SPSS Version 11.01). For each of the phrases listed, the frequency of responses of the respondents to each rating scale was determined, as were the mean scores with their standard deviations. Overall mean scores were also determined according to profession, gender, marital status, and length of service. Student t-tests were applied to determine the significance of the social variables while AOV was used for the length of the service. The level of significance was set at p-value ≤0.05.

The dimensionality of the 20-item measure was analyzed using a varimax rotated principal component analysis. Three criteria were used to determine the number of factors to rotate: the a priori hypothesis that the measure was one-dimensional, the Scree test, and the interpretability of the test solution. Item analyses were further conducted on the 20 items after the factor analysis. Initially, each item was correlated with its own scale (with the item removed) and with other factors. Assessment of the convergent and discriminant validity of newly-derived factors was made by correlating each item with its own scale (with item removed) and with the other scales. Coefficient alphas were computed to obtain internal consistency estimates of reliability for these factors.

Results

One hundred and thirty-four (134) health workers participated in the survey. Seventy six (56.7%) were nurses and 58 (43.3%) respondents were medical doctors. The age range of the respondents was between 20 to 58 years with a mean of 38.11 (±9.13) years. There were 73 (54.5%) females and 61 (45.5%) males. The majority of the workers (60.4%) had put in more than 10 years of professional practice. Respondents were predominantly Christians (96.3%). Ninety-two (68.7%) were married. Most (83.6%) of the respondents were Yorubas, the predominant tribe in the South-Western Nigeria where the study was conducted (Table 1).

Table 1 Frequency distribution of the respondents (health professionals in OAUTHC, Ile-Ife, Nigeria) by socio-demographic characteristics
Full size table

As Table 2 shows, wife battering was the most recognized form of VAW with a mean score of 8.94 ± 2.28 followed by verbal abuse (7.63 ± 2.91). Sex selective abortion was the least recognized form with a mean score of 6.19 ± 4.12. The respondents had adequate knowledge (mean score of 7 and above) in 11 out of the 18 factors (61.1%). It was noted that female genital mutilation, forced contraception, and lack of right of women to determine when to get pregnant were the lowest recognized factors. The attitudes/practice regarding persons experiencing IPV was generally poor with only five out of the 12 listed items showing mean scores greater than 7 (Table 3). The most positive attitude was in respect of the education of the trainee health professionals to regard IPV as a medical problem (8.75 ± 2.52). Most doctors would not encourage the victims of IPV to leave the violent situation (mean score of 3.9 ± 3.62), and they would not want to confront the victim if battering is suspected when the person affected is trying to conceal the fact.

Table 2 Frequency distribution of health professionals in OAUTHC, Ile-Ife, Nigeria by their knowledge of what constitutes violence against women
Full size table
Table 3 Frequency distribution of health professionals in OAUTHC, Ile-Ife by their attitudes in supporting selected beliefs/practices towards victims of IPV
Full size table

In Table 4, the extent of opposition to some selected attitudes is shown. Respondents (76%) were mostly averse to the idea that a husband should hit his wife even when he has been provoked; while the majority was also of the view that wife battering is caused by the victim’s psychological problem.

Table 4 Frequency distribution of health professionals in OAUTHC, Ile-Ife by their attitudes in opposing selected beliefs/practices towards victims of IPV
Full size table

It was only in 42% of instances that the health workers would take affirmative decision in favor of the patients as Table 3 has shown (only in five out 12 instances was the mean score up to 7 and above). And in only one out eight (12.5%) instances that the respondents took the right stand against what was obviously a wrong attitude to management of victims of IPV.

Table 5 shows the comparative analyses of total mean scores for different groups of respondent. There was no statistically significant difference between the knowledge, attitudes, and practices of the health professionals with regard to their gender, marital status, length of service, and whether they were doctors or nurses.

Table 5 Comparative analyses between the socio-demographic characteristics and the scores on knowledge about IPV among the health professionals (n = 134)
Full size table

The scree plot indicated that our initial hypothesis of one-dimensionality was incorrect. Consequently, using the varimax rotation procedure, factor analysis yielded three interpretable factors (Table 6) explaining 35.3% of the total variance. Factor one accounted for 16.0% of the item variance, factor two 11.2% of the item variance, and factor three 8.1% of the variance. From initial item analysis of each of these three factors, items 4, 15, and 16 were removed from factor one; items 5 and 7 from factor two and item 21 from factor three (Table 6). Item 20 was found to be more highly correlated with other items in factor one and therefore, was put in factor one. In support of the measure’s convergent and discriminant validity, all items were consistently more highly correlated with their own scale than with the other scales (Table 6). Coefficient alphas for factors one, two and three were 0.66, 0.74, and 0.45 respectively. Internal consistency reliability coefficient (Cronbach’s alpha) for the total scale was 0.62. The factors, loadings, eigenvalues, and the internal consistency coefficients of each of the factors are shown in Table 6.

Table 6 Factor analysis and item correlations of attitudes/practices towards victims of IPV among the health professionals
Full size table

Discussion

Many studies have been reported in Nigeria on VAW and IPV (Odimegwu and Okengbo 2003; Odujinrin 1993; Ogunjuyigbe et al. 2005). However, those focusing on health professional’s knowledge, attitude, and practices as related to their competence in supporting people affected by IPV, are rare. This study provides information relating to the ability of health professionals to respond to the need of IPV victims in the Nigerian environment.

Our results showed that nurses and doctors in our tertiary health institution were generally informed about what constitutes VAW/IPV as the mean scores of the extent of agreement with the list of different forms of VAW among the respondents ranged between 6.19 to 8.94. This indicates that at least about 62% of them will recognize any of the features of VAW most of the time. However, when it comes to adequacy of the knowledge, only 11 out of 18 (61.1%) listed characteristic features of VAW that were adequately (mean score of 7.0 and above) recognized. In general, this is a better performance when compared with the ability of the same sample of the professional to recognize the indicators or the occult manifestations of IPV in a previous study (Adeyemi et al. 2005.) where only 41% of them were observed to have adequate knowledge. This observation implies that the respondents in the study find it much easier to diagnose VAW when the information is obtained directly through the client’s history or physical examination, rather than when the victims present with some somatic or psychological manifestation of the hidden problem.

Our results are consistent with reports in the literature that wife battering is the most common and classical manifestation of VAW. Verbal abuse was found to rank next in importance to IPV. Verbal abuse includes exchanges of words with threats or insinuations sufficient to cause the person involved psychological disturbances like weeping, sadness, loss of appetite, or insomnia. What constitutes verbal abuse is culturally determined, and varies across environments. The relatively low rating of sex selective abortion as a form of VAW may have to do with the fact that the level of sophistication in pre-natal diagnosis of sex is generally low in the study setting as facilities are scarce. This may also be a very critical issue because of the pronatalist nature of the Yoruba tribe who place high premium on childbearing regardless of the sex of the child although there may still be preference for one to have both sexes.

This present study shows that degree of knowledge does not necessarily translate to the correct attitude. When it comes to taking affirmative decision in favor of women affected by IPV, it was observed that it was in five out of 12 instances (41.7%) that the performance of the respondents was adequate (i.e., mean score of 7.0 and above). The attitude of the health professionals was even worse with respect to taking a stand or an opposing view against a practice unfavorable to the victim of IPV. It was only in one out of eight instances (12.5%) that 71% of respondents took the right stand. This was when they disagreed with the fact that there is nothing wrong for the husband to hit his wife if he has really been provoked (mean score 7.30 ± 3.36). This attitude lends credence to the popular proverb in this environment that “patience is the main ingredient for being the husband to a woman.”

Noting that our study was conducted in a tertiary health institution, our result might not be representative of what the situation would be among doctors and nurses at lower levels of our health system. This is because health professionals in tertiary institutions in our environment generally have better exposure to educational opportunities and information technology than their counterparts working at primary and secondary health care facilities, where most clients who need health attention seek care. There is a need to conduct similar studies at the other levels of health care in the country.

The findings from this study have implications for the reproductive health of Nigerian women. To meet the challenge of IPV, and its possible contribution to women’s health morbidity and mortality, there is the need for doctors and nurses to be trained in the routine screening for VAW, and to be able to provide appropriate care to victims. As most of the victims will not directly give information about experiences of abuse and violence, the need for a high index of suspicion on the part of the health workers is important. Health care providers are strategic in the advocacy, prevention, and effective management of IPV. Thus, their capacity and competence need to be improved to respond holistically to the challenges of primary, secondary, and tertiary prevention of IPV (Berenson et al. 1994; Friend 1999). In addition to in-service training of existing health workers, it has become imperative to incorporate the issue of violence prevention, detection, and management into the curricula of health professionals from the undergraduate level up to the post-graduate level. Findings from this study could contribute to the development of appropriate training tools that would address relevant areas of knowledge, attitude, and practices on the part of health workers in Nigeria.