Primary health care physicians' approach toward domestic violence in Tehran, Iran.

BACKGROUND: Primary health care physicians (PHCPs) are the first in the clinic to detect and help victims of intimate partner violence (IPV). Therefore, their attitude and practice toward domestic violence (DV) are important to manage this problem. The aim of current study was to compare the behavior and attitude of PHCPs about DV versus other health risk factors in Tehran, Iran.
METHODS: A convenience sample of 220 PHCPs was evaluated. The study was carried out in April 2012. Two self-administered questionnaires were used to identify physicians' beliefs and behaviors on screening and intervention of DV and other health risk factors. All analyses were performed using SPSS version 18.0 (SPSS, Inc. Chicago, IL).
RESULTS: One hundred and ninety eight questionnaires were analyzed. PHCPs' mean age was 39.06 (±7.5) years. Participants were just reported 10% screening of regular patients for DV compared with 29% to 48% for other health risk factors. Mean age of PHCPs was not associated with their approach toward the DV. Compared to male physicians, females spared more time for DV victims. Major of physicians (96%) believed that DV is not a private problem and is something that needs to be addressed cautiously.
CONCLUSION: The results of this study indicated that DV screening occurs less than that of other health risk factors. Attitude of majority of PHCPs was positive for addressing this problem.

Introduction

The Declaration on the Elimination of Violence against Women (1993) defines violence against women as “any act of gender-based violence that results in, or is likely to result in physical, sexual, or mental harm or suffering to women, including threats of such acts, coercion or arbitrary deprivation of liberty, whether occurring in public or private life” (1).

It has been shown between 10% to 50% of women were abused at some stage by an intimate partner worldwide (2,3). Domestic violence (DV) is a major source of morbidity among women, with an estimated 36% physical intimate partner violence (IPV) in women aged 18 and older in Tehran, Iran (4). Prevalence of physical violence during pregnancy was reported as 10.7% in a study carried out among inpatient women in Tehran, Iran (5).

Abused women have poor mental and physical health and use more health care services than non-abused women (6-8). Women who have abused by their partner suffer a wide range of problems with a profound impact on their health (9). It has been shown that exposure to violence can be a risk factor for chronic pain syndromes, gastrointestinal disorders, and irritable bowel syndrome in victims (10-12).

Primary health care physicians (PHCPs) are in a unique position to detect and help victims of intimate partner violence (IPV). PHCPs are the first in the clinic to manage the problem of victims of DV. AbuTaleb et al. in a study carried out in Kuwait showed that the knowledge of PHCPs about the prevalence of DV is poor (13). They discussed that shortage of knowledge about the prevalence of DV could be considered as a barrier to manage these victims effectively. It has been shown just 7-15% of victims of DV are detected in clinical practice and 60% to 90% managed inappropriately (14). An important barrier to adequately manage victims of DV is lack of training and specific education on IPV among health professionals (15). Fikree et al. showed about 37% of obstetricians did not know how to ask about DV during their practice (16).

In Iran, according to the prevalence of DV and importance of prevention of this problem, detection of victims is necessary. The role of PHCP in detection and management of victims of DV is significant. To identify and efficiently manage the problem, PHCPs should receive required knowledge and training. The knowledge, attitude and practice of physicians about DV has not been investigated in this setting. The aim of current study was to compare the attitude and behavior of PHCPs about DV versus other health risk factors in Tehran, Iran.

Methods

Population and sample

The health care centers in Iran are divided to urban and rural ones. In these centers the primary health care is provided for clients. General practitioners are the physicians who practice in these centers. Using a cross-sectional design in April 2012 we offered two questionnaires to a convenience sample of 220 practitioners. Our subjects were divided into two groups. The first group was selected from PHCPs in health care centers. In each center a research assistant offered the questionnaires to the physicians and then collected the returned ones. We selected all urban and rural health centers under supervision of Shahid Beheshti University of Medical Sciences located in North, North East and West of Tehran. The second group of samples included PHCPs who had participated in annual continuing medical education specific for general practitioners. We asked about the work place in each questionnaire, therefore duplicate data were excluded. The time to filling out questionnaire was about 10 minutes. Participation in the study was voluntary. We excluded physicians who did not work as a PHCP. Physicians were informed that their answers would be kept confidential. Totally, 205 questionnaires were returned and finally 198 completed questionnaires were analyzed (response rate 90%). The Ethics Committee of school of Medicine in Iran University of Medical Sciences approved the study.

Questionnaire and items

We used a self-administered questionnaire about the physicians’ beliefs and behaviors on screening and intervention for patients’ health risk factors conducted by Gerbert et al. (17). The questionnaire had four sections: 1) domestic violence, 2) drug use, 3) alcohol use, and 4) HIV/STD risks. Each section had 9 items framed in a five-point Likert scale from never to always. The tool evaluated the frequency of screening new and routine patients for each risk. The assessment about each risk included frequency of intervention behaviors, the number of patients with whom they had consulted each risk during past three months, and the length of time in minutes spent consulting each risk. In this questionnaire there were some items about the attitudes of physicians on screening and interventions according knowledge, effectiveness, and available resources.

The original questionnaire was translated into Persian by an independent health professional fluent in both English and Persian. Another translator who had no knowledge of the original instrument back translated the questionnaire into English. Final adjustments were made following reviews by all investigators. The final version was completed by 15 physicians and results showed that all they easily understood the items. The Cronbach’s alpha for violence, alcohol, drug, and HIV/STD sections were 0.85. 0.83, 0.89, and 0.89, respectively.

We also used a questionnaire to determine the attitude of PHCPs about DV. This questionnaire appeared in an unpublished work in Iran and included 15 items to assess attitude of population about the DV. Each item was on a five point Likert scale from strongly agree to strongly disagree. Evaluating attitude of PHCPs was an adjunct objective of the current study. We presented data as agree/disagree for simplicity of interpretation of responses. In this questionnaire, DV was presented as IPV against women specifically. Cronbach’s alpha for this questionnaire was 0.70.

Statistical analysis

All analyses were performed using SPSS version 18.0 (SPSS, Inc. Chicago, IL) in 2013. Cronbach’s alpha was used to assess reliability of questionnaires. We used table of frequencies for presenting categorical variables and mean and standard deviation (SD) to describe numeric variables. Chi-square test was used to compare relative frequencies across the physicians’ beliefs and behaviors on screening and intervention for health risk factors. Spearman rho correlation coefficient was used to assess association between screenings of DV with other health risk factors. Also, we used logistic regression to determine independent effects of screening of other health risk factors on DV as a dependent variable. Result of model reported using Odds ratio (OR) and 95% confidence interval. Significant level was set at 0.05.

Results

General findings

One hundred and ninety eight questionnaires were entered in final analysis. Demographic characteristics and other measured variables are illustrated in Table 1. The age of subjects ranged from 26 to 63 years with mean and SD of 39.06 and 7.5, respectively. Out of 198 PHCPs, 52% were female. Majority of them (80.9%) were married and 40% were worked in governmental centers. Fifty five percent had more than ten years of professional experience. The range of professional experience was from one to 30 years. About one fifth of physicians were graduated from Tehran University of medical Sciences (data was not shown).

Table 1

Demographic and general characteristics of PHCPs (n=198*)

Variable	Number	percent
Age≤ 3030 – 50≥ 50	2715412	1479.86.2
SexFemaleMale	10393	52.647.4
Marital statusMarriedSingleOther	157352	80.918.01.0
Profession experience (years)Less than 10More than 10	86106	44.855.2
WorkplaceRural health centersUrban health centersPrivate clinicHospital clinicOther governmental clinics	4420665216	221033268

*There was missing values for some variables

Physicians’ characteristics and screening for DV

There was an almost significant association between mean of professional experience of physicians with screening new patients for DV. Physicians with professional experience of 14.1 (± 1.7) years were more likely to screen DV than those with 11.6 (± 0.49) years which the relation was partially significant (p= 0.07). The association for screening of regular patients about DV was also significant. Physicians with professional experience of 15 (± 1.6) years were more likely to screen DV in regular patients versus physicians with 11.5 (±0.49) (p= 0.03). We could not reveal any other significant association between professional experience of physicians and other practice of physicians about DV. Mean age of PHCPs was not associated with their approach toward domestic violence.

Female PHCPs were more likely to document DV (21.3%) than males (9.8%) significantly (p= 0.03). Providing information about shelters and other services to victims of domestic violence was more in female physicians (40%) versus male physicians (26%) significantly (p= 0.04).

Blue: New patients, Red: Regular

We asked about the time that physicians provided for consult with victims of DV. It was shown that 75.3% of female physicians provided more than 10 minutes for victims of DV versus 57.3% of male physicians. This difference was statistically significant (p= 0.009).

Physicians in private health centers were more likely to screen new patients for DV (13.8%) than physicians in governmental health centers (5.1%) significantly (p=0.04).

Seventy three percent of physicians who worked in private health centers provided more than 10 minutes for victims of DV versus 57.9% of physicians who worked in governmental health centers. This difference was statistically significant (p= 0.03).

Practice for DV versus other health risk factors

Reporting screening of regular patients about DV was fewer than other health risk factors in our samples. It was 10% versus 48%, 29% and 30% for drug use, alcohol and HIV/STD risk factors (Fig. 1). These differences were statistically significant (p<0.01). Documentation of DV was less likely than other health risk factors (Table 1). Providing counseling to increase patient safety was more for HIV/STD risk compared to other health risk factors (60% versus 23% to 48%). Also providing information about supporting services for HIV/STD risk (55%) and drug use (54%) was more than other health risk factors (22% to 32%). Arrange for follow-up visits or calls and refer to another resource were more for HIV/STD risk than other health risk factors (Table 2).

Fig. 1

Table 2

Frequency of physicians practice about DV and other health risk factors for regular patients (n=198)

Item	DV		Drug		Alcohol		HIV/STD		p
	AO*	N/R/S†	AO	N/R/S	AO	N/R/S	AO	N/R/S
How often do you ask regular patients about …….risks?	20 (10)	178 (90)	95 (48)	103 (52)	57 (29)	141 (71)	59 (30)	139 (70)	< 0.01
When problem is first identified, how often do you do eachof the following?
- Document the problem	30 15)	168 (85)	67 (34)	131 (66)	49 (25)	149 (75)	105 (53)	93 (47)	< 0.01
- Provide counseling to increase patient safety	85 (43)	113 (57)	95 (48)	103 (52)	45 (23)	153 (77)	118 (60)	80 (40)	< 0.01
- Provide information about supporting services	63 (32)	135 (68)	108 (54)	90 (46)	44 (22)	154 (78)	110 (55)	88 (45)	< 0.03
- Arrange for follow-up visits or calls	63 (32)	135 (68)	83 (42)	119 (58)	47 (24)	151 (42)	98 (49)	100 (51)	< 0.05
- Refer to another resource	79 (40)	119 (60)	110 (55)	88 (45)	43 (22)	155 (78)	124 (63)	74 (37)	< 0.01

*:Always/often; †: Never/rarely/sometimes

The physicians’ trend toward asking about DV versus other health risk factors for new and regular patients

Blue: New patients, Red: Regular

Frequency of more than ten patients consulted over the past three months by physicians about DV was 12.5% versus 46.6%, 25.5%, and 22% about drug use, alcohol and HIV/STD risks respectively. We showed that PHCPs spent more than ten minutes to counsel about drug use (67.4%), DV (67%); alcohol (36.2%), and HIV/STD (49.7%) risks. Eighty four percent of physicians reported they ask about occurrence of DV in suspected patients. This proportion was 89% for other health risk factors.

There was a positive association between screening of regular patients for DV with drug use (Spearman rho= 0.22, p= 0.02), alcohol (Spearman rho= 0.25, p= 0.001), and HIV/AIDS (Spearman rho= 0.17, p= 0.01) risks.

In regression analysis, we showed that just asking about alcohol risk has a positive, and significant independent effect on screening of DV, adjusted physicians’ age, sex, and professional experience (OR= 3.5; CI95%: 1.02 – 11.9).

Attitude about DV factors

Table 3 illustrates the attitude of physicians about DV. Almost all physicians believed that DV against women is a critical public health problem (97%). Eighty three percent of respondents believed that those men who commit violence against women has some kind of mental problem. Almost all of physicians (96%) did not believe that DV is a private problem and it is better do not ask about it.

Table 3

Attitude of PHCPs about DV (n=198)

Item	Agreen (%)	Disagreen (%)
DV against women is an important public health problem	192 (97)	6 (3)
Men who do violence against women have a mental problem	165 (83)	33 (17)
The cause of majority of mental health problems of women is DV	136 (69)	62 (31)
The rate of DV is low in religious families	54 (27)	144 (73)
DV often interrupts during pregnancy	54 (27)	144 (73)
Bad behaviors and nags of women are causes of violence against them	92 (46)	106 (54)
The cause of majority of physical problems of women is DV	101 (51)	97 (49)
I do not have a good feeling to ask my patients about DV	51 (26)	147 (74)
DV is a private problem and it is better do not ask about it	8 (4)	190 (96)
Asking about DV put my patients in a bad situation	15 (7)	183 (93)
Women should tolerate violence due to keep family’s mystery	17 (8)	181 (92)
DV has not an important effect on the family’s health during a long time	10 (5)	188 (95)
If DV is severe, patients tell about it without asking	50 (25)	148 (75)
DV is an ordinary way in case of uncontrolled problems for men	62 (31)	136 (69)
DV is an acceptable act in majority of cultures	72 (36)	126 (64)

About 95% of physicians were not agree about the DV has not an important effect on the family’s health during a long time” item. About 46% of physicians reported that bad behaviors of women is a cause of DV. This rate was 37% and 60% for female and male respondents with statistically significant difference (p= 0.0001). About 31% of married physicians and 51% of single physicians were agree about acceptability of DV in majority of cultures (p= 0.02).

Physicians who were agree about the belief that “If DV is severe, patients tell about it without asking”, screened less patients on DV than those disagreed about it (8% versus 12%, p= 0.50). Men (36.4%) were more agree with this belief than women (16.8%) significantly (p= 0.002). Also, physicians who reported they had not a good feeling to ask patients about DV, did less screening of patients (4%) for DV than others (12.5%) who had not this belief (p= 0.08).

Discussion

The aim of current study was to determine the approach of PHCPs about DV compared to other health risk factors. The response rate was 90 percent, and 198 questionnaires were entered to final analysis. This rate of responsiveness is perfect compared to other similar studies on physicians. This rate was reported from 20 % to 63% in similar reports (13,18-20). This could be explained by novelty of subject for physicians and their interest to know more about this problem. Recently, screening of this public health problem in primary health care setting has been offered by some experts and investigators in Iran and it is under assessment by health system policy makers.

Majority of physicians who participated in current study were between 30 to 50 years old. It seems that our participants were younger than physicians in studies of AbuTaleb et al and Gerbert et al (13,17). It is due to selection of majority of physicians who had participated in annual continuing medical education specific for general practitioners and from primary health care centers. Most of physicians in these two settings are in middle ages. The gender distribution of physicians and their work experience seemed almost acceptable.

The main finding of current study was revealed difference between physicians approach and attitude on screening DV and other health risk factors. Participants were just reported 10% screening of regular patients for DV compared with 29% to 48% for other health risk factors. Also, documentation of DV was less than other health risk factors. Although these findings are similar to Gerbert et al study (13), but screening and documentation of all health risk factors included for DV was more in their study compared to current one. They reported screening of DV, alcohol and HIV/STD, 19%, 90% and 47% respectively. One explanation for less reported screening of alcohol and HIV/STD in current study could be the low rate of HIV (of course not STD), stigma and alcohol use in our population.

Nevertheless, the difference between attitude and practice of PHCPs regarding DV compared to other health risk factors could be explained by different content of these public health problems. DV for a long time considered as a social problem. Therefore, before focusing on social determinants of health (SDH), this problem was considered out of interesting of physicians. Todays, PHCPs should address the SDH as similar as other determinants of health. They could manage the SDH with advocacy and participation of other relevant organizations.

DV which seems to be a private family problem, carries a social stigma for a long time, but majority of participants in current study reported their tendency to ask about it for the sake of patients’ safety. Although, the majority of physicians spent little time to manage this problem.

Other interesting finding in current study was female physicians more than males documented incidence of DV. Also, providing information about supporting services to victims were reported more in female compared to male physicians. This finding could be explained by female physicians’ sympathy to the same gender, or by their deep sense of responsibility toward victims of DV.

Regarding of attitude of physicians about DV, our findings were noticeable. Almost all participants reported that DV is an important public health problem. It shows they knew and believed in importance of DV and its effect on health.

Bad behaviors and nags of women as a trigger of occurrence of violence against women were reported by 46% of participants. In a study conducted by Shearer et al, 32% of male and 47% of female chiropractors were strongly agree that victims have triggered something to cause abuse (21). It is an expecting result. Female physicians do not agree with fault of women as a cause of DV as often as male physicians. However, this factor should be considered as a cause of DV. Some behavioral interventions focused on behavior of women to control DV. We experienced one of this interventions with implementation of life skills workshop for victims of DV and we could show the decreasing rate of DV after intervention (data was not reported).

Majority of participants (74%) in current study did not have a bad feeling to ask their patients about DV. It is a hopeful finding of current study. In a study by Fikree et al (16), about 72% of Pakistani obstetricians were not afraid of patients’ offending when asking about DV.

The other interesting finding was the association between beliefs and practices of physicians about screening of DV. Physicians who believed asking about DV is not necessary as a routine, did less screening of patients about this risk. Also, having bad feeling asking about DV was in a negative association with screening of DV. Although, these two associations were not statistically significant, but the difference between proportions were noticeable and had implicit message.

The strength point of current study was the response rate and the study innovation in examining PHCPs in Iran. Regarding the results of this study we could understand the attitude and approach of the PHCPs about DV. Our findings become useful for health policy makers while they think of PHCPs collaboration in DV reducing programs. Besides, our findings about PHCPs’ approach toward alcohol, drug, and HIV/STD risk factors are considerable.

The present study had some limitations. Our findings could not be generalized to the entire population of PHCPs due to the used convenience sampling method. Although, selection of some physicians from participants of annual continuing medical education specific for general practitioners could dilute this limitation relatively. Moreover, some factors related to physicians’ behavior about health risks were not taken into account in current study, but suggested to be addressed in future studies.

Conclusion

The results of this study indicated that DV screening rate is less than other health risk factors. We also showed documentation of DV in primary health care setting is less than other health risk factors. Attitude of majority of PHCPs were positive for addressing this problem and most of them confirmed importance of DV as a public health problem. Regarding our results, providing suitable educational programs for PHCPs and integrating screening of this problem in PHC could be considered.

Primary health care setting could be the first line to manage and control DV. PHCPs majorly showed a positive attitude about control and managing of DV. Although, the policy makers should provide incentives to encourage them to manage and screen this important public health problem. PHCPs need more time to manage this problem, therefore their activities in health care setting should be revised.

Conflict of Interest

We declare that we have neither financial disclosure nor conflict of interest in this manuscript.