Health-seeking behaviour of women reporting symptoms of reproductive tract infections.

Durr-E-Nayab

A woman's access to health care, in physical, social, and psychological contexts, depends on her health beliefs and her socio-economic and demographic background. As in most developing countries, the health system in Pakistan is a combination of modern and traditional medicine, and the nature of care sought again depends on the individual's health beliefs and background characteristics. This paper thus not only focuses on whether women seek help or not when sick, but also on the differentials that exist in the health-seeking behaviour among women with different backgrounds. It finds that less than half the women reporting any symptom related to reproductive tract infections seek help, while for some symptoms the proportion seeking help goes down to a mere one-fifth. The decision to seek help depends on a woman's educational and economic status, the extent to which she is worried about the symptom, duration of experiencing the symptom, and interspousal communication about the symptom. Lack of finances to access any health service and considering the symptom as something common not needing attention are the two main reasons for not seeking help. The choice of the health-provider consulted for a symptom is linked to the perceived cause of the symptom, but allopathic doctors are preferred by the majority of women seeking health care.

**********

In patriarchal societies, like that of Pakistan, women are considered responsible for taking care of the health of the family members, young or old. And as very aptly put by Kabira, et al., the "paradox of entrusting the woman with the responsibility of health and at the same time denying her the opportunities to influence policies remains a major obstacle" (1997, p. 23). In a broader perspective, these policies are not only those made at the national level but also the ones applicable at the personal and household level. Women are usually subservient to the decisions and authority of males in the household in matters relating health, despite being assigned the role of health carets. The issue is further complicated if the health problem is related to anything having sexual connotations in regions where it is deemed a taboo topic.

Evidence from the South Asian region shows that women's traditionally determined roles could greatly undermine their health, including reproductive health, and affect their use of health services, with majority not seeking any proper care for their problems. The present case study looks into the health seeking behaviour of women reporting symptoms associated with reproductive tract infections (RTIs) in urban Pakistan. RTIs in many cases can be asymptomatic, but at the same time reporting of any symptom associated with reproductive tract infections does not necessarily mean presence of infection. The health seeking behaviour of women would thus be analysed regardless of the medical accuracy or otherwise of the self-reported symptoms, because it is their perceived ill health that determines their behaviour, not the actual presence or absence of disease.

A woman's health seeking behaviour is a result of syncretism between her health beliefs and her socio-economic and demographic background, which affects factors defining her access to healthcare, in physical, social and psychological context. In Pakistan, as in most developing countries, the health system is a combination of modern and traditional medicine, and the nature of care sought depends on the health beliefs and socio-economic and demographic background of the concerned person. Thus, our interest here is not only in whether women seek help or not, but also in the differentials that exist in health seeking among women with different backgrounds.

The present paper thus investigates:

1. Whether women reporting symptoms associated with RTIs seek help or not, and reasons for those not seeking help.

2. The nature of treatment they seek and how it varies by different symptoms.

3. Differentials in seeking help for RTI related symptoms by women having different socio-economic and demographic characteristics.

METHODOLOGY

Locale and Sample

As the name of the study, Rawalpindi Reproductive Tract Infections Study 2001-2002 (RRTIS 2001-2002), suggests, the study was conducted in the city of Rawalpindi, a major urban area of Pakistan. Investigating the behavioural factors affecting women's health seeking process was one of the objectives of the study, and these factors could be better understood at a locale where absence or inaccessibility of health services, as in most rural areas of Pakistan, is not the main factor affecting the decision to seek treatment. Thus, conducting the study in a major urban area with sufficient medical facilities removes this reason for not seeking help in case of illness.

Currently married women aged 15-49 years, having their husbands living with them, comprised the sample. Sexual activity, especially for women, in the country usually take place within marriage so being currently married was of importance to be included in the sample because if women were not in a current union they were unlikely to be sexually active or using contraceptives, which are factors linked to having RTIs and thus of interest to this study. Similar reasons led to the decision to include only those women whose husbands were living with them.

Using Federal Bureau of Statistics' primary sampling units (PSUs) of Rawalpindi, a representative sample of 500 households was drawn based on the economic status of the households. It was premised that differences in economic background bring with them differentials in factors that affect the health-seeking behaviour of women, including factors like education level, health-seeking behaviour, health perceptions, etc. Twenty-five PSUs were randomly selected, covering the economic composition of the city. From these sampling units, 20 households each were selected randomly to give a sample total of 500 households. Of these five hundred sampled households, 490 households were interviewed successfully. Since some households had more than one eligible woman, the final sample comprised 508 eligible women.

Survey Tool

Drawing on the relevant literature, including questionnaires for demographic and health surveys, a mainly open-ended questionnaire was developed for the study. An open-ended approach was preferred because a lot of information can be left unrecorded or concealed if the responses are given in the confines of narrow options provided as answers, as in a standardised structured questionnaire. However, pre-coded questions were asked in instances where the responses could not in any case be other than the options given. On reaching the field and doing a pilot study some problems were found with the questionnaire, and then some further changes were suggested by the doctors who were consulted in Rawalpindi. The questionnaire was amended keeping these suggestions in consideration. Since doctors deal with patients on an everyday basis, their help was also taken to translate the questionnaire in the Urdu language, using terms that were easy to understand for the respondents and carried the meaning that they were supposed to convey. The final version of the questionnaire inquired about:

(a) Social and economic characteristics: including age, sex, marital status, education, occupation, income type of family, age at marriage, and water and sanitation facilities at house.

(b) Obstetric and gynaecological history: number of pregnancies and their outcome, any complications faced during pregnancy or delivery, person attending the delivery and any health problem with the newborn. Being associated with RTIs, questions regarding infertility were also part of the questionnaire.

(c) Contraceptive history: including whether or not any method was ever or currently being used, and any problems experienced while using it.

(d) Hygiene practices: general state of hygiene, like number of baths taken per week, but specifically the menstrual hygiene of women.

(e) Health status: any health problem being faced by women and whether they were seeking help for it and from whom.

(f) Knowledge regarding RTIs: including their perceptions of the causes, consequences and ways of treating the infections.

(g) Experience of RTI symptoms: their personal experiences regarding RTIs at the time of the survey in particular, and over their lifetime in general. To give women a time reference to report for their current experiences a period of one week preceding the survey was given. However, for two symptoms, those of menstrual irregularity and pain during menstruation, a period of three months preceding the survey was given to women for reporting their current experiences. This decision was taken due to the cyclic nature of the process and because a period of three months is taken to medically define a menstrual problem as a disease. Along with reporting their experiences, women were asked about the severity and duration of these symptoms and the way these symptoms affected their daily routine.

(h) Health seeking behaviour: if women were experiencing any RTI related symptom were they seeking help, if yes, what type of treatment and from whom, and their impressions about the treatment provided to them. If they do not seek help what are the actual or perceived barriers to it.

(i) Inter-spousal communication: whether women discussed about their RTI related experiences with their husbands.

(j) Autonomy: including her say in decision-making regarding different matters of the household, control over household income and independence to leave the four walls of her house without seeking permission. Questions were also asked about any abuse the women were a target of.

Measurement of Self-reported Symptoms

To measure the self-reported prevalence of RTIs, women were asked questions based on the symptoms associated with different reproductive tract infections. These symptoms were:

(1) Abnormal vaginal discharge: discharge that was not usual to the woman in colour, texture, odour or consistency, and if it caused an itch in the genitals.

(2) Lower abdominal pain: nature, duration and severity of pain in the lower abdomen.

(3) Menstrual irregularity: changes in duration, quantity, cyclicality or consistency of blood during menstruation.

(4) Dysmenorrhoea: pain during menstruation.

(5) Sores and ulcers on the genitals.

(6) Dysuria: painful urination or burning sensation during urination.

(7) Dyspareunia: painful intercourse, bleeding or bad odour after intercourse.

(8) Lower backache: only if it was reported accompanying any of the above symptoms.

Women were asked if they were experiencing these symptoms at the time of the interview, or in the last three months regarding the two menstruations related symptoms. This constituted the measure of self-reported prevalence of RTI related symptoms. Some of these symptoms could be present due to factors other than RTIs but since they are associated with one or more RTIs they were included in the questionnaire. The possible association of reported symptoms with different reproductive tract infections are shown in Table 1.

RESULTS

Women's health seeking behaviour differed according to the symptom being reported. As Table 2 shows, the proportion seeking help varies from one fifth of the women reporting a particular symptom, as in the case of dyspareunia, to one half doing the same, as in the case of dysuria. The proportion seeking help is lowest for women reporting the symptoms of dyspareunia (21 percent) and sores/ulcers on genitals (25 percent), which could be due to the sense of shame associated with these symptoms.

It is easier to understand these figures if we look into the reasons stated by women for not seeking help for each symptom. As Table 3 shows, lack of resources, and taking the symptom being experienced as part of womanhood are the two main reasons for not seeking help for all the symptoms. Women do not feel the need to seek help because for all these RTI related symptoms they feel that, "it is a normal thing to happen to women". Interestingly, the proportion feeling so is highest for dyspareunia (51 percent). It is not only lack of knowledge regarding RTIs that lead women to maintain the beliefs they have but also the socialisation process they go through. Another stated reason, that of "did not find time", which features for most of the symptoms, also implies lack of importance given by women to these symptoms. Time could be found if a thing is deemed important enough to give attention to. The "lack of resources" reason along with being a result of a general poor economic condition of the household could also be due to factors associated with money allocation within the household. Spending money on alleviating the health of women with these symptoms might not be a priority in a situation of economic constraints.

Contraceptive use also features consistently as a reason perceived to be causing the experienced symptoms, thus not warranting any treatment (Table 3). Contrary to what could be expected in the given circumstances, constraints due to lack of permission by husband or in-laws do not appear to be among the major reasons affecting woman's lack of action regarding her symptoms, while a feeling of shame, as expected, was among the major reasons for not seeking help for sores/ulcers on genitals (17 percent) and dyspareunia (28 percent). Dissatisfaction with prior treatments is also among the major reasons given by respondents for not seeking help for most of the symptoms, especially abnormal vaginal discharge (20 percent), which was a chronic problem for many women. The dissatisfaction stems from not just the failure of the treatment to help alleviate the problem but also the attitude of health providers during the course of consultation (Table 3).

It is interesting to note that self-medication, in the absence of seeking help from anyone else, is high for symptoms associated with any kind of pain. As Table 3 shows, self-medication is highest among women experiencing dysmenorrhoea (27 percent), lower abdominal pain (14 percent) and lower backache (14 percent). The highest rate for self-medication in case of dysmenorrhoea is of especial importance because it shows a rather low rate for seeking help otherwise (Table 2), as against lower abdominal pain and lower backache for which health providers are consulted more often. Pain associated with menstruation is considered a routine thing by women, not motivating them enough to consult any health provider.

The factors affecting women's decision to seek help thus are a combination of their beliefs regarding the cause of the symptom, the perceived consequences of the symptom, and the constraints they face to seek help in situations where they decide to seek help but cannot. To elaborate this synergy, we can take the example of the health seeking behaviour of women experiencing abnormal vaginal discharge. The rate for seeking help was lower for women who thought weakness (37 percent), use of contraceptives (20 percent), or sex after marriage (zero percent) was the cause of their having the discharge, compared to those who thought they had a problem with their uterus (83 percent) or those who did not know why they were having the symptom (56 percent). The rate was also understandably low for women who considered this experience as a common thing (25 percent). Similarly, the perceived consequences, expressed as the reasons for worrying on having a symptom, affected women's health seeking behaviour. Women had a higher rate of seeking help for abnormal vaginal discharge if they thought that the symptom could lead to something more serious (77 percent), will cause infection in their uterus (60 percent), than if they thought that it can lead to weakness and backache (33 percent). These perceptions and fears accompanied by the constraints offered by their socioeconomic and demographic conditions define the course of action they take in response to the experienced symptom.

Choice of Healthcare Sought

Allopathic treatment is the most commonly reported choice of healthcare sought by women in the study sample, followed by the traditional doctors, (1) for all the RTI related symptoms (Table 4). Not many women approach other health care providers, including nurses/lady health visitors/family welfare workers or the traditional birth attendants/dais to get relief from the symptoms they are experiencing. An even smaller proportion consults their husbands who get them medicines, without any formal direct consultation of women with any health provider (Table 4). It is however to be pointed out that those considered "doctors" by women could be in some cases practitioners not having any professionally qualified allopathic training.

It is a common place happening in Pakistan, especially in rural areas but not absent in urban areas either, that persons having any exposure to medical procedures while working as a dispenser, compounder, nurse assistant, or an assistant in a pathological laboratory, after gaining experience for some time open clinics by themselves and dispense medical advice and prescribe medicines. This was evident from what one of the respondents said on being asked about what was she advised by the "doctor" she consulted for her problem. The advice she got was, "he told that I have 'heat' inside my body and gave me two kinds of tablets to eat". This advice shows a combination of treatment, finding a cause that lies in the domain of Hikmat (2) but giving allopathic medicines, which may or may not help. Receiving advice of similar was reported by many other respondents which raise doubt about the authenticity of the "allopathic" doctors consulted by women. If such instances were found in a study conducted in a major urban area, where some sort of a monitoring system exists to check such malpractices, it could be imagined how big the problem of lay practitioners and quacks could be in the country otherwise, especially in rural areas, and the impact it might be having on the health of the population.

Discounting the data for those having sores/ulcers on genitals as there are too few cases to base any valid inference on, the proportion seeking help from traditional healers is highest for those experiencing abnormal vaginal discharge, notwithstanding (Table 4). The choice of treatment could be linked to the perceived cause of having the symptom. The proportion of women perceiving reasons like weakness (41 percent) and body heat/hot food (8 percent) was highest for those reporting abnormal vaginal discharge, reasons that fall more in the realm of traditional medicine than modern medical practices. Several studies in developing countries have shown this link between the perceived cause of illness and the choice of the type of treatment sought [Lambert (1996); Erwin (1993); Pool (1987) and Colson (1971)].

Gould (1965) in his study in India found that traditional medicine was used for illnesses that were not considered incapacitating, and doctors were consulted in situations where the illness was deemed incapacitating. The choice of treatment in the present study also hints at a similar trend. The seriousness and gravity with which a symptom was perceived was reflected in the decision to seek or not seek help and in the nature of treatment sought. For example, dysuria was considered the most worrying symptom being experienced by women and as Table 4 shows, the proportion seeking help for it is higher than any other symptom (49 percent). An examination of the relation between socio-economic and demographic characteristics of women and their choice of the type of treatment sought shows that more educated women and those belonging to the upper economic group tend to opt for doctors slightly more frequently than the uneducated women and those belonging to the lower economic group, however the differences are not so significant to warrant any detailed discussion. More than the choice of health provider, it is the basic decision of whether to seek help or not that confronts women, reflected in the differences in their health seeking behaviour.

Differentials in Seeking Treatment

Indicators specific to each symptom show a more significant relation with women's decision to seek help, than the socio-economic, demographic and autonomy indicators (Table 5). For all the RTI related symptoms included in the study, duration of experiencing a symptom, worry related to that experience and inter-spousal communication about it are strongly associated to women's consulting someone for her condition. Other factors having strong relation include women's educational and economic status, and their control over household income, with all having a positive association with their ability to seek help for the symptom they are experiencing (Table 5).

Socio-economic Indicators

Women's age does not generally show a statistically significant relation with her decision to seek help but increasing age does show an increasing trend for most symptoms (Table 5). Women aged over 34 years have the highest rate of seeking help for symptoms of menstrual irregularity (48 percent), dysmenorrhoea (31 percent), dysuria (56 percent), and lower backache (45 percent). The incidence of sores and ulcers was too low to draw any significant inference, while the lower rate of seeking help for dyspareunia (11 percent) among women aged over 34 years could be because of the lack of desire for sexual intercourse. On the other hand, women in the youngest age group have the lowest rate of seeking help for almost all RTI related symptoms. This pattern supports the prevailing belief that women gain power and authority with age. One example in this regard is the increasing proportion of older women deciding by themselves to seek help when unwell. For women aged less than 25 years, the proportion is 34 percent, increasing to 59 percent for the 25-34 age group and 67 percent for those aged 35 and over. (4)

The impact that the initial years can have on a person's life is evident from the higher rate of seeking help by women having urban background, as compared to those having a rural background (Table 5). Although the relation was not always statistically significant, the same pattern, of urban background women seeking help more often, exists for health seeking for all the symptoms. On the other hand, no trend is found for the type of family structure the woman is part of and her health seeking behaviour across different RTI related symptoms. For some symptoms, those living in nuclear households have a higher rate of consulting a health provider while for others it is those living in the extended/joint family households that are more likely to seek help (Table 5). This finding dispels the belief that extended/joint family households are essentially an obstacle for women's decision-making and behaviour, including those related to health.

Increasing level of education, as expected, has a positive relation with women seeking help for the symptoms they report to be experiencing (Table 5). Women with more than ten years of education were more likely to consult a health provider for alleviating their symptoms than those who had never been to school. The relation is not statistically significant for most symptoms but there is a pattern of more educated women being more likely to seek health care, a trend more visible if we look into the differentials between women who have been to school compared to those who have never been to school, instead of the figures for their level of schooling (Table 5). Education, as noted by Chatterjee (1990), affects an individual's perception of health needs, knowledge of services and the ability to access them. He considers the recognition of need as a first step for the utilisation of health services. Experiences from the present study show that illiterate women, specially those belonging to the lower economic group, were even unable to use the available health facilities that were almost free to avail, comprising primarily government run hospitals, not only because of lack of information but also due to lack of confidence to approach doctors. As Caldwell points out, an educated woman "is more likely to be listened to by doctors and nurses. She can demand their attention even when their reluctance to do anything more would completely rebuff an illiterate. She is more likely to know where the right facilities are and to regard them as part of her world and to regard their use as a right and not a boon" (1979, p. 410).

Economic status of women is significantly associated with their ability to seek help for most RTI related symptoms (Table 5). Women belonging to the lower economic group have the lowest rate of seeking help for all the symptoms. The relation between health disadvantage and economic disadvantage has been a theme of many recent studies, including those of Braveman and Tarimo (2002); Heuveline, et al. (2002); Gwatkin (2000); Wagstaff and Doorslaer (2000); Ecob and Davey-Smith (1999). These studies highlight the inequalities that exist between the health of the poor and the rich, at all levels, that is, both between countries and within countries. The poor not only get sick more often but they also have lower accessibility to health services. In their study on social stratification and health in Pakistan, Hadden, Pappas and Khan (2003) found that the disease burden is borne differentially by individuals at different levels of economic status. Reproductive tract infections in this study show a similar trend. Women in the lower economic group, as we saw earlier, had a higher rate of infections and, as Table 5 shows, they are least likely to seek help for their problems.

Demographic Indicators

Health seeking differentials based on selected demographic characteristics, including duration of marriage, number of pregnancies, number of children and interspousal age difference, show no statistically significant pattern across symptoms among women (Table 5).

Symptom-specific Indicators

Most of the indicators measuring different aspects of current symptoms, including their duration, severity, whether they were a source of worry, and if they were discussed with husbands show a statistically significant association with women's decision to seek help for most of the symptoms (Table 5). Women are more likely to seek help if they are having multiple symptoms, except for dysuria and menstrual irregularity, for which health care is sought more often even without the co-existence of any other symptom. Dysuria is a symptom that prompted most women to take action and is also considered a source of worry by the biggest proportion reporting any symptom. Past experience with the symptom does not have any statistically significant association with women's health seeking behaviour (Table 5), but for majority of the symptoms any similar experience in the past has a negative affect on women's current health seeking behaviour. This could be a result of a feeling of dissatisfaction with prior treatment experiences or the fact that women reconcile having the symptom as something they must endure. Other factors, especially financial constraints, might also be a barrier to seek help repeatedly for the same symptom, especially if there is a perception that the treatment did not work on previous occasions. An example of this could be some of the women who were chronic sufferers of abnormal vaginal discharge. When asked why they were not seeking help for the symptom, most of them responded, "It is an old problem. If we go to see a doctor every time, how would we manage other household expenses?".

The length of time a symptom has been experienced is strongly associated with the women's decision to seek help for all the RTI related symptoms (Table 5). The duration between symptom recognition and seeking help can have important repercussions for women's health as obviously the more promptly help is sought the better it is for women, and by implications also their husbands. In the present study, the proportion seeking help within 30 days of experiencing a symptom is lowest, increasing with the passage of time, and is highest when a symptom is over 90 days old. Bhatia and Cleland (1995), in their study in South India on the health seeking behaviour of women with gynaecological morbidities, also found that there is a higher probability of seeking treatment among women who have been experiencing a symptom for a longer time than for those whose experience is a more recent one.

In the present study, the promptness with which treatment is sought varies between symptoms. Women with dysuria (30 percent) were most likely to seek help within 30 days of recognising the symptom (Table 5). While the promptness to deal with lower abdominal pain (19 percent), abnormal vaginal discharge (17 percent) and lower backache (17 percent) is not as much as with dysuria, but it is for menstruation irregularity, dysmenorrhoea and sores/ulcers on the genitals that women appear to delay seeking treatment the most, with no woman seeking help within 30 days. The cyclic nature of menstruation related symptoms could be a reason for women waiting for the next cycle to see if the symptom stays or goes away, inhibiting them to seek help promptly. Otherwise too, dysmenorrhoea was considered a normal thing by a big proportion of women and not many sought help for it at all, leave alone taking a prompt action. Perceived non-serious nature of symptom can delay seeking treatment by women. As Evans and Lambert (1997) found in their study in India, women sought help more promptly for symptoms they perceived as acute and debilitating, but delayed it for ambiguous symptoms, like vaginal discharge and menstrual disorders.

Whether a symptom is perceived as a source of worry or not by women is significantly associated with the decision to seek help, with those getting more worried about the symptom more likely to seek treatment, for all the symptoms (Table 5). The results conform to the existing evidence that suggests this pattern of higher health seeking behaviour among women who perceive their illness as something serious compared to those for whom it is not a source of worry [Crombz, et al. (1999); Evans and Lambert (1997); Younis, et al. (1993)]. As Table 5 shows, the proportion seeking treatment is highest for women who were "worried a lot" about the symptom, followed by those who were "somewhat worried" and those who were not worried at all, with the latter having the lowest rate. The differentials between the health-seeking behaviour for sores and ulcers on genitals, for any of the selected characteristics of women, do not show any pattern or significant relation because of the very few women reporting the symptom (eight) and even fewer seeking help (two).

Another aspect of the experienced symptom that can affect the health seeking behaviour is the perceived severity of the symptom, and the way it affects women's daily activities. In the present study, women were predominantly housewives so the affected activity tended to be carrying out of household chores. As Table 5 shows, women had a higher rate for seeking treatment if the experienced symptoms were perceived as being so severe as to affect the carrying out of daily household activities, compared to those who perceived them as severe but not to a level to affect daily chores. The rate is lowest for those who perceived the symptoms as not severe at all, a trend common among all the symptoms.

One of the strongest associations is found between women's communicating about their symptoms with their husbands and their health seeking behaviour (Table 5). Women talking about their experiences regarding the RTI related symptoms have a much higher rate of seeking help than those who do not. Santhya and Dasvarma (2002) in their study on reproductive illness among women in rural south India also found a significant impact of spousal communication about the illness on the curative behaviour of women regarding their problem. As they point out:

"Gender differences in access to and control over, key material and social resources and resultant inequalities in power, knowledge and the capacity to make independent decisions and to act on them underlie the poor reproductive health status of women in most societies. In gender-stratified societies, men's attitudes and behaviours impact on women's ability to exercise reproductive choice and attain positive sexual and reproductive health outcomes".

[Santhya and Dasvarma (2002), p. 223]

Along with these reasons, in many cases women have restricted mobility and need to take permission from their husbands before leaving the house for any purpose. They might also need to ask their husbands to take them to a health service. Given this scenario, it is understandable that women discussing their problems with their husbands have higher rates of seeking help than their counterparts.

Autonomy Indicators

Women's autonomy in mobility, decision-making and financial matters can have important repercussions for their ability to seek treatment for any health problem they face. Findings of the present study however do not generally show a significant relation between women's health seeking behaviour and their indicators of autonomy (Table 5). When women were asked, who takes the decision to consult or not to consult when they have any health problem, 58 percent said that they take that decision themselves, while 30 percent are a part of the decision-making process carried out jointly by women, their husbands and/or elders of the family, and for 12 percent the decision is made by others, mainly the husband and/or his family. However, this ability to take decision on their own also does not always materialise in actually seeking help, because there are accompanying issues of mobility and money. When asked specifically about their freedom in mobility regarding going to any health service, less than one third (27 percent) said that they did not need permission for it, while 62 percent always needed permission and for 11 percent it depended on the kind/distance of the service they were thinking to access. With women having lower autonomy status but discussing their symptoms more often with their husbands, again highlights the same motivation for communication discussed earlier, that is the need to have consent from the spouse to take action regarding the symptom rather than just talking about the symptom for the sake of sharing information.

The only autonomy indicator showing a regular pattern for health-seeking behaviour, also statistically significant in some cases across the eight symptoms, is for women having at least some control over household income (Table 5). Women involved in controlling household income are more likely to seek help than those who are not. Her control over household income gives her more power to spend money where she wants to, including her access to health care if she decides to do so, removing at least one obstacle in seeking help. Gaining autonomy within the household is just part of the equation, and as reflected by restrictive mobility, it is the general social attitude towards women that form the whole picture, shared even by women themselves. The following statement by one of the respondents summarise this whole idea, when she says, "It is not right for women to go to hospital alone. Why should they want to go alone? Why should they give an opportunity for others to talk bad about them? They should always go with their husbands or some other family member so that others do not get any suspicious ideas in their mind".

Given the strength of influence inter-spousal communication about current symptoms has on women's health seeking behaviour, it would be of interest to see what factors affect this exchange of information between spouses. As Table 6 shows, level of spousal communication varies with the type of symptom. Women with dyspareunia (67.3) were most likely to discuss the problem with their husbands, followed by those reporting dysuria (59 percent) and lower backache (59 percent). The proportion being highest for women with dyspareunia is understandable considering the nature of the problem, and the time when they experience it. Dysuria was the symptom that worried women the most, and this worry could lead them to discuss the problem with their husbands. The symptom that is least discussed with the spouses is dysmenorrhoea (31.5), a symptom that was considered "normal/common thing" by a big proportion of women and for which only around one fourth sought treatment. The proportion of women talking about their experiences with their husbands for the remaining symptoms range between 37 to 47 percent (Table 6).

No statistically significant association is found between a women's age and talking about the symptoms with their husbands, but younger women, that is those in the under 25 age group, had the smallest proportion communicating for majority of the symptoms. The proportion is highest for women in the middle age group, that is 25-34 years, for most symptoms (Table 6). Santhya and Dasvarma (2002) in their study also found younger women to be less likely to tell their husbands about their problems than older women. Contrary to what might be expected, level of education does not show any trend with inter-spousal communication on the current experiences regarding RTI related symptoms (Table 6). For some symptoms it is the more educated women who have a higher rate of communication with their husbands about the issue while in others it is those who have never been to school, with the later generally communicating more on the symptoms associated with pain, like lower backache, lower abdominal pain and dysmenorrhoea (Table 6). A similar irregular pattern is found when we take into account the background area of women, with the proportion talking to their husbands about the problem being higher for those having urban background for some symptoms and rural background for others.

No statistically significant variation is found between women talking about their symptoms and the selected indicators of their demographic characteristics. As Table 6 shows, no discernible pattern of communication is found for different durations of marital union, inter-spousal age difference and the number of children a woman has.

Although statistically not significant, women living in nuclear households are more likely to talk about their problems with their husbands than those living in joint/extended families for majority of the symptoms (Table 6). This could be because of the age effect on the household structure, with younger women more likely to be living in joint/extended arrangements whom we saw talked less about these matters with their husbands, but also due to the generally held idea of lack of privacy between couples in joint/extended households. Women living in such families might be talking about their problems with their mother-in-laws and sister-in-laws instead of their husbands, something quite common in such kind of household structures. Women's economic group, an indicator that has been shown to be significant in most cases, is not significant with regard to inter-spousal communication on RTI related symptoms (Table 6). There is no pattern found across different symptoms, as for some symptoms it is women belonging to the upper economic group who are talking more to their husbands about the symptoms and for others it is those belonging to the lower economic group having more communication.

The factors that seem to govern women's decision to talk about their symptoms with their husbands appear to be more symptom-specific, as can be seen from Table 6. The number of symptoms women are experiencing does not show any significant association with their decision to communicate with their husbands about the problem, but the duration, severity and worry associated with the symptoms have a significant influence on their behaviour. Women who have experienced the reported symptom for longer periods, that is more than 90 days, are more likely to talk about the problem with their husbands than those whose experience is less than 30 days old (Table 6). Likewise, the more a woman is worried about the symptom she is experiencing the more likely she is to discuss it with her husband, as would those who perceive their symptoms so severe so as to affect their daily household routine

Women's autonomy indicators do not show any pattern of association, leave alone a statistically significant one, with inter-spousal communication about the symptoms (Table 6). In some cases it is women with lower autonomy status discussing the problem with their husbands and in others the ones having more autonomy. All four autonomy indicators in the study share this characteristic of no trend between women having a symptom and their discussing it with their husbands.

It could be inferred from the discussion above that women are generally prompted to talk to their husbands about the symptoms they are experiencing when enough time has passed and when they are worried about it. It would not be wrong to say that it is not just a matter of sharing the information with their husbands but a need to take action that make them talk about their problem. As we saw earlier (Table 5), the two most significant factors for women's decision to seek help were the duration of the experience regarding the symptom and the level of worry associated with it, which are factors similar to those significant for their communicating about the problem with their husbands. This gives credence to the inference that it is the need for her decision to seek help to be sanctioned by her husband, which urges her to talk about the problem with him. In some cases she actually needs permission to leave home, in others she might need money for the consultation, and in still others she might need her husband to take her to a health service. No matter what the situation, husband's consent and cooperation is needed in one way or another. Once that support is there, women are more likely to seek help because their decision to seek help is complemented by their husbands' support for them to seek help.

Determinants of Women's Health-seeking Behaviour: A Multivariate Analysis

The aforementioned discussion shows different patterns of health-seeking behaviour among women having different socio-economic and demographic backgrounds. In order to determine the factors most likely to influence this behaviour we will analyse the data using the logistic regression technique. The dependent variable being a dichotomy, that is, seeking treatment versus not seeking treatment, logistic regression is the most suitable method. We will only look into the determinants for seeking help for abnormal vaginal discharge here. This symptom is selected for two reasons. One, it is the symptom most representative of RTIs, and two, it is the most reported symptom in the present study. Similar procedures were carried out for other symptoms as well but the analysis suffered due to small number of cases in many explanatory categories, making the analysis futile. Two models were created for this purpose. Model 1 included all factors that were considered to have a possible impact on the health seeking behaviour of women and Model 2 was based on stepwise forward conditional logistic regression method applied on Model 1, keeping the entry criterion for a variable at .05 and removal criterion at 0.1. Table 7 shows Model 2 to be a better model, as having much fewer variables it correctly predicts the variance in the health seeking behaviour almost as much as Model 1 does, with many more variables. The factors significantly associated with explaining the variance in women's decision to seek treatment for abnormal vaginal discharge include their level of education, their economic group, the length of time they had experienced the symptom, their level of worry about the symptom, talking about the symptom with their husbands and their control over household income (Table 7).

A positive relationship between education and better health has been shown by a number of studies, and the present study also shows that women were five times more likely to seek help if they had up to 10 year of schooling, increasing to six times with more than 10 years of schooling, compared to women who had never been to school, in Model 2 (Table 7). Likewise, the likelihood of seeking help increases dramatically (11 times) for women belonging to the upper economic group compared to those in the lower economic group, corroborating the findings of the bivariate analysis.

The increasing duration of current episode of abnormal vaginal discharge influences women's health seeking behaviour significantly in the multivariate analysis, as was also found in the bivariate analysis. The likelihood of seeking treatment increases with increasing duration, with women experiencing the symptom for 31-90 days being three times, and those for over 90 days being seven times more likely to seek help than those who had experienced it for under 30 days (Model 2, Table 7). A similar relation exists between the worry associated with experiencing abnormal vaginal discharge and women's decision to seek help. Women who are "worried a lot" about the symptom are eight times and those who are "somewhat worried" are three times more likely to seek help than those who are not worried at all about the experience (Model 2). Severity of the symptom, which has a significant association in the bivariate analysis, does not show a significant relationship in the multivariate analysis, but the likelihood of seeking help increases with increasing severity (Model 1). Although not a significant relation, the number of co-existing symptoms along with abnormal vaginal discharge and health seeking behaviour show an interesting pattern of relationship. The likelihood of seeking help increases by four times if a woman has one additional symptom accompanying abnormal vaginal discharge, but decreases with the increasing number of symptoms (Model 1). Inter-spousal communication, as in bivariate analysis, shows a strong statistically significant positive relationship with women seeking treatment. The likelihood increases by 8 times for women discussing their symptom with their husbands than for those who do not (Model 2).

The autonomy indicators, except for women's involvement in controlling households income, do not show any significant relation with women's health seeking behaviour for abnormal vaginal discharge (Table 7). Women who are included in controlling households' income are 6 times more likely to seek help than those who have no control (Model 2). However, despite not being significant, Model 1 does show women with more autonomy being more likely to seek help for all the indicators. Example in this regard being the health seeking behaviour of women with more say in household affairs (twice as likely than those having no say), contented women (2 times more likely than battered women) and women with free mobility (4 times more likely than those who always need permission to go out).

CONCLUSIONS

Less than half the women reporting any symptom seek help, while for some symptoms the proportion seeking help goes down to a mere one fifth. The decision to seek help is a result of a number of factors, important of these being woman's educational and economic status, the extent to which she is worried about the symptom, duration of experiencing the symptom and inter-spousal communication about the symptom. Lack of resources to access any health service and taking the symptom as something common, not needing attention, are the two main reasons for not seeking help. The choice of the health-provider consulted for a symptom is linked to the perceived cause of the symptom, but allopathic doctors are the most commonly reported choice of health providers. The authenticity of these allopathic doctors could at times be dubious but on part of women, it shows their preference for allopathic treatment in most cases.

Having appropriate knowledge plays an effective role in not only prevention of disease but also in the health-seeking process. As findings of this study showed, motivation to consult a health provider for a symptom, and the kind of help sought, also depended on the health beliefs women had about the symptom, so having the right information can help them make more informed decisions. This association extends to conformity to the recommended treatment as well. As Horne and Weinman (1998) point out, health beliefs are stronger predictors of reported adherence to prescribed treatment than clinical and socio-demographic factors. They believe that, "many patients engage in an implicit cost-benefit analysis in which beliefs about the necessity of their medication are weighed against concerns about the potential adverse effects of taking it and that these beliefs are related to medication adherence" [Horne and Weinman (1998), p. 555].

Public campaigns should use local vocabulary and idioms, and refrain from making assumptions that any foreign symbols or words would carry the intended message. With many women not able to read, special consideration should be given to the need of the illiterate. Messages should be concise, unambiguous, and pretested, and should focus on providing information instead of using scare tactics. Given the interest shown in watching television by women, in the present study, it promises to be the best medium to convey such messages. Within the cultural constraints prevalent in Pakistan regarding such issues, devising a relevant public campaign is a difficult task, but not an impossible one.

Improved training of health professionals is the need of the hour as any improvement in health services would be of no avail without appropriate training and education of all professionals involved. Findings of this study showed that doctors, whom women were consulting, did not always give medically sound advice. This finding is supported by the study done by Khandwalla, et al. (2000) about knowledge, attitudes and practices regarding STIs among general practitioners and medical specialists in Karachi, Pakistan. They found doctors, especially GPs, lacking in skills and knowledge to manage and counsel STI patients. Among specialists, they found urologists and dermatologists to be better equipped to manage STIs than gynaecologists, a finding having serious repercussions for the health of women. Most women, if consulting a specialist, would generally go to a gynaecologist, and not a urologist or dermatologist, and it is also gynaecologists who are present in FP/MCH centres. Another study in the country, done on the quality of care provided by private practitioners, showed poor prescribing practices among the health providers [Thaver, et al. (1998)]. Better trained health providers at lower rungs of the health delivery structure would improve the existing poor patient referral system, as also found by Siddiqi, et al. (2001), and Zaidi (1994), and reduce pressure on tertiary health services.

Author's Note: This paper is part of a larger study conducted for my PhD dissertation at the Australian National University, Canberra, Australia. I am indebted to my supervisor, Professor Terence Hull, for his invaluable guidance, and to anonymous referees of this journal who gave comments on an earlier version of this paper. Thanks are also due to Dr Naushin Mahmood for her constant encouragement.

Annex-I
Distribution of Women by Selected Socio-economic and
Demographic Characteristics

Characteristics Number Percentage

Total 508 100.0
Age of Woman (Years)
 <25 96 18.9
 25-34 214 42.1
 34< 198 39.0
Ever been to School
 Yes
 No
Level of Education
 More than 10 Years
 1-10 Years
 No Education
Background Area
 Urban 390 76.8
 Rural 118 23.2
Family Type
 Nuclear 363 71.5
 Joint/Extended 145 28.5
Economic Group
 Upper 116 22.8
 Middle 235 46.3
 Lower 157 30.9
Inter-spousal Age Difference
 Wife Older 15 3.0
 Same Age 30 5.9
 Husband 1-10 Years Older 401 78.9
 Husband >10 Years Older 62 12.2
Duration of Marriage
 One Year or Less 29 5.7
 2-5 Years 105 20.7
 6-15 Years 195 38.4
 16 Years or More 179 35.2
Number of Pregnancies
 None 23 4.5
 1-2 146 28.7
 3-4 148 29.1
 5 or More 191 37.6
Number of Children
 None 41 8.1
 1-2 185 36.4
 3-4 164 32.3
 5 or More 118 23.2
Decision-making Authority
 No Say at All 36 7.1
 Moderate 94 18.5
 Substantial 242 47.6
 Major 136 26.8
Freedom from Threat
 Afraid and Beaten 87 17.1
 Afraid but not Beaten 150 29.5
 Not Afraid but Beaten 57 11.2
 Neither Afraid or Beaten 214 42.1
Freedom of Mobility
 Needs Permission
 Always 314 61.8
 Never 140 27.6
 Depends 54 10.6
Control over Household Income
 Has Control 364 71.7
 Does not have Control 144 28.3

Source: RRTIS 2001-2002.

REFERENCES

Bhatia, J. C., and J. Cleland (1995) Self-Reported Symptoms of Gynecological Morbidity and Their Treatment in South India. Studies in Family Planning 26:4, 203-216.

Bhatti, L. I., and F. F. Fikree (2002) Health-seeking Behaviour of Karachi Women with Reproductive Tract Infections. Social Science and Medicine 54:1, 105-117.

Braveman, P., and E. Tarimo (2002) Social Inequalities in Health within Countries: Not Only an Issue for Affluent Nations. Social Science and Medicine 54, 1621-1635.

Caldwell (1979) Education as Factor in Mortality Decline: An Examination of Nigerian Data. Population Studies 33:3, 395-413.

Chatterjee, M. (1990) Indian Women--Their Health and Economic Productivity. The World Bank, Washington, D. C. (World Bank Discussion Paper No. 109.)

Colson, A. C. (1971) The Differential Use of Medical Resources in Developing Countries. Journal of Health Social Behaviour 12:3, 226-237.

Crombz, G., C. Eccleston, F. Baeyens, B. V. Houdenhove, and A. V. D. Broeck (1999) Attention to Chronic Pain Is Dependent Upon Pain-Related Fear. Journal of Psychosomatic Research 47:5, 403-410.

Ecob, R., and G. Davey-Smith (1999) Income and Health: What Is the Nature of the Relationship? Social Science and Medicine 48, 693-705.

Erwin, J. O. (1993) Reproductive Tract Infections Among Women in Ado-Ekiti, Nigeria: Symptom Recognition, Perceived Causes and Treatment Choices. Health Transition Review 3:(Supplementary Issue), 135-149.

Evans, C., and H. Lambert (1997) Health-seeking Strategies and Sexual Health Among Female Sex Workers in Urban India: Implications for Research and Service Provision. Social Science and Medicine 44, 1791-1803.

Gould, H. A. (1965) Modern Medicine and Folk Cognition in Rural India. Human Organisation 24, 201-208.

Gwatkin, D. R. (2000) Health Inequalities and the Health of the Poor: What Do We Know? What Can We Do? Bulletin of the WHO 78:1, 3-17.

Hadden, W. C., G. Pappas, and A. Q. Khan (2003) Social Stratification, Development and Health in Pakistan: An Empirical Exploration of Relationships in Population-based National Health Examination Survey Data. Social Science and Medicine 57, 1863-1874.

Heuveline, P., M. Guillot, and D. R. Gwatkin (2002) The Uneven Tides of the Health Transition. Social Science and Medicine 55, 313-322.

Horne, R., and J. Weinman (1998) Patients' Beliefs About Prescribed Medicines and their Role in Adherence to Treatment in Chronic Physical Illness. Journal of Psychosomatic Research 47:6, 555-567.

Kabira, W. M., E. W. Gachukia, and F. O. Matiangi (1997) The Effect of Women's Role on Health: The Paradox. International Journal of Gynecology and Obstetrics 58, 23-34.

Khandwalla, H. E., S. Luby, and S. Rahman (2000) Knowledge, Attitudes and Practices Regarding Sexually Transmitted Infections Among General Practitioners and Medical Specialists in Karachi. Sexually Transmitted Infections 76:5, 383-385.

Lambert, H. (1996) Popular Therapeutics and Medical Preferences in Rural North India. Lancet 348:9043, 1706-1709.

Pool, R. (1987) Hot and Cold as an Explanatory Model: The Example of Bharuch District Gujarat, India. Social Science and Medicine 25:4, 389-399.

Population Council (2001) Reproductive Health and Family Planning: An Overview. New York: Population Council.

Reproductive Health Outlook (2001) Reproductive Tract Infections. Website: www.rho.org.

Said, H. (1983) The Unani System of Health and Medicare. In R. Bannerman, J. Burton, and C. Wen-Chieh (eds.) Traditional Medicine and Health Care Coverage. Geneva: WHO. 61-67.

Santhya, K. G., and G. L. Dasvarma (2002) Spousal Communication on Reproductive Illness among Rural Women in Southern India. Culture. Health and Sexuality 4:2, 223-236.

Siddiqi, S., A. A. Kielmann, M. S. Khan, N. Ali, A. Ghaffar, U. Sheikh, and Z. Mumtaz (2001) The Effectiveness of Patient Referral in Pakistan. Health Policy and Planning 16:2, 193-198.

Thaver, I. H., T. Harpham, B. Mcpake, and P. Garner (1998) Private Practitioners in the Slums of Karachi: What Quality of Care Do They Offer? Social Science and Medicine 46:11, 1441-1449.

Wagstaff, A., and E. van Doorslaer (2000) Income Inequality and Health: What Does the Literature Tell Us? Annual Review of Public Health 21, 543-567.

Younis, N., H. Khattab, H. Zurayk, M. el-Mouelhy, M. F. Amin, and A. M. Farag (1993) A Community Study of Gynecological and Related Morbidities in Rural Egypt. Studies in Family Planning 24:3, 175-186.

Zaidi, S. A. (1994) Planning in the Health Sector: For Whom, by Whom? Social Science and Medicine 39:9, 1385-1393.

Zikria, B. A. (1967) Islamic Medicine. American Philosophical Society 57:3.

(1) Traditional doctors' or healers here refers to hakims, homeopathic doctors and spiritual healers together. Though having different philosophies towards treatment they have been put together here because of fewer numbers of cases and showing them as a method parallel to allopathic treatment. 'Hakims' give importance to the temperament of food and liquids and usually derive their medicines from herbs. Treatment recommended by them usually entails a strict diet pattern, stressing avoidance of certain types of food. 'Homeopaths' base their treatment on the philosophy that "'poison kills poison" and treat diseases by administering minute doses of drugs that would cause, in a healthy person, symptoms like those they aim to treat. 'Spiritual healers' on the other hand provide treatment by using religious verses or approaching the spirits [Zikria (1967); Said (1983); Bhatti and Fikree (2002)].

(2) 'Hikmat' is the school of medicine followed by 'Hakims' (see footnote 1 for the description of hakims).

(3) For a simple presentation of frequencies for the predictor variables used in Table 5, 6 and 7 see Annex I.

(4) This question was asked with reference to seeking help for any health problem they had, not specifically RTI related symptoms, but could also be applicable to the latter.

Durr-e-Nayab is Research Anthropologist at the Pakistan Institute of Development Economics, Islamabad.

Table 1
Symptom and Possible Possible Link to RTIs

Abnormal Vaginal Bacterial vaginosis, candidiasis,
 Discharge trichomoniasis, chlamydia, gonorrhoea
Lower Abdominal Pain Chlamydia, gonorrhoea
Menstrual Irregularity Chlamydia, gonorrhoea
Dysmenorrhoea Chlamydia, gonorrhoea
Sores, Warts, Ulcers Genital herpes, chancroid, syphilis
 on Genitals (primary), HPV
Dyspareunia Chlamydia, gonorrhoea
Dysuria Chlamydia, gonorrhoea, trichomoniasis
Lower Backache Bacterial vaginosis, candidiasis,
 trichomoniasis

Source: Population Council (2001); Reproductive Health Outlook (2001).

Table 2
Proportion Seeking Help for Each Reported Symptom

 Number % Seeking
Symptom Reporting * Help

Menstrual Irregularity 105 41.9
Dysmenorrhoea 92 26.1
Dysuria 118 49.2
Lower Abdominal Pain 133 36.8
Abnormal Vaginal Discharge 201 45.8
Sores/Ulcers on Genitals 8 25.0
Dyspareunia 111 20.9
Lower Backache 183 37.1

Source: Rawalpindi Reproductive Tract Infections
Study 2001-2002 (RRTIS 2001-2002).

* From among the X08 women included in the sample.

Table 3
Reasons for Not Seeking Helper Each Symptom (%)

 Menstrual
Reasons Irregularity Dysmenorrhoea Dysuria

Lack of Resources 33.4 20.6 43.3
Did not Find Time 6.7 6.9 8.3
Do nut Feel the Need 20.0 31.8 24.0
Know it is because of
 Contraceptive Use 11.7 88.0 --
Am Tired of Getting
 Treatment, it Does not
 Help Doctors Attitude 8.3 10.3 6.7
Do not Get Permission
 from Husband In-laws 8.3 29.0 5.0
Feel Ashamed -- -- --
Know it is because of Age 11.7 2.9 6.7
Self-medication -- 26.6 7.6
Other 10.0 74.0 6.0

 Lower Abnormal Sores
 Abdominal Vaginal Ulcers on
Reasons Pain Discharge Genitals

Lack of Resources 40.5 26.8 60.0
Did not Find Time 10.7 2.7 --
Do nut Feel the Need 28.3 31.6 --
Know it is because of
 Contraceptive Use 6.0 2.7 16.7
Am Tired of Getting
 Treatment, it Does not
 Help Doctors Attitude -- 20.4 --
Do not Get Permission
 from Husband In-laws 4.8 2.7 --
Feel Ashamed -- 9.3 16.7
Know it is because of Age -- --
Self-medication 14.3 6.6 --
Other 6.0 3.7 167.0

 Lower
Reasons Dyspareunia Backache

Lack of Resources 19.6 31.6
Did not Find Time -- 3.4
Do nut Feel the Need 60.6 23.9
Know it is because of
 Contraceptive Use 4.6 9.4
Am Tired of Getting
 Treatment, it Does not
 Help Doctors Attitude -- 4.3
Do not Get Permission
 from Husband In-laws -- 1.7
Feel Ashamed 27.6 --
Know it is because of Age -- 6.8
Self-medication 3.4 14.4
Other 3.4 9.4

Source: RRTIS 2001-2002.

Note: Includes multiple responses given by women not seeking help
for menstrual irregularity (61), dysmenorrhoea (68), dysuria (60),
lower abdominal pain (84), abnormal vaginal discharge (109), sores
ulcers on genitals (6), dyspareunia (88), and lower backache (116).

Table 4
Type of First Treatment Sought for Each Symptom

 % Seeking First Help from

 Women Doctor Traditional Nurse/
 Seeking Healers LHV/
 Help FWW
Symptom (%)

Menstrual Irregularity 41.9 38.1 1.9 1.0
Dysmenorrhoea 26.1 22.8 -- 1.1
Dysuria 49.2 42.4 3.4 0.8
Lower Abdominal Pain 36.8 32.3 1.5 1.5
Abnormal Vaginal Discharge 45.8 31.3 10.9 1.5
Sores/Ulcers on Genitals 25.0 12.5 12.5 --
Dyspareunia 20.9 19.1 0.9 --
Lower Backache 37.1 33.3 2.2 0.5

 % Seeking First Help from

Symptom TBA/Dai Husband Relative

Menstrual Irregularity 1.0 -- --
Dysmenorrhoea -- 2.2 --
Dysuria 0.8 1.6 --
Lower Abdominal Pain 1.5 -- --
Abnormal Vaginal Discharge 0.5 -- 1.5
Sores/Ulcers on Genitals -- -- --
Dyspareunia -- 0.9 --
Lower Backache 1.1 -- --

Source: RRTIS 2001-2002

Table 5
Differentials in Seeking Help for RTI-related Symptoms
by Selected Background Characteristics of Women (3)

 Women Seeking Help (1) (%)

 Menstrual Dysmenor- Dysuria
Characteristics Irregularity rhoea

Total 41.9 26.1 49.2

Age of Woman (Years)
 <25 25.0 19.0 47.8
 25-34 39.5 25.7 42.2
 34< 48.0 30.6 56.0

Ever been to School
 Yes 46.8 19.0 50.0
 No 34.9 38.2 48.3

Level of Education
 More than 10 Years 54.5 23.1 64.7
 1-10 Years 42.5 17.8 44.2
 No Education 34.9 381.0 48.3

Background Area
 Urban 43.8 33.8 51.9
 Rural 36.0 4.2 43.2
 (**)
Family Type
 Nuclear 43.2 26.8 47.6
 Joint/Extended 37.5 23.8 52.9

Economic Croup
 Upper 48.0 27.8 71.4
 Middle 54.8 27.5 51.0
 Lower 23.7 23.5 41.8
 (**) (*)

Inter-spousal Age
Difference
 Wife Older 100.0 50.0 100.0
 Same Age 33.3 0.0 42.9
 Husband 1-10 Years Older 41.3 26.0 48.8
 Husband >10 Years Older 42.9 30.8 47.8

Duration of Marriage
 One Year or Less 33.3 20.0 50.0
 2-5 Years 37.5 15.0 30.0
 6-15 Years 40.0 22.5 55.3
 16 Years or More 45.7 40.7 51.9

Number of Pregnancies
 None 27.3 30.0 0.0
 1-2 31.8 13.0 53.8
 3-4 62.5 26.1 48.3
 5 or More 39.6 33.3 52.6

Number of Children
 None 46.7 33.3 20.0
 1-2 24.0 15.4 47.1
 3-4 45.0 24.2 52.9
 5 or More 52.0 38.1 55.0

Number of Other Symptoms
Experienced Simultaneously

 Only this Symptom 58.8 0.0 57.1
 1 More 44.4 23.5 52.9
 2-3 More 41.2 26.5 56.0
 [greater than or equal 33.3 32.4 38.6
 to] 4 More

Had Symptom in the
Past too
 Yes 42.9 23.5 47.1
 No 40.0 27.6 52.1

Duration of Experiencing
the Symptom
 < 30 Days 0.0 0.0 29.7
 31-90 Days 33.2 l62 37.8
 > 90 Days 63.3 54.2 75.0
 (**) (***) (***)

Were you Worried about it
 Yes, a Lot 56.4 37.5 53.9
 Yes, Somewhat 5.0 25.0 59.5
 No 15.6 12.5 24.0
 (***) (*) (*)

Severity of the Symptom
 Very Severe, Could nut
 Do House Chores 58.6 38.5 65.2
 Severe, but Could Do
 House Chores 42.3 30.0 46.7
 Not Very Severe 27.3 13.8 44.0
 (*)

Inter-spousal
Communication
About the Symptom
 Yes 81.2 55.2 74.3
 No 18.2 12.7 12.5
 (***) (***) (***)

Decision-making Authority
 No Say at All 50.0 8.3 35.7
 Moderate 40.9 38.1 39.3
 Substantial 45.9 27.5 49.0
 Major 35.3 21.1 68.0

Freedom from Threat
 Afraid and Beaten 42.3 21.4 37.5
 Afraid but not Beaten 38.5 30.0 57.9
 Not Afraid but Beaten 10.5 25.0 38.5
 Neither Afraid or Beaten 61.8 25.0 54.3

Freedom of Mobility
 Needs Permission
 Always 41.8 23.9 42.2
 Never 50.0 35.0 66.7
 Depends 12.5 20.0 62.5

Control over Household
Income
 Has Control 53.8 29.8 54.9
 Does not have Control 22.5 22.2 40.4
 (***)

 Women Seeking Help (1) (%)

 Lower Abnormal Sores/
 Abdominal Vaginal Ulcers on
Characteristics Pain Discharge Genitals

Total 36.8 45.8 25.0

Age of Woman (Years)
 <25 33.3 38.9 0.0
 25-34 41.1 48.4 50.0
 34< 33.3 45.7 20.0

Ever been to School
 Yes 38.9 54.8 50.0
 No 34.4 31.2 16.7

Level of Education
 More than 10 Years 43.5 54.1 0.0
 1-10 Years 36.7 55.2 100.0
 No Education 34.4 31.2 16.7
 (**)
Background Area
 Urban 36.3 47.3 28.6
 Rural 38.7 41.2 0.0

Family Type
 Nuclear 36.4 46.9 16.7
 Joint/Extended 38.5 43.1 50.0

Economic Croup
 Upper 36.4 66.7 --
 Middle 43.1 47.1 50.0
 Lower 28.3 34.3 16.7
 (**)

Inter-spousal Age
Difference
 Wife Older 100.0 14.3 --
 Same Age 33.3 33.3 --
 Husband 1-10 Years Older 35.8 51.3 25.0
 Husband >10 Years Older 29.4 29.0 --
 (*)
Duration of Marriage
 One Year or Less 33.3 62.5 --
 2-5 Years 48.1 29.3 0.0
 6-15 Years 32.6 56.2 33.3
 16 Years or More 35.2 39.7 25.0
 (**)
Number of Pregnancies
 None 40.0 14.3 --
 1-2 48.4 45.7 50.0
 3-4 38.9 51.5 33.3
 5 or More 29.5 43.9 0.0

Number of Children
 None 63.6 20.0 --
 1-2 36.4 43.7 33.3
 3-4 33.3 54.7 50.0
 5 or More 33.3 40.0 0.0

Number of Other Symptoms
Experienced Simultaneously

 Only this Symptom 23.1 42.5 0.0
 1 More 50.0 56.3 --
 2-3 More 41.3 47.1 66.7
 [greater than or equal 30.0 34.9 0.0
 to] 4 More

Had Symptom in the
Past too
 Yes 28.6 45.5 28.6
 No 37.8 66.7 0.0

Duration of Experiencing
the Symptom
 < 30 Days 19.0 17.4 0.0
 31-90 Days 28.1 38.5 0.0
 > 90 Days 49.4 52.5 66.7
 (**) (**)

Were you Worried about it
 Yes, a Lot 48.8 66.7 50.0
 Yes, Somewhat 38.0 52.4 0.0
 No 22.5 25.9 33.3
 (*) (***)

Severity of the Symptom
 Very Severe, Could nut
 Do House Chores 53.3 60.0 0.0
 Severe, but Could Do
 House Chores 42.6 51.2 25.0
 Not Very Severe 28.1 40.9 33.3
 (*)

Inter-spousal
Communication
About the Symptom
 Yes 69.4 68.0 33.3
 No 8.5 32.5 20.0
 (***) (***)

Decision-making Authority
 No Say at All 20.0 46.7 100.0
 Moderate 41.2 38.9 --
 Substantial 39.1 43.8 0.0
 Major 32.0 52.5 50.0

Freedom from Threat
 Afraid and Beaten 20.7 34.8 0.0
 Afraid but not Beaten 35.7 48.4 50.0
 Not Afraid but Beaten 26.7 43.5 100.0
 Neither Afraid or Beaten 51.1 51.5 0.0
 (*)
Freedom of Mobility
 Needs Permission
 Always 28.6 40.7 25.0
 Never 53.8 55.1 25.0
 Depends 56.3 58.8 --
 (*)
Control over Household
Income
 Has Control 36.7 51.9 20.0
 Does not have Control 37.0 33.3 33.3
 (**)

 Women Seeking Help (1) (%)

 Dyspareunia Lover
Characteristics Backache

Total 20.9 37.1

Age of Woman (Years)
 <25 7.7 29.6
 25-34 32.1 33.0
 34< 10.7 45.1
 (*)
Ever been to School
 Yes 27.1 40.0
 No 10.0 32.9

Level of Education
 More than 10 Years 31.8 51.9
 1-10 Years 25.0 36.1
 No Education 10.0 32.9
 (*)
Background Area
 Urban 25.0 40.6
 Rural 7.7 28.3

Family Type
 Nuclear 15.2 34.9
 Joint/Extended 29.5 42.1

Economic Croup
 Upper 18.8 56.5
 Middle 25.0 37.5
 Lower 15.8 30.7

Inter-spousal Age
Difference
 Wife Older 25.0 42.9
 Same Age 0.0 12.5
 Husband 1-10 Years Older 24.7 39.5
 Husband >10 Years Older 0.0 29.2

Duration of Marriage
 One Year or Less 7.1 27.3
 2-5 Years 25.9 23.3
 6-15 Years 29.3 36.0
 16 Years or More 10.7 45.7

Number of Pregnancies
 None 40.0 14.3
 1-2 9.1 36.6
 3-4 34.5 35.5
 5 or More 15.8 40.8

Number of Children
 None 31.3 23.1
 1-2 15.0 29.7
 3-4 28.1 46.2
 5 or More 13.6 38.6

Number of Other Symptoms
Experienced Simultaneously

 Only this Symptom 0.0 43.8
 1 More 36.4 32.4
 2-3 More 26.8 46.2
 [greater than or equal 9.8 25.9
 to] 4 More (*)

Had Symptom in the
Past too
 Yes 20.8 -- (a)
 No 22.2

Duration of Experiencing
the Symptom
 < 30 Days 2.9 17.5
 31-90 Days 14.0 25.0
 > 90 Days 50.0 53.3
 (***) (***)

Were you Worried about it
 Yes, a Lot 36.0 46.4
 Yes, Somewhat 27.3 42.6
 No 12.7 2.9
 (*) (***)

Severity of the Symptom
 Very Severe, Could nut
 Do House Chores 40.0 58.8
 Severe, but Could Do
 House Chores 26.1 37.3
 Not Very Severe 16.9 31.3
 (*) (*)

Inter-spousal
Communication
About the Symptom
 Yes 28.4 55.5
 No 5.6 10.5
 (**) (***)

Decision-making Authority
 No Say at All 7.1 25.0
 Moderate 21.4 28.6
 Substantial 23.6 36.3
 Major 23.1 50.0

Freedom from Threat
 Afraid and Beaten 8.0 27.7
 Afraid but not Beaten 23.5 34.9
 Not Afraid but Beaten 18.2 36.4
 Neither Afraid or Beaten 27.5 48.1

Freedom of Mobility
 Needs Permission
 Always 16.0 32.9
 Never 333.0 51.4
 Depends 25.0 44.4

Control over Household
Income
 Has Control 24.2 42.2
 Does not have Control 15.9 28.6
 (*)

Source: RRTIS 2001-2002.

Note: (1) From among those reporting a particular symptom, that is,
menstrual irregularity (105), dysmenorrhoea (92), dysuria (118),
lower abdominal pain (133), abnormal vaginal discharge (20l), sores
and ulcers on genitals (8), dyspareunia (111) and lower backache (183).

(a) "Experienced in the past too" was not calculated for
backache because its definition included the presence of any other
symptom at the same time too, which would be difficult to measure
due to recall problems.

Chi-square/Fisher's Exact test significance levels:
*** p<.001, ** p<.01, and * p<.05. for having/not having
any infection.

Table 6
later-spousal Communication on Current Symptoms by Women (1)

 Women Talking about their Experience
 of a Symptom with their Husbands (%)

 Menstrual Dysmenor- Dysuria
Characteristics Irregularity rhoea

Total 37.1 31.5 59.3
Age of Woman (fears)
 <25 33.3 28.6 52.2
 25-34 32.6 34.3 64.4
 34 42.0 30.6 58.0

Level of Education
 > 10 Years 54.6 23.1 70.6
 1-10 Years 32.5 24.4 48.8
 No Education 32.6 44.1 63.8

Background Area
 Urban 40.0 36.8 59.3
 Rural 28.0 16.7 59.5

Family Type
 Nuclear 39.5 32.4 61.9
 Joint/Extended 29.2 28.6 52.9

Economic Croup
 Upper 36.0 33.3 71.4
 Middle 47.0 24.0 63.3
 Lower 26.3 38.2 52.7

Inter-spousal Age
Difference
 Wife Older 50.0 50.0 50.0
 Same Age 33.3 0.0 57.1
 Husband 1-10 Years Older 37.5 30.1 55.8
 Husband >10 Years Older 33.3 38.5 69.6

Duration of Marriage
 One Year or Less 33.3 20.0 62.5
 2-5 Years 37.5 20.0 10.0
 6-15 Years 37.5 42.5 71.1
 16 Years or More 37.0 259.0 57.7

Number of Children
 None 46.7 33.3 40.0
 1-2 32.0 38.5 52.9
 3-4 42.5 27.3 67.6
 5 or More 28.0 28.6 62.5

Number of Other
Symptoms Experienced
Simultaneously
Only this Symptom 52.9 28.6 57.1
 1 More 38.9 23.5 64.7
 2-3 More 35.3 35.3 68.0
 [greater than or 30.6 32.4 47.7
 equal to] 4 More

Duration of Experiencing
the Symptom
 < 30 Days 18.8 0.0 59.5
 31-90 Days 34.9 26.5 51.4
 > 90 Days 60.0 45.8 65.9

Were you Worried
About it
 Yes, a Lot 56.4 31.3 56.9
 Yes, Somewhat 32.4 38.9 66.7
 No 18.8 20.8 52.0
 (**)

Severity of the Symptom
 Very Severe, Could not
 do House Chores 46.8 53.8 65.2
 Severe, but Could do
 House Chores 39.3 30.0 60.0
 Not Very Severe 31.7 24.1 56.0

Decision-making
Authority
 No Say at All 41.7 33.3 57.1
 Moderate 45.5 47.6 57.1
 Substantial 35.1 25.0 52.9
 Major 32.4 26.3 76.0

Freedom from Threat
 Afraid and Beaten 34.6 28.6 43.8
 Afraid but not Beaten 57.7 30.0 71.1
 Not Afraid but Beaten 10.5 33.3 53.8
 Neither Afraid or Beaten 38.2 33.3 62.9
 (**)
Freedom of Mobility
 Needs Permission
 Always 37.3 35.8 61.4
 Never 43.3 20.0 51.9
 Depends 12.5 20.0 62.5

Control over Household
Income
 Has Control 10.0 29.8 62.0
 Does not have Control 32.5 33.3 55.3

 Women Talking about their Experience
 of a Symptom with their Husbands (%)

 Lower Abnormal Sores/
 Abdominal Vaginal Ulcers on
Characteristics Pain Discharge Genitals

Total 46.6 37.3 37.5
Age of Woman (fears)
 <25 33.3 30.6 0.0
 25-34 55.2 45.3 50.0
 34 42.1 30.0 40.0

Level of Education
 > 10 Years 39.1 24.3 0.0
 1-10 Years 44.9 33.3 100.0
 No Education 50.8 48.1 33.3
 (**)

Background Area
 Urban 45.1 32.0 28.6
 Rural 51.6 52.9 100.0
 (**)

Family Type
 Nuclear 45.8 41.3 33.3
 Joint/Extended 50.0 27.6 50.0
 (*)

Economic Croup
 Upper 50.0 16.7 --
 Middle 44.6 37.5 0.0
 Lower 47.8 46.3 50.0
 *

Inter-spousal Age
Difference
 Wife Older 50.0 14.3 --
 Same Age 33.3 33.3 --
 Husband 1-10 Years Older 45.3 35.1 37.5
 Husband >10 Years Older 47.1 54.8 --

Duration of Marriage
 One Year or Less 33.3 37.5 --
 2-5 Years 40.7 34.1 0.0
 6-15 Years 50.0 43.8 66.7
 16 Years or More 48.1 30.2 25.0

Number of Children
 None 63.6 30.0 --
 1-2 45.5 43.7 66.7
 3-4 38.1 32.0 --
 5 or More 52.8 37.8 33.3

Number of Other
Symptoms Experienced
Simultaneously
Only this Symptom 30.8 20.0 0.0
 1 More 54.2 37.5 --
 2-3 More 50.0 42.9 66.7
 [greater than or 44.0 44.2 25.0
 equal to] 4 More

Duration of Experiencing
the Symptom
 < 30 Days 17.2 24.1 0.0
 31-90 Days 47.5 29.6 33.3
 > 90 Days 59.4 42.4 50.0

Were you Worried
About it
 Yes, a Lot 60.5 56.l 66.7
 Yes, Somewhat 56.0 47.6 0.0
 No 20.0 16.0 50.0
 (***) (***)

Severity of the Symptom
 Very Severe, Could not
 do House Chores 66.7 50.0 100.0
 Severe, but Could do
 House Chores 57.4 53.5 25.0
 Not Very Severe 32.8 22.7 33.3

Decision-making
Authority
 No Say at All 40.0 60.0 100.0
 Moderate 52.9 50.0 --
 Substantial 43.8 28.1 20.0
 Major 48.0 37.7 0.0
 (*)

Freedom from Threat
 Afraid and Beaten 44.8 30.4 0.0
 Afraid but not Beaten 42.9 40.6 50.0
 Not Afraid but Beaten 533.0 43.5 0.0
 Neither Afraid or Beaten 48.9 36.8 100.0

Freedom of Mobility
 Needs Permission
 Always 45.1 43.0 50.0
 Never 53.8 18.4 25.0
 Depends 43.8 47.1 --
 (**)

Control over Household
Income
 Has Control 48.1 34.8 40.0
 Does not have Control 44.4 42.4 33.3

 Women Talking about their Experience
 of a Symptom with their Husbands (%)

 Dyspareunia Lower
Characteristics Backache

Total 67.3 59.1
Age of Woman (fears)
 <25 61.5 51.9
 25-34 71.4 58.0
 34 64.3 63.4

Level of Education
 > 10 Years 72.7 59.3
 1-10 Years 60.4 51.8
 No Education 72.5 67.1

Background Area
 Urban 63.1 59.4
 Rural 80.8 58.5

Family Type
 Nuclear 69.7 60.5
 Joint/Extended 63.6 56.1

Economic Croup
 Upper 62.5 60.9
 Middle 66.1 56.8
 Lower 71.1 61.3

Inter-spousal Age
Difference
 Wife Older 25.0 71.4
 Same Age 33.3 25.0
 Husband 1-10 Years Older 74.2 60.5
 Husband >10 Years Older 45.5 58.3

Duration of Marriage
 One Year or Less 64.3 36.4
 2-5 Years 70.4 53.3
 6-15 Years 61.0 58.7
 16 Years or More 75.0 65.7

Number of Children
 None 43.8 61.5
 1-2 75.0 57.8
 3-4 62.5 53.8
 5 or More 77.3 68.2

Number of Other
Symptoms Experienced
Simultaneously
Only this Symptom 66.7 56.3
 1 More 77.3 64.7
 2-3 More 73.2 60.3
 [greater than or 56.1 55.2
 equal to] 4 More

Duration of Experiencing
the Symptom
 < 30 Days 48.6 35.0
 31-90 Days 69.8 50.0
 > 90 Days 84.4 75.6

Were you Worried
About it
 Yes, a Lot 76.0 67.9
 Yes, Somewhat 68.2 63.2
 No 63.5 29.4
 (***)

Severity of the Symptom
 Very Severe, Could not
 do House Chores 70.0 70.6
 Severe, but Could do
 House Chores 78.3 63.7
 Not Very Severe 63.6 49.3

Decision-making
Authority
 No Say at All 84.6 37.5
 Moderate 42.9 50.0
 Substantial 7(l.9 65.0
 Major 85.7 64.6
 (**)

Freedom from Threat
 Afraid and Beaten 72.0 46.8
 Afraid but not Beaten 70.6 57.1
 Not Afraid but Beaten 63.6 77.3
 Neither Afraid or Beaten 62.5 64.8

Freedom of Mobility
 Needs Permission
 Always 64.0 58.6
 Never 77.8 59.5
 Depends 62.5 66.7

Control over Household
Income
 Has Control 71.2 64.7
 Does not have Control 61.4 50.0
 (*)

Source: RRTIS 2001=2002.

Note: (1) Figures shorn proportion of women, with a specific
symptom at the time of survey. communicating about their
experiences with their husbands That is proportion of women having
inter-spousal communication from those reporting menstrual
irregularity (101), dysmenorrhoea (92), dysuria (118), lower
abdominal pain (133). abnormal vaginal discharge (201), sores and
ulcers on genitals (8), dyspareunia (111) and lower backache (183).
Chi-square/Fisher's Exact test significance levels: *** p<.001,
** p<.01, and * p<.05, for having not having any infection.

Table 7
Logistic Regression Analysis of Women's Health-seeking Behaviour on
Experiencing Abnormal Vaginal Discharge

 Model 1

Predictor Variable Co-efficient Odds Ratio

Age of Women
 >25 (a)
 25-34 -1.571 0.21
 34< -.697 0.50

Level of Education
 Never been to School (a)
 1-10 Years 1.050 3.86 *
 11 or More Years 1.894 6.64 *

Family Structure
 Joint/Extended (a)
 Nuclear .233 0.79

Background Area
 Rural (a)
 Urban -.259 0.77

Duration of Marriage
 1 Year or Less (a)
 2-5 Years -1.848 0.16
 6-15 Years .092 1.10
 16 Years or More .549 0.58

Economic Group
 Lower (a)
 Middle .383 1.47
 Upper 2.232 9.32 **

Inter-spousal Age Difference
 Husband > 10 Years Older (a)
 Same Age 1.117 3.06
 Wife Older -.145 0.87
 Husband 1-10 Years Older 1.697 5.46

Number of Pregnancies
 None (a)
 1-2 2.643 4.05
 3-4 1.824 6.20
 5 or More 2.161 8.68

Duration of Experiencing the
 Symptom
 [less than or equal to] 30 Days (a)
 31-90 Days 1.693 5.44 *
 > 90 Days 2.490 13.06 ***

Number of Symptoms
 Only this One Symptom (a)
 One More Symptom 1.428 4.17
 2-3 More Symptoms -.055 0.95
 4 or More Other Symptoms -.825 0.44

Worry About the Symptom
 No (a)
 Yes, a Lot 2.523 12.47 ***
 Yes, Somewhat 1.540 4.67 **

Severity of Symptom
 Not Severe, Could Do House Chores (a)
 Very Severe, Could not Do House
 Chores .328 1.39
 Severe, But Could Do House Chores .229 1.26
Inter-spousal Communication About
 the Symptom
 No (a)
 Yes 2.46 11.74 ***
Decision-making Authority
 No Say at All (a)
 Moderate Say -.122 0.89
 Substantial Say -.461 0.63
 Major Say .121 1.13
Freedom from Threat
 Afraid and Beaten (a)
 Afraid but not Beaten .503 1.65
 Not Afraid but Beaten -.564 0.57
 Neither Afraid or Beaten .528 1.70
Freedom of Mobility
 Needs Permission
 Always (a)
 Never 1.396 4.04
 Depends .091 1.10
Control Over household Income
 Does Not Have Control (a)
 Has Control 1.284 3.61 *
Constant -8.959 ***
Model Chi-Square 126.678 ***
Degrees of Freedom 36
R-square 62.5%
Reporting Predictive Correctly 85.6%
Hosmer-Lemeshow Test .280
Number of Cases 201

 Model 2

Predictor Variable Co-efficient Odds Ratio

Age of Women
 >25 (a)
 25-34 -- --
 34< -- --

Level of Education
 Never been to School (a)
 1-10 Years 1.525 4.60 ***
 11 or More Years 1.751 5.76 **

Family Structure
 Joint/Extended (a)
 Nuclear -- --

Background Area
 Rural (a)
 Urban -- --

Duration of Marriage
 1 Year or Less (a)
 2-5 Years -- --
 6-15 Years -- --
 16 Years or More -- --

Economic Group
 Lower (a)
 Middle .411 1.51
 Upper 2.351 10.50 ***

Inter-spousal Age Difference
 Husband > 10 Years Older (a)
 Same Age -- --
 Wife Older -- --
 Husband 1-10 Years Older -- --

Number of Pregnancies
 None (a)
 1-2 -- --
 3-4 -- --
 5 or More -- --

Duration of Experiencing the
 Symptom
 [less than or equal to] 30 Days (a)
 31-90 Days 1.066 2.90
 > 90 Days 1.988 7.16 **

Number of Symptoms
 Only this One Symptom (a)
 One More Symptom --
 2-3 More Symptoms --
 4 or More Other Symptoms --

Worry About the Symptom
 No (a)
 Yes, a Lot 2.115 8.30 ***
 Yes, Somewhat 1.084 2.97 *

Severity of Symptom
 Not Severe, Could Do House Chores (a)
 Very Severe, Could not Do House
 Chores -- --
 Severe, But Could Do House Chores -- --
Inter-spousal Communication About
 the Symptom
 No (a)
 Yes 3.132 8.44 ***
Decision-making Authority
 No Say at All (a)
 Moderate Say -- --
 Substantial Say -- --
 Major Say -- --
Freedom from Threat
 Afraid and Beaten (a)
 Afraid but not Beaten -- --
 Not Afraid but Beaten -- --
 Neither Afraid or Beaten -- --
Freedom of Mobility
 Needs Permission
 Always (a)
 Never -- --
 Depends -- --
Control Over household Income
 Does Not Have Control (a)
 Has Control 1.718 5.57 ***
Constant -6.217 ***
Model Chi-Square 97.802 ***
Degrees of Freedom 10
R-square 55%
Reporting Predictive Correctly 84.1%
Hosmer-Lemeshow Test .807
Number of Cases 201

Source: RRTIS 2001-2002.

Note: Chi-square/Fisher's Exact test significance levels:
*** p < .001, ** p < .01, and * p < .05, for having/not
having any infection.

(a) Reference category.