Sexual harassment is a serious problem for women workers. There is extensive anecdotal evidence indicating its pervasiveness, but it remains hidden by the veil of silence surrounding the issue.
In India, reports suggest that women who report sexual harassment are doubly victimised: first when they are harassed and subsequently through the protracted and traumatic process of redress. The victim is blamed and stigmatised and her prospects of continuing work are affected (1, 2). Women remain silent from fear of having the event trivialised (3, 4) or losing employment (5). In most cases reported, the harasser is the employer or in the top rung of the management hierarchy (6, 4).  Moreover, mechanisms of redress are slow. Rupan Deol Bajaj, an IAS officer, waited almost 10 years for a Supreme Court verdict (1).

Sexual harassment in medical settings
Sexual harassment is a human rights violation. It is also a serious cause for concern in health care institutions that train students, employ women in various capacities and also cater to health needs of men and women. It affects the attitudes, behaviours, and learning capabilities of medical students. It results in a hostile atmosphere at work, interferes with work performance (7) and can affect patient care.
Available studies on harassment of medical professionals have focused on developed country experiences (8, 9, 10).  These indicate that a significant proportion experience bullying in some form (8). Minority groups, women and people lower down in the workplace hierarchy are more likely to be victims (8). 
In an anonymous survey of residents and interns, three quarters of women respondents reported at least one episode of harassment. Women in academic medicine indicate that such experiences continue in their professional lives.  Few reported sexual harassment to authorities, believing that it would be detrimental to their careers (7). Clearly, women medical professionals remain at risk of sexual harassment despite the power they acquire through medical training. They are also vulnerable to sexual harassment from patients;  more than three fourths of the women responding to an anonymous survey reported some form of sexual harassment by a patient (11).
In this country, media reports focussed public attention on the subject and led to advocacy initiatives. As a result, in 1997, the Supreme Court's judgement in the Vishaka Vs State of Rajasthan laid down a clear definition of sexual harassment:
"Sexual harassment includes such unwelcome sexually determined behaviour (whether directly or by implication) as: a) physical contact and advances; b) a demand or request for sexual favours; c) sexually coloured remarks; d) showing pornography; e) any other unwelcome physical, verbal or non-verbal conduct of sexual nature.
"Where any of these acts is committed in circumstances whereunder the victim of such conduct has a reasonable apprehension that in relation to the victim's employment or work whether she is drawing salary, or honorarium or voluntary, whether in government, public or private enterprise such conduct can be humiliating and may constitute a health and safety problem. It is discriminatory for instance when the woman has reasonable grounds to believe that her objection would disadvantage her in connection with her employment or work including recruiting or promotion or when it creates a hostile work environment. Adverse consequences might be visited if the victim does not consent to the conduct in question or raises any objection thereto." (12)
The Supreme Court's judgement requires institutions to take action against harassment.
We could not identify Indian studies looking at sexual harassment in medical institutions. However, there have been many press reports on sexual harassment in hospitals. To illustrate: a nurse at a leading Mumbai hospital was raped by hospital staff (13); a professor was accused of harassing women faculty (14) and patients have been sexually assaulted (15, 16). 
There has been a steady increase in the number of women reporting sexual harassment, from 4,756 in 1995 to 11,024 in
2000 (5). A survey by the National Women's Commission reports that 46.58% of women report sexual harassment in the work place; only about 3.54% report the matter to authorities; 1.4% reported it to the police (5).  In 2001, a five-state survey of workplace sexual harassment undertaken by Sakshi, a NGO in New Delhi, reported that 80% of the respondents said sexual harassment existed in their work place (17).  Only 23% had heard of the Vishaka Guidelines; 66% of these said that the institutions had not effectively implemented these guidelines. When they had been implemented, redress seemed to be biased.
Women must complain in an "extremely hostile environment, with risk of backlash, humiliation, injury - mental and physical -- and complete loss of confidentiality"(4). Trade unions were not involved in the original Supreme Court judgement (6) and have not been particularly sympathetic to complaints. At times they have even agitated against the complaining woman worker (2, 4).
In this context, we attempted to explore the issue of sexual harassment in the medical workplace setting.

Ethical and methodological issues
Any study on sexual harassment must ensure confidentiality for respondents. Further, direct interviews in the workplace (after obtaining informed consent) may not yield reliable information, as all parties are identifiable. This is a problem even if they are guaranteed confidentiality.
Before this survey was conducted, some of us had attempted to undertake a survey on the felt need for training on ethics and gender in health service delivery among the staff in a medical institution. The survey included questions on sexual harassment, as informal anecdotal information indicated that it did exist. However, none of the few completed schedules that were returned from face-to-face interviews included responses to the questions on sexual harassment.
An alternative is to conduct the survey in a number of institutions. But anonymous surveys have lower response rates (18).  Further, while they allow participants anonymity, institution-based surveys can compromise institutions' anonymity. Finally, medical institutions are unlikely to permit surveys of sexual harassment, as any identification of sexual harassment in the institution would affect their credibility and also affect patient inflow.
Mail-in questionnaires would not identify either the individuals or the institution concerned, and people would also be able to respond freely. However, this method permits multiple counting of the same event reported by different people, and bogus reporting.  Further, the mean response rate to mail-in surveys published in medical journals is just 60 per cent (18). One might also receive responses referring to non-medical institutions; of course such responses would be valuable as they relate to sexual harassment in the workplace.
It can be presumed that women occupying low rungs of the occupational hierarchy are most vulnerable to sexual harassment. The survey should reach these vulnerable groups. However, such women would also have the most to lose by identification and would therefore be most reluctant to respond to surveys.
Web-based surveys for medical personnel may be an alternative to traditional surveys, though the response rate can be lower than that of mail-in surveys (19).  Internet users are presumably higher up in the socio-economic hierarchy. If we found events of sexual harassment reported among this group, it would be reasonable to believe it was more common among women in lower-paid jobs within such institutions. 

Method adopted
A survey was developed containing key open-ended questions on incidents of workplace-related sexual harassment, institutional mechanisms and their efficacy. This was posted on the Indian Journal of Medical Ethics website and  also printed in the April 2004 issue of the journal. Respondents could send the completed questionnaire to the journal through the website or by mail to the journal's executive editor. (It was decided that attaching a self-addressed, stamped envelop to the mail-in questionnaire was an undue inducement.) Only the executive editor and the web editor had direct access to the responses. These were anonymised and forwarded to us for analysis.
The study was reviewed by the Institutional Ethics Committee of the Sree Chitra Tirunal Institute for Medical Sciences and Technology (SCTIMST) and cleared with the caveat that it should be reported in a peer-reviewed journal with a strong ethical orientation. After the survey was put up on the website of the Indian Journal of Medical Ethics in April 2004, the journal's executive editor received calls that the wording of the questionnaire prevented the reporting of sexual harassment over one year old. We therefore reworded the relevant question to read: 'Do you have any knowledge of any event of sexual harassment in your work setting that ever happened to you or to one of your acquaintances?' The changed question was resubmitted to the IEC and put up on the website. Responses from April to November 2004 were considered for analysis.

Responses
There were 23 web-based responses to the study and one mailed-in response (excluding test responses).  Three were from non-medical settings but were included for consideration.
There were 11 responses reporting harassment of men, in all cases by a male abuser. All had been reported to the appropriate committees in the institutions and action taken. However, we had doubts about the authenticity of these responses as existing rules do not include inquiring into sexual harassment of men. Our doubts were confirmed when scrutinising the completed questionnaires. The similarity of responses and the pattern in these 11 cases (compared to the 13 others) strongly suggested that they were frivolous. For this reason, we confined our analysis to the remaining 13 responses.
In 11 of these 13 responses, women were reported to be the victims of harassment. In two cases the respondents did not mention the sex of the harassed person. The persons involved in nine of the 11 reported cases were men in supervisory or senior positions. In two cases they were co-trainees or transport personnel in the same organisation (occupying lower levels in the hierarchy). 
We treated each of the 13 schedules (including those referring to non-medical situations) as case reports and analysed them for clues about the nature of sexual harassment and the potential for redress for the victims. Four narratives were particularly illustrative of the situations faced by women.

Who is harassed
The victims were mostly young and/or relatively powerless women, such as rural women seeking care in urban health facilities, post-graduate students, field staff and contract employees.
A faculty member of a medical teaching institution demanded oral sex of a girl who had come to the hospital from a near-by village. Her father tried to protest to the hospital authorities but the case was quashed. (Reported by a male non-medical professional)
However, women in higher positions can also be vulnerable, though they may not initially acknowledge that they have been victims.
A senior woman government servant on election duty reported that the financial observer made physical advances, demanded sexual favours and made sexually coloured remarks. She reported this in a TV interview to break the denial syndrome and highlight the fact that women in higher positions are also subject to sexual harassment. (Reported by a senior woman government officer)

Types of harassment
The most frequent type of harassment seems to be physical contact and advances (eight responses) and sexually coloured remarks (eight), other unwelcome physical, verbal or non-verbal conduct of a sexual nature (five), and demands for sexual favours (four). There are also reports of voyeuristic behaviour and one report of a display of pornography.

Redress mechanisms
Of the 13 persons whose cases were reported, four did not have institutional mechanisms of redress, or they did not know of them. Of the nine women who reported the experience of sexual harassment, eight reported the matter to the authorities and resolved the problems by either removing the abusing  person or by restraining inappropriate behaviours.
Abuse is likely to remain unreported when the victim is relatively powerless or not from within the system. Thus, female users of health facilities, who are already vulnerable because of the illness for which they seek care, could be further victimised by abusive staff. Women from the system who complained did so only after a prolonged period of self doubt. Complaints registered collectively rather than by an individual seemed to have a better chance for redress.
The teacher used to try and make physical contact and advances during the duty. He would also make sexually coloured remarks to post-graduate students. The students refrained from complaining as they believed no action would be taken and they would be blamed unnecessarily. However, later they collected enough courage to complain to the authorities. The teacher was asked to resign.
(Reported by a male medical officer)
Redress also seems to be swift if the abuser is relatively powerless. In one report a visiting trainee indulging in voyeurism was asked to leave the programme and a strongly-worded note sent to his parent institution.
It is also possible that women who have been abused once continue to be victimised. Women who complain of abuse related to career advancement are less likely to be viewed sympathetically.
The senior academic kept promising to get the junior staff member a permanent job. All the while he continued to make verbal and physical advances. One day he took her out on his two-wheeler, supposedly to get her employment status regularised. Instead, he drove out to a lonely place. When he started making vulgar comments she jumped from the moving two-wheeler and was injured. She complained,  an enquiry was conducted and the man was found guilty. However, authorities decided to retain him as he was a  permanent employee with a powerful position in the institution. She lost her job. (Reported by a university teacher)

What next?
There are many questions one might ask of a web-based survey which received only 24 responses, of which 11 were judged to be frivolous. It might be that a web-based survey is not appropriate for this topic, particularly in India where even educators and researchers have limited use of the internet. Potential respondents may not trust that a web-based survey will protect their anonymity. It is possible that information about the survey was not conveyed to the target population. There were also no mechanisms for follow-up to improve the number of responses.
It may also be argued that the methodology permitted fraudulent and multiple responses. Also, the survey results say nothing that is not already known. However, it is worth noting these responses have emerged from a relatively systematic effort to record people's experiences. There is a need to explore alternative methods to study sexual harassment in the work place, particularly if one wishes to quantify its occurrence or develop a quantified understanding of its determinants.
The picture that emerged from the responses supported our perception of the situation. This perception was based on anecdotes collected through informal interactions. The responses suggest that sexual harassment is, indeed, a reality in medical workplaces, and that both employees and patients are vulnerable. This has to be taken cognisance of when discussing workplace sexual harassment.