Contents - Previous - Next

Exploring women's health-seeking behaviors

Unit 16. Informant Interviewing: Use of Illness Scenarios with Local Practitioners

Purpose

This task has several purposes. It is intended to elicit data on practitioner's views of (i) how they should respond to an illness episode, including home care as well as seeking help outside the home; (ii) how soon signs and symptoms are expected to improve after an action has been taken; and (iii) what sequence of actions may follow when these expectations are not met. The scenarios are also useful for obtaining views of health care providers concerning their expectations about what women usually do and what they should do for illness episodes.

Data Collection

1. Find informants who are local healers that serve the community. These should include both traditional and allopathic health providers. The basic format of the procedure is the presentation of hypothetical cases to the respondent, followed by a series of questions. In the case of key informants and women respondents, the first question asks the individual to give advice about what the sick woman should do. This is followed by additional questions concerning next steps and events after the initial action.

By casting the questions in the form of hypothetical cases, the intention is to elicit responses about the "best" or most appropriate course of behavior as the respondent sees it. The questions are framed in terms of giving advice to someone else. Answering hypothetical cases concerning another woman is intended to remove inhibitions that may arise from a respondent's unique economic/social situation.

By varying the severity of symptoms described in the episode, the responses to the scenarios also provide an indication of whether women recognize danger signs and change their advice in the presence of these signs. By varying the age of the woman in the scenarios, the influence of these factors on perceptions can also be examined.

In addition to data on cultural norms of household management and expectations about the course of women's illnesses, this procedure provides an opportunity to compare cultural ideals with the reported choices obtained from narratives of past illness episodes. This will permit a rough assessment of the differences between what women feel they should do, and what they usually do.

2. Selecting the type of scenarios depends on the organization's interests.

In order to make the stories believable and permit analysis of the results in relation to local disease terms, modify the basic format by:

a) Giving the woman a name, and other characteristics that will increase interest without changing the nature of the problem.

b) Using local terms for signs and symptoms.

c) Try not to glue either local or biomedical illness-categories in the hypothetical case that would represent a pre-judgement or "leading question" that is better avoided.

3. The texts of the scenarios vary the woman's age and the signs and symptoms. The basic format of the scenarios, with suggestions for the type of "local context" are given below. This context will need to be added in order to make the stories believable. The terms for specific signs and symptoms may have to be altered to conform with local cultural perceptions. In making adjustments to the scenario content given below, try to maintain the same duration of symptoms and ages of the women.

EXAMPLE
Scenario for Teenage Girl with Weakness

"Your neighbor, Mrs. Pandit, has a daughter, Kumari, who is fourteen years old. Two days ago Kumari started feeling tired and unable to work. She eats very little during meals and is very slow about her chores. Suppose that Mrs. Pandit comes to ask you for advice about what she should do. What would you tell her?"

Write Several Scenarios For Different Key Illnesses:

(1) _________________________________________

(2) _________________________________________

(3) _________________________________________

4. Suggested Wording

Generally local health practitioners respond well to requests to comment on hypothetical cases. but be prepared to give a specific rationale for presenting this task to respondents. It might be useful to stress that while a lot has been learned about the kind of problems women experience, it would be helpful to know more about the way women should treat these problems.

Do not present all scenarios to every respondent (2-3 per respondent is sufficient) as this would become very tedious and time consuming. Every respondent should be given at least one serious and one mild case.

When giving the scenarios, be sure to randomize the order of presentation. Don't start every interview with the youngest woman, or every case with signs of "white water" or "weakness."

5. Follow-Up Questions

In addition to the scenarios, which conclude with a general question about what the woman should do, there are a series of additional questions and probes that accompany the presentation of the cases. The specific sequence of these questions may vary, depending on the answers the woman initially gives.

'These should include the following:

a. Is there anything else it would be helpful to know about (name of woman) in order to give her advice?

b. What illness do you think the woman has?

If the respondent suggests a home remedy only:

a. Is there anything else - any other sign - the woman should watch for?

b. How soon should the woman get better after she gets ____ (remedy)? What are the signs that she is getting better?

c. If the (named signs) don't improve in (time specified by respondent) what should the woman do next?

If the respondent suggests going to an allopathic practitioner:

a. Is there any treatment the doctor (nurse) should give the woman at the clinic (office)?

b. What kind of treatment do you think the doctor (nurse) will tell the woman to do at home?

c. How soon should the woman get better after she gets ____ (remedy)? What are the signs that she is getting better?

d. If the (observed signs) (or the woman) don't improve in (time specified by respondent) what should the woman do next?

6. Procedures for Recording the Responses

Use Form 16.1 to record the responses of the respondents. Remember, this is not an interview form. Record the information in the relevant section and probe for further details. Become thoroughly familiar with the recording form so that it is easy to find the right place to write a piece of information. If a tape recorder is being used, this component of the interview can be taped since it can be time-consuming to write the full response, particularly for more talkative respondents. If possible, a second investigator could take notes.

Be sure to record as much as possible of the additional comments the respondents make in the context of giving and discussing their advice. It may be that this exercise is a good vehicle for learning about problems in access to health care, problems of intrahousehold decision making about when and where to seek help, and other social dynamics of health seeking behavior.

Data Analysis

1. For each of the scenarios, tabulate the answers in terms of the number of times each practitioner/ health service was selected. (Form 16.2)

2. For the home care practices, create a code to classify the responses. Then, tabulate the answers for each of the scenarios (Form 16.3)

3. Analyzing Responses to Open-ended Questions

The responses to the open-ended questions can be tabulated and listed much as the answers to the "Why?" questions in the paired comparisons of Unit 7 were categorized and grouped. Also, you may wish to code them using the coding scheme.

a. Make a list of all the individual answers to each probe question. For example, the responses to the probe: "Is there anything else helpful to know about (name of woman) in order to give her advice?" could include the following: (i) "does the woman have diarrhea?" (ii) "is the husband home?"

b. Group the answers into categories that are similar in their general meaning. When in doubt, list all the answers separately.

c. Tabulate the answers and summarize by relevant classifications, such as "severity of symptoms."

Interpretation Suggestions

1. Does women's care-seeking behavior appear to vary with differing severity of illness? by differing life stages?

2. Are there clear-cut frequent answers by informants considering the scenarios? Which ones seem to be more equivocal?

DATA COLLECTION FORM 16.1
Scenario Recording Form (sample)

Scenario No. and Type: 3 mother of two children with safed paani

Diagnosis:

What the woman should do:

Try home remedies and then see an herbalist

What illness the woman has:

Safed paani

Other information respondent would like to have:

How long she has been ill? Did she eat a lot of hot foods prior to getting sick?

Home Remedies:

What should be done:

Eat cold foods

How soon to see response:

A few days

Evidence she is getting better:

White discharge will stop; bachache will stop

Care-Seeking:

Place and type of provider:

Should see an herbalist

Treatment expected:

Poultice of the leave of the Guyana tree

How soon to see response:

Maybe about 5 or 6 days

Next steps if woman does not improve:

Go to the hospital

DATA ANALYSIS FORM 16.2
Tabulation Sheet for Diagnosis with Scenarios (sample)

DATA ANALYSIS FORM 16.3
Tabulation Sheet for Home Remedy Treatments with Scenarios (sample)

Continue

Contents - Previous - Next