Why study narrative?

Dear Sir

Your recent series on narrative based medicine [1] adds to the debate
concerning qualitative methods in medical research [3]. However, this
useful contribution belies the emerging paradigm in qualitative research
which has elevated the status of the spoken word (be it self-report, focus
group, narrative or interview). Participant observation and more direct
ethnographic data collection methods data are rarely part of the modern
qualitative researcher's repertoire [4].

Yet, direct observation and description of events by skilled
researchers provide contextual data which augments our understanding of
social phenomena. One prime example is that of AIDS and injecting drug
use, where observation of injecting practices has fine-tuned health
promotion and prevention interventions, contributing to reducing the
spread of infection [4]. We need to ensure ethnography's unique
contribution is not side-lined in favour of less direct methods of data
collection.

To illustrate these points I refer to the evaluation of an
intervention run by Gay Men Fighting AIDS (GMFA). GMFA provided condoms
and other health promotion materials to men using an open air public sex
environment (PSE) in London. One practical problems was public complaints
about the amount of "litter" created by men discarding used condoms and
other materials. In response, regular "litter picnics" were held, where
volunteers scoured the area, clearing the debris. Myself and another
researcher participated in one of these events. Each volunteer was given a
black plastic bag (and one rubber glove!). Previous participant
observation sessions took place at night. But, in broad daylight, it
became apparent that here was a rich source of data. After two hours
rubbish collecting we randomly selected one of the ten bags and collated
the contents.

Table 1: Materials recovered from the litter bag
used condoms as distributed by GMFA (n=47);
torn GMFA condom packs (77);
other used condom brands (4);
other torn condom packs (45);
GMFA lubricant packs (57);
other lubricant packs (6);
large quantities of GMFA "green" tissue paper;
sundry items (including 6 used amyl nitrite bottles and items of
clothing).

These data, serendipitously collected during an observation session,
strengthened the process evaluation, confirming and augmenting findings
from self-reports and interviews. In this light, five conclusions were
drawn. One, the "litter collection" data indicated sex was taking place in
the area. Two, GMFA packs were being distributed and reaching the target
group. Three, the identifiable contents of the GMFA packs (condoms, water-
based lubricant, tissues) were being used in situ. Four, condoms and water
-based lubricants were being used by men at the PSE. Five, men were not
merely "relying" on GMFA provision, but were bringing to the area (and
using) other brands of condoms and lubricants. A sixth, and more general,
conclusion could be that there is more to qualitative research than the
spoken and recorded word.

[1] Greenhalgh T, Hurwitz B: Why study narrative? Brit Med J 1999,
318: 48-50.

[2] Pope C, Mays N: Reaching the parts other methods cannot reach: an
introduction to qualitative methods in health and health services
research. Brit Med J 1995, 311: 42-45.

[3] Denzin N, Lincoln Y. Handbook of qualitative research. London:
New Left Books; 1975.

[4] Grund J, Kaplan C, Adriaans N, Blanken P: Drug sharing and HIV
transmission risks: the practice of frontloading in the Dutch injecting
drug user population. J Psych Drugs 1991, 23: 1-10.

Yours faithfully

Robert Power PhD
Senior Lecturer in Medical Sociology
Department of Sexually Transmitted Diseases
Royal Free and University College Medical School
The Mortimer Market Centre
Mortimer Market
London WC1E 6AU