The study was undertaken in Khayelitsha, a periurban settlement in Cape Town, which has grown substantially because of recent rural-to-urban migration (7). Population estimates vary from 330,000 to 1 million (21,22). Most residents live in poverty: 55% live below the poverty line compared with the Cape Town average of 25% (21), and although some find employment, 41% are economically inactive (21), surviving on the marginal informal economy (8).

The School of Public Health at the University of the Western Cape initiated the Khayelitsha noncommunicable disease prevention initiative in 2000 after being approached by community health workers who were concerned about the levels of diabetes and hypertension in their community. The first task was to work with the community health workers to enhance their understanding of the risk factors and determinants of noncommunicable diseases. The community health workers then led a prevention intervention that included public information events, group fun walks that facilitated safe outdoor activities, and screening for high blood pressure and diabetes. In 2005, weekly health clubs were established to broaden the community focus (13,23). The initial focus of the clubs was lifestyle change through discussions, cooking demonstrations, and exercise sessions. The aim, however, was to expand the focus to take account of the broader determinants of noncommunicable diseases. To achieve this, a clearer understanding of the community's circumstances was deemed necessary.

Data were collected in 2 phases — a quantitative study described elsewhere (22) and a qualitative study, which is the focus of this article. The qualitative study was necessary to understand the respondents' perceptions about the circumstances and effect of their migration. We used 3 approaches to enhance the data collection and allow for triangulation. These approaches were in-depth interviews (conducted in February and March 2007), participatory action and reflection groups (described below, conducted in June 2007), and focus group discussions (conducted in September 2007). All components of the qualitative study took place in the community.

We used purposive sampling to identify 55 people (45 women and 10 men for the study). (Women were not deliberately oversampled.) Each respondent participated in only 1 activity, and no one declined to take part. We selected 10 respondents from the quantitative study for the interviews (3 of whom were club members). The community health workers selected the remaining respondents from their programs. The only selection criteria were that all respondents had to have lived in Khayelitsha for 5 years or less, and some respondents had to be club members. Given this sampling procedure, some respondents would have had some knowledge of noncommunicable disease prevention.

One bilingual author led all 3 approaches so that she could build on the data and probe for clarification as required. The local language, isiXhosa, was used throughout. We used in-depth interviews to gain an initial understanding, focusing on changes in diet and physical activity, what helped participants cope, and the barriers they faced. The interviews averaged 45 minutes to 1 hour.

We then conducted 2 participatory action and reflection groups: 1 of club members (10 people) and 1 of nonmembers (15 people). Because the focus of the researcher was on facilitating the groups, we created a more dynamic, respondent-led environment. These groups lasted from 1.5 to 2 hours and consisted of 3 activities: Venn diagrams, which explored social interactions and sources of information; ranking and sorting, which discussed coping mechanisms; and problem-tree analyses, which probed for reasons for weight gain. Finally, we led two 45-minute focus group discussions, 1 of club members and 1 of nonmembers (10 people in each). These discussions maximized respondent interaction, focusing on social integration, challenges, and reasons for weight gain. The interviews and focus groups were recorded for accuracy, transcribed, and translated by 2 people independently. A researcher checked the accuracy of the discussions with the respondents at the end of each session. We identified themes and priorities through content analysis of the transcripts and the notes and charts from the action and reflection groups. No statistical software was used. All respondents consented to take part and knew they could withdraw without repercussion. We obtained ethical clearance from the University of the Western Cape.

Overwhelmingly, people left a life of poverty and hard work to come to Khayelitsha. Most were farm workers, and women — who made up most respondents — also fetched water, carried firewood, and attended to their homes and families. Job security was a major concern because options were limited beyond the agricultural sector.

Access to health services was limited, and transportation was difficult.

Part of the lure of the city was the perception of an easier and more comfortable lifestyle, and for those who found employment, circumstances improved. Typically, however, circumstances remained difficult.

Furthermore, participants were concerned about personal safety, which limited how much people were willing to venture beyond their homes. Social integration was also more difficult than they had anticipated. Several respondents described being lonely and isolated on arrival and missing friends and family.

Despite these concerns, most respondents were pleased to be in the city. This seemingly contradictory view related to the relative ease of their lives and their growing integration into a new community.

Initial contacts with family expanded to include new friendships and broader social networks. Neighbors — an accessible source of local knowledge — were particularly valuable. Respondents also learned about formal institutions such as health services and the church, which provided support and information.

The respondents were learning to negotiate a new environment with opportunities that did not exist in rural areas. They were becoming part of an urban society that, while difficult, provided diversity and the potential for personal development. These achievements created a sense of well-being and satisfaction.

The adoption of an urban lifestyle affected the diet and physical activity of the respondents. They ate more often and spent less time preparing food. However, the food they ate was often high in sugar and fat. Food choices were limited by budget and the poor range of food available locally. The exposure to, and desire for, a more urban diet, including fast food, was often perceived as progress from rural diets. Food could also be bought on credit from local street vendors.

Physical activity, part of daily life in the rural areas, was now reduced, because of a combination of reality and because of their current circumstances. Although physical activity was considered to be a requirement or disadvantage of their past life, respondents no longer perceived it as necessary. Basic amenities were more accessible, and for people with jobs, work was less strenuous. The fear of violence during outside activity was also prevalent, particularly among women.

Walking also became a choice instead of a necessity because other transportation was available.

Respondents self-reported a substantial amount of weight gain, but weight gain was not a concern. On the contrary, weight gain was seen as symbolic of their success in the city. A loss in weight indicated problems in one's life. A frequent comment was that respondents were overweight because they were happy.

Respondents who attended health clubs appreciated the information about risk factors of diabetes and hypertension because they were aware of the increase in health problems in their community.

But the information was difficult to act on. Apart from practical constraints, family members were an obstacle because they resisted giving up the lifestyle they had become accustomed to. Family members were also tempted by the elevated status of obesity. The discussions in the club helped people address these dilemmas and provided role models of people who had lost weight. The clubs also provided social and emotional support, and the club members described a sense of personal growth and belonging.