Effect of Cross-level Interaction between Individual and Neighborhood Socioeconomic Status on Adult Mortality Rates
Abstract overview:
This study is particularly important because it is a 17-year follow up to determine whether the socioeconomic level of a neighborhood influences death rates differently for women and men who have different levels of income or education (measures of socioeconomic status, or SES). It is a population-based study that followed 4,476 women and 3,721 men aged 25-74 from 82 neighborhoods in four California cities: Monterey, Salinas, Modesto, and San Luis Obispo. Careful note was made of the availability of goods and services in the various neighborhoods, as well as the geographic boundaries that characterized each neighborhood. The large sample size, survey data on participants’ risk factors, and long-term follow-up made this study one of the most rigorous to date to examine the relationship between neighborhood SES and the death rates for both high and low SES residents.
The study had two major findings: (1) Unexpectedly, death rates for low SES women were highest for those living in high SES neighborhoods, lower for those in moderate SES neighborhoods and lowest for those in low SES neighborhoods. The mortality rates for men followed a similar pattern. (2) The health behaviors and risk factors of the low SES study participants, such as obesity and smoking, did not vary substantially for low income people across neighborhoods and therefore do not appear to explain the disparities in death rates. Additionally, living close to neighborhood amenities that might enhance health— like primary care physician offices, grocery stores, parks, or schools— did not vary across neighborhoods for low-income people.
The specific death rates for low SES women in high SES neighborhoods were 1907 per 100,000 person years after accounting for age. This compared with death rates of 1323 for low SES women in moderate SES neighborhoods and 1128 for women in low SES neighborhoods. The death rates for low SES men were 1928, 1646, and 1590 respectively in high, moderate and low SES neighborhoods. The consistency of the results, which were found in each of the four cities studied and among white non-Hispanic respondents (the largest group in the sample), adds legitimacy to the findings.
The authors used survival curves to estimate at what time during the 17-year study follow-up mortality differences first became apparent. At the beginning of the follow-up, there were few differences when people died. But between 10 and 15 years of follow-up, changes became apparent. Low SES women and men in high SES neighborhoods began to fare poorly at this mark, and the probability of dying at 10-15 years of follow-up was about 12% for low SES women in high SES neighborhoods. This compared with 7% and 3% for low SES women in moderate and low SES neighborhoods, respectively. For low SES men, the probability of dying was about 10%, 8% and 7% in high, moderate, and low SES neighborhoods, respectively.
These findings echo the death rate findings of two smaller studies, one in the United States and one in Canada.
The authors considered two possible explanations for the findings: Given the increased living expenses necessary to maintain residence in a higher income neighborhood, lower income people living in high SES neighborhoods may have less disposable income and/or time to take advantage of the goods and services that surround them, such as healthier foods, good doctors, and private exercise facilities. They may also have to work longer, and have less time to maintain or adopt healthy behaviors like regular exercise or healthy meal preparation. Therefore, their relative deprivation may be greater than those living around them.
Psychosocial explanations may also account for the study findings. It has long been suggested that the discrepancy between an individual’s social position relative to others in his or her community may influence risk of death. Low social position may be associated with fewer resources to cope with stressful life events, lack of social support, and/or low sense of control which may result in real or perceived social isolation, discrimination, and/or other psychosocial stressors. If lower income people in higher income neighborhoods see how little they have compared with their neighbors, this may affect their health, either indirectly by influencing health behaviors, or directly by influencing neuroendocrine or immune functioning.
Secondary findings of the study:
Death rates for moderate and high SES women and men showed no clear pattern according to the neighborhood in which they lived, but their death rates were substantially lower than death rates for low SES women and men.
Implications and recommendations from the study:
It is hoped that the higher resources and knowledge generally associated with higher SES neighborhoods would reach all residents, regardless of their incomes or educations. The findings suggest that this may not be happening. There is a demonstrated need for health professionals to understand the neighborhoods they serve and to be aware of the specific health concerns of all residents.
While public health strategies should continue to focus on the lowest SES adults in lowest SES neighborhoods -- those experiencing the greatest deprivation -- this study suggests a need to focus attention on low SES adults living in more advantaged neighborhoods who may represent a potentially “hidden” high-risk group of people.
Proposed news articles and related information:
- What can policy makers, health professionals, and community residents do so higher income neighborhoods better serve those residents who are poorer or have less education?
- What do doctors and other health professionals in higher income neighborhoods need to be aware of when treating their patients, especially in regard to those with lower incomes or educations?
