SPH 200 Unit 3 All Modules Lecture Notes - Winter 2025 PDF

SPH 200 Unit 3 Module 1 Measuring the health of populations Measuring health  Measuring health Defining Historically, infectious diseases plagued populations and thus, shaped the definition of health to exclusionary, the absence of disease. As our understa...

SPH 200 Unit 3 Module 1 Measuring the health of populations Measuring health  Measuring health Defining Historically, infectious diseases plagued populations and thus, shaped the definition of health to exclusionary, the absence of disease. As our understanding of disease and ability to combat disease via medical and public health interventions advanced, health became more positive. Defined in the 1946 WHO constitution as a state of complete physical, mental and social well-being.  Measuring health state of complete physical, mental and social well-being Wellness without disease Is this as healthy as we can get? or injury Illness with disease Wellness with disease or injury or injury Illness without disease or injury  Measuring health Health can be improved, and thus it should not be measured in a finite way that does not allow for advancements in physical, mental or social determinants of health Should we agree that there should be some minimal acceptable level of health? In 1978 the WHO revised its definition, calling for a level of health that permits people to lead socially and economically productive lives.  Measuring health Quantitative and qualitative factors are clearly important to consider as health metrics Measuring health needs to include qualitative elements that will be influenced by social and cultural factors  Measuring health Despite the recognition that health is much more than just the presence or absence of disease, many of our health metrics remain disease-based Most often, we default to reporting mortality, which is an especially crude and binary measurement of ‘health’ Mortality is easy to measure, and often easy to attribute Mortality measurements can be simply death rates, but can also be presented as life expectancy data  Mortality-based measurements Life expectancy at birth together with infant mortality rates are the two most comment mortality-based measurements used Mortality-based measurements Morbidity, disability and quality measurements Consideration of morbidity and disability that impact on functioning but do not cause death The Disability-Adjusted Life Year (DALY) is a common unit of measurement for health burden as is the Years of Healthy Life Lost (YHLL) measurement In 1998 the average US life expectancy was 76.7 years, but the YHLL modifier was 11.5 years of life (limitations of major life activities). A average 65 year span of healthy life is a better way to describe years of healthy life Morbidity, disability and quality measurements 11.1 YHLL 13.1 YHLL 14.3 YHLL The morbidity and disability metrics can be applied to mortality datasets to expand comparisons of health outcomes across various descriptors And, can be used to assess the impact various diseases have on health of a population by quantifying lost healthy years due to a disease which can even be translated into an economic cost. SPH 200 Unit 3 Module 2 Measuring the health of populations  Influences on health In 1996 the US CDC included the prevalence of cigarette smoking to the list of diseases and conditions to be reported by states to the CDC This was the first time in US history that a health behaviour, rather than an illness or disease, was nationally reportable Canada followed suit in 2000 with the adoption of the Tobacco Reporting Regulations This represents a fundamental shift in the focus of public health efforts Not just reporting on on disease, but now including the underlying causes of disease Now there is data on a negative health behaviour (a risk factor) that is amenable to intervention  Risk factors Many factors can influence the health of a population Because these factors are part of a chain that leads to particular health outcomes, tracking the factors themselves can provide an indication of the direction that population health trends will move These factors are broadly termed ‘risk factors’  Risk factors Biological factors Genetic endowment, aging Environmental factors Food, air, water and exposure to infectious diseases Lifestyle factors Diet, injury avoidance, smoking Psychosocial factors Poverty, stress, personality and culture Access to health services Risk factors are often interrelated Disease Functional Economic SMOKING risk capacity strain STRESS DIET POVERTY Risk factors are often interrelated While sometimes the relationships between risk factors can be vague, other times it is very well established Unintentional injuries for example Risk factors include: Accessibility to firearms Alcohol consumption Seatbelt use Heart disease Risk factors include: Tobacco use Hypertension Overnutrition (obesity) Diabetes  Not all risk factors are equal Risk factors associated with biology, environment and behaviour are often relatively easy to attribute to specific outcomes Understanding risk associated with social and cultural factors can be more difficult, often because it can be tricky to decide on what is appropriate to measure Socioeconomic status What defines social class from a health perspective? Imprecise measurements may inaccurately estimate health differences between economic groups Absolute wealth vs relative wealth Particularly within developed nations, it is the countries with the narrowest income differentials that actually have the longest life expectancy, not the richest countries  Income differential and life expectancy The Gini coefficient is a measurement of income inequality A Gini of 0 means perfect equality where all income values within a population are the same A Gini of 1 (or 100%) means maximal inequality De Vogli et al, 2005 SPH 200 Unit 3 Module 3 Measuring the health of populations  Epidemiology The study of the distribution and determinants of of disease frequency in human populations Disease: defined broadly as ‘health outcome’ – can sometimes be clear (gunshot wounds) and sometimes less clear (diarrhea) DISEASE FREQUENCY: important to understand the number of individuals with the defined disease, as well the number within the specific population that do not have the disease Incidence: rate of new cases over time in a defined population Prevalence: total number of cases existing in a defined population at a specific time DISTRIBUTION: who, when and where WHO: characterize the disease victims (age, sex, economic status for example) WHEN: trends in disease frequency over time (is incidence increasing, decreasing or stable?) WHERE: compare disease frequencies between different locations Understanding distribution leads to knowing more about determinants  Epidemiology The study of the distribution and determinants of of disease frequency in human populations DETERMINANTS: evaluating relationships within the data categorized in who, when and where For example who and when questions clearly illustrate the link between smoking and lung cancer Males started smoking in early 20th century and lung cancer rates began to rise among the smoking male population 20 years later Females started smoking in large numbers in the 1940s and 1950s and lung cancer rates began to rise in the female smoking population in the 1960s HUMAN POPULATION: epidemiologists study human populations often using observational, rather than experimental methods The exception is intervention studies  Intervention studies Often the type of study used to test a new treatment such as a drug or vaccine Clinical trials are examples Expose one group to the treatment and another group to a control (placebo) Wait to observe whether there is a response Often the most convincing clinical trials are called randomized double-blind trials Randomized means that each participant is assigned randomly to the group Double-blind means that neither the patient nor the investigator know which group the patient is in Often this type of study is required before a new treatment is approved for use in humans  Cohort studies A type of study that allows the epidemiologist to link exposures to results by observation alone Typically… Large population required Often this group is a normal segment of the population, typically healthy but exposed to some risk factor at the beginning of the study The population is then followed over time and the researchers wait to see whether those exposed to the risk factor are more likely to develop disease than those not exposed Good for measuring the strength of an association between a risk factor and a health outcome – relative risk  Relative risk Relative risk for development of lung cancer in heavy smokers vs non smokers Exposure category Lung cancer death rate per 100,000 persons Heavy smokers 166 Nonsmokers 7 Relative risk 166/7 = 23.7 Doll and Hill, BMJ, 1956  Case control studies Start with people that are already ill and then look back to determine their exposure to a risk factor Because they can be accomplished retroactively, they are often efficient to undertake and can include smaller populations These studies compare ‘cases’ – individuals with the health outcome being measured, to ‘controls’ – healthy individuals selected to match the case group as closely as possible (age, sex and other relevant factors) Often the output is an odds ratio An estimate of what the relative risk would have been if a cohort study had been undertaken Calculated by dividing the ratio of exposed to the non-exposed in the case group by the ratio of the exposed to the non-exposed in the control group  Odds ratio Number of cases Controls Exposed 26 53 Not exposed 1 87 Total 27 140 26/1 26 x 87 2,262 = = = 42.7 53/87 53 x 1 53 SPH 200 Unit 3 Module 4 Measuring the health of populations  Studying humans is hard Hypothesis: A low fat diet reduces the risk of heart disease Intervention study (randomized control trial) One group eats an average diet, the other eats a strict low-fat diet Heart disease isn’t an acute health outcome so the timeline of this study is 5 years Will people participate for this long and adhere to the study? Cohort study Large group of people and ask them detailed questions about diets Over the next five years, compare the health of those eating low fat diet to those eating an average diet Are those eating a low fat diet already more health conscious? Case control study Chose a group of people that already have heart disease and interview them A comparable group that did not suffer from heart disease would serve as the control group Both groups questioned about their diets over the past five years Will people remember their diets that far back? Will participants respond truthfully?  Sources of error Sample size Observed association is due to chance Cause and effect may not be obvious Timeframe between exposure and outcome can be long Confounding variables Factors associated with the exposure that can independently affect the risk of developing the disease (outcome) being measured  Bias Selection bias An issue that arises during the selection of study participants for a study (case control study most often) The most satisfied or dissatisfied individuals tend to participate Cohort studies can also suffer from this form of bias as they are long term and participants can drop out Reporting bias Sometimes called recall bias – common in case control studies Occurs when the study group and control group systematically report differently, even if exposure is the same  Cause and effect Proving cause and effect is incredibly challenging In a few cases, odds ratios like 42.7 arise and suggest a very strong association, but often they are lower (1.7, 2.3) which leaves room for error to have influenced the results Epidemiological data can be strengthened if there is a known biological relationship (from a toxicology study for example) Reproducibility and consistency build confidence