Tuberculosis is regarded as a disease of poverty and many aspects of low SES, for example, overcrowding and malnutrition, are accepted individual- and household-level risk factors for the disease. Inequities can be explained in terms of differences in socioeconomic status and other structural factors that influence exposure to risk, vulnerability, and the ability to recover after becoming ill [23].

As with many other diseases, the TB burden follows a clear socioeconomic gradient, with the poorest at the most elevated risk [24, 25]. This most fundamental of socioeconomic measures, poverty, is ineluctably paired with individual or household income. Yet poverty is multidimensional, including material well-being but also absence of infrastructure or a lack of input [26]. In the TB literature, markers of poverty range from individual indicators of household poverty status, to aggregate indices across geographical areas such as neighborhoods, to select populations who are considered socially vulnerable.

In Vienna in the early 1900s, the wealthiest portion of the city had a death rate from tuberculosis of 11 per 10,000 of the population; the income tax payers amounted to 25 % of the population, and the illegitimate births 0.8 per 1000, whereas in the poorest section of the city, the death rate from tuberculosis was 67 per 10,000; the income tax payers 9.2 % of the population, and the illegitimate births 9.2 per 1000 [27]. Almost a century later, the incidence of TB in central Harlem in 1990 was 32 times that of neighboring and more affluent sections of Manhattan [28]. In the United States, in both New York City and Seattle, neighborhood poverty has recently been strongly associated with TB incidence [29, 30]. Ecologic studies conducted in both the United States and Britain have found crude associations between tuberculosis rates across areas and low levels of education, high levels of poverty, less government social support, social deprivation, and income inequality [31–34]. For example, Fig. 11.2a, b shows both TB incidence rates, as well as the percentage of persons below the poverty level per zip code tabulation area (ZCTA) for an urban county in Georgia. A strong correlation is observed between the poorest ZCTAs and the highest TB incidence rates. Evidence from ecologic and multilevel studies in Brazil, South Africa, and other countries supports the existence of this relationship in middle- and low-income countries [35–41]. Finally, individual-level studies of the link between low SES and high risk of tuberculosis have found associations in poorer, high tuberculosis-burden settings [42, 43]. Similarly, the magnitude of all TB death rates has been found highest in low-household income areas, followed by middle- and high-income areas [44]. The association between infection and SES is not as clear. Tuberculin skin test positivity was least frequent in households with higher educational level, income, skilled occupations, and room size [45, 46]. Other studies have found that the risk of tuberculin skin test positivity was not associated with socioeconomic indicators [47, 48] or that infection as measured through the Quantiferon blood test actually increased with SES, possibly because wealthier homes were less well ventilated [49].

One would expect that greater density would allow for higher contact rates with an infectious individual, and thus elevated risk of disease transmission [50]. In 1996–2000, after adjusting for sociodemographic factors, Wanyeki et al. [5] used residential addresses to geocode active TB cases reported in Montreal. They found that dwelling and building features, particularly dwellings in taller and new buildings, with lower resale value, and dwellings on blocks with high residential density as well as crowding were associated with TB occurrence. Similarly, in New Zealand, TB incidence has shown to be associated with household crowding [51], and TB case rates were significantly higher in the highest crowding quartile of zip codes in the US [52]. Children living in areas of the Bronx in which over 12 % of homes were severely overcrowded were over fivefold more likely to develop active TB [53]. In First Nations communities in Canada, higher TB disease rates were observed in communities with more people living in a room (housing density) and an increase in risk of >2 TB cases occurring for every 0.1 increase in persons per room was observed [54]. While greater population density might equate with more shared air with a TB case, communities with overcrowded housing may also experience a higher prevalence of latent TB infection, or risk factors for progression from TB infection to disease. However, Myers et al. [55] found a protective effective for crowding (after adjustment for race, ethnicity, immigration, and socioeconomic factors) and no effect for population density in pediatric TB cases within California. They explained this as partly due to correlation with other variables that better explained the variability in tuberculosis cases, such as race/ethnicity, lower median incomes, and immigration. As well, to be discussed further below, crowding may be associated with a more tightly woven social network that could protect against disease [25].

Unemployment, a factor expressing lower social class, is associated with disparate TB rates as well. Among the Inuit, it has been described as one of the major determinants of risk with those on social funds or unemployed over four times more likely to have a TB infection than workers or students [56]. The greatest difference in active TB rates in British Columbia is that between employed and unemployed men [57]. Retired patients in Brazil were three times more likely to be infected with cluster-pattern strains than patients with any other occupation [58]. Occupations that have contact with infected cases (health care workers), those with silica exposure and silicosis (mining and construction), and low SES have higher TB mortality based upon National Center for Health Statistics data [59–61]. US TB case-fatality rates among unskilled white laborers were nearly seven times higher than among professional persons [62] and certain professions such as mining production are associated with elevated TB incidence rates [63]. Poor economic conditions such as self or even spousal unemployment are associated with greater health risks in general, including mortality, especially for those of working age [64].

The homeless have long been at increased risk of infection and progression to active TB due to a combination of poverty, poor nutrition, substance abuse, a lack of affordable housing, and increased exposure to public places [65, 66]. TB outbreaks among the homeless are often attributed to lack of health insurance and treatment delay [67, 68]. As early as the late 1930s, focused X-ray screenings took place in poor populations, with 47,160 homeless men screened, 2250 (5.3 %) of whom were diagnosed with active tuberculosis; additionally 1919 (2.9 %) of 65,459 Harlem relief recipients were shown with active disease [69, 70]. In 1954, X-rays of almost 2000 men who were homeless revealed a detection rate of 4 %, or more than 15 times that of the general population [71]. In landmark studies in San Francisco, nearly tenfold higher infection levels were seen among homeless people with the TB case rate among African American and other non-white homeless people 3.5 times greater than among the general population [72]. In general, incidence rates among the homeless have been difficult to assess, given the lack of a true homeless census.

Based on a thorough homeless count, Feske et al. [73] have shown that more transient housing status is linked to TB incidence that is almost 100 times the US population average, with homelessness more closely associated with social determinants of health rather than disease comorbidities in multivariate analyses. Increased genotypic clustering, a surrogate for disease transmission, had also been associated with transient housing. Additionally, given lack of other transportation options, homeless persons with TB were more likely than nonhomeless to report weekly bus ridership. Buses and other forms of public transit have been shown to be effective means of TB transmission [74–77].

There is a strong association between higher TB incidence in countries and lower gross domestic product per capital [22]. National income levels per capita and income inequality are also important predictors for TB incidence and prevalence in Europe [78].

Within many countries, the distribution of TB has been shown to be higher among the poor than the nonpoor. In the Philippines, for example, the prevalence rate of smear-positive TB was found to be 1.6 times higher in urban poor communities than in nonpoor urban communities [79]. In China, 78 % of TB patients and their families were found to have per capita annual family income lower than the average for the locality [80]. The TB mortality rate in poor rural China was found to be nearly three times higher than that in more developed urban areas. A study in northern Vietnam observed that three times as many TB patients belonged to the lowest income quintile compared to those in the general study population [81].

While not explicitly testing the poverty–TB association, higher rates of TB in refugee camps worldwide have also indirectly demonstrated the role of social conditions on likelihood of acquiring disease. For example, in Kenya, the incidence of smear-positive TB was four times greater among refugee camp residents than for the local population [82].

Worse outcomes of disease have been noted recently in varying regions of the world among the poor. For example, in Ivanovo, Russia, TB case fatality rate (during treatment) was higher among Russian Federation homeless patients than among other patients [83]. In Kenya, primarily female TB patients’ major barrier to complying with treatment protocol was financial hardship [84]. The burden from losing a whole day’s income to visit the clinic for medication resulted in lower compliance with the drug regimen.

“Environmental” social determinants, such as housing conditions, social networks, and social support, are strong drivers of TB epidemics. Molecular tools have helped to discover complex social networks through which infection spreads [85–88]. High numbers of indoor contacts and intergenerational social mixing in households and transport likely contributed to high rates of TB transmission reported in a South African community [89]. Increasing numbers of social contacts occurred throughout childhood, adolescence, and young adulthood, predominantly in settings such as schools and public transport, paralleling the increasing TB infection rates during childhood and young adulthood reported in this community [90].

On the other hand, analyses reveal strong correlations between social capital and self-rated health on the aggregate level [91]. Social support mechanisms are instrumental in influencing health-seeking behavior, adherence, and TB patient well-being [92]. Indeed, social capital was strongly correlated with decreased TB case rates at the state level in the US [25]. Social networks can also positively influence adherence to TB drug therapy [93]. Similarly, in India, social infrastructure development which led to social capital generation was associated with decreasing TB incidence rates [94].

Given its airborne transmission route, we would expect TB to predominate in indoor environments with less air exchange and poor ventilation. Indeed, in homes with poor natura l ventilation in rural South Africa, estimated TB transmission risk was quite high [95]. Similarly, the possibility of an association between household ventilation (room volume, air change rates) and TB transmission has been examined in other studies [96, 97]. Evidence from healthcare facilities indicates that natural ventilation; that is, use of open doors and windows, greatly reduces the risk of airborne transmission [98]. In general, despite seeing disparities by ventilation, there is currently lack of sufficient data on the specification and quantification of minimum ventilation requirements in relation to the spread of airborne diseases [99].

While little work has taken place to examine the salutary effects of different housing designs on TB, housing designs intended to lower TB risks are now being implemented in the rebuilding of Haiti [100]. Factors that may inhibit increased ventilation in a house include the outdoor temperature, noise, comfort, energy costs, the condition of windows or doors, or cultural and personal habits. A poorly ventilated home may suffer from high humidity and condensation, resulting in mold growth. Indeed, while not directly linked with TB infection, presence of mold may result in increased susceptibility to respiratory infection [101].

In addition to upstream determinants such as poverty, there is also increased awareness of the contribution of intermediary risk factors such as HIV, undernutrition, smoking, harmful use of alcohol, diabetes, and indoor air pollution to TB [102]. For example, children who had contact with index cases who were smokers showed a higher infection rate than those in contact with index cases who were nonsmokers [103]. These factors have all been linked to poverty, with strong associations documented at the individual level between these risk factors and poverty across medium- and low-income WHO-defined subregions [104] as well as in more urban areas of the US [105]. The population attributable fraction for TB in high-burden countries (which comprise 80 % of the global TB burden) for each of these risk factors has been estimated at between 8 and 27 % [102].

Poverty and low socioeconomic status are generally associated with worse treatment outcomes for those with TB [102] and differential access to care and health service delivery is specifically implicated in generating TB health outcome disparities, especially for the poorest and most vulnerable people [106, 107]. For example, TB patients in Georgia with lower household income were at greater risk of poor TB treatment outcomes [108]. Living in disadvantaged neighborhoods also reduces the likelihood of having a usual source of care and of obtaining recommended preventive services [109]. Limited access to care, in turn, including proximity, cost, service acceptability, and presence of public clinics and transportation, is likely to result in greater disease severity and transmission [110]. A recent systematic review showed that TB patients and households in sub-Saharan Africa often incurred costs of more than 10 % of their per capita income when utilizing TB treatment and care [111].

In a prospective cohort, longer patient delays, defined as the number of days from first TB symptoms to first medical visit specifically for those symptoms, were associated with nonwhite race and less education [112]. Longer healthcare delays, defined as days from first consultation to the initiation of treatment, were associated with presentation to a private physician or those receiving a different respiratory illness diagnosis prior to TB diagnosis. Delay in diagnosis as well as total treatment delay was associated with greater transmission of infection to contacts among US-born participants [112, 113]. Delays due to missed diagnoses among HIV-coinfected TB patients have also been documented [114]. Low educational status, living in a more rural area, and limited availability of resources such as personal finances, health insurance, time, access to qualified health workers, and social support systems are a source of delay across the globe [115–117]. In the US, the proportion of advanced pulmonary TB increased the greatest in the last 15 years among whites, the employed, and the U.S. born, implying that low-incidence groups traditionally seen as being at low risk for TB disease were in fact receiving delayed diagnoses [118]. Additionally, individuals living in lower SES areas do not necessarily have more severe pulmonary disease at diagnosis [119].

The leading causes of death and disability have a disproportionate impact on African Americans, Alaska Natives, American Indians, Asian Americans, Hispanic Americans, and Pacific Islanders and TB is no exception [120]. Indeed, dating back to the 1930s, extensive study in the United States showed that low SES was an important contributor to increased risk of TB disease among blacks in the US [121]. In a survey impressive in its scope undertaken from 1931 to 1934, the authors found that the “economically more fortunate” Ballard Normal School African Americans had more rooms per home, fewer persons per room, more windows per room, and greater home ownership, than the general Macon, Georgia African American population. As well, this group had lower prevalence of TB, less history of contact in the household, and less positive history of TB in the family [121].

To this day, in the US, TB is largely a problem among both Hispanic and black populations, with rates eight to nine times that of white populations [122] (Fig. 11.3). At the zip-code level, Hispanics and African Americans have been shown to be exposed to risk factors such as poverty and crowded housing that may facilitate TB transmission [123]. The burden of pediatric TB is largely borne among the minority population in many parts of the US—heightened transmission among US-born non-Hispanic black adults results in subsequent infection of non-Hispanic black children [18]. Hispanic ethnicity and black race also independently predict clustering in molecular epidemiologic studies in the US [72, 124]. A recent molecular analysis found that younger age, fewer years of education, use of public transportation, and inner city residence were independently associated with black race among TB patients [125].

The disproportionate burden of TB among racial and ethnic groups is largely due to differences in living and social conditions [126]. For example, adjusting for six socioeconomic indicators, low SES accounted for approximately half of the increased risk for TB among blacks, Hispanics, and Native Americans [52]. TB risk factors such as HIV, substance abuse, and homelessness are not evenly distributed across racial groups and contribute to both increased exposure and progression from latent infection to active disease [125]. Racial minorities are also more likely to be uninsured, or to have other comorbidities, increasing barriers to health-care access [127].

There is an overall paucity of high-quality data, disaggregated surveillance data that would allow for the estimation of TB case rates in different groups. A recent systematic review found that indigenous peoples in high-, middle-, and low-income countries continue to bear a high and disproportionate burden of TB [128]. Groups most burdened by TB are located in small regions of Latin America, India, and Africa and in the US; TB case rates for American Indians are more than five times greater than for non-Hispanic white people and 13 times as great among Pacific Islanders [129]. Based on surveillance data, Native Americans and Alaska Natives were also more likely than other racial/ethnic groups to be homeless, excessively use alcohol, and come from places with a greater proportion of people living in poverty and without health insurance, all of which increase risk of TB disease.

An estimated 11 % of all TB cases worldwide occur in children younger than age 15 years [130] with infection and disease acquisition common because of the high likelihood that children have frequent and close contact with adults with infectious TB. In high incidence communities, increased exposure to adults with pulmonary disease occurs to due to sociodemographic factors such as overcrowding.

The disproportionate degree to which children have TB often occurs because TB goes undetected. In most of the world, sputum microscopy is still the gold standard for TB diagnosis, yet young children generally are unable to produce a sputum sample, and if they do the sample may not be sufficient and often contains no detectable bacteria [131]. Providers thus often have to rely on clinical criteria, chest radiography, and skin testing alone [132]. Additionally, new drug development for treatment of children has lagged because of the difficulty of confirming active TB, concerns about adverse effects, complexities in involving children in drug development [133].

The WHO ’s “Towards Zero TB Deaths in Children” is advocating for viewing childhood TB in the context of a family illness, providing outreach to children with HIV, better integrating TB services with maternal and child health services, and actively reaching out and finding individuals with active TB [134]. Additionally, there has been a movement to include children in the picture when testing new rapid diagnostics and shorter, safer medication regimens that might actively benefit not only adults. Recent studies have shown that raising awareness about the risk of childhood TB among health workers and teaching them to use a scoring card for TB symptoms increases detection of childhood TB by almost three times [135]. Not only is poverty associated with increased risk of a child being exposed to TB, but it also influences risk of becoming infected and of then developing disease. Increased attention to childhood nutrition and improvement in the socioeconomic and environmental conditions of communities is likely to have a significant impact on new TB diagnoses and transmission to children [136].

Nearly twice as many men as women have been diagnosed with TB globally [137]. Men have higher mortality from TB, are at greater risk for treatment failure, and treatment default [138, 139]. Explanations for these imbalances have varied, with one hypothesis positing that gender-specific social roles may require men to have more social contact, thereby increasing TB exposure [140] and another differences in immunity due to levels of sex hormones that result in greater susceptibility among men [141]. In most cases, the observed disparities in outcomes are not due to disparities in health care [142], although undernotification of women due to greater difficulties in gaining access to care may partly explain the disparity [143].

The inmate population has been reported to have TB rates as much as 100 times higher than the civilian population [144]. One notable reason for the high rates of TB in correctional institutions is the greater proportion of persons who are at high risk for TB but who cannot access standard public health interventions such as universal HIV testing [145]. The fundamental relationship between SES and TB risk thus holds among vulnerable populations such as inmates [146]. Transmission risks particular to correctional institutions include close living quarters, poor ventilation, and overcrowding [147, 148]. Health disparities in treatment outcomes are particularly prevalent in this population. Inmates are much less likely to complete treatment [147]. This is a cause for concern both for the health of those individuals who did not receive a full course of curative therapy and for the communities in which they live. In Arkansas, it was discovered that the majority of persons (83 %) who later had TB had not received any TB screening while in jail [149]. A recent systematic review estimated the median estimated fraction of TB in the general population attributable to the exposure in prisons for TB as 8.5 % in high- and 6.3 % in middle/low-income countries [150].

Patients in behavioral high-risk groups are likely to delay seeking timely medical care and not adhere to TB treatment, causing a prolonged period of infectivity and possible outbreaks [151–153]. Transmission among frequent alcohol or drug users may be common because of the inability or reluctance of patients to share information and behavioral patterns [87, 154].

In large cities, tuberculosis mortality in both sexes has been shown to be three times higher in lowest than in the highest socioeconomic group among 35-year-olds and under, with a ratio of six to one among men over 35 [156]. A study in Denmark found that TB incidence rates in urban areas were twice as high as were incidence rates in rural areas [157] and TB in major cities has been shown to account for more than one-third of all US patients with TB [68]. Social conditions that affect urban areas such as homelessness, HIV, suboptimal access to healthcare, and migratory patterns are associated with higher TB incidence [158–160].

On the flip side, living in rural areas farther from available healthcare also matters. Individuals in indigenous First Nations communities who were more isolated and further removed from services had higher risks of incident TB [54]. Residents in rural Chin a, Vietnam, Kenya, among many other settings, have shown low case-finding rates and high rates of TB disease and transmission [161–163].

A number of studies have shown spatial clustering of TB cases, with significant associations of clustering found with social indices as well as unemployment, overcrowding, and income [30, 35, 36]. Similarly, areas that experienced the greatest child HIV/TB mortality were those without any health facility [164]. Uneven spatial distribution of cases has been documented in most continents [8, 165–167] and worldwide [168]. TB incidence in neighboring townships has even shown to have an effect on the TB incidence in a given township [169]. GIS-based screening based on multiple comorbidities, including TB, has been suggested as a tool to effectively penetrate populations with high disease burden and poor healthcare access [170]. Modeling studies have shown how high-incidence hotspots play an important role in propagating TB epidemics and the community- and city-wide impact in reducing the TB transmission rate in these hotspots [171].

Genotypic clusters of isolates often serve as surrogates for recently transmitted TB disease. A number of studies have found that molecular clusters occur in geographically distinct areas of communities and that they account for a high proportion of TB cases [172–175] (Fig. 11.4). Disparities in transmission are such that certain high transmission neighborhoods have been shown to overlap with areas of high incidence, and to include or exclude transmission across various population groups [172]. The combination of molecular and geographic tools has also been used to document community transmission of multidrug and extremely drug-resistant TB [176, 177].

In North America, groups at greatest risk for recent transmission appear to be men, persons born in the US or Canada, members of a minority race or ethnic group, persons who abuse substances, and the homeless [174, 178–181]. Based on these findings, authors have advocated for location-based control efforts for the early identification of persons with latent TB infection and undiagnosed TB cases [173, 182–184].

Migrants are disproportionately affected by TB, often due to high TB incidence in their original hometown, and the limited access to healthcare and infrastructure both on their journey and destination, as well as poverty and social exclusion in their new home [185]. Language barriers and immigration status can be additional barriers to ameliorating TB disparities and inequality [22, 186]. Migrants from rural to urban areas or across countries have delayed diagnoses that result in low treatment cure rates [187–189]. Since persons who were born in countries where TB prevalence is high might have acquired TB before immigrating [190], migrants may develop disease many years after arrival, mainly as a result of reactivation of latent tuberculosis [191]. Approximately half of new TB cases in the United States occur among foreign-born persons and the TB rate in foreign-born persons was approximately ten times that of persons born in the United States [4]. TB rates among foreign-born adolescents in the US were nearly 20 times as high as among US born [192]. An increase in the proportion of homeless who are foreign born has also been reported in Toronto, Canada [193].

Interestingly, in several settings in the US and Europe, it has been noted that disease from recent transmission of TB rarely exists among individuals born abroad but this is not true among the native-born population, where behavioral or social risk factors often predominate [194–196]. Additionally, low SES is only weakly associated with TB among foreign-born persons in the United States [197]. These findings support the hypothesis that TB rates among the foreign-born are more strongly influenced by experiences in their countries of origin than by their environments in their adopted country [198, 199]. Authors have suggested that future studies could explore the association of TB rates, SES, and country of birth based on differential immigrant settlement patterns [197].

National guidelines recommend that identification of local at-risk populations, increased knowledge of issues affecting immigrants and foreign-born persons, and modification of existing TB programs to meet the needs of these communities will help to reduce TB rates among foreign-born communities [4]. More broadly, the World Health Assembly stated that it is necessary “to formulate and implement strategies for improving the health of migrants” [200]. Blumberg has noted that in addition to focusing on the health needs of vulnerable migrant populations, the broader need is to invest in global tuberculosis control, and in development of better tools to enhance tuberculosis control [201].