Table 1 Description of the variables of socioeconomic stratification and control

 Maximum educational level of parents^a

Highest educational level of parents.

Ordinal variable: 1) university degree or higher; 2) bachelor’s degree or technical career; 3) some level of high school; and 4) some level of primary school or uneducated.

 Monthly family income level^a

Monthly family income reported in the Register of Childhood Cancers, converted into United States dollars (USD), on December 31, 2015, with an exchange rate of 20.6194 pesos per dollar [31]. Ordinal variable: grouped in quartiles of monthly income, 1) richest population to 4) poorest population.

 Availability of basic services in the home^a

Characteristics present in the domicile: potable water; indoor toilet (seat, cover, flushing system); and cement or finished floor. Dichotomous variable: 1) with all three characteristics; 0) absence of one of the characteristics.

 Maximum occupational level (specialization of parents^a)

Groupings based on the International Standard Classification of Occupations (ISCO) 2008 [32]. Ordinal variable:

1) Managers and professional occupations (includes ISCO group 1: managers and ISCO group 2: professionals);

2) Mid-level and support occupations: (includes ISCO group 3: Technicians and associate professionals and ISCO group 4: Clerical support workers);

3) Services, sales and skilled workers (includes ISCO group 5: Service and sales workers, ISCO group 6: Skilled agricultural, forestry and fishery workers, and ISCO group 7: Craft and related trades workers);

4) Machine operators; unskilled laborers and unemployed (includes ISCO group 8: Plant and machine operators and assemblers and ISCO group 9: Elementary occupations).

Proportion of minor’s life path that was IMSS-insured prior to ALL diagnosis ^b

Proportion of days of minor’s life path (from date of birth to date of diagnosis of cancer) during which the minor was insured by social security.

Ordinal variable: 1) 80–100%; 2) 50–80%; 3) 25–50%; and 4) < 25% of minor’s life path.

 Age group at ALL diagnosis^a

Years between minor’s date of birth and date of diagnosis.

Ordinal variable: 1) < 1; 2) 1 to < 5; 3) 5 to < 10; and 4) 10 to < 15 years old.

 Sex^a

Dichotomous variable: 1) male; 0) female.

 ALL risk diagnosis ^a

Presence of clinical characteristics at the time of diagnosis of acute lymphoblastic leukemia [23]. For high-risk diagnosis: age < 1 year or > 10 years; hyperleukocytosis; immunophenotype of T cells; and any infiltration, mediastinal mass, or Down syndrome. For standard-risk diagnosis: absence of all the characteristics of high-risk diagnosis.

Dichotomous variable: 1) high risk diagnosis; 0) standard risk diagnosis.

 Time lag in diagnosis^a

Months between parents’ awareness of symptoms and date of diagnosis in a tertiary hospital [26]. Ordinal variable: 1) < 1 months; 2) 1 to 4 months; and 3) ≥4 months.

 Diagnosis cohort^a

Year in which diagnosis was confirmed by a specialist. Ordinary variable: 2007, 2008, 2009.

 Proportion of treatment period of IMSS-insured minor (after ALL diagnosis) ^b

Proportion of the treatment period of the minor (from date of cancer diagnosis to five-year follow-up) during which the minor was insured by social security (adjusted for date of death, when required).

Ordinal variable: 1) 75–-100%; 2) 50 to < 75%; 3) 25 to < 50%; and 4) < 25%.

 Area of residence prior to ALL-diagnosis^a

Area where patient resided prior to the date of diagnosis.

Dichotomous variable: 1) Metropolitan (Mexico City and State of Mexico), 0) Provincial (states of Chiapas, Guerrero, Hidalgo, Morelos, and Querétaro).

 Level of marginalization of the municipality of residence^a

Marginalization, estimated by CONAPO [27] for the area of residence on the date of diagnosis, is based on the percentages of the following: population (≥15 years old) that is either illiterate or did not complete primary school; occupants in homes A) without indoor plumbing or toilet, B) without electric energy, C) without piped-in potable water, D) with some level of overcrowding, or E) with dirt floors; populations in locations with < 5000 inhabitants; and populations earning up to two minimal salaries. Ordinal variable: 1) very low or low; 2) medium; and 3) high or very high.

 Network of services^a

Medical service network corresponding to the location of family residence. Dichotomous variable: 1) SXXI service network covers the populations from the states of Chiapas, Guerrero, Morelos, Querétaro, and the southern part of Mexico City; 0) La Raza service network covers population of the states of Hidalgo, México, and the northern part of Mexico City.

 Distance from tertiary hospital^c

Distance (km) from the medical unit of primary care to the tertiary hospital that corresponds to the residence of patient. The distance was calculated by using Google maps [28] to determine the shortest route for a privately owned car. Ordinal variable): 1) < 20; 2) 20 to < 50; 3) 50 to < 200; and 4) ≥200 km.

 Type of secondary hospital^d

Type of resources available at the secondary hospital. The “Hospital General de Subzona” (HGS), has the four basic specialties plus an emergency department, 30–72 beds. The “Hospital General de Zona” (HGZ), has the same services as the HGS, as well as 72–144 beds and other specialties, such as trauma, ophthalmology, otorhinolaryngology, and subspecialties. The “Hospital General Regional” (HGR), provides medical attention to the population referred from the HGS and to some patients from the HGZ, provides the basic specialties and various subspecialties, and has > 200 beds. Ordinal variable: 1) HGR; 2) HGZ; and 3) HGS.

 Distance from secondary hospital^c

Distance (km) from primary-level medical unit to secondary hospital corresponding to location of patient’s residence. The distance was calculated by using Google maps [28] to determine the shortest travel distance for a privately owned car. Ordinal variable: 1) < 5; 2) 5 to < 10 km: 3) 10 to < 20 km; and 4) ≥20 km.