Epidemiological & Biostatistics Exercises:

a) Demographic indices (Morbidity, Mortality & Fertility rates)

Q1) In a population of 10,000 . There were 40 new cases of measles in the year 2018. 10 died
among new measles cases. Calculate.
a) Incidence in 2018
b) Case fatality rate.
Ans:
a) Incidence =
No. of new cases (of disease at a given time)/ (Population at risk during that period) x
1000
Population at risk = 10000
Incidence = 40/10000 x 1000
= 4/ 1000 population per year

b) Case fatality rate = Total No. of deaths (due to disease)/ (No. of cases of same disease) x
100
= 10/40 x 100
= 25%

Q2) A PHC catering 30,000 population has given the data about tuberculosis from January 2021
to December 2021

Particulars Tuberculin +ve Sputum +ve

Old cases 11,160 75
New cases 540 25
Total 11,700 100

Calculate the relevant epidemiological indices for sputum + ve and write the validity of
these indices.
Ans : Epidemiological indices of tuberculosis
Incidence of disease = Number of new sputum positive cases x1000
Total population
= 25 x 1000
30,000
= 0.8 per thousand population per year
Prevalence of disease= Number of sputum positive(old+new) casesx100
Total population
 =75+25x100
30,000
=100 x100
30,000
= 0.3 percent

Q3) The given figure depicts the occurrence of tuberculosis cases at different time perods in a
population of 300. Assuming the population is same at any point of time, calculate the
following rates for Tb:
a) Point prevalence on 1st Jan 2023
b) Point prevalence 31st December 2023
c) Period prevalence in year 2023
d) Incidence in year 2023

Solution :
a) Point prevalence on 1st Jan 2023
(No of TB cases on 1st Jan/ population on 1st Jan)*100
No of TB cases on 1st Jan = case 1, case 3 and 9 = 3 cases
Population on 1st Jan = 300(at any point of time)
Point prevalence on 1st Jan 2023 = (3/300)*100 = 1%
 b) Point prevalence on 31st Dec 2023
(No of TB cases on 31st Dec / population on 31st Dec)*100
No of TB cases on 31st Dec = case 1, 4,5 and 7 = 4 cases
Population on 31st Dec = 300(at any point of time)
Point prevalence on 31st Dec 2023 = (4/300)*100 = 1.33%

c) Period prevalence in year 2023=

(No of new and old cases of Tb in year 2023 / population at the start of 2023)*100
No of new cases = case 2, 4, 5,6 , 7 and 8 = 6 cases
No of old cases = case 1, 3 and 9= 3 cases
Period prevalence in year 2023= ((6+3)/300)*100=3%

d) Incidence in year 2023=

(No of new cases of Tb in year 2023 / population at at risk at the start in year
2023)*1000
Thus incidence in year 2021 =( 6/300)*1000= 20%

b) Mortality Problems
Q4) A slum with a population of 40000 experienced an outbreak of cholera in 2023. 800 cases
were recorded and there were 16 deaths due to cholera. There were 400 deaths due to all
causes in that year calculate the following

a) Cause specific death rate.

b) Proportional mortality rate.
c) Crude death rate.
d) Case fatality rate of cholera

ANS. Total Population of slum in 2023 = 40,000

Total Cholera Cases = 800

Total Deaths due to Cholera = 16
Total deaths = 400
 a) Cause Specfice Death Rate = Total deaths due to a disease in a yr x 1000

Mid-year population

= 16 / 40000 X 1000 = 0.4 deaths per 1000 population

b) Proportional Mortality Rate = Total deaths due to a disease in a yr x 100

Total deaths in that area during same time

= 16 / 400 X 100 = 4%

C ) Crude Death rate = Total No of deaths x 1000

Mid-year Population

= 400/ 40000 x 1000 = 10 deaths per 1000 population

d) Case Fatality Rate = Total deaths due to a disease in a yr x 100

Total Cases of that disease

= 16/ 800 x 100 = 2

Q5) Given data for Telangana in 2023: Total deaths: 10,000. Intepret various PMRs
a) Deaths due to Tuberculosis (TB): 500
b) Deaths due to Cancer: 1,500
c) Deaths due to Coronary heart disease: 2,000
d) Deaths due to Road traffic accidents: 1,000
e) Deaths due to Malaria: 100
Solution:To calculate the Proportional mortality rate for each cause, we use the formula
𝑝𝑟𝑜𝑝𝑜𝑟𝑡𝑖𝑜𝑛𝑎𝑙 𝑚𝑜𝑟𝑡𝑎𝑙𝑖𝑡𝑦
= ( 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑑𝑒𝑎𝑡ℎ𝑠 𝑑𝑢𝑒 𝑡𝑜 𝑡ℎ𝑒 𝑐𝑎𝑢𝑠𝑒 ÷ 𝑡𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑑𝑒𝑎𝑡ℎ𝑠)
× 100
a) Tuberculosis(TB):
Poportional mortality rate for TB = (500/10,000)×100=5%
 b) Cancer:
Poportional mortality rate for Cancer=(1,500/10,000)×100=15%
c) Proportional mortality rate for Coronary heart disease=(2,000/10,000)×100=20%
d) Poportional mortality rate for Road traffic accidents=(1,000/10,000)×100=10%
e) Poportional mortality rate for Malaria=(100/10,000)×100=1%
Therefore, the proportional mortality rates for each cause of death in Telangana in the year 2023
are as follows:
• Tuberculosis (TB): 5% , Cancer: 15% , Coronary heart disease: 20% , Road traffic accidents:
10% , Malaria: 1%
• The major cause of deaths in Telangana is CHD and least is Malaria
Q6) In a study conducted in a certain region, the total population was found to be 1,00,000
people. Over the course of a year, a total of 500 deaths were recorded. Among these deaths,
100 were attributed to heart disease, 150 to respiratory diseases, 50 to cancer, and 30 to
diabetes, with the remaining attributed to various other causes.
Calculate the cause-specific mortality rates for heart disease, respiratory diseases, cancer, and
diabetes per 1000 individuals in the population for the specified year.
Ans: formula: Cause-Specific Mortality Rate = (Number of deaths from specific cause / Total
population) * 1000
Given data:
Total population = 1,00,000
Deaths from heart disease = 100
Deaths from respiratory diseases = 150
Deaths from cancer = 50
Deaths from diabetes = 30

Calculations
Heart disease: Cause-Specific Mortality Rate = (100 / 1,00,000) * 1000 = 1 deaths per 1000
individuals
Respiratory Diseases: Cause-Specific Mortality Rate = (150 / 1,00,000) * 1000 = 1.5 deaths
per 1000 individuals
Cancer:
Cause-Specific Mortality Rate = (50 / 1,00,000) * 1000 = 0.5 deaths per 1000 individuals
Diabetes:
Cause-Specific Mortality Rate = (30 / 1,00,000) * 1000 = 0.3 deaths per 1000 individuals
These rates indicate the number of deaths for each cause per 1000 individuals in the
population for that year
 Q7) Mortality observed in villages of India, Singapore and USA is given. Calculate the
proportational mortality rate and comment

Total deaths 0-5 yr death

Singapore 400 10

India 450 143

USA 50 1

Solution:

Proportional mortality rate( 0-5yrs)= number of deaths in 0-5 yrsX 100

Total number of deaths
Proportional mortality rate of Singapore = 10/400 X 100= 2.5%

Proportional mortality rate of INDIA = 143/450 X 100= 31.75%

Proportional mortality rate of USA = 1/50 x 100 = 2%

Inference: Proportional mortality rate of under-fives in India is approximately 15 times higher

than in Singapore and USA.
Q8) In a town with midyear population of 2 lakhs there were 6050 live births, 110 still births, 250 deaths
within 1 week after birth and 480 deaths in 1st month of life and 750 deaths in 1st year of life.
Calculate Perinatal Mortality Rate (PMR) , Infant mortality rate (IMR)
Ans: a) Perinatal mortality Rate = (Still births+Early neonatal mortality rate)/(Live births +still
births ) x 1000
= (110+250)/(6050+110) x 1000
= 360/6160 x 1000 = 58.4
PMR = 58.4 / 1000 live Births

a) Infant mortality Rate = No. of deaths (within 1year age)/(Total No.of live births) x
1000
= 750/ (6050) x 1000
= 0.1239 x 1000
IMR = 123.9 / 1000 live births
Q9) In a community, with a mid-year population of 136,000 there were 3576 live births in year
2020. Following were the number of deaths:
No. of fetal deaths (≥28 weeks gestation) = 24
Deaths in less than 7 days of birth = 39
Deaths during 7-28 days of birth = 34
Deaths during 29 days to less than 1 year of age = 58
Maternal deaths = 7
Deaths in the under-five age group = 170
Calculate Maternal mortality ratio, Still birth rate, Perinatal mortality rate, Neonatal
mortality rate, Infant mortality rate and Under 5 mortality rate
Ans:

a. MATERNAL MORTALITY RATIO = No. of maternal deaths x 100,000

No of live births

= (7/3576) x 100,000 = 196 per 100,000 live births

b. STILLBIRTH RATE = No. of still births (late fetal deaths) x 1000

No of live births + stillbirths

= [ (24) / (3576 + 24)] x 1000 = 6.7 deaths per 1000 total births

c. PERINATAL MORTALITY RATE = No. of perinatal deaths (late fetal deaths + deaths in less
than 7 days of birth) x 1000

No of live births + stillbirths

= [(24 + 39) / (3576 + 24)] x 1000 = 17.5 deaths per 1000 total births

d. NEONATAL MORTALITY RATE = No. of neonatal deaths (within 28 days of birth) x 1000
No. of live births
= [(39 + 34) / 3576] x 1000 = (73/3576) x 1000 =20.4 deaths per 1000 live births
 e. INFANT MORTALITY RATE = No. of infant deaths ( less than 1 year) x 1000
No. of live births
= [(39 + 34 + 58) / 3576] x 1000 = 36.7 deaths per 1000 live births
f. UNDER 5 MORTALITY RATE = No. of deaths among children less than 5 years x 1000
No. of live births

Q10) In a town with mid-year population of 150,000 following vital events occurred
Total live births 3,200
Total deaths : 1,400
Infant deaths : 270
Maternal deaths: 10
Calculate (1) Crude birth rate (2) Crude death rate (3) Infant mortality rate and (4) Maternal
mortality ratio. What is the Infant Mortality Rate and Maternal Mortality Ratio of India
according To SRS 2020?
Solution:
Crude birth rate (CBR) = Number of live births during the year in the town / Mid-year
population of the town x 1,000
= 3,200 / 150,000 x1,000
= 21.3 per thousand population per year

Crude death rate (CDR) = Number of deaths in given place during the year / Mid-year
population of the same place and time x 1000
= 1,400 / 150,000 x1000
= 9.3 per thousand population per year

Infant mortality rate (IMR) = Number of deaths of infants under 1 year of age / Number of
live births x 1000
= 270 / 3,200 x1000
= 84.1 per thousand live births
Maternal mortality ratio (MMR) = Total number of deaths of women during pregnancy delivery
or within 42 days after delivery in a given year / Total number of live births in that area x 1,00,000
=10*100000/3200
= 311/lakh live births per year
 Sample registration system in India is the source of Infant Mortality Rate and Maternal
Mortality Ratio in India.
Latest report was published in 2020,
According to which, Infant mortality rate in India is 28 Infant Deaths per thousand live births
and Maternal Mortality Ratio is 97 maternal deaths per 1,00,000 live births

c) Fertility rates
Q11) A primary health centre with 30,000 population, gives the following data of 1 year

Age groups Number of women Number of live births in 1

year
15-24 2000 500
25-34 1800 250
35-44 1400 90
Total 5200 840

Calculate general fertility rate, age specific fertility rate, Total fertility rate.
Solution:
General Fertility Rate:
= No. of Live Births in the year / No. of Women of (15-44yrs) x 1000
= 840 / 5200 x 1000 = 161.5 per 1000 women

Age-specific fertility rate: No. of live births in specific age / Mid-year population of same
age x 1000

15-24 years, 500 / 2000 x 1000 = 250

25-34 years, 250 / 1800 x 1000 = 138.8
35-44 years, 90 / 1400 x 1000 = 64.3

Total fertility rate (TFR): Sum of all age-specific fertility rates x 5 (International standard) /
1000

453.1 / 1000 x 5 = 2.26

Thus, a woman would have 2.26 children if she passes the same fertility rate in each age
group.
Q12)Following are the statistics for a city with MYP of 2,10,000 for the year 2021.Assuming that there
is no mortality in women of the reproductive age group, calculate Age Specific Fertility Rates and
Total Fertility Rates.
Indicate whether this population has achieved replacement-level fertility? Compare this Total
Fertility Rates with India’s Total Fertility Rates According To SRS 2020.

Age Group (Years) Mid-Year Population of Women Total Live Births

15-19 9,400 710
20-24 9,300 1,250
25-29 8,500 1,450
30-34 8,300 790
35-39 7,950 770
40-44 5,950 560
45-49 6,600 370
Total 56,000 5,900
Solution:
TFR = 5 x Σ ASFR for live births (from age 15-19 to 45-49) / 1000

ASFR = No. of live births in a specific age group in a year / Mid-year population of women
in the same age group x 1000

ASFR for women aged 15-19 years = (710/9400) x 1000 = 75.5 per 1000 women
ASFR for women aged 20-24 years = (1250/9300) x 1000 = 134.4 per 1000 women
ASFR for women aged 25-29 years = (1450/8500) x 1000 = 170.6 per 1000 women
ASFR for women aged 30-34 years = (790/8300) x 1000 = 95.2 per 1000 women
ASFR for women aged 35-39 years = (770/7950) x 1000 = 96.9 per 1000 women
ASFR for women aged 40-44 years = (560/5950) x 1000 = 94.1 per 1000 women
ASFR for women aged 45-49 years = (370/6600) x 1000 = 56.1 per 1000 women

TFR = 5 x (75.5 + 134.4 + 170.6 + 95.2 + 96.9 + 94.1 + 56.1) / 1000

TFR = 5 x 722.8 / 1000 = 3.6

TFR = 3.6 live births per woman in the reproductive age group.

To achieve replacement-level fertility, TFR should be 2.1, but for the given population, it is
3.6; thus, the population has not achieved the replacement-level fertility. As per SRS 2020,
 TFR of India for 2021 is 2.0; thus, the TFR of the given population is higher than India’s
TFR.
Q13) The following are the statistics for a city for the year 2023

Age Group Mid-Year Population of Total Female Live

(Years) Women Births
15-19 10,000 350
20-24 9,000 630
25-29 8,700 710
30-34 8,300 380
35-39 8,000 370
40-44 6,000 260
45-49 5,000 150
Total 55,000 2,850

Calculate Gross reproduction rate (GRR) assuming that there is no mortality in women of the
reproductive age group. Compare this GRR with India's GRR According to SRS 2020.

Solution:
GRR= 5 x Σ ASFR for female live births (from age 15-19 to 45-49) / 1000
For this, we need to calculate ASFR by using the number of female births in different age
groups:
Here ASFR = No. of female live births in a specific age group / Mid-year population of women
in the same age group x 1000
ASFR for women aged 15–19 years = (350/10,000) × 1000 = 35 per 1000 women
ASFR for women aged 20–24 years = (630/9000) × 1000 = 70 per 1000 women
ASFR for women aged 25–29 years = (710/8700) × 1000 = 81.6 per 1000 women
ASFR for women aged 30–34 years = (380/8300) × 1000 = 45.8 per 1000 women
ASFR for women aged 35–39 years = (370/8000) × 1000 = 46.3 per 1000 women
ASFR for women aged 40–44 years = (260/6000) × 1000 = 43.3 per 1000 women
ASFR for women aged 45–49 years = (150/5000) × 1000 = 30 per 1000 women

GROSS REPRODUCTION RATE = 5 × (35 + 70 + 81.6 + 45.8 + 46.3 + 43.3 + 30) / 1000
= 1760 / 1000
= 1.76
= 1.76 female live births per woman in the reproductive age group
 As per sample registration system (SRS), GRR of India for 2020 is 0.9; thus, the GRR of given
population is higher than India’s GRR.

Q14) In the year 2022, the MYP of town A was 83,000 including 20,000 women in the reproductive
age group. There were total 2500 live births in the same year. Calculate Crude birth rate and
GFR. Compare these with CBR and GFR of India.
Solution:
CRUDE BIRTH RATE = No. of live births in an area in a given year / Mid-year population × 1000
= (2500 / 83,000) × 1000 = 30.1 live births per 1000 mid-year population
CRUDE BIRTH RATE = 30.1 live births per 1000 mid-year population
GFR = No. of live births in an area in a given year / Mid-year population of women in the
reproductive age group × 1000
= (2500 / 20,000) × 1000
= 125 live births per 1000 women in the reproductive age group
GFR = 125 live births per 1000 women in the reproductive age group.
As per SRS, CRUDE BIRTH RATE of India for 2020 is 19.5 live births per 1000 mid-year population
and GFR is 67 Live Births per 1000 women thus, both indicators in given population are higher
than India's CRUDE BIRTH RATE and GFR.
 d) Infectious disease epidemiology

Q15) A routine clinical survey for filariasis was carried out in a community health center,
serving 1 lakh population; data collected is as follows:

Night blood smears collected 30000.

Persons showing only mf positive 300.

Persons showing signs of filariasis 80.

Persons showing both mf positive and signs 10.

Calculate the possible filarial indices. And suggest control measures.

Ans. CALCULATION OF FILARIAL INDICES

number showing mf positive

Microfilaria rate (mf) = number of persons slides examined × 100

300+10
= × 100 = 1.03%
30000

number showing filarial disease symptoms

Filarial disease rate = number of persons examined
× 100

80+10
= 30000 × 100 = 0.3%

signs + number of mf positives + Both

Filarial endemicity rate = number of persons examined
× 100

80+10+300
= × 100 = 1.3%
30000

CONTROL MEASURES:

The current strategy of filariasis control is based on

1. Chemotherapy

2. Vector control
Chemotherapy

Drug – Diethylcarbamazine (DEC) – 6 mg/kg/day divided doses after meal.

Duration – 6 days in a week for 2 weeks, i.e. 12 days

Total dose: 72 mg/kg

FILARIAL CONTROL IN THE COMMUNITY

i) Preventive chemotherapy

Albendazole 400mg + Ivermectin (150-200 mcg/kg)

Or

Albendazole 400mg + DEC (6mg/kg)

The above drugs are given annually to an entire high-risk population for about 4-6 years.

ii) Selective treatment

In areas of low endemicity DEC is given only to those who are Mf positive

iii) DEC Medicated salt: Common salt medicated with 1- 4 gm of DEC/Kg

VECTOR CONTROL

Antilarval measure:

A) Chemical methods: 1. Mosquito larvicidal oil (MLO), 2. Pyrosene oil – E, 3. Organophosphorus

larvicides (e.g. temephos, fenthion)

B) Removal of Pistia plant: Mansonia mosquitoes can be controlled.

C) Minor environmental measures: filling up of ditches and cesspools, drainage of stagnant

water etc...

Anti – adult measures: Pyrethrum as a space spray, using mosquito nets.

Q16) In the year 2022, a subcentre under a PHC covering a population of 5000 reported a total
of 600 fever cases which were examined for malaria parasites. 400 were positive among
which 50 were falciparum cases & the rest were vivax cases.

Calculate API, ABER, Slide positivity rate, pf% and SFR.

Ans.
confirmed cases during one year
API : × 1000
Population under surveillance
400
= × 1000 = 80
5000

𝑁𝑜.𝑜𝑓 𝑠𝑙𝑖𝑑𝑒𝑠 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒 𝑓𝑜𝑟 𝑚𝑎𝑙𝑎𝑟𝑖𝑎 𝑝𝑎𝑟𝑎𝑠𝑖𝑡𝑒

SPR: × 100
𝑁𝑜.𝑜𝑓 𝑠𝑙𝑖𝑑𝑒𝑠 𝑒𝑥𝑎𝑚𝑖𝑛𝑒𝑑
400
= 600 × 100 = 66.6 %

𝑁𝑜.𝑜𝑓 𝑠𝑙𝑖𝑑𝑒𝑠 𝑒𝑥𝑎𝑚𝑖𝑛𝑒𝑑

ABER: 𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑢𝑛𝑑𝑒𝑟 𝑠𝑢𝑟𝑣𝑒𝑙𝑙𝑖𝑛𝑎𝑐𝑒 × 100
600
= 5000 × 100 = 12%

Plasmodium falciparum percentage (pf%)

𝑇𝑜𝑡𝑎𝑙 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒 𝑓𝑜𝑟 𝑝. 𝑓𝑎𝑙𝑐𝑖𝑝𝑎𝑟𝑢𝑚
× 100
𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑙𝑖𝑑𝑒𝑠 𝑠ℎ𝑜𝑤𝑖𝑛𝑔 𝑝𝑎𝑟𝑎𝑠𝑖𝑡𝑒𝑠
50
= × 100 = 12.5%
400
𝑆𝑙𝑖𝑑𝑒 𝑝𝑜𝑠𝑖𝑡𝑖𝑣𝑒 𝑓𝑜𝑟 𝑓𝑎𝑙𝑐𝑖𝑝𝑎𝑟𝑢𝑚 𝑚𝑎𝑙𝑎𝑟𝑖𝑎 𝑝𝑎𝑟𝑎𝑠𝑖𝑡𝑒
Slide falciparum rate (SFR) = × 100
𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠𝑙𝑖𝑑𝑒𝑠 𝑒𝑥𝑎𝑚𝑖𝑛𝑒𝑑
50
× 100 = 8.3%
600
Q17) In a population of 50 children, 10 children are immunized against chicken pox. 5
Children developed chicken pox on 1st March 2021. Other 28 children developed chicken pox
after 2 weeks. Calculate the attack rate and Secondary attack rate of chicken pox.
Ans:
𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑛𝑒𝑤 𝑐𝑎𝑠𝑒𝑠 𝑜𝑓 𝑎 𝑠𝑝𝑒𝑐𝑖𝑓𝑖𝑒𝑑 𝑑𝑖𝑠𝑒𝑎𝑠𝑒 𝑑𝑢𝑟𝑖𝑛𝑔 𝑎 𝑠𝑝𝑒𝑐𝑖𝑓𝑖𝑒𝑑 𝑡𝑖𝑚𝑒 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙
Attack rate = × 100
𝑇𝑜𝑡𝑎𝑙 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑎𝑡 𝑟𝑖𝑠𝑘 𝑑𝑢𝑟𝑖𝑛𝑔 𝑡ℎ𝑒 𝑠𝑎𝑚𝑒 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙
28
× 100 = 70%
40
𝑁𝑜.𝑜𝑓 𝑒𝑥𝑝𝑜𝑠𝑒𝑑 𝑝𝑒𝑟𝑠𝑜𝑛𝑠 𝑑𝑒𝑣𝑒𝑙𝑜𝑝𝑖𝑛𝑔 𝑡ℎ𝑒 𝑑𝑖𝑠𝑒𝑎𝑠𝑒 𝑤𝑖𝑡ℎ 𝑖𝑛 𝑡ℎ𝑒 𝑟𝑎𝑛𝑔𝑒 𝑜𝑓 𝑖𝑛𝑐𝑢𝑏𝑎𝑡𝑖𝑜𝑛 𝑝𝑒𝑟𝑖𝑜𝑑
SAR = × 100
𝑇𝑜𝑡𝑎𝑙 𝑛𝑜.𝑜𝑓 𝑒𝑥𝑝𝑜𝑠𝑒𝑑 𝑜𝑟 𝑠𝑢𝑠𝑐𝑒𝑝𝑡𝑖𝑏𝑙𝑒 𝑐𝑜𝑛𝑡𝑎𝑐𝑡𝑠

Primary cases should be excluded from both numerator and denominator.

10 children are immunized so, they are not susceptible.
No. of susceptible contacts are 50-10 = 40
Denominator should not include primary cases so, in the given formula Total no. of susceptible
contacts are 40-5 = 35
28
SAR= 35 × 100 = 80%

Q18) In an area with total population of less than five-year age group is 200, 5 measles cases
reported on 10th February,2020. In next 10 days another 25 children developed measles in the
same area. After history taking it was found that 20 children were immunized against
measles. Calculate the secondary attack rate (SAR). Give the outlines of measles outbreak
control.
Ans.
𝑁𝑜.𝑜𝑓 𝑒𝑥𝑝𝑜𝑠𝑒𝑑 𝑝𝑒𝑟𝑠𝑜𝑛𝑠 𝑑𝑒𝑣𝑒𝑙𝑜𝑝𝑖𝑛𝑔 𝑡ℎ𝑒 𝑑𝑖𝑠𝑒𝑎𝑠𝑒 𝑤𝑖𝑡ℎ 𝑖𝑛 𝑡ℎ𝑒 𝑟𝑎𝑛𝑔𝑒 𝑜𝑓 𝑖𝑛𝑐𝑢𝑏𝑎𝑡𝑖𝑜𝑛 𝑝𝑒𝑟𝑖𝑜𝑑
SAR = × 100
𝑇𝑜𝑡𝑎𝑙 𝑛𝑜.𝑜𝑓 𝑒𝑥𝑝𝑜𝑠𝑒𝑑 𝑜𝑟 𝑠𝑢𝑠𝑐𝑒𝑝𝑡𝑖𝑏𝑙𝑒 𝑐𝑜𝑛𝑡𝑎𝑐𝑡𝑠

Primary cases should be excluded from both numerator and denominator.

Children who are immunized to measles should not be included.
25
= 175 × 100 = 14.2%

Human is the only natural host for measles and there is no carrier state.
Transmission of measles is by droplet infection.
Period of communicability is 4 days before and after appearance of rash.
In case of an outbreak, first rapid search for cases should be done and source of infection
should be identified.
• All the cases should be isolated till 7 days after appearance of rash.
• Immunization status of all the children in that area should be checked.
• To the susceptible active immunization within 2 days of exposure (if vaccine is
contraindicated, immunoglobulin should be given within 3-4 days of exposure)
• Prompt immunization at the beginning of an epidemic is essential to limit the spread.
• This vaccine gives protection only given within 3 days of exposure.
• All cases of severe measles should be treated with vitamin A, a high dose should be
given immediately after diagnosis, and repeated next day. <6 months of age-50,000 IU;
6-11 months -1,00,000 IU; >12 months -2,00,000 IU. Child with clinical signs of vit. A
deficiency should be given third dose 4-6 weeks later.
 e) Vaccination and nutrition related problems

Q19) A Mandal PHC of 30,000 Population with CBR 20, IMR 40/1000 live
births, has to implement UIP. As a medical officer, calculate the vaccine requirement.

Ans: As a medical officer, I would like to calculate:

i. No. of pregnant mothers = total population * CBR =

30,000*20/1000 = 600

ii. No. of infants = total population * birth rate (1- IMR) =

30,000*20/1000 (1-0.04) =600*0.96 =576

iii. Estimation of vaccine requirement = no. of doses required

iv. No. of doses required = no. of eligible* no of doses* wastage factor

• Wastage factor for DPT, TT, HEP-B & OPV = 1.33

• Wastage factor for MEASLES & BCG = 2 iv.

• No. of TT doses = 600*2*1.33 = 1596

• No. of BCG doses = 576*1*2 = 1152

• No. of OPV doses = 576*3*1.33 = 2298.24 = 2299

• No. of DPT doses = 576*3*1.33 = 2299

• No. of MEASLES doses = 576*1*2 = 1152

• No. of HEP-B doses = 576*3*1.33 = 2299

v. No. of doses in each vial = BCG = 20

MEASLES = 10

DPT=10

POLIO=20 , HEP-B=10
 vi. No. of vials of BCG required =57.6 = 58

• No. of vials of DPT required =229.9 = 230

• No. of vials of MEASLES required= 115.2= 116

• No. of vials of OPV required = 114.95 = 115

• No. of vials of HEP-B required= 229.9= 23

Q20) A vaccine trial was conducted among 6000 children; of these, 3000 children received the
vaccine and 3000 received the placebo. After a specific period of follow up, it was found that
400 children developed the disease among the vaccinated group while 1030 children developed
the disease among the placebo group. Calculate the vaccine efficacy.

Ans:
Incidence in Unvaccinated – Incidence in Vaccinated
Vaccine Efficacy = -------------------------------------------------------------------- * 100

Incidence in Unvaccinated

Disease No Disease Total

Vaccinated 400 (a) 2600 (b) 3000 (a+b)
Unvaccinated 1030 (c) 1970 (d) 3000 (c+d)

c/c+d - a/a+b

Vaccine Efficacy = ------------------------------ × 100

c/c+d

1030/3000 – 400/3000 630

Vaccine Efficacy = ---------------------------------------- × 100 = ------------- × 100

1030/3000 1030

Vaccine Efficacy = 61.16%

Q21)Prescribe a balanced diet for a 50 year old diabetic patient who is a sedentary worker &

whose BMI is 32.5.

Ans: Balanced diet for a 50 YO diabetic sedentary person

• Adult Men-Sedentary- 2110 Kcal , Moderate work - 2719Kcal,

Heavy work - 3470Kcal

• Avoid Foods with high glycemic index like:

White rice, potatoes, white bread, candy, Sweets, sweetened beverages

a. Carbohydrates: Moderate consumption, Processed and refined carbohydrates should be

avoided.

b. Protein: Protein can be obtained from lean sources such as chicken, turkey, fish, legumes,

nuts, and seeds.

c. Healthy Fats: Healthy fats should be incorporated in the diet such as nuts, seeds,

avocados, olive oil, and fatty fish.

d. Fiber: Foods high in fiber such as whole grains, fruits, vegetables, and legumes should be

consumed regularly.

e. Vitamins and minerals: Diabetic patients may have a higher risk of nutrient deficiencies,
so a variety of colorful fruits and vegetables should be included in the diet.

* Split meals: Divide the whole calorie requirement of the day into 5-6 meals or

snacks(nuts,fruits,sprouts) in a controlled and planned manner.

 f) Measures of Central Tendency and Dispersion
Q22) Mid arm circumference of 11 male children aged 18 months are given below:
12, 14, 11, 12, 13, 15, 12, 13, 15, 14, 12
Calculate mean, median, mode.

x
Solution: a) Mean x = () n
143
= = 13 cm
11

()
Mean x = 13 cm

𝑛+1 𝑡ℎ
b) Median = ( ) value
2

11 + 1 12
= =
2 2
= 6th value
Median = 13 cm

c) Mode = frequently recurring number in the series

= 12 cm.
Q23) If in a moderately asymmetrical frequency distribution, the values of median and mean are 72
and 78 respectively, estimate the value of the mode.
Solution : Mode = 3 Median – 2 Mean
= 3 X 72 – 2 X 78
= 216 – 156 = 60.
Mode = 60
Q24) Following are the weights of 10 medical students
60, 40, 50, 56, 48, 64, 72, 38, 80, 76

Calculate measures of Dispersion

Solution: Measures of Disperssion

a) The Range = Maximum Value –Minimum Value

Maximum Value =80 Minimum Value = 38
Range = 80-38 = 42
b) Mean Deviation = Average of the deviation from Arithmetic Mean = ∑(X-X) / n = 120/
10 = 12
Mean = 584 / 10 = 58.4

Wt in kg’s(x) Arithmetic Mean (X) Deviation from theMean(x-X) (x-X) 2

60 58.4 1.6 2.56

40 58.4 -18.4 338.56
50 58.4 -8.4 70.56
56 58.4 -2.4 5.76
48 58.4 -10.4 108.16
64 58.4 5.6 31.36
72 58.4 13.6 184.96
38 58.4 -20.4 416.16
80 58.4 21.6 466.56
76 58.4 17.6 309.76
Total = 584 Total = 120 Total = 1934.4

c) Standard Deviation = √∑(X-X)2 / n-1

= √1934.4/ Ƞ -1 =√1934.4/10-1

= 14.66
 Q25) In a clinical study comparing the efficacy of two different insulin dosages for patients with
Type 2 Diabetes, the blood glucose levels before and after treatment were recorded for two
groups. Group A received a standard insulin dosage, while Group B received a higher dosage. The
mean blood glucose levels for the two groups before treatment were 4 mmol/L for Group A and
3 mmol/L for Group B, with standard deviations of 1.6 mmol/L for both groups.
Considering the variability in blood glucose levels before treatment, which group exhibits greater
variability in response to the insulin dosage?

Solution: Coefficient of Variation (CV) = SD x 100

Mean

GRP A CV = 1. 6 X 100 = 40
4

GRP B CV = 1. 6 X 100 = 53.33

3
Grp B is more variable than Grp A.

Q26) Find 95% Confidence limits for mean height (cm) of 625 children of age 18 years studied.
Details given Mean: 172.4; S.D: 6.25; Sample Size (n): 625.
𝑆⋅𝐷
Solution : ̅̅̅̅̅ +
𝑥 − 1.96 ( 𝑛 ) ,
√
6.25
= 172.4 +
− 1.96 ( )
√625

∴ 171.91 < 172.4 < 172.89

Q27) In a group of 100 under five children attending pediatric OPD, the mean weight is 15kg. The
standard deviation is 2. In what range 95% of children’s weight will lie in the sample?
Solution: Range in which 95% children’s weight in the sample will lie: 95% reference range = mean +/2SD

̅̅̅̅̅ +
𝑥 − 1.96 SD

= 11-19Kg
Q28) Suppose 2 classes of 40 and 50 strength have mean Biostatistics marks of 52 and 61
respectively. Find out the combined mean of the 2 groups.

Solution:

Given data are N1=40, N2=50, 𝑥̅ 1= 52, 𝑥̅ 2= 61

̅𝟏+𝐍𝟐 . ̅𝒙𝟐
𝐍𝟏 .𝒙
Combined Mean = = 5130/90
𝐍𝟏+𝐍𝟐

Then Mean of 2 groups = 57.

Q29) In a school health check-up, the hemoglobin level was estimated in 300 children. Data is given
below.
Hb level (gm%) Number of children
6–8 150
9 – 11 140
12 - 14 10
Total 300

Calculate the mean Hb level of the school children.

Solution:
Class Intervals Midpoint of class Frequency (F) Cumulative (mF)
interval (m)
6–8 7 150 1050
9 – 11 10 140 1400
12 – 14 13 10 130
Total N = 300 ∑mf = 2580

Mean Hb of the group = ∑mF / N = 2580 / 300

= 8.6 gm%.
 g) Epidemiological study designs

COHORT STUDY

Q30) In a study to assess the relationship between birth weight and infant mortality. Following
data was captured 800 babies among 5400 weighed below 2500 grams at birth. They were
followed up till they completed their first birthday. Out of 5400 infants, 460 infants deaths were
registered. In which 320 deaths occurred among babies with Normal birth weight.
a. What type of study is this?
b. Construct a 2 x 2 table
c. Calculate relative risk and attributable risk
d. Interpret the results
Ans. a. Cohort study
b. 2 x 2 table
Cases- Infant death Control- No infant
death
Low birth weight 140 (a) 660 (b) 800 (a+b)
Normal birth 320 (c) 4280 (d) 4600 (c+d)
weight
460 (a+c) 4940 (b+d) 5400
(a+b+c+d)

c. Relative risk = Incidence rate among exposed

Incidence rate among non-exposed
= a/(a+b)
c/(c+d)
= 140 x 4600/ 320 x 800 = 2.52

Attributable risk (AR)

(I R among exposed – I R among unexposed)

= ------------------------------------------------------- X 100
I R among exposed
 (a/a+b)-(c/c+d) (140/800)-(320/4600)
= ------------------ X 100 = -------------------------X100
a/a+b 140/800
0.175-0.0695
=----------------------------X100 = 60.1%
0.175
d. Interpretation:
i. Low birth weight infants are 2.52 times at greater risk of developing infant
mortality than normal weight infants or the incidence of infant mortality is 2.52
times higher in low-birth-weight infants than in normal weight infants.
ii. 60.1% of infant mortality among infants was due to their low birth weight.

Q31) In a cohort study, out of 60 Hypertensive 30 were using OCP. Among 70 non-hypertensives,
25 were using oral contraceptives.
a. Construct a 2 x 2 table
b. Calculate the Incidence rates among exposed and unexposed , Relative risk and
Attributable risk.
c. Comment on the results.
Ans. a. 2 x 2 table
Cases- Hypertensive Control- Non Hypertensive
OCP (using) 30 (a) 25 (b)
OCP (not using) 30 (c) 45 (d)
60 (a+c) 70 (b+d)

b. Incidence Rates

(i) Among OCP users = a/(a+b) x 100

= 30 /55 X 100
= 54.5%
54.5% OCP Users developed Hypertension
 (ii) Among OCP Non-users = c/(c+d) x 100
= 30/ 75 X 100
= 40%
40% of OCP Nonusers developed Hypertension

Relative risk = Incidence rate among exposed

Incidence rate among non-exposed

= 30/55 = 0.5 = 1.25

30/75 0.4

Attributable risk (AR)

(I R among exposed – I R among unexposed)

= ------------------------------------------------------- X 100
I R among exposed
(a/a+b)-(c/c+d) (30/55)-(30/75)
= ------------------ X 100 = -------------------------X100
a/a+b 30/55
0.5-0.4
=----------------------------X100 = 20%
0.5
c. Comment:
i. Incidence rate of Hypertension is higher among OCP users compared to OCP
Nonusers.
ii. OCP users are 1.25 times at greater risk of developing HTN than OCP nonusers
or the incidence of HTN is 1.25 times higher in OCP users than in OCP non-
users.
iii. 20% of hypertension among OCP users was due to their OCP usage
CASE CONTROL STUDY

Q32) In a matched case control study of 200 cases of lung cancer and 200 non lung cancer cases
as controls were included. Out of 200 cases 170 having history of smoking for more than 5
cigarettes a day and out of 200 controls only 20 having history of smoking more than 5 cigarettes
a day

a. Construct a 2x2 table and

b. Calculate exposure rate, odds ratio and comment

Ans.

Cases (with Controls (without Total

lung cancer) lung cancer)
Smokers (> 5 170 (a) 20 (b) 88 (a+b)
cigarettes a day)
Non -Smokers 30 (c) 180 (d) 29 (c+d)

200 (a+c) 200 (b+d) 400 (a+b+c+d)

Exposure rates

i. Cases = a/a+c =170/200 = 0.85 = 0.85 percent

ii. Controls = b(b+d) = 20/ 200 = 0.1 = 10 percent

Odds ratio = ad/bc

= 170 x 180

20 X 30

= 51

Interpretation – Lung cancer patients were 51 times more likely to have smoked > 5 cigarettes
per day when compared to controls
Q33) In a study, out of 100 diabetic 50 were obese and among 100 non-diabetics 10 were obese by
using this data:

Diabetes Non Diabetes Total

Obese 50 (a) 10 (b) 60 (a+b)

Normal 50 (c) 90 (d) 140 (c+d)

100 (a+c) 100 (b+d) 200 (a+b+c+d)

Solution:

Exposure rates

i. Cases = a/a+c =50/100 = 0.5 = 0.50 percent

ii. Controls = b(b+d) = 10/ 100 = 0.1 = 10 percent

Odds ratio = ad/bc

= 50 x 90

50 x 10

=9

Interpretation – Diabetes people are 9 times likely to be obese than normal people.
 h) SCREENING RELATED PROBLEMS

Q34) The results obtained after screening 1000 population for diabetes mellitus with random
blood glucose estimations are as follows-

Screening Test- Diabetes mellitus Total

RBS
+ve -ve
Positive 7 10 17
Negative 8 975 983
Total 15 985 1000

Calculate sensitivity, specificity, Positive predictive value, negative predictive value, false
negative percentage and false positive percentage.

• Sensitivity = a/ (a+c) X 100 = 7 / 15 X 100 = 46.67%

• Specificity = d / (b+d) X 100 = 975 / 985 X 100 = 98.9%
• Positive predictive power = a / (a+b) X 100 = 7 / 17 X 100 = 41.18%
• Negative predictive power ==d/ (c+d) x 100 =975/(8+975) = 99.1%
• False negative percentage = (c/a+c) x 100 or 100-Sensitivity in % =(8/7+8) x 100 = 53.33%
• False positive percentage = (b/b+d) x 100 or 100-Specificity in % =(10/10+975) x 100 =
1.01%
COMMENT: RBS as a screening test for DM is highly specific test and hence there is very
little chance that a normal person being dubbed as diabetic. RBS is a low sensitivity test
and hence, more than half of the actual diabetics will not be detected with this test.
Q35)A new screening test for a certain disease was administered to 480 persons, 60 of whom are
known to have the disease. The test was positive in 50 of the persons with disease as well as in
20 persons without the disease.
• Construct a 2x2 table?
• Calculate Sensitivity, Specificity, Positive predictive value?
Ans.
Screening Test Disease Total
+ve -ve
Positive 50 (a) 20 (b) 70 (a+b)
Negative 10 (c) 400 (d) 410 (c+d)
Total 60 420 (b+d) 480 (a+b+c+d)
(a+c)

• Sensitivity = a/ (a+c) X 100 = 50 / 60 X 100 = 83.33%

• Specificity =d / (b+d) X 100 = 400 / 420 X 100 = 95.23%
• Positive predictive power = a / (a+b) X 100 = 50 / 70 X 100 =71.43%
 i) Hypothesis testing & Tests of Significance

(T test, Z test – test of means)

Z Test for Single Mean & Two Means (Sample Size > 30)

Q36)In a city, the mean Hb level of the women population was 11.5 g%. From the same city, mean
Hb level of a sample comprising 300 women was 11.1 g%, with SD of 1.8 g%. Find whether the
difference between mean Hb level of the sample and the population is significantly different.
(Tabulated value of test statistic at 5% level is 1.96.)

Solution

Null hypothesis: There is no difference in mean Hb level between sample & population.

𝑥̅ −µ
Z = 𝑆𝐷
√𝑛

µ = population mean = 11.5

𝑥̅ = sample mean = 11.1
SD = 1.8
N = 300
𝑆𝐷 1.8
Standard error of mean = = = 0.1039
√𝑛 √300

̅̅̅̅̅̅̅̅
| 11.1−11.5 | 0.4
Z= = 0.1039 = 3.849
0.1039

• Calculated Z value is more than the table value.

• Interpretation: Reject null hypothesis – There is a statistically significant difference in
the mean Hb level of the sample & the population.
Q37)Mean weight of rural girls (N=105) was 58.6 kg with SD-8.3 and that of urban girls (N=101)
was 56 kg with SD-6.2. Is the difference in weight between rural and urban girls significant?

(Tabulated value of test statistic at 5% level is 1.96)

Solution

Null hypothesis: There is no difference in weight between urban & rural girls.

𝑥̅ 1 – 𝑥̅ 1
Z=
𝑆𝐸 𝑂𝑓 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑜𝑓 𝑚𝑒𝑎𝑛

𝑥̅1 = mean weight of rural girls = 58.6

𝑥̅2 = mean weight of urban girls = 56
N = 101
𝑆𝐸 𝑂𝑓 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑜𝑓 𝑚𝑒𝑎𝑛 = Standard error of the difference of mean

𝑆𝐷12 𝑆𝐷22
𝑆𝐸𝑥1−𝑥2 = √ +
𝑛1 𝑛2

= √0.65 + 0.38 = √1.03

̅̅̅̅̅̅̅̅
| 58.6−56 | 2.6
Z= = 1.03 = 2.52
1.02

• Z value @ 5% level is 1.96

• Calculated Z value is more than the table value.
• Interpretation: Reject null hypothesis – There is a statistically significant difference in
the weight between urban & rural girls
 j) STUDENTS t TEST/INDEPENDENT t TEST/UNPAIRED T TEST - TWO INDEPENDENT MEANS

Q38)In a study, mean weights of subjects in two groups A and B were 59.2 and 56.5 kg with SD of 2.9
and 3.6, respectively. Group A had 17 subjects, whereas group B involved 15 subjects.Find out
whether the weights of these two groups differ significantly. Assume that the distribution of
weight is normal In both groups. (Tabulated value of the test statistic at required d.f. is 2.042.)

Ans:

Null hypothesis: There is no significant difference in mean weight between group A & group
B

Group A: 𝑥̅1 = 59.2; SD1 = 2.9; n1 = 17

Group B: 𝑥̅2 = 56.5; SD2 = 3.6; n2 = 15

𝑥̅ 1 – 𝑥̅ 2
t = 𝑆𝐸
𝑂𝑓 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑜𝑓 𝑚𝑒𝑎𝑛

𝑆𝐷12 𝑆𝐷22
𝑆𝐸𝑥1 −𝑥2 = √ +
𝑛1 𝑛2

2.92 3.62
= √ 17 + = √0.49 + 0.864 = √1.354 = 1.16
15

59.2−56.5
t= = 2.32
1.16

• Inference: Calculated t value is more than table value. Hence null hypothesis is
rejected.
• There is a statistically significant difference in mean weight between group A & group
B
 k) PAIRED TEST FOR DIFFERENCE BETWEEN TWO DEPENDENT MEANS (BEFORE AND AFTER
MEANS)
Q39)The following table presents the mean cumulative weight loss in grams for 12 patients receiving
propranolol and for 11 control patients following sweating during insulin induced hypoglycaemia.
Degree of freedom is 21 (t value for 21 degrees of freedom at 1% level is 2.831).

n Mean wt. loss in g SD

Propranolol 12 120 10.0
Control 11 70 8.0
n1 = 12
n2 = 11
𝑥
̅̅̅1 = 120 𝑥2 = 70
̅̅̅
SD1 = 10 SD2 = 10

𝑥̅ 1 – 𝑥̅ 2
t = 𝑆𝐸
𝑂𝑓 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑜𝑓 𝑚𝑒𝑎𝑛

𝑆𝐷12 𝑆𝐷22
𝑆𝐸𝑥1 −𝑥2 = √ +
𝑛1 𝑛2

102 82
= √ 12 + 11 = 3.76

120−70
t= = 13.29
3.76

• So, t value of 21 degrees of freedom @ 1% level = 2.831.

• Calculated value is more than table value.
• Null hypotheses is rejected.
• The difference in weight with propranolol is statistically significant
Q40) Mean gain in haemoglobin level after iron supplementation was 1.86 gm/dl with S.D of 1.10
gm/dl. Numbers of children under experiment were 14. Check the inference. (Test at 1% 0.s.
3.01).
Ans:
Null hypothesis (H0)= there is no significant mean gain in Hemoglobin level after iron
supplementation.

Alternate hypothesis (H1) = there is significant mean gain in Hemoglobin level after
iron supplementation.

|𝑑̅| |1.86|
Test Statistic (t) = ( 𝑆⋅𝐷 )=( 1.10 )
√𝑛 √14

= 1.86/0.2939 = 6.32
Conclusion: Reject null hypothesis at 1% l.o.s.
 l) Chi-square test

Q41) From the following data, use 𝑋 2 -test and conclude whether inoculation is effective in
preventing tuberculosis.

GROUP ATTACKED NOT ATTACKED TOTAL

Inoculated 10 90 100
Not inoculated 26 74 100
Total 36 164 200

(The table value of 𝑋 2 at α= 0.05 with 1 degree of freedom is 3.84)

Ans. Let the Null hypothesis be that inoculation is not effective in preventing TB infection.
An alternate hypothesis is that inoculation is effective in preventing TB infection. The expected
frequencies are tabulated below.

GROUP ATTACKED NOT ATTACKED TOTAL

O E O E
Inoculated 10 (A) 18 90 (C) 82 100
Not inoculated 26 (B) 18 74 (D) 82 100
Total 36 164 200
𝑅𝑜𝑤 𝑡𝑜𝑡𝑎𝑙×𝑐𝑜𝑙𝑢𝑚𝑛 𝑡𝑜𝑡𝑎𝑙
Expected frequency = 𝐺𝑟𝑎𝑛𝑑 𝑡𝑜𝑡𝑎𝑙
100×36
Expected frequency of cell (A) = = 18
200
100×36
Expected frequency of cell (B) = = 18
200
100×164
Expected frequency of cell (C) = = 82
200
100×164
Expected frequency of cell (D) = = 82
200
Calculation of 𝑋 2 :

CELL OBSERVED EXPECTED (O − E)2 (O − E)2

FREQUENCY (O) FREQUENCY 𝐸
(E)
A 10 18 (−8)2 64⁄18= 3.555
B 26 18 82 64⁄18= 3.555
C 90 82 82 64⁄82= 0.780
D 74 82 (−8)2 64⁄82= 0.780
Total (O − E)2
∑
𝐸
= 8.67
(O−E)2
Assuming that 𝐻0 is true, the test statistic is 𝑋 2 = ∑ = 8.67
𝐸

d.f. = (2-1) (2-1) = 1

Given that the table value of 𝑋 2 at α= 0.05 with 1 degree of freedom is 3.84
Since 8.67 > 3.84, the null hypothesis is rejected and conclude that inoculation is effective.

Q42) The distribution of 40 individuals by presence or absence of cancer and smoking habits
is given below: (𝑿𝟐 – value at a = 0.05 with 1 d.f. is 3.84)
Disease Smokers Nonsmokers Total
Cancer 8 (A) 2 (C) 10
Non cancer 4 (B) 26 (D) 30
Total 12 28 40
Test whether there is any significance association between cancer and smoking.
Solution: The null hypothesis, 𝐻0 : no association between cancer and smoking
The alternate hypothesis 𝐻1 : there is an association between cancer and smoking.
The expected frequency corresponding to each cell is calculated as
10×12
E (A) = 40
=3
10×28
E (C) = 40
=7
30×12
E (B) = 40
=9
30×28
E (D) = 40
= 21
Here, E (8) = 3 Which is less than 5. Hence the Yate’s correction for continuity is applied i.e.,
(|O−E|−0.5)2
𝑋2 = ∑ 𝐸

(|8−3|−0.5)2 (|2−7|−0.5)2 (|4−9|−0.5)2 (|26−21|−0.5)2

𝑋2 = ∑ 3
+ 7
+ 9
+ 21

20.25 20.25 20.25 20.25

𝑋2= 3
+ 7 + 9 + 21

𝑋 2 = 6.75 + 2.89 + 2.25 + 0.96 = 12.85

d.f. = (2-1) (2-1) = 1
The 𝑋 2 – value at a = 0.05 with 1 d.f. is 3.84
2
Since 𝑋 2 calculated = 12.85 > 𝑋0.05 = 3.84
Therefore 𝐻0 Is rejected and conclude that there is a significant association between cancer and
smoking (P < 0.05)

Q43) Among 200 alcoholics, 50 have developed cirrhosis. Among 300 non-alcoholics, 50 have
developed cirrhosis in a cohort study. Find out is there any association between alcohol and
cirrhosis.

(𝑿𝟐 – value at a = 0.05 with 1 d.f. is 3.84)

a. Formulate 2*2 table.
b. Define null hypothesis.
c. Do Chi-Square test & interpret.
a.
Cirrhosis -ve
Alcohol 50 (A) 150 (C) 200
b. Non-alcohol 50 (B) 250 (D) 300
100 400 500

Null hypothesis: There is no significant association between alcohol & cirrhosis.

 c. Expected table:
200∗100 200∗400
(A) = 40 (C) = 160
500 500
300∗100 300∗400
(B) = 60 (D) = 240
500 500

∑(𝑂−𝐸)2
X2 =
𝐸

(50−40)2 (150−160)2 (50−60)2 (250−240)2

= 40
+ 160
+ 60
+ 240

= 2.5 + 0.625 + 1.67 + 0.416

X2 = 5.211

df = (r-1) (c-1) = 1

• Table value for df=1 is 3.84

• Calculated value > table value
• Reject null hypothesis and accept alternate hypothesis
• There is significant association between consumption of alcohol and
cirrhosis

Q44). Out of 2500 diarrhoea cases with mild dehydration, 1500 were treated with homemade ORS
and rest 1000 were treated with low osmolar ORS, 300 did not recover from each group. Find out
if there is any difference between the treatment outcomes in the two interventions. (𝑿𝟐 – value
at a = 0.05 with 1 d.f. is 3.84)
a. Formulate 2*2 table.
b. Define null hypothesis.

c. Do Chi-Square test & interpret.

a.
Recovered Not recovered Total
Homemade ORS 1200 (a) 300 (c) 1500
Low osmolarity ORS 700 (b) 300 (d) 1000
Total 1900 600 2500

Null hypothesis: There is no significant difference between outcome when treated with home-
based ORS or low osmolarity ORS.
Expected table:

1500∗1900 1500∗600
= 1140 (a) = 360 (c)
2500 2500
1000∗1900 1000∗600
= 760 (b) =240 (d)
2500 2500

∑(𝑂−𝐸)2
X2 = 𝐸

(1200−1140)2 (300−360)2 (700−760)2 (300−240)2

= 1140
+ 360
+ 760
+ 240

= 3.15 + 10 + 4.73 + 15

X2 = 32.88
df = (r-1) (c-1) = 1

• Table value for df=1 is 3.84

• Calculated value > table value
• Reject null hypothesis and accept alternative hypothesis.
• Conclusion: Homemade ORS is more effective in treating diarrhoea when
compared to low osmolarity ORS