**Part 1**

Importance of Census Data

With regards to census data as applicable to your country:

- Briefly explain why census data are critical to public health in your country. Be sure to provide examples of specific public health issues, needs or policies, which might rely on the census data.
- You have been appointed as the Head of Public Health Services for the whole of your country. What changes would you introduce for the next census in order to improve:
- The comprehensiveness of the data (e.g. collecting more variables and data collection from targeted populations (e.g. indigenous, minorities, immigrants, ageing population, etc.);
- the quality of the data; and
- the health-related information that will help you to plan and monitor public health?

Please provide justification for these changes.

**Part 2**

Standardisation

As a public health researcher, you wish to compare mortality rates for cancer recorded in the area/district that you are employed, with that of the national average of the country for years 2010-2011.

- Why is it not appropriate to compare the crude mortality rates within your area/district population with those of the national average?
- What method of standardisation is used to calculate the SMR?
- Can SMRs for different areas in the country be directly compared?
- From the table below, calculate the expected deaths for your area/district.Table showing deaths, population and mortality rates for cancer for 2010-2011

Age group | Deaths in Your area/district | Population in your area/district | National mortality rate (per 1,000/year) | Deaths ‘expected’ in your area/district |

0–14 | 2 | 950 | 1.61 | |

15–24 | 1 | 724 | 0.98 | |

25–44 | 5 | 932 | 2.32 | |

45–64 | 12 | 870 | 13.64 | |

65–84 | 26 | 423 | 64.45 | |

85 + | 8 | 24 | 187.13 | |

Total | 54 | 3,923 | 12.41 |

- Calculate and interpret the SMR for your area/district based on the table above.
- Calculate and interpret the confidence interval for the SMR using the formulae below. For clarification purposes, please review the definitions of standard error (SE), standard deviation (SD) and confidence interval (CI):Standard error (SE) (also known as SE of the mean or SEM)is used in inferential statistics to estimate how the mean of the sample is related to the mean of the underlying population. The SE depends on both the standard deviation (SD) and the sample size. It is calculated as SE = SD/√(n or sample size).Standard deviation (SD) describes the variability in a set of data – a summary of how dispersed the values of a variable are around its mean. When we calculate the SD of a sample, we use it as an estimate of the variability of the population from which the sample was drawn. Therefore, the larger the SD, the higher the variability within the sample.Confidence interval (CI) describes a range of values for a measure (e.g. rate) or variable of interest constructed so that the range has a specified probability of including the true value of the measure or the variable. The confidence interval gives an indication of the precision of say a mean or an SMR. The CI is expressed as two numbers, known as the confidence limits with a range. This range, with a certain level of confidence, carries the true but unknown value of the measured variable in the population. Most often the CIs are set at the 95% level. This essentially means that 95% of the time properly constructed confidence intervals should contain the true value of the measure. (In other words, there is a 5% probability that the true value might lie either below or above the two confidence limits.)There are well-accepted formulae for a CI of the mean and for a proportion. However, a number of different formulas are available for risk ratios and odds ratios. Regardless of the measure, the interpretation of a confidence interval is the same: the narrower the interval, the more precise the estimate.In statistics, the 95% CI is equal to 1.96 times the standard error (SE) of the estimate: 95% CI = 1.96 x SE. The lower limit of the 95% confidence interval = mean minus 1.96 × standard error and the upper limit of the 95% confidence interval = mean plus 1.96 × standard error.

The formula for Confidence Interval for SMR:

Where

Hence, the formula for the Upper limit of 95% CI for SMR:

The formula for Lower limit of 95% CI for SMR: