Health Indicators Surveillance System (HISS): Methodology

Design

The Health Indicators Surveillance System (HISS) is a short cross-sectional Phone interviews conducted in all 13 administrative   regions of Saudi Arabia on a quarterly basis. It started as a pilot between September – December 2019 in 5 administrative Saudi Regions on 4009 participants.

Each interview lasted approximately 4 minutes, conducted by a trained data collector. This study used the QPlatform® data collection system (ZdataCloud), which had integrated eligibility and sampling modules, to control the distribution of the sample.1 All questions had to be answered for the questionnaire to be successfully submitted to the database. All data were coded and stored on the ZdataCloud database.1

Sampling and sample size

This study used the proportional quota sampling technique to get an equal distribution of participants, stratified by age and gender, within and across the 13 administrative regions of Saudi Arabia. We used two age groups based on the Saudi Arabia median age of 36 years. This led to a quota of 52 for this study.

The sample size was calculated based on the depth of the sub analysis we wanted to reach, which compares the age and gender groups across regions with a medium effect size of approximately 0.3 with 80% power and 95% confidence level.2 Thus, each quota required 100 participants, and a total sample of 400 per region, to form a grand total of 5,200 participants/wave. Once the quota sample was reached, participants with similar characteristics were not eligible to participate in the study. The quota sampling is an automated process with no human interference, as the sampling process is controlled automatically by the data collection system.1

Update:

In the 4th quarter on 2020 we increased the sample to 134 per quota, which produce 536 participants per region and a grand total of 6,968 participants, with a small effect size of approximately 0.24 with 80% power and 95% confidence level.2 This increase were decided to improve the sample power within regions to allow for better comparison. 

Participants and recruitment

Participant recruitment was limited to Arabic-speaking Saudi residents who were ≥18 years old. A random phone number list was generated from the Sharik Association for Health Research to identify potential participants. The Sharik database is composed of individuals who are interested in participating in future research projects and contain a growing number of registered participants that have reached more than 74,000 distributed across the 13 regions of Saudi Arabia. Participants were contacted by phone on up to three occasions. If not responded, a new number with similar demographics will be generated from the database until the quota is completed and closed automatically. After obtaining consent to participate, the interviewer assessed the eligibility, based on the above-mentioned quota completion criteria.

Questionnaire design & validation

After providing verbal consent, participants were asked about their age to determine eligibility. Then the data collector will record the gender and the region. As shown in the HISS data model (Figure 1) the dataset includes behavioral risk factors (diet, physical activity, and smoking), diagnosed intermediate risk factors (high blood pressure, and high cholesterol) and obesity measured as a body mass-index (BMI) using height and weight, and finally some diagnosed chronic conditions in which the participants currently receiving treatment for, including (diabetes, cardiovascular diseases, stroke, cancer, lung diseases, liver disease and genetic diseases).

Participants were asked to provide height in cm and weight in kg. We also calculated participants’ body mass indices (BMI). BMI was calculated using the following formula:

BMI =  (Weights (kg))/(Height ^2).3

We used the Center for Disease Control and Prevention (CDC) BMI category status, which specifies the BMI below 18.5 as underweight, from 18.5 to 24.9 as normal, from 25 to 29.9 as overweight and 30 and above as obese.3

Following the questionnaire development, linguistic validation was done to ensure that the participant understand the questions as intended and can provide the right answers. A focus group of eight participants were asked to discuss and answer the survey as one group. This process was repeated a second time with the same people to test the reliability. Afterward, 8 new participants were interviewed to ensure the accuracy of the original meaning and to better understand the modified questions. Accordingly, the questionnaire was edited further, and the final version of the questionnaire was produced.

In terms of outcomes validation, HISS  results are reviewed and compared against other results published by other national organizations such as the Saudi general authority of Statistics, and with other national level research projects conducted by Sharik association for health research.

Figure 1:

Notes on HISS methodology

This surveillance system is a work in progress; however, it is generating unprecedented national level data in a frequent basis, and the only one of a kind currently in Saudi Arabia. However, it could be criticized for using quota sampling, with its attendant risk of selection bias, rather than a random probability sampling approach. However, the costs of probabilistic sampling are significantly greater and for this project intention and the type of variables the risk of some (generally low level)4,5 bias was considered acceptable. In addition, using proportional large sample with large number of quotas plays an important role in reducing the selection bias. Currently in Saudi Arabia the only way to generate random national level sample is via household survey which also has some significant limitations due to sociocultural factors. Our preliminary comparison with probabilistic sampling national projects showed very slight differences some was not related to the sampling technique but to the way the data collected or measured. We will keep updating this page with the results of our outcomes validation as soon as we possible.

Finally, as mentioned above the HISS is a work in progress and we are striving to improve our data quality and methodologies, thus, we may release new version of this document highlighting the changes and the reasons behind it.

To Cite this Article please use:

NF Bindhim, NA Althumairi, MH Basyouni. Health Indicators Surveillance System (HISS): Methodology. Sharik Association For Health Research. 2019. https://sharikhealth.com/phm/method.html

References

1.         NF BinDhim. Smart Health Project. 2012; https://shproject.net/. Accessed March, 2020.

2.         Cohen J. Statistical power analysis for the behavioral sciences. Academic press; 2013.

3.         Center for Disease Control and Prevention. About Adult BMI.  https://www.cdc.gov/healthyweight/assessing/bmi/adult_bmi/index.html. Accessed March, 2020.

4.         Groves RM, Fowler Jr FJ, Couper MP, Lepkowski JM, Singer E, Tourangeau R. Survey methodology. Vol 561: John Wiley & Sons; 2011.

5.         Marsh C, Scarbrough E. TESTING 9 HYPOTHESES ABOUT QUOTA SAMPLING. Journal of the Market Research Society. 1990;32(4):485-506.