Not all parts of an organisation carry the same substance use risk. A distribution warehouse may have a different risk profile to a corporate office. A remote mining site may face different challenges to an urban construction project. Treating the entire organisation as a single, uniform risk group means spreading your testing resources evenly when they should be concentrated where they are needed most.

Analysing test result trends by department, location, and site allows you to identify where elevated risk exists, investigate the underlying causes, and deploy targeted interventions that are more effective and more efficient than a one-size-fits-all approach.

How to Analyse by Department and Location

The starting point is disaggregation — breaking your aggregate testing data into meaningful subgroups. The most useful breakdowns typically include:

• Department or business unit — Operations, maintenance, logistics, administration, and so on.
• Location or site — For multi-site organisations, comparing results across locations can reveal significant variation.
• Shift — Day shift, night shift, and swing shift workers may have different risk profiles.
• Employment type — Permanent, casual, contractor, and labour hire workers may show different patterns.
• Role type — Safety-critical vs non-safety-critical roles.

Data Requirements

Meaningful analysis requires that your testing data is tagged with enough contextual information to allow disaggregation. If your records only capture the employee name and result, you cannot analyse by department or location. Ensure your testing management system captures:

• The employee’s department, location, and role at the time of testing
• The testing type (random, for-cause, post-incident, pre-employment)
• The date, time, and substance(s) detected (if any)

This metadata is essential. Without it, your data is a flat list of results rather than an analysable dataset.

Statistical Significance Considerations

Before drawing conclusions from departmental data, consider whether the sample size is sufficient to support the conclusion. This is where many organisations make errors in interpretation.

• Small sample sizes — A department of 10 people tested twice per year produces 20 data points annually. A single positive result gives a 5% positive rate. Two positives gives 10%. The difference between those two scenarios is one person — hardly a statistically significant trend.
• Larger sample sizes — A site with 500 employees tested quarterly produces 2,000 data points per year. A shift from a 2% positive rate to a 3.5% positive rate across that sample is meaningful and warrants investigation.
• Time period — Aggregate data over multiple quarters or years before comparing departments. Quarterly snapshots are too volatile for small groups.

The practical implication is that for smaller departments or sites, you may need to combine data over longer periods — or combine related departments — to achieve a sample size that supports reliable analysis.

What Elevated Rates Might Indicate

When a department or site consistently shows a higher positive rate than the organisational average, it is a signal — not a verdict. The elevated rate could reflect several different underlying issues.

Workplace Culture

Some teams develop norms that tolerate or even normalise substance use. This is particularly common in industries with a strong social drinking culture or in teams where long hours and physical labour are the norm. The culture within a specific team may diverge significantly from the organisational culture.

Leadership and Supervision

Teams with inconsistent supervision, absent leadership, or managers who are themselves disengaged from the safety program may have higher rates. The correlation between leadership quality and safety outcomes is well established.

Working Conditions

Departments or sites with poor working conditions — excessive overtime, inadequate facilities, high-pressure production targets, or limited access to amenities — may see higher substance use as workers attempt to cope with the demands.

Geographic Factors

For organisations operating across multiple locations, regional differences in substance availability and social norms can influence positive rates. A site in a region experiencing a methamphetamine crisis may show different patterns to a site in a metropolitan area.

Demographic Factors

Workforce composition — age distribution, gender mix, employment type — varies across departments and can influence positive rates. These factors should be considered when interpreting data, though they do not change the organisation’s obligation to manage the risk.

Intervention Strategies

Once you have identified a high-risk area and investigated the likely contributing factors, targeted interventions can be deployed.

• Increased testing frequency — Temporarily increase the random testing rate for the affected department or site. This has both a deterrent effect and a detection function. Ensure this is permitted under your policy and communicated transparently.
• Targeted education — Deliver substance awareness training tailored to the specific substances being detected and the specific workforce. Generic training is less effective than targeted messaging.
• Leadership intervention — If the issue is leadership-related, address it directly. This may involve coaching the supervisor, changing the reporting structure, or providing additional management support.
• Working condition improvements — If the root cause is workplace stress, excessive hours, or poor conditions, address those factors. Testing alone will not solve a problem caused by the work environment.
• Enhanced support services — Consider deploying EAP resources on site, holding wellbeing sessions, or making counselling more accessible for the affected group.

Privacy in Reporting

When reporting on departmental or site-level trends, privacy must be carefully managed. The smaller the group, the greater the risk that aggregate data could identify individuals.

• Minimum group size — Do not report positive rates for groups smaller than a defined threshold (commonly 20-30 people). Below this size, a single positive result can be easily attributed to a specific individual.
• Report rates, not counts — A report that says “Department X had a 4% positive rate” is less identifying than “Department X had 2 positive results.” In a department of 50, rates are safer. In a department of 10, even rates can be identifying.
• Restrict access — Departmental-level reports should be available only to those with a legitimate need — senior management, WHS, and HR. Line managers should receive their own department’s data only.
• Focus on the trend, not the individual — The purpose of departmental analysis is to identify systemic issues, not to locate specific employees. Keep the reporting and discussion at the aggregate level.

Building a Continuous Improvement Cycle

Departmental trend analysis should not be a one-off exercise. Build it into your program’s regular review cycle:

1. Quarterly analysis — Disaggregate results by department and location each quarter.
2. Identify outliers — Flag any department or site where the positive rate is materially above the organisational average.
3. Investigate — Determine the likely contributing factors for any outlier.
4. Intervene — Deploy targeted measures appropriate to the cause.
5. Monitor — Track the impact of interventions in subsequent quarters.
6. Report — Include departmental trends in your board and senior management reporting.

This cycle ensures that your testing program is not just generating data — it is using that data to drive continuous improvement in safety outcomes.

Ready to identify where your highest risks are? Start a free trial of FairTest and use department-level dashboards and trend analysis to focus your testing resources where they matter most.