How to Measure Social Connection
How to Measure Social Connection: A Plain-English Guide to WHO-5, UCLA-3, and PHQ-ADS
If you're evaluating a social prescribing vendor, a behavioral health program, or any solution that claims to improve well-being, you'll run into an alphabet of assessment tools pretty quickly. Vendors cite scores, point to percentage improvements, and reference scales by their acronyms as if everyone already knows what they measure. It's worth knowing, because these tools are how you tell the difference between a program that's genuinely tracking outcomes and one that's using soft, unfalsifiable language to describe impact.
Here's what the three most common instruments in this space actually measure, and how to read the numbers.
WHO-5: overall well-being
The WHO-5 Well-Being Index is a five-item, self-report questionnaire developed by the World Health Organization to measure psychological well-being over the past two weeks. It asks people to rate things like how cheerful, calm, and engaged they've felt, and it produces a score from 0 to 100, where higher is better.
A few reference points matter here. A score of 50 or below suggests possible depression, and scores at or below 28 approximate a clinical depression diagnosis based on systematic reviews of the literature. The commonly cited threshold for a meaningful, clinically relevant change in WHO-5 score is a 10-point increase. So if a vendor tells you their members improved "on average," ask what the baseline was and how many points the improvement represents. A 3-point shift is a rounding error. A 15 to 20 point shift, sustained over multiple follow-ups, is a substantial and clinically relevant change.
UCLA-3: loneliness
The UCLA-3 Loneliness Scale is a short, three-item version of the longer UCLA Loneliness Scale, one of the most widely validated tools in loneliness research. It asks people to rate how often they feel a lack of companionship, feel left out, and feel isolated from others. It's designed to capture the subjective, felt experience of loneliness, which is distinct from social isolation (the objective absence of social contact). Someone can be surrounded by people and still score high on loneliness; someone who lives alone by choice and stays connected through other means might not.
Because it's short and easy to administer repeatedly, the UCLA-3 is a practical tool for tracking loneliness over time, at intake, and at multiple points during an intervention, rather than just once.
PHQ-ADS: anxiety and depression combined
PHQ-ADS is a composite score combining two other well-established tools: the PHQ-2 (a two-item depression screener drawn from the longer PHQ-9) and the GAD-2 (a two-item generalized anxiety screener drawn from the GAD-7). Combining them into a single ADS (anxiety-depression severity) score gives a broader picture of behavioral health status without requiring the longer, full-length versions of each instrument.
When a program reports outcomes "based on the PHQ-ADS," it's telling you it's tracking both anxiety and depression symptoms together, which is useful, since the two conditions are highly correlated and often need to be addressed as a pair rather than in isolation.
What to actually ask a vendor
Knowing the tools is only half the value. When you're evaluating a program that reports outcomes using any of these instruments, a few questions will tell you a lot fast:
What's the baseline, and what's the follow-up interval? A single post-intervention score without a baseline tells you almost nothing.
Is the sample the full population, or just completers? Attrition matters. Programs with high dropout can look more effective than they are if only engaged members are counted.
Are they using validated cutoffs, or just directional language? "Members reported improved well-being" is vague. "Members saw an average 18-point increase on the WHO-5, more than the established 10-point threshold for meaningful change" is a claim you can evaluate.
How often is the assessment administered? One-time surveys tell you where someone started. Repeated measures tell you whether a program is actually working over time.
These three instruments, WHO-5, UCLA-3, and PHQ-ADS, show up across social prescribing, behavioral health, and population health programs precisely because they're validated, short enough to administer repeatedly without burdening members, and specific enough to give you something real to evaluate. Understanding them puts you in a much stronger position, whether you're comparing vendors or reading your own program's results.
