Ask workers directly and a third say the job hurt them

**Approximately 2.8 million non-fatal workplace illnesses and injuries were
reported in the United States in 2018.** That figure comes from employer
records, and **current surveillance methods might underestimate the prevalence
of occupational injuries and illnesses.**

So NIOSH asked the workers.

**Overall, 35.1% of employed respondents had ever experienced a work-related
health problem** (95% CI 33.0–37.3).

More than one in three. **The most commonly reported problem was back pain —
19.4%** (95% CI 17.6–21.2), which is on its own nearly a fifth of the working
population.

Among industries, construction was highest at 48.6% (95% CI 36.5–60.6) —
approaching half, with a wide interval that reflects a smaller sample rather
than a doubtful finding.

Why the two numbers are so different

They are not measuring the same thing, and the difference is the point.

The 2.8 million counts incidents an employer recorded in a year. The 35.1%
counts **people who believe, over a whole working life, that their job caused
or contributed to a health problem** — which the definition makes explicit:
occupational exposures might **directly cause, contribute to, exacerbate, or
predispose** workers to a health problem.

Back pain is exactly the kind of thing that falls between them. It rarely
arrives as an incident on a date, so it rarely becomes a record, and it is
overwhelmingly what people say when asked.

What the survey is

SummerStyles, an annual online survey run by Porter Novelli Public Services
with panellists recruited by probability-based sampling, running since

1995. **SummerStyles data have been shown to be valid for reporting health
      outcomes when compared with the Behavioral Risk Factor Surveillance System.**

Sent to 5,584 panellists; 73.2% responded, giving 4,088 adults. The
work questions went only to those describing themselves as **full-time paid,
part-time paid, or self-employed** — 2,425 people in this analysis.

**Prevalence varies by employment situation, industry and some demographic
characteristics**, which is the second half of the finding: not just how many,
but that the burden is not evenly spread.

What follows

Workplace injury and illness prevention programmes are needed — with two
targets named: specific problems such as back pain, which is the single
largest contributor, and higher-risk industries such as construction, where
a worker is about as likely as not to say the job has hurt them.

Source: Centers for Disease Control and Prevention, MMWR (NIOSH).