The phone call from my insurance carrier came in unexpectedly one morning. “We see that you’ve just been discharged from the hospital …”

I blinked. I had neither been admitted nor discharged from any hospital. When I pressed for details, the caller claimed her records showed I’d been hospitalized for more than a month. Why? For which condition? She didn’t have that information, she told me. Only the dates.

My first thought: It must be a scam. I ended the phone call. However, a few days later a letter arrived from my insurance company, reading, in part: “Our records show you have just been admitted to a facility …” They really did believe I’d been hospitalized for a month.

Second thought, which was far more chilling: Was I dealing with identity theft? Given all the data breaches in health care systems in the recent past, it wouldn’t be out of the realm of possibility, I feared. I called my insurer’s billing department and we went through every single claim they’d processed under my name for the last six months. Fortunately, there wasn’t anything suspicious.

I delved into the reasons behind the outreach. It turned out that my regular IVIG treatment for common variable immune deficiency (CVID) – which took place in a hospital outpatient clinic – was being misinterpreted as repeated inpatient admissions. IVIG is a lifetime treatment for CVID, and I’d been receiving infusions regularly since 2018 at the same clinic with the same doctor. The insurance company had processed these claims without incident for years. However, their newest claim scanning tools were apparently flagging the word “hospital” without nuance. I was told it was likely that the tools would trigger more alerts in the future, and all they could really do was add me to transitional care’s “do not call” list.

It wasn’t the first time I’d been on the receiving end of misguided outreach from my insurance carrier. At one point I received information on their asthma care management program: a leaflet stating “asthma is a disease where you can’t breathe” and asking me to “pledge to take your medication.” I tossed the leaflet into the recycling bin, and I am guessing a lot of other asthmatics had the same response. The pity was that since I’ve been a diagnosed asthmatic since the age of 15, and am always interested in improving my care and preserving lung function, if the outreach had included anything helpful, I would have eagerly read it.

Other completely unhelpful outreach efforts I’ve received have included postcards telling me to get my children vaccinated. Since I have no children and absolutely no intention of changing that, it’s been amusing.

What’s the point of patient outreach when it’s misguided, unhelpful and in some cases directed to the wrong people entirely? States and insurance carriers spend countless dollars on patient education and outreach, but when there’s so little quality control, are they just throwing good money after bad? When efforts are not targeted appropriately, patient populations are not accurately identified and there’s nothing beneficial included in the outreach, it wastes everyone’s time and money.

Insurance companies and public health agencies may claim that patient populations resist outreach. Hey, they’ll say, we tried. What they neglect to question is whether the outreach was ignored because the patients who received it weren’t in the intended demographic at all or found it unhelpful.

Outreach can, and should, be helpful if it’s targeted and written appropriately. In the case of the asthma management program, if they’d sent a peak flow meter, a starter symptom and medication diary and some tips on identifying asthma triggers or reducing household allergens, it might have reached more people. Perhaps they could have included some pamphlets from the American Lung Association or information on local Zoom or in-person support groups for newly diagnosed asthmatics.

In the case of the transitional care contacts, every minute they spent sending letters and making phone calls to me was time taken away from someone who actually needed those services. Their AI tools should have been taught to look for more than the word “hospital” when scanning claims, and their findings should have been verified by a human before being sent on to outreach. There was nothing in my claims history or medical record supporting the inpatient admission flag, after all.

As AI becomes more prevalent, states are pressed for funding and jobs are cut, we’re likely to see more misguided efforts. However, it’s exactly why care needs to be taken to ensure outreach is accurate and effective – it’s more critical than ever to ensure those health care dollars are not wasted with the wrong message, for the wrong audiences.

Denise Reich is a patient advocate.