Follow-up: challenge and opportunity in HIV testing and treatment

Here's a nice, short article from the New York Times this week, which brings up an important point about HIV diagnosis and care:

Follow-up is key.

And, oh, so tricky.

...

Our study team works very hard to maintain close follow-up with our patients. We see a rural population, so the average patient travels a long distance - sometimes a whole day - just to keep an appointment. Also, since our study population is extremely poor, the price of travel is a considerable barrier. The trip to the hospital can cost over 20,000 Ugx (7 USD) in each direction. That is more than 10x the average daily income, dispensed twice per visit - not to mention the opportunity cost of missing a day in the fields, at work, or at home with the family. We reimburse up to a fraction of this expense, but the money we give cannot replace all the money spent (nor should it, ethically - else we border on coercion). The only additional help we offer is a hospital bed where a patient may spend the night, for free, before traveling back home all day the next day.

Mothers and babies leaving the hospital - how many miles to go?

Considering so many obstacles to care, every visit feels like a minor miracle. So you can see why this (recent data from our interim analysis) feels like a MAJOR miracle:  99% of our patients complete all, or almost all, of their 7 monthly visits. Amazing! They come to get prescriptions refilled, to have a trained ear listen to their heart and lungs. But more than that, I think they come to have a caring hand placed on their shoulder, that hand belonging to someone who knows them well and cares for them, who wants to hear about their pains, their nightmares, their dreams, their families, their hopes for the future. I credit our clinical staff and community health worker for the VERY strong relationships that they create with our patients. This personal brand of medicine brings the doctors and patients closer, and gives the patients a reason to keep coming back.

...

Now, of course, things could be better. We do not have point-of-care testing in Kiboga, as was shown to be so helpful for that clinic in Mozambique, so there is an inherent delay between (a) testing the blood for HIV and CD4 counts, and (b) starting the patient on appropriate ARVs.

In the general ART clinic, the median delay to treatment is just under 3 months. That statistic is obtained by looking at all patients who actually started ART, i.e. those who reached point (b). Meanwhile, a quick glance into the clinic records room would reveal to you a collection of patient files that gather dust in a "pre-ART" section, i.e. the patients who passed point (a) but then got lost. Did they move, transfer, forget, decline, or die before starting HIV meds? We don't know. In truth, many probably died, but these are the hardest patients to find, and their deaths can be very hard to confirm. With such losses occurring but not reflected in the "median 3 month delay", you can see why it is critical to shorten and track patients through the (a) to (b) time.

Row 1, transferred out; Row 2, lost on ART; Rows 3-4, lost pre-ART

In our study, by scheduling the initial 2 visits close together, we have reduced the delay to ART to 1-2 months. Perhaps more important, we keep track of every patient recruited for the study - phone calls when they have a mobile, and "neighbor watch" or home visits if needed - so we lose almost no one between testing and treatment. In the coming weeks I will be returning to the general ART clinic records room to sift through the "pre-ART" files, and try to track those patients to find out what happened to them. Then, with the past outcomes counted, we will be able to estimate how many deaths have been averted with closer follow-up. Looking at those dusty folders, filling 2 whole rows...I think it will be many. I will let you know what I find out!

Filing folders for patients in our study

...

To bring it back to the article:  the shorter the delay between HIV testing and treatment, the greater the number of patients who will actually get the care they need. Innovations like rapid CD4 counts allow for fewer visits, especially in the treatment initiation stage when patients are sickest and most vulnerable. This could save many, many lives.

Still, in the absence of such new tests, our study is proof that the provision of quality medical care - with individualized patient visits and continuity with providers - can be enough to motivate patients to seek care in the most adverse of circumstances.

Combine the strengths of thoughtful, new technologies and focused, age-old patient care, and we may finally see how much better things can get.