Wednesday, September 27, 2017

Quick take: delirium + clinical trials

My hat’s off to Drew Rosielle for his recent post about delirium at Pallimed, and for this passage in particular:

“...delirium is an international health crisis, it is real, it can be devastating (lead to permanent cognitive changes), leads to far worse outcomes for our patients (longer hospital stays, not being discharged home), costs billions of dollars, sucks shit for the patients and families going through it, and we don’t have a real inkling about actual, effective drug treatment for it.”

Pow.

Rosielle’s focus is the report in JAMA on a study conducted at MD Anderson Cancer Center that compared haloperidol + lorazepam versus haloperidol + placebo as treatment for hyperactive (agitated) delirium at end of life. It’s a worthwhile read, and one that I’ll return to in a later post.

His overall concern is the lack of firm guidance for the treatment of delirium with drugs based on double-blind randomized clinical trials (RCT), the gold-standard for evidence-based practice. So, here’s a brief discussion about clinical trials, and about another less-rigorous way that we can know our actions are based on something more than crazy ideas shared around a late-night campfire.


Saturday, September 23, 2017

Can one number tell us if this hospice is any good?

Boo Boo loves data

How would you respond to someone who asked, "What single factor can help me decide if this hospice is any good?"

I posed that question to Boo Boo, and she replied without hesitation, "First, I'd ask the agency how many patients and families they usually care for who live in their own homes. I think they call it 'the average daily census for home based patients,' or something like that. It's a different group of hospice patients than those who reside in nursing homes, assisted living facilities, hospice houses, or inpatient hospice units. Anyway, I'm looking for a number, very simple."

She continued, "Next, I'd use this handy little chart to find how many full time hospice RN case managers (FT RNCM) are needed to care for that number of patients and families. These are the nurses with an important 40-hour, Monday through Friday job. An RNCM is the glue that binds the rest of the team. Let's say the agency tells me they usually care for about 300 patients and families living in their own homes. The corresponding number to 300 is 28, so I'd ask if they had 28 full time hospice RN case managers on staff. Their answer to that one simple question could say a lot about whether or not the hospice is any good."

Boo Boo soon fell fast asleep. Here's the chart she was talking about:


figure 1

Some background

A colleague and I had a conversation recently about staffing, specifically the ideal average caseload for a full time hospice RN case manager working with patients and families in their homes.

My colleague and I met several years ago, at the hospice branch of the home care service group of a large health system. Many of our patients were referred from the system's two comprehensive cancer centers, so acuity was generally high, and the length of stay was often brief. We were pretty busy case managers. The health system sold off their hospice in 2011, and we went with it for a while.

Our discussion led us to the website of the hospice industry trade group, the National Hospice and Palliative Care Association (NHPCO). Their monthly member newsletter from March, 2011 includes a cover story on the agency staffing guidelines they developed. (pdf)

We followed a link at the NHPCO website, but ended up at a members only page, so we couldn't access the staffing tool and supporting information.

We also tracked down a 90-minute workshop at their management and leadership conference: "New NHPCO Staffing Guidelines - Methods and strategies for success," and even found a 29-page document with 57 PowerPoint slides from the presenters.

A handout like that is probably the single worst way to share important information, but I digress...

We read through the bullet points and truncated phrases and arrived at a bottom line: the NHPCO staffing guidelines from 2011 recommend a caseload of between 10 and 13 home-based patients per FT RNCM, or 10-13/RN.

figure 2 - from page 20 of the handout

figure 3 - from page 21

figure 4 - from page 22

figure 5 - from page 24

figure 6 - from page 26

The guidelines include several factors that may increase or decrease the suggested caseloads, like whether or not RNCM's also admit patients or provide off-hours coverage, and note that staff turnover can also affect caseloads.

These numbers and those conditions made sense to us both, because they're consistent with our own experience. A caseload of 9 to 13 patients allows for efficient and effective quality care, including documentation and collaborating with the team. We seldom, if ever, had fewer than 10. Things get harder to juggle after about a dozen.

It can take up to 4 hours to conduct and document an admission, time which could otherwise be spent visiting two homes, or maybe even three.

Responding to off-hours events interferes with the RNCM's Monday-Friday responsibilities, no matter if the response is by phone or in person, and regardless of how often they provide the off-hours coverage.

High staff turnover means more work for RNCMs who must fill the holes left by the departed while also mentoring the new nurses, if any.

It can be a real mess, and often is. That was the upshot of the conversation with my colleague, anyway.

Boo Boo crunches the numbers

When she woke up, Boo Boo explained how she developed her chart.

"I based it on an average caseload of 11 patients for each FT RNCM. That's the median in NHPCO's guidelines (see figure 2). I just divide the average daily census (ADC) by 11, to get the number of FT RNCM's needed."

"I'd have to think hard about working with an agency that didn't have enough staff to handle their census. Maybe there's an explanation, and I'd be open to hearing more. But I wouldn't want to end up on the short end of a stick, if you know what I mean."

Boo Boo also pointed to the document, "Suggested Questions to Ask When Choosing a Hospice," at the Hospice Compare website (pdf). "They suggest some interesting questions for patients and families to ask a hospice. Two of them are really astute: 'Will I have the same hospice nurse?' and, 'How many patients are assigned to each hospice nurse?' That last one gets right to the heart of FT RNCM caseload."

figure 7 - from Hospice Compare at medicare.gov

I wondered how an agency would feel about being asked these sorts of things - how many staff they have, how many patients and families they serve. Could an agency refuse to give a clear answer? Would they treat these numbers like a trade secret? Would they respond using Weaselspeak, the language of hinky dodging, or with bland assurances that they know what they're doing, so we should all just shut up, sit down, forget about any silly numbers, and trust them?

"That would tell me more about the agency than any number ever could," Boo Boo said. "That would tell me all I needed to know, and I'd just cross them off my list."

Boo Boo can be harsh sometimes, but in my experience she's always been fair.