The future of healthcareLearn More

It is time for primary care to enter the triggering business

It has been suggested, some would say demonstrated that doctors know very little about their patients. If you are a doctor could you identify your patient’s partner from a line-up of strangers (other than people you see as a couple)?  Or could you tell without seeing the name on the document if this bank statement belonged to that patient? Or whether that utility bill was from where that person lives? Is this internet search history theirs? Do you know how much they spend on lottery tickets? Alcohol? Vegetables?

A few years ago our team then based in the UK was evaluating an intervention to increase access to general practitioners. If the intervention worked we would have to demonstrate improvement over the course of a whole year. Here’s the thing, we noted that year after year there was a pattern to the demand for same day (emergency) appointments- with definite peaks and troughs. So if the intervention worked it would have to be sustained during both the peaks and the troughs. It did. The data on out-of-hours services exhibited very similar patterns- with definite peaks and troughs and at unexpected times of the year. We could not explain the patterns but noted that when the meteorological office recorded  22 hours or more of sunshine in the week the demand for appointments dropped. Not the prevalence of viral or other community pathogens but sunshine of all things! Okay may it was some factor that we hadn’t modelled in the analysis but there was a definite pattern that we could not immediately explain on the basis of what seemed plausible at the time. We called it the Spring Cleaning Effect– we hypothesised that people in the UK were less likely to attend doctors in general practice when there was a run of sunny days on which to do outdoorsy things. We didn’t anticipate this- nor did clinic managers because the patterns of demand were not used to inform the scheduling of doctors’ on-call rosters. It was clear that they were blind to a phenomenon nobody understood fully.

More recently I reviewed some data on certification for low back pain and noted the pattern that as unemployment rates in a locality increased the rates of certification dropped and then plateaued.

Our team is now investigating similar data from a large employers’ records. We hypothesise that rates of submission of sickness certification will show a sharp drop when vacancy rates fall and other markers of economic health decline. People may be far less likely to take time off sick if they are fearful of upsetting their supervisor. With respect to primary care, it is unlikely that doctors will know everything that impacts on their patient’s choices. Time spent with the patient in discovering these things is unlikely to increase as it comes at a financial cost. Therefore doctors will never fully anticipate all the drivers to patient behaviour. Why does that obese person fail to take action on weight management? Why does this other person take ‘medication holidays’ when they need to take the treatment consistently to benefit? Why does the next person refuse to have an X-ray? Why is there a rush of people with relatively minor conditions demanding appointments this week and not last?

Some drivers lead people to behave in unexpected ways as I have commented here previously. Not only that but as Mullainathan and Shafir have postulated people are often unable or perhaps unwilling to follow doctor’s advice. In the end, the best we can hope is to trigger the relevant behaviour in people who are already motivated and seek teachable moments to inspire people to act for their benefit. Primary care may be more about recognising or fishing for opportunities and much less ‘educating’ for change. Such triggers need to fit within the final moments of a 15-minute consult. The work to develop and evaluate such triggers is only beginning. Counselling patients to stop smoking will yield 1:20 quits in a year, showing them a trigger (in less than 5 minutes) that appeals to their vanity results in 1:7 quits. A substantial number (1:5) of obese people will lose weight in 6 months if they are shown what difference that would make to their appearance without having to be extensively counselled on diet and exercise.

Picture by Aimee Rivers

Can the patient relay what was done for them?

A perennial source of dissatisfaction in healthcare (as documented here and here) is the poor flow of information from one sector to another. ‘Joe’ (speaking here– video from BMJ open) couldn’t tell me, his doctor, anything helpful about what had been done while he had been in hospital. That means we have to schedule several appointments to try to unpack it all. He was an in-patient for two weeks and someone had decided one Thursday morning that it was time for him to go home. It wasn’t really clear to Joe or to me why that particular morning or what was to happen when he got home other than that he should contact his ‘local GP’. A letter would follow some time in the future. There may have been good or bad reasons for sending him home. We could only guess what was in the mind of the person who made the decision:

We needed the bed. Joe was fine. His observations were normal, he was ambulant his wife was happy to take him home.

But of course Joe comes home with lots of questions, which I now struggle to answer without making phone calls to track down the busy medical team. The problem is articulated by several ‘stakeholders’ members of the ‘multidisciplinary team’ on the ward none of whom feel they own the problem of telling this man what he needs to know. There is only one constant in this story- Joe. If Joe can collect the information we need during the course of his hospital stay we might begin to improve the outcome:

In addition to increasing the burden on GPs, it engenders a need for a subsequent GP appointment; it limits GP capacity to respond to patient concerns and queries, at least on one occasion; it may result in a re-referral to the specialist; and it increases GP dissatisfaction with the care provided to the patient by the hospital. BMJ

The problem is Joe often does not know what he needs to know by the end of his hospital stay. It isn’t impossible to work out how to trigger questions for Joe to ask throughout his hospitalisation. What is far more difficult is to motivate every hospital ward and every discipline in a team to address this challenge consistently. It is ‘easier’ to nudge one individual than enlist the cooperation of the dozens of health professionals who will come into contact with Joe. Making people active in healthcare processes has achieved results before:

Influence at Work, a training and consultancy company that Cialdini founded, worked with the United Kingdom’s National Health Service (NHS) in a set of studies aimed at reducing the number of patients who fail to show up for medical appointments. They did this by simply making patients more involved in the appointment-making process, such as asking the patient to write down the details of the appointment themselves rather than simply receiving an appointment card. Sleek

Picture by Michael Coghlan

Doctors need better tools to help people recognise danger

Doctors see it all the time. The fifty-year-old with a BMI of 28, the teenager who is developing a taste for cigarettes, the twenty-year-old who now binge drinks every weekend, the soon-to-be-mum who is ‘eating for two’. Small choices that may become habits and habits that lead to consequences. Where I work the average consultation is fifteen minutes. In that time we address whatever symptoms or problems have been tabled. The list may be long. Occasionally it’s possible to raise a topic that I’m worried about. The problem is the patient may not be worried about that issue.

Afterall doctor I don’t drink any more than my mates do or I don’t really eat that much.

What’s needed are tools that help frame the issue from the perspective of the patient, not the practitioner. Tools that help us address public health priorities that speak TO that person, not AT everyone. Before making any changes the person needs to agree that their choices might blight their hopes for the future. These are not inconsiderable challenges given the gloomy predictions for the future.

At the other end of the malnutrition scale, obesity is one of today’s most blatantly visible – yet most neglected – public health problems. Paradoxically coexisting with undernutrition, an escalating global epidemic of overweight and obesity – “globesity” – is taking over many parts of the world. If immediate action is not taken, millions will suffer from an array of serious health disorders. The World Health Organisation

Diabetes is likely to cement its place as the fastest growing epidemic in history. The Medical Journal of Australia

In addition, youthful drinking is associated with an increased likelihood of developing alcohol abuse or dependence later in life. Early intervention is essential to prevent the development of serious alcohol problems among youth between the ages of 12 and 20. NIH

Picture by Marcelo Nava

Perspective is crucial when considering changes in healthcare policy

It was summer of 2010 in Australia. I had been working hard in the garden one Sunday afternoon. Feeling the need for a little mindless entertainment I suggested to our then 14 year old that we might rent a sci-fi movie. It was around 7pm and  getting dark. My son jumped into the car beside me and we took off toward the video shop. That’s when I noticed that the car headlights were far too dim. I turned on full beam- but it didn’t help. I ranted for a good five minutes about how difficult it was going to be to get a mechanic to look at the car so close to Christmas. I was also a bit put out that our teenager was showing no concern for my predicament. At this point he quietly reached across and took my sunglasses off my face.

There. Problem solved dad.

I learned something that day not least what it would cost me if that story was not be retold to his brothers.

I love the work of Deana McDonagh and Joyce Thomas, especially their thinking on empathic design. Deana and Joyce begin their sessions on empathic design by inviting participants to try on their designer glasses- the ones that demonstrate what it must feel like to have tunnel vision. They’ve written about it in the Australasian Medical Journal. I keep those glasses in my office to remind myself and visitors of the valuable insights they offer but also as a treasured momento of a fun workshop generously organised by a brilliant team.

Their work came to mind later when we were investigating the attitude to self-management of a condition that is progressive and for which there is no cure. Patients and doctors in an Asian setting were interviewed. We recorded poignant stories about the impact of this condition on people’s lives- resulting in social isolation, self loathing and a need to feel supported by a health practitioner:

Both patients and doctors were against the adoption of self-management strategies. This is contrary to recommendations for the management of COPD by many studies and guidelines. However, another study has similarly shown that self-management skills were not rated as important by patients. Furthermore, the psychosocial impact of their disease such as fear limited their ability to manage their own symptoms. A lack of knowledge may also contribute to their dependence on doctors and health care providers.

We concluded:

In reality, patients have to conduct self-management daily and it is not feasible for physicians to provide all of the management needs that patients have during their day-to-day lives. Therefore, self-management remains an aspect of overall COPD care. However, it should not be the only focus and future interventions should also examine ways to improve access to health care.

On reflection we noted something similar with patients in Australia. Those who had an established medical condition were much more likely to ‘trust’ their doctor than those who were not currently unwell or those from higher socioeconomic groups. Innovating requires the ability to see people as heterogenous having very different perceptions on the need to be in charge of their own health, perceptions that are liable to change with circumstances. I also wonder if policy makers consider what it must be like to implement their big ideas from this perspective:

Picture by Redfishingboat

Why people will sack a plumber but won’t sue their doctor

Everyday somebody somewhere summons a plumber. The drain is blocked, the boiler isn’t working, there’s a leak under the sink. The problem is obvious the solution is technical and everyone knows when the job is done. If it’s not fixed asap the plumber is sacked.

That’s rarely what it’s like in medicine. Not everything is a blockage or a break. Not everything can be fixed by sitting quietly with a tool box and following the instructions in the manual.

Many of the commonest problems in healthcare don’t have an easy fix.

  1.  The pain of ‘tennis elbow’ can last for months despite treatment

Patients with tennis elbow can be reassured that most cases will improve in the long term when given information and ergonomic advice about their condition. Bassett et al.

2. Plantar warts don’t always respond to cryotherapy

Little evidence exists for the efficacy of cryotherapy and no consistent evidence for the efficacy of all the other treatments reviewed. Gibbs et al

3. Lung cancer is incurable in most cases

Lung cancer is the main cancer in the world today, whether considered in terms of numbers of cases (1.35 million) or deaths (1.18 million), because of the high case fatality (ratio of mortality to incidence, 0.87). Parkin et al. 

4. Anti hypertensives aren’t guaranteed to prevent stroke

Our analysis indicated that the absolute benefits of antihypertensive therapy depended on BP as well as the presence or absence of additional cardiovascular disease risk factors and the presence or absence of preexisting clinical cardiovascular disease or target organ damage. Ogden et al

5. Mild depression can be hard to treat

The magnitude of benefit of antidepressant medication compared with placebo increases with severity of depression symptoms and may be minimal or nonexistent, on average, in patients with mild or moderate symptoms. Fournier et al 

6. There is no cure for the common cold ( You don’t need a link for this)

These and most other problems in healthcare cannot be ‘fixed’. They can be diagnosed and they can be ‘managed’ but they can’t be fixed in the way that faulty plumbing can be fixed. Therefore that queue of people in the waiting room is saying something more than ‘I’m here to be fixed.’ Patients are saying:

  1. I am in pain
  2. I am anxious
  3. I am unhappy
  4. I am bored
  5. I am angry
  6. I am confused
  7. I am lonely
  8. I don’t like my job
  9. I can’t pay my bills
  10. I need tablets or surgery

The job of the doctor is to work out which and then to fix what can be fixed and help the patient to live with the rest until their perspective or their circumstances change.

  1. Most people won’t take their tablets as prescribed.

    Because non-compliance remains a major health care problem, high quality research studies are needed to assess these aspects and systematic reviews are required to investigate compliance-enhancing inteventions. Let us hope that the need will be met by 2031. Vermeire et al 

  2. Most people won’t benefit substantially from health promotion advice.

Exercise-referral schemes have a small effect on increasing physical activity in sedentary people. The key challenge, if future exercise-referral schemes are to be commissioned by the NHS, is to increase uptake and improve adherence by addressing the barriers described in these studies. Williams et al

3. Most people get better in spite of treatment and not because of it.

Disease mongering can include turning ordinary ailments into medical problems, seeing mild symptoms as serious, treating personal problems as medical, seeing risks as diseases, and framing prevalence estimates to maximise potential markets. Moynihan et al

But then most people will be deeply grateful to their family doctor because they don’t have to respond a certain way to be treated with respect and they don’t expect a ‘cure’ and won’t ask for their money back when things don’t work out. The doctor’s role is to be there, to encourage, to educate, to accept and to walk with their patient through all the challenges that life has to offer.

Picture by Vicki

How can medicine compete against the new normal?

Joanna was exactly the person we are being urged to help. In her forties, overweight verging on obese. Hypertensive, asymptomatic but well on her way to chronic diseases. We discussed her diet.

I like salt. So my food tends to be salty. Also most people in my house are my size. I thought about reducing my portions but I like meat, lots of meat. I’m a member of a gym but I rarely go there.

We talked about her risk of heart disease and encouraged her to banish the salt cellar from the table, perhaps think again about reducing the portion size and making time to go to the gym. She looked at me pityingly her eyes said

Well that ain’t gonna happen

This was not a teachable moment. She was not ready to make an investment in changing her habits. She could not see that she was at risk. She was ‘normal’ as far as she could see. So she was not going to change her diet to deal with a problem that she did not perceive as real.

There are many things that are regarded as ‘normal’.

It is now normal:

  1. To have to wear extra large clothes.
  2. To be offered larger portion sizes when we dine out
  3. For more than one in three Australians aged 14 years and over to consume alcohol on a weekly basis
  4. For friends or acquaintances to be the most likely sources of alcohol for 12–17 year olds (45.4%), with parents being the second most likely source (29.3%)
  5. For more than one in three Australians aged 14 years and over to have used cannabis one or more times in their life
  6. For more one in ten people to drink and drive
  7. For one in three people to lose their virginity before the age of 16 ( i.e. before the age of consent) and also to have multiple partners
  8. For 66 percent of all men and 41 percent of women to view pornography at least once a month, and that an estimated 50 percent of internet traffic is sex-related.
  9. For most people who join a gym to never use it

These and many other trends dictate what is ‘normal’ to the average person. It’s OK to eat and drink far too much because everyone else does. It’s OK to be promiscuous, watch pornography and take risks because that’s what people see happening all around them.

Against these trends the challenge is to seek opportunities when ‘normal’ is seen as risky and hopefully before that risk has manifested as pathology.

Picture by Mario Antonio Pean Zapat

Why wearables don’t work and people don’t floss their teeth

Wearable devices are now a billion dollar business:

Fitbit Reports $712M Q415 and $1.86B FY15 Revenue; Guides to $2.4 to $2.5B Revenue in FY16. Press release.

In a wonderful article from the Washington Post the author reports:

Another friend, a woman in her 40s, explained: “I realized that there were a couple weeks where I took it off because it was making me feel bad when I was ‘failing,’ so why do that to myself?” Steven Petrow

Associated Press pointed out that:

One research firm, Endeavour Partners, estimates that about a third of these trackers get abandoned after six months. A health care investment fund, Rock Health, says Fitbit’s regulatory filings suggest that only half of Fitbit’s nearly 20 million registered users were still active as of the first quarter of 2015. Anick Jesdanun

This is consistent with what my patients are telling me. I’ve seen the same trend with relatives. But it is all very predictable because these devices fail on one fundamental count. People are not logical. Information alone does not lead people to make choices. Humans are driven by emotion and not just information. If that were not the case people would floss their teeth, not text while driving or borrow more than they can afford to pay.

Innovations that rely on people acting on information to improve healthcare outcomes have no longterm future. If we want people to change their choices we need to accept that information alone does not lead to behaviour change. Functional Magnetic Resonance Imaging studies of the human brain have identified that our brains are resistant to change even when the change might be in our best interests. Habits drive our behaviours and are as an old pair of slippers, comfortable, familiar and easy. Change requires us to activate other parts of our brain, expend energy, learn and adopt new habits. Change requires effort which most people find uncomfortable. As a result, change is avoided and the easiest thing is to refuse to heed the message and bin the device.

There are three stages to adopting new behaviours:

  1. Unfreezing current patterns/unlearning old behaviours.
  2. Changing/applying new behaviours.
  3. Embedding new behaviours.

Of these wearables provide information that might get us underway with the first step by getting us to question the status quo. However that is far from what is required to get us to adopt a diet and exercise regimen. This so-called ‘disconfirming data’ is not enough – we can easily dismiss it, ignore it, or deny its validity. Which most people seem to be doing because it isn’t enough to generate new habits. Two other factors that are also essential to get us to the next stage:

  1. We need to accept that something is wrong and
  2. We need to believe that we can do what is necessary.

The ‘something wrong’ is the problem. Many people who are overweight or obese don’t see themselves are having a problem because in most cases the condition is asymptomatic. They may be surrounded by people who are of a similar body habitus and are therefore resistant to any notion that this body shape is in any sense abnormal. Finally for many people the idea that they might be able to change their shape is hard to swallow as in many cases they do not see results after weeks of effort.

For innovators a fundamental message is that there is no quick fix to healthcare problems because fundamentally humans are feeling not thinking creatures and therefore not responsive to messages that only tackle part of the drivers for change.

Picture by Philippe Put

Please don’t ask for my time as refusal may offend

In an article from the Economist, the authors advise that we appear to perceive ourselves as short of time. Every minute of the day can be filled ‘doing’ something or several things simultaneously, often badly. We are also increasingly wedded to the demands of our mobile devices. People spend time texting while eating, driving and even on the loo. In fact it is estimated that we spend one third of our waking hours interacting with our phone.

In an experiment carried out at the University of Toronto, two different groups of people were asked to listen to the same passage of music. Before the song, one group was asked to gauge their hourly wage. The participants who made this calculation ended up feeling less happy and more impatient while the music was playing. Participants reported experiencing greater impatience while listening to the music when prompted to think about their time in terms of money.

This has some important resonance in healthcare where practitioners who earn a living through fee for service may, and often do, become impatient about engaging with researchers or academics. This is a significant problem because in countries where primary care is the first port of call for most people we desperately need research to ensure that the service offered is both evidence based and cost effective. But what is the order priority: appointments with patients, results of tests, administrative paperwork, continuing medical education, family commitments, surveys or recruitment to research? There is an opportunity cost to all of these activities.

Research in primary care is challenging not least because as has been observed repeatedly it is very difficult to recruit participants to trials in primary care. Practitioners are not funded by the paymaster to do research. Costing the recruitment to adequately compensate for the time taken to seek informed consent is prohibitive and the medical defense organisations do not indemnify doctors to do research.

Therefore, it is often if not almost always, very challenging to conduct randomised clinical trials in that setting. Other methodologies, such as prospective observational studies may be more likely to yield data whilst at the same time acknowledging that this would not be classed as level one evidence. Nonetheless, we need to work within a paradigm in which neither patient nor practitioner is likely to participate or if they do attrition is substantial and threaten the validity and generalisability of the data.

We need more creative ways to test hypotheses and collect data in primary care. Methods that capture data from a substantial number of representative participants. Methods that generate a minimum disruption to the business of doctoring. Traditional research approaches are failing or are only really possible where doctors and patients have unlimited funded time. I don’t know where that is, meantime remain sceptical about results of trials that appear to have recruited and retained thousands of participants but where the practices were involved in a significant effort to recruit to research. In that scenario somebody has paid, always ask who and why.

Picture by M01229

Business R&D approaches may be the salvation of healthcare

There are lots of get rich quick ideas around. Just as there are lots of people who will tell you they can fix the healthcare system. But as the song goes:

There’s a guy works down the chip shop swears he’s Elvis
But he’s a liar and I’m not sure about you. KM

What really works requires knowledge of the business you want to fix, especially when you know what happens when the rubber hits the road, not a decade ago, not even five years ago but last Friday afternoon.

There are opportunities to improve every aspect of healthcare but that requires accepting one fundamental point:

People don’t really care whether their actions will please you or not.

People won’t change to please their doctor, their pharmacist, their mother or their government. They will adopt an intervention because it helps them do what they want to do for themselves. The opportunities lie in understanding how to work within these parameters. This is fundamental when it comes to business and people have made significant headway adopting this paradigm. Here are some ideas that demonstrate the power of knowing something about the needs and wants of the target audience:

  1.  Peerby from Amsterdam enables people to borrow expensive items from their neighbours, rather than splashing out on new products.
  2. Fortaleza Tour in Panama City is a walking tour set up by rehabilitated graduates of the Esperanza Social Venture Club — an organization dedicated to demobilizing Panama’s street gangs, integrating their members into society, and improving the area’s economy.
  3. Peru’s black vultures are well known locally for their natural aptitude for garbage location. In that country by fitting a flock of them with GoPros, the authorities collect real-time GPS data and enable the people to find the illegal dumps across the city of Lima.
  4. The UNPF is currently flying condoms, birth control pill and other medical supplies to the Upper East Region of Ghana using a fleet of long-flying drones.
  5. Many of the hosts on Airbnb are vacation property managers with multiple lettings. There are a number of startups offering management platforms and services that enable them to optimize their sub-letting business. Now, Parakeet is a platform that enables hosts to manage and monitor their property remotely via a cloud-based dashboard and keyless entrance system.

There are numerous examples of such out-of-the-box thinking. These innovations allow people to continue as before, to access equipment they hardly ever use but sometimes need, use their knowledge of a neighbourhood to make a living, use nature to monitor the environment, deploy technology to allow people to make personal choices and facilitate ownership of investment properties. The key aspect in each case is keen observation and insight.

This is needed in healthcare, local solutions that can be scaled to improve outcomes without imposing burdens on patients or practitioners. Nobody is as well placed to make these advances than those who already deliver and or avail of the services.

Picture by Cris

Why conclusions about doctor performance can’t be based on outcomes only

It sometimes seems ‘obvious’ why things go wrong in practice. For example, the proportion of people with diabetes prescribed a cholesterol reducing drug is low…. because? You might have your favorite answer at the ready. Others certainly do and will climb their hobby horse with little or no encouragement. ‘Prescribers don’t accept the guidelines‘, ‘patients don’t take their medicines‘, ‘people can’t afford the drugs‘ or ‘doctors don’t monitor patients‘. The truth may encompass any or all of these.

Let’s do the maths with reference to Glasziou and Haynes.

Let’s assume 80% is true in each of the following points:

1. Doctors are aware of the guidelines.
2. Doctors accept the evidence underlying these guidelines.
3. Doctors remember to apply the guidelines when the relevant patients present.
4. It is possible to do something practical to comply with the guidelines.
5. Doctors act to prescribe the relevant treatment.
6. Doctors and patients agree on the need for that treatment.
7. Patients comply with the treatment.

If these statements are true 80% of the time then 21% of people with the relevant problem will be managed according to the guidelines (0.8x 0.8x 0.8x 0.8x 0.8x 0.8x 0.8= 0.21). Experience tells us that in many, if not most, conditions only 1 in 5 people will be managed as per research evidence.

A quick review of the literature confirms this.

1. Only 17% of patients with diabetes were screened for sexual dysfunction despite it being a common complication of this condition.

2. A primary care study has shown that despite an active education program over two years the proportion of treated patients whose blood pressure was controlled to < 160/90 mm Hg remained at only 33%.

3. When examining the referral origin of all Colorectal cancer patients diagnosed in one study only 24% had been referred on a pathway that was consistent with national guidelines.

An agile and cost effective solution could be devised for any problem in medicine if the innovator has a firm grasp of the context in which an innovation is to be deployed. Doctors therefore already have some of the necessary credentials. And also, if a solution is generated that changes the way everyone involved feels when they do the needful. These are a function of the right brain (intuition and imagination) as opposed to the left brain (logic).

Lean innovators know that an innovation will only change outcomes if that treatment is less of a threat than any alternative, and or if it increases good feelings. We also know that the actions of doctors alone are never enough to result in a particular outcome. If medicine is a partnership then so are the results. As always context is everything.

Picture by algona81