A FEW MONTHS AGO I woke up on a weekday morning with a most intense, unbearable pain in my head and neck.

By the time it had gone, I had come to understand that the word “migraine” means something much, much more than a bad headache; to grasp something of what the crisis in mental health services in Australia has meant for its inner city hospitals; and to see that while hospitals are full of brilliant medical staff, injecting common sense into patient care can nevertheless be a struggle.

After being admitted to the hospital, I was left in a chair in a brightly lit hallway, where I was discovered by a six-foot, barefoot, dreadlocked, transgendered woman, who wanted me to waltz with her.

AAP/Hugh Macknight/PA Wire

Once rescued by a very small, determined nurse, I was put into a bed beside a man in a wheelchair, who kept getting out of the wheelchair, Little Britain style, whenever the nursing staff disappeared. He also spent a lot of time in the bathroom. Desperate for a quiet, darkened room, I was instead lying in an open ward forced to listen to an exasperated doctor repeatedly ask the man why he has been brought to the hospital. What was he doing here? He finally replied, “Oh, love, I’m raging bipolar, I am.”

I tell you these things to emphasise that when you are sick the very last place you can recuperate is a busy inner-city hospital ward. You simply must get yourself home if you are in any state to do so.

But figuring out what state you are in, seems difficult for the doctors.

I came in with a migraine. I had a brain scan; it was clear. I was given intravenous migraine medication that made me feel much, much better. Thus I concluded I was probably suffering from a migraine. I wanted to go home, a desire that only intensified when I realised that Little Britain had spent all that time in the bathroom smearing faeces all over the tiled walls.

Not so fast, said the doctor! There was an extremely slight chance I was having a stroke and should therefore seriously consider having a lumbar puncture — a spinal tap, a big needle in my backbone — to see if my brain was bleeding into my spinal fluid. (To be fair, both my parents had suffered strokes by the age of 30, so I probably had him spooked.) “Do strokes usually respond to migraine medication?” I asked, speaking very deliberately through the peripheral pain of the migraine. “No, they don’t,” he replied. Right. I’m off.

Afterward, I tell this story to two doctors.

One recent medical graduate insists it was entirely reasonable, offering a mother-of-two like me, with my medical history, a lumbar puncture. He thinks me reckless for refusing.

The other doctor is Dr Ranjana Srivastava, a consultant oncologist who sees general hospital patients as well as those needing cancer care. When I tell her about my hospital experience, she understands why I left that day, happy to take the 1 per cent chance that I was having a stroke.

Srivastava’s recently published essay, Dying for a Chat — The Communication Breakdown Between Doctors and Patients, looks at why patients end up having unnecessary, expensive tests. She also questions why the old and dying are often subject to relentless, painful treatments in their final days.

The author is a kind woman, frank and intelligent. She has written widely on the challenges of doctor-patient communication, including a book, Tell Me The Truth — Conversations With My Patients About Life and Death.

When she tells me that just days ago before our interview, she’d diagnosed a relatively young man with terminal cancer, she does so quietly and directly. She has seen a lot of death growing up in India; more than most Australians will ever see. And now as an oncologist she strives not to avoid the subject in the Melbourne hospital where she works. You can almost imagine how she would appear across a desk, looking you in the eye and gently telling you the worst news.

Yet her essay, published by Penguin, starts with a mea culpa about one of her own patients, who should have been allowed to die in peace.

Ella Rubeli/The Global Mail

An elderly, frail woman — Mrs Johnson — was brought to hospital with a mild infection. Srivastava was overseeing her care. As the woman’s condition deteriorated she became the unfortunate subject of intensive, invasive and inevitably futile medical heroics by an array of specialists.

“I kick myself for not having been more diligent, by safeguarding her against the assault of so many opinions and interventions,” Srivastava writes. “But it’s hard to do. It’s very difficult to hint to another physician that at 90, a patient needs less not more.”

Srivastava recounts an earlier event in Mrs Johnson’s stay at the hospital. A group of doctors, including the author, all crowded around Mrs Johnson’s bed. While they all observed the patient’s growing distress, they were too busy talking about oxygen levels and intravenous antibiotics to allow the old woman to speak.

“She turns to me and catches my eye. A single tear escapes to her cheek. Reluctant to start the day with regrets I sit down in the chair next to her,’’ Srivastava writes.

“Is something bothering you, Mrs Johnson?’’

“‘Please can someone remove my bedpan from under me? Its overflowing and I’m cold. I just want to be comfortable.’”

Mrs Johnson had watched her husband spend three “awful” months in hospital before he died, and she had told her own adult children, “If I become like that, just let me go’’. But she had never given any more precise instructions about how she wanted to be treated at the end of her life.

Srivastava writes there was one final opportunity to really talk to Mrs Johnson about her wishes. But somehow that conversation went awry too.

“Her latest numbers do look bad,” she writes, recalling the sudden deterioration in her patient’s condition. “Her kidneys are protesting and something will need to be done urgently.”

She imagines what she could have said to the patient:

“Mrs Johnson, there has been a serious change in your health.”

“Mrs Johnson, we need to talk about a difficult subject.”

“Mrs Johnson, we will need your help with this.”

“These are the things she needs to hear,” Srivastava continues. “Instead I hear myself saying, ‘Mrs Johnson, we will need to work on your kidneys a bit.’”

“I hate [my] deliberate vagueness … Tomorrow I will pull up a chair, sit down and ask her how she really is. Tomorrow, I promise myself, things will be different — this is not the way I want to practise medicine.”

But a midnight phone call to Srivastava’s home brings news of Mrs Johnson’s heart attack. A crisis. Too late for conversation. Asked what they want to do, her family reply that — of course — they want everything done to save her.

After the weeks in hospital and despite the scepticism of the doctors about any chance of survival, Mrs Johnson spends a final 11 harrowing days in intensive care — with artificial feeding, one device to stand in for her breathing, another to do the job of her kidneys — before her inevitable death.

“Long after she has died, Mrs Johnson’s case keeps turning in my mind, the taste of dissatisfaction and an unfulfilled contract of care rancid in my mouth,” Srivastava writes. “Was it my fault?

“If the patient is dead, the family is dissatisfied, the doctors are disillusioned … who is it that modern medicine claims to serve?”

In this audio interview Srivastava argues that with an aging population — on some projections, by 2050 life expectancy for Australians will be 93-plus years for men and 96-plus for women — doctors and patients need to find a better way to talk about their medical choices throughout their lives and as they die.

We began by discussing Dr Srivastava’s experience with Mrs Johnson.