On Wednesday 23 March last year, Penny Campbell, my partner of 19 years and the mother of our beautiful son, Joseph, had an injection for haemorrhoids. "Was it all right?" I asked her on the phone that night. "You try having a big needle stuck up your bum and see if you like it," she replied. "Of course it wasn't all right, it was bloody uncomfortable." Piles was never a subject Penny liked to dwell upon. I got the message and we changed the subject, as invariably we did when I was away for work, to what Joseph had got up to that day.
Six days later, I was sitting in a black cab leaving the Royal London Hospital thinking about how I would tell a six-year-old boy that he would never see his mummy again, asking myself whether it would, in the longer term, be good for him to see Penny's bruised and bloated corpse lying lifeless in an intensive care room, her lovely smiling face transformed into a grotesquely deformed version of Marlon Brando's Godfather.
And in the haze of shock, disbelief and grief, I wondered how the life of a healthy, articulate 41-year-old could, in less than a week, have slipped through the widening cracks of our National Health Service.
On one level the culprit is easily identified: a bacteria that goes by the name of Group A Streptococcus. GAS is a Dr Jekyll and Mr Hyde sort of a bug. You can carry it on your skin or in your nose without it doing you any harm. But if it gets into your bloodstream it becomes an assassin, an "extremely virulent pathogen", in the words of Professor Susannah Eykyn, the clinical microbiologist who provided expert testimony at Penny's inquest last week. But like a playground bully, GAS is not so fearsome if it is stood up to. In a punch-up with almost any kind of antibiotic it will lose, provided they are administered in time. Penny died because that did not happen.
Her post-mortem showed the haemorrhoids procedure had been carried out without error. But because Penny must have been a GAS carrier the injection inoculated the bacteria into her bloodstream. By the following afternoon it was beginning to make her feel queasy and feverish. She woke on the morning of Good Friday between sheets drenched in sweat. Over the course of the next 48 hours, a rash, clamminess and constant, increasing pain in the abdomen and groin were added to the symptoms she reported to doctors working for Camidoc, the service that provides out-of-hours cover for our Islington GP and most of north London. At different points in the chain she reported a temperature of 41 degrees and that she could hardly walk. Yet it was not until she spoke to an eighth Camidoc doctor, Dr De Choudary, that she was referred to hospital, on Easter Monday morning. "She had severe, ongoing pain," Dr Choudary told the inquest. "It was obvious she had to be seen urgently." If only it had been so obvious to his Camidoc colleagues.
When Penny walked through the doors of A&E at the Royal London, her kidneys were already starting to fail. Within two hours, I knew she was going to die. We were moving up the old hospital's hierarchy too quickly, the just-out-of-college juniors in their sky-blue pyjamas giving way to consultants with suits on under their white coats. "I'm very worried about her," I overheard one of the house doctors saying on the internal phone. Penny was as lucid as ever as she was rigged up to various drips. "Don't let Angus see any needles, he'll faint," she warned the nurses, knowing I was in earshot. A few minutes later, one of the consultants took me into a corridor. "You have to understand, Penny is very, very ill. She could die."
How do you find the words when you know in your heart they will be the last you ever speak to someone you have loved for your entire adult life? Penny and I had been a couple since I was 19, a second-year politics and modern history student at Manchester University. She was doing a post-graduate MSc in science and technology policy. On the day we met, I was wearing bleached jeans with rips in them and had the most ill-advised of mid-Eighties hairstyles. But Penny was a forgiving person and our relationship grew stronger on our travels from Manchester to Nice, through London, Paris, Brussels, where Joseph was born, and Hong Kong. By the time we got back to London in 2002, ripped jeans and mullets were back in fashion. It is a lot to sum up in the seconds before an anaesthetic kicks in. "I love you and Joseph loves you," I told her and she responded with one of the contented little hmmphs with which she always replied to hearing me say that.
I believe Penny did not think she was going to die and she was not in pain when she did, at 7.50am the following day. It would have been consolation of a kind if her death had been inevitable. A car crash, a heart attack or cancer: the pain involved in all of those would have been easier to cope with than the thought that she was betrayed by a scarcely credible combination of organisational chaos and gross incompetence on the part of the service that had her life in their hands over that Easter weekend. Betrayed, too, by my failure to recognise what was going on before all the ingredients of the fatal cocktail were exposed in the Victorian chambers of Poplar Coroner's Court last week.
Penny considered herself a lucky person. That sometimes irritated me. She used it to play down the intelligence and charm that made her able to adapt and thrive over the course of a career in which she was, variously, a telecoms systems engineer, a media analyst and a teacher of both degree-level maths and basic literacy, before settling on journalism and rising to become an associate editor at the European edition of Time magazine. From the moment she picked up the phone to call a doctor on Good Friday, it was clear Lady Luck was no longer her friend.
When Penny began to feel ill, on the Thursday, she blamed a lunchtime fish chowder. But it was the colorectal surgeon who had injected her, Miss Sue Clark, she called first, anxious to rule out a reaction to the previous day's injection. Of all the doctors Penny consulted over that holiday weekend, Miss Clark would have been the only one certain to be aware that serious septicaemic infection is a possible, if very rare, complication of the procedure. Consequently, she told Penny it was unlikely her symptoms were linked to the injection but, she revealed at the inquest, would have arranged to see her again if she had not been due to leave the country on holiday that evening. Instead, and fatally, Penny was placed in Camidoc's care.
Everything proceeded smoothly enough to start with. A call to Dr S El Kinani, in which she complained of stomach pains and nausea, resulted in her being advised to come in to the Camidoc centre at St Pancras hospital to be seen face-to-face by Dr Michael Fitzpatrick, a GP with nearly 30 years' experience. He noted that Penny was febrile but did not record her temperature or her blood pressure and diagnosed a probable viral infection coincidental to her injection. Penny's fever that night was sufficient to ensure the sweat-soaked sheets had to be changed in the morning.
In a properly functioning system, Dr Fitzpatrick's failure to record Penny's temperature or blood pressure might have hampered the work of the next doctor she consulted. Only "might have" because, in reality, none of the six lines of notes he scribbled on the Camidoc call sheet were ever going to reach Dr Ranko Vucevic, who spent eight minutes talking to Penny over the phone on Saturday afternoon. Dr Vucevic was working at a different Camidoc centre, at Homerton Hospital in Hackney, which meant he had no access to the notes made by either Dr El Kinani or Dr Fitzpatrick, unless he specifically requested them to be faxed over. He didn't and thus he was able to record, on the basis of Penny's own account, that her abdomen was "slightly tender to the touch" but not the fact that this amounted to an overnight change in her condition: Dr Fitzpatrick had reported the abdomen to be "soft". At the inquest, Dr Vucevic stated that he had had a "suspicion something was going on", but, having ruled out appendicitis, he diagnosed flu.
In his written statement to the coroner, Dr Vucevic chose to stress that Penny "did not at any point request that she should be seen either at home or at the Camidoc centre". Was the unspoken implication of that recollection, made six weeks after her death, that Penny, a confident, fluent speaker, would have known whether she needed to be seen? I can't be sure but it was not the only point in her treatment where her eloquence, even while in pain, appeared to result in the symptoms she was describing being downplayed in the mind of the doctor at the other end of the line.
Like Dr Vucevic, Dr Kathleen Wenaden, whom Penny spoke to on the phone just before 11pm that night, had no access to the notes from the previous consultations. That meant it fell to Penny to inform her that the rash that had appeared on her chest was a new symptom. Penny did not, however, tell her about her injection of the previous Wednesday. Unlike Dr Fitzpatrick and Dr Vucevic, Dr Wenaden did not offer a diagnosis, telling the inquest her role was restricted to the exclusion of serious illness, a duty she felt she had discharged by checking that the rash was non-blanching, ie not indicative of meningitis. She "agreed" with Penny (was that her articulate nature taking effect again?) that she should phone the following day and "should be seen".
It was shortly after this conversation that I arrived home. Penny was obviously sick, uncharacteristically irritable and desperate to get off to sleep. Maybe that was why she did not mention to me that Dr Wenaden had told her she needed to see a doctor. As Dr Wenaden was at Homerton, not St Pancras, no recording was made of the consultation, contrary to the requirement set down by the Government in its guidelines for such centres when they decided to take out-of-hours care out of the hands of GPs from the start of last year.
According to Camidoc's records, which contradict my recollection of calling earlier, the first contact we had with the service the following day was at 11.40. Ten minutes later, Penny reported to Dr Sylvia Laquer a temperature of 39 degrees, a rash, nausea, oral thrush and abdominal pain which meant she "can hardly walk", which was underscored in the notes. Like the other doctors, Laquer had no previous notes to consult before she recommended a "non-urgent" home visit. Asked to explain that at the inquest, Dr Laquer recalled that Penny had not sounded "very distressed", yet she could not recollect whether she had elicited any information about the nature of the rash reported to her. One month after she spoke to Penny, 17 months before the inquest, Dr Laquer had submitted a written statement to the coroner in which she said she could not recall the consultation at all.
Dr Laquer's non-urgent visit request resulted in Dr Bengi Beyzade arriving at our house four hours later. We had waited twice as long as we should have done under Camidoc's own targets but such delays were common, Dr Beyzade told the inquest. After a 15-minute examination of Penny in our bedroom, Dr Beyzade recorded that she had earlier been running a temperature of 41 degrees. When he checked, it was normal (37.3 degrees) but he noted that Penny's hand was "clammy" to the touch. That is a symptom that can indicate a drop in blood pressure that happens when a blood infection begins tipping into full-blown toxic shock. It was not, however, a possibility that apparently occurred to Dr Beyzade, who failed to check Penny's blood pressure.
He said at the inquest he had been reassured by a normal pulse reading, a finding he failed to record in his notes of the visit. Nor did Dr Beyzade register the rash Penny had last reported four hours earlier while confirming that he had not inspected the front of her chest below the level of the nightdress she was wearing. His diagnosis was food poisoning, and as he left the house he reassured me that, yes, that could explain the rash. Both suggestions were dismissed by Professor Eykyn. Rash is not part of the general clinical description of food poisoning, she told the coroner's court, before questioning how food poisoning could be diagnosed in a patient with no history of diarrhoea who had vomited only once in four days.
It seems so obvious now, but at the time, it did not occur to me to be sceptical. With the benefit of hindsight and second opinions, I now know that, at the very moment I was feeling a quasi-euphoric sense of relief that everything was OK, Penny's chances of survival were slipping away. Sent to hospital at that point, given antibiotics and organ support, she would have survived, Professor Eykyn believes. Maybe she would even have survived if she had got to hospital five hours earlier than she did.
But that was a possibility that disappeared when she spoke to a Dr TP Chuah just before 5am on the morning of Easter Monday. Dr Chuah, like his colleague Dr Laquer the previous day, recalled that Penny had not "sounded" distressed. Yet he felt able to record an "impression/ working diagnosis" of colic as an explanation for her severe abdominal pain. He did not elicit from Penny when her pain - first reported on Friday evening - had started. To have got that information from anyone other than Penny, Dr Chuah would have had to find the reference numbers for six different call sheets and requested faxes from two other centres. It was not just in Penny's case that it did not happen, it routinely did not happen.
So who do I blame? Where do I send the bill for the extravagant presents with which I've tried to make two motherless birthdays bearable for Joseph? How do you compensate Penny's mother, Pat Campbell, 72, for spending the rest of her life without a daughter she regarded as her best friend? The civil courts will assess the extent of the liability, if any, of the individual doctors involved for their errors of judgement and failure to elicit basic information that may have compensated for the fact that the Camidoc system could not ensure respect for one of the founding principles of the NHS: continuity of care. At the inquest, Camidoc's chief executive, Michael Golding, accepted that the system had malfunctioned and claimed an internal investigation into Penny's case had acted as a "spur" to the computerisation of patient records.
Yet, 18 months after Penny died, there has been no review of the performance of the doctors who treated her and the coroner's judgement that the issues raised by her tragic case warranted referral to the Secretary of State for Health has yet to elicit a response from Patricia Hewitt.
It is not hard to understand why. Taking responsibility for out-of-hours care away from GPs is beginning to look like Labour's own clinical negligence case. Introduced without any kind of pilot trial, it is a reform that has failed on every count.
The National Audit Office reported in May that the new system had cost £70m (22 per cent) more than anticipated in its first financial year of operation. As few as 15 per cent of primary care trusts were meeting targets on the speed of consultations, although the NAO confirmed the picture was clouded by poor record-keeping. What is clear, however, is that serious patient complaints are on the rise - up 66 per cent between 2002 and last year, according to the Medical Defence Union, the body that helps defend doctors against allegations of malpractice. And how many cases like Penny's have not come to light because the victims, unlike me, did not have the means to employ solicitors and barristers to try to decipher the doctors' notes?
So why did Tony Blair let this happen? The political rationale for this reform appears to have been simply that something had to be done to raise morale among doctors. In return for giving up £6,000 a year (offset by other reforms that boosted their income), GPs were allowed to stop working nights and weekends. But why shouldn't they? Young journalists are always told, "If you want to work banking hours, go and work in a bank." Medicine, like journalism, is supposed to be a vocation. A system that reflected that principle might have been capable of preventing Penny's needless death.
How GPs cut their hours
* Forty years ago, GPs cared for their own patients at all hours, including nights and weekends, visiting them at home when they were too ill to go to the surgery.
* They were responsible for their patients 24 hours a day and shared cover out of hours with colleagues in the same or neighbouring practices.
* During the 1970s, commercial deputising services grew up offering to provide cover for GPs on some nights and weekends, in return for a fee paid by the GP.
* The deputising services were staffed by qualified doctors wanting temporary work or who wanted to earn extra cash.
* But GPs were banned from contracting out all their night and weekend cover on the grounds that it was an essential part of their training to see families in extremis at home.
* Through the 1980s and 1990s, GPs did progressively less out-of-hours work themselves and contracted out more and more to deputising services, but remained responsible for their patients 24 hours a day.
* In January 2005, the Government bowed to pressure from GPs who wanted to shed their round-the-clock responsibility and passed it to primary care trusts, who contract the work out to private companies.
* A report by the National Audit Office last May found the new system was a shambles, with slow response times and rising costs.
* A doctor could earn £4,000 for a weekend shift, the NAO report found, and foreign doctors were flying in to the UK to take advantage of the generous rates.
Join our new commenting forum
Join thought-provoking conversations, follow other Independent readers and see their replies