The night my daughter was born, I was filled with conflicting emotions. I had dreamed, hoped and prayed for a baby and now here she was, at last. While I was overjoyed, I still could not believe that this longed-for, perfect child was actually mine. Caroline, now 15, will always be extra special to me because I never thought I'd be lucky enough to have her. After years of tests to find out why I was not getting pregnant, followed by invasive medical treatment and devastating miscarriages, my beautiful baby, my own miracle, had entered the world.

Nowadays, the IVF treatment I underwent to conceive her is almost commonplace, but back then women like me still felt like pioneers. The birth of the first IVF baby, Louise Brown, had happened only a dozen years earlier. I didn't know anyone who had had a test-tube baby, as they were then called, and the whole experience felt like a leap in the dark. I was undeterred though: like most infertile women, I was driven by an all-consuming need to hold my own child in my arms, whatever the cost.

Yet for some, according to research published last week, the cost could be high. A study at the University of California appears to show that children born by IVF are more likely to be diagnosed with autism, childhood cancers and cerebral palsy. The researchers used more than 19,000 medical records to compare the pregnancies, births and child health outcomes of fertile women with those of women who conceived after fertility treatment and those who conceived naturally after a long period of infertility. They also found that children born to infertile women were 40 per cent more likely to have attention deficit hyperactivity disorder and 20 per cent more likely to be born with low birth weight.

Some of the problems with children's health may be caused by medical problems in their infertile parents, the doctors say. But fertility treatments may also be a factor. So should couples be warned about the possible adverse affects of fertility treatment? And how much would such a warning affect their decisions?

I'd been married three years and been trying for a baby without success. My GP advised me to take my temperature every day to find out when I was ovulating and time intercourse appropriately. First problem: I didn't appear to be ovulating. My husband went for tests: his sperm mobility and motility were fine. When the drug clomiphene failed to kick-start my ovulation, an operation showed that my fallopian tubes were severely blocked, and that our only hope was IVF.

I was taught to inject myself with drugs that would force my ovaries to produce the requisite amount of eggs. In the two cycles I went through, I experienced what the doctors assured me was only "mild" ovarian hyper-stimulation, though I was in agony with a grossly swollen abdomen. Both cycles ended in success and then failure - pregnancy and then miscarriage. Third time round, we got lucky and our baby was born.

Three years later, my husband and I decided to try for a sibling for little Caroline. We had frozen embryos in storage, but they failed to "take". After a year off, to give my body time to recuperate, we started the long and arduous process of IVF again. Thankfully, I only had to endure one cycle this time, resulting in the birth of our second gorgeous daughter, Dora.

Since then, I have barely given those years of treatment a thought. My girls are happy and healthy, and life with them is so busy that mulling over the past seemed an indulgence. And I had almost forgotten about it, until the day I idly did a Google search on the word "IVF" and discovered alarming reports of health risks for children like mine.

I was aware that the main risk of artificial reproductive techniques (ART) comes from multiple births, leading to premature babies who have an increased risk of cerebral palsy, of being underweight, being delivered by Caesarean and being more likely to die around the time of their birth. I also knew that the Human Fertilisation and Embryology Authority (HFEA) wants clinics to offer patients single embryo transfers, because of this.

But the University of California study is not the first to raise questions about other health effects on children born after IVF. I discovered a US study which showed that full-term singleton babies conceived through ART were more than twice as likely as infants in the general population to be born underweight. I found out that IVF babies, according to a Swedish study, have three times the rate of cerebral palsy compared with children in the general population, whether they are twins, triplets or singletons of normal birth weight.

In my blissful ignorance, I hadn't realised that an Australian study (published in the prestigious New England Journal of Medicine) has found that babies conceived through assisted conception are more than twice as likely as naturally conceived infants to be diagnosed with multiple birth defects in their first year of life. These infants had more major cardiovascular, genitourinary, chromosomal and musculoskeletal defects.

I learnt that the use of frozen embryos raises the risk of an ectopic pregnancy 17-fold, and that intracytoplasmic sperm injection (ICSI), in which a single sperm cell is injected directly into an egg, helps men with low sperm counts but also bypasses the body's natural mechanisms for weeding out weak sperm, meaning the sperm has a greatly increased risk of carrying damaged genetic material.

Previous studies have also suggested that IVF children may be more prone to developing cancers. Dutch researchers found that these babies have a five to seven times higher chance of developing a rare form of eye cancer than those from naturally occurring pregnancies.

A recent UK study claimed that IVF children run four times the normal risk of a rare genetic imprinting disorder, Beckwith-Wiedemann syndrome, which causes too much growth, kidney abnormalities and an increased risk of tumours. The authors argue that children conceived by artificial techniques should be followed up during the first years of life.

Dr Wolf Reik, one of the study's authors, added that other research has suggested an increased risk of Angelman syndrome, an imprinting disorder that causes neurobehavioural symptoms, in children born after IVF. He said: "Obviously, the vast majority of IVF babies are born healthy and happy, but parents should be informed of all potential risks. I also feel that the IVF community has not taken on board the need for study of these children's long-term health. For example, I know of no study of cancer rates among artificially conceived children. Research on such basic questions is sorely needed."

Professor Lord Robert Winston has spoken of "lowering clouds" over assisted-conception therapies he pioneered, with women and children being used in a "mass experiment" by the medical community, trying out techniques before they have been properly researched. As to the mums concerned, one research project found that infertile women who had taken fertility drugs had 2.7 times the risk of developing ovarian cancer of those who had never taken them. University College London is currently investigating this.

Dr Richard Kennedy is a consultant obstetrician and gynaecologist, and a spokesperson for the British Fertility Society. He agrees that, with up to 4 per cent of IVF babies having something wrong with them compared with 2-3 per cent of all babies, it's important to discover the cause of the discrepancy. "Monitoring of IVF offspring for the next 20 to 30 years is long overdue. I'm broadly confident there are no serious long-term risks to mums from the treatment, but as for babies, we're just not sure. Remember, though - nothing in medicine is risk-free."

Comment: Nothing in allopathic medecine is risk-free because it always comes at a price.

But why is this "long overdue" follow-up of IVF babies not in place already, as we come up to three decades since the first birth? Professor Peter Brodie, a specialist in reproductive medicine at Guy's Hospital, says: "The lack of it has been a big problem for the UK, because patients do need to be reassured. But the HFEA, which keeps the database of who's had treatment, is prevented by law from sharing that information with, for example, cancer and death registers. It's common sense that that situation should change."

Now this legislation is under review, with the HFEA recommending that restrictions on its database are lifted to enable proper monitoring to take place. However, a spokesperson warns: "Many IVF parents still feel there's stigma attached to infertility, don't want people to know they had treatment and will refuse to take part in a follow-up."

Letting your emotions cloud your judgment in this way is typical of those of us touched by infertility. According to a Scottish survey, women awaiting IVF are more willing to have a child with a chronic adverse birth outcome including cerebral palsy and cognitive and visual impairment than have no child at all. I empathise with this desperation, but we parents have buried our heads in the sand for long enough. An IVF baby may be an everyday miracle, but we need the hard facts to find out what we're letting ourselves in for.

IVF: the facts
  • Around 30,000 patients have IVF in the UK, according to the latest published figures (2004).
  • Over 10,000 babies were born as a result - from 8,250 births.
  • Of 100 couples trying to conceive naturally, 20 will conceive in a month, 70 in six months and 95 in two years.
  • The National Institute for Clinical Excellence (NICE) defines infertility as failing to get pregnant after having regular unprotected sex for two years.
  • Success rates for IVF average 28.2 per cent for women under 35, and 10.6 per cent for women aged 40-42.
  • NICE recommended couples should be offered three cycles of treatment on the NHS if the women is aged between 23 and 39, if there is an identified cause of their infertility or they have not conceived for three years.
  • The Government said from 1 April, 2005 all women with appropriate clinical need should have at least one cycle of treatment paid for by the NHS but many Primary Care Trusts are still refusing to fund it.
  • Typical costs of private treatment are £3,000 per cycle plus the price of consultations, drugs and tests.
  • There are 85 licensed clinics in the UK of which 52 see NHS patients.
  • The single biggest risk of treatment is multiple births which are more likely to be premature.
  • A review in the US journal Pediatrics (2004) concluded that the risk of birth defects in IVF children was 40-100 per cent higher than in the general population.
  • Third bullet item.
  • Jeremy Laurance