A prospective study of nausea and vomiting in pregnancy

Roger Gadsby, BSc, MB, ChB, DCH, DRCOG, FRCGP
General Practitioner, Redroofs, Senior Lecturer in Primary Care, Warwick University, Warwick, Medical Student Teacher, Leicester University Medical School, Leicester, U.K.

My co-researcher and retired senior partner, Tony Barnie-Adshead, first became interested in nausea and vomiting of pregnancy (NVP) in the early 1960s. A patient came to him who was severely afflicted with NVP and asked him what was causing it. She stated it was so bad that she didn't want any more children. He looked into the literature of the time and the only writings in the British texts in those days proposed the theory that NVP was of a psychodynamic origin where the mother was rejecting the foetus. So Tony started to try and find out what could be the true cause of NVP. It was apparent to me, when I joined the practice in 1979, that nobody had clearly described the natural history of the condition. Furthermore, we felt that in British general practice we had the opportunity to do that. The study I will present is thus a descriptive, prospective study of the condition of NVP, as seen from the community perspective.

Few studies published describe the clinical history and daily symptom pattern of nausea and vomiting in pregnancy. In the UK, GPs have a fixed list of registered patients, usually of mixed socioeconomic class, who usually present early when they become pregnant. Our practice includes approximately 14,500 patients, with 5% comprised of ethnic minorities and the remainder being Caucasians.

Our study consisted of a study cohort of 435 women over a two-year period who consulted about a possible pregnancy and who were not requesting a termination.1 All patients included in the analysis (363 women) delivered a single live baby. Seventy-two patients were excluded from the study due to presentation after day 84 from last menstrual period (LMP), miscarriage, twins, treatment for symptoms of nausea, stillbirth, hypothyroidism, ectopic pregnancy, and miscellaneous disorders and reasons. Each patient was visited at home by the study midwife who obtained patient details and taught patients how to keep the daily symptom diary. Close contact was maintained, i.e., home visits every 14 days until symptoms ceased, to ensure the diaries were kept accurately. The pregnancy was dated by an accurate LMP, and confirmed by an ultrasound scan and the date of confinement. No drug therapy was prescribed for the NVP.

The mean age of our patients was 26 (range 15 to 40). There were 37% primagravidae, 42% gravida 2, 13% gravida 3, and 8% gravida 4 or more. The median day from LMP to the initial interview was 57, therefore there was only a small amount of retrospective data on the first days of patients' NVP, the majority of the data being prospective.

The results of the study showed that 20% had no symptoms, 28% had nausea only, and 52% had both nausea and vomiting. The mean number of days from LMP to onset and from LMP to cessation of symptoms were 39 and 84 respectively. In 40% of our patients symptoms ended abruptly. Cessation occurred in the majority of patients at around the same time whether symptoms began early or late or whether they were mild or severe. However, a group of patients were still having symptoms at day 120 and beyond. Other researchers have also identified a group of patients who seem to have persistent symptoms. It may therefore be postulated that there is a separate etiology for onset versus cessation of NVP, and perhaps the individuals who continue to have persistent nausea late in their pregnancy may have a problem with the mechanism involved in the cessation of NVP. The mean number of hours of nausea in the 292 women with symptoms was 56, with peak symptoms occurring in week 9. Women described their nausea as episodic; 85% experienced days with at least two episodes; 53% of vomits occurred between 06:00 and 12:00 with symptom incidence scattered throughout the waking hours. This time relationship may corroborate the use of the term "morning sickness."

In a subsequent review of our study cohort, we reported on the relationship between the total hours of nausea in early pregnancy, as a continuous variable, and various factors in women's obstetric and personal histories.2 Preliminary analysis showed higher levels of nausea were found in women who:

No significant associations were found between duration of nausea and:

Thirty-five percent of women with symptoms spent a mean of 62 hours away from paid work because of their symptoms. The study population included 206 women in paid employment. Applying this percentage to 1989 birth data in Britain gives a figure of 8.6 million hours per year lost to paid work because of NVP. We also had the patients record the hours lost to housework. The data there is much less accurate, especially since women reported not that they didn't do housework, but that they did it at a different time of day, when they were free of symptoms.

In conclusion, the condition is best described as episodic, daytime pregnancy nausea and vomiting. It has a significant socioeconomic impact through time lost to paid employment. We may postulate a hormonal etiology, with a suggestion that a different agent may be involved in the initiation of NVP versus the one involved in its cessation.


  1. Gadsby R, Barnie-Adshead AM, Jagger C. A prospective study of nausea and vomiting during pregnancy. Br J Gen Pract 1993;43:245-8.
  2. Gadsby R, Barnie-Adshead AM, Jagger C. Pregnancy nausea related to women's obstetric and personal histories. Gynecol Obstet Invest 1997;43:108-11.

    Gadsby R. Pregnancy sickness and symptoms: your questions answered. Prof Care Mother Child 1994;4:16-7.

Back to home