Primary Care

Twenty five years of requests for euthanasia and physician assisted suicide in Dutch general practice: trend analysis

BMJ 2003; 327 doi: (Published 24 July 2003) Cite this as: BMJ 2003;327:201
  1. R L Marquet, senior investigator (r.marquet{at},
  2. A Bartelds, sentinel network director1,
  3. G J Visser, senior investigator1,
  4. P Spreeuwenberg, statistician1,
  5. L Peters, senior investigator1
  1. 1Netherlands Institute for Health Services Research (NIVEL), Utrecht, Netherlands
  1. Correspondence to: R L Marquet
  • Accepted 6 June 2003


Concerns have been expressed that the Dutch policy on euthanasia (E) and physician assisted suicide (PAS) may lead to an exponential increase in the number of requests and use.1 Many Dutch general practitioners, nursing home physicians, and pharmacists have a fairly positive attitude and have become more tolerant over the years.2 We investigated the effect of increasing acceptance on the number of and underlying reasons for requests for E/PAS inDutch general practice from 1977 to 2001.

Methods and results

The data were derived from the Dutch Sentinel Practice Network, which constitutes a sample of about 60 general practitioners, covers about 1% of the Dutch population, and is fairly representative of the total population (16 million) with regard to age, sex, geographical distribution, and level of urbanisation.3 Every year general practitioners reported data on requests for E/PAS from terminally ill patients, including age, sex, underlying disease, reasons for request, and presence of a living will. We estimated trends with multilevel analysis.

Over the 25 years the network received 915 requests (equivalent to 3660 requests/year; 2.6 requests/year/10 000 inhabitants). General practitioners in cities received 3.3 requests per 10 000 patients compared with 1.9 requests in rural areas. Over half (503) of the requests were from men (mean age 67 years, range 31-96 years); the mean age of women was 68 years (range 32-88 years). Most patients (769) were nursed at home; 503 had stated a wish for E/PAS in a living will (increasing from 15% in 1984 to 87% in 2001), with no differences in age and sex. The mean proportion of requests for PAS, separately recorded from 1987 onwards, was 6.7%, decreasing from 9.5% in 1988 to none in 2001.

The number of requests increased from 1600 in 1979 to 4000 in 1985 (extrapolated data, figure). The number then stabilised at about 5000 requests a year. Most patients (74%) had cancer, mainly gastrointestinal and lung cancer. Less common were cardiovascular diseases (7%) and chronic obstructive pulmonary disease (5%). Among the other diseases, those of the musculoskeletal system, neurological diseases, and AIDS were most frequently mentioned. Fear of pain (37%), deterioration (31%), hopelessness (22%), and dyspnoea (15%) were the most important reasons for requests. The figure shows the trends in the reasons for requests. Pain became significantly less important, whereas deterioration became more important. The other trends (hopelessness and dyspnoea) were not significant.


Number of requests for euthanasia/physician assisted suicide (E/PAS) per 10 000 patients (raw scores and trend; multilevel continuous regression techniques used to estimate trends with 95% confidence intervals calculated on basis of variance between years) and major reasons for requesting E/PAS (multilevel regression technique used to estimate trends)

The responses from the same practice over the years were positively correlated. In addition, the data were unbalanced because several sentinel practices participated only during a certain number of years. The correlation between sentinels indicated that 18% of the total unexplained variance in the dependent variable (after we controlled for the yearly trend) was due to differences between practices.


The number of requests for euthanasia or physician assisted suicide increased in the first decade of registration in the Netherlands, but from 1995 onwards stabilised at about 5000 requests per year. The increase probably reflected the process of liberalisation in the early years, boosted by broad publicity on lawsuits and the foundation of the Dutch Society for Voluntary Euthanasia. Mainly due to the activities of this society the number of living wills has increased substantially over the years.

The importance of pain in such requests decreased significantly, paralleled by a proportional increase in the importance of deteriorating health. Improvements in pain management and the increasing importance of feelings like self esteem are obvious reasons for these changes.4 Over the past decades the willingness of both physicians and the general public in the Netherlands to accept E/PAS has increased. This attitude resulted in the acceptance of a law tolerating E/PAS performed in compliance with strict regulations.5 Some people feared that the lives of increasing numbers of patients would end through medical intervention, without their consent and before all palliative options were exhausted. Our results, albeit based on requests only, suggest that this fear is not justified.


We thank the general practitioners from the Dutch Sentinel Network for their invaluable contributions over the years and Marianne Heshusius for her secretarial support.


  • Contributors RLM was responsible for the overall conduct of the study and for writing the paper. AB initiated the survey and is guarantor. GJV was responsible for data analysis. PS undertook the statistical analysis. LP was the data manager.

  • Funding None.

  • Competing interests None declared.

  • Ethical approval The ethical committees of the Dutch General Practitioners Association and NIVEL approved the investigation


View Abstract