Clinical Review Lesson of the week

Danger of stereotyping in suspected osteomalaciaCommentary: Stereotyping may delay proper and systematic investigation

BMJ 2001; 323 doi: https://doi.org/10.1136/bmj.323.7305.149 (Published 21 July 2001) Cite this as: BMJ 2001;323:149

Clinical findings, not ethnic origin, should drive the diagnostic evaluation for suspected osteomalacia

Danger of stereotyping in suspected osteomalacia

  1. Sarah Sheikh, medical studenta,
  2. Ken Williamson, general practitionerb,
  3. Karen Kearley, general practitionerb,
  4. Sophie Bassindale, general practitionerb,
  5. Tim Lancaster (tim.lancaster{at}dphpc.ox.ac.uk), clinical reader in general practicec
  1. a Corpus Christi College, Oxford OX1 4JF
  2. b Jericho Health Centre, Oxford OX2 6NW
  3. c Department of Primary Health Care, Institute of Health Sciences, Oxford OX3 7LF
  4. Department of Primary Care and General Practice, Medical School, Birmingham B15 2TT
  1. Correspondence to: T Lancaster
  • Accepted 25 April 2001

Clinical findings, not ethnic origin, should drive the diagnostic evaluation for suspected osteomalacia

Clinicians frequently encourage medical students to recognise typical case vignettes (the travelling salesman with sexually transmitted disease, the New Forest hiker with Lyme disease). Pattern recognition is a widely used diagnostic strategy, but its distinction from stereotyping is not always clear cut. We report, from the perspective of general practice, on two patients in whom spurious pattern recognition led to delay in reaching the correct diagnosis.

Case reports

Case 1

A 21 year old English woman, whose parents had emigrated from Pakistan, presented to an accident and emergency department with fatigue and a painful, swollen leg. A venogram was negative. Investigations showed iron deficiency anaemia (haemoglobin 87 g/l, mean cell volume 69 fl), and she was treated with ferrous sulphate. She continued to have pains in the wrist and knee. She was not a vegetarian and ate a balanced diet that included fewer than three pieces of unleavened bread a week. She worked full time outside the home and wore dress that did not shade her face or always cover her arms. Serum levels of calcium (2.38 mmol/l), phosphate (0.91 mmol/l), and alkaline phosphatase (160 IU/l) were in the normal range, but a 25-hydroxycholecalciferol level was 6 µg/l (reference range 7-50). At a hospital outpatient clinic “Asian osteomalacia” was diagnosed. She was discharged to the Jericho Health Centre in Oxford with the recommendation to prescribe ergocalciferol and calcium.

Her pains improved, but she continued to complain of fatigue and developed a new problem of frequent loose stools. She was referred to a gastroenterologist, and a duodenal biopsy showed partial villous atrophy, consistent with coeliac disease.

Case 2

A 67 year old widow had emigrated from Pakistan to England 13 years ago. She was not a vegetarian and ate less than three pieces of unleavened bread a week. Her clothing usually covered her arms and shaded her face. After complaining of chest pains during an asthma exacerbation, she was admitted to hospital, where chest wall pain was diagnosed. Blood tests showed a high level of alkaline phosphatase (1146 IU/l), with low levels of calcium (1.8 mmol/l), phosphate (0.68 mmol/l), and albumin (34 g/l). Renal function was normal. The parathyroid hormone level was 24 pg/l (reference range 0.9-5.4); 25-hydroxycholecalciferol was 5.9 ng/ml (reference range 5-30 ng/ml) and 1,25-hydroxycholecalciferol was 13 pg/ml (20-50 ng/ml). She was discharged to the Jericho Health Centre with the diagnosis Asian osteomalacia.

Despite taking ergocalciferol and calcium she complained of pains in the chest and shoulders. Repeat investigations showed normal calcium and phosphate levels but a continued high alkaline phosphatase level (718 IU/l). She was referred to a physician for further investigation. Radiography and a radioisotope bone scan showed Paget's disease, corresponding to the areas where she complained of pain.

Discussion

Concerned by the delay in reaching a correct diagnosis in these two patients, we set out to learn more about the causes and presentation of Asian osteomalacia. We searched Medline using strategies for studies of causation and aetiology described by McKibbon.1 Our search yielded no prospective cohort studies. Most of the clinical studies that we identified were case series or controlled cross sectional studies.

Osteomalacia can usually be diagnosed by considering clinical, biochemical, and radiological findings. Nisbet et al found raised parathyroid hormone and alkaline phosphatase levels (without hypercalcaemia) to have the closest correlation with the gold standard of bone biopsy.2 Our first patient had a low level of 25-hydroxycholecalciferol, but her main problem was coeliac disease. Our second patient may have had a degree of osteomalacia, but further investigation showed that she also had Paget's disease. In both cases Asian osteomalacia was diagnosed on inadequate evidence.

In the 1960s rickets and osteomalacia emerged as a public health problem among communities who had emigrated from India and Pakistan to British cities.3 The reasons for this are still not fully understood. Most research focused on dietary factors and exposure to ultraviolet light. In a study of south Asian women, including 27 with a history of osteomalacia and 77 controls, Henderson et al found a higher risk of abnormalities of vitamin D among vegetarians, particularly in lactovegetarians who did not eat eggs.4 Similarly, in a study of 175 south Asian adults attending medical outpatients clinics, Finch et al found that the risk of osteomalacia increased with stricter vegetarian diets and was higher in female patients.5 On the other hand, Solanki et al, in a study of 54 Asians and 34 white controls, was unable to detect an increased risk of reduced vitamin D levels from a vegetarian diet.6

Reported exposure to ultraviolet light was an independent risk factor in some but not all studies. 4 5 Dunnigan and Henderson proposed an epidemiological model in which low exposure to ultraviolet light is not sufficient to cause osteomalacia but may increase the likelihood that it will develop in combination with dietary factors.7

Evidence is limited for the prevalent perception that intake of chapatis (a form of unleavened bread) increases the risk of osteomalacia. Phytate (inositol hexaphosphate) impairs the absorption of dietary calcium, and in small uncontrolled trials some patients improved on a diet that excluded chapatis.8 Henderson et al, however, failed to identify chapati intake as a risk factor for adult osteomalacia.4 High intakes of phytate may have a role but probably only in combination with other factors.7

Genetically determined variations in vitamin D metabolism could explain different levels of susceptibility to osteomalacia. One recent study found increased activity of 25-hydroxycholecalciferol-1a-hydroxylase, the rate limiting enzyme for degradation of 25-hydroxycholecalciferol in cultured skin fibroblasts from healthy Indian Asians compared with age matched white controls, but we found little other research into genetic factors.9

We were interested to determine whether people of Asian descent born in the United Kingdom are at higher risk than other ethnic groups (as expected if there is a substantial genetic contribution). We found no prospective studies that compared prevalence of osteomalacia in different ethnic groups. Data from hospital discharge summaries in Glasgow from 1968 to 1994 show that the prevalence of osteomalacia has fallen. This suggests that environmental factors, particularly dietary, contribute to risk.7

Conclusion

The epidemiological identification of an increased risk of osteomalacia has been important for improving public health and clinical care in south Asian communities in the United Kingdom. An epidemiological risk factor, however, is not the same as a disease entity. We could find no evidence to confirm or refute whether British Asians face a greater current risk than any other population group. Together with our recent clinical experience, this leads us to question the value of Asian osteomalacia as a clinical diagnosis. The diagnostic evaluation for suspected osteomalacia should be driven by the clinical findings not the patient's ethnic origin. Clinicians must be alert to the limitations of diagnoses based on stereotypical pattern recognition.

Acknowledgments

Contributors: SS took the histories, performed the literature search, and drafted the manuscript. TL obtained the patients' consent and supervised the literature search and the writing. SB, KK, and KW discussed the content at the planning stage and contributed to development and revision of the manuscript. TL is the guarantor for the paper.

Footnotes

  • Funding None.

  • Competing interests None declared.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.

Commentary: Stereotyping may delay proper and systematic investigation

  1. Joe Kai (j.p.kai{at}bham.ac.uk), clinical senior lecturer
  1. a Corpus Christi College, Oxford OX1 4JF
  2. b Jericho Health Centre, Oxford OX2 6NW
  3. c Department of Primary Health Care, Institute of Health Sciences, Oxford OX3 7LF
  4. Department of Primary Care and General Practice, Medical School, Birmingham B15 2TT

    These cases warn of the danger of labelling people. Stereotyping the patients on the basis of their ethnicity may have delayed proper and systematic investigation. The outcomes were a wrong diagnosis in the first case and an incomplete diagnosis in the second.

    Unfortunately we all have a tendency to stereotype others, in particular those we perceive as different from ourselves. Developing stereotypes seems to be a way coping with the uncertainty of the world and an attempt to make life more predictable. Yet in being stereotyped, people become victims of narrow judgments and prejudice.

    How can we become more aware of our tendency to stereotype others and avoid doing so? Better training may help. Medical education commonly reinforces stereotypes by emphasising ethnic differences in disease patterns at the expense of promoting reflection on personal attitudes to diversity and responding to people as individuals.1 The latter is more challenging for students and teachers alike but needs to be encouraged.2 A focus on the passive acquisition of knowledge about cultural differences still tends to predominate.

    Not surprisingly we are often more preoccupied with the modest differences between ethnic groups than with their greater similarities. Osteomalacia is an example of how public health priorities can become distorted by this myopia. Recognition of the increased risk of osteomalacia and rickets in south Asians (people born in, or descended from, those born in Pakistan, Bangladesh, or India) has led to major public health initiatives in the past.3 In comparison the growing importance of coronary heart disease in these communities remained neglected.4

    Many clinicians will recognise the common presentation of patients with non-specific tiredness and musculoskeletal pain as described in case 1. Diagnosis and management is often challenging. This is compounded if aspects of ethnicity such as class, sex, culture, and language are not shared and frustrate communication. It may then become easy for the demon of stereotyping to surface.

    So is “Asian osteomalacia” a useful term? Its currency may have jeopardised the care of the patients described. It is perhaps more common, however, to attribute similar patients' symptoms to unhappiness or distress, particularly when initial tests are normal. This risks applying another stereotype to “Asians” of “somatising” psychological ill health. The full syndrome of osteomalacia (proximal myopathy, diffuse bone pain, with attendant biochemical and radiological evidence) is now uncommon in the United Kingdom. Yet an awareness of so-called Asian osteomalacia might still serve as a helpful reminder to avoid making assumptions about patients. As with most clinical problems the use of low technology interventions, such as careful listening and examination before investigating all patients appropriately, is crucial.

    Sheikh and colleagues challenge clinicians to consider the boundary between diagnostic pattern recognition and stereotyping. I think the former involves being aware of summary patterns. It would be perverse to ignore them; they can alert us to possible explanations, and sometimes they are helpful. But stereotyping occurs when we fail to respond to people as individuals and inappropriately assume these patterns apply. We owe it to our patients and ourselves to be clear about the distinction.

    Footnotes

    • Competing interests None declared.

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    View Abstract

    Sign in

    Log in through your institution

    Subscribe