correspondence
Although financial incentives could change doctors’ behavior,2 it is difficult to ascertain whether the achievement reported by Doran et al. is due solely to incentives or to an improvement in clinical practice in general, since there is no control group and there are no baseline data. In our area, organizational care indicators for diabetes, such as data recording, have increased dramatically, but clinical indicators, such as cholesterol levels and glycated hemoglobin values, have revealed a smooth increase that might be due to other factors, such as the use of national targets and the active dissemination of guidelines. Pay for performance could result in a loss of the holistic approach to patient care,3 and patients with diseases that are not included in the contract could be put at a disadvantage.3 Incentives may need to increase with time to maintain targets. Pay for performance may be a good idea, but it should be implemented with caution. We would recommend that when this approach is introduced into a new area it be started as a pilot, so that some comparisons with conventional care as a control can be made. Abd A. Tahrani, M.D., M.R.C.P. Royal Shrewsbury Hospital Shrewsbury SY3 8XQ, United Kingdom
[email protected]
George I. Varughese, M.R.C.P. University Hospital of North Staffordshire Stoke-on-Trent ST4 6QG, United Kingdom
fects of financial incentives on medical practice: results from a systematic review of the literature and methodological issues. Int J Qual Health Care 2000;12:133-42. 3. Roland M. Linking physicians’ pay to the quality of care — a major experiment in the United kingdom. N Engl J Med 2004; 351:1448-54.
The editorialist replies: I agree in general with Tahrani et al. As noted in my editorial, the findings reported by Doran et al. could reflect a number of different factors other than improved performance as prompted by the payment incentives. And surely there are a number of reasons to have modest expectations for the improvement in quality associated with pay-for-performance programs and to be wary of the potentially deleterious side effects they may inspire. There have been relatively few studies of pay for performance in health care.1,2 On the whole, their findings are not encouraging, although most of the programs studied may not be comparable to the large efforts now envisioned. Numerous pay-for-performance programs are under way in the private sector, and although few have been formally analyzed, anecdotal information has not pointed to large negative consequences. Many aspects of pay for performance make intuitive sense. Thus, it seems to me to be reasonable to bolster efforts in this direction, so long as we maintain moderate expectations and monitor the programs carefully, with an eye to making appropriate modifications. Arnold M. Epstein, M.D. Harvard School of Public Health Boston, MA 02115
Andrew F. Macleod, M.D. Royal Shrewsbury Hospital Shrewsbury SY3 8XQ, United Kingdom
1. Rosenthal MB, Frank RG, Li Z, Epstein AM. Early experience
1. Epstein AM. Paying for performance in the United States
and abroad. N Engl J Med 2006;355:406-8. 2. Chaix-Couturier C, Durand-Zaleski I, Jolly D, Durieux P. Ef-
with pay-for-performance: from concept to practice. JAMA 2005; 294:1788-93. 2. Rosenthal MB, Frank RG. What is the empirical basis for paying for quality in health care? Med Care Res Rev 2006;63:135-57.
Causes of Chronic Diarrhea To the Editor: In the table about congenital di- NHE4 (which houses SLC9A4), but even this locus arrheal disorders in the Perspective by Binder (July has not been firmly implicated.3,4 20 issue),1 congenital sodium diarrhea is attributed Daniele Focosi, M.D. to mutations in the gene encoding the sodium– Azienda Ospedaliera Universitaria Santa Chiara hydrogen exchanger (NHE) isoform 3 (SLC9A3, 56100 Pisa, Italy also known as NHE3). Although NHE3-knockout
[email protected] mice (those deficient in Slc9a3) are the only avail- 1. Binder HJ. Causes of chronic diarrhea. N Engl J Med 2006; able animal model of congenital diarrhea,2 genetic 355:236-9. 2. Schultheis PJ, Clarke LL, Meneton P, et al. Renal and intestianalyses of patients with congenital sodium diar- nal absorptive defects in mice lacking the NHE3 Na+/H+ exrhea have excluded all mapped NHE loci except changer. Nat Genet 1998;19:282-5.
n engl j med 355;17
www.nejm.org
october 26, 2006
1833
The
n e w e ng l a n d j o u r na l
of
m e dic i n e
study sodium–hydrogen exchange in affected family members, who had partial villous atrophy. However, it seems likely that these children had congenital sodium diarrhea with different etiologic factors than those with a functional defect in sodium–hydrogen exchange and histopathologically normal jejunum.4 Furthermore, their results do not exclude other explanations for congenital soThe author replies: Focosi is correct that a mo- dium diarrhea that would involve NHE3, such as lecular explanation for congenital sodium diarrhea a defect in a regulatory protein responsible for has not been identified. Several circumstantial ob- NHE3 trafficking to the apical membrane. servations suggest that a mutation in SLC9A3 (en- Henry J. Binder, M.D. coding NHE3) is likely. First, NHE2 and NHE3 are Yale University the only sodium–hydrogen exchangers identified New Haven, CT 06520 in small intestinal apical membranes. Second, mice 1. Yun CH, Gurubhagavatula S, Levine SA, et al. Glucocorticoid lacking NHE3, but not those lacking NHE2, have stimulation of ileal Na+ absorptive cell brush border Na+/H+ exwith an increase in message for NHE-3, diarrhea. Third, glucocorticoids increase both in- change and association an epithelial Na+/H+ exchanger isoform. J Biol Chem 1993;268: testinal sodium absorption secondary to enhanced 206-11. sodium–hydrogen exchange function and NHE3 2. Zachos NC, Tse M, Donowitz M. Molecular physiology of in+ + (but not NHE2) messenger RNA and protein.1 testinal Na /H exchange. Annu Rev Physiol 2005;67:411-43. 3. Muller T, Wijmenga C, Phillips AD, et al. Congenital sodium Thus, NHE3 is generally believed to be responsible diarrhea is an autosomal recessive disorder of sodium/proton for intestinal sodium absorption.2 exchange but unrelated to known candidate genes. GastroenterHowever, Muller et al.3 did not find any defects ology 2000;119:1506-13. 4. Booth IW, Stange G, Murer H, Fenton TR, Milla PJ. Defective in known NHE genes in two Austrian families with jejunal brush-border Na+/H+ exchange: a cause of congenital secongenital sodium diarrhea. These authors did not cretory diarrhoea. Lancet 1985;1:1066-9. 3. Muller T, Wijmenga C, Phillips AD, et al. Congenital sodium
diarrhea is an autosomal recessive disorder of sodium/proton exchange but unrelated to known candidate genes. Gastroenterology 2000;119:1506-13. 4. Online Mendelian Inheritance in Man (OMIM) database. Diarrhea 3, secretory sodium, congenital. (Accessed October 5, 2006, at http://www.ncbi.nlm.nih.gov/entrez/dispomim.cgi? id = 270420.)
Proton-Pump Inhibitors and Hypomagnesemic Hypoparathyroidism To the Editor: We report two cases of hypomagnesemic hypoparathyroidism associated with the use of proton-pump inhibitors, in which patients presented with carpopedal spasm in association with severe hypomagnesemia and hypocalcemia without an appropriate increase in the level of parathyroid hormone. Patient 1 was a 51-year-old premenopausal woman who had been taking omeprazole for more than a year (at a dose of 20 mg twice daily) and who presented with carpopedal and truncal spasm. She began receiving 2.4 g of elemental calcium per day and, later, high-dose magnesium (Fig. 1A). Fourteen months later, omeprazole was discontinued, and ranitidine initiated. The levels of magnesium in the patient’s urine and serum rapidly normalized. However, another proton-pump inhibitor (esomeprazole) was prescribed; within 2 weeks,
1834
n engl j med 355;17
the levels of magnesium in the patient’s urine and serum again fell and remained low until ranitidine was again substituted. All magnesium supplements were then withdrawn, and the patient’s magnesium level remained normal while she was taking ranitidine (at a maximum dose of 900 mg daily). Gastroscopy, colonoscopy, biopsy of the small bowel, and an upper gastrointestinal series with small-bowel follow-through revealed Barrett’s esophagitis but no other abnormality. Patient 2 was an 80-year-old man who had been taking omeprazole (at a dose of 20 mg daily) for several years and who presented with symptoms similar to those of Patient 1 (Fig. 1B). After 10 months of high-dose magnesium supplementation, omeprazole was stopped, and levels of magnesium in both urine and serum rose rapidly to the normal range. The patient began taking ranitidine
www.nejm.org
october 26, 2006
