I feel that this study leaves me with a sense of being on safari to catch a glimpse of the 'Big Five' - as a health professional from South Africa I recognize that it is only but interesting to note the first world trends in pre-eclampsia, but it does not provide any sustainable short or long term treatment modalities, where it is so urgently required.

Hypertension during pregnancy has been attributed to one of the top five causes of maternal deaths in South Africa (Blaauw and Penn-Kekana, 2010). The South African Health Review, 2010, admits that despite an increase in antenatal care coverage and skilled birth attendance, maternal mortality is still on the rise (Blaauw and Penn-Kekana, 2010). While other interventions like low dose asprin and multi-vitamin supplementation do play a role, in this study elective delivery before the due date was one of the primary causes attributed to the trend in the decrease of pre- eclampsia. I am not convinced that elective caesarean sections is the ideal, as this method of delivery may contribute to other causes of morbidity and mortality, not to mention increased costs of surgical delivery, especially in the African setting.

The abovementioned article makes some valuable contributions to the understanding of international trends in pre-eclampsia. However, once again it raises the issue of what we interpret as 'international'. My concern about these trends being acknowledged as international is that the study takes place across borders but it does not necessarily apply to the global context.

The study lacks inclusion of developing nations, and thus does not take in consideration the countries urgently seeking solutions, in their urgency to reach Millenium Development Goals and reduce maternal mortality (MDGs 4 and 5). I would urge that cognizance be taken of the location of the countries used in the study.

References:

Blaauw, D & Penn-Kekana, M. 2010. In: Fonn, S., Padarath, A. South African Health Review. Durban: Health Systems Trust. Available online: http://www.hst.org.za/publications/876

Conflict of Interest:

None declared

Dear Editor,

This paper from Villalbi and colleagues use a before-after design without control group to analyse deaths due to Acute Myocardial Infarction (AMI) in Spain from 2004-2007 and concludes "the extension of smoke-free regulations in Spain [came into force in January 2006] was associated with a reduction in AMI mortality, especially among the elderly". While we are clearly in favour of this law, their immediate impact on the population coronary health is doubtful, and this paper does not help to raise doubts.

First, coronary deaths have been declining in Spain, as in other developed countries, during the last years. Raw rates for men have changed from 109/100,000 in 2001 to 89 in 2008 (77 to 69 for women). The most important relative change for men was between 2004 and 2003 (before the law implementation). Authors argument that the 2003 heat wave distort the mortality patterns, but previous years maintain a similar pattern than 2003(1). AMI mortality rates in Spain (age-adjusted to European standard population) for 100,000 men were: 2001: 89.5; 2002: 87.5; 2003: 87.9; 2004: 81.2; 2005: 80.0; 2006: 74,6; 2007: 72.7; 2008: 67.8; similar figures for women were: 39.0; 38.8; 38.5; 36.2; 35.3; 31.9; 31.0; 29.2).

Second, the law did not have an additional effect in the prevalence of cigarettes consumption. According to the Spanish National Survey on Drug Use (EDADES, biannual periodicity) the prevalence began to decline years before the smoke-free regulation(2). For adult men: 2001: 51.5%, 2003: 53.0 2005: 47.2%, 2007: 46.0% (for adult women in the same years: 40.5%, 42.6, 37.5%, 37.6%).

Third, while the main effect of the Spanish 2006 law was the smoking ban in workplaces, the paper reports a higher mortality reduction in elderly (retired people who no longer work).

Fourth, in the paper the decline of mortality rates were steeply falling in the first 12 months after the implementation of the law. In contrast, other study that has evaluated the impact of anti-smoke laws in coronary mortality rates reported a modest non significant effect in the short term(3) (1.6% CI95%: -4.0%;7.0%), but larger effects after the first 12 months (18.6% CI95%=13.6%;23.3%).

In the 70s and 80s in Spain the number of storks decreased dramatically, and contemporary, the birth rate fell among the lowest in the world. Policies on these birds were able to recover their number in the 90's. In this period the Spanish birth rate increased again. The association of uncontrolled temporal trends -specially with a very short trend as in this study- should not be interpreted as causal (unless we are willing to accept that babies come from Paris).

Therefore, in the absence of other evidences we should be cautious about establishing causal relationships between the decrease in coronary deaths and one specific policy. As much as we favour that policy, scientific causality has other rules.

References:

1.Instituto de Salud Carlos III. Ministerio de Ciencia e Innovacion. Mortalidad Espana y comunidades autonomas. Available: http://www.isciii.es/htdocs/centros/epidemiologia/anexos/ww9201_cau.htm

2. Ministerio de Sanidad, Politica Social e Igualdad. Encuesta Domiciliaria Sobre Alcohol y Drogas en Espana (EDADES). Available: http://www.pnsd.msc.es/Categoria2/observa/estudios/home.htm

3.Dove MS, Dockery DW, Mittleman MA, Schwartz J, Sullivan EM, Keithly L, et al. The impact of Massachusetts' smoke-free workplace laws on acute myocardial infarction deaths. Am J Public Health. 2010;100:2206-12.

Conflict of Interest:

None declared

Whilst acknowledging the previous comment's valid concerns over what constitutes an international survey, I wanted to correct the assumption that all elective deliveries must be Caesarean sections. Most elective deliveries prior to term are likely to be inductions of labour as per the HYPITAT study. This may still not be relevant for the South African setting but it is very different concept than suggested in the response.

Reference

Koopmans CM, Bijlenga D, Groen H, Vijgen SM, Aarnoudse JG, Bekedam DJ, van den Berg PP, de Boer K, Burggraaff JM, Bloemenkamp KW, Drogtrop AP, Franx A, de Groot CJ, Huisjes AJ, Kwee A, van Loon AJ, Lub A, Papatsonis DN, van der Post JA, Roumen FJ, Scheepers HC, Willekes C, Mol BW, van Pampus MG; HYPITAT study group. Induction of labour versus expectant monitoring for gestational hypertension or mild pre-eclampsia after 36 weeks' gestation (HYPITAT): a multicentre, open-label randomised controlled trial. Lancet. 2009 Sep 19;374(9694):979-88. Epub 2009 Aug 3.

Conflict of Interest:

None declared

Dear Editor,

We read with interest the article 'Evaluation of the incremental cost to the National Health Service of prescribing analogue insulin'(1). The stated aim of this study was to characterize the pattern of insulin prescriptions dispensed between 2000 and 2009 in UK and to evaluate the financial cost to the National Health Service (NHS) of using analogue insulin instead of its equivalent human insulin preparations. Whilst we agree with the summary point including the fact that the use of insulin analogues has had a substantial financial impact to the NHS, we feel that it is important to urge for a more cautious analysis of the estimates provided, to take into account the bigger picture of diabetes treatment prescribing costs in order to derive a more holistic conclusion to this analysis.

The long acting insulin analogue has a smooth peak-less action profile and clinical trials comparing insulin analogue with NPH insulin have shown similar effects on HbA1c but significant reductions in nocturnal hypoglycaemia when using insulin analogue(2,3). The Quality, Innovation, Productivity and Prevention (QIPP) agenda aims to make efficiency savings in the NHS and one of its key performance indicators is to report the ratio of insulin analogue use in relation to the amount of NPH usage for the treatment of patients with type 2 diabetes. This is based on the notion that NPH insulin is cheaper and hence represents a cost saving to the health service. Whilst this is true for the direct cost of insulin, the broader costs of treatment and utilization of health care resources need to also be considered. When taking all these factors into account, two previous studies have shown that the cost differential between NPH insulin and glargine is less clear cut(4,5). In one of the study, the largest differences in resource use between the glargine group and the NPH group, was in the bolus of insulin and amount of blood glucose self-testing strips used(4). In the glargine group, 54% of respondents used bolus insulin compared with 80% in the NPH group. The glargine group used, on average 1.40 test strips and 0.66 lancets per patient per day, while the NPH group used 2.44 and 0.74 test strips and lancets respectively per patients per day. This is particularly relevant given the rising cost for prescribing blood glucose monitoring strips in England and Wales. In the East Midlands SHA alone, the total cost of prescribing diabetes treatment in 2010 was 26.2 million pounds, 12.2 million pounds of which was spent on blood glucose monitoring strips and devices.

Dr Holden and colleagues also argued that if all patients using insulin analogues could have received human insulin (a 100% reduction in the use of insulin analogue), the overall incremental cost of analogue insulin was 625 million pounds. Again, we would caution against these estimates because the conversion failed to take into account all factors, was based on many unidentified assumptions over a 10 year period and may occur at the expense of increased hypoglycaemia complications(2-5) and subsequent cardiovascular events6. Among other assumptions also not considered were, the change of hazard over time, total costs of all interventions including use of bolus insulin, glucose monitoring devices and intensity of health-care resource utilizations. Notwithstanding the ambiguities reported in this study, we agree that the use of a pharmacotherapy based purely on marketing reports rather than a thorough review of their evidence base is not cost-effective and should be discouraged. As such, we accept that the current percentage of analogue versus human insulin for total insulin cost is indeed high. In the East Midlands, this figure for year 2010 was 78.6%, which was lower than that observed in this present study (84%) and the 2010 national average of 85.3%. Lowering this figure to 20% would mean a saving of around 2.5 million pounds per year in the East Midlands. This 'aspirational' target for use of human insulin analogue is in-line with the Better Care Better Values prescribing indicators available. However, we believe a more realistic figure is 40%. This figure whilst arbitrary is more safe in terms of avoiding the potential rise in hypoglycaemia(7) as well as achievable since the policy from the Medicine Management groups advised that patients are not 'switched' from present regime unless there is a clinical concern and that new patients are initiated on analogue insulin only as per NICE guidelines.

In conclusion, we encourage primary care physicians and health care managers in the UK to critically evaluate cost-benefit analysis of human versus analogue insulin to take into account other confounding factors, which are relevant to clinical practice at patient and population level. In particular, the cost-benefit of insulin analogue should be viewed in the context of escalating costs of prescribing blood monitoring strips and devices as this plays a major impact on the total cost of prescribing diabetes treatment.

References

1 Holden SE, Poole CD, Morgan CL, Currie CJ. Evaluation of the incremental cost to the National Health Service of prescribing analogue insulin BMJ Open 2011; 1: e000 258 doi:10.1136/bmjopen-2011-000258 2 Home, P.D., Fritsche, A., Schinzel, S., Massi-Benedetti, M. 2010. Meta -analysis of individual patient data to assess the risk of hypoglycaemia in people with type 2 diabetes using NPH insulin or insulin glargine. Diabetes, Obesity and Metabolism, 12: 772-79 3 Hermansen K, Davies M, Derezinski T, Martinez Ravn G, Clauson P, Home P. 2006 A 26-week, randomized, parallel, treat-to-target trial comparing insulin detemir with NPH insulin as add-on therapy to oral glucose- lowering drugs in insulin-naive people with type 2 diabetes. Diabetes Care. 29:1269-74 4 Schoffski, O., Breitscheidel, L., Benter, U., Dippel, F.W., Muller, M., Volk, M., Pfohl, M. 2008. Resource utilization and costs in patients with type 2 diabetes mellitus treated with insulin glargine or conventional basal insulin under real-world conditions in Germany: LIVE- SPP study. Journal of Medical Economics, 11: 695-712 5 Rhoads GG, Dain MP, Zhang Q, Kennedy L. 2011 Two-year glycaemic control and healthcare expenditures following initiation of insulin glargine versus neutral protamine Hagedorn insulin in type 2 diabetes Diabetes Obes Metab. 13:711-7. doi: 10.1111/j.1463-1326.2011.01394.x 6 Dluhy RG, McMahon GT 2008 Intensive glycemic control in the ACCORD and ADVANCE trials. N Engl J Med. 358:2630-3 7 UK Hypoglycaemia study group. 2007 Risk of hypoglycaemia in types 1 and 2 diabetes: effects of treatment modalities and their duration Diabetologia. 50:1140-7.

Conflict of Interest:

II and/or MD have received research grants from Eli Lilly, Sanofi Aventis and Novo Nordisk