- Email: [email protected]
Total cost comparison of standard antenatal care with a weight gain restriction programme for obese pregnant women
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
available at www.sciencedirect.com
Public Health journal homepage: www.elsevier.com/puhe
Original Research
Total cost comparison of standard antenatal care with a weight gain restriction programme for obese pregnant women N. de Keyser a, A. Josefsson a, W.G. Monfils a, I.-M. Claesson a, P. Carlsson b, A. Sydsjo¨ a, G. Sydsjo¨ a,* a
Division of Obstetrics and Gynaecology, Department of Clinical and Experimental Medicine, Faculty of Health Sciences, Linko¨ping University, SE-581 85 Linko¨ping, Sweden b Centre for Medical Technology Assessment, Department of Medical and Health Sciences, Faculty of Health Sciences, Linko¨ping University, Linko¨ping, Sweden
article info
summary
Article history:
Objective: To perform a cost comparison of a weight gain restriction programme for obese
Received 3 May 2010
pregnant women with standard antenatal care, and to identify if there were differences in
Received in revised form
healthcare costs within the intervention group related to degree of gestational weight gain
1 December 2010
or degree of obesity at programme entry.
Accepted 9 February 2011
Study design: A comparison of mean healthcare costs for participants of an intervention study at antenatal care clinics with controls in south-east Sweden. Methods: In total, 155 women in an intervention group attempted to restrict their gesta-
Keywords:
tional weight gain to <7 kg. The control group comprised 193 women. Mean costs during
Pregnancy
pregnancy, delivery and the neonatal period were compared with the costs of standard
Weight restriction
care. Costs were converted from Swedish Kronor to Euros (V).
Intervention
Results: Healthcare costs during pregnancy were lower in the intervention group. There was
Cost-effectiveness
no significant difference in total healthcare costs (i.e. sum of costs during pregnancy, delivery and the neonatal period) between the intervention group and the control group. Within the intervention group, the subgroup that gained 4.5e9.5 kg had the lowest costs. The total cost, including intervention costs, was V1283 more per woman/infant in the intervention group compared with the control group (P ¼ 0.025). The degree of obesity at programme entry had no bearing on the outcome. Conclusions: The weight gain restriction programme for obese pregnant women was effective in restricting gestational weight gain to <7 kg, but had a higher total cost compared with standard antenatal care. ª 2011 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: þ46 13 22 20 00; fax: þ46 13 14 81 56. E-mail address: [email protected] (G. Sydsjo¨). 0033-3506/$ e see front matter ª 2011 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.puhe.2011.02.004
312
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
Introduction The obesity epidemic is not confined to a single subset of the world’s population, but is spread across all races, age groups, genders and socio-economic classes, including pregnant women.1e3 From 1978 to 2001, the proportion of overweight women of childbearing age tripled in Sweden, and obesity in that population increased five-fold.4 The increase in body mass index (BMI) brings with it an increased risk for complications during pregnancy, delivery and the postpartum period.5e9 The infant of an obese woman is at increased risk for antepartum stillbirth, neonatal death, birth defects and neonatal complications.10e14 Recent studies have shown that some of these risks can be averted if a woman’s pregnancy weight gain is restricted.15e17 Few studies have provided quantitative estimations of the related increase in the use of healthcare services for obese pregnant women. General healthcare costs for obese pregnant women were studied by Chu et al.,18 who found that obesity during pregnancy was associated with significantly more examinations, more obstetric ultrasonographic examinations and increased use of medication. Chu et al. also found that obese women had more prenatal visits with physicians, a longer hospital stay after delivery, and more telephone calls to obstetrics and gynaecology departments. Most of the increase in length of stay associated with higher BMI was related to increased rates of caesarean delivery and obesity-related high-risk conditions. Few interventions exist to control excessive weight gain during pregnancy, and those that do exist have shown poor results.19e21 Policy makers and healthcare planners are aware that even a small increase in the cost of health care associated with obesity during pregnancy for the woman, and later her child, will have substantial economic implications for the future. This makes intervention programmes and assessment of efficacy and economic viability of interventions essential. In order to make an informed decision about which interventions to implement, healthcare providers and policy makers must have information about which programmes maximize health, given everpresent budget constraints. The purpose of this study was to compare the total costs during pregnancy, delivery and the neonatal period between women in the index group of a weight gain restriction programme for obese pregnant women conducted at the University Hospital of Linko¨ping, Sweden and a cohort of women given standard antenatal care.
were invited to a 30-min session every week. The midwife worked according to the following schedule: (1) a thorough assessment of the pregnant woman’s knowledge of obesity as a risk factor during pregnancy, childbirth and for the child; (2) if the woman lacked sufficient knowledge, she was offered the information and given accurate facts; (3) the woman was informed about the potential consequences of behaviour and cognition associated with eating and food intake, and written information was supplied if needed; (4) all successive sessions included weight control and a supportive talk regarding how to deal with food intake, motivational change of behaviour, how to deal with partner and family, supporting exercises and behavioural pitfalls; and (5) all women were invited to participate in a water-gymnastics programme designed for obese women once or twice a week.22 The target weight gain during pregnancy was <7 kg. The women in the control group were offered standard antenatal care. The intervention group consisted of 155 women with a BMI 30 chosen consecutively from all pregnant women attending regular antenatal care at the University Hospital of Linko¨ping, Sweden between November 2003 and December 2005. Exclusion criteria were inability to speak and understand Swedish, diagnosed with diabetes or disease related to thyroid function, and current treatment with neuroleptic drugs. Forty-five women met these criteria and were excluded from the study. In addition, 13 women who had an early miscarriage or a legal abortion, and 29 women who moved out of the catchment area in early pregnancy were also excluded. Two hundred and thirty obese women were thus eligible and invited to participate. Out of these, 70 women refrained from participation and five women dropped out during the intervention. Thus, a total of 155 obese women (67.4%) agreed to participate and completed the intervention. To form the control group, women with the same inclusion and exclusion criteria were consecutively invited from the ACCs in two nearby cities over the same time period. Three hundred and eighty-five women were invited to participate, 177 women refrained from participation and 15 women dropped out during the study period. Thus, 193 (50.1%) women agreed to participate and completed the study as controls. The obese women in the control group attended the routine antenatal care programme. All three cities share similar demographics such as age, parity, marital status and smoking,22 and lie within south-east Sweden (Fig. 1).
Costs Methods The Swedish antenatal healthcare system and delivery care reaches almost 100% of pregnant women. Antenatal and delivery care is free of charge. At antenatal care clinics (ACCs), healthy pregnant women are recommended to attend the regular antenatal programme with seven to nine visits to a midwife, and, if needed, extra appointments with an obstetrician and/or with a midwife. In the intervention study, pregnancy weight gain was controlled in the intervention group through weekly motivational meetings with a cognitive and behavioural change approach conducted by a specially trained midwife. All women who agreed to take part in the intervention
All costs have been converted from Swedish Kronor to Euros (V) using a 10:1 ratio.
Cost of the intervention programme A motivational meeting with a midwife cost V82 (national average cost of a 30-min meeting with a midwife). Each meeting was documented in the patient’s journal. The rental cost of the water-gymnastics pool was approximately V60/h. Each water-gymnastics session was supervised by a midwife, with a concurrent cost of V28/session. Costs for loss of productivity due to participation in the programme were calculated using the mean monthly income
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
313
of the cost of an appointment (V99). An appointment with a midwife cost V82. In the event of inpatient care during pregnancy, the appropriate DRG costs were assigned.
Costs during delivery The guidelines for the choice of DRG during delivery were as follows: vaginal delivery, uncomplicated; vaginal delivery, complicated (if vaginal delivery required the use of instrumental aid, or resulted in a Grade IIIeIV vaginal rupture, placental retention or bleeding >1000 ml); caesarean section, uncomplicated; and caesarean section, complicated (acute caesarean section).
Costs of neonatal care
Fig. 1 e Subjects in the study.
in 2005 for all Swedish women (V2210). Fifty percent was added for the employer’s contribution to the national social insurance system, giving a total of V3315.23 Given that approximately two-thirds of women in the intervention group were employed at the beginning of the programme, the total monthly income was multiplied by 2/3 to reach a sum of V2210. There are approximately 22 working days per month. The hourly cost of productivity loss was therefore calculated as follows: V2210/22 giving a daily income of approximately V100, then divided by 8 working hours (100/8) ¼ V12.56. Each motivational meeting or water-gymnastics session was assumed to take 1.5 h, including travel time, and thus contributed V19 to the cost of the programme. The number of motivational meetings attended was obtained from standardized patient journals. The number of water-gymnastics sessions attended was obtained from a sign-in sheet.
Healthcare costs Hospital costs during pregnancy, delivery and the neonatal period were calculated using national averages as presented by NordDRG. Diagnosis-related group (DRG) uses a system of averaging costs for diagnosis, procedures performed, gender, age, length of hospitalization and status at discharge to classify patients into categories with similar use of resources.24 The following inclusions and assumptions were made when assigning each diagnosis into each DRG.
Costs during pregnancy Each appointment with a physician, whether as an outpatient at the hospital or at a public ACC, cost V297. The cost for a midwife consulting a physician was estimated to be one-third
The assigned DRG for each case depended on the birth weight and number of diagnoses/complications that were noted in the patient’s journal. In the case of a diagnosis of ‘large for gestational age’, ‘postmature’ or ‘hyperbilirubinaemia’, the diagnosis was not considered to be a complication if the infant was not treated in the neonatal ward. An uncomplicated pregnancy, vaginal delivery and a healthy child costs V4400. A short hospitalization during pregnancy, such as might result in the case of an emergency caesarean section and an infant requiring limited hospital care, costs approximately V12,000.
Analyses Mean total healthcare costs and costs during pregnancy, delivery and the neonatal period for the intervention group were compared with costs for the control group. The weight gain subgroups of women who gained <7 kg (target maximum weight gain of the programme) and 7 kg were compared. Since weight gain recommendations are a topic of current debate, the groups were further divided as follows: (1) those who gained <4.5 kg during pregnancy; (2) those whose gained 4.5e9.5 kg during pregnancy; and (3) those who gained >9.5 kg during pregnancy. Cost variables within each of these subgroups were compared with the control group. Weight gain was calculated using the last registered weight during a motivational meeting minus weight at programme entry. Each subgroup was further compared with their respective weight gain interval in the control group to determine if any potential differences were possibly attributable to factors other than weight gain. Additionally, the outcomes based on BMI at programme entry (stratified into 30e34.9 kg/m2, 35e39.9 kg/m2 and 40 kg/m2) were compared to elucidate any possible effect of the degree of obesity on the results of the programme.
Statistical methods Mean total healthcare cost and costs during pregnancy, delivery and the neonatal period were compared. Additionally, intervention costs were added to the total cost and compared with the total average cost for a pregnant woman in the control group. The costs were compared using Student’s t-test.23 All
314
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
tests were two-tailed and 95% confidence intervals were calculated. All calculations were performed using Statistical Package for the Social Sciences Version 15.0 (SPSS Inc., Chicago, IL, USA).
cost of health care and intervention. No difference in total healthcare cost was found between the groups. However, the intervention group had significantly lower costs (approximately V348/woman) during pregnancy than the control group.
Results Total cost of health care and intervention Brief summary of the results of the intervention programme (for more details on obstetric and neonatal outcomes, see Claesson et al.22) The mean weight gain for the intervention group was 2.6 kg less than that of the control group (8.7 kg vs. 11.3 kg; P < 0.001). Significantly more women in the intervention group gained <7 kg compared with the control group (P ¼ 0.003), and there were more nulliparous women in the intervention group than the control group (P ¼ .018). The women in the intervention group had a significantly lower BMI at the postnatal check-up compared with the control group (P < 0.001). Neonatal outcomes such as birth weight, gestational age and mode of delivery/complications did not differ between the groups. The same held true when the analyses were restricted to women who gained <7 kg during pregnancy.22
As seen in Table 1, the intervention cost per woman was V1616. During the 27 months of the intervention programme, 155 women took part, resulting in a total cost of V250,511. The mean per woman/infant difference in total cost, including intervention costs, was V1283 more for the intervention group than the control group (P < 0.025).
Intervention subgroups compared with control group For subgroup analyses within the intervention group, the women who gained <7 kg (target maximum weight gain for the intervention group) were compared with the controls, and the women in the intervention group who gained >7 kg were compared with the controls. There was no difference in total cost (Table 2).
Intervention subgroups <4.5, 4.5e9.5 and >9.5 kg compared with control group
Total healthcare costs Two extremely preterm infants (one from the intervention group, one from the control group) were hospitalized for approximately 90 days. Both cases were excluded from the calculation of neonatal costs, total healthcare cost, and total
The intervention subgroup who gained <4.5 kg showed increased costs for neonatal care and total cost (P ¼ 0.16 and 0.20, respectively) compared with the control group. However,
Table 1 e Average costs (in Euros) per woman or woman/child pair: intervention and control groups.
Healthcare costs Pregnancy Physician appointment Physician consultation Midwife appointment Hospitalization (DRG) Total Delivery Hospitalization (DRG)
Intervention (n ¼ 155-1a)
Control (n ¼ 193-1a)
Mean (SD)
Mean (SD)
778 (592) 94 (131) 658 (125) 169 (619) 1699 (1040)
P-value
95% CI
928 (866) 67 (104) 713 (173) 263 (616) 2044 (1199)
0.057 0.039 0.001 0.157 0.004
304 to 4 1e53 86 to 23 226 to 37 582 to 109
0.390
122 to 313
2397 (1012)
2302 (1039)
Neonatal care (One child in each group is missing from the analysis) Hospitalization (DRG)
(n ¼ 154)
(n ¼ 192)
3187 (4550)
3559 (4732)
0.885
1062 to 917
Total healthcare costs
7258 (5117)
7595 (5390)
0.556
785
Intervention costs Motivational meetings Loss of productivity due to meetings Water-gymnastics Loss of productivity due to water-gymnastics Total intervention costs
1156 248 83 131 1616 7595 (5390)
0.025
164e2403
Total cost of health care and intervention (per woman/child pair)
8878 (5095)
SD, standard deviation; CI, confidence interval; and DRG, diagnosis-related group. a An extremely preterm child has been excluded from both Intervention and Control group.
315
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
Table 2 e Average costs (in Euros) per woman or woman/child paira: intervention subgroups <7 kg and ‡7 kg and control group. Intervention subgroup
Control (n ¼ 193-1)b
Mean (SD)
Mean (SD)
Subgroup healthcare cost Subgroup: weight gain <7 kg Pregnancy Delivery Neonatal care Total healthcare cost
n ¼ 57-1b 1663 (528) 2216 (932) 3414 (5140) 7987 (8047)
Subgroup: weight gain 7 kg Pregnancy Delivery Neonatal care Total healthcare cost
n ¼ 98 1719 (1037) 2503 (1045) 3057 (4198) 7278 (4803)
P-value
95% CI
2021 (880) 2306 (1230) 2927 (4103) 7255 (4804)
0.05 0.56 0.47 0.52
715 to 1 392 to 214 841 to 1814 1008 to 2471
2021 (880) 2306 (1030) 2927 (3803) 7255 (4804)
0.04 0.14 0.81 0.97
592 to 13 62 to 454 902 to 980 1174 to 1220
SD, standard deviation; and CI, confidence interval. a For the women with singleton pregnancies whose weight was assessed at the start of the intervention and 1e2 weeks before delivery. b For the exclusion of the extremely preterm child.
after excluding the extremely preterm cases, the results showed no such trend (P ¼ 0.88 and 0.95, respectively). There was no difference in costs for the subgroup that gained >9.5 kg compared with the control group. However, in the subgroup that gained 4.5e9.5 kg, the mean cost during pregnancy was V574 less than that for the control group (P < 0.001) (Table 3).
Intervention subgroups compared with control subgroups Total healthcare costs for all subgroups in the intervention group and the control group (e.g. <7 kg intervention compared with <7 kg control) showed no significant differences. However, differences were found between the DRG costs during pregnancy. A statistically significant reduction in costs was found for the intervention subgroups <7 kg and
4.5e9.5 kg. A similar trend was found within the other subgroups (data not shown).
Stratified BMI at programme entry Women in the intervention group with an initial BMI of 30e34.9 kg/m2 (n ¼ 100) had a mean total healthcare cost, excluding intervention costs, of V7948, while women in the control group with the same initial BMI (n ¼ 112) had a mean cost of V7023; this difference was not significant (P ¼ 0.226). Women in the intervention group with an initial BMI of 35e39.9 kg/m2 (n ¼ 36) had a mean total healthcare cost of V6136, while women in the control group with the same initial BMI (n ¼ 39) had a cost of V7007; again, this difference was not significant (P ¼ 0.249). Finally, women in the intervention
Table 3 e Average costs (in Euros) per woman or woman/child paira: intervention subgroups and control groups. Intervention subgroup
Control (n ¼ 193-1)b
Mean (SD)
Mean (SD)
Subgroup healthcare cost Subgroup: weight gain <4.5 kg Pregnancy Delivery Neonatal care Total healthcare cost
n ¼ 33-1b 1997 (1247) 2375 (990) 4300 (8429) 8671 (9227)
2021 2306 2927 7255
Subgroup: weight gain 4.5e9.5 kg Pregnancy Delivery Neonatal care Total healthcare cost
n ¼ 54 1448 (845) 2343 (980) 3385 (5145) 7175 (5666)
Subgroup: weight gain >9.5 kg Pregnancy Delivery Neonatal care Total healthcare cost
n ¼ 68 1766 (1049) 2453 (1059) 3082 (3971) 7301 (4648)
(880) (1029) (4103) (4804)
2021 (880) 2306 (1029) 2927 (4103)
2021 2306 2927 7255
(880) (1029) (4103) (4804)
P-value
0.92 0.73 0.16 0.20
95% CI
491 319 532 751
to to to to
443 458 3277 3584
<0.001 0.82 0.50 0.92
920 to e227 272 to 345 871 to 1785 1604 to 1444
0.11 0.33 0.79 0.95
569 to 58 147 to 442 999 to 1309 1304 to 1397
SD, standard deviation; and CI, confidence interval. a For the women with singleton pregnancies whose weight was assessed at the start of the intervention and 1e2 weeks before delivery. b For the exclusion of the extremely preterm child.
316
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
group with the highest BMI (n ¼ 19) had a mean total healthcare cost of V8043, while women in the control group with the same initial BMI (n ¼ 21) had a mean cost of V8952; once again, this difference was not significant (P ¼ 0.723). All costs contributing to the total cost (i.e. pregnancy, delivery, neonatal care) were analysed in the same manner. No significant differences were found.
Frequency of exercise Attendance of water-gymnastics sessions during pregnancy ranged from zero to 28, with a mean of seven and a mode of five. A small negative correlation (0.15, P < 0.07) was found between frequency of attendance of water-gymnastics sessions and gestational weight gain; however, no correlation was found between exercise and costs.
Lost weight during pregnancy Eleven women lost weight during their pregnancy. No differences were found in costs associated with pregnancy, delivery, neonatal care or total cost compared with the control group.
Discussion This study is, to the authors’ knowledge, the first analysis of the total cost of an intervention programme to limit weight gain in obese women during pregnancy. The cost comparison analyses and interpretation of results in the present study were conducted by individuals who were not connected with the intervention programme, thereby preventing conflicts of interest. The results show that the intervention programme had no economic benefit as a whole, although the intervention group had lower costs during pregnancy (approximately V348). Recommendations for weight gain by obese women during pregnancy vary, and are a topic of current debate. The Institute of Medicine25 has recently changed its recommendation from a minimum weight gain of 6.8 kg to weight gain between 5 and 9 kg, while a study published by Cedergren suggests that weight gain of <6 kg is appropriate for obese women.26 The gestational weight gain limits for BMI categories in Cedergren’s large population-based cohort study from Swedish Medical Registers showed that a decreased risk of adverse obstetric and neonatal outcomes was associated with lower gestational weight gain than was recommended previously, especially among obese women. These findings are supported by the results of a recent randomized clinical trial concerning perinatal outcomes in nutritionally monitored obese pregnant women.16 Regarding the debate concerning recommendations for weight gain by obese women during pregnancy, further studies are warranted to ensure the safety and efficacy of future intervention programmes. The benefits of reduced costs during pregnancy in this particular intervention study were outweighed by the cost of the intervention (V1616/participant). Data analyses found that the pregnancy weight gain that provided the greatest economic saving was between 4.5 and 9.5 kg. Even given the best case scenario under which all women in the intervention group kept their weight gain to 4.5e9.5 kg, a net loss of
approximately V1002/participant would be expected. The cost of this particular intervention programme therefore precludes any potential economic benefit. Considering that the motivational meetings and loss of productivity due to the meetings accounted for 87% (V1404) of the intervention costs, a sizeable cost reduction here may make cost-effectiveness achievable. Say, for instance, that the motivational meetings could be conducted effectively in groups of six women, every other week as opposed to individual, weekly meetings. The total intervention cost could thus be reduced to approximately V330/participant. For women who gain 4.5e9.5 kg, a net saving of approximately V249 could be made. Such an adjustment may be warranted for future intervention studies, given that approximately half of the women in the intervention group stated that there were too many motivational meetings, and 72% reported that the social character of the water-gymnastics (a group session) was one of the most positive aspects of the programme.27 This intervention study had both strengths and limitations. The intervention study was not randomized, which can be seen as an important limitation. Randomization is a powerful tool and a good general rule is to randomize whenever possible.28,29 In all scientific research, it is important to control external factors and to make an effort to minimize situational contaminants. The environment has been found to exert a powerful influence on people’s emotions and behaviours,29 and requires careful consideration if intervention and control groups will get treatment and care at the same setting. A continuing intervention programme with a new routine might influence the staff to change their behaviour and regimens, even for those who are supposed to be controls and receive standard care. The ACC in Linko¨ping is structured as one large unit. As the intervention programme designers were determined to ensure that no contamination occurred between the intervention and control groups, ACCs in two nearby cities were used as controls. The antenatal programme in Sweden is standardized and almost identical in terms of pregnancy management, which ensures similar care at different ACCs. Therefore, the design was consecutive inclusion into the intervention group, with two other cities and their ACCs used for controls. Additionally, background characteristics were analysed in the intervention study to correct for possible confounding factors. However, there may still have been small differences in management and routines that may have affected the results. The present results are based on univariate statistical analyses. These results may, however, be interpreted to be a reasonable representation of the wider population as the degree of weight gain in the intervention and control groups was shown to be independent of all demographic characteristics when analysed using ANCOVA.22 Subgroup analyses were performed shortly before all women had given birth, thus the numbers of individuals in the intervention and control groups are lower than those used for analysis of the programme as a whole. This small reduction is not expected to have any impact on the results. The standard deviation of the cost factors was extremely wide; as such, the statistical power to find any difference in costs, if they exist, is low. However, since the mean costs during pregnancy, delivery and the neonatal period were very
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
close in the intervention and control groups, it can be assumed that no difference exists. Approximately 110,000 children are born in Sweden each year,30 and about 11,000 of these are born to obese women.4 If one-third of these obese women could keep their weight gain to 4.5e9.5 kg during pregnancy, a total direct saving to society of around V1,000,000 could be expected. With other health benefits that follow from a healthy change in lifestyle, the benefits would certainly exceed this modest short-term saving. Since better health for woman and child, and budget streamlining are natural goals for policy makers and healthcare providers, further efforts to create a viable weight gain control programme during pregnancy are warranted.
10. 11. 12.
13.
14.
15.
Ethical approval Human Research Ethics Committee, Faculty of Health Sciences, Linko¨ping University, Sweden.
16.
Funding
17.
Grants from the Health Research Council in south-east Sweden. 18.
Competing interests None declared.
19.
references
20.
1. The Swedish Council on Technology Assessment in Health Care. Fetma e problem och a˚tga¨rder, En systematisk litteraturo¨versikt. Stockholm: SBU; 2002. 2. Yeh J, Shelton JA. Increasing pre-pregnancy body mass index: analysis of trends and contributing variables. Am J Obstet Gynecol 2005;193:1994e8. 3. La Coursiere DY, Bloebaum L, Duncan JD, Varner MW. Population-based trends and correlates of maternal overweight and obesity, Utah 1991e2001. Am J Obstet Gynecol 2005;192:832e9. 4. Brynhildsen J, Sydsjo¨ A, Norinder E, Ekholm Selling K, Sydsjo¨ G. Trends in body mass index during early pregnancy in Swedish women 1978e2001. Public Health 2006;120:393e9. 5. Dempsey JC, Ashiny Z, Qiu C-F, Miller RS, Sorensen TK, Williams MA. Maternal pre-pregnancy overweight status and obesity as risk factors for caesarean delivery. J Matern Fetal Neonatal Med 2005;17:179e85. 6. Sebire NJ, Jolly M, Harris JP, Wadsworth J, Joffe M, Beard RW, et al. Maternal obesity and pregnancy outcome: a study of 287,213 pregnancies in London. Int J Obes Relat Metab Disord 2001;25:1175e82. 7. Kumari AS. Pregnancy outcome in women with morbid obesity. Int J Gynaecol Obstet 2001;73:101e7. 8. Baeten JM, Bukasi EA, Lambe M. Pregnancy complications and outcomes among overweight and obese nulliparous women. Am J Public Health 2001;91:436e40. 9. Vahratian A, Zhang J, Troendle JF, Savitz DA, Siega-Riz AM. Maternal prepregnancy overweight and obesity and the
21.
22.
23. 24.
25.
26. 27.
28. 29. 30.
317
pattern of labor progression in term nulliparous women. Obstet Gynecol 2004;104:943e51. Cedergren MI. Maternal morbid obesity and the risk of adverse pregnancy outcome. Obstet Gynecol 2004;103:219e24. Cedergren MI, Ka¨lle´n BAJ. Maternal obesity and infant heart defects. Obes Res 2003;11:1065e71. Kristensen J, Vestergaard M, Wisborg K, Kesmodel U, Secher NJ. Pre-pregnancy weight and the risk of stillbirth and neonatal death. BJOG 2005;112:403e8. Stephansson O, Dickman PW, Johansson A, Cnattingius S. Maternal weight, pregnancy weight gain, and the risk of antepartum stillbirth. Am J Obstet Gynecol 2001;184:463e9. Cnattingius S, Bergstro¨m R, Lipworth L, Kramer MS. Prepregnancy weight and the risk of adverse pregnancy outcomes. N Engl J Med 1998;338:191e2. Cedergren M. Effects of gestational weight gain and body mass index on obstetric outcome in Sweden. Int J Gynaecol Obstet 2006;93:269e74. Thorton YS, Smarkola C, Kopacz SM, Ishoof SB. Perinatal outcomes in nutritionally monitored obese pregnant women: a randomised clinical trial. J Natl Med Assoc 2009; 101:569e77. Crane JMG, White J, Murphy P, Burrage L, Hutchens D. The effect of gestational weight gain by body mass index on maternal and neonatal outcomes. J Obstet Gynaecol Can 2009; 31:28e35. Chu SY, Bachman DJ, Callaghan WM, Whitlock EP, Dietz PM, Berg CJ, et al. Association between obesity during pregnancy and increased use of health care. N Engl J Med 2008;358: 1444e53. Polley BA, Wing RR, Sims CJ. Randomized controlled trial to prevent excessive weight gain in pregnant women. Int J Obes Relat Metab Disord 2002;26:1494e502. Olson CM, Strawderman MS, Reed RG. Efficacy of an intervention to prevent excessive gestational weight gain. Am J Obstet Gynecol 2004;191:530e6. Gray-Donald K, Robinson E, Collier A, David K, Renaud L, Rodrigues S. Intervening to reduce weight gain in pregnancy and gestational diabetes mellitus in Cree communities: an evaluation. CMAJ 2000;163:1247e51. Claesson I-M, Sydsjo¨ G, Brynhildsen J, Cedergren M, Jeppsson A, Nystro¨m F, et al. Weight gain restriction for obese pregnant women: a caseecontrol intervention study. BJOG 2008;115:44e50. Statistics Sweden. Available at: http://www.scb.se/templates/ tableOrChart____149077.asp [accessed 07.11.06]. Nord DRG users’ manual: updated 2004-03-05. Available at: http://norddrg.kuntaliitto.fi/manual_2004_Eng_Full [accessed 03.11.06]. Institute of Medicine. Weight gain during pregnancy: reexamining the guidelines. Washington: National Academies Press; 2009. Cedergren MI. Optimal gestational weight gain for body mass index categories. Obstet Gynecol 2007;110:459e64. Claesson IM, Josefsson A, Cedergren M, Brynhildsen J, Jeppsson A, Nystro¨m F, et al. Consumer satisfaction with a weight-gain intervention programme for obese pregnant women. Midwifery 2008;24:163e7. Graziano AM, Raulin ML. Research methods: a process of inquiry (with student CD-ROM). Boston: Pearson; 2004. Polit DF, Hungler BP. Nursing research: principles and methods. Philadelphia: Lippincott; 1991. Statistics Sweden. Live births in the country 2009. Available at: www.ssd.scb.se [accessed 28.04.10].
available at www.sciencedirect.com
Public Health journal homepage: www.elsevier.com/puhe
Original Research
Total cost comparison of standard antenatal care with a weight gain restriction programme for obese pregnant women N. de Keyser a, A. Josefsson a, W.G. Monfils a, I.-M. Claesson a, P. Carlsson b, A. Sydsjo¨ a, G. Sydsjo¨ a,* a
Division of Obstetrics and Gynaecology, Department of Clinical and Experimental Medicine, Faculty of Health Sciences, Linko¨ping University, SE-581 85 Linko¨ping, Sweden b Centre for Medical Technology Assessment, Department of Medical and Health Sciences, Faculty of Health Sciences, Linko¨ping University, Linko¨ping, Sweden
article info
summary
Article history:
Objective: To perform a cost comparison of a weight gain restriction programme for obese
Received 3 May 2010
pregnant women with standard antenatal care, and to identify if there were differences in
Received in revised form
healthcare costs within the intervention group related to degree of gestational weight gain
1 December 2010
or degree of obesity at programme entry.
Accepted 9 February 2011
Study design: A comparison of mean healthcare costs for participants of an intervention study at antenatal care clinics with controls in south-east Sweden. Methods: In total, 155 women in an intervention group attempted to restrict their gesta-
Keywords:
tional weight gain to <7 kg. The control group comprised 193 women. Mean costs during
Pregnancy
pregnancy, delivery and the neonatal period were compared with the costs of standard
Weight restriction
care. Costs were converted from Swedish Kronor to Euros (V).
Intervention
Results: Healthcare costs during pregnancy were lower in the intervention group. There was
Cost-effectiveness
no significant difference in total healthcare costs (i.e. sum of costs during pregnancy, delivery and the neonatal period) between the intervention group and the control group. Within the intervention group, the subgroup that gained 4.5e9.5 kg had the lowest costs. The total cost, including intervention costs, was V1283 more per woman/infant in the intervention group compared with the control group (P ¼ 0.025). The degree of obesity at programme entry had no bearing on the outcome. Conclusions: The weight gain restriction programme for obese pregnant women was effective in restricting gestational weight gain to <7 kg, but had a higher total cost compared with standard antenatal care. ª 2011 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: þ46 13 22 20 00; fax: þ46 13 14 81 56. E-mail address: [email protected] (G. Sydsjo¨). 0033-3506/$ e see front matter ª 2011 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.puhe.2011.02.004
312
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
Introduction The obesity epidemic is not confined to a single subset of the world’s population, but is spread across all races, age groups, genders and socio-economic classes, including pregnant women.1e3 From 1978 to 2001, the proportion of overweight women of childbearing age tripled in Sweden, and obesity in that population increased five-fold.4 The increase in body mass index (BMI) brings with it an increased risk for complications during pregnancy, delivery and the postpartum period.5e9 The infant of an obese woman is at increased risk for antepartum stillbirth, neonatal death, birth defects and neonatal complications.10e14 Recent studies have shown that some of these risks can be averted if a woman’s pregnancy weight gain is restricted.15e17 Few studies have provided quantitative estimations of the related increase in the use of healthcare services for obese pregnant women. General healthcare costs for obese pregnant women were studied by Chu et al.,18 who found that obesity during pregnancy was associated with significantly more examinations, more obstetric ultrasonographic examinations and increased use of medication. Chu et al. also found that obese women had more prenatal visits with physicians, a longer hospital stay after delivery, and more telephone calls to obstetrics and gynaecology departments. Most of the increase in length of stay associated with higher BMI was related to increased rates of caesarean delivery and obesity-related high-risk conditions. Few interventions exist to control excessive weight gain during pregnancy, and those that do exist have shown poor results.19e21 Policy makers and healthcare planners are aware that even a small increase in the cost of health care associated with obesity during pregnancy for the woman, and later her child, will have substantial economic implications for the future. This makes intervention programmes and assessment of efficacy and economic viability of interventions essential. In order to make an informed decision about which interventions to implement, healthcare providers and policy makers must have information about which programmes maximize health, given everpresent budget constraints. The purpose of this study was to compare the total costs during pregnancy, delivery and the neonatal period between women in the index group of a weight gain restriction programme for obese pregnant women conducted at the University Hospital of Linko¨ping, Sweden and a cohort of women given standard antenatal care.
were invited to a 30-min session every week. The midwife worked according to the following schedule: (1) a thorough assessment of the pregnant woman’s knowledge of obesity as a risk factor during pregnancy, childbirth and for the child; (2) if the woman lacked sufficient knowledge, she was offered the information and given accurate facts; (3) the woman was informed about the potential consequences of behaviour and cognition associated with eating and food intake, and written information was supplied if needed; (4) all successive sessions included weight control and a supportive talk regarding how to deal with food intake, motivational change of behaviour, how to deal with partner and family, supporting exercises and behavioural pitfalls; and (5) all women were invited to participate in a water-gymnastics programme designed for obese women once or twice a week.22 The target weight gain during pregnancy was <7 kg. The women in the control group were offered standard antenatal care. The intervention group consisted of 155 women with a BMI 30 chosen consecutively from all pregnant women attending regular antenatal care at the University Hospital of Linko¨ping, Sweden between November 2003 and December 2005. Exclusion criteria were inability to speak and understand Swedish, diagnosed with diabetes or disease related to thyroid function, and current treatment with neuroleptic drugs. Forty-five women met these criteria and were excluded from the study. In addition, 13 women who had an early miscarriage or a legal abortion, and 29 women who moved out of the catchment area in early pregnancy were also excluded. Two hundred and thirty obese women were thus eligible and invited to participate. Out of these, 70 women refrained from participation and five women dropped out during the intervention. Thus, a total of 155 obese women (67.4%) agreed to participate and completed the intervention. To form the control group, women with the same inclusion and exclusion criteria were consecutively invited from the ACCs in two nearby cities over the same time period. Three hundred and eighty-five women were invited to participate, 177 women refrained from participation and 15 women dropped out during the study period. Thus, 193 (50.1%) women agreed to participate and completed the study as controls. The obese women in the control group attended the routine antenatal care programme. All three cities share similar demographics such as age, parity, marital status and smoking,22 and lie within south-east Sweden (Fig. 1).
Costs Methods The Swedish antenatal healthcare system and delivery care reaches almost 100% of pregnant women. Antenatal and delivery care is free of charge. At antenatal care clinics (ACCs), healthy pregnant women are recommended to attend the regular antenatal programme with seven to nine visits to a midwife, and, if needed, extra appointments with an obstetrician and/or with a midwife. In the intervention study, pregnancy weight gain was controlled in the intervention group through weekly motivational meetings with a cognitive and behavioural change approach conducted by a specially trained midwife. All women who agreed to take part in the intervention
All costs have been converted from Swedish Kronor to Euros (V) using a 10:1 ratio.
Cost of the intervention programme A motivational meeting with a midwife cost V82 (national average cost of a 30-min meeting with a midwife). Each meeting was documented in the patient’s journal. The rental cost of the water-gymnastics pool was approximately V60/h. Each water-gymnastics session was supervised by a midwife, with a concurrent cost of V28/session. Costs for loss of productivity due to participation in the programme were calculated using the mean monthly income
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
313
of the cost of an appointment (V99). An appointment with a midwife cost V82. In the event of inpatient care during pregnancy, the appropriate DRG costs were assigned.
Costs during delivery The guidelines for the choice of DRG during delivery were as follows: vaginal delivery, uncomplicated; vaginal delivery, complicated (if vaginal delivery required the use of instrumental aid, or resulted in a Grade IIIeIV vaginal rupture, placental retention or bleeding >1000 ml); caesarean section, uncomplicated; and caesarean section, complicated (acute caesarean section).
Costs of neonatal care
Fig. 1 e Subjects in the study.
in 2005 for all Swedish women (V2210). Fifty percent was added for the employer’s contribution to the national social insurance system, giving a total of V3315.23 Given that approximately two-thirds of women in the intervention group were employed at the beginning of the programme, the total monthly income was multiplied by 2/3 to reach a sum of V2210. There are approximately 22 working days per month. The hourly cost of productivity loss was therefore calculated as follows: V2210/22 giving a daily income of approximately V100, then divided by 8 working hours (100/8) ¼ V12.56. Each motivational meeting or water-gymnastics session was assumed to take 1.5 h, including travel time, and thus contributed V19 to the cost of the programme. The number of motivational meetings attended was obtained from standardized patient journals. The number of water-gymnastics sessions attended was obtained from a sign-in sheet.
Healthcare costs Hospital costs during pregnancy, delivery and the neonatal period were calculated using national averages as presented by NordDRG. Diagnosis-related group (DRG) uses a system of averaging costs for diagnosis, procedures performed, gender, age, length of hospitalization and status at discharge to classify patients into categories with similar use of resources.24 The following inclusions and assumptions were made when assigning each diagnosis into each DRG.
Costs during pregnancy Each appointment with a physician, whether as an outpatient at the hospital or at a public ACC, cost V297. The cost for a midwife consulting a physician was estimated to be one-third
The assigned DRG for each case depended on the birth weight and number of diagnoses/complications that were noted in the patient’s journal. In the case of a diagnosis of ‘large for gestational age’, ‘postmature’ or ‘hyperbilirubinaemia’, the diagnosis was not considered to be a complication if the infant was not treated in the neonatal ward. An uncomplicated pregnancy, vaginal delivery and a healthy child costs V4400. A short hospitalization during pregnancy, such as might result in the case of an emergency caesarean section and an infant requiring limited hospital care, costs approximately V12,000.
Analyses Mean total healthcare costs and costs during pregnancy, delivery and the neonatal period for the intervention group were compared with costs for the control group. The weight gain subgroups of women who gained <7 kg (target maximum weight gain of the programme) and 7 kg were compared. Since weight gain recommendations are a topic of current debate, the groups were further divided as follows: (1) those who gained <4.5 kg during pregnancy; (2) those whose gained 4.5e9.5 kg during pregnancy; and (3) those who gained >9.5 kg during pregnancy. Cost variables within each of these subgroups were compared with the control group. Weight gain was calculated using the last registered weight during a motivational meeting minus weight at programme entry. Each subgroup was further compared with their respective weight gain interval in the control group to determine if any potential differences were possibly attributable to factors other than weight gain. Additionally, the outcomes based on BMI at programme entry (stratified into 30e34.9 kg/m2, 35e39.9 kg/m2 and 40 kg/m2) were compared to elucidate any possible effect of the degree of obesity on the results of the programme.
Statistical methods Mean total healthcare cost and costs during pregnancy, delivery and the neonatal period were compared. Additionally, intervention costs were added to the total cost and compared with the total average cost for a pregnant woman in the control group. The costs were compared using Student’s t-test.23 All
314
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
tests were two-tailed and 95% confidence intervals were calculated. All calculations were performed using Statistical Package for the Social Sciences Version 15.0 (SPSS Inc., Chicago, IL, USA).
cost of health care and intervention. No difference in total healthcare cost was found between the groups. However, the intervention group had significantly lower costs (approximately V348/woman) during pregnancy than the control group.
Results Total cost of health care and intervention Brief summary of the results of the intervention programme (for more details on obstetric and neonatal outcomes, see Claesson et al.22) The mean weight gain for the intervention group was 2.6 kg less than that of the control group (8.7 kg vs. 11.3 kg; P < 0.001). Significantly more women in the intervention group gained <7 kg compared with the control group (P ¼ 0.003), and there were more nulliparous women in the intervention group than the control group (P ¼ .018). The women in the intervention group had a significantly lower BMI at the postnatal check-up compared with the control group (P < 0.001). Neonatal outcomes such as birth weight, gestational age and mode of delivery/complications did not differ between the groups. The same held true when the analyses were restricted to women who gained <7 kg during pregnancy.22
As seen in Table 1, the intervention cost per woman was V1616. During the 27 months of the intervention programme, 155 women took part, resulting in a total cost of V250,511. The mean per woman/infant difference in total cost, including intervention costs, was V1283 more for the intervention group than the control group (P < 0.025).
Intervention subgroups compared with control group For subgroup analyses within the intervention group, the women who gained <7 kg (target maximum weight gain for the intervention group) were compared with the controls, and the women in the intervention group who gained >7 kg were compared with the controls. There was no difference in total cost (Table 2).
Intervention subgroups <4.5, 4.5e9.5 and >9.5 kg compared with control group
Total healthcare costs Two extremely preterm infants (one from the intervention group, one from the control group) were hospitalized for approximately 90 days. Both cases were excluded from the calculation of neonatal costs, total healthcare cost, and total
The intervention subgroup who gained <4.5 kg showed increased costs for neonatal care and total cost (P ¼ 0.16 and 0.20, respectively) compared with the control group. However,
Table 1 e Average costs (in Euros) per woman or woman/child pair: intervention and control groups.
Healthcare costs Pregnancy Physician appointment Physician consultation Midwife appointment Hospitalization (DRG) Total Delivery Hospitalization (DRG)
Intervention (n ¼ 155-1a)
Control (n ¼ 193-1a)
Mean (SD)
Mean (SD)
778 (592) 94 (131) 658 (125) 169 (619) 1699 (1040)
P-value
95% CI
928 (866) 67 (104) 713 (173) 263 (616) 2044 (1199)
0.057 0.039 0.001 0.157 0.004
304 to 4 1e53 86 to 23 226 to 37 582 to 109
0.390
122 to 313
2397 (1012)
2302 (1039)
Neonatal care (One child in each group is missing from the analysis) Hospitalization (DRG)
(n ¼ 154)
(n ¼ 192)
3187 (4550)
3559 (4732)
0.885
1062 to 917
Total healthcare costs
7258 (5117)
7595 (5390)
0.556
785
Intervention costs Motivational meetings Loss of productivity due to meetings Water-gymnastics Loss of productivity due to water-gymnastics Total intervention costs
1156 248 83 131 1616 7595 (5390)
0.025
164e2403
Total cost of health care and intervention (per woman/child pair)
8878 (5095)
SD, standard deviation; CI, confidence interval; and DRG, diagnosis-related group. a An extremely preterm child has been excluded from both Intervention and Control group.
315
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
Table 2 e Average costs (in Euros) per woman or woman/child paira: intervention subgroups <7 kg and ‡7 kg and control group. Intervention subgroup
Control (n ¼ 193-1)b
Mean (SD)
Mean (SD)
Subgroup healthcare cost Subgroup: weight gain <7 kg Pregnancy Delivery Neonatal care Total healthcare cost
n ¼ 57-1b 1663 (528) 2216 (932) 3414 (5140) 7987 (8047)
Subgroup: weight gain 7 kg Pregnancy Delivery Neonatal care Total healthcare cost
n ¼ 98 1719 (1037) 2503 (1045) 3057 (4198) 7278 (4803)
P-value
95% CI
2021 (880) 2306 (1230) 2927 (4103) 7255 (4804)
0.05 0.56 0.47 0.52
715 to 1 392 to 214 841 to 1814 1008 to 2471
2021 (880) 2306 (1030) 2927 (3803) 7255 (4804)
0.04 0.14 0.81 0.97
592 to 13 62 to 454 902 to 980 1174 to 1220
SD, standard deviation; and CI, confidence interval. a For the women with singleton pregnancies whose weight was assessed at the start of the intervention and 1e2 weeks before delivery. b For the exclusion of the extremely preterm child.
after excluding the extremely preterm cases, the results showed no such trend (P ¼ 0.88 and 0.95, respectively). There was no difference in costs for the subgroup that gained >9.5 kg compared with the control group. However, in the subgroup that gained 4.5e9.5 kg, the mean cost during pregnancy was V574 less than that for the control group (P < 0.001) (Table 3).
Intervention subgroups compared with control subgroups Total healthcare costs for all subgroups in the intervention group and the control group (e.g. <7 kg intervention compared with <7 kg control) showed no significant differences. However, differences were found between the DRG costs during pregnancy. A statistically significant reduction in costs was found for the intervention subgroups <7 kg and
4.5e9.5 kg. A similar trend was found within the other subgroups (data not shown).
Stratified BMI at programme entry Women in the intervention group with an initial BMI of 30e34.9 kg/m2 (n ¼ 100) had a mean total healthcare cost, excluding intervention costs, of V7948, while women in the control group with the same initial BMI (n ¼ 112) had a mean cost of V7023; this difference was not significant (P ¼ 0.226). Women in the intervention group with an initial BMI of 35e39.9 kg/m2 (n ¼ 36) had a mean total healthcare cost of V6136, while women in the control group with the same initial BMI (n ¼ 39) had a cost of V7007; again, this difference was not significant (P ¼ 0.249). Finally, women in the intervention
Table 3 e Average costs (in Euros) per woman or woman/child paira: intervention subgroups and control groups. Intervention subgroup
Control (n ¼ 193-1)b
Mean (SD)
Mean (SD)
Subgroup healthcare cost Subgroup: weight gain <4.5 kg Pregnancy Delivery Neonatal care Total healthcare cost
n ¼ 33-1b 1997 (1247) 2375 (990) 4300 (8429) 8671 (9227)
2021 2306 2927 7255
Subgroup: weight gain 4.5e9.5 kg Pregnancy Delivery Neonatal care Total healthcare cost
n ¼ 54 1448 (845) 2343 (980) 3385 (5145) 7175 (5666)
Subgroup: weight gain >9.5 kg Pregnancy Delivery Neonatal care Total healthcare cost
n ¼ 68 1766 (1049) 2453 (1059) 3082 (3971) 7301 (4648)
(880) (1029) (4103) (4804)
2021 (880) 2306 (1029) 2927 (4103)
2021 2306 2927 7255
(880) (1029) (4103) (4804)
P-value
0.92 0.73 0.16 0.20
95% CI
491 319 532 751
to to to to
443 458 3277 3584
<0.001 0.82 0.50 0.92
920 to e227 272 to 345 871 to 1785 1604 to 1444
0.11 0.33 0.79 0.95
569 to 58 147 to 442 999 to 1309 1304 to 1397
SD, standard deviation; and CI, confidence interval. a For the women with singleton pregnancies whose weight was assessed at the start of the intervention and 1e2 weeks before delivery. b For the exclusion of the extremely preterm child.
316
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
group with the highest BMI (n ¼ 19) had a mean total healthcare cost of V8043, while women in the control group with the same initial BMI (n ¼ 21) had a mean cost of V8952; once again, this difference was not significant (P ¼ 0.723). All costs contributing to the total cost (i.e. pregnancy, delivery, neonatal care) were analysed in the same manner. No significant differences were found.
Frequency of exercise Attendance of water-gymnastics sessions during pregnancy ranged from zero to 28, with a mean of seven and a mode of five. A small negative correlation (0.15, P < 0.07) was found between frequency of attendance of water-gymnastics sessions and gestational weight gain; however, no correlation was found between exercise and costs.
Lost weight during pregnancy Eleven women lost weight during their pregnancy. No differences were found in costs associated with pregnancy, delivery, neonatal care or total cost compared with the control group.
Discussion This study is, to the authors’ knowledge, the first analysis of the total cost of an intervention programme to limit weight gain in obese women during pregnancy. The cost comparison analyses and interpretation of results in the present study were conducted by individuals who were not connected with the intervention programme, thereby preventing conflicts of interest. The results show that the intervention programme had no economic benefit as a whole, although the intervention group had lower costs during pregnancy (approximately V348). Recommendations for weight gain by obese women during pregnancy vary, and are a topic of current debate. The Institute of Medicine25 has recently changed its recommendation from a minimum weight gain of 6.8 kg to weight gain between 5 and 9 kg, while a study published by Cedergren suggests that weight gain of <6 kg is appropriate for obese women.26 The gestational weight gain limits for BMI categories in Cedergren’s large population-based cohort study from Swedish Medical Registers showed that a decreased risk of adverse obstetric and neonatal outcomes was associated with lower gestational weight gain than was recommended previously, especially among obese women. These findings are supported by the results of a recent randomized clinical trial concerning perinatal outcomes in nutritionally monitored obese pregnant women.16 Regarding the debate concerning recommendations for weight gain by obese women during pregnancy, further studies are warranted to ensure the safety and efficacy of future intervention programmes. The benefits of reduced costs during pregnancy in this particular intervention study were outweighed by the cost of the intervention (V1616/participant). Data analyses found that the pregnancy weight gain that provided the greatest economic saving was between 4.5 and 9.5 kg. Even given the best case scenario under which all women in the intervention group kept their weight gain to 4.5e9.5 kg, a net loss of
approximately V1002/participant would be expected. The cost of this particular intervention programme therefore precludes any potential economic benefit. Considering that the motivational meetings and loss of productivity due to the meetings accounted for 87% (V1404) of the intervention costs, a sizeable cost reduction here may make cost-effectiveness achievable. Say, for instance, that the motivational meetings could be conducted effectively in groups of six women, every other week as opposed to individual, weekly meetings. The total intervention cost could thus be reduced to approximately V330/participant. For women who gain 4.5e9.5 kg, a net saving of approximately V249 could be made. Such an adjustment may be warranted for future intervention studies, given that approximately half of the women in the intervention group stated that there were too many motivational meetings, and 72% reported that the social character of the water-gymnastics (a group session) was one of the most positive aspects of the programme.27 This intervention study had both strengths and limitations. The intervention study was not randomized, which can be seen as an important limitation. Randomization is a powerful tool and a good general rule is to randomize whenever possible.28,29 In all scientific research, it is important to control external factors and to make an effort to minimize situational contaminants. The environment has been found to exert a powerful influence on people’s emotions and behaviours,29 and requires careful consideration if intervention and control groups will get treatment and care at the same setting. A continuing intervention programme with a new routine might influence the staff to change their behaviour and regimens, even for those who are supposed to be controls and receive standard care. The ACC in Linko¨ping is structured as one large unit. As the intervention programme designers were determined to ensure that no contamination occurred between the intervention and control groups, ACCs in two nearby cities were used as controls. The antenatal programme in Sweden is standardized and almost identical in terms of pregnancy management, which ensures similar care at different ACCs. Therefore, the design was consecutive inclusion into the intervention group, with two other cities and their ACCs used for controls. Additionally, background characteristics were analysed in the intervention study to correct for possible confounding factors. However, there may still have been small differences in management and routines that may have affected the results. The present results are based on univariate statistical analyses. These results may, however, be interpreted to be a reasonable representation of the wider population as the degree of weight gain in the intervention and control groups was shown to be independent of all demographic characteristics when analysed using ANCOVA.22 Subgroup analyses were performed shortly before all women had given birth, thus the numbers of individuals in the intervention and control groups are lower than those used for analysis of the programme as a whole. This small reduction is not expected to have any impact on the results. The standard deviation of the cost factors was extremely wide; as such, the statistical power to find any difference in costs, if they exist, is low. However, since the mean costs during pregnancy, delivery and the neonatal period were very
p u b l i c h e a l t h 1 2 5 ( 2 0 1 1 ) 3 1 1 e3 1 7
close in the intervention and control groups, it can be assumed that no difference exists. Approximately 110,000 children are born in Sweden each year,30 and about 11,000 of these are born to obese women.4 If one-third of these obese women could keep their weight gain to 4.5e9.5 kg during pregnancy, a total direct saving to society of around V1,000,000 could be expected. With other health benefits that follow from a healthy change in lifestyle, the benefits would certainly exceed this modest short-term saving. Since better health for woman and child, and budget streamlining are natural goals for policy makers and healthcare providers, further efforts to create a viable weight gain control programme during pregnancy are warranted.
10. 11. 12.
13.
14.
15.
Ethical approval Human Research Ethics Committee, Faculty of Health Sciences, Linko¨ping University, Sweden.
16.
Funding
17.
Grants from the Health Research Council in south-east Sweden. 18.
Competing interests None declared.
19.
references
20.
1. The Swedish Council on Technology Assessment in Health Care. Fetma e problem och a˚tga¨rder, En systematisk litteraturo¨versikt. Stockholm: SBU; 2002. 2. Yeh J, Shelton JA. Increasing pre-pregnancy body mass index: analysis of trends and contributing variables. Am J Obstet Gynecol 2005;193:1994e8. 3. La Coursiere DY, Bloebaum L, Duncan JD, Varner MW. Population-based trends and correlates of maternal overweight and obesity, Utah 1991e2001. Am J Obstet Gynecol 2005;192:832e9. 4. Brynhildsen J, Sydsjo¨ A, Norinder E, Ekholm Selling K, Sydsjo¨ G. Trends in body mass index during early pregnancy in Swedish women 1978e2001. Public Health 2006;120:393e9. 5. Dempsey JC, Ashiny Z, Qiu C-F, Miller RS, Sorensen TK, Williams MA. Maternal pre-pregnancy overweight status and obesity as risk factors for caesarean delivery. J Matern Fetal Neonatal Med 2005;17:179e85. 6. Sebire NJ, Jolly M, Harris JP, Wadsworth J, Joffe M, Beard RW, et al. Maternal obesity and pregnancy outcome: a study of 287,213 pregnancies in London. Int J Obes Relat Metab Disord 2001;25:1175e82. 7. Kumari AS. Pregnancy outcome in women with morbid obesity. Int J Gynaecol Obstet 2001;73:101e7. 8. Baeten JM, Bukasi EA, Lambe M. Pregnancy complications and outcomes among overweight and obese nulliparous women. Am J Public Health 2001;91:436e40. 9. Vahratian A, Zhang J, Troendle JF, Savitz DA, Siega-Riz AM. Maternal prepregnancy overweight and obesity and the
21.
22.
23. 24.
25.
26. 27.
28. 29. 30.
317
pattern of labor progression in term nulliparous women. Obstet Gynecol 2004;104:943e51. Cedergren MI. Maternal morbid obesity and the risk of adverse pregnancy outcome. Obstet Gynecol 2004;103:219e24. Cedergren MI, Ka¨lle´n BAJ. Maternal obesity and infant heart defects. Obes Res 2003;11:1065e71. Kristensen J, Vestergaard M, Wisborg K, Kesmodel U, Secher NJ. Pre-pregnancy weight and the risk of stillbirth and neonatal death. BJOG 2005;112:403e8. Stephansson O, Dickman PW, Johansson A, Cnattingius S. Maternal weight, pregnancy weight gain, and the risk of antepartum stillbirth. Am J Obstet Gynecol 2001;184:463e9. Cnattingius S, Bergstro¨m R, Lipworth L, Kramer MS. Prepregnancy weight and the risk of adverse pregnancy outcomes. N Engl J Med 1998;338:191e2. Cedergren M. Effects of gestational weight gain and body mass index on obstetric outcome in Sweden. Int J Gynaecol Obstet 2006;93:269e74. Thorton YS, Smarkola C, Kopacz SM, Ishoof SB. Perinatal outcomes in nutritionally monitored obese pregnant women: a randomised clinical trial. J Natl Med Assoc 2009; 101:569e77. Crane JMG, White J, Murphy P, Burrage L, Hutchens D. The effect of gestational weight gain by body mass index on maternal and neonatal outcomes. J Obstet Gynaecol Can 2009; 31:28e35. Chu SY, Bachman DJ, Callaghan WM, Whitlock EP, Dietz PM, Berg CJ, et al. Association between obesity during pregnancy and increased use of health care. N Engl J Med 2008;358: 1444e53. Polley BA, Wing RR, Sims CJ. Randomized controlled trial to prevent excessive weight gain in pregnant women. Int J Obes Relat Metab Disord 2002;26:1494e502. Olson CM, Strawderman MS, Reed RG. Efficacy of an intervention to prevent excessive gestational weight gain. Am J Obstet Gynecol 2004;191:530e6. Gray-Donald K, Robinson E, Collier A, David K, Renaud L, Rodrigues S. Intervening to reduce weight gain in pregnancy and gestational diabetes mellitus in Cree communities: an evaluation. CMAJ 2000;163:1247e51. Claesson I-M, Sydsjo¨ G, Brynhildsen J, Cedergren M, Jeppsson A, Nystro¨m F, et al. Weight gain restriction for obese pregnant women: a caseecontrol intervention study. BJOG 2008;115:44e50. Statistics Sweden. Available at: http://www.scb.se/templates/ tableOrChart____149077.asp [accessed 07.11.06]. Nord DRG users’ manual: updated 2004-03-05. Available at: http://norddrg.kuntaliitto.fi/manual_2004_Eng_Full [accessed 03.11.06]. Institute of Medicine. Weight gain during pregnancy: reexamining the guidelines. Washington: National Academies Press; 2009. Cedergren MI. Optimal gestational weight gain for body mass index categories. Obstet Gynecol 2007;110:459e64. Claesson IM, Josefsson A, Cedergren M, Brynhildsen J, Jeppsson A, Nystro¨m F, et al. Consumer satisfaction with a weight-gain intervention programme for obese pregnant women. Midwifery 2008;24:163e7. Graziano AM, Raulin ML. Research methods: a process of inquiry (with student CD-ROM). Boston: Pearson; 2004. Polit DF, Hungler BP. Nursing research: principles and methods. Philadelphia: Lippincott; 1991. Statistics Sweden. Live births in the country 2009. Available at: www.ssd.scb.se [accessed 28.04.10].