Obesity
Original Article EPIDEMIOLOGY/GENETICS
Multimorbidity in a Prospective Cohort: Prevalence and Associations with Weight Loss and Health Status in Severely Obese Patients Calypse B. Agborsangaya1,2, Sumit R. Majumdar1,2, Arya M. Sharma1,2, Edward W. Gregg3, and Raj S. Padwal1,2
Objective: To examine the prevalence of multimorbidity (2 chronic conditions) in severely obese patients and its associations with weight loss and health status over 2 years. Methods: In a prospective cohort including 500 severely obese adults, self-reported prevalence of 20 chronic conditions was calculated at baseline and 2 years. Multivariable logistic regression models were fitted to test the covariate-adjusted associations between 5% weight reduction and reduction in multimorbidity and the association between health status (visual analogue scale [VAS]) and reduction in multimorbidity over 2 years. Results: After 2 years, mean weight change was 212.9 618.7 kg, 53% had 5% weight reduction, mean change in VAS was 11.5 6 21.2, and 53.5% had 10% increase in VAS. Multimorbidity was reported in 95.4% and 92.8% patients at baseline and 2 years, respectively. Weight loss (5%) over 2 years was associated with reduction in multimorbidity (adjusted OR 5 1.7, 95% CI 1.1-2.7). Reduction in multimorbidity was associated with clinically important improvements (10% increase in VAS) in health status (adjusted OR 5 2.5, 95% CI 1.6, 4.0). Conclusions: Multimorbidity is common in severely obese patients. Having 5% weight reduction over 2 years was associated with a reduction in multimorbidity, which was also associated with improvements in health status. Obesity (2015) 23, 707–712. doi:10.1002/oby.21008
Introduction Multimorbidity, defined as the concurrent occurrence of two or more chronic conditions in an individual, is increasingly common (1). It has been linked to adverse health outcomes such as frequent and longer hospitalization, higher healthcare costs, readmission, reduced quality of life, and mortality (2-4). The prevalence of multimorbidity varies markedly depending on the patient population studied, ranging from about 17% to 90% (1,5-7). This variation in prevalence results in part from the absence of a single, standardized definition. In particular, the inclusion of obesity in the list of chronic conditions increases multimorbidity prevalence substantially because obesity is very common, affecting one-quarter to one-third of individuals in developed nations (8). Obesity is also an important propagator of multimorbidity because it leads to the development of numerous medical complications (9-15). Individuals with severe obesity (BMI 40 kg/m2) are at increased
risk of developing these complications (14). Severe obesity is common, affecting 3% of adult Canadians and 6% of adult Americans (13) and is growing rapidly (16). Despite these well-established links between obesity and the development of comorbidities, few data exist characterizing multimorbidity in patients with severe obesity or the response of multimorbidity to weight loss (17). Weight loss in patients with severe obesity may lead to chronic disease remission (18,19). In a recent study, Courcoulas and colleagues (18) investigated the effect of surgical interventions on the occurrence of chronic conditions in patients with severe obesity after a 3-year follow-up and noted up to 68% partial remission in patients with diabetes, 62% in patients with dyslipidemia and 38% in patients with hypertension. Despite the overarching effect of excess weight on chronic multimorbidity, this study examined only three chronic conditions. In fact, most studies evaluating the effect of weight loss are limited to a single or few health outcomes (20,21).
1 Department of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada. Correspondence: Calypse B. Agborsangaya (agborsan@ ualberta.ca) 2 Alberta Diabetes Institute, Edmonton, Alberta, Canada 3 Division of Diabetes Translation, Centers for Disease Control and Prevention, Atlanta, Georgia, USA.
Disclosure: The authors declared no conflict of interest. Author contributions: SRM, AMS, EWG, ACB, and RSP conceived the study and design. CBA completed the analysis, drafted the first manuscript, and completed manuscript edits. SRM, AMS, EWG, and RSP also contributed to manuscript editing, and along with ACB, had final approval of the submitted and published versions. Received: 27 June 2014; Accepted: 23 November 2014; Published online 6 February 2015. doi:10.1002/oby.21008
www.obesityjournal.org
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
707
Obesity
Multimorbidity in Severely Obese Patients Agborsangaya et al.
Accordingly, the objectives of this study were first, to characterize multimorbidity at baseline, including its prevalence and most common comorbid conditions in a prospective cohort of 500 severely obese patients recruited from a population-based regional bariatric program and followed for 2 years Second, to determine the association between change in weight and change in multimorbidity status. Third, we studied the association between change in multimorbidity and change in health status. We hypothesized that weight loss over 2 years would be associated with a reduction in multimorbidity, and that reductions in multimorbidity would be associated improvement in health status.
Methods Setting and subjects We analyzed data from the Alberta Population-based Prospective Evaluation of the Quality of Life Outcomes and Economic Impact of Bariatric Surgery (APPLES) study, a 500-patient cohort study in which consecutive, consenting adults with BMI levels 35 kg/m2 were recruited from the Adult Edmonton Weight Wise regional bariatric program. A description of this study has been published in a detailed study protocol, approved by the University of Alberta Health Research Ethics Board (Pro00003594) (15). Briefly, Edmonton Weight Wise is a comprehensive bariatric program structured such that patients progress from a wait list to intensive medical management to bariatric surgery (in interested and eligible candidates). The 500 patients enrolled were between 18 and 60 years old and were either wait-listed (n 5 150), beginning intensive medical treatment (n 5 200) or had just been approved for bariatric surgery (n 5 150). The patient characteristics and main study findings are published elsewhere; percentage weight reductions over 2 years were 1%, 3%, and 16% for wait-listed, medically treated and surgical-treated groups, respectively (22). For the present study, our objective was to characterize multimorbidity in the entire study population; thus, all 500 patients were combined and initial study group (wait-list, medical, surgical) was controlled for in the analysis (as detailed below).
Measurement and data collection Detailed case report forms have been previously published (15). Baseline data collection included socio-demographic characteristics and medical history. Participants’ body weight was measured using a calibrated bariatric scale and recorded to the nearest 0.1kg. Height was measured to the nearest 0.1 cm using a wall-mounted stadiometer. Finally, health status was measured at all study time points using a visual analogue scale (VAS), a component of EuroQol’s EQ-5D quality of life instrument that ranges from 0 to 100, with 0 representing death and 100, a state of complete health; previous work has established that the minimal important difference on this previously validated scale is 10 (23).
Multimorbidity definitions Multimorbidity is commonly defined as the presence of more than one chronic condition in the same individual (1). For purposes of comparison, an alternative definition (defined as 3 chronic conditions) used in the Canadian Community Health Survey (CCHS) (5) was also examined. The number of chronic conditions per patient was determined by a simple count of the following 20 self-reported chronic conditions: anxiety or depression, asthma, bipolar disorder,
708
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
chronic kidney disease, chronic urinal/fecal incontinence, diabetes, dyslipidemia, fatty liver disease, fibromyalgia, gastrointestinal reflux, heart failure, hypertension, hypothyroidism, obesity hypoventilation, lymphedema, joint pain (on the basis of osteoarthritis), polycystic ovary syndrome, sleep apnea, stroke, and venous stasis (6). Three conditions (hypertension, diabetes, and dyslipidemia) were defined based on self-report and clinical parameters. Hypertension was considered present if self-reported blood pressure levels were 140/90 mm Hg (130 mm Hg in patients with diabetes), or if antihypertensive medications were prescribed. An A1c over 6.5% or a fasting glucose 7.0 mmol/L were used as diagnostic criteria for diabetes. Dyslipidemia was considered present if one of the following conditions was met: total cholesterol 6.2 mmol/L, low-density lipoprotein (LDL) cholesterol 4.1 mmol/L, high-density lipoprotein (HDL) cholesterol < 1.0 mmol/L, or a triglyceride 2.3 mmol/L.
Statistical analysis Descriptive analyses were first conducted, including calculation of means, medians, and proportions. Mean VAS scores were calculated according to the number of baseline chronic conditions (categorized from 0 to 8). ANOVA was used to compare the mean VAS score in subjects with different numbers of comorbidities (ranging from 1 to 8) to the VAS scores in subjects with no comorbidity. To examine predictors of multimorbidity reduction over the followup period, we grouped subjects into binary categories according to whether or not their number of chronic conditions was reduced over 2 years. Binary logistic regression was used to examine predictors of a reduced number of chronic conditions over the follow-up period (yes/no). The major predictor variable of interest was 2-year 5% change in weight, classified as a binary variable (5% vs. > 5%). Greater than a 5% reduction in weight was chosen because it is thought to be clinically relevant threshold indicating change in disease status (24,25). In a sensitivity analysis, we tested the association between 7% weight loss and multimorbidity reduction and noted no significant change in our findings. To evaluate the association between change in multimorbidity and change in health status, VAS was dichotomized to indicate clinically important improvement ( 10% versus < 10%). All models were adjusted for age, sex, income status and baseline BMI. Because we wanted to isolate the effect of weight loss on change in number of chronic conditions, analysis was performed in the combined sample while adjusting for treatment status (waitlist vs medical vs surgical). No adjustments for multiple testing were performed and a P-value < 0.05 was considered statistically significant. At 2 years, analyses were restricted to 406 patients with evaluable data (i.e., a completers’ analysis). All analyses were performed using STATA software version 12 (Stata, TX, USA).
Results Of 500 enrolled patients, 406 (81%) had complete baseline and 2-year weight data. The average baseline weight did not differ for completers (mean 5 131.5 kg, SD 25.0) and noncompleters (mean 5 133.9 kg, SD 25.3), P 5 0.4. The socio-demographic characteristics of the 406 completers are presented in Table 1. The majority was female (88.2%) and white (93.6%), with a mean age of 46.1 years (SD 9.5). The mean BMI at baseline was 43.1 kg/m2 (SD 8.0). At 2 years the mean weight change was 212.9 kg (SD 18.7), and 215 (53.0%) patients had at least a 5% decrease in weight (39.4%
www.obesityjournal.org
Original Article
Obesity
EPIDEMIOLOGY/GENETICS
TABLE 1 Sample baseline characteristics
Variable Age [years, mean (SD)] Weight [kg, mean (SD)] BMI [kg/m2, (SD)] Sex (females, %) Ethnicity (%) White Non-white Annual income CAD (%) 80,000 Education (%) Some high school or less High school diploma Some post-secondary Completed post-secondary Marital status (%) Married or common-law Others Smoking status (%) Current smoker Former smoker Never smoker
Total (n 5 406)
Average number of chronic conditions (mean, SD)
46.1 (9.5) 131.5 (25.0) 43.1 (9.0) 88.2
4.8 (2.2)
93.6 6.4
4.9 (2.2) 5.2 (1.7)
6.9 10.0 14.6 25.6 42.8
5.5 5.6 4.7 4.8 4.9
(2.2) (2.1) (2.3) (2.3) (2.1)
6.7 17.5 17.5 58.4
5.4 4.9 4.4 5.0
(2.7) (2.3) (2.2) (2.1)
57.1 42.9
5.0 (2.2) 4.8 (2.2)
6.9 44.6 48.5
5.9 (2.4) 4.9 (2.1) 4.6 (2.1)
Figure 1 Percentage morbidity (number of chronic conditions) by age group at baseline in APPLES. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Prevalence of specific chronic conditions The most common single chronic conditions at baseline were joint pain (72.2%), anxiety or depression (65.4%), hypertension (63.4%), dyslipidemia (60.4%), diabetes mellitus (44.6%), gastrointestinal reflux disease (35.4%), and sleep apnea (33.5%). Among persons with three chronic conditions, the most common triads were diabetes-hypertension-dyslipidemia (12.2%), dysipidemia-joint pain-anxiety/depression (10.6%), and diabetes-dyslipidemia-anxiety/depression (4.6%). The intersecting prevalence between chronic conditions is summarized in Figure 3.
Change in weight and change in multimorbidity Over 2 years, the average (mean) change in the number of chronic conditions was 20.4 (SD 1.9) and 212.9 kg (SD 18.7) for weight. At 2 years, 41.8% (n 5 209) of the patients reported a reduction in
BMI, body mass index; SD, standard deviation; CAD, Canadian dollars.
had 10% weight change). The average number of chronic conditions in the study population was higher at baseline (mean 5 4.9, SD 2.2) compared to that at 2-year follow-up (mean 5 4.5, SD 2.3), P < 0.02. At baseline the mean (SD) VAS was 57 (20), the mean VAS change over 2 years was 11.5 (21.2), and 53.5% had a clinically important 10-point or greater improvement in VAS.
Prevalence of multimorbidity At study enrollment, the proportion of patients who reported 2 and 3 chronic conditions was 95.4% and 85.8%, respectively. Patients were more likely to report higher number of chronic conditions if they were older (Figure 1). In fact, all patients aged 50 years and over reported having at least one chronic condition. Likewise, the proportion of patients who reported multiple chronic conditions was greater among patients with higher BMI (Figure 2). For instance, 46% of patients with BMI less than 40 kg/m2 reported five or more chronic conditions, compared to 55% among those with BMI 60 kg/m2 (Figure 2). At 2-year follow-up, the proportion of patients who reported 2 and 3 chronic conditions were 92.8% and 83.2%, respectively. There was a 8.2% reduction in the mean number of chronic conditions at 2 years [mean (SD) was 4.9 (2.1) versus 4.5 (2.3)].
www.obesityjournal.org
Figure 2 Percentage morbidity by BMI categories in APPLES. BMI, body mass Index. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
709
Obesity
Multimorbidity in Severely Obese Patients Agborsangaya et al.
Figure 3 Concomitant comorbidity prevalence grid at baseline in APPLES. For a given condition (listed in each row), the size of the circle indicates how often the second condition (listed in each column) is concomitantly present. (Chronic conditions are diabetes, hypertension, dyslipidemia, sleep apnea, gastroesophageal reflux disease, hypothyroidism, anxiety or depression, and osteoarthritis or chronic pain). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
the number of chronic conditions. 53.2% who lost weight also reported a reduction in number of chronic conditions compared to 35.5% among those who either did not lose or gain weight (P < 0.001). The proportion of change in number of chronic conditions by percentage weight change categories are presented in Figure 4. Having 5% weight loss was associated with increased likelihood of reporting a reduction in multimorbidity at 2-year follow-up (adjusted OR 5 1.7, 95% CI 1.1-2.7). Other covariates of interest are presented in Table 2. The patterns of reduction in specific chronic conditions differed across treatment groups. For example, in the surgery group, the top three chronic conditions that decreased in preva-
lence over follow-up were sleep apnea (43% at baseline vs. 25% at 2 years, P 5 0.001), dyslipidemia (60% vs. 47%, P 5 0.02), and anxiety or depression (59% vs. 47%, P 5 0.04); in the medically treated group anxiety or depression (69% vs. 57%, P 5 0.02) and joint pain (77% vs. 67%, P 5 0.04); and none in the wait-listed group.
Health status and multimorbidity At enrollment, the mean VAS score was lower for patients with higher number of chronic conditions at baseline; subjects with 8 or more chronic conditions had significantly lower mean VAS scores than subjects with no chronic conditions (8 conditions vs. none, difference 5 228.4 (95% CI 246.6, 210.3), P 5 0.001. The mean VAS scores declined in a direct and graded manner as the number of chronic conditions increased, ranging from 75.2 (SD 5 19.8) for those with no condition to 46.7 (SD 5 20.3) among those with 8 chronic conditions (Ptrend < 0.0001). The mean VAS score was significantly higher at 2 years (mean 5 69.0, SD 18.7) compared to baseline (mean 5 57.0, SD 20.4), P < 0.0001. At 2-year follow-up, 53.4% reported clinically important (10%) increase in VAS. Independent of weight loss, having a reduction in multimorbidity was associated with a higher likelihood of reporting clinically important improvement in health status (OR 5 2.5, 95% CI 1.6, 4.0) (Table 3).
Discussion Figure 4 Percentage change in number of chronic conditions by percent change weight categories. Dx no-change/Gain means “no change or increase in number of chronic conditions” and Dx reduction means “decrease in number of chronic conditions” over 2 years Multimorbidity reduction is higher for persons with higher weight loss. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
710
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
In summary, we found that multimorbidity was almost ubiquitously present in patients with severe obesity and reductions in multimorbidity were associated with clinically important improvements in health status. Furthermore, clinically important weight loss was associated with a greater likelihood of exhibiting a reduction in the number of chronic conditions.
www.obesityjournal.org
Original Article
Obesity
EPIDEMIOLOGY/GENETICS
TABLE 2 Predictors of reduction in multimorbidity at 2 years in
APPLES Univariate analysis Reduction in number of chronic conditions (yes/no) Percent weight change < 5% loss 5% loss Study groups Wait-listed Medical Surgical Sex Females Males Income (CAD) < 15,000 15,000-29,999 30,000-49,999 50,000-79,999 80,000
Multivariable modela
Odds ratio
95% CI
Odds ratio
95% CI
1.0 2.2
1.5-3.3
1.0 1.7
1.1-2.7
1.0 2.1 4.1
1.2-3.5 2.4-7.0
1.0 1.8 3.1
1.0-3.1 1.7-5.6
1.0 1.9
1.0-3.5
1.0 1.7
0.9-3.3
1.0 2.0 2.0 2.5 2.3
0.6-6.0 0.7-5.9 0.9-6.8 1.0-6.1
1.0 1.7 1.8 2.2 1.7
0.5-5.5 0.6-5.3 0.8-6.2 0.6-4.6
a
Adjusted for age and baseline BMI.
The prevalence of multimorbidity was 95% in this study, over five times higher than the 17% estimate for the general population of Alberta (6). Although the high prevalence in our study may to some extent reflect referral bias (whereby sicker patients are being referred for bariatric care), the impact of severe obesity in promoting the development of chronic conditions should not be discounted (5,11). An additional factor that may have raised the prevalence is that we included 20 chronic conditions in the multimorbidity calculation. An increase in the number of chronic conditions used to generate a multimorbidity index heightens the prevalence of multimorbidity (5). Nevertheless, in our sample, the presence of severe obesity was nearly synonymous with multimorbidity development. This underscores the importance of multiple chronic disease management strategies for severely obese individuals. Three of the most common chronic conditions were hypertension, dyslipidemia, and diabetes, important components of the metabolic syndrome, a clustering of metabolic and clinical conditions associated with cardiovascular disease (26). Obesity is a prominent component of conditions that define the metabolic syndrome (27) and the prevalence of metabolic syndrome rises as obesity severity increases (28). In a previous study among persons seeking obesity treatment (28), over half of the study sample met minimum criteria for metabolic syndrome. Importantly, the major components of the metabolic syndrome are all amenable to weight loss; thus, successful weight management should lead to reduced multimorbidity based upon regression of metabolic syndrome components alone (29). Over 2 years there was about a 8% reduction in number of chronic conditions across our study population. We defined clinically
www.obesityjournal.org
important weight loss as 5% reduction in baseline weight, and this degree of weight loss was associated with about a two-fold increased likelihood of multimorbidity reduction. Moreover, the reduction in number of chronic conditions was likewise associated with improvement in health status. The finding that disease remission is associated with improvement health status has been reported in patient populations such as those with depression and heart failure (30), cancers (31,32), and diabetes (33). Indeed, we found that the reduction in health status associated with reduction in number of chronic conditions is independent of weight loss. Despite its strengths, the present study has several limitations. First, most of the chronic conditions indicated are self-reported. Selfreported chronic disease status is subject to response and recall bias that tends to lead to comorbidity under-reporting (34). It is also possible that patients with no reported multimorbidity may have other unlisted chronic conditions that were not among the list of predefined conditions. This study, however, included the core chronic conditions recommended for inclusion in measures of multimorbidity (35). Second, while we documented the presence or absence of disease, we have no real measures of the “severity” of each condition and considered any one chronic condition to be equivalent to any other. Third, we assumed (and adjusted all analyses on these grounds) that all weight loss was equivalent, whether achieved by medical or surgical means. Fourth, the observational longitudinal study design provides associative evidence but does not indicate a causative relationship between variables, and reverse causality might be an issue. For example, it could be that treatment of anxiety or depression during follow-up led to increases in weight loss, rather than the converse. Finally, although we studied patients referred to a population-based bariatric program, our findings still require caution TABLE 3 Predictors of improvement in health status at 2 years in APPLES
Univariate analysis 10% or more change in health status Multimorbidity reduction No Yes Study groups Wait-listed Medical Surgical Sex Females Males Income (CAD) < 15,000 15,000-29,999 30,000-49,999 50,000-79,999 80,000
Multivariable modela
Odds ratio
95% CI
Odds ratio
95% CI
1.0 2.4
1.6-3.5
1.0 2.5
1.6-4.0
1.0 1.7 1.5
1.1-2.8 0.9-2.5
1.0 1.3 0.7
0.8-2.3 0.4-1.4
1.0 1.0
0.5-1.8
1.0 1.6
0.8-3.2
1.0 2.2 0.9 1.0 1.5
0.8-5.9 0.4-2.3 0.4-2.2 0.6-3.3
1.0 2.3 0.8 0.8 1.4
0.8-6.7 0.3-2.2 0.3-2.1 0.6-3.3
a
Adjusted for age, weight change, and baseline BMI.
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
711
Obesity
Multimorbidity in Severely Obese Patients Agborsangaya et al.
when generalizing to dissimilar health care jurisdictions and settings. In conclusion, we found a very high burden of multimorbidity in patients with severe obesity, and the severity of multimorbidity was associated with a reduction in health status in a direct and graded manner. Over 2 years of follow-up, we were also able to determine that clinically important weight loss was associated with a reduction in multimorbidity, and that multimorbidity reduction is associated with improvement in health status. We feel that these findings increase overall understanding of multimorbidity in this patient population and reinforce the importance of weight reduction as one of the means to reduce the burden of multimorbidity and improve health status in these high-risk patients. O C 2015 The Obesity Society V
References 1. Fortin M, Bravo G, Hudon C, Vanasse A, Lapointe L. Prevalence of multimorbidity among adults seen in family practice. Ann Fam Med 2005;3:223-228. 2. Condelius A, Edberg AK, Jakobsson U, Hallberg IR. Hospital admissions among people 651 related to multimorbidity, municipal and outpatient care. Arch Gerontol Geriatr 2008;46:6. 3. Fortin M, Bravo G, Hudon C, et al. Relationship between multimorbidity and health-related quality of life of patients in primary care. Qual Life Res 2006;15:8391. 4. Tooth L, Hockey R, Byles J, Dobson A. Weighted multimorbidity indexes predicted mortality, health service use, and health-related quality of life in older women. J Clin Epidemiol 2008;61:151-159. 5. Fortin M, Hudon C, Haggerty J, Akker M, Almirall J. Prevalence estimates of multimorbidity: A comparative study of two sources. BMC Health Serv Res 2010; 10:1. 6. Agborsangaya CB, Lau D, Lahtinen M, Cooke T, Johnson JA. Multimorbidity prevalence and patterns across socioeconomic determinants: A cross-sectional survey. BMC Public Health 2012;12:201. 7. Britt HC, Harrison CM, Miller GC, Knox SA. Prevalence and patterns of multimorbidity in Australia. Med J Aust 2008;189:72-77. 8. Haslam DW, James WP. Obesity. Lancet 2005;366:1197-1209. 9. Barnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: A cross-sectional study. Lancet 2012;380:37-43. 10. Wang F, Xu S, Shen X, Guo X, Shen R. Epidemiology of multimorbidity. Lancet 2012;380:1382-1383; author reply 3-4. 11. Sailer D. Obesity: Entrance port to multimorbidity. Wien Med Wochenschr 1998; 148:388-392. 12. Dong HJ, Unosson M, Wressle E, Marcusson J. Health consequences associated with being overweight or obese: A Swedish population-based study of 85-year-olds. J Am Geriatr Soc 2012;60:243-250. 13. Shields M, Carroll MD, Ogden CL. Adult obesity prevalence in Canada and the United States. NCHS Data Brief 2011:1-8. 14. Hensrud DD, Klein S. Extreme obesity: A new medical crisis in the United States. Mayo Clin Proc 2006;81:S5-10.
712
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
15. Padwal RS, Majumdar SR, Klarenbach S, et al. The Alberta population-based prospective evaluation of the quality of life outcomes and economic impact of bariatric surgery (APPLES) study: Background, design and rationale. BMC Health Serv Res 2010;10:284. 16. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. Jama 2014;311:806-814. 17. Chang SH, Stoll CR, Song J, Varela JE, Eagon CJ, Colditz GA. The effectiveness and risks of bariatric surgery: An updated systematic review and meta-analysis, 2003-2012. JAMA Surg 2014;149:275-287. 18. Courcoulas AP, Christian NJ, Belle SH, et al. Weight change and health outcomes at 3 years after bariatric surgery among individuals with severe obesity. Jama 2013; 310:2416-2425. 19. Bolen SD, Chang HY, Weiner JP, et al. Clinical outcomes after bariatric surgery: A five-year matched cohort analysis in seven US states. Obes Surg 2012;22:749-763. 20. Flores L, Vidal J, Nunez I, Rueda S, Viaplana J, Esmatjes E. Longitudinal changes of blood pressure after weight loss: factors involved. Surg Obes Relat Dis 2014. 21. Aminian A, Brethauer SA, Daigle CR, et al. Outcomes of bariatric surgery in type 2 diabetic patients with diminished pancreatic secretory reserve. Acta Diabetol 2014;51:1077-1079. 22. Padwal RS, Rueda-Clausen CF, Sharma AM, et al. Weight loss and outcomes in wait-listed, medically managed, and surgically treated patients enrolled in a population-based bariatric program: Prospective cohort study. Med Care 2014;52: 208-215. 23. Luo N, Johnson J, Coons SJ. Using instrument-defined health state transitions to estimate minimally important differences for four preference-based health-related quality of life instruments. Med Care 2010;48:365-371. 24. Lau DC, Douketis JD, Morrison KM, Hramiak IM, Sharma AM, Ur E. 2006 Canadian clinical practice guidelines on the management and prevention of obesity in adults and children [summary]. Cmaj 2007;176:S1-13. 25. Stevens J, Truesdale KP, McClain JE, Cai J. The definition of weight maintenance. Int J Obes (Lond) 2006;30:391-399. 26. Galassi A, Reynolds K, He J. Metabolic syndrome and risk of cardiovascular disease: a meta-analysis. Am J Med 2006;119:812-819. 27. Maison P, Byrne CD, Hales CN, Day NE, Wareham NJ. Do different dimensions of the metabolic syndrome change together over time? Evidence supporting obesity as the central feature. Diabetes Care 2001;24:1758-1763. 28. Marchesini G, Melchionda N, Apolone G, et al. The metabolic syndrome in treatment-seeking obese persons. Metabolism 2004;53:435-440. 29. Pi-Sunyer X, Blackburn G, Brancati FL, et al. Reduction in weight and cardiovascular disease risk factors in individuals with type 2 diabetes: One-year results of the look AHEAD trial. Diabetes Care 2007;30:1374-1383. 30. Xiong GL, Fiuzat M, Kuchibhatla M, Krishnan R, O’Connor CM, Jiang W. Health status and depression remission in patients with chronic heart failure: Patientreported outcomes from the SADHART-CHF trial. Circ Heart Fail 2012;5:688-692. 31. Ramsey SD, Berry K, Moinpour C, Giedzinska A, Andersen MR. Quality of life in long term survivors of colorectal cancer. Am J Gastroenterol 2002;97:1228-1234. 32. Ohno S, Sumiyoshi Y, Hashine K, Shirato A, Kyo S, Inoue M. Quality of life improvements among cancer patients in remission following the consumption of Agaricus blazei Murill mushroom extract. Complement Ther Med 2013;21:460-467. 33. Weiner S, Neugehauer EA. Quality of life of diabetic patients with medical or surgical treatment. Nutr Hosp 2013;28(Suppl 2):66-77. 34. Gross R, Bentur N, Elhayany A, Sherf M, Epstein L. The validity of self-reports on chronic disease: Characteristics of underreporters and implications for the planning of services. Public Health Rev 1996;24:17. 35. Diederichs C, Berger K, Bartels DB. The measurement of multiple chronic diseases–a systematic review on existing multimorbidity indices. J Gerontol A Biol Sci Med Sci 2011;66:301-311.
www.obesityjournal.org
Original Article EPIDEMIOLOGY/GENETICS
Multimorbidity in a Prospective Cohort: Prevalence and Associations with Weight Loss and Health Status in Severely Obese Patients Calypse B. Agborsangaya1,2, Sumit R. Majumdar1,2, Arya M. Sharma1,2, Edward W. Gregg3, and Raj S. Padwal1,2
Objective: To examine the prevalence of multimorbidity (2 chronic conditions) in severely obese patients and its associations with weight loss and health status over 2 years. Methods: In a prospective cohort including 500 severely obese adults, self-reported prevalence of 20 chronic conditions was calculated at baseline and 2 years. Multivariable logistic regression models were fitted to test the covariate-adjusted associations between 5% weight reduction and reduction in multimorbidity and the association between health status (visual analogue scale [VAS]) and reduction in multimorbidity over 2 years. Results: After 2 years, mean weight change was 212.9 618.7 kg, 53% had 5% weight reduction, mean change in VAS was 11.5 6 21.2, and 53.5% had 10% increase in VAS. Multimorbidity was reported in 95.4% and 92.8% patients at baseline and 2 years, respectively. Weight loss (5%) over 2 years was associated with reduction in multimorbidity (adjusted OR 5 1.7, 95% CI 1.1-2.7). Reduction in multimorbidity was associated with clinically important improvements (10% increase in VAS) in health status (adjusted OR 5 2.5, 95% CI 1.6, 4.0). Conclusions: Multimorbidity is common in severely obese patients. Having 5% weight reduction over 2 years was associated with a reduction in multimorbidity, which was also associated with improvements in health status. Obesity (2015) 23, 707–712. doi:10.1002/oby.21008
Introduction Multimorbidity, defined as the concurrent occurrence of two or more chronic conditions in an individual, is increasingly common (1). It has been linked to adverse health outcomes such as frequent and longer hospitalization, higher healthcare costs, readmission, reduced quality of life, and mortality (2-4). The prevalence of multimorbidity varies markedly depending on the patient population studied, ranging from about 17% to 90% (1,5-7). This variation in prevalence results in part from the absence of a single, standardized definition. In particular, the inclusion of obesity in the list of chronic conditions increases multimorbidity prevalence substantially because obesity is very common, affecting one-quarter to one-third of individuals in developed nations (8). Obesity is also an important propagator of multimorbidity because it leads to the development of numerous medical complications (9-15). Individuals with severe obesity (BMI 40 kg/m2) are at increased
risk of developing these complications (14). Severe obesity is common, affecting 3% of adult Canadians and 6% of adult Americans (13) and is growing rapidly (16). Despite these well-established links between obesity and the development of comorbidities, few data exist characterizing multimorbidity in patients with severe obesity or the response of multimorbidity to weight loss (17). Weight loss in patients with severe obesity may lead to chronic disease remission (18,19). In a recent study, Courcoulas and colleagues (18) investigated the effect of surgical interventions on the occurrence of chronic conditions in patients with severe obesity after a 3-year follow-up and noted up to 68% partial remission in patients with diabetes, 62% in patients with dyslipidemia and 38% in patients with hypertension. Despite the overarching effect of excess weight on chronic multimorbidity, this study examined only three chronic conditions. In fact, most studies evaluating the effect of weight loss are limited to a single or few health outcomes (20,21).
1 Department of Medicine and Dentistry, University of Alberta, Edmonton, Alberta, Canada. Correspondence: Calypse B. Agborsangaya (agborsan@ ualberta.ca) 2 Alberta Diabetes Institute, Edmonton, Alberta, Canada 3 Division of Diabetes Translation, Centers for Disease Control and Prevention, Atlanta, Georgia, USA.
Disclosure: The authors declared no conflict of interest. Author contributions: SRM, AMS, EWG, ACB, and RSP conceived the study and design. CBA completed the analysis, drafted the first manuscript, and completed manuscript edits. SRM, AMS, EWG, and RSP also contributed to manuscript editing, and along with ACB, had final approval of the submitted and published versions. Received: 27 June 2014; Accepted: 23 November 2014; Published online 6 February 2015. doi:10.1002/oby.21008
www.obesityjournal.org
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
707
Obesity
Multimorbidity in Severely Obese Patients Agborsangaya et al.
Accordingly, the objectives of this study were first, to characterize multimorbidity at baseline, including its prevalence and most common comorbid conditions in a prospective cohort of 500 severely obese patients recruited from a population-based regional bariatric program and followed for 2 years Second, to determine the association between change in weight and change in multimorbidity status. Third, we studied the association between change in multimorbidity and change in health status. We hypothesized that weight loss over 2 years would be associated with a reduction in multimorbidity, and that reductions in multimorbidity would be associated improvement in health status.
Methods Setting and subjects We analyzed data from the Alberta Population-based Prospective Evaluation of the Quality of Life Outcomes and Economic Impact of Bariatric Surgery (APPLES) study, a 500-patient cohort study in which consecutive, consenting adults with BMI levels 35 kg/m2 were recruited from the Adult Edmonton Weight Wise regional bariatric program. A description of this study has been published in a detailed study protocol, approved by the University of Alberta Health Research Ethics Board (Pro00003594) (15). Briefly, Edmonton Weight Wise is a comprehensive bariatric program structured such that patients progress from a wait list to intensive medical management to bariatric surgery (in interested and eligible candidates). The 500 patients enrolled were between 18 and 60 years old and were either wait-listed (n 5 150), beginning intensive medical treatment (n 5 200) or had just been approved for bariatric surgery (n 5 150). The patient characteristics and main study findings are published elsewhere; percentage weight reductions over 2 years were 1%, 3%, and 16% for wait-listed, medically treated and surgical-treated groups, respectively (22). For the present study, our objective was to characterize multimorbidity in the entire study population; thus, all 500 patients were combined and initial study group (wait-list, medical, surgical) was controlled for in the analysis (as detailed below).
Measurement and data collection Detailed case report forms have been previously published (15). Baseline data collection included socio-demographic characteristics and medical history. Participants’ body weight was measured using a calibrated bariatric scale and recorded to the nearest 0.1kg. Height was measured to the nearest 0.1 cm using a wall-mounted stadiometer. Finally, health status was measured at all study time points using a visual analogue scale (VAS), a component of EuroQol’s EQ-5D quality of life instrument that ranges from 0 to 100, with 0 representing death and 100, a state of complete health; previous work has established that the minimal important difference on this previously validated scale is 10 (23).
Multimorbidity definitions Multimorbidity is commonly defined as the presence of more than one chronic condition in the same individual (1). For purposes of comparison, an alternative definition (defined as 3 chronic conditions) used in the Canadian Community Health Survey (CCHS) (5) was also examined. The number of chronic conditions per patient was determined by a simple count of the following 20 self-reported chronic conditions: anxiety or depression, asthma, bipolar disorder,
708
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
chronic kidney disease, chronic urinal/fecal incontinence, diabetes, dyslipidemia, fatty liver disease, fibromyalgia, gastrointestinal reflux, heart failure, hypertension, hypothyroidism, obesity hypoventilation, lymphedema, joint pain (on the basis of osteoarthritis), polycystic ovary syndrome, sleep apnea, stroke, and venous stasis (6). Three conditions (hypertension, diabetes, and dyslipidemia) were defined based on self-report and clinical parameters. Hypertension was considered present if self-reported blood pressure levels were 140/90 mm Hg (130 mm Hg in patients with diabetes), or if antihypertensive medications were prescribed. An A1c over 6.5% or a fasting glucose 7.0 mmol/L were used as diagnostic criteria for diabetes. Dyslipidemia was considered present if one of the following conditions was met: total cholesterol 6.2 mmol/L, low-density lipoprotein (LDL) cholesterol 4.1 mmol/L, high-density lipoprotein (HDL) cholesterol < 1.0 mmol/L, or a triglyceride 2.3 mmol/L.
Statistical analysis Descriptive analyses were first conducted, including calculation of means, medians, and proportions. Mean VAS scores were calculated according to the number of baseline chronic conditions (categorized from 0 to 8). ANOVA was used to compare the mean VAS score in subjects with different numbers of comorbidities (ranging from 1 to 8) to the VAS scores in subjects with no comorbidity. To examine predictors of multimorbidity reduction over the followup period, we grouped subjects into binary categories according to whether or not their number of chronic conditions was reduced over 2 years. Binary logistic regression was used to examine predictors of a reduced number of chronic conditions over the follow-up period (yes/no). The major predictor variable of interest was 2-year 5% change in weight, classified as a binary variable (5% vs. > 5%). Greater than a 5% reduction in weight was chosen because it is thought to be clinically relevant threshold indicating change in disease status (24,25). In a sensitivity analysis, we tested the association between 7% weight loss and multimorbidity reduction and noted no significant change in our findings. To evaluate the association between change in multimorbidity and change in health status, VAS was dichotomized to indicate clinically important improvement ( 10% versus < 10%). All models were adjusted for age, sex, income status and baseline BMI. Because we wanted to isolate the effect of weight loss on change in number of chronic conditions, analysis was performed in the combined sample while adjusting for treatment status (waitlist vs medical vs surgical). No adjustments for multiple testing were performed and a P-value < 0.05 was considered statistically significant. At 2 years, analyses were restricted to 406 patients with evaluable data (i.e., a completers’ analysis). All analyses were performed using STATA software version 12 (Stata, TX, USA).
Results Of 500 enrolled patients, 406 (81%) had complete baseline and 2-year weight data. The average baseline weight did not differ for completers (mean 5 131.5 kg, SD 25.0) and noncompleters (mean 5 133.9 kg, SD 25.3), P 5 0.4. The socio-demographic characteristics of the 406 completers are presented in Table 1. The majority was female (88.2%) and white (93.6%), with a mean age of 46.1 years (SD 9.5). The mean BMI at baseline was 43.1 kg/m2 (SD 8.0). At 2 years the mean weight change was 212.9 kg (SD 18.7), and 215 (53.0%) patients had at least a 5% decrease in weight (39.4%
www.obesityjournal.org
Original Article
Obesity
EPIDEMIOLOGY/GENETICS
TABLE 1 Sample baseline characteristics
Variable Age [years, mean (SD)] Weight [kg, mean (SD)] BMI [kg/m2, (SD)] Sex (females, %) Ethnicity (%) White Non-white Annual income CAD (%) 80,000 Education (%) Some high school or less High school diploma Some post-secondary Completed post-secondary Marital status (%) Married or common-law Others Smoking status (%) Current smoker Former smoker Never smoker
Total (n 5 406)
Average number of chronic conditions (mean, SD)
46.1 (9.5) 131.5 (25.0) 43.1 (9.0) 88.2
4.8 (2.2)
93.6 6.4
4.9 (2.2) 5.2 (1.7)
6.9 10.0 14.6 25.6 42.8
5.5 5.6 4.7 4.8 4.9
(2.2) (2.1) (2.3) (2.3) (2.1)
6.7 17.5 17.5 58.4
5.4 4.9 4.4 5.0
(2.7) (2.3) (2.2) (2.1)
57.1 42.9
5.0 (2.2) 4.8 (2.2)
6.9 44.6 48.5
5.9 (2.4) 4.9 (2.1) 4.6 (2.1)
Figure 1 Percentage morbidity (number of chronic conditions) by age group at baseline in APPLES. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Prevalence of specific chronic conditions The most common single chronic conditions at baseline were joint pain (72.2%), anxiety or depression (65.4%), hypertension (63.4%), dyslipidemia (60.4%), diabetes mellitus (44.6%), gastrointestinal reflux disease (35.4%), and sleep apnea (33.5%). Among persons with three chronic conditions, the most common triads were diabetes-hypertension-dyslipidemia (12.2%), dysipidemia-joint pain-anxiety/depression (10.6%), and diabetes-dyslipidemia-anxiety/depression (4.6%). The intersecting prevalence between chronic conditions is summarized in Figure 3.
Change in weight and change in multimorbidity Over 2 years, the average (mean) change in the number of chronic conditions was 20.4 (SD 1.9) and 212.9 kg (SD 18.7) for weight. At 2 years, 41.8% (n 5 209) of the patients reported a reduction in
BMI, body mass index; SD, standard deviation; CAD, Canadian dollars.
had 10% weight change). The average number of chronic conditions in the study population was higher at baseline (mean 5 4.9, SD 2.2) compared to that at 2-year follow-up (mean 5 4.5, SD 2.3), P < 0.02. At baseline the mean (SD) VAS was 57 (20), the mean VAS change over 2 years was 11.5 (21.2), and 53.5% had a clinically important 10-point or greater improvement in VAS.
Prevalence of multimorbidity At study enrollment, the proportion of patients who reported 2 and 3 chronic conditions was 95.4% and 85.8%, respectively. Patients were more likely to report higher number of chronic conditions if they were older (Figure 1). In fact, all patients aged 50 years and over reported having at least one chronic condition. Likewise, the proportion of patients who reported multiple chronic conditions was greater among patients with higher BMI (Figure 2). For instance, 46% of patients with BMI less than 40 kg/m2 reported five or more chronic conditions, compared to 55% among those with BMI 60 kg/m2 (Figure 2). At 2-year follow-up, the proportion of patients who reported 2 and 3 chronic conditions were 92.8% and 83.2%, respectively. There was a 8.2% reduction in the mean number of chronic conditions at 2 years [mean (SD) was 4.9 (2.1) versus 4.5 (2.3)].
www.obesityjournal.org
Figure 2 Percentage morbidity by BMI categories in APPLES. BMI, body mass Index. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
709
Obesity
Multimorbidity in Severely Obese Patients Agborsangaya et al.
Figure 3 Concomitant comorbidity prevalence grid at baseline in APPLES. For a given condition (listed in each row), the size of the circle indicates how often the second condition (listed in each column) is concomitantly present. (Chronic conditions are diabetes, hypertension, dyslipidemia, sleep apnea, gastroesophageal reflux disease, hypothyroidism, anxiety or depression, and osteoarthritis or chronic pain). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
the number of chronic conditions. 53.2% who lost weight also reported a reduction in number of chronic conditions compared to 35.5% among those who either did not lose or gain weight (P < 0.001). The proportion of change in number of chronic conditions by percentage weight change categories are presented in Figure 4. Having 5% weight loss was associated with increased likelihood of reporting a reduction in multimorbidity at 2-year follow-up (adjusted OR 5 1.7, 95% CI 1.1-2.7). Other covariates of interest are presented in Table 2. The patterns of reduction in specific chronic conditions differed across treatment groups. For example, in the surgery group, the top three chronic conditions that decreased in preva-
lence over follow-up were sleep apnea (43% at baseline vs. 25% at 2 years, P 5 0.001), dyslipidemia (60% vs. 47%, P 5 0.02), and anxiety or depression (59% vs. 47%, P 5 0.04); in the medically treated group anxiety or depression (69% vs. 57%, P 5 0.02) and joint pain (77% vs. 67%, P 5 0.04); and none in the wait-listed group.
Health status and multimorbidity At enrollment, the mean VAS score was lower for patients with higher number of chronic conditions at baseline; subjects with 8 or more chronic conditions had significantly lower mean VAS scores than subjects with no chronic conditions (8 conditions vs. none, difference 5 228.4 (95% CI 246.6, 210.3), P 5 0.001. The mean VAS scores declined in a direct and graded manner as the number of chronic conditions increased, ranging from 75.2 (SD 5 19.8) for those with no condition to 46.7 (SD 5 20.3) among those with 8 chronic conditions (Ptrend < 0.0001). The mean VAS score was significantly higher at 2 years (mean 5 69.0, SD 18.7) compared to baseline (mean 5 57.0, SD 20.4), P < 0.0001. At 2-year follow-up, 53.4% reported clinically important (10%) increase in VAS. Independent of weight loss, having a reduction in multimorbidity was associated with a higher likelihood of reporting clinically important improvement in health status (OR 5 2.5, 95% CI 1.6, 4.0) (Table 3).
Discussion Figure 4 Percentage change in number of chronic conditions by percent change weight categories. Dx no-change/Gain means “no change or increase in number of chronic conditions” and Dx reduction means “decrease in number of chronic conditions” over 2 years Multimorbidity reduction is higher for persons with higher weight loss. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
710
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
In summary, we found that multimorbidity was almost ubiquitously present in patients with severe obesity and reductions in multimorbidity were associated with clinically important improvements in health status. Furthermore, clinically important weight loss was associated with a greater likelihood of exhibiting a reduction in the number of chronic conditions.
www.obesityjournal.org
Original Article
Obesity
EPIDEMIOLOGY/GENETICS
TABLE 2 Predictors of reduction in multimorbidity at 2 years in
APPLES Univariate analysis Reduction in number of chronic conditions (yes/no) Percent weight change < 5% loss 5% loss Study groups Wait-listed Medical Surgical Sex Females Males Income (CAD) < 15,000 15,000-29,999 30,000-49,999 50,000-79,999 80,000
Multivariable modela
Odds ratio
95% CI
Odds ratio
95% CI
1.0 2.2
1.5-3.3
1.0 1.7
1.1-2.7
1.0 2.1 4.1
1.2-3.5 2.4-7.0
1.0 1.8 3.1
1.0-3.1 1.7-5.6
1.0 1.9
1.0-3.5
1.0 1.7
0.9-3.3
1.0 2.0 2.0 2.5 2.3
0.6-6.0 0.7-5.9 0.9-6.8 1.0-6.1
1.0 1.7 1.8 2.2 1.7
0.5-5.5 0.6-5.3 0.8-6.2 0.6-4.6
a
Adjusted for age and baseline BMI.
The prevalence of multimorbidity was 95% in this study, over five times higher than the 17% estimate for the general population of Alberta (6). Although the high prevalence in our study may to some extent reflect referral bias (whereby sicker patients are being referred for bariatric care), the impact of severe obesity in promoting the development of chronic conditions should not be discounted (5,11). An additional factor that may have raised the prevalence is that we included 20 chronic conditions in the multimorbidity calculation. An increase in the number of chronic conditions used to generate a multimorbidity index heightens the prevalence of multimorbidity (5). Nevertheless, in our sample, the presence of severe obesity was nearly synonymous with multimorbidity development. This underscores the importance of multiple chronic disease management strategies for severely obese individuals. Three of the most common chronic conditions were hypertension, dyslipidemia, and diabetes, important components of the metabolic syndrome, a clustering of metabolic and clinical conditions associated with cardiovascular disease (26). Obesity is a prominent component of conditions that define the metabolic syndrome (27) and the prevalence of metabolic syndrome rises as obesity severity increases (28). In a previous study among persons seeking obesity treatment (28), over half of the study sample met minimum criteria for metabolic syndrome. Importantly, the major components of the metabolic syndrome are all amenable to weight loss; thus, successful weight management should lead to reduced multimorbidity based upon regression of metabolic syndrome components alone (29). Over 2 years there was about a 8% reduction in number of chronic conditions across our study population. We defined clinically
www.obesityjournal.org
important weight loss as 5% reduction in baseline weight, and this degree of weight loss was associated with about a two-fold increased likelihood of multimorbidity reduction. Moreover, the reduction in number of chronic conditions was likewise associated with improvement in health status. The finding that disease remission is associated with improvement health status has been reported in patient populations such as those with depression and heart failure (30), cancers (31,32), and diabetes (33). Indeed, we found that the reduction in health status associated with reduction in number of chronic conditions is independent of weight loss. Despite its strengths, the present study has several limitations. First, most of the chronic conditions indicated are self-reported. Selfreported chronic disease status is subject to response and recall bias that tends to lead to comorbidity under-reporting (34). It is also possible that patients with no reported multimorbidity may have other unlisted chronic conditions that were not among the list of predefined conditions. This study, however, included the core chronic conditions recommended for inclusion in measures of multimorbidity (35). Second, while we documented the presence or absence of disease, we have no real measures of the “severity” of each condition and considered any one chronic condition to be equivalent to any other. Third, we assumed (and adjusted all analyses on these grounds) that all weight loss was equivalent, whether achieved by medical or surgical means. Fourth, the observational longitudinal study design provides associative evidence but does not indicate a causative relationship between variables, and reverse causality might be an issue. For example, it could be that treatment of anxiety or depression during follow-up led to increases in weight loss, rather than the converse. Finally, although we studied patients referred to a population-based bariatric program, our findings still require caution TABLE 3 Predictors of improvement in health status at 2 years in APPLES
Univariate analysis 10% or more change in health status Multimorbidity reduction No Yes Study groups Wait-listed Medical Surgical Sex Females Males Income (CAD) < 15,000 15,000-29,999 30,000-49,999 50,000-79,999 80,000
Multivariable modela
Odds ratio
95% CI
Odds ratio
95% CI
1.0 2.4
1.6-3.5
1.0 2.5
1.6-4.0
1.0 1.7 1.5
1.1-2.8 0.9-2.5
1.0 1.3 0.7
0.8-2.3 0.4-1.4
1.0 1.0
0.5-1.8
1.0 1.6
0.8-3.2
1.0 2.2 0.9 1.0 1.5
0.8-5.9 0.4-2.3 0.4-2.2 0.6-3.3
1.0 2.3 0.8 0.8 1.4
0.8-6.7 0.3-2.2 0.3-2.1 0.6-3.3
a
Adjusted for age, weight change, and baseline BMI.
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
711
Obesity
Multimorbidity in Severely Obese Patients Agborsangaya et al.
when generalizing to dissimilar health care jurisdictions and settings. In conclusion, we found a very high burden of multimorbidity in patients with severe obesity, and the severity of multimorbidity was associated with a reduction in health status in a direct and graded manner. Over 2 years of follow-up, we were also able to determine that clinically important weight loss was associated with a reduction in multimorbidity, and that multimorbidity reduction is associated with improvement in health status. We feel that these findings increase overall understanding of multimorbidity in this patient population and reinforce the importance of weight reduction as one of the means to reduce the burden of multimorbidity and improve health status in these high-risk patients. O C 2015 The Obesity Society V
References 1. Fortin M, Bravo G, Hudon C, Vanasse A, Lapointe L. Prevalence of multimorbidity among adults seen in family practice. Ann Fam Med 2005;3:223-228. 2. Condelius A, Edberg AK, Jakobsson U, Hallberg IR. Hospital admissions among people 651 related to multimorbidity, municipal and outpatient care. Arch Gerontol Geriatr 2008;46:6. 3. Fortin M, Bravo G, Hudon C, et al. Relationship between multimorbidity and health-related quality of life of patients in primary care. Qual Life Res 2006;15:8391. 4. Tooth L, Hockey R, Byles J, Dobson A. Weighted multimorbidity indexes predicted mortality, health service use, and health-related quality of life in older women. J Clin Epidemiol 2008;61:151-159. 5. Fortin M, Hudon C, Haggerty J, Akker M, Almirall J. Prevalence estimates of multimorbidity: A comparative study of two sources. BMC Health Serv Res 2010; 10:1. 6. Agborsangaya CB, Lau D, Lahtinen M, Cooke T, Johnson JA. Multimorbidity prevalence and patterns across socioeconomic determinants: A cross-sectional survey. BMC Public Health 2012;12:201. 7. Britt HC, Harrison CM, Miller GC, Knox SA. Prevalence and patterns of multimorbidity in Australia. Med J Aust 2008;189:72-77. 8. Haslam DW, James WP. Obesity. Lancet 2005;366:1197-1209. 9. Barnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: A cross-sectional study. Lancet 2012;380:37-43. 10. Wang F, Xu S, Shen X, Guo X, Shen R. Epidemiology of multimorbidity. Lancet 2012;380:1382-1383; author reply 3-4. 11. Sailer D. Obesity: Entrance port to multimorbidity. Wien Med Wochenschr 1998; 148:388-392. 12. Dong HJ, Unosson M, Wressle E, Marcusson J. Health consequences associated with being overweight or obese: A Swedish population-based study of 85-year-olds. J Am Geriatr Soc 2012;60:243-250. 13. Shields M, Carroll MD, Ogden CL. Adult obesity prevalence in Canada and the United States. NCHS Data Brief 2011:1-8. 14. Hensrud DD, Klein S. Extreme obesity: A new medical crisis in the United States. Mayo Clin Proc 2006;81:S5-10.
712
Obesity | VOLUME 23 | NUMBER 3 | MARCH 2015
15. Padwal RS, Majumdar SR, Klarenbach S, et al. The Alberta population-based prospective evaluation of the quality of life outcomes and economic impact of bariatric surgery (APPLES) study: Background, design and rationale. BMC Health Serv Res 2010;10:284. 16. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. Jama 2014;311:806-814. 17. Chang SH, Stoll CR, Song J, Varela JE, Eagon CJ, Colditz GA. The effectiveness and risks of bariatric surgery: An updated systematic review and meta-analysis, 2003-2012. JAMA Surg 2014;149:275-287. 18. Courcoulas AP, Christian NJ, Belle SH, et al. Weight change and health outcomes at 3 years after bariatric surgery among individuals with severe obesity. Jama 2013; 310:2416-2425. 19. Bolen SD, Chang HY, Weiner JP, et al. Clinical outcomes after bariatric surgery: A five-year matched cohort analysis in seven US states. Obes Surg 2012;22:749-763. 20. Flores L, Vidal J, Nunez I, Rueda S, Viaplana J, Esmatjes E. Longitudinal changes of blood pressure after weight loss: factors involved. Surg Obes Relat Dis 2014. 21. Aminian A, Brethauer SA, Daigle CR, et al. Outcomes of bariatric surgery in type 2 diabetic patients with diminished pancreatic secretory reserve. Acta Diabetol 2014;51:1077-1079. 22. Padwal RS, Rueda-Clausen CF, Sharma AM, et al. Weight loss and outcomes in wait-listed, medically managed, and surgically treated patients enrolled in a population-based bariatric program: Prospective cohort study. Med Care 2014;52: 208-215. 23. Luo N, Johnson J, Coons SJ. Using instrument-defined health state transitions to estimate minimally important differences for four preference-based health-related quality of life instruments. Med Care 2010;48:365-371. 24. Lau DC, Douketis JD, Morrison KM, Hramiak IM, Sharma AM, Ur E. 2006 Canadian clinical practice guidelines on the management and prevention of obesity in adults and children [summary]. Cmaj 2007;176:S1-13. 25. Stevens J, Truesdale KP, McClain JE, Cai J. The definition of weight maintenance. Int J Obes (Lond) 2006;30:391-399. 26. Galassi A, Reynolds K, He J. Metabolic syndrome and risk of cardiovascular disease: a meta-analysis. Am J Med 2006;119:812-819. 27. Maison P, Byrne CD, Hales CN, Day NE, Wareham NJ. Do different dimensions of the metabolic syndrome change together over time? Evidence supporting obesity as the central feature. Diabetes Care 2001;24:1758-1763. 28. Marchesini G, Melchionda N, Apolone G, et al. The metabolic syndrome in treatment-seeking obese persons. Metabolism 2004;53:435-440. 29. Pi-Sunyer X, Blackburn G, Brancati FL, et al. Reduction in weight and cardiovascular disease risk factors in individuals with type 2 diabetes: One-year results of the look AHEAD trial. Diabetes Care 2007;30:1374-1383. 30. Xiong GL, Fiuzat M, Kuchibhatla M, Krishnan R, O’Connor CM, Jiang W. Health status and depression remission in patients with chronic heart failure: Patientreported outcomes from the SADHART-CHF trial. Circ Heart Fail 2012;5:688-692. 31. Ramsey SD, Berry K, Moinpour C, Giedzinska A, Andersen MR. Quality of life in long term survivors of colorectal cancer. Am J Gastroenterol 2002;97:1228-1234. 32. Ohno S, Sumiyoshi Y, Hashine K, Shirato A, Kyo S, Inoue M. Quality of life improvements among cancer patients in remission following the consumption of Agaricus blazei Murill mushroom extract. Complement Ther Med 2013;21:460-467. 33. Weiner S, Neugehauer EA. Quality of life of diabetic patients with medical or surgical treatment. Nutr Hosp 2013;28(Suppl 2):66-77. 34. Gross R, Bentur N, Elhayany A, Sherf M, Epstein L. The validity of self-reports on chronic disease: Characteristics of underreporters and implications for the planning of services. Public Health Rev 1996;24:17. 35. Diederichs C, Berger K, Bartels DB. The measurement of multiple chronic diseases–a systematic review on existing multimorbidity indices. J Gerontol A Biol Sci Med Sci 2011;66:301-311.
www.obesityjournal.org
