PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Arch Intern Med. Author manuscript; available in PMC 2013 August 26.
Published in final edited form as:
PMCID: PMC3752594
NIHMSID: NIHMS498809

Smoking and Weight Change After New Health Diagnoses in Older Adults

Abstract

Background

Smoking and patterns of diet and activity are the 2 leading underlying causes of death in the United States, yet the factors that prompt individuals to adopt healthier habits are not well understood.

Methods

This study was undertaken to determine whether individuals who have experienced recent adverse health events are more likely to quit smoking or to lose weight than those without recent events using Health and Retirement Study panel survey data for 20 221 overweight or obese individuals younger than 75 years and 7764 smokers from 1992 to 2000.

Results

In multivariate analyses, adults with recent diagnoses of stroke, cancer, lung disease, heart disease, or diabetes mellitus were 3.2 times more likely to quit smoking than were individuals without new diagnoses (P <.001). Among overweight or obese individuals younger than 75 years, those with recent diagnoses of lung disease, heart disease, or diabetes mellitus lost −0.35 U of body mass index (calculated as weight in kilograms divided by height in meters squared) compared with those without these new diagnoses (P<.001). Smokers with multiple new diagnoses were 6 times more likely to quit smoking compared with those with no new diagnoses. The odds of quitting smoking were 5 times greater in individuals with a new diagnosis of heart disease, and body mass index declined by 0.6 U in overweight or obese individuals with a new diagnosis of diabetes mellitus (P <.001).

Conclusions

Across a range of health conditions, new diagnoses can serve as a window of opportunity that prompts older adults to change health habits, in particular, to quit smoking. Quality improvement efforts targeting secondary as well as primary prevention through the health care system are likely well founded.

Smoking and patterns of diet and activity were the 2 leading underlying contributors to deaths in 1990 and 2000.13 Although rates have declined, one-fifth of adults older than 25 years smoke. The share of the population who are overweight or obese is increasing; two-thirds of adults aged 20 to 75 years are overweight (33%) or obese (34%).4 Surgeon General reports and federal guidelines recommend weight loss and smoking cessation in overweight or obese individuals as primary and secondary preventive measures against health risks. These reports also recommend that physicians and other practitioners counsel patients about weight loss and smoking cessation.58

Until recently, efforts to target individuals for health behavior change after health events occur (secondary prevention) have been limited compared with other quality-improvement goals. For example, although studies show that smoking cessation counseling is effective, national rates of providing cessation counseling to Medicare patients hospitalized because of acute myocardial infarction declined by 3.6 percentage points between 1994–1995 and 1998–1999, from 40.8% to 37.2%, and remained unchanged in 2000–2001.9,10 In contrast, rates of providing β-blockers at discharge to patients after an acute myocardial infarction who were receiving Medicare increased by 20.4 percentage points, from 50.3% to 70.7% between 1994–1995 and 1998–1999 and increased by an additional 7.0 percentage points by 2000–2001.9,10 Among individuals trying to lose or maintain weight, 43% of obese individuals and 16% of overweight individuals reported receiving advice to lose weight from a health care professional in 2000.11 Reported rates of receiving counseling about exercise during a physician visit are also low.12

Several factors point to the potential value of promoting health behavior change after adverse health events. Secondary prevention is increasingly important because individuals are living longer after onset of acute and chronic conditions. In addition, aging of the baby boom generation will place increased demands on the health care system and contribute to rising health care spending. Secondary prevention may be particularly relevant for individuals who are middle aged or older because they are more likely to experience adverse health events.13 Furthermore, recent Medicare quality improvement and coverage policies could facilitate further improvement in secondary prevention.

This study was conducted to examine whether individuals are more receptive to health behavior change after adverse health events. The findings of several studies and review articles suggest that individuals who have experienced adverse health events are more likely to quit smoking than individuals who have not experienced adverse health events,1417 yet effects on weight change and effects of particular health conditions are less well understood. Thus, this analysis examined weight change and smoking cessation as outcomes and considered whether health-habit changes vary by number and type of new diagnosis. The study used a unique panel survey data set of middle-aged and older adults, who are likely to experience adverse health events.

METHODS

DATA AND SAMPLE

This study analyzed data from the Health and Retirement Study, a longitudinal survey of middle-aged and older individuals using data sets created by the RAND Corp (Santa Monica, California).18,19 The survey tracks 4 cohorts of middle-aged and older individuals and their spouses and collects a rich set of information about health, work, and retirement including questions about whether a physician has ever diagnosed various health conditions. The first survey was conducted in 1992 and has continued every other year since.

The smoker sample included individuals who reported smoking in the previous period. The sample for weight-change analyses included individuals younger than 75 years with a body mass index (BMI) (calculated as weight in kilograms divided by height in meters squared) of 25 or higher (overweight or obese) in the previous period. Research suggests that the relationship between BMI and health risk is approximately linear in adults before age 75 years, whereas after age 75 years, it seems to lessen and is less well understood.2022 The sample included individuals with at least 2 observations from the first survey year through 2000. The unit of observation was person-years; thus, an individual surveyed in 3 waves would account for 3 observations and 2 potential health-habit changes.

OUTCOMES AND NEW DIAGNOSES

The quit-smoking outcome was defined as yes if smokers reported that they had quit smoking since the previous period. The weight-change outcome was measured as change in units of BMI.

Health conditions were selected separately for smoking and weight-change analyses. For smokers, any new diagnosis refers to whether a smoker received 1 or more of the following new diagnoses since the last survey wave: stroke, cancer, lung disease (eg, chronic obstructive pulmonary disease), heart disease (eg, angina, congestive heart failure, myocardial infarction, or other heart condition), or diabetes mellitus (DM). Smoking is a risk factor, and smoking cessation is recommended as secondary prevention of all of these diagnoses.

For BMI change, diagnoses included DM, lung disease, and heart disease. Being overweight is a risk factor for DM and heart disease but not lung disease, and weight loss is recommended as secondary prevention of these 2 conditions. Cancer and stroke were omitted from the BMI change analyses. Although overweight is associated with increased cancer risk,23 cancer was omitted because these data showed that individuals with normal weight and those who were overweight lost weight after cancer diagnosis, which suggests that weight loss may be a consequence of cancer or its treatment rather than an attempt to achieve a recommended weight. No other diagnoses were significantly associated with weight loss in individuals with appropriate weight. However, stroke was also excluded from BMI change analyses to rule out the possibility that weight reduction in overweight and obese individuals was due to unobserved differences in severity of stroke. This exclusion resulted in more conservative estimates.

ANALYSIS

The analysis compared health-habit changes in individuals during 2 years encompassing a new health diagnosis compared with changes in individuals with no new diagnoses. Health-habit changes were assessed relative to changes in individuals without these diagnoses to account for any secular changes in the population at-large. The quit-smoking regression analysis was run as a logit, with coefficients displayed as odds ratios. Regression in change in BMI was performed using the least squares method. Separate regression analyses were performed for each health-habit change and type of new diagnosis, as described (see the “Outcomes and New Diagnoses” subsection of the “Methods” section). In all analyses, standard errors were clustered to account for correlation across multiple observations in individuals in the sample. Estimates were weighted using survey weights for the first observation for each person. Analyses were performed using commercially available software (STATA, version 8.2; SPSS, Inc, Chicago, Illinois) by using survey commands to account for survey design.

The smoking regression analysis included controls for age, sex, race/ethnicity, educational status, BMI, health status, marital status (including separate controls for marriage to a non-smoker or a smoker), whether currently employed, government or job-based health insurance, quartile of income (total household income including, eg, wages, government transfers, and pensions), and wealth (net value of household assets including, eg, real estate and savings). The BMI-change regression analysis included age, sex, race/ethnicity, educational status, BMI,5 health status, marital status (including separate controls for marriage to a spouse who is overweight or obese or who has normal weight), smoking status, whether currently employed, government or job-based health insurance, quartile of income, and wealth. Body mass index, health status, income, and wealth controls reflect characteristics before the diagnosis. Adding controls for census region and for the number of physician visits in the previous period did not alter the results of smoking or BMI-change analyses (data not shown).

RESULTS

HEALTH-HABIT CHANGES AND NEW DIAGNOSES

Overall, 18% of smokers quit, and BMI increased by 0.04 U in overweight or obese adults younger than 75 years (Table 1). Approximately 13% of smokers received at least 1 of the 5 diagnoses (stroke, cancer, lung disease, heart disease, and DM), and 8% of overweight individuals received 1 of 3 new diagnoses (excluding stroke and cancer).

Table 1
Descriptive Statistics

PREDICTORS OF HEALTH-HABIT CHANGES

Individuals with new diagnoses were more likely to adopt healthier habits than those without recent new diagnoses (Table 2). The odds of smoking cessation were 3.2 times greater in individuals who received at least 1 of the 5 new diagnoses compared with individuals with no new diagnoses (P <.001). Compared with individuals married to nonsmokers, those married to smokers were half as likely to quit smoking (P <.001). Non-Hispanic white individuals were 20% less likely to quit smoking than were African American or Hispanic individuals (P <.05). Individuals whose previous health status was excellent or very good were 32% more likely to quit smoking (P <.05). Odds of smoking cessation were significantly greater for individuals older than 70 years compared with individuals aged 50 to 55 years. Body mass index was not significantly associated with smoking cessation. In addition, sex, educational status, income, and wealth were insignificantly associated with smoking cessation or BMI change (results not shown).

Table 2
Health-Habit Changes and New Health Diagnosis

Overweight or obese individuals younger than 75 years with 1 or more new diagnoses of DM, lung disease, or heart disease lost an additional −0.35 U of BMI (2–3 lb [to convert pounds to kilograms, multiply by 0.45]) compared with overweight or obese individuals younger than 75 years with no new diagnoses (P <.001; Table 2). Among individuals with overweight or obese spouses or who were not married, BMI change was 0.18 and 0.25 U greater, respectively, compared with individuals married to spouses with normal weight (P <.001). Body mass index change was 0.16 U greater in individuals in good vs fair or poor previous health status (P <.05). Compared with nonsmokers, BMI change was 0.47 U greater in individuals who quit smoking during the period and was −0.18 U lower in smokers. Being obese (vs overweight) was associated with a reduction in BMI of −0.31 U (P <.001). Reduction in BMI was greater in individuals older than 65 years compared with individuals aged 50 to 55 years. Race/ethnicity was not significantly associated with smoking cessation. In addition, sex, educational status, income, and wealth were also insignificantly associated with quitting smoking (results not shown).

HEALTH-HABIT CHANGES BY NUMBER AND TYPE OF NEW DIAGNOSIS

Multiple diagnoses, as opposed to 1 new diagnosis, were associated with a greater magnitude of health-habit changes (Table 3). Compared with individuals with no new diagnoses, the odds of smoking cessation in individuals with 1 new diagnosis were 2.9 times greater and in those with multiple new diagnoses were 6.1 times greater (P <.001). One new diagnosis was associated with a reduction in BMI of −0.34 U in overweight and obese individuals (P <.001). Multiple new diagnoses were associated with a reduction in BMI of −0.64 U compared with no new diagnoses; however, this result was not statistically significant. These regression analyses included the same control variables as given in Table 2; coefficients were similar and are not shown.

Table 3
Health-Habit Changes With 1 or Multiple New Health Diagnoses

The odds of smoking cessation varied by new diagnosis (Table 4). Compared with individuals with no new diagnoses, the odds of smoking cessation were 2 to 5 times greater in individuals with a new diagnosis of stroke (odds ratio, 4.3), cancer (4.3), lung disease (2.3), or heart disease (5.2) (all P <.001). Individuals with a new diagnosis of DM were 70% more likely to quit smoking compared with individuals with no new diagnoses (P <.05). For BMI change, the magnitude of reduction associated with a new diagnosis of heart disease was −0.27 U of BMI (P <.05). A new diagnosis of DM was associated with a reduction in BMI of −0.60 U (P <.001); a new diagnosis of lung disease was not significantly associated with BMI change.

Table 4
Health-Habit Changes by Type of New Diagnosis

COMMENT

The results of the present study suggest that across a range of acute and chronic conditions, new diagnoses can serve as a window of opportunity that prompts individuals to change health habits, in particular, to quit smoking. Health diagnoses were associated with weight reduction and smoking cessation in adults who were middle aged or older. Changes were particularly pronounced in smokers with stroke, cancer, or heart disease and in overweight individuals with DM. Targeting individuals with recent new diagnoses may be particularly effective in middle-aged and older individuals, who are increasingly likely to receive a major diagnosis or to be hospitalized as they age. Individuals with new adverse health events are accessible through contact with the health care system or through the Internet or other written information about their disease, and this study suggests that they are more motivated to change health habits.

A disproportionate share of smokers with adverse health events quit smoking, which suggests that they may be particularly receptive to cessation interventions at the time of an adverse health event. Smokers with 1 of the 5 included diagnoses or who were hospitalized because of any reason accounted for 31% of smokers but 50% of smokers who quit smoking (Figure). Still, considerable opportunity remains to reduce smoking in older adults. If all of the 31% of smokers with these health events were to quit smoking, the overall share of older smokers who quit smoking would increase from 18% to 40%.

Figure
Share of smokers with and without adverse health events who quit smoking, calculated from Health and Retirement Survey panel data.18 New diagnoses included stroke, cancer, lung disease, heart disease, and diabetes mellitus.

Similarly, overweight and obese individuals may be motivated to attempt weight loss after receiving a new diagnosis, yet there is substantial potential for further reductions. Individuals with new diagnoses of lung disease, heart disease, or DM lost, on average, −0.31 U of BMI, whereas BMI increased slightly by 0.06 U in those without new diagnoses (data not shown). Further weight reductions could be needed to be clinically significant in lowering future health risks. For example, one study cites weight reduction of 5% to 10% as clinically meaningful.22 In additional analyses, individuals with diagnoses of DM, lung disease, or heart disease were not significantly more likely to lose 5% or 10% of body weight compared with individuals without these conditions (P >.10; data not shown).

The results suggest that quality-improvement measures that target individuals at the time of an adverse health event are likely to be well founded. Medicare hospital public reporting includes measures of whether smoking cessation counseling is provided to patients with heart failure, acute myocardial infarction, or pneumonia.24 Guidelines and evidence reviews recommend smoking cessation counseling and pharmacotherapy for older smokers and conclude that older smokers derive health benefits from smoking cessation.7,25 Several studies find smoking cessation treatments to be cost-effective.26,27 Similarly, counseling about diet and exercise is recommended in overweight and obese individuals.5 Previous studies found that rates of discussing weight loss or smoking cessation in clinical settings are low,911,28 although a recent study found improvement in rates of cessation counseling in patients with heart failure, acute myocardial infarction, or pneumonia in hospitals accredited by The Joint Commission.29

Recent Medicare coverage policies also focus on secondary prevention from adopting healthier habits and, thus, could facilitate further improvement during this potential window of opportunity. In the last several years, Medicare has added coverage for treatment of obesity related to treatment of medical conditions such as DM (2004), smoking cessation counseling in patients with diseases linked to tobacco (2005), cardiac rehabilitation services (2006), and bariatric surgery in obese patients with a comorbid condition related to obesity (2006).30

The study results show that individuals made health-habit changes in response to conditions in which the health habits are a risk factor and did not change in response to other conditions. Individuals quit smoking in response to diagnoses of stroke, cancer, lung disease, heart disease, and DM and lost weight in response to diagnoses of DM and heart disease, all conditions in which health habits are risk factors for developing the disease. In contrast, a new diagnosis of lung disease was not associated with weight change. Further analyses showed no association with weight change or smoking cessation after a diagnosis of arthritis or a psychiatric disorder, 2 additional conditions in which health habits are not risk factors (data not shown). These patterns could reflect clinician counseling about recommended health habits or increased concern by individuals about health risks from smoking or being overweight after receiving a new diagnosis.

This study has several issues to consider. The findings cannot rule out the possibility that health behavior changes precede new diagnoses rather than the reverse. This consideration seems less likely for acute conditions in which the timing of onset is difficult to predict. In addition, weight and smoking were self-reported, which could bias the results if individuals with new adverse health events systematically reported more positive behavior changes than individuals without new adverse health events. Use of self-reported data about health behaviors is common in the research literature, however.11,14,16,23 Furthermore, changes in question wording and sample size limitations prevented analysis of changes in other important health behaviors such as changes in exercise patterns and in alcohol consumption among heavy drinkers. However, this study did examine 2 key behaviors of interest, weight change and smoking cessation. This study cannot speak to the questions of which secondary prevention interventions would be most effective or whether primary vs secondary prevention efforts would be more cost-effective in older adults.

In conclusion, adverse health events were associated with losing weight and smoking cessation in middle-aged and older adults. At the same time, low overall rates of behavior change suggest that the health care system could do more to target the potential window of opportunity for individuals to adopt healthier habits after experiencing adverse health events. Smoking and overweight or obesity contribute to increased morbidity and mortality and to health care costs. Policy makers and clinicians might consider additional efforts to target information about adopting healthier habits as a component of secondary, as well as primary, prevention efforts.

Acknowledgments

Funding/Support: This study was supported by grants T32-AG00186 and R01 AG027045-02 from the National Institute on Aging, National Institutes of Health.

Footnotes

Author Contributions: Dr Keenan had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Financial Disclosure: None reported.

Additional Contributions: Robert Blendon, ScD, Daniel Carpenter, PhD, David Cutler, PhD, Peter Neumann, ScD, Joseph Newhouse, PhD, and Jody Sindelar, PhD, provided comments on previous versions of this article.

References

1. McGinnis JM, Foege WH. Actual causes of death in the United States. JAMA. 1993;270(18):2207–2212. [PubMed]
2. Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States, 2000. JAMA. 2004;291(10):1238–1245. [PubMed]
3. Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Correction: actual causes of death in the United States, 2000 [published correction appears in JAMA. 2004;291(10):1238–1245] JAMA. 2005;293(3):293–294. [PubMed]
4. National Center for Health Statistics. Health, United States 2007. Hyattsville, MD: National Center for Health Statistics; 2007.
5. US Department of Health and Human Services. The Surgeon General’s Call to Action to Prevent and Decrease Overweight and Obesity. Washington, DC: US Dept of Health and Human Services; 2001.
6. US Department of Health and Human Services. The Health Consequences of Smoking: A Report of the Surgeon General. Washington, DC: US Dept of Health and Human Services; 2004.
7. Fiore MC, Bailey WC, Cohen SJ, et al. Treating Tobacco Use and Dependence: Clinical Practice Guideline. Rockville, MD: US Dept of Health and Human Services; 2000.
8. US Preventive Services Task Force. Counseling to Prevent Tobacco Use and Tobacco-Caused Disease: Recommendation Statement. Rockville, MD: US Preventive Services Task Force, Agency for Healthcare Research and Quality; 2003. AHRQ No. 04–0526.
9. Burwen DR, Galusha DH, Lewis JM, et al. National and state trends in quality of care for acute myocardial infarction between 1994–1995 and 1998–1999: the Medicare Health Care Quality Improvement Program. Arch Intern Med. 2003;163 (12):1430–1439. [PubMed]
10. Jencks SF, Huff ED, Cuerdon T. Change in the quality of care delivered to Medicare beneficiaries, 1998–1999 to 2000–2001 [published correction appears in JAMA. 2002;289(20):2649] JAMA. 2003;289(3):305–312. [PubMed]
11. Mokdad AH, Bowman BA, Ford ES, Vinicor F, Marks JS, Koplan JP. The continuing epidemics of obesity and diabetes in the United States. JAMA. 2001;286(10):1195–1200. [PubMed]
12. Wee CC, McCarthy EP, Davis RB, Phillips RS. Physician counseling about exercise. JAMA. 1999;282(16):1583–1588. [PubMed]
13. Smith J. Consequences and Predictors of New Health Events. Cambridge, MA: National Bureau of Economic Research, Inc; Nov, 2003. NBER Working Paper 10063.
14. Clark A, Etilé F. Do health changes affect smoking? evidence from British panel data. J Health Econ. 2002;21(4):533–562. [PubMed]
15. Dale LC, Olsen DA, Patten CA, et al. Predictors of smoking cessation among elderly smokers treated for nicotine dependence. Tob Control. 1997;6(3):181–187. [PMC free article] [PubMed]
16. Falba T. Health events and the smoking cessation of middle aged Americans. J Behav Med. 2005;28(1):21–33. [PubMed]
17. Wagner EH, Curry SJ, Grothaus L, Saunders KW, McBride CM. The impact of smoking and quitting on health care use. Arch Intern Med. 1995;155(16):1789–1795. [PubMed]
18. Juster FT, Suzman R. An overview of the Health and Retirement Study. J Hum Resources. 1995;30(special issue):S7–S56. doi: 10.2307/146277. [Cross Ref]
19. RAND Center for the Study of Aging. RAND Health and Retirement Study Data, Version D. Santa Monica, CA: Rand Corp; 2004.
20. Heiat A, Vaccarino V, Krumholz HM. An evidence-based assessment of federal guidelines for overweight and obesity as they apply to elderly persons. Arch Intern Med. 2001;161(9):1194–1203. [PubMed]
21. Stevens J, Cai J, Pamuk ER, Williamson DF, Thun MJ, Wood JL. The effect of age on the association between body-mass index and mortality. N Engl J Med. 1998;338(1):1–7. [PubMed]
22. Willett WC, Dietz WH, Colditz GA. Guidelines for healthy weight. N Engl J Med. 1999;341(6):427–434. [PubMed]
23. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of US adults. N Engl J Med. 2003;348(17):1625–1638. [PubMed]
24. Centers for Medicare & Medicaid Services. [Accessed January 21, 2008];Hospital Compare. http://www.hospitalcompare.hhs.gov.
25. RAND Corp. Evidence report and evidence-based recommendations: interventions to promote smoking cessation in the Medicare population. Santa Monica, CA: RAND Corp; 2003. Prepared for the US Department of Health and Human Services, Health Care Financing Administration. Contract No. 500–98–0281.
26. Cromwell J, Bartosch WJ, Fiore MC, Hasselblad V, Baker T. Cost-effectiveness of the clinical practice recommendations in the AHCPR guideline for smoking cessation. Agency for Health Care Policy and Research. JAMA. 1997;278(21):1759–1766. [PubMed]
27. Curry SJ, Grothaus LC, McAfee T, Pabiniak C. Use and cost effectiveness of smoking-cessation services under four insurance plans in a health maintenance organization. N Engl J Med. 1998;339(10):673–679. [PubMed]
28. Doescher MP, Saver BG. Physicians’ advice to quit smoking: the glass remains half empty. J Fam Pract. 2000;49(6):543–547. [PubMed]
29. Williams SC, Schmaltz SP, Morton DJ, Koss RG, Loeb JM. Quality of care in US hospitals as reflected by standardized measures, 2002–2004. N Engl J Med. 2005;353(3):255–264. [PubMed]
30. Centers for Medicare & Medicaid Services. [Accessed January 6, 2008];National Coverage Analyses. http://www.cms.hhs.gov/mcd/index_list.asp?list_type=nca#closed.