Free Medical Journals: Health status as a risk factor in cardiovascular disease: A systematic review of current evidence

Health status as a risk factor in cardiovascular disease: A systematic review of current evidence

Paula M.C. Mommersteeg PhD^a^,, Johan Denollet PhD^a, John A. Spertus MD, MPH^b and Susanne S. Pedersen PhD^a^,^,^,

^aCenter of Research on Psychology in Somatic Diseases (CoRPS), Tilburg University, Tilburg, The Netherlands

^bMid-America Heart Institute and University of Missouri–Kansas City, Kansas City, Missouri

Received 25 July 2008;

accepted 26 September 2008.

Available online 19 December 2008.

Background

Patient-perceived health status is receiving increased recognition as a patient-centered outcome in chronic heart failure (CHF) and coronary artery disease (CAD), but poor health status is also associated with adverse prognosis. In this systematic review, we examined current evidence on the influence of health status on prognosis in CHF and CAD.

Methods

We conducted a search of PubMed using a set of a priori–defined search terms, the Web of Science for newly cited articles, and the reference lists of eligible articles, resulting in 34 articles.

Results

Poor physical health status was a significant predictor for adverse health outcomes in patients with CHF and CAD. In CHF, poor physical health status seemed to be a stronger predictor of hospitalization than mortality. Little evidence was found that poor mental health status is associated with adverse prognosis in CHF and CAD. A disease-specific measure was a better predictor in CHF, but not in CAD. The majority of studies adjusted for an objective measure of disease severity. Neither the index event nor time to follow-up appeared to influence the predictive value of health status.

Conclusions

Poor physical health status is associated with adverse CAD and CHF prognosis. Heterogeneity across studies makes definitive conclusions difficult as to which components of health status may be detrimental to patients' health, and how health status as a potential risk factor should be assessed, monitored, and intervened upon in clinical practice.

Article Outline

Methods
Results
Discussion
References

Health status is receiving increased recognition as an important patient-centered outcome in patients with cardiovascular disease (CVD).¹ In patients with established disease, health status refers to “the range of manifestation of disease in a given patient including symptoms, functional limitation, and quality of life, in which quality of life is the discrepancy between actual and desired function.”² Self-rated health, encompassing health status and quality of life, has been associated with both incident fatal and nonfatal heart disease in population-based studies³ and with hospital readmissions and mortality in patients with CVD.^[4]^,^[5]^,^[6]^,^[7]^and^[8]

Patients' evaluation of their health status does not necessarily concur with that of physicians,^[9]^and^[10] with physicians often failing to identify functional disabilities reported by patients.⁹ Patient-reported health status also seems to be a risk factor for mortality in patients with CVD of equal or greater magnitude than traditional biomedical risk factors.⁸ Taken together, this suggests that knowledge of health status, as perceived by the patient, may both provide a unique window into patients' clinical status and be prognostically important, such that it can enhance secondary prevention interventions as a relatively low-cost assessment with which to identify high-risk patients.

In a 2002 editorial, Rumsfeld² suggested that health status should be added to the amalgam of clinical factors used for risk stratification in clinical practice. Six years later, there is little evidence that health status measurements have become part of standard assessment, even though its routine measurement has been established as a performance measure of high-quality care by the American College of Cardiology, the American Heart Association, the American Medical Association,¹¹ and the National Quality Forum (http://www.qualityforum.org/projects/ongoing/ambulatory). One of the critical barriers to the adoption of health status assessments in clinical care could be ambivalence about its utility in risk stratification. To better clarify the potential role of routine health status assessment, the objective of this systematic review was to evaluate the evidence on the influence of health status on prognosis in CVD patients and describe the implications for clinical practice and future research.

Methods

PubMed was searched in the period between January 1980 and August 2007, using the following search terms: Quality of life, health status, health status indicators, functioning, coronary artery disease, CAD, heart failure, cardiovascular disease, CVD, CABG, percutaneous coronary intervention, PTCA, myocardial revascularization, coronary artery bypass, patient readmission, hospitalization, mortality, survival analysis, and prognosis. The search was limited to articles published in peer-reviewed English medical journals in adults (>18 years). Studies were included in the review if they focused on patients with established coronary artery disease (CAD) or chronic heart failure (CHF), used a standardized measurement of quality of life or health status, or asked patients to rate their health using a purpose-designed question, and analyzed the association between health status and (all-cause) mortality, hospital readmission, or a combination of mortality and readmission. Studies using a mixed sample of patients with CAD or CHF with other disease entities (eg, stroke, diabetes, or heart transplantation) focusing on hospital length of stay as the only outcome were excluded.

Figure 1 provides an overview of the search and selection of articles. The title and abstract of the 1,561 articles were reviewed by one author (P.M.C.M.), whereas the 52 articles meeting the inclusion criteria were reviewed by the first and last author (P.M.C.M. and S.S.P.). An additional search on Web of Science for newly cited articles based on the shortlist (forward citation, November 2007) led to 4 additional articles.^[12]^,^[13]^,^[14]^and^[15] A hand search of the reference list of these 31 articles identified 3 additional articles,^[16]^,^[17]^and^[18] with the review being based on 34 articles. Although the initial objective of this study was to perform a meta-analysis, this was not possible given the heterogeneity between studies.

Full-size image (33K)

Figure 1. Flowchart of the selection of studies.

Description of studies

The 34 identified articles (Table I) were all prospective cohort studies that used self-report measures of health status at baseline, or within a few days (72 hours) to weeks after the index event (ie, hospitalization for CHF, myocardial infarction [MI], percutaneous coronary intervention [PCI], or coronary artery bypass graft surgery [CABG]). In some instances, CHF outpatients were also included. Two studies used retrospective information on health status.^[13]^and^[19]

Table I.

Study descriptives and confounders

Reference	Ref. no.	n	Population	Quality of life	Outcome measure	Follow-up length	Confounders
							Demographic		Disease severity				Comorbidity
							Age/Sex	Smoking	Hypertension^‡	Disease severity^§	Hypercholesterolaemia	Previous hospitalization^¶	Renal functioning^#	Diabetes	Comorbidity^††	Medication use^‡‡
Alla et al, 2002	[38]	108	CHF	HRQoL	M, H	18 m	A/S			DS	Hypercholesterolaemia	PH	RF		CM	M
Bosworth et al, 1999	[24]	2885	CAD, CABG, PCI	SR-health, SF36, DASI	M	3.5 y	A/S	Smoking	Hy	DS			RF	Diabetes	CM
Bouvy et al, 2003	[16]	152	CHF	MLHFQ	M	18 m	A/S		Hy	DS		PH	RF	Diabetes	CM	M
Chocron et al, 2000	[17]	209	CABG	NHP	M	3 y	A/S			DS		PH	RF	Diabetes	CM
Curtis et al, 2002	[25]	1778	CABG	SF36	M	1 m	A/S		Hy	DS	Hypercholesterolaemia	PH	RF	Diabetes	CM
Deaton et al, 1998	[26]	93	CABG	HSQ12	H	3 m	A/S	Smoking	Hy	DS		PH		Diabetes
Dixon et al, 2001	[31]	945	MI, CHF, IHD	MacNew	M/H	12 m	A/S	Smoking	Hy			PH			CM
Faller et al, 2007	[12]	231	CHF	SF36, KCCQ	M	2.7 y	A/S		Hy	DS	Hypercholesterolaemia		RF	Diabetes		M
Formiga et al, 2006	[13]	88	CHF	ADL	M/H	12 m	A/S	Smoking	Hy	DS	Hypercholesterolaemia	PH	RF	Diabetes
Havranek et al, 2001	[40]	545	CHF	SR-health	M/H	2.8 m	A/S			DS						M
Heidenreich et al, 2006	[34]	505	CHF	KCCQ	M/H	12 m	A/S		Hy	DS		PH	RF	Diabetes	CM	M
Herlitz et al, 1998	[27]	1290	CABG	NHP	M	5 y	A/S	Smoking	Hy	DS		PH	RF	Diabetes	CM
Ho et al, 2005	[14]	3160	CABG, valve replacement	SF36	M	6 m	A/S	Smoking	Hy	DS		PH	RF	Diabetes	CM
Hülsmann et al, 2002	[15]	96	CHF	MLHFQ	M/H	12 m	A/S			DS						M
Koch et al, 2007	[4]	6305	CABG	DASI	M	8.6 y	A/S	Smoking	Hy	DS		PH	RF	Diabetes	CM
Konstam et al, 1996	[35]	5025	CHF	HRQoL	M, H	3.0 y	A/S		Hy	DS		PH		Diabetes	CM
Kosiborod et al, 2007	[5]	1358	MI, CHF	KCCQ	M, M/H	14 m	A/S		Hy	DS	Hypercholesterolaemia	PH	RF	Diabetes	CM	M
Lenzen et al, 2007	[28]	3786	CAD, MI, CABG, PCI	EQ-5D	M	12 m	A/S	Smoking	Hy	DS	Hypercholesterolaemia	PH		Diabetes		M
Lim et al, 1998	[32]	375	MI	MacNew	M/H	18 m	A/S	Smoking	Hy	DS		PH				M
Mayer et al, 2003	[19]	123	CABG	SF36	M	18 m	A/S		Hy				RF	Diabetes	CM
Mejhert et al, 2004, 2006	[20] and [21]	208	CHF	NHP	M	3.1 y	A/S		Hy	DS		PH	RF	Diabetes	CM	M
Mozaffarian et al, 2003	[33]	8908	CAD	SAQ	M	24 m	A/S	Smoking	Hy	DS		PH	RF	Diabetes	CM
Murberg et al, 1999	[36]	119	CHF	HRQoL	M	24 m	A/S		Hy	DS		PH				M
Pedersen et al, 2007	[6]	667	PCI	MacNew	M/H	<6>	A/S	Smoking		DS		PH		Diabetes	CM	M
Piotrowicz et al, 2007	[23]	1058	MI	SF12	M, H and M/H	3.0 y	A/S	Smoking	Hy	DS				Diabetes		M
Rodriguez-Artalejo et al, 2005	[7]	394	CHF	SF36, MLHFQ	M, H	6.6 m, 6.2 m	A/S			DS		PH	RF	Diabetes	CM	M
Rumsfeld et al, 1999	[29]	2480	CABG	SF36	M	6 m	A/S	Smoking	Hy	DS		PH	RF	Diabetes	CM	M
Schwarz et al, 2003	[39]	156	CHF	FS	M, H	3 m	A/S		Hy	DS		PH	RF	Diabetes	CM
Soto et al, 2004	[22]	1516	MI	KCCQ	M, M/H	12 m	A/S	Smoking	Hy	DS	Hypercholesterolaemia	PH	RF	Diabetes	CM	M
Spertus et al, 2002	[8]	4484	CAD	SAQ	M, H	12 m	A/S	Smoking	Hy	DS		PH	RF	Diabetes	CM
Stull et al, 2001	[18]	3884	CHF	HRQoL, PsyQoL	H	3.1 y	A/S		Hy	DS		PH	RF	Diabetes	CM
Subramanian et al, 2007	[37]	459	CHF	SF36, CHQ, KCCQ	M	5 y	A/S		Hy	DS	Hypercholesterolaemia	PH	RF	Diabetes	CM	M
Westin et al, 2005	[30]	349	MI, CABG, PCI	HRQoL	M	10 y	A/S	Smoking	Hy		Hypercholesterolaemia					M

Full-size table

ADL, Activity of daily living; CHQ, McMaster Chronic Heart Failure Questionnaire; DASI, Duke Activity Scale Index; EQ-5D, EuroQoL; FS, functional status; HSQ12, health status questionnaire 12; HRQoL, health-related quality of life; IHD, ischemic heart disease; KCCQ, Kansas City Cardiomyopathy questionnaire; MacNew, MacNew heart disease quality of life instrument; MLHFQ, Minnesota living with heart failure; NHP, Nottingham Health Profile; PsyQoL, Psychosocial quality of life; SAQ, Seattle Angina Questionnaire; SF-12/36, Short Form Health Survey 12/36 items; SR-health, Self-rated health.

M, Mortality; H, hospitalization; M/H, mortality or hospitalization.

^‡Hy, Hypertension or systolic blood pressure, diastolic blood pressure.

^§DS, Disease severity; left ventricular ejection fraction, multivessel disease, Killip Class, NYHA class, or angina severity.

^¶PH, Previous hospitalization or event.

^#RF, Renal functioning; creatinine.

^††C, Comorbidity other, eg, COPD, PI, or Charlson comorbidity index.

^‡‡M, Medication use.

Sample sizes and populations

The median number of patients included in the studies was 545 (mean ± standard deviation 1,628 ± 2,119), with a range from 88 to 8,908. The sample sizes are depicted in Figure 2.

Full-size image (10K)

Figure 2. Frequency curve of the sample sizes for the included studies (n = 33) (study no. 20/21 is included once).

To detect potential differences between types of cardiovascular disease (CVD), we stratified studies by CAD and CHF (Table II and Table III). In total, 47% (n = 16) of the studies focused on patients with CAD. Although 1 study excluded patients with MI or angina complaints,^[20]^and^[21] three studies combined CHF with MI,^[5]^,^[22]^and^[23] which are listed under CHF. The index event was CABG in 33% (n = 11) of the studies,^[4]^,^[14]^,