Research ArticleOriginal Research

Predicting Incident Multimorbidity

Luke T. A. Mounce, John L. Campbell, William E. Henley, Maria C. Tejerina Arreal, Ian Porter and Jose M. Valderas
The Annals of Family Medicine July 2018, 16 (4) 322-329; DOI: https://doi.org/10.1370/afm.2271

Article Figures & Data

Figures

  • Figure 1
    Figure 1

    Prevalence of comorbidities on the x-axis for participants who have the condition on the y-axis, with associated relative risks (data from 2012-2013).

    COPD = chronic obstructive pulmonary disease; IHD = ischemic heart disease.

    Note: Conditions with a prevalence <5% were excluded (see Table 2). Relative risks are for having the comorbidity on the x-axis given the individual has the condition on the y-axis.

Tables

  • Table 1

    Participants’ Characteristics at Baseline (2002-2003) and Outcomes at End of Study (2012-2013)

    CharacteristicTotal SampleNo Condition at Baseline1 Condition at Baseline≥2 Conditions (Multimorbidity) at Baseline
    All eligible, No. (%)4,564 (100.0)1,477 (32.4)1,534 (33.6)1,553 (34.0)
    Multimorbidity by 2012-2013, No. (%)2,897 (63.5)377 (25.5)1,001 (65.3)NA
    ≥1 Additional condition by 2012-2013, No. (%)2,845 (62.3)901 (61.0)1,001 (65.3)943 (60.7)
    Age-group, No. (%)
     50-54 y1011 (22.2)462 (31.3)347 (22.6)202 (13.0)
     55-59 y1093 (24)392 (26.5)372 (24.3)329 (21.2)
     60-64 y792 (17.4)245 (16.6)272 (17.7)275 (17.7)
     65-69 y762 (16.7)182 (12.3)266 (17.3)314 (20.2)
     70-74 y518 (11.4)130 (8.8)158 (10.3)230 (14.8)
     75-79 y267 (5.9)48 (3.3)86 (5.6)133 (8.6)
     80-84 y103 (2.3)15 (1.0)27 (1.8)61 (3.9)
     ≥85 y18 (0.4)3 (0.2)6 (0.4)9 (0.6)
    Female, No. (%)2,570 (56.3)737 (49.9)856 (55.8)977 (62.9)
    Total wealth, median (IQR), £166,899.7 (82,689-303,170.1)190,500 (105,039.6-337,505)181,000 (893,45.5-315,573.5)134,742.6 (57,158.52-254,200)
    Wealth quintile, No. (%)
     1 Wealthiest896 (19.6)354 (24.0)329 (21.5)213 (13.7)
     2896 (19.6)290 (19.6)325 (21.2)281 (18.1)
     3896 (19.6)333 (22.6)284 (18.5)279 (18.0)
     4898 (19.7)258 (17.5)298 (19.4)342 (22.0)
     5 Least wealthy897 (19.7)214 (14.5)268 (17.5)415 (26.7)
     Missing81 (1.8)28 (1.9)30 (2.0)23 (1.5)
    Educational attainment, No. (%)a
     Undergraduate degree or higher1,342 (29.4)508 (34.4)451 (29.4)383 (24.7)
     Intermediate1,801 (39.5)601 (40.7)616 (40.2)584 (37.6)
     No qualifications1,421 (31.1)368 (24.9)467 (30.4)586 (37.7)
    Lives alone, No. (%)2,823 (61.9)921 (62.4)930 (60.3)972 (62.6)
    BMI category, No. (%)
     Underweight25 (0.6)10 (0.7)9 (0.6)6 (0.4)
     Normal weight1,086 (23.8)428 (29.0)371 (24.2)287 (18.5)
     Overweight1,835 (40.2)626 (42.4)623 (40.6)586 (37.7)
     Obese1,187 (26.0)293 (19.8)389 (25.4)505 (32.5)
     Missing431 (9.4)120 (8.1)142 (9.3)169 (10.9)
    Smoking behavior, No. (%)
     Never smoked1,772 (38.8)604 (40.9)613 (40.0)555 (35.7)
     Smoked in past2,101 (46.0)642 (43.5)696 (45.4)763 (49.1)
     Current smoker691 (15.1)231 (15.6)225 (14.7)235 (15.1)
    Physical activity, No. (%)b
     Sedentary162 (3.6)31 (2.1)44 (2.9)87 (5.6)
     Low877 (19.2)201 (13.6)267 (17.4)409 (26.3)
     Medium2456 (53.8)810 (54.8)849 (55.4)797 (51.3)
     High1067 (23.4)434 (29.4)374 (24.4)259 (16.7)
     Missing2 (0.0)1 (0.1)0 (0.0)1 (0.1)
    Social detachment, No. (%)c414 (9.1)113 (7.7)126 (8.2)175 (11.3)
     Missing395 (8.7)123 (8.3)128 (8.3)144 (9.3)
    External locus of control, No. (%)d2979 (65.3)906 (61.3)1000 (65.2)1073 (69.1)
     Missing268 (5.9)78 (5.3)82 (5.4)108 (7.0)

    • BMI = body mass index; IQR = interquartile range; NA = not applicable.

    • Note: Characteristics were assessed in 2002-2003, except for BMI, which was assessed in the 2004-2005 nurse visit. Covariates with no row for missing data had no missing data.

    • ↵a Educational attainment was categorized into higher education/degree, intermediate (secondary education/high school), or no qualifications.

    • ↵b Physical activity was assessed using methodology from the Allied Dunbar fitness survey.

    • ↵c Participants were considered “socially detached” if they were detached from at least 3 out of 4 assessed domains; civic participation, leisure activities, cultural engagement, and social networks (Supplemental Appendix, available at http://www.AnnFamMed.org/content/16/4/322/suppl/DC1/).

    • ↵d External locus of control means believing that life events are outside of one’s control.

  • Table 2

    Prevalence of Multimorbidity and Individual Conditions From 2002-2003 to 2012-2013 in the Full Sample (N = 4,564)

    Condition2002-20032004-20052006-20072008-20092010-20112012-2013
    Multimorbidity, No. (%)1,553 (34.0)1,885 (41.3)2,166 (47.5)2,445 (53.6)2,695 (59.1)2,897 (63.5)
     Mean (SD), No.1.21 (1.16)1.42 (1.26)1.62 (1.35)1.81 (1.41)2.01 (1.49)2.17 (1.53)
    Hypertension, No. (%)1,564 (34.3)1,844 (40.4)2,067 (45.3)2,230 (48.9)2,377 (52.1)2,472 (54.2)
    Ischemic heart disease, No. (%)391 (8.6)454 (10.0)512 (11.2)571 (12.5)627 (13.7)670 (14.7)
    Congestive heart failure, No. (%)16 (0.4)18 (0.4)22 (0.5)23 (0.5)33 (0.7)44 (1.0)
    Arrhythmia, No. (%)250 (5.5)350 (7.7)420 (9.2)495 (10.9)581 (12.7)663 (14.5)
    Diabetes, No. (%)241 (5.3)295 (6.5)379 (8.3)456 (10.0)530 (11.6)582 (12.8)
    Stroke, No. (%)112 (2.5)149 (3.3)174 (3.8)216 (4.7)258 (5.7)304 (6.7)
    COPD, No. (%)209 (4.6)257 (5.6)289 (6.3)328 (7.2)381 (8.4)427 (9.4)
    Asthma, No. (%)534 (11.7)591 (13.0)621 (13.6)656 (14.4)689 (15.1)707 (15.5)
    Arthritis, No. (%)1,364 (29.9)1,631 (35.7)1,848 (40.5)2,023 (44.3)2,179 (47.7)2,324 (50.9)
    Osteoporosis, No. (%)196 (4.3)277 (6.1)346 (7.6)419 (9.2)509 (11.2)582 (12.8)
    Cancer, No. (%)231 (5.1)305 (6.7)361 (7.9)415 (9.1)527 (11.7)604 (13.2)
    Parkinson’s disease, No. (%)5 (0.1)12 (0.3)20 (0.4)24 (0.5)31 (0.7)39 (0.9)
    Affective MHC,a No. (%)362 (7.9)270 (5.9)304 (6.7)334 (7.3)364 (8.0)381 (8.4)
    Psychotic MHC, No. (%)24 (0.5)26 (0.6)33 (0.7)37 (0.8)45 (1.0)53 (1.2)
    Dementia, No. (%)12 (0.3)15 (0.3)19 (0.4)28 (0.6)41 (0.9)62 (1.4)

    • COPD = chronic obstructive pulmonary disease; MHC = mental health condition.

    • ↵a We ascertained at each wave whether participants still had their affective mental health condition; hence, rates may fluctuate across time.

  • Table 3

    Effect of Baseline Characteristics on Time to Developing Incident Multimorbidity Between 2002-2003 and 2012-2013: Complementary Log-Log Models

    CharacteristicNo Condition in 2002-2003 (n = 1,477) and ≥2 Incident Conditions (Multimorbidity) by 2012-2013Any Number of Conditions in 2002- 2003 (n = 4,564) and ≥1 Incident Conditions by 2012-2013
    HR (95% CI)P ValueHR (95% CI)P Value
    Age, yLinear trend<.001Linear trend<.001
     50-541.00 [Reference]…1.00 [Reference]…
     55-591.44 (1.05-1.99)a.025a1.29 (1.14-1.45)a<.001a
     60-641.85 (1.31-2.61)a<.001a1.42 (1.25-1.62)a<.001a
     65-692.93 (2.08-4.13)a<.001a1.49 (1.31-1.70)a<.001a
     ≥702.58 (1.83-3.64)a<.001a1.49 (1.31-1.70)a<.001a
    Female (vs male)1.14 (0.86-1.50).3601.10 (0.99-1.21).072
    Wealth quintilesLinear trend.001aLinear trend.002a
     1 Wealthiest1.00 [Reference]…1.00 [Reference]…
     21.27 (0.89-1.80).1850.97 (0.85-1.10).620
     31.41 (1.00-1.97).0491.08 (0.95-1.23).215
     41.23 (0.85-1.78).2691.11 (0.97-1.26).120
     5 Least wealthy2.19 (1.50-3.19)a<.001a1.22 (1.06-1.39)a.005a
    Educational attainmentLinear trend.058Linear trend.923
     Degree /higher ed1.00 [Reference]…1.00 [Reference]…
     Intermediate0.95 (0.73-1.23).7011.04 (0.94-1.15).413
     No qualification0.73 (0.53-1.00).0511.00 (0.89-1.12).993
    Lives alone (vs cohabits)0.93 (0.71-1.21).5801.08 (0.98-1.19).128
    BMI category (2004)Linear trend<.001aLinear trend<.001a
     Underweight or normal weight1.00 [Reference]…1.00 [Reference]…
     Overweight1.15 (0.87-1.52).3391.10 (0.99-1.21).070
     Obese1.92 (1.43-2.59)a<.001a1.27 (1.14-1.42)a<.001a
    Smoking behavior
     Never smoked1.00 [Reference]…1.00 [Reference]…
     Smoked in past1.12 (0.88-1.43).3591.09 (1.00-1.19).050
     Current smoker1.22 (0.87-1.70).2481.21 (1.07-1.36)a.002a
    Physical activityLinear trend.031aLinear trend.004a
     High1.00 [Reference]…1.00 [Reference]…
     Medium1.3 (1.00-1.70).0511.08 (0.97-1.19).149
     Low1.43 (1.02-2.00)a.039a1.19 (1.06-1.35)a.004a
    Social detachment (vs none)1.16 (0.77-1.72).4801.07 (0.93-1.22).360
    External locus of control (vs internal)1.41 (1.10-1.82)a.007a1.13 (1.03-1.23)a.010a

    • BMI = body mass index; HR = hazard ratio.

    • Note: Complementary log-log models are the discrete time equivalent of Cox proportional hazards models for continuous time. The reference category for each covariate is the first category listed. A HR >1.00 indicates increased risk, whereas an HR of <1.00 indicates reduced risk. Models were corrected for differential nonresponse using longitudinal sample weighting.

    • ↵a Significant association.

  • Table 4

    Relative Association Between the Presence of Individual Conditions on Time to Developing Incident Multimorbidity Between 2002-2003 and 2012-2013: Complementary Log-Log Model

    ConditionHR (95% CI)P Value
    Hypertension1.00 [Reference]…
    Ischemic heart disease1.27 (0.91-1.78).162
    Arrhythmia1.55 (1.06-2.26)a.024a
    Diabetes mellitus1.06 (0.63-1.78).839
    COPD2.32 (1.55-3.46)a<.001a
    Asthma1.33 (1.05-1.70)a<.019a
    Arthritis (any)0.98 (0.83-1.16).819
    Osteoporosis1.32 (0.87-2.01).185
    Cancer (any)1.19 (0.86-1.63).295
    Affective mental health condition (any)0.97 (0.71-1.31).831
    Otherb1.21 (0.64-2.30).552

    • COPD = chronic obstructive pulmonary disease; HR = hazard ratio.

    • Note: From a sample of participants having 1 condition in 2002-2003 (n = 1,534), and an outcome of 1 or more incident conditions. Complementary log-log models are discrete time equivalent of Cox proportional hazards models for continuous time. This model adjusted for all baseline patient characteristics, and differential nonresponse using longitudinal sample weighting. HR >1.00 represents increased risk; an HR of <1.00 indicates reduced risk.

    • ↵a Conditions significantly more more likely than hypertension to be associated with incident multimorbidity.

    • ↵b Includes participants with conditions with low prevalence; congestive heart failure, stroke, psychotic mental health condition, and dementia/Alzheimer’s (Table 2). Combined, these individuals in the “other” category accounted for <1% of the sample.

Additional Files

  • Supplemental Appendix

    Supplemental Appendix

    Files in this Data Supplement:

  • The Article in Brief

    Predicting Incident Multimorbidity

    Jose M. Valderas , and colleagues

    Background If multimorbidity (multiple medical conditions in an individual) is to be prevented, it is important to identify the characteristics that contribute to it.

    What This Study Found Results of a 10-year study find that, for adults age 50 and older, risks of developing multimorbidity are positively associated with age and are higher for those with low socioeconomic status, obesity, low level of physical activity, or an external locus of control (believing that life events are outside of their control). No significant associations were observed for sex, educational attainment, or social detachment.

    Implications

    • The authors suggest that future work to reduce the incidence of multimorbidity should promote healthy lifestyles while targeting an internal locus of control in order to empower patients to achieve and maintain behavior change with the potential for synergistic effects.
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Get Permissions

Jump to section

Related Articles

Cited By...

More in this TOC Section

Similar Articles

Keywords