Examining the association between pain severity and quality-of-life, work-productivity loss, and healthcare resource use among European adults diagnosed with pain

Figures & data

Table 1. Demographic characteristics.

	Diagnosed pain(n = 14 459)		No pain diagnosis(n = 47 541)		Total NHWS(n = 62 000)
	%	n	%	n	%	n	ϕc
Age							0.10
18 to <25	5.7%	823	8.0%	3822	7.5%	4,645
25 to <35	15.0%	2166	21.4%	10,153	19.9%	12,319
35 to <45	17.7%	2561	19.2%	9139	18.9%	11,700
45 to <55	23.1%	3337	18.9%	8968	19.8%	12,305
55 to <65	19.1%	2756	14.1%	6692	15.2%	9448
65+	19.5%	2816	18.4%	8767	18.7%	11,583
% Female	60.1%	8691	51.8%	24,614	53.7%	33,305	0.07
% Married	63.7%	9211	62.3%	29,595	62.6%	38,806	0.01
% University Degree	34.1%	4928	41.2%	19,577	39.5%	24,505	0.06
Income (Euros)							0.06
<20k	29.8%	4316	24.4%	11,605	25.7%	15,921
20k to 50k	44.6%	6445	45.0%	21,410	44.9%	27,855
50k+	13.8%	1990	16.6%	7877	15.9%	9867
Decline	11.8%	1708	14.0%	6649	13.5%	8357
Body Mass Index (BMI)							0.12
underweight	2.7%	385	3.1%	1472	3.0%	1857
normal weight	37.0%	5349	46.3%	22,001	44.1%	27,350
overweight	32.7%	4728	32.4%	15,407	32.5%	20,135
obese	25.2%	3638	15.2%	7241	17.5%	10,879
unknown	2.5%	359	3.0%	1420	2.9%	1779
Smoking Status							0.09
Current	29.7%	4288	23.6%	11,236	25.0%	15,524
Former	34.1%	4933	30.3%	14,392	31.2%	19,325
Never	36.2%	5238	46.1%	21,913	43.8%	27,151
% Currently Drink Alcohol	76.2%	11,022	78.0%	37,092	77.6%	48,114	0.02
% Currently Exercise	53.6%	7754	61.5%	29,251	59.7%	37,005	0.07
Charlson Comorbidity Index (CCI)							0.19
0	67.5%	9761	84.5%	40,180	80.6%	49,941
1	20.4%	2948	11.1%	5254	13.2%	8202
2	7.1%	1021	3.0%	1418	3.9%	2439
3+	5.0%	729	1.4%	689	2.3%	1418
Pain Severity
0 No Pain	5.6%	808	–	–	–	–
1	6.4%	920	–	–	–	–
2	9.5%	1369	–	–	–	–
3	11.0%	1597	–	–	–	–
4	9.4%	1364	–	–	–	–
5	12.7%	1840	–	–	–	–
6	12.8%	1844	–	–	–	–
7	15.0%	2162	–	–	–	–
8	11.1%	1610	–	–	–	–
9	4.3%	618	–	–	–	–
10 Pain as bad as you can imagine	2.3%	327	–	–	–	–

All comparisons between the “Diagnosed Pain” and “No Pain Diagnosis” groups were statistically significant at p < 0.001.

ϕ_c, Cramer’s V effect size estimate.

Table 2. Correlations between demographics and severity.

	Pain the past week (r)	VIF
Age	0.103	1.126
% Female*	−0.093	1.065
% Married*	−0.003	1.063
% University Degree*	−0.094	1.047
Income (Euros)	−0.075	1.066
BMI	0.114	1.070
Smoking Status	−0.049	1.046
% Currently Drink Alcohol*	−0.116	1.039
% Currently Exercise	−0.126	1.066
CCI	0.137	1.056

All correlations are significant at p < 0.01.

* Point Bi-serial correlation, all other correlations are Pearson correlations.

BMI, Body Mass Index; CCI, Charlson Comorbidity Index.

Figure 1. Predicted health-related quality-of-life. Note: p-values reported are for highest order significant interaction term. These are cubic (health utilities), quadratic (mental component summary), and quadratic (physical component summary), respectively. All models included the following variables as covariates: age, gender, education, and CCI score (comorbidity burden).

Figure 2. Predicted work productivity impairment by pain severity (past 6 months). Note: p-values reported are for highest significant interaction term. These are linear (absenteeism) and quadratic (presenteeism), respectively. All models included the following variables as covariates: age, gender, education, and CCI score (comorbidity burden).

Figure 3. Predicted healthcare resource use by pain severity (past 6 months). Note: p-values reported are for the highest significant interaction term. These are quadratic (ER visits), linear (hospital visits), and linear (primary care provider visits), respectively. All models included the following variables as covariates: age, gender, education, and CCI score (comorbidity burden).

Table 3. General form of equations.

Health outcome	Equation
Health utilities†
Mental component summary†
Physical component summary†
Absenteeism†
Presenteeism†
ER visits†
Hospital visits†
Primary care provider visits†

Means for each variable in the models are as follows: age =49.04, male =0.40, education =0.34, CCI =0.56.

† Final terms were only included if they were statistically significant.

Table 4. Unstandardized regression coefficients (b values) from multiple regression models by outcome.

Outcome	Intercept	bage	bmale	beducation	bCCI	bseverity	bseverity^3	bseverity^3
PCS	52.58	−0.14	−0.24	1.41	−1.51	−1.86	−0.10
MCS	33.72	0.20	0.99	0.76	−0.96	−1.25	−0.03
Health utilities	0.63	0.00	0.01	0.01	−0.02	−0.02	−0.00	−0.00
Absenteeism	2.56	−0.01	0.04	−0.24	0.21	0.21
Presenteeism	3.60	−0.01	0.01	−0.07	0.12	0.18	−0.01
ER visits	−0.46	−0.02	0.07	0.14	0.27	0.14	0.01
Hospital visits	−1.69	−0.01	0.16	0.05	0.33	0.13
PCP visits	1.83	0.00	−0.13	−0.03	0.17	0.08

Coefficients reported are from the regression model with the highest-order significant interaction term. Therefore, models that do not have a coefficient for b_severity³ or for b_severity³ are models in which those terms were not significant and, thus, were not included in the final model.

Table 5. Adjusted/estimated means for all study outcomes by level of pain severity in the past week.

Pain severity score in the past week	Physical component summary (PCS)	Mental component summary (MCS)	Health utilities	Absenteeism	Presenteeism	No of ER visits	No of hospital visits	No of PCP visits
0	45.16	43.48	0.64	8.3%	25.9%	0.26	0.18	7.29
1	43.20	42.21	0.62	10.3%	30.7%	0.30	0.20	7.88
2	41.05	40.87	0.59	12.6%	35.7%	0.35	0.23	8.53
3	38.69	39.48	0.57	15.5%	40.6%	0.43	0.26	9.22
4	36.13	38.02	0.53	19.1%	45.3%	0.53	0.29	9.97
5	33.36	36.51	0.50	23.6%	49.4%	0.66	0.33	10.79
6	30.40	34.94	0.46	29.0%	52.8%	0.85	0.38	11.67
7	27.24	33.31	0.42	35.7%	55.2%	1.11	0.43	12.62
8	23.87	31.62	0.37	43.9%	55.6%	1.48	0.49	13.65
9	20.31	29.87	0.31	54.1%	56.8%	2.02	0.55	14.76
10	16.54	28.07	0.25	66.6%	55.7%	2.81	0.63	15.97

Values reported are adjusted/estimated means for each outcome by pain severity score controlling for age, gender, education, and CCI score.